S31?13S ?WNIJISS IG3 .V ~ ~ ~ ~-.. .-...-. [ep Slo 4Npul t.l x lUJaA a aa % 9 _'I ' V. 2 t t t e V '7i-, V -UOIL,Uug NUL19~~~~~~ I--P O - - 1 asuq:)inuumep - -uauzdztzraGr insriqx <42Ot :7:2 risy l.I1UOISV juAJOAQdI aATiOllV 04p S Vt~~~~~~~~Vftu uc E -lp EDI SEMINAR SERS The Desigrn and Management of Sustainable Projects to Alleviate Poverty in South Asia Collected Papers from an EDI Seminar Held in Bangalore, India, July 31 to August 9, 1991 Edited by Michael Bamberger Abdul Aziz The World Bank Washingtn, D. C. I~~~~~~~~~~~~~~~~~~~~ 01993 The International Bank for Recontrucion and Development I THE WORLD BANK 1818 H Street, N.W. WashinBton, D.C. 20433, U.S.A. All nghts reserved Manufactured in the United States of Amica irrSt printig August 1993 The EconomicDevelopmentInstitute (EDI) wasestablised by the World Bankin 1955 to trainofficialsconcncdwift development planning, polcymaking. invatment nalysi and project inplemetation in memberdeveloping countries. At present the subsAance of the EDrs wwork emphasiza macroeconomic and wctorao economic poficy analysis. hrogli a variety of courses. seminars, and workshops, must of wbich are given overseas in coopemtion with local instutions, the SDI seeks to sharpen analytical skills used io policy aa:dysis and to broaden undwasnding of th experience of individual countries with economicdewvepmentAlthough theEDrspublicatiousaredesigpod tosuppoztits training activities many are of interest to a much broader audience. EDI materials, including any finding, interpretaions, and conclusions, are entily those of the authors and should not be auributed m any manner to the World Banir, to its affiliated organizations. or to members of its Board of Execufive DirectDrs or the countries they represent Becmuseoftlleinformalityoftbisseriesandtomalethepublicainavailablewihthtelestpossibledelay,theumasscript has not been edited as fully as would be thecase withamore formaldocument, and the WorldBankacpts no responsibilit for errom The material in this publicaton is copyrighted. Requests forprwssion to reproduce portions of it shoud be sent to the Officeof the Publisher atthe address sbown in ihecopy3gbtnotice above Tbe World Bank encourages dieinion f its work and will normally give permision promptly and. wben the reproduction isfornoncommercial purposes, without asking afee. Pemmission to copyportions forclassroomuseisgranted through theCopyrightClearance Center. 27Congress Street, Sam, Massachuses 01970, U.S.A ThecompletebacklistofpublicationsfromtheWoridBankis shown iatheannualndexofPfNlauio. which contains an alphabetcal title list (with full ordering information) and indexes of suibjects, auos, and counries and regions. The laesteditionis available freecfcargefrotbe Distributior.Unit, Office ofthe Publisher, TbeWorldBank. 1818H Street. N.W., Washington, D.C. 20433. U.SA.. or from Publicatons, Banque mondiale, 66, avenue d'Idna, 75116 Paris, France. Mihael Bawberger is senior sociologist in the Human Resources Diviion of tneWorld Bank's Economic Development Institute Abdul Aziz is a professor and head of the Economics Unit at the hnsttute for Social and Eci.omic Change. Bangaore, South India. Library of Congress Cataogngia-Publieaton Date The design and management of sustainable projects to aleviate poverty in South Asia: collected papers fom an EDJ seminarheld in Bangalore, India. July31 to August 9, 1991 1 edited byMic, ; Bamberger., Abdul Aziz. p. cm-D-(EDI seminar series, ISSN 1013-2015) ISBN 0-8213-2472-1 1. Poor-,South Asia-Congresses. 2. Income distribution-South Asia-Congresses. 3. South Asia-Economic policy-Congresses. 4. Sustainable development-South Asia-Congresses. L Bamberger. Michael IL AziZ, AbduL 1937- . I. Series HC430.629P613 1993 33E96S-dC20 93-237 CD N EDI Catalog No. 520/1009 CONTENTS Foreword vii Preface ix Executive Summary xi PART 1. POVERTY AND STRATEGIES FOR POVERTY ALLEVIATION 1. Poverty Alleviation Strategies and the Role of Targeted Programs for the Poor Micdael Bamberger and Abdid Aziz 3 2. The Asian Poverty Senaro: Perspectives from the 1990 and 1992 World Development Reports Miciael Bamberger 19 PART 2. SECTOR EXPERIENCE WITH POVERTY AILEVIATION IN ASIA 29 Editor's note 31 3. The Accessibility of Population, Health, and Nutrion Services to the Poor: -The Experience of South Asia J.K Satia 35 4. Rural Development and Poverty Alleviaion The Experience of the Asian Development Bank Dieter Buricer and Keih Johnson 61 5. The Urban Poor and Urban Basic Infastucture Services in Asia: Past Approaches and Emerging Chailenges Yue-Man Yeung 77 6. The Role of Credit in Poverty Alleviation: The Asian Experience Ismael GCUg, Jr. 103 7. Gender lssues in Anti-Poverty Programs m Asia: Experiences and Issues Noeleen Heyzer 127 ii PART 3. DECENTRALZATION, TARGETED POVERTY PROJECTS, AND SURVIVAL STRATEGIES OF POOR HOUSEHOLDS: LESSONS FROM KARNATAKA 149 8. Targeted Poverty Programs for the Poor: Lessons from Project Visits AbdulAziz 151 9. Decentmlized Planning and Poverty Alleviation: The Kamataka (India) Experience AbdzdAziz 163 10. Survival Stategies of Poor Households: A Comparison of Kanmatal and Tamil Nadu with Expeience from Asia, Latin America, and Africa Abdul Aziz and Michael Ramberger 177 PART 4. GUIDELINES FOR THE SELECTION, DESIGN, IMPLEMENTATION AND SUSTAINABILITY OF POVERTY ALLEVIATION PROGRAMS 199 Editor's Note 201 11. The Interface between Poverty Projects and the Policy Environment AbdulAziz 203 12. Identification and Design of Poverty Reduction Projects Neil O'Sulvan 211 13. Implementing Poverty Reduction Projects Neil 0'5ulv z 235 14. Designing Gender-ensitive Poverty Reduction Projects Neil O'SWfivan 247 15. Susainabiity of Poverty Alleviation Projects Neil O'SuUl v0a 261 PART 5. SUMMARY AND CONCLUSION'S 271 16. Sectoral Expiences in Poverty Alleviaion Midcae Banbffger 273 17. Conclusions and P i ons: Improving the Performance of Povery Alleviation Projects Michad Bamberger and Abdad Aziz 289 About the Authors 297 iv -:7-..~ - - .-.... . 'L ~~~~ ~ ~~~ ~~~~~~~~~~~~~~~ ->~. t;4 ; ---* Anx 1: Seminar Participants and Resource Persons 299 Ann 2: Seminar Progrm 303 References 311 v FOREWORD This document is one of a series being produced under EDI's training program on poverty alleviatiox These reports present the findings and cenclusions of semiars and workshops at which policymnaks, planners, and managers from govermental and nongovermmental organizations have reviewed their exeriences with policies and programs designed to benefit poor and vulnerable groups and hep teem integrate into the mainstream of economic and social development. The present document reports on a seina held in Bangalor. in August 1991 to review the expeience of South Asia in designing and managing sustinble poverty alleviation projects. The seminar was attended by country teams from India, Bangladesh, Pakistan, Nepal, and Sri i anka. Amnon Golan, Director Economic Development Institute i%~. 2fl 3t-sA ~ / PREFACE This report is the product of a South Asia regional semina, "The Design and Management of Sustinable Poverty Alleviation Projects," which was organized as part of EDI's training program on poverty alleviation. The seminar took place in the Institute for Social and Economic Change (ISECQ in Bangalore, ndia, from July 31 to August 9, 1991, and was jointly sponsored by the Economic Development Institute (EDI) of the World Banlk and the Asian Development Bank (ADB). Additional support was provided by the Austrlian Interational Development Assistance Bureau (AIDAB). Seminar discussions reflected the growing concern of donor agencies and goverments about the need for more cost-effective and equitable methods of increasing the participatxon of the poor in the benefits of social and econonmc development. As is widely ackmowledged, the path to this goal lies in improving the design and implementation of poverty programs and projects. Any such endeavors must obviously iclude traing for policymakers and for those who implement such projects. The seminar consisted of three parts: different counties in Asia first shared their experiences with poverty alleviation programs and projects; next, participants visited project sites; and third, they held a workshop on how to improve the design and management of poverty alleviation projects and programs. Participants presented country papers, while experts in differen fields offered review papers on the experience to date with the design and management of sectoral poverty projects anc programs. Before participants visited the project sites, ISEC faculty presented six project profiles, based on extensive field surveys. A separate ISEC publication contis the project profiles and the conclusions of the project visits. The country experiences and sector review papers were presented during the first week of the seminar. ISEC aculty chaired the sessions and participated as discsants. The burden of recording the discussion and summarzing the proceedings of the previous day was again shouldered by ISEC faculty. Professor M. N. Srinivas opened the discussions, and brief introductory remarks were made by Erling Dessau, UNDP representative in India, Dieter Bucher on behalf of ADB, and Michael Bamberger on behalf of EDI. The seminar was attended by twenty-seven participants from govemental and nongovemmental organizations in India, Pakistn, Nepal, Bangladesh, and Sri Lanka (see Annex 1). The Director of the Institute for Social and Economic Change, T. R. Satish Chandran, along with Registrar M. Nageswara Rao, Accounts Officer V. Ramamurdhy, and administrative and library staff, provided excellent logistical support. C. Charles Nelson, M. Devendra Babu, V. N. Venkawesha, and M. S. Nagavalli of the Economics Unit, ISEC, prepared the seminar material. K. Narasimba Murthy reviewed the seminar material and provided valuable editorial assistane. Nelida Rasson of EDI was responsible for participant and consultant administration; while Hamia Sarkees, Faythe Calandra, Marie Cristine Theodore, and Patricia Rodriguez greatly assisted with the preparation and revision of the various drafts of the report. The document was copyedited by Venka V. Mcliyre. ix EXECUTIVE SUMMARY Despite high and sustained rates of economic growth, approximately 420 million people in Asia live in conditions of absolute poverty, suffer from mitritional deficiencies, and do not ear enough income to obtain even the basic necessities of life. Not surprisingly, massive poverty remains among the leading political and development concerns for mnost countries of the region. Although the alleviation of poverty is a primary policy objective of all South Asian governments, there is still considerable debate as to the most effective way to achieve this objective. Wish the realization in the 1970s that high economic growth does not automatically ensure the equitable distribution of benefits, attention turned to the distributive consequences of different paterns of growth. Investments in human capital (education, health, and related social services) also came to be recognized as a key component of sustained economic growth. Yet, sstantial portions of public subsidies went to middle- and even upper-income groups, while large sections of the poor and vulnrable continued to be excluded. As a result, new programs were developed specifically for vulnerable groups such as undernourished children, landless laborers, and female-headed households. The oil crises of 1973-74 and 1979-80 and the increasing debt burdens and the slowing of growth that followed indicated a need for broad structural reforms to reduce financial imbalances and to stimulate growth. Many of the ensuing reforms reduced public penditures, however, and caused concern about the impacts of the so-called social costs of adjustment on poor and vulnerable groups. Thus, in recent years "social safety nets" have beeu designed to protect these groups and to guarantee them access to a minimum level of essential services. The evolution of these approaches to poverty alleviation was reflected in the 1990 World Development Report (WDR) on poverty (World Bank 1990) and in a number of subsequent publications by the World Bank (World Bank 1992) and other intenational organizations (UNDP 1990, 1991). The WDR argued that poverty can only be effectively attackd through an integrated approach that * Promotes sustained economic growth by providing opportmities for the poor and permitting them to parficipate in growth. * Delivers social services to the poor. * Creates social safety nets and targeted programs for the poor and vulnerable. In August 1991 the Economic Development Insitte of the World Bank and the Asian Development Bank jointly sponsored a South Asian seminar on the design and management of sustainable poverty alleviation projects. The seminar was organied in cooperation with the Institue for Social and Economic Change in Bangalore. Its purpose was to review experiences with different approaches to poverty alleviation and to draw from them general lessons at could be used to improve the design and management of future poverty alleviation projects and progrms. This report presents the findings and recmendations arising from this seminar. xi Part I. Poverty and Strategies for Poverty Alleviaton In Chapter 1, Michael Bamberger draws on the 1990 World Development Report on poverty and the various publications that have followed it (particularly the UNDP reports on human development) to describe the evolution of strategies for alleviating poverty. These strategies sprang from the need for an integrated approach to poverty alleviation, in which poverty-supportive macroeconomic growth policies are combined with systematic efforts to universalize education, health, and otier basic services and with social safety nets and targeted programs for groups unable to benefit from the increased economic opportunities. Bamberger compares the methods used to alleviate poverty in South Asia, Latin America, and Africa and considers the lessons to be learned from each region. He also presents a typology of approaches to poverty alleviation and shows how the chapters of this book relate to it. In Chapter 2, Bamberger examines the characteristics of Asian poverty. In 1985 South Asia contained an estimated 29.7 percent of the total population of the developing world, but slighty more than half (50.6 percent) of the developing world's poor. In 1988 an estimated 817.5 million people in South Asia (74.5 percent of the total population) were living in rural areas. Approximately 441.5 million of these (54 percent) were living below the poverty line. In contrast, only 23 percent of the population in urban areas fall into this category. That is to say, in 1988 approximately 85 percent of South Asia's poor were living in rural areas. Because of the preponderance of rural poverty, it has overshadowed the plight of the 135 million people living in abject poverty in Asia's cities. Bamberger looks at the reasons for the lack of attention to urban poverty and some of the links between urban and rural poverty. He also discusses future trens in poverty. Part HI. Sector Exeience with Poverty Alleviation in Asia Chapter 3, by J. K. Satia, turns to the accesibility of population, health, and nutrition services to the poor. As Satia points out, the health status of the poor has improved in South Asia, but to a lesser degree than that of the population as a whole. Despite the massive expansion of peripheral population, health, and nutrition services (PS), they are not being delivered to the poor at the expected levels owing to the lack of continuing political commitment, the neglect of the processes of implementation, underfimding, and inadequatz management capacity. Nongovemnmen organizations (NGOs) have generally been more successful in reaching the poor, although their total involvement is smaU in comparison with the govermnents' PS delivery systems. Much more learning is necessary to forge successful partnerships between the goverment and NGOs. To be sustainable, a PS delivery system must first have a strong base of information about the poor. Second, policies need to focus on the poor and on improved targeting procedures. Third, programs should strengt managemant's capacity for designing and impleme PS delivery systems. Fourth, systematic research shlould be carried out at the local and regional level to rine methods of providing PS to the poor. Chapter 4, by Dieter Bucker and Keith Johnson, looks at the experience of the Asian Development Bank (ADB) in helping to promote rural development and alleviate poverty. Rural poverty, the authors point out, is caused by the combined effects of excluding the poor from normal levels of production and consumption, by the inhibiting effect of their inheren disadvantages (poor health and education, higher birth rates, and so on), and by the excessive risks to well-being that they face. The prevalence of poverty in South Asia can also be related to a mnumber of overlapping sucturl causes, and thus the authors argue for an integrated xii Executive Sranmwy multisectoral approach in the fight against poverty. Rural development projects typically consist of an integrated bundle of investment and technical assistance measures presumed to be mutually reinforcing. As such, they are considered to be especially well suited to combating poverty and providing the basis for sustainable development. Although in theory the integrated approach has a number of advantages, ex post assessments indicate wide divergences in practices and achievements. Too often integrated rural development projects have proven to be complex and difficult to implement and sustain. Also, many of them have failed to improve the lot of the poor or to provide the expected level of benefits. The chapter also reviews the ADB's approach to the preparation, design, and implemnentation of rural development programs and discusses the results of an evaluation of fifty such programs. The authors conclude that the concepuWal framework for rural development is still evolving as practitioners devise and implement new approaches to project formulation and implementation. An innovation of recent years has been to link rural development directly to poverty alleviation goals. Although difficulties abound, there is rnewed optimism that rural development projects can be cost-effective for development agencies, affordable to beneficiary groups, and sustainable over a long period of time. The key to good project design and smooth implementation lies in understanding and harnessing the energies and aspirations of the poor. A targeted approach that allows beneficiaries to participate in project preparation and implementation is therefore vital for project sustainability. Rural projects cannot be too complex, yet must address the various faces of poverty. Bucher and Johnson favor a phased approach with some flexibility in project implementation, but they note that it can only be carried out if rural development and poverty alleviation are treated as long-term processes. Practitioners therefore need the time and scope to devise and apply area-specific planing and implementation procedures, backed by a firm high-level commitment to the achievement of human development objectives. In Chapter 5, Yue-Man Yeung describes the poverty and basic infrastructure services in the urban areas of Asia. Since the end of World War II, rapid urbanization has shaped development policies in Asia's cities. The lack of basic in e services continues to be a serious problem, especially in the low-income communities of these urban areas, and is expected to grow even worse in the futre. The urban poor-however inprecisely defined-tend to be concentrated in certain parts of Asia, the highest concentation being in South Asia. Improving their access to basic urban services is a major challenge for planners and policymakers. Although various methods have been devised to deliver urban services to the poor, they have met with varying degrees of success. The trend now is toward maltifaceted and integrated projects. with active beneficiary participation. Of the many lessons to be learned from project experience in Asia, some of the more salient ones that the author points out are the failure of top-down planing and the need for commnity participation, stronger local govements, investment in leadership and training, and increased women's contributions. The Asian experience raises a host of policy questions, the foremost being who is to pay for the services? Urban governments are ill prepared to taclde this issue in view of their low reveme base. A number of other important trade-offs exist, inchluding: cost recovery versus subsidization, centralization versus decentralization, equity versus efficiency goals, public versus private sector provision, formal versus informal sector roles, and high versus appropriate technologies. xiii -*~~~~~~~....,.. ..;- - &ecitie Summary Chapter 6 turns to the Role of Credit in Poverty Alleviation. Ismael Getubig, Jr., opens the discussion with a brief review of credit programs for the poor in Asia and shows that most conventional credit approaches (for example, IRDP, cooperatives, and subsidized agricultural credit) have had failed in terms of both repayrent and outreach, in large part because the design, management, and implementation of the program neglected to account for the unique condition and circumstances in which the poor live. At the same time, inovative credit programs have emerged in various countries of Asia and elsewhere that have adapted their design, managemtnt, and implementation systems to the special needs of the poor. These programs have shown remarkable success in reaching the real poor while maintaining high repayment rates (ranging from 90 to 99 percent). Getubig provides the details of four such programs that have succeeded in significantly scaling up their credit activities: an intermediary-type program of the Working Women's Forum of India, specialized programs of formal-sector banks illustrated by the KUPEDES program in Indonesia, specialized banks for the poor (illustrated by Grameen Bank of Bangladesh), and a package-type program in Pakistan (the Aga Khan Rural Support Program). The successful credit programs illustted in the paper share some of the following charactenstics: 3 effective targeting * borrowers organized into solidarity groups that provido mutal support and replace the collateral or guarantor requirements of conventional banks so as to fit the credit delivery system to the needs of the poor * credit services that observe proper pricing practices (that is, trge market intrest rates) * self-selectdd income-generatin activities * assurance of bigger subsequent loan upon repayment of old loan * effective savings mobilization *- adequate and appropriate traimng and prepartion of borrowers * accountability and effective management information system, and 3 appropriate implementing institution. The author concludes that credit is one of the most potent instruments for alleviating poverty and can be made viable, sustainable, and effective if appropriately delivered by programs that are well designed, managed and implemented. However, credit programs have their limits. For one thing, projects undertaken by poor borrowers are usually small, the technology is limited, and the market is also small. Any hope for alleviating and eventually eliminting poverty throughout society rests on the economic policy environment. In Chapter 7, Noeleen Heyze -akes up gender issues in Asia's antipoverty programs. She finds poverty is more acute among women than men because of gender-based discriminaion, which can be seen in the unequal sharing of food, inadequate medical care or lack of it and thus higher incidence of illness and health hazards among women, underpayment, long hours, tedious and hazardous wori, and loss of employment or lower productivity due to illness. Many women are thus forced to sell or pawn their meager belongings and have difficulty finding shelter, fuel, and secrity. Furthermore, a large percentage of female-headed families have litle or no land or other property, live below the poverty line, and suffer persistent verbal and physical abuse, which lea.ds some to commit suicide. xiv Executive Summary According to Heyzer, these problems can be remedied by drawing women into nonagricultural pursuits outside the women-oriented areas, shifting emphasis from cash to food crops, providing improved technologies and training women in their use, building credit facilities and extension services, establishing cooperatives and marketing outlets, and developing cottage industries. New technologies can have adverse effects, however. They can increase the work load and thus affect the welfare of women and children, cause girls to withdraw from school, and deprive women of their traditional means of iivelihood. To be of real benefit to women, poverty-alleviating measures must be introduced in a paclkge rather than piecemeal and must cover the entire process of production and marketing. Special institutions like cooperatives for women would avoid sibordinating women's interests and allow women to retain a leadership role. Cash income by itself may not raise women's standing at home. The size of income, mode of control df family budget, and cultural constraints are also important. If more women become employed, however, this may affect the employment position of men and may give rise to conflicts at home. The success of these measures depends on the influence of women's organizations and the willingness of women to strive for a better quality of life. The growing trend toward urbanization in Asia is due mainly to the influx of rural migrants, who are now dominated by women in many countries, particularly in the fifteen to twenty-four age group. The international migration of young girls seeking jobs as donestic helpers is developing into a big business. Heyzer also points out hierarchical differences in the employment of women: (a) there is a labor aristocracy of small groups of educated women; (b) a large group of skilled young women is present in export industries; but (c) the informal sector is the main avenue of employment, created by the growth of a large middle class. The informal sector offers three types of employment: self-employment, which absorbs two-thirds of migrants; outwork, which is mainly done by women and children; and domestic work, which is an extension of work aL home. Assistance to women may take the form of skill upgrading, higher prices for products, easier credit, and the like. Informal banking institutions, family and kdnship systems, newly created aid networks, and patron-client relationships can also be of help, particularly in times of hardship. Although basic services like water supply, sanitation, health care, and rural energy benefit primarily women, thei; views and needs are still not being considered and us they do not obtain the intended benefits in full. To do so, women must make known their views and needs, and they must be given a greater say in development programs. Part . Decenlizaon, Targeted Poverty Projects, and Suival Strategies of the Poor, Lessons from Karnataka Seminar discussions drew heavily on local experence in Karnaak (the state in which the seminar took place) and Tamil Nadu (the location of the project visits). In Chapter 8, Abdul Aziz goes over the lessons from the visits to targeted poverty projects. Aziz describes six poverty alleviation projects visited in and around Bangalore and Madras: rag-picking for destitute children, lacquerware production, sericulture development, the Working Women's Forum, sites and services and slum upgrading, and the Tamil Nadu Integrated Nutrition program. xv T.~~~~~~~~ ~~ ~~~ ~ ,,& ;~t7 -. -:s .- --- . - ,. .- ' , .; t : .. * 0-.0w :; . .-J:: ,: E .! - - -<--.C. C .: 2 - .i- iu = . t :* -; 'e'--;.: ;~~~~~~~~~~~~~~~J Erve Swumavy The following lessons emr:erged from the project visits: * Although household income can continue to be an important criterion for identifing f poverty groups, the activities they engage in (for example, rag-picking) and their caste (for example, handicraft artisans) could also be used for this purpose. Levels of mntrition, health condition, and oppressive social conditions could also form a basis for identifying target groups. * The appropriate beneficiaries are most likely to be overlooked where the subsidy is high, as in the case of the sites and services project and the lacquerware development project. Rigorously planned and executed household surveys by independent or voluntary agencies (as in the rag-picker project) will certainly reduce misidentificationproblems, but the cost of such procedures is unduly high. Perhaps a better alternative is to have community and neighborhood members identify beneficiaries. * Projects and benefits must be decided on with local resources in mind. At the same time, certain social imperatives should not be overlooked. High-rise housing projects, for example, may not be acceptable for the poor because the residents cannot keep animals, have a workshop, or a set up a petty shop. Similarly, if benefits like nutritional supplements are prepared by low-caste nutrition workers, this may not be acceptable to higb-caste beneficiaries. * Some basic econoamic imperatives also need to be considered. Projects that intefere with the cmuren family earning potential (such as rag-picking) are likely to meet with resistance from the heads of famfilies. This difficulty can be overcome by compensaig the family for opportuity cost. Even here, a purely business approach to evaluatng the opportunity cost does not work unless the community or parents are informed of the benefits to their children's future eaming potential. 3 The income- and employment-generatng intervention alooe-say, credit for self-employment or work under a wage employment scheme-is not likely to ensure poverty atleviation. More effective results would be obtained from a package of benefits such as training, schooling, and placement, as in the case of rag pickers; training, input supply, and markeig of the product, as in lacquerware and sericulture projects; and infrastructure and amenities, as in sites and services projects. A simultaneous attack on the symptomatic condition like malnutrition and disease appears to be a more appropriate strategy of targeting the benefits than tacing them individually and sequentially. * NGOs appear to be more willing and effective in transformming a given form of government assistance (such as a grant) into an integrated, multiservice poverty alleviation program (for example, a rag-pickers' project) than a govermnent organization. The cost-effectiveness of such projects appear to be high. * Integated multiservice projects seem more liely to have cost-effective delivery systems. A single service such as credit could be provided in a more cost-effective way by inviting xvi Ereadive Suwmary community members to help identify beneficiaries and admiister the delivery system. For bureaucracy-managed programs like lacquervMe production, performance can be improved through better coordinaion between the agencies involved (such as KSHDC, commercial banks, and an artisan taining institute). * Project implementation can be monitored more effectively by the peer group or the people's representatives than by the bureaucracy. * When the assets created under the project bring a high rent income and generate capital gains, it is difficult to expect the beneficiaries to retain the asset (sites and services), If controls are not introduced, after a lapse of time assets such as housing are likely to fall into the hands of individuals who are better off. * Providing benefits free of cost inevitably reduces the life span of a project because project authorities are unable to sustain and replicate it. Conversely, wherever cost recovery is high, sustainability and replication are likely to be high. * 2One way to improve cost recovery and project sustainability is to link the sale of the product to loan recovery through a government channel (as in the lacquerware project). Another way is to use peer group pressures to encourage beneficiaries to repay loans (as in the Worldng Women's Forum in Madras). Perhaps a more enduring way is to create a policy environment that enforces loan recovery and promotes the collection of user charges. Such actions require political will and clout on the part of the government. In Chapter 9, Aziz considers the lessons from Karnataka regarding decentalized planing as a strategy of poverty alleviation. He begins by defining the concept and the methodology of decentralized planning, and the preconditions for such a process. Next, he descnbes and evaluates various antipoverty programs designed and implemented in Karnataka by the decentralized governments at the district and village cluster levels. The author concludes that despite political pressures, continued problems with coordiation, and inconsistencies between the projects and the policy enviromnent, denralition has made programs more accessible to the pocrest secors of society. This is mainly due to the fact that the local community is actively involved in decisionmaking, in controlling who receives benfits, and in ensuring tht assets are retained. Chapter 10, by Abdul Aziz and Michael Bamnberger, focuses on survival straeges of the poor. The authors compare the role of interhousehold transfers, comomity support systems, and other strategies in Karnaak, other parts of Asia, Africa, and Latin America. The poor may use a variety of survival strategies to redc the dastic effcts of unstable economic, polical, and social conditions. A knowledge of these strAteies can help authorities to formulate appropriate policies to furfier improve the survival efforts of the poor; frame a comprehensive strategy for poverty alleviation acceptable to the poor; and become familiar with the c!ltexts in which specific strategies are adopted by the poor. Aziz and Bamberger note that other rearchers have suggested that a 'culture of poverty" exists among poor households whereby they provide informal credit services and even share food, mil -. - 2) Exesw - clothes, and tizrniture. In Cuidad Guyana, Venezuela, such mutual assistance extends to house- building and maintenance and the supply of water. Such economic reciprocities operate among both kin and nonkin households, although they tend to be stronger among the former. Some have called this phenomenon a 'continuum of exchange" or reciprocit and have identified several patterns of reciprocity. In some cases, goods are given free, with no expectation of having them returned. Reciprocity in the slums of Mexico City is said to be related to social, economic, psychological, and physical distance, while in El Salvador it is the poor households with no steady income or not enough for their basic needs that cooperate with each other. It appears that "givers" and "receivers3 constanty change places as their fortunes change. A special form of reciprcityprevailing in the Philpines is called UCang ng loob, meanng debt of gratitude, which remains an obligation of the receiving family until it is repaid-sometimes one or two generations later. Malasaak (or cbarity) is another form of aid given to persons in indigent circumstances and, therefore, need not be retred. Female-headed households-because of their lack of steady income, limited numbers of wage eamers, and lack of access to fornal credit market-tend to rely heavily on informal support networks. Assistance and reciprocity consists of the exchange of money, food, and clothing and frequently the transfer of children to households with a higher income or with better access to schools or places of employment. In Karnatak and Tamil Nadu, various surval stategies operate in the marlet context, at the individual level: casual workers, for example, prefer piecework in order to maxnimze earnings; poor household workers desire payment in kind to counter the efficts of inflation; and poor workers such as maids prefer regular, though low-paid, worlc At the household level, during drought or distress, many women and children offer to work for wages to supplement the low income of the male-earner and migrate to big cities in search of jobs. These findings have important policy implications for publicly financed intergenerational transfers such as pensions and child welfare benefits. They also affect how household income is estimated and the basis on which the capacity to pay for services is calculated. In addition, they suggest that attctive projects can mobilize greater-than-expected resources from within poor communities. Part IV. Suiddines for die Selection. Design, Imple ton, and aility of Poverty AMlevon Programs Part IV proposes operational guideines to ensure ttm the selection, design, implementation, and operation of poverty alleviation progams respond to the characteristics of the target population and to the economic, social, and political envirome in which the poor live and work. It is argued that if poverty alleviation projects are designed according to conventional project cycle procedures, the projects are less likely to benefit large sections of the target population, will be poorly implemend, and will not be sustined. The design and management of poverty alleviationprojects carry substantal costs in terms of staff time, the information required, and preparation. Such projects are usually smaller than the typical investment project and thus have higher supervision costs per dollar loaned. In many cases, some of these costs can be absorbed by NGOs or community organizations themselves. xviil Fxecutive Summary Additional investment is required to reorient govermnent bureaucrats (and private officials for credit projects, for example) to the needs of the poor. This will frequently involve training, structuawA reorganization, and intensive cooperation on small pilot projects. It may also be necessary to create new coordinating mechanisms to ensure effective liaison between participating public and private sector agencies, and between the government and the beneficiaries. With appropriate design and implementation strategies, however, the costs are likely to be offset by significant benefits. First, projects are more likely to reach the intended groups. Even a 5 to 10 percent increase in this regard may more than offset the additional start-up and supervision costs. Second, the quality of project design and implementation will improve when beneficiaries participate in project organization. Third, additional savings (to the government or donor) may occur if communities contribute their labor to the project (which they are more likely to do if they have been involved in the project design). Fourth, the klelihood that a project will be sustained is greatly inceased if beneficiaries help operate and maintain it. Community involvement is particularly important to the life of irrigation projects and social infrastructure. Many benefits are also likely to accrue from investments in training and institutional restrucuring introduced to adapt government, and where necessary private, agencies to the culture of poverty- respons:ve administraton. The five chapters in this section emphasize a number of general principles. First, they indicati the need for a conducive policy environment and for close integration between public polices and poverty programs and projects. That is to say, decisionmaking will have to move closer to local imple agencies and beneficiaries in order to build the necessary public sector culture and institutional structure. At present few South Asian countries are politically committed to sustinable and financially viable poverty programs. Even those that do offer services to the poor provide little support for effective cost-recovery, and thus they have been unable to sustain many of their poverty projects. Second, an effort must be made to modify the process of collection and intemretation of data for planning, monitoring, and evaluation. The analysis must be broadened from the conventional focus on economic data to include social, culural, political, and institutional factors. Rapid assessment techniques should be developed to reduce the costs and teh time of data collection and to make the analysis more responsive to commuity characteristics and needs. The interpreration of the data sbould be carried out with the help of beneficiaries rather than in a government office. Third, beneficiaries should participate in all stages of project identification, design, imtplementation, and operat-n. This requres a considerable investment of staff time as frequent meetings st be held with beneficiaries and the project must be modified in response to these discions. Fourth, the project design and implementation strategy must be adapted to the characteristics and needs of the beneficiary population. Project are likely to be smaller and simpler and to make greater use of local technology and implementation practices. Greater flexibility wii also be required in the desig financial control, and implementation schedules. It is vital for the project to be considered a "learning process" rather than one in which a predetermined blueprin" is being followed. Consequently significant modifications can be expected as the project proceeds and as beneficiaries come to understand exactly what the project involves. Although these changes and delays would be considered costs in the conventional xix - ~ - - - . - Eecuive Suwmay project cycle, they are among the principal benefits of the "process approach" in that they stegthen the capacity of local communities to identify and implement futuire programs. In Chapter 11, Abdul Aziz explains how targeted poverty projects might be integrated into national development policies. He shows how approaches to poverty alleviation evolved from simply a component of strategies for accelerating growth to the recognition of the persistence of poverty even during periods of high growth. Policies now seek to make simultaneous and sustained interventions at various points of the poverty cycle. He emphasizes the need for a conducive policy environment, using the evolution of poverty alleviation policies in India to illustrate his points. Chapter 12, by Neil O'Sullivan, identifies three important lessons that can be drawn from past experience in designing poverty reduction projects. First, projects are most likely to succeed if their objectives match the demands of the poor and if the poor are direcdy involved in the projects. Second, sLuch projects cannot succeed without concentrated attention over a long period of time. However, the resources available to implementig agencies are invariably limited and are likely to shrink rather than to expand over time, with the result that little attention is likely to be given to conditions after the project has been completed. Third, even where both political will and government resources have been adequate, the results-particularly sustainability-have still been ointing. O'Sullivan therefore points to the need for economic and administative realism in identifying and designing projects. He describes procedures for rapid social assessment and their links to participatory project design. In Chapter 13, O'Sullivan provides guidelines for implementing poverty projects. He argues that beneficiary organization is a key factor in the design of poverty projects. He also calls for greater attention to institutional factors such as the reorientation of bureaucracies, the selection and training of staff, the building of flexibility, and effective coordination. As the author notes, appropriately designed and evaluated pilot projects can play an important role in the development of major proams. He also emphasizes targeting and the involvement of NGOs. In Chapter 14, O'Sullivan considers the gender-sensitive issues that can arise in designing poverty reduction projects. Like Noeleen Heyzer (in Chapter 7), he argues that both economic and equity issues need to be addressed to ensure the ful participation of women. Women, he notes, are in effect the managers of many poverty projects, most of which would fail to achieve their objectives if they were not gender-sensitive. O'Sullivan proposes methods for encouraging the full participation of women and relates gender issues to the rapid social analysis and participtory project design described in the two previous chapters. In Chapter 15, O'Sullivan discusses sustainability, which has been a continuing problem for development planers in many parts of Asia. Although projects fail for various reasons, usually it is because planners have paid inadequate attention to institutional arrangements, both on the supply and the demand side, or did not muster sufficient "political will" to overcome the inevitable social, economic, and technial barriers that separate the marginalized poor from the mainseam of society. Sustinability is also affected by other design flaws, the lack of attention to monitoring and evaluation procedures, and institutional weaknesses. xx Erecaive Sumnaiy Part V. Summary and Condlusions Chapter 16, by Michael Bamberger, sunmarizes the findings and conclusions of the earlier chapters on population, health, and nutrition; rurl development; the urban poor and urban basic services; the role of credit in poverty alleviation; gender issues in poverty alleviation; and decentralization and poverty alleviation. Chapter 17, by Bamberger and Aziz, brings togedier all the lessons discussed at the seminar. The authors first explain why many poverty projects fail and then propose guidelines for inproving performance. Why Many Poverty Projects Fail With a few notable exceptions, the performance of poverty alleviation programs and projects has been disappointing. They have not reached the poor in an equitable and cost-effective manner or produced a significant and sustainable impact on the lives of a significant proportion of the target groups. These disappointing results can be traced to many factors, notably a weak policy environment, poor project design and management, insufficient attention to administrative and organizational details, the opposition of powerful stakeholders, and the lack of attention to gender issues. The Policy Environment Although most of the poverty alleviation programs discussed during the seminar were initiated by the cental or state govemments, in many cases the political will was not strong enough to ensure that programs would be implemented as intended or that they could be sustained. Political will was lacldng in the following areas in particular: * Few governments have been willing to take the necessary measures to enforce loan recovery or the payment of user fees and service charges. This has been reonsible in part for the poor maintenance of most projects and for the lack of resources for new equipment and supplies. Becamse loan recovery rates have been low, credit agencies have not been able to recycle these fimds trough new loans. * Govermments have been reluctant to relinquish the finacial control necessary to implement effective dec lizion policies. In many cases, cental plaming agencies have not been willing to promote a genuine decentrlization of the planing function. * While recognizing the potential contrbutions that NGOs can make in the fields of project planning and management, governments have been reluctant to provide them with enough financial and operational autonomy to operate effectively. Littde attention has been given to project susainability. Financial provision is rarely adequate to cover operaional and maintenance costs. Although an effort has been made to carefully choose the agencies responsible for project implem=ion and to provide xxi :~~~~~ ~ .= -. .~ . -. - -~. . . , = ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~- . Excutive Swnmry them with the necessary institutional and administrative support, litte thought has been given to the selection of the agencies responsible for managing the project once it becomes operational. Project Design ad Management Although there is ample documentation to show that poverty projects should be designed and implemented with their unique characteristics in mind, this is mrely done. * The success of a poverty project depends not only on the adequacy of its financial, economic, and technical design, but also on the way it addresses culural, social, institutional, political, and environmental issues. However, the appropriate forms of social analysis are rarely incorporated into project analysis and design in a systematic way. * It is now widely acknowledged that the intended beneficiaries must participate in project identification, design, and operation if successful implementation, operation, and maintenance are to be achieved. Yet few projects make systmatic use of beneficiary assessment techniques or beneficiary participation, or extend the length of the project preparation phase to allow for more thorough consultations. * Most projects continue to be designed and managed from the top down with a "blueprint" approach. That is to say, important decisions on the cheice and design of projects are made at the central level. Project design is seldom flexible enough in its financial and implementation aspects to allow for significant modifiaions as the project evolves. * The procedures used in identifying and targetig the poor are often weak or poorly administered, and a considerable proportion of project resources frequently benefit wealthier groups. Monitoring procedures are usually weak, so it is difficut for management to obtain rapid feedback when problems of this kind arise. * The potential contribution of NGOs in project planing and implementation is frequently overlooked. The mutual suspicion and distrust that often exists among government and nongovermnent agencies must be overcome and an effort made to develop more flexible administrative procedures that could allow the goverment to exercise the necessary degree of financial control over public resources without reducing flexibiliy, which is one of the main strengths of NGOs. Administrative and Organiatonad Isues The size and complexity of govemments and the wide range of responsibilites thy must assume often make it difficult for them to ensure the administrative fliility needed to deliver services to the poor in a cost-effective and equitable manner. Th following are some of the typical problems and issues that arise in the manent of poverty programs: nxil Exective Swmnmy * Government agencies prefer standard procedures that simplify opeations and make it easier to define staff responsibilities and monitor their performance. Consequently, many of these agencies find it difficult to build the flexibility required to adapt to local social and economic conditions. * For similar reasons, govremments like to define targets (number of loans authorized, number of children vaccinated, and so on) and like to give their staff specific achievement targets to aim for. These targets have become a problem, particularly in credit programs where extension workers are often accused of trying to force people to accept a loan for the acquisition of an animal, agricultural input, or machine that they do not wish to have. * Governments seek to ensure strict control of how funds are used and conseq y require that expenditures be precisely defined and controlled. Because it is extremely difficult to change the way in which approved funds are used, there has been little flexibility in the way funds in poverty projects can be spent. Overcoming Opposition from Powerful Stkehods Powerful economic and political groups create two kinds of problems for poverty projects First, these groups often try to divert resources to themselves or their followers. Significant proportions of low-interest loans, subsidized agrcultral iputs, low-cost housing, and subsidized food are fruently diverted to powerfu local interests. Given the influence of these goups, it may be difficult to control even well-documented cases of mppriation of funds. Second, many political or economic mterest groups see poverty programs as a threat, especially if they involve the reallocation of land, permit the entry of poor producers into markets previously controlled by the local elites, or empower members of the electorate whose votes had previously been -controlled by a political party. Also, projects that offer economic or legal empowerment to women or low-caste groups may pose a threat to male workers, spouses, or religious groups. In many cases, the threatened groups will try to sabotage the project through political pressure, and in some cases may even resort to physical violence. Many projec and programs bave failed because of this kind of opposition. Lad* ofAdequate Geder Analyst Despite the fact that women are disproportionately represented among the poor and that the economic contribution of women is seriously umdervalued, gender issues are often largely ignored in project planing and implemnation. xxiii .. . .. . . . .~~~~~~~~iil : . . .X~~~~~~~~~~~~. Eraive SUMm" mproving the Perfoance of Poverty Ablviation Projects The hIponrance of an Integrated, Muilevel Approach The findings of the semina strongly support the three-pronged approach to poverty alleviation proposed in the 1990 World Development Report. This is an integrated approach with simultaneous and coordinated actions at the policy, sectoral/program, and project levels. It requires effiective intersectoral coordination. Poverty cannot be alleviated in an effective and sustained manner unless actions at each of these levels are given equal emphasis. Creating a Conduive Policy Eniromnent A strong and sustined political commitment to poverty alleviation is also essential. However, good intentions are not enough. The policy environment must also be conducive to poverty alleviation. Seminar participants pointed out a number of examples in which well- intentionedpolicies actually prevented projects from achieving sustainable results. Such has been the case where services are free or subsidized, because they make beneficiaries feel that they are not responsible for the maintenance of the ftcilMies and services pmvided. This also eliminates the possibility of generting revenue to cover operations and maintenance costs. Another example is the assumption by many policymakers that most women do not have primary economic rponsibility for their household. This attitude helps reinforce wage and employment dfiscriminatio.L To crea and sustain a conducive policy environment, authorities mst meet some of the following rqirements: * See that subsidies are reappraised, along with their impact on the long-term susainbility of the programs they support and their overall impact on the poor. Many subsidies should be eliminated or significantly reduced. * Put greater emphasis on botom-up planning and give the poor have an active role in the planning and imple on processes (see Chapter 12). * Support policies that combine financial and administrative decentralizaion. A strong political commitment Ls required so that a real effort is made to decenalize decisiomakig. Valuable lessons can be leamed from the decentalizaton experience in Karataka * Ensure that progms and projects are sustained. Most governments and donor agencies bave a relatively short time horizon and are more concerned with invesg in new programs than providmg financal and other resources to ensure that services are sustained at an accptable level and quality. Political wil is partiulaly important in a federal systm such as that in India, where both the central and state goverments tend to pass the responsibility for operations and maintenance on to lower-level authorities who do not have the resources to adequately cover these costs. Ereaave SLey Developing an Effecive Data Base To be effctive, poverty policies and programs must have access to reliable and timely information on the magnitude, c istics, and distnbution of poverty over time, and on the impact and effectiveness of difft poverty allevaton poLcies and programs. Therefore, accurate and frquent estimates must be made of the charactedstics of poverty. Such estmates are already being made in India and to a lesser extent in other countries. The South Asian countries also need to put more emphasis on monitoring and evaluating poverty programs. Creating a Poverty-Responsive Instiudonal and Administraive Structure As already mentioned, policies and programs must be organized and administr in ways that are more flexble and responsive to the interests of the poor. Some of the institutional requirements are * Bottom- planning and management prooedures that include systematic - smechansms for consultig with, and obtaining feedback ftom, the poor and other local groups and organizations. * Dcentalization of decisionmaking anti program management. * Decentlization of financial management and responsibility, which can be achieved by delegating the control of financial resources, allowing greater fleibdility in how funds are managed (with more emphasis on ex post evaluation of how funls were used rather than on complicated ex ante approval procedures), and by increasing the revenue sources available to local authorities (through service charges, higher taxes, and the like). * A greater role for NGOs at the policy and ipla ion levels. * A greater role for the private sector in the delivery of services * The use of training and otber management techiques to provide public officals with the skldls and attitudes requed to implement participatory appches. * Stronger interagency coordination to permit the integrated delivery of a wide range of services. Stegtheng Management Capaity At present, most poverty programs are badly managed. One of the problems is that poverty programs are hfquety considered to be social welfare sevces rather than ways to promote economic growth through investments in human development Conseqently, xxv - - :- - . Execauive Summary management accountability is often weak. The management of poverty programs can be strengthened in a number of ways: * Provide a clearer definition of progrm objectives and the criteria for evaluating success. * Recognize the importance of rigorous project analysis and design. Chapters 12-15 provide guidelines on how to identify and design projects more effectively through greater administrative and economic realism. They also stress the need to adapt conventional project design to the special characteristics of poverty projects. Most projects will be smaller, simpler, and of a shorter duration than conventional economic and infrstucture investment projects. * Much greater use should be made of market studies to ensure that tere is a market for the goods or services to be produced through income and employment generation pmrjects. * More emphasis should be placed on demand-driven as opposed to supply-driven projects. Projects should respond to the needs and priorities of beneficiaries rather than the wishes of central planners. * Systems of performance evaluation should be introduced so that future budget allocations are based on results (volume and quality of outputs and impacts) and not simply on past expenditures. * More effective and management-responsive monitoring and evaluation systems must be introduced to obtain rapid feedback on project progress and problems. These should not simply be an admstrative reportng system for funding agencies. * Private sector agencies should play a greater role in the provision of services. With their involvement, a greater effort should be mad to recover the costs of services and to generate the necessary resources to cover operations and * NGOS should be encouraged to participate in poverty alleviation. Public sector agencies should use analytical procedures such as cost-ectiveness analysis to identify areas in which NGOs appear to have a comparative advantage in the provision of different services. * Higher priority should be given to esurg susinability, for example, by incorporating sustinability assessment into project analysis, involving beneficiaries in aU stages of project design and implementation, and incorporating mechanisms to -generate the resources required for operations and maintenance. xxvi Ercdw -mar Use of Appropriate Targeting MecwLss Considerable corversy surrounds the merits of mnarow" targeting. Some analysts argue that most targetd programs are expnsive, difficult to adinister, and do not reach the most needy groups. They also contend that categorizig certain groups as poor or weak may dete them from seeking work or becoming self-reliant. Oths, however, say that at least in the Saudi Asian context, narowly targeted programs are te only efective way to overcome the effect of illiteracy, the uneWal distibution of economic and political power, cultral and religious barriers, and the unequal rgonal resource endowments and to ensure that resources and services reach the poorest and most vunerable groups. The issues involved are extremely complex. However, the folowig guidelines can be used to assess the potenal benfits and costs of targeting in a given circumstae and to select the most effective targeng mechanism for a particular program: * Incidence analysis can be used to assess who is benefiting from programs. If a partcular program or service ap s to be accessible to all major low-income and vulnerable groups, then the design of a narrowly targeted program is prdbbly not justified. However, if it can be demonstat that certain groups are excuded on the basis of sex, region, race or similar factors, then more nrrowly taged projects may be required. * An analysis sbould be made to detmine why certain groups are excluded. In some cases, the problems are dministatve and can be corrected without hie need for a special prgrm. In other case, the problems may be more complex and a particular program may be unable to reach major sectos of the target group. * Where targeting is to be used, a careful compariso should be made of the alterate approaches and teir probable cost-efatveness for a particular progran Ther are three main approaches to targetig: individu entitlement on the basis of economic or social c of an individual or household; geographical trgeg to sectors (low-income urban or mral populations), regions (resowue poor geographical reions) or states; and self-targetng, whereby an individual, household, or community decides whether to pardcipate in aprograin or to solicit a service. EBach of these approaches and their many subvariants have advantages and da es in terms of cost, complity of opraton, and effectiveness in reaching dif&rert grups. The recent work of Mararet Grosh (1992) assing t mechnsms in Latin America provides some usef guidelies on the methods of targeig. Integratng Women into Poverty Programs One of the principal theme of this report is that a strong gender-bias exists in the formulatonand imple iof many devel policies and programs, including povety -J- . .x *:-: >--.-- i- --.s ;- Ekalin S alleviadon programs. Consequedy, it is essental to adopt, select, design, and manage all poverty alleviation pogrms witi a view to detectg an eliminating any gender biases. The following steps should be take by policymakers and planneLs in this regard: * Recognize tat the econoiic contribudon of women is underestimated and promote the use of gender analysis in policy formulation and project planning and implementation. * Develop political and economic organizations to aid women in the labor market, help them protect their land rights, and help them fight fbr political, economic, and social equality. * Promote flexible and tageted credit programs. Such programs play a particularly important role in providing women with access to the capital and the technical assistance thiy need to break out of their economic and social dependence on their family and male relatives. Many of the more successful credit programs have provided an entry point for a broader range of legal, economic, and political suport. * Encourage programs that reduce the tme women spend on collecting water and fuel so as to give them more time to work as agricultural producers. * Understand and protect women's existng sources of livelihood. * Eliminate discriminatry laws on ownership and on the control of productive assets. * Ptomote equitable access to agricultral inputs. X Help strngthen urban networks and support groups that women can use as a surval strategy. nyviii Part 1 PovERTY AND STRATEGIEs FOR POVERTY ALLEVIATION 1 POVERTY ALLEVIATION STRATEGiES AND THE ROLE OF TARGETED PROGRAMS FOR TEE POOR Michael Bamberger and Abd Aziz The Evolution of Approaches to Poverty Alleviation While the alleviation (reduction) of poverty has become a major objective of development policy throughout the world, the debate continues as to the most effective way to achieve this objecive. During the 1960s and early 1970s the predominant (but by no means unchallenged) development philosophy was that poverty should be attacked throujgh sustained economic growth. The Brazilian "economic miracle," with its heavy capital investments and promotion of industrial development, was often cited as an example of this approach. When it became clear that high economic growth did not automatically ensure the wide distnrbution of benefits, attention turned to distrbutive consequences of differt pattens of growth (Cbenery, Holls, and others 1974).' During -the 1970s and early 1980s increasing emphasis was also placed on the development of human capital as a key component of sustained economic growth, and consequently there was a greater call for investments in health, education, and other social secors. The oil crises of the late 1970s and early 1980s, increasing debt burdens, and the subsequent slowing of growth led policymakers to believe that broad structural reforms were needed to reduce financial imbalances and to stimulate growth. Because many of the strucural adjustment programs that were introduced reqired goverments to drasically reduce public expenditures, the new concern was that poor and vulnerble groups would be subjected to excessive strain (as a result of the so-called social costs of adjustment), namely, the loss of employment, cuts in social services, and increases in the prices of basic commodities during the period of economic and strucural reform (Cornia, Jolly, and Stewart 1987). This concern intensified whe planners saw that it would take far longer than expected for the benefits of adjustnent policies to be widely enjoyed. They also recognized that, even during normal times, a high proportion of public subsidies go to nonpoor while large numbers of the poor and vulnerable continue to be excluded.2 Thus they became more interested in targeting progrms for specific vulnerable groups, such as unde ed children, landless laborers, or female-headed households (Grosh 1992). The evolution of these different approaches to poverty alleviation were reflected in the 1990 World Development Report (WDR) an poverty (World Bank 1990b) and a number of subsequen publications by the World Bank (World Bank 1992a) and by other intenational organizions (UNDP 1990, 1991). The WDR argued that poverty can only be effectively attacked through an intgrated approach focusing on three main endeavors: 1. Refrences cited in the chapters by the editrs arm in in mte bibliograhy at the emd of the report 2. Sudies in Jazica, Algeia, and Sri L have fod tat only 14 pecn of ie beiefis of genel fod subsidies anneud to te bout. incme quintlc. In the cue of public heath cern in Argtina, Cbile, and Pru, only 51, 22, and 11 pent of beIfis rspectively, acd to fth lowest quintie (Crh 199). 3 4 Michae Bamberger and Abd,d Aziz a. Promoting sustained econonic growth Through strategies that provide opporznesfor the poor andpernit them to pafliczpate in growth. This activity consists of three broad tasks: encouraging rural and urban employment; increasing the participation of the poor in grovth by increasing their access to land, credit, public infrastructure and services; and giving special attention to resource-poor areas, both to promote outmigration and to provide subsidized public investments to meet basic needs, maintain or increase yields, and preserve natural resources. b. Delivering social services to thepoor. Investments in health, education, nutrition, and other basic social services direcly improve the welfare of the poor and are also an essential condition for increased productivity and sustined economic growth Coeny, an investment in people is an essential component of a poverty alleviation strategy. A key issue in the ensuing discussions has been whether the poor will be reached by providing universal access to health, education, and related services, or whether this goal is unrealistic in many poor countries where educational partcipation rates and access to health services are low. Narrowly targeted programs may be a better ortion in countries wher universal access is not feasible (in the medium term). C. Provdg social sfety ne and targetedprograms for the poor and vulnerabe. Not .A the poor will benefit from the two previous policies. It may take a long dme for the working poor and those in remote or resource-poor areas to benefit from these policies, and some groups-such as the old, infirm, children, or clturally underprivileged groups-may never benefit. Other groups may continue to be vulnerable even when the new policies are filly operational. The only answer in the first case may be some form of income transfers or the direct prowision offree or subsidzed goods and services; while others may need some kid of income assurance or direct provson of goods and services to help dtm through short-term stress and calamites. These three approaches have been referred to as the 'three legs (or conmponents) of the WDR poverty alleviaton strategy. Figure 1.1 shows the links between these components. The policy environment affects the ability of human resource development programs to achieve their objectives. For example, the interventions in credit, labor, and product markets will affect the outcome of programs itended to create new economic opportumities. They will also affect health and education programs by changing the prices of basic consumption goods and will affect the supiy of teachers and medical staff through changig wage levels. Similarly, the efficiency with which education, health, and other human resource development progrms are managed will affa:t the operation of social safety nets (for example, those used to detmine which groups are excluded from regular health and nutrition programs) and targeted programs. One of the main reasons that many targeted health and nutrition programs operat relatively well in counties such as Chile and Costa Rica is that the health sector is well managed and these countries have the capacity to conduct the necssary health, nutrition, and ecoomic surveys and admhiister the new programs. Figure 1.1 shows that the links also operte in the other direction. f many targeted programs fiil, for exaWple, this will put pressure on the national health and education systems and may even cause major changes in national policy. Many economic reform programs in Asia. f:~ ~ ~ - - - S Po Aleviaon Strategies and the Role of Targeted Prognuns for the Poor 5 Africa, and Latin America have had to be drastcally modified becmse targeted food and income programs have been unable to relieve the pressure on some of the groups most affected. The failure of food distribution programs can cause riots and even the overthrow of governments. Likewise, the failure of social secor ministries to recover pat of the cost of social services can significantly affect national financial policies and attitudes to questions such as administive and financial decentralization. Although it is generally accepted that poverty canmot be reduced unless economic opportunities for the poor are promoted as part of a strategy for sustained economic growth, opinions differ as to the short and long-term impacts of economic growth on different low-income groups. In particular, the likely outcomes of promoting economic liberalization in Africa have been widely debated. In addition, questions have been raised about the feasibility of ensuring the general availability of basic social services in many of the poor countries. As Safia argues in Chapter 3. the goal of "health for all by the year 2000" may, in fact, divert attention and resources away from ensuring that primary health services are made available to the poor. Opinions also differ with espect to the third component of the WDR poverty alleviation strategy. Some argue that the poorest and most vulnerable groups rill not be able to compete with economically and politically more powerfil groups and hence must be protected through the provision of targeted services such as credit programs restricted to the poor (Getubig [992), employment and training exclusively for women (Heyzer 1992; Armnachalam 1989), or nutrition programs exclusively for the malnourished (Cornia, Jolly, and Stewart 1987; UNDP 1990). Others argue char targeted programs tend to be expensive, to provide poor quality services, and to be ineffective in reaching the poor as resources are siphoned off by influential groups (Sen 1992). Some also claim that tgeted programs provide disincentives to work bese beneficiaries must publiclj acknowledge that they are poor?3 A significant contnbution to the debate was a recent review of targetig experiences in the health, mtrition, and education fields in Latn America and the Canbbean (Grosh 1992). The review showed that a wide range of targeting mechanisms have now been developed to improve the equity and cost-effectiveness of delivery of health, nutrition, education, and other services to the poor and vulnerable groups. Although many questions concerning the most effective poverty alleviation strategies remain unresolved, the consensus is that poverty will continue to be a major development concern. Despite the imnpressive declines in the proportion of the population below the povertY line in most developing regions, the 1992 WDR estimated that more than 1, 100 million people are likely to be below the povertr line by the year 2000. Even though tis represents a proportional decline from 30.5 percent in 1985 to 24.1 percent in the year 2000, poverty will clearly contiue to be a primary developme cbjective for at least the early decades of the next cenDtry. 3. A recent example is te Janasviya progm in Sri Lanka This was designed in 1989 as a crash program for poverty alleviation. it aimed to cover all households enrlled in fte food stamp progm. Poorhhotebolds were to receive a montly gRmt of Rs 2,50 over a tweny-fur-month period. When th program began, pardciP were receving a monthly conumption grant of Rs 1,458. but the mnont mu wag was aruud Rs 1.000. Consequently, the program povided a stong disentivc to wtr. Other public emplment guant progms in India and lsewberc have sought to ensure that wages or food am eW to. or lOWer than market wags to avoid suh a disincentve effec. FIGURE I-1 THE ELEMENTS OF AN INTEGRATED POVERTY ALLEVIATION STRATEGY AND THE INTERFACES BETWEEN THE DIFFERENT LEVELS CREATING A CONDUC' ." PROMOTING HUMAN SOCIAL SAFETY NETS AND POLICY ENVIRONMENT RESOURCE DEVELOPMENT TARGETING PROGRAMS *Devaluation through FOR THE POOR AND *Trade Increasing access to pilmary VULNERABLE *Fiscal and monetary policy health and education *Mitigating 6ocial costs of *Public expenditure manage- Promoting economic adjustment ment opportunities *Protecting those unable to *Market and Institutional reform Improving living conditions protect themselves A *Local capacity building j}>.-- and empowerment - * ~Investing Inprmary Strengthening Promoting Impacts on Labor, Credit and health and education social small Product Markets *Increasing ownership infrastructure businesses of productive assets Upgrading not - destruction of housing Ipcson welfare of poor *Sustainability households Poverty AUeviaton Sraegies and the Role of Targeted Progrms for the Poor 7 Poerty and Approaches to Poverty Alleviation in Asia, Africa, and Latin America Although the Bangalore seminar focused exclusively on Asia (and mainly South Asia), the discussions should be seen in the broader framework of poverty alleviation experience in Africa and Latin Amenca. A brief review of the approaches to poverty alleviation there will serve to indicate what they have in colmmon with the Asian experience and therefore to identify problems of a general nature. Such details will also draw to the attention of Asian readers some of the material and experiences from Africa and Latin Amenca that might be relevant to Asian poverty programs. Probably the most important characteristic of South Asia is that it conins half of all the poor in the developing world. In 1985 South Asia was home to 29.7 percent of the population of the developing world but to 50.7 percent of the population below the poverty line. In contrast, 17.5 percent of the developing world's poor are in Africa (11.1 percent of the population) and 8.3 percent (11.2 percent of the population) in Latin America. By the year 2000 it is estimated that South Asia will still contain 46.1 per cent of teh poor. Consequently, South Asia is facing poverty on a more massive scale than the other two regions and must develop larger and more complex programs to alleviate ft. .Untl receny, most South Asian countries practiced central planning, which meant that the central governmen played a major role in the formulation and implementation of poverty programs. During much of the post-independence period, die predominant philosophy in these countries had been one of entitlements, which gave the poor the right to receive an extensive package of free or subsidized services. It was argued that a combination of economic, political, and cultura constraints made it necessary to protect the poor and help them to cross the poverty line. Many of these coutries also face serious ethnic, religious, and cultural conflicts that limit the opportunities for economic and social development of significant sectors of the population. Women have continued to be one of the most disadvantaged groups, and many special programs have been developed for them. A notable difference between India and the other countries of South Asia is tbat poverty programs in Bangladesh, Nepal, and Sri Lanka have been largely financed by international donor agencies and have a strong project focus, whereas in India donor support for the social sectors has been proportionately less important until recently. These considerations, combined with a huge populaton living in poverty has led Indian authorities to focus on broadly based, largely government-funded programs rather than on discrete, donor-assisted, projects. In contrast to Idia, Pakistan, and Bangladesh, most African countries have relatively small, widely dispersed populations. After promising performances in the post-independence period, most Afican countries expenenced negative rates of growth in the 1980s combined with rapidly increasing debt problems (Obidegwu 1990, 1992). Many of these countries have therefore been forced to carry out broad stucual reforms, which have had serious iplications for poverty alleviaton strategies. To reduce deficits, they have had to reduce public sector employment and expenditmres an basic social services such as health, nutition. and education (Grosh 1990b; Sahn 1992; Pinstnup-Anderson, Jaramillo, and Stewart 1987). In response to these budgetary constraints, donors have been providig substatial support for programs specifically targeted for poor and vulnerable groups. These have take the form of Social Action Programs -that frequently include a large nmber of components targeted for specific groups such as undernourished children, women farimers, dislocated govenment worlers as well as projects to protect social infrastructure such as schools and water sources (Sldllings, Marc, and Makonnen 8 Micael Bamberger ard Abdud Aziz 1991).' Considerable interest has also been expressed in trying to adapt the experience of the Latin American Soial Emergency Funds to the African context5 One of the questions being debated throughout Sub-Saharan Afica concerns the appropriateness of structural adjustment programs and their impacts on the poor. Some Arican governments and development experts believe that the liberalization and privatzation strategies advocated by many international donors would expose the weak export and productive sectors of many economies to excessive oDmpetition, with negative consequences for many of the weaker countries, particularly for the poor. Many African countries in the throes of struchtral adjustment have large numbers of displaced public, and to some extent private, sector workers in need of assistance. In some countries, tens of thousands of government and military employees bave lost their jobs, ad this trend has caused political as well as social and economic problems. These displaced govermnent workers are usually vocal and politically well organized so that although they may not be the poorest sectors affected by adjustment, they have received a considerable proportion of the resources provided under the social costs of adjustment programs (Callaghy 1990; Graham 1992a). ConseWendy, in Africa an important distinction nmst be made between the poor and the vunerable (Graham 1992b) and long-term strategies for the alleviadon of strnucuralpoverty must be distinguished from short-term programs to combat fra:icional or conjectural poverty caused by structural adjustment programs. In practice, the distinctionbetween these two has become clouded as many of the structral adjustment programs have not produced the hoped-for rapid economic unaround, with the result that the social costs of adjustment are usually felt over a much longer perod of time and may affect a larger number of people than anticipated. Because most African nations also have a weak administraive and project implenion capacity, it is difficult for them to apply many of the approaches to poverty alleviation used in Asia and Latin Ameica (Bamberger, Yahie and Matovu 1993). These other regions have made extensive use of social surveys, for example, to identify the poor, to determine eligibility for programs and to monitor progrm impact (Grosh 1992). Most Sub-Saharan African countries do not have the administative capacity to conduct or to analyze the ldnds of socioeconomic surveys recommended as part of the social dimensions of adjustment programs.' Another point to note is that the African experience cannot be fully understood without some idea of the poitical economy ofadjustment(Nelson 1990). The economic and administrative reforms proposed under many adjusent programs have run into strong opposition from political and economic elites, who see these reforms as a threat to their control over goverment policies. At the same time, powerfUl local groups have been able to siphon off significant portions of the resources itended for the poor (Grahnam 1992a, 1992b). It has thus become clear that poverty alleviation has political as well as economic dimensions that must be takn into account in policy and program design and implementation. Governance, local capacity building, the role of NGOs, 4. The Pr_gme for de Mitigin of Povety ad the Socil Costs of Adjusmen (PAMSCAD) in Ghana and the Program for fie Aileviation of Poverty and Social Costs of Adjustmen (PAPSCA) in Uganda both included me than twenty cmponents. Bodi programs have been crized for having too many componen and for not having a clear satee for selecting eslablithing the priority of compoanms. In both countries. the numbr of cmpont is likely w be reduced in subsequent progms. S. Zambia is one cunty that is fying to develop a social investment fimd based specifically on the Bolivian Social Emew and Soci Investmen Funds. 6. Se' imr example, Grootaert and Marchant (1991) for an indication of the kinds of da requimes being proposed. …- - _ . . - S ."., . - - - L .' a; . . -_ f _C . _- Poverty Aleviaion Strategies and the Role Of Targeted Programs for the Poor 9 and d alizaon are some of the issues now receiving greater consideration in response to these concerns. In much of Latin America the 1980s aic considered 'the lost decade" because many of the social and economic gains of the 1970s subsequently vanished. The debt crisis obliged many countries to cut back drastically on spending on social services even, though efforts were made to protect the social sectors from disproportionate cuts (Grosh, 1990b). Beginning with the Bolivian Social Emergency Fund (Jorgensen, Grosh, and Shacter 1992), many countries have sought to develop fast disbursing grant or credit programs to permit projects to be rapidly implemented to protect the poor and vulnerable groups from the impacts of social expenditure cuts and increasing unemployment resulting from structural adjustment programs. Innovative delivery systems have been developed for targeting essential services and increasing the cost- effectiveness with which health, nutrition, basic urban services, and education, among other services, are provided to the poor. Studies (Pfefferman and Griffin 1989) have shown that a well targeted program can in some cases provide a given level of services to the poor at as little as 10 percent of the cost of providing the same service thrugh a general food subsidy or untargeted program. Latin America has a number of advantages over the other regions with regard to poverty deviation. First, the overall proportion of the population below the poverty line is much lower than that in South Asia or Africa, so it is easier to develop narrowly targeted programs for the poor? May Laidn American countries also have the administrative capacity to plan and implement targeted programs.8 Furthrmore, poverty in Latin America is more of an urban problem than in either of the other two regions. in some ways, this makes it easier to identify the poor and lowers the administrative costs of delivering services. At the same time the grater visibility and more threatening nature of the urban poor increases the pressure on govemments to take action. Areas in Which Latin Amenca and Aica Could Benefit frn South Asia's Experience Despite the differences just descnibed, the experience of South Asia holds some important lessons for other regions and has already influenced credprogramsfor thepoor. The experience of the Grameen Bank and organizations such as BRAC, the Working Women's Forum, and the Small Farmers Development Programs have influenced the design of credit programs in Africa, Latin America, East Asia, and even the United States (Getubig 1992). These credit programs also provide an effective delivery system for a wide range of development programs for women (A1machalam 1989). Women's weak political status and lack of authority within the family is largely caused by their lack of control over economic resources or access to employment. Consequently, many of these credit programs have had a strong impact 7. in 1985 approxmaey 19 percet of de Latin Ameica populaion were below the poverty line compared with 47 percet for Sub_Saran Africa. 51 parent for South Asia in general, and 55 percent for India (Worid Dewlopnw Report 1990: 29, Table 2.1, p. 29). 8. Chile (Ractynsi 1993). Brazil (Pffrman and Griffin 1989), Venezuela ( Garcia and LEvy 1992), and Medco (Pfeffnan and Griffi 1989) are examples of counes ht have vied difet medhos of targeting health, nutriio and educati progamns for the poor, witfh somc success. A number of other couties-notably, Bolivia, Ecador, UTmrguay Colombia, and El Salvador-have developed or are developing poverty maps to improve the use of geographical waring. 10 MidcaeL Bamberger ard Abdul Aziz on the social, legal, and political position of women, as well as on their more immediate economic condition. This has been particularly true of programs such as the Self-Employed Womrn's Association (SEWA) and the Worling Women's Forum (WWF) in India, and the Grameen Bank and BRAC in Bangladesh. South Asia is also one of the areas in which NGOs have come to play a leading role in proving servzces to the poor and in strengthening the planning and management capacty of local organizaions. The experience of organizations such as BRAC offers valuable lessons for Africa and perhaps Latin America. NGOs in Uganda, Ghana, and Burkina Faso, for example, are rapidly expanding their services to the poor and would find it usefl to know something about the way the NGO sector has evolved in countries such as Bangladesh and India (Clarke 1990; Wiliams 1990). The experience of women's organizations could be of particular value to Africa, where women's movements are gaining in stregth but have less experience in many of the legal, economic, and political areas in which their counparts in South Asia have been working for many years. Of particular interest is the evolution of government-NGO relations and the kinds of issues that arise within the NGO sector as it expands and tries to establish its own internl organizatonal and regulatory mechanisms (Bamberger and Shams 1989). Asian NGOs also have an imnportant role to play in the delivery of services such as health (Chaterjee 1990; see also Chapter 3), housing (see Chapter 5), and credit (see Chapter 6). Furthermore, counties such as Sr Lanka and India can offer many lessons vnth respect to the creation of social sqety nets and efforts to provide a high level of social services to all sectors of the population. Sri LIanka is frequently cited as one of the low-income countries to have been successful in achieving low infant mortality, high literacy, and high scores on other social indicators (UNDP 1991; UNICEF 1987) while maintaiing a reasonable rate of economic growdL One of its most important programs in the 1960s was the Food StaWp Program, which provided a subsidy to the entire population. Other social service programs provided assistance in the areas of education, health, and housing. After the relatively prosperous 1950s and 1960s, however, these programs began to drain government resources and were gradually reduced or phased out. In 1989 an atempt was made to revitalize this social welfare approach thrugh the Janasaviya, but it had to be scaled back drastically because of budgetay pressures. Albough these programs contributed significantly to the impressive social indicator scores of recent years, some questions have been raised about Sri Lanka's experience (Bruton and others 1992). Many of these programs, particularly the income transfer programs such as Janasaviya, created disincentives to work (in some cases, the cash grant was higher than the market wage). Many of the programs were also not very effective in reaching the poor and in many instances a high proportion of the most needy groups did not benefit, as indicated by the steady increase in malnutrition over the past two decades. Bmton and others (1992) have also suggested that the wefre policies may have discouraged economic growth and hence may have had long-term negative impacts on the poor owing to the lack of employment opporunities. Bhalla and Glewwe (1986) have argued that during the period of direct policy measures (1960 - 78) the rate of economic growth was not exceptional and was due largely to favorable initial conditions. In the period of growth-promoting polcies and fewer direct welfare programs (1977 - 84), economic growth rates more than doubled, expenditure inequality did not choge significantly, consumption expendiur of thxe general population and of the poor increased, and several living standard indicators continued to imwprove. Many valuable lessons can also be drawn from the broad socal welfare policies of ILia. Since independence, the central and state goverments have been responsible for a wide range of social and economic services (for example, food subsidies) to the poor and often to much Poverty Aleiton Strogies and he Role of Targeted Programs for uhe Poor 11 broader groups. While these programs have had a significant impact on the welfare of large sectors of the low-income population, their cost has become a heavy burden for the state. Also, many of the programs have proved costly to administ or have failed to reach the poorest groups. Most of the integrated rural development programs (in other South Asian countries, as well as lndia) have failed to reach the poorest farmers (World Bank 1989; see also Chapters 4 and 6 of this report), health programs have had limited success in reaching the poor (see chapter 3) in general and women in particular (Chatterjee 1990; World Bank 1989b, 1989c), most credit programs have not reached the poor (see Chapter 6), and the formal sector has had limited success in providing housing and basic urban services for the poor (see Chapter 5). One program that has attracted widespread interest and has been replicated in a mnmber of other counties (particularly in Africa) is the Rural Employment Guarantee Program. It has sought to guarantee employment to the rural population through public works programs. The goal has been to reduce seasonal unemployment, but with the exception of the Maharashtra Employment Guarntee Program, which was able to geerate almost 600,000jobs per day in a single state (World Bank 1989c: 99), the scale of most of these programs has been quite modest (Plnning Commission 1987), their costs quite high, and the quality of the works in most cases quite low. Despite many political and adminishave difficulties, the experience in Mabarshtra shows that a well planned and administered program can have a significant impact on seasonal unemployment. India also offers valuable lessons conceing the planning, management, and targeting of social services. Althoughmany of its programs have been poorly managed and have had difficulty In reaching the poor, there are also manay examples of well-managed public and private programs, mtably in the population, health, and nutrition sectors (Heaver 1989). The Tamil Nadu Integrated Nutrition Program (discussed in Chapter 8) is frently cited as an example of a well- targeted and managed program. SouthAsia also has developed many nnovative approaches to strengthning the economic, socal, and political panicipation of womet In almost every Asian country women account for a disproportionately large percentage of the poor and the very poor. There is increasing evidence that women do not automancally beefit from antipoverty programs, and that many growth strategies may even worsen the conditions of significant mmibers of poor women if certan adjustments are not made in regard to planing assumptions and implementation methodologies (Heyzer 1987; Ng, Hamid, and Ali 1987). Heyzer (1992) argues that women's experience of poverty may be different and even morc acute than that of men because of gender-based forms of exclusion. The primary gender- based problems of the poor are the unequal sharing of food, inadequate medical care leading to illness and health hazards, underpayment, long hours, tedious and hazardous work, and loss of eWloyment or lower producevity because of illness (which has forced them to sell or pawn their meager belongings). Women also face shelter, fuel, and security problems. A large percentage of female-headed families, having little or no laads or other property, live below the poverty line and suffer persistent verbal and physical abuse, which has led some to commit suicide. Policies and programs to reduce poverty need to be sensitive tv gender issues within and among households. Within households, efforts to fulfill basic needs must take into account the intrahousehold structures and dynmics that affect the use of income and decisions over resource allocation along gender lines. Female-headed households with young children tend to be among the poorest in the rural areas. Too ofte, national data on poverty conceals these differences and does not reflect reality at the local level. Policies and programs need to take these stuctu more seriously. Women and men clearly experience the state of poverty differently and often unequaly and they become impoverished through factors that are not always the same and therefore should 12 Midcael Baunberger and Abdl Adiz be considered in policies and programs. Indicators need to be found that are capable of addressing tde differen life circumstances of women and men such as the differing significantce of marriage and child-ben in their lives, the greater social constraints on women's mobility, women's unpaid contributions in income-replacing activities, and in their role as managers of meager resources of poorer households. A large number of innovative programs have already contributed significantly to strengthenig women's economic, social, and political roles. The Woridng Women's Forum (Azad 1986) and the Self-Employed Women's Association are two eff-ectve women's organizations in India that combine political, social, legal, and economic activities. Significant progress has also been made in developing national gender strategies in counties such as India (World Bank 1989b) and Bangladesh (World Bank 1990a). These approaches could prove usefil in both Africa and Latin America. In addition, Asia has some valuabie experiences - both positive and negative - regarding the targeting ofprograms for the poor. Many of the major rural development programs have had relatively limited success in reaching the poor and providing them with credit, basic services, and agricultural inputs. Efforts to establish umnerical targets have frequendy created unintended bureaucratic rigidities and have resulted in small farmers and other target groups being pressured to accept loans, animals or productive inputs they did not want. At the same time, there have been many successful experiences in using local organizations and NGOs to improve targeting of health, mtrition and credit progrms. A number of the nutrition and health targeting systems, such as the Tamil Nadu Integmted Nutrition Program, would be of considerable iterest to Latn Amencan countries, many of which (for example, El Salvador, Colombia, Ecuador, Bolivia, and Argentina) are developing targeted nutition programs as part of an integrted social sector development strategy. The experiences with the involvement of community organizations would be of particular interest. LemonS for South Asia frm the Experience of Lati America and Africa In turn, South Asia can learn a great deal from Lain America about social emergency and social investment funds (Jorgensen, ;rosh, and Shacter 1992) and also in the area of innovative and cost-effective approaches to targeting services for the poor (Grosh 1992; Pfefferman and Griffin 1989; Baker 1992). Lain American efforts to increase the role of the private sector as a provider of education, health, housing, and urban services can also provide some interestig insights for South Asian countries such as India, which are debating the role of privatzatn in social service delivery. Latn America has also had eesive experience in combating urban poverty and in the provision of services to the urban poor. The FUNDASAL low-cost housing programs m El Salvador (Bamberger, Gonzalez-Polio, and Sac-Hau 1982) is one of a number of community- based housing. programs that have served as models for many counties in Afica and Asia. The region has also had many sucesfiu exp ces with the use of credit to promote both housing and microeterprises. Africa, too, has some important lessons to offer, althugh many of its programs have as yet had limited success because they are still in their infincy. Of particular interest are Africa's approach to AJDS control and to developing system for protecting AIDS orphans and widows. Countries such as Uganda and Tanzania, which are facing massive AIDS epidemics and have very lnmted resources, are seeling ways to enable the extended fmily and the community to protect orphans and widows. Poverty Aeiaon Stnegies and the ARle of Targded Progams for the Poor 13 Becmse of the severi of its environmental problems, Africa has also eperim with a number of innovative approaches to involving the comqunity, particularly women, in the management of environental resources. The participatory environmetal management programs in Burkin Faso are an intresting example. Local communities develop and are involved in the implementation of environmental management plans for local terroirs (World Developmnt Report 1992: 146). A Frmework for D ssing Apprac to Poverty All_evati To be successfil, poverty alleviation strategies at the national level must seek the following objecdves: (a) a conducive policy environment, (b) sectoral policies and programs that ensure services are accessible to the poor and vulnerable, and (c) temporary relief or *safety nets' for tmes of catastrophe or economic shock and pemanent nets for the chronically poor or vulnerable (see Table 1.1). A conducvepolicy ewnironment refers to the broad developme models underlying and integratng all of the approaches to poverty alleviation in a particular country. It also refers to the policy inme that will povide the appropnate Incentives and regulatory controls to ensure that policy objectives are reflected in sectoral strategies, progrms, and project. The policy framework inludes guidelines on the kinds of actions poverty policies should undertake, such as assessig the mpt of national policy interventons (such as devaluation, trade, fiscal and monetary policies, and market and administratve refonms) on the operation of labor, product, and credit markets and on their accessibility to the poor; supportng labor creation strategies to Promote sustaned econnmic growth; promoting ivafimafion; and maintaining broad social saety nets to ensure unversal acess to basic socia services. It also identifies te broad instruments to be used to ensure that poLicies are implemented: public expenditu managment to prmote poverty responsive secoral policies and progam (Lacey 1989; UNDP 1991), the identfication and elimination of gender biases in developmt programs (Heyzer 1992), financial incentives to promote the participation of the private sector in the provision of basic health services, and turgeting mecaism to increase the cost- effectiveness of service delivery to the poor Prnomotng wman resource developmen by broadng accs to socid and economc servces refers to sectoral program and policies at broaden the provision and susainability of social and economic services in such a way that they are increasingly acessible to the poor. Examples include national adult literacy programs thfat provide literacy and related vocational and life skills twrough a particuar region or on a national basis and that continue over a long period of time; integrated child developmet services that provide primary health care, nutrition, and related services to children over a wide geographical area; and itegrated rural development programs that provide a wide range of economic and social services to smal frmers and the landless on a contuing basis. Socal safety nets and targetngprogramsfor thepoor and weabkle are intended to (a) mitigate the social costs of structu adstment, (b) protect those umne to help themselves, and (c) build local capacity and e. Many of these tgeted projects are time-bound sets of actvities wit a single source of finacing and vwth dearly defined and montorable objectives. Frequently such projects are incorporated into broader programs. For eamle, a trgeted nutrition project for a particular age group that is fumded by an exteral agency and lasts for three to four years can be part of a broader sbtewide or national child nutrition progam, as in the case of the Tamnil Nadu Integrated Nutrition Program discssed in Chapter 8. Another example would 14 Midcad anaberger and Abdl Adz Table 1.1 The Components of an Integrated National Poverty Alleviation Strategy Level Component Example 1. Creating a Promoting econonic * Using public expenditure magem to conducive policy growth through labor ensure that investmet policies are pro-poor environment intensive policies and * Eliminatng gender biase with due attenion to * Eliminating distortions that promote capital- distributive impacts intensive production methods 2. Promoting human * Incsing access to * Giving investment priority to construction of resource development primary health and primary schools and healthcenters. by broadeing access education 6 Promoting access of girls to health and to social and educado economic services * Promoting economic * Increasing ownership of physical assets, opportunities particularly land * Raising productivity of assets owned or used by the poor * Improving living * Promotng sites and services and squatter conditon upgrading housing * Giving greater priority to the sustainality of social inhfratucfture 3. Social safety nets 0 Mitigating th socal * TaWrted health and nutrition prgram and targeting costs of strunnl 0 Employment guaratee programs programs for the poor adjustmet * Targeted credit prgrams and vulnerable 0 Protecng those * Subsidies for food, housing and basic services unable to help * Targeted credit and econonmc prgrams for themselves women * Safety to cover essential health services * Building local 0 Bottom-up, participatory planning capacity and 0 Leadership training empowernent 0 Strengthening the role of NGOs. ,, I .- f ,-: -', . D'' ' 'h P- n - : Povefy Alleation Straegis and the oe of Targeted Pvogma for the Poor 15 be a sites and services or slum upgrading project that is part of a municipal or statewide popular housing program, such as the Madras sites and services project discussed in Chapter 8. Table 1.2 shows how the tiemes discussed in this report fit into this framework. Ta 1.2 Po,ury AisdomPolicks, Program ad frjcts nias in Ths Book COrouavg PwLJCr EIIVIlCN1d PCIJLI AND PDOSRAEU PIOMOITL4O SOaAL SAPFElY WMiTAN HUMAN 3Dcx3R DEV.ELouIN TARGE ftOOKAMI ReaM ad IMrkI.x (ampur 3) .1. piou omof psiuuy heftd I. muhnim1 appoahas 1.Inruhr mge myism 2. Pnrmodng priet secto 2. Providins gm*etr ecess to dem poor 2. Spporemg hbgratd n3tidon peoject 3. Sfeagd3eaiz ml of NOOs 3. Improviqg Ur efficiecy of larli.smIle delivmy sysmm (Chaptor 9) 4. Prmoating womesmnacess to health (Chapter 4. Stzeqdngibel td role of tmdidonal medicin S. Promoda pardclpairy appruahes RuM Dcv,rpme (Chpter 4) 1. EvolutIn of poverty alleviation strategies 1. Promoting flexibe "leamnln approache I. Adapding die project cycle to Mhe * ~~~2. Ensring opramiromi ad fImancial sruunabilty 2. SociW analysis ci aMstancesof the poor 3. Importanc of daim bas and sound analysis 3. Lending modalkuuds 2. Making nml development project accessible I. ~~~~~~~~~~~~~~~~~4. Stzrngftening die inolvemerli of NGOs to women '0% 5g1. Defing nd measuring poverty HoSpin and Urban Development (Chapter ) , 1. Partcipatory approaches 1. Iden,ing the uban poor 1. Madru sites aj s'ices a._ slum 2. Intgmated multisetorhl approaches 2. ResvIewing and evalutn main program approaches upgradin (ChapterS8) 3. StrengthenIng local govemnment 3. EmpbasIzing lrrtltutiona arrangements 2. Incme gcneration for fth urban poor 4. Affordability, subsidies and cost recvery. 4. Assessing affodability an cost meovery (Chapter B) S Land tenure S. Involving NOos 3. MakIng housing and urban progTams * ~~~~~~~6. FinancIng urban services sccesible to women 7. Centrliaon vemus denttllztion S. Public versus private sector ; ~~~~~~~~~~~~~. .... 9. Environmentalsse *o.: .?.. he Rate of aredis In Poverty Allevation (Chapter 6) 1. Involving commrca binds 3. Evaluatg the main kinds of credit progrms 1. Design of credit delivery systems 2. EvaluNtg the prs and cons of subsidized credit 2. ESring access to bt poor 2. Gmss roots organization 3. Loan recovery 3. Loan pocedures 4. Loen pmoductvity 4. A noci ny and support services S. Specized credit prgram for women. WorkiNg Womens Formm Table 1.2 (continued) Gender Issues In Poverty Alevigton (Oczr at . 1. Gender biases in development strtegies 1. Gender bias in agriculture programs 1. Improving design of rural pmjects 2. Undesndlg suIvlval strategies of female headed 2. Asoing poverty programs for women 2. Imroving design ofuxbm pmojecs households (Chapter 10) 3. Gender biases In uban development . 3. Impmving Icome and employment- 4. Promoting women's economic opportunities in the informal generating projecs sector 4. ImprovIng design of health projects Design and MnageMM OfPoverty Pwgnun and Projects (Chapters 11-15) 1. he interface between poverty projects and tbe 1. Prmoting participatory approaches to development 1. Ydenfiication of projects policy environment 2. Rhodeoring bureaucracies 2. Social analysis 3. Stengthening tb role of N1ON 3. Sutnghenlng die mle of NHOO 4. Targetg Projet deign 6. Cost ecovery 7. OrgariEadomI sces. S. B cayo n 9. OGeder oenisve aproches 10. DesInI um_ieb project 2 THE AsIAN POVERTY- SCENARIO: PERSPECTIVES FROM THE 1990 AND 1992 WORLD DEVELOPMENT REPORTS Michael Bamnbeger The World Devoment Report (WDR) Deumition of Poverty A basis for comparing poverty levels througout the world was established by the 1990 World Development Report (WDR).' A range of poverty is identified by using both a "high and a "low' level of anml per capita income levels measured in 1985 "purchasing power parity" dollars: the lower esfimate is $275 per person a year; and the higher esfimate is $370 per person a year (equivalent to $420 in 1990 prices). This range spans poverty level definitions used in a numer of countres, inluding India, Bangladesh, Egypt, Indonesia. Kenya, Morocco, and Tanznia. It should be emphasiied that these definitons are only used for the purposes of interational comparison. Country-specific definitions are used when esimating trends within a particular country. The main purpose of this income-based definition is to permit innational comparisons. The 1990 WDR emphasized that poverty is mulinsional and this consumption-based poverty measure is complemented by information on nutrition, life expectncy, infant morWity (among cbildren less than five years of age), and school enrollment rates. The UNDP Human Development Report (UNDP 1990, 1991) has proposed that several of these indicators be developed into a composite index that combines -income, life exptancy, and literacy. Although this index has created considercble interest, some questions have been raised about the methodology used in consucin the first version of this index. Comparing Poverty in the Devloping Regions Table 2.1 shows that in 1985 an estimated 30.5 percent of the populion of the developing world was living below the poverty line of $370 per capita per year. The region with the highest poverty rate was South Asia. In 1985 an estimated 51.8 percent of its population was living below the poverty line of $370 per capita pez year. The poverty rate for Sub-Saharan Africa was only slightly lower, at 47.6 percent. The Middle East and North Africa had approximatly one-ird (30.6 percent) of ther population below the poverty line, while the percentages for Latin Amerca, East Asia and East Asia were 22.4 percent, 13.2 percent, and 7.1 percen, resectively. 1. This daptr is basnd an nrl dmwn from dIe 1990 and 1992 World DecvPmec Rep and odher recant World Bank studies. It is inmded to provide ody a brief itoduction to Asian poverty, and the reader is referrd to dt sour for more conplete infixmadon. 2. Many of dhsec aiticims are discussed in the 1991 Hina Dew{opna Rpr (UNDP 1991). 19 20 Mihaed Ramberger Table 2.1 Populabon and Poverty in the Developing Regions, 1985 Percentage of Percentge of Percentage of the Developing Developing Popuaion below Countries Countes Povert9 Povert Lint Popukdon Arl Developing 100.0 100.0 30.5 Countries South Asia 29.7 50.7 51.8 East Asia 40.2 17.3 13.2 Sub-Saharan 11.1 17.5 47.6 Africa Latin America 11.2 8.3 22A and the Caribbean North Africa and 5.5 - 5.7 30.6 the Middle Eaat Easte Europe 2.2 0.5 7.1 a World Development Report 1990, Fi'gur I (p. 2). See World Develpopmnt Report 1992, Table 1 (p. 218), for the population of Eastern Europe. I WorldDevelopmnt Report 1992, Table 1.1, p. 30. c 1The two previous sources. In 1985 South Asia had an estimated 29.7 percent of the total developing county populaton, but it was home to more than half (50.7 percent) of the developing world's poor.3 In contrast, East Asia contained an estimated 40.2 percent of the developing world's popuation but only 17.3 percent of this group could be classified as poor. Although Sub-Saharan Africa had an extemely high poverty rate, owing to its much smaller population base, it accounted for only 17.5 percent of the total poor. Latin America with 11.2 percent of the population had 8.3 percent of the poor. North Africa and the Middle had 5.5 percent of the populaion and 5.7 percent of the poor. Eastem Europe, with 2.2 percent of the populaton, accounted for only 0.5 percen of the poor. East Asia's low percentage of poor is reflected in its consistently high real growth rates of per capita GDP over the past twemy-fiveyears (5.1 per per year during 1965-73, 4.7 percent per year during 1973-80, and 6.7 percent per year duing 1980-89). During the same 3. Poor is defied as earning p:r capita income o ess ha $370 per year. wVey poorW is deied as earning Iss than $275 per caita per year in 985 prchase powe par- doars. The Asian PovJty Senwio 21 periods, the real per capita GDP for South Asia grev at a much more modest 1.2 percent, 1.7 percent, and 3.2 percent, reSpectively.4 The Asin Poverty Scenairio in 1985 Table 2.2 smmarizes the estimates for East and South Asia on differet indicators used by the World Development Repor to assess poverty in 1985. A comparison is also given for alU developing countries. A distinction is made between the wetrenely poor" (annual per capita income below $275) and "poor' (annual per capita income below $370). According to these income definitions, in 1985 a total of 420 million Asians were elassified as extremly poor (9 percent were in East Asia and 29 percent in South Asia) and 800 million were classified as poor (20 percent in EastAsia and 51 percent in South Asia). In global tenr, East and South Asia accounted for 18 perct of the exatremdy poor and 33 percent of the poor of the worid's developing countries as a whole. Table 2.2 also estimates a 'poverty gap," which is the aggregate income shortfall of the poor as a percentage of required minimum consumption. In India, for example, the total aggregate income of the poor would have to be increased by 12 percen to permit all poor households to achieve the ininmun required consumption levels. Informaion is also provided on mortality among te ppulation less than five years of age, life expecrncy, and net primary school enroDlment rates. On each of these three indicators, the situation is much less satisfactory in South Asia than in East Asia. South Asia is closer to Sub-Saharan Africa on mortality rates of those less than five and on life expecancy, although school enrollment rates are signfintly higher in South Asia than in Africa. 4. World Devopme Report 1990, Table 1.2, p. 11. 22 Midcad alwubeger Table 2.2 Indicators of Poverty in 1985 Used in the 1990 World Developmnt Report East Asia Oina South Asia India All Developing Countries Exremely Poor Nunmer 120 80 300 250 633 (millions) Headcount 9 8 29 33 18 index ( percent) Poverty gap 0.4 1 4 3 1 Poor (including extremely poor)' Number 280 210 520 420 1,116 (milions) [182] __ [532Y [11,107 Headcount 20 20 - 51 55 33 index C Social ndicators_ Under 5 96 58 172 199 121 mor| ity (per thousand) Life 67 69 56 57 62 expectancy (years)__ _ _ _ _ _ _ _ Net pimary 96 93 74 81 83 school enrolaet Note: The estimates for the numbers of poor differ fom those presemed m Table 2.6 as the latter use reestiuated figures included in the 1992 WDR. The 1992 estmates are given i parentheses for comparison. Source: World Development Report 1990. The Asian Povt Scoawfo 23 Rurad Poverty In 1988 an estimated 817.5 million people in South Asia were living in rural areas; this group represented approximately 74.5 percent of the total population C(able 2.3). About 441.5 million of these people (54 percent) were living below the poverty line. Information on the rual populaton below the poverty line was only available for four countries: the highest proportion (86 percent) resided in Bangladesh, followed by Nepal (61 percent), and India (51 percent). Pakistan had only 29 percent below the poverty line. The incidence of poverty is considerably higher in rural areas, where 54 percent of the populon faills below the poverty line, compared with only 23 percent in urban areas (see the next section). The higher incidence, combined with the fact that the majority of the popuation is rural, means that in 1988 approximately 85 percent of South Asia's poor were living in rural areas. In Chapter 4, Bucker and Johnson examine the distinctive features of life for the rral poor in South Asia. First, most of them are excluded from normal levels of production and consumption. They have very Umited access to productive assets and tend to occupy a peripheral position in the labor market, with the result that they must depend on a mix of marginal economic activities. Poor families are also frquently forced to spend a high proportion of their income on food. Second, many of the poor live in resource-poor areas and suffer from limited access to basic public services such as education and health. Third, the poor tend to be subjected to excessively high levels of risk and uncertainty because they have to work the poorest soils and to Jive in arem with the least favorable climatic conditions. In view of the unfavorable circumstances under which most of the rural poor live, rral development strategies must adopt an ingrat approach seekig to improve a broad range of socioeconomic conditions as well as to provide direct productive resources and employment opportuies. Urban Poverty hi Asia Although 80 percew of Asia's poor live in rural areas, more than 40 percent of all of the developing world's urban poor live in the cities of Asia. There are probably at least 135 million people living in conditions of absolute poverty in the cities of Asia, with over 110 million in the cities of India, the Philippines, and China (Table 2.4). These figures compare with approximately 77 million urban poor in Latin America and the Caribbean, 59 million in the Arab cuntries and East Europe, and 55 million in Africa.5 This vast number of urban poor in Asia live in squatter settlements, run-down tenements, or on pavements. They have received far less attention from govmets and intnational agencies than the urban poor in Latin America and Africa. The experence of India is perhaps typical of the approach in many South Asian countries. Although issu relating to rural poverty were explicitly addressed at least from the time of the Third Five-Year Plan (1961-66), it was probably not until the Seventh Plan (1985-90) that the problems of urban poverty were 5. Estimates tken frim Urban Devdopmt Division FY89 Sear Review of Urban Development Opeuaion: Reaching the poor bougb Urban Opedms,' Wad BDank (1989t: Tabk A.3). 