DISCUSSION PAPER Report No.: UDD-36 PARTICIPANT - OBSERVER EVALUATION OF URBAN PRGJECTS IN LA PAZ, BOLIVIA AND GUAYAQUIL, ECUADOR by Lawrence F. Salmen November 1983 Water Supply and Urban Development Department Operations Policy Staff The World Bank The views presented here are those of the author, and they should not be interpreted as reflecting those of the World Bank. The author, Lawrence F. Salmen, is an urban planner who has also worked as an administrator of development programs and engaged in anthro- pological field methods. He has been a consultant to the World Bank, attached to the Latin America and Caribbean Regional office and the Water Supply and Urban Development Department, since January, 1982. Research Title: "Participant-Observer Evaluation of Urban Projects." Number: 672-5§. ABSTRACT The author conducted evaluative research by living with project beneficiaries for 14 months, divided equally between a low-income settlement being upgraded in La Paz, Bolivia and a new low-cost housing project in Guayaquil, Ecuador. In addition, structured interviews were held with artisans and home improvers who had taken out loans from Bank-supported credit programs in each city. As a result of this participant-observer qualitative evaluation, a ntmber of findings emerged which should help both project per- formance and design. These include: the importance of project implementors promoting a project not only with community leaders but with the general popu- lation to be affected as well; the far greater importance placed on delays in project execution by beneficiaries as compared to administrators of projects; the identity which residents of lower-income settlements feel with their neighborhood which may be seen in the provision of physical infrastructure to an area; and the differing benefits received by renters and home-owners in an upgrading area. Each of the projects being reviewed, in La Paz and Ecuador, has already been modified as a result of feedback provided by this evalu- ation. Furthermore, project managers have become sufficiently convirnced of the utility of the approach that local participant observers are being con- tracted to continue this kind of evaluation in each city. Table of Contents Acknowledgements ...................... i .................. i I. Introduction....... .... 0e*.e.. . . . . - . 1 II. Projects .... i..i ................ .... 3 Ai The Setting......................... . . .i . 3 B. Findings ...................................... 5 III. Upgrading.. .................... . .. .... . i . . i. . . 7 Ai Communication ...... ........................ 7 B. Catalytic Effects. ., .... .... i .......... 9 C. The Community ...................... ................ 10 Di Intangible Benefits ........................11 IV. Credit Programs................. i........................ 12 A. Artisan Credit., . .......... . ....... ......... 13 B. Home Improvement ................................... .14 V. New Housing .................. ...... 15 Ai The Before Situation.......... .................... 16 BX The House..a.. . . ... ... d a a ?, . .. . ...... 16 C. The Settlement . . .................... ...... 17 VI Conclusion.; .. ........ .... . . . . . 23 A. The Approach. .................. > O.. 23 B. The Projects. ... a . a.............i* 24. . a . . . . ...... . . Annex: Methodology.. ........................................... . 27 A. The Approach ...... ...... .............* 27 B. Practice............................i.............. 28 C. Replicability .............. ......... ...... s...... 30 ACKNOWLEDGMENTS This report is the product or many individuals and organizations. The idea to do an evaluation of urban projects using anthropological, participant-observer methods originated with David Beckmann, in close collaboration with Harold Dunkerley, of the World Bank's Projects Department, Latin America and Caribbean (LCP) and Urban Development Department, respectively. Funding for the research was provided by the World Bank's Research Committee (2/3) and by the Governments of Bolivia and Ecuador (1/6) each). Supervision was carried out by David Beckmann and Neil Boyle (LCP). Work was guided at the World Bank by a Steering Committee chaired by Gregory Ingram and consisting of Harold Dunkerley, Ian Scott, Maritta Koch-Weser and Michael Bamberger. Mahmood Ayub, the Bank's Representative in La Paz was generous with office space and administrative support. Many other individuals from a number of offices and departments, including Agriculture and Rural Development, the Economic Development Institute, the office of the Regional Vice-President for Latin America and the Caribbean, Programs and -Projects Departments from various regions, and the Operations Evaluation Staff, all