24 Michel Bwmberger Table 2.3 Estimated Inddence of Rural Poverty in South Asia, 1988 Rural Percentage of Perentage of population rural Rirad Popution below powerty popuion Cowmby POpulon Rurl (ndmions line (millions belowpoverty (milions 1988) 1988) 1988) line (l977T- Afghanistan 12.6 79.0 n.a. n.a Bangladsh 95.7 87.0 82.3 86.0 Bhutaa 1A 95.0 n.a n.a. India 598.6 73.0 305.3 51.0 Nepal 16.4 91.0 10.0 61.0 Pakistan 79.4 69.0 23.0 29.0 Sri Lanka 13A4 79.0 n.a. n.a. South Asia 817.5 74.5 441.S 54.0' a. Mean based on countres for which data wer available. Sourre. Adapted from Table 4.1, Chapter 4. 1 systematically addressed.6 A number of iniatives were launched at this time a number, such as the Self-Employment Programme for the Urban Poor and the Basic Urban Services Programme. Similarly, many of the leading social research institutions in South Asia have focused largely on problems of rural development and rual poverty. Even now that the exstence of urban poverty is gradually being ackmowledged, many studies still downplay its importance because they think the urban labor market has a strong absorptive capacity and a self-regulating mechanism in that people feely choose to migrate to the cities and would not continue to do so if the conditions there were not better than in the rural areas they came from. One obvious reason that urban poverty been ignored by policymakers and even by social researchers in most parts of Asia is that the urban poor are greatly outnumbered by the rural poor. Only 25 percent of the Asian population lives in cities, whereas in Latin Ameica more than 60 percent of the population lives in urban areas and the urban ratio is rapidly approaching 75 percent.' Also, the urban poorhave rarely beenperceived as a serious political and secui threat 6. Natioml itu of Urban Affais (199o 3s). 7. Diego C arion, Mario, Vasnez and Jorg Gai AceAss by tbe Urban Poor to Basic nfk r Services: Lain Ameri Regio Paper." prepard fior the hftinrr and Urban Devdeopment Division of EDI (November 1989). pp. 6-7. The Asian Poven Scenario 25 Table 2.4 Urban roverty in Some Asian Countres, 1988 and 2000 (thousands) 1988 2000 Urban Urban Percenag Urban Urban Percenta Populaton Poor e Urban Populaidon Poor ge Urban Poor Poor Bangladesh 13.50 7.56 0.56 26.62 11.32 0.43 China 225.82 14.90 0.07 314.61 5.42 0.02 India 214.54 81.74 0.38 330.02 117.75 0.36 Indonesia 49.95 6.54 0.13 77.07 i5.06 0.20 Korea, 30.36 3.67 0.12 40.59 0.00 0.00 Republic of Malaysia 6.77 1.99 0.29 10.32 1.95 0.19 Nepal 1.59 0.75 0.47 3.30 1.66 0.50 Philippins 24.25 12.12 0.50 36.29 18.20 0.50 Sri L nka 3.17 2.04 0.64 4.75 1.49 0.31 Thailand 11.84 1.78 0.15 19.28 3.36 0.17 All Asia 591.00 136.53 Avenge 0.23 0.20 Source: World Bank, Urban Development Divsion (impubhshed). to Asian societies m :he way that the Latin American poor have been feared from tine to time in cities such as Rio de Janeiro, Caracas, Lima, Bogota, and Ciudad de Mexico. Some would also argue that urban poverty is more dificult to measure. Despite the huge and highly conspicuous urban slums in cities such as Manila, Bombay, and Karachi, many of the urban poor are almost invisible. While it is relaively easy to defin and measure rural poverty, because most of the poor live in clearly definable areas and the variations between family econonic conditions are relatively small (given the limited range of economic activity), the same is by no means trwe of urban poverty. In most cities, the majority of the poor do not live in the large and conspicuous slums but are scattered throughout the city, living in a wide variety of tenements and other kinds of crowded and substndard housing, or else have no housing at all. Even in the United States, the Census Bureau has ackmowledged that millions of homeless people in the cities were not detected or measured in the 1990 Census-despite the fact that large-scale special studies were mounted specifically to identify and count the homeless. If the poor canot be located by the U.S. Censsds Bureau, what hance do cities such as Karachi, CalMta, or Manila have of producing accurate estimates? 26 Midcae Ramberger In 1988 approximately 23.1 percent of the urban population in Asian cities had incomes below the minimum subsistence level (Table 2.5), compared with 41.6 percent of the urban population of Africa, 33.9 percent of that in the Europe, the Middle East, and North Africa region as a whole (called the EMENA region by the World Bank), and 26.5 percent in Latin America and the Caribbean (Urban Development Division 1989). Despite substantial variations from one country to another, the overall proportion of the Asian urban populafion classified as poor was lower than for the other developing regions. Over the period 1988-2000, however, the total number of urban poor in Asia is expected to grow faster than in Latin America and EMENA (2.4 percent per year compared with -0.8 percent and 1.3 percent, respectively). One of the contnbuting factors is the higher projected population growth rate in Asia (3.3 percent a year) than in North Africa, the Middle East, and Eastem Europe (2.8 percent) and Latin America (2.8 perce). Table 2.5 Magnitude and Trends in Urban Poverty in Asia, 1988-2000 Urban Poverty 1988 Urban Poverty 2000 AwaGrowth us Urban PopuFon and Urban Poverty, .1988 2000 Region Thousands Percentage Thousands Percenrage Percentage 1 Percentge of Urban of Urban Change in Change in _______j Populaion Popukaion Urban Urbn _____i___ .________ .__________________ ,________ Popu ation Poor Asia 136.5. 23.1 181.2 20.6 3.3 2.4 Africa 55.4 55.4 41.6 41.5 5.4 5.4 ENENA 59.5 59.5 33.9 69.3 2.8 1.3 Latin 77.2 77.3 26.5 70.4 2.8 -0.8 America . Source World Urban Dievelopmei ivision (1989), Tables Al and A3. Nevertheless, the proportion of the urban population living in poverty is projected to decline for Asia as a whole, from 23.1 percent in 1988 to 20.6 percent in 2000. Future Trends in Asian Poverty Table 2.6 summarizes the 1992 World Development Report estmates for the likely regional trends in poverty between 1985 and 2000. It is projected that for all developing regions the proportion of the population living below the poverty line will fail from 30.5 percent to 24.1 percent. Significant declines are projected for both East Asia (where the poverty proportion is 7he Asian Poveny Scnario 27 excted to faill from 13.2 percent to 4.2 percent) and South Asia (where the proportion is expected to fall from 51.8 percent to 36.9 percent). In all the other regions no significant change is expected in the proportion of poor, although the _nubers of poor in Sub-Saharan Africa will increase from 216 million in 1985 to 304 million in 2000, those in North Africa and the Middle East will increase from 60 million to 89 million, and those in Latin Amenica from 87 million to 127 million. Table 2.6 Poverty in 2000 by Developing Regionr Percenage of the Populaion below Numbers of Poor (numlons) the Poverty Line Region 1985 2000 1985 2000 East Asia 132 4.2 182 73 South Asia 51.8 36.9 532 511 Sub-Saharan 47.6 49.7 184 304 Africa East Europe 7.1 5.8 5 4 Middle East, 30.6 30.6 60 89 North Africa Latin America 22.4 24.9 87 126 All Developing 30.5 24.1 1,051 1,107 Countries Source: Ravailmon, Dat, and Chen (1992). Cited in World Development Report 1992, Table 1.1, p. 30. Table 2.5 showed a less dramatic projected decline in urban poverty in Asia. Although the urban poor will grow less rapidly tan the urban population between now and the year 2000 (2.4 percent increase in the poor compared with a 3.3 perent increase in the urban population), the share of the urban population projected to be living below the poverty line in the year 2000 will only have fallen to 20.6 -percent compared with 23.1 percent in 1988. - - - I - -.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Part 2 SECTOR ExPERINCE WITH POVERTY ALLEVIATION IN AsIA Sector Experience with Poverty Alleviation in Asia Editor's Note Ths section containsfive review chapters commissionedfor the semnwa. Three examine the povert alleviation experience in rural developmnt, homing and urban delopmet, and primy healt care. The fourth assesses te role of credit programs in povert alleviation, and the fifth disusses gender issues in poverty alletion. The emphasis in pover aleviation strategies in these sectors has been tofornulae a sectorpolicy, which has then been mplemented through broadprograms and specjfic targetedprojects. A wnuber of common thes and issues run through tese five chapters. fPrst, thy aUl recognize the importance of a conducivepolicy ezn ent thatfocuses spedficafly on ensuring ta services andprograms are accessible to the poc- and are designed to respond to their needs. In the case of housing, tis requires appropriate pricing and land tenure policies, as well as measures tha drectly involve the private sector. Emphosis must also be placed on upgrading radwer an denolitdon. In dhe area of primy health, these goals can only be achieved if there is a polWical conmnmient to guarantee a sufflient level of fundg to susaDn serve deli very and acceptance of the concept of cost shaing and a more active promodon of the role of the prie sector f(th its conunercial eleme and NGOs). In rural development and credit, i is necessary to accept dha the poor are "bankable" and thatprograms should be rnm on a solid commeria basis with emphasis on cost recovery, carefid selection of economically viabk projects, and guaranteed folow-up loans. Here, too, polieial commitment is vital to ensure programs are targetedfor the poor and that beneft are not uaphoned off by powerfid rural interest groqps. Another key policy requreme is ta gender be recognized as an inportat issue and tha specdi polides be designed to eiinate gender biases and to ensure that programs reond to the speciic needs and potenia of women. 7he secondpoint ephasied throughoutthis section is the needforpartdpatory, bottom- up approachs. All five chapters stress that top-dow project planning and mmnagemn approaches in the social sectors have by and &rgefailee he argue that benefic mast be actively involved in the selecon, design, and the management ofprojects whether these be lw-cost hoig, nmnor irrigation works, creditprograms, managem f localf alth servses, or the planning and manement of basic urban servces (which are the primy concern of women - yet their intaeret arefrequently not taken into conside n). An essental ingredient of this approach is the investment of resources and time in loca capaciy bcuding, at boh the community and local government leveLs, and in partidpatory plaing, which in many cases requires a dramatic change in he tradonal role of government planning and mpmetatin agnaes. A third theme is thatadmiistrative andfianda responsibility needs to be decrizeS Altough many countries have eperimented with adinistrative deetalzaton, few have delegatedfncial responsbity or have had the polical will to cede control to regional and local authorities. 31 Fourw/, management practices at all levels are in great need of improvement. Poverty programs, party because of their welfare orientation, have on the whole been poorly managed, whether at the policy/planning, program, or project levels. More effort needs to be put into uiproving accowntability, management informaton systems, financial control, and the feedback on projectperformance. This cannot be done effectively without delegating authority andfinancdal control to managers so that they will have the necessary autonomy they need to manage. Rfth, prograns should be more demand driven. Programs should move away from centrally defined targets and should use participatory planning and market mechanisms (see below) to give the intended beneficaries a greater role in defining their priorties. Tris has importn implications for how programs are managed as it is essential to have sufficient managementflexibility to be able to respond to rapidly changing demands for different services. Sixth, greater use should be made of market mechanisms. This means the private sector (both commercial enterprises and NGOs) should play a greater role in providing services, and that some consumers should pay the market price for some servces. The question of which services should be provided free, which subsidized, and who should be eligible f&r subsidies raises a complex range of issues that are only bnefly addressed in this secion. Getubig argues in Chapter 6 that subsidized credit should be eliminated or substantially reduced. In Chapter 5 Yeung notes that pricing and cost-recovery consitute one of the key issues in housing. The move toward market prcing is recommended even for government-provided services (such as primary health or eduation). An interesting application of marketing is the use of socal marketing to promote health programs, particularly fanmly planning. Seventh, NGOs should play a more active role. There is considerable evidence on the comparative advanages hat NGOs enjoy in plkaning and managing many kinds ofprogranmfor the poor. Their grass roots approach, flexibility, and emphasis on efficent management and accuntability make the stronger NGOs well suited for assisting in the design and mangement of small-scale projects in varying envronments andfor catering to populations dut have limited access to, and confidence in, regular government programs. Every sector contains examples of NGOs ta have been abk to put together cost-effective delivery systems reaching out to large populations groups excluded from most public sector programs. Among the highly successfud NGOs are: in the credit sector Grameen Bant, BRAC, and the Working Women's Forum; in low- cost housing the Orangi Project in Karaci; ASAG in Ahmedabad, and Freedom to Build in Manila. In the area of health, Bangladesh's NGOs provide 40 percent of contraceptive services and organizaons such as BRACplay a maor role in oral rehydration therapy and otherprimary health servces. However, further research is clearl needed to conduct systematic comparisons of the cost-effectveness and accessibility of government and NGO programs (see Chapter 3for a discussion of evidence from the health sector). Eighth, aU the chapters mention the needfor targetedprograms and conclude that despite the confinuing debate concerning the merits of programs targetedfor the poor (see Chapter 1), targeting is justie wider some circumstances. Stai (Chapter 3) argues that the goal of providing universal healh services has diveed atenion from ensuring that these servces are accessible to the poor. He stresses the needfor targetedprograns for the poor, many of which should be dmand, ratherthan supply, driven. Bucher and Johnson (Chapter 4) state tha a target group orientation is esset in rural developmen4 both to prevent resources from being siphoned off by powerfid groups and to ensure at programs adapt to the characteristics of nucroregions. Getubig (Chapter 6) states that many of the most successful credit programs (such as Grameen Bank) are based on thefirm beliejr that targetedprograms are essential to protect resources from being siphoned off by more powerful groups. Successful credit programs usualy operate through Part i Editor's Note 32 solidaty groups that are based on strict eligibility criteria to ensure that only the poor benefit. Heyzer (Chapter 7) also points out that it is essenial for programs to respond speciically to the needs of women in general, and poor women in particular. She cautions, however, against the kind of central numerical targeting, common in both rural development andprimary health, that is used to assess the performance of extension workers. Under such systems families are frequently pressured to accept loans or assess they do not want, or extension workers may limit their work to those activities that can be quantified ffor emple, the number of children vaccinated or number of houses visited). Nnth, conventonal project design needs to be adapted to the special requirements of poverny allemation. To be effeive, poverty alieviation projects must place grea,er emphasis on participatory planning and design, greaterfexbility in the implemntadon schedule and deftnition of objectives, andflexibility in the allocation offinancidal resources. It may also be necessary to have smaller and simpler projects. Therefore it will be paricularty important to ensure that project design responds to thze needs and capacities of beneficiares in all sectors and that they are actively involved at all stages of planning and implemntion. Tenth, integrated servce delivery is essential. Although ts remans a controversid issue, it is widety recognized that poverty is multifaceted and dtat little change can be produced in the overall quality of life of a poor household through the provisbn of a single service such as nutriton, beter housing, or credit f the broader causes ofpoverty are not atced. In practce, however, it has proved to be administratively dfficult to irplement integrated rural development programs, health and urban services, or credit, markt=ig, and technical assistance. Many programs that attempt to do so have become extremety bureaucratc d inefficient. Integrated rural development programs, in partica-ar, have been widely criticizedfor their ineffectiveaess. Nevertheless, the integrated approach to poverty allenation seem; desirable, even though individu servces may be delivered separately and possit!y by different orgaitions. It was stressed that NGOs and local organizatons have an importn role to play in geneag deman4ddefining priorities, and inproving coordination between diferent servce delivry agenies. Allfive chapters mess the needfor effective coordination between thepolicy, program, and priect levels. A.though these chaptersfocus largely on fie project andprogrm levels, they all support, implictly or explictly, the "three-legged" approach to poveny alleviaon advocatid by the 1990 World Development Report. Bucher and Johnson speczfically advocate ths integrated approach for rural development, Getubig emphasizes the need to integrate creditprogramsfor the poor into an overall commercial credr policy, and Heyzer notes the importance o,f creating economic opportunities for women and of ensuring ttitese programs are consistent with broader labor maaket and economic growth strategies. Chapter 3 on health by Sata and Chapter 5 on urban homsing by Yeung pay less attnion to the macroeconomic environment Part 7 Editor's Note 33 3 THE ACCESSlBLMr OF POPULATION, HEALTH, AND NUTRITION SERVICEs TO THE POOR: THE ExPERiENCE OF SOUTH AsIA J. K. SSa Investment inn human capital-particulkay healt, education, and nrwiion-has come o be accepted as a key comwonent of any s inepoverty alleviaon strategy.' Internaional agenda have t fdweore srongly supported human resource development programs and governmn arond the wrld have responded with appropriate aciom. Following the Alna Asa declarion (1978) of hweth for alP and the adoon of a pnmary health care approach, the South Asian countries rqpidy exanded dthir penpheral population, hea, and nurition servce (PS) eliveay systems. In paricn&z, thy built heah caters and snbcenters orposts and established cmmutyhoedheath servces provided through commnity health workers. Despite a massive cpnsion ofjfcles and staff, the PS systems have not performed wel when it comes to heping the poor. Internatona aplerience suggests thatsustabk PS delivey sysms designed to reach the poor shoud provide effecive servce deivery at the village lewl with back-up support from higher leves, should natonalize responsilitiesjfor warious servces, ehance people orientaton of sernce provids, emhasize dmad creaion, and bui trong Wcommuny ns. PS delivery systems in Sout Asia are pursuing ths deald, b there are serous shortfalLs. Nongovernment organizaions (NGOs) have generay been more successd in reacing te poor. Alhough growng, their total involveme is small conpared with the PS deivery systems tha governmets operate. Muc more lernng is necessary belore it wil be possible to form a successfupartnership between the governent d N GOs. Four obstacle& stil need to be overcome before sustable PS delivery systems can be bilt to meet the needs of the poor: the lack of a conring potical comtmin the neglect of the process aspects of impulenatdon, undefowding. and ndequate management capaciy. As coverage evels improve, targeted strategies would be ncessawy to supplment PS delivery systems. Depndng on the areal distrIbuton of the poor and characteristics ofthe PS delivery system, it may be necesary to udeake geographic targeting, promote at-rsk approaches, wide chces for the pwoor. ad eploy multsetoral qpproae. 1. This chaptr borows hei fim - eale paper ppaed for fte Wod dank, Providing Perpheal Ppulatn. hah, and Nidan Services to die Poor-What Is eeded.' Mhe ar is grN o Mihe ambeg for conmunts on tis drft and to Bulatao, C. Ohauberiin, Jame Ott., Jlea-Isis Labaa, A. R. Mshean4 and R Sto for ti cms one eadier drfts 35 36 J.Saia Four actions are necessary to build sustainable PS delivery system. Firs the base of ifonnaion tabou the poor needs to be strengthened. Secon4 poliaes need tofocus on the poor. Third programs should strengthe managemt capad4y for designing and iplementing PS delivery systems. Fouih, ssteanic research efforts are needed at the local and regional levels to hDp refine responses to the problems of providing PS to the poor. Before tde 1970s, health services in Souti Asia largely consisted of curative institutions. In the 1970s, pressure developed for basic health care for a larger populon through the establishmen of health centers, posts, or s:.bcetr. These sevices were accessible to only a few. Preventive services were also inadequate, and the bias of service providers was toward secondary care. Significant gains in health status came mainly from the control of malaria and other epidemics and the eradicadon of smallpox. By the mid-1970s, a primary health care (PHC) movement was born, spurred by the remarkable buccess of small-scale, largely nongove oins' (INGO) projects and by the expericenes of some -countries (notably China). These pressures culminated in the Alma Ata declaration (1978) of the goal of "health for all." More than a decade later, there is still some confusion regading the meaning of PHC. PHC is defined not only as essential health care made universally available to individuals and families in die community by means acceptable to henm, but it also implies their ful partiipaton. Furhiermore, intersectral action is emphasized to improve health status i a holistic sense. Many, however, tbink that such comprehensive and communitybased health care is too idealistic a concept and impossible to achive. Tangible decreases in morbidity and mortality comparable to small-scale projects have not been reported (vanl Der Geest, Specknin, and Sretand 1990). The most concrete action, since the Alma decaation and the accepance of the goal of "health for all, was that all the South Asian countries expanded their periphral population, health, and nutrition service delivery in ue by addig health cemers and subcenters, and by establishing conmumity-based healdth sevces tugh community healdth workers. In India, the number of subcents- more than tripled, from around 35,000 in 1975 to 120,000 in 1989. In Pastan, the number of public servce outlets increased from a few hundred to about 5,000. The PS are defined as essenially outreach or outptent services roudnely delivered by the health system. Most often these incude s petaning to matemal and child health (MCH), preventive health, and oi d the tment of mior ailments and health eduion. In some countries they may also include growth mon ng and supplem y nutridon of younger children. Generaly, several ministries 2re involved in delivering sevices, and thus their activities need to -e coordinaed. Although the health status of the poor has improved in response to these serces, it has not reached the level of the population as a whole. PS coverage, particularly MCH and first-level 2. Howev. Sri LaU. ha wldeeloped preveie heat cae sevi ad odhr counri ad recognizdt ipoim of prmay health care. For insgamce, Ebt lmomue (I4 inrcm dd a sysem fs dedivay of puinmy haft care fr ndia. z .. . ;. The Accessibilty of Popuaon, Health and Nutrton Programs ro tFe Poor 37 treatment for minor ailments, has certainly increased.3 Also, with the expansion of PS delivery systems and overall socioeconomic development, infant and child mortality has declined and contraceptive use has inrease-d. These rates coninue to show a favorable trend. The problem, however, is that governments have been mainly concerned with improving the coverage and utilization of PS delivery systems and have neglected to pay enough attention to "who has not been reached and why. " To answer this question, it is necessary to review experiences in South Asia in order to assess the role of PS in the provision of sustainable health services to the poor. Perfonnance in Relation to the Poor Reacing the Poor Despite the massive expansion of physical facilities, incremental policy refoms, and program improvements, the performance of the PS in relation to the poor has been disapoinig (Justice 1986; Khan 1988; World Bank 1989c; Murthy and others 1990) (Box 3.1). Current evidence suggests that much remains to be done to provide adequate PS coverage to the poor. Although income and education certainly have an effect on health status, PS coverage has been shown to bring about significant improvements in health.4 Problems in Reaching the Poor Several factors have made it difficult for PS to reach the poor. Distance is a major deteminant of service accessibility, as can be sew in three districts of Madhya Pradesh, India. Only 26 percent of the population use health facfilities that are more than 5 kilometers away, in comparison with 38 percent of those who live closer. A study of three upazillas in Bangladesh found that post-neonatal infant mortality (28-365 days after birth) was 26 per thousand live births for areas in which a qualified allepathic doctor was available within 2 kilometers, in contrast to 45 or more in other areas (Paul 1991). Despite improved physical access to PS, considerable iewqualities persist among geographical areas because of biases in the location and staffing of facilities. Usually the income and education level of households determine the ability to use PS. Studies in Bangladesh and India have found that low-income households rely more on home or traditional remedies, whereas higher income households preferred to be treated by modem physicians. In Sri Lanka, where literacy levels are high, the proportion of children with health cards increased from 65.7 percent among mothers with no education to 86.8 percen among mothers with more than secondary education (Sri Lanka 1987). 3. In a wider sense, health service coveage would mean adequate access to and actual utili2aion of effective and efficiemly delivered services. This is quite different fmn adhinisuave definition that may mean no nwm than a number of health facilides and staff in some statisdcal relatsip to the popuation (Gish 1990). 4. The poor have become more vunrable, as the traditional safety nets and mchanism for survival, in case of catstrophic events of ilss, are weakeig. Consequently, teir depende on PS divery sstem is growng. Modem PS deliey sems. rater than enable existing cultral and community mechasms, have sought to supplant then. Some of the populaion, health and nutrition technlogies introduced by PS delivery stms have yielded results. 38 J.K.Satia Box 3.1 How Well Do PS Deivery Systems Reach the Poor? Data on PS coverage of the poor is fragmentary because the concern for reaching "all" has clouded the concern for reaching the poor. Although, in theory, PS delivery systems are expected to cover everyone, in practice there are severe shortfalls. In Bangladesh, for instance, only 17 percent of the households having operational landholding of less than 0.5 acres sought illness care from government semvices and the remaining 83 percent sought it from either private allopathic or traditional practitioners (Khan 1988). Uneven coverage of PS delivery systems has been well recognized in South Asia. Although not systematically analyzed, PS delivery systems tend to mirror social inequities. Generally, those who have higher incomes, are educated, are closer to health facilities, or have privileged access (say civil servants) are served better covered th those at the periphery. In India, only 70 percent of the required numnber of priary health centers and subcenters are established in trbal areas in comparison with about 85 percent in the country (Iuda Yearbook 1988). Similarly in Bangladesh, vacant positions for family welfare assistants totaled about 10 percent in more accessible areas compared with about 30 perent in remote areas, with a comparable difference in performance (ICDDRD 1990). Even in Sri I anka, attention by trained personnel at delivery ranged from 77 percent to 98 percent in the provinces (Sri Lanka 1987). The poor consequently suffer from greater health problems. For instance, 23 percent of the population residing in the socioeconomically backward states of India account for 40 percent of all infant deaths. Gender differences also persist; generally, women and female children suffer the most. Consequendy, fmae life expectancy at birth in South Asian countries, except Sri Lanka, is lower than male life expectancy, although these differences may be declining. For a number of reasons, the poor fail to use these services even when they are accessible (ADB 1986). Those on the supply-side include the lack of medicines, improper behavior of the health staff, low skills and motivation of staff, and lack of community involvement. Demand is governed by health behavior and beliefs, the lack of resources to provide access to facilities, impediments created by the social structure, and the poor quality of services (Srinivasan and others 1986; Talwar and Bhatia 1985; Khan and others 1987; Khan 1988). The switch between various service providers, or medical pluralism, because of uncerainty in the efficacy of treatment, complicates user behavior. A careful evaluation of provider and consumer incives is necessary to identify the banriers to increasing PS coverage of the poor. The benefits to providers can be increased by giving greater weight to coverage in their performance evaluation and linldng it with a package of suitable monetary and non-onitary innives. At the same time, the costs to providers can : . - - . The Accessbilty of Populaion, Heath and Nutrition Program to the Poor 39 be reduced by improving mobility and increasing travel allowance. Consumer benefits would be higher if the service mix included those perceived to be of high priority; if information, education, and co cation (IEC) activities were strengthened; and if there was peer pressure for behavior change and the quality of care improved. Similarly, the costs to consumers using PS delivery systems can be reduced by improving geographical access, reducing waiting time, and making the system more reliable. Sustainable PS Ddlrery System International Expeience There is considerable evidence to indicate that PS delivery systems are more likely to reach the poor than are higher levels of health care. It is therefore imnportant to determine exactly how much we know about establishing sustinable PS delivery systems and what can be done to increase their coverage.5 A review of international experiences suggests that five areas need attention: service delivery structure, organizational arrangements, people orientation of service providers, demand creation, and communty links (see Box 3.2). Developing Swwinble PS Delivery Systens in Soutlz Asia Through successive reforms1 South Asian PS delivery systems have been attempting to attain the goals outlined above but serious problems persist. STRucTuRE OF PS DEUVERY SysrEms. Several countries have developed three-tier PS delivery structures for rural areas. These consist of a village health worker (VHW, also referred to as community healdt worker, auxiliary midwife, volunteer) at te community level; health subcenters or posts (also known as dispensaries) sering a population of 3,000 to 10,000; and a health center designed to serve a population of about 30,000. South Asian versions of this servce delivery structre vary considerably, as can be seen from Table 3.1. The great difficulty in most of these cases is that village service delivery remains weak. Furthermore, although a large number of health centers/subcenters now exist, the population does not seem to make full use of them (ADB 1986). This underutilization has several causes: inaccessibility, poor training, personnel problems, irregular drug supplies, low staff skills and motivation, inadequate supervision, more attractive medical facilities in nearby areas, and contnued use of indigenous health care resources. Much can be done to improve the use of these centers by improving their fimctiomng, encouraging community involvement, and planning their expansion by taking into account the other health services in the area. While most NGO programs and many counies have been able to use VHWs to effectively build links between health staff and the poor (Berman and others 1987; Walt 1988), the South Asia VHW progms have not been too successful. In India, the VHWs, are part-time volunters who are paid a small honorarium But selection problems (whether they should be 5. Ile term 'susinability of PS is used to imply contied deliveIy of services and production of benefits, mainenace of physical frastuce, long-nm insttutional caacity and suppon from key smakeholders Bainbrger and chlem 1990). 40 J.K.Siaia Box 3.2 Sustanable PS Delivery Systims tonal Eie A review of selected country experiences and interventions to increase coverage suggests that sustainable PS delivery systems have the following characteristics: * They provide effective service delivery consisting of village level presence, appropriate structure of health services, and suitably located health facilities. * They have a sound organizational base through rationalized responsibilities and commited middle management. * The enourage serce providers to become onented toward people through work programming, in-service training, supportive supervision, coverage-oriented monitoring systems and fimctioning logistics support. * Tney pay mention to demand creation activities by attempting to understand bineficiary needs, offer an appropriate mix of services, improve the quality of care and sensitize the community through IEC and socil marketing. * They develop strong commuity links through a mix of comnnmity volunteers, special groups, and village health communities. Each of these elements makes a vital contribution, but more inportant, they inteact yngically to increase PS coverage. male or female and a concern that the decisions may be politically motivated) and vaillatng policies regaruing the financing of VHWs have reduced their effectiveness. These kinds of problems also showed up in Nepal's MCH-Family Planning Progam, especially among Panchayat-based health workers. VHWs have bad a poor record of success for several reasons. First, doctors and paramedical workers who focus on technical approaches to health care tend to have little inteest in ivolvn comnumty-based VHWs in health issues. Second, VHWs need to understand the 6. Aldbugh most NGO pogum frored a fanale worker who had strong links th comnmy, ffiere has be consWiderable debe over who constutes a better VHW-male or fkmale. married or single, youmg or old. But the cvidmce an this issue is poor. -7te Accessibility of Populaion, Health and Nutriton Progrwas to the Poor 41 reationship between poverty and health and what they can do about it. It is naive to assume that VHWs can play a political role where there are considerable inequities in the society. In most such cases, they will be perceived as extensions of existing health services. Finally, the quality of VEW activities may have bee generally poor, given the problems in their selection, trining, and supervision and in the logistics of providing their services. The potential of VHWs can only be realized if their work is realistically targeted and they are given better support and supervision. Furthermore, they would need more financial and managerial resources than was originally anticipated. In the uncertain economic climate of today, a case can be made for reforming and consolidating rather than expanding existing programs. Sri Lanka has experimented with purely volunry VHWs, who obtained no financial support from the government. However, the desertion rate among unpaid VHWs was high (walt 1989). Most national programs pay their VHWs either a salary or an honorarium, but few proide sustained communty financing of VHWs. According to a recent WHO draft document, there is litte evidence that the mobilization of volunteers in national programs is an effective policy (WHO 1987). TRADITIONAL BRTH ATrENDANTS. In South Asia, traditional birth attendams (TBAs) provide some antenal care and usually assist with the delivery, even when health services are available nearby. Therefore, almost without exception, national programs' have been launched to train TBAs to upgrade their delivery services, but rarely other services. Such training has shown to improve antenatal care and immunization. In one snall expeiment, TBAs were trained to function as VHWs and were found effective (Ford Foundation 1987). Although TBAs are recognized as a means of expanding MCH and family planning (FP) service coverage, no accounts of successful national programs of susained TBA involvement have been found. TuDmoNAL PRACTrnoNERs. South Asian countries have a large number of traditional practitioners, but there are vast differences in their skills and methods of treatnent. Bangladesh, for example, has 4,500 qualified allopathic practitioners along with an estimated 2,000 Unani, 4,000 Ayurvedic, and 10,000 homeopaths. There is some evidence that popularity of traditional medicine is declining. Most goverments have formal Enks with traditionalpractitioners. Despite the high level of availability, accessibility, acceptability, and afforlability of traditional medicine among rual people, actual collaborative programs between traditional and modern biomedical practitioners have been few and the question of how to utilize their potential in improving health services still remains unresolved (Thapa 1986). Urban PS delivery systems have been neglected. In urban areas, the PS delivery system often consists of local government dispensaries. Since distances are not large, it is assmed that the poor in urban areas will be able to use the available health infrastructure. Fragmentary evidence suggests, however, that distance and other barers to contact do operate and that health status of the urban poor may be only marginally better ffian that of the nual poor. A three-tier PS delivery strucntre, comparable to that in the rural armas,' could be expensive, particularly in view of the already existing private and public service delivery structure. Perhaps only the most peripheral part of PS-the community worker-needs to be added and other existng facilities rationalized, as the experience of an urban health project in Calcutta shows (Heaver 1989). 7. By 1988, m- hn 25.000 TBAs hi Bagaesh and 580,000 in India wter. reported to have been traed. S. For instance, the Knmn Commitee (1982) in Iudi prepared an urban vamping sceme under which a fenale voluntiy health wo*rer for a popuad of 2,000, and a health post for 50,000 popation staffed by a doctor ant nltpwpowe outech workers was recon1nded. 42 J.K.Sada Table 3.1 PS Delivery Stnrcture in Some South Asian Countries Level I Level 2 Lvel 3 Country Type Popula- Type Popula- Type Popula- tion tion tion Bangladesh HealIh Village HFWC 22,000 Upazilla 200,000 and FWA India VHG 1,000 Sub-cente 3,000- PHC 20,000- 5,000 30,000 Pakistan Disp/ 1,000- MHC/Basic 5,000- RHC 25,000- Subc 2,000 health unit 10,000 D100,00 Sri Lanka GHC 3,00 Subdiv. 20,000 DHC 60,000 Nepal Health 1,500- Health post 10,000- Center NA worker 3.000 20,000 Note: Seice delivery system in Sri L nka and Nepal is being rationalized and the stmcte indicated here is tentative. Source: ADB (1986); World Bank (1989c); and Tuladhar (1989). Therefore, a flexible service delivery patten, appropriate for local needs and situations, is called for in urban slums. ORGANIZAI1ONAL ARRANGEMENTS. By the mid-1970s smalpox had been eradicated and incifence of malaria had declnd, with the resuk that most counties turned their attention to integrating various health programs, inchuling family welfare (see Box 3.3). However, several difficulties arose in rationalizing the rganiional responsibiities at various levels of health services and in delivering the integrated range e! health serices. In India, the field functionaries of several vertical programs were mtegrted into a new cadre of multipurpose workers,. but it took many years to work out salaries and orgaiztional responsibilities. Several programs contiued to opeate vertically, although responsibility for them was inegrated at the distrct level. In Bangladesh, the ministries of health and population control were reorganized several times. Health and family planng have been fiuly inrated at the upazilla level and below (Badrud Duza 1989). In Nepal in 1987, the government decided to imegrate all services, including family planing, by integrating all vertical projects in seventy-five districts, which were expected to cease to have a vertical identity by 1990. The ministry's organization was restructured to ac te most of the vertical projects' staff members. Integrted health services are to be made available through the District Public Health Office (uladhar 1989). ln Pakistan, integration still appears to be at a preliminary stage (World Bank 1989c). Sri Lanka, The Accessibility of Population, Health and Nutrtoion Program to the Poor 43 which has well-established preventive health services, is still debating about and exper i with appropriate modes of integration of all health services. Thus, scholars seem to be divided on the issue of integration. However, all the countries have favored integrating services to rationalize the nunber of field workers and reduce their area of operation. Because the poor are unable to adequately integrate services from different channels, they are more likely to benefit from integrated, but more accessible, services. aearly integrated programs require more sophisticated planning, greater staff motivation, coordination, and control and their success will depend on how these managerial processes are strengthened. Once the responsibilities for different health programs are rationalized, particularly at the district level, more attention can be paid to overall health and to pooling resources among different programs. This does not mean tbat a single worker, facility, or agency should provide all the services. PEOPLE ORIENTAON. It is not enough to set up a PS structure; health staff need to be oriented toward providing services to people. Most countries find it difficult to reorient health professionals and govemment bureaucracy to PS. Thus a major task of PS managers is to increase the people orientation of the service providers, and thereby help improve provider-user interactions. For instance, the interational Center for Diarrhoeal Disease Research in Bangladesh worked with the government to enhance the people orientation of service providers by providing assistance in training, logistical support, organizational development, and record keeping. Otber problems that also needed attention in this case were the shortage of supplies and essental facilities, laclk of well-defined priorities among the tasks of workers, an unwillingness to make the extra effort required to ensure systematic service delivery, inadequate supervision, and unrealistic number of clients to be served (Koblinsky and others 1989). Several lessons have been learned about orienting health staff toward providing services to the poor:' * Realistic work progrmmg increases outreach. * Well-designed and field-oriented in-servioe trainnig helps to improve the performance of field staff. * Monitoring systems need to focus on simple indicators of coverage and quality. * Management information systems cannot be improved in isolation; other relevant management activities must be included in any improvement efforts. * Although the lack of supportive, problem-solving supervision and low worker motivation have been often cited as reasons for the low coverage of PS delivery systems,any corrective actions in this regard are not well docunented. 9. Heaver (19893 draws seveal lessons for impvin the ourach of PS from the Bank prjes in India. Frst succssfi projects limted the masks of peipheral workers by ensuring a reasonable number of clients to be sewed by a worker, directed the focus of workers on a limited nuaiber of service and defined dteir work rouines targetg on priority clients. Second, they selected loci workers, provided them with regular in-wservice trining, and ensurd reasonable supervisor/worker ratos. Third, tiey involved diets in program implnenmion. Fouth, dty paid considerable atnion to details of program stucor and day-to-day dlivery services. 44 J.K.Sata * Inadequate medical supplies discourage people from usng health services. Many successful programs have paid considerable attention to Ehese problems. For instance, in the Tamil Nadu Integrated Nutrition Project (Berg 1987; Subbarao 1989), administaive superoion of field community nutrition workers is separated from technical support, which includes pre-service as well as in-service training on the job and is provided by instuctresses. Training is provided at the local level so that the content can be tailored to local needs. This approach promotes close coordination between on-the-job training by supervisors and rn-service training and can be targeted to specific worker needs based upon an appraisal of his or her performance. Careful attention is also paid to seleting workers at the local level. In each community nutrition center, selected community data and performance indicators are shown on a biackboard so that the community members, workers, and supervisors can easily assess the acivities of the center and the nutrition status of children under three years of age. The commnity nutrition worker forms a women's group to assist in several priority activities. More inpotan, the group plays a role in communicatng both the content and philosophy of the program to the communty and thus helps increase the participation of potnial beneficianes. EMPHASIS ON GENERAMTNG DEMAND. National programs generally assume either that demand already exist or can be easily created once service delivery is in place. The problem with tis view is that the poor may not perceive these services as their most-needed ones and thereore may not be willing to use them. The quality of services may also be so poor that people do not wish to use them. While expanding service coverage, some projects put considerable emphasis on increasing the quantity of services to the detimen of their quality. This strategy eventually constrains demand. How to insfill a concemr for quality among widely dispes PS delivery staff remains a problem that has not received much attention in the literature. Ministries may not be well enough equipped to provide the necessary d lizaion, interorgarizational coordination, and range of twclical sklfls for effective comication. For insmce, although many World Bank- asisted projects have included information, education, and comnication (IEC) components for several years, project staff in most cases assme that the government has been unable to implement effective IEC activities. In a variety of settings, social markeing' through commercial or other channels (such as traditonal healers), has met some of the demand for ontraception, complementing family planning program efforts (Sheon, Schellstede, and Derr 1987). In Bangladesh, India, and Nepal, social marketing has been contributing about half of condom and a fourth of pill usage. The range of social marketing could be expanded, however, as recendy shown in Egypt, where it was used -to promote oral rehydration salts (NCDDP 1986). The Indonesian Nutrition Project supported by the World Bank and the suceeding project support by USAID demonstrate that social marketig approaches can help change feedng practices and improve nutrition staus (Heaver 1988). Several examples from qualtative research show that local perceptions of diseases and their causes may differ signifianly from those of health professionals. Generally speaking, increased knowledge about the people can lead to improvements in PS delivery (Salmen 1987). However, studies of health practices and care-seeking behavior are often not translated ito 10. Bmadly speaing, socia makedig refers to dte usc of modam narking, wih its compens of market and cmmer reseah, advehiig, and disLin and is amy of sales prnmoton devices for social purposes. TIhe premise is ta ft cowmnier is the key and comsmer pecpion is die fmdmema wisdom (Mianoff 1985). The Accessibity of Popuation, Heafh and Nuarnion Pflgrams to the Poor 45 Box 3.3 Community Lins It has been very difficult to build strong links between the communities and PS delivery systems in South Asia: a. Volwureer-worker link. VHWs are the most important means for linking national programs with communities. But, as noted earlier, they do not seem to have succeeded in South Asia. b. Groups. Most programs envisage the formation of women's grups, and a considerable amount of experience is being gained in activating such groups. However, large-scale national success has not been reported. c. illage comamnees. Examples fiom goverment progams in India (Bhatt 1988) and Bangladesh (ESCAP 1989) show that the committee-type stuctr cannot suddenly be superimposed on communities and be expected to function effectivdy. Thus, the available evidence suggests that formation of women's groups with link community voluntes is the most suitable way in which kS delivery systems in South Asia can strengthen their community links. project design, and the result is less-than-optimal performance. MAITAIN G COMMUNnY IINKS. Community ivolvement can be sought by strenthening the lnks with the commity, although the extent and mode of this endeavor may vary, depending on the socioeconomic conte (see Box 3.3). The stronger the links with the community, the larger the coverage. However, it may not be possible to build broad, self-sustaining community participation through health service activities alone (Rifkin 1986). Experience shows a correlation between high participation and an integrated aprah that is attmed to communty piorities and not only related to health care. Since apprpriate community actions depend upon the social, cultural, political, adminisative, and economic sitation in the country, what can be done in a small NGO program may be impossible to repeat on a large scale. Role of NGOs NGOs have generally been more successful in delivering services to the poor than other groUps and their involvement has been increasing. In Bangladesh, by the end of 1991, 23,000 emale field workers were employed in the govrnmen program while about 6,000 female field workers were employed in rual NG0 family planing programs. They provide almost 40 percent of all contraceptive semvi and a majority of MC.H services in urban areas (ICDDRB 1990). In Pakistan, NGOs have been successful in delivering primary care, albeit to relaively small 46 J.K.Sauia numbers. NGOs are more active than goverment agencies in developing new initiatives in family planning -ervice delivery (Fazil 1989; World Bank 1989c). NGOs are active in many areas of India. Their share in total care is significant, and they have experimented with innovative forms of service delivery as well as assisted the government in implementing programs (Ford Foundation 1987). The health, family planning, and nutrition services provided by NGOs have often been the leading founs of "modern" (as opposed to "traditional") care in rural areas of Nepal. Recently, NGOs have started services in urban areas as well and have become agents of innovation in this sector (Thapa 1989). With a history of conmnunity involvement and well- developed government health services, the role played by NGOs has been small in the health sector in Sri Lanka, although it has been significant in family planning. NGOs: Strenghs and Weaknesses The signifiant features of NGO experiments have been their responsiveness to community needs, paramedic/village-based service delivery, and participatory processes in the planning and implementation of programs. Having long worked with the poor, NGOs are better able to help them. Because of the participatory nature of NGOs, they not only deliver needed services, but also help the poor to help themselves. They are more flexible and experimental than goverument agencies and therefore can more easily adapt their activities to local needs. At the same time, most NGOs are small, isolated, and dependent on one or two charismatic leaders. The initiative often comes from outside and not from within the community. Even when an initiative emerges within the community, it tends to come from the more educated and elite members of the community. As a result, the poor may come to depend on NGOs (Bhatt and others 1987). NGOs use several methods, in addition to expanding their own coverage and range of services, to upscale service delivery (IEMA 1985). NGOs are also getting involved with government systems in a technical and administrative capacity. They assist the goverment by providing various services-training, planning, monitoring and evaluation, and a low-cost supply of medicines. They also play a linking role either by encouraging the people to make denands or by assisting them in making use of the government services. Finally, at the national level, many of them attempt to influence government policy by playing an advocacy role. The govemment's main concem in replicating NGO activities ias been cost. The evidence on this issue is mixed and data on the cost of community health projects are scarce. A study of such projects in Maharashtra, India (Gupta and Duggal 1986), showed that their costs were not substantially higher than the amount the government may have been spending on rural health care. Another study (Berman and Dave 1990) estimared that the cost per contact for curative care and immunization in selected NGO programs was within the range of costs reported by government services, although NGOs were spending 5 omewhat more per capita on primary health care than the government. In Bangladesh, the Matlab program was estimated to have at least four times the worker population density of the government program (Badrud Duz 1989). And in Nepal, several NGOs were reported to be spending more resources than the government program (Thapa 1989). Government-NGO Pa,mership Governments generally have greater capacity for organizing service delivery on a large scale and have the resources to sustain them. In corntrast, NGOs are more oriented toward people, are able to create demand, and enable the poor to integrate various available govemment services. The Accessibility of Population, HealLh and Nutnrion Programs to the Poor 47 Box 3.4 Modes of Governnent-NGO Partnership NGOs INNOVATE; GOVERNMEN REPUCATE. Often NGOs are able to test and refine innovative approaches. The government then uses these models to upscale the programs. However, performance usually deteriorates during the process of upscaling. NGOs can remedy the situation through training and other support, which can reduce the gap in perfonrance between small and large programs. NETwoRKs, FEDERATIONS, and COORDINAnON CoMMurTEEs. Federations and intermediary organizations are often created to coordinate the activities of a large number of small NGOs and channel government funding to them. However, many smaller NGOs jealously guard their autonomy. Such coordinating bodies have not become strong, well organized and effective, except in a few instances. DIVSON OF LABOR. The government offers services, and NGOs take up the responsibility of creating demand and enabling the poor to make use of those services. This model has potential for conflict. To maintain a reinforcing relationship with the government, a careful balance needs to be observed that allows the state machinery to be pushed, but not too hard. EAuMARix AREA RESPONsITEs. When both the government and NGO services are being provided in an area, it is often desirable to allocate areas for specific operations, so as to avoid duplication and confusion. NGOs TAKE OVER GovERNMENT UNIS. Several NGOs have been handed over government programs in specific areas. But generally more success is achieved when both technical and administative flexibility rests with the NGOs. To achieve a sustained partnership, the government and NGOs need to not only increase mutual trust but also strengthen the institutional capacity for learning from each other. Is it possible to zombine their strengths for reaching the poor with government capacity for operating on a larger scale? What modes of collaboration have been tried and with what results (see Box 3.4)? Operating Large-scale Peripheral Service Delivery Systems In the absence of data on coverage, cost, and quality, it is difficult to comment on the success of the country programs In general, PS expansion has had a greater impact on health status where other social and economic conditions have been favorable or were improving. 48 .K.Saria However, PS delivery systems have faced many obstacles in their efforts to increase their coverage. Some resarchers do not thiink implemenation is the reason that the PS delivery system has failed to reach the poor. They claim that PS delivery will continue to mirror social inequality ind, unless actions are taken on a broad front to reduce such inequities in the society itself, significant improvement in its outeach cannot be expected (Mburu 1989). Nevertheless, governments have expanded their PS delivery structures to reduce such inequities. Heaver (1988) reviewed Bank projects concerned with population, health, and nutrition sectors to determine how much emphasis they put on reaching people at the periphery. He found that these projects had incorporated many elements required to reach the periphery, but that they were deficient in two areas-the design was not flexible or based on client needs, and targeting was needed to increase provider incentive to reach the periphery. Since people at the penphery are distant, differentiated, and voiceless, one needs to make a special effort to adapt services to their needs. This not only requires flexibility but also the capacity to receive feodback and to learn the lessons of past experience. . When lessons from small and large projects and country programs abound, why have these not been exptiously translated into practice? Why is the pace of improvement in PS delivery so slow? Four factors seem to have kept PS delivery systems from adequately responding to the health neds of the poor: a weak political commitment, inadequate attention to the process Of ilementaton, underfunding, and overstretched management capacity in the ministries of health. Poldca Commt PS delivery systems grew out of a national political commitment to PS delivery and to equal access for the poor. Consequendly, budgets allocated to them increased in real terms. A part of this increase came from their increased share in overall healin resources, as in India. However, budgetay constri and the lack of political wil have made it difficult to improve PS delivery systems in South Asia. Operairg large-scale PS delivery system requires a continuing political commitment. Process of Implemenm ion The experience of small-scale projects has revealed that the success of any program depends on effective implementation. It is not only "what is done" but "how it is done" that determines success. PS delivery systems should be demand-led and softwae-driven, and PS delivery structures should be flexble. In small-scale projects, considerable attention has been given to sdmulating demand by sensitizing commnities, establishing village health committees, and sdlecting an appropriate mix of services before expanding the service delivery structure. While establishing e service delivery structure, many planner have emphasized training, supervision, and logistical support. The resulting project designs have been flexible, depending on the task and organizational context, and-in keeping with the principle of learing while doing-have been modified where necessary. In contrast, national programs have been supply-led, concerned with plant and equipment, and have followed a uniform pattern in the structure of service delivery. Governments lave conentrated on building necessary clinics and hospitals, training VHWs and other manpower, and provdg supplies, equipme, and vehicles to make the system work. The govement - - . -~~~~~~~~M The Accessibility of Populsion, Heal and Nustition Progrums to the Poor 49 bureaucacies have tended to be rigid, have focused on curative services, and are facility oriented. In such a system, providers are likely to have little incentive to reach the peiphery. Undernding of PS Delivery Systems Reliable estimates of expenditure on PS delivery systems across countries are difficult to make because comparable data are not available. Roughly, 20 to 30 percent of the total health expenditure (including the private sector) or 0.3 to 0.5 percent of GNP is spent on PS delivery systems. The per capita expenditure on PS varies from US$0.45 in Bangladesb to US$1.60 in Sri Lanka (Griffin 1990). One assumption behind the rapid expansion of PS delivery systems is that such systems do not require large resources. In fact, the fiancial requirements of PS delivery have been underestimated. In the 1980s, governments targeted spending toward rural areas but because of low overall funding and in the absence of radical redistribution of resources from other paru of the health system, PS delivery systems were underfunded. The result was poor trainimg, weak supervision, low mobility, lack of supplies, and inadequate communication and community support activities. Consequently, the quality of PS services was diluted and often, even the poor preferred to pay for private services. There is growing evidence that aMdonal inandal resources wil be requiredfor effective PS delivery, bxt the extent of underfnding ai not clear. For instance, it has been estimated that in real terms, Bangladesh would require an ince of 74 percent in its health budget to implemnent current plans for expanding and improving PS operations, but this would only partly elininate some of the known problems (Griffin 1990). Therefore, an effort must be made to increase either the overall resources allocated for health or the share of PS delivery systems in health budgets, or both. One possible solution would be to instite user charges in the hospital sector, which would take away 50 to 70 percent of the total heaitn resources and redirect available resources to PS delivery. However, examples of user charges and an effective means test to deterine the ability to pay that are rdministratively efficient and widely acceptable are difficult to find. Also their distribution effects across different income groups and their impact on oher providers need to be considered. For instance, a study of treatment behavior for cildren's illneses in urban Pakistan showed that although the poor are more price-responsive than the general population, the reduced use of government clinics forlowing a price rise leads to greater use of private care providers rather than to an increase in self-care or the forgoing of tare (Alderman and Gertler 1989). It has also been argued that PS systems need not be funded exclusively by the public sector. Several financing mechanisms such as health iurance in poor areas and community ficing are being tested. But these have not yet proven sustinable. Management Capacity Ministries of health in many countries are or ionally overstretched. In the 1960s, they concentrated on developing hospital-based care. In the 1970s, when auention shifted to basic health service, new organizational capabilities were required. The need to forge closer links with village conmmnities further stained these capabilities in the 1980s. 50 J. K.Sa:.;a Because of the charactenstics of their dients anzd low level of technology, providing PS to the poor imzposes certain burdens on the ministries of health. The tasks in managing PS delivery differ from those in managing health care institutions in terms of variety, dispersal of personnel, professional orientation, the need to create demand, and emphasis on software. These differences need to be recognized by using appropriate structures and processes for managing PS delivery systems. The issues that need to be addressed are decentralization, strategic planning capacity, human resources management, and technical support to the periphery to ensure quality of care and communication activities. Strategies Specificaly Desiged to Reach the Poor Despite the attention currently being given to the issues just outlined, it is still uncertain that government PS delivery systems would have the necessary adaptability to provide the poor with adequate coverage. Targeted Strategies As coverage levels increase, it becomes necessary to target services tO the underserved population. Experience with such strategies is not well documented. The following discussion is thus exploratory and suggests what direction furfther investigations might take. Targeting strategies would depend on the areal concentration of the poor and the characteristics of the PS delivery system: Vhen concentration and PS delivery system then appropriate strtegy of poor is dcaracteristics are to reach poor is High PS delivery is inadequate and faces Geographic targeting may problems Medium PS delivery system exists but poor Promote at-risk approach are not geting services Low Alternative channels for services are Widen choices for poor available Ultra-poor Low demand for PS Multisectoral approaches In the context of nutrition programs, Kennedy and Alderman (1989) observe that geographic targeting worked well in the Philippines because there were many families with food deficits and malnourished preschoolers in the villages chosen for treatment. In contrast, the geographic targeting to low-income areas in Mexico City was less effective, mainly growth retardation was not as prevalent there. In such cases effectiveness can be enhanced with the aid of screening measures, although it should also be pointed out that targeting PS delivery to the The Accessibility of Populaion, Heah and Nutrition Programs to the Poor 51 poor in areas where their proportion is very low may not be feasible, in which case their range of choices for service channels can be expanded. PS may not be an inportant felt need of the ultra-poor. In these areas, many NGOs have used multisectoral approaches. GEOGRAPHIC TARGEr1NG. Geographic targeting to socioeconominclly backward areas not only implies a selective expansion of the PS structure but also its adaptation to local needs. Although not necessarily economical, geographic targeting would clearly improve equity by directing resources to those most in need and would help create an appropriate differential response to poor areas. However, such targeting is often limited to administrative units of a reasonable size in which all the beneficiaries are not necessarily poor. The services may still be disproportionately used by better-off segmerts of the society. Yet, if smaller geographic areas were used for targeting, the administrative overhead would increase. PRoMOG AT-RmSK APPROACHES. The at-risk principle is well accepted in the health field. That is to say, depending on the severity of a problem, it may be referred to higher levels of the health system. However, at-risk criteria are seldom used to prioritize primary services Directing services to those most in need by first screening them is likely to shift available services in favor of the poor and help promote equity. Critics argue that the supposed efficiency gains of these approaches over near-universal coverage may be neutralized in part by the costs of screening, particularly if the screenig critena are complicated and involve costly procedures. The poverty level of the household could be used as a factor for screening, here, too, difficulties arise. Means test are not always reliable and there is bound to be some leakage of effort. More important, although poverty predisposes people to higher health risks, others may also be affected." In general, at-risk approaches work well when the proportion at risk is neither too low nor too high. Most monitoring systems do not encourage screening by service providers because such efforts are considered wasteful in comparison with the actual services. Thus at-risk approaches are more likely to succeed if they are s-upported by continuming training and supervision, if screening is used as one of the performance criteria, and if those found to be at risk obtain quality treatment that produces the desired effect. WIDENING THE CHoIcE FOR THE POOR. The limited purchasing power of the poor often restricts, even nullifies, their ability to choose the organization from which to obtain services. Increasing their ability to choose is likely to introduce competition among different service providers and consequently improve service quality. For instance, rudeness on the part of providers has deterred many poor people from seeking their services. Subsidized third-party payment mechanisms or the use of vouchers can widen the choices for the poor and, at the same time, ensure that the money goes to the organizations that can attract them and provide the services they need. As Salmen (1989) says, "When poor are given choice among organizations that serve them, they are empowered. Rather than having to endure indignities and inefficiencies, they can 'exit' the offending organizations and 'enter' another one". Such mehanisms have been widely used in developed countries, perhaps, because the concentration of poor is low and a plurality of service institutions already exists. Although this strategy has limited applicability in South Asia, it could be more widely used than it is at presen. 11. The Tamil Nadu Intgxatcd Nutriton Project (Berg 1987) in L-xdia, uses a growth fatering criterion to supplemert children with beraedc nutritioL The projec has been able to cut seere malnurkion levels in half and is thought to have twice as much impact at kss cost compared with a similar program ta atempts to provide near- univeal coveage. 52 - J.K.Sada MuLnsEcToRAL APPROACHEs. The ultra-poor (bottom 10 to 20 percent of the population) have special needs. Income and employment are their most serious concerns and health sevces have low priority. The health sector is not equipped to orchestrate the nmltisectoral approaches that would be required to reach them. Many NGO programs have attempted to reach the ultra-poor by responding to their special needs. These programs began their activities by attending to irrigation or income generation because they were the primary needs of the people in these marginal areas. Some programs that focused on health services found that even if the level of ignorance is reduced and health services are made available, there can only be limited improvements in health unless poverty is alleviated. In the Comprehensive Health and Development Project at Pachod, India (Ford Foundation 1987), poverty explained the difficulties encountered in reducing the malnutrition of children below a certain level. Consequetly, the program broadened its activities to include afforestation and biogas proams. The Child in Need Institute in Calcutta, India, recogized the link between income, literacy, and the capacity of a woman to look after her child. It therefore added literacy and income-generation activities to those designed for women's groups. Perhaps NGOs have more success with the multisectoral approach because of their flexibility and commitmaet, whereas, the goverment rns into problems of coordination. Governments have begun to recognize the distinctive strengt of NGOs and are developmg mechanism to encourage such NGO activities. Role of the Privat Sector A large proportion of health care is sought from the pnvate commercial sector, which comprises private practitioners of allopathic and other systems of medicine. Griffin (1989) argues that economic efficiency gains are likely to occur if developing country governments increase their coopeation with the private health care sector and nurure thir-party payment mechaisms t finance the minx of semces their citize currendy use. The government could then improve the targeting of subsidies and direct services to those who are disadvantaged or at risk The burden on the PS delivery system would be reduced if the private sector could provide services to those who can afford them and if public services were direaed toward those who cannot. Vanous institutions and health care delivery systems often need to be restructureJ to target the public PS delivery system to the poor.?2 However, this strategy, which appears an effective way of targedng health expenditures to the poor, needs to be accompanied by a conmmitment to provide the poor with good quality services and to protect expendiures for them. But before the public and the private sector can perform complementay roles, the policy on the role of the private sector must be clearly defied. 12. To target public hea services to dt poor, te govmuen of Chile first expanded and improved die PS ddevy systa, emphasing them- vulnerable groups-mothers and yog children living in nwal sneas, where a map of extane povrty showed the lowes indicas of health stus to be. The govnmen then decentralized the nra health services into twenty-sx regional amoms seri ransrng die rewponsibility for the PS delivery system to the municipalities. The role of th ministry as the policymaking, normative, supevisoy, and valuatng agent was also strengthened Major instituonal and financial rfm for the health sector were acomplished by creatig a financial stituon for die ealth sector and chargig k with the collecdon of al revenues hon the natoal budget as well as payroll dedctios, and paying the provider of services in accordane with ffie services rered and aowimg workers te option of choosig between publc or print health insuance fumds (Casaneda 1969). The Accasibility of Popudon, Health anid Nut on Prognums to the Poor 53 Condusom The health status of the populaton of South Asia, and of the poor in particular, can only be improved if PS coverage is improved. More attention must be placed on creating demand, increasing tie people oriention of service providers, and encouaging flexibility at the local level. To this end, it will b- important to address the issues surrounig the proce of implementation, underfnding of PS delivery systems, management capacity at the higher level, partership with NGOs, and the role of the private sector. Such efforts must be backed by a political comitment to enhaing PS coverage of the poor. Where coverage levels are moderate to high, targeted strategies would be necessary to reach the poor, and the specifics of targeting would depend on the context. Four actions are necessary to buid susainable PS delivery systems and implement targeted strategies to meet the health needs of the poor. Fust, the infrmation base about the poor needs to be stegthened. Second, policies need to focus on the poor. Third, programs should increase their manaemnt capacity for designing and implenting PS delivy SYstes. Fourth, we need to learn more about specific constraints and how to overcome them through systematic research efforts. Strengning the Base of Inbrnmation about the Poor A special effort needs to be made to understand the poor. Since they have no voice either through the maket or through links with service provis, their demanxs are not properly understood. The cental planners are often far away and have preconceived notions about the poor that may not coincide with reality. Even when contacted, the poor are suspicious, inarticulate. and rarely organized to speak as a group. There is nreasing evidence that survey data do not prvde an adequae base of infration for undstaing the poor. Yet the survey is mhe most common medtod of coliecting data. One of its problems is that it involves a relatively large nmber of respondents and rather thm information frm individuals. Program managers trying to reach the poor are becoming aware of the value of cultural information about them (Salmen 1987). Qualtative research consting of participant observation, openeded in-depth iterviews, and focus groups are tiUS needed to gain insight to health behavior, its de , and its cultural and ecological constraint. A combination of quantitative and quaitatve methodologies can be used to develop a data base on the poor. But a one-tim altempt at unestnding the poor will not suffice. As project and program services interact with beneficiaries, both the program objecives and opaions may have to chang. Therefore, it is iportant to aim for flexibility of design with a built-in autonomy for local managers based upon a learning process approach (Korten 1984). This approach reur constant feedback during project inplc ion to allow the project to become adapted to its environment. Ibis feedack shoud come from service sttistics, longitudinal surveys, and process docmetaton. The mon indicators should assess not oDny the input of staff and facilities, coverage of beneficiaies from different socioeconomic groups, quality of sen ces, and impact on bealth status of the intended beneficiary groups, but also the involvement !54 J.X.Saia of the poor and the people orientation of service providers. Finay, project ipacts should be measured through surveys and the reactions of beneficiaries. 13 Policy Response: Focus on the Poor A policy focus on the poof implies allocating greater resources to underserved population groups, involving the poor in decisionmaking, and providing a suitable role for NMOs and the private sector: * Resources should be allocated so as to correct the bias in locating facilities and staff both in favor of the peripheral part of the services and the poor. * the involvement of the poor should be encouraged through the formation of beneficiary groups and link volunteers, perhaps the most suitable mode of involvement, given the sociopolitical context of South Asia. Governmments need to accept and respect the role of NGOs as a movement or a pressure group a the policymaking process. * the private sector could provide services to those who can afford them, and public serces could be directed to the poor. However, the role of the private sector needs to be clearly defined. Program Respoe: Strengtn Managenwm Cqaaciy How to strengthen management's capacity to design and implement sustainable PS delivery systems has been a major concern of researchers and administrators alike. Unfortmately, there are no easy answers. A recent World Bank-assisted project in Sri Lanka emphasized the following steps in developing management capacity: (a) dec tion achieved by strngthening provincial and local capabilities; (b) improved manegement of human resources through training, studies of health manpower, and s of the human resource management fimction; (c) strategic planing lthgh the establishment of a management development and plamiing unit in the ministry, reponal and prvica health planning, health sector policy development studies, and informon systemrs; (d) techical support for a higher quality of care through improvements in dhe medical sWpplies department and in the staard procedures for diagnosis and treatment, and support for communication tbrough a strengthened IEC cell. In addition, policy issues connected with the struct of service delivery, role of NGOs and the private sector, and u fin need to be addressed. 13. Sevea medKds of rural apid appasl and pc y ea have been developed during thelast decade, by Rihad Chambers an his assos (Chambe 1986. The Accessty of Popladn, Healh and Nanidon Programs to the Poor 55 A Research Agenda Although a great deal has been wrtte about the functioning of PS delivery systems, the factors that have constrained such systems from reaching the poor are not well understood. Research along these lines should concetate on comparative case studies of PS delivery systems to identify the constaints in reaching the poor; and on possi1ble actions that might be used to overcome these constraints at the district level. Attention also needs to be paid to actions required at the higher level, to enable and motivate the distnct level to reach the poor. Since these are not amenable to research at the local level, knowledge can only be expanded through cross-country investigations in South Asia of the following questions: * What is the extent of underfunding of PS delivery systems? * How can the management capacity of the minstries be improved? * What role can the private sector and NGOs play in different country contexts? * What can we leam fron the countries that have succeeded in reaching the poorest two decile of the population? The agenda for research, suggested bere, can help refine responses to the problem of providing PS to the poor. 56 . JSadia Reerenes Alderman, Harold, and Paul Gertler. 1989. "The Substitutability of Public and Private Health Care for the Treatment of Children in Pakistan." World Bank Living Sadards Measuem Study, Working Paper 57. Asian Development Bank (ADB). 1986. Use of Rural Health Serices. Poedings of the Regional Seminar on the Use of Rural Health Services, 20-25 January, Manila, Philippis. Badrud Duza, M. 1989. "Overview of Findings." In South Asia Study of Populaion Programmes: Bangladesh. UNFPA. Bamberger, M., and S. Cbeema. 1990. Case Studies of Project Sustanbility: mplications for Policy and Opermions from Asian Experience. Washing, D.C.: World Bank, Economic Development Insicue. Berg, A. 1987. MaLaition: Wh Can Be Done? Baltimore, Md-: John Hopkins University Press. Benman, Peter E., Davidson R. Gwadkn, and Susan E. Burger. 1987. "Community-Based Health Workers: Head Start or False Start Towards Health for All." Socia Scence and Medicine 25(5):443-59. Berman, P., and P. Dave. 1990. "Experiences in Paying for Health Care in India's Voluntary Sector.' Paper presented at the Conference on Strategic Action in Health: The NGO Expeence, sponsored by the Ford Foumdation, New Delhi, and held at Goa, April. Bhatt, A. 1988. "Comnmnity Participation in Govenment Managed Health Programs in Ladit: An Overview.' Populaion Manager 2(1):19-23. Bhatt, A. and othes. 1987. Builngfrom Below: Local IniavesforDecntraizedDevelopnw in Asia and Pacific. Vols. 1 and 2. Asian and Pacific Development Center and Konrad Adenauer stiftung. Bhore Committee. 1946. Reponr of the Health Swwy and Development Conittee. New Delhi: Govenmnent of India. Catneda, T. 1989. "Innovative Social Policies for Reducing Poverty: Chile in the 1980s." Report prepared for the World Bank Chambers R. 1986. "Shortcut Methods in Social Informaon Gathering for Rural Development Projects." In Rapid Rural Appraisal. Bangkok, Thailand: fhon Kaen University. ESCAP. 1989. Organizatonal ssues in Comunzity Particaton in Naional Faly Planning Programs: A Compative Analysis of Five Countries in the ESCAP Region. Bangkok. The Accesibity of Population, Healt awd Nutition Progrms to die Poor 57 Fazil, S. 1989. "Lessons from Successf Experiences. " In South Asia Study of Populaon Poicy and Progrmmes: Pastn. UNFPA. Ford Foundation. 1987. 'Anubav Series Experiences in Community Health Docmenting Child in Need Institute (CM), Comprehensive Health and Developnent Proiact at Pachod, Strrehitkarini, Rural Unit for Health and Social Affairs (RUHSA), Action for Social Welfare and Awakening (AWARE) and Six Other NGO projects.' New Delhi, India. Gish, 0. 1990. "Some Links between Successful Implementation of Primary Health Care Intervention and the Overall Utiliation of Health Services." Social Scence and Medicine 30(4):401-5. Griffin, Charles C. 1989. "Stengthening Health Services in Developing Coumnies through the Private Sector." Intermational Finmce Corporation Discussion Paper 4. WashiWon D.C. . 1990. Heafth Finncing in Asia Report to the World Bank, May. Gupta, Jesani, and R. Duggal. 1986. NGOs in Rral Health Care. Bombay: Foundation for Reseach in Community Health. Heaver, Richard. 1988. Reaching People at the Periphery-Can the World Bank's Population, Healk, and Ntion Operations Do Beaer ? World Bank Working Paper 81. Washigton, D.C. . 1989. Improving Family Plandng, Healh and Nuttion Outreach in India:.Experiece from Some World Bank Assisted Progrwns. World Bank Report 7689-IN. Wasngton, D.C. ICDDRB. 1990. "Coordination betw NGOs and the MOHP in the Delivery of Family Plannin Rural Services in Rural Bangladesh." MCH-FP Extension Project Briefing Paper 12. India. 1988. Fawiy Wefre Year Book 1987-P8. New Delhi: YMistry of Health and Family Welfare, Deparmnt of Family Welfare. Indian Institute of Maageme. 1985. Regional Workshop on Scaling Up for Child Suiwat Activitie. Vols. 1 and 2. Sponsored by the Ford Foundtion, New Delhi, and held at Ahmedabad, August 20-23. Justice, J. 1986. Polides, Plans and People. Berkeley: University of California Press. Kennedy, Eileen T., and Harold H. Alderman. 1989. "Compaative Analysis of the Nutritional Effectiveness of Food Subsidies and Other Food-related Intentions: Conclusions." Foo4 wad Nutrition Bulin 11(1):74-76. 58 J.K.Saria Kban, M. E., and others. 1987. "Access to Health and Family Plaunn Services in Rual Uttar Pradesh-A Micro Level Evaluation." World Bank Operations Research Group, Washington, D.C. Khan, M. R., ed. 1988. Evaluion of Primay Health Care and Fanmiy Planning Facliies and Their Limitations Specialy Rural Areas of Bangladesh. Reseach Monograph 7. Dhaka: Bangladesh Institute of Development Studies. Koblinsky, Maijorie A., Susan J. Griffey Brechin. Samuel D. Clark, and Yousuf M. Hasan. 1989. "Work Schedules of Field Workers in Rural Bangladesh." Studies in Famly Plang 20(4):225-34. Korten, David C. 1984."Rural Development Programming: The Learning Process Approach." In D. C. Korten and R. Klnsus, eds., People Centered Development. West Hartfor. Conn. Kumarian Press. Manoff, R. K. 1985. Social Marketing: A New Imperative for Public Health. New York: Praeger. Mburu, F. M. 1989. wWhither Communty-Based Health Care ?" Soial Science and Medicine. 28(10): 1073-79. Murthy, N., I. Hirway, P. R. Panchmuk1i, J. K. Satia. 1990. How WeU Do India's Social Service Programs Serve the Poor? World Bank Working Paper 491. Washington, D.C. NCDDP. 1986. "NCDDP-This Is the Egyptian Oral Rehydration Project." Cairo. Paul, Bnima Kanti. 1991. "Healh Service Resources as Detminants of Ifant Death in Rural Bangiadesh: An Empirical Study." Social Seiezce and Mediine 32(1):43-49. Rifl;in, Susan B. 1986. "Lessons from Community Participation in Healffi Programmes." Health Poicy and Planing 1(3):240-49. Sahnen, L. 1987. Listen to the People: Participant-ObserverEaluaion ofDeveopment Projects. Oxford University Press. _ . 1989. InstitonalDimnions of Poverty ReJdion. WashoM, D.C.: World Bank. Sheon, Amy, William Scheistede, William, and. Bonnie Derr. 1987. "Contraceptive Soca Marketing." InRobert J. Lapham andGeorge B. Simmons, eds., OrganizingforEffective Family PlainWng Programs. Was_hgto D.C.: National Academy Press. Sri Lanka. 1987. Demograpic and Healt Survey. Colombo: Departmcnt of Census and Statistics. The Accessibility of Populaion. Health aNutrition Programs to tle Poor 59 Srinivasan, K., Tara Kanitkar, D. Misma, and M. Hola. 1986. Demographic Situaton and Utilization of Health and Famoiy We(fare Serices in Bihar-A Baseline Suvey Report. Bombay: Intemational Institute of Population Studies. Subbarao, K. 1989. rImproving Nutrition in India." World Bank Disussion Paper 49. Washington, D.C. Talwar, P. P., and P. S. Bhatia. 1985. Demographic Sictation and Utilizaion of Heah and Family Welfare Servces in Rajasthan-A Baseline Survey Report. New Delhi: National Isitute of Health and Family Welfare. Thapa, Poonam. 1989. "Family Planing Success Stories in Nepal: Their Implications on HMG and NGO Relationships." In South Asia Study of Population Policies and Programmes: Nepal. UNFPA. Thapa, Rita. 1986. "Role of Traditional Medicine in the Demand and Supply of Rural Healft Services." In Proceedings of the Regional Sminar on the Use of Rural Health Serices. Manila, Philippines: Asia Development Bank. Tuladbar, J.M. 1989. "Supply Aspects of Meetg Demand for Family Planning." In South Asia Study of Population Policies and Programs: Nepal. UNPFA. Van Der Geest, S., J. D. Spec, and P. H. Streefland. 1990."Primary Health Care in a Multi-level Perspectie: Towards a Research Agenda." Social Science and Medicne 30(9):1025-34. Walt, Gil, Myrte Perera, and Kris Heggenhougen 1989. "Are Large-Scale Volunter Communky Health Worker Programmes Feasible? The Case of Sri Lanka." Social Science and Medicine 29(5):599-608. Walt, Gill. 1988. "CHWs: Are National Programs in Crisis?" HeahPolicyand Planning 3(1):1- -21. World Health Orpnization (WHO). 1987. Communwty Health Workers: Piasfor HealhforAli. Report of the interregional Conference, Yaounde. World Bank 1989a. Nepal Socil Secor Strategy Review. Vols. 1 and 2. Washington, D.C. . 1989b. Popuaton, Health and Nurition-FY88 Annual Sector Revew. Working Paper 273. Population and Human Resources Department. Washington D.C. World Bank. 1989c. Women in Pakistan: An Economic and Social Strategy. Country Study. Washington, D.C. 4 RURAL DEVELOPMENT AND POVERTY ALLEVIATION: THE ExPERIENCE OF THE ASIAN DEVE-LOPMENT BANK Dieter Bucher and Keith Johnson Summary Ths chapter reviews the approach and role of the Asia Development Bank (ADB) in alleviating poverty.' The ADB's initial strategy of promodng rural development through intersectoral lnkages between agriculure and off-farm sectoral development with supportive land reform measures and new agricultural technologies has been strengthne by the willingness to provide a package of inputs and social amenities. The purpose of these acivties kas been to promote and support -rural development projects wih multisecroral coverage and an integrated approach. Pursuant to dsw policy, the ADB has supported a number of rural devdopment projects in Asia and provided technical asistace. Evaluations of these projects suggest that successful implmentation depends on (a) an appropriate policy envronment in dt aid-receiving country, (b) flexible planning and implementaton procedures based on past experence, (c) beneficiary paracipanon and involvement of local govemments, (d) loa counterpart undig, and (e) commined quaity leadership and project staff. Past experience suggests that the ADB wll be expected to particioate directly and more rigorously in promoting rural development as a means of poverty alkvaon. The basic premise in tis chiapter is that a muesectoral integrated approach to poverty alleviton is desirable, ta a top-down approach to implementation is iappropriate, that these projects ought to be target oriented, and thattheparticpatoryplanningand impleenonapproac topovertyalleation projects should be followed. The chapter also emphasizes social analyst in project planning, fleiblity in project mplemntaon to allow lessons from pat exerence to gude future acton, the recovery of project costs from benfiaries to make projects sustainable and involve local governments, and greater reliance on NGOs to make Ahe projects cost-effecive. Itoduction Because the prevalence of poverty can often be related to a number of overlapping tnictural causes, the figt against poverty should be based on an integrated multisectoral strategy. Rural development projects typically consist of an intated bundle of investment and technical assistance measu presumed to be mutually reinforcing. As such they are considered to be especially well-suited to combatig poverty and providing the basis for sustainable development. Notwithstndig the conceptual advantages of the integraed approach, post facto 1. Jie view expressed in this chapt are tho of the au nd are not nectssariy dt of the Asian Dedopmnet Dank. 61 62 Dieter Bucker aRd Keith Johnson assessments indicate wide divergences in practices and achievements. Too often it seems that integrated rural development projects have proved to be too complex and difficult to implement and sustain. At the same time, projects have frequently failed to uplift the poor, and the overall level of benefits has failed to meet expectations. The main objective of this chapter is to assess the experience gained in the design of rural development projects, with special reference to the evolution of the policies and practices of the Asian Development Barnk (ADB). The core of the chapter is based on a recent working paper (ADB 1988a) and the ADB's in-house evaluation of its project portfolio. The chapter is divided1 into three sections. The first examines the challenges posed by the scale and nature of rural poverty; the second discusses the evolution of the concept of rural development for poverty alleviation and assesses ADB's achievements; and the fftird part provides some suggestions for improving the impact, targeting, and sustainability of rural development measures. The Scale and Nature of Rural Poverty in South Asia The Scale of the Problem The concept of poverty, like that of sustainability, is elusive; yet, it is clear that both are now central components of development strategies. Much has been written on the problem of defining appropriate poverty indicators (for an excellent summary, see Quibria 1990), and this chapter does not discuss the problem further. Broad estimates by the World Bank (1990) suggest that about 520 million South Asians could be classified as poor in 1985, and 300 million of this group could be regarded as "extremely poor." From UNDP (1990) data for 1988, it appears that 570 million out of a total population of 1.150 million in South Asia fell below the poverty line. Some 440 million, or 78 percent of the poor, lived in rual areas. India accounted for about 70 percent of the rural poor (see Table 4.1). The trends in the incidence of povet indicate that South Asia is maldng significant progress toward its alleviation. In India, for example, the national head count index of poverty fell from 54 percent to 43 percent between 1972 and 1983 -(World Bank 1990), and the comparable ratio for rural poverty fell from 50.5 to 40.4 percent (Bandyopadhyay 1988). However, such broad aggregates may conceal significant regional, income, and temporal variations. At the same time, methodological difficulties and data inadequacies give ample room for debate. Reviewing the available evidence, Singh (1990) concludes that sustained declines in the incidence of rural poverty have been strongly associated with increases in agricultural output per capita and have been confined to a relatively small range of geographical regions that have favorable agrarian structures. Elsewhere-particularly in Nepal, eastern India, and Bangladesh- the rural poor have experienced stagnant or declining living standards. The poorest groups, in particular, have bmvone further disadvantaged. Furthermore, the population pressures in many countries are causing a steady rise in the absolute number of poor. Thus it is clear that the task of designing sustainable approaches to pover alleviation is both complex and pressing. The Nature of Rural Poverty In the esh-rt to define and characterize rural poverty more clearly for the purposes of project design, IL is helpful to start with the word "poor." If someone or something is poor, it may also be described as needy/deficient, inferior/defective, or weaktat risk. These synonyms suggest that poverty has three main characteristics relating to (a) the partial exclusion of the poor Rual Development and Poverty Alleviadon 63 from participation in normal levels of production and consumption, (b) the presence of inherent disabilities that restrict opportunities for advancement, and (c) the prevalence of excessive risks relating to the maintenance and improvement of well-being. EXCLUSION OF THE POOR FROM NORMAL LEVELS OF PRODUCTION AND CONSUMFrION. Looking first at the exclusion or mnarginalization of the poor, three contributory factors can be distinguished relating to asset availability, access to sources of income, and the restriction and distortion of expenditure patternis. Asset availabilify. Access to land is a. primary detenninant of household prosperity in rural South Asia. The poor are generally agricultural laborers and fanners with small owned or tenanted holdings of 2.5 acres or less. In India, about 60 percent of the agrictultural laborer households can be classified as poor, and there is a clear relaionship between land assets aTnd levels of poverty. Among the extreme poor in rural Bangladesh, 29 percenit are landiess and a fiuther 55 percent "functionally landless," owning less than 0.5 acres (Hossain and others 1991). At the same time, the landholding poor are less likely to have access to irrigation facilities and generally occupy farmland that is marginal in quality or location. Population pressures are also leading to the increasing subdivision of holdings and the degradation of common land, raising in many areas the proportion of nonviable fann units and the incidence of landlessness. The distribution of nonland assets tends to follow the pattern of landholding, with poo;r farmers fequeatly unable to meet expenditures for variable inputs such as fertilizer or pesticides. Access to sources of income. The poor frequently depend on a mix of marginal activities to sustain themselves. Lacking savings kad access to credit, they tend to take up seasonal and intermittent work to meet expenditure shortfalls. However, the livelihood of both hired laborers and the self-employed is heaviiv influenced by the overall health of the rural economy, which in tun depends in part on govermment policies in- the agricuitral sector. In India, real wage levels Table 4.1 Estimated Incidence of Rural Poverty in South Asia, 1988 Rural population Percenage rural Percentage below poverty population below Rural population Population Rural line poverty line Country (milions) (millions) (millions) (1977-87) Afghanistan 12.6 79.0 n.a. n.a. Bangladesh 95.7 87.0 82.3 86.0 Bhutan 1.4 95.0 n.a. n.a. India 598.6 73.0 305.3 51.0 Nepal 16.4 91.0 10.0 61.0 Pakistan 79.4 69.0 23.0 29.0 Sri Lanka 13.4 79.0 n.a. n.a. South Asia 817.5 74.5 #441.5 #54.0 Note: # Mean based on countries for which data are available. Poverty line is defined as "tbat incomc 'evel below which a nutritionally adequate diet plus essentil nonfood requrem a not available." Soxrce: UNDP, Human Development Report 1990; human developmelt indicators. ~~~~-. __ __ .-_.. ' , ' . . . ! . '~~1 -- - 64 Dieter Bauher and Kefth Johnson for agricultural laborers have been rsing in many regions (Acharya 1989, quoted in Mahendra and others 19I), but this trend does not appear to be shared by Bangladesh (Hossain and others 1991). Rural nonfairm employment is generally becoming more important, reflecting a tendency for the development of a more complex economic structure in nural areas, with infonnal sector services becoming more prevalent. At the same time, many rurl dwellers depend heavily on remittances from urban wage earners and use transfers made by relatives and friends to meet deficits and fmance lumpy expenditures. Restriction and distortion of xpenditure paterns. Poor families spend a large share of their mcomes on food and tend to depend heavily on strchy food staples. Agricultural pricing policies can therefore exert a considerable influence on the welfare of the poor, with food subsidies frequently playing an important role in sustaining poor families (Radhakrisna and Indrakant 1988). Malnutrition is an obvious consequence of low incomes and poverty, with its most severe manifestations in the stmted physical and mental development of the young which in tam perpetuates poverty. In some instances, the poor may even develop cultural or religious taboos against more nutritious but expensive foodstuff which as a result are not consumed even when opporunities permit (Cutting 1977). The frequent distortion of expenditure patterns away from foodsauffi and mecies in male-dominated households is also notewotahy and confirms the special role th female employment and influence can play in attackig poverty. Role of inherent disabilites in prejudicing advancemenr. The poor are frequently demarcated as a special group by their prevmiling socioeconomic characteristics. They generally have unsatisactory access to health and education services and suffer from poor housing and low levels of personal security. In some cases, as with the ijri Musehars, a subcaste in the Glazipur District of WIia (GCate 1984), poverty and class are inextricably linked, and thus economic development cannot proceed without socia change. Because most por families have higher birth rates, lower levels of health, and higher dependency ratios than other social groups, they are less able to aval themselves of new opporamities for generating income. Poor households are also more likely to have a smaller proportion of income eamers, to consist of single-parent families, or to lack access to extended family or other social service support networks. The exploitation of chid labor is common in such households.A culture of poverty is evident in India, where the sbare of scheduled castes and tibes in total rmal poverty is ising (Mahendra Dev and others 1991). Prevaence of excessive risky to well-being. Being at the margin of.the economic system, the poor suffer disproportionately fom uncertainty. Both laborers and small farmers face periods of intensive agricultural activity dictated by the seasons, followed by long periods of underemployment or unemployment. With few opportmities for savig and limited access to credit, poor households face recurrent financial, nutritional, and health risks. The death or incapacittion ofbadwinners may also bring destitution. Drought, exacerbated by environmental degradation, is anotier source of uncerainty. Studies of central India for 1976-83 (World Bank 1990) show that the proportion of newly poor in total rural poor households was three times hier in the worst drought year than in the most favorable year. The influence of severe natural disastc.r is also readily apparet. In Bangladesh, for example, the national headcount ratio for poverty rose frm 41 percent in 1985-86 to 60 perent in 1987-88 as a result of unprecedented flooding (Hossain and others 1991). Swings in agricaltural prices (sometimes reflecting the strucatal adjustment of national policies) can also have a damaging short-terim impact on income sewty. Rul Demeopmtw ad Poverty Alwiation 65 Rural Development as a Tool for Sustainable Poverty Alleviation Evolution of the Concevtual Framework for Rural Development Rural activities have always provided the main source of livelihood in developing Asia, and the economic viability of the rural sector exerts a strong influence on national growth rates. During the 1950s and 1960s rural development was viewed ^ - important but secondary element in national development strategies, which tended to emphasize centrally planned investments in basic industries and physical infrastructure. Although "development" was regarded as generally desirable within a regional planing framework, it was frequently assumed that the "trickle-down" effects from the development of the industrial base would provide the main source of advancement to the rural poor. During the 1970s agricultural development was given greater emphasis, partly in response to the first oil crisis, which led many countries to see food self-sufficiency as an important goal. At the same time, a 'reformist doctrine' promulgated by the World Bank under the presidency of Robert S. McNamara had begun targeng investments designed to assist the poor. However, the new integrated rnual area development and basic needs projects tended to retain a "center- down" bias and did little to motivate and enmpower the beneficiary groups. Rural development projects in the 1970s ran into a number of problems, however. Projects were generaI!y too ambitious and complex and paid insufficient attention to wider agricultural pricing and policy issues (World Bank 1990). The level of sustainability of rurl development projects was also low, as a result of their essentially enclave character and their marked depeodence on expatriate expertise (Lewis and others 1988). Furthermore, the benefits of projects of this type-particularly where they were linked to heavy investments in irrigation and the introduction of "Green Revolution" technology were often siphoned off by landlords and the rural elite and did not reach the poor section of the rural society (Waller 1987). In the 1980s, a number of new strands were woven into rural development project designs. First, much more emphasis was placed on geing agricultural policies and prices right. Both the World Bank and the ADB provided finance to underpin the process of sectoral policy reform in Asia. Second, it had become evident that the poor are 'bankable." Carefully targeted projects that provide appropriate technologies and credit within a sympathetic local institutional structure can have a major impact on rural poverty. Over the past decade such examples as the Amul dairy coopeative system in India, the Grameen Bank System in Bangladesh, the Small Farmer Development Program in Nepal, the Aga Khan Rural Support Project in Paldstan, and the intoduction of participatory irngation management in Sri Lank prove that good project design can pay off handsomely (Uphoff 1988). At the sam time, it has been recagp'zed that an exclusive commitment to "bottom-up" or grass roots development is rarely jusfiable. In most cases economic growth creates new opportunities for the poor and varous existing chamnels can be exploited to help poor groups. These chas include policy reforms, bureaucratic intevention, and the activities of NGOs. More recenly, attention has turned to the need for susinable approachs to rral development. Also, numerous attempts have been made to define the compass of project sustainability. The Brundtland Commission, for example, defines the concept of project sustaibility as "a process of change in which the exploitation of resources, the direction of investments, the onentation of technological development, and institutional change are all in harmony and enhance both current and fitre poteial to meet human needs and aspirations" (ADB 1990). :C ~ ~ ~ ~~~~- .D - -, 4 ' '-<. ,.-. t..4 66 Dieter Bucher and Keith Johnson In a discussion of the econometric dimensions of different sustainabiliry ethics, Pezzy (1989) draws attention to the need to differentiate sustainability from optimality and survivability. A more pragmatic and practical approach to sustainability has been taken by Bamberger and Cheema (1990), who argue that project sustainabiliry should be evaluated in four dimiensions: the continued delivery of services and production of benefits; the maintenance of physical infrastructure; long-term institutional capacity; and political support. Clearly, more is at stake in such evaluations than the continued flow of benefit streams and the mainnace of stocks of physical, institutional, and political capital. It is necessary to look at macroeconomic, fiscal, and environmental effects, and to take some measure of a project's impact on the survivability of target beneficiaries and their capacity for self-help. If a project or program has a negative impact on the empowerment and motivation of the poor, it matters little that it is sustainable in the conventional sense. ADB 's Initial Approach to Rural Development The establshment of the ADB in 1967 coincided with the start of the Green Revolution in Asia. Consequently, the ADB commissioned a review of agricultural pr,spects in Asia (ADB 1967). The recommendations of that survey by and lazmge supported the "production" approac- widely favored at that time. Recognizing the problems associated with the uneven impact of the Green Revolution in the 1970s, the Bank undertook a second survey of Asian agriculture (ADB 1976). This report stressed the importance of understanding how institutional factors interact with technological factors in the development process to produce different patterns of growth and income distnbution. The report pointed out the prevalence of low purchasing power and weak intersectoral links in Asian economies. It recommended that ADB assistance in mral development should focus on supporing land reform, creating off-farm employment in rural areas, and introducing appropnate new agricultural technologies. The rural development projects that emerged during the late 1970s and the early 1980s had many features in common and were regarded as a strong tool for addressing poverty in ADB's developing member countries (DMCs). Project planning and formulation usually consisted of the following activities: choosing physical areas and population target groups for specific projects (in terms of geogphy and occupation, but not income class); identifying resource development opportunities that represent the main potential source of economic growth for the areas and the people concerned; and deciding the extent to which (that is, how much, when, and in what sequence) interlinked productive activities in different economic sectors in the same geographic area should be simultaneously stimulated, and the extent to which nonproductive social infucture and services should be developed in addition to the economnic activities (ADB 1988a). Given that poverty could not generally be related to a single cause and that a multisectoral approach appeared necessary, most rural development projects were designed to incorporate a number of components, the intention being to implement them in an integrated manner. Thus, a t-ypical rura development project ivolved investments in more th one form of agricultural activity. It also supported the provision of inputs, credit, and assistance in the marketing and processing of crop and livestock products. Projects of this type sometimes included infrastructural components that were only indirectly related to the agricultural components, such as transport, commnications and rural electricity and, more rarely, environmeal protection activities (for example, afforestation and social forestry). In some instnces the development of social services such as the provision of health, nutrition, family planning and education services, clean drining Ruanz Development and Poverty Alleviadon 67 water supplies and public sanitation facilities, and the establishment of programs aimed at cultural and community institution building were also incorporated. The institutions involved in rural development projects and the roles they performed varied widely among ADB's DMCs. Govermnent departments and public sector agencies were generally the executing agencies. In some countries a large number of agencies were involved in the. implementation process, becaise of the diverse subsector activities in the projects. In other cases, efforts were made to concentrate on specific activities and to entrust the complex task of project inplementation to newly created independent project management units (ADB 1988b). Elduation of ADB's Experience As of December 31, 1990, the ADB had provided sixty-nine loans totaling $1.9 billion in support of rural development. Substantial sums had also been alocated to project preparatory technical assistance and institution building. The ADB has therefore built up a good deal of experience in ali facets of rural deveiopment. In 1988, the ADB subjected fifty of its completed mral development projects to postevaluation review. These projects were generaEly approved in the 1970s and early 1980s. All were multisectoral in coverage and 'integrated" in their approach. More recently, the ADB has conducted a special study of agriculture and rural development projects that reached maturity between 1985 and 1990. The following observations arc based on the two postevaluation exercises DATA INADEQUACIES- Both of the ADB postevaluation studies point to the difficulties that arise in evaluating the performance of rural development projects. Although only 50 percent of the projects evaluated in 1988 were found to be "generally successful," it was apparent that difficulties in obtaming data-in part because there were no adequate project benefit and monitoring studies in earlier projects-may have led to an understatement of benefits. However, evaluations using standard benefit-cost methodologies sometimes failed to adequately reflect the longer- term benefits that stem from institution building, applied agricultral research, and the *empowerment" of poor groups through the development process. IMPORTANCE OF ApmOPiUATE PoLicms. The ADB's experience strongly supports the view that even the best-planned projects will not prosper in an adverse policy environment. If, as a result of government intervention, agricultural prices fail to provide adequate production incentives, little purpose is served by making new seeds and agricultural inputs available to farmers. On the other hand, as noted in the ADB's evaluation of its Livestock Development Project in Nepal and its Livestock Services Training Project in Bangladesh, the subsidization of inputs such as veterinary drugs can cause waste and disrupt the market. TSE NEED FOR A LEXIBLE "LEARNING APOACH. Rural development projects appear to be particularly difficult to design and implement in 'blueprintw form, where the components are spelled out in advance and tigbt schedules are set for implementation. As a result of this approach, a number of ADB's rural development projects have been significantly delayed as the plans run up against field realities. Indeed, projects of this type are frequently placed m the "problem" category as disbursement begins to lag. The clear lesson here is that flexible planning and implementation procedures are necessary to incorporate past lessons and to mobilize the expertise and enthusiasm of beneficiary groups. For example, the ADB found in the review of its Nepal Livestock Development Project that mistakes were duplicated because insufficient time bad elapsed in which to draw lessons from an earlier p -4ect in the same sector. SOCIAL DEvELoPMET AND VALUE FoRDIA¶-.w . As already explained, many rural development projects have been marred by excessive complexity. In many instznces, costly :, ~ ~ ~ ~ ~ ~ ~ 4 .:- . - 0 < ' j ' i 0 00 * = y t j i;, > , . > 0 : --. ' i5 68 DinrerBtcher and Keith Johnson investments in a wide variety of "hardware" facilities such as irrigation systems, roads, and agroprocessing plants preceded or supplanted work on the less tangible "software" aspects of the devlopment process. The consensus now is that it is essential to increase beneficiary participation and to allow local executing agencies to become involved. Too often, the proposed scheduling of investments has proved to be overoptimistic when sociocultural constraints have been neglected. Local communities often have little say in the development process when the fimds and technical assistance ae channeled through government agencies (iha 1987). As a result, plans to introduce new forms of agricultual technology and cropping practices have often been abortive. On the other hand, the ADB found that although its Integrated Rural Development Project in Nepal bad been only partly successful in achievnng its economic objectives, it had effectively mobilized the poorer sections of the com_mty by taing a more gentle and pragmatic approach to overcoming sociocultural obstcles. PRovIsioN FOR OPERATIONAL AND FINANCIAL SSTAINAITY. Although the ADB's postevaluation e-ercses have generally focused on the period between appraisal and the physical compietion of a project, the ADB recognizes that many rural development projects have a low level of sustainability. As already menioned, this siuation stems in pat from the low level of participation by potential beneries in project planning and implementaton Some problems are also due to shortfalls of local currency or "counterpart funds" for both invest=ent and rmw vn expenditures. iadequate finding arangement are particularly detimenl to ural development projects becamse they tend to have high local costs as well as heavy recurrent costs. Problems with conterpart fuds usually arise when bureaucratic procedures govern the flow of funds to line nmistries and local administrative units, and to some extent when the central government is not fully committed to the objectives of rural development. QUAUTY AND COMMMENT OF LEADERP AN PROECT STAFF. It has become apparent over the years that the best rural development and poverty alleviation projects are those initted and guided by competent and dedicated private individuals. Examples of such projects are the Aga Rhan Rural Suppert Program in Pakistan, the Grameen Bank m Bangladesh, and the activities of the Bangladesh Rural Advancement Committee. Postevaluation results of projects such as the ADB-financed Small Fanner Development Project in Nepal, indicate that skilful and commited staff in executing agences can lift projects into the highly successil category. Similarly, field staff who live and work with coummnities; like Sn Lanka's "change agents," have a valuable role to play in guiding the processes of project design and implementaton. Rural Development for Poverty Alleviation-The ADB Pespective General Considrandons ADB's POVERTY ALLEVIATION ACTION PRPOGRAM. In recent years there have been a nmber of significant changes in the approach to poverty levition in the operational policies of the multilateral development ageies, including the ADB. The ADB used to deal with povety alleviation indirecty, by financing projects that contrbuted to economic growth and the efficient use of resources. Although it is still accepted that poverty cannot be eliminated without sound and sustaible economic growth, experience has shown that growth is an important but not necessarily a sufficient conditon of poverty alleviation. The ADB is therefore placing greater emphasis on poverty alleviation issues in its operations, parncularly on direct initiatives that address the livelihood needs of the poor. The ADB's new policy approach to poverty alleviation was adopted in 1988 and has two Rad Deweopmot aid Poverly AWllwoxn 69 main elements. First, the ADB is seeking to enhance the quy and scope of its conventional projects by paying greater atteion to issues relating to poverty during country programmng and project processing activities. Although some projects will always be pred y concened with meeting economic growth objectives, an effort is being made to ensure that whenever possible conventional projects will be onented toward alleviatng poverty.Project designs must therefore make a point of mimizing adverse impacts and accentuating outcomes that favor the por. Second, the ADB is emphasizing direct iniatives in its strategies for alleviaig poverty. Such direct initatives can take two main forms: * Providing greater encouragement and support for DMC programs and policies aimed directly at poverty alleviation, including the provision of program and sector lending to underpin policy adjustments that either help the poor participate in the growth process or improve their access to essential facilities of the social infras . * Developing projects and project components ("hardcore" activities) that are specifically designed to improve the livelihood prospects and living standards of targeted groups of poor people. Projects of this type will generally seek to create susntable ncome-earning opporunities and meet the basic needs of beneficiaes of physical and social infastrucure services. Rural development projects can form a major element of this direct approach to poverty alleviation. After the ADB refocsed its approach to poverty alleviation, commitments for direct poverty alleviation projects and policy reform programs benefiing mral areas rose from $152 million in 1988 to $183 million in 1989 and $214 million in 1990. About $582 million was expected to go to projects of this type durig 1991. Nepal, Pakistan, Bangladesh, and Sri Lank have been among the principal beneficiaries of these initiatives in South Asia. STATUS OF RURAL DELMENT Although substantial experience has been gained in the design and implementtion of mral development projects since the early 1960s, conceptul aproaches to rur! development are still evolving. Past experience has indicated that many rural development projects have had only limited success in reduciug pverty and in promotig broadly based and sustinable socioeconomic progress in rural areas. It is now widely recognized that direct action is required to contain and reduce the icidence of rural povery in Asia. The ADB therefore decided to reevaluate its approach with a view to improving the perfoumance of rural development projects. This review arived at four main conclusions: * Rural development projects provide a useful cnceptual hamework for grass roots, broadly based, socioeconomic rural development, and that a multisectral. integrated approach is generally apropriate for mral poverty alleviation initiatives. * Rural development projects cannot be implemented from the "top down or from the "outside" alone but need to be initiated and sustined by the participating population itself with assistance of govenment istuton, prvate development orga.izations, and eventually eternal donors (principle of self-help). * Rural development projects need to be onened to target groups. The access of these groups to productive resources, services, and marets needs to be improved. However, S-' f '- . -; --l 70 Dieter Budcer and Keith Johnson target groups must be able to use their own initiative to obtain access to the means of production, information, and markets (prnciple of subsidiary). To this end, target groups nmst be organized and mobilized, and planning mechanisms must be oriented toward those groups and their participation. * More work needs to be done to clarif the factors that underpin the success of particular projects, with a view to providing practical operational models that can be used in the processes of participatory development and project design and implementation. REGIONAL AND SuBREGIONAL DIFERENuTTON. There are significant differences in the rural development requirements of individual countries within the ADB's field of operations. In particular, conditions in the relatively poor DMCs of South Asia are markedly different from those in Asia's newly industrialized economies. In most South Asian DMCs, GDP per capita and sial development indicators have improved over the past decade but large numbers of people still remain below the 'poverty line" and most of the poor live in rural areas. Here infucta facilities and services are often lacding and institutional capability is relatively weak. However, there are exceptions. For example, India's rural support institutions are relatively well develnped in some states and the rural population in Sri Lanm enjoys good access to reliatively high standads of social services. South Asian DMCs support a wide range of rumal development activities, including physical infrastructure development, social services provision and employment creation, as well as directly productive activities both in the agricultural and nonaricultural sectors. Particular emphasis must, therefore, be placed on institution building in support of antipoverty progras. The ADB is curently involved in rural development projects throughout South Asia (with the exception of India) and is putting increasing emphasis on poverty alleviation in its policy dialogues with DMCs, based on its Country Strategy and Country Programing work. Operational Approaches The ADB now views rural development as a basic weapon in the direct attack on poverty. It also stresses distinct poverty alleviation objectives, benefit targeting, participatory development, and sustinability in its operational approaches. The practical issues gaing increasing attention are new proc&0ures for project preparation and design, project imple ion, and institution biding. PRtoTa PREPARATION AND DESIGN. In the area of project preparation and design, the focus is on participatory planning and social analysis. Panwpatory pklnig. The ADB's expeience suggests that more time needs to be allowted to project planing. In particular, it is essential to ensure that local institutions and potential beneficiaries participate adequately in the preparation and design of a proposed rural development project. Feasibility studies sould identify iportant development constraints, account for relevant indigenous experiences, and gauge the development potential of the project area and the wilingness of the local population to participate in project activities. Before these tasks can be carried out and resources committed to feasiliy studies, it will be necesary to conduct an in-dth analysis of local development needs. Second, early and sustained contacts, both formal and informal, will have to be established between the project planning team and the local population (oth the project beneficiaries and executing agency staff). This will help to Rarda Development and Poverry Alleviation 71 generate a higher degree of govemment and beneficiary commitment to a project. At the outset, rural development projects should focus on the needs and capabilities of targeted beneficiary groups. The core component of any scheme should be pro-esses that add value to the resources controlled by the rural poor. In general, the poorer an area in terms of services and the less developed the country, the more consideration should be given to off-fanm production activities for generating income, creating employment opportmities, and establishing social infrastucture facilities (such as potable water supplies, sanitation systems, medical clinics, and schools and community centers). In the initial stages of a project, services and facilities that will have a measurable and immediate impact on the standard of living of the very poor should be emphasized. Such confidence-building measures will both reinforce the attraction of the project to the beneficiaries and encourage their participation in its productive activities. Soda) analysis. The ADB's approach to participatory development. which is now commonly regarded as a prerequisite for rural development projects aiming at poverty alleviation, has recently been strengthened as a result of the adoption of and dissemination of guidelines for the social analysis of projects (ADB 1991). The guidelines commit the ADB to (a) taldng due account of the sociocultural dimensions of the development process, (b) following people-oriented and people-responsive development approaches, and (c) involving beneficiary groups in the processes of project design and implementation. The ADB will pay special attention to social issues durng the processing of loan in order to identify the potential consequences of a project for different population segments and mold the design of the project to mitigate adverse impacts and ensure that the expet benefits are realized in an effective and timely manner. Whenever a project is likely to have significant positive or negative impacts, a rapid social assessment should be conducted during fact-findig and project preparation. The implementation of this proposal will necessitate action to make suitably qualified staff available or provide resources for technical assistance. In cases where problems may arise in mitigating adverse impacts or providing a framework for beneficiary participation, a social design study (SDS) will be prepared. An SDS may thus help planners to (a) optimize the project design, (b) develop an appropriate implementation strategy, and (c) specify appropriate covenants and policy conditionality provisions. Additional technical assistance funds will be needed to implenent this approach. Whenever feasible, SDS work will be coordinated with or will incorporate the preparation of baseline data for project benefit and monitonng purposes. PRoJccT IMPLEMENTATIoN. The central concerns under tin headig of project implementation are project complexity, flexibility, scale of costs and cost recovery, lending modacSides, and project implementation period. Conq p1exity. Past experience has illustrated the problems associated with complex project designs. To begin with, fully Wintegated" projects proved difficult to implement. Adopting an integrated approach does not, however, necessarily mean that action in all of the sectors identified for intervention must be implemented simultaneously. Although several factors may be contributing to the underdevelopment of resources and the overall poverty of the population m any given area, only one or two of these factors may require immediate attention under a sequential approach. A project should concentrate on only a few entry activities aimed at removing identifiable constraints. The choice of entry acties should center on those that will secure early productivity benefits for the project participants or provide them with a much-needed- service or facility that secures their commitment. Other activities can then be introduced, to allow the total project activity mix in the area to gradually become more comprehensive. P2biNfity. Rural development projects must also be flexible enough to allow the .4 - .- 72 Dieter BHicher and Keith Jokhson experience gained dunag the earlier stgFs of implementaton to be reflected in the designi of subsequent phases. Project designs shoud allow project implementation to proceed in relatively short, distinct phases. These phases should include action-oriented or in-service training of both local govenment staffand beneficiary group leaders. New technologies, organizational structures, and delivery systems should first be tested through applied research tials with appropriate incentves to ensure the participation of all concemed. As discussed previously, a "blueprint" approach to project design is unsuitable for rural development projects seeking to ignite a process of social change. The nature and extent of the changes required in institutions and in the social behavior of target groups cannot be anticipated or specified in advance. The achievement of targets and the ultimate impact of rual development projects therefore depend on securing the active and contiued participation of the idenified beneficiaries in project activities, the majority of whom are usually poor, illiterate, and frequently isolated. Alt hogh project designs may still be required to contain certain specifications regarding long-term development targets for each major component (in connection with the sitng of roads, the coverage of veteriay programs, and so on) allocations for detailed ex ante planning should be curtailed. Project preparation resources can then be devoted to (a) planning for the mobilization of local inputs and-rercurces at the beginning of project implemen , (b) field testing new activities, and (c) establishing the or onal framework for the initial participation of executing agencies and beneficiaries, and then, 7.miliarization with the concept of the project and their potential role in it. This "process approach," althouigh time consuming at the onset, provides a basis for long-term sustinability. The scale and iniece of costs and cost recovery. With the increasing shortage of domestic resources for development needs, project sustainability has become more dependent on the scale and incidence of costs and the potential for cost recovery. The total cost per beneficiary of a rural development activity should be affordable. Ohrwise, the social impact of the project will be relatively restricted and its susinabiity endangered. Project benefits should also be widely distrbuted, reaching as large a number of the rural poor as possible. Only then can the project be considered a rqplicable model for rural development. In additon, an effort must be made to recover the costs of project investments from the beneficianies- to ensure that project faciLities are operated and maintined on a sustainable basis. Ledng modalizes. Rural developxment project differ from other projects both in their nature and scope becase of the mnltiseetoral dimensions of mral problems and the combinaion of short-term and long-term objectives of such projects. It is often desirable to trigger an immediate impact on the productivity and income of selected social groups while establishing a base for the progressive reduction of rural poverty on a broad front. Lending poicies must therefore be flexable. Many old-style ADB-flmnced rual development projects were unable to proceed with implementation as originally scheduled, and progress in the first few years was slow. This slow progress can be partly attnrbud to rigid financing procedures both within the ADB and the DMCs. Greater financing flenbility is needed to safeguard both the inancing agency's and the recipient country's authority over project direction and content. It is therefore necessary to develop appropriate financing formulas that restrict detailed advance budgeting to major investment components, while putting aside adequate portions of committed funds to meet unllocated budgetay items under discretionary spendg by local executng agencies, or even beirficiary oions in some ci mst Proect mplementaon penod In general, rural development projects require reltively Rural Devwtopmez wad Poverty AUeviaton 73 long innmentation periods, for examp_., seven to ten years. Depending on various factors (such as the institutional framework), some projects may take as long as fifteen years to implement. Donor agencies should therefore be prepared to finc certain categones of costs beyond the expected completion dates of infrastucture components. Furthermore, the completion of "hardware" investments should not necessarily define the end of a project's implementation period. The definition of an appropriate period for project financing depends on the nature of the project and the ability of the govemment and project beneficiaries to finance the contination of the development process. INsITnONAL CONSDERATIONS. Instkutional points of concem are the characteristics of the project implementation agency, the degree of decentralization and local management, and cooperation with NGOs. Project implemeion agency. Rural developmet projects are difficult to implement at a single administrative level or through an individual implementing agency. The appropriate form of administrative organition for rural development varies among the DMCs, and donors must therefore adjust their approaches to the prevailing administrative system. A plethora of investment components and executing agencies is generally a recipe for disaster. The ADB, like other donors, is now designing less complex and more effective projects that can be implemented and phased more simply. Nevertheless, substantial coordinadon at the national and regional level will sdti be required. Decentralization and local level management. Expience with ADB-financed projects suggests a substarlal measure of management decentalization is needed to ensure that local govermunt bodies and other development-onented institutions are involved in project management. Decentalization is most important when a complex, multisectoral mix of actvities has to be integrated and scheduled. Local management provides the functional flexbility that is needed to modify projects and programs as circumstances warrant. Furthermore, the greater authority, responsibility, and accuntability at the local level encourage prospective beneficiaries to participate actively in the decisionmaking process and stimulates teem to inie complementary activities using their own resources. Cooperaion mith NcOs. In several of the ADB's DMCs, NGOs have the capacity to prnvide substantial development assistance at the local community level, particularly with regard to proect preparation and design, manpower training, technology ransfer, and project implementation. Because the recognition of training value, social mobilization, and technology transfer are vital objectives in most rural development projects, the ADB now initiates dialogues with relevant NGOs at an early stage in project formulation. NGOs are also encouraged to play a major role in sN g instiuon-building efforts, etablising effe conication with beneficiaries, and assisting in value formation and the social mobilization of beneficiaries, and to participate, where appropriate, in the implementation of specific project components. Concision The conceptual framework for rural development contimues to evolve as practitioners attempt to devise and implement new approaches to project formulation and implementaton A recent inovation has been to link rural development to direct poverty alleviation goals. Although difficulties still abound, there is renewed optimism that nual development projects can be bankable for development agences, affordable to beneficiary groups, and sustinable over a loqg period of tine. The keys to good project design and smooth imple ion lie in undrstanding and harnesing the energies and aspirations of the poor. A targeted approach that allows N, . ~ .-~- X - - - 74 Dieter Budcer and Keith Johnson beneficiaries to participate in project preparaion and implementation is, therefore, required for project sustainability. Such projects cannot be too complex, yet they nmst be integrated in nature in order to address the various ficets of poverty. A phased approach allowing flexibility in project implementation therefore seems to be desirable, but can only be achieved by addressing poverty issues directly and by treating rural development and poverty alleviation as long-term processes. Withm this context, practitioners need the time and scope to devise and utilize case- appropriate planning and implementation procedures, backed by a firm high-level commitnent to the achievement of human development objectives. Rural Developmmt ard Poverty Alleiaon 75 References Asian Development Bank (ADB). 1967. Asian Agiiahural Surve. Manila. . 1976. Asian Agriculture. Manila. . 1988a. Sector Paper on Rural Development. Mnia. . 1988b. Cooperation with NGOs in riculture and Rural Development. Mania. . 1990. Economic Pollcies for Sustainable Development Manila. . 1991. Guidelines for the Socal Analysis of Development Projects. Manila. Bamberger, M., and S. Cheea. 1990. Case Studies of Project Susanability: Implicationsfor Policy and Operatons from Asian Experience. Washingtor, D.C.: World Bank, Economic Development Institute. Bandyopadhyay, D. 1988. -Direct Inervention Programmes for Poverty Alleviation -An Appraisal." Economic and Political Weekly. New Delhi, June 25. Cutdng, W. 1977. 'Nutrition-Making the Best Use of Available Food in South India. in E. Stamp, ed.. Growing Out of Poverty. Oxford University Press. Ghate, P. 1984. Direct Atacks on Rural Poverty-Policy Programmes and Imp ementaion. New Delhi: Concept Publishing. Hossain, M., R. Mannan, H. Z. Rahban, D. P. Roy, and B. Sen. 1991. "Priority Issues in Rural Poverty Alleviation in Bangladesh." In ADB Study on Priority Issues and Poicy Measures to Alleviate Rural Poverty. Manila. Forthcoming. [a, S. C. 1987. "Rural Development in Asia: Issnes and Perspectives." Asian Development Review 5(1). Manila. Lewis, J. P., and others. 1988. Strengdtnig the Poor: What Have We Leaned? New Brunswick Overseas Development Council. Mahendra, Dev S., Kirit S. Parikh, and M. H. Suryanarayaa. 1991. *ural Poverty in India- Incidence, Issues and Policies.' InADB Study on Priority Issues and Poliey Measures to Allevia Rural Povert Measures to Alleviate Rural Poverty. Manila. Forthcoming. Pezzey, J. 1989. "Economic Analysis of Sustainable Growth and Sustainable Development." Working Paper 15. World Baink, Environment Deparment, Washington, D.C. 76 Diner Buher and Kaih Johmon Quibria, M. G. 1990. 'Poverty-Conceptual and Measurement Issues." Paper presented at the ADB-EDI semiar, "Poverty Alleviation through Agricuhltral Projects," Dbaka, October 1990. Radhakrislma, R., and S. Indrakant. 1988. 'Effects of Rice Market Intervention Policies in India: The Case of Andhra Pradesh. In ADB, Eviuatdng Rice Market Intervenion Poliaes- Some Asian ExaWles. Maila. Singh, 1. 1990. The GreatAscent-zhe Rural Poor in Sout* Asia. Baltimore, Md.: John Hopkins University Press. United Nations Development roamme (NDP). 1990. Human Development Report. New York. Uphoff, N. 1988. 'Assisted Self-Reliance: Working with Rather than for the Poor," In J. P. Lewis and others, eds., Srengthing the Poor. ITht Have We Leared?' New Brunswick. Transaction Books Waller, P. P. 1987. Probleme der Umsetzuug des Grundbeduerfniskonzeptes: das Beispiel der Regionalentwicklungsprojekte." In D. Schwefel2 ed., Soziale Wirlamgen von Projekten in der Driten Wel. Baden-Baden. World Bank. 1990. World Devdopment Report. New York Oxford University Press. 5 THE URBAN POOR AND URBAN BRASIC INFRASTRUCTuRE SERviCES iN ASmA: PAST APPROACHES AND EMERGING CHALLENGES Yue-Man Yewng Sice the end of World War II, rapid urbanizton has been a sent factor in shwing development policies in the cities of Asia The lack of basic irastruaure servces has been a senous problem in Asian cries, fek most acutely by eke poor in low-income communes. The problem is exected to grow worse in the flare. The urban poor, hJver imprecisely dejine4 tend to exhbit a regional concenaion in Asia, which is most pronounced in South Asia. The chauengeforplaners and poicyimakers, now as wel as in the future, is to provde the poor with better access to basic uba services. Various methds have been used to deliver urba serWces, with vaying degrees of success in reaching the urban poor. The current trend is toward mltjkceted and integrated projects, with beneicy participation. Counry and project experience in Asia offers mwiy lessons, both of success andfailure. Some of the more salient ones are the failure of top-down planing, the need for comunit parficpaion, the merit of strengtheing local govermes, he wgentcy of investin leadership and trainng, and the need to increase women's contrbutons. An eminat of the Asian experience in delivering basic urban servces hrow into relief a host of key policy issues and aternatives. Foremost among these is the question of who pays for the services. This is an issue that Urba govewnmen are partiadarly iU pred to tckle in view of eir low revenue base. Many otier issues may be concepuazed in a fawimy of dichotomes or tradeoffs, such as cost recovery vers subsidzation, centraliaion versus decentralatio.n, eqity versus encey goals, public versus private sector provision, formal versus informal sector roles, high versus appropriate techologies, and so on. Tis chapter rases a number of quesions th need to be addressed if Asian cowries are to come to terms with thr emerging chalenges. Introdon Cental to much of the character and pace of development in cities of the developing world m the postwar period has been the unrelenting and rapid urbanization. Urban populations have been grwig at twice the rate of national populaons. For the past three decades, urban populadon in Africa, Asia, and Latin America has been growing at the rate of 3.49 to 5.10 perent a year, brnging the total urban population in developing countries to 1.19 biillonby 1988 (see Table 5.1). Among the many problem that rapid urbanizaton has posed for cities in developing counies has been how to provide a reonable level of basic ifastructure services compatible with hnuman decency and basic needs. While the present inadequaies in such service 77 - Z 78 Yue-Man Yewng ~Table 5.1 Urbanization Trends in Deveoping Regions, 1988-000 Region Urban Pop. Urban Pop. Change Annual (thousands) (thousands) +() Growth 1988 2 _ (percent) (p@rcen) Africa 133,245 250,625 117,380 5.4 ASIA 591,908 877,650 285,742 3.3 EMENA 175,140 245,162 70,022 2.8 LAC 291,656 404,621 112,965 2.8 Total | 1,191,949 1,778,058 586,109 3.4 Note: The regions correspond to the World Bank's geographical regions Source: United Nations (1988). provision are already obvious and serious, the challenges ahead for planners and policymakes in these counties are even more daunting. For the remaindr of this ccntuy, the increase in urban populaion is expected to moderate in certain developing regions, but for Asia it is exected to remain high at 3.3 percent annually for the period 1988 to 2000. The Asian rate will ease up from 3.65 and 3.63 for the 1975-80 and 1980-90 peiods, respectively. However, for the period 1988-2000 the urban population in developing countries wil increase by 586 million, which represents almost three- fifths of the total population increase in dtese countries (Table 5.1). Asia alone will accoumt for almost half of the increase, and its totl urban population of 877 million will also represent approximately half of the total urban popuon in developing countries at centLy end. Another indicator of the nature of Asian urbanization is that by 2000, that coninet will have 148 cities, each with a population of more than I million, a vast increase from 69 in 1980. Also, of the world's largest 22 at tbat time, Asia will have the lion's share of 13, each with a population of more than 10 million. Rapid urbtion, by itself is no case for concern, but when it is accompanied, not by econtnic restucturing and greater employment opportuities, but rather by growing social inequities and economic inefficiences, it becomes a phenomenon of increasing magnitude confronting policymakers, planners, and administrators in developing countries. To add to the problem, a large proportion of the new urban population is poor; hence, the incidence of poverty, is shifting from rural areas to urban aras. The growing incidence of urban poverty is intensifying The Urban Poor and Urban Basic IrsmWre Serices in Asia 79 an already serious situation with regard to employment, the environment, and physical and social services. In the large cities of Asia, where large segrients of the population are living in abject poverty, the severe inadequacy of basic inrastrcture services is magnified. In this chapter the term "basic in ture services"is used intdie conventional sense to refer to services such as housing, transport, and water supply. It is not extended to other services that may be viewed basic to the needs of urban dwellers. If the basic needs of urban dwellers are considered, one might conceivably include food, energy, land, and perhaps other services. The notion of access encompasses both the availability of services, however these are provided, and the affordability of services by the inhabitants. In brief, the physical provision and financing of services are equally perdint to the consideration of access by the poor to urban infrastructural services. Efforts havc been made to improve that access at international, national, and city levels. After almost two decades of trial and experimentation, a growing stock of experience, both positi]e and negative, has accumulated in the delivery of basic services to the urban poor. This chapter reviews the available literature, case studies, and the author's experience in the cities of Asia to better understand the urban poor, highlight past successful and unsuccessful approaches to basic infrastructure service delivery, take stock of the lessons leamed, outine key policy issues, and provide some thoughts on the challenges of the fimre. The Urban Poor Despite the still unsettled debate surrounding the conceptual and operational definitions of urban poverty, one fact is clear: cities in developing countries are being saddled with large segments of their populations that to all intents and purposes can be viewed as poor. Any realistic urban policy must take into account this large and growing proportion of poor urban populations. Even in developed countries, urban poverty is a persistent social condition that has failed to disappear with national and general economic progress and prosperity. The inescapable conclusion is that urban poverty wil be with Asian cities for as far as we can see into the future. Many students of social conditions have atempted to grapple with the concept of poverty. Although most would agree that the notion of subsistence poverty is arbitrary and circular, for it involves a subjective judgment about the monetary and physical needs of individuals, the idea of absolute and relative poverty seems to have gained currency. There is support for the notion that poverty is a general form of relative deprivation, the outcome of the maldistribution of resources. Maldistbution not only involves income, but also other resources such as capital assets, occupational fringe benefits, access to public seri.ces, and pnrvate income in kind. In the final analysis, it is the household's effective command over society's resources that would determine its relative position in the society of which it is a part. A household in poverty is one that has resources, opporhtities, and economic surplus substantially removed. This immediately raises questions concering social and economic inequality as part and parcel of the concept of poverty. Notwithstanding the imprecise and relative nature of the concept of poverty, politicians and planners have tried to apply it in order to assist, or sometimes appear to assist, the poor. Policies designed to narow the "poverty gap" or "income deficit," and the World Bank's "lower 40 percent" principle are examples of puttig into practice measures to alleviate poverty. Income aintce programs in several developed countries are varias on the same theme. National and intrnaional agencies have often adopted the concept of absolute poverty that is expressed m some monetary value equivalent to a minimum level of consumption necessary to sustain 80 Yue-Man Yea; physical health. However, problems of data collection, comparability across countries and cities, rVpidly changing economic circumstances, coupled with religious, etnic, and ecological factors, have made the concept of limited value. In the absence of a morz manageable concept or adequate data, absolute poverty is often applied with the -poverty line' to delineate the size of the poor population. Another conumon practice is to array the population by income levels in percentage terms or some statistical groupings, whereby policies designed to rcach a certain proportion of the poor population can be tailor-made and more accurately targeted, at least theoretically. Economists, including Simon Kuzmets, have observed for some time that the upper- income groups in poor counties tend to receive a larger share of the total income than their counterparts in rich countries. In other words, incomes tend to be distributed more unequally in low- ffian in higher-income countries. World Bank data on social indicators for Asia in 1980 indicate that the percentage of income received by the richest 5 percent in selected countries was as follows: 16.7 in Bangladesh, 26.3 in India, 17.1 in the Republic of Korea, 27.0 in Malaysia, 17.8 in Pakistan, 28.8 in the Philippines, 18.6 in Sri Lanka, and 23.8 in Thailand. The percentage of income received by the poorest 20 percent was 7.9 in Bangladesh, 6.7 in India, 7.1 in Korea, 3.3 in Malaysia, 8.0 in Pakstan, 3.7 in the Philippines, 7.5 in Sri Lanka, and 6.1 in Thailand. it is clear from these figures that income distribution is extremely uneven in some Asian countries, but the situation varies from country to country. For instance, income distr'bution is more unequal in Malaysia and the Philippines than in Bangladesh, Korea, Paldstan, and Sri Lanka. Within Asia, the proportion of urban population below poverty income varies by subregion and country in tems of per capita gross national product. The World Bank social indicators of 1985 depicted the following distribution of urban populations below income level: Bangladesh (86.0 percent), Burma (40.0 percent), India (40.3 percent), Indonesia (26.0 percent), Nepal (55.0 percent), Pakistan (32.0 percent), Sri Lana (26.0 percent), South Korea (18.0 percent), Malaysia (12.6 percent), the Philippines (32.0 percent), and Thailand (15.0 percent). The diffences in the level of urban poverty implies that the intensity of the problem varies, as do the policy instrments. The problem of urban poverty appears to be most serious in South Asia, whose urban populations and by extension the urban poor have been increasing rapidly. It has been estimated that South Asia's share of the world's urban poor households will increase from 31 to 44 perent in the period 1975-2000. By the end of the century, South Asia will have an esfimated 32.6 million urban poor households, which means more than 100 milion people in poverty. Consequently, the sitation in Asia clearly calls for a multiplicity of approaches to alleviate poverty and to provide the poor with better access to basic iatutue services. Past Approaches to Defiveing Urban Services Extemely diverse mechanms have evolved for delivering basic services to target populations. Only the more predominant ones can be covered here. It should be pointed out that by and large they are ones in which the urban poor themselves have played key roles. Two multicountry projects in Asia were funded by the Intemational Development Research Centre (IDRC) of Cnnada The first project, called Participatory U.-ban Services in Asia, involved Hong Kong, Indonesia, the Republic of Korea, Malaysia, and the Philippines. The project focused on diffirt ways the urban poor improved their physical and social environment when basic services were found to be deficient. As in other parts of the developing world, a queuing system of some kind is at play in these five economies, and for lack of political clout, vocal support, and appropriate orgaon, low-income communiies are served only after The Urban Poor and Urban Basic hfrastrecmue Sevics in Asia 81 higher-income areas. The poor have therefore organized themselves and delivered the needed services. The hamessing of community resources in participatory, self-help, and cooperative styles of participation has helped narrow the gap between the demand and supply of urban services in low-income conmmunities in these countries. The project also revealed the general ineffeciveness of a "service delivery" model (that is, government-provided services) and reaffirmed the worth of experimental and innovative efforts to mobilize people's resources toward improving the urban living environment (Yeung 1985; Yeung and McGee 1986). One of the questions investigated under this project concemed the geographical and demographic unit that would be a fiuncionally efficient basis on which to organize participatory urban services. Is there a minimum threshold size? In Indonesia and the Philippines, the lowest administrative units-namely, the keluroiha and the barangay-were found to be convenient building blocks for extending or organizng urban services because they possess an effective leadership structure. Second, the study found a positive relationship between strong leadership and successful delivery. Leadership structures are especially well-developed in the barangay in the Philippines, which appears to be a better system than the one prevailing in die Indonesian kmpungs, in which leadership is centralized in the lurak. The bnr, an appointed person, takes all key decisions in implemenng sevice programs and, as such, is not entirely effective in delivering services to his community. Third, it was found that government program goals did not match residents' needs in most of the delivery systems. In Penang, where goverment programs had been slow to react to the needs and problems of young workers, they tuned to a host of nongovernmental organizations, which turned out to be far more successful in identfying and meeting the needs of a new work force. Another finding was that the needs and choice of activities for community participation cannot be adequately articulated unless infrmation is available about the target population, their characteristics, and aspirations. Ideally, information should flow freely among four different levels-individual, household, community, and stae-so that appropriate strategies can be adopted to meet the socioeconomic needs of each urban community. Programs Directed at Aleviaing Povet The second comparative project, representing the concerted action of the national and city governments, is concerned with a whole range of strategies aimed at alleviating poverty m general rather than urban services per se. This line of action rests on the notion that a direct attack on poverty is needed to improve the lot of the urban poor and plays down the trickle-down effects of economic growth on this populion group. The study was comparative in only a loose sense as it was carried out at different times witho-ut a strictly comparative hamework. It covered Metro Manila, Korea, and Malaysia (Yeung 1988). One conclusion reached in this case was that the degree of awareness among the cites in poverty programs and their utilization rates are closely linked. That being the case, it is vital for the agencies concened to increase public awareness. Second, the problem of insufficient funding for the progrms is a serious one, reflecting the country's and the city's priority in econmic growth or other social programs. It is thus necessary to rank services and target groups in order of priorrty when resources are limited. Third, the duplication of efforts is a general problem but one that especially bedeviled Manila's numerous programs. An effort should teefore be made to mount these programs in a way that will avoid duplication but ensure maxmum efficiency and filA coverage of the target populaon, with built-in onitong and evaluation. FinaLly, it was found that organizational fiators are closely related to program performance. Given the geraUy low educational levels . . .- 82 Yue-Man Yeng of the target groups, the organizaional structure should be simplified, cost-effective, and designed in such a way that the programs can be easily used by, and identified with, the needs of the target population. Serlemenu aUpgrading Programs Over the past two decades, Asian countries have built up a store of rich experiences in upgrading slum and squatter setdements. Even a cursory examination of seme of these programs would highlight their country-specific strengths and weakness. The largest settlement upgrading program in the world is the Kampung Improvement Program (KMP in Indonesia, which started in 1969 as an imdigenous initiative to mprove the worst living conditions in Jakarta but has flowered into a gigantic, multilatel program, with substantial support from agencies such as the Worid Bank and the Asian Development Bank By the Third Plan (1980-84), KIP had expanded to 200 cities, including many medium and semall cities, and was benefiting 3.5 million inhabitants. The Fourth Plan (1985-89)extended the program to a total of 398 cities. KIP, adopting a "public works" physical infuctre approach to settlement upgrading, is essentially an environmental program and is not designed to alleviate urban poverty. In fact, an obvious weakness of KIP is its almost total neglect of the social needs of the inhabitant, which it has left to organizations such as the United Nations Children's Fund (UNICEF), which developed the Urban Kampung Services Program for delivering social services in low-income kampungs through communty participation. The program covered the penod 1979-84 in four cities: Cirebon, Yogyakarta, Surabaya, and Ujung Pandang. Land issues are not touched in the upgrading process, as more than 95 percent of the houses in Jakarta are not physically affected. Although a recent study on KIP in Bandung reported a new strategy of recovering costs through a property tax, service costs are generally not recovered from the beneficiaries. In tbis manner, the program has reached a huge number of poorly serviced urban communities across the country within a relatively short period (Devas 1981; Soegijoko 1985). It should be noted, however, that from 1984 to 1989 KIP accounted for less than 10 percent of the total development expenditures on the public works sbsector, with water supply and roads occupying 35.4 and 30.7 per cent, respectively, as the largest spenders. In Thailand, the Slum linprovement Program is implemented by the National Housing Authority (NHA). Some 26,000 dwelling 'mits were included in the program during the 1978-82 period. Experience to date points to problems in land acquisition, cooperation from slum dwellers, and the question of standards and physical design. A typical project here is the King Petchproject in Bangkok, which covers half of its total costs with funds from aWorld Bank loan to the Thai government. This program devotes equal atention to physical and socioeconomic conditions, both of which are in need of rehabilitation. The experience in King Petch mirrors a process that has been going on in many other slum areas in Bangkok. By varyg degrees, the program is being extended to other cites in Thailand and, in concert with other concurrent efforts, has been able to reach more of the poor whose housing conditions are in great need of improvement. In Metro Manila, the NHA similarly implements a slum improvement program known as -the Zonal Improvement Program (ZIP), which is designed to provide land tenure and services to more hn 1.8 million people in 300 depressed areas over twelve years. It is a "total" approach to settlement upgrading, for it is based on the eleven basic needs approach announced in 1979. Complementg the ZIP are two upgradig programs called the Metro Manila astucture Udlities and Engineering Program (MINUTE) and the Program for Removing Sewage from Steets (PROGRESS), designed by the The Urdon Poor and Urban Basic Infrastrfre Services in Asia 83 Mimistry of Public Works as a "network" approach to reinforce the "area' approach of ZIP (Taylor and Williams 1982). In India, urban reforms since 1972 have substantially improved living conditions for the three million people who live in bustees. In the new relationship between the state and the bustees, the former has intervened and used its power to ensure that the bustee dwellers are protected from unfair property control by the Thika tenants and private landlords. Calcutta is no longer gripped by an air of desperation about its bustees and basic utilities. It has upgraded environmental conditions in large tracts of private rental housing through redistributive public finance (Pugh 1989). In other Indian cities, shelter policy has been based on experience in upgrading settlements in Delhi, Chandigarh, Ahmedabad, and Hyderabad and has focused on issues such as program content and coverage, the informal economy, affordability, security of tenure, and beneficiary participation (Datta 1987). Mutzfaceted, Comprehesive Approach During the past decade, Kuala Lumpur has evolved an integrated, multidimensional approach to '-pgrading squatter settement. The squatter problem began to attract public attention in the 1960s because of the rapid growth of the squatter population. Between 1974 and 1980, the squatter population grew at 9.7 percent per year and, in 1982, reached a total of 243,200, occupying 7.3 percent of the total city in 177 areas. In 1979, the comprehensive Nadi program was organized. It was an outgrowth of the previous efforts of the Sang Kancil (named after the clever mousedeer in popular children's stories in Malaysia) antipoverty project fumded by UNICEF and the expanded National Family Planing Board Program, which combined parasitic control for children with family planning. As an integrated program, it was designed to provide a multiagency delivery of services and resource utization, group and area specificity, a sharp focus on the family as the unit of development, and the direct participation of beneficiary groups. Eighteen agencies have bee involved in delivering services, which can be grouped into a community-based operation with 173 difrferent tasks (Diaz 1982). The bulk of the funding comes from the Malaysian government, with technical support from UNICEF in the early stages. The main services provided under the Nadi program are comprehensive health services, basic infrastructr services, and communty and family development activities. In actual practice, the services that have received the lion's share of fumding are electricity supply, the Sang Kancil project, parasitic control, and provision of standpipes. The Sang Kancil project was the first to use squatter participation in project design and to reflect official awareness of the multifaceted needs of the poor. Lim (1985, 1988) has fully documented and analyzed the Nadi program and emphasized the need to develop administrative structures and the capability to implement new strategies for delivering services to the poor. Appropriate structures, in turn, require prior identification of problems and adjustment to changing socioeconomic conditions and priorities. The program has not achieved a snowballing effect in Kuala Lumpur, but the experience to date has certainly been positive. The reason for the relatively limited impact of the Nadi program is that, not being designed as a self-cntained program of any particlar agency, it is dependent upon the contnbutions and cooperation of a multitude of existing agencies. Thus, the Ministry of Federal Territory under which a steering committee is set up to imWlement the program, has no earmarked resourcs to expand its coverage in tenns of the range and adequacy of services, or the number of squatter settlements. - . - - .- 84 Ye-Maon Ywufg Integrated Social Servces Program in the rapidly strializRepublic of Korea, a new program, with an accen on social rather than physical services, bas been developed in Bongehun dong, one of the largest squatter areas located in southern Seoul. The idea of an integrted services program for the poor in the urban district stemmed from a series of workshops organized by UNICEF in connection with its contribution to the nation's Fifth Five-Year Development Plan (1982-86). The main objective of the project is to foster the developmet potential of the urban district through the provision of a package of services with the cooperation of government and private organiztions. In this sense, the program depends on a mixture of "top- down" and 'bottom-up" approaches. The high motivation and professionalism of nongovernmental organizaion (NGO) staff have elicited a positive response from the residents and the program has achieved a measure of success. Three tp of community or ons-govermnt-backed popular o ions, the community development committee, and project organizations-have been insumental in building a consensus among the beneficiaries. The exprimenl project augers well for residents in the area, but there is no comprehensive, long-range plan for providing services for the urban poor in the city or the country as a whole (Whang 1988). Despite its rapid economic growth, Korea is yet to devote more attention and resources to poverty alleviaton. The situation is not as serious as in soni other Asian counties, but a findamental change in the thbing of planners and politicians is desirable for the sake of equity and balanced development. Goernmet-Proddd Community Developmet Mnoter variant of the community participaion approach to delivering urban services can be seen in the Community Development Councils established in the slums and shanty gardens of Colombo during the past decade. In 1978, the government's detnion to improve social developmen and physical infastructre within the urban areas gave rise to the Slum and Shanty Development Unit, established under the Ministry of Local Government's Urban Developmen Auhority. In the following year, the Envionental Health and Commuity Development project was launched in the slums and shanties of Colombo with UNICEF's assistance. A three-tier community development council system was developtd, drawing representatives from the nuclear level in the slum gardens, from the district level, and from the city level. The 1979-43 five-year program focused on recruiig and training a new cadre of field workers calied Health Wardens, who were to acquire knowledge and expeience in community development, primary health care, nurition education, and envirotal sanitation. Under their mobilization, 291 community dve-lopment councils were established by 1981, covering 15 percent of the Colombo slum and shanty populon. By 1981, 723 latrnes, 340 bathoms, and 543 stndpipes in 285 shanty gardens had also been mwproved. Clearly, the inrastl improvements were iWmpressive. Municipal authorities were so satisfied with this progress that they took over the payment of the wardens' salaries in full, after the initial No-year commitment by UNICEF came to an end, and the goverment planned to extend the new method of amenity upgrading to other cities in Sri Lanka (CasiM 1982). A smiWr program, but one highly dependent on community efforts, was the successful Million House Program cvering the period 1983-89. It covered different types of slums and squatter setflements and employed the methodologies of upgrading, minimal relocation, sites and servies, and planned plots. It relied heavily on community participaiocin action planning, on- The Urban Poor and Urban Basic lrramtawe Srvices in Asia 85 site blocking out of land, contract system, mainnance of infrastructure, enterprise proam, and training leaders. This program was so well received that the 1.5 Million Houses Program is being implemened for the period 1990-95, following the outline and themes of the previous program. Another project that evolved on a similar mode of govemment-community cooperation is the Urban Community Development Project in Hydeabad, India. Although initiated by a department of the municipal government, the project was aimed at improving housing conditions, child and mother care services, preschool education, and incomes of the urban poor. After beginning in a small locality in Hyd&abad, the project has since been extended to the entire city. As a result of the success of tie project, national and intional sources have provided substantial financial support for replicating the program m other Indian cities. Aided Self-Help A rmuldtitude of inovative and resourceful self-help approaches to improving basic urban senrces have also bee tried out in Asian counties. Most such efforts have been intiated by NGOs. The Building Together project carried out in Bangkok in 1978 is an outstanding example of erecting a new neighborhood on the basis of mutual assistance. With professional assistance available on a voluntary and consultative basis, residents were allowed to purchase their own land, plan the site, finance the project, and eventually build the houses. The impact on the residents in question was so favorable that the project attacted funding and support by aid agencies from other countries. The design, imaplementaion, monitoring, and coordination of a rehabilitation project providing shelter for flood victims m Jabanirpur has been the responsibility of the DeNhi Catholic Archdiocese in Delhi. The project succeeded in meeti the requirements of the target group. In addition to shelter, it has introduced other componets, such as self-employment schemes, adult education, and legal aid. Similar success was attained by the Ainmedabad Study Action Group (ASAG), an NGO that took immedSite action in 1973, when twenty-three settlements involving 3,000 households were affeded by floods at Vasna, Abmedabad, after the River Sabarmati overflowed. The Iaegrted Urban Development Project (IUDF) was designed to respond to both the physical and socioeconomic needs of the flood victims. The IUDP is notable for its people-based approach and action-rearch methodology. In the Bogum Jahi area of Seoul, Korea, evicted squatters constucted a community of 170 houses in five months in 1976. The stiking feature of this building program is the manner in which the people were motivated and the action iniated for community development and inprovement. The leadership in the project came from two individuals, a priest and a university student who had lived in Seoul's largest slum for more than two years before the eviction notices were served. The project underlined the importance of peonal relwaionships in effective commnmty participaton. On a larger scale is the Orangi Pilot Project, which strted in 1980 as an NGO effort to improve sanitation in the Orangi settlement in Karachi and has sprung up more or less unplanned over an area of 5,000 acres. Through social organizaion and technical extension, community participats have installed sanitary latrines in the area's houses, underground sewerage in the lanes, and secondary or collector drains. The sanitation and general conditions have improved considerably as a result. In Manila, Freedom to Build is an NGO that disdnguishes itself merdy by the way it organizes itself and helps people buld their own houses. It started off in the mid- 1970s with a project in the Dusmarinas Resettlement Project, a 2344Uecte site 34 kilomet outside Manila, where approximately 4,000 evicted squatts had beea relocated since 1974. Freedom to Build ran a buildig supply store and provided setters with other assistance they needed to build their homes. 86 Yue-Masz Yeung Hong Kong's boat dwellers, who were upgraded to land squatters in 1982, owed much to the efforts of the Society for Community Organization, which began in 1970 as an informal group. A society project initiated in 1971 to help resettle boat dwellers became a public housing issue after the plight of Yaumatei boat dwellers gained public attention in the territory. Finally, Kampung Sawah was one of two slums affected in 1982 by the construction of a highway in West Jakarta. Residents received assistance from the Panca Bati, a Jakarta-based NGO, in their relocation to Cikumpa Depok and Sidomukti Depok, where they planned to build a new environment, and to establish a cooperative once the families were resettled. Although the relocated settlement in Depok would mean increased transport costs for some families, it was an opportunity to start life anew with proper land titles that would have been beyond the capability of many households without the assistance of the NGO. A distinctive feature of all these projects is the organization and management expertise provided by NGOs, which have been successful in rallying inhabitants to a common goal. To attain these goals requires consummate skdll, relevant experience, and expert knowledge about how to deal with the bureaucracy, all of which these inabitants lack. The NGOs not only provided the initial spark of interest in crystallizing a critical problem, but also saw the project to its completion. Essendally, the people themselves did the work, but the NGO provided a vital institutional framework in which to reach a comnmon goal. Recent experience speaks well for this approach in Asia. Lessons lAnd Several general lessons emerge from past experience with delivering urban services in Asia. IstitonalArrangements As the demands for basic services become more complex and pressing, with more people and land area to serve, city governments have experimented with a range of institutional arrangements in their endeavor to improve delivery. Cheema (1988) has identified three phases of institutional evolution. At first, most Asian cities atempted to design, finance, and manage infrastructre services sectoradly. Large-scale projects in housing, water, sanitation and the like were the responsibility of sectoral authorities independent of municipal governments. They succeeded in expanding urban infructure and services, but because short-tenm sectoral planning was implemented through semiautonomous government agencies, little attention was paid to the possible long-term impact of government intervention in service provision. Another disadvantage of these projects was that they expanded the role of appointed officials in the management of the city. With the increasing sophistication and development of urban services, the authorities in question were unable to make opfimum use of their resources or coordinate their activities with other agencies. As a result, these programs did not reach the urban poor. Metropolitan-wide authorities came into vogue in the 1960s and 1970s when the limitations of sectoral responses became clear. The Calcutta Metropolitan Planning Organization (1961) and the subsequent Calcutta Metropolitan Development Authority, the Karachi Development Authonty (1962), the Bombay Regional Development Authority, and the Metro- Manila Commission (1971) were all established with the objective of formulating, financing, coordinating, and supervising the implementation of multisectoral projects through sectoral and local authorides. Initially, these authorities were only involved in coordinating and supervising program activities, but gradually they asmed the fimctions of sectoral agencies and local The Urban Poor and Urban Bsic Infrairructure Service n Asia 87 governments themselves, under the prodding of international and donor agencies, in the interest of speedy implementation of foreign-funded projects. As a result, these command-type authorities led to the furiher centralization of powers. Multisectoral plannwig and programming functions were also relatively neglected. The third type of institutional arrangement centered on the creation of metropolitan-wide government with special provincial powers and status. The Bangkok Metropolitan Administration (1972) and the Special Capital Territory of Jakarta (1964) are exanples of this genre, which had the power to plan and coordinate activities in contiguous areas. For Asian cities, six types of metropolitan management structures may be distinguished: centrally controlled (Shanghai, Beijing), special province (Bangkok, Jakarta), two-tier systen (Manila, Tokyo), development authority (Delhi, Bombay, Karachi, Colombo), single-tier city/metropolitan government (Kuala lumpur, Surabaya), and intermunicipality cooperation (Calcutta). Failure of Top-down Planng At great pain and cost, most Asian municipalities have come to recognize that a centrist, top-down approach to service delivery, without support from other actors, is doomed to failure. The futility of the centralized, government delivery orientation was manifested in Seoul's attempt in 1973 to resettle 76,650 squatter families to satellite Sungam New Town. Seoul had no more success in enhancing the welfare of slum dwellers through urban renewal. As much as 43 percent of the new houses were occupied by people not originally in the renewed area, as the target households could not afford the minimum purchase price. Both projects failed because they had failed to focus on urban poverty, or even on the proper management of services to the poor, but were simply concemed with eradicating illegal housing units. The projects were designed to satisfy aesthetic considerations and physical planning standards and gave lower priority to the service needs of the disadvantaged groups (Whang 1985). Needfor Commrnity Paricpation Many of the projects concerned with improving urban services to the poor have demonstrated the critical role played by beneficiaries themselves in any such endeavor. Locally led and energized initiatives are particularly essential in the initial phase when priorities are set, as well as at the implementation stage. Community participation may be affected through official support (such as the community development councils in Colombo), the efforts of an NGO (such as the Panca Bakti, which helped in fte relocation of Kampung Sawah), or by the inhabitants themselves (as m Urban Saemul Undong in Seoul). Effective project implementation-which encompasses mobilization, replication, and self-reliance-cannot take plane unless positive responses are elicited from the affected populabon. Note that while the tradition of commuity participation is strong in ural Asia in satisfying the basic needs of the rural population, urban dwellers have recenty developed people-based m anisms to address the problems in infrastructure provision by the govenment. In Indonesia and Malaysia, the age-old principle of gotong-royaong (self-help) has been extended to urban kampungs. Likewise, the Saemaul Undong and Sarvodaya movements in Korea and Sri Lanka, respectively, have found the cities in those countries ferile gromund for improving basic liig conditions. 88 Yae-Man Yang Toward Greater Integration Urban infrastructure services to the poor may be improved through integration in two ways. Although sectoral considerations are important in some circumstances and in certain low- income communities, many municipalities have chosen to design integrated infrastructure packages that would benefit target groups in multiple ways. Most of the delivery approaches reviewed earlier in this chapter have their roots in this philosophy. At the sme time, development planners and decisionmakers have come to recognize that basic urban services can only be improved if planning, implementation, and monitoring structures among actors and across different leveLs are better coordinated, streamlined, and, to a degree, inegrated. Integration does not necessarily mean greater control from central authorities, however. Theoretically, it is quite feasible for ideas, plans, and projects to geminate at the grass roots and to spread upward, but they do have to be compatible with development plans at higher levels. Thus there is a need for greater integration of plans and implementation between levels and among sectors. Tecnical Assistance as Partners- Many of the larger integrated projects in Asia have depended on intemnational agencies and donors for assistance with their various activities, from conceptual design, funding, and training to implementation. Outside technical assistance is obviously vital in countries that have not achieved rapid inustrial and economic progress. The United Nations system of agencies that evolved in the 1960s was among the early ones to respond to the newly identified needs of developing countries. Much of the assisance was in the form of technical assistance (with some ramining), pre-investment studies, and capital assistance. Ofien, authorities in the aided countries prefrred high-cost urban technologies under the mistaken impression that a high level of external capital assistance would be available. Unfortunately, competition for capital assistance among these countries inhibited the development of a less expie infrastructure, notably in sanitation, housing, and trnsport. Since the early 1970s, many of the urban assistance programs by the intemational aid agencies have focused more on alleviating urban poverty. They have developed their own style and emphasis in attpting to improve the material welfare and livg conditions of the urban poor in developing countries. Each in its own way-the development bans (notably IBRD and the Asian Development Bank), USAID, UNDP/UNCHS, ADAB. UNICEF, ELO, IDRC, and so forth-have provided much-welcomed and well-intentioned assistance and have worked in partnrship with Asian counties to alleviate the plight of the urban poor. Sirengthening Local Governnts Dependence on foreign assistance must proceed hand in hand with measures to sctngthe local governmns in the long-rn interest of the countries themselves. Several Asian countries have recently made some progress in this respect (Wegelin 1989). Between 1974 and 1979 lIdia's central goverment made conditional soft loans available to state governments under the ntegrated Urban Development Program. The loan provision took the form of matching funds for a specified investment package with die balance coming from the state and local goverments. The program, designed for cities with a populon of 300,000 and above, was suppled by the Integrted Developmen of Small and Medim Towns (IDSMT) in 1979, in an attempt to extend the facility to other cities. In Indonesia, a nationwide approach toward delivering urban services, called Integrated Urban Infrastucture Development Program lUmP), was mounted in The Urban Poor and Urban Basic Irfrastncure Serices in Asia 89 1985, in which the responsibility for planning, programming, and implementation of urban services was placed at the local level, and various components of infrructure provision were integrated. The program has been viewed as a bold departure from fte hitherto top-down approach in falvor of bottom-up planning for 300 local governments in twenty-seven provinces. In the Philippines, the Program for Essential Municipal Infratructure, Utilities, Maintenance, and Engineering Development (PREMIUMED) was launched in 1984, to cover sixteen regional centers outside Metro Manila, as a strategy to strengthen local govermnents. Likewise, the Sri Lankan government embarked on an ambitious program in 1985 to improve municipal management in its fifty-one urban local authorities within the policy of decentralization and strengthening local governments. Also, Thailand's Regional Cities Development Program launched in 1985 in four major municipalities has recently been extended to several additional urban centers, with the implementation of urban infastructure as a high priority. Recent programs clearly indicate a growing realization of the positive economic functions cities can perform through large-scale investments in infrastructure, urban services, and shelter improvement. These new functions can be financed out of revenues generated in the cities themselves, provided local urban management/institutional capabilities are drastically improved. An assumption underlying all these programs is that cities must become less dependent on national fiscal resources and must try to generate more local revenues to finance the development of more cost-effective urban infrastucture/services. All programs rely heavily on technical assistance, including taining. Also, the programs attempt to break out of a traditionally "top- down" policy and planning environment, by strengtening the "bottom-up" approach to urban services delivery, within certain rules, implying a changing balance of power and access to funds between different levels of government. These are innovative strategies for improvement, but it is still too early judge their efficacy. String for a Consensus Only a low level of consensus exists at present concerning the goals and methodology suitable for improving infastructure in low-income communities in Asian cities (Angel 1983). The different perceptions of householders, environmental engineers, communty builders, politicians, intenational funders, and the people themselves are grounded in divergent self- interests and preferences. It would be unrealistic to expect these varied standpoints to suddenly converge, but continued dialogue and articulation of any party's views would help narrow the differences and would be in the interest of the betterment of basic services for the urban poor. Seeking Temurial Security The participats in infraucture upgrading in Asian cities differ most sharply on the question of the extnt to which squatters should be recognized with respect to the illegal land they occupy. Land is a commodity to which the urban poor have no access. Many municipal govemments have been loath to improve infrsucre services in low-income settlements with uncertain teimial status, as any such effort might be interpreted as tacit approval of "squatter rights" and give the residents a strong claim to security of tenure. In any event, it is almost universally true dtat, where tenurial rights are regularized, considerable improvements have been made in the living environment, especially individual shelters, as demonstrated in the Tondo foreshore project in Manila since 1974. A 1981 evaluahion showed that, about 97.5 percent of the households had improved their dwellings, with 12.5 percent building enfirely new stauctres. 90 Yae-Man Yeung Also, in Bhopal, India, slum clearance and improvement programs have been implemented since 1957, but it was not until 1984 that legal tenure was granted to households in existing squatter settlements. Shelter consolidation has been in progress from the beginning. Thus it appears that it is the improvement in perceived tenurial security rather fian the legality of tenure itself, that can induce infrastructure improvement (Mitra 1988). Affordability and Cost Recovery Affordability and cost recovery are two closely related aspects of urban service provision that are linked to the questions of access by the poor, subsidization, and replicability. The ability of governments to extend urban services to a large number of people is contingent upon measures of cost recovery through community contnbution, user charges, taxation, and the like. The interplay of these factors is connected with the larger issues of urbanization policy, public expenditure, building standards, income levels, and the ability or willingness of different income groups to pay. In World Bank-assisted shelter projects it was discovered that many families were willing to spend up to 40 percent of their income, not 25 peicent as originally estimated, with substantial income transfer from extended families to help morthly payments. Many families perceived improved housing as a form of investment. This accounted for a 10 percent increase in families offering space for rent in Tondo in 1981. Another example of an affordable and cost recoverable urban service is low-cost sanitation improvement project in ndia. Since the latrine had been used in Idia for more than fbrty years, the goverment decided to provide low-cost pour-flush latrines on a large scale in urban areas of less than 100,000 inhabitants each. Since 1979, household sanitation has greatly improved in towns in five states: Gujarat, Tamil Nadu, Kerala, Madhya Pradesh, and Uttar Pradesh. The low-cost option has been propagated as an attractive investment of approximately $150 to $200 to households, as distinct from the conventional piped sewerage system viewed as a public works project (Urbn Edge, December 1987). Unrealistiay High Standas There is an inherent conflict between attempts to maintain high physical standards and the low incomes of urban populations. When projects are designed at stadards beyond the reach of the urban poor, they are unable to pay the costs of operafing, maintaining, and servicing the debt on the new infastructure services. Cases abound in which settlement upgradig programs in Asian cities have directly caused the target group to move out simply because they could not afford the costs of new facilities. The original settlers failed to benefit from upgrading policies in government housing programs in Calcutta (Pugh 1989), in the KIP in Indonesia (Soegijoko 1985), in an urban renewal project in Seoul (Whang 1985), and in the Tondo foreshore project in Manila (Laquian 1983). Many residents have had to sell their priority assignment or have had difficulty meeting the costs of livimng in an improved environment. The poorest of the poor have thus been eluded in the i ement process. As a result, such projects give rise to increased subsidies, distort market factors, and fail to mobilize fully the potential of urban centers. Linn (1983:153) has summed the problem up in this way: "Experience has shown over and over that where public service standards are set at levels unrealistically high in relation to the poor beneficiaries' ability and willingness to pay, most of the intended recipients of the services have generally not received any service at all." Subway construion in Calcutta provides a good example. Between 1972 and 1978, transport investments in Calcutta reached $50 million a year, The Urban Poor and Urban Basic If2shwtrure Serces in Asia 91 which represented 48 percent of the government's budget for infrastructure investmnents. The initial projected investment has to be multiplied by a factor of ten to $1 billion if the project is to be completed, because oi the increase in costs in the interin. The system also will need at least $1 million each year to operate. Heavy investments and subsidies notwithstanding, the urban poor-the city's single largest group- cannot afford to ride the subway (Brown and Jacobson 1987:42). Investing a Year in Leadership ad Training One of the worst bottlenecks in development administration in Asian cities is caused by the shortage of tmined personnel. One city of four million inhabitants reports that it does not have one qualified accountant in the entire municipal administration. The staff for planning there totals fifty and ranges from senior officers to draftsmen; perhaps five are professionals with adequate training and leadership qualities (PADCO 1976:13). It has repeatedly been shown that strong leadership and successful service delivery are positively related. Leadersbip may be formal or informal, elected or installed, but leaders with dynamism, adequate traning, and good communication skills are hard to find everywhere. Leaders must be trained, nurtred, and identified. In view of their shortage in Asian cities for the task of improving infrastucture services, it is imperative that investment in human resources be increased. In the long nm, its contribution to poverty alleviation will be indiect but critical in the realization of policy goals. Recognizing Women's Contribuiaons Since urban poverty is not gender-specific, women constitute one of the most vulnerable groups, along with children, in low-income communities. Programs designed by UNICEF and other bodies have recognized the problem and have begun to pay more attention to women's specific economic and household roles in these communities. Women have mz4e many contnbutions to the development of low-income communities. For example, female wardens are especially effective in mobilizing women's participation in preventive and cuatve health progms in the environmental health and commnity development project in the slums and shanties of Colombo. Activities organized by housewives in urban lampungs in Indonesia are more successful and better attended. Such favorable perceptions of the woman's role in the lives of the urban poor have prompted expanded trainng opporunities for women. One such example is a participatory training progrm for women pavement dwellers in Calcutta, where women learn to resolve their problems iD a noncrisis situation. Women also play a large role in the Self- Employed Women's Association in Ahmedabad, which is a trade union of poor women workers, many of them illiterate, that helps members stabilize and increase their incomes (a bank was formed), develop their skills, and enbance social interaction. Recognition of the positive roles of women in economic affairs and family welfare is reflected in the planned activities of the Orangi Pilot Project in Karachi. Apart from the more visible and by far the most successful low-cost sanitation program, the project design does include women's health education and a women's work center program. The progress to date in these two components relating to women is uneven and slower an anticipated. However, the potential is certainly present, as many women have expressed interest L woridng at the centers. 92 Yu-Ma YTwig Key Policy su and Alteatives A number of policy issues and options need to be considered by the cities of Asia. Fnancdng Urban Serices The foremost problem most Asian cities face in tyig to provide infrastucture services to their inhabitanmts is that they do not have the hinds to finance them. Per capita spending by local government tends to be higher in larger cities, but they have heavier responsibilities than smaller cities in providing a long list of services. The responsibilities of local, provincial, and central authorities also overlap gready,and it is not uncommon to find the same service provided by them all. And because poverty is so prevalent, it is difficult to raise the necessary revenues at every level of government. As a result, large sections of Asian cities have no service or are underserviced, and it is the urban poor who suffer the bnmt of these deficiencies. The revenue structure of Asian cities may be divided into two components: local revenue, in the form of locay raised taxces (notably property taxes, except in Jakarta), user charges, and locally raised revenue, such as license fees, penalties, and stamp duties; and extemal sources of local finance, through borrowing and grants from provincial or cenral governments, including shared taxes. The bulk of urban finances in Asian cities is derived from local sources, up to 70 to 80 percent. Within the local sources, Karachi, Jakarta, Manila, and Madras have relied beavily on local taxes, whereas Ahmedabad, Bombay, and Seoul have exhibited a more balanced distribution between local taxes and self-financing service revenues Linn 1981). Cities may also raise loans in the intenational capital markets, although, it appes that cities in India and Pakistan are permitted more fle:ibility m this than cities in Korea. Given the grave urban service problems in Asian cities, there is much scope for expanding the degree of grant and share-tax financing with higher-level governments. Indeed, this is the approach taken by Bangkok, which has an exceptionally narrow local fiscal base, with only four taxes-a house and land tax, land development tax, signboard tax, and animal slaughter tax-coming under the Bangkok Metropolitan Administration. Its fiscal powers are not much stronger, since prcperty, land, and cs are grossly undertaxed. In 1981, the average revenue of the city government was 741 baht per pesn, which was one-hird the national average. Consequeny, the cental govemment every year contibutes the lion's share of Bangkok's fiscal expenses. Financing urban services is tbus basically an intergo Iernmental proess in which inteational agencies have a vital role to play, where domestic financial resources arc unable to match needs. Cost Recovery versus Subsion The policy dilemma for many governments is whether to opt for cost recovery or subsidzation Cost recovery refers not only to the measures used to finance initial capital outlays, but also to cover long-run maintnae and opeting expenditures. The higher the cost recovery, the lower the drain on the public coffers. The fiscal devices used to recover costs mclude land pricing, taxation on land and buildings, user dharges, and tax and noutax revenues. If urban services are provided on a strictly cost-recoverable basis, many urban poor would not be able to afford them. Subsidies, if adopted as fulfilling a redistbutive goal or poverty alleviaton objective, can be justified. Liwise, subsidies for refuse disposal are often required because of extenalities of pollution and the difficulty of controlling unauthorized dumping. Selective subsidizaon aes to be a prudent policy choice which is prefeable to indiscnmiate or heavy The Urban Poor and Urban Basic batnhucmnre Saices in Asia 93 subsidizaion, as both can produce inefficiency i service provision. The subsidized but inefficient bus systems in some Asian cities mentioned earlier are a somber reminder of the merit of keeping the provision of certain urban services competitive and cost-ffctive between the public and the private sectors. The experience to date in developing countries suggests that the donor commuity should avoid general subsidies in the design and imlem ion of urban development projects. Shiafing Polices Three broad types of policies may be used in providing basic services to the urban poor. One type is based on laissez-faire or market mechanisms and ignores low-income settlements in the belief that they are an aberration in the development process and, will disappear with economic development. The second type is the restrictive policy that aims at reducing the size of these settlements or, where feasible, relocating them to the urban periphery. These two types of policies have by and large been discredited, leaving the dtird and most widely adopted type, supportive policy, which is designed to improve and expand on present mfnstu conditions in cooperation with local residents (Rondinelli and Cheema 1988). The World Bank's successful expeience m employing market mecnms in its projects to reach the urban poor should also be mentioned. Many Bank projects combine laissez-faire and interventionist principles, with an effective reach of target groups. Cetraiaton versus Decealizadon At the core of these policy choices are concerns about resource mobilization, financial capability, and management efficiency at the local leve. Most Asian countries have long favored a high degree of political and financial cn , to the detiment of municipal auhorities, which are starved of the requisite resources, power, and political will to delver basic urban services. To be sure, the magnitude of the urban problem is beyond the financial and administaive capabiities of many Asian nations. A 1977 study of twenty-five Asian countries suggested that, even if their total saving were mobilized for the provision of urban housing and infrasuctur for additional migrants, they wold not be enough. For most countries in the region, though, several decdes of development experience have convinced politicians and planners that declization of authority, resources, and responsibility to local govermnts promises to be a wise course for development administration to follow. The rece strategies designed to strengthen local governments in five Asian countries outlined in the previous section have built-in performance critria at the local level to ensure the desired results. In a move to improve mumicipal managememt in its urb local authorities, for example, Sri Lanka has begun issung government pants to those municipalities that have scored high on selected performance improvement indicators. In other words, those local govermen are being strengthened that are showing inceased public acuntability and responsiveness. Admittedly, it is far harder to implement decentralization objectives than might be imagined. Equity versus Effidenty Goals Many ar_gm about urban service provision make reference to a tradeoff between equity and efficiency goals. Evidence suggests, however, that these two objecives are acually compatible. Indeed, the rates of retum on poverty-oriented projects supported by assistance agencies l the World Bank have not been significntly different from the rtuns on untargeted 94 Tue-Ma Yomwg ones. Moreover, it has been argued that these goals can best be achieved by avoiding increased subsidies to urban services through reliance on local taxes and user charges (World Bank 1988:119-20). On a larger plane, the equity issue impinges on the urban poor in relation to other higher-income groups in the city. Biases against the disadvantaged groups in infrastructure provision, whether they be pricing policy, attitudinal orientadon, or strucural irresponsiveness, must be eliminated. Among the urban poor, too, equity is still an issue in projects having a differential impact on strata within the target population. Shelter projects are a prime exmple. Upgrading is less effective in fostering greater equity in that it provides benefits to households with relatively higher incomes in the affected settlement. Without a sharp targeting mechanism, the very poorest are unable to benefit from the project and equity goals will remain unmet. By contrast, sites-and-services and core housing projects can be designed to achieve greater equity when affordability criteria can be enforced. Even then, such projects have yet to reach the poorest 20 percent of the urban population. Modalities of Service Delivery The debate on the effectiveness of different modalities of urban service delivery actlly has to do with the broader issues of resource mobilization and the extent of decentralization of power to local governments. Many infrastructure services are still based on a "service delivery" model, that is, govermnent-provided services. However,in the present climate of increasing emphasis on beneficiary participation, many Asian cities have turned to other approaches, with support from interntional orgaions, semi-public or private agencies, NGOs, community self- help, and so on. Generally speaking, participatory, self-help, and community programs are conceived as "bottom-up" approaches. Clearly, no one mode of service delivery can meet every need. Each has a useful role to play, and a responsi'ble and responsive local government should bend over backward to give support to local efforts in planning, setting norms and priorities, and evaluating projects. Local groups are often poorly organized and have no resources and thus need all the encouragement, support, and funding that governments and other bodies may be able to provide. In the end, they all help to make the urban habitat a bettr place in which to live. Public versus Private Service Provision Economies of scale, extermalities, and the possibility of monopoly conditions under private sector provision are often used to justify public provision of utility services in urban areas. The present regulatory framework tends to protect public agencies with established monopolies. For some urban services, however, such protection is not well placed and results in inefficiency. Transport service is a case in point, as illusatd by the experiences of CalcLtta and Bangkok. In both cities, public bus companies covered about 60 and 72 percent of their total costs with their revenues and incurred subsidies to the tune of $8 million in 1976 in Calcutta and $17.6 million in 1978 in Bangkok. Both cities are also served by private buses charging the samne fare and carrying 75 percent and 40-50 per cent of bus riders in Calcutta and Bangkok, respectvely. Private operators are plainly more efficient. By operatn 50-60 percent less expensively dtn public buses they contnbute substantial revenues to pubLic coffers. The disparity in efficiency is so obvious that it has been suggested that all nationalized bus monopolies be open to private operators of either buses or minibuses (Feibel and Walters 1980). in low-cost housing, too, the private sector has played an important role, notably in Malaysia. Official recognition of the private sector in providing housing for the poor is again reflected in the planned constuction by T7he Urban Poor and Urban Basic I*astnwhcre Services in Asia 95 the private sector of 374,100 low-cost dwelling units in the Fifth Malaysia plan (1986-90). The Malaysian example illustates the wide scope that exists for closer cooperation between the public and private sectors in delivering urban services in Asia. Supply of and Demandfor Urban Services Much of the infrastructure provided by public or quasi-public bodies is supply-led and bears little direct relation to the nature and size of demand by the urban poor. One of the more effective ways of increasing the access of the poor to urban services is, in fact, to promote employment generation activities. With higher incomes, the disadvantaged population would create a greater effective demand for basic services, which could be satisfied through a variety of mechanisms. In many Asian cities, the prevailing high levels of unemployment and underemployment and low incomes among the urban poor are translated into their inability to pay for needed urban services directly through user charges or other methods, or indirectly through taxes. Projects designed to improve the livelihood of the urban poor may be embedded in integrated infrastructure approaches that have been reviewed previously, or may be income geneation-specific, such as the Program for Investment in the Small Capital Enteprises Sector (PISCES) supported by the U.S. Agency for International Development (USAID). The demand for specific urban services may also be met by financing programs such as the cornmity mortgage program in the Philippines, which is an innovative concept of low-income home financg tiat enables several beneficianes to acquire an undivided tract of land through conmunity ownership. The landless urban poor can thus secure land for house construction, though the financing program is administered by the National Home Mortgage Finance Corporation. It also administs the Unified Home Lending Program as part of the national objective of providing social housing to the lower 30 percent of the population through sites and services and resettlement projects. Fonnal versus Informal Sector Roles Asian cities, like those in other parts of the developing world, are creations of planners and politicians who cater to the interests of the rich and powerful, first and foremost. Cities for the poor do not exist in the minds of planners. As a consequence, the regulatory framework, resource allocation, and political support are everywhere in favor of the formal sector. However, the extent to which informal sector provision can directly benefit the lives of the urban poor should not be neglected. In a nmber of urban services, such as housing improvement, transport, water supply and public securty, informal sector contributions have kept costs low and have been effective. For example, the informal market that exists in many squater settlements,facilitates mutual help, the procurement of building materials, and the dissemination of information and know-how. Strong social networks in these commuities are often sufficient to deter crime and, where necessary, a mutual aid system of vigilant corps can be devdoped an the basis of voluntary contribution of labor. Despite the rhetoric for admnisrative and financial decentralization and devolution, excessive regulations and restrictions still impede the acivities of the informal sector. A case in point is Tamil Nadu, Iodia, where it is illegal for the private sector to attempt to duplicate the successful sites-and-services projects supported by the World Bank in the state. It has simply forced the private sector to operate lunderground. 96 ne-MaD YTig High vns Appropriate Technologies The choice of technology to be used in providing urban service depends on the fonnal and informal sector roles, and the issues of affordability, cost recovery, and replicability. The Calcutta subway project served to underline the absurdity of pursuig a high-tech option without regard for the needs of the urban poor. The challenge here is for Asian cities to devise low-cost, appropriate technologies that are not only innovative and indigenous but can also be afforded by most of the disadvantaged population groups. Large versus Seconday Cities Given the increasing concentration of population in Asian cities, particularly in the large centers, and the attendant pressure on irastructure services, economically conscious scholars and administrators have raised the question of the relative costs of service provision in cities of different sizes. Of course, urban agglomeration goes with economies of scale, but, beyond a certain city size, the gain in cost provision for some urban service disappears. Evidence tends to indicate that urban infrastruca services can be econonically provided in cities as small as 100,000 in population and the per capita costs in providing many urban services are lower in Intermediate-sized cities than in the largest metopolitan areas. At what size do diseconomies of scale begin to set in? The question of urban efficiency in relation to city size is an age-old and compIxc one. The prospect of a constellation of megacities emerging in Asia has revived the debate and spurred thinking on national urbanization policy and the roles large and secondary cities can play in the continued quest for better infrastucture services for the populace. Emg9g Chbaeng This study has attempted to demonstrate the diversity of approaches to delivering basic infrastructure services in Asian cities and the limited access of the urban poor to these services within the present administrative and planning famework. If ways are to be found to meet the emerging challenges, answers must be found to the following questions: Wha problems do Asian cties face in proviing basic ifrastrncture servces? Asian cities are facing deteriorating levels of basic infrastucture service provision because supply has failed to catch up with demand. Given the trends in Asian urbanization, with growth rates remaing high and the persistce and mcrease of urban poverty, the challenge to planers and policymakers is to search for innovative ways of delivering at least the present level of services to larger numbers of urban dwellers within the present financial and administrative coaints. Many problems of the psent system of delivery have been identified and these obstacles need to be removed. For example, standards of service are rarealistically high, technologies cannot be afforded by the inabitants, the legal famework is preudiced agt the interests of the urban poor, there is administrative and organizational inefficiency in the delivering ageni, weak local (city) governments do not have the poltical clout and financial ability to carry out effective programs, channels for ariculaing needs are feeble, and so on. It is clear that many city govmuents in Asia are in an uphiU battle to maintain their present inadequate levels of service provision, let alone upgrade them.. The Urban Poor and Urban Basic Iftzsnutre Servces in Asia 97 What should be the scope of urban baske infrastructure services? This chapter has adopted a rather orthodox and conventional definition of infrastrucure services, including such services as housing, transport, water supply, and waste disposal. These are surely key services needed to sustain the life and the well-being of individuals and families. It has been argued, however, that if one were to take a broader basic needs approach, other services-such as food, energy, and land-could be taken into consideration in certain circumstances. Of course, some of these additional services are more land-consuming and thus are more difficult to provide for in view of escalating and competing demands on urban land in Asian cities. Perhaps the point to emphasize. is that the basic urban service provided by the authorities must be one that is reaUy desired and affordable by the inhabitants. In this case, authorities should try to be particularly sensitive to the varied demands of different income groups. Who pays for basic urban services? Financing urban sevices in Asia is one of the most chalenging tasks for administratrs and planners and one that appears most difficult to undertake becas of institutional rigiditY, bureaucratic inertia, and ingrained biases. Many Asian cities have narrow and limited revenue bases from which to finance the needed services for their populace. Jakarta and Bangkok, for example, depend heavily on their cental government for transfer payments to foot their nmnicipal bills. Indian cities are so financially strapped that even routine maintenance of essential urban services is reportedly being neglreted. The situation in slums and similar settlements has been described as appalling. If the city government cannot raise enough revenue from its inhabitants, who are themselves by and large poor, there is every reason to question the general efficacy of the govemment-provided, or "service delivery" model of service provision, at least insofar as the urban poor are conced. Altenaives to the conventional model of delivery must be explored, with all shades of self-help, participatory, community modes of delivery that are known to be more cost-effective and to be capable of meeting the needs of the popuation groups concened. Even through the established methods of service delivery, appropriate technology, realistic standards, and improved user charges could go some distance in lightening the financial burden of the authorities to deliver basic urban services. How can nongovernmental organizaions (VGOs) help in imroving urban service delivery? Consideing the city government's inability to provide the necessary basic infstructure service to its inhabitants, it is almost incumbent on nongovenmental organizations of every description to assist in mitigating the plight of vulnerable population grups, particular that of the urban poor living in slums and squatter settlements. There are also coMMunitY-based organizations (CBOs) that aim primarily at improving their mmbers' situation. Because both NGOs and CBOs are close to the grass roots, thy are more effectve than governmenteal organiztions in identifying real service needs and have more innovave, flexible, and often cost- effective ways of meeting them. The Building Together project in Bangkok, the Orangi Pilot project in Karachi, and Freedom to Build in Manila are examples of volunhtary and Private initiatives to improve housing, sanitation, and en conditions of the inhabitants in 98 Yue-Man Yewag question. In the same vein are the increasingly recognized economic and household roles that women can play in uplifting the well-being of low-income communities. Women's contributions have been particularly acknowledged in the environmental health and community development project in slums and shanties in Colombo. Who eLse can help in improving urban basic infrastructure serices ? Higher levels of government and international organizations can greatly assist the strapped city governments. To begin with, the state government can decentralize responsibility, authority, and resources to the city government to expand service provision. In addition, national governments can redesign the contributions of central and local governments in delivering urban services, demonstrate renewed political will to alleviate urban poverty and basic service deficiency through budgetary allocation and institutional restructuing, and increase spending on human resources development for urban service delivery. Regional Institutions can likewise promote the exchange of ideas, visits, experience, and informationthrougha program of training, conferences, and research. At the same fime, they can use pilot schemes or demonstration projects to disseminate new insights and technologies in building more humane communities. Finally, international nizions can assist in policy transformation to improve urban service delivery in conjunction with national goverments, refocus efforts and resources on poverty alleviation, and broaden the range of activities in support of poverty goals. What is the best institutonal and management arrangement to improve urban service delivery? Although the authorities have been ineffective in providing infrastructure services in Asian cities while commnmity-based groups have helped urban dwellers to help themselves, the two groups must not allow themselves to work at cross purposes or independently. Clearly, neither group can be entirely successful on its own and the emphasis should be on mutual understanding, coopration, and assistance. The Environmental Health and Community Development Project is an exmple of such cooperation. In this case, UNICEF, the Colombo city government, and people themselves worked well together in aproject to upgrade urban amenities. Another example of cooperaton between government and private organizations in improving social services is that achieved in Bongchnn dong, a large squatter area in Seoul, referred to earlier. It must be remembered that NGOs by themselves have poor resources and are fragmented in their orientation. Ultimately, only governments have the resources and authority to create the necessary conditions for fl-scale grass roots mobilimtion. Are emnronmenal issues relevant in urban serice provision? At first glance, the link between providing urban services and the environment may not be obvious. In fact, the choice of shelter strategy, tansport technology, waste disposal methods, source of water supply, and energy choice all have vital bearing on the status of the urban environment. Of these basic urban services, energy choice and transport technology are perhaps the most significant China's heavy dependence on burning coal for cooking and heating, especially in its northern areas, and the widespread use in Asian cities of the automobile, which relies on a carbon-based fuel,both cause serious air pollution. They are directy responsible for lhe Urban Poor and Urba Basic Jfautcre Serwc in Asia 99 the production of increasig amounts of carbon dioxide, methane, nitrous oxide, and chorofluorocarbons (CFCs)-greehouse' gases-hat have led to alarming warming changes in the earth's climate and potendally disastrous and irreversible effects on the ecosystems of the world. Most cities in Asia pay little attention to such issues in their energy policy. Their planners and administrators must become more env mentally conscious in choosig technologies for the services they provide. What is the root cause for the increasing gap between the demandfor. and supply of, wur services? The extent to which the provision of urban services is considered adequate depends on the size of the population being served. In most Asian cities, the growth of urban services is betg outpaced by the growth of the population. Consequently, financial and other resources camot be marshaled fast enough to provide the necessary additional services. The root cause of the problem lies in the excessive concentration of people in the cities, as a result of which demand is rapidly outstippmg supply. In general, urban populations grow at twice the rate of the national population. The population in many low-income communities within Asian cities grows even fastr than the overall urban population. Thus it is not surprsing fiat the gap in service provision in these commuities tends to widen over time compared with other parts of the city. Social measures are obviously needed to keep the population down in urban areas, but there is also a need for a national urbanizaton policy that rationalizes the roles, functions, and population distribution of cities of different sizes.As already mentioned, the debate centers on te relative efficiency of large versus secondary cities. Has the pkmning of Asian cities caused the prvision of bsc iifrasrture services to deteriorat? It is probably unfair to blame the present sad state of wevice provision in Asian cities on planners or on the plannng process. Te present predicament can be taced to many factors.A signficant point to note, however, is that the planig of Asian cities has been conceptually and physically influenced by thinkng in developed countries. That is where many Asian planners have obtained their taining. Many exting plannig practices and procedures are based on legal norms and architectural standards more in tume with developed economies. Not uncommonly, Asian plann are imbued with visions of the modern city, in which the urban poor have no place. The failure of the urban renwal project in Seoul and of various settlment upgrading policies in Calcut, Indonesian cities, and Tondo in Manila clearly demonstate that planmers and policymakers tend to be insensitive to the needs of the urban poor. The removal of squatters from the Inrmuras and Tondo in Manila to Sapang Palay and teb constuction of the subway in Calcut are still other emples in which policy decisions have been made without taking the interests of the poor into account Some cynics would even veture to say that the Asian cities of today are for the rich; cities for the common people and the poor have yet to be designed and constuctedL 100 Yue-Man Yaeng References Angel, Shlomo. 1983. "Upgrading Slum Infrastucture: Divergent Objectives in Search of a Consensus.' Third World Planning Review 5(1):5-22. Brown, Lester R., and Jodi L. Jacobson. 1987. The Future of Urbanizaton: Facing he Ecological and Econonic Constrains. Worldwatch Paper 77. Washington, D.C. Cassim, Jeizn K. and others. 1982. "Development Councils for Participatory Urban Planning." Assignment Children 57158:157-87. Cheema, G. Shabbir. 1988. "The Extended Metropolis in Asia: Implications for Urban Minagement." Paper presented at the Conference on the Extended Metropolis in Asia, East-West Environment and Policy Institute, Honolulu. Dana, Abbijit. 1987. 'Alternative Approaches to Shelter for the Urban Poor in India.' Cities 4(1):35-42. Des, Nick. 1981. "Indonesia's Kampung sIprovement Progranm: An Evaluative Case Study. E'Bsics 48(286):19-36. Diaz, Ralph. 1982. "Restructuring Services to Reach the Urban Poor in Kuala Lumpur. Assignmt Children 57/58:135-56. FlMbel, Charles, and A. A. Walters. 1980. Ownership and Efficency in Urban Buses. World Bank Staff Working Paper 37. Washington, D.C. Laqulan, Aprodicio A. 1983. Basic Housing : Policesfor Urban Sites and Servces and Shelter in Developing Countries. Ottawa. International Developmen Research Centre. Lim, Hong Hai. 1985. "Nadi Integrated Soci Services Programme, Kuala Lumpur." Regional Development Dialogue 6(2):123-47. . 1988. 'Urban Service Provision in a Plural Society: Approaches in Malysia." In D. A. Rondineli and G. S. Cheema, eds., Urban Servces in Developing Countries. London: Macmilan. Linn, Johames F. 1981. "Urban Finnces in Developing Countries." In Roy Bahl, ed., Urba Governmnent Finnc Emerging 7rends. Beverly His, Calif.: Sage Publication. _. 1983. Ciaies in the Developing World: Polides for Ther Equitable and Effent Gro,Mh. New York: Oxford University Press. Mitra, Banasbree C. 1988. 'Impact of Tenure ulaization and Environmental Upgrding Programmes on Shelter Consideration in Squatter Settlements in Bhopal." Indian Hwnan Settements Programme, Resarch Report. 3, Delhi. The Urban Poor and Urban Basic Ifrsinv uire Servces in Asia 101 PADCO. 1976. Guidelines for Fonmudating Projects to Benejft the Urban Poor in the Developing Coutries. Vol. 1. Washigton, D.C. Pugh, Cedric. 1989. 'The World Bank and Urban Shelter in Calcutta. Clries 6(2):103-18. Rondinelli, Denis A., and G. S. Cheema, ads. 1988. Urban Services in Developing Countres : Public and Privae Roles in Urban Development, London: Macmillan. Soegijoko, Sugijanto. 1985."Managing the Delivery of Urban Services for the Poor in Indonesia: Case Study of KIP in Bandung.' Regional Developmt Dialogue 6(2):78 103. Taylor, John L., and David G.Williarns. 1982. "Upgrading of Low-Income Residential Areas in Jakarta and Manila.' In John L. Taylor and David G. Williams, eds., Urban Planing Pracdce in Developing Counwties. Oxford: Pergamon. United Nations. 1988. The Prospects of World Urbanzaton: Revised as of 1984-85. New York. Wagelin, Emie A. 1989. "New Approaches in Urban Service Delivery: A Comparison of Recent Experience in Selected Asian Countries." Discussion Paper, Centre of Urban Studies and Urban Planning, University of Hong Kong. Whang, In-Joung. 1985. "Policy Issues in Managing Urban Services for the Poor: The Case of Squatter Improvement in Seoul, Republic of Korea." Regional Development Dialogue 6(2):104-22. 1988. "Social Services Programmes for the Poor in a Newly Industrialzing Country: Experience in South Korea.' In Dennis A. Rondinelli and G. S. Cheema, ads., Urbn Services in Developing Couanries. London: Macmillan. World Bank. 1988. World Development Report New York: Oxford University Press. Yeung, Yue-man. 1985. "Provision of Urban Services in Asia: The Role of People-Based Mechanisms." Regional Development Dialogue 6(2):148-69. . 1988. "Livelihoods for the Urban Poor: Case for a Greater Role by MetropoLitan Governments." Regional Development Dialogue 9(4):40-60. Yetmg, Yue-Man, and T.G. McGee, eds. 1986. Community Panriciaion in Delivering Urban Serices in Asia. Oetawa: Iernational Development Research Centre. 6 TH ROLE OF CREDrr IN POVERTY ALLEVATION: THE AsIAN ExPERIENCE Jsmael Geubig, Jr. Summary This chaper briefly revews credit programs for the poor in Asia and shows tha most conventional credit approaches (for erample, IRDP, cooperaives, and subsidized agricultural credit) have had serious failings in terms of both repayment and outreach. The main reasons for tdisfailure are shortconungs in the design, management, and implementation of he program and a lack of attenton to the unique condition and circumstnes in which the poor live. At the same time, innovative credit prograus have emerged in various parts of Asia and elsewhere, that have adated their design, management and implementation systems to the special needs of the poor. hiese programs have shown remarkable success in reachng the real poor whUie maintainng high repayment rates, ranging from 90 to 99 percent. The chapter reviews in some detal four such programs -that have succeeded in signaficantly scaling up their credit actvites. Each ofthefourrepresentsa uneapproadztoprovling creditforthepoor, busztey also share some charcteristics in their design, management, and implementauion systems. Thze four models and the respective prograns discussed to illustrate them are (a) the intermediary type, illustrated by the Working Women's Forum of India (b) specializedprograms offormal-sector banks, illstrated by the KUPEDESprogram in Indonesia, (c) spedalized bans for the poor, illustred by Grameen Bank of Bangladesh. and (d) paocage-ype programs, illustrated by the Aga Rkhan Rural Support Program of Pafastan 44KSP). Some of the characteristics shared by most, if not all, of the successful credit programs illustrated in the chater include effective targeing; borrowers organized into solidarity groups at provide mutual supportand replace die collateral or guarantor requirements of conventional banks, fitng the credit delirvery system to the needs of the poor; properly pnced credit services (for example, interest charged at market rates); self-selected icome-generating acvites; assurance of bigger subsequent loans upon repaymnt of old loans; effective savings moblizaion,7 adequate and appropriate training and preparation of borrowers; accountabily and eective management information system, and appropriate irlementing instituton. The chapter concludes that credit is one of the most potent instruments for alleviating poverty and can be made viable, sustainable, and effective fappropnately delivered by programs tha are wel designed, managed, and inqenented. However, credirprogramsfor the alleviation of poverny have their bralt-in limits, such as the small scale of projects undertaken by poor borrowers, technology limitions, and the limited size of the market. The way to allevate and evetualy eradicate povery mnay actually be to create a conducive economicand sociopolical environmentthaspromotes equitablegrowth tIrough klabor- intensive productive employmentpolices; pro-poorfzscal, monetary, trade, exchange rate, price, and sectoral policies; and through social development ezpenrditures that enhance the human capital of thepoor by improving their education, heath, and nutridon,; and through investent in rural infrastru es, as the expenence of the newly industrializing economies has clearly demonstrated. 103 104 Ismad Geubig, Jr. Over the years, many countries in Asia have instituted various programs aimed at improving the condition of the poor, pardy in recognition of the failure of inequitable growth to pull the poor up the socioeconomic ladder, or simply to compensate for their disadvantaged position in society. The provision of credit to the poor has been a leading component of many of these programs because the lack of access to productive capital is thought to be one of the main factors preventing the poor from brealdng away from the "poverty trap." This trap, it is argued, makes it extremely difficult for the poor to overcome poverty without outside intervention. In its simplest form, this trap can be viewed in terms of the poor person's low capacity to generate income, savings, and investment in an economic environment that offers limited employment opportunities, which thus leaves the poor in perpetual poverty. It is argued that providing the poor with credit can break this barrier to economic emancipation by giving them access to resources, in particular, to income-producing or enhancing assets for the small- scale farmers and fishermen in rural areas, the small artisans, petty traders, and microentrepreneurs in urban areas. With their labor largely underemployed or employed in unremunerative work, the poor can make more productive use of their labor through credit and in the process, progressively increase their income, savings, and investments through self- employment and thus cross the poverty line over tinei Credit Progrms for the Poor Most goverments in Asia have allocated considerable amounts of resources in the form of subsidized credit for the alleviation of poverty, especially in rural areas. The rationale for subsidizing this credit is that the poor have limited capacity to repay because they have limited investment opportunities and providing them with credit services entails high transaction costs. Subsidized credit often takes the form of goverunment-sponsored rural cooperatives, agricultural credit for the small farmers, and specialized programs like the integrated rumral development programs common in South Asia. The number of poor who have crossed the poverty line as a result of these programs is difficult to ascerain. The undifferentiated impact of many of these programs (for example, in terms of value added or returns on investment) is often positive and significant. But according to many studies that have evaluated these programs, their benefits have largely accrued to the rural elites who have been attracted by the credit subsidies and used their influence over the administration of the credit delivery system to "corner" most of the credit resources at the expense of the rural poor (Islam 1935; Mukhopadhyay, 1985). Even the cooperative approach to credit, which had earlier been so successfully tried in Europe and North America in reaching the small farmers, and much later quickly and widely implemented in Asia largely through government initiatives, floundered badly with few exceptions (for example, the cooperatives in Japan, Korea, and Taiwan [China] and some NGO-initiated cooperatives elsewhere). Even the highly acclaimed Comilla cooperative model, which attined considerable success during its early years when it restricted itself to a limited geographical coverage, on the whole fadiled to reach the intended beneficiaries-the small farmers-and suffered serious repayment and viability problems when the government replicated it nationwide. In the case of some other credit programs, the failure is riot so much in reaching the target group but in the low repayment rates and the failure of the programs to sustain the significant positive impact of credit through the provision of a follow-up credit to deserving borrowers. A good example is that of India's Integrated Rual Development Programme (IDRP), T7he Role of Credit in Poverty Alleviton 105 probably the most massive credit program of its kdnd targeted exclusively at the poor. Since its inception in 1978, IRDP has channeled more than Rs 71 billion in credit and Rs 27 million in subsidies to mostly poor households. A World Bank study (Pulley 1989) showed that 97 percent of the IRDP beneficiaries were poor, many of whom were very poor to extremely poor. However, default rates in the program have been high (68 percent in 1984) and it was able to provide only I percent of its beneficiaries with a second loan. In the absence of an effective mechanism for providing follow-up credit - sustain capital requirements of their income- generating projects, the initial gains in income by the beneficiaries were lost after two to four years, and their assets gradually decapitalized to meet consumption and emergency needs. Other programs showed much promise during the early stages, but could not sustain their high level of performance, as can be seen in the Small Farmers Development Program introduced in Nepal in 1976. The SFDP is a "package type" credit program based on the guidelines developed by the Food and Agriculture Organization (PAO) whereby credit is channeled through farmers' organizations and supplemented by support services such as extension and technical assistance. Aside from Nepal, SFDP-type programs were initiated in Thailand, the Phili1ppines, and India. The SFDA program in Nepal had gained more prominence than others because of its initial success in reaching a large number of small farmers while maintining a high repayment rate (more than 90 percent). As it expanded, however, credit leakages to the nonpoor and the default rates increased. By 1989, SFDP's repayment rate was only 48 percent (Holt and Ribe 1990). Governments trying to provide poor farmers with access to credit often require conventional banks to eannark a certain percentage of their outstanding loans at subsidized interest rates for poor borrowers for whom collateral requirement is waived. Loan guarantees are normally provided for by the government to reduce the high risks borne by the banks in lending to the poor. This arrangement often fails to achieve its goals. The almost universal response of banks to such a program is either to ignore it where the progrm ilemention mechanism allows it to circumvent the regulations, or to keep its involvement to a minium, just enough to give the appearance of compliance as a gesture of support. The Differential Interest Rate (DIR) Program in India, another massive credit subsidy program for the poor, is a good example of this. While DIR has reached a relatively large number of poor borrowers, this accounted for only a small proportion of the potential borrowers it was designed to reach, and its repayment rates have been low. Reasons for the Failure of Credit Programs What went wrong with the credit programs that failed? Could they have performed better if they had incorporated the appropriate changes. Or is it that the poor are not bankable because providing credit to them entails uneconomically high transaction costs and risks? It is difficult to pinpoint the reasons for the failure of many credit programs, because is as much as a host of interrelated factors that often come into play. However, it is possible to identify some of the underlying factors that contnbuted to failure. Much of the failure of past credit programs sponsored by governments stems from shortcomgs in the design, management, and administration of the credit delivery system. One of the most serious design flaws, for example, was the lxck of a targeting mechanism, such as realistic andholding and income ceilings, so that only the trly poor could qualify for credit under the program. This permitted the entry of non- poor borrowers who were attracted by the generally low interest rates charged by these programs. With their greater socioeconomic and political power, the non-poor were able to get access to 106 Ismea Getebig, Jr. credit delivered in this manner much more easily than their poor counterparts, thus crowding out the latter. General credit programs for agriculture and rural development often faill under this trap. But even some specialized credit programs for the poor (for example, the Special Agricultural Credit Program in Bangladesh and the Differential Interest Rate Scheme of India) have also suffered because of this weaknwe (McGregor 1988; Berger 1989). In the case of many credit cooperatives that failed, a host of contributing factors acting together brought about their demise. Often cited are inadequate preparation of members, control and domination of the cooperatives by powerfil local elites who often borrowed an inordinate proportion of the loan funds, weak management, the top-down manner in which the cooperatives had been set up and run by the government, and overdependence on outside fimds rather than on local savings mobilization (Huppi and Feder 1989; Mukhopadhyay 1985). Programs ftat have attempted to tap the resources of conventional banks to provide credit to poor borrowers through some form of anrangement with the government, often in the form of directives, have generally failed because they did not provide adequate incentives for the banks to make adequate profit from their loans to the poor. The primary objectve of conventional banks is to make a profit on their loans. But providing credit to the poor is staff-intensive and entails high transaction costs and risks that must be covered by interest earnings from leJing and service fees. Since these programs often put a ceiling on the interest rates tne banks are allowed to charge, the banks are unable to make sufficient profit from this program comparable what they can obtain in alternative capital outlets. Understandably, most banks would not be wiling participants in such programs and would accordingly try to avoid it or cut dowi tvwir losses or opportunity costs by minimizing their involvement. The lack of an appropriate credit delivery mechanism that clearly specifies the procedures and requirements for personnel and support services and provides for effective enforcement mechanisms also contributes to the ineffectiveness of the credit-for-poor programs of conventional banks. For example, the commercial ban through which the Special Agricultural Credit Program in Bangladesh channeled its loans took the easy way out by abdicating its responsibility to irplement the program, letting the rual elites and politicians identify and recommend borrowers. It also inisted on a collateral requirement despite the program's directive to waive it, as also happened with coopering commercial banks in India's DIR programs (McGregor 1988; Singh and Kelles-Viitanen 1987). At the level of the bank staff assigned to implement the credit-for-poor programs, there is generally no provision for incentives to compensate for the more difficult nature of their work. Often, this work is added on or mixed with their regular work with nonpoor borrowers. As a resul;, bank officers have low motivation and morale and prefer to concentrate on the nonpoor borrowers, who are less difficult to service.This belps to explain their poor performance in servicing the poor borrowers. Any credit program that experiences susined low repaymet rates will eventually become insolvent and collapse. Repayment rates are therefore, the most important and simplest indicator of the program's viability and sustainabiity, provided the interest charged is high enough to cover cost and a reasonable margin of profit. It reflects the borrower's capacity and willingness to pay back the loan. The borrower's capacity to repay is, in tun, a funaion of the productivity of investment for which the loan is utilized. It is now well documented that the poor have the capability to use credit productively using the skills and knowledge they have and that they can d.merate rates of return comparable with, and often exceeding, those earned on much bigger investmets by the nonpoor in the formal sector. Therefore, the most important factor affecting repayment rates in a credit-for-poor The Rok of Credit in PFn ALlvaion 107 program is the wilingness of the borrower to pay back. This Li turn is affected by many factors. An important element is the incetive system built into the program that rewards the borrowers for prompt repayment of loans. One of the most important incentives to poor borrowers is the assurance of a foUow-up loan upon repayment of an outstanding loan. This incentive is ing in many programs and is one of the main reasons for the low repayment rates, as in the case of the IRDP and DIR in India and BIMAS in Indonesia. The effect of such an incentive on repayment is apparent when one considers that the poor, by virtue of their extremely weak and vulnerable economic security, need credit support to be sustained long enough to allow their assets and income to become stable. Second, very often, the credit-for-poor program is the only access to credit available to the program participants. Even their access to the village moneylender is very limited and credit often available only at exorbitant rates. A poor borrower becomes even less incined to pay if a follow-p credit facility is closed because these poverty- oriented credit programs often do not require collateral, which the borrower would stand to lose if he defaulted. As discussed later, most successful credit programs have assured borrowers of a subsequent loaa upon repayment of the previous debt. Even when a program is appropriately designed, it can still flounder over time if the rigor in implementing the procedures essential to its success (for example, targeting, financial discipline and accountability, and supervision) is relaxed over time. This is one reason for the deterioration in repayment rates of Nepal's Small Farmers Development Program as well as others. Some Success Stories In the light of these past failures, many expeiments have tumed to innovative aproaches to credit delive;y sysms (CDS) for the poor. The resultng CDSs have in large part been highly efective in achiievg the major objectives of credit- for-poor programs. Success Criteria The success of a credit program designed to alleviate poverty would have to be evahlated in the light of the program objectives. Some of these programs may incorporate a broad set of objectives in addition to credit, such as mnpowenent, "conscientization, education, and sMills training. Five criteria are often used to assess the effectiveness of credit-for-poor programs. * Exent to whih the program has read the tn'4y poor. This can be measured both i terms of the proportion of beneficiaries who are truly poor and the number of poor borrowers reached. * Loan recovery. This is measured by the repayment rate. Sustned high loan recovery represents the simplest and clearest indicator of a program's success as it reflects the productivity or profitability of the loan, as well as the borrower's satisfaction and support of the program. * Productivity of the loan. This measure the ilt of the investment into which the loan was applied by the borrower. It is normally measured in terms of retun on investment or capital-output Adio. This criterion is often difficult to measure accrately when applied to the small-scale investmes of poor borrowers because part of the income keaks into consption and thereby oduces a downward bi on the estimated returns on investm. Thus, a less precise but easy-to-measure proxy, asset mention by the borrower, is often used. e impact on borrowerIs noe. This impact refs to the extent to which the borrower's real income has ireased as a result of the credit provided. As the income base of the poor borrowers is very low, the incrase in absolute mcome arsig from very small investments (eneral!y less than US$100, and some as low as US$20) is almost univemsally small. For this reason, it may be more meagul as a measure of progress to use dte relative increase in income, in respect of the pr-credit iwombase or the poverty line to be crossed. This citerion is probably the most difficult to measure by convenional "before-after" or "with-withou" techiques. This diffiulty stems from the problem of isolaing te dfect of credit from many interveing variables (in the case of 'before-afer) and the comparability betwe the expemental and control group and the imnssibility of deermining the countfictnal (in the case of 'with-without). * Smsaiiity. Here it is useful to ditgish sustanb at the indidual borrower level in contrast to susminabiliy at the program level. The forme can be mesured in terms of project vablity as dissed earlier, wheeas the latter relates to the ability of the credit progm to be self-rliant in term of resources, ledership, and so on. Specifically, it rfers to its capacity to become istionalized into a fnancially self- sp-ing program, able to cover aUl costs and to generate sufficient profits from its opeons. Thus it is affeed by such vaiables as interest rae charged on loam, repaymen erat, an tansaction costs. Sustinabiift also refers to the ability of the ;istittion to carry on in the absence of the dynamic charismatic leaders that are ofte bhid many of these progras (for example. Gramteen Bank Aga Kban Rural Suport Program, th Self-Employed Woring Women's Asociaton, Workig Wome's Form). * Noneonoc aiteia. Pehaps, the most difficult task of all is to develop success crit for the noneomic objectives of a credt program-for ecample, the epowerment of poor ral women i sing their stams in the fiuly and the comunity, developing their self-confidence, and so on. These, however, are legitinae objectives that mst also be coasider in assessing the efeveness of programs, espdcially those targeted at poor wme borrowers. A numbe of successful credit progams for alleviatig povertY can be identfieid by applying fte above crteria. It is not possible in ftis brief chapter, however, to provide deied information on all of these criteria, for the varous sources reviewed have not applied them consiteny. Neerdheles, a mber of sues have emned many of these uovve credit progms in Asia and elsewhere and assessed their effectiveness on the basis of te criteria. Some of tese prgms, along with their essntial ch ristics, are listed m Table 6.1. Typology Mos of the programs lised in Table 6.1 coud be classified mnder th following 27w Role of Chdit in Povenry Alkaion 109 Type A. Intemeary Type B. Speciaized programs of formal-sector bans Type C. Spealized bamks for the poor Type D. "Packge" prgra.m IDnTmDRY. An Intermediary-type credit program is one that plays the role of a broker between the forma lerndig institution and the poor beneficiares who do not have access to these institutions when left on their own. There are many variations of this way of playing the brokenng role. 'Te most common is for the intermediary to simply facilitate accessing credit facilities that commercal bank provide, by government mandate, for poor borrowers (for example, SEWA and WWF in their early years). In this type of program, responsibility is often divided between the intermediary and the banks to enable them to take advantage of each other's strengths and minimize their weaknesses: the bank administers the loan disbursements and repaymen while the program assists in processing the loan, collecting repaymnts, following up on defatuters, and the like. In this way, the bank is able to minimize the transacdon costs associated with lending to poor borrowers, while the intermediary-whose strength lies in familiarity with the borrowers and grass roots organizatons-is able to handile the transactions involved much more efficiently than the ban's own persomnel could. In one variant of this model, the intmdiy mobilizes savig from the beneficiaries, which are thu deposited in a commercial bank. The commeral banks open a revolving credit account, guaranteed by the beneficianes' savings, which can then be loaned out either directly to the beneficiares or indirecdy via the intermediary administering the loan SPEAED CREO N PRO M OF PRMAL-SER lBANKS. Prams olr this cae M o"a administered directy by bas in the formal sector without the use of intermediaries. This form of CDS recogmzes that conventona commercial banking procedur are not suitable for senrcmg poor borrowers as they are designed primarily for weil-off cLent. C)S theref adjusts is equimets and procedures to make it more effective in both mobilizing the savings of poor beneficiaries and providing small amounts of credit to them The crt-for-poor part of their operation, however, constit only a small proportion of their baing operation in the fomal sector. One of the best-known examples of this is the BKK, KUPEDES, and S[MPEDF2 programs of Bank Rakyat Indonesia. SPEcMUzD BANKS FOR TEm PooR This cind of credit delivery system is stil very rare, perhaps becuse it poses the most difficult challenge of all. It goes beyond adjustments within eax g formal-sector banking institutions to rectify the inadequacy of their conventional practi to service credit needs of the poor, by setting up an entirely new bank institution that caers exclusively to the poor. It recognis that the poor have unique needs arisig from their socioeononic-political disadvantaged rpsiton in society, and ths requires an alternative credit delivery system designed specifically fbr them. Table 6.1 Selected Credit Programs Directed at the Poor in Asiit Total Number Average Repayment Percentage of Program Countty TY;pd Year of Year of of Percent Loan Size Rate Loan Funds in Reference Operation Beneficiaries Womea U$ % Arrears BKK' Indonesia Qh,-oenet 1982 10 2,700,000 60 527 -19 KUPEDES rndonesia Govemnment 1989 5 1,380,000 25 426 95 5.4 MbM Inidonesia NOO 1987 10 3,500 s0 207 80-90 - Y1S Indonesia NGO) 1988 16 5,700 -170' 99- BRAC Bangladesh NGO 1987 18 169,000 93 31 92.3- Grmenen Blank Bangladesh NG3O 1990 9 852,000 91 38 98 - PCRW Nepal Government 1989 a 6,640 100 59 90 - SEWA Inia NrGO 1982 14 40,000 100 72 90 - WWF India NGO 1989 12 50,000 100 90-95 - World Vislon Sri Lanka NGO 1987 - 700 Village - 20 98 -~ Note: The abbreviated naimes stand for the following: BICK, Badan Kredit Kecamatan; MBM, Maha Bohoga Marga; YIS, Yayasan Indonesia Soyahtera; BRAC; Bangladesh Rural Action Committee; PCRW, Production Credit for Rural Women-, SEWA, Self-Employed.Women's Association; WWF, Working Women's Forum, a. Per group of 30 members. Source: Holt and Ribe (1990) and Remienyi (1991). Tse Role of Credir in Povry Aleviaion 111 The central concer in this approach is to determine the conditions under which the poor live and the nature of their credit needs. The resulting infonnation is then used to design a credit delivery system to meet those needs, subject of course to certain norms to ensure that the institution is viable and sustainable and complies with legal requirements. Unlike the specialized credit-for-poor programs of the fornal-sector bancs, a specialized bank for the poor is completely unfettered by the inertia of conventional rules and procedures that the former has to grapple with, creating instead innovative ones that have a good fit with the needs of poor borrowers. And because it caters exclusively for poor clientele, it is able to focus all its resources and energies on just one challenge-how to make banking for the poor work. The Grameen Bank of Bangladesh is probably the only major CDS of this kind in Asia, and perhaps in the world. PACKAGE PROGRAM . This is a very common approachthat has been triedby most countries in Asia. It became popular when the so-called integrated development program was in vogue among donors and development agencies.Credit isjust one of several components in this package, which often includes any combination of the following: infrastucture development (for example, iigation and drainage canals), technology (for example, improved crop varieties and practices), training and extension, and market information and facilities (transport and processing). The track record of this approach in terms of reaching the poor and sustaining a viable credit program is rather dismal. It has a high default rate and a large proportion of the benefits have accnued to nonpoor beneficiaries. A notable exception is the Aga Khan Rural Support Program in northern Pakistm. BRAC of Bangladesh and SFDP in several countries also fall under this category. Credit-for-Poor Programs: Some Cases For lack of data and space, we shall limit our discussion to only four credit-for-poor programs, one for each of the four general types just discussed. Thlee of these are rural-based because there are more credit-for-poor programs in rural areas. The Working Women's Forum and SEWA of India are two of the beter-known examnples of urban programs with a strong credt coomponent. Credit-for-poor programs in urban areas are not as well documented as those that cater to the rural areas. Thke Intermediary: Working Women's Fonum (India) The Working Women's Forunm (WWF) began in 1977 as an experiment on financial intermediation on behalf of urban poor women who were petty traders in a slum area of Madras. The catalyst in this case was a political and social female activst who wanted to find a mechanism by which to enable poor self-employed women to take advantage of the Differentil Interest Rate (DIR) scheme of the Idian government, which mandated that the nationalized banks were to extend subsidized credit at 4 percent interest rate to the poor but had not effectively reached poor women borrowers. WWF organizes women workers into neighborhood groups and helps them apply for loans for income-generating activities from the nationalized banks. PERFoRMANCE. Since 1978, the WWF has mobilized 50,000 poor women borrowers into groups that have been able to access smafll loans averaging US$40. It has maintinled repayment rates of 90-95 percent for its credit intemdiation programs, compared with 30 percent for the nonintermediated loans of DIR for all of India. The WWF loans have had a significant positive impact on employment and income. An iadepth study of WWF borrowers in Madras has shown 112 Ismael Gtubig, Jr. that 34 percent of them have been able to initiate a new business activity as a result of the loan (Noponen 1987). Increases in income have been brought about by shifting from low-paid piece- rate work to more remunerative self-employment, or through increases in the daily earings of women traders by freeing themselves from the exploitative practices of the moneylenders and middlemen. The increase in income experienced by WWF borrowers arising from the loan varies between categories of worln more than 150 percent for women who moved up from being seasonal agricultial laborers to self-employed members of dairy cooperatives; 100 percent for agarbati workers in Bangalore and 36 percent for beedi roliers in Vellore who were enabled by the loan to bypass the middlemen. No figures were available for the entire WWF credit program but dtese bits of evidence clearly indicate that the impact on the borrower's income is significant. While the WWF-assisted poor borrowers have demonsuated the viability of their enterprises as reflected in their high repayment rate and increases in income, the long-nm viability of WWF itself as a program is less clear since it is not earning enough to cover the cost of operation. The difficulties it experienced in being a pure credit intermediary prompted the WWF to set up its own Women's Cooperative and Credit Service Society in 1981. A similar strategy was adopted by another credit intermediary for poor women, the Self-Employed Women's Association of Amedfabad, which has also set up a cooperative bank for the poor, enabling it to mobilize saving fiom its members and making a profit on its lending operation at market rates of interest (17 percent in 1989). Design of Credit Delivery System TARGETlNG. The credit indiation activities of WWF focus on the urban, self-employed poor women egaged in petty trading, small-scale home-based production, and low-paid workers and micn p , most of whom live in slum areas. By choosing to work with these groups, the WWF did not fid it necessary to scrutinize the income and asset base of its mmber- borrowers to ascertain their poverty status. Furthermore, the small size of the loans that must be paid fiequently and the membership eligibility that requires regular attendance of group meetigs, consistent loan repayment, and serving as a mutual guarantor for the loans of all group members would probably be sufficient to deter the nonpoor from participating. Therefore almost all the women beneficiaries of the WWF credit intermediation are likely to be poor. GRASS ROOTS ORGANIATION. The beneficiaries of WWF are organized into a grop of ten to twenty members. The group selects its own members but they must live in the same neighborhood. Groups are normally formed by a potential group leader who has become familiar with the WWF program and calls together a number of potential members to explain how the WWF works. Once a sufficient number have agreed to join, the group elects its leader and registers with WWF. The group convenes frequent meetings to discuss its problems and provides mutual support tO overcome them. The group leader plays a critical role, being responsible for assessing each member before recommending her to the WWF and for collecting and depositing repayments. It is through these groups that the WWF facilitates the application for and processing of loans as well as prompt repayment. This form of grass roots organization has contnbuted to the high repaynt rate of the loans through peer pressure and mutual support. LOAN CONDMTION. Loans provided by the nationalized banks under the DIR scheme are mandated to be colateral and guarantee-free at 4 percent interest. The loans are given in cash for icome-grting activities that the borrower herself selects. WWF requires its beneficiaries to repay the loan in ten monthly installments. LOAN PROCEDURE. Being just an intermediary, WWF has to conform to the standard 7Te Role of Credit in Poveny Alevidon 113 requirements of the bank for loans to poor borrowers. This means borrowers must fill out application and loan disbursement forms, which can be a very time-consuming and complex process for the poor, who are usually illiterate women borrowers on their own. The administrative intermediation of WWF is particularly valuable at this stage for it provides assistance in filling out the application form, following up on the loan application, and accompanying the beneficiary to the bank to fill in the forms required during loan disbursement. INCENm AND PENALTY SYsra. WWF assists its members in obtainimg a subsequent loans following the complete repayment of the previous one. CONIROL AND SUPERVISION. The WWF loan groups are essentally self-policing. The group leader's responsibility is to collect and deposit the monthly repayments from the members. Only in cases of default is a WWF staff member involved in ensuring compliance with the repayment obligation. As a result, the continued viability of the group depends on the leadership quality of the person heading it. Again, it is peer pressure and the familiarity of the group members with the activities of the others that form the basis of control and supervision in WWF. ANCILLARY AND SUPPORT AcTnIVES. WWF considers credit simply an entry point for organizing poor women to improve their economic condition. It believes in gradually moving into the broader social and political obstacles that working women face in India. WWF is involved in social service projects of high value to its members. These include the seting up of day care centers, night classes, sklfls training, health and family planing, and social action such as lobbying govermnent officials to make public goods and services more equitable and accessible to poor women. In 1982 it also organized the National Union of Women Workers to suppon the union-oriented work of WWF. LEADERSHP. A strong and effective leader has been instumental in the founding and growth of WWF since its inception in 1978. It is not clear whether WWF could sustain its dynamism and performance in the absence of its founder-president. Spealized Program of Formnl-Sector Bank: he KJPEDES Program in Indonesia KUPEDES is an iteresting example of how formal lending institutions could respond, with some adaptation in the design of their CDS, to the challenge of banking with the poor. It is a specialized rural credit program designed to provide credit and savings facilities to small borrowers, but not the smallest ones. It is administered by the largest national banking system in Indonesia, the People's Bank or Bank Rakyat Indonesia (BRI), with 300 branches and 2,850 village-based sub-branches spread throughout the country. It has demonstrated the tremendous potental of commercil banks with an established network of branches distibuted nationwide to make a quick and substantial impact on smaull borrowers. PERFORMANCE. Since its inception in 1984, KUPEDES has grown rapidly. By the end of 1989 it was providing US$470 million in outstanding loans to some 1.5 million borrowers, 25 percent of these being women. This group may not represent the poorest, say, 20 percent of the country because the average loan given amount to US$426. Although this is very small by commercial baiDng standards, it tends to be much bigger than the loans provided by the programs targeted at the poorest levels. A survey by Bank Rakyat Indonesia in 1989. however, showed that 70 percent of the KUPEDES borrowers owned less than half an acre of nceland, which suggests that they are likely to be reatively poor, but not the bottom poor. KUPEDES has also succeeded in maintining a high repayment rate, over 95 percent i 1989, and this has contributed to its viability as a program Indeed, it started to make a profit within eighteen months of operation and since then bas consistey made a healthy and growing 114 Lmad Getubig, Jr. profit every year. Therefore, KUPEDES is a finanialy sustinable credit program with the potential to reach a large nmber of small borrowers that other credit-for-poor programs are not likely to cover. SuccEss FAcroRs. A number of factors have contnbuted to the impressive performance of KUPEDES. Most of these relate to design, management, and implementation aspects of the program. It's success has been attributed to the following factors (Biggs, Snodgrass, and Srivastava 1991): 1. A strong instituional foundation. The large mnmber of well established branches and village units of BRI through which KUPEDES has opexated enabled the program to spread rapidly once it gained confidence in its CDS. Overhead costs have also been kept to a minimum as KUPEDES did not have to set up its own separate branches and village unit. 2. Firm rules and a system-wide set of incentives to the implementing staff. 3. Decentalized decisiomnakidg a: the local level. 4. An efficient management inoDrmation system to monitor performance of branch and unit managers. 5. Suitable loan terms and conditions. KUPEDES has had the advantage of learing from the earlier highly successful credit program of Badan Kredit Kacamatan (BK), a specialized CDS under the supervision of a goverment-owned provincial ban in Central Java, which is targeted at the poor, including the poorest Many of the features that contributed to the success of BKK have been incorporated by KUPEDES. These include such thigs as charging interest (about 23-31 percent annually), which covers trnaction costs (thereby ensuring financial sustainability), assurance of a bigger subsequent loan upon complete repayment of the previous one, frequent and small repayments, cash loans, lending for a wide range of borrower-selected, fbst-turnover and mostly nonagricultural income-generating activities. Furthermore, it does not get involved in providing other support services like extension, training, and market information. which raises transaction costs. Unlike the BKK, however, which did not require collateral, KUPEDES requires its borrowers to put up colateral but in a form that most small famers can afford, such as houses and house lots. 6. Savings mobilization. The secure and convenient saving program (SIMPEDES) that KUPEDES has set up alongside its credit services attracted potential rural savers as well as borrowers to the program. The huge amount of savings thus mobilized enabled KUPEDES to rely on intemally generated sources of funds for on-lending to their rural cientele. 7. Requitin a collateral. The fear of losing collateral may have worked as a deterrent to wilful default and contributed to the high payment rate enjoyed by the program. The Role of aedit in Poverty Allenation 115 Specialized Bank for the Poor: The Grameen Bmak (Bangladesh) Grameen Bank started as an experiment in 1976 by an economics professor in Chittagong to see if a credit delivery system adapted to the needs and conditions of poor basket weavers and, subsequently, other microentrepreneus was feasible. In its early years, it tried its unique approach to credit for the poor as a specialized window of nationalized commercial banks, using the bank's facilities and manpower, but this did not go well for various reasons. So it became a specialized credit program for the poor with funding suwort largely by the Bangladesh Bank, as well as from donor agencies, notably the Ford Foundation and the Intemational Fund for Agricultural Development. It became a bank in 1983 and has since spread its branches throughout Bangladesh. In December 1990, Grameen Bank operated in 19,240 villages in ten out of twenty- one districts in the county, providing small loans (the average is US$58) to very poor borrowers (Table 6.2). PERFORMANCE. In November 1990. Grameen Bank was serving more than 850,000 borrowers, of whom 91 percent were women. Almost all of the Grameen Bank beneficiaries are poor. Eligibility requirements in respect to land assets and anmnal income is strictly enforced, thus ruing out entry by the nonpoor. The bank has consistently mained a repayment rate of over 98 percent from the start. Studies conducted by the Bangladesh Institute of Development Studies and others showed that the investments of its borrowers have yielded high rates of returns, enling them to repay their loans and earn a reasonable profit (Hossain 1988). Impact studies by the World Bank from a sample of 800 borrowers showed an incrtase in income of 28 percent as a result of the loan (Remenyi 1991). The sustainbility of Grameen Bank is less clear and more difficult to assess, pardy because of its rapid growth, which tends to make operating costs rise (new branches take about five years to become profitable) and on account of subsmatl funds canmeled to it from the donor agencies at highly concessionary interest rates. However, bank estmates show that it can operate at a profit within a few years if it stops epanding even if interest subsidies are removed. Transaction costs of a Grameen Bank branch show a marked decline as it approaches maturity, diminishing from 16 to 25 percent of total loans outanding for new branches (less than a year) to 6 per for older branches (three 3 years or more) (Holt 1990). Older branches operate at a profit, so as a whole the bank has the potential to operate profitably, when most of its branches reach the mature stage. DESIGN FEATURES OF GRAMEEN BANK Among the more important design features that have contrbuted to the bank's excellent performance are its method of targeting, grass roots organition, loan conditions, incentives/penalty system. insistence on accountability, nature of investments and support services, savings mobilizaion, autonomy, and leadership and institutional support (see Gibbons and Kabim 1990). 116 Umad Gtubig, Jr. Table 6.2 Consolidated Cuuative Statement of Grameen Ban as of November 1990 (in Tab 100,000) Particulas Male Female Total 1. Amoun disbursed 13,571.48 60,285.35 73,856.83 2. Amount repaid 12,172.40 49,182.70 61,355.10 3. Amomzunrepaid after one year (%)- 3.19 4. Amount overdue (%)t - 1.24 S. Group find savings' 1,186.01 4,995.32 6,181.33 6. Emgency fumd savings 297.08 926.29 1,223.37 7. Loans fom group fund 997.78 2,487.89 3,485.68 8. Number of members 77,075.00 775,547.00 852,622.00 9. Number of centers 3,133.00 30,572.00 33,705.00 10. Number of villages covered - 19,240.00 11. Number of braces in operati (units) 770.W0 - NGt available. L. Tis figu does - include TK 7,821.65 Lh and TK 9.779.36 tk (rl 1k 17,60102 Lath) dibursed so fr as houfing loans and loans for acvites involving higher tcbogy. b. Ovadue means aMou remn u id after two yeas c. A group member can borrow fom Ife group fu wit die contof rein grop members. Loans fiwm this fud can be used for bosh consmmptiu and investui pupseL d. This is a sort of isurne find This fBnd, when open.onaL wi cover tMe members in caLse of accideu deads, and disster SOn Gacmeefant.U Targeting. Grameen uses an unambiguous eligibility cnterion that ensures that only the poor and very poor can participate. Thus, to be ligible to borrow from the banlk, a borrower's land ass must not exceed 0.50 of an acre and his ime must be less than the market value of one acre of arable land. Grameen's bank workers make a thorough invetigation of the potential borrower's asset and income stats by visiting his or her home and making inquiries in the communt. Grass roots organizaion. When Grameen Bank establishes a new branch, its bank workers first undertake an information dissemination campaign among potential borrowers and motivate them to organize temselves into groups of five lik-minded members and to make one m each group the group leader. The groups then form themselves into a "center," which compries six groups. A center serves as the basic operatig unit of Grameen Bank, conducing wreeldy meetings to discuss loan prposals opey and accept weey repayments and comulsory The Role of Credit in Poert AUeviation 117 svigs deposits. While the loans are made to individual members, the group assumes collective responsibility for the regular repayment of the loan of each member and promotes mutual support among its members. Because the groups and centes are small, the members know each other's activities intimately, and dtis reduces any chance of misconduct. This form of grass roots organition promotes solidarity and participation among the members at the group and center level; it also encourages mutual support and peer pressure to ensure that the loaus are properly utilized and repayments made promptly. Grouping the borrowers into a center that meets weekly also facilitates the process of loan processing and reduces transaction costs, thereby contributing to the ban's viability. Loan conditions. Grameen's loans charge market rates of intrest (16 percnt) and are strictly restncted to self-selected income-generatn activity. The loan reqmres no collateral or guarantor, removing one of the biggest obstacles to institutional credit and enabling even the poorest to borrow. The amount given is determined by the requirements of the investment selected, tends to be small, is given in the form of cash, and is payable in equal weeldy installments over a one-year period. The smal weekly repayment contributes to the high repayment rate enjoyed by the bank. The application and processing procedure of Grameen Bank loans is very simple and straightforward once the loan proposal is endorsed at the center meeting. The borrowers are not asked to fill out complicated forms because there are no collateral or guarar requirements. Illiterate borrowers, especially women, are not handicapped by the bank's requirements and thus it is open to an even greater nuber of poor borrowers. Incendves/penalty system. The bank's most powerfil incentive/pealty feature is the asmrance of a bigger subsequent loan upon complete repayment of the loan, and the denial of further loans to any defaulting member and his group. Because borrowers highly value this access to a reliable and sustained source of reasonably priced credit, -hey go out of their way -to make sure tbat they pay their loans on time. This is specially true for poor women borrowers who have no altemative access to credit, often not even to the money lenders. By belonging to Grameen's grass roots organization, borrowers not only enjoy mutual support and solidarity, but also socially beneficial actvies like preschool services, which are therefore an added incentive for them to remain in good standing with the group and center. A new inceive the bank introduced in recent years is the provision of a housing loan to the older members who have proven their creditworthiness. Accountability. The structure of Grameen Bank clearly delineates everyone's responsibility and accountabiliW, all the way from borrowecrs to bank workers; to the staff at the branch, zone, and headquarters offices; and to the managing diretor and the Board of Management. The board consists mostly of democratically elected represves of borrowers who are able to articulate the needs and view of their fellow borrowers. This not only empowers the poor borrowers but also helps the bank keep its operations efficient and free of corruption in an environmen in which corruption is pervasive. Nature of investme and support serices. The investments that Grameen Bank loans finance c tstically go toward labor-itensive small businesses in trade, processing, manufacture, and services with a fast turnover. Only a few investment go into small-scale agriculture. Almost all the activities are ones that the borrowers are already fatmilia with in terms of the skills required and the market for its inputs and products. As a result, te is little concern about support services such as training, market information, and extension and Grameen is able to concentrate on just its credit operation and keep its costs lower thm would otherwise be the case. Furthermore, the risk of failure is minimized when the borrower is already experienced in 118 Ismael Geubig, Jr. operating his small enterprise, so there is less chance of default. Savings mobilization. A compulsory savings requirement is a key feature of Grameen. An automatic deduction of 5 percent of the loan amount is made at the outset which goes into the group's savings account. Thereafter, each member of the group is required to pay one taba (or US$0.03) at each weeldy meeting. This group fimd is intended to provide ready access to cash by members of the group, with the group's approval, for consumption or invesunent. The group fund, along with the emergency fund to which menibers also contribute, thus serves as a kind of insurance against illness, accidents and other unpredictable events (for example, the death of a credit-financed milchcow). In this way, the meibers not only learn the discipline of saving but also enhance their stake and sense of ownership of Grameen, all of which strengthens their support of the bank. Three years after Graneen became a bank in 1983, the borrowers increased their equity participation from 40 to 87.5 percent, the remaining share being held by the Bangladesh Bank. Autonomy. While Grameen is largely funded by the government of Bangladesh and donor agencies, it has operated almost autonomously, and these two sponsors have made no effort to interfere in the way it administers its credit operation, as described above. Over the years, this autonomy has enabled Grameen Bank to experiment with various methods of carrying out its operations, discarding what did. not work and improving on those that did. Thus, the autonomy enabled bank to be flexible and to develop key design frnures of its credit delivery system that have been most effective in providing a viable banking service suitable for the poor. Leadership and instiuaional support. Another factor that has contributed to the bank's success is the leadership provided by its founder and current managing director. Under less able hands, it is doubtful that Grameen Bank oDuld have evolved into its present form. But now that its credit delivery system has become institutionalized into a bank with a legal personality of its own it is probably now also capable of surviving even without its founder. The sustained support provided by the Bangladesh Bank through loan funds and its willingness to allow GB to operate freely have also helped see the bank through its various phases of evolution. The considerable financial resources of the Bangladesh Bank has also been a big factor in the rapid expansion of Grameen Bank in the past eight years. "Package" Program: The Aga KIamn Rural Support Program KSP) of Pakistan AXRSP is a rural development program with a strong poverty focus. Unlike the other credit programs discussed here,it looks at the poverty problem from a holistc viewpoint, taking into consideration the organizational, institutional, technical, capital and social needs of the entire community. For this reason, AKRSP has adopted a "package" approach, which entails setting up village-level organizations (VOs), developing productive physical infrastructure (PPI) such as irrigation canals and village roads, and the provision of support services such as credit, training, extension and agricultural inputs (Table 6.3). Started in late 1982 as a private entity with support from the Aga Khan Foundation, the government of Paldstan and, later, a number of intemational donors, by 1989 it had readced some 216,707 largely poor beneficiaries in 53,000 households livmg in impoverished mountainous areas of the provinces of Gilgit, Chitral, and Baltistan in nortbem Pakistan. PERFoRMANcE. Areportbythe WorldBank in 1990evaluatingtheAKRSP program shows that it has been highly effective in its goals of poverty alleviation, institution building, and The Role of Credit in Poeny Alleviation 119 infrastrucure development. The credit delivery system of AKRSP is particularly interesting. In 1986 the program provided Rs 8 million (US$0.34 million) to 10,666 very poor farmers in the project area in short-term loans for fertilizer, seeds, marketing, and so on. In 1989 it provided medium-term loans totaling Rs 46.58 million (US$1.97 million) to 21,428 borrowers, mostly for land development and the purchase of machinery. The program was also able to mobilize Rs.60.2 million (US$2.54 million) in savings deposits from the participating farmers in 1989. The default rates on these loans have been exemplary, standing at less than 2 percent for short-term loans and only I percent for mediun-term loans. It is extremely difficult to determine the impact of the credit program on account of the 'package" nature of the services provided by AKRSP. However, its overall impact in alleviating poverty can be inferred from increases in productivity and income from the participating firmers' lands with the introduction of irrigation, fertilizer, high-yielding and disease-resistant crop varieties, and better access to the markets. Since the beneficiaries are economically homogeneous-that is, they are equally poor- it is probably safe to assume that the benefits are equitably distributed among them. The program is designed to become redundant after a period of tire (initially set at ten years). The hope is that the village organizations that AKRSP helps establish will be able to keep the development momentum going on their own. SucCSS FACTORS. A 1987 evaluation of AKRSP by the World Bank attributes the success of the program to its appropriate design,which has matched the development activities of the program to the needs of the communities serviced, as well as to effective management style and imiplementation procedures. Also, the emphasis on establishing PPIs in an area where these were lacding but valued highly by the conmmnity, and the setting up of highly participatory and democratically run VOs through agreements between the villagers and AKRSP in program- assisted activities have greatly encouraged the villagers to participate in a sustained way. The participatory, give-and-take approach used by AKRSP in dealing with the VOs-for example, villagers select the PPI themselves, and villagers are asked to 'pay" for their share of the cost of the PPI by contributing free labor and are epected to manage these themselves-has also contributed to the program's success. Otber contributing factors include a flexible, decentalized m1anagement style with clearly identified objectives and procedures; an effective management iformaion system that enables the program to respond to needs and problems quickly; close contact between the program and the VOs; and highly committed and stror leadership at the top. The VO fom of grass roots organization also plays an importan role in ensuring that credit is properly used and repaid. The tremendous social pressure exerted by the VO on defaulting members and the fact that the savings of the VO are required as collateral must have contrbued to the high repayment rate of AKSRP credit. 120 kmad Gerbig, Jr. Table 6.3 Summay Data an AKRSP Progrim Performance, 1983 to June 1989 (cumulive, unkss stated). 1983 1984 1985 1986 1987 1988 2L989' ViUaghe org s 131 379 477 571 762 993 1,087b es bleshed Women's organizations 10 72 100 133 174 248 271 estabisbed'c fraoqive physical Projects identified 363 706 826 1,045 1.249 1,346 1,426 Projects completed 23 114 195 256 375 514 560 5avings and credit (Rs millions) Deposits by village and 0.8 6A 11.9 18.0 34.3 51.3 66.1 women's organia-tions Credit disbursedlyeazd 1.0 3.2 8.3 12.5 25.7 34.6 30.8 Marketing Village organizaIons 11 28 45 164 191 215 73 Partciatinglyear Number of courseslyeaz' 4 8 14 16 24 37 19 Beeficiares Village organizaton 12.1 31.0 35.8 40.0 45.2 51.3 53.0 Me,nbrsIbfastructure project Penrcmpm1 of mral 12 32 36 40 46 52 54 housebolWs Average members per 92 82 75 70 59 52 49 group Women's organization .6 4.2 5A 6.8 8.3 9.7 10.3 Memberskip (wrands) Average members per 60 58 S4 51 48 39 38 group Acnculzwal credityre 4.7 13.1 122 39.4 40.3 61.0 36.1 (tli;ousuld househords7 Mrkeznear (thousand 0.5 03 1.1 4A 6.6 8.5 2.8 First-time trainges/year 91 179 275 347 401 770 370 a. To June 30. b. ndJudes Me y villg orgnizatons in Gilgit frmed simce 197 by members splitting off Som exisdig village organizat.o c. Jucludes women's sectons of vilage ora tions. d. Short and medium te. f. Also, My>-seven rel rcourselshave been held, making a tDtal of 179 courss and 2,433 ainees attending. g. Total number of ral households is 98,200. h. Includes double colmdig of short- and medium-tm lending to VOs. Some: World Punk (1990). Thc Role of Cre in Poerty Alvion 121 Factors Contribuing to Success: A Synthesis What lessons can be drawn from the failure of the past credit-for-poor programs and the success of many innovative credit approaches to poverty alleviation, such as the four described above? To begin with, there is no single model of credit delivery system most suitable for poverty alleviation. It all depends on the characteristics of the target poverty groups and the economic, social, political, and physical environments in which these groups reside. Developing a suitable CDS that has a good fit with the intended beneficiaies requires some experimentation with what is perceived to be desirable featres in the design, management, and implementation of a credit-for-poor program. What is striking about the four examples described above is that they have all evolved through trial and error. AKRSP, for example, has drawn much from the experience of successful cooperatives in Europe and East Asia, and the Comilla progrm in Bangladesh from which it directly evolved; the KUPEDES can be seen to have evolved from the BKK established much earlier; and Grameen Bank grew out of a pilot projectlprogram of long sandig). This suggests that these successful credit innovaflons may not easily lend themselves to replication, in the sense of blind imitation. However, certainunderlying lessons or "principles" pertainng to the design, managment and, implem on of a CDS could be gleaned frm the experience of effective credit programs. The essential elements of the Grameen Bank approach, for example, have been successfully replicated with minimal changes in Malaysia (Amanah Ikhuiar Malaysia), te Philppines (Ahon sa Hirap) and Indonesia (Kuya Usaha Mandiri). Amanah Ikhtiar Malaysia has now become a fu-fledged Grameen-Bank-type credit program for the poor covering all the country's thirteen stes. It has successfilly mained a repayment rate of 99 percent since its inception as a pilot project in 1986. Efecdve Targeting If a CDS is meant to benefit the poor exclusively, an unambiguous method needs to be incorporated in program design. This was done by using income and asset criteria in the case of Grameen Bank, by confining itself to petty fmale traders and producers in the case of WWF, and by working with poor villages in the case of AKRSP. KUPEDES has done this in a limited way by focusing on the small farmers, although its program was not designed to cater exclusively to the poor. Fralng CDS to the Needs of the Poor CDS attends to the needs of the poor by replacing conventional collateral and guarantor with nonconventional ones like mntual guarantee of solidarty groups; simplified loan application ad processing; small loans, small and frequent repayments; providing loan in cash, rather an in kind; ad arranging a comvenient time and place for the loan transacdon. Orgazing Borrowers into Groups When programs organi borrowers into groups, they promote solidarity and -mual support among the borrowers and there is also more peer preur to perforn well. KUPEDES w6rowers are not organized into groups but the program requires collateral as a detrent to default. 122 Ismae Getubig. Jr. Property Pricing Credit Servces Inerest on loans should cover tansaction costs if the program is to be viable, self-reliant, and sustainable. Appropruite Types of Investmet Finnced Projects undertaken by poor borrowers are more likely to be viable if they are self- selected, designed strictly for income-generating purposes, and have a fast tunover. Providing Effecaive Incentive/Penalby Structure One of the most powerful incentives to the borrower to remain in good standing is to offer the chance of obtaning a larger subsequent loanL The formation of grass roots organintions that provide valuable social and/or economic services is also an effective inentive. ffective Savings Mobilzation Poor borrowers are capable of saving and can be induced to do so with appropriate savings mechanisms that safeuard the deposit and also let interest build on it. This also helps the credit program eam more profit and become more self-reliant in gerating its Ivan fimds. Adeque and Appropriate Training and Preparaon Program staff and borrowers both need to be appropriately trained and prepared to pursue their objectives, procedures, and responsibility. Financial and other forms of incentives for the staff are often effective in motivang them to do a good job, but it is equally important is to instfill in them a strong sense of dedication and social concern by exposing them early on during raining to the plight of the poor and giving them an opportunity to do something about it through the program, as Grameen Bank has done. Accounwbifly and Effective M gemen Informaion System Accountability ensures that the program is responsive, the staff performs efficiendy, morale is high, and there is no misallocation of resources. Accountability with respect to the borrowers will ensure prompt use and repayment of the loan. Approprie Implementing Inritions The instituion established to oversee the credit program must be able to carry out the credit delivery system that embodies the desired design, management, and implementaion fratures. As mentioned earlier, this can take many forms, and no one blueprint should be foliowed. -. . - . . .~ The Rok of Credit in Poverty Alleuiadon 123 Condusion Despite the apparent lack of success of past credit programs for the poor in Asia, many innovativre programs have emerged in the region and elsewhere in recent years that have reached a large nmmber of truly poor beneficiaries, have been economically viable in their loan investments, and have produced income. These programs show that giving the poor access to credit at reasonable terms can help them become economically more productive, and thereby increase their incomes quickly. The manner in which credit is provided to the poor, however, is a critical factor that determines the success, or failure, of the credit program. To be sustinable and to ensure long-term impact, credit should be directed to viable income-generating activities that enable the poor to increase their income and pay back the loan. At the program level, sustinability depends on proper pricing of the credit services,such as the loan interest and service fees so that the program will recover the cost of delivering the services. In addition, the program needs to have a suitable and stable institution through which it could be implemented in a sustained manner. Different types of institutions have been successful in delivering credit services to the poor, as exemplified by the four programs described in ftis chapter. Their effectiveness has much to do with the design of the credit delivery system and how this is managed and implemented by the prGgram staff. Some of the key elements in the design, management, and imple on of an effective. program will be the targeting employed, beneficiary organization, incentives and penalty structure, the manner in which the program staff and beneficiaries are prepared for their roles, the management information system, pricing of services provided, leadership, and institutional support. The creditworthness of the poor and the fact that only a tiny proportion of them have access to credit at reasonable terms make it extremely important to find ways for most of them to obtain credit in a way that is economically sustainable and that encourages them to eventally become self-reliant. In short, access to credit, rather than credit subsidy, is what the poor need most. What is the scope for bringing this about? It appears that the two most promising models for helping a maxmum number of poor obtain credit are the (a) specialized and autonomous credit program targeted exclusively at the poor, such as the Grameen Bank, and (b) a specialized credit program implemented by a commercial bank with a widespread network of branches, typified by KUPEDES and BKK of Indonesia. Programs like that of the Grameen Bank represent an alternative that is exclusively a credit delivery system havmg a very flexible delivery system and havng its management and implementation tailored to the special needs and cusa xs of the poor. But it takes a long time to establish such an institution nationwide, as the Grameen Bank experience shows. The KUPEDES-BKK model represents the structural reform of commercial lending institutions, through a specialized credit delivery system for that part of its operation targeted at small, low- income borrowers. Having countrywide branches through which to channel its credit program for the poor, it is more cost-effective and is able to scale up its outreach much more rapidly than the Grameen-type programs (at least in the short runm). Finally, it is likely to be easier for borrowers of specialized programs of commercil banks to 'graduate" toward borrowing bigger amounts from the banks' normal lending programs once the borrowers establish their creditworthiness under the- specialized program. Indeed one of the frustatons of Grameen Bank is the difficulty it has encountered in getting its borrowers with impeccable repayment records to bormw from commercial banks. This may not be a problem with KUPEDES-type programs, since the "graduation" process is confined within the same bank. Te Rok of Crad in Povert Alleviadon 125 Rderenaces Berger, M. 1989. "GWing Women Credit: The Strengths and Limitations of Credit as a Tool for Alleviating Poverty." WorlDevelopment 17(7):1017-32. Biggp, T. S., D. R. Snodgrass, and P. Srivastava. 1991. "On Minimalist Credit Programs." Savings and Deyelopment 15(1):39-52. Holt, S. L., and H. Ribe. 1990. "Developing Financial Institutions for the Poor: Reducing Gender Barriers." In Polfcy, Research and Exteral Affairs. Washington, D.C.: World Bank. Hossain, M. 1988. Credft for Allevation of Rural Poverty: The Grameen Bmzk in Bangadesh. IFPRI Research Report 65. Washington D.C. Huppi, M., and G. Feder. 1989. "The Role of Groups and Credit Cooperatives in Rural Lending. " World Bank Policy, Planning, and Research Working Paper 2814. Washington, D.C. Islam, R., ad. 1985. Strategies for Aleviating Poverty in Rural Asia. Geneva: Intemationa Labour Office. McGregor, J. A. 1988. "Credit and the Rural Poor: The Changing Policy Environment in Bangladesh." Public Adsmaton and Development 8(4):467-82. Mukhopadhyay, S., ed. 1985. The Poor in Asia: Productiviy-Raising Programs and Stmegies. Kuala Lumpur: APDC. Noponen, H. 1987. "Organizing Women Petty Traders and Home-based Producers: A Case Study of Working Women's Fornm, India." In A. M. Singh and A. Kelles-Viitanen, eds., Invisible Hands: Women in Home-Based Production. New Delhi: Sage Publicatons. Pulley, R. V. 1989. "Making the Poor Credit-Worthy: A Case Study of the Integrated Rural Development Program in lIfdia." World Bank Discussion Pper 58, Washington, D.C. Remenyi, 1. 1991. Where Credt Is Due: Income-Generating Programs for the Poor in Developing Cowutries. London: IT Publications. Singh, A. M., and A. Kelles-Viitanen, eds., 1987. Inviible Hands: Women in Home-Based Prodion. New Delhi: Sage Publications. World Bank. 1990. The Aga Khan Rural Support Progr Pakisto A Second Interim Evaluaion. Washington, D.C. :~~~~~~~- 7 GENDER ISSUES IN ANTIPovERTY PROGRAMS IN ASIA: ExERIENCES AND ISSUES Noeleen Heyzer 9uxMmsry Although poverty is commoon to both men and women, it is more acute among women because of gender-based discrmination. The leading gender-based problems of the poor are unequal sharing offoo4 inadequate medical care and the resulting illness and health hazards; wadepaymnt; long hours and tedious and hazardous work; loss of employment or lower produvity due to illness, which forces the poor to sell orpawn eir meager belongings; shelter, fuel, and secrty problems; no land or property, particularly among a large percentage of female-headed families; a standard of living beow the poverty line; and persistent verbal and physical abuse of women, often ending in suicides. Many measures have been proposed to remedy these problems: divert women to nonagricultural pursuits outside the women-onented areas, shift enphasis from cash to food crops, provide uiproved technologies and train women in their use, provide creditfacilies and ension services, establish cooperatives and marketing outles, and develop cottage indutries. New technology has some adverse effects, however. It may increase the workload and thus affect the welfre of women wad children, may force girls to witdraw from school, and may depnve women of their traditonal means of livelidood. To be of real benefit to women, die measures for mz nmust be designed as a package coveng the enireprocess fproduction and marketng. Special insiuons lke cooperatvesfor women would avoid subordnating women's interests and prevent a loss of female leadership. Cash income may not raise women's standing at home by itself, a great deal aLso depends on the size of uicome, mode of control of the family budget, and culural constraits. At the same time, increasing employment of women may affect the employment position of men and may give rise to conflicts at home The success of thee measures depends on the influence of women's organizations and the wll of women to strive to improve their lives. Asia is stilpredomnanty rural, buta here is a growing trend tward urbanizaion, main because of the influx of rural migranss. Contray to the traditional male-domnated nmgrants, womnn nmgrants are now ounwnbeng men and their nbers are subsnti4 particuarly in the ffteen to twenty-four age group. The iternational migration of young girs seeking jobs as domestic helpers is developing into a big business. Hierarchical fferences exist in the employment of wome: (a) there is a labor aristocracy of smal groups of educated woen; (b) a large group of sklled young women is now working in export industries; (c) the informal sector is the main avenue of employm created by the growth of a large midde class. The informal sector offers three types of emplyment: self- employment, which absorbs two-thirds of migranus; outwork which is mainly done by women and chldlren and domestc work, wich is like an etension of work at home. 127 128 Noeleen Heyzer Assistance to women may take the form of skill upgrading, higher pnces for procs, easier credit, among other measres. Nomfowbanlng institutions,fa,ily anD kinship systems, newly created aid networks, and patron-client relationships can assist, particularly in times of great hardship. Basic sertces like water supply, sanation and health care, and rural energy are prmaly of benefit to women. Stll women are not getting theirfull benefits because their views and needs are not considered. The sustomability of development programs would be greatly improved if the field experience of women was made known, if their vews and needs were asceruzined and if they were given greater control over such program. Orny then can women influence and really benefit from these programs. Jntrducion After more than four decades of development effort in Asia and a high rate of grwth in many of its regions, a large number of its people continue to live in poverty. The nuMber is estmated at more than 500 million. The several anti-poverty programs implemented so far have without doubt alleviated poverty to some extent, but they have also created new inequalities or reinforced the exising ones. There is increing evidence that women do not automatically benefit from anti-poverty programs, and that many growth strategies may make the conditions of significant numbers of women in povert groups worse, if certain adjustmen are not made in plannig assumptions and implementation methodologies. In almost every Asian country, women make up a disproportionately large percentage of the poor and the very poor. There are proportionaely more feae casual workers than male. For example, in Ixdia, according to the 1981 census, half of all rmal female workers, against one-fourth of all rural male workers, belong to this category. The reason for this disproportionately large percentage needs to be fully understood and measures taken to overcome them. Women's experience of poverty may be different and even more acute than that of men because of gender-based forms of exclusion. Women may become impoverished in three distnct ways.First, women may become poor (or conversely, improve their economic situation) together with the rest of the household. Second, they may become poorer through a worsening of the household access to resources. In this case their interests are interd t with the collective interests of the household. Third, they can also become poorer with the breakdown of the flkmily unit itself or their loss of male support. At the same tme, women's lives are governed by more complex social constaints, roles, and responsibilities and are focused more on the nonmoneted sector of society ham those of men. For this reason, policies and programs to reduce poverty need to be sensitive to gender issues within and among households. Within households, efforts to provide the fulfillment of basic needs must take into account the intrahousehold strctures and dynamics that affect the use of Income and decisions over resource allocation along gender lines. Among households, those headed by women and that have young children tend to be among the poorest in the rural areas. Very often country data on poverty conceals these difernces and overlooks reality at the local level. Policies and programs must bqgin takng these strucmes more seriously. They must also recognize that women and men experience the state of poverty differtly and often unequally and that sometimes-but not always-they become impoverished for diet reasons. Indicators need to be found that are capable of addressing the different life circunmances of women and men such as the different effects of marage and child-bearing in their lives, the greater social Gender Isses in Anurpovwy Programs 129 constraint on women's mobility, women's unpaid contributions in income-replacing activities, and their role as managers of meager resources of poorer households. Main Gender-Based Problems Affecting Poor Families In general, the poor appear to lack the basic means by which to ensure their physical well-being. They suffer from severe malnutrition and illness and their life expectancy at birth is among the lowest in the world. Physical survival and well-being is at the core of the poverty phenomenon and the struggle to stay alive, the overriding priority of the poor. Concern with the main problems of physical well-being must therefore be centrl to any aappt to deal with poverty. Once physical survival is ensured, other basic needs come to the fore-the need for shelter, fuel, clothing, security, and self-respect. At the etreme levels of deprivation, the main concen is likely to be for survival and security, so that such needs as autonomy and self-respect will often take a secondary place in the coping strategies of the very poor. In fact, the former set of needs can often only be ensured at the expense of the latter and this explains why poor families seek to become clients of powerfil local patrons and why first wives are often forced to accept their husbands' emrriages. Poverty is indeed generally associated with powerlessness, manifested in random violence, harassment, and uncertainties that characterize the lives of the poor and reinforce their exclusion from the decisiomiaking structures of society. The major problems facing the poor can be identified as physical weakness, powerlessness, vulnerability, physical and psychological isolation, lack of access to and control over development resources like credit, skills, education, land, new opportumities, channels of decsionmaking, as well as the lack of access to the more secre sections of the segmented urban labor market. It is vital to look at these problems through the gender perspective in order to design more effective policies and programs. Hence, this chapter examines in greater detail the burden of poverty between male and female household members in the context of male-headed households, and the nature of poverty in female-headed households. Unequal Sharing of Food and Womern's Econonuc Worth While the burden of poverty relates first and foremost to the problems of basic physical well-being, it is also affected by relationships and social stuctures within and outside the family. Food is the most important item of consumption in poor households. Data on the intrahousehold sharing of food provide some evidence of the iequality in food intake relative to requirements. A number of studies from sweval regions of Asia indicate that female members get a nuch lower share than male members in food quality and quantity, after adjustments are made for height, weight, age, activity level and, for the female, whether she is pregnant or lactating. The most pressmg concen in dealig with widespread malnuttion is simply that people do not have enough food. The -intpan of the unequal sharing of female food intake is complicated by the interaction betwee nder and economic factors. Practices that lead to the inequitable gender sharing of food in the family include feeding males first, particularly adult males, and giving them the choicest and largest servigs. The norms and values justifying such practices are ascribed to by women as well as men. The main items in the daily household diet are another indication of the economic sitation. Food occupies an important role in the social life of many societies and there is a disnt hierarchy m the value attached to different items of food. In many parts of Asia, rice, especially the finer grain variety, is usually prefenred to all flour-based bread and represents the 130 Noeleen Heyzer most satisfactoxy basis of the daily diet. Most other items can be ranked on a scale ranging from status food to fiaine or poverty food. Meat and certin large varieties of fish, for instance, can only be afforded by the affluent. Poverty diets, on the other hand, include meals such as rice, chili and salt only, a gruel made of rice and water, and rice cooked only with lentils. In poor households it is not uncommon for younger women and girls to subsist on poverty diets while resemving better-quality food for the males and the elderly. Studies of landless and landed households indicate that in general girls from landed households are less undernourished than those from landless households (Agarwal 1986). Also, female ifat are breast-fed for a shorter time and given less supplementary milk and solid foods. The discrimination against fimale children sharpens in times of economic distess. The extent of these practices against females is due to historical, cultural, and economic practices. Historically, for istance, female infanticide was practiced m several parts of South Asia and East Asia. The explanations commonly given for this practice are heavy dowry expenditures, the preference for sons, and the interaction between economic and cultural forces. Several researchers (Bardha 1984; Miller 1981) have argued that the eaming capacity of females affects the economic value placed on women and female childre in the household. Many writers have hypothesized that the differential survival chances of a femae child would relate to the expected employment or earing opportunities of female adults. The actual economic worth of worling would depend on the exisftng economic contribuon of women to the household income. For a rise in women's employment to make an impact on the survival chances of female children in relation to male children there would need to be a distinct increase in employment and income oppormuities such that the advantages are dearly perceived and can affect the behavior and attitude toward girls. The perceived economic worth of females to the household is likely to be related not merely to whether or not they work in productive tass but whether or not they ean, which would make their worth more economically visible. The physical viility of women's work, whether field-based or home-based, shoud also be important insofar as this affects the social recognition accorded to it. In households where both men and women are eaming, the differentials in their earings are also likely to infringe on the relative valuation of males to females in the household. Health and Women's Life Cawaces There is indeed a link between health and poverty. Inasmuch as health care requires time, financial resources, and information, the poor are at a disadvantage. Chambers (Chambers 1983), for instane, points to sickness in sitations of poverty where costs enaibin in dealing with ill health lead to the loss or sale of assets. Illness can thus have pmfound effects on the lives of the poor, both immediat and over the longer term. Several studies have pointed out that illness is a nonnal part of life for people who are very poor, rather than an exceptional circumstance. The unsanitary conditions in which the poor attempt to regain their health combine with the inability to pay for adequate treatment and exacerbate what might have been relatively minor illnesses. Furthermore, in houseolds where the main source of income is the physical labor of members, days of employment lost or productivity lowered by illness often mean the starting point of the household's gral descent ito greater poverty. Many have to sell off the family's land, pawn their jewery or the livestock to pay for an illness and are unable to recover their property becu the illness has undmined the household total eaming potential. ;-~ I. X,-~ - - GCder Lsrw in Atipow Prognams 131 Many studies have suggested that there is a strong gender discrimination in access to health care. A study relating to Bangladesh (Chen, Huq. and D'Souza 1981) has shown that even in an area where free clinical care is available, males may be more likely tc be brought to the clinic and hospitalized than women. Other studies in Bangladesh also sbowed that among an equal number of boys and girls who were afflicted with diarrhoea, 66 percent more boys than girls were taken to health facilities for treatment. In a study of child mortality rates, it was demonstraed chat female biological advantage is only evident in the first month after birth when it may be assumed that social disctionhas not yet had its effect. By the end of the firse year the biological patten is reversed. In the first four years of life, balf as many more girls are likely to die as boys. In fact, apart from death by drowning, UNICEE notes that young girls tend to account for a larger nunber of deaths from all other causes. Because women's work is undervalued either in the household or workplace, women are expected to work longer hours than men and to do tedious, hazardous jobs using less technologically advanced equipment. The concept of women as mothers, producers, and rearers of children leads to expectations that women should continue to bear children even when their bodies are exhausted or their health is at risk. Apart from the lack of facilities, women face other problems that limit their access to health services. Some of the obstacles documented are lack of information about the actual health semces available, difficulty in attndg clinics due to employment, housework and child care, inappropriate clinic scheduling, reluctance to be examined or treated by men, lack of privacy in clinics, la k of money for tnsportion costs, dlinic and drug fees. The actul state of health of women's bodies is reflected indirectly in the health of the babies the bear. The m int of women i Bangldesh (Asian and Pacific Development Centre 1990), for example, has been blamed for the fact that 50 percent of all babies born are of low birth weight-below 2.5 kilograms-which in tum can ince the chances ef infant h or of serious iness. A baby with low birth weight is thirty times more likely to die before the age of six weeks than a baby of normal weight. South Asia has the highest percentage of low birth-weight babies in the world-31 percent. About 10 million low birth-weight babies are bom annually in Bangladlesh, India, and Pakistan. Women who give birth to babies of insufficient weight are often those who themselves were underweight and poorly nourished before pregnancy. Duing pregnancy. they did not consume enough calories for both themselves and the fetus. This condition can even be traced back to childhood and adolescence when discriminatory food practices in the family meant less food for girls. Because the extent of women's work in the field, workplace, and the home has gone unrecognized, the myth that women spend less energy in work and therefore need less food than men has come to be widely accepted. The health problems of agricultua workers, pariculaly the hazards that affect women, have been a neglected area of study and action. The use of pesicides is a widespread danger to women workers and can cause cancer, miscarriage, and genetic defects. Dust from nce husks can cse asthmna and eye growths. Since women in agricultur are the lowest-paid workers, they could end up with jobs that no-one else would do and often are the most hazardous, such as pesdcide-spaying. A health project by the Foundation for Research in Commity Health in India in the early 1980s (Asian and Pacific Development Centre, 1990) found, uexpectedly, a high incidence of stillbirths, premature births, and deaths during the peak rice cultivation season. At dtis ime, when rice is being transplated, everyone, including women whose pregnancies are almost full-term, are in the fields the whole day. Their work involves squatng and bending for hours. Such physical strains and pressure m the uterus can lead to premature labor as wefl as stillbihs. 132 Nodeen Heyzer Women worldng in plantions face many sitations that can endanger their health. And many women sprayers sdll complain of sore eyes, rashes, burnt fingernails, and disruptions in their st n period. The Malaysian Ministry of Health in 1988 (Asian and Pacific Development Centre 1990) attributed incidents of pesticide poisonng to unsafe working practices such as mixing pesticides with their hands, blowing the nozzle of the sprayer with the mouth to remove blockage, and carrying out spraying operations without regard to basic safety precautions for handling toxic pesticides. Shelter, Fuel, and Secwuy Since shelter and fuel are joint fonnrs of consumption, gender disadvantage is less likely to be disernble in ousehold comparisons and more likely to show up in comparisons for male- and female-headed households. The housing enjoyed by a family detennines not only the degree of shelter provided from the natu-al elements but is also a mark of social status. The homes of the poor tend to be precarious structures constucted with flimsy materials-walls of mud and jute sticks and roofs thatched with rice straw that have to be repaired constantly and are etremely vulneable to adverse climatic conditions. Recent studies have shown the growing scarcity of fuel sources and its implications for the well-being of the poor. Access to a patron's land or to common ecological reserves-the tadiional source of fuel for the poor-are drying up, and women as well as children spend a great deal of time seekig new sources (Howes and Jabbar 1986; Douglas 1982). In the face of growing scarcity, the poor often take steps to achieve economies in fuel consumption-by cutting down on the number of times a fire is lighted, eaing half-cooked or warmed food, storing dried dung cakes and leaves and twigs under their beds for monsoon days. The poverty of a household can be deducted from the kind of fuel it uses since there appears to be a fairly well recognized order of preferences among villagers, ranked roughly accordig to comfort and convenience of use, the effort required in the collection process or in storage, and whete or not the fuel has some value to nonfuel end use. Another area of concern is personal serity and self-esteem. The mai problems faced by the poor are closely intertwined. Some aspects of the basic needs discussed earlier touch on the questions of security enjoyed by the household. The problem discussed here relates to another dimension of well-being-the security of the person, that is, vulnerability to violence and harassment, which appears to be endemic to the condition of poverty in many parts of Asia. The intense compeion for dwinling resources among the very poor as well as the struggle by the elite to defend their place m the exsti power structures result in random and systematic intimidation, police harassment, false evictions, violent assaults, and murder, and it is the poor who are most frequeny the victims. Gender violence has been less researched, although it is referred to again and again by women's groups. Verbal and physical intimidation of women is a common domestic phenmenon, part of a strategy by male family members to maintin their authority within kin-based biearchies. Husbands, brothers, and fathers figum most frequeny as being responsible for violence. When violence and ill tatment are persistent features of a woman's life, suicide represents the final escape and the tragic revenge. Many development workers can confirm tat the lists of suicide or incidental death through drinking poison in many parts of Asia are predominantly female. Violence against women may be an aspect of class-based violence as well, sice women, particularly poor women without male protection, are amog the most vulneable groups in poor communities. Semal harassment and physical violence, together wih the ability Geder Iuu in Anovzeny Programs 133 to manipulate community noms and customs, are firther weapons in the hands of those who already enjoy power and privilege in the community. Violence, both systematic and random, is part of the condition of poverty inasmuch as poverty is associated with relative powerlessness, and the poor are least able to defend themselves or to remove themselves from threatening situations. Poor women are often forced to stay with violent husbands or to accept sexual abuse at the hand of powerful landlords becase they have no other options. Female-Headed Households and Poverty Anoter major problem is the poverty proneness of female-headed households (F), especially those with young children. Several studies have shown that female-headed households in relation to male-headed households either have no land or a nominal amount of land per capita and have a higher child-dependency ratio, that is, those fourteen years or younger in relation to those in the fifeen to fifty-nine age groups (Visaria 1980). A significantly higher percentage of female heads of households in relation to male heads are in a higher age group, especially those over sixty. They are dependent on wage labor, self-employme in nonagricultural activities, and contract work. Usually EHHs report no full participation in the labor force. Instead they are employed on a short-term basis (for example, weey), they are unemployed, or are unable to find work on days when they seek work, they have a low level of education and high illiteracy rate. The picture that emerges from seveal studies (Partasarathy 1982; Youssef and Hailer 1981) indicates that female-headed households have poorer survival chances than male-headed ones, given that they have less control over land resources and depend more on wage income, they have higher rates of involuntary unemploymt, and as households heads have lower levels of education and literacy. Also many female heads are over the working age of thirty-nine. Although the smaller size of female-headed households is an advantage in that it makes for a higher per capita income, it also implies a lower availability of household labor and overall less labor at the command of female households heads. This can negatively affect the ability of female heads to be successful in self-employment ventmes. All these aspects, when viewed in connection with the observatxon that a higher percentage of female-headed households are m the lower-income deciles, indicate that female- headed households are more poverty prone. A study by Parthasarthy carred out in 1980 in the District of Andhra Pradesh (Prthaarathy 1982), notes that the percentage of female-headed households in the lowest income group of less than Rs 175, which the author terms the "poorest of the poor," is higher than that of male-headed hoseholds. This is the case for every caste where the data are cross-classified by caste groups. A much higher percentage of households among the scheduled castes has female heads than do other caste groups. Among households in the bottom-most decile, 57.5 percent of female-headed households are dependent on wage labor for a livelihood, 37.7 percent of them depend on agricultural wage labor, while only 36 percnt of all heads, male plus female, are so dependent. In other words, female-headed households in this area constitute a much more margilized group even among the poorest of the poor. 134 Noekeen Heyzer Anti-poverty Programs from the Perspective of Women In this section the strategies most commonly followed by anti-poverty programs are evaluated from the perspective of women. The strategies most commonly followed are integrated rural development programs that aim to generate income and employment opportunities for the poor; area development programs for impoverished and remote areas; special credit programs for target groups, for example, small farmers, small business, and the landless; decentralized administrative systems to encourage bottom-up planning and better coordination among government agencies that deliver resources and services to disadvantaged groups; land and tenure reforms requiring the redistribution of land or the establishment of more secure tenancy rights; the provision of basic needs; relief or dole especially in the wake of natural disasters; development of household economies and small industries; and large-scale land-settlement schemes. These programs have some serious shortcomings from the perspective of women. Women's interests are assumed to be included in the various groups that govermnents plan for, such as farmers and the poor. Yet a closer ex; nination reveals that these groups are differentiated and that the lives of men and women within each group are structured in fundamentally different ways. A semal division of labor exists that allocates to women the most tedious and labor- inteive work and limits women's access to control over development resources. The generation of employment tends to build on this existig division of labor and may even intensify it. Even the most sympathetic thinlkng on women's position still focuses on remedial action after the crucial resource allocation decisions have been taken and the direction of economic development has been laid down. There are many untested assumptions that guide anti-poverty programs that can work to the detiment of womens position. The first of these assumptions is about the soial responsibilities of men and women. Goverments plan as though men support poor families when, in reality, it is men together with women who do so, and frquently it is women who do so alone. The fact that most male wage workers are paid less than subsistence wages has been amply demonstated. In these situations women's ime is essential to the survival of the household. It is also essential in migrant male households in wich the women are left behind to care for the family. Female-supported households are clearly on the increase and the poorer the household, the greater the reliance on the eaming capacity of women. Even with the most effective economic development policies, most poor families would not be able to survive without a major contribution fiom their female members. As a first step in remedying this situlation, planners should stop using the term "household" alone as the preferred unit of analysis, but should also use "household members." By emphasizing the household planners tend to ignore the economic and social behavior that occuS both within and without the household. Whether planners Lke it or not, there are gender differnces in the interhousehold allocation of production and consumption. These gender complexities and the handling of income affect the quality of family life as well as the quality of food intake to children, their education and employment, and household stability (conflicts about income are one of the primary sources of household tension). If the household is made the unit for planning, no attention is paid to bargaining, power relations, and the possibility of conflict between members or to the interests of those women who are less articulate. In order to capture intrahousehold differences it is essentia to complement household data with data on household members. Gender Issues in An4povn Programs 135 A closely related problem concerns the concept of work. In the analysis of work and its reward, sharp distinctions are made between domestic and nondomestic spheres of activity. Yet, for many groups of women the boundaries of the two are not so clearly defined. For women in the subsistence sector and nonwage sectors of society, the domestic and nondomestic spheres form a single system and it is often difficult to separate work directed at household members ano work meant for the market. Gender Equity and Poverty Alleviation in Sectoral Ptogrms This section deals with the effect of sectoral programs on gender issues and poverty. Agricultural Development MAJOR APPROACHEs AND AssumPTIONS. Rural women began receiving more attention in the early 1970s, when there was widespread disenchantment with the effects of development policies on the agrarian sectors of developing countries. In some countries, many of these policies allowed food production to stagnate, mnurition to decline, and rmal communities to become destructured, with a consequent massive migation from rural to urban regions. The problems of absolute poverty and of rural and urban unemployment and underemployment then began gaining more attention. As the emphasis in policymaking shifted from modernizaton to the provision of the poorest people's basic needs-for food, shelter, health -distributional issues increasingly appeared on the agenda but failed to take into account the effects of gender inequality beyond those of class membership. Meanwhile, a growing body of research documented the counterproducive efforts of ignoring rural women's contribution and their specific needs both from the point of view of agricultural productivity and of the overall welfare of rual families, as policy proposals related to rumral women became intimately linked to an ongoing assessment of strategies of agricultural development. This section provides an overview of the most commonly held assumptions about the desirability of making rural women the target of direct policy measures. It also examines the primary goals of such endeavors and the means advocated to achieve these goals. The approaches can be evaluated in terms of which of women's triple roles-that of producer, child-rearer, or homemaker-they enmpbasize and which gender need is met. Since the 1950s various interventions have been formulated reflecting changes in national economic and social policy approaches to development. The approach toward women has shifted from welfare to equity to anti-poverty and recently, efficiency and enpowennent (Moser 1989). Diffierent institutions may favor different approaches and even a particular agency may use different strategies simultaneously to meet the needs of different constituencies. In many countries the welfare approach is the oldest and still the most popular method of helping women achieve social development. The programs generated were specifically targeted for 'vulnerable groups" and would come into play when the normal structure of supply, the family, and the market break down. The welfare approach is based on two assumptions: first, that women are passive recipients of development, rather than participants in the development process; second, that the main roles tdat women play are those of homemaker and child-rearer. Welfare is provided in the form of top-down handouts of free goods and services. By the 1970s, there was widespread dissatisfaction with the welfare approach. Many women pmfessionals and researhers had become concemed about the incrasing evidence that development projects for developing counies were having a negative effect on women. These criticisms resulted in the United 136 Nocken HRy7a Nations Decade for Women in 1976, which formally put women on the development agenda" and gave impetus to a diversity of women groups in both the developed and the developing world. During die decade, women's groups and the agencies they ifluenced shifted the emphasis in the appmach to women to eqpity, anti-poverty, efficiency, and empowerment. Several studies in the mid-1970s showed that although women were often producers of development, particularly in agriculture, their economic contribution was ignored in national statistics and in the planning and implementation of development projects. At the same dme, new modernization projects, with innovative agricultural methods and sophisticatedtechnologies, were negatively affecting women, displacing them from their traditional productive functions and diminiishing their income, status, and power. The equity approach grew out of these studies and sought to integrate woren into the development processes of their country through equal access to employment and the marketplace. This approach recognizes that women are active participants in the development process and focuses on reducing the lost ground of won compared to men in the development process. It makes the economic independence of women synonymous with equity. It argues that women should be given their fair share of the benefits of developmeat. However, the equity approach encountered problems from the outset because many thought it interfered with country's traditions of male-female reladons. Nevertheless, its official endorsement in 1975 ensured that it would continue to provide an important fawork for those working in government to improve the status of women through official legislation, particularly in areas such as divorce rights, custody of children, property, and credit. More acceptable has been the efficiency approach, which shifts the emphasis away from women toward development on the grounds that wou;en's economic participation in development links efficiency and eqity. The approach has been formuated by several rural development programs as simply how best to tap the economic resources represented by rumral women in developing countries, and rural women themselves have been defined as resources in the rual sector. The assumption here is that there are vast human resources lying idle and untpped. An equally prominent but somewhat contradictory view is that women are already overburdened with work and that the most pressing need is to alleviate the drudgery involved in providing the basic necessities of life such as fetching water from long distaces, time-consuming food processing, firewood collection, and so on. An ILO report (Kandiyati undated) suggests that most women in developing countries, especially in rual areas, are overworked rather than underemployed and they need to incorporate a more productive tehology in the tasks they perform to save on labor.Tbe objective here is to improve the quality of their enployment rather than create empioymenL Policy interventions aimed at rural women have gained relatively widespread acceptance. Some of the concrete areas in which such interventions are expected to produce beneficial results are ferdlity control, health, food production, and the alleviation of absolute rural poverty through expanded opporunities for geneaing income. It is said that the most promising way to break the circle of rural poverty and high fertility is to upgrade women's productive activities from the household and subsisnce setors to income-generating employment outside the home. The argument here is that nonagricultural employment for women may, at one and the same time, promote unral development, raise the status of women, and aiter reproductive bebavior. Many writers also believe the solution to world hunger is to place appropriate technology for subsistence agrculture wbere it properly belongs-in dte hands of women. .i. Gender Issues in Antpover Programs 137 These views gave rise to the anti-poverty approach based on the concept that many mral women are "the poorest of the poor." Low-income women were identified as a particular target group to be assisted in escaping absolute dprivation. Women's poverty and inequality with men are attributed to their lack of access to ownership of land and capital, and to sexual discrimination in the labor market. Consequently, the goal of this approach is to increase the employment and income-generating option of low-income women through better access to productive resources such as credit, markets, raw materials, and new opportunities and skills. The three main stategies for meeting the world's food crisis are to increase production, increase income-producing activities, and reduce postbarvest food losses. These coincide with strategies for halping poor rural women. The introduction of new technologies, mzny writers have argued, has so far been geared to producing to cash crops, theby drawing off land and labor from food crops. Hence the level of food intake has tended to fail even where cash incomes have increased. If the food crisis strategies are to achieve their goal of feeding the world, women must not only be included in planning, they must be central to it. In the introduction of new technologies, women must be consulted m the selection of these new technologies, trained in their use, and given the means to control those technologies most related to their spheres of economic activities. Assistig poor rural women is a crucial element in guaranteeig both development equity and food in the developing world. Women are food producers in many developing countries. The main constraints on their productivity are related to the libor time involved in their daily household mantenain tasks, which limits the amount of time women have to invest in income-generating actvities. A growing nmrber of time-budget studies indicate that tasks such as fetching water, collecting fuel, preparin and processing food can account for the better part of an adult woman's exremely long working day. Allocating resources to better sanitation, easy access to water points, cheap sources of fhel, improved means of porterage and transportation would have immediate beficial consequences, as would the reduction of laborious food-rocessing operatons through the introduction of apopriate labor- and tme-saving technologies. Better access to productive resources, improved chniques and tools for cultivation, access to agricultural extension services, and the provision of farm credit to women are some of the measures envisaged in the area of agriculural production. As far as nonfarm income-generating activities are concerned, interventions may cover a mynriad of rural-based cottage industies from food caming and textile manufcture to soap- and brickmaing, depending on local skills and resources. Again, education and taning, as well as the establishment of womens organiations such as cooperaives to help them obtain credit and find marketig outlets are considered desirable and realistic objectives. In short, the evidence pointing to increasig levels of female poverty and the implications for community welfare have undoubtedly been isumental in promotig a wide-rangig reassessment of ural women's access to resources, access to land and water, to agricultal inputs, credits and services, to education, training and extension, and to instiutions and organizations. Ensuaing that women will have greater rights of appropraton over resours and their own labor has emerged as an important policy objective for which a variety of measus have been advocated. These measures may be roughly summaized as follows: (a) protect women's existng sources of livelihood; (b) eliminate disciminatory laws on the ownership and contol of productive assets; (c) promote equitable access to agricultural inputs, credit, extension services, and education; (d) support extra household forms of organzation of women's labor; and (e) help inrease woenm's apacity for political empowerment and organization. 138 Noeteen Heyzer The fifth policy approach to women is called the empowermet approach. This approach questions the fundamental assumptions concernmg the interrelationship between power and development that underlie previous approaches. It is argued that women should be able to increase their intermal strength and right to make their own choices in life and to influence the direction of change through control over crucial material and nonmaterial resources. Long-term strategies are needed to break down the inequality between genders, classes, and nations. Empowerment is seen as a slow process that can only move ahead if women develop diverse organizations, movements, networks, and alances to promote the necessary changes. These changes lie in the cultural sphere as much as the econoiic and social one and are aimed at eradicating specific forms of women's subordination- Hence, besides focusing on issues of access and productivity, the empowerment approach also emphasizes legal changes, mobilization, consciousness-raising and popular education. Facors Affecting the Success and Failure of Programs Each of these apparendy straightforward measures presents policymakers with importanit choices and dilemmas. First, governments and planners must become more wary of development measures that actually result in the loss of female control over earings that were traditionally theirs. There are numerous examples of the displacement of female labor and a consequent loss of income in agriculture, maminacturing, and trading. A case in point is the introduction of motorized rice hullers and modem rice mills, which are seen as more commercially viable. The problem here is that the people who used to hand-pound the rice are not the ones to receive subsidized credit for the purchase of the hullers. If this were the case, they would not only retain their employment opportunities but would be reed from drudgery and long hours of work and would have improved their income. Instead, it is the wealthier villagers and urban traders with money to invest who benefited most from the subsidized rice hullers and commercialized nce mil. The new system has made the women who were formerly involved in the hand-pounding of rice redundant and forced them to subsist by subdividing availablejobs by makin themlabor- intsive and by lowering their wage level further. The spiral of rural change in the regions has increased the labor intensity of work and drudgery of some of the poorest rural women without necessarily increasing their income. Thus it is particularly important to provide women with the necessary tainig, organitional skis, and credit resources to retain activities over which they already have some control. Second, a move must be made to eliminate legislation barring women from access to productive assets m terms of rights of inheritance, the ownership and control of property, and the adoption of positive measures to ensure their equitable access to land, livestock and other productive resources. Legal access to resources is a vital step in this direction, but it does not in itself ensure control. Piecemeal legal measures, especially if they are confined to titde ownership, may create new possibilities but will tend to have a limited impact unless they are backed up by a package of measures covering every stage of the production process, including marketing. Third, the inequality of women's access to agricultural inputs, credit, and services, as wel as to extension and education, has been well documented. Access to credit and to favorable terms of credit is a problem for all economically and politically disadvantaged groups. In the case of rural women, it is compounded by the fict that in many places women are not yet recognized as legal adults.In some societies women do have traditional credit arrangements among themselves, but the collective resources they are able to pool are neager and often tend to be more of a crisis fund. The measures proposed to enhance rual women's access to credit range . .. .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Gender Isus in Auipoerny Prognums 139 from the support of grass roots self-help networks to the foundation of cooperatives and the introduction of supportive government legal systems. There is a growing awareness that measures designed to enhance women's options as producers, access to resources, credit, extensions, education, and services must be matched with parallel efforts to reduce their domestic responsibilities. An important factor to consider in this respect is that the creation of new opportumities for one category of women in a household should not work to the detiment of others. In other words, it is essential to look at the household as a total system and to evaluate the impact of changes in the life options of any member in terns of their effect on others. When, for example, opportunities arise for women to eam and control cash income, their daughters should not have to withdraw from school to take care of household tasks and their younger siblings. It is not difficult to imagine other situations in which the intensification of women's productive labor makes the mother-daughter relationship a central labor relation in the household. The spread of schooling has the opposite effect, in that it increasingly removes children, especially female children who are mothers' helpers, fom the household and makes their labor unvailable to women, whose burden therefore increases. It may sometimes be in women's interests to hold their daughters back, and they in fact do. Given their subordinate position within the household, women have little control over labor other than their own, and the only area of latitude they have might be in relation to their daughters, daughters-in-law, mothers, and other female relatives living with them. Policies that are geared to the abstract category of women but that do not consider their current responsibilities within the household and how these are going to be substituted for, run the risk of merdy sharng certain burdens differently in each group. When new demands are made on women's time, something has to be given up, and more often than not that something is anotber woman's time. Yet another factor affecting the success of a program is whether specific institutions are established to organize rural women's work A strong case is made, for instance, for having separate women's organizations such as cooperadves for women only. The choice is justified on several grounds: such insttutions build on already exsting female networks or modes of cooperation, avoid confrontation with cultual patterns that oppose the mixing of unrelated men and women, and do not submerge women's interests and loss of leadership to men, as occurs all too frequently in cooperatives with household membership. There may stiU be- an act of faith involved in the notion that providing women with a cash income will automatically ensure a significant improvement in their standing within the household. Admittedly, this may hardly be relevant when women's added income barely ensures survival. In other cases it will greatly depend on the nature of the work; the regularity, security, and amount of the wage earned; and the nature of domestic arrangements, including modes of budget control and the culural coaints on women. Another issue that merits fxrther investigation is whether in situations of male dominance in which cultral systems emphasize male responsibilities there are short-term or long-term increases in domestic violence as well as other manifestations of stress and conflict when employment opportumities are created for women, especialy in the context of male unemployment or underemployment. Conversely, where men's opportunities for eamng wages are far ahead of those of women and women are perceived to be an economic liability, they may suffer great abuse, as in the case of Tndia where soaring dowry rates have been related to the increasing incidence of dowry deaths and even to female infanticide. 140 Nodcen Heyzer On the whole, there is relatively little detailed information about the effects of changes in women's options on sexual dynamics in the household and the workplace in different culnual setings. These effects should not be expected to be uniform or unilateral but should be explored in their own nght since they may spoil the difference between increasing levels of harassment and abuse, or on the contrary, greater autonomy and well-being. Yet another important factor to consider concerns rural women's organizational capacity. It is useless to hope for significant advances in the rights of appropriation and access to services without a certain measure of political empowerment and partcipation. Many cultures today have an array of women's organizations and solidarity networks outside the home. The extent to which women's networks have been used to help them survive, to maintain existing prvileges, to resist unfair treatment or to create political change is also being increasingly documented. Several studies, however, have shown that class, caste, and ethnic divisions may introduce strong competing loyalties as well as different interests. In conclusion, policies and programs for rural women have come from interationai development agencies as well as governments and there have been successes and failures as well as ambiguities and conadictions. This is partly due to the fact that many of the programs proposed to assist rural women are, at the same time, stop-gap measures to tackle some of the more visible outcomes of underdevelopment, such as rural poverty and maginalon, the food crisis, rural-to-urban migration. The need to persuade economic planners of the cerainty of payoffs involved in assisting rural women on their own terms and within their own frame of reference reiforces the mage of women as instruments of development rather than social and political subjects in their own right. Although the package of trget group approach to rural women might produce effects in the short-term, it can be no substitute for development strategies with a serious commitment to tackling the mechanlsms responible for inequality and poverty. These strategies need to be informed by an acute awareness of the culturally and historically specific forms that women's subordination takes and animated by the will to struggle of the women themselves. Urban Development Although the majority of the popuation still live n the rural areas, urbanization has greatly increased in the Asian and Pacific region during the past three decades. Figures charting this growth both globally and regionally are staggering. In the thirty-five years since 1959, the number of people living in cities almost tripled, increasing by 1.25 billion (World Commission on Emvronment and Development 1987). While the urban population in the more developed world doubled in those thirty-five years, that of the developing world quadrupled during the same time frame. By 2025, the world's populon is expected to reach aboat 8.5 billion, and much of this will be in cities of the developing world; of this projected population, an estimated 60 percent will be urba (World Commission on Environment and Development 1987). In Asia, increasinmly large proportions of the total populatin live in cities and urban setings. For example, the populaton of Jakarta, Indonesia, which had a population of 1.5 million in 1950, is exected to increase to 12.8 millionby the year 2000, while Manila, Philippines, with a populaijon of 1.78 million in 1950 is expected to reach 11.1 million by the same year (World Commission on Environment and Development 1987). This is the growing trend throughout Asia, with the exception of China and India whose urban populations only make up 32 and 24 percent, respectively, of the total population (Brown and otbers 1987). GenderLuue in Antovey Programs 141 Owing to the speed of urbanizaon, Asian cities are experiencing ovecwding, uncontrolled growth in their slums; an ncrease in urban poverty, traffic congeston, inadequate sanitation, water, and waste agclrt; and a prolifeation of disease and crime. More than half of this growth in urban populations is due to ntral urban growth and uearly half to ural- urban migrtion. There has been a marked increase in feimale migrants. In fact, the rura-urban migration rate for women in the fiften to twenty-four age group has increased substtally so that there are more women than men migrating in this age grp for many countres in South East Asia. This pattern of migration is a break from traditional male-dominated migration in the Asian region. A new trend is also that of the international migration of young women sedeing work as domestic helpers. This has become big business. It is commonly argued that poor urban women are better off than poor rural women bese it is assumed that they migrate from free choice and there is a free labor market in operadon. In reality, to understand the migration process it is important to consider the strural forces that shape the organization of production as wefl as the dynamics of the household and kinship networks in influencing the choice of young women to migrate. ft is well-known that with asymmetical power relaions of households and kinship networks, young girls and young women are among the least powerful. They are usually controlled by famly rules, the family development cycle, and kinship obligabons. Research on the importance of young rural women to the cities has shown that the changng prosduon structure and the poverty that follows such changes lead fmiies to send their daughters and young women to town to earn extra money to support the family or household unt. The earigs of these young women are often cunil to the surval of the family. Thize is sufficient evidence to indicate that female migation is not only reated to factors of production but, equally IMpornt, to social relations to gender. Male violence, breakdowns in maage, and polygamy are all pressures that have driven women ito FACToRs AFFTG WoME WoR IN URBA AREAs. The work that women do in the udrm areas depends on the wide economic siuation, which deermies differentil labor absorption in the urban areas as well as dependence on the stage of the wome's life cycle and their educationa backgrund. The econonies of many Asian countries are at a stage of development in which the growth of stable wage employment is insufficient to absorb the high growth in population so that certain groups, particularly women who are migrants, are incorporated at the margins of the urban economy. It has often been argued that in such a situation hierarchical differes are used to organize the conditions of production and accumulation. To understand the major problems faced by women in the urban areas, it is necessary to look at some of these hierarchical differences and how they contribute to the concentration of women in the low-income nonwage sector. First, a labor aritocacy has developed in many Asian countries with high wages for small numbers of workers in the face of an excess labor supply. Tie labor maietthus becomes highly snmented, for the labor force is seped into submarkets wih different employment conditions and wages. This segmentation is not a passing phenomenon that will disappear 'as econmic growth tricldes down.' Urban poverty wili contine to coexist with protected labor markets and will continue to be the life -experience of many urban women and migrant womnL The issue of schooling and skills in bargan tes on a new significance with the growth of a protected labor market, particuarly for employment in international, ma and clerical jobs in newly created bureaucracies. In these labor mats, schooling and the credenials it provides are used as selective mechanisms and as a measure of labor-strainability 142 Noewa Heyf and discipline. Access to high-wage jobs depends not only being competent in certain skills but also on the protection and bargaiing power of organized labor. Second, consideration should be given to the cration of an age-specific feiale work force. In several Asian countries the internadonal fragmenotatio of the labor process has led to the differential labor selection of women in different sages of their life cycle, that is, to the creation of an age-specific labor force. There has been an overwhemning concentration of young women in world market facorie, that is, in factories that export to the inatonal market. The women who are selectively absorbed in stable wage employment in urban areas are those with educatonal qualifications and those who possess the dcrctrscs demanded by modern sector employment. These are usually male and young female workers. For many women this employmert has brought about a break with the past, an opportunity for earni cash imcome and greater independence. However, in terms of woring conditions, many of the intricate mechanisms of gender-based subordinaion contnue to operate, even though the framework is more heterogeneous and complex. The group most discriminated against were the older, often marmed, womnen with few educational qualificaions or forma documens. ASSISING WOMEN IN THE URBAN INFORMAL SECTOR. In many Asian counties the avenue of employment for these women lies in the urban informal secor. Even for informal sector employment. competition is inten and most of tXa women are employed in activities that tke specific forms. In designing policies and program to assist women in the urban informal sector, it is imorta to determine why the economic participation of these women takes its specific form. In many Asian countuies whether women participe in the informal sector depends on their ability to manage multiple roles. Women's work in this sectr differs fom dot of men in a variety of ways. Fir, they are concenat in areas of the nonwage sector that are compatible with their reproductive role, particularly, child ring, and often extensions of their domestic responsibities witdin the household. Second, they are, with few exeptions, concentrated in areas that reqir little capital ouday and in areas with less growth potential. Studies that have been done on the semual division of labor in a nunber of infonmal sector activities in urban centers of Asia found that the kind of nonwage labor in which low-icome women are concentrated are those that have involved skils developed wihn the households. The task entailed in unpaid domestic labor, that is, work carried out maidy by women in the home, are transferred to the wider economy where production is for exchange. Men, on the other hand, do work that tends to involve new skills, new to the traditoal Asian village society learned in the formal or onforma educaonal sysems, or on the fringes of the modern industrial econmy, for exmple, machine rearng. In addition a higher capital investment is required for men's economic activities in the informal sector, such as owning a smal shop, or owning a ti or van. Women are mainly found in the domestc services, in petty trading (seli fits and vegetbles or cooked food, as well as in the tetile and tailoring secor. Most of this work can be caried out by women in their own homes and be part of fanily cowumption. Many women cook within their own premise and hence the trading of cooked food is a frequent choice for women, particularly if they are heads of their households. It requires very little initial capital outlay, makes use of tadal skl, allows a woman to combine her livelihood with her child-raising inctions, and is more lucrative than nmany of the jobs open to women with little foral edtuation. If we athe maretg activities of wom with those of men, here is a tendeny for women to concentrate in areas tat provide lower retums. In many parts of Asia women are concentrated m small-scale trading and smal businesses while men are involved with large-scale tading and larger businese. The main reason for this is not only the lack of access to market Gender hsues in Antipoty Progmums 143 and capital, trading concepts and transportation but also the fact that large-scale tradin,g and businesses require a level of flexibility and mobility that women will not have unless they are freed from their household chores. While a rigid sexual division of labor does not exist in small- scale marketing and trading, it is the women who usualy help out in the husband's small retail shop as the unpaid family worker together with her children. Three major forms of employment can be identified in the informal sector: self- employment, wage or contract labor, and family labor (which is often unrecognized). Nearly two- thirds of women in Asia in the urban and informal sector are self-employed. In fact, in many countries, urbanization may help women escape some tadional social barriers to employment and even weaken patriarchal control in their daily lives. However, new problems emerge. Many of these women trade on pavements, indicating both the high degree of uncertinty of location as well as inability on their part to make structural and technological improvement. They are also the group most vulnerable to police harassment. Women in outwork fonns another very large percentage of women in the informal sector. Outwork involves putting out all or part of the production process from a central point to several small units. This process is part of the strategy to cut labor costs, to overcome problems of capital investment, and to survive in a competitive market. By putting out its work, the central mterprises tale advantage not only of low wages, insecurity of employment, and lack of social benefits but also of long working hours. Outworkers in Asia are predominantly women and children. Domestic service is another foxm of work that is a complete extension of tasks performed by women within the household. This form of employment is one in which women are already skilled as a result of task allocation by the seal division of labor within the houshold. The tasks are practcally the sam ones women perform as unpaid workers wiin the family- cleaning, washing, ironing, and taking care of children. Although domestic service is a relatively stable wage employment and camnot be strictly descnbed as informal sector employment, it has some featumes silar to most ionnfal secr jobs. For intance, it is work that is often done without a contract and therefore the employee may be subjected to instnt dismissal. I is work that is seldom protected by any govermnent regulations. Domestic servce is a frequent job choice for female migrants, and poor urban women are readily available becase of the growth of a large middle class. in developing programs and strategies to help women in the informal sector, an important question is why some men in the informal sector manage to be self-employed and even be smnll- scale employers, whereas women seldom achieve the status of independent produers despite their skills. In order to assist women, the current strategies have focused on si ng women's existing econonmc actvitkes by sdll upgradg; improvig productivity; obtainng better selling pnces, lower costs of producton, and easier tms of credit; and providing support sermces like space, equipment, health care, child care, and introducing completely new acvities that have a relatively high economic retr An importa intervention is providing access to credit. This means understanding and removing consaints on access to baks and other credit systems. This applies not only to fiXed capital but also to working capital. It is in this area that nonformal banking systems can step in to help these women. To encurage the expansion of small business, daily or weeldy system of lending for working capial could be eolved m colaboration with non-govemmentl organizations to prevent the misuse of funds. Structural and technological changes are also important. Both of these require acces to credit as much as access to a fixed location. 144 Noeken Heze7r WOME'S NErwoRKs AS A SURVIVAL STRATEGY. This section deals wi how life goes on for households in urban poverty. The reproduction of life is precarious in these households. Besides trning to income-generating activities, women-especialy those in crisis situations--tun to complex networks of kin and neighbors as a survival strategy. It may be that some women in extremely impoverished conditions are still able to raise healthy children while others are not, mainly because the former group has access to various survival networks while the second group does not. In South Asia, three patterns appear to be at work: 1. Although the family system has frequendy been a system of sexual inequality and many women have sought to escape from oppressive family relationships, it is often the family and kinship systems that are recreated by women in their atteWts to deal with the problems of poverty and survival. 2. Besides the family and kinship system, new social groups are created for mutWal support. These social groups take various forms. They may be neighborhood groupings of supportive households; they may be people joined together by a shared, newly created ideology; they may be people who are linked together by a common religion. 3. Vertical links may be established with social groups that are financially more stable. The most common form is the patron-client relationship. A number of studies have shown that relationships are activated and used for acts of exchange. The urban poor, as a survival strategy, organize their relationships as social capital ffiat also substitutes for the credendials and organized careers of the formal economy. In a situation of poverty, where life is extremely insecure, it is important to have automatic support in times of crisis. Family and kiship systems have commonly been used as clear indications of who had the right to assistance in times of need. These systems have not been without conflict, domination, and violence, but they at least provide their members with some social insurance. These survival strategies are especially created in times of economic crisis and adjustment, when the increased inability of the modern sector to absorb the large supply of available labor means a further reduction in the generation of employment and a fail in the standard of living of the poor because of price increases, a reduction of services, and a drastic reduction in real income. This has often been accompanied by a parllel niion of daily life and the emergence of various coping mechanisms by women to deal with conditions of precarious living in resource-poor situations. Many women in poor households are forced to make drastic adjustments in the household consumptionpatterns, reduce the quanity and quality of food intake and basic sevices, and go into debt. In the absence of a welfre state it is at the level of the household where the fierce struggle of survival of the poor takes place and women, as immediate managers of household resources, are at the center of this process. Pressures are set up to concentrate only on the most urgent needs and to neglect others such as home upkeep and repairs, as well as to look for lower food prices in markets. This often means the coordination of shoppig with other family membern, including the use of extended faidy networks, especily women members of their networks. These networks are also relied upon as a result of the intensificationof domestic work fiom daily shopping due to more restrictive budgets, to increased cleaning, fiEing, mending, and sewing done at home-all of which has increased the workload of women and the time they use. In other words, in the absence of a welfare state and the Gender suts in Anlipoveny Programs 145 decrease in governmenl services and subsidies, the fiamily and kinship networks are often the only source of support and of alterative strategies. The informal sector is able to provide a basic minimum standard of livig to at least a section of poor women in the urban areas. Although the kinds of work available to them are governed by structural and sexual divisions of labor, women are seldom passive agents of these forces. Their economic behavior in response to these forces takes different forms. They interpret and organize their social experience and a resistance to larger processes, which may range from highly individualistic competitive solutions-for example, individual sm-scale traders and fruit sellers competing in the marketplace-to solutions that are managed and controlled in a collective manner, such as the setting up of cooperatives and the organization of self-employed workers. Most common are privately created systems. These range from family and kinship systems and newly created mutual-aid networks of different sorts to patron-client relationships. Although these networks do not bring about any long-term fumdamental changes in emancipaton problems and even poverty, in the short-term they at least pmvide some guaantee of survival, particularly during periods of great hardship. Programs designed to protect women in the informal sector could build on these networks side by side with ways to provide women with training, credit, and other support required tr diversify production to improve their productivity and income. Condusion Toward Soal Sstainability The development of basic services is without doubt of benefit to women. Local water supplies, sanitation, roads, rural energy programs, and family health care can all do much to reduce women's work and improve the health conditions of women and children. The available data suggests, however, that in the planning of these basic services women's needs and perspectives have been excluded, although women are the prime users of such services. As a result, these services may not be used by those for whom they are intended. More and more decisions are taken by bureaucracies, and women have little, if any, control over the directon that planning takes within these organizations. In the dismssion of larger issues of sustainability, it is important not to forget the need for social sustainability. In the final analysis, the best guarantee for sustimbility of development programs is greater self-reliance, in this case, women's. This means allowing women to have control or to gain greater control, have a say, and be listened to, be able to define and respond to their needs and perspectives, be able to influence social choices and decisions affecig the program, and be recognized and respected as equal citizens with a contrbution to make. Programs can provide these opportunities for women if they examine the ways in which they are worling with women in the planning and implementation stages. Planners and implementors need to ask how the relationship with decisionmakes in the program can be changed to allow women to act as definers or creators of the program. This means exmnining the possible role of women and involvement in given projects to ensure -that they have more control and are not just doers. The program should allow women to feed back into the system, instead of receivmig istructions or ideas firm the system and implementing them on its behalf. The field experience of women who were doing the actual day-to-day inplemntation and management of the program should be used as a resource by decisionakers. Their perception of the needs and problems, particularly of women, could go back into the system. Hence, in designing programs, more thought should be given to how women could report back into the program, project implementation, and evaluation process. As a resu, the program would be more responsive to 146 Noely HeAyzer women's input-an important improvement. There will also be a change in the sharing of knowledge within the project. Programs are more ihkely to be susuinable if they are responsive to the real needs of people. Women could establish lines of communication and feedback on their own during meetings with various program offices by presening ideas and suggestions. Women should not just be told what the problem is but be able to teli government departments and officials what problems they have found. Women should also help to set priorities. These are but a few eamples of ways in which women can influence and gain more benefits from programs and projects. ader Issus i Anpovty Program 147 References Agarwal, Bina. 1986. "Women, Poverty and Agrialltural Growth in India." Joumnal of Peasant Stdes 13(4). Asian and Pacific Development Centre. 1990. Asian and Paczfic Women's Resource and Action Series: Health. Kuala Lumpur, Malaysia. Bardhan, Pranab. 1984. "On Life and Death Questions; Poverty and Child Mortality." in Land, Labour and Rural Poverty: Essays in Development Economics. Delhi: Oxford University Press. Chambers,R. 1983. Rural Development: Putting the Last First. London: Longman. Chen. L. C., E. Huq, and S. D'Souza. 1981. "Sex Bias in the Family Allocation of Food and Health Care in Rural Bangladesh." Popuotn and Development Revew 7(1). Douglas, J. J. 1982. -Traditional Fuel Usage and the Rural Poor in Bangladesh." World Development 10(8). Howes, M., and M. A. Jabbar. 1986. "Rural Fuel Shortages in Bangladesh: The Evidence from Four Villages." Discussion Paper 213. I)D2. Sussex. Kandiyati, Deniz. (undated). Women in Rural Production Systems; Problems and Poiaces, UNESCO. lester Brown, and others. 1987. State of the World. New York: W. W. Norton. Miller, Barbara. 1981. The Endangered Sex-Neglect of Femnale Chdren in Rural North India. Ithaca, N.Y.: Cornell University Press. Moser, Caroline. 1989. "Gender Planning in the Third World: Meeting Practical and Strategic Gender Needs." World Development 17(11). Heyzer, N. 1986. Working Women in South-East Asia; Development, Subordnation and Enciaon. Milton Keynes, U.K.: Open University Press. Parthsarthy, G. 1982. Rural Poverty and Female Heads of Households: Need for Quantitanve Analysis. India: Institute of Social Studies Trust. Visaria, P. 1980. Poverty and Living Sandards in Asiat An Owvrew of the Main Results and Lessons of SelectedHousehold Surveys. World Bank WorkingPaper 2. Washigton, D.C. World Commnission on Enviroment and Development. 1987. Our Common Futre. 148 Nodee,, Heym7er YwSef, Nadi4 and Carol Hatder. 1981. Women-Headed Households and Rural Poverty: 1T Do We Know? Geneva: Inteamional Labour Office. Part 3 DECENTRALIZATION, TARGETED POVERTY PROJECTS, AND SURVIVAL STRATEGIES OF POOR HOUSEHOLDS: LESSONS FROM KARNATAKA 8 TARGETED POVERTY PRoJECTS FOR THE POOR: LESSONS FROM PROJECT VISITS Ab& Aziz Diwith the idea that poverty cold be alleviaed by promoting ecomic growth gave r to three new strategs: (a) the target group apprch, which aims development projects scficay at the poor and emphasizes anployent and income generadon; (b) dhe paricPatoiY approach, which is concened par1tic y with urban services and with diveng services in a more cost-effective maimer-in tms of both time and money----ough a government ddivery yeand (c) the aided self-help approaches inated by NGOs, under which the NGOs proide technical, oranional, and other resources to enable the poor to help thensdves. ExPience has shown, however, that the benefts and impacts of such progms and projects have been. less than ecte. The are a number of reasons for this disappointing perfrman: amismatch betwee project design and the national policy eviromnent; inadequate prOject planning, particulady in regard to assessing Prect viability and sustainability; iappropriate logies; inabiliy to develop locay apprpriate oaniat with provision for people's particaon, which could sustain, manage, ane ilement progrm with a minimum of political interferc, leakage of benefits owing to a ms fica ion of beneficaries and misuse of resoures by the benefciaries; unproductive nature of the assets acquirecreted and nonrdetetiGn of these assefs by te beneficiaries; and a lack of political will and clout to recover loans so as tn recle funds for sustaining the life of the project. In other words, many projects fail to deliver the itnded benefits to tie poor becase their design and mana are inadequate. This problem nust be fully understood before corrctive murs can be developed. The path to such u snding is firsthand knowledge of the manner in which such projects ar dsigned and implemeted at the field level. Bangalore seminar participants vere given an oppomnity to gin such knowledge throu visits to six projects. They were enoraged to discuss all aspects of t project with planning and implementig personnd and with the beneficiaries. The sim projects revolved around rag piking, lacquewar, seicuture, a women's credit and poitical edu;atio program, sites and serc housing and shm upgraing, and nutritiom. The first thee are located in Kanataka and th last tbree in Tamil Nadu. Tlhe rag pickers and the women's program are NGO-managed; the others are nmaaged by the govmnent. Each of them pwvides services to a specific target group- Semar partcipants were divided into three groups on the basis of their field of in,rest, ascetained by means of a short questionnaire. Each group was led by an ISEC faculty member 151 152 Abdu AZiz who had prepared the project profile.' The profile was prepared in advance and highlighted such aspects of the project as its design and management, the nature of the target group, the organizational struture of the project, the nature of the benefits to be delivered, and the delivery system. The profile also provided a brief evaluation of the project. At the project site, the participants held discussions with the plamning and implementi agency and with some beneficiaries. Participants then discussed their findings among themselves and presented a report to the other seminar participants. These exercises produced some interesting findings and useful lessons. This chapter gives a brief account of the projects visited and the lessons drawn from the project visits. Project Profiles Rag-picing Children's Rehabilitation Project It is a well-known fact that the urban poor depend on infonnal economic activities for their livelihood. For a large segment of the migrant population and those unable to find work in the organized sector, the urban informal sector offers unlimited opportunities (although less remrumerative). One such activity that has attracted the urban poor, especially the very young and the very old, is rag picking. In towns and cities, it is a common sight to see children swarming around public dust bins day in and day out to scrape and to pick discarded tins, paper, rags, and such other waste material for which there is a secondary market from the junk traders. The rag-picking activity provides opportunities for children to supplement the family income; but the social (and even the prvate) cost of such activity is quite high because the children who ought to be in school and in play grounds are engaged in hazardous work. The work is hazardous in the sense that most of the children engaged in rag-pcking suffer from skin diseases and respiratory ailments. Being members of poor households, they are also undernourished. An NGO in Bangalore formulated and implemented a project related to rag picking. The project is funded in part by the Union Ministry of Labour and in part by donations from voluntary agencies. Its operations are periodically shifted from one slum to another where rag- pickdng childrrn are concentaed. The project objective is to wean the rag-picking children away from this hazardous activity and to rehabilitate them. Toward this end, project staff identify the target group and extend the following benefits: medical aid, nutrition, and recreadonal facilides for their mental and physical development; nonfornal education designed to help them achieve funcional literacy; vocational tramining to enable them to take up alternativejob opportunities; and motivation for formal education. In the first stage of innlementaion, the staff carries out a household survey in the selected slun and identifies the rag-picking children. Those thirteen to fourteen years of age are then given placements in city hotels, cycle shops, petrol bunks, garages, and workshops. These children are not only taken out of the rag-picking activity, but are also placed injobs that provide opportunities to build a career. 1. The falty rmeber who pcpard ad pesened die prXje pr,fies wer Id Frabladachar, P. Thippiab, S. N- Sangiba, S. A. Pasha, K. 0. Gayahri Devi. Shlt Mohae, SmaI Johnson, P. H. R;appa. an! M Nageshwam Pao. Tuhe dulled profiles prepared by txse fruity membe ae presented in smmy form her. Lssons Frm Projea Kirrs 153 Of the children who are younger than rteen, a few who have an aptitude for vocational jobs are selected for vocatiora taining. At the end of te training period, project management places these children in appropriate organizations. Tle remaining ones are allowed to pick rags up to 2:00 p.m. every day but are then required to assemble for tbree hours at predetermined locations, where they receive nutritious food, instruction in personal hygiene, assistance with fuctional literacy, orientation about the potential markets for their products, and voional training. These children are gradually weaned away from the rag-picking trade and admitted into formal schools. Thereafter, they are given books, school uniforms, and some cash-for a total worth of about Rs 150-315 depending on the class they enter. Project management uses different strategies to motivate children and parents to make the switchover. Video films ame one means used to attract children to the program. To persuade parenb to agree to the switchover, project staff bold a direct dialogue with the parents in their residential area; hold a fortnightly "mothers' meting"; and during the vocational training period when the faIwiy loses the child's income, the opportuy cost inurred by the parents is compensated by a payment of Rs 100 per mont-half of which is given to the parents and the otier half deposited m a bank account for the child, who may use this amount after the training ends to purc*ase the equipment and tools needed to carry out an independent businus. The project also combats some of the exploitaive mechanisms characteristic of the junk trade. The dire poverty m which rag-picling cildren and their parents live forces them to sell each day's collection of material to the junk taders on the latter's terms. The junk traders also preempt the chilren's right to sell elsewhere using tactics like advance payment and facility of petty loan. In order to extricate the beneficiary children from the clutches of such exploitation, the project stff settles children's dues with traders and provides facilities for stocking their picked material in the project go down. Angements are also made to sell the material in dte open market for a better price. In the beginng the project was mainly concemed with ipprovig the livin and health conditions of the target group while they continued worldng in the same trade. Subsequently the focus shifted tD imparting vocational trainnig so that the children would be prepared for taking up betr jobs. This change is said to have been initiated in response to demand fiom the target group. Those of the rag-picking children who were earmarked and prepared for formal education began going to schools regularly and completely disassociated themselves from rag pickng- much to the satisfaction to the project staff. Drawing on their experence and encuraged by the success, the project staff have replicated the program in four more areas and are planning to extend their area of operation. These plans are being held up, however, by delays in getting govmen grants and inadequate funds, all of which is causing severe strain on the human and financial resourc of the project. Lacqzrware Handicrafts Project Traditional hanicraft artisans are regarded as one of the poorest sections of Indian society. One such group consiss of the lacquerware artisams in Karnataka,, who are concenated in the block of Bangaore rural distict. They are engaged m producig lacquerware products such as toys, flower vases, bangles, smoking pipes, beads made out of the locally available special wood known as Haale and exotic chemical colors. There is a good market for these products in India and abroad. But the economic conditionsof the lacquerware arisans deteriorated recrenty because the main input, Hale wood, became scarce when the forest 154 AbdulAziz area where it is grown beaTme depleted. Lac, which is imported, had also become expensive. The artisans had barely enough financial resources to buy and stock these expensive raw materials. The skills of the artisans and the production technology employed were not in tune with the tastes of the modern- day consumers. Moreover, they had come to depend on the exploitative middlemen both for the purchase of raw materials and the sale of their products. To improve the socioeconomic conditions of the poor artisans the government of India and the government of Karnataka recently intervened in the production and marketing spheres through a comprehensive development program for the poor artisans. One of dte first steps taken under this progran was to establish separate lacquerware training institutes for boys and girls with a view to upgrading the skldls of artisans. The Regional Design and Technology Development Institute established at Bangalore develops and supplies tools and equipment to the artisans and thereby helps propagate new technologies. The commercial banks have been directed to advance loans to the needy artians under the Differential Interest Rate Scheme; government subsidies are also available to the poor artisans. Through a depot in Channapatt the Kanataka State Handicrafts Development Corporation (KSHDC)in Bangalore supplies scarce raw materials like Haale wood, lac, and colors at reasonable prices. Also under a separate scheme, the state provides common service facilides and house-cum-worksheds on tms that are easy and attractive. Project implementation also includes periodic artisan household surveys carried out to ascertain problems relating to skills, techniologies, tools and equWments, the supply of raw material, credit, and prduct market conditions. These data are used as inputs to work out artisan development strategies. The first point of intervention is the triing of youth. Trainees are idenified through open application from boys and girls belonging largely to weaker sections. The training period lasts six to twelve months, and during this dme the tainees are paid a stipend ranging from Rs 150 to 250 per month. The training curricula and teaching methods are designed to impart both practical and theoretical knowledge. In addition to technical skills of production, the trainees are taught managerial and entepreneurial skills. Along with such regular taining programs, these ititutes offer short duration tining courses for the practicing ardsans and master craftsmen with a view to demonsng new technologies and designs developed in the national laboratories and design institutes. The traied candidates are linked to programs eligible for subsidized credit from the commercial banks to enable them to start their own enterprises. The required raw material is supplied by KSHDC. The common facilties center offers services for processig wood and manufacturing and coloring the articles at reasonable charges. The finished product is purchased by KSHDC and sold through its network of depots and showrooms. The house-cum-worksheds built in 1988 with assistance from the Netherlands Development Agency are available to poor artisans. Each unit costs Rs 30,000 of which 10 percent is subsidized by the government of Karnataka. The cost of the house is recovered from proceeds on the sale of their products through the KSHDC. The project is designed to provide a package of benefits to the artisans. As such, it offers the aran ample time and energy to devote himself to the production of this product. Second, the proceeds from the sale of his products are linked to loan repayment and recovery of rhe cost of materials supplied and service rendered, all of which ensures that the project is sustained. The available data suggest that the training facilities have been taken advantage of by the weaker sections and minorities. More than 50 percent of the trainees were scheduled castes (SCs) and scheduled tribes (STs), and a quarter were muslims. Half of the tainees were women Lessons From Project V-uin 155 However, only a small number of ardisans interviewed had obtained bank assistance (the rest having borrowed from private sources); only a smal proportion of total output produced had been bought by KSHDC, leaving the arfisans at the mercy of middlemen; because of fixed working hours and frequent internuption of electricity, the common facilities center could not ensure the required service and the artisans were forced to go to private landowners. The small number of house-cum-worksheds (126) built was inadequate. In addition, there were complaints that ineligible persons-artisans having their own houses and persons who were not in lacquerware trade-had been allotted sheds. Krnaaka Seiculzure Project Modem sericulture has a history of 200 years in Kaataka. Thanks to the patronage of Tipu Sultan, ruler of Mysore State at that time. Today, sericulture is an important economic activity of rural Kanataka. Although the fortnmes of the Indian silk idustry in the intet.ntional market nave been flucuating owmg to competition from China, Japan, and Italy, India ranks second only to China in silk production. Among the leading sericulture states in India, Karnataka is number one in the production of cocoons and raw silk. Alhough Karnatakm has great poentia for sericulture development-because of its f-avorable climatic and soil conditions, skiled famers, cocoon rearers, and silk reelers- sericulture has been confined to a few pockets in the state and mostly to big landowning classes. Its high employment potential and its scale-neutral production would certainly have opened up possibilities for using sericulture as an insument for alleviatfing poverty, But sericture in -Karnataka was sfill too backward for this to happen. It lacked the technology, infrastructure, soil conditions, and administrative structure needed to develop sericulture to its full potential. In order to remove these obstacles, the govement of Karnataka in 1980 launched the massive Kamataka Sericulture Project with the World Bank's assistance. The project was initially planned for a five- year period,but subsequently its terms were extended up to 1988. The project objectives were to increase raw silk production by 1,600 tones a year; to bring 14,000 hectares of irrigated land under mulberry in new areas (this was subsequently revised to upward of 18,500 becures); to introduce modem processig facities and methods that would upgrade raw and spun silk to export quality; and to introduce the latest tehologies from leading silk-producing countries, to expand local researh, and to introduce regular teaching programs in the universities. To achieve these objectives, the governme undertook a massive effort to establish and strenthen infrastructure facilities. For ample, it established the Kamataka State Sericulture Development Institute to carry out applied research, and it also established additional graiages that were to produce disease-free laying. This would permit private grinages to develop, and would provide more extension and infrastructral support for silkworm rearing by establishing community hatcheries (kmown as Chawki rearing centers) and technical service centers, and would thereby strengthen the existing markets and establish regulated cocoon markets. It would also start common facility centers for the benefit of reelers, establish silk exchanges to facilitate exploitation-through the free purchase and sale of silk yam-and would designate certain cooperative banks, ral banks, and commerial banks to finance sericulture activities. Although- the project is considered an agricultural development project, it has an implicit antipovert thrust. The project is sposed to generate incomes and employment oppornides for the poorer sections such as small and marginl farmers, landless agricultural laborers, women, and the SC-ST population. The project seeks to reach dtese sections by providing them credit, extension, and common serice facilities. On-fam credit for the constuction of sheds and 156 Abdu AziZ purchase of rearing equipxmet is provided under ERDP to all categories of poor sericulturists and under Special Component Plan to SC-ST serclturists. The latter also are entitled to a free supply of disinfectants, trays, and chandrikes (bamboo containers where the silk worms weave cocoons around themselves). In addition, the bivoltine rearers are entitled to some cash incentives. Extension service is provided to all categories of sericulturists. In some areas community development centers have been established to provide rearnng accommodation exclusively to SC- ST sericulturists. The main benefit of the project is that land under mulberry cultivation has been extended and there has been a substanial increase in cocoon production. Sericulture activity has spread beyond e five districts to which it was confined and is now practiced in all the districts of Kaka. The project has increased employment opportunities and income generation, particularly in the weaker sections. Mulberry has become a popular crop to cultivate among small and marginal farmers and is giving them an opportunity to ean higher incomes. A high proportion of participants come from disadvantaged grops such as women, minorities, and SC-ST populations in sericulture activity. As evaluations of the project point out, however. there have also been some failures. Crop losses have been frequent owing to silkworm disease. In addition, credit bas been misused, loan repayment is poor, and cocoon markets opened in new areas have not attrcted sufficient quanities of high-quality cocoons. Worilng Women's Forum (WW9 Women's earnngs, however small, contribute signifcantdy to the needs of their families. Female participation in the labor force is also high. In reality, however, women are among the most disadvantaged and oppressed groups, in addition to being victims of the existing caste and class oppression. Their caste affiliation restricts their mobilty, members of their cas have limited access to productive resources, and because of their gender role they are unable to participate in many profitable economic activities. Consequently, a large proportion of women belonging to poorer sections end up worling in low-paying inmal sector jobs. In cities like Madras they are engaged in preparng and selling raw and processed foodstuffs, they serve in households as maid servants, end animals, sew and clean, keep small shops and work as small contract manufacturers of beedies (local cigaretts), incense sticks, paper bags, and the like. Because these women cannot obtain credit support from formal institutions, they are highly depedent on informal sources of credit sources and are therefore continuously indebted to moneylenders, landlords, and middlemen who charge exorbitant rates of interest. They do not even havre basic amenities available to them, like drinking water, sanitation, headth care, schooling, and electricity. In short, their living coditions are miserable. Harassment by petty officials, policemen, and local s Utmen is a daily fiature of their lives. The wayward drunkard and indebted husbands add to tieir burdens. The Working Women's Forum was founded in Madras in 1978 by a social worker of repute in an effort to rescue women from such exploitation and oppression. The basic objectives of dtis organization are to provide oraizational support to women workers on trade uimon lines, to raise their level of political consciousness, to defend their rights for equal opportumities, to fight discrimination and harssment by petty officials and local strongmen, to improve their icome levels and enhance their access to basic development resources, and to provide an Lessons From Project Visi 157 intensive health and family planning program with a view to strengthening women's productive role and restricting their reproductive role. To help organize women and sustain group initiatives, the forum has created neighborhood credit groups and has encouraged beneficiaries to repay their loans through peer group pressure. Initially, the WWF acted as an intrmediary to obtain credit from nationalized commercial banks for its members under the differential rate of interest scheme by standing guarantee and ensuring repayment. Subsequently, it established its own Working Women's Credit Cooperative Society in 1981. The society helped women cope with a variety of problems, such as delays in the processing of loans, cumbersome procedures, and the red-tape characteristic of the nationaizd bankdng system. The society is managed by the women members and over the ten years since it was founded, it has disbursed Rs 9.63 million in the Madras branch alone. The loan recovery rate is said to be about 95 pernt. The WWF workers carry out surveys of slums and shanties where informal sector working women are concentrated and identfy those women who are poor, exploited, illiterate, and powerless and those who lack access to resources and opportunities and belong to lower castes. In the selection process, priority is given to those who are victims of ignorance, natrl disasters, and physical incapacities. Using the participatory method, role playing, and pictorial charts, the WWF makes prospective beneficiaries aware not only of the dangers of exploitaion, but also informs them about acdvities that can have a positive impact on their life, such as cooperation, family planning, literacy, hygiene, and financial self- reliance. By the time the woman member is trained, made politicaUy aware, and linked with the credit cooperative society, she has transformed herself into a self-reliant small entrepreneur who can manage her own affairs with confidence and courage. The WWF has etended its activities from Tamil Nadu to Karnatama and Andhra Pradesh, and from urban informal sector activities such as vegetable vendig, the preparation and sale of foodstuffs, beedi rolling, and agarbathi ing to nonfarm sectors in rural areas such as fisheries. In all these cases, the entry point is the provision of low-interest credit with a view to starting and strengthening self-employment opportunities for its members. Subsequently, health services and civic and political activities are imtrcduced to make the members better citizms This organization helped women achieve a better quality of lfe and have their basic needs met, such as children's education, housing and health, and asset building. With credit readily available to them, women's enterpises have gained econmic idependence. Thus it has allowed them to stop pawnmg household articles to moneylenders and has freed them from continuous indebtedness to informal sector financiers. On the intangible side, these women have strengthened their position in the family by participaig in decisionmakling, improving relations with other family members, and increasing their social consciousness. They have learned to operate through group processes, and have thereby increased their solidarity and clout in fightig for their rights. This has enabled them to obtain better civric facilities, and helped protect them from people who harass, exploit, and victimize women. With their newly won bargaining power, they have sucded in secunng markets for their work and toilet facilities both in the home and the workplace. The Tanil Nadu Integrated Niurion Project (H7NP) Because of their extreme poverty, populations in developing countres have lower intakes of food than those in more developed countries. Withi counties nutritional status varies from region to region. In Tamil Nadu the food intake is lower than the national average. In such conditions, the sharing of food within the household may be ineitable and the children may be 158 Abdul Aziz particularly hard hit. The Tamil Nadu Food Habits Survey found severe deficits in food consumption among children between six and thry months of age. It is therefore not surprising that many young children there suffer from malnutrition and related health problems, including. a wide range of diseases. Malnutrition and ill-health reinforce each other and cause irreversible damage to the physical and psychological development of the child. Under the circumstances, the state has taken responsibility for controlling malnutrition by introducing supplementary feeding programs for children and may include pregnant and lactating mothers. At least the political commitment to nutrition has always been strong in Tamil Nadu. Three projects are being implemented at present. For the seminar field visit we selected TINP which is distinctive in several respects. Apart from being rural-oriented, this project selects its beneficiaries with cost-effectiveness in mind; for example, it plans to admit undernourished children into the program and then to exclude them from supplementary feeding once their nutritional status has improved. The project also integrates nutrition and health delivery services and seeks to improve knowledge, attitudes, and practice of women in areas of nutrition and health so as to generate benefits on a sustainable basis. In addition, it has a mechanism for close monitoring and evaluation throughout the life of the project so that corrections can be made in midcourse. The project has three components: nutrition, health, and communication. Under the first component, children aged six to thirty-six months are periodically weighed by the community nutrition worker and those who are underweight are enrolled for the supplementary feeding program. Those whose weight retuns to normal are phased out and those showing no response are referred to the village health worker or medical officer at the primary health center. Pregnant and nursing women are also selected for feeding on the basis of objective criteria such as family income and their health status. The health care delivery system, which is the second component of the project, involves periodic house-to-house visits by the health worker for immunizing the pregnant women and young children and for distrbuting iron and folic acid tables to the expectant women and nursing mothers. The health worker also assists in deliveries. The third component, communication, entails propagating nutrition and health practices among the rual community through interpersonal contacts and elements of the mass media, such as posters, films, cassettes, and hoardings. In this way, project workers motivate rural community members to bestow better nutrition care on six- to thirty-six-month old children and expectant and nursing mothers. The evaluations conducted have shown that there has been a steep reduction in severe malnutrition, improvement in the nutritional status of beneficiary children, a reduction in infat mortality rates, better coverage under immnization programs, and better integration of nutrition and health components. However, the field visits revealed soi0 ' drawbacks in project management. There was lack of adequate coordination with other ong&ing nutrition programs. Tne result was that children in some age groups were left out. Even where different agencies worked for the same age group,there were some gaps in coordination, particularly between the nutrition- and health-adminiserg agencies. Some difficult questions concerning targeting were also brought to our attention: Is it appropriate to exclude those children who are slighdy (not substantially) above the cut-off weight point? If the weight of a child coming out of the program later goes down, would that not have some adverse impact on its health? If laddus (enriched biscuits) are given to the pregnant mother and if she shares them with her children, would that amount to misuse of benefits? Should one take action to prevent such misuse? Lessons From Project VAsits 159 Sites and Servces, and Slum Upgrading Projecr With the rapid urbanization initiated by rural immigrant, the demand for housing in the urban areas tends to grow at a much faster rate than housing supply. Many immigrant workers tend to live on pavements and vacant lands and to put up makeshift shelters, huts, and the like and create slums and shanties where extremely unhygienic conditions prevail. Although it is the responsibility of the state to provide shelter meeting minimum living standards, this responsibility puts a heavy strain on state resources. Given the constraints on the state's financial resources, it is difficult to expect it to provide free housing even to the poor. Hence there is need to develop a housing scheme with cost recovery. The cost recovery approach to housing for the poor is justified on the principles of replicability, affordability (that is, willingness and ability of the beneficiary to pay for shelter), and the feeling of ownership of the house. In keeping with the above housing philosophy, authonties have developed a series of housing projects. With the financial and technical support of the World Bank, the Madras Metropolitan Development Authority (MMDA) initiated the Madras Urban Development Projects consisting of (1) a sites and sevices program and (2) a program for upgrading slums. Both programs proposed to improve housing of the urban poor and slum dwellers. The sites and services program started in 1977 in the Arunbakkam area of Madras city, where it provided 2,804 sites, of which 74 percent went to economically weaker sections and the rest to lrw- and middle-income groups. The program provided a package of services consisting of(a) infrastructure such as water supply, sewerage and electric connection, street lighting, storm water drains, roads, paved paths, and (b) amenities such as community halls, schools, medicare centers, shops-aim-residences, service industies, bus terminals, workshops, and so on. Sites were ailotted by means of a lottery system. The beneficiaries also received material loans, house building loans, municipal development approval, and technical advice. The cost of the site was to be recovered through an initial block payment and a monthly installment payment spread over fifteen to twenty years. The response to the scheme was good: houses were put up promptly and loan repayment was impressive. Subsequently, however, it was estimated that nearly half of these houses had been rentcd out as they fetched attractive rents and that with the rise in land prices in the area, the proportion of sites/houses being sold had increased significanty in recet times. In one sense, poor families certaily benefited from the project as it brought them capital gains (when they sold houses) and regular income (when they rented houses). Although this is helpful from a purely economic point of view, it may be less positive with regard to social development. The poor are expected to live in the houses allotted to them and to upgrade their living conditions. If, however, they use the proceeds to acquire new income-yielding assets or to fbrm human capital, that is a different matter. A second problem noticed was that as soon as middle-income groups purchased houses and moved in, the roads laid out earlier proved to be too narrow to meet the growing volume of traffic. This fact suggests a lack of foresight on the part of the project designing agency. At the same time that the above program was initiated, the scheme for upgrading slums was initiated in a slum in the Annanagar area. The Tamil Nadu Slum Clearance Board decided to improve works such as land leveling, road and street formation, drainage, water supply, provision of street lights, and other amenities. The hut dwellers-the original residents-who had been shifted to a nearby area were then given lease-cumsale deeds for their sites; they were expected to make an initial deposit and a monthly installment payment for twenty years. Thy 160 Abdl Aziz were also provided with "home inprovement loans" that would be released in the stages of house construction. There was a good response to the scheme, and loan recovery was said to be satisfactory. Encouraged by the success of the schene, the MMDA later undertook 322 slum upgrading programs, which benefited about 75,000 households. The problem with the "home improvement loan" was that not many persons came forward to use it. The reason was that under the scheme the amount advanced was very low and was given in stages of house construction. As a result, many preferred to raise funds by selling jewelry rather than going through the cumbersome procedures required to obtain a loan. Some Lessons Effective design and management of a poverty alleviation project is achieved through (a) effective targeting and (b) attention to ensuring project sustainability and replication. Targeting should fulfill the following conditions: the project beneficiary should be clearly and appropriately defined, cost-effective systems for identijring the target group should be developed, an effective system of benefit delivery should be in place and monitoring should be carried out to check on who is actually receiving benefits as the project is implemented. Project sustainability and replication are ensured if benefits continue to be delivered to the intended target groups through the delivery mechaiism evolved and if the life of the project is prolonged through the generation of intemal finance and its recycling. This means assets must be created and retained, and that the cost of the project must be recovered from the beneficiary. According to these conditions, the project studied yield the following lessons: 1. PoLHcymakers must pay particular attention to defining poor households. Household income has traditionally been the most widely used indicator of poverty, but experience indicates that it is more difficult to measure the income of urban than rural households in view of the varied sources of income obtained by the forner. Since the sources of income are restricted and specific to land size in rural areas, it is easier to esfimate income there. For example, a rough estimate of income can be made from information on the size of a landholding. 2. With information about the nature and quantity of assets held or occupation engaged in, it should be possible to obtain a better estimate of the economic position of a given urban household. Therefore, while household income can continue to be an important basis for defining the poverty goups, the activity engaged in (rag-picking or lacquerware, for example) and caste could also become a basis for defining these groups. Nutritional levels, health condition, and oppressive social conditions faced by the groups in question could also form a basis for determinng target groups. Such supplementary meares are important because income alone does not determine the quality of life. It is important to take into consideration access to social services such as health, education, nutrition, and housing. 3. Beneficiaries are most likely to be misidentified where subsidies are high, as in the case of the sites and services project and lacquerware development project. Rigorously planned and executed household surveys by independent or voluntary agencies (rag-picker project) will certainly reduce the misidentification, but the cost of such procedures may prove unduly high. Lessons From Projea Visfs 161 A better alternative is perhaps to involve community members, especially the neighborhood members (WWF), in identifying beneficiaries. 4. While designing the projects and identifying the benefits for targeting, undoubtedly local resource availability, household human resource stock, skills, and managerial capabilities must be kept in mind. At the same time, certain social imratives ought not to be overlooked. Thus, high-rise housing projects may not be acceptable to the poor for they will not be able to keep animals there or have a workshop or a small shop with which to obtain supplementary income (as they would with sites and services and slum up-grading projects). Similarly, if benefits like nutritional supplements are prepared by low-caste nutrition workers, this may not be acceptable to high-caste beneficiaries (TINP) on account of the social stigma aetached to low-caste workers. Perhaps workers belonging to this category should be posted to villages that have predominantly the same category of population. This may go against the government's anticaste policy but may be justified by practical considerations. 5. Several basic economic imperatives need to be considered as well. Wherever the project tends to interfere with the fimily earning potential (rag pickers), resistance and protests from the heads of families are likely to be encountered. This difficulty can be overcome by compensating for the family opportunity cost. Even here, a purely business approach to evaluaing the opportunity cost does not work unless parents are educated about the future increase in their children's earning potential through schooling and training. 6. The income- and employment-generating intervention alone-say, credit for self- employment or work under wage employment scheme-is not likely to alleviate poverty. This measure should be complemented by a package of benefits that includes traiing, schooling, placement (as in the case of rag pickers), taining, input supply, marketing of the product (as in the case of lacquerware and sericulture projects), infrastructure, and amenities (as in sites and services). A simultaneous attack on the symptomatic condition, such as malmtrition and disease (TNIP), appears to be a more approprate strategy of targetig the benefits than taclding them individually and sequentially. 7. NGOs appear to be more willing to, and effective in, transforming a given govemment assistance such as a grant into an integrated, multiservice, poverty alleviation program (rag- pickers project) thm a govement organization. In such cases the cost-effectiveness of the projects is always high. But a problem faced by the NGOs (as in the rag-pickers project) is that they work under financial constraints. Under these circumstances, there is, perhaps, a strong case for NGO-govemment cooperation in poverty alleviation programs so that each can complement the other. 8. The integrated multiservice projects seem to have better potential for developing a cost- effective delivery system. Where even one single service is provided (for example, credit), it could still be made more cost-effective if community members were involved in the identification of beneficiaries and administration of the delivery system (WWF). Even under the bureaucrcy- managed programs like lacquerware, this objective can be achieved if there is better coordination between the agencies involved (such as KSHDC, the commercial banks, and Artisan Training Institue). 162 Abdd AziZ 9. Project implementation can be monitored more effectively by the peer group (WWF) and by people's representatives (lacquerware) than by the bureaucracy. 10. When the assets created under the project bring a high rent income and generate capital gains, it is difIlt to expect the beneficiaries to both use the asset and to retain it (sites and services). Some kind of legal restriction that applies more stringently against the transfer of the asset should be devised and enforced. Otherwise, benefits would reach the target groups at the start of the project and then fall into the hands of the better-off individuals after a lapse of time. 11. Providing benefits free of cost inevitably affects the life span of a project and, therefore, the project authorities wiU find it difficult to sustain and replicate the project (rag pickers). Where cost recovery is found to be very high (WVWF and sites and services), project sustainability and its replication have also been high. The sustainablity of some kinds of projects can also be ensured by having a connal stream of beneficiaries enering ito and graduating from the scheme (TINP). 12. Linking the sale of the product through government channels to loan recovery (lacquerware) would be a more effective way of ensuring cost recovery and project sustainability. Such is the case when peer group pressure is used to encourage the beneficiaries to repay loans (WWF). Perhaps a more enduring way is to establish a conducive policy environment that would enable the authorities to enforce loan recovery and collect user charges, but this approach calls for political will and clout on the part of the government. 9 DECENTRALIZED PLANNING AND POVERTY ALLEVIATION: THE KARNATAKA (INDIA) EXPERIENCE Ab Aziz Deceralized governance is beeved by mmy to be an effecie mews of plznning and inplemenng povey aUeviaonprogras. Insofar as it is ndertakn at he gross roots level (us dhe village or vilWage sters) and by die people or thr represenatives, decentralized pknning it le4 to be more responsive to people's needs and more effetive in managing progranm, deiverigbens its, and mobiing rsourcs. Pwuingforpowery alleviation caried out by decntaled governments is herefore epected to reap dte adwztages of decetraLized planning. lJis chapter ewniner the Kaactaa erce with decenralzedplanningforpoverty alleo ni, in the above pearective. The ditria and tde vilage cluer gowrnments azia paihad and nman panchayaLs, respectvely) were constited in 1987 wth elected people's representatives (including prqrsemadon of weaker sections ke women, SCs and STS, and backward classes through resemtion). Thee local bodies are in chargc ofplumaingforpovert alleviado AU thre tpes f pover alleation straegies fisted in Chater 1-4=te1, promotion of laor inesie economic grow, broadeing access to social serWces, and specific povet alleviton projects-have bee handled by the decaralizd governments wnder study. An anaSis of the pattern of resource allocaton mUong mzdalpwchayats show tha socia andemc i c , uich is nmre eaphoymn producton oriented, has received highprioity. The on-site empnt opportnes generate, the stngnng ofsocial servces such as hel and educaton, the provi of drig water, and stavy faaiti have improved the living conditon of the poorer sectons of th area The Union governent's progran for creating wage emplymen (Jawahar r Yojana, JP.19, dth self-empl nta suppordng Integrated Rul Developmenm Program (IRDP). and the central- and staedflad Speial Component Plan eclJivey for SCs and STs are managed by the marndpandcoats. From ther ownufmJ, the mandalpanday alsopkt and implement the 18percent sdme xasiveyjfor the benfi of SCs and STs. In addition they hande specific prjects such as those tatprovide buods and carts, lghting, irrigation wells, smokeless stoves, and th like, for poor hosds. Projects for the sociay drssed dasses lie SCs and S7s sffer itte leakage and misidntication beawuse of the sf-selecdan process that is implied in such programs. In JRY and fffDP, laZkages have been kept to a rndniw because befiiaie are udntfW systemaUcally in open wllage assembly mee_s to wzich all the ctizens are inied 1. Ths rsmcb u_mi of Daw_a Babu in dt prqa_dw of dis cI is gmlu acdkowlg. 163 164 Abd,dAz Introducton Over the years two approaches have been sequentially followed in India to alleviate poverty. First, the national approach of growth maximization, which was followed until the early 1970s, sought to allocate more and more resources to those sectors, regions, and sections of the commuity that were well-endowed and thus had higher productive potential than others. When this development strategy produced growth of the enclave type-that is, growth was only achieved in some areas-and when it caused regional, sectoral, and sectional dispariles and did not reduce the incidence of poverty and unemployment, an alternative strategy was developed. That was the target group approach, which sought to provide direct succor to the poor. Under the target group approach, projects were designed and managed to either increase the poor man's ownership of physical assets such as land; or to raise the productivity of assets owned such as land and livestock; develop human resources through health care, training and education, improved living conditions through belter housing, disease control, and family planning; or endow him with income yielding new assets through bank finnce and government subsidy. The target group approah to poverty alleviation, which sought to provide self- and wage- employment opportunities to the poorer sections, dominated the Indian scene during the late 1970s and the whole of the 1980s. A detailed study (Rao and Aziz 1990) of the relevance and impact of these projects on community asset formation and on income and employment generation at the household level of the poor pointed to the following gaps: 1. The schemes applied to poor households do not coincide with the skills available there or with the infrasture in the villages. 2. Owing to the kind of power structure in the rural areas, beneficiaries have not been correctly identified, with the result that in many cases nonpoor households have obtained benefits under tie antipoverty progrms. 3. Benefits and assets have leaked out of the program because of corruption among the bureaucracy and local leaders. Furthermore, loans have had a low recovery rate, with the result that project sustinability has been poor. This state of affairs is blamed in part on the fact that planning for poverty alleviation is primarily done by bureaucrats sitting at the center of power, remote from the people, and that neither people nor their representatives participate m designing and managing such projects. Decentralized planning of poverty alleviation projects has recently emerged as an alternative to these other two strategies. Decenualized plaming is said to have several advantages: it is more responsive to the needs and aspiraions of the people; local problem are better understood and resolved by utilizing local resources; people, particularly the beneficiaries, are involved in the design of the projects and thus there is less likelihood of the mismatch problem; since the people are involved in imple tion, beneficiaries are less likely to be misidentified; with better monitoring of the program, asset reteion and loan recovery will be more satsfactory and hence susiiunability of the project will be ensured; and most important, overhead costs will be kept to a mu' '&nm and tmus the delivery system will be more cost-effective. This chapter examins the effectiveness of decenralzed planng as a strategy for alleviatig poverty and illusaes at least some of the above propositions in the light of field information gathered from a study of the Karataka case. Since the d alid method of planning has been Decentralized Planning and Poverty Alevifion: The Karna1ka Case 165 in use for only a limited period and not much literature is available on the subject, the impact of such a plan strategy cannot be discussed in depth. Nevertheless, our field observations have given rise to some distinct impressions. Deenrald Planing: Concept and Logistcs Many would agree that planng camed oLut below the state level could be treatd as decentralized plan process, as against,the macro planning carried out at the national and state levels. Below the state level, planing is done at district, block, and village cluster levels, although in some states either the block or the village cluster is skipped for operational reasons. The Working Group on District Planning, in its report submitted to the Planning Cowmission, noted that decentralized plan stategies arise from the need to supplement the national and state plans with a more detailed examination of the resources, problems, and potential of local areas, so that investment progms can be tailored to the particular needs of each of the lower spatial units (Working Group on District Planning 1984:22). From this broad definitk' the following implications for decentrlized plan formulation process may be drawn: First, the methodology used in planning for the lower spatial units should consist of developing a mechanism that (a) identifies the local needs of the people, (b) ascertin the grm.h potential of the area on the basis of a assessment of the available local resources and budgeary allocations from the state, (c) draws up priorities for the programs and schemes, and (d) helps in formulating a strategy for maxmizing the outcome. Second, before the decentalized approach can be applied, autorities must set up decentralized data-ollectn machinery, a decentraized political institution that can aiculate people's needs and aspirations, and a decentr-aized administrative structure to formlate and implement the dcentalized plans. Third, since decentralized planning is seen as a subsystem of the multilevel plan process, local plans must be dovetailed with the plans at higher levels, so that integrated planning is promoted in the state economy. DecentralizedPlanning in Karnataka Although decentralized planning has been practiced in almost all the states of India, the case of Kamataka is looked upon as a unique and itesdng one for the following reasons: Karataka had a good start in decentralied planning, thanks to the keen interest of the government of arataka, which in the early 1970s created a Department of Planning manned largely by economists and supervised by the economic advis3r to the goverment. To impar professionalism to the decentrlize plan exrises, the university and college teachers who were specially tained fot the purpose were associated with the district bureacracy in the preparation of district plan documents. Although this arangeme has since been discontinued, subsequenty qualiied persons were drawn from diferent departmen to fill the positions of distict planmng officers who prded professional inpub to the district plan exercises. Second, Kanamak has a gbod record with respect to political decetraization, manifested first in the village panchayat system and now in zilla paisbad and mandal panchayats, which ensures an instituional arrangement for arlating people's needs and aspirations. It is this association of professionalism with plan formulation and of polit!cal den which makes decenalimzed planning an appropriate alternative for poverty alltviation projects. tsl~~~~~~~ - - 166 Abdul Aziz With the passing of the Karnataka Zilla Parshads, Talulc Panchayat Samithis, Mandal Panchayats and Nyaya Panchayats Act in July 1985, the earlier three-tier panchayat raj (PR) institutional structure was replaced by a systan comprising zilla parishads, taluk panchayat samithis, and mandal panchayats. These institutions came into being following the elections of January 1987. The members of the zilla parishads and mandal panchayats are directly elected, as was the case with the erstwhile Taluk Development Board and the viUlage panchayats. Of these members, 25 percent and 18 percent are elected from constituencies reserved respectively for women and for scheduled castes (SCs) and scheduled tribes (STs). The taluk panchayat samiti, like the earlier district development council, has ex-officio members and is only an advisory body. The zilla parishad has, in addition to elected members, nominated members consisting of the local MPs, MLAs, and MLCs with voting rights. As the apex body of rural self-government, the zilla parishad (ZP) is expected to supervise, coordinate, and integrate development schemes at taluk and district levels and to prepare and implement the district development plan; promote measures for kicreasing agricultural production and strengthening animal husbandry; promote the welfare of Schedule Castes (sc) and Scheduled Tribes (T); undertake the consruction, repair and maintenance of district roads and buildings; manage hospitals and dispensaries; promote a land-propagaing cooperativc effort; foster general and social education; develop fisheries and animal husbandry; distibute essential commod;ities; and prsue the scbeme of rural electrification. These fnctions are carried out under the supervision and control of the elected president, who enjoys the sttus of a minister of state, and the elected vice-president, who has the status of a deputy minister. With the estabfishment of the new PR insitutions, all the development deparments and various development agencies of the district formerly involved in formulatng and implementing the district plans, have been brought under the control of zilla parishads. The executive head of ZP is the district chief secretary-a senior (Indian Admiistrative Service (lAS) officer, who is assisted by one or two deputy secretaries and four to five subject specialists in the field of planing, statistics, and accounts. The traditional regulatory functions are now performed by a separate district administrative head-the deputy commissioner-who is placed slightly below the position of the district chief secretary. Two important merits of this rearrangement are that it ensures an effective mechanism for identifying and articulating people's needs and aspirations, and it helps coordinate the efforts and resources of the various district development departments. Perhaps even more important, the distr.ct chief development officers and other functionaries are now accountable to the elected representative, namely, the president of the ZP, and not to their superiors at the state level, as was the case earlier. The block development officer, who is assisted by the extension officers, is the executive head of the pancbayat samiti, and the secretary a nongazetted officer-is his countrpart at the mandal pandcayat (MP). Under the new legislation, the panchayat samiti is established at the taluk level, but it has only a supervisory, review, and coordinting role and not a planing one. It also does not have any direct mplementationpower and its members are not directly elected, but act in an ax-officio capacity, having been drawn from MPs and ZPs. The samiti is expected to advise MPs about the discharge of their functions, review their work from time to time, provide guidance, and coordinate their work where it relates to more than one mandal. The gram sabha, which constitutes the lowest rung on the ladder, is composed of all persons on the electoral roLls of ZP peraining to the village. Its faunctions are to prepare and promote village development schemes, organize village sanitation and drainage schemes, mobilize voluntary labor and contributions in kind or cash for the com_muity welfare programs, and assist the MP in implementing developmental schemes pertaining to the village concemed. Decetralized Planng and Poveny Alleviaion: The Kamnaaka Case 167 Owing to the nature and composition of the samiti and the gram sabba, these bodies are not given plannng and implementation functions. Therefore, for all practical purposes, the planning unit below the district is the mandal-cluster of villages with a population of 10,000 to 15,000. Every 500 persons or less is represented by one elected member in the MP, with reservations in favor of women, backward classes, scheduled castes, and trbes. A pradhan and a upa pradhan are elected fom these members to preside over the wcrt. of the MP. The functions assigned to MPs relate to sanitation and health, public works and amenities, agriculture and animal husbandry, welfare of scheduled castes,and tribes and backward classes, the preparation of plans for the development of the Mandal areas, maintenance of records relating to the survey of village sites and public and private properties, and to organizing and mobilizing the village youth for constuctive and productive purposes. A major function of the PR institutions is to fonmllate and to implement district and mandal plans. Schemes and programs that fall under the jurisdiction of the state and the PR institutions are ciParhy distinguished. Two criteria are used for identifying the PR institution schemes and programs: they can be planned and implemented at the district and mandal levels, and they must benefit the people living in the district and the mandal concerned. Accordigly, these institutions are g;i-n responsibility to plan and implement schemes in sectors such as agriculture, hotiulture, animal husbandry, cooperation, irrigation and groundwater resources, public health, education except university education, district and runal roads, small and cottage industries, welfare of SCs and STs, and so on. The plan process actually starts when the State Planing Department indicates to ZPs and MPs the fcial ceiling under which their annual plan should be formulated for the next year. As a first Step in thIs process, the vilage assembly identifies programs for the development of the village and presents them to the MP. On the basis of this feedback, the latter formulates plans relatig to local works such as water supply, sanitation, and rural communication and sends them onto the ZP. The ZP in turm prepares a district plan, including in it the MP plans, and sends it to the state departmnft heads and state planning department. At the State Development Council chaired by the chief minister, to which the ZP presidents and state department heads are invited, the district plans are discussed and an effort is made to ensure that they are consistent with and integratd with the state schmes before they are finalized. The state departments consolidate the district plan schemes and build them into the various state sector programs. The resultant draft annual plan is sent to the national Planning Commission, which is the final authority that determines the state plan outlay for the year. Depending on what changes the Planning Comission makes in the state plan outlay, if any, the district plans and their outlay are changed accordingly. The above description of the district and mandal. plan process irdicaes how the PR institutions are integrated into the state system of plannng. Poverty Alleviation Projects The data on poverty alleviatio projects in Karnataka presented in this section are drawn from a study carried out on planning in eight selected mandal panchayats of Bangalore Rural District (Aziz 1991). As aeady mentioned, the decentralized governments are charged with planning and implementing development programs in the various sectors such as agricultre, animal husbandry, public health, and education. Together, these programs are likely to promote oveal development of- the substate regions such as districts and mandals. By promoting sustained growth in subregions, such programs provide employment opporunites and thereby enable the poor to share in the benefits of growth. At the same time, the development of the subregions by itself 168 AbdulA& cannot alleviate poverty among the various sections of the community. These sections have different socioeconomic characteristics and, as a result, they differ in their access to the benefits of growth. Poverty alleviation projects must therefore be designed to fit the needs of the different groups among the poor. The decentralized governments have seen the merit of tailoring poverty alleviation programs to each of these poverty groups, defined on the basis of economic, social, and gender criteria. Some of these programs are designed by the union/state govemments but are implemented by the decentalized governments. There are also some that are exclusively designed and implemient by the decentralizel governments. The poverty alleviation projects that were formerly designed by the union government and managed by the state government but now are managed by the decentralized governments are (a) the wage employment Jawahar Rozgar Yojana; (b) the self-employment hItegrated Rural Development Program; (c) the sectoral programs that endow the poor with skills (training) and improve iheir living conditions by providing houses, lighting, smokeless stoves, and so on; and (d) the special component plan for SC and ST population. The 18 Percent Plan for SCs and STs, and the progr of smokeless stoves are designed and manag exclusively by the decentralized governments. Jawahar Rozgar Yojana The Jawahar Rozgar Yojana scheme introduced in 1989 is sponsored by the government of India. Under this scheme, projects aimed at creating durable assets are taken up in the rural areas to generate wage employment opportunities for the rural poor. This scheme merged the National Rual Employment Program and the Rural Landless Employment Guarantee Program- both of which were meant to provide wage employment to the rural poor during the off-season. The cost of the project is shared by the centai and the state governments in the ratio of 80:20. The funds are dcnneled to ZPs through the state and from there to MPs. Ihe shares going to ZPs and MPs are 20 percent and 80 percent, respectively. Out of the total funds allocated, 10 percent goes to spiilover works and 5 percent toward administrative expenses. The renaining funds are meant to be used for imple new projects. As for the MPs, they are expected to conform to the following norms while designing the proects and spending the funds: 35 percent of the total allocation goes to the creation of profitable assets such as shops and hotel buildings; 25 percent is for roads, culverts, and school and hospital buildings; 25 percent is for social forestry-for planting saplings and protecting them on roadsides, tank bunds, wastelands, and forestlands; and 15 percent is for residential houses and sites to be given to SC and ST households and for drainage in their colonies. It is stipulated that the works should be carried out by- the block development officer, with the assistance of two junior engineers, and that the wage and the material components should be in the ratio of 60:40. The daily wage rate for both the male and female workers is Rs 12.80,2 which includes the value of 1½ kilograms of rice (Rs 5.10) that is given to the worker. Each year, the secretary of the mandal panchayat prepares a list of willing workers who could be drawn upon for employment under the scheme, and this list is updated every year. It is stipulated that only the pest among the poor should be drafted for employment and that of the total workes 50 percent should belong to SC and ST households and 30 percent should be women. The actual works to be carried out are decided on by the members of the mandal panchayat Each 2. Sice April 1991, fte wag rate has boee revised pward to Rs 16. Decentralized Planing and Poverty AlleiMoion: The Karnataka Case 169 mandal panchayat member is asked to identify the needs of the village or village cluster that he represents and the list of such works, which is usually very long, is placed before the mandal panchayat meeting for discussion and approval. Depending on the availability of funds, works are selected on a priority basis through consultation and negotiation. Those projects that could not be taken up because the fuids were lacling, would be deferred to the future years. The works taken up and completed by the mandal panchayats under study are given in Table 9-1. As the table shows, housing for SCs and STs, school and shop buildings, link roads, bridges, and culverts are the most popular projects under this program. Social forestry and f irrigation are of low priority in the mandals under study. Integrated Rural Development Program The IRDP was first implemented in 1979 and is meant to assist rural households below the poverty line, which is defined as an annual income of less than Rs 3,500 at 1978-79 prices. The assistance consists of bank loan at 10 percent interest and a government subsidy that ranges from 25 to 50 percen of the assistance, depending on the social and occupational status of the beneficiary. The beneficiary is not given cash; he or she is given an asset equivalent to the value of the assistance. The "subsidy" part of the istaunce is equally shared by the center and the states. This amount is channeled to the ZP, which in turn credits it to the service area banks in the district in the ratio of the number of beneficiaries to be assisted in each area. This program was earlier designed and managed by the District Rural Development Society-a goverment organ functioning under the supervision of the distict chief executive. Since formation of the new panchayat raj system, it has been transferred to the mandal panchayats. Following the allotment of resources under this head of account from the zilla parishad, the mandal panchayat distributes pamphlets informing people about council meetings (gram sabbas) in each village or group of villages to which all the eligible voters, government functionaries, and people's representatives are invited. The government functionaries are the block development officer or the credit officer, bank manager, and the imple sectoral department officers. The people's rereseives-such as the pradhan and the members of the mandal panchayat, ZP members, and local leaders-are also invited to the meeting. This meeting is held to consider the list of households identified as living below the poverty line. The list is prepared by the secretary of the mandal panchayat through a formal household survey. The required number of persons are then chosen from among these households for program assistance. Generally speaking, a good deal of discussion and debate takes place before a beneficiary is identified in the open meeting. Once the benefcianes are selected, they are asked to indicate the type of asset to be given to them. This decision reached by the village council meeting and is based on the background and skills of the beneficiary, and also on the availability of local inras and market facilities. Next, the Purchase Commitee-consisting of the block veterinary officer, the block agriclture officer, the block development officer or the credit officer, the pradhan or secretary, and the beneficiary-visits marketplaces to purchase the asset. The asset is normally purchased with the full concurrence of the beneficiary.An effort is also made to avoid paying the seller an exorbitant and arbitray price. The Purchase Committee normally avoids buying the asset in the local market; rathcr, it goes to a distant marketplace to ensure anonymity and to avoid corrupt practices. Each of the mandals under Bagalore Rural ZP is allotted forty-three units of assistance, of which 50 percent are stipulatr be given to SC and ST households and 30 percent to women. The nature of the assets givLcn to the beneficiaries is indicated in Table 9-2. It is evident that the animal husbandry units such as milch animals, sheep, and pigs are high on *- .: . . - - - . :~~~~ .. .; 170 AbdulAzi Table 9.1 Works Undertaken by Mandal Panchayas under Jawabar RKA-Mar Yojana Name of the Maudal Naure of Work Taken Up and Completed Pwwchayat Harohalli Drainage, earthwork, shops, Nazeer Awas houses for SCs and STs, school buildings, and drainage Honganur Shops and hotels, drainage, school building repairs, construction of school building and Nazeer Awas houses for SCs and STs Lakshmisagara Nazeer Awas houses for SCs and STs, school buildings, dainage, link roads, culverts, and laying of strees in ST colowy Ka-naswadi School buildings, link roads, culverts and bridges, Nazeer Awas houses for SC and ST population, samtation in their colonies and repar of irrigation tank Yentaganahalli Social forestry, school and office buildings, drainage, village santation, Nazeer Awas houses for SCs and STs Kudur Shop buildings, pLay ground preparation (repair), link roads, drainage, desilting of irrigation tanks Nandagudi Nazeer Awas houses for SC and ST households, lidira Awas for SCs and STs, school buildings, link complex, social forestry, and school buildings Avathi Village sanitation, link roads, school buildings, community center, drainage, culverts, Nazeer Awas houses for SCs and STs, shop buildings, residental quarters, and social forestry the list, followed by petty businmess units and ardsan-related assets. Specia Componen Plm The projects under this program are targeted exclusively for the benefit of SC and ST households. The central government and the state both allocate funds for these schemes. These funds are transfenred to ZPs. which then fomulate and implement the projects under this program Decentralized Planing and Poverly Alleviaon: The Karnataka Case 171 Table 9.2 Assets (Projects) Provided under the Integrated Rural Development Program Name of the Mwdal Panhayat Asset (Project) Provided Harohalli Sheep and milch animals Honganur Petty shops, mich animal piggery Lakshmisagara Milch animal Kanaswadi Milch animal, bullock cart, sheep piggery, sewig machine, carpentry and embroidery Yentaganahalli Milch animal, piggery, sheep, petty shop, cycle shop, teashop, and bullock cart Kudur Milch aninal, petty shop, cycle shop, and looms Nandagudi Milch animals, bullock carts, sheep, silk charakha, petty shop, and cycle shop Avathi Milch animal and bullock carts in consultation with the MPs. The Special Component Plan is a sectoral plan in the sense that the various distrct sectoral departments design and implement projects that benefit the SC and ST population- The assistance available to them under this program consists of a free supply of agricultural implements; fertilizers; bullocks and carts; frit and vegetable seedlings; soil conservation assistance; an allotment of sites and houses; a supply of milch animals, sheep, and pigs; a special subsidy on long-term loans for asset creation given by land development banks; individual and community wells; and sheds to be used for living and work In the mandals under study, the assistance given consists of community and individual borewells, improved agriculntur implements, improved seeds, coconut and mango seedlings, sericulture seedlings, and sheds for cocoon rearing. The role of the MPs is limited to identfying the beneficianes and recommending their cases to the sectoral department heads. Twenty Percent Schm The Twenty Percent scheme is funded purely from the finances of manaa "anchayats and is designed to improve the living conditions of the SC and ST population. This scheme is so named because, out of the total resources of the MP, 20 percent is set aside to promnote the well-being of the SC and ST poaton, which incidentally accounts for about 20 percent of the total 172 AMid AZf Table 9.3 Assista Given under the 20 Percent Scheme NIame of the Mandal Assistance Given Panchayat Harohalli Free lighing, low cost housing, lavatories, drainage and uniforms for school children. Honganur Free lighting and drainage in Harijan Colony Lakshmisagara Free ligng and community building (a portion of the cost was covered by the mandal panchayat) Kanaswadi Mandal jyothi, houses, astra stove, drainage and iron plows Yentaganahalli Astra stove, mandal jyothi, compensation for accidentally bured houses, sites and houses, sarees and dhotis Kudur Street lights, tiles and beams, drainage, agricultural inputs such as seeds and ferdlizers, and repair of a drinkig- water well platform Nandagudi Mandal jyothi, tiles and bem, and sanitation Avathi Astra stoves, mandal jyothi, tiles and beams, drums on cultural and religious occasions, and house sites population of the mandals. There is what is called a Mandal Yyothi scheme, under which free electric lighting is provided and the nmndal panchayat meets the servicing charges for fixing the light poits while the beneficiary meets the recuring electricity carges. Second, there is a low- cost housing scheme under which the beneficiaries who own their own sites can build houses and get the cost reimbursed frm the nmndal panchayat. Otherwise, the mandal panchayat meets the cost of drainage in the SC/ST colony, the cost of uniforms for school children, and so on. Table 9-3 indicates the assistance given to SCs and STs under this scheme in different mandal pmnchayats. Free lghting, sites and houses, tiles and beams to enable the beneficiaries to build houses, and astra stoves (that is, smokeless stoves) are the most popular kinds of assistance in the mandals under study. Decenalzed Planndng and Poverty Alleviaton: Te Kamataka Case 173 Other Programs The local govemments also implement other schemes that benefit the mral poor in general. Thus under the Neg-Iu Bhagya scheme, the poor fanner is given free-of-cost equipment such as iron plows, digging equipment, fertilizers, and the like. Under the 100-Wells scheme, the small and the marginal firmers are given financial assistance for the digging of irrigation wells. Under another scheme, smokeless stoves are given to all categories of poor people free of cost. In this case, although the beneficiary bears the cost of bricks, labor charges, and other material, the mandal panchayat meets the cost of the cement stove and a smoke outlet pipe. There are also special development programs for women. In these cases, women are given trainig in tailoring, mushroom cultivation, baske ng, pemaking, seed preservation, and poultry animal management. The trained women are sponsored under the Differential Iterest Rate (DIR) scheme or IRDP for assistance in purchasing the required asset. Quite a large number of womeen have been trained under this scheme in the Honganur mandal panchayat, where the Pradhan is a woman. Some Observations The first point to note about the poverty alleviation programs is that some are sponsored by the government of Idia and some are generated by the zilla parishads and randal panchayats thmelves. In these programs the ZPs and MPs are given broad guidelines as to what types of projects are to be designed and managed for the benefit of the poor. These bodies take into 2---.unt the local needs, and the people's aspirations are articulated by their repesetatves, although theoreically it is the people themselves in the annual village council (gram sa) meeings who should identify their needs and aspirations. The usual procedure is as follows. The mandal panchayat isues notices and pamphlets about the biamnual gram sabha meeting mdicating the date, time, and venue of the meeting. Apart from the voters, the MP functionaries and people's representatives, the bank and the block development officers (if it is the IRDP) will be present. The development programs for the village or the cluster of villages concerned and the beneficiaries will be identified durng this open meeting. The consolidated list of projects and beneficiaries is disussed in the MP meeting. Normally, the number of projects suggested by the people and the names of beneficiaries identified, in the first instnce, is very large. Hence a process of sifting and choosing also takes place. This task is not easy because by now the members of the MP would have commimed tbemselves to some projects and some beneficiares. There are always complaints of favoritism and of some villages and some beeeficiaries being neglected. In such cases a compromise is usually reached giving some projects to each village in the current year aDd deferrng odters to a future date, and fixing the quota of beneficiaries to each of the members. Whether or not the projects selected and allotted to villages, and the beneficiarie identified and assisted through this process conform to the norms expected is a moot poin Nevertheless, this appears to be the only satisfactory way of articuatng people's needs and aspirations, as the whole exercise takes place within the framework of a political process. A related question that arises concers the apprpria of the priority structure devised by the d alized govenmes. Some critics argue that the local governments have given piority to building social and economic infiructure such as roads and buildings rather hn to the production-oriented projects and that such project and the amount spent on them is overwhelmingly large. Some also point out that since more money goes into these kinds of -~ -. --- , - - I 174 Abdul Aziz projects than in the production-c-iented ones, the people's represertatives have an opportunity to make some money from them and to patronize cleir friends, relatives, and supporters by offering work contracts. To be fair to the local governments, however, it may be that the people's representatives favor infrastructure projects because they first want to build the required human and physical capital before a concerted effort is put into building the productive capacities of the production- oriented sectors. It can also be argued that the local pressures may be such that the decentralized governments are, perhaps, forced to provide the people with social and economic infrastructure on a priority basis. Also. it can be seen that in the spending guidelines of specific programs such as Jawahar Rozgar Yojana, projects relating to social and economic infrastructure take the largest share. Hence, it is difficult to conclude that the design of poverty alleviation projects evolved by the decentralized governments is guided solely by the desire of the people's representatives to make money. Whatever the intention behind the priorities that have evolved, the decentralized governments seem to have provided help to the rural poor by ensuring wage and self-employment projects and also by providing other amenities to make their lives more comfortable. As for the management of poverty alleviation projects, conceptually, management is a process of translati the program plan into action. Ideaily, this process should extend the benefits to the poor in the most cost- effe-tive manner possible. If that is the guiding principle, then the function of the implementation process should be to achieve. interdepartmental coordination; avoid mismatching the scheme and the skills of the beneficinit.. avoid misidentif3ring target groups and niisallocating resources; and monitor progress in a way that ensures the asset is retained and not misused and the loan is repaid, so that the poverty alleviation project is sustained. Iplementation is conducted by a series of district departments. Therefore, if implementation is to be effective, there should be some degree of coordination among them. Before the new panchayat raj system was developed in the state, there was little coordination across different district development departments, primarily because each of them was directly under the state heads of departnents. In the new system, all the district development departments are under the jurisdiction of the ZP. As a result, the district heads of departments are now answerable to the ZP chief secretary and the ZP president. This. arrangement has provided a mechanism for coordinating the activities of the varicus district development departments. Whenever there is an interdepartmental coordination problem, it is resolved in the monthly meetings held under the auspices of the ZP. Earlier, the beneficiaries to be assisted under the IRDP were identified under the supervision of tle District Rural Development Society. The list of households below the poverty line, identified by the block functionaries and subsequently scrutinized and approved by the gram sabha, formed a data bank fom which the beneficiaries would be chosen. In somc cases, beneficiaries who were not identified by means of the household survey, were still recommended by the BDO and the bank managers or sponsored by the District Industries Center, especially for industry, service, and business projects. Obviously, the block development fuctionaries played an important role in beneficiary identification in the earlier system. The general complain at that time was that the block functionaries, not being in touch with the mwasses, tended to choose beneficiaries who were either friends and relatives of, or those sponsored by, the village landlords or local MiLA. Hence, there was scope for misidentification under that system. Although the block and bank functionaries still figure in the identification process in the new system, the MP members (people's representatives) play by far the dominant role. Since there is a miember representing almost every village, he will press to get assisaance to at least one individual from his village; and in all probability, the beneficiary selected would belong to a poor Decentralized Pklnning and Poverty Allevation: The Karnalca Case 175 household. Because the identification process is carried out in the open village assembly in the presence of voters and other members of the MP, he certainly would not like to take chances when recommending a case for assistance. In other programs such as Jawahar Rozgar Yojana, the Special Component Plan, and the Twenty Percent scheme, it is true that beneficiary identification is done in the MP meetings and not in the open village council meeting. But, here also, the fear of criticism from the opposition party and of losing face in the village community, in case they deviated from the norms, would force the people's representatives to be fair to the extent possible. Notwithstanding these practices, some still complain that MP nmebers show favoritism to their friends, relatives, and supporters. However, the aggrieved persons today live under the hope and assurance that they wil soon have a tum at obtaining assistance. Moreover, today there is a fair chance for all villages to get the projects they merit, in contrast to the earlier anrangement, when it was the benevolence of the local MLAs that brought projects to the villages. There was always a chance that some villages would be left out. Also, it appears that the poor are able to obtain assistance the village community is able to obtain services much faster now. Today the IRDP beneficiaries get their loans and subsidies much faster because of the pressure exerted by the MP members. Similarly, replacement of village street lightbulbs, repairs to drinking water handpumps, and other works are carried out more promptly and people's needs are met without delay. A case in point is the water scarcity following the collapse of the village drinkng well in Avathi and the submergence of the pump and motor. The villagers UgheIraoedN (surrounded the MP members) and demanded immediate action. The ZP chief secretary, upon receipt of information at the district headquarters, immediately bad a borewell sunk in the village. We were told that such a miracle would not have happened under the earlier system, for few people would have gone to taluk, much less to district headquarters, to press their case. Normally, beneficiaries tend to misuse the assistance by diverting funds for purposes other than the one prescribed. To discourage such activity, assistance is normally given i kind. Even in such cases, some beneficiaries misuse the assistance by selling the assets. Under the earlier system, the distance faor and lack of personal contact between the beneficiary and the government machinery (usually at the district and taluk levels) made it easier for the beneficiary to get away with this. But under the new system, the beneficiary lives in close proximity to the mandal funcionaries and the people's representaives are kept posted about the goings-on in the village. Any move by the beneficiary to sell the asset would easily be deteted. On such occasions, beneficiaries have been issued a waring and in some cases police reports have been filed. One beneficiary in Kudur UP is reported to have sold the asset (bullocks and cart) and purchased a television set from the proceeds. The mandal members who got wind of this swung into action immediately and helped the bank manager recover the entire loan that he had made. The fear of detection, especially by the MP members who helped the beneficiaries obtain assistance, is an important deterrent now. Besides, mandals have devised a procedure that makes it difficult to dispose of the asset, as can be illustrated by an episode involving the smokeless stove project. It was found that beneficiaries would either sell the smoke outlet pipe that is provided along with the stove or use it as a water outlet in their paddy fields. To ensure that the asset is not misused, some mandals give the smokeless stove only to those who have just completed constucting a house, and even in these cases the mandal is responsible for fixing the stove and the pipe. The beneficiary would be expected to bear the labor charges and cost of bricks and sand, and the mandal would provide cement and othtr materials. In the matter of loan recovery, too, the new system provides some degree of support to bank employees who go arumd to collect dues from the recipients of loans. More often than not, e i'EX,,, 4 0 :-U0f ,S 0 ^''' < N 176 Abdul Aziz MP members accompany the bank recovery officers-a phenomenon unknown under the earlier arrangement-and put pressure on the loanees to pay what is due. Unfortunately, it was found during our field trips that the loan waiver fever had spread to the village, and it had become more difficult for both the bank officers and the MP members to recover dues. The general impression among the public today is that loans from government and semigoverument organizations are not meant to be repaid at all. Although there are always a few cases of conscientious beneficiaries repaying loans, under the new mood, project sustainability through the recycling of fumds has become very doubtful. Nevertheless, the Karnatal experience points to several benefits of using decentralized planning as an instrument of poverty alleviation. In general, a decentralized system pays close attention to people's needs and aspirations. In Kanataka, thcse needs tended to be related to the economic and social infrastructure ( notably, roads, school and hospital buildings, rural water supply and sanitation) and housing for the SC and ST population. The community assets created in the process by and large meet the basic needs of the poor. Furthermore, the wage employment opportuimties generated through the community asset building benefit the poor directly. Therefore, it appears that dec ed plaming, generates projects that potendally benefit the poor more than the better- off sectors. Furtemre, the poverty alleviation projects are found to be managed better under decentalized governance than under cantral bureaucracies. There is also less risk of misidentifying the beneficiary and of the latter misusing the assistance. In addition, the prospects for retaiing assets and recovering loans appear to have improved, although there is scope for further improvement. All in all, decentralized peming and local participation could be an effective insument for alleviating poverty. 10 SURVIVAL STRATEGIES OF POOR HOUSEHOLDS: A COMPARISON OF KARNATAKA AND TAMIL NADU WITH EXPERIENCE FROM ASIA, LAIN AMERICA, AND AFRICA Abdd Aziz and Michael Bamberger 9ommawl In dteir eortto inprove their lot, thepooradopt several surva strategies. A knwledge Of these strategies will help policmakers formulate podicies to further improve the survval effort of the poor, frame a comprehensive strategy for povery alleviaton acceptable to the poor. and understand the contexts in which specific strategies are adopted by the poor. lhus chpter presents a brief account of the survival strategies of the poor, taken from field experience in Latin America cAfrca, Sow East Asiz and India. OscarLewus refers to a 'culture ofpoverty among poorhouseholds whereby they provide informal credit services and even share food, clodes, furniure, and other items. In Cuidad Guyana, Venezuela, such mutual assistance also inclde house-buiding and maintenance, and water supply. Such economic recprocites operate among both kin and nontdn households, although they tend to be stronger among the former. Among bladc households In midwestern ANolh Atmerican cties, such muta assisance tended to be confined to households that hd the ability to recdprocate (to give as well as receive assistance). Marshal SatUns refers to a 'continuon of exchange' or reaprocity and mentions several paerns of recprocity. bi some cases, goods are given free, wkth no eWectation of their return. In others, thy are given with no precondition for their reun. These are called 'genralized reaproaties.' When exchange takes place in such a way that each party aims to maximie its gains, it is called 'balanced reciprocity. ' If exckanges are condued without social or moral considerations, they are called 'negative reciprocities. ' Laissa Lomnisz found in a shanty town ouside Mexico City th people in dire circumstces pooled their resources and provided muta assistance. She explains these reciprocities with reference to four dimensions of distance, namely, social, economic, psychological, and physical distance and concludes ht 'the less the distance between households, the greater the chances of reciprocity. - Dani Kauftnan observed a special knd of reciprocty among Salvadoran households called contractariaim. In this case, famdies with no sady income or not enoughfor their basic needs agree to cooperate with each other. So there will be consa 'givers" and "receves " among them. Sone prudent "givers' may traser resources even when they themselves are short of them, hoping to get them back in times of greater need In some cases, the 'givers" and 'receivers' constantly change places as theirfortnes change. 177 : - - -:-: - ; - - -=~~~~~~~~~~- 178 Abdul Azi wnd Michael Bamrberger A specialform of reciprocity in the Philippines is called Uang na 1, meaning debt of gratitude. The receipt of assistance creates a "social debt, " which remains an obligation of the receitving family urntil it is repaid- sometmes even one or two generations later. If they are not recaprocatei4 the receiver will sffer hiva or shame. Malasaldt (or chaity) is anotherform of aid given to indigent persons and dterefore need not be returned. Femle-headed househiolds, because they lack a steady .icome, have limited numbers of wage earners, and lack access to formal credit markets, tend to rely particularly heavily on informal support networks. Assistance and reciprocity involves the exchange of noney, food, and clothing, andfrequently the transfer of children to households with a higher income or with better access to schools or places of employment. In India, the survival strategies identifledfrom case studies in Karnataa and Tamil Nadu are of special interest. Exmples of these strategies in the market contea, at the individual level incle casud workers who prefer piecework in order to maximize earnings; poor household workers who desire payment in kind to counter the effects of iladion, andpoor workers like maid servants who prefer regular, though lowpiud, work. Examples of strategies at the household level are dunng drought or distress, even women and children offer zo workfor wages to supplement the low income of the male-earner and mugrate to big ities in search ofJjobs. Informal networks in the nonmarker context are those built by reltives and friends, nezghbors, former enployers, political workers, and volwary agenaes. These iformal netvorks are of great importne in underdeveloped counries where governments with limited resources are unable to provide even the minimum level of social welfare to the poor. lItroduction In his Nobel lecture, T. W. Schultz (1980:37) observed: "Poor people are no less concerned about improving their lot and that of their children than those of us who have incomparably greater advantages. Nor are poor people any less competent in obtaning the maximum benefit from their limiited resources.' If the poor are capable of managing their poverty problem, a question tat arises is, What are their survival strategies? Understanding the survival strategies of the poor is extremely important to policymaking on three counts. First, it provides feedback to the policymaker who wishes to formulate and orient his policies toward augmening the capabilities of the -poor to survive the way they do. Second, it provides a fiamework for building a comprehensive strategy of poverty alleviatior-a strategy that draws from the experience of the poor themselves, which is most likely to find acceptance among them. Such feedback to policymaking is extremely important and timely, because many poverty alleviation programs iplemented so far have not been all that fruitful. Th!rd, it also provides us with the contexts in which the poor seek to follow specific strategies; and these contexts would provide clues to the root causes of poverty-an understanding of which would be a boon to policymaking. Much of what has been written about poverty talks about the "vicious circle of poverty". It is argued that the poor are incapable of generating their own fiancial resources to build an income-yielding asset (physical capital), in view of their low income levels. What is more, even with a given level of capital, they cannot generate adequate incomes because of the presence of surplus-extracting individuals or institutions under whom the poor are obliged to work. Therefore, it was thought that any strategy for alleviting poverty ought to help the poor build their own assets to come out from under these forms of exploitation. Surviva Strategies of Poor Housdeolds 179 Although such a perception about the genesis and alleviation of poverty appears to have some merit, it also rmsses certain other points that become clear from the experience in the fieid. Thus, it may be logical to treat poverty among agricultural laborers, tenants, small cultivators, and urban petty traders and hawkers as the product of their low resource-endowment, on the one hand, and the exploitative mechanics of the market, on the other. But poverty and extreme destitution are also found to be the result of noneconomic factor. such as death or sickness of the family breadwinner, his involvement in legal cases, imprisonment, addiction to liquor and gambling, exposure to political and communal violence ana such other factors that cause households to slide down the poverty line. Here, poverty does not necessarily emerge as the product of a lack of capital or presence of a surplus-extracting system. There are circumstances that cause even those households endowed with a sufficient amount of resources to become poor. Therefore, the mere provision of capital and market regulation in such cases will not always alleviate poverty. The strategy then will have to be different and also specific to each set of cases in point. On the question of what these strategies should be, the answer is that we should undersd the strategies evolved by the poor themselves under different circumstances and try to build on these strategies to make them more effective. This chapter presents a brief account of the survival strategies of the poor drawn from field experience. It opens with a review of the experience of countries in Latin America, Africa. and Southeast Asia and thien considers the Indian context to complete the picture. The chapter concludes with a discussion of the implications of these experiences for the formulation and implementation of poverty alleviation policies and programs. The Experience of Lati America, Africa, and Asia Mual Assistance The importance of social networis for the survival of their individual households has been a topic of interest in the ethnographic literature on the urban poor since the 1950s. I Oscar Lewis's pioneering studies of poor city-dwelling families in Mexico City, Puerto Rico, and New York's ghettos in the 1950s and 1960s focused on the "culture of poverty" in many underdeveloped areas. In describing the characteristics of this phenomenon, he noted that groups of poor households regularly provided informal credit services for each other's support. Clothes, food, and furniture in the domestic domain were shared by families actg in collective interest. Lewis noted that the social, economic, and affective ties between Ida- and nonldn-based households in urban barrios were extremely meanngful. In the abseoce of formal support from fir Acial istitutions and in the face of irregular remmeration from cycles of employment and unemployment, the extended family or fictive kin ties between households provided insurance for individual households that their basic needs would be met. In his opinion, recourse to relatives for economic, social, and emotional support reduced the risks that households would be subject to dissolution born of poverty, lack of work, and the tensions inhering in these economic crises (Lewis 1965, 1968). 1. This section draws heavily on Swvival Sraegies and SiWporT Nemo*s :MAn A4ndawopogkl Perspecie bi Scot Parris (1984). Paurs was the anhmpologist on te study of surviv sres of low-inome hiusehIds m Carrgna, Codombia. to u*ich frequent refeec wil be made. 180 Abdl Aziz and Midca Bamberger A host of other scholars, while not accepting the notion of a "culture of poverty" as an explanatory device for the socioeconomic characteristics Lewis reprted, also documented the occurrence of the network support system in a wide variety of settings. Lisa Peattie (1969) analyzed forms of intensive social cooperation in an urbanizing barrio in Ciudad Guyana, Venezuela. She showed that mutual assistance was not restricted to the domais discused by Lewis, but embraced tasks in house-building and maintenance, supplying the barrio with water and constructing other infrastructure. Carol Stack (1976) described similar pattems of intense mutual aid among poor black households in North American cities of the Midwest. In both cases, however, households that were unable to sustain expectations for giving and receiving aid such that all parties benefited, were cut off from, or failed to develop, reciprocity networks. Reciprocity Relaionships About the same time that Oscar Lewis was detailing social, economic, and affective exchange relatioxLahips, Marshall Salins published a theoretical essay accounting for exchange relationships in general (1968). Sablins believed that patterns of transfers between households or larger social groups could be expressed along a continmum of exchange, or reciprocity. Ae one extreme of the contimnum, commodities transferred to receiving units are given as donations, with no expectation of return. Such givers are providing the transfers as expressions of altuim. Moving along the contimnum away from altruism, some 'givers" transfer resources to others with no specifications or schedule for retum. Resources are given with the Un ting that if and when they can be repaid in some form, they will be reciprocated. Such exchanges exemplfied "generlized reciprocity," typified by the resources parents give children. In exchange for raising them, parents expec their children to help them at a fture time. Moving further along the condmtnu, exchanges between socal units reach a point where each party aims to maximize its own gain or utility from the transaction. Such "market-oriented" exchanges are examples of balanced reciprocity." Beyond the point of balanced reciprocity, exchanges are no longer limited by social or even morl considerations. Parties anempt to etract commodities or goods from each other by force. These exchanges can be considered instances of "negative reciprocity." Sablins's account of patterns of reciprocity influenced a generation of socal scientists, who attempted to apply and refine the cOntium in particular ethomgraphic and socioeconomic contexts. Looldng primarily at social relationships in Mexican communities, Enc Wolf (1965) agreed that many relations of kinship and fiendship, and patron-client ties, could be understood by exmining the kinds of reciprocities among them. He believed that economic ties between kin, especially genealogically "close" family members, were reinforced by cultural sanctions binding relatives to help each other on a moral basis. Economic exchanges between nonlkin of similar socioeconomic status, especially friends or fictive kin (conqxdr), were not endowed with the same moral emphasis as those found between blood -elations. However useful the ties were betwen economically coopeaing friends and compadres, they were apt to be subject to economic and social pressures not felt to the same degree by kin, and thus more likely to break apart. If or when these nonkdn households detmined they would be better off economically, socially, or affectively without their exchMange partners, they severed ties with them. A similar pattern occured with patrons and clients. By virtue of their diferent hierarch status, both groups sought to receive and provide services of value to the other. Each group SWarvwu Srmegiesof Poor Household 181 assessed the value of the services rendered to the other oveir time, and either strengthened, continued, weakened, or broke off the ties between them. One of WoWs followers, Larissa Loumitz (1977h, studied patterns of reciprocity among kin, compradres, and neighbors in the shanty town called Cerrada del Condor outside Mexico City. Like Oscar Lewis, she found that economic necessity prompted many Mexican households to tur to each other for economic and social assistance. Low remuneration from work, which itself was fluctuating owing to high rates of unemployment and underemployment in fth formal and informial sectors of the Mexican economy, caused many members of Cerrada del Condor social networks, especially those based on kinship ties, to pool their resources to keep all households involved afloat. The mutual aid provided in cash, goods, services, and emotional support buttressed individual households whose incomes regularly fell short of meeting their basic need--. The primary motivation in this support system was thus to reduce the risk of indigence or household dissolution due to lack of resources. As Lomnnitz (1977: 190) puts it, These social networks are structure which maximize an important commodity, namely securty.... Whenever the physical or social survival of a group is threatened, people mobilize their social resources for whatever economic value they may afford. The Spani.sh saying "Hoy por ti, manan por mi" (your turn today, my turn tomorrow) summarrizes the principle of reciprocal exchange in a situation of balanced scarc-ity assumed to persist indefinitely for both persons. In accounting for the presence of reciprocity relationships among Cerrada del Condor households, Lomnitz indicated that four dimensions of distance or closeness could help explain whether regular exchanges would take place. These four dimensions are formal social distanc, economic distance, psychological distance or conjianza (that is, mutuWaltust), and physical distanc. Taken together, the less the udistance are" between households, the greater the chances that the households are cooperating in reciprocal fashion. L-ow-income households that are economically "close" in levels of income are more likely to enter into exchange relationships than those of signific-antly different earning pattern, given the former's need to find partner to help tide them over in difficult periods. Lomnnitz notes, as Wolf had previously, that pressures to redistribute income and remain socialy and emotionally close can and do become wearing on the households involved, If the households involved are not bound by true kinship, the ties are likey to be subject to suspicion that each is not doing enough to support the other, that is, that the households are concerned more about their own welfar than that of their exchange partner. These affectively loaded relatonships, involving drinking partnerships between male household heads, often cannot sustain the initial pattern of intense male spending and subsequently dissipate or dissolve. Linked to these firt two "distance" feature is psychological closeness or confiwaza. This mutual trus must be present to some degree between households, or they will not risk trading resources with each ofthr. Conflanzza, however, is also subject to srgteigor weakening, dpning- on the behavior of the exchangers Finally, Lomnif suggests that the greater the spatia neares between households, the more likely they are to reciprocate with each other than with those who are physically distant. As such, neighbrs are more likey to engage in some kind of transfer relationships than are people -who do not live nar each other or in the same barrio.- LomnitZ alSO found that socioeconomic and affective ties between parents and children and between siblings are often stronger than those between husbands and wives (legally married or not), and other affine (ties based or traced through marrage). While nmarages may last for 182 AbdMd Aziz and Midael amberger decades, kin ties last forever, even if these are subject to lapses in frequent or intense social contact. Both because of the closeness of many sibling ties from childhood onward, and because of the patterns of divorce, Cerrada del Condor women are liable to rely on their brothers (and hence, the rest of the extended family) more over time than they do on their husbands for long- term economic and social assistance. Spouses do not have identical relationships with other conjugal pairs in their social networks: men appear to emphasize the importance of ties to their cuates (closest friends, drinkng partners, and the like), while women emphasize ties to their own extended family and to female helpmates living near them. The upshot of these differential patterns is that reciprocity networks are subject to change over time, as men and women's investments of time, money, and support to them changes. Inspired in part by the redistributive theory of social justice (Rawls 1973), by the findings of Lomnitz (1977), and by World Bank studies on transfer activities among Salvadoran households aAufmnann and Lindauer 1980), Dani Kaufmann (1982) has expounded a theory on a particular form of reciprocity he terms "contractarianism.* In Sablits's continuum of reciprocal exchange, Kaufmann's conuctarianism wouldcomebetweengeneralized reciprocityandbalanced reciprocity. As Kaufiann sees it, households engzge in contractarianism as a strategy to reduce the risk that any participating household will fail to have its basic needs met at any particular point in time. They devise an implicit insurance system against having the income of member households fall below a certain tbreshold. He does not specify any necessary basis for households to enter into "conactarian" relationships, although it is apparent that segments or the whole of an extended family are likely to compose this form of reciprocity network. Conactaian households, in effect, provide a safety net for each other- In the absence of a secure, steady income from formal sector sources, contrian units recognize that their fluctating sources of income, usually from the informal sector sale of homemade goods, resale of other goods, simple household enterprises, or sale of services, are not apt to provide for all their needs or those of their networks over time. As a result, they agree to cooperate to see each other through difficult times When one or more households have a surplus of resources, they in effect promise to transfer a portion of these to households in their network experiencing shortfalls at that time. These latter households may have helped them through their own financial shortfalls earlier. In retumrn, the latter households will send resources back to the "gi-vers" when their incomes permit or when the "givers" need such resources to secure household necessities. Put another way, Kaufmann sees poor households acting as constant "givers" and "receivers" of resources among members of their networks. Mis theory is general enough to include identifiably different types or subtypes of contractarianism. Some households, for example, might serve as constant "givers' to members of their network who are in the position of constant "receivers." The givers in this subtype acquire enough resources to provide for their own household's needs and part of the needs of others. The givers acknowledge, however, that their mcomes might not always cover their needs; thus they are giving to the "receivers" as an investment with the expectation that the receivers will, if able, return transfers to them. In this subtype, conrctarian exchange is a kind of genealized reciprocity. In other instnces, "given and "receivers" will be constantly changing places. Households giving resources to another in one month may require resources the next and may receive transfers from the "original" receivers. The periods of giving and receiving may last for considerably longer periods of time, however. This subtype of contracariani ppears to embrace the concept in its tnuest or fUllest form, followin the Spanish proverb "hoy por ti, manana por mi." SurviW Strategies of Poor Houseolds 183 In stil other instances, households may give and receive transfers within the same period of time. So frequent are the incoming and outgoing exchanges that the households may be termed transfer "users" (see also Bamberger and Parris 1984:v). Households in this subtype seek to reallocate resourcs among the network to even out income among the constituent households, and they may be termed practitioners of "balanced contactrianian" This form of reciprocity resembles a redistnbutive policy among the households in the network. While Kaufman himself does not identify these subtypes of contractarianism explicitly, he does note that different kinds of social networks will have their own perceptions regarding when and how it is appropriate for them to help, or receive help from, each other. These different perceptions will be related to the networks' demographic composition (numbers of persons in the households, nmbers of households with f;male heads, and so on), their employment and income patterns, and other characteristics such as confianza. Depending on these factors and others such as frequency of visiting, residential closeness, and extent of the day-to-day knowledge of what the others need or want, households will be in a more or less good position to "earnmk" transfers for particular uses. In a sociological study of low-income urban households in the Alagoas District of Salvador, Brazil, William Norns (197T) brought additional ethnographic dimensions to the contrbutions made by Sahlins, Wolf, and Lomnitz although some of his conclusions differed from theirs. Norris's investigation also confirms that Kaufmann's treatnent of conhtactarianism can be found "on the ground" in various forms among households comprising reciprocity networks. The same was found in social support networks in the Southeast Zone of Cartagena (Bamberger and Parris 1984). Norris observed that diferent forms of the reciprocity SahlinS described were in fact in operation in the Alagoas. Some households, among the poorest in the area, are constant, long- term recipients of transfers from otier family members, neighbors, or fiends. Many of these households were headed by women. As Norris puts it, transfers tended to flow consistently from the less poor to the more poor, often over long periods of tme. In the ternninology employed above, networks evinced generalized reciprocity. This pattern does not mean, however, that no form of contractianism exists among Alagoas households. Reciprocal exchanges occur within segments of household networks, that is, between men or women and tansfer parmes of the same sex. Patrons and clients, moreover, exchange services of value to each, even if the transactions reflect the differential status between the two groups. The clients in Alagoas are usually women, linked to both female and male patrons. Somewhat contrary to Wolfs and Lomnitz's findsigs, Norris found that conjugal ties and ties between friends appeared to be more enduring in the Brazlian case than in the Mexican ones. Absent fathers continued to support their estranged children, at least in economic terms; although poorer people in Norris's sample had fewer stable marital-type relationships than better- off Alagoans, this result was as much the product of the death of male partners as it was of the abandonment of women. Noms documted the extent to which husbands and wives shared their network ties (thereby creating 'zones of inclusive density") or pursued these on their own (and thereby generating independent sector zones), and analyzed how these zones of density and independent sectors were shaped by the male- and female-headed households' mode of participation in Salvador's economy. Stifying the results by income (roughly above, at, or below $40 per month), he proposed the seven types of household-network linkages given in Box 10. 1. 184 Abda Az= and Midwcel Bamberger BOX 10.1 Nomis 's Cassificaion of Networks by Economic Level of Households and :h Basis of the Network 7les Upper Strata: 1. Male breadwinner household with kinship ties predominating 2. Male breadwinner household with political ties predominating Middle Strata: 3. Two-breadwimner household with kinship ties predominating 4. Male-breadwinner household with religious ties predominating 5. Two-breadwinner household with religious ties predominating Lower Strata: 6. Female breadwinner household with neighborhood ties predominating 7. Female breadwimer household deficient in network ties Surce: Norris (1977). As Norris observed, the upper-strata households most closely resemble the traditional working-class patern found in both developed and developing countries, where a man's steady work and his (and his wife's) predominant kinship and political ties largely determine network activity. In the middle strata, two breadwinners are usually necessary to keep households afloat, again using primanly kinhip and this time religious-based ties as links to a network. The bottom straa point to a pattern observed by Lewis, Stack, Rainwater, and numerous other anthropologists and sociologists workig with low-income Afro-American and Ibero-American urban communities: Female-headed households, however successful in engaging with others to share resources, still suffer economically by comparison withr male-headed households. The worst-off households are those headed by women without the ties to generate reciprocal exchanges with other units. Whfle Norris does not document how widespread the generl "downward pattern of transfers is from better-off to less-well-off households, his work demonstates how diverse household network strategies are among the lower strata of urban populations, and how valuable contacts with networks are in hmselves and as vehicles for economic, social, and affective security. Other Forms f RezprocUy and Assistance: ug ng oob and malasakit A different form of reciprocity reported from the Pbilppines is utang na loob, or debt of graitude. As Reforma and Obusan (1981) describe it, such social indebtedness occurs when a household receives assistance in the form of goods or services from another linked to it by kinship or by some other form of close social relationship. The recipient is compelled by wag na loob iterally, a debt inside oneself) to return the assistance or favor with interest, to be sure of not being in the original donor's permanent debt. SuWieal Staties of Poor Horholdr 185 As noted above, repaying utang na loob can take years. In one case described by Reforma and Obusan, a young man who was taken into the home of his aunt and who received care and educational support while he was studying, many years later sought to raise money himself and from other relatives for a surgical operation his aunt needed. Even though the woman involved had children of her own who were willing to help pay for the operation, the nephew would feel culturally-sanctioned hiya or shame if he did not use this opportunity to reciprocate his aunt's help to him. Utang na boob may also be incurred among households not related by kinship. In another instance recounted by Reforma and Obusan, a lawyer helped a couple obtain a residential lot. Years earlier, the wife's father had assisted the lawyer financially in his own educational career. When the father died at a relatively early age, the lawyer went so far as to transfer the wmg na lob to the man's children, providing both monetary assistance and aid in the form of semces. Malasakt in the Philippines resembles the concepts of altruism or charity found in other parts of the world. Donors give malasakit to members of their family or nonkdn because of the latter's less fortunate socioeconomic circumstances. Recipients of mabasakit (literally, "commiseration') are not expected to make restitution to the givers. In a case of malasakit among kin-linked households described by Reforma and Obusn (1981), an extended family regularly provides assistance to one of its member households. The latter household is composed of a marginally employed man, his chronically ill wife, and their children. The wife has always been the "baby' of her family of orientation; though married, her parents and siblings continue to provide her with regular financial aid and services becase they recognize that her husband cannot by himself generate all the income needed to support the fianily. Incapable of returning even small financial favors to his wife's family, the husband suffers hiya and avoids his in-laws. As in many other social situations, including those found in developed countries, the acceptance of nzalasakft on a continual basis can foster both dependence and a sense of shame in the receiving household. Gender Issues in Socia Network Analysis A growing number of writers have emphasized the importance of social networks for female-headed households and for women in general. Because women fqently have less access to formal labor-market, political, and finmcial networks, they frequently rely more on informal networks. Clarke (1993) in her review of household economic strategies and support networks in Kenya, documens the extensive social and economic functons of informal social networks for women (Box 10.2). A Colombian scholar, Nohra Segura de Camacho, emphasizes the economic struggles that proletaianization and rapid urbanization bring to poor households, whose members must launch daily efforts to provide the resources to reproduce themselves socially. Such households thus tun to others, especially to those linked to them by kinship, to generate the income needed to sustain them collectively. Shared poverty, however, still makes the family unit open to instability: In general, the low levels of remertion, the instability of employment, the conditions of work and life associted with poverty oblige (poor urban) sectors to develop economic survival strtegies, through, for example, the greater participation of members of the [extended] fmily in the generation of income; [such] economic precariousness demands a degree of solidarity and confluence of eff-orts greater than that of other social sectors, 186 AW Aziz and Micdael Bamberger in that (the former) depend upon each one of (their members) as a generation of income with the aim of maintaning the family group, but at the same time these very conditions of poverty ac! (in such a way as to) make the stability and the internal cohesion of the family groups more vulnerable. (Segura de Camacho 1982:88) Segura de Camacho argues that women bear the brunt of keeping their own households together as a unit. In her view, both men and children place less priority on their household's stability than do women. While some of men's socioeconomic priorities are aimed at productive purposes such as economic support of the extended family, other priorities take the form of extrafamilial personal consumption, such as drinkdng or gamnbling. In situations like these, especially where male household heads fail to generate the income need to sustain the household, male control of the domestic domain tends to dissipate. As a result, women often take over the role of household head to keep the unit intact (Segura de Camacho 1982: 87-88). Other Latin and North American feminists also see women as the principal motivations of interhousehold cooperation for mutual economic sustenance. Citng a host of case studies from all over Latin America, particularly Mexico and Brazil, Mariane Schmink asserts that extra- domestic exchange patterns observed empirically among networks of poor households throughout the continent serve primarily to substitute for goods and services that higher-sector households can afford to purchase outright. Women are the predomiant sources of such reciprocity (Schmink 1982:17). There is significant evidence that over time women are obliged to assume greater responsibility for their household's support owing to the absence of steady male providers (Sandrunii 19 -:83). It is in these cases that female-headed households may cooperate economicaily in the forms observed or reported by Stack (1974), Sclmink (1982), and Segura de Camacho (1982). Interhousehold Transfer of Persons The above discussion has emphasized forms of transfe of cash, goods, and services through interhousehold networks, but people may also be 'transferred" through these networks. Some scholars believe, moreover, that households themselves can be considered the product of antecedent distributions of persons, goods, and services in social networks traced back over time. In other words, households are shifting coalitions providing vital functions for groups of people dispersed m different dwellings, all of whom recognize some form of social and economic responsibility toward the rest of the coalition (Peattie 1981; Nelson 1978). The reasons for transferring persons temporarily or permanently to different households in social networks have been accounted for in various ways. In one study from El Salvador, children were frequently moved from one household to another both to ensure their proper care by persons able to look after them and to free parents or guardians to pursue remuetive work (Nieves 1979). In Kenya, Nci Nelson has theorized that children are often sent to live w.:h relaives in rural areas so that the children's mothers might be permitted to work with fewer responsibilities of child care (1978). Bamberger and Parris (1984:40-42) also observed that a redistributive mechanism appeared to be at work among at least some networis in the South East Zone (SEZ) of Cartagena with respect to child care. Household heads or co-heads "adoptig" additional children into their homes fequently stated that the chdildren would have suffered neglect if allowed to continue living in their former domiciles. These households did so even though they were raising or had raised children of their own. Those adopted children, however, were a potential source of added domestic labor for the receiving household, as well as representing Survival Straregies of Poor Households 187 BOX 10.2 The Social and Economic Functions of Informal Social Networks for Women in Kenya * Women beer brewers in Mathare Valley are largely cut off from relatives and rely heavily on networks to provide police protection and child care, to assist in the acquisition of raw material and production, to mobilize credit, and to attract customers (Nelson 1979) * Women who run a business together sometimes pool their resources to form a household composed of nonkin work mates (Obbo 1980). * In Mathare Vallcy in Nairobi, some of the women who head households send their chidren to live with their maternal grandparents, both to provide help for the gradparents and to ensure that the children have a safe place to grow up. This also frees the mothers from child-care responsibilities so that they can remit more money to the village (Nelson 1978). * The tradition among many Kenyan tnlbes is to place children in the families of relatives or tnrbesmnen during times of economic stress Source: Clarke (1993). potential wage earners for that household's benefit in the future. Yet children are not the only persons to come into or leave SEZ households. Some units took in working-age "strangers" because the latter needed a place to live in excbange for performing tasks or earning income for the household. Other households sent out' members, primarily to work (and thus remit income) and less often to study. On balance, however, persons residing temporarily in SEZ households generated less income for the host households an did members leaving SEZ households for areas with greater employment and incomen-generatig opportunities. Two trends, in short, appear to be at work in the transfer of pesms in and out of SEZ networks. People, especially children, are on the one hand brought into households able to care for them, if not to prepare them for engaging eventually in productive work. There does then seem to be a strong sense of community solidarity in loolkig after cbildren's welfare. On the other hand, being a zone of few lucrative income-generating enterprises, sigificant numbers of members of SEZ households leave the area to seek jobs elsewhere. Many go to Venezuela or other parts of Colombia such as Barranqilla, where employment opporimities appear to be better paying hn in Catgena (Bamberger and Pauns 1984). Estmating the Magnitude and Determinants of Network Transfers Some studies from Asia, Africa,and Latin America have attempted to estimate the magnitude of interhousehold transfers and their quantitative significance for recipient (particularly low-income) households. It is clear from these studies that interhousehold transfers mrke a 188 Adi Az and Micdh Bmbgger significant contribution to the income of poor households in countries in all regions. For exmple, in Tanzania (1955) 20 percent of anmual cash needs of rural areas in southern Tanzania were cover by cash brougbt by returning migrants; in Kenya (1974) the poorest 20 percent of families received 41 percent of their income in the form of transfers; in Idia (1976) remittances represented on average 6.5 percent of the income of rural areas; and in Cartgena, Colonbia (1980), 50 percent of the lowest thi of households received transfers that represented on average 41 percent of their household income. The sties are based either on existing national survey data or on specially conducted surveys focusing on particular comnumities or regions. Many of the national surveys contain data on income and expenditure patterns of individual households so that statistical analysis can be used to examine relationships between household characteristics such as income, number of breadwinners, number of school-age children, sex of household head, and other such details and the magnitude of transfers given and received. An example of this approach is Cox and Jimenez's (1989) analysis of the Peru Living Standards Survey. After using multivariate analysis to assess the magnitude and detrminants of transfers, they then compare the effectiveness of private tasfers and government programs as ways of providmg security for vulnerable groups such as the old, young children, and the sick. Knowles and Anker's (1981) study is- another example of the use of national survey data. Odter studies use more aggregated data, which only permit an estimate of the total amount of transfers between Darticular regions (or between urban and rural areas). Most specially designed surveys focus on one particular community or subregion and collect data at one point in time on household social and economic characteristics and on the magnitude, soures and destination, determinants, and utilization of transfers. One of the most comprehensive studies of this kind was conducted in the South East Zone (SEZ) of Cartagena, Colombia. The study was conducted at a time when a large squatter upgrading project was being iniated and its goal was to descrbe in detail the characteristics and detenninants- of interhousehold transfers and to assess how these transfers might be affected by, and might affect, the implemention of the housing project. The project, which imvolved upgrading roads, drainage, and water supply in a series of low-income communities requred fmilies to invest considerable sums of money (some of which could be obtained through loans from the project) to upgrade their houses. Approximately 35 percent of all households in the Zona Sur reported having received a monetry transfer durng the month prior to the survey. When transfers in-kind are included, almost one-half of the households received some type of transfer. For the lowest income quintile, transfers represented 51.6 percent of total household income for female-headed households and 40.3 percent for male-headed households; and for the second lowest quintile, 32.2 percent and 18.9 percent, respectively. The Indian Expience Indian society is the product of a well-knit social and kin-relation orgaic body with a long and enduring tradition of supportig all social strata. Like other societies, it has developed informal support systems for the survial of the poor. Although there is no special reason to expect Lndia's support systems to be drasticaUy differnt from those identified in the above paragraphs, it is our intetion here to look for some unique cases, if any. As part of this exercise, at the end of the chapter we present some brief case studies of selected households from SurvWvl Smuegies of Poor Households 189 Kamataka and Tamil Nadu and draw from them some broad inferences on the nature of the support systems there2 The area chosen for fieldwork is the slums in Madras City and Bangalore City, and a village in Kamataka. A brief questionnaire especially prepared for the purpose was conducted among the selected households representing the following categories: (1) migrant, (2) female- headed, (3) destitute and old, and (4) normal poor households. Although some quantitative data were collected from the respondents, we felt that it would not be meaningful to carry out rigorous statistical analysis of these data considering the small number of respondents covered. Our research effort was primarily directed at identifying and illustrating the survival strategies of the poor households rather than statistically verifying the tendencies implied in the behavior pattern of these households vis-k-vis poverty. The case studies presented in the appendix to this chapter point to two broad levels at which the poor would adopt survival strategies: one at the individual level and the other at the household level. Further, at each of these levels the strategies evolved by the poor were deployed both in the market and the nomnarket contexts with a view to augmentig (a) earings and (b) capacity for eaming more. We may now illustrate these strategies by drawing on the evidence provided by the cases presented. nhdividual strategies have been defined as those that are employed by the poor in their individual capacity within the constraints imposed by the socioeconomic setting. Given that these constraints cannot be released by the individual poor persons, they nevertheless adopt a course of action that they think will malk the best out of the prevailing circmstances. Thus a poor casual worker like Krishna (case I1) may prefer to and insist upon worldng on a piece-rate basis instead of a time-te basis and, with intensified effort, he may seek to maximize his earnings. The poor household workers may also like to get in-kind payments at least in part to beat the continually rising prices and falling real incomes. This is more common among women who work as maid-servants. Also, to judge by the prefrence shown by Jaya (case 9), who wished to work as a maid servant with evn lower wages than as a casual worker with higher wages, the poor seem to be iterested in maxmizng the annual rather than the daily wages. Work tbroughout te yFx, the associated nmonetary benefits like food and used clothes, and the moral support of the employer get into the utlity index of the fenale-headed household worker and induce her to prefer the regular low-paid household work to the uncertain high wage casual work. Household strategies also evolved within the constraints imposed by the. socioeconomic conditions. One important strategy is to vary-albeit increase-the work participation rates of the members of the family, especially of women and chdildren. During times of stress, drought conditions, and emergencies, women and children enter into the labor market as wage earners with a view to suppleme_n family income. The cases in point are Raju's wife and daughters (case 1), Mary (case 4) and Elisa (case 10), who, when the main breadwinner's wages were found to be inadequate, dedded to take up work. Of course, in this procss, the burden of earning additonal income for the family will undoubtedly fall more severely on females than on males! But the idian social system seems to work that way, particularly among people belonging to the lower sta Further, when the gomg becames dificult in the birthplace, large numbers of the poor migrate to cities, where more work opportumities are available. This was the strategy followed by many of our respondents. They migrated from rural to metropolitan cities, even 2. We are gratefi to Charles Ndson and Devendaa Babu for the rarch assince in the pration of his pert ef the chaper. _ - : . a 190 Abdul Aziz and Michad Rwaberger though this meant living in an alien and unhealthy enviromnent like the slums, where work is available, however. Survival Strategies in the Nonmarket Context The poor also function in a system outside the market context and try to develop survival strategies for the conditions there. This system comprises of a series of informal networks that have come to the rescue of the poor. These networks can be classified into four categories on the basis of the nature of social relations the poor have developed. The network most frequently used for help is, of course, the one built up by relatives and friends. Next in importance is the neighborhood network. The third is that made up of former employers and acquaintances. And the last one, which seems to be gaining ground in Indian Society, is made up of the political workers and the voluntary agencies dedicated to the cause of the poor, disabled, and destitute. Help received from friends and relatives is in the form of interest-free loans and outright nonreturnable monetary contributions (cases 7, 10, and 11), and clothes on festival occasions (case 4). Sending children to relatives who would assume full responsibility for bringing them up (case 1) is a strategy that the network of friends and relatives has encouraged. Apart from lightening the blurden of bearing the expenses of feeding children, this course of action on the part ot the poor ensures better care for their children. Coming to neighbors, they contribute cash, interest-free loans, and free transport during emervencies (case 12), and look after the baby when the mother is at work (case 9). The former employers, as in the case of Shobha and Palani, not only extend financial help but also give moral support. A network that has emerged recently and is as important as that of friends and relatives is the one built by the local polidcians and the charitable institutions. Their role in Indian society is likely to expand with the breaking up of the extended families, and the fast-growing rural to urban migration-both of which are causing physical and mentl distance to widen among friends and relatives. The fact that the state is now doling out benefits of all conceivable kinds opens up possibilities for the local politicians and social workers to informally link the poor with such benefits, which may not be intended for them anyway. Venkat, on the recommendation of a local councilor, gettng free food from a school that runs the midday meal scheme for the school children, is an example. Such a help goes a long way in sustaining the life of the destitute persons. Charitable institutions such as the Sai Baba Service Society that provide succor to the flood-hit poor households (case 9) is an example of what sort of help the nongovernment organizations could render to the poor. It is rather difficult to estimnate the extent of help in cash and in kind that flows from the informal networks to the poor. However, from what we have observed durnng our field visits it is evident that the help emanating from this source has lightened the burden of the poor and has even given them an opporunity to live better and, at times, to live with the hope for a good future. Implications for Policies and Program The findings reported above can have important implications for poverty alleviation at both the policy and project levels. A major concern at the policy level is to find the most efficient and equitable way to provide special security for vulnerable groups such as the aged, young children, hapless wome, the unemployed, and the sick. The main policy instrment is redistnrbution through taxation. A tax may be imposed on workers, consumers, or propery Survival rsegies of Poor Househwods 191 owners to generate resources, which are then paid out in the form of pensions, unemployment benefits, subsidies for children, or the provision of free or subsidized health services. As Cox and Jimenez (1989) have pointed out, substantial private transfers are already being used for precisely this purpose when children provide for their aged parents or when grandparents help cover the education and health costs of their grandchildren, or when neighbors assist each other in times of need. It is important therefore for planners and policymakers to assess the interactions between private and public transfers. If private transfers decline when the state assumes responsibility for the protection of vulnerable groups, the net redistributive effect of public transfers would be significantly reduced. In some extreme cases, the net impact of public transfers could be negative Cif, for example, the taxes collected by the state are used less efficiently than the money that would have been used through private trnsfers. The following are some of the policy questions to be investigated: * Are there any services (health, education expenses, provisic n for the sick, nutritional supplements for undemourished children) that are more efficiently provided through private rather than public asfers? * Are there any groups (ethnic minorities, difflicult to identify vulnerable groups, female household heads) that are reached more effectively through pnvate transfers? * What are the net effects (after the impacts on private transfers are taken into consideration) of public transfers on the provision of different services and accessibility to different groups? * How can public transfers be used most effectively to leverage rather than reduce private transfers? * Are there any specific areas in the unorganized labor market that require state intervention? At theprojecr level it is important to assess the volume of private transfer resources that can be mobilized in support of different kinds of projects. It is also necessary to evaluate the impact of these transfers on the consumption of basic needs. Several questions need to be considered here as well: * How can estimates be made of the liely volume and distribution of private transfers that will be generated by diferent kinds of projects? * How can transfers be leveraged so as to mobilize private investment resources for social or physical inf s projects such as housing and the construction and operation of schools, clinics, and minor irrigation projects? * What are the impacts of these transfers on the consumption of basic necesities such as food, clothing, and medicine? * How effective are social networks in identifying vulnerable groups and in ensuring that these groups maintain a minimum standard of living? 192 AbdAlziz and Midcael BambeTger * Are there any services that could be more effectivdy managed by social networks than through public services? * How could social networks be integrated into the planning and management of different kinds of social security and social welfare programs? It is apparent that any attempt to systematically work with and through social networks would encounter a wide range of methodological, political, administrative, and even ethical questions; and that any proposal to do so should be studied and monitored very carefilly. However, the fact remains that the resources available to governments in most developing countries are far too inadequate to guarante even a minimum level of social welfare to the poor and vulnerable. Consequently, any possible way of mobilizing private transfers to complement public resources merits serious consideration. Case Studies of Survival Stratgies in Kamataka and Tamil Nadu3 Case I Raju ran away from his home in Pondicherry at the age of twelve and came down to Madras. He worked in a small hotel and subsequently switched over to a better job. He settled down in a slum, acquired a small plot, and built a house on it for his dwelling. Raju got married at the age of twenty-seven and now has four children-two daughters and two sons. He has not been keeping well for the past eleven years and cannot go out to work. As a result, the burden of rnning the family has fallen on his wife and the daughters. The oldest daughter eams RS 60 per month by tailoring and the other one, aged sixteen, is earing Rs 12 per day workiog as a casual laborer in a foundry. The sons are studying in school. Because they are poor, one of the sons has been sent to Poongavapuram where he is looked after well by Raju's relatives. That has given some relief to the family. Case 2 Mali, a poor scheduled-caste girl, was married at the age of sixteen to Raghu, a vegetable vendor. Both the husband and wife would go to the city market early in the morning to purchase vegetables and retail them in a push cart along the streets of the residential area in Madras. A baby girl was born in the house and the husband and wife were happy. One day Raghu went to the vegetable market alone and never returned for he had met with an accident and died. His wife borrowed money (Rs 200) from others at an interest rate of 10 percen a month in order to have him cremated. Now, Mali and her daughter are staying in the slm. Mali continues to be in the vegetable vending business and earns Rs 15 a day. Her daughter is working as a maid-servant in a house and earming Rs 50 a month. The big worry for the mother is to get her daughter, who has already grown up to be a young woman of twenty-four, married. 3. At thc ime of dte study tie exchange rte was Ptp. 25 w the dollar. Surival Stnzregies of Poor Households 193 Case 3 Veniat migratd to Madras from Dindivanam during his childhood and settled down in a slum in Madras. Before he reached the age of twenty, his father arTanged for him to be married to a particular girl, but owing to some problem the marriage did not take place. Venkat remained a bachelor all his life. As a young man, Venkat picked up skills as a mason and led a comfortable life. Having saved some money, he built a house far himself in the slum. One day, when he was arranging bamboo poles on the roof of the thatched house, a bamboo pole suddenly slipped and injured his eyes. He was rushed to the general hospital, where an operation was performed, but unfortunately, he lost his eyesight. Now, at the age of sixty and without his eyesight, he cannot do any work and therefore cannot earn a living. The councillor of the area in which Venkat is staying has arranged for an old-age pension from the governmnt. He is getting a pewnion of Rs 50 per month and also Rs 20 as rent from a small hut that he had acquired many years ago. He also gets food from a school that runs a midday meal scheme for its children. The teacher, out of compassion and on the recommendation of the councillor, parcels out some food for the old man. Because he is not in a position to collect the food personaly, his neighbors fetch it every day on his behalf. He keeps part of the food brougit from the midday meal center for his supper. He thus lives on in this way. Case 4 Gopu, a young man of scheduled caste, migrated to Madras from a village near Dindivanam dunng his twenties and seltled down in a slum in Madras. He had lost his parents and cane to Madras in search of work as a casual laborer. He purchased a bullock cart and a pair of bullocks and subsequently got married to a Christian girl, Mary, who lived in the same slum. The Gopus have four children. Hhag out the bullock cart, he would earn Rs 25 daily-part of which was spent on maintainig the buluocks. With four children to be fed at home, this eaing was found to be quite inadequte. Hence, Mary decided to work. She now works as a head load carrier and earns Rs 8 per day. But this work is seasonal and hence she remains without work during most of the year. Since Mary married outside her com=mnity and to a member of the scheduled caste, her relatives did not want to have anythn to do with her. However, one of her cousins who is relatively better off and broad-mindedt gives clothes to her and her children during Christmas and other festivals. A Catholic priest to whom she explained her condition is supporting heix son in school. This help has lighned the burden of the family a little. Case 5 Shobha, aged about thirt-four, lives in a North Madrs slum. She got married to a person from the same slum seventeen years ago. Although she had four children by this marriage, her married life is miserable because her husband is a habitual drunkard and gambler. To make matters wor2e he also has a bad heart. Being a good-for-nothing fellow, he would daily pick a quarrel with Shobha and pester her for money. Not being able to get monetary help fom- her husband, she set up a small petty shop in the residetal house. She cooks snacks, particularly idli, and sells it to the residents in the slum. She discovered that the unsold idli could be served 194 Abdd Aziz and Michael Banberger as lunch to herself and her children and she could even make a small profit, which was usually snatched away by her husband for his personal expenses. The oldest child-a daughter-is studying in IX standard and has been awarded a scholarship that covers her educational expenses. Her second and third children-boys-are studying in school. Their educational expenses must bc met. The fourth child-a boy-is mentally retarded; he needs a great deal of attention. On top of everything, the husband, without Sobha's knowledge, borrowed money from a shopkeeper by mortgaging the house. There is now pressure on her either to Spay the loan or to vacate the house. When she confronted her husband with this piece of information, he absconded. When Sobha approached some relatively better-off persons in the area for financial help, they wanted to take advantage of her helpless position. Some even suggested that before the money was given, she should send her daughter to serve as a maid-servant in their house. She was suspicious about this proposal and approached a lady doctor for whom she had worked as a maid-servant before she got married. The doctor is a kind woman and is helping her financially and also giving her moral support. Although the mortgage on the house causes her great anxiety, she is managing the day-to-day subsistence problem with her IdaW business. She is even seting apart some money daily so that someday she can re-lease the house from the clutches of the moneylending shopkeeper. She is living with this hope. Case 6 Krishna, aged about twenty-five, and his family have been living on a footpath in Bangalore City. His wife, aged about seventeen, and his brother, aged about twenty, constitute the other members of his family. He migrated to the city frm a village in Malur taluk five years ago. He worked on the family landholding for about two years. They used to get four bags of ragi (a local millet) from the land during normal, and barely one bag of ragi under partial rahifall. As a result, he was forced to work as an agricultural laborer to supplement the family eanings from the land. Since the entire taluk is dependent on monsoons, which are not regulr, agriculture had become uncertain. Added to this, wage employment, too, had becme uncertain because of inadequate rainfall. He was, therefore, forced to migrate to Bangalore in search of work. Since he does not own a house and cannot afford to take a rented house, the family is living on the footpath. They take refuge in a nearby bus shelter when it rains. Both the brothers go out to work. Tbey dig the foundations of residential buildings on a piece-rate basis. The time rate earned per day of work is Rs 20 per head. If they worked on a piece rate, they earn an average income of Rs 40 to 45 per day per person. Hence they prefer, and at times insist upon, worldng on the piece-rate basis. They save part of the earnings and remit this to the village where their old parents live. It is, perhaps, this commitment to the well- being of the parents that has compelled them to live on the footpath. Case 7 Chenna was one of five brothers born and brought up in a village near Gulbarga His parents owned 3 acres of dry land. Inme from the land was quite inadequate for a family of seven. He worked as a watchman in the canal irrigation department for about three years where he earned a monthly income of Rs 150. When his services were terminated, he found living in the village too difficult. Since his uncles and other relatives were living in Bangalore, he boldly decided to migrate to Bangalore City along with his wife, leaving behind his parents and brothers. Suraidva Sirategies of Poor HouseFotds 195 When the couple came down to Bangalore City and settled down at Vasanthnagar slum, they were physically sturdy and fit for bard work. So both of them worked at breaking stone. About four years earlier, they were struck by a flood. Their hut was washed away and they lost all their belongings. The city corporation officials shifted them to another place at Lingarajapuram. The Bangalore Development Authority provided them with a site, five poles, and some asbestos sheets. No sooner had they built a new house and settled down there, than the family faced a new problem. His wife lost one of her hands when a stone accidentally fell on it. In the course of time, Chenna got a sprain in his shoulders and on his back. Although he was operated on, he was left disabled. His brother paid for the treatment which cost about Rs 5,000. He got the money by selling some land. In addition, Chenna pawned his wife's sarees and some vessels for Rs 150, which he also used for his medical expenses; Then, when his father-in-law died, be had to borrow Rs 800 from a local friend to perform the funeral ceremonies. Now that both husband and wife are disabled, life has become miserable. Consequently, the woman has turned to begging as a means of living. Case 8 Palani, aged fifty-six years, came to Bangalore from Ooty on the request of his ex- employer who owns race horses. He came to look after horses, with his wife and they left bebind his three daughters and a son. He lived in a house near Vasanthanagar slum. He led a happy life on a saiary of Rs 500 per month. In 1987, he suffered a paralytic stroke. In the beginning his employer gave him Rs 500 for treatment. But he did not recover from the stroke. He has become permanently disabled. Also in 1987 the entire slum was flooded with water after heavy rains. The dwellers lost their belongings and their huts. After some dme, people living in the slum were shifted to Lingarajapuram by the City Corporation officials. The Bangalore Development Authority formed a new layout for the benefit of the flood-affected poor and the City Corporation provided asbestos sheets and bamboo poles to construct huts. Being one of the beneficiaries of this program, Palani constructed a small hut, and at present he is staying there. Having been completely paralyzed, Palani is unable to support himself. His children visit him once a year but do not support him financially. Now he depends entirely on the earnings of his wife, who works as a casual laborer. Her earnings are supplemented by a monthly rent of Rs 30, which he ears by renting a portion of his house to a teashop owner. The tenant gives him a free cup of tea every moming. Friends and relatives give used shirs, lungis, and sarees. The daughter of his former employer who visited him recenty gave him Rs 100, which cheered him up. Case 9 Jaya is a poor scheduled-caste woman who came to Bangalore from Gudiyatam in Tamil Nadu five years ago, along with ber husband, in search of work. Her mother was working as an Aaya in Bangalore City. First, both of them stayed in a slum close to Cantonment railway station. The husband worked as a casual laborer and earned Rs 10 per day. One day her husband came home drunk. Her itoubles began that day because afterward he spent most of the money he earned on liquor and gave little to the wife to nm the house. 196 Abdul Ai* and Michad Bamberger One day, he went out for work and never returned. She has been searchig for him for the past four years. She is staying in a hut in the slum along with her four-year-old daughter. As a daily wage worker, she was eamrig Rs 7 to Rs 10, which was barely enough for her and her daughter's subsistence. Unfortunately, her hut was washed away in the flood of 1987. Having lost her hut, she temporarily settled down on the Cantonment railway station platform. During this time, the Sai Baba service Society supplied her food for a week and gave her a saree and a blanket. She went out to work every morning and did not return to the railway station until the evening. She used to cook for herself only in the evening in the open space. After some days she returned to her old slum and put up a hut with the assistance of the Sai Baba Service Society, in the form of poles and thatches. The neighborhood hut dwellers helped her for about six months by looldng after her baby when she went out for work. After that, an old lady looked after the child. Now, she works as a housenaid. Aithougliher monthly salary is Rs 125, which is far less than what she would get as a casual worker, she says there is an advantage in this. There is work throughout the year and she eats her meal in the employer's house and also brgs food for her child. Old clothes such as sarees for herself and used dresses for her daughter are also given to her by the lady of the house where she works. Case 10 Rm, who is about forty years old, married a Cbristian girl-Elisa-in the Frazer town slum of Bangalore City some twenty years back. Even though he is a Hindu, their married life went on very smoothly. He would go to the temple and she would go to church. They have five children-two of whom are studying in schools. Raim works as a casual worker and earms Rs 20 to Rs 25 daily. But of late he has been drining (alcohol). As a result, the amount of money he provide to maintain the family has been reduced. Hence Elisa felt that she should also work for survival. With the Rs 50 that she had saved, she sated fruit vending. She goes to the city market and buys fruit and sells it in the slum area. On an average, she earns Rs 10 a day from this source, which keeps the hearth warm. One day, when she was cooking food, a fire broke out accidentally and she was burned. She was rushed to the hospital and luckily she recovered from the burmns. Her friends and relatives helped her with Rs 500 for her treatment; otherwise, there was no other way of meetng this unexpected expenditure. With her small eargs, she is surviving with her husbnnd and childre Sometimes, her children may not get anyting to eat. On such occasions, neighbo take pity on the children and feed them. Her ambition in life is to give all of her children a good education, so that they do not go the way of their father. Case 11 Rama, aged about 'forty-seven, has been livmng with his wife in his native village of Gottigere near Bangalore. They have no children. He has two brothers, one at Bannerghatta looking after a petty shop and another in the same village worldng as a truck driver. Rama has no land and his main occupation is agricultual labor. His subsidiary occupation is rearing cows and buffalo. Earlier his wife also worked as a casual laborer. San7 Stregies of Poor Homeholds 197 While working in the field, one day Rama accidently slipped suffered a dislocation and frature. This necessitated two operations-one on the thigh and another on the waist. Now he cannot walk properly and has become a physically handicapped person. Since his brothers are living separately and are also poor, they were unable to offer him any support. However, help came from other quarters. His cousin, who is a professor of ophthalmology and is residing in the same village, helped him in many ways. He covered Rama's medical expenses, amounting to Rs 500. His friend who is an agriculturist from the same village gave him a loan of Rs 200 to buy food and clothing. On account of the influence of his cousin, he obtained a loan of Rs 5,000 from Vijaya Bank, Bannerghatta, under the program of credit for the physically handicapped to take up business. With this loan he built up a good business and was able to repay the entire loan and interest to the bank. Besides, he had pawned some household articles for Rs 500 and earrings for Rs 800 to meet the medical expenses. His cousin also got him a monthly pension of Rs 50 fom the government, which is generally offered to the physically handicapped. Case 12 Giri, aged about sity, has bee living in Gottigere village since 1966. He is married and has two children- a boy of about eight and a girl of seven. Both children are studying in the local primary school. Giri migrated to this village from a village in Hosur taluk of Tamil Nadu. There, he lived on income from cultivating govement land and working as a casual agriculural laborer. During the 1966 drought, he was unable to find work in the vicinity of his village and the land yielded no crop. Life was so miserable that he was forced to migrate to a place where work was -vailable. Thus, he landed in a coconmt garden near Gottigere, where he is residing at present. After a few years at Gottigere, his wife died from smallpox. Later he married another woman, who also died tbree years later owing to a prenatal problem. When his second wife fell ill, his financial problems did not permit him to take her to hospital in time to save her. However, on his request, one of the villagers helped him with a loan of Rs 200. An autorickshaw owner provided free transport for the woman to go up to the government hospital in Bangalore City. In spite of all this effort, she died. The vilagers contributed Rs 200 toward the funeral expenses. Giri's two children are studying in the local primary school. Unable to meet their school expenses, he considered withdrawing them from school at one point. However, when the school came forward with an offer of free uniforms and textbooks, he allowed them to continme their education. He lives in a small old deserted Maih on the outskirts of the village. It has no doors and one of the walls is badly cracked. He is physically weak and us unable to tke up hard work. For the past four years he has been engaged in retailing coriander leaves m the Bangalore city market, which brngs in about Rs 12 per day. But since coriander leaves are not available on a regular basis, he faces worldess days occasionally. Before the death of his second wife, she too was working as a laborer and supvlementing the frnfily income. Now, with the wife gone, he has no one to help him. However, during emergencies he gets support in cash and in kid from his fellow villags. Part 4 GUIDELINES FOR THE SELECTION, DESIGN, IMPLEMENTATION, AND SUSTAINABILITY OF POVERTY ALLEVIATION PROGRAMS Guidelines for the Selection, Design, Implementation, and Susana-biity of Poverty Alleviation Programs Ediors' Note Parts 1, 2, and 3 have demonstrated the unique characteristics of Poverty alleviation programs and have argued that the conventonal project cycle needs to be able to respond to these characteristics. Part IVproposes operational 8uidelines to ensure that the selection, design, implementation, and operation of powvery alleviation programs take into account the chaacteristics of the target population and the economic, social, and politcal eviomnment in vwhch the poor live and work. It is argued that if poverty allevitin projects are designed according to conventonal project cycle procedures, the will not benefit large sections of the target populton,4 stand a greater risk of being poorly implemented, and wil be difficult to sustai. The design and management ofpoverty alleviaton proj ects implies somne additional costs in terms of staff tim, isvformauion reqiurements, and Preparation time Poverty projects wdil frequently be smaller tha conventional investment projects, and this, too, will increase the supervison costsper dollar loaned. In manyj cases, however, some of these costs can be absorbed by NGOs or cormmwuiy organiztons themelves. An addiional investment is required to reorient government bureaucrats (and private offidcals for projects such as credit) wand to create a government calture tha is responsive to the needs of the poor. This willfrequenztly involve training, modifications to organizationalstructure, and intensive cooperative efforts on smallpilot projects. It may also require the creation of new coordinatng mechnimsm to ensure effective liaison between Participating public and Privat sector agencies, and between the government and beneficiaries. Thie urse of appropriate design and implementaton strategies is, howeve, likely to produce significant bewflts to offset the costs. First, the project benefit are much more likely to reach the intended target groups. Even a modest 5 to 10 percent increase in the proportion of Project benefits reaching the intended targetrgroups may more than offset the aMddtonal stat-up and supersion costs. Secnd the quality of project design and implemetio willfrequently improve when the expertise and organizzational cuapaiy of beocneiares is utilized. Thrd, there nmy be additional cost savings (to the government or donor) if communitie contribute their laor (wvhich they are much more likely to do if the have been involved in the project design).- Fourth, the lkelihood that a project wdil be sustained is greatly inceased if beneficiaries are activelY involved in project operations and maintenane. The life of irrgation projects and social infasruc rreare two examples in which the increased project life due to communty involvement in maintenace can more than offset the more intensive project design and superviion reqided. Benefts are also likely to accrue from the ionvetents in raining and mistiutonal resruturngto adapt government, and where necssay private4 agences to the culture of povert-responsive admninistration. A number ofgeneral principles are advocated througho uttthe fiVe chapers in this section. First, it is vita to develop a conducive policy environment and to ensure cloe integraton between national and sectoral policies and the design and implmenttin of povert programs and projects. This has implications for both the public sector cultue and the instiutina 201 structure. Decentralizton and dedisio ng that is close to locl icqleen ng agences and beneficianes is one importantreqrmentwith many orgztionaimlicapions. Aiz mphasir the importance of a conducivepolicy enmronment in hspessimisic assessment in Chapter 11 of the low levd ofpoitical cmmient i most South Asian cownies to sustainable andfnacaly viable poverty programs. Although there is strong support for subsidizing services to the poor, there is very little support for effective cost recovery, and this is one of the reasons that many poverty projects have not been sustainable. Second, the process of coUecdng and interpreting for plning, monitoring, and evaltion must be modfied. The emphas wU be on rapid assessment tedniques, whih in addition to convenional ewnomic data ll also focus more on social, cuural, politcal, and unsdiutonal factors. The data will be interpreted with the actve particpaton of befciaries rarher than in a government office. Turd, a key requrement i that benfiarnes participate in all stages of project idendfiotion, design, implemnaon, and operanon. is requres a much higher investn of staff time asfrequent meetings must be held with benfiaries and the project must be modified, as izecessary, in response to thse discussions. Fourth, the project design and implem ion strateg must be adapted to te characteristics and needs of the beefciary populaon. Projects are lkely to be smaler and sumpler and will miake greater use of loca technology and implementaton pracices. Greater flexibility will also be required in the design, flinan control, ad mpl ion schedues. The project must be considered a Ulearning process" rathr ta the implmtaton of a predeemined blrint Consequely, it is ikely that there wil be signmifit modifications as the projea proceeds and benicaes come to udrstand exctly what the project imolves. While these changes and delays would be considered costs wthn the convonal project cycle, they are often considered among the greatest benfts within the 'process approach" as te strengthen the capciy of local communities to idenf and Jplemetfture programs. Part IV Editor's Note 202 11 THE INTERFACE BETWEEN POVERTY PROJECTS AND nE POucCY ENVIRONMENT Abdl At Until recenty, poverty was considered a product of inadequate growth of the national economy. Hence, poverty alleviation efforts teaxed to be part of the endeavor to accelerate economic growth. When it was found that accelerated growth did little to bring poverty down from its old levels, a new approach to poverty alleviation was developed known as the target group approach. It focuses government poverty projects not at the national or state level, but at the viage or urban household level Poor households are idenified and given inputs that will help them escape from the vicious circle of poverty and launch them into self-sustined development. The strategy of poverty alleviation worked out under this approach was to operate at one point along the vicious circle of poverty. To illustrate, the poor have low incomes, and the low incomes do not permit them to save and to build productive capital. Since the stock of capital is low, their productivity levels are also low. Hence, hecy again end up with low incomes The poor being thus caught in the vicious circle of poverty are unable to come out of it unless they receive some external help. The literature on the subject suggests tiat one way to help the poor escape from the vicious circle of poverty is to break it at some point, preferably at the point of capital bnmation. As a result, state intervention in developing counties has taken the form of finacial assistance to the poor to enable them to build sufficient amounts of capital at the household level. It has been found, however, that a one-time inevention by the state at any one point along the vicious circle of poverty is just not enough to sustain the poverty alleviation effort.' Financial assistance for building capital at the household level does not automatically alleviate poverty unless other thigs follow. To begin with, the capital or asset formed out of capital must be retained and put to pioductive use. Second, the output or service generated must be marketed- -and marketed at a r e prce. Third, part of the income generated must be saved and plowed back to augment caal. Only when all these measures are taken is it possible to successfully eradicate poverty. Therefore policymakers have of late recognized that poverty alleviation strategies should not only call for direct intervention by the state but that such interventions should take place at different levels and points on the vicious circle of poverty. This means that the poverty alleviation project should provide for a package of inputs and services that make it sustainable. Sustabiailitry is futher -ensured if such projects have links with the wider poverty alleviation policies and strategies. hI fact, the poverty reducto effort will be lightened somewhat if (a) a policy environmen conducive for desiging and managing poverty alleviation projects is created, and (b) the poverty projects are linked to the national poverty alleviation policies and 1. SEc Cbapt 1. particuarly Fipre 1.1, for a discussio of dte two-way link between tbe national policy nvirmnem, sector policies and program, and she design and managemet of tageted powvrt projects. 203 204 AbdAz strategies. This means that the poverty alleviation projects should be identified, evaluated, and implemented in the wider perspective of the national poverty reduction policies and strategies. Such a course of action sustains the poverty project by providing the needed inputs and services and by ensuring that the project is formulated and implemented in a conducive environment. This dcapter examnes the extent to which the required policy environment for poverty alleviation is in place in the South Asian countries and offers a few guidelines for integrating poverty projects into the national poverty alleviation policies and strategies. But it was gradually recognized that a one-time intervention by the state at any point along the vicious circle of poverty was not enough to sustain the poverty alleviation effort. Policymakers saw that not only should there be direct intervention by the state, but that such intervention should occur at different levels and points on the vicious circle of poverty. A policy enviromment conducive for designing and managing poverty alleviation projects was needed to ensure that this intervention would become a sustained effort. Only then, it was argued, would such projects have a sustainable impact on poverty alleviation and the living conditions of the poor. The available literature on the subject suggests that such a policy environment should have the following contours: 1. Antrural or andurban policy biases should be eliminated wherever they exist. 2. Rural and urban development and agricultural policies of the govcernent should foms specifically on poverty. 3. It is vital to tamrget services and benefits for the poor and vulnerable groups. 4. Rural and urbmn development must be seen as a process requiring long-term integrated planning. 5. There should be greater emphasis on questions relating to project and program susanbility. 6. There must be emugh politcal will and clout to ensure that programs are implemented aaud sstained as intended. The concemed authorities should wield political will to ensure that (a) measures are taken to enforce loan recovery or the payment of user fees and service charges so that funds are recycled through new loans and projects can be mainned; (b) adeqate financial autonomy is given to decentalized project planing bodies; and (c) NGOs are allowed adequate financial and opeaonal autonomy. Thus many conditions must be met in order to create a conducive policy environment, some of which will be difficult to achieve. One problem lies in developing a framework for the inrface between the national policy level and the project framework. For example, if the national pricing policy is to benefit the consumers m general, then no useful purpose would be served by developmg specially targeted projects to promote production among small farmers-as they would already benefit from the broader policies. Similarly, the power strucure favors the nch land elite, any attmpt to m4prove the condition of poor faiers through decentalized planing is likely to be fruitiess unless it incorporates measures to protect their interests from the The Interfac Betwen Poveny Projects ad the Policy Ewronmne 205 powerful economic elites. This then calls for an integrated approach that meshes the poverty alleviation concerns implied in the project witli the objectives of the macroeconomic policies directed at managing the economy in general. In the face of such conshtaints it would be a tall order to expect that a policy environment for poverty alleviation would ever meet the level and degree expected. Nevertheless, it ts important to consider what characteristics of a policy environment are conducive to the design and management of the kinds of poverty alleviation projects implemented in the countries of South Asia. In India, the work of Michael Lipton led to a debate on the question of urban bias in planning. Much later (during the National Front government), this found expression in the restructuring of plan priorities in favor of rural development. With the Congress government back in power, this debate seems to have taken the back seat. Nevertheless, the commitment to poverty alleviation of the governments of India and other South Asian countries has been fully reflected in a series of programs that were developed to assist rural poverty groups such as small landholders, artisans, tenants, and agricultural laborers. The Integrated Rural Development Programs (IRDP), Training of Rural Youth for Self-Employment (TRYSEM), National Rural Employment Program (NREP), and the Rural Employment Guarantee Program (RLEGP) in India, Small Farmer Development Program (SFDP) in Nepal and Bangladesh, Village Aid Rural Works Program and Peon'Js Work Program in Paistan are cases in point. The programs have been the outcome of the government's enduring interest in and concern for alleviating poverty through well-structured antipoverty policies. The historic shift of emphasis from the traditional growth-oriented planning strategy to targeting benefits of such growth is a testimony to this. Along with the debate on the urban bias in plaming, the issue of urban-bias policies, particularly relating to prices, has also attracted attention. Many Asian governments, in their eagerness to keep prices under control, have adopted physical controls and rationing. Although these measures do play an important role in allocating scarce consumer goods to the weaker sections at reasonable prices, they could also generate an unintended side effecs since the products under control, particularly food products, are the ones that are produced by the small and marginal farmers. The policy of price control in respect of food grains may unintentionally contribute to inequities. Such inequities may become accentuated if a majority of the beneficiaries of this policy tum out to live in urban areas- In India, however, the commodities subject to price contols have in recent years been made available to the rurl poor as well Moreover, since the small and marginal farmers are found to generate a zero marketable surplus, their interests as producers may not be greatly affeced by price controls. It follows that even urban-bias policies could be tolerated if such policies also tend to the interests of the rural poor, as in the above case. While the various programs referred to above are targeted to take care of rural poverty, until recendy urban poverty problems had not recerved the same level of aetion. One important reason for this ambivalence was the belief that urban poverty is an extension of rural poverty in that it is a result of the migration of the rural poor to urban areas in search of work. Hence, it was argued that poverty alleviation strategies should focus on nrual rather than urban poverty. However, the appalling living conditions of the poor in the urban settlements and the political threat such conditions posed to the governments in power obliged them to shed this bias and to develop specific policy measumes for alleviating urban poverty. In the initial stages, the policy concerning the urban poor focused on providing the basic minimum needs like housing, health education, nutrition, and the like. Since then, policy has shifted toward supporting programs that generate income and employment The SEPUP in India clearly illustrates this shift. 206 Abdil Aziz In general, therefore, the rural and the urban poor such as the small holders, tenants, agricultural laborers, artisans, urban informal sector workers, and slum dwellers have been the main concern of policymakers. In recent years, howevcr, the more vulnerable of these groups, especially the old, the destitute, women, and persons belonging to socially disadvantaged seecions of society are being specificaly targeted under the antipoverty policies and programs. With this change, the policy has become focused on the poorest of the poor-a development that is quite welcome considering that the ultimate goal of planning is to reach the benefits of development "unto the last." One of the most important ingredients of the policy environment for poverty alleviation is the realization of the need to help poor and vulnerable groups accumulate income-yielding assets. While doing so, the government should formulate policies that strengthen the ability of the poor to accumulate physical and human capital. Some progress has been made in this area in South Asia. With a view to promoting human capital formation, primary education has been made universal. To achieve this goal, the government has provided various incentives to school children such as a free supply of books, uniforms for the weaker sections, and midday meals. Special programs have been designed and implemented to encourage girls and minority groups to become literate. Family planning programs are also being rigorously implemented by providing easy access to family planning services. In addition, standards for health care, hygiene, food supplements, znd medications have been rigorously propagated and provided to the poor. Also social safety nets for the old, disabled, and destitute have been provided. These and many other policies oriented toward building hmman capital among poor households have added to the required environment. With respect to financial capital, the earlier policies were in the nature of providing credit at subsidized rates to farmers, artisans, and agricultural and other laborers. Since experience showed that there were high transaction costs owmg to formal loan reqirements (and the emergence of middlemen), the emphasis now seems to be on encouraging nmutal cooperation either through self-help or with the help of nongovernment agencies. As a result of this change in policy, a large number of social workers and NGOs have emerged to organize the poor for mutual help in the area of credit and financial assistance. Two examples are the Working Women's Forum in South India and the Proshika organization in Bangladesh. Conferrng ownership rights on the tenat, regulating rent, fixing tenure, and giving land to the landless by means of land reforms is another important staategy for endowing the poor with income-yielding assets. While such measures ensure tenurial security to the rural poor, experience has shown that this can also increase land productivity. A progressive land reform policy is therefore considered an important instrument of state intervention in poverty alleviation. The South Asian countries, of course, did enunciate land reform policies with two main components: one is the regulatory policy seeking to regulate tenancy and rent and the other is redistribution. The latter sougbt to identify surplus land and to distribute it among the landless and the marginal landowners. There was also a supplementary policy that aimed at providing credit and other forms of assistance to the beneficiaries of the land reforms to enable them to improve their land and achieve higher levels of productivity. An important breakthrough in addressing the policy environment question has been the recognition in recent years that poverty alleviation efforts would have greater impact if carried out in the framework of integrated development.That is, activities, groups of people, and the spatial location of amenities and infrastructure facilities should be integrated in a way that creates The Interface Beween Povet Projea and the Polcy EnWronmet 207 backward and forward links. The rural and urban development policies of the South Asian nations now seek to implement this idea through appropriate action plans. The policymakers, having thus realized the need to target services and benefits like credit, have also provided for mechanisms to identify the target groups, to assess their needs, and to meet those needs in adequate measure. There are two approaches to targeting benefits. The efficacy of targeting depends on how well these approaches are followed. The first approach is to use some indicators for screening the beneficiaries, while the second is a self-selection process whereby the applicants screen themselves. Under the first, the income of the applicant is generally used as the principal targeting indicator. In the absence of, or in addition to, income, other indicators such as geographic location (backward areas, hilly areas, flood or drought-prone areas) and demographic features such as age, sex, and caste are used. Under the second method, inclusion or exclusion is automatic and is done by the applicants themselves. The program keeps in view the requirements of the poor, such as low-cost dwellings, low-status foods, water and sanitation facilities, and the nonpoor do not become involved in these programs through a lack of interest in what is being offered. The required institutional structures have also been built up. Many South Asian countries (notably ndia and Nepal) have introduced decentralization policies that give people and their representatives a major role in planning and implementingpoverty-alleviating projects. The NGOs have also been encouraged to design and manage such projects with or without government support, financial and otherwise. Thus some ingredients of a conducive policy environment are present in the South Asian countries. Unfortunately, another important ingredient, political will, is frequently absent. Progam sustaiability is ensured if the assets created at the beneficiary household level are properly mainutained and retained, and if the program cost in the form of loans given and service charges are recovered so that funds are recycled to keep the project in place. Unfortunately, poor cost-recovery performance and lack-luster govemment efforts to recover these dues, and, to top al this, the loan waiver spree noticed all over the country suggest that the government of India is not interested in sustaining the financial viability of poverty alleviation projects. Given such an ambivalenc policy environment, it is difficult indeed to motivate the beneficiaries to use the fiuds for the intended purpose, to maintain and retain the assets created under the project, and to repay loam. Second, the shift in emphasis in favor of urban poverty alleviation programs is no doubt a welcome step in view of the ft that in the South Asian countries the urbanization process is proceeding rapidly and that during this process more and more of the urban population will be sucked into the poverty trap. However, it is essential to recognize that poverty is not specific to either rmual or urban areas. Nor can rural poverty be taclded without considering what is happening in the urban settlements, and vice versa. Rural and urban poverty are interrelated through the channels of migration. This calls for an integrated approach to poverty alleviation, with the focus on tackling rual and urban poverty simultaneously, and not sequentially, nor to the exclusion of one or the other. However, an integrated approach to rural and urban poverty alleviation is not yet being implemented in the South Asian countries. In the light of the new economic policies being enunciated and implemented, state efforts at allevian poverty by means of specific programs have become sunounded by some unctinty. The market-friendly approach to promoting development and resolving economic ills is likely to put a damper on poverty programs followed to date. The current debate is concerned with the need to design and implement antpoverty programs. Such programs iMpose heavy burdens on the exchequer as they call for huge subsidies, in addition to heavy amounts being 208 Abdul Azi spent to administr the schemes. In the present mood of aversion to any kind of subsidy, the antipoverty program that have a substantial subsidy element will naturally be seen in a poor light. There is, therefore, a need for a clear enunciation of policy toward poverty alleviation under the new economic policies, which give primary attention to market forces. Some Guideines Soudt Asian countries have obviously developed some ingredients of a policy environment conducive to poverty alleviation. Poverty projects could be still more effective in their impact and reach if they were integrated into national strategies of poverty alleviation. Such a result becomes a possibility if in the first stage there is a conducive policy enviromnent for evolving and implementing the poverty projects and in the second stage these projects are integrated into the national strategies for poverty alleviation. Since the efficacy of the project depends on how effectively it is integrated into the national strategy, the manner of integrating the two becomes crucial, which raises the question of what guidelines may be evolved to help the policy maker. The following guidelines may be considered: 1 . To reap the full benefits of intervention, the poor must be provided a package of benefits designed to alleviate poverty. This means not only that poverty projects should provide all the inputs needed by the poor household-such as land, capital (credit), and market facilities- but that all their benefits should be consistent with national poverty policies. A poverty project that offers poor households an animal husbandry unit should also allot land, through land-reform policies for raising fodder. A poor household that obtamins land under the land reform program should be entitled to credit from the banking system for improving land productivity, The off- season wage employment provided to the poor must be consistent with the national wage policy that ensures a reasonable minimum wage to the workers, and so on. 2. Poverty projects-especially those that give production units such as milch animals, goats and sheep, horticulture and agricultural enterpnses, bullock carts, and irrigation wells-should be planned in an integrated way that consists of backward and forward links. Here the idea is to ensure markets for the products and services rendered by the project beneficiaries so as to make the project sustainable. To this end, it will be necessary to identify the market and estimate demand for such products and services. The number of asset units to be given to the poor should be tailored to market availability. Also, since links will have to be established within the local market, asset units should be given to the poor in a carefiully planned sequence. For instance, if milch animals are to be given, fodder-growing units must be given first, and not the other way round. Again, the poverty project must take note of national price policies. Thus a project that produces a product whose price is controlled as a matter of national policy will be of no help to the beneficiary unless either the inpUts are subsidized or a compensatory payment is made. 3. There should be bottom-up and top-down links with regard to poverty projects and poverty policies and strategies. In other words, the poverty project should be formulated and implemented in the framework of the poverty alleviation policies and strategies formulated at the national level so that both become mutually complementary. The cople requirement implied in the links between the project and the policy can be ensured if project fornmlation at the national level and policy formulation at the local level are designed to be complementary to each other. To illustrate this point, let us assume that the national strategy for reducing poverty The merface Bewoeen Poveny Projects end the Polly Envronmet 209 is to direct investment into sectors that are libor-intensive and sectors that produce wage goods. Since the national strategy largely applies to sectors rather than regions, and resource allocation is mostly guided by the desire to maximize employment and wage goods produced in certain regions, subsectors, and population groups, particularly those that are backward, will get little atention. To fill such gaps, local povertv projects specifically tailored and targeted to such regions and population groups ought to be formulated and implemented. These projects will not produce desired results unless there is a support system based in the physical and social infrastructure. That system should arise from national strategies for developing the infrastructure with a view to alleviating poverty. 4. In order to mount a simultaneous, rather than a sequential, attack on rural and urban poverty, an attempt should be made to develop an integrated approach. Moreover, such an approach should aim at stalling the migration of the rural poor to urban areas. One way of doing this is to create work opportmities in rural areas. While the development of the farm sector will hold out good prospects for employment to the rural poor, development of the nonfarm sectors will add to such prospects. Merely creating nonfarm work opportunities will not help the poor unless they have the required skills to perform the necessary tasks. Therefore, 'lie national strategy of creatng nonfarm jobs should be preceded by local projects tLat impt skills to the rural poor. 5. National educafion, health, and housing policies- have been spelk I cut by many governments in South Asia. These policies usually provide some benefits for the poor in the sense that they include the poor and provide these benefits at subsidized rates. Unless appropriate projects that focus specifically on the poor are formulated and implemented at the local level, the poor may not benefit from such policies. There is, therefore, a need to formulate target-oriented projects that utilize opportunities afforded by the relevant national policies. Conversely, the local- level health. education, and housing projects should be fornuLated to be consistent with the focus and philosophy of the national policies relating to these service areas. If the national housing policy emphasizes upgrading slums and providing sites and services, for example, local projects that involve the construction of housing complexes for the poor will not receive support from state and federal agencies to ensure their sustainability. Similarly, if the national goal is to impart relevant skills to the poor, no purpose will be served by a locaL project whose goal is to impart only general education. A~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 12 IDENTIFICATION AND DEsIGN OF POVERTY REDUCTION PROJECTS Neil O'Sullivan Lning fxom the Past From the coundess eflorts and eenditures that have been directed at poverty reduction in Asia, many lessons have been extracted and can now be reintroduced into the design of new projects, as described later in this chapter.' Three important lessons, however, frequently rn either overlooked or ignored: 1. Projects are most likely to succeed if their objectives match the priority demands of the poor and if they elicit the active and ongoing participation of the poor. 2. Successful poverty alleviation projects normally require intensive and time-consumig attention; however, the resources available to inplementing agencies are invariably Limited and are likely to shrink rather than to expand over time with the result that post- project scenarios are likely to receive little, if any, attention. 3. Even where both political will and government resources have been adequate, the results, particularly in terms of sustinability, have frequently remained disappointing. One important implication of the above observations is that strategies for alleviating poverty must shift away from the traditional focus on the provision of govrnmment support services toward a focus on sdtmuling the institutional capacity of the poor to analyze their needs, initiate their own efforts, and articulate their demands. This conclusion reflects the accumlated experience of the World Bank (1988c) and many other observers, which has shown that one of the most importan deterninants of project success is the atenton given to institional arrangements, particularly with respect to the receiving side. The importance and potential of fostering beneficiary organizaion and participation has long been recognized, and over the past decade, many successful efforts have been made in this direction by both government and nongovernment development agencies. The fact that this approach has not yet been widely adopted, however, reflects both the insttutional difficulties of implementation on a wide scale and the predictable social difficulties of or ming bureaucracies, organizng poor communities, and transferring greater power to them. In recognition of the above problem, this chapter proposes an analytical framework for the design of people-based poverty reduction projects and notes the principal factors that are vital to successful project outcomes. It is well recognized, of coumse, that not all projects lend 1. The refrnces cied in Chaptm 12-15 are included in ht bibliograpby at the end of this publication. 211 212 Nei OS'Sadiwa themselves to a fully participatory approach. For some types of large or national projects and programns the necessary intensity of the participatory approach may be either impractical or prohibitively expensive. Similarly, the application of a participatory approach is not necessarily relevant when introducing national polcy changes that are frequently an essential part of antipoverty strategy. However, in these cases, increased attention would normally be given to monitoring the impacts on the target population, as descnbed later in the chapter. Foremost among those factors crucial to the design of successfil poverty interventions is an understanding of the very nature of poverty itself which, all too often, has been seen as simply a deprived economic situation. Where project designs likewise restrict themselves to a purely economic focus, they run the risk of a leakage of benefits and poor sustainability. Although the nature of poverty varies in every situation, there are few cases in which a depressed economic state is not accompanied (or perhaps even caused) by a depressed psychological state. The severity of this laIter problem is most apparent in rural areas and appears related to the proximity to the poor community of more sophisticated, more powerful, or more exploitative- communities. In extreme cases, as in the aboriginal communities of modem-day Australia. many years of massive government expenditure have proved fruitless in improving the desperate socioeconomic conditons of aboriginal people. More positive results were obtained only when guvernment strategy introduced measures to help aborigimals srngthe and protect their own culture and idenity.2 In more typical and less exreme situations, the poor communities of Asia are still characterized by low levels of sel-confidence and, even in conditions of dire poverty, are more likely to opt ' r traditional, outmoded practices rather than newer, more appropriate responses. In such situations, project designs, no mater how rational, are likely to meet with little success unless measures have first been taken to develop sufficient confidence in the target community so that thiy will acually develop a demand for the proposed changes. Conversely, expience has shown that most antipoverty initiatives are short-lived if they rely on the supply of goods or services from some extral agency, whetber government or nongovemmen. Unfortnately, unless techniques such as those described below are used, poverty project planners, from the moment they begin to talk of "target groups," run the risk of seeing the poor as passive objects of externa help rather than as the subjects of their own self-reliance. In many parts of Asia, the legacy of centuries of feudalism has also left many poor communities with a low level of self-respect,which together with low self-confidence creates a strong sense of passivity and iertia. Such communities are also likely to be fiuther burdened with varios forms of exploitation such as indebtedness to moneylenders, low wages, tenancy, and the corruption of minor officials with whom they must deal. The above problems are unlkely to be overcome by resource allocations made in the traditional manner. Resources channeled indiscriminately into a distorted socioeconomic situation may simply end up tratig symptoms rather than causes and may ultimately entech an already ineqitable situation. What is more likely to be required is a mixture of careful and sensitive planning based on a detailed and empathic knowledge of both the target group and those surrounding groups that would be likely to help or hinder their development. 2. The prgrssv gramting of land gts has conbutd much to the ovuall wel-beg of radioa Aboigah, while for Abgas living m smi-urban enviro ss, dn oppormity of managng e om health and eation rices hbs ld to mch hig leves of pudatpn and effective nd efficit scce provsiaL Idhenficauion and Design of Poveny Reduction Projects 213 While the above descrbes the siatation of the poor in most taditional mral areas, sigicant differences have emerged both in landless commities and more especially in urban areas. In the case of the landless, organizatons such as BRAC (Bangladesh Rural Advancment Committee) have observed that in a situation of endemic and intractable landlessness and poverty such as exists in many areas of Bangladesh, the poor may actually be somewhat liberated by the very desperation of their plight and be prepared to accept new ideas, enterprises, and forms of social organization. (Such characteistics have been both successfuly encouraged and developed by organizations such as BRAC and Grameen Bank in Bangladesh.) A similar liberation has taken place among the urban poor, although in this case it was brought about more by the drastically different social relationships and levels of awaress and opporunmity that prevail even in urban slums. In such a situation, the design of poverty reduction projects can normally rely on a more straightforward and positive response from the target group toward any opportunity for economic betterment that is offered. Here, the primary focus in project design is more likely to on maximizing beneficiary participation in the selection of appropriate options and arrangements. ldentfying Poverty Reduction Projects The basic concept for individual poverty projects may emerge firom many sources: political imperatives, donor pressure, national plans, and the like. Although most of these sources could reaonably be expecd to have a clear idea of the extent and urgency of tackling poverty in a particular location or sector, the details and priorities for an appropriate intervention shategy are much less clear, particularly if the strategy is to mirror the priority needs and demands of the target group. Nevertheless, it is inevitable that for the foreseeable futr, initial project concet will be generated outside the communities for which they are designed. For this reason, considerable attention must be given to systematic and progressive testing and refinement of the orginal concept on the basis of grass roots realities. Poverty alleviaton srteges can generally be divided into three types: * Those that careflly target direct transfers or subsidies to the poorest or most vulnerable sectors of the populaion. * Those that increase access to basic social services such as education, health, nutrition, and family planning. * Those that increase the productive use of whatever assets the poor may have (with incsing landlessness, labor automatically becomes a main focus). In pracce, most national poverty alleviation policies will be reqired to combine all three strategies. In this chapter, however, the last stra receives the most attention, because (a) despite its frequent past disappointments, this approach offers the greatest potential for dramatic positive results, and (b) in the longer term, it is vital to ensure an economicaly producive base on which to sustain essential social services. 214 Nei O 'Suivau If poverty projects are to lead to the efficient and effective use of scarce public resources, it is important at the outset to examine key demographic and strategic factors that should be used to select the most promising project concepts. Demographic Factors As a guiding pnnciple, poverty alleviation projects should atempt to improve the income levels of the maximum number of poor people within the limits of the available budget. In many developing counties the greatest concentration of poverty is often to be found within a specific sector-whether geographic, ethnic, or social. For example, pockets of poverty often occur in isolated, mountainous or border areas, among tribal minorities, people of low caste, and women- It is important to understand the basic character of these groups since their project design requirements will vary significantly. The folowmg are some common scenarios: * Geographically and ethnically based poverty is likely to be the result of past isolation and neglect. Given the probable social homogeneity existing in such situations, a project package focusing on ifrastructure, social services, and credit and using local institutions would be xpected to have excellent prospects for success. * Where the above scenario has been disrupted by the intrusion of outside factors, such as logging, trawler-fishing, lowlander settlement, agnbusiness, oriurbaiaton, additional attention would need to be givento stre ening local organizations to enable them to cope on more equal terms with the usually more powerful and sophisticated external forces. 6 Many poor urban areas are within the minseam of society and likely to have at least theoretical access to local infastrure and developmet resources. In this case the main obstacles to progress are more likely to be social or psychological, and any physical inputs would therefore need to be preceded by an appropriate o tional or social preparation phase (see Chapter 6). Targeting In many broadly based development p. zjects such as those involving the provision of essential services, the primary concem winth respect to poverty reduction is to ensure that the poor are not put at a disadvantage by gaining access to such services as primary health care and education. In most specific, antpoverty projects, however, especially those involving direct transfers or subsidies, the implications for equity and efficiency make effective targeting a much more important concern. In an ideal system, the benefits of such projects would only be directed toward those living below some agreed poverty line and the level of support would be proportional to their distance below the poverty line Many NGOs have been able to come close to the above ideal because of their close proximity to the target group and their generally small, operational area. On a larger scale, NGOs have often been successfl in encouragg target communities themselves to take responsibility for selectng beneficianes fom among their peers. In large-scale, government-implemented programs, however, the likelihood of political and otier local distortions is much greater, and it is often difficult to arrive at an equitable and operationally viable method of direct targeting. detificaion and Design of Poveyy Reduction Projecrs 215 In terms of cost, each additional degree of targeting precision required is likely to add significantly to the project's administatve costs (it should be noted, however, that in some programs such investment can help improve project sustainablity; see Chapter 15). In certain situations a too perfectly targeted program may generate an external backlash sufficient to undermine the political support necessary for the project to continue. A useful. means of assessing the effectiveness of a given targeting procedure is that based on statistical theory, which considers two ypes of error: type I, which refers to the percentage of intended beneficiaries who are excluded, and Type II, which refers to the percentage of those who are not entitled to the benefits, according to the selection criteria. From among the basic options available, the use of income or landholding levels as the selection criterion has been conmon, although difficulties in measuring these levels have led to many anomalies. As a result, this approach has been abandoned in many places in favor of low- cost, politcally acceptable approaches such as targeting specific, low-income, geographic regions. Income and landholding criteria are, however, successfully used by many NGOs or other external agencies that, by transferring identification responsibility to the poor, are generally confident that beneficiares will be selected on the basis of both acacuate local knowledge and a reasonably compassionate inte on of the rules.3 In primary health programs, where targetng was not originally applied, advantages in efficieny and equity are increasing being seen in applying geographical and "at-risk" targeting approaches.4 In programs involving food subsidies, effictive targeting can at times be achieved by selectng those commodities that are preferentially consumed by the poor and by concentrating ration shops only in poor areas. The basis for selection of a particular food (or any other service) is the share consumed by the poor as a proportion of the total consumption. Thus if a food is consumed exclusively by the target group, the subsidy will be very efficient. The other common form of targeting is based on food for work or similar measures that assume the poor have a lower opportunity cost of time and that the better-off will not be tempted to make the contnrbutions of time or effort required to qualify. Such measures become increasingly atractive when one of their outputs is an increased level of skill that means the recipient may not require futher support in the fusture Where the objective of targeting is to ewsure that the very poor are adequately covered, it is important to have a sufficiently detailed knowledge of the target group to understand the varioous strategies the poor themselves use to assist in their own survival. While the interal trader of resources thrgh informal networks is an important, if hidden, phenomenon in many 3. The Gramea Bank pku die majority of Bangladesh NGOs taget thcir acdvities on the fimclonay landless by applyig common crira. For example. the maximm landholding is 0.5 ac an family inme comes nwsdy frm die sale of own labor. 4. By monitoring for inadeqae eight gain infin, the Tamil Nadu lItegrated Nutridon Projec was able to halve severe nmouridan levels and achieve twice the impact at less cost than a comparable program providing near univers coveage. As pointed at by Saia, hoer. both and at-risk approches have some limitaion. In the case of geographical trti health services may stil be consumed by tie non-poor wthmin a priorw rgi. whil the wat rsk approach may be polically unpopula and wDll genray involve high s costs For hinher detals see Chaptr 3 of the prsent reporL S. The Vulneble Gru Dement (VGD) Ptogamme in Bangdesh povides a reglar gain tion to destuWte women toughout the coaunry, btt in sevemrl sccefl pilet projects, the emphasis has been shifted to providing this group with potntial livelihood skills suh as polty. seriwre, and vegetable growing. 216 Neil 0'SuWi a poor communities, there are invanably certain subgroups that cannot avail themselves of such support and are thus particularly vulnerable. A large and growing subsector that frequently falls into this categozy is that of female-headed households. The needs of those households will also vary greatly, however, depending on the circumstances of the husband's departure. Strategic Factors The ultimate success of projects will be determined not only by the care with which they are designed and implemented, but also by the degree of congruence between the project's objectives and design and the national policy environment in which they will opeate. Apart from the formal or legislative situation, the phenomenon generally referred to as political will is widely considered the most significant determinant of success of a povert reduction )roject. Where governments at both the central and local level are united and determined to ensure the success of a project, few external problems are likely to be encountered. Conversely, more politically contentious projects will need much greater care and consultation if potentially disruptive forces are to be kept at bay. In many developing countries, the nature and extent of political will in relation to the poor tends to be a reflection of historically derived attitudes based on vested (and usually patraistic) interests, ratier than of any democratically expressed preferences. Although in the short term some limited leverage can at times be applied by external development agencies, in the long term the most susuinable strategy imnolves inicasig the level of awareness, and srengthening the voice, of the poor themselves. Project designs must as far as possible be in harmony with prevailing or planned national and sectoral development strategies. In urban areas, overall policy decisions on urban-rural links and planed growth rates will have a great influence on most projects that provide services for the poor. In rural areas, project concepts should similarly reflect a goverment's current rural strategy on the removal of price distortions, improvement of market mechanisms, export development, and so on. Likewise, in questions of institutionaI reform, if a goverment has recognized and accepted the inefficiencies of an approach based on increasing the resources and technical sophistication of line agency extension programs, then identifying altemative bottom-up strategies will be relatively straightforward. Regardless of the.official policy position, however, would-be implementing agencies tend to cling to their own culture, usually because of its past status, political pressures, or intemal leadership. Where a progressive leadership or instiutional culture can be identied in advance, it is advisable to develop a project concept that can make use of such an agency to spearhead new project approaches. Any new goverment initiatives in areas such as administrative decent on and land reform (including regulariation of tenure in slums) can be expected to create a range of positive opportunities for improving both growth and eqpity and thus setting the stage for potential project designs. Institonal reform, such as the creadon of a specialist Rura Development Bank would obviously provide opportunities in areas such as micro-credit. Where powerful political leaders or ministers are enthusiastic about a particular policy direction, much can often be achieved for a relatively small expenditure. Conversely, specific projects may be better postponed if they run counter to prevailing political passions. For example, it would be better to defer the setting up of a credit project if the prevailing mood is one of debt forgiveness. The stats of women and the potental for change in the sus of women must also be carefully considered duing project identification. ldenfticadon and Design of Poely Redion Project 217 kt is now widely accepted that women frequetly bear an unequal burden of labor yet are denied equal access to basic resources such as educaion and health. On the positive side, experience throughout Asia has demonstrated that in both urban and rural projects, women are more enthusiastic and reliable participants than are men. The latter dcaacteristic, coupled with women's usual role as managers of the household health and finances, can provide an important potential focus for successful project design, especially in those projects involving cost recovery, maintenance, or credit functions. The question of designing gender-sensitive poverty reduction projects is dealt with more fully in Chapter 7. Effective schemes to reduce poverty should, in principle, reduce the pressure on the environment and naural resources. In some circumstances, however, poverty alleviation strategies can actually accelerate environmental damage (as in the case of grant- or credit-based projects that might unintentionally encourage the growing of cash crops on marginal, sloping land). Although such practices may be detected by monitoring and evaluation procedures, they may be difficult to prevent. Experience has shown that the poor are normally just as concemed about environmental proicon as other sectors of society but that their first priority will always be to ensure their own survival. Key Factors in Project Design An otherwise excellent, participatory-based project design may either coilapse or fail to produce las;tng benefits if certain key factors are neglected. A number of such factors are described below. Administraive Reaism Many failed rural development and poverty alleviation projects were initated in a mood of great optimism. Generally a basic design was drawn up, frequently in conjunction with an exteral donor, and it was not until later that an assessment was made of the administrative capacity of the chosen agency to implement this design. At the time of agreeing on the basic concept, the parties involved were frequently preoccupied with the range of isttutional (and sometimes personal) benefits that would flow from the project and tended to downplay the many administrative problems that would need to be resolved if the project, as conceive d, was to be successful. Officials considered potental administrative weaknesses a conseuence of inadequate resources that could be resolved via an injection of capital, equipment, and perhaps a little overseas training for senior staff. In prctice, the "instittional wealmess was fiequently later reveaed to involve a systemic, low productivity brought about by a combination of low morale, frequent staff transfers, the promotion of incompetent staff, bad lederhp. unreliable statistical data and exaggeraed performance, corruption, poor commnications, and the like. Even worse were the external administrative problems caused by unreliable contractors, the lack of communication facilities, local political intferenc, corrupt suppliers, lack of inteacy coordination and cooperation. By focusig on the ideal rather than the reality, many past projects were doomed even before they began. Unfortmately, many of the same factors continue to predispose decisiomakers toward the same mistake of overptimism with a resultant compromise in ultimate performance. This danger is compunded in pmjects dealing with the poor both because the target group is more vulnrble and the process of stimating self-reliance is itself extremely slow and 218 Neil O'SWlivan sensitive. Some of tie factors that influence the selection and performance of impleme agencies are covered in Chapter 13. Because the prospect of a new project (especialy if extemally fimded) represen potetial access to increased power and resources, it is not uncommon for bureaucrats to give more attention to attracting a new project than to solving the problems of an earlier project having the same, or similar, objectives. For similar reasons, initial project designs tend to pay close attention to the early, input-rich phase of a project and comensurately less to the opeation and maintenance aspects, while the all-important, post-project scenario is often ignored entirely. An obvious lesson from this is that no new project concept should be developed until an objective analysis has been made of the problems and shortfills of any prior comparable project. In the past most infastructure projects were undertaken on a sectoral basis independent of local government. The result was poor sustinablity, inadequate participation, and faulty coordination between sectors. Responses to these problems have included the creation of metropolitan-wide authorities, as in Manila and Calcutta, and even specially empowered metmpolitan governments, as in Bangkok and Jakarta. One of the many advantages to emerge from a better-integrated managemen structe is the ability to recogniz, and respond to, the multifaceted needs of the poor; for example, improved health and shelter have been achieved via functional education and microenterprise programs. Ironically, one of the legacies of the disappointng Integrated Rural Development programs of the 1970s has been a tendency among some development agencies to opt for straightforward projects with a minimum of components, an approach that almost automatically bypasses the poor, whose needs are invariably complex and intertwined. Some observers argue that appropriate administrative structures cannot be establisheduntl the more serious problems of the poor have been clearly identified and meshed with overall socioeconomic priorities (Lim 1988: 125-46). Economic Reaism For the sake of efficiency and thus ultimately, sustainability, projects should seek to take maximum advantage of market forces and of the pnvate sector. In this way, they will be able to increase resource-use efficiency and reduce their long-term dependence on governmen support. Untl recendy, unrealistically high standards were commonplace in the design of service projects throughout Asia, for planners and politicians alike were fequently attracted to modem, high-technology investments. The impact of this pheomnon on poverty reduction is probably more significant than is generally realized. As Lim (1988) has observed, "Experience has shown over and over that where public service stndards are set at levels unrealistically high in relation to the poor beneficiary's ability and willingn to pay, most of the intended recipients of the services have generally not received any service at all." Examples abound of settlement upgrading programs that, because of their high standards, ultimately succeeded not only in displacing the original inhabitants but also in dissipating scarce subsidies and distorting market factors. In a related if more subtle manner, project planmers frequenly find it more attractive to puse options such as a new water-supply system rather than allocate less funds to rectify the typical 25-50 pexcent leakage that affects Asian cities. (Again. there would normally be little interest in the furtier possiblity that scarce funds could be used even more effectively by supporting a health and sanitation education canmp among poor consumers.) Ienfication and Design of Povery Raction Projects 219 In earlier chapters both Satia and Yeamg claimed that many inefficiencies in service provision also stem from the present regulatory framework that tends to protect public agencies with a virtal monopoly stats. Evidence from Calcutta and Bangkok, for example, suggests that, in the area of public trasport, the introduction of private operators can lead to considerable saving. Given that Asian cities are typically growing at more than double the national population growth rate and that the shortfall in providing essential services is acualaly increasing in most regions, observers such as Yeung have questioned the assumption that the poor will be best served by traditional govemment-sponsored service delivery models. While the development of effective private sector alternatives may necessarily be slow, a logical starting point would be to increase the support for all forms of self-help and community-based initiatives and to encourage appropriate commnercial enterprises. In addition, it is suggested that the existing large and vital informal sector be given much greater recognition. The informal sector already plays an effective and low-cost role in areas such as housing, transort, water-supply, and public secrity. In many situations, however, the benefits of increased efficiency may not automatically flow from the entrepreneur to the producer, particularly if there is a wide disparity between the two in terms of bargaining power and certain strategic interventions become necessary. In the area of urban housing, Indian authorities have successfully intervened in protecting bustee dwellers from exploitation by private landlords while at the same time investing large sums in upgrading environmental conditions in private rental housing areas. Similarly, it is often difficult to predict market response to the initiatives of the poor, and careful monitoring of market equity must be maintined by the implementing Agency in situations where the poor are developing direct links with the market for the first time. In nural areas in the past many "farm-to-market" road projects beame, in fact, "market-to-fam roads with the terms of trade tdited in favor of the merchant or midleman. Although well-organized producer groups can eventually rectify this situation via cooperative marketing arrangements, for example, it is not realisfic to expect this degree of sophistication from a beneficiary group until quite some years alter its formation. nquitable trading arrangemes are especialy common where the local merchant has also functioned as the moneylender and established a tradition of discounted prices paid for goods on which an advance has been provided. In this type of sitation, apart from intensifying the knowledge and organizational skills of the beneficiary orgaizations, a more direct intervention will frequently be called for. In many cases, the introduction of a rudimentary market intelligence system may be sufficient, although in more entrenched situatons, the implementing agency may need to establish alternative market links until such time as the market adjusts.6 Problems with economic realim can also easily occur where project planners rely on market forces to generate employment or entrepreneurial development projects. Where such projects have been undertaken in a mood of optimism rather than on the basis of detailed surveys of the availability of markets, tr tation and storage facilities, credit mechanisms, seasonal 6. This ype of interventio was iaoduced, for example, by Thaiand's Bak for Agiutue and agica cooperatives to ovrcome a situaion in 1985 in which heavily indebted paddy producers were being forced to sell tir crop immeditly after harvest at a time whe the prce was kept artificially low by merhants. BAAC'S 'paddy pledgig schme (shor-tm loans given for up to 80 percent of the value of barvest) allowed the fanrns to decide when and to wbom to sell. Wiin hrce years the market had been rguarized and fLrmers no longer needed to resort to such loans. 220 Nel O'Sidilvan availability of labor, and so on, the chances of faiure are inevitably high. In Sri lanka's Janasaviya self-employment program, for example, rural handcrafts were produced in such numbers by unemployed villagers that the government was itself forced to establish purchasing centers and warehouses to absorb stock that could not be sold commercially. In urban projects, similarly detailed surveys will invariably be required to assess the housing market and land availability, or the preexistent distribution and usage of services such as health, water supply, or waste disposal. Cost Recovery Mechanisms In order to maximize the impact of scarce public resources, project design should seek to recover as much of the investment capital and running costs as is realstically fair and possible. Where a project design does not seriously address the question of cost recovery, it is likely to encounter the two problems that bedevil most services provided in subsidized form. First, the benefits of such a project will probably be regarded as a service provided by eternal parties for uncertin or even dubious reasons and thus not sometig that recipients should either value or maintain (the universal problem of public, water-supply hand-pump maintenance is a classic example of this phenomenon). The second problem relates to the inficiencies and distortions frequenty brought about by subsidies. The mmaerous problems caused by subsidized rural credit are now weli known (see Chapter 6). However, more subtle problems can afflict other programs such as public health schemes where, as Satia points out, attempts to apply the notion of 'health for all" within the limibtations of a finite health budget have probably been to the detriment of the poorest, that is, those least able to gain access to, or make demands on, public health services (see Chapter 3). The amount of study and consultation reqired dunng the design phase to confim the appropriatess and acceptability of the planned cost recovery approah will be detmined largely by the prevailn social expectations. Where, local politicians have traditionally sought to win popular support by promising an unrealistic level of free governmen services, for example, services sbould not be provided unril recipient communities have demonstrated their commitment, usually in the form of some agreed level of equity contribution. Where the needs and demands of each target subpopulation have been clearly cfirmed during the rapid social analysis, or its equivalent, cost recovery design should be as flexible as possible in order to cater to the differing capaciies and resures available in differing communities. Among very poor commumiies, for example, the ability to make cash contributions may be very limited; however, widt an often lower opporunity cost of time, such communities, once organized, may be willing and able to substantially reduce the cost of govermment services by contributing their volumary labor. The choice of technical options has a major impact on the affordability of a project for the poor and dws on likely cost recovery performance. A project that responds to what the target community sees as a high priority and at the same time provides an appropriate design, is likely to have a high level of contbution, as noted in World Bank-assisted shelter projects in which many families were able, with the help of income transfers fom extended families, to allocate up to 40 percen of their income to shelter, rather thn the 25 percent, as ariginaly anticipated. The decision by a number of Indian statesto provide low-cost pour flush latrines as an alternative to expensive piped sewerage mean that substantially more urban poor would benefit from the expenditure of public fimds. , _ - . . ,, . -. . _ Ideeflcadon and Design of Pover Reduction Projec 221 Project Size and Duration Whether in the search for increased efficiency, or as a result of staff budget limitations, both aid donors and developing country governments have, in recent times, been pressured "to do more with less." Unforunately, at the field level this well-intentioned pressure has frequently been translated into an imperative to "move money,' with the rate of transfer, rather than the project's impact, being the basis of evaluation. The end result has bee an increasing pressure for larger projects and programs. Two assumptions underlie this trend, the first being that the principle of "economies of scale" can be comfortably applied to antipoverty projects and the second that, regardless of project setting, "development" can be treated as a kind of machine that transforms inputs into outputs. In problems of poverty, the existing "deelopment machinery" is itself part of the problem. Consequently, susimabledevelopment involves creating and transforming organizaions and stctures so tha they genuinely serve the needs of the poor, as wel as the rest of the population. Such a task cannot be readily achieved within the type of blueprint approach used when designing a bridge, nor can the scope or speed of development pmjects be expanded simply via larger injections of funding. Becuse poverty-reduction projects involve tackling both physical and social problems simultneously, they face a multifaceted set of management problems of the kind that bedeviled the integrated rural development projects of past decades. By following a participatory approach and foamg on building the institudonal strength of the poor, many of these problems may ultiately be solved by the poor themselves. However, there wil still remain a complex tangle of bureaucratic, technical, and political concerns that can only be coordinated and solved progressively by the implementng agency. Until such time as effective standard approaches are available to cover target groups in n c= ities and regions, the principle of economy of scale simply camnot be assumed in taklding poverty problems. In fact, given the above fictors, together with the basic umpredictability of poverty-focused projects, a strong argume must be made for keeping the unit size of such projects to a minimum.7 Becas projects based on the development of human capital cannot readily be accelerated, it is generally prudent to allocate the maximum amount of time possible for project completion. A major poverty reduction project involving a number of regions and subpoplations will invariably require several years to set up its infrastructuLre and begin a social preparation phase, a firther three years or so to establish local acceptance, and perfaps four to six years before the social and economic impact is filly visible and cbanges can be mstiutionalized. Although interim evahluations can be made, it will normally take 10 to 15 years to confidently assess such a project. Developmt agencies and planns who atempt to apply the conventional project cycle and compress poverty reduction projects mto the traditonal ime rame of three to five years run the risk of building an expensive struc on foundations that are notoriously shaky. Chapter 6 raises similar concerns and therefore calls for more flexible approaches, such as basing project managemenlt on an initial pilot project and introducing the main project in a phased manner. 7. Some experts sub as Nonran Upboff (1990: 1401-41) suggest tba large projects are so vulnrable to bod imerna and extemal dismupdoms dm4 "eta dzabgs bing equ4L shar are &i!diiz revn la projac size. 222 Neil O'Suluvan Rapid Social Assessment In most countries of Asia, the performance of past poverty projects indicates recurring problems with time-overruns, cost recovery, sustainability, and a leakage of benefits. Many of these problems reflect a lack of understanding of the target group, its likely response to the proposed project, and the possibility of external interference. Many project designs also repeat the error of the discredited Community Development Movement of the 1960s, which assumed village (or urban) communities were a single harmonious entity.3 The above problems may be tackled in a variety of ways. For example, in 1987 in the Indian state of Karnataka introduced administrative reforms designed to decentralize development planning functions and ensure greater beneficiary participation in the design and implementation of poverty reCuction projects (see Chapter 9). The effectiveness of such an approach, however, is limited by the degree to which the poor are able to develop and articulate a critical awareness of their priority needs and their capacity to ensure an appropriate response from local govemment. Where it is impossible to base project design on an informed and democratically articulated voice of the poor, as is normally the case, the initial project concept should first be subjected to an appropriate form of social analysis. If beneficiary participation is required, rapid social assessment (RSA), as detailed below, is one relatively fast and simple technique that not only reduces the risk of negative outcomes but helps refine and crystalize the project design and also sets the stage for a more participatory mode of project implementation. The RSA is not geared to the detailed design or justification of a specific project; its mn flnction is to clearly identify the target groups and to predict the potential need or demand of different subpopulations for a proposed project, given their level of development and absorptive capacity. The RSA also identifies the likely sources and levels of support and opposition to the project. A major outcome of the RSA will be to confirm the need for a more detailed participatory project design (PPD) phase and, if required, to define its scope and priorities. (The PPD, see below, is the detailed participatory analysis undertaken in order to optimize project design and implementation strategies.) Rapid social assessment is conducted at the earliest possible stage following conceptualization of the project. Depending on the nature and scope of the proposal, several weeks of fieldwork may be required to gather the necessary data. The approach used in RSA consists of six steps: identifying the target population, rating the level of development of each subpopulation, assessing the target population's need or demand for a project, assessing absorption capacity, assessing relationships, and reporting and interpreting the RSA. Iden*fig the Target Popuklion This task covers nontarget, affected popultons, as well as those populations that are directly targeted. The process consists of S. See, for example, Lar-Brmad (1970), who descrbes the community devrlopment movm as inevitably being taken over by dte existing lites and institutionalized as a vehicle for increasmig dteir coaol over power and reSoiies in thir kcal communiy. Idenftficaton and Design of Poverty Reduction Projects 223 * developing a siWle input/output model to assess potential socioeconomic and culural impas of a proposed project; * collectng demographic data on the overall target populaion from formal (govermnent, academic) and informal (NGO, private sector) sources; * identifying major subpopulions within the target area, on the basis of land and capital holdings, employment status, level of social organization, ethnic or culaual divisions, and access to government services (see Table 12-1 for a sample classification). Rang the Level of Development of Each Subpopulation The following are some guidelines for raing the level of development: High: Those who own surplus-producing land; have access to a prmary and secondary education, institutional credit, agricultural inputs, political power, and health services; and have solid, spacious housing. Medium: Small fanners producing small or negligible surplus and limited access to the above services. Modest housing. Low: Marginal fanners or landless people who gain an income from lowly paid casual labor, have no access to formal credit, and little or no access to education and health. Assessing the Target Population's Need or Demandfor a Project Whether the target population needs a project can be determined by the following measures: * Field visits to the project area * Direct observation of the living conditions of the beneficiaries (, 1 cross-check data and ratings * onsuatons with benefidaiwes to ascertain their general development priorities (this does not involve direct discussion of the proposed project * Analysis of thefelt need for a project (the degree to which the project addresses the target community's priority concerss) and the demand for a project (the degree to which the trget community are prepared to contribute in order to become involved in a project) * Rating of the level of need/demand for each subpopulation. 224 Ncil O'&SllMm 0 Rating of the level of need/demar I for each supopulation. Suggested guidelines for ratig the level of need/demand are as follows: High: Stong and spontaneous expressions of need related to the development problem that the proposed project is inded to address. Obvious preparedness to involve and contribute. Mediutm: Some intrest/need in relation to the development problem being addressed by the proposed project but not as a first priority. Limited contribution likely. Low: Preoccupation with development problems other than the one the proposed project is intended to address. Little likelihood of contributions. Assessing Absorptive Capacity During the site visit, several fuither factors are analyzed in order to determine whether a target community is likely to be able to absorb proposed project inputs. The fictors to be considered here are as foliows: * Level of social cohesion, as indicated by evidence of cooperation, coordinadon, equity, factionalism, conflict, and exploitation * Sprit of sef-relfance as indicated by previous conmmuity initiatives, loan repayment history, use or misuse of past development funds, and so on * Lvel of development of locai community organizations, if any * Level of biwledge and skills of the target populations. The above information is then analyzed for each subpopuladon in order to rate its likely absorptive capacity. Suggested guidelines for ratng are as follows: High: Homogeneous society (socially cohesive), positive experiencelrecord with loans and financial m, active and representative local o ions, positive liks with govemmet agencies, technically inovative (new crops, enteprs). Med/ian: Some social stratification, mixed success with past development programs, lifeless local oanizations; ccasional meeting with goverment agenies, technically static. Low: Social stratification and disunity, poor record with past development programs (for example, financial m agement), local organiions absent or defunct. Idenification and Design of Poveny Redaion Projects 225 Assessing Reladonships Given the critical nature of the relationship beween the target communty and both the communiity at large and the proposed imple agey, it is necessary during the RSA to clearly assess the mutual feelingslattitudes that aeist between the poor target community and all relevant govemment agencies and better-off subppltions, particularly local elites, if they exist. These data are reported in summary form in the mann shown below. Reporfing ad Interpreting the RSA The results of the RSA are normally reported in a format resembling the one illustrtd in Table 12.1. Table 12.1 RSA Report Form , populauiau Lero of Le' of Need Len/of Percnta e Of e Devepment iDemand Absorpive T. arget -: -- Tribas Low Highlaw Medium 15 Close-kit, but passie . ~~~~~~~~~~~~~~communt Subsien Low MediumALow Low 40 Vrry low lvel farmers of orgaization and mondle Small Med Hgh/Medigbm Medium 30 Keen but muginal disorzed Lar High Medium/ILw High 15 Poor hbistory of The above report form also includes a record of the percentage of te total target population represented by each subpopulation (sine this should be a basic factor in determiing the relative levels of resource allocation) and brief menton is made of any significant characteristics. These data are supplemented by an additional narrative report detailing the relationhps that exist between the above subpopulations and other groups or actors likely to influence the outcome of the project (see Table 12.2). Table 12.2 Relatnship Report ThxrgaG~m. IForezrDea *: flclw*3t :. r L.g. FwmrjiEl. Tribak . Hostile Lile conact Litle contct Neualhostdile Subsismm frmers Little conne hosdie Littl contact Lilde contact Neutral/ost :sain :rel: Li:tle contact Supporive Neutral Neutal I Supportive 226 Nel O'Sulivan The results of the RSA may then be used to reassess: the approprateness of project goals and purposes; the likelihood that subpopulations in the project area will be willig and able to accept or participate in the proposed project; which subpopultons and actors need to be involved in the formulation of the detailed project design; that a participatory project design (PPD) phase needs to be undertaken to maximize positive impact; what the focus of the PPD should be. The need for the PPD is determined as follows: RS4 Result Response a. Where all of the following conditions prevail: No PPD * no negative impact * high level need/demand * high absorptive capacity b. Where any of the following conditions prevail: PPD reqired * limited need/demand * defects in absorptive capacity * potential negative impacts for some subpopulations C. Where there are high levels of potential negative Relocation or impact, significant opposition, or major rehinking of limitations to absorptive capacity project concept The focus for the PPf is also determiined on the basis of the results of the RSA: a. The PPD would focus on means of refining the design to maximize partiipation of those with medium and low levels of development (usually the numerically larger subpopulations)- b. Similarly, where the RSA raises doubts as to the ability of particular, targeted subpopuIatons to gain equal access to project benefits (for cample, tenant farner PPD would focus on the need to develop methnds for directly targeting this group. c. Where the RSA exposes weaknesses in absorptive capacity, the PPD would focus on desigmng a strategy for development of local commuty organizatins (social preparation). d. Where, during the RSA, a subpopulation is assessed as being technically undeveloped, the PPD would focus on working with the himlementng agency to design public informaton campaigns and training/education programs, as well as assess the maimum amount of technical innovation that can be absorbed by the beneficianes. .. :- . ;. Idenaficadon and Design of Poveny Reduction Projecs 227 Use of the RSA in Urban Sttings Although the same principles apply to the use of the RSA in urban situations, the task is frequently made simpler by a number of factors, including greater accessibility, higher levels of awareness, increased homogeneity and greater likelihood of existing local organizations. In applying the RSA to a proposed sites and services project, for example, the subpopulations to be surveyed might typically become, legal titleholders, recent illegal squatters, established illegal squatters and tenants. Paricpatory Project Design The objective of the Participatory Project Design (PPD) Phase is to optimize project design by integrating target-group preferences and piiorities while developing realistic but appropriate project implementation strategies to which all parties have made a comitment. The PPD, a recommended method for achieving this elusive goal, combines the more detiled social analysis as indicated by the RSA with Zopp-style participatory problem-solving techniques-9 Because the PPD involves intensive investigations and consultations, it may take some months to complete and would frequently be conducted in conjunction with economic, instional, and environmental analyses, if required. The PPD is normally carried out independerily of the proposed implementing agency by social analysts on the planning team. In view of the inevitable problems of interdepartmetl coordination, consultation workshops are held in order to build team spirit and reach a consensus on all key matters. 'Me task of the planing team can be broken down into four basic steps: Step Purpose Problem analysis To confirm or amend the project concept by identifying the core problem of the target group and depicting the causes and effects of this problem visually in the form of a problem tree. Objectives analysis To develop specific objectives that relate directly to the problems identified above. Alternatives analysis To assess alterative or competing strategies so as to achieve the previously agreed objectives in the most efficient and equitable manner. 9. Zopp, the acoym for Zielodentierse Projet Planug, or goal-oriented prject planning, is a technique poularizd by die Gean Deveopmnt Agency, GIZ and mom reccnty adpted by many othies as a means of genrating great understndig and conunitmct toward a project desigL The basic approach of dte ZOPP method is similar to that outlined i this dcapter. For more detais on the pradical appliation of the ZOPP method, see (GZ (198. 228 Neil O'wZvan Planning matx To spell out detafled action plans to achieve the above objecdves and identify indicators to measure the progress in achieving these objectives. Problem Analysis The first task is to undertake in-depth and systematic consultations with a represenie sample of each subpopulation and all actors identfied during the RSA. This analysis must be conducted in a genuiely consultative manner so as to ensure eventual consensus and conimtment to the end-product. The objective of problem analysis is to identify core and related problems of the primary target groups that are to be addressed by the project and to identify their respective causes, effects, and nxtuelationships. For each subpopulation or actor, the above interrelationships are recorded on cards in the form of a problem tree (see Figure 12.1), which because of its simple visual nature, becomes a key tool in later co unicang ad assessing the merits of the different perspectives of the various groups involved in the project. In each case the consultation involves a problem analysis fom tie perspecive of those bang intenmiewed. For example, the problem tree shown below would probably not adequately depict the position of tnibals or subsistence farmers who might be unwiling to invest in their most valuable asset, land, because of the insecurity of tenure. In keeping with the team-building approach, workshops are held for all of the parties to compare their peceptions of the problem and to seek agreement on a common analysis. With inCreaSig levels of poverty, particlarly psychosocial poverty, genuine partcipation in any joint consultation will be increasingly difficult to achieve for some groups. In the design of poverty- alleviation projects, a substantial amount of time and attenion mIuSt be given to the poorest subpopulations since they are ivariably the most dispersed, disorganized, and inarticulate. For these groups, joint consultation may need to be defenred until the design is almost finalized In the interim, however, intensive direct consultation would be carried out and would typically involve participatory surveys (of human, physical and economic resources, qualty of social services, cutural and behavioral factors, villagelevel, institional develpme, and ogoing development of projects and programs; discussion and analysis of the survey results and the development context in the project area (constraints on poorer groups, producton, consumption and savings, social welfare, organiztional capability, problems, potential solutions); and disusions related to the project concept (past experiences, risks, potential, benefictary wilingness to participate or contnbute, technological preferences, preferred credit conditions and regulations, preferred institutional aran s): Objectives Analysis The purpose of the objectives analysis is to descnbe a future siuation that will be achieved by resolving one or more of the problems identified; thus, the hierarchy of problems (proble btree) is tansformed into a hierarchy of objectives (objectives tree; see Figure 12.2). This exercise helps identify the potential altnative interventions available to resolve the problems. Idemzfcation and Design of Pover Reducton Prjet 229 Depig& on N DOei | ff& I a No 1 1 oNoActo | Af*iMkz Fi or 12.1Bak"roblefmftTree . S~~~~~So MO DOM- La I I IEa,O 1 ~ ~~G_ompc L oWbw4Sur Fig=r 12.1 Problem Tree Alternatives Analysis The purpose of this stage is to idenify altemtive interventons and staegies to help achieve the agreed objectives, and to evahlate these alteratives using social, envimnl, technical, financial, and institutional criteria so as to identify the optimal intentions and strategy. 230 Neil O'Sualm Referring to f' previous example, if all parties agreed to the same analysis- of both problem and objectives, their strategic preferences would still differ because of their different pespectives. For example, the preferred project strategy might vary from increasing bank staff, subsidizing credit, and licensing moneylenders to perhaps creating a new credit institution. From its earlier analysis, the planning team should generally be able to predict and discount the influence of "vested interests" in the strategies proposed and to steer the planning workshops toward the most rational and equitable resolution. (In feudal or otherwise polarized social situations, some 'tradeoff' may be necessary in order to gain the support of local power groups.) OPTIMNG PROJECT DESGN. The separate fimding of financial and environmentalanalysis will at this stage be incorporated into the final design. In practice, however, the greatest problems are likely to be encountered on the human institutional side, for example, in selecting the best options for incorporating community priorities and preferences. maximizing absorptive capacity, and selecting appropriate tehnical strategies. a. Integrating Target Priorities and Preferences * Project goals are ranked according to their place in the priority ranking of each target sub-population. * The participation levels of each subpopulation are estimated on the basis of the above ranking. eTradeoffs are calclated in toxms of the number of beneficiaries from each subpopulation who may be expected to participate according to the strategy chosen. b. Selecting Appropriate Technical Options Cl general, teChnical options need to be examined in terms of familiarity, size, location, timing, and complexity so as to idetify the project design that will most accurately reflect target population priorities and absorptive capacity. * If the proposed project involves the introduction of new technologies, first assess the degree to which the innovation accords with, or departs from, the estng practices of each subpopulatuon. Major changes should not be introduced to vuleable subpopulations unless the technology has already been verified with other comparable commumities and approprate educational support is available. Innovatons should reduce the income risks of the poor, not increase them. * Where the absorptive capacity of one or more of the targeted subpopulations cannot be adequately increased in the short term, consideration should be given to limiting the activities/components of the proposed project or to providing inputs in the form of preintegrated packages. Idnfwadon and Desgn oF PoVeny Redun Projects 231 * If benficiary patcipation is likely to be reduced or disnptd as a result of structural inequaliti in a target population, project design strategies need to be developed to overcoe these problems (for example, direct targeting, indivisible inputs, om ory c. Maximizing Absorptive Capacity * Where a subpopulation has little, if any, expeience in self-reliant development iniiatives, the project design would need to provide for the mobiliation, organization, and education of beneficiaries to encouage local initiative and develop a sense of responsibility prior to the introduction of external inputs.In other words, a social prepartion phase shoud be incorporated (refer Capter 6). * Where local commnity orgaizaons do not t are corrupt or inactive or do not reresent marginal subopulatiom (for example, the landless, women, ethnic connmities), the project design would need to provide for the development of commnity organtons, particularly among the poorest. * Where the relationship between the proposed implementing agency and potental beneficiaries is marred by defects that might affect project imple ion and stainility, project design would need to explore methods of oveoing these problems (for example, NGOS could be used as intemediaries, extension staff could be appoited from within the taget group itself, or an altenave implementng agency could be established). * Where beneficiaries have low' levels of knowledge or skill in new technologies, consideration may need to be given to overcoming these limitations by training education components designed to increase absorptive capacity. EVAuJAIG PosIE ALTE ES. Whe anative sttgies aear possible, each shaoud be eiamined against critical factors of the kind set out in the matix below: 232 Nel O'&dgvan Altematives Evaluation Matrix institutional Appwpri ageny to Wil community How upesennve wil Will community Anangeme_a: undertke beneficiary orgpnizadons have organinom be? organiztions be Afuly orgniztion? accss t chnical or self-relant? L Deman Side other traing? b. Delivery Side Do involved agencies Adete budget? Full coverage of taget Cost per benefiay have a convincing trck Polidcal support? grup likdy? record? Permanence of Key Coordinatimn stan? Taeing Are reliabic selecion Participaory selection Percentag of poorest in Percenta lkahg crilter available? basis? rg oup? likly? Technical Proven elsewher in Simpiciy? Need for Bariers to enty for IkI-ly rAr of return of Appropriatencss compable situations? cxtenl support? poorcst, ehnic investmen? minoritcs, women? Compatibl with market forces? Cost Recovery Acceptable aget Wil project grCne Charges pmpornal 10 mhUplicky of conectio ump? Precedent? adequae pofit for level of benficiais? uit cos? chagsrupynu f al subsaidies involved? Risks Lel of poliical Levd of echial Levd of lakage? supportioposition? sophistication? Cerainty of market _____________________ price? The Projea Plan MatXr The final outcome of the PPD is the prqect planning matrix or logical frameork, which becomes the reference point for the project throughout its life. Because of the vulnebility of the poor it is certain that, over time, the project stategy will need to be periodically modified in line with changig circumstances. The plannig matrix thus becomes crucial in helping to ensure that the original hierarchy of problems remais the central focus of the revised strategy. The planning matrix provides a summy of why a project is carried out, what the project is expected to achieve, how the project is going to achieve these results, which extemal factors are crucial for the success of the project, how the success of the project can be assessed, where the data can be found to assess the success of the project, and what the project will cost (see Table 12.3).1o 10. Detaied guiddins for drawing up thc maix cm be obtaincd fom various resource documents, for example. NORA) (1990). Idenficadon an Design of Povy Rection Project 233 Table 12.3. Project Planning Matrix . - . ... . . -: . . Overall gVal7 * e d u c e d koms Lvels Income surveys Crdit used wisIY poverty in Expendiue Leve Expenditr surveys Additinal cme used wisely Region XYZ S:sinbleecreditprogmtu ,' "'' - S .*:_ SSavings/nvesment levels Savins Investment suveys Projectuxpors Affordable. Pevenu;ge of families takig Credit records Cedit used prdo anay by accessible loans the poor(no external f. . credi schem interference) Effectivc for x,ooo moivaional program poor famhis in Region f; ,, . .... . .,'f igt9a _ _ _ _ mgk ~~years ___ Ret - -* G r o u p Type of loans issued BoHrwer records Gronp credit discipline b a s e d denbped borrowing scheme . n. introduced High repaynrut rAtes Projea records Raw mterials and marmts * Viable accessible enterprise available for die poor No vilages borrowing Projea records Bks prepaed to coperate '.f..::..:* Vil.ge. based credit H . . '.. * operatitn- . ..:'..... '' established -:. E:..' ... ., ... .: , A,,,l vutjies, * Group Level of saving Project records Vilaes coniet to borrowv scheme haroduced * Villae Number and type of loan Prject rcords Committed, credible minuer .... cxensionl applcati available progrm inmoduend ......... . eVillager Number meetings held Project =cans Effective mouivators avalb momovadonl Number grups fourd Project records 'EmE. *'',.. tinig Number leads baied Project records Absear of disuptive demeats .stmblisdbd 13 IMPLEMENTIG POVERTY REDUCTION PROJECTS Neil 0Sufli%a As mentoned in previous *ptes, pove recton projecs cnt be implemened effectively unless thepoor dmrelvs are able to id* and atiulae teir dem as the bosis for suable projec design Whre ere are no preexistet organizaons of th poor (as s generay dte ase), thefirststp of ipmenrton istherfore tie challenging tk of orgnizng the proposed beneficares. Be_iary Organizatio The World Bank and varnous other bodies have consistently identified insmitonal faors as the most significant determinant of a pxoects success. In poverty-related projects, probably the most important insituonal factor is the absorptive capacity of the poor as represented by the stength of the organizatons controlled by lem. Unfrtunately, however, the poor are seldom orzanize. Because setting up new organizations for the poor is normally difficult and time, consuiing, a common tendency is eiter to fall back on existg struces, such as official village-lvel administrative units or to allocate the task of setfing up such organizations to those govemment msdtutions (for example, baks, agricultural depatmn) a already bave a widespread network m the target am. In terms of representing the interest of the poor, the record of most forms of local goverment is very poor. As Lewis (1988) has noted, local council tend to be eltist and maniated frm above; moreover, the delegaton of cental authority to locally selected governments is apt to wide the inquaes between classes because traditional elites are likely to be most transigent witin their own territory. For similar reasons, local cooperatives, whether indeped or quai-govrmment orgamizations, have rarely been an effective means of reaching the poor. Similarly, most raditonl insttuts such as banks and government deparments tend to reflect a alre that has dveloped in both training and practice thrugh a concern for the problems of the more afflue and bette- educated clietele. In eteime cases, as inm traditiona rural credit programs (see Chapters 4 and 6), atdes to the poor border on being opedy hostile, whie even in more neutr situaions regular employees canmot nonnally be expected to have much aptitde or enthiasm for a t:ask that reques usual levels of ene, entprise, and dedicationL The responsMility for beneficiary organization should be undertaken by a goverment agency only if that agency can demonst the necessary flexibility, sensitivity, and commimnt. Altbough more and more govermet agnci are showmig tis sensitivity (iculaly in urbn 235 236 Neil O&Slkiin areas),' in general, they have difficulty gaing the trust and cooperation of vulnerable groups. Tbus this demanding task is ftrquently subcontracted to an experienced nongovermuewt organization. If no appropriate govement agency or NGO is available, it is sometimes possibe to create a separate unit within a government agency under the leadership of a person of vision and commitment. The most successful organizations of the poor are those that focus on carrying out specific functions the poor consider important. Predictably, such organizaions usually concentrate on income generation, especialy in rural areas, where employment opportunities are both more limited and fickle. For this reason, communty-based savings and credit schemes have now become a common basis for organizing those in the community without assets. In urban areas, where a weaker sense of community usualy exists, such credit programs have yet to emerge on a large scale1. With the greater awareness and social mobility existing among the urban poor, organization has frequently been issue-based, and community organizaions such as slum improvement programs have arisen almost spontaeously to tackle threats or problems. Although some of these o ions have evolved into effective and broadly based organizations of the poor, more often they have been co-opted by a political party or have suffered internal divisiveness. A Modelfor Benftdwry Organizaton An important first step in implementing a poverty alleviation project is to undertake some form of soci preparaton.This is essentially a pre-investment phase designed to strengthen the absorptive capacity of communities, which, by dint of their poverty, isolation, or social stus, have been margnaized to the point where they are unable or unwilng to participate in mainsteam development activities. As described in Chapter 5, the urban poor are generally quie willing and able to participate if projects are appropriately designed. Thus the following model applies more particularly to the rural poor, although the general prmciples apply in almost all situations. The objective of social prepaation is to reinsfili in such communities the confidence, hope, and motivation they need to participate in development activities. This process normally involves helping communities to form local (grass roots) organations that can serve their development needs on a coninuing basis. Social preparation may take from three to twelve months, depending on the ompleity of the paicula environment. Many examples exist in which agcies impatient to see progress bhve taken shortcuts. The results have imvariably been disappoinig, just as they have when development agencies create beneficiary groups simply as an absorptive mechanism for funds already earmarked for their use. Even if a potential group has been advised it will only receive funding at some future time when vaious conditions have been met, there is a strong likelihood 1. Schems su as die Uban Communi Devdopm Prgam in Hyderaad and dhe Karpug povem Progm i Inone are csidered to live developd suh flexble and sensitive mechnsm for reachig the poor. In nual areas, where menoris are log, evn die most canst atmp by govme such as Sri L's Janasaviy Pgam nmay be nulified by bad experiences in the oast. 2. Sever impora cceptions such as the SEWA (Sdf-EmployM Womes Associatio) and te Working Wamens Fonrm have aegd m dia, but m each ca the deped on addina cmmon ficrs (bemg ftnale and self-mployed) to crae a strog sane of cmmuny. Impmeztdng Povey Projects 237 that on the day of judgment the beneficiary group will have fulfilled all the required conditions, but not as a result of its own natural maturation. The best moment to supply external resources is when the local organization has developed sufficient capacities of its own to digest the assistance as a suitable complement to the resources (members' contributions, labor, and materials) it has successfully deployed in various opertions. Providing funds before an organization has proved itself makes it all the more dependent, and the agency's resources are poured into an inappropriate vessel with a high likelihood of mishap. Social preparation usually consists of four phases: identification, mobilization, organization, and institutionalization. IDENTJFICATJON. In this initial phase, the task is to define the specific nature and location of the target group. Particular attenion is given to the level of social stratification that exists, for experience has shown that in stratified societies, beneficiary groups must be homogeneous (for example, all landless laborers, all women). Women, in particular, respond far more enthusiastically in separate women's groups than in mixed groups. MoBa1zAN. Because the poor are kely to be submerged in an environment of inertia and a sense of hopelessness, an outside catalyst will be required to stimulate change. The catalyst will be the development agency's community organizer (CO) or communty development worker (CDW) who takes up residence in the target villages. (Each CO or CDW is likely to be responsible for five to ten village-level communities at any given time.) The mode of stimulation depends on the local culture and the degree of the commumity's social and economic deprivation. In Illiterate communities, a functional literacy course may be very effective; in tribal cmmunities, cultural performances have been used to anger, cajole, and challenge the poor to restore their dignity and demand their rights. All stimnulation processes will result in a greater critical awareness among participants. Once stimulated, the interest and energy'of the community must be directed toward some constructive activity. One of the first acts of the development worker will be to select an "entry point' to the community. Although this entry point may be in a specific sector (such as health or drinking water supply), a broader approach is more useful. Many workers choose the medium of a village survey. By visitng the home of the poor on a hut-by-hut basis, the field worker gains the confidence of the community and identifies key people and problems in a relatively short time. The reawakened hope and energy of the communty is then immediately channeled into a fledglin organizational structure. Normally, this takes the form of an informal group of fifteen to fifty participants who meet at least weeldy to analyze their problems and consider solutions. ORGANIATiON. Once moblization has been achieved and a group formed, the task is to formalize it in order to increase its strength and susainability. Simultaneously, the group is encouraged to take its first joint action, for example, arranging, or if necessary demanding, a meedng with the local agricultural extension officer or agricultural bank manager. The objective during this stage is to inculcate a sense of cooperation, mutual discipline, and unity that can witstand internal power struggles and extenal interference or subversion. The formation of responsible leadership is a key eleme of this phase. The growth and development of the grovp will depend on whether it takes action from an early stage. Since members are bound to be poor, one of their main interests will be to generate income. Consequenty, most early meetn will be preoccupied with discussing the merits and risks of various options that might be available. It is at this stage that the group becomes ready to interact with outsideforces- -government agencie, banks, politicians, and the like. 238 NMU O'Sullvan One of the most popular and successfil strategies for group development at this stage is the group savings scheme. Since most poor people have had some experience of high-interest credit from moneylenders, the possibility of having access to cheap credit from within their own community normally generates considerable support. At the same time, the requirements for setting up even a simple revolving loan are substantial and will necessarily strengthen confidence, cohesion, discipline, and decisionmaking capacity within the poor community. Where a savings scheme has been successfully introduced, a logical extension is to formalize the scheme by linking it to a broader credit institution that has the capacity to provide access to higher levels of investmnent. Many community credit schemes depend on the prior formation of borrowers' organizations, not only to providejoint moral collateral (as an altemative to physical collateral) but also to reduce transaction costs (see Chapter 6). By advancing to this stage, the local organization is automatically strengdtened and its overall absorptive capacity is expanded.3 The purpose of the ultimate transitional stage is to ensure that the newly formed organization has its own integrity and that it is not simply a short-term vehicle created to serve the needs of some external agency. To this end, the field worker encourages maximum autonomy and avoids the imposition of any specific goals or timetables from outside. Ironically, beneficiaries are usually organized by an outside body with a specific outcome in mind. Although this approach may appear to conflict with the idea that groups need to establish their own programs and priorities, a mutually desired objective is likely to emerge if the planning team has carried out carefiul research and consultation during the project's design stage. A problem that is more likely to arise is that technical inputs and budget allocations will be difficult to coordinate with the unpredictable pace of the maturation of organizations from differing communities and locations. Such problems are best dealt with via underprogramming, as explained in the section on flexibility, below. INSTTUTiONALZAnON. This final phase recognzes the linitations and vulnerability of small pinmary units and attempts to overcome this problem by linking primary groups with secondary structures via a cluster Gr dibtrict-level organization to which each local level organization sends one or two representatives. This secondary level of organization is still small, usually consisting of about ten groups (that is, approximately 250 members). However, it does provide small communities with a means of articulating their needs, coordinating their efforts, and arranging more complex inputs (training, credit, and so on). Ideally, this secondary level will be extended to a teriary level and tdus allow development strategies to emerge at a greater pace and with more sophistication. A potentially powerful people's organization will have taken shape at this stage, and it will be welcomed by those governments that recognize their own resource limitations. However, it may be either coopted or, perhaps opposed, by those that are less certain of their own position and performance. Institutiol Factors Whereas the above comments pertain to the institutional development of the target group, particularly their ability to caystallize their demands, and increase their absorptive capacity, the 3. Strategies of tis tpe are followed by development agencies in Bangladesh such as Grameen Bank and BRAC. Invarably, these orgaaons observe a ramge of additional bcnefits over and above access to crdiL Increases in family planning acceptance, childrens ducation and income-generating inidatives are typical. Inyplomezdg Povert Projects 239 more common concem in the implementation of poverty projects is the institutional arrangements ffiat affect the supply of project inputs. Predictably, the meshing of supply and demand creates many challenges at the institutional level and fequently calls for a major reorientation of the attitudes and practices of the implementation agency. Reorienting Burweacracies Experience in many developing countries has repeatedly highlighted the problems implementing agencies face in readjusting their style of operation in order to cope with the special needs of the poor. Bureaucracies often develop a culture that is against the flexibility and sensitivity necessary for establishing effective working relatonships with the poor. During the design of a poverty reduction project, the planning team would have gained a clear understanding of the likely relationship between the proposed target group and implementing agency and would thus be able to judge the relative likelihood of compatibility or conflict. Where significant potential for conflict does exist, either the implementing agency must be appropriately transformed or an alternative agency sought. In the past there has been a tendency to assign sensitive or special projects to a project implementation unit, but experience has shown that when the time comes for handing back the project to a line agency, problems are likely to occur that may well jeopardize the sustinabiity of the project (see Chapter 15). Although it is temping to bypass the existing bureaucracy, autonomous project organizations are frequently criticized for allowing external investors or national politicians to take vital resources from other areas in order to create an unsustinabk. enclave in which impressive, but essentially short-term, gains can be made. At times, however, the goals may be so important, or the alternative chanels so compromised, that an autonomous body will be the best option. If such a course is taken, 'arefillly detailed strategies must be devised at the outset so as to minimize the disruption and loss of momentum that is likely to occur when the project is eventually reintegrated into a line agency. Even if no tension exists, it is essential to provide a special training program to sensitize field staff to the most fruitful ways of woridng with the poor. Experience has shown, however, that the trining of individual staff has little lasting impact until there has been some basic eorientation of the entire senior management. To date, there is a lack of proven tecques for transforming bureaucracies, although expeience in Nepal and the Philippines' has indicated the importance of (a) assigning the task to an experienced and highly committed leader, (b) using a highly participatory process to persuade staff members of existing perfbrmance shortfalls and of the political, budgetary, and institutional imperative of developing extension systm driven by the demands of the poor; using successful pilot projects to provide intensive on-t-job training for existing staff; ensuring that appropriate measures have been taken at the policy level to reinforce desired changes. One of the guiding principles in reonenting bureucracies sbould be to encourage staff to develop a client orientation in the same way that private enterprise does if it wishes to remain viable. The public service staff of many developing countries still retain, as the legacy of a feudal past, an exclusively upward orientation. 4. Experience in the Dhading pilot project of the Small Farmes Development Proect in Nepal and te Phlippne National Irrigation Admin (see Konn and Sly 1989). 240 Neil O'Suivan One of the principal desired outcomes in the reorientation of the buraucracy is that its style of planning and management move from a bluepdrt approach to what Korten (1980: 509- 17) has described as a process or learnWng approach. Most hierarchical or ons tend naturHy to gravitate toward a blueprint approach, which, however, is rarely successful in dealing with the poor since it -stumes (a) goals have been clearly specified in advance of design; (b) the design is comprehensive and is appropriate to the needs of each subpopulation in each distinct region at every given time; (c) reliable monitoring information is available and authorities have full control over the activities and performance of field staff, and (d) there is a stable external environment devoid of political, bureaucratic, or climatic disruptions. By contrast, a learning approach assumes that external situations cannot be controlled but rather that responsible organizations mst develop the capacity to continually analyze changing or variable field siuations with respect to the organization's basic objectives and to modify responses accordingly. Under this approach, planning is based on actal field experience and thus is continuous. That is to say, it operates from a reliable, two-way information system based on a common commitment to a clearly understood set of goals. The difficult but unavoidable challenge of reorienting supply-side institutions in this way parallels the task of constructing dynamic, beneficiary institutions. In both cases, a similar capacity is required to be able to cope with the onion-like layers of problems that emerge as the development process moves forward. As in beneficiary organzations, this capacity is created not just by acquiring technical knowledge but by developing fimdamental attributes, including leadership, group commitment, and concerns for equity. In certain situations it will become clear that existing institutions cannot realistically be transformed to meet the needs of the poor or that the new function required is somewhat different from that performed by traditional agencies. In these situations a strong case can usually be made for constructing an alternative channel, provided that it is seen as a long-term undertaking. In commercial banks operaig in rural areas, for example, much of the taditional function of meeting the credit needs of larger farmers must be nuintined, and in many societies the concerns of poor borrowers will never be given adequate pnonty. In such situations more positive results are much more likely to be achieved by creating totaly new instutions dedicated to the speciic needs of the poor. In the field of rural credit, most Asian counties have experimented in this direction, with probably the best performance being recorded by the Grameen Bank, the institution that has by far the tightest poverty focus (borrowers must have less than 0.5 acre of cultivable land or equivalent value of assets) and, thanks probably to its founder, a strong pro-poor culture. Saffing The critical link between a poverty reduction project and its target group is undoubtedly the project's- contact staff. If project field staff do not come from the same commnity as the beneficiaries or have difficulty in empathzing with them, problems are bound to arise. The severnty of these problems will in part depend on the "social distance" between the staff and target group and in part on the basic level of disempowerment fek by the latter. Where a significant gulf cannot be overcome, for example, by employing members of the target community itself or by applying special criteria for the selection or training of outide staff, a mutually acceptable intermediary should be considered. In many cases NGOs can be contracted to perform this function or, as in the case of one area of the Nepal Small Farmer Development Iiplrming Poverty Projeas 241 Program, persons nominated by the beneficiaries can act as volunteer intenmediaries between the project and the community.5 Poverty reduction projects often require a high level of committed and inspirational leadership if they are to overcome the normal obstacles. Major poverty reduction projects should give considerable attention to the recruitnent of such a person(s) and to their active involvement in finalizing the project design. One of the desirable attributes in the project leadership is a capacity to manage public relations and to ensure continuing support for the project until demonstrable positive results begin to emerge. Where appropriate leadership has been recruited, it is essential to try to keep it intact until an appropriate institutional culture has developed and key staff can be moved to other areas without destabilizing the project-the first three or four years is critical in this regard. Even if high-quality leadership is available, it is still vital to ensure that substantial effort is dedicated to training those second-line leaders who will ultimately assume responsibility for the project. Many poverty reduction projects have difficulty recruiting and retaining suitable staff, particularly in remote locations or in areas that offer low salaries. Although a positive institutional culture and local recruitment (unemployed youth are now commonly available even at the graduate level) may help in resolve some of these problems, appropriate rewards should be available to compensate for the frequent travel, long hours, and difficult conditions that are likely to be an essential part of the job. For staff in general, adequate training, motivation, and supervision are essential. If, as is normally the case, female staff are needed to reach the female nembers of the target group, special provision must be made to overcome the normal cultural bamers faced by women (for example, they can travel in pairs or be given on-site accommodation). Various techniques apart from recruitment, taining, and motivation can be used to maximize staff efficiency in the field (Heaver 1988). These techniques include establishing clear objectives for extensicn work, limiting the range of tasks, and using regular work routines to assist in locating and contacting community members. Fledbilizy Becuse of the extreme vulnerability of the poor, working with them requires a great degree of flexibility. At the same time, budgets and work plans must be scheduled well in advance even for normal project planning. Although a more flexible approach to budgets and schedules is higbly desirable, public administraton policy is itself frequendy inflexible; in such cases, however, , projects can be implemented inphases (with a separate budget for each phase) so reliable information to can be gathered before progressing to subseqent phases. Whereer possible, funding must be flexible enough to cope with ievitable new and unforeseen problems without the need to reformulate and resubmit budgets for approval. For the sake of sustainability, most poverty reduction projects seek to respond to the d&nand of the target group; however, the rate at which this demand is generated is notoriously difficult to predict and much damage has been done by harassed implementing agencies resorting 5. in the Da pg pilot project, connities wer invted to selected a nmre. reliable reseztaie with sufficient dme to act as dh volunty link with thie prject adminion. Whem such pero wored with only soci approbaio as reward, excellet trust was-established. Wherever financial reward was i the community teded to be.m distruf eve of teir own nomiee. 242 NAeil O0'Suli to supply-side targets as a means of confirming their work output.6 This problem can usually be handled by deliberately underprogramming in the initial stages so that, at any given time, extension activitdes are actuly taking place in slighldy more communities than have been budgeted for in terms of follow-up services. In this way, the imple g agency is never forced (in order to meet targets) to initiate activities in a place where beneficiaries are not yet fully prepared and anxious to participate. Unde tpro i is useful for dealing with ny aspects of poverty reduction projects for it helps provide managers with the possibility of exercising choice in situations where it is difficult to specify in advance, exactly who or what will be required at some later date. For example in contact-staff training programs, the cost of training an additional, say, 20 percent of candidates is usually minimal in terms of overall budgets; however, the ability to identify and dispose of inappropriate trainees before they are due for deployment can add enormously to the quality and morale of the project. Needless to say, it is essential to have sufficient budget flexibility to use underprogramming techniques. As a matter of prnciple, projects should, wherever possible, be launched first in those locations where the chances of success are highest, and thereby would provide concrete evidence of the practicability of the project to recalcitrant conmiunities. In most poverty-focused projects, flexibility will be improved by decentralzing authority as far as is possible, particularly where a wide diversity of comnTaities and locations are to be covered. The tradition in many Asian countries, however, has been to equate 'decentrlization" with "centralized planning" accompanied by "decentralized implemenation. " This attitude, when applied at the g.-ass roots level, is lhkely to result in reduced participation by giving beneficiaries the impression that the project is in reality not "theirs" but is rather the property of some remote authority and should be treated accordingy. In keeping with the desire to maximize the agency's responsiveness as well as its learning capacity, supervision should be kept simple and chins of command kept short. Where the project's goals have been fully understood and accepted by both the staff and the target group and a high level of participation has been generated, much of the day-to-day supervision will be automatically taken care of by the beneficiaries themselves. Consequny, external supervision and management infonnation will be able to concentrate on "exception reporting-" Coordination As a guiding principle, the chances of a project's success can be considered to be inversely proportional to the number of separate agencies involved in its implementation. Problems of interagency coordination are notorious in most developing countries. Thus, wherever possible, project control (and financial resources) should be vested in a single agency. Unfortunately, if one agency is appointed to be the lead agency ahead of other potential candidates, the unsuccessful agencies will sometimes lose interest in the project, even though their cooperation and input may still be vital. Where coordination difficulties are anticipated, two basic approache can serve to reduce difficulties. First, at the design stage, all interested agencies should have been identified and directly involved i finlizig the project design. Wherever this 6. in th Tndian IegrAed Rwrl D pme Poect, amitr worki I difficult regions have fre ly resored to highly supaficial. la meks (canival) i order wo mem loan distrifon margrs. Recovery of loans disauted in dtis way is reporedly as low as 20-25 parce (see Pulley 1989). ImWpfmeaing Povery Projecs 243 has been done, the subsequent commitment to the project from all agencies has generally been high. At the same time, basic management practices must be followed zealously to ensure that there is a clear and agreed definition of responsibility for every element of the program and that appropriate and inalienable budgets are available. Once such commitment has been established, it is then important to devise appropriate coordinating mechanisms to provide regular feedback on actual performance and to facilitate prompt remedial action. Pilot Projects Apart from its obvious objective of verifying overall project strategy, a pilot phase offers the following advantages: allows unforeseen obstacles or bottlenecks to be identified; develops a capacity for identifying and testing a limited range of inputs rather than resorting to a specific blueprint that may not be appropriate for many communities or situations; helps identify staff with special talents or commitment who can assume key roles in the expanded project; provides general field staff with practical training before the main project starts up; provides an opportunity to test alternative approaches to community organizing. Given the general scarcity of public resources and the basic vulnerability of the poor, unless there are already clearly established and successful precedents, the use of an appropriate pilot project should be considered mandatory in most poverty reduction projects. Arguments that pilot projects are not replicable and take too long to complete can usually be countered by ensuring that the pilot project is itself of significant size and is kept ahead of the main project by a minimmm time factor (the time lag will vary according to the degree of confidence in a particlar operation but, in general, problems in the pilot project can be quickly detected via an ntensive monitoring system). Targets In an effort to improve efficiency and output, most antipoverty projects maintain strict monitoring based on performance as measured against specific physical targets, typically using measures such as the number of families covered or the volume of goods or services disbursed. Although these parnmeters provide a fast and cheap method of monitoring progress, they arc essentially supply-side indicators and will only be reliable if the target group already has a high level of awareness and absorptive capacity.7 In terms of their ability to assist in predicting likely project sustainability, supply indicators are by and large irrelevant. In the more common situation, where the poor are neither aware nor organized, supply- type indicators have two serious limitations. First, as with many official statistics gleaned from a distant network of sources, each with a variety of vested interests, the end result is notoriously unreliable. More important, however, is the situation that arises when project statistics are fairly reliable and the staff use these data as the basis for their evaluations. Here, the staff is amnost irresistbly drawn to shoiL-term, quantitative matters such as coverage rather than qualitative concerns such as participation and taiing. This latter trend imvarably demoralizes committed 7. In the Indian states of Kerala and West Bengal, for cxampl4 wbere both formal and poEitcal awaeness is highly developed, the Integrated Rural Development Program eczountered much cess distortio an recover=is thm in most other s (Ptey 19E8. 244 Neil O'Suwvan staff, nreases the insensitivity and inflexibility of project adminstration, and reduces the likelihood of sustainability. Nongoverment Organizatons In many developing countries, N(GOs have long played an important role in supplementing govermnent services, particlarly to the poor and disadvantaged. More recently, however, many nongovernment development organizations (NGDOs) have emerged with a strong commitment to the empowerment of poor communities with which they work. Although these organizaions are more likely to be professional, rather than voluntary in nature, they are still able to claim a high level of effectiveness and efficiency in their work because of their capacity to geerate m mm benefciary participation and volunty energy.' The increasing involvement of NGOs reflects the generally accepted belief that such organizations have a comparative advantage over traditional govemment agencies in those situations where the enthusiastic participation of the target community is required to achieve a sustinable impact. The key characteristics exhibited by sucsfid NGDOs include their proxnmty, both physical and social, to the poor; trust and credbilbty, usually only developed over a period of dme; commmn to enduring the frustrations of workig with communities that will iniiall y be suspicious and demoralzed; and theirflexibility, and thus a wilingness and capacity, to modify project strategies in accordance with the frequently changing requirements of vulnerable communities. In recognition of the above characterstics, both govemments and external development agencies are increasingly making use of NGDOs in poverty-related projects. NGDOs often play a role in communty surveys at the design stage, in community organizing, and in monioring and evaluation. Although NGDOs are frequently subcontracted to perform discrete elements of a project, their oft-expressed preference is that, as far as possible, they be trated as a partner and given an oppornity to be involved in a project from its inception, in keeping with the NGO's wide-ranging development concerns. The advantage of such an arrangement is dtat following the nominal conclusion of a project and the witdrwal of the iwplementing agency to other areas or projects, it is quite likely that an involved NGO would be keen to continue working in the area unDi a basic level of genuine self-reliance had been reached. Most development NGOs are highly sensitive to what they descrbe as the develpment process and insist that sustainability depends on the method by which economic advancement is achieved.9 In situations where environmental degradation or political harassment becomes a major concer of the poor, for example, development NGOs will frequeny take up an advocary role on behalf of the poor. In fict, a small number of NGDOs focus exclusively on this task. Because nongovernnent agencies cherish their independence, wheneas govermment agencies are 8. One of the mn dmanic examples of te productive use of wbomwy asry is to be found in the lare micro-credt prM such as those of BRAC and Gamn Bak in Ba_s. Hat otewise prohbitively high bankig sion cost can be made magable by orgazig smnal borvwer groups to assue muc of the survisory fimcion nmmufly undakn by highly paid bank staff. 9. Developme NGOs def th developmen proce as '"he process by which idivihus idedfy tiemselves as a co_onmity and coiecvedy acquire fte necesay owwledgM power. vaues and -s to irversibly sa and expd tha commuity's reources for the benefic of all its members whit beifg at the expes of ete communiies or of dhe evironmen' lmpklemdng Poerty Pjes 245 actmed to excering total control, tensions are likely to arise during project implmnentation unless a clear understanding has been forged at the outset as to their respective roles. 14 DESIGNING GENDER-SENSrnE POVERTY REDUCTION PROJECTS Neil O'SWvapn ntroduction Earlier chapters emphasized disaggregating "comnmities" into subpopulations early in the project design stage. Gendr anaysis repusents a firther dissection of subpopulations undertaken in order to separately identify the likely levels of participation and relative impact that a particular design will produce for women and men, respectively. The reason for incorporAtg gender analysis into project design relates to concems for both equity and economics. In the case of equity, it is now weil recognized that, even among the poor, women are often especially disadvantaged in terms of worldoads, wage rates, literacy, and health. Projects aimed at poverty reduction are themselves an atempt to introduce greater equity into the general social situafion and must, to be intemrally consistent, also attenpt to advance the relative position of women within the target group. At the very least, the design of poverty reduction projects must ensure that the position of women is not in any way worsened, as frequently happens, for example, via increasing the workload or reducing reurns Even to achieve this much, however, planners must have a detailed understandg of women's overall role in a prospeeive project area. In a purely economic sense, women play a key role in many activities on which a poverty project is likely to focus, for example, enterprise development and credit, crop improvement and maintenance, water and sanitation, health and education. Accordingly, the final project design can only ensure women's participation by taking into account their preferences and the competing denands for their time and energy. Wherever a major decisionmaking or implemeation role is played by women in these activities, project design must be based on effective consultation with them. Finally, poverty reduction projects in rual areas must take into account the increasingly common problem of female-headed households, which now account for 20 percent of families in most parts of rural Asia. A recurrent cause of the lack of female pardcipation in poverty reduction projects has been the incorrect assumptions made by (usually male) project planners and iplemeers about women's roles in nonhousehold enterprises and about what project beneficiaries consider "acceptable" forms of female participation. Planners frequety relegate women to the domestc or household production sphere and assume that their role in agricultur or off-farm income generation is inconsequential compared with that in the home and family management. In reality, a poor household functions as a unit and makes a range of management decisions regarding the production and marketing of goods and the deployment of its labor. Heyzer (see Chapter 7) describes a number of additional factors that underline the iportance of gender issues in antpoverty programs. Among ffiese concerns is the fear that not only are existing divisions of labor allocating the most tedious and labor-intensive asks to women but that new employment initiatives tend autatically to build on and even inensiify this distortion. Similarly, conventional Wprches to project design frequently fail to address 247 248 Neil O'Suivan care, physical stress, and so on) and that are likely to affect the equitable distnnution of any new resources generated by external projects. In response to this situation, an effective project response must normally address a number of issues ranging from the protection of women's existing sources of status and income; the promotion of equitable access to essendal services, credit, and agricultural inputs; the reduction of time spent on routine domestic chores such as water and fuel collection; the development of women's support organizations and networks; and finally, elimination of all laws that discriminate against women's ownership and control of productivre assets. The above stereotyping, coupled with an assumption that women are little interested in matters outside the home, has meant that many poverty projects restrict thfe role of women to that of 'clients" in health and family plannig programs while ordaining men as the 'target group' in all production- related matters. One of the reasons for this predisposition on the part of project planners is the tiradition of viewing the (male-dominated) household as the basic economic unit rather than assessing the individual roles of its members. Further, it is likely that senior (and therefore older) planners, whose attitudes have been formed in an earlier era, have failed to keep pace with the rapid changes that are tacing place in the lives of rural women throughout Asia Attempts are sometimes made to incorporate gender-sensitivity in a project design by adding on special 'women's components" to an otherwise male-oriented project or to provide a special budget for women's training etc. The far preferable approach, however, is to treat women as a specific subpopulation and so ensure that appropriate consultations and design strategies are evolved that will become an integrated part of the final project design. For this reason, this chapter provides guidelines that can be applied both during the initial rapid social assessme (lSA) phase as well as in the more detailed participatory project design (PPD) phase. The alternative approach to integrated projects is to design special poverty projects for women. In practice, both sttegies are normally reqired for even if all future poverty reduction projects are designed in an appropriately gender-sensitive manner, there remains an additional need for investment to address the present status of women, who remain disadvantaged even among the poor. In certain situations, it is also strategically desirable to focus specifically on women a fact that has been recognized in the country gender studies undertaken recently by both the Asian Development Bank and the World Bank.' Benfit of Gender-Seusitive Project Design Apart from concems for equity, the first and most obvious benefit of gender-sensitive project design is that it will ensure a higher level of beneficiary participation, greater impact, and more certain sustinabhlity of project benefits. Project designs that deliberately facilitate the involvement of women benefit in a number of tangible, if little-recognized, areas. For example, women are in effect the managers of poverty since, they are generally the ones who must try to spread a meager icome over the many conflicting necessities for a family's survival. In this role, 1. A typical case is that of the Self-Employed Women's association (SEWA) in Almnexad, whosm initial attempts to imprve the posin of femak worke focusad on in_tgraimg women int exisig tctilc us. Because of the stngth and lack of sesitivity and sympathy of the exisig bodies the iniial satew was evually abandond in favor of setng up ectlusivdy women's - Demignng Geder-Sensitive Povety Reduction Projects 249 women invariably become highly responsible in relation to the handling of money, the wise use of any additional income, and the cautious choice of any possible investment opporunmities3 The managers of those projects in which women are involved frequently describe women's performance as being more responsive and more reliable than their male counterparts. This greater responsiveness of women is attributed to the fact that m many situations a development project frequently represents a much-valued opportmity for women to move outside a traditionally restrictive environment in which she has litle economic opporunity, an inense frustration at being unable to adequately feed and clothe her children, a high level of dependence on her husband, and often a strong sense of social inferiority and vulnerbility. At a purely practical level, the integration of women in a project virtually doubles the number of potential participants, thus providing a greater level of "competition" for involvement in the project and a better means of ensuring that the project is responding to local demand. Similarly, women's involvement not only expands the pool of available talent but also increases group strength (since women are generally more cenrally located and are more likely to be able or willing to attend group meetngs). Many observers also consider women to be more task- oriented and less liable to be distracted by matters of competition and status. Finally, women are said to be very "level headed" and less likely to embark on overambitious or foolhardy development schemes. Encouraging Women's Participation Strengthening the absorptive capacity of any weak or vulnerable subpopulation involves the development of some form of commumity organzation for that group. If in the promotion of village-level organizations; however, representadon is based on the principle of 'one household, one vote," males tend to assume tat position regardless of whether or not it is more appropriate. Such situations can be avoided in a number of ways, first by establishing separate men's and women's organizations as, for example, in savings and credit clubs. As described in Chapter 5, the normal socially liberating effects of urban life mean that the need for separate womn's groups among the urban poor is much less pronouncedc Where separate organizations are culturally unacceptable or where the sexual division of labor is such that women play a significant role, as they do in most aspects of subsistence agriculture or water use, membership may be organized on a dual basis, that is, husband and wife together. However, unless meeting bmes and places are carefully selected, it is quite possible that women may still be prevented from attendig by virtue of their pervasive domestic responsibilities. Alternatively, as a reflection of the actual relationship between men and women in carrying out their income-generatng activities, membership may be arranged on an altenae basis. In this way, differences between busbands and wives can be reduced in response to the organization's demands on their resources. Women's involvement at meetings can, for example, 2. The exein of micrcdit projects throughout Asia has rpeaedly demons thr e supcei rpaymet peformn of famale borrws - a key for in Ie development of succss lasge-sck crdt progrm such as BRAC amd the Gramen Bank in Bangladsh. Alkhgh some attribute the better repayment peformance to women's choice of less risky (and perhaps less remuneative) projects. it is significant that an expeienced credit organization like BRAC (Bangladesh Rural Advancmcnt Comme) rtinely allows women to borrw vrey soon after poinmg a savigs grups whereas m ar required to wait several unonths durig whichi it is assumed they wil absorb and accpt the basic caudon and discline necessary for a sustainable credit program. 250 Neif O'Su lvan help to extract firmer commitments from members to pay membership fees and other dues since it is commnldy the woman who is responsible for the household's financial obligations and general expenditure. The question of role models has also been shown to be important in encouraging women not only to participate but also to assme positions of leadership. Where in the past women have not customarily taken leadership positions, it will frequently be necessary for the project's community organizers or contact staff to drectly intervene by encouraging and promoting the involvement of suitable female candidates. Once the precedent has been established, thereafter talented women will normally be selected purely on the basis of their potential contnbution to the task at hand. Where cultural traditions are likely to inhibit the active participation of women even though they have been granted nominal membership of a primary group, involvement can be stzwngthened by setting up women's subcomtrees and progressively allocating greater levels of responsibility to such groups as their confidence and strength increases. Alternatively, copleme=wy groups may be established in which responsibility for an overall task is allocated between men and woren on the basis of separate interests or capacities (for example, men may accwt responsibility for the construction of a water supply system with women taking responsibility for the operation and maintenance). Rapid Social Asessment During the rapid social assessment (RSA), as descrbed in Chapter 12, the project design tean seeks to clearly idenfy the target groups and to predict the potecntial need or demand of different subpopulations for a proposed project, given their levels of development and absorptive capacity. It should be wted, however, that RSA is but a preliminay step before the derailed design phase and most often relies on frequently inadequate and disaggregated data. In order to gain the necessary detailed data for the final design, the RSA would be used to identiy the necessary surveys and analyses that would be further required? The RSA also identifies the likely sources and levels of support and opposition toward the project. In order to ensure chat the eventual design is truly gender-sensitve, the RSA treats women as dinct subpopulations. As mentoned earlier, the RSA approach consists of five steps: a. Identification of the target population Here the process involves: 0 Development of a simle iznuput/pt nmodel to assess the potental socioeconomic and cultmal impacts of a proposed project on women, as distinct from men. 3. Addion work tat wold typically be requird might indudt (1) use of labor mrket surves to exmin female parpuzion in diffare conomic secmu orw eanmie wage difterendals; () usc of household iwo and exediture and credit surveys to assess diltTes between a nd fmale aondp patens and to examme th efrcz of inrased fnale earnig n househ dd consumptin pmes; and (3) health surveys and measums to asses differetal morbidiy and moiy paes or diffeetial manution Designing G;ender-Sensitive Poverty Reduction Projects 251 * The collection of demographic data on women within the overall arget population from formal (government, academic) and informal (NGO, private sector) sources. This information would include ownership of resources, as well as health, educational, and soc J status. * ldentifcation of the relaive positions of women within the major subpopuldons in the target area. Basic subpopulations will already have been identified on the basis of land and capital heldings, employment status, level of social organization, ethnic cultural divisions, and so on. Ideally, each subpopulation will then be divided further into male and female subsections (see Table 14.1). In certain situations-for example, where there are fewer differences between women from differing subpopulations than between their male counterparts-it may be possible to group women into a smaller number of subpopulations. b. Rating the Level of Development of Each Subpopulation Suggested guidelines for rating the level of development of each female subgroup are as folows: High: Those who own or have access to resources in their own right; participate fully in all decisionmaking affecting their own lives and that of their families; have received a secondary education, have unrestncted access to institutional credit, political power, health services, and the like; and have solid spacious housing. Medium: Those with limited decisionmaking powers and access to resources in their own right, primary education and limited access to the above services. Modest housing. Low: Those who do not have decisiomnaking powers and access to resources in their own right, have not completed primary education, have no access to formal credit and little or no access to health facilities. c. Assessing the Target Population's Need or Demand for a Project Several steps must be taken to assess the target population's need or demand for a project: * Afield visit to the project area * Direct observaion of the living conditions of women (to cross-check data and ratings under steps a and b) o Consultations uth women to ascertain their general development pnorities (this does not involve direct discussion of the proposed project) * Analyzing t/e felr need for a project (the degree to which the project addresses women's priority concerns) and the dend for a project (the degree to which 252 Nel O0'Sadiw women from that subpopulation would be prepared to contnbute in order to become involved in a project) * Rating of the level of need or demand for each subpopulation. Suggested guidelines for radng the level of need or demand are as follows: High: Strng and spontaneous expressions of need related to the development problem that the proposed project is intended to address. Obvious preparedness to involve and contnibute. Mediwn: Some interest or need in relation to the development problem being addressed by the proposed project but not as a first priority. Limited contribution likely. Low: Preoccupation with development problems other than those that the proposed project is intended to address. Little liklihood of cotions. d. Assessing Absorive Cpacity During the site visit, seveal furthr fators are analyzed in order to assess whether women in the target subpopulation are likely to be able to absorb proposed project inputs. Seveal factors need to be considered: * The level of sodra) coheion among women, as imdicated by evidence of cooperation, equity, and mutual protection of the rights of women. * A spiet of self-reliance, as indicated by the fequency of previous wome's self-help iiiatives, lan repayment history, use or misus of past development opportnities, and so on. * Existing role of women in local connais organzaton, if any. * Te level of rdevant knwledge and skils of women in comparison with men in the target population The above information is then analyzed for each subpulaop on in order to rate its likely absorptive capaciq. Suggested guidelines for rating are as follows: High: Homogeneous women's community (socially cohesive), active and positive experience with self-help initiaives, proven capacity for financial u, acdve as offiares in local organizatons, direct linis with government agencies, technically innovative (new eterprises, succesfl kitche-gardens, effective health and sanition systems, and so on. Med : Limited social uity, mixed success with past development programs, minor involvement in local organizations, occasional direct meetings with government agencies, technically Static. Deigning Geder-Senfsiw Povt Reddon Projecs 253 Low: Lack of unity, no history of joint women's activities or actions, minimal direct involvement in past development programs, no involvement in local organizations. c. Assessing Relationships Given the cntical nature of the relationship between the target communty and both the commuity at large and the proposed implemenig agency, it is necessary during the RSA to clearly assess the mutual feelings and attitudes that exist between women in the target poor community, males in the same community, and relevant govermnen agencies and outsiders. For a mral project aimed, for eample, at providing an !ztegrated package of ints for impoverished women, these data would be reported in summary form, as shown in Table 14.1. £ Reporting and Interpreting the RSA for Women Table 14.1 RSA Won's Report Form .W T OFd 1 2 Levl _L _ dr ein ht *C0CeSqR :;*h Dr ". :e . : :: ... ... :: ; rgt. .* . Triba women LOW H*Whlow Medium 15 Stron aDd dlowmekit, but uneducated; m d by suiuuew' Low Hiow LOW 40 Very poor health and faireS education kvds low sel- e.e. 30 paces hosehol Smalllmarginal Low EIgh/medium Medim 30 Ke but disranized and farmes unaware; 18 percent .houoldbheds lac fammrs Medium MeaGianamedin Medium 15 Socially ismlated; limitd education The above report form is suplemened by an additional narratve report detailing the relationships that exist betwee women from the above subpopdations and other groups or acors likely to influence ffie outcome of the project (see following exmple). 254 Neil O'Suflivan Table 14.2 Relationship Report (for Women) ..... - -- . ..... .: .n.: :.an Lr p- .-e - Tribal womnen HsieLittle- contact Nfo contact NeuuiliuintidWatng Subsistenc farmes little contatotile Little conact No contact Neutarintidating Smll farmers Little con Little cona Lite contctw NeraEUsppontive The results of tbe RSA may then be used to reassess: the appropriateness of project goals and purposes from the perspective of women; the likelihood that women in the different subtapulations will be willing and able to accept or participate in the proposed project; whether a participatory project design (PPD) phase needs to be undertken to maxmize participation and a positive impact in respect of women; what the focus of the PPD should be so as to ensure the effective paicipation of women. From the perspective of women, the rneed for the PPD is determined as follows: RSA Result Response a. Where all of the followig coditions prevail: No PPD * no negative impact on women * high level of need or demand by women * high absorptive capacity of women b. Where my of the following conditions prevail: PPD required * limitations in need or demand of women * defects in absorptive capacity of women * potental negative impacts for women in some sub- populaifons c. Where there are high levels of potential Relcation or negative impact on women, significant opposition, retinkiig of or major limitations to absorptve capacity project concept The focus for the PPD is also determined according to the results of the RSA: * The PPD would nomaily focus on means of refining the design to pardcipation of female subpopulations with medium and low levels of development. Designing Gender-Sensidve Poverty Redaion Projects 255 * Similarly, where the RSA raises doubts as to the ability of particular, female subpopulations to gain equal access to project benefits, the PPD would focus on the need to develop methods for direcdy targeting this group. In practice, this would normally involve a detailed mapping and analysis of women's productive and reproductive roles within each subpopulation. * Where the RSA exposes weaknesses in the absorptive capacity of women, the PPD would focus on designing a strategy for the development of local women's organizations (social preparation). * Where, during the RSA, a female subpopulation is assessed as being technically undeveloped, the PPD would focus on working with the implementing agency to design campaigns and women's training and education programs, as well as assess the maximum amount of technical innovation that could be appreciated and absorbed by the women in question. Participatory Project Desi In relationship to women, the objective of the participatory project design (PPD) is to optimize project design by integrating women's preferences and priorities while at the same time developing realistic but appropriate project implementation strategies to which all parties have made a commitment. The PPD involves intensive invetigations and consultations with women and is undertaken in conjunction with economic, institutional, and Environmental analyses, where required. The PPD is nonmally carried out independently of the proposed implementing agency by gender analysts in the planning team who have both the sklls and the commitment to ensure that a genuine consultation with women takes place. In view of the likelihood that a traditional male orientation in at least some of the bureaucracies will be involved in the project, consultation workshops are used wherever possible to increase the level of sensitization in reion to women and to reach a consensus on effective means of positivttly involving women in the project. The task of the planning team is broken down, as described in Cbapter 12, into four basic steps, problem analysis, objecdves analysis, atemnadves analysis, and development of aplanning matrix. Since the basic details of the approach taken in each of these stages has already been described, what follows are specific concerns that must be pursued at each stage to ensure a final, gender-sensitive design. Problem Analysis This stage involves in-depth and systematic consultations with a representative sample of women from each subpopulation. The consultation will focus on identifying the core and relaxed problems as seen by women; however, the emphasis will be on concerns highlighted during the RSA. Agan problem trees should be developed for each of the female supopuiations (see Figure 14. 1, which is a gender-based expansion of that illustrated in Chapt er 12). Workshops are then held for each of the parties to compare their perceptions of the problem and to seek 256 Ndr O'Sullvan agreement on common analysis. In the case of female populations with low levels of developme, joint consultation may be difficult to achieve and wilU require the allocation of a disproportionate amount of time and attention. For these groups, joint consulation may need to be deferred undl the design is close to finalization. In the interim, the planing team (which should obviously include women), may itself need to need to directly represent the concerns of women. Meanwhile, intensive direct consultation would be carried out and would typically involve: * A detailed mapping of women's productive and reproductive roles (including household management) and analysis of resultant opportunities and constraints. * Participatory surveys (quality of social services accessible to women, cultural constraint on women's activities both real and perceived, role in village-level insitutons, and continuig projects and programs for women). * A discussion or analysis of the survey results and the development contxt in the project area (constraint on women's, production, consumption and savings, social welfare, orional capability, problems, poteial solutions). * Discussions related to the project concept (past experiences, risks, willingss to participate or contnbute, technological preferences, preferred conditions, and institutional Objectives Analysis As before, the purpose of the objecti analysis is to identfy the potential altemadve interventons available to resolve the problems and to thus transform the problem re into a hierarchy of objecives, or objectives tree (see Figure 14.2). Alternatives Analysis This stage identifies and evahlates those altemative interventions and strategies that would involve women in helping to achieve the project's agreed objecives. Referring to the previous example, it is inevitable that the perspectives of women would give rise to diffeing strategic preferences, for exmmple, the deployment of female extesion staff, or inclusion of a fimuctional literacy component. From its earlier analysis, the planning team should generally be able to predict and discount male antipathy to such measures and to steer the planning workshops toward the most participatory and equitable resolution. OPThrG PROJECT DE5ON. This stage involves selecdtg the best options for incorporatig women's priorities and preferences, maximizig the absorptive capacity of women, and selecting technical strategies appropriate to women. Figure 12.1 Problem Tree Applied to Gender Anlalysis - -- | | ~~No SUi l No Ae to Affordobl INo Ca&W No Bankleib of mm tof . -I 1" NoHio Dft ^ Y ] O da As N h lboomo~~~~~~~~N E' . aw a_ | IN°l~~~~Rte tdf | LOW" Soo c ld LOGEm l lu | ;1Dho || L | Noto U bz ll | I~~~~~~~~~cmas mwl mak No C1t W Ith 258 Neil O'SuMivan a. Iltegrating Target Priorities and Preferences By ranking project goals according to their place in the priority of respective female subpopulations, it will be possible to predict the likely levels of participation for each and thus to consider what variations or expauion of the project components would maximize participation. b. Maximizing Absorptive Capacity Female subpopulations among the poor will frequently have little, if any, experience in self-reliant, development initiatives and the project design would most likely need to provide for the mobilization and education of the female target group and the development of appropriate community organizations prior to the introduction of external inputs. That is, a social preparation phase should be incorporated (see Chapter 12). Where there are deficiencies in the relationship between the proposed implementing agency and the female target group that could affect their participation, project design would need to explore mechanisms for overcoming these problems. Where the problem is simply a lack of expenence in dealing with women, apprpriate institutional strengthening would involve the appointment of significant numbers of female field staff and the provision of sensitivity training for male field staff. In projects that do not focus primarily on women's participation, the sensitivity of contact staff to women's aspirations and interests in the project becomes crucial. c. The Use of NGOs as Intermediaries Where antipathy toward women is likely to be encountered within the proposed implementing agency, the use of NGOs as intermediaries, the-appointment of extension staff from within the target group itself, or the appointment or establishment of an alternative implementing agency may be necessary. Where women have been assessed as having "low" levels of knowledge or skill with regard to new technologies, consideration must be given to overomig these limitations by training and education components designed to increase absorptive capacity. Such training, however, should itself be undertaken in a gender-sensitive manner (for example, provision of a series of short, on-site training programs for women rather than a single, longer-term program for the whole community at a distant campus). Each of the above options has budgetary implications and it is important that they be explicitly recognized and included in the project budget. d. Selectn Appropriate Technical Options In general, technical options need to be examined in terms of familiarity, size, location, timing, and complexity so as to identify the project design that will most accurately reflect-the priorities and absorptive capacity of the female target group. As a rule, the options sought should be compatiUble with those preexistent daily and seasonal demands that should have been idenified in the process of mapping women's productive and reproductive roles. If the proposed project involves the introduction of new technologies, the degree to which the innovation accords with, or departs from, the existing practices of women and their likely reaction must be carefully assessed. Whatwer benefits the proposed project provides, it will Figure 14.2 Objectives Tree Applied to Gender Analysis I-- AQet Mad" h CI PA I I . l~~~~~~~~~~~ w , RPoa FI wHxbR* wHoumMe_| Xm I I I - i ll SUPAVdbbFe wdvmw syt _ l . I hdSdl 259 260 Neil OSdli&wan invariably change the tilationship between men and women in terns of the relative proportion Where a new household technology or activity provides more income for the male and a increased (nomuati) workload for the female, then project sustainability is unliely. Care must similaily be taken when introducing new technologies that have not already been successfiully intrduced among similar female subpopulations elsewhere. (if, for example, there has been no prior experience with a proposed new crop or enterprise, consultation with women will not be able to reveal the amount of time that will be required for processing or marketing a scheme will remain uncertain.) If women's participation is likely to be reduced or disrupted as a result of social constrais in a target population, project design strategies will need to be developed to overcome ese problems (for examWle, direct targeting, indivisible inputs, compensatory components). In all cases it is import to know to what extent women will retain control over any increases in their earnings. Zhe Projed Plan Matrir As before, the final outcome of the PPD is the project planning matrix or logical framework that is crucial in helping to ensure that the original hierarchy of problems remains the cental focus of the project strategy even when, as is inevitable, ongoing monitoring and evaluation reveal that need for modifications i strategy so as to ensure the involvement and benefit of vulnerable groups, such as women. 15 SUSTALNABIT OF POVERTY ALLEVIATION PROJECTS Neil O'Suaivwan The poor susrainbilty f poverty reduction projects contus to frustrate development pklners in many pas of Asia. Although the reasons for these failures vary, they are usually the result of inadequate attention to i ostional arrangemes at the design stage. both on the supply and the derand side or to the lack of politicat will needed to negotiate the inevitable sodal, economic, and techmical barriers that separate the margindized poor from the mainstream of socety. Problems can also aise from varnous other design flaws, as weUl as from the lad of aention to monitoring and evluation. Insitutional Factors Developmena Orienation Viewed from a broad perspective, the intrinsic vulnerability of the poor means that no matter how successful a project may be initially, the narrower its focus, the more likely it is to evenuay fail when confronted with the wide range of hnman or technical problems that sooner or later emerge. Although every project must have a specific, tangible objective, the manner in which it is implemented will have signficant bearing on the oveall empowerment of the target groups and thus on their ability to cope with unexpected problems and to sustain the specif benefits derived from the project. In practice, what is required of an implemeing agency is to establiish and maintain an instional cultUre that encourages all staff, i their dealings with the poor, to avoid traditional welfare-tYpe attitudes and to conceate instead on dealig with the poor only in ways that increase their confidence and self-reliance.' The achievement and maintenance of this development attitude will largely depend on the strength of leadership and the availabilit of on- going train and incentives for field staff. The agreed definition of what, in practice, constitutes a "development" approach may vary in diffeent contexts (see, for example, the definiton presented in Chapter 12). A woring definition, however, should not only specify goals concerning a sustinable increase in hnuman and financial resources but should recognize that these resources must be acquired wihiin a framework that acknowledges tht- need for reinforcing agreed social values such as equity, cooperation, and environmental protectic 1. The Grameen Bank, for oxamplk, today fuhns strictly as a finaucial itimion. Since its incepim however, Gramcen has establisbed a clear develpmnt cultum which both staff and bonowers are a ty encouraged to maintain a certin standard of disciplined behavior. This cature has lbstred an attude of mutual support but also saoe potenaly dagin practies. For example, bonwers '..io Iend to ouwsiders at exmobitant iftres rles are aueomaticay su*ende2. 261 262 Nei O'Sulivan Beneftciay Organiations No matter how wefl designed a project strategy may be and how high the level of partcipaion, sooner or later the beneficiary will encounter problems with replacing supplies, uncoordinated inputs, market access, political interference, and so on. As long as the project infrastructure remains intact, as in the case of most basic health, education, and urban services projects, it may well be possible to remedy these problems. However, wherever that support structure is evenutafly withdrawn, as normally must happen in ruraI income-generating projects, it is extremely difficult for a poor household acting alone to sustain the new-found benefits. Ideally, weaknesses in the absorptive capacity of the target groups will have been identified at the first stage of the design process. Unless appropriately sensitive organizational sttegies and channels are subsequently employed and adequate gestation time provided, the resultant beneficiary organizations will be vulnerable to distortion and will have limited effectiveness. Although the amount ofpreinvestment required will vary according to the degree of social staification and intensity of poverty in a particular situation, there are few sitations in which even a village or slum-level organization will be able to sustain, let alone expand, the benefits of a project in the long term. Because long-term links with a specific agency are rarely possible, the best way to ensure that benefits will cotinue and expand is for the local-level organization to be affiliated with similar groups in the same neighborhood so as to allow a more powerful and SophistiCated, secondary level of organization_ As with primary-levei beneficiary organizations, the sustainability of secondary structes depends basically on two factors: the degree to which the organization fulfills important priorities of its members and the extent to which the por feeltbtkitiseirorganiztion(asdistinct from that of some goverment department, local politician, or NGO). To ensure such an outcome, it is important to keep the role of local politicians or elites to a minimum, provide intensive training for beneficiaries and potential leaders (see below), promote multiple leadership (care must be taken to avoid cenalizing excessive information and power in the hands of a small emergent elite within the poor community), and maximize the level of democracy and transparency in all decisionmaking processes. Training Training has long been considered an important element of institutional development. HoweVer, training has tended to focus almost exclusively on raising the technical skills of implementing agency staff in accordance with overseas standards. In reality, the major training needs of the staff should frequently be reoriented toward human development skills but should also include a thorough grounding in the principles of sustinability itself, since this concept affects virtually every aspect of policy formion and project design and management. Finally, the training needs of the target group, which are normally given little attention or budget 2. In satdons with a udalu-tpe background, he amrgence of an orgaid poor communit with greaw bargainig pow will not be welcomed by uradhiomu clites or their polical representatives. Governm and nongoveramenT agencies that bave been given the task of communiy organizing must therfe exrcisc extune matuity and sensity and must aveid promotg any politica party. &wlainabiity of Poveny Aleviaion Pjr= 263 allocation, should be greatly expanded. With the prospec of shifting public expendiue budgets, formal extension training for huge numbes of poor on an individual basis is no longer realstic. However, the taining needs of the poor are generally very modest and can usually be imparted economically wherever beneficiary organizations have been formed. Training needs for the poor normally fall into three areas: Human development skills Include leadership training and cooperative enterprise Functional skills Basic bookkeeping. record-keeping, and management for small groups. Where literacy and neacy levels are low, functional education may also be ef&ctive as an entry point to com_mity organzig Technical skills These are general related to income- ge initiatives or preventive health care needs of individuals or small groups. Considerable debate still surrounds the question of how to provide technical skills to the poor on a wide scale. In general, individual absorptive capacities of the poor are extremely low. and large-scale schemes such as ILdia's IRDP have been crticized for assuming that millions of poor people can be turned into successful entre simply by a small injection of cash and rudimery training. Experience fm Graeen Bank in Bangladesh and the Bangadesh Rural Advancement Committee (BRAC) suggests that, particularly in mercantile-type clures, the informal market can successfilly absorb surrisingly large numbers of micro-entrepreneurs, as long as their area of enterprise is closely akin to their traditional field of activities. Where new skills must be impanted or new markets found, both investment costs and failure rates have been disappointingly high. Past experience suggests that where ining budgets are limited, it seems advisable to concentrae on supporting the institutional development and absorptive capacity of the beneficiary organizations and to facilitate the more productive use of existing skills-rather dan introduce a wide range of new skills-by ensuring access to essentid inputs, improving quality, and providing access to market inforimation systems. Many project designs allow for, and encourage, the "gradation" of beneficiaries to mainstream support agencies (for example, beneficianes who have made successful use of a project's initial credit facility are enwouraged to then move toward a more susainable credit source such as a bank). This process, however, seems practical only for those who are marginaUy poor. For the very poor, particularly those whose poverty has a strong social and financial dimension, graduation to traditional institutional links is unlikely to occur until the nmber of poor in this situation reaches a critical mass or a more compatible institution emerges to deal with such people. 3. In both Granmen But and MRAC cdk pmmms, sm loams have primrly ben used to fimce pety tg. 264 Nei Otiliwv Continuity of Liaison in poverty reduction projects, the pressure to maintain coverage scbedules often means dht the iplementing agency leavu the moniting and evaluation tcam to reew all aspects of pmformance and impacts. If this team is closely inrlocked with overall project management, ongoing lessons cn be absored and misakes corrected. Howev, the momtorig and evaluation funcion is frequently carried out belatedy or at too great a distance from project management. Over and above the formal project review process, it is incmben on project management to continualy review performance specifically from the perspective of sustinability. At the time the initial project design was finalized, sttegic plans should have been drawn up for progressively reducig depedence on extemal assistance within the lifetime of the project. Whether this plan depends on automous benefciaty oraIIzatons or on link with preexiste marke or service organzatons, it is almost axiomatic that problems wil occr and changes will be necessary. SusaNy therfore demands that project nnae m tain an ongoing system of direct consltation with the target goup (for example, by continuing throughout the project the type of consultative process descrnbed in Chapter 13) so as to be able to identify such problem and make appropriate adjustmn to thc long-term stratey for achieving susaiabiity. Where a specia project ilemenion unit has bee established to ensure efficient installaton of a proje but management responsibility is reaft handed back to de normal Ime agencis, serious problems will normally occu. To keep such problem to a minimm, a mignificant overlap perod must be allowed for, staff inteanges should take place durig the life of the project, and some reasonable incentives (for example, vehicles or equipment) should be transferred over so as to reduce the predictable resetment of line agencies that are expected to take over the responslity of what they see as more glamoromus bodies with better resources. Where an obviu increase in workload is likly to accrue to the line agenCyes), the orgial project budget shDld include a special 'wind-down" allocation to cover the cost of additional staffing and expenses for the agency during a reasonable nid on period. Althugh breaucraticaly difficult, this transition must be smooth or much of what was achieved at great -pense during the original project may be rapidly reversed. Design Factors The sustainability of povcrty reducio projects depends on a wide range of desin elemens, foremost of which is dte basic viability of the scheme. The viabilit of a project is in tum deteruned largely by ias cost efficiency, whih thus takes ito account such factors as subsidy levels, cost recovery mchnms, acmay of targetig, amd economc rationalsm. Beyond these macro i lies a network of te a fators, any one of whic, if iot corectly assessed and monlored, might well- mdeine the sustainability of the project. This category incles factors such as the initial asesment of a project's tchnical and financial viability, the adequacy of marktg anaysis and mechamsms, and protection against co-opton by eies. Subsidies are commonly provided in poverty reducinprojcts on the grounds ditb they nase the likly r nsiveess of the poor and provide an elent ofjsuse to tho who have otherwise received little benefit from the naton's overal economic prgress. By allocatig significant subsidies Co poverty secors, govmet bave hoped to stinmmlat growth and self- Sustawbiity ofr Povmrty Ailiadon Projecs 265 reliance in pnority areas. Unfortunately, where self-reliance has been a goal, the allocation of subsidies has rarely resulted in gains at the levels hoped-for, has usually caused cost efficiency to decline, and in many cases, has actually reduced project sustainabiity. Where self-reliance of the beneficiaries is not the direct goal of a project, as in primary education, health-sector, or rural works employment programs, then the above concerns do not arise and the appropriateness of the subsidy need only be justified against the basic economic rationale and sustainability of the project in contrast to other options that might be able to achieve the same goals more efficiently. Although the availability of a subsidy element frequently has (in the manner of most welfare handouts) a negative psychological effect on the recipient's sense of self-reliance, by far the more serious problem arises from the market distortions and fase messages that frequently accompany subsidies.4 A further problem arises in more feudal situations where widespread subsidies attract the attention of those more powerful elements in society that believe they have an autmatic right to control or share in such government largesse.' From a design perspective, the allocation of subsidies is frequently a shortcut method of attracting the participation of the poor, participation that would otherwise require a substantial investment of time and energy in order to persade the target group of the benefits of the scheme being proposed. Needless to say, participation that is "purchased' in this manner is unlikely to lead to sustainable project outcomes. While the allocation of subsidies is a simple and politically popular strategy, the same funds diverted to social preparation and institution building among the poor would invariably lead to a more productive and sustainable outcome. In rural credit programs, for example, interest subsidies are frequently given to small farmers with a wide range of negative consequences. Were market interest rates to be charged instead, the previous *subsidy" could be treated as a "sunk cost" and applied on a once-off basis to the formation and iraning of small savings and credit groups-an approach now proven to be highly sustainable- Cast Recovery Cost recovery has a profound effect on project sustanability. The level of its success will be deternined both by the appropriateness of the inital technical design and by its perceived value to the consumer and the mechanisms used to actually recover costs. Deeisions regarding the type and extent of cost recovery mechanisms to be used should be made at the initial design stage. The predetermined subsidy levels and fee charges should together cover the maintenance costs of a particular sevice. With regard to essential services for the poor, the proportion of the subsidy is normally based on some notion of the target group's "capacity to pay," a factor that will have been gauged during the rapid social assessment phase. Where long-term recurrent costs are involved, the sustainability of project benefits is normally contingent on a reasonable proportion of cost recovery being achieved. Where cost recovery 4. The destructive impact of cheap r crEdit has been wel documented. in the Indian IRDP, where the levels of subsidy varied accding to the type of entepnsc bemg undeuk subsequn studies (Puley 1989) found no posiive correlation beween the level of subsidy and the profbility of the enteprise undertaken 5. Food-for-work programs are notorious for the corruption of ther supevis and conurcos; howevr, mor seious problems occr m subsidized IRDP sdcemes swc as livestock projects, where benefiaries are commonly required to provide illegal inducements to back officials, veterinary offices, and odhers for what are fiequendy unhealthy or substandard animas icpable of providing a viable retun on the funds bonowed. 266 Nei O'Sutlivaw mechanis_m are based on a clearly established demand of the target group, realistic expectations will normally be met as long as continuity of the desired service is naintained (one of the main reasons that loans in micro-credit programs are not repaid is the uncertain availability of repeat loans). Where community or beneficiary organizations have been estlished during, or prior to, project implementaton, such organizations normally have the potential to play a key role in ensuring cost recovery, as has been demonstrated in water-user type groups in many parts of Asia. Targeting Although project planne are often pressed to economize in the development of finely tned and exclusive targetng criteria there are certain situations in which such cntena can direcy assist in ensurng the susainabiliy of overall project benefits. This is the case when targetig acts to reinforce the emergence of homogeneous beneficiary groups and to strengthen their confidence in dealing with more advanced communities. In the successful micro-credit progams in rural Bangladesh, for example, carefil targeting has excluded all but the landless and assedless and thus allowed previously unthinkable energy and initative to emerge widtin borrowers' groups. Economnic Realsm Given the enonrous political pressure on governments to tackle poverty rapidly and on a gand scale, it is hardly sunprising that projects are sometmes supported that, viewed from hindsight, were not economically rational.6 With the passage of time, such projects lose mo_mtm and are ultimately abandoned or replaced with some alterative scheme. Apart from their basic political imperative, many such projects are unfortunately based on the view that the poor are incapable of helping themselves and therefore require wefre support, albeit in the guise of income-generating projects or similar services from government. As discussed in Chapter 13, well-designed projects will seek to take maximm advantage of market forces and of the private sector and wil thereby increase resoLrce-use efficiency and reduce the long-term dependence on govemment support. Such a scenario assumes, however, that the poor are able to compete in the marletplace on reasonably equitable terms, a situation that is seldom the case. Where such a potential problem has been confirmed during the design phase, special attention will need to be given to instutil st ng among the poor if direct market or similar private sector links are envisaged. Women The performance of women in poverty-focused, crdit programs is now widely recognized as superor to that of men in trms of carefilnes and reliability. Similarly, in water, health, and education projects, women, when given genmune encouragement to participate, have regularly 6. Throug dhe use of a genmus loan-plus-gtat pakage, Sri Ianks ambitious Janaya Progam eancuged widepead handirat producto as a means of reducig nual mnmplcymeL Such an artficial stimulus was not, bowver, based on market surveys or demand. with te t that the govermem was forced to estbLish speca prhasiWg wadis m Orter to prvete impoveriShed producas fm becoming dillusioned. Snutainability of Povely Aileviadon Frojectrs 267 shown themselves to be enthusiastic and dependable. Whre project strategy involves the formation of beneficiary organizations, special care should be taken to include women so that the resultant organizations will have a more practical and consistent focus and dtus greater dcance of achieving sustainability. The paths to this goal were discussed more fully in Chapterl4 Although the approach depends in large part on the particular cultural settng, project management should employ a maximum number of women on the field staff and to ensure that training programs are scheduled at times and locations convenient to the worldng patterns of local women. Envronmental Factors In the long term, the best method of sustaining the natural resource base of the poor is to raise the level of consciousness about the environment and to simultaneously provide viable income-generating alternatives. in the short term, clear policy decisions are somedmes necessary at the initial design phase so that the poor will perceive clear-cut advantages in their behaving in an environmentally sound manner. For example, in areas of deforestation, project design should attempt to provide incentives that will reinforce the idea that forest protection is a more atractive proposition than clearing forest or selling firewood. If the project design has attended to the above factors, then routine monitoring and evaluation should alert the miplementing agency to any unexpected environmental destruction. Since it is rarely feasible for iplemeing agencies to have an effective communication with individual beneficiarics on issues such as environmental protection, the only realistic channel for raising awareness of such issues is, once again, the beneficiary organizations. If such organizations have been genuinely set up with a broad development mandate and not just as a channel for project inputs, it is likely that these organizations will need little encouragement to be concerned about environmental damage. In such cases, properly organized communities will automatically seek to apply whatever solutions are within their power. Ap,propriate Technologies Provided the design phase was conducted with sufficient care, technical design options will have been chosen that are as close as possible to those that the target group is already familiar with and that maximize the affordability of the particular service. Where significant changes are necessary, however, or where absorptive capacity is particularly weak, both additional training and time will be necessary, as will more intensive monitoring, if the chances of sustainabilty are to be maximized. Implementation One of the main tasks during implementtion is to ensure that the beneficiary organizations have the capacity to continue the project theselves. Therefore it is important to devise a realistic, post-project design in conjunction with beneficiary groups and to see that the final nhase of the implementing agency's work is concemed with establishing the training, systen -.. dnd links needed to ensure the success of that design. During the implementation of poverty reducion projects, certain problems are likely to arise simply because certain actions and reactions are imnariably stimulated during a successful empowering of the poor. Freedom and access to resources, for example, are bound to be 268 Neil O'Stdlimn exploited by a minority of the target group themselves. At the same time, powerful outsiders will frequently seek to distort the project to their own benefit or, if that is not possible, to sabotage it. For the sake of the project's repuitation and continuity, such incidents must be anticipated and prevented wherever possible by continuing social analysis and more intensive social preparation where necessary. Recognizing, however, that even with the most carefil of preparations, human problems will continue to cause project disruptions, all project staff should be encouraged to be alert to such dangers and, wherever their presence is suaspected, to call for an intensified monitoring surveillance.? At a broader level, project management must recognize that at times their project wili come under close scrutiny by potentially powerful adversaries and that it will be necessary (particularly in the early stages where benefits camot readily be demGontrated) to maintain a strong public relations effort. Where the basic strategy of the poverty reduction project has been to provide employment opportunities or social services, f will not be possible to hand over responsibility in the same manner as in self-employment type projects. Even here, however, the same princtple applies. Tnat is, beneficiary groups must be involved, prior to the transition, in devising strategies that will ensure the continuation of benefits and thus ensure their ongoing participation. In the case of rural-works employment projec, there should from the outset be a commitment to helping the rget group establish self-reliant labor organizations that will eventually be capable of ixkpcndenty bidding for government road contracts and ocher jobs. MonkiOring and Evaluation As implied earlier, the difference between poverty reduction and nonmal development projects is that the former rqe more intensive monitoring and evaluation. The main reasor for this is that the desired outcomes of poverty reduction projects are both physical and social. Social outcomes, apart from being notoriously difficult to measure, are also highly susceptible to exteral factors and backlashes that are likely to be generated simply by pursuing the project's goals. Poverty reduction projects are thus highly fluid, dynamic, and unpredictable and, unless close and consistent attention is given to all aspects of their design and lementation, rapid and -someties irreversible damage can result. (It should also he remembered that a well-intentioned project can sometimes make the poor worse off, and in such cases ft will be doubly difficult for any future project to regain the confidenc of the poor, no matter how good or appropriate it might be.) f monitoring and evaluation are to assist in ensuring a project's sustainability, these activities should not odly be carried out intensively, but the resuts produced should contimnously be integrated into project design. In this way, the uncertn path that a project may take can be cont y reassessed and prwmpt, corrective action tken when necessary. In general, the most effective means of integrating de monitorng and evaluation system into the project's man 7. The Indin hegscd Rur Dvelopm Progam. for exampk, mnt - insve and independ concurre ahtion of all aspecs of the proe's prfonumnce and inmp Where specifc pblems arc noed-fo exmpl, if the offic puras cmmlee is foud ecu ng unofficl es and so jeopariLzig t viability of a beneficiay's mvestue-ie projt managemn progrealy inaes alnrsive anmngemet in this case, te bcneficiaries received a direct cah payme so dt Ihey coud makme eir own purchases. This change has, of course. intoduced a new se otf oFtenl prblems that mist also be specfically monitord. --~~ ~ ~~~~~ ~ ~~~~~~~~~~~~ . :.- Sustainabitity of Povety Alleaion ProjecU 269 system is to follow a participatory approach,as suggested for the initial design. Where both primaxy and secondary beneficiary groups have been formed, these would become an increasingly important source of comaunication and advice to project management and would nonnally be the focal point for most monitoring and evaluation. In cases where the above type of strong benefciary groups have not emerged, it is important to devise an alternative arrangement that will provide long-term feedback to the concerned agency. In the past, monitoring arangements were usually terminated immediately after the project was implemented. When dealing with the poor, however, it is essential to establish a monitoring system that continues to provide feedback throughout the operating life of the project. Part 5 SUMMAARY AND CONCLUSIONS y -~~- X 1 - ,;f;.- --.. 16 SECTORAL ExPERIENCES IN POVERTY ALLEVIATION Michael Bamberger This chater swnmarizes the findings and condusions from the 1991 Bangalore Sminar wih respect to sectorae expenences wiah poverty alkiaton programs d projeas in South Asia. It draws on the sector review papers presented in Chapters 3-7 and 10, the presenations by seminar pancpants, the senar discussions, and the lessons drawn from project vsits (presented in ChWter 8). of the Poor to Pop-ladion, HeaWJh and Nutrition Services (See Chapter 3 by J. K. Until the 1970s, it was pdssible produce sianifi=ct and even dramatc gain in hezl!h through the ccotrol of malaria and of epidemics as well as the eradiation of smallpox. The impact of these i n on and related programs diverted attention from the disapponting performance of health services in reaching the masses of the poor and satisf ir basic health needs. Consequent to the Alma Ata declaration (1978) of "health for all" and the adoption of the primary health care approach the South Asiaircounties rapidly expanded perpheral popuation, heath, and nutrition service (PS) deliveiy systems by adding health centes and subcenters or posts and establishing community-based health swevices thogh community health workers. The expansion of these services has reduced ifant and cbild mortality and incsed contaceptive use. Despite the massive expansion of PS fcilities and staff, policy reforms, and program improvements, the health status of the poor has improved to a lesser degree than that of the pation as a whole. All efforts to improve accessibility to health services and to improve the health status of the poor are ficed with three intTer ed problems: resource constraints, inadequate service delivery systems, and limited demand for many available services. Resource constraint Not only are South Asian counties among the poorest in the world, but most of them have rapidly growig and widely dispersed and inaccessible populations. Even with the most cost- effective systems for delivering health smvices, available resources are extrmly limited and must be continuously protected from the demands of otber equally important development sectors such as agriculture, in e, and tansport and industial development. If health status is to continue to improve, PS coverage of the poor needs to be substantialy inreased. Distance, the nrde behavior of sevice providers, the lack of medicines, and perceived low efficacy of servces al deter the poor. Existing health behavior and practices 273 274 MfidaeL Bamberger at-! the low priority assigned to these services by the poor reduce their demand. Significant gender diferences also persist. International experences suggest that in order to reach the poor, PS delivery systems should provide effective service delivery at the village level with back-up support from higher levels, rationalize responsibdilities for various services, enhance the people orientation of service providers, emphasize demand creation, and build strong lik to the community. PS delivery systems in South Asia are attempting to approximate this ideal, but there are serious shortfalls. The service delivery structure is still weak at the most peripheral part, the village level. Although there has been progtess toward integrating services, organizational arrangements, particularly at the village level, still need rationalization. Lack of Atteion to Dmand Geneation Although service delivery has been rapidly expanding, realistic programming of work, in-service training, coverage-riented monitoring systems, and supportive proble-solving supervision have been neglected and now require urgent attentioL It was assumed that once services were provided, people would use them. Consequently, little emphasis was placed on generating demand and the quality of cocmmcation activities was poor. Village Health Worker (VHW) programs have generaly been inadequate and unsuccessful, and links to the commumity remain weak Forming women's groups with commty volunteers provides the best opportmity to strengdten commity uinks. Health services for urban slums have been neglected, and apropriate service delivery stucrs need to be established. Demand generation is also constrained by social and cultural varables. In many parts of the subcontinent, families are less willing to itvest in ensuring that women and girls are able to benefit from available health and education services. In some areas female infanticide (either direct or through neglect) is still common. Religious and caste bariers also continue to bar equal access to health services. During the visit to the Tamil Nadu Integrated Nutrition Project, seminar participants were able to observe the problems of ensuring that nutritional supplements are accessible to low-caste families. The Contibuton of Nongoverment Orgaaizadons Nogoverne orgtons (NGOs) have generally been more successful in reaching the poor than govermment agencies. The quantitative impact of NGOs in the health field may still be quite limited (although growing) in India, but it has become very importan in Bangladesh. The NGOs have succeeded by being responsive to community needs, establishing paramedic- based service delivery at the village level, and allowing beneficiaries to participate in decisionmaking. But many of them remain small and continue to rely to a large extent on outside funding. Their success in creating self-reliant commnities is also liited. NGOs are active in assisting goverments in upscaling -their innovative and successful programs by training and providing other iplementation support as well as assistig communities in using the available government services. Although mrany would agree that NGO activities have been more effective than government efforts i many healdth-related areas, the relative cost-effectvess of NGOs remains a hotly debated ise. Many observers dispute the -"1im of NGOs that their delivery systems are more cost-effctive than those of govement and _-.ntend that NGOa have disguised costs such as unpaid staff and volunteers provided by outside agencies. The issue of co-effectiveness wfll Secboral Eperfc in Pover Allevation 275 become icreasingly important as interest grows in ways of scaling up some of the more successful NGO experiences. All of the countries in South Asia are trying to establish mechanisms for collaborating with NGOs, for example, by forming coordination comnmitees, delineating area responsibilities, and sharing tasks (NGOs create demand and the government provides the services). Much more learning is necessary, however,to arrive at patterns of successful partnership. Factors Affecting the Susatainbity of Health Delivery Systems Sustainable PS delivery systems to meet the needs of the poor have been difficult to build for a number of reasons. First, it takes a continuing political commitment to operate large-scale PS delivery systems for the poor. Second, it is not only "what" is done but "how it is done" that determines success; therefore, implementatin requires close attention. Third, PS delivery systems have been underfunded. Fourth, the management capacity within the ministries needs to be strengthened. Because of the chraceistics and low level of technology of their clients, the ministries of health must meet certain requirements in providing PS to the poor. In particular, they must endeavor to promote decentralizaion, build strategic planning capacity, pay more attention to hun resources develoment, and try to provide beter technical su-ppor to the periphery. As coverage levels improve, targeted strategies are becoming necessary to supplement PS delivery system. Although inreasing attention is being given to the above-mentioned issues, it is still uncertain whether goverment PS delivery systms would have the necessary adaptability to adequately cover the poor. Depending on the spatial distbution of the poor and characteristics of the PS delivery system, geographic targeting, promoting at-risk approaches, widening choices for the poor, and multisectoral approaches may have to be followed. The role of the private sector as a provider of health services to higher-income groups needs to be expanded. In conclusion, four actions are necessary to build sustinable PS delivery systems and implement dtageted strategies to reach the poor. First, the information base about the poor needs to be strengthened. Second, policies need to focus on the poor. Third, programs should stiengthen management's capacity for desiging and implementng PS delivery systems. Fourth, we need to learn more about specific constraints and how to overcome them through systematic research efforts at local and regional levels to help refine responses to the problems of providing PS to the poor. Poverty Alleviation Experience in Rural Development (See Chaper 4 by Dieter Bucher and Keith Johnson) Approaches to rural development have evolved in response to lessons leamed as well as changing economic and tchnologIical conditions. In the 1950s and 1960s the emphasis was on the commuity development approach and on the s of commuty o ions. This approach came under inreasing criticism because of its lmited impact on the economic conditions of local commumities. Many programs also fiiled either because of opposition from local elites who saw the programs as a tbreat or becaus these elites were able to siphon off most of the benefits. Following the advent of the "Green Revolution' in the late 1960s, attention over the next decad turned to increasig production and the consequent need to focus on broader areas in developmet projects. This lead to a vertically integrated apprach to devlopmt and the belief 276 Mic6ae Ramberger that the poor would benefit from increased productivity and the ensuing multiplier effects throughout the rural population. Aklhough the "Green Revolution" had a significant impact on overall agricultural productivity, it had becam clear that the poor did not automatically benefit and that in many cases ther might even be worse off. The nonagicultural rural population also began to multiply at a fast pace. Following Robert McNamaras Nairobi speech in 1972, poverty alleviation became a cental theme of rural development, and increasing attention was paid to ensung that development projects specifically addressed the needs of the poor. This led many to recognize there was a need for inegrated rural development projects that had an area focus and included nterined productive activities, comprehensive social components, and a comktmen to stening rural infrastructue and developng local inttutions. Integrated mral development programs and projects condnue to use this framework and to emphasize the integrated approach. Initial evaluations of the Asian Development Bank's (ADB) Rural Development projects suggested that their impacts and benefits were less than expected owing to a comination of inadquate project preparation, inappropriatechnologies, and inapprOiate or ns. However, a more recent evaluation by ADB's Post Evaluadon Office covering projects completed between 1985 and 1990 was more positive and indicated that potental positive outcomes have often been underesmated because of difficulties in obtain adequate data, and the negatve impacts of extnous factors such as weather and government policies, which disiminate st agriculture. The World Bank (World Bank 1988c) concrs with many of these condusions and stress that millions of rural people have benefited from rural infrastructure investments, food Production, and support to subsitece farmers. The asssment recognzes, however. that the sutegy was to assist smallholders and those with productive assets rather than to reach the "poorest of the poor." The assessment by the Bangalore seminar participant was frequently more pessimistic than the donor evaluations. In the case of India, the 1989 Conurrent Evaluation estmated that only 17.63 percent of poor fEnilies (below an income level defined in terms of calorie consumpion) who had been assisted had been able to cross the poverty line (Department of Rmal Development 1989). The ointing results could be due to the unproductive nature of the assets provided, to the fact that the incrmental income from the assets was not sufficient to help families cross the poverty line, or to the fact that assets were given to families with incomes above the cut-off line (leakages). The results of rural development (iRD) programs in Pakistan have also been disappointing. Village Aid, Rural Works Programme, People's WoLk Program, and IRDP all failed to have a sustainable impact on the social and economic conditions of the poor. Some of the problems here were that these programs were unable to develop local oranons that owld sustain them, they were poorly managed and implemented, and there was strong political interference. Although government-sponsored and ad_mnstered credit programs for the rural poor clearly fadiled to meet thei goals in many cases (see Chapter 6), there are numerous examples of NGO or public programs with a strong commuity participation that have been successfull. The Small Farmers Development programs in Nepal and Bangladesh, and BRAC and Grameen Bank in Bangladesh (see chapter 6) are a few examples. Seminar partiipants frequently cited the cornuption and the influence of local elites as one of the reasos that programs do not benefit the mral poor. Ehtes either sabotage programs (partcularly programs tha stcg"- &fnnrv or community o or c up Seaoral Experiencs in Povery Aleiaion 277 indepenent purchasing and marketing chanels) or that succeed m appropriating a significant proportion of the resources. Corruption, often referred to as "grease" or "leakages" forces beneficiaries to make payments to intermediaries, thus increasing the real cost of borrowing. Lessons Learned 1. Rual Development projects require a more comprehensive identification and appraisal process with the systematic involvement of -local organizations and a thorough analysis and understanding of local conomic, social, and political conditions. Full advantage should be taken of NIGOs' in-depth knowledge of local conditions. 2. Greater attention needs to be given to the collection, analysis, and utilization of data during the project preparation, monitoring, and evaluation phases. 3. Many projects fail because of poor project management. Some of the problems at the project level include inexperienced management, frequent transfers, lack of incenives to work in mral developmnent, and lack of coordination with local organizadons. 4. At the program level, a major stumbling block has been the difficulty of coordinatng and integring different project components, and the need to tailor overall program planning and coordination to local conditions and the priorities of intended beneficiaries. A frequent problem is that targets are centrally defined and thus encourage extension workers to pressure beneficiaries to accept loans or assets they do not wish to have. A related problem is market mechanisms are often ignored, with the result that farmers may be encouraged to produce unprofitable crps or that economically nonviable off-fkm activities are promoted. 5. There is a need to get away from top-down "blue-print" approaches to a more participatory and flexible planning approach in which project are allowed to evolve gradualy and to be modified in the light of experience and changing circmtances. 6. Projects should be simplified by reducing the number and complexity of components and breaking them down into phases. 7. The provision of technical assistance reqires more carefiul planning to ensure that it is available when reqired and that it is appropriate. 8. Project financing needs to be more flexible and innovative. The difficulties in mobilizing counterpart funds also need to be addressed. This can be done by stening cost recovery and asking farmers and other beneficiaries to pay the economic cost of productive assets such as irrigated water, fertilizers, and equipment. 9. Greater attention must be given to institutional development through staff development, decentalizaon, and a broader role for NGOs and communityions. 10. Projects and programs can only have a sustinable impact on poverty alleviation if they operate within a conducive policy vronment. To build such an environment, it will be necessary to eliminaee andrral policy biases, focus specifically on poverty in rural development 278 MAidid BwBbrger and agriculural policies, target services and benefits for poor and vulnerable groups, recognze that rural development is a process requirig a long-term commitment, and put greater emphasis on project and program sasainabiity. Access of the Poor to Housing and Basic Urban Services (See Chapter 5 by Yue-Man Yeung) In recent decades the uxban population has grown at approximately twice the rate of the total population. A number of approache have been adopted to improve the provision of services to the urban poor and alleviate urban poverty. Panidpotory urban serices have been adopted in the Republic of Korea, Indonesia, the Philippines, Hong Kong, and Malaysia whereby the urban poor organized themselves to improve their urban services. Some of these progrms were quite effective in upgrading services in a more cost-effective and timely manner than through government service delivery programs. Poveny-addressing programs rely on an itegrated approach in which national or stae governments together with municipal authorities provide a wide range of services to low-income communities. Programs of this kid have been tried in Korea, Metro Manila, and Malaysia. None of these programs has made a major impact on poverty in part because of a lack of fimding, but also because of the difficulty of coordinating many different agencies and of motivating low- income households to participate in the organizaionally complec programs. Slun upgradlingprograms are seen as a complement to the more expensive and complex provisionof low-cost housing throughsites and serWces projects. Slum upgrading programs have made a significant quanive impact in Idia, the Philippmes, and Indonesia. In a number of other countries-Thailand is one-the impact has been reduced by adminisraive (land acquisition) and orgnizational (involving intended beneficiaries) problems. Wnere the community organiztion is well prepared, coordination between govenment agencies is ensured and there is a long-tenm goverment commitment, slum improvement programs can be a cost-effective way to improve the physical conditions of the poor. Integrated approaches have been developed in Malaysia and Korea. Here, agencies cooperate in providing a wide range of infrastructure and social services. So far, the results have been promismg, but these approaches have only been tied on a small scale. It is unclear whether resources would be availUable for larger-scale replication or whe -ir the same level of interagency coordination could be maintained on a larger scale. Govevnmet-prodded comnwity devdopment programs have been tried out in Sri Lanka and in India. The Sri Lankan experience has been quite successfil, although heavily dependent ,n govermnent support to promote and mainain community participation. In Hyderabad the Urban Commuty Development Program (UCD) has been effective in inmroving coordiaton between municipal agencies in the delivery of services, in mobilizing commuty counterpart fuids, and in s t community organions (Shah 1990). The Urban Basic Services Program (UBS) begun in India in 1986, which later became the Urban Basic Services for the Poor (UBSP) is an example of this approach. The main objective here was to develop communty stucus for elicitg "felt needs" of the slum dwellers at the neighborhood level and ensure their participationi m developmental programs such as low-cost water supply and sanitation, immuization, child-care, preschool education, and women's development activities. X 4 Seconal Eperences in Povelny Aeviaion 279 NGOs have initated a wide -nge of aided self-help approaches in which communties are provided with technical, organizational, and other resources in order to be able to help themselves. Numerous examples can be found in South Asia, including an integrated urban development project initiated by the Alnelabad Social Action Group (ASAG) for 3,0M0 households affected by flooding and a project rehabilitation organized by the Delhi Catholic Archdiocese providing aid to flood victims in Jahangirpuri. A notable characteristic of most housing and urban developmeat strategies is that the emphasis is almost exclusively on the provision of services, and vely little ateion is given to issues relating to poverty aeviaion.In contrast, ural development progams have made poverty alleviation one of their prime objectives. For example, the main indicator used to evaluate the Indian IRDP program is the number of families who have been helped to cross the poverty line. Urban poverty and urban poverty alleviation have, until recendy, received very litde systematic attention in most South Asian countries. In the case of India, for example, while issues relating to rural poverty were explicitly addressed at least from the time of the Third Firve-Year Plan (1961-66), it was not until the Seventh Plan (1985-90) that the problems of urban poverty were directly addressed. Lessons Leamed I. It has proved vitually impossible to provide the - s of urban poor with adequate basic services through a 'top-down" approach. Government deliver; rsystems are too expensive and often have difficulties in reaching the poor. 2. Services can be delivered most effectively through decenralized approaches in which both administrative and financial power is delegated to local agencies. This approach can only be effective if resources are invested in traing local authorities and strening their institions. Decentralization should be complemented by paicpatory approaches that include beneficiaries in the planning and implementation of projects. 3. NGOs also have a mnnber of important roles to play: (a) they can act as an intermediay between the community and the govenmnent, (b) they can help com mnities strengthen their capacity for project planning and implentation, and (c) they can work directly as a project executing agency. Certain NGOs can also make an important contribution in research, evaluation, and policy formulation-although these functions are generally less accepted by goverment and donors. 4. Careful attention must be given to the targefing of programs and services to ensure they reach the poor. Because urban services are scarce, higher-income groups may use their influence to divert reources intended for the poor, or they may buy out the poor. The poor are also difficult to identify and target because a high proportion do not live in easily identfiable slums or low-income housing areas but are scattred thrughout the city. 5. Urban planners must undetand the important contribution of the informal sector to the provision of housing and urban services for the poor. Even though many of the informal enterprises providing these services are illegal or quasi-legal, they play an important role and their existnce must be recognwed and factored into urban developmet planng. 280 Midkad Bamberger 6. It is also important to understand the multiple functions of housing. For a poor family, the house is a consumption good, an investment, and a place of work. The design of housing projects needs to recognize these multiple functions. 7. Mechanisms need to be developed for achieving greater and more effective itgion of service delivery. 8. The mechanisms through which technical assistance is provided need to be improved. Although outside expertise is often required, it should be given in a participatory manner with both government agencies and community organizations directly involved. Care should also be taken to ensure that assistance is available when required. In many cases assistance is only available during the project design stage. 9. Security of tenure is a crucial element of any participatory urban development approach. This is important both so that residents will be willing to invest their own resources and also to eliminate any administrative barriers to the provision of public services. 10. A more pragmatic and realistic approach to cost recovery is required. Despite the growing evidence that even very poor households are prepared to pay for valued serices (and that in fact they are already paying for many services through the informal market), many ideological, political, and administrative barriers still stand in the way of effective cost recovery - and hence make it difficult to entend services to a broader range of households. Cost recovery should be accompanied by affordable services and standards that are not unrealstically high. 11. The role and administration of housing credit must be better understood. Many low- interest credit schemes are underutilized because of a low-ceiling on loan size or cumbersome administrative procedures. Flexible approaches to loan ceilings are required in recognition of the changing economicucircumstances of the household. Flexible repayment schedules may also be required. 12. The role of women in the urban housing market must be better understood. One of the consequences of urban migration is a greater degree of economic and social independence of women. This means that women may become house owners (or co-owners) and that they may be actively imvolved in the financing and construction of houses. Women are major users of urban services such as water and tranwort, and women's needs and preferences should be fully taken into consideration in the planning of these services. 13. Much greater attention must be paid to the suainability of housing and urban service programs. In many case authorities fail to adequately provide for the financing and administration of operations and maintenance once houses are occupied. As a consequence, roads, drainage, and other public services deteriorate rapidly. The Role of Credit in Poverty Alleviadon (See Chapter 6 by Ismael Getubig) It is widely recognized that many policies seeking to improve the economic conditions of the poor, either through the promotion of economic growth or through direct employment creation, have failed to reach or to benefit the majority of the rural and urban poor. Pardy as a . - I . , Sectoral Ereiiences in Poverty AlieWaiion 281 consequence, many countries have sought to use credit as a means of helping the poor generate their own economically productive activities. With few exceptions, the poor have no access to normal commercial credit. Consequently, governments and NGOs have developed special credit programs for the poor, usually with below-market interest rates. Most government-sponsored credit programs have failed to reach the poor or to develop economically viable programs with a significant and sustainable impact on the poor. This has been true for programs directly managed by the state (such as the integrated Rural Development Program in India) and programs in which commercial banks arc encouraged or required to provide low-interest loans (such as the Differential Interest Rate program in hdia). A number of fctors have contributed to the low success rate of these programs. First, the targeting procedures have frequently been poorly applied, so that a high proportion of the low-interest loans have gone to local elites. Second, poor planning and a lack of technical support have resulted in the approval of many loans that were no economically viable or loans that were too small to produce any significant impact on the economic situation of the beneficiary. Third, the asset retention raie has been low- both because of a lack of supervision (West Bengal farmers sell their buffalo to Bangladesh, for exawmple) and because the loan did not generate sufficient income to permit the asset to be profitable and hence retained. Fourth, the limited avrailability of second loans has meant that even when benelfits and impacts are produced they cannot be sustained. Finally, owing to the lack of motivation, poor administration, and political pressures, the cost-recovery rates of most programs have been extremely low. In view of all these problems, most conmnercial banks have been extremely reluctant to participate in poverty aleviation projects. Their transaction costs are high and the small size of the loans offers little potential profitL Frequently, they only participate because they are required to do so by the govement. Successfil Experiences Despite these many problems, examples can be found of highly successful credit programs for the poor. The success of a program should be defined in terms of (1) the extent to which it has reached the truly poor, (2) its loan recovery, (3) productivity of the loan, (4) impact on borrowers' income, (5) sustainability, and (6) noneconomic criteria such as poLitical empowerment. Programs can be classified into four main types. * Programs that act as an intermediay between commercial banls and low-income borrowers. The working Women's Forum in Madras is an example. * Specalized credir programs of formal-sector banks. The BKK program in Indonesia provides one of the few successful examples of this approach. * Specialized bans for dhe poor have been established to work exclusively with the poor. The Grameen Bank in Bangladesh is one of the best-known and most successful examples- * Package programs offer credit as one of a number of services. The integrated development programs represent the largest, but a mainly unsuccessfu example of this approach. 282 Midwel Bamberger Factors Contrbung to the Success of Creit Programs for the Poor Getubig's analysis demonstates thatdifferent programs have achieved success in different ways. Some programs achieve high cost recovery by using group pressure and the promise of a second loan, but do not require collateral (Working Women's Forun and Grameen Bank). Others, such as DICK, require collateral and use hiis, along with efficient administration, to ensure a good recovery rate. Some programs use credit as an entry point for a broad political and social program (Working Women's Forum) whereas otners focus exclusively on credit (Grameen Bank and BK. One way to reduce costs is to work through community-based groups (Grameen Bankc), but it is also possible to integrate the special program for the poor into the regular banking system and thus avoid the additional costs of special branches and services. The review of successful and unsuccessful experiences suggests that the success of a credit program for the poor depends on several factors: * Programs must offer repeat loans in order to ensure a sustainable impact. The possibility of a repea' loan motivates beneficiaries to repay the first loan. * Programs are most successful when loam are used for economic ventures selected by the borrowers. Success rates are even higher when fellow borrowers have some say in the approval of the project. Programs are least successful when projects are selected by the government and imposed on borrowers. * A high cost-recovery rate is an essential component of a successful program, both becase it recycles fimds into new loans and because it obliges borrowers to take more seriously the selection of projects. * Targeig mechanisms are essentia to ensure that loans go to the poor. This can be achieved through strict criteria for participation in solidarity groups through whom loans are approved (for example, a ceilig can be placed on the amount of land that can be owned), by targeting programs toward generally poor groups (such as self-employed women), or by making the loans small enough to be unauractive to higher-income groups. Gendr lssues m Antipovert Progams (See Chapter 7 by Noeleen Heyzer) It is freqwey assumed that the interests of poor women will automatically be taken care of in rural development, primary health, low-cost housing, and other programs intended for the poor. However, women often benefit much less from poverty programs, wuless they are specifically designed to take into account the special needs and socioecononic cuaacterstics of women. As Heyzer says: There is increasimg evidence that women do not automatically benefit from antipoverty programs, and that many growth strategies may make the conditions of significant numbers of women in poverty groups worse. Secoral Expernces in Povewy A aliion 283 A sexual division of labor exists that allocates to women the most tedious and labor-intensive work and limits women's access to and control over developmnmt resources. The generation of employment tends to build on this existng division of labor and even intensify it. As Kabeer (1991:1) points out with reference to Bangladesh, poverty- oriented programs operate within a context of male dominance and female exploitation that is backed by strong religious, social, and legal pressures: Throughout their lives they have been brain-washed into believing that their pr-ordained role is to serve and please men. If they fail, they can be discarded as they no longer have any value. This social order is perceived as a 'natumal order-women have nothing to remind them that they are bom equal with men and that given the will they have the power to change the social order and take control of their own lives. There are a number of ways in which women are affected by their unequal position in South Asian societies. There is often an unequal sharing of food, as a result of which women and girls suffer significany higher levels of malnutrition and infant mortality than males. Women are also exposed to special health hazards, including more limited access to health clinics (even when services are free), exposure to hazardous chemicals (particularly fertilizers and pesticides), and physical injury from carrying heavy burdens such as water and firewood. The hazards have a cumulative effect on life expectancy, particularly of young girls. Women also have signficantly different employment pattrns and are often cncentrated in the informal sector, where they enjoy much less protection from labor legislation. Also the kinds of assistance women need to increase their economic productivity are often different from those required by men. Shelter, Fuel, and Secwrty Even though shelter and fuel are normally considered items of joint consumption, women still experience frequent discrimination. As fuel becomes increasingly scarce, women tend to use less of it for cooking their own food (which will often be consumed half-cold). Even though a high percentage of low-income urban households have a female head, women frequently find it more difficult than m-n to acquire housing, either for legal reasons or because they have more difficulty raising the necessary finances. Personal security and violence against women remain almost undocumented, even though these are major concers in any women's discssion group in Asia. Fenale-Headed Households and Poverty Female-headed households are significanly more poverty-prone than male-headed households, partly becas they usually do not have a male worker. As a result, a woman who is also household head is like',* to be faced with even more problems than a male counterpart is ensurig the basic survival and welfare of the fsmily for which she is responsible. A woman in such circumstances usually has to work ectremely long hours in addition to carrying out heavy household chores. 284 Midcad Bambffger Plnners Underesimae the Economic Contribution of Women Governmes continue to plan as if men provide the only support A,. poor families, when in fact women make a major economic contnbution to the survival and welfare of poor households. Migrant households rely on female earnigs even more. The emphasis on the household (rather than household members) as the basic economic unit is one of the reasons that the sexual division of labor, along with power relations within the household, and women's interests, have received so little attenon. Another consequence of misunderstanding the sexual division of labor is that women's work and its economic contribution have been greatly underestimated. As a result, women's role in the development process is greatly undervalued, both socially and politically. Gender Issues in Agricultural Development Although some studies have shown the negative impacts of the increased agnculural productivity on the poor (the best-known example being the 'Green Revolution"), less attention has been given to the impacts of agricultural modernization on women (see Heyzer 1987). For example, the introduction of motorized rice hullers and modem rice mills have increased agricultural productivity, but the benefits have mainly gone to landowners and other higher- income groups that were able to purchase the machines. In this, as in other cases, women have been left with increasgly labor-intensive production tasks and have been left worse off rather than better off. Agriltural planners have fiequently failed to understand how women's triple roles as producer, child-bearer, and hom-maker affect their ability to benefit from new technologies. At the same time, women's major role as agricultural producers is frquentIy not recognized, and consequently programs are not designed to respond to their special needs and potentials. Bamiers to Women in Agnrcdtlral Prodcion A number of factors mitigate against policies and programs that would increase women's role in agricultural production. To begin with, the problem is poorly understood (for example,it is assumed that women play a limited role in agricultual production or that they benefit equally with men from tochnical assistance and agicultural inputs provided to the household). Second, many of the required measures would challenge the accepted male role as household head or would pose a threat to local elites through political empowerment. Gender Issues in Urban Development and in Urban Labor Markets Female mral-to-urban migration contiues to increase steadily, and in some Asian countries (manly in South Asia) occurs at a hiher rate than for men. This is fiuther increased by itaional migration, particularly among young women leaving their countries to work in service industries in the Persian Gulf or Europe. - Some labor market economists argue that migration is the result of a freely made and economicaly rational decision to move to areas of higher economic opportuniy This is an oversimplification of the situation. in many cases young women are pressured by their families to move to the city in order to remit money to the rral area. In other cases women migrate to escape from male violence or dominance. Seceoru Etprences in Poveoy Aliarion 285 It should also be undertood that urban labor markets are fequently less free and rational than assumed by the labor economists. In fact, many labor markets are rigidly segmented by umions and other organizations protecting the interests of workers in the modem sector and ensuring that wages will not be lowered by an influx of labor. Also, many forms of sexual exploitation are carried over into the workplace. Female employment is frequendy segmented by age, with yomng, unmarried woren having greater opportunities in the modem, export sector, and with older women being confined to the unskilled labor force in more traditional, and less growth-oriented industries. Frequently employment is closely linked to domestic activities. Women are also constrained by their triple role, which limits their flexibility and geographical mobility. Outwork and domestic service are two principal sources of female employment that illustrate this exploitation. Many women are involved in service industries where they tend to be involved in enterprises requiring less skill and capital than their male counteart. Women's employment oppornmities are frequently limited by their need to maintain teir multiple role. Ofteka as the head of a poor household without a regular male wage contribution, they are so preoccupied with the necessities of survival that they have little tinm or resources to invest in economic or educational improvement. ItmaUmUzadon and Poverty Alleviation (See Chapter 9 by Abdul Aziz) During the 1960s, 1970s, and 1980s, approaches to poverty alleviation tended to focus either on macroeconomic growth-promoting strategies or on the targeting of programs for poor and vulnerable groups. Although macroeconomic policies were able to promote significant economic growth in many counties, their emphasis on the allocation of resources to high- productivity regions and groups tended to marginalize the less productive areas and groups. At the same time, the impact of the targeted approaches (established specifically to reach the groups that bad not benefited from growth promotion) was much less than had been expected. This was due to a combination of weaknesses in top-down planning that caused mismatches between local needs and capacities, on the one hand, and the kinds of projects being promoted, the misappropriation of resources by local elites, and general finefficiencies and corruption in the administraion of the programs, on the other. Furthennore, these approaches tended to be standaized, inflexible ("blueprint-) exercises that required govemment extension workers to achieve cenally determined numerical targets. A growing literaure documents the potential contributions of decentrazed development (Shams and Siedenopf 1987; Uphoff 1988; Moser 1988; Martin 1988; Bamberger and Shams 1989; Getubig and Shams 1991). The advocates of dalization argued that it is more responsive to local needs, that it provides a better match of local needs and government resource, and that people's involvement increases the efficiency of implementation and susinability. Improved monitoring by local govemumei and beneficiaries should improve asset reteion and improve loan recovery. The programs should also be made more cost-effective through reduced overheads. While this LIteratu supports the potentially imrtant contrbutioms of decentralizaion, it also highlights the administrative and organizational difficulties involved in its implementation, the inability of local agencies to effectively plan and manage decentralization, and the cotimed influence of local elites. 286 Midaed BRmbeiger Shams and Siedentopf (1987) argue that in fact most South Asian govemments have achieved very little effective decentraizaion and that their preference has been for deconcenradon or delegation, both of which allow the central and state governments to continue playing a strong role (this was referred to in the seminar as "centralized planning with decentralized implementation-). Most of the significant progress toward decentralization has occunred where nongovernment organizatons have been given a more active role and closer ties have been established between them, the govenment, and international agencies. A major issue raised in the seminar, as well as in the literature (see Bamberger and Shams 1989) is the extent to which successfil small-scale NGO programs can be replicated on a larger scale and in cost- effective manner. the Experience of Karnataka Kamataka was one of the first Indian states to introduce decentralizton, and a strong commitment to decentralization has existed at least since the present system was put into place soon after enabling legislation in 1985. Under the new system, the chief development officers and other local officials are now accountable to the elected representative, the president of the zilla paishad (ZP), and not to state officials. Although the plannig fnction has been decntalized to the district level, the ZP still has only limited fcial independence for it relies mainly on grants-i-aid from the state, of which approximately 86 percent is allocated for committed expendirs (salaries, ongoing state schemes, minimum needs programs, and the like). As a result, the ZP can only decide the allocation of 14 percent (in addition to the modest financial resources it is able to mobilize direcdy). Four main poverty programs are affected by the decentralization: (a) wage employment programs, (b) self-employment promotion through integrated rural development, (c) skdlls taining and the mprovement of living conditions, and (d) specia programs including skills training and provision of housing and basic services for scheduled castes and scheduled tnrbes. A great effort is made to ensure that each of these programs responds to locally perceived needs and that the beneficiaries are the most needy. Despite some accsations of favoritism and corruption, the selection of projects and the idetification of those in need of aid by the villagers themselves rather than by distant state officials, does seem to mcrease equity and responsiveness to local needs. There is also evidence of inreased efficiency in the monitoring of asset retention and improved loan recovery. Thus, prelimEiary findings suggest that the decentralization policies offer some promise of increasing the efficiency and equity of poverty alleviation programs-at least in states in which there is a strong political commiment to these objectives. Lessons and LIsues 1. It is not yet clear how much effective financial authority has been delegated to local organiztions, nor even whether central and state governments are willing to reduce their financial control over local authorities. 2. The complex issues of how to reconcile the need for coordinaed development strategies with greatly increased local control over how resources are allocated and programs managed at the local level have not yet been resolved. Sedoral Experiencs in Powvert Alleviation 287 3. Although it is widely accepted that NGOs have a major contribution to make in ensurng effective local participation in project planning and implementation, relatively little progress has been made introducing effective ways to institutionali their relationships with government. 4. Very litde systematic analytical work has been done to assess the cost-effectiveness and overall efficiency of different approaches to decentralzation, either through government programs or in cooperation with NGOs. Few data are available to support either the claim that NGO service delivery systems are much more cost-effective than govermment delivery systems or the opposing claim that NGO systems, while providing a high-quality product, have higher unit costs. 5. More effective systems are needed to monitor the implementation and operation of decentralization programs. 6. Urgent attention must be given to insttion building at the local level and to strengtheningthe capacity of local govermment agencies and community organizations for planning and imple g d ralization policies. 17- CONCLUSIONS AND RECOMMENDATIONS: IMPROVING THE PERFORMANCE OF POVERTY ALLEviATION PROGRAMS AND PROJECTS Michael Banberger and Abdul Azi Factors Affecthn the Sccess of Poverty Alleviaion Projecs With a few notable exceptions, the performance of poverty alleviation progrms and projects has been quite disapointng in terms of their ability to reach the poor in an eqlitable and cost-effective manner and to produce a significat and sustinable impact on the lives of a significant proportion of the target groups (Getubig and Ledesma 1988; Getubig and Sbams 1991). This ointing performance can be traced to the policy environment, project design and management, administrative and organzatonal issues, opposition from powerfil stakeholders, and the lack of adequate gender analysis, among other factors. T7he Poicy Environnent While most of the poverty alleviation progms discussed duing the seminar were iitiated by the central or state govements, in many cases the politcal will was not strong enough to ensure that programs were implemented as intnded or that they could be sustained.For example, 0 Very few governments have been willing to take the necessary measures to enforce loan recovery or the payment of user fees and service charges (see Chapters 3 and 4). Pardy as a consequence of this, most projects have been poorly mnined or lackd the resources to purcase new equpment and supples. Poor loan recovery rates have also mean that credit agencies have not have been able to recycle these fimds through new lan2s. * Govemuents have been reluctant to relinquish the financial control necessary to implemeat effective deetalization policies (Reddy 1992). In many cases central planning agencies have not been willing to promote a genuine decentralization of the planning function (Chapter 9; Shams 1991). * While recogning the potental contnrbutions that NGOs can make to project planning and management, govenments have been reluctant to allow NGOs to have the financial and opraional autonomy they need to operate effectively. * Inadequate attention has been given to factors affecting project sustainability (Bamberger and Cheema 1990). Financiaprovisionrarely oDvers operational and mantenane cos. Akthough attention is paid to ensuring that the agencies responsEble for project implementation are carety chosen and have the necessary institutional and 289 290 Michad Bamberger and Abd Aziz administrative support, much less attention is given to selecting and strengthening the agencies that will be responsible for the management of a project once it becomes operational. Project Design and Manageme There is ample documentation to show that poverty projects have a number of unique characterisics that must be taken into consideration in their design and implementation (Bhatt 1991; Morato 1991; Shams 1991; Yahie 1993). However, this is rarely done. 0 While the success of a poverty project certainly depends on the adequacy of its financial, economic, and technical design, equally important is the way the project addresses cultural, social, institutional, political, and environmental issues. However, the appropriate forms of social analysis are rarely incorporated in a systematic way into project analysis and design (Asian Development Bank 1991). * It is now widely acknowledged that the successful implementation, operation, and sustainability of a poverty project (as well as many other kinds of projects) greatly depend on the extent and adequacy of the participation of the intended beneficiaries during project identification, design, and operation. Yet, most projects do not use the techniques of beneficiary participation or assessment in any systematic way (Salmen 1992; D'Silva and Bysouth 1992). Even fewer projects are willng to extend the length of the project preparation phase to allow for more thorough consultations and capacity building. * Most projects continue to follow a top-down 'blue-print' approach in which major decisions on the choice and design of projects are made at the centrl level. Project design rarely includes sufficient finaial and implemetation flexibility to allow for significant modifications as the project evolves. * The procedures used to identify and target the poor are often weak or poorly administered, and a considerable proportion of project resources are frequently diverted to the benefit wealthier groups (BRAC 1983). Monitoring procedures are usually weak, so it is difficult for management to obtain rapid feedback when problems of this kind arise (Ahmed and Bamberger 1989; Khan 1989, 1990). * The potential contribution of NGOs in project planning and implentation is freuently underutlized. It is vital to overcome the mutual suspicien and distrust that often exists and to develop more flexible adminisatdve procedmes to allow government to exercise the necessary degree of financial contol over public resources without taking away the flen1Diity of action that is one of the main strengths of NGOs (Bamberger and Shams 1989). Adminisvtrtve and Organizaioa Issues The size and complexity of governments and the wide range of responsibilities they must assume often make it difficult for them to ensure the administrative fleibility needed to deliver Improbong the Perfonrance of Poverty Alvation Projecs and Program 291 services to the poor in a cost-effective and equitable manner. The following are some of the typical problems and issues that arise concernin, the management of poverty programs: * Government agencies prefer to have standard procedures to simplify operations and to facilitate the definition of staff responsibilities and to monitor their performance. Consequently, many of these agencies find it difficult to build in the flexibility required to adapt to local social and economic conditions. * For similar reasons, govenments like to define targets (fLor numbers of loans authorized, numbers of children vaccinated, and so on) and set quantitative achievement targets for their staff. These targets have become a major problem, particularly in credit programs where extensiun workers are often accused of trying to force people to accept a loan for the acquisition of an animal, agricultural input, or machine which they do not wish to have. * Governments seek to ensure strict control of how funds are utilized and consequently require tha expenditures are precisdly defined and controlled. It is extremely difficult to change the way in which approved fumds are utilized, which is one of the reasons why little effort has been made to increase the flexibility with which funds in poverty projects can be utilized. Overcoming Opposkion from Powerjd Swkeholders Many poverty projects face two kinds of problems concerning powerful economic and political groups. First, these groups often try to divert resources intended for the poor to themselves or their followers (BRAC 1983) . Significant proportions of low-interest loans, subsidized agricultural iputs, low-cost housing, and subsidized food are frequently diverted to powerful local nersts. Given the influence of these groups, it may be difficult to control the misappropriation of funds even when such activity is welt documented. The second problem is that many poverty programs are seen as a threat to political or econozuc interest groups. Some projccts involve the reallocation of land, while others permit the entry of poor producers into markets previously controlled by the local elites (see Chapter 12). Other projects involve direct political and economic empowerment of workers, squatters, or sectors of the electorate whose votes had previously been controlled by a political party. Still other projects involve the economic or legal empowerment of women or low-caste groups-again posing a threat to male workers, spouses, or religious groups. In many cases the threatened interest groups will bring considerable pressures to bear to sabotage the project-in some cases even resorting to physical violence. Many projects and programs have failed because of this kdnd of opposition. Lack of Adequate Gender Analysis Despite the fact that women are disproportionately represented among the poor, and that the economic contribution of women is seriously undervalued, gender issues are often largely ignored in project planning and implementation (Cbapter 7; D'Silva and Bysouth 1992). 292 MXCidiael Bmbefer and Abdul Aziz Improviog the Perfoae of Povery Alleviation Projects The Inportance of an Integrated Multilevel Approach The fmdings of the seminar strongly support the thee-component approach to poverty alleviation proposed in the 1990 World Development Report, namely: a. Promoting sustained economic growth through strategies that provide opportnities for the poor and permit them to participate in growth. b. Deliverig social services to the poor. c. Social safety nets and other targeted programs for the poor and vulnerable groups and those who are not able to participate in the beneits of economic growth. This reures an integrated approach with simultaneous and coordiated actions at the policy, sectorlprogram, and project levels. It also requires effective intersectoral coordination. Actions on each of these levels are of equal importance to the achievement of effetive and sustined povery alleviation. Creating a Condrive Policy Environmt A strong and sustained political commiment to poverty alleviation is essential (D'Silva and Bysouth 1992). However, good intentions are not enough. The strengthening of political will must be accpanied by the creation of a policy environment conducive to poverty alleviation (Chapter 11). A number of examples cited in the seminar demonstatd how well-intentioned policies can actually detract from susained efforts to reduce poverty. The provision of free or subsidized services is a case in point. This does not allow beneficiaries to feel that they are responsilble for ensuring the maintenance of the facilities and services provided. It also eliminates the possibility of geneating revenue to cover operations and maintenance costs. Another example is the assumption by many policymakers that most women do not have primary economic responsibility for their household. This attitude helps perpetua wage and employment discrimination (Chapter 7). A number of steps could be Uken to create a conducrve policy environment: a. Subsidy programs should be reappraised and an assessment made of their impact on the long-term sustainability of the programs they support and on the poor. In many cases the subsidies should be eliminated or significandy reduced. b. Greater emphasis should be placed on bottom-up planning methods that give the poor have an active part in the planning and implementation processes (Cbapter 12). c. Authorities should provide greater support for d alizon policies that combine financial as well as adminisaive decenazation. A strong political commitment is required to ensure that the decetralization of decisiomnki is Improving the Perfonnwce of Powvrty Ariaon Projcas and Programs 293 real and not simply symbolic. The discussion of decentralization experience in Kanataka (Chapter 9) illusrates the many barriers to effective dalization. d. A much stronger commitaent should be made to enuin the sustainability of programs and projects. Most governments and donor agencies have a relatively short time-horizon and are more concened about investing in new programs than in providing ongoing financial and other resources to ensure dtat programs are susained and continue to provide their services at an acceptable level and quality. The political will to follow through with such programs is particularly important in a federal system such as that of India, where both the central and state governments tend to pass the responsibility for operations and maintenance on to lower-level authorities who do not have the resources to adequately cDver these costs. Developing an Effective Data Base The formulation and implementation of effective poverty policies and programs requires acess to reliable and timely information on the magnitude, characteristics, distribution, and changes in poverty over time; and on the impacts and effectiveness of different poverty alleviation policies and programs. These exercises must be based on accuate and frequent estimates of the characteriics of poverty-which already exist to a reasonable degree in India and to a lesser exte in other countries.The monitoring and evaluation of poverty programs also needs to be more effiective.The South Asian countries are particularly deficient in this area (Ahined and Bamberger 1989; Khan 1989) Creatng a Poventy-Responsive Instizionad and Adminstraive Stucture Poverty-responsive policies and prngram quie organational structures and adminisative procedures tha are more flexible and responsive to the interests of the poor (Tilakaratn 1991). The following are some of the institutional requirements: a. Bottom-up planng and management procedures that include systmatic mechanisms for consulting with, and obtaining feedback from, the poor and othr local groups and organiztio. b. Decentalized decisionmaking and program management. c. Decentraized financialn and responsibility. That is say, local groups muTst have more say in the control of financial resources, there should be greater flexibdity in how fauds are managed (with a move toward ea-post evalumion of how funds were used rather than complicated ex-ante approval procedures), and the revenue sources available locally (for example, the authority to introduce service carges, to raise taxes) should be increased. d. Greater role for NGOs in policymaking and implementaton. 294 Michadl Bamberger and Abdrd A2fz e. An increased role for the private sector in the delivery of services (Batalla and Morato 1991). f. Use of training and other management techniques to provide public officials with the sldlls and attitudes required to implement participatory approaches (Chapter 12; Shams 1991). g. Greater interagency coordination so as to permit the integrated delivery of a wide range of services. Sengthedng Managment Caday Poverty programs have a reputation for being badly managed. This is pary due to the fact that poverty programs are frequently considered to be social welfare semces rather than ways to promote economic growth through investments in human development. ConsequPncly, management accountability is often quite weak and tIn quality of programs suffers. The management of poverty programs could be strengthened in a number of ways: a. Clearer defnition of objectives and criteria for definig and evaluating success (Shams 1991) b. More rigorous project analysis and design. Chapters 12-15 provide guidelines for more effective project identification and design that emphasize the need for greater administrative and economic realism. They also stress the need to adapt conventional project design to the special c ristics of poverty projects. In most cases projects will be smaller, simpler. and of a shorter duration than conventional economic and infase investment projects (Bamberger and Shams 1989:chap. 4). C. Greater use of market studies to ensure the existece of a market for the goods or services to be produced through income and employment-generation projects (Chapters 12 and 13). d. More emphasis on demand-driven as opposed to suly-driven projects. Projects should respond to beneficiary needs and priorities rather than be cetrally plamed. c. More emphasis on systems of evaluntg performance to ensure that futue budget allocations are based on results (volume and quaity of ouus and impacts) and not simply on past expendits. f. More effective and management-responsive monitoring and evaluation systems. These should be considered a mgement tool providing rapid feedback on progress and problems, and not simply an edministrativ reporting system for fiunding agencies. Improing the Performance of Poveny Alkwiation Projects and Programs 295 g. Greater role for private sector agencies in providing services. This implies that greater effort should be made to recover the costs of services and to generate the necessary resources to cover operations and maintenance. h. The active participation of NGOS. Public sector agencies should use analytical procedures such as cost-effectiveness analysis to identify areas in which NGOs appear to have a comparative advantage in the provision of different services (Chapters 3 and 5). i. Greater attention to achieving sustainable programs and projects. This means, among other things, incorporating sustainability assessment into project analysis, involving beneficiaries in all stages of project design and implementation, and incorporating mechanisms to generate the resources required for operations and maintenance (Bamberger and Cheema 1990). Use of Appropriate Targeftng Mechanisms Frequent reference has been made to the debate on the merits of "narrow" targeting. Critics have argued that most targeted programs are expensive and difficult to admiister, and in most cases fail to reach the most needy groups. It is also argued that categorizing certamn groups as poor or weak may produce a stigma or disincentives to seek work and thus prevent these groups from becoming self-reliant. However, others have argued that at least in the South Asian context, narrowly targeted programs are the only way to overcome the effects of illiteracy, unequal distribution of economic and political power, cultual and religious barriers, and unequal regional resource endowmets-and to ensure that resources and services reach the poorest and most vulnerable groups. The issues involved are extremely complex. However, the following guidelines can be used to assess the potential benefits and costs of targeting in a given circumsac and to select the most effective targeting mechanism for a particular program (Grosh 1992). a. incidence analysis can be used to assess who is and who is not benefiting from programs. If a particular program or service appears to be accessible to all major low-income and vulnerable groups, the design of a narrowly targeted program is probably not justified. However, if it can be demonstrated that certain groups are excluded on the basis of sex, region, race, or similar factors, then more narrowly targeted projects may be required. b. An analysis should be made of why cerain groups are excluded. In some cases the problems are administrative and can be corrected without the need for a special program. In other cases the problems may be more complex, and a particular program may be unable to reach major sectors of the target group. c. Where targeting is to be used, a careful comparison should be made of the alternative approaches and their probable cost-effectiveness for a paicular program. There are three main approaches to targeting: individual enitlement on the basis of economic or social characteristics of an individual or household; geographical targeting to sectors (low-income urban or rual populations), regions 296 Mid ad Bamberger and Ad Azrk (resource poor geographical regions) or states; and self-targeting, whereby an individual, household, or community decides whether to participate in a program or to solicit a service. Each of these approaches, and their many variants, have advantages and disadvantages in terms of cost, complexity of operation, and ability to reaching different groups. Studies such as the recent work of Margaret Grosh (1992) assessing targeting mechanisms in Latin America provide some usefil guidelines for choosing between these approaches. Integrating Womzen into Poveny Programs It has been emphasized throughout this report that there is a systematic gender bias in the formulation and implementation of many development policies and programs (Chapter 7). These biases are equally evident in many poverty alleviation programs. Consequently, it is essential to adopt a proactive approach in the selection, design, and management of all poverty alleviation programs to detect an eliminate any gender biases. The following guidelines would be useful in this regard: a. Increase the awareness of policymakers and planners of the fact that the economic contribution of women is underestimated and promote the use of gender analysis in policy formulation and project planning and implementation (Chapter 14). b. Develop political and economic organizations to assist women in the labor market, help them protect their land rights, and help them fight for political, economic, and social equality (Shreshta 1991; Bhatt and Vyas 1991). c. Strive for flexible and targeted credit programs. Such programs play a particularly important role in providing women with access to the capital and technical assistance they need to break out of their economic and social dependence on their families and male relatives. Many of the more successfil credit programs have provided an entry point for a broader range of legal, economic, and political support (Chapter 8; Arunachalam 1989; Azad 1986). d. Promote programs that reduce the time women spend on collecting water and fuel so as to free up more time for work as agriculural producers. e. Endeavor to understand and protect women's existing sources of livelihood (D'Silva and Bysouth 1992). f. Eliminate discriminaory laws governing ownership and the control of productive assets. g. Promote the concept of equitable access to agricultural inputs. h. Help srengthn women's urban networks and support groups, which can be used as a survival strategy (Arunacbalam 1989; Clarke 1993). ABOUT THE AUTHoRs AbddAziz is a professor and head of the Economics Unit in the Institute for Social and Economic Change in Bangalore. South India. Earlier he worked in the Shri Ram Centre for Industrial Relations and Human Resources and taught economics at the University of Mysore. His special interest is industrial economics and local-level planning, and his publications in this field include a dozen research monographs and numerous journal articles. He has also edited three seminar volumes. He is associated with the academic programs of several Indian universities. Midhael Bamberger is senior sociologist in the Human Resources Development Division of the Economic Development Institute (EDI) of the World Bank and coordinator of EDI's Poverty Allviation Traiing Program. From 1985 to 1992 he coordinated EDI's taining programs in Asia, and previously from 1978 to 1985 he was adviser oni monito&ng and evaluation to the World Bank's housing and urban development programs. Dieter Bucher is a senior project specialist in the Social Dimensions Unit of the Asian Development Bank. Before joining the ADB, he held positions with Deutsche Gessellschaft fur Technische Zu narbeit (GT-Z) and the World Bank. Keith Johnson is a former senior development policy officer with the Asian Development Bank His responsibilities there included the monitoring and fiuther evolution of the ADB's poverty alleviation program and the development of its approach to the social analysis of projects. Earlier he was a senior economist with ADB's Economics and Development Resource Center. He is currently a senior policy analyst and unt manager of resource economics with the Ministry of Commerce, New Zealand, dealing with resource pricing and management issues. Ismaei Getubig is the coordinator, Poverty Alleviation and Rural Development Programme at the Asian and Pacific Development Centre, Kuala Lumpur. Since 1985 he has been involved in rural credit, replication of the Grameen Bank approach to credit for the poor, and studies of famers' organizations, the design and delivery systms of poverty alleviation programs, and the impact of modern fishing technologies on artisan fishermen. Nc>efeen Heyzer is the gender and development program coordinator for the Asian and Pacific Developmen Centre (APDC) based in Kuala Lumpur. She has edited a number of books on gender issues in development, including Daughters in Industy and Women Farmers and Rural Change in Asia, and has directed many research and training programs throughout Asia on gender and development issues. Neil O'Sulnr is a senior consulant with Interational Development Support Services, an Australian-based company specialiing in poverty alleviation and related consultanIy work. For most of the last two decades he has been intimately involved with the appraisal and evaluation of grass roots development projects throughout South and Southeast Asia. In recent years, hc has worked extensively with the Asian Developme Bank and other agencies in developing policies and strategies designed to apply successful grass roots developmet lessons at the national level. 297 Jayanil K Sada is a professor at the Indian Institute of Management in Ahmedabad. He has worked extensively in the areas of operations management, information systems, and management and their application to the health and population sectors. He has conducted numerous interational consulting assignments in the field of training and health management in Asia and Africa. Yue-Mcan Yeung is professor of geography and director of the Hong Kong Institute of Asia-Pacific Studies at the Chinese University of Hong Kong. Earlier he taught at the University of Singapore and woiked for the International Development Research Centre of Canada. He has published nmunerous works on Asian cities, low-cost how ing, and the informal sector. His latest publications include China's Coastal Cities (University of Hawaii Press, 1992) and Changing Cities of Pacific Asia (Hong Kong, Chinese University Press, 1990). 298 ANNEX 1 LIST OF SEMINAR PARTICIPANTS AND RESOURCE PEPSONS I. PARTICIPANTS Prof. K. V. Palanidurai Professor, District Planning Unit India Lal Bahadhur Shastri National Acadeny of Administration Smt. Lalitha Venkateshan Deputy Manager Mr. K. Neeladri Raju NABARD Deputy Director (Economics) National Institute of Rural Developmnt Smt. V. R. Sundaram Under Secretary Mrt. K. N. Patel Ministry of Urban Development Secretary and Law Officer Government of Tudia CIDCO, Legal Advisor, B.M.R.D.A. Mr. Ramchanda F. Dainde Mr. A. Theerthakarai Project Co-ordinator Deputy Planner Gram Vikas Sanstha, DEVRUIH Madras Metropolitan Development Authority Mr. H. S. Sethi Deputy Conmissioner (IRD) Ministry of Rurl Development Bangadh Goverment of India Mr. Dewan Ali Haider Alamgir Mr. Satish Tripathi Manager Secretary Palli Karma-Sahayak Foundation Urban Development DeparUment Govenment of Maharashtra Mr. Md. Muzibur Rabman Deputy Chief Dr. R. M. Kummur Planing Commission Development Policy Department NABARD Dr. Ekraamul Ahsan Member, Directing Staff (Research & Dr. N. Mohanan Consultancy) Deputy Director (co-op) Bangladesh Public Adminstration National Instiute of Rural Development Traiing Centre (BPATC) 299 30 Parficipwus and Resource Patsons for the Bangalore Seminar Banglzdesh continued Ms. Usha Rawal Women Development Officer Md Shahadat Hossain Karmali Zone, Jumla Regional Manager Women Development Division Rural Credit Project (RCP) Bangladesh Rural Advancement Committee Mr. Nanda Man Sthapit (BRAC) Senior Health Education Officer Public Health Division Mr. Muhanmad Kamaluddin Ministry of Health Deputy Secretary (World Bank Branch) Economic Relations Division Palksn Ministry of Finance Ms. Aysha M. Siddiqi Ms. Wahida Huq Deputy Director Senior Program Officer (Agriculture) Mmistzy of Local Government and Rural Resident Missin of the World Bank Development Bangaesh Ms. Hilal Bangash Ms. Tahera Yasmin Hug Director Planming and Projects Program Diector Rural Development Support Program Saptagram Nari Swanirvar Parishad Ms. Samia Raoof Ali Mr. Md Mazibar Rahman Senior Program Officer Deputy Direcor Sangi Development Foundation Directorate of Relief Government of Bangladesh Sri Lanka N.pal Mr. A. M. M. Sidique Additional Government Agent Dr. Rajenda B Shrestba Akurana Senior Economist Centre for Economic Development and Administraon (CEDA) II. RESOURCE PERSONS Tribhuban University Mr. Dieter Bucher Mr. Shekhar Bahadur Pradhan Asian Development Bank Deputy General Mana Internal Control Agriculture Development Bank Dr. Ismael Getubig Asian and Pacific Development Center Dr.. Noeleen Heyzer Asian and Pacific Development Centre Parfcipan and Resoure c Pc for the angalore Senar 301 Resource Persons confrmed Dr. (Mrs.) HemIata Rao (ISEC) Mr. Neil O'Sulrvan Dr. D. M. Nanjundappa ISEC) Intrnational Development Support Services (1DSS), Austrlia Dr. Dieter Bucher Dr. Michael Banberger Dr. J. L. Satia Mr. Neil O'Sullivan Professor (Public System Group) Dr. knmael Getubig Indian Institute of Managenet Dr. Md E. Ahsan Ahmedabad Mr. K. N. Patel Dr. Rajendra B. Shrestba Ms. Kaye Bysouth Ms. Aysba Siddiqi Inteational Development Support Services (IDSS). Australia V. DISCUSSANTS: ISEC Dr. Michael Bamberger Economic Developme Instihute Dr. A. S. Seethraml The World Bank Dr. T. Krishna Kumar Washington, D.C. Dr. Vinod Vyasulu Dr. P. H. Rayappa ir. Erling Dessau Dr. H. G. Hanumappa Resident Repreentative Dr. M. V. Nadkani UlNDP )Dr. N. Bhaskara ho India Dr. Abdul Azi VI. RAPPORTEURS: ISEC Institte for Social and Economic Change Bangalore Mr. Mathiyazhagan Dr. M. Prabladacdar Ell. PROJECT PROFILES Dr. R. Kanbargi PREPARED BY ISEC FACULTY Dr. K. N. Raju Dr. K. N. Nman 1. Dr. S. N. Sangita and Mr. Syed Mr. V. S. Pardasarathy Ajmal Pasha Dr. 0. K. Karanth 2. Dr. M. Prnhldachar and Mr. P. Mr. Johnson Samuel Thippaiah Ms. Mabel Koilpillai 3. Mr. Johmson Samuel and Dr P. Dr. -S. N. Sangita Hanunanthapayappa Mr. B. G. Kulkani 4. Dr. Shanta Mohan Dr. C. S. Nagaraju 5. Mr. M. Nageswma Rao Dr. B. S. Bhargava 6. Dr. (Mrs.) Gayatri Devi Mr. M. Venkata Reddy Dr. M. R. Narayana IV. CHAIRPERSONS Mr. M. Vivekanada Dr. V. S. Sathyaiya Mr. T. R. Satish Chandran (ISEC) . ;,0 w ANNEX 2 SEMINAR PROGRAM South Asia Seminar on the Design ad Management Sustainable Poverty Alleviation Projects (July 29-August 9, 1991) Bangalore, India MONDAY JULY 29 9.00 - 9.45 Inrucaory Session with Panicipas 1. Objecdves and organization of the seminar 2. Review of the program 3. Admiistr arrangements 4. Introducig paticipants and resource persons 10.00 - 11.00 Inaugural Session 1. Introductory remarks by: Mr. T. R. Satish Chanudran, Director. ISEC Dr. Dieter Bucher, Asian Develamem Bank Dr. Michael Bamberger, Economic Development Inttute, World Bank Dr. Erling Dessan, Resident Representative1 UNDP, India Professor Abdul Aziz, Head, Economics Unit, ISEC and Seminar Coordinator 2. Inaugural Address by: Professor M. N. Srinivas, Chahman, ISEC 3. Vote of tanks by Mr. M'Nageswara Rao, Registrar, ISEC 11.00- 11.30 TEA BREAK MODULE 1: SEMIM4R FPRAMEWORK, OVERVEWAND IDENTFCA7ION OF KEY ISSUES 11.30 - 12.45 1. "The South Asia Poverty Scenario' Presentaion: Dr. Michael Bamberger 303 304 Senarr Pmgmm 2. [The UNDP Human Developmen Approach to Poverty Alleviation" Presenaion: Dr. Erling Den, Resident Representative UNDP, India Chairman: Mr. T. R. Satsh Clandran Discussant: Dr. A. S. Seeharamu 12.45 - 2.00 LUNCH BREAK 2.00 - 3.15 *A Framework for the Assessment of Poverty Alleviaon Programs: Key Issues to Be Discussed during the Seminar" Presentaon: Dr. Michael Bamberger CDainan: Dr. D. M. Nanjundappa Discussant: Dr. M. V. Nadknri 3.15 -3.30 TEA BREAK 3.30 - 5.00 Reports on group discussions followed by general disussion. Rapporteurs: M. Matiyazhagan Dr. M. Prahkadachar TUESDAY JULY 30 9.00 - 9.15 Review of Day I Presentation: Rapporteurs MODULE 2: POVERTYALLEVLW ON PROGRMMS IN SOT AS14: EXPERENCE AND ISSUES 9.15 - 10.30 "Poverty Projects in Health and Nutrition: A Review of South Asian Expence" Presenation: Professor J. Satia, Indian Institute of Management, Ahmeddad Chairman: Dr. Diete Bucer Discussant: Dr. P. H. Rayappa 10.30 - 10.45 TEA BREAK 304 Seia P8rognm 305 10.45 - 12.00 Poverty Projects in Health and Nutrition: contned aCwir: Sri Lank Presentation of case studies by 2 participants, followed by general discussion. 12.00 - 12.30 Briefings on the projects to be visited in the aftemoon Chaiiperson: Dr. (Mrs.) Hemlata Rao Project: A Study of Jayarijendra Rag Pickers Project Brieings by: Dr. M. Prahladachar and Mr. P. Thippaiah Project Lacquerware Eandicraft Project Briefing by: Dr. S. N. Sangitha and S. A. Pasha Project: Poverty Alleviation through Senculture Briifing by: Dr. K. G. Gayathri Devei 12.30 - 1.45 LUNCH 2.00 -5.00 Project visits: Each group will be given an outline of the quesdons they should address in their presentaion the following mo. Evening Groups mee informally at their own convenience to review their visit and to orniz the presentaons they will make the next moring. Apportewrs: Dr. R. Kanbangi Dr. K. N. M. Raju WMEDNESDAY JULY 31 9.00 - 9.45 Reports on project visits 9.45 - 10.30 "The role of credit in poverty allUevi a review of Asian expience hi nual and urban areas" Prenaion. Ismael Getubig: Asia and Pacific Development Centre C7zir: Dr. Michael Bamberger 306 Se hmr Prgram Discussant: Dr. Vinod Vyasulu 10.30- 11.00 TEABREAK 11.00 - 12.15 Role of Credit in Poverty Alleviation (cotinued) Presentation of case stuies by 2 participants, followed by general discussion O7ir: Nepal 12.15 - 2.00 LUNCH BREAK 2.00 - 3.15 'Poverty Projects in Housing and Urban Development: A Review of South Asian Experience' Presenaion: Michael Bamberger, Economic Development ltime adr Pakistan Discussant: Dr. Krishna Kumar Followed by general discssion 3.15 - 3.30 TEA BREAK 3.30 - 5.00 'Survival Straegies of Low-income Households' Presentafion: Professor Abdul Aziz (ISEC) Owia Dr. Getubig Discussant: Dr. N. Bhaskara Rao Rqpportear: Dr G. K. Kazanth, Mr. V. S. Parthasarathy lfHUSDAY AUGUST 1 9.00 - 9.15 Review of previous day Presentcoin: Rapporteurs 9.15 - 10.1 'Rumal Development and Poverty Alleviation: The Experience of the Asian Development Bank' Presentadon: Dieter Bucher, Asian Development Bank Oar India Discussant: Dr H. G. Hanumappa Sanr Program 307 10.15 - 10.45 TEA BREAK 10.45 - 12.00 Rural Development and Poverty Aileviation (continued) Presentation of case studies by participants followed by general discussion. Chair: Banglade RapDorteUr: Dr. K. N. Ninan 12.00 - 12.45 LUNCH 1.00 Departue for train station 1.50 - 8.00 Train journey to Madras FRIDAY AUGUST 2 MODULE 3: FIELD VISIT AND PRACTCAL EXERCISES IN MADRAS Grou 1. Workng Women's Forun Group 2. Tamil Nadu bntegted Nutrition Project Group 3. Sites and Services anld Slum Upgrading SATURDAY AUGUST 3 GROUP DISCUSSION ON PROJECT VISITS MONDAY AUGUST 5 MODULE 4: WORKSHOPONLTHE PLANNING AND DESIGN OF POVERTY PROJECIS (Module dircted by the Intermational Development Support Services) 9.00 - 10.15 Introduction 10.15 - 10.45 TEA BREAK 10.45 - 12.15 SmaB group discussion on factors conbing to povertY 12.15 - 2.00 LUNCH BREAK 2.00 - 3.15 Presentation and discuion of the concepts of Social Analysis 3.15 - 3.30 TEA BREAK 308 Seminar Pogmm 3.30 - 5.00 Presentaton and disssion on project social analysis RaQ[rteurs: Ms. Mabel Koilpillii Dr. S. N. Sangitha TUESDAY AUGUST 6 9.00 - 10.15 Presentation and discussion on gender analysis Presentation: Noeleen Heyzer (APDC) Disc nt: (nme) Followed by general discussion 10.15 - 10.45 TEA BREAK 10.45 - 12.15 Group discussions on the incorporation of gender and social analysis in poverty alleviation projects. 12.15 - 2.00 LUNCH BREAK 2.00 - 3.15 Case study on thc ZOPP approach to project design. 3.15 - 3.30 TEABREAK 3.30 - 5.00 Case study/exercise on ZOPP Rppgrtgu: Mr. B. G. Kulkarni Dr. C. S. Nagraju WEDNESDAY AUGUST 7 9.00 - 10.15 Psentation of exercises on ZOPP 10.15 - 10.45 TEA BREAK 10.45 - 12.15 Discussion on ZOPP contimues 2.00 - 3.15 Decentralised Planning for Povery Alleiation" Presentation: Dr. Abdul Az 3.15 - 3.30 TEA BREAK 3.30 - 5.00 Discussion on decentralised planning continues Rappogreu: Dr. B. S. Bhargava Mr. M. Venkata Reddy Seminw Progrmi 309 THURSDAY AUGUST 8 9.00 - 10.15 Presentation and dicuion on NGO/Government coIlaboration models. 10.5 - 10.45 TEA BREAK 10.45 - 12.15 Presentation/discussion on flexiblelresponsive project designs for poverty alleviation 12.15 - 2.00 Lunch Break 2.00 - 5.00 Group exercises on the application of the planning tecbniques. Bgrteur: Dr. M. R. Narayana Mr. M. Vivekmnada 7.00 - 9.30 Closing supper and cultural performance: West End Hotel FRIDAY AUGUST 9 9.00 - 10.15 Presentaions on group exercises 10.15 - 10.45 TEA BREAK 10.45 - 12.15 Group discussions on specific recommendations for improving the design and management of sustainability poverty alleviation projects 12.15-2.00 Lunch Break MODULE 5: CONCLUSIONS AND 2.00 - 3.15 Presentation and ditussion on groUp reicodations Rpotur;nv Dr V. S. Satyapriya 3.15 - 3.30 TEABREAK 3.30 - 4.30 Seminar evaluation and discussion of follow-up Presentation of certficates REFERENCES Abmed, Viqar, and Michael Bamberger. 1989. Monitoring and Evaluating Development Project: The South Asia Erpenence. Economic Development Institute Seminar Series. Washington, D.C.: World Banlc Arunachalam, Jaya. 1989. "Case Study of te National Union of Working Women: An Indian Experiment." Course Note . World Bank, Economic Development Institute, Wasnon, D.C. Asian Development Bank. 1991. Guideinefor Social Analysis of Development Projects. Manila: Asian Development Bank. Azad, Nandini. 1986. Empowering Women Workers: TIe Working Women's Forum Experiment in Indian Cies. New Delhi: UNICEF. Aziz, Abdul. 1991. 'Decentalised Planing: The Karnataka Experiment." Istitute for Social and Economic Change, Bangalore. Baldwin, George. 1990. 'Non-govemmental Organzations and Afican Development: An Enquinry" In World Bank, The Long-Tnmn Penpcive Stud o.f SbSaharan Africa. Vol. 3: Insitutonal and Sociopobical Issues. Washigton, D.C. Baker , Judy. 1992. 'Dominican Republic:Targeting Maternal-Child Health Care through PROMI" in Margaret Grosh From Plaitudes to Practice: Targeting Socal Programs in Latn America. Latin America and the Caribbena Technical DeparDment. Washington D.C: World Bank. Bamberger, Michael, and Shabbir Cheema. 1990. Case Sudies of Project ustanability: Implcaionsfor Polcy and Operatins from Asin Etpenence. Economic Development Insttute Seminar Series. Washingt D.C.: World Bank. Bamberger, Gonzalez-Polio, and Sae-Hau 1982. Evauaton of Sites and Services Projects: The Edence from El SalvAdor Staff Working paper No. 549. Washgton,D.C: World Bank Bamberger, Michael, and Scott Parris. 1984. 'The Structure of Social Networks in the Zona Sur-Oriental of Cartagena.' Uxban and Regional Report UDD-50. World Bank, Urban and Regional Economics Dqeatmet, Washington, D.C. Bamberger, Micbael, and Khalid Shams. 1989. Comnity Participation in Project Management: The Aszan Expeience. Kuala Lumper. Asian and Pacific Development Centre. 311 312 References Banberger, Michael, Abdullahi Yabie, and George Matovu. 1993. The Design and Managemet of Targeted Poverty Aleviation Projects in Anglophone Africa. Washington, D.C.: World Bank, Economic Development Institute. (Forthcoming). Bangladesh Rural Advancement Committee (BRAC). 1983. The Net: Power Strucure in Ten Wll,ages. Dh: BRAC. BaUtla, Eric, and Eduardo Morato. 1991. "Harnessing Private Sector Resources for Poverty Alleviation: The Philippine Business for Social Progress. La Isnsel Getubig and Kbid Shams, eds., Reaching Out Effectivey: Improving the Design, Management and ISmplemion of Poverty Allevion Programntes. Kuala Lumpur: Asian and Pacific Development Centre. Bhllia, Surjit and Paul Glewwe, 1986. "Growth and Equity in Developing Countries: A Reintrpretation of the Sri Lanka Exrienc" World Bank Economic Review (September) Volume 1 Number 1:35-64 Bhatt, Anil. 1991. "Critical Elements in the Design of Poverty Alleviation Programmes." In ismael Getubig and Kbalid Shams, eds., Reaching Out Effectively: Improving the Design Manageeg and Implemenaon of Poverty Alevaon Programm es. Kuala Lumpur: Asian and Pacific Development Centre. Bbatt, Anil, and Vyas Jayshree. 1991. SEWA Cooperative Bank. Design and Delivery of Bankig Services for Poor Women." In Ismael Getubig and Khalid Shams, eds., Reching Out Effectivey: ImWroving the Design, Management and Impemntaon of Povery Alevation Programmes. Kuala Lumpur: Asian and Pacific Development Centre. Bruton, Henry and associates. 1992. The Political Econony of Poverty, Equty and Growth Sri Lanka and Malaysia. New York: Oxford University Press. Callaghy. 1990. "Lost between State and Market: the Politics of Economic Adjustment in Ghana, Zambia and Nigeria." In Joan Nelson, ed., Economic Cisis and Polciy Choices: the Politcs of Adustment in the Third World Princeton, N.J.: Carrion, Diego, Mario Vasconez, and Jorge Garcia. 1989. 'Access by the Urban Poor to Basic Inastrct Services: Latin America Region Paper." Prepared for the World Bank, Economic Development Institute, Infastucture and Urban Development Division, Washingo D.C. Chattejee,l Meera. 1990. Indian Womter Their Health and Economuc Productivity. World Bank Discussion Paper No. 109. Washigton D.C: World Bank Chenery, Hollis, and others. 1974. Redstribuion with Growth. New York: Oxford University Press. References 313 Clark, John. 1990. Democratizing Development: The Role of Voluntary Organizatons. Conn.: Kumarian Press. Clarke, Mari. 1993. House/old Economic Stegies and Supporn Networis of the Poor in Kenya a Literature Review. Economic Development Institute Worldng Paper. Washington, D.C. Cornia, Giovanni, Richard Jolly, and Frances Stewart, eds. 1987. Adjustment with a Hmnan Face: Protecting the Vulnerable and Promoting Growth. 2 vols. A UNICEF Study. Oxford: Oxford University Press. Cox, Donald, and Emmanuel Jimenez. 1989. Privare Tranfers and Public Policy in Developing Countries: A Case Study from Peru. World Bank Policy, Planning, and Research Working Paper 345. Washington, D.C. Department of Rural Development. 1989. Concurrent Evluation of Rural Development Programs. Ministry of Agriculture. New Delhi. Government of India. D'Silva, Emmanuel, and Kaye Bysouth. 1992. Povery Alleviation through Agricultua Projects. Economic Development Institute Seminar Report 30. Washington, D.C.: World Bank. Garcia, Haydee and Alberto Levy 1992 "Venezuela: El PAMIv (Veneuzela: the mtrition program for mothers and children) in Margaret Grosh From Pkltudes to Practice: Targeting Soda! Programs in Latin America. Latin America and the Caribbena Technical Department. Washington D.C: World BanI Germani Agency for Technical Co-operation (GTZ). 1987. ZOPP: An Introduction to the Method. Frankfurt am Main. Gentbig, lsmad. 1992. The Role of Credit in Poverty Allevaon: A Reiew of Asian Exerience. Washington, D.C.: World Bank, Economic Development nstitute. Getubig, Ismael and Antonio Ledesma. 1988. Voices from the Culture of Silence: The Most Disadwaaged Groups in Asian Agriculure. Kuala Lumpur: Asian and Pacific Development Centre. Getubig, Isnael, and Khalid Shams. 1991. Reahng Owt effectively: Inqroving the Design. Management and Iplementton of Poverty Alleviaton Programs. Kuala Lumpur: Asian and Pacific Development Centre. Govenment of Karntaka. 1988. Distict Governments and Decentralised Plannig. Bangalore. 314 References Graham, Carol. 1992a. "The Politics of Adjustnent and Poverty in Zambia: The Hour Has Come.' Photocopy. - 1992b. "The Vocal versus the Needy in Senegal." Photocopy. Grosh, Margaret. 1990a. Social Spending ir . attn America: The Story of rhe 1980s. World Bank Discussion Paper 106. Washington, D.C. . 1992. From Pltudes to Practice: Targeting Socida Programs in Latin America. 2 vols. Washington, D.C.: World Bank. Grootaert, Christiaan, and Timothy Marcham. 1991. The Sodl Dimensions of Adjastment Priority Survey. Poverty and Social Policy Division. Africa Techn ical Department. Washington, D.C.: World Bank. Heaver, Richard. 1988. Reaching People at the Peripery-Can the Bank's Popion, Healh and NuMrition Operaions Do Better? World Bank Staff Working Paper 81. Washington, D.C. Heyzer, Noeleen, ed. 1987 Women Farmers and Rural C/ange in Asia. Kuala LJupur: Asian and Pacific Development Centre. 1 1992. Gender Isues in Paverty Alleviation: Experience from Asia. Washington, D.C.: World Bank, Economic Development Institute. India, Planing Commission. 1984. Report of the Working Group on Disict Plwaning. Vol. 2. New Dehi: Govenument of India. Jorgensen, Steen, Margaret Grosh, and Mark Schacter. 1992. BoliWa's Answer to Poverty, Econonmc Crises and Adjustment The Emergency Social Fund. World Bank, Regional and Sectoral Studies. WashWgton, D.C. Kabeer, Naila. 1989. "Monitoring Pcoverty As If Gender Mattered: A Methodology of Rural Bangladesh. "Discussion Paper. Institute of Developmean Studies, University of Sussex, Brighton. Kaufmann, Daniel. 1982. "Social lnteractions as a Strategy of Survival among the Urban Poor: A Theory and Some Evidence." Ph.D. diss., Harvard University, Cambridge, Mass. Kaufinmn Dani, and David Lindauer. 1980. "Basic Needs, Interhousehold Tramsfers and the Extended Family. " Urban and Regional Report 80-15. World Bank, Urban and Regional Eonomics Division, Washington, D.C. Referencs 315 Kafnn, Daniel and David Lindauer. 1986. "A Model of Inoe Transfers for the Urban Poor". Joural of Deveopment Economics 22 (July-Augst):337-50 Khan, Adil. 1989. "A South Asian Regional Sita; on Current Thoughts and Practces in Monitoring and Evaluation." Course Note 645/015. World Bank, Economic Development Inste, Wasdngton, D.C. . 1990. "Monitoring and Evaluation of Developmet Projects in South East Asia- The Experience of Indonesa, Malaysia, the Philippines and hailand." Course Note 6451019. World Bankr Economics Developmet Intue, Washigon, D.C. Knowles, James, and Richard Anker. 1981. An Analysis of Income Trnsfers in a Developing Country: The Case of Kenya." Jouwd of Deveopment Enomics 8:205-326. Korten, David. 1980. "Comwnmity O n and Rural DevelopmenL A Learning Process Approach." Pidc Adminaton Review 40(5):509-17. Korten, Frances, and Robert Sly. 1989. lrwwjobning A Buraumy. lanila: Almo De Manila Univemsity Press. Lacey, Robert. 1989. Managing Pkblic Expendture: An Evolvng World B*an Perspectiv. World Bank Discussion Pae 56. Washington, D.C. Lar-Brand J. 1970. Our Claim on hie Fure. New York: Friendship Press. . Lewis, John P. 1988. Strengdzing the Poor. VWiar Have We Leaned? Washington_ D.C.: Ovee Development Council. Lewis, Oscar. 1965. La Vida A Puerto Rican Family in the Culture of Poety-San Juan and New York. New York: Vintage Books. 1 1968. A Sudy of Slam Culures: 1Bac*rwd to "La VUda. w New Yotk: Random Hnuse. Limn, Hong Rai. 1988."Urban Service Prvision in Plua Society: Approaches in Malaysia." In D. A. Rondinelli and G. S. Cheemn, ads., Urban Seruic in Deveoping Counies. London: Macmila. Lomnitz, Larissa. 1977. Net Worb and Marg*iity: Life in a Mxcan Shantytowm. New York: Academic Prs. Marc, Aexa , Marc-Shacter, and Caol Gain 192. -Social Acon Pogm and Social Funds: A Review of Design and Impl i Sub-Saharan Afica." Poverty and Social Policy Division. Africa Technial I1 >ament World Bank. Photco.- 316 References Martin, Patricia. 1988. "Community Participation in Health and Population Programs." Course Note 076/004. World Bank, Economic Development Institute, Washington, D.C. Morato, Eduardo. 1991. "Alternative Delivery Systems for Poverty Alleviation Programmes." In Ismnai Getubig and Khulid Shams, eds., Reaching Out Effecdvely: biprovng the Design, Managemet and Implemenaion of Poverty Alleviation Programmes. Kuala Lunpur: Asian and Pacific Development Cete. Moser, Caroline. 1988. "Approaches to Community Participation in Urban Development Programs in Third World Countries." Comse Note 076/003. World Bank, Economic Development Insttute, Washington, D.C. National Institute of Urban Affairs. 1990. The Challenge of Urban Poverty. New Delhi. Nelson, Joan, ed. 1990. Economic Crisis and Polciy Choices: The Politics of Adjustmen in the Thd World. Princeton, N.J.: University of Priceton Press. Nelson Nici. 1978. "Female-Centered Families: Changing Patterns of Marriage and Family AonDg Buzaa Brewers of Matiare Vailey." African Urban Sudes 3(Winter). Michigan State University. 1979. 'WVomen Must Help Each Othern The Operation of Personal Networks Among Buaa Beer Brewers in Maftare Valley, Kenya.' In Women Uniteit Women DiMde Compative Studies of Ten Contmporay Clures. Indiana University es.s. Ng Cecilia Siti Nor Hamid, and Syed Husin All. 1987. 'Rural Development Programmes, Womep's Participation and Orgnzations in Malaysia.' In Noeleen Heyzer, ed., Women Farmen and Rural Change in Asia: Towards Seaml Access and Equay. Kuala Lumpur: Asian and Pacific Development Centre. Nieves, Isobel. 1979. "Household Arrangs and Multiple Jobs in San Salvador.' Signs: Journd of Women in Cure and Sciety 15(1). Norris, William. 1977. "Suvival Stategies and Social Structure in an Urban Squatter Setlemen: Housiold Networks of the Urban Sea Dwellers of Bahia, Brazil." Ph.D. diss., Dqertment of Sociology, Harvard University, Cambridge, Mass. Norwegian Agency for Developmet Co aon (NORAD) (1990). 7he Logical Frmnework Approa* HWlanWoocfor Objectives-orientd Projea Plwning. Oslo- Obbo, Christine. 1980. Affccan Women: Their Strgglefor Eonomc Independence London. Zed Press References 317 Obidegwu, Chkwuma. 1990. -Adjustment Proams and Economic Change in Sub-Sahara Africa?- World Bank, Strategic Planng and Review Deparme, Wasibngton, D.C. . 1992. 'Recent Economic Trends, Adjustment and Poverty in Sub-Sahara Africa." Paper presented at the Economic Development Institute Semina, "The Design and Management of Targeted Poverty Allevation Programs in Anglophone Africa.' Kampala, May. Parris, Scott. 1984. "Survival Stategies and Support Networks: An Anthropological Perspective.' Urban and Regional Report UDD-58. World Bank, Urban and Regional Economics Division, WasWngton, D.C. Peattie, Lisa. 1969. The View fom the Barrio. Ann Arbor, University of Michigan Press. ____. 1981. "Marginal Settlements in Developing Countries? Anna) Revew of Socolog Volume 7:157-75. Pffrman, Guy, and Charles Griffin 1989. Metrion and Health Progrmns in Latin America: Targeing Social Ependitures. Washingo, D.C.: World Bank. Pinsup-Anderm i, Per, Mauice Jaramiflo, and Francis Stewart. "The Impact on Govemment Expendiur." In Cornia, Jolly and Stewart, eds., Adjusmen wieh a Human Face: Proteding the Vube-abkeand Promoting Growth. Vol. 1. A UNICEF Std. Oxford: Oxford University Pre. Pryor, Frederic L. 1990. The Poltical Ecowmy of Poverty, Eqity and Growtk Malaw and Madagascar. New Yorc: Oxford University Press. Purey, Robet V. 1989. Making the Poor Credit Worthy. World Bank Disussion Paper 58. Washington, D.C. Raczynskd, Dagmar. 1993. *La Focalizaci6n do Pograms de Salud y Nutrici6n para Madres y Nitos de Escasos Recursos: Lecciones de la Experiencia de Chile- (Targeting Health and Nutrition Programs for Poor Women sad Children: Lessons frm the Eperince of Chile). Economic Developme Institute. Working Paper. Washidngton D.C: World Bank. Rao, V. M., and Abdul Aziz, eds. 1990. Poverty AUeaton in Inda. New Delhi: Ashish Ravallion, Mautin, Gaurav Datt, and Shaohma Chen. 1992. "New Estimates of Agggate Poverty Measus for the Developing World 1985-1989." World Bank, Population and Human Resources Department, Washingto, D.C. Rawls 1973. A Theory of Justice. Cambridge Mass. Harvard University Press. 318 References Reddy, K. N. 1992. "Health Expendiurs i India." Working Paper 14. National Institte of Public Finance and Policy, New Delhi Reforna, Mila, and Ricci Obusmn. 1981. "Household Networks and Survival Strategies among the Urban Poor. Monetary and Non-monetary Transfers among Selected Families in Tondo." Urban and Regional Report 81-22. World Bank, Urban and Regional Economics Division, Washington, D.C. Sahlins, Marshall. 1968. -On the Sociology of Primitive Exchange." -i M. Banton, ed., The Relevance of Modets for SociaAnthropology. ASA Monographs 1. London: Tavistock. Sahn, David. 1992. "Public Expenditire in Sub-Saharan Africa during a Period of Economic Reforms." World Developmezt 20(5):673-93. Salmen, Lawrence. 1992. "Beneficiary Assessnt: An Appach Described." World Bank. Washington, D.C. Sandroni, Paulo. 1982. "La Proletrizaion de la Mujer en Colombia despues de 1945" (The proletarizaton of women in Colombia since 1945). In Magdalena Leon, ed., La Realidad Colomia vol. I (Debate Sobre la Mujer en America Latna Y el Carzbe) (The debate on the status of women in Latin America and the Canbbean). Bogota, Colombia: ACEP. Schmink, Mariane. 1982. "La Mujer en la Economia Urbana de America latina' (Women in the urban economy in Latin America). In Magdala Leo, ed., La Realidad Colombia vol. 1 Pebate Sobre la Mujer en America Lata y el Carbe). Bogota, Colombia: ACEP Schultz, T.W. 1980. "The Economics of Being Poor." Buetn o the Atomic Scientists. Segmua de Camacho, Ndbra. 1982. "La Reproduccion Social, Famila y Trabajo" (Social reproducton, family, and work). In Magdalena Leon, ed., La Realidad Colomhbiaa, voL 1 (Debate Sobre la Mujer en America Latia y el Carie) Bogota. Bogota, Colombia: ACEP. Sen, Amartya. 1992. "Political Economy Issues and Targeting.' Paper presented at the World Bank Conferece, "Public Exenditues and the Poor Icidence and Targeting,U Washington, D.C., June 17-19. Shah, Kirtee. 1990. "Conmuny PartIon in the Hyderabad Urban Commnity Development Project.- Course Note 0761007 World Bank Economic Development Institute, Washigton, D.C. R4fereces 319 Shams, Khalid. 1991. "Essential Design, Mngemet and Imlementation Feaures of Poverty Alleviation Programmes." In Ismael Getubig and Khalid Shams, eds., Reacing Out Effectively: Improvng th Design, Managem and Imploemeaon of Povert Alleviation Progranmes. Kuala Lumpur: Asian and Pacific Developmen Centre. Shams, Khalid, and Neinrich Siedeopf, eds. 1987. Building frm Below. Loal Initatives for Decentraled Development n Asia and Pacfic. Kuala Lump Asian and Pacific Development Centre. Shresita, Rajendra. 1991. 'Production Credit for Rural Women (PCRW) in Nepal.' In Ismael Getubig and Khalid Shams, eds. Reacng Ot Effecivdy: Imroving t Dign, -Management and Implemnauton of Povey Aleviaion Programmes. Kuala Lumpur Asian and Pacific Development Centre. Skilhings, Robert, Alexandre Marc, and Roza MakonnnL 1991. "The Design of Socia Action Programs under Structural Adjustment. " World Bank, Afica Reion, Socil Dimensions of Adjusmnt Unit, Wasbigton, D.C. Soto, Reman de. 1989. The Ote Path: The Iviible Revolution n th Third World New York: Harper and Row. Stack, Carol. 1976. All Our Kin: Strategies For Swnivul in a Black Cy. New Yot Harper and Row. Tilakaratna S. 1991. 'Capacity Building of the PDor. The Importan of an ed, Multi- level Approach.' In Ismael Getubig and Khalid Shams, ads. Readcing Out Effectively: IWroing t Design Managem and kImrnadon of Paver-y Aleviaion Programmes. Kuala Lumpur: Asian and Pacific Developmnt Centre. United Nations Deve!opink Programme (UNDP). 1990. Hum Deveopment Report 1990. Geneva. . 1991. Human Development Report 1991. Geneva. Uphoff, Norman. 1988. "Approaches to Commuity Participation in Agriculture and Rural Development" Course Note 0761005. World Bank, FEnomic Development nute, Washington, D.C. Uphoff, Norman. 1991. "Fitting Projects to People' in Micbael Cernea (ditor) Pating People First: Sociological Variable in Rual Development. (Second Edition). New York. Oxford Univcrsity Press. Urban Development Division. 1989. FY8r9 Sector Revew of Urba Devdopment Operationsr Reaching the Poor dtrough Urban Operations. Wasbington, D.C.: World BantL 320 References Williams, Aubrey. 1990. 'A Growing Role For NGOs in Development" Finance and Deveopment, December. Wolf, Eric. 1965. 'Kinship, Friendship and Patron-Client Relations in Complex Societies." In Michael Banton, ed., 7he Social Anthropology of Complex Soceties. ASA Monographs 4. London: Tavistock. World Bank. 1988a. Targeted Programs for the Poor during SructuralAdjustment: A Swnmry of a Symposium on Poverty and Adjustment. Washington, D.C. . 1988b. The World Bank's Supportfor the Alleviaton of Poverty. Washington, D.C. . 1988c. 'PRural Development: World Bank Experience 1965-86.' Operations Evaluation Deatment, Washington, D.C. * 1989a. 'Chinat Women in Development: Country Assessment Paper." China epartment, Washington, D.C. 1989b. "Gendir and Poverty in India: Issues and Opportmities Concering Women in the Indian Ecomy: Summary Report." Population and Human Resources Department, Women in Development Division, Washington, D.C. 1990a. "Bangladesh: Strategy Paper on Women in Development' Population and Human Resocs Departme Report 7899-BD. Asia County Department 1, Wasington, D.C. . 1990b. World Development Report: Poverty. New York. Oxford University Prs. * 1992. Povety Reducton Handbook. Washington, D.C. 1 1992b. World Development Report: Development and the Environment. New York. Oxford University Press. Yahie, Abdullahi. 1993. The Design and Mnagement of Poverty Aeion Projects in Aflica. Evolving Guidelines Based on Experience. Economic Development Inaztte Worldng Paper. Washington. D.C.: World Bank stibutors of World Bank Publicaions AXONIWA fluMEdlIMtCIMomw UXVA SOW AI bOfAU CaImHIu*.S 4L5u0S A,9mkIn EM 5A rc&uL Gcs.a. Ca. Qawmne&eMbas OdUSnt1fl _sZ5,4|b6U8- ~ ~ ~ ~ ~ i DmIuU mes_a FAMm_VUW_ RNUN OK m DMM1911IC LUIIL41LFrJUDLU NSM .. m14 MU&mmm RAIsaN,JdSpps caps TkCepnNIS AUIBAUAJINUVVDKA PA41D H3 l 223G KOI.WUU AjznouNNUDh SCZ nla 0 t mnDtxmkCmpm hrdfrhawd. DAJuISnMMI&Mm MCAH So"t P.O. a"am Eu S.. at WaS l G_kA 1i1u.inSm I Qa.ndlis MAALAVU LExMM CREANY P-amdhPmSln N s. SAwt. CwMrdQC6 UNO?Vkg mm Khla Ib37 Cb060M :S5*IdcmrA8 - JWtld ANUaDES HONG KONOJEACAO Ap:nd.PmF0lJI G IC..t3K ~.hiI.D.'Sup.a AiMSXDUID IS. 1418 fl.~uOL 1OPT. USp _r AamSedqCDAM 445WySmUISS hUmSl VgSIN W _ HgaAIb SILANKAANDTIEMIWVM CbasS it/Ann 2511ra 0 -.wbINwwm twSNcm imflp Chit7/ I CmaINhqSXaa P.CDM123 FI Dm261 7SA msS.aI.rOt.INIpAm. rdY , III.UI_ AIDS iaam ftbLUL NEWZLALND C1_IM2 luatCamSAn hbslh_S EBSOONZUL CIWprElCS dai-SDU PtinMdIqSSIC - S N-wl For ghS 74LDAa 4w CAuIA - m I-kdf fI4U D5 1..1mm%MS RSIfltfta W Eesi*UN Maui RifS27lEd lpUUk WUtsud. N.WMb-fOwm ted IMU.OIDmI 541 235am. CANAIDA ICIdUIjmAWfII NOEVAAY IaDtsm cakomS-7111111 Nanm.IUISn&BCad 5WAIflhAND CU.S3Ir.A.mpke leak dFwrhoaAk Umdmi.uIaoti Japum Mmii dCq P.O. Ioic6ISRAd Iabal Jean SOiNSbS.Gasdbluqm N4iEoom CHampLhf Cm= _A_TAN b"VeWrsA. ssraaaa naABwyFrdwimdw nSM Cwpa.pasbDbSh Oma&inm i&au&WhJDSD.*YBUE PEUU ftMMawqmm~ Ahnbbd'h-UOin EidfaMDsuibMA 1NALANC aktDabDoaqJe ApuSeJCCADVfftt CWD Luaw- NWUUn S hk&WdnMZi..6 hIarIAdAgkCmIw U?1INDAD&TCUASU.ANuIMIA pDL rd buw rOhUMC&CUDMDAMUYANA, cOflWWOI33 coaflaSa.d inZ&NEVIS.SLLUA, Ahmi.t. Wl RISuDai Sy Csjna - in tiIsmasMs POLAN B_-5611ULWUb22 - aiW_iP cmu Sc,Ww 3m --.HX_CvmsSupsAa .b-30 '.fsb KLW ADblqm Vg D5Hbtmman Su CAN" PADmSh '.3.2 UL1arn3 3lamDi. cvnm UMLAM~Mwwuaw.m -ux EMAS ~~~~~~Yonuumftlibag POKIIICAI.iNlDS S.sdnd.U PA iUs LhuiaftaiZ b.msumfhll TdAvwdN RJa DCa7074Dm WTALYT DOImCANIWJC . Lba.CaasmmaSwA SAUM AZAlmA. QATS EdtI.flkCpwA. Rb Us. cIOaktvl S_fairtk5 IndIA U 3mbtImmelmhdhetmS coSa MSu P.O. MN335 LbWsmai Ni. ApiraCs_nZ rSlSt Elyit 21AS.. JAN UNICA)OSATAIWAN, A!ab. HonpSCb-w SW4MfI13 lafeuh. SUd _ican Tkyao VitaLS. cdm ~~~~~~~~~~~coMm WNS leSs EL Film"mmiios 0 c - -.slnn The World Bank f"Ok", 14 e 'z 41,4 7V va4 ons OXLe,- e, 62' 4, -o j3:, 45'W RLID56M q_ 41 CY 'In e t 'y c m r 71 Fl Zi O',f6' m e 6d'a IT6 tY 0(3 4v zr , __'t,47 520/009 E9189' ISBN 0-8213-2472-1