gave encouragement and support. The sponsor institutions in Bolivia and Ecuador which allowed themselves to be exposed to this intensive look by an alien body were most cooperative at every step of the way. Key individuals were: in Bolivia, Jaime Medrano and Freddy Quiton of HAM-BIRF, Gustavo Medeiros of BISA, and Sammy Vera of BANVI; and in.Ecuador, Enrique Blacio and Sandra Luque of BEV, Jaime Vera and Nelly Valarezo of the Municipality of Guayaquil and Alexis de Aguilar and Rafael Cuesta of the Banco del Pacifico. This was a learning experience for all of us. A number of individuals, from Latin America and the United States, were kind enough to read and comment on the major reports as they came out, thus contributing to this research as it evolved. These were all experts in urban development with particular interest in applying research effectively to action. They include: from Brazil, Vilma Faria (CEBRAP), Carlos Nelson (IBAM), Lenina Pomeranz (Government of Sao Paulo) and Ephim Schluger (IPLAN-Rio); from Mexico, Maria Elena Colcero and Leonel Lechuga (University of Mexico); and from Peru, Jose Matos Mar. In the United States, the primary interlocutors were Anthony Leeds (Boston-University) and Lisa Peattie (HIT). Special recognition is due to my research assistants, Yolanda Barriga and Luisa Fernanda Rada in La Paz and Hilda Sanchez de Gaviria in Guayaquil. To them the baton is passed for successive research efforts similar to that described here in Bolivia and Ecuador. It is conventional in social research to end acknowledgments with the people in the comminities under study. In this case, these people were really the beginning. This is their story. I thank them for sharing so many moments with me so that I may tell it here. I. INTRODUCTION Since 1974, the World Bank has innovated with strategies to reach the urban poor, notably sites-and-services, slum upgrading, and credit for small enterprises. These have increasingly been adopted by many developing countries and by other development agencies. Large-scale evaluatiorts of these strategies with sophisticated statistical analyses have been conducted on a number of urban projects. Learning from these evaluations has been incorporated in the design of new projects and in policy formulation. As the Bank seeks to refine and broaden learning about projects, there is concern regarding the cost of these evaluations and the nature of the information collected by them, based, as it has been, almost solely on extensive questionnaire surveys. There is felt to be a range of issues, from beneficiary attitudes towards project-induced changes, to communication between beneficiaries and implementing agencies, to the effect of local power structures on the distribution of benefits in a community, which may best lend themselves to more intensive an-hropological methods of inquiry. The Bank's intention at the time of func:;.g this research was not to supplant but rather to complement the existing quantitative impact evaluations. The Bank was interested in gaining additional understanding about the two urban projects selected. It also wished to experiment with the method itself. The Bank sees the potential for this qualitative, participant-observer approach to project evaluation to provide timely, directed and cogent feedback to project managers that can help provide guidance' for ongoing project activities and lead to improved design of subsequent projects; at relatively low cost. The research reported on in this document followed an approach which is new to the World Bank. The purpose of this inquiry has been to learn enough about the perspective of the beneficiaries to provide feedback to project managers that would help them improve project performance. Methods and findings were thus fashioned to meet the practical concerns and needs of decision makers rather than to provide evldence for or against abstract models or hypotheses. The methods employed reflected the importance given to understanding human behavior within the dynamic context of development project's, rather than to collecting objectively verifiable and static data. The focus of this kind of research is on the process of project-induced change as experienced at the level, of the barrio, or community of beneficiaries. (The first person is used throughout this report as a natural expression of this methodology). Qualitative evaluation derives from a paradigm which stresses the meaning people give to their actions and hence recognizes a degree of unpredictability inxhuman behaviour. Within this paradigm, the concrete context in which a person or group acts is seen to have a critical bea-ring on the nature of those acts. This intellectual framework, with its contextual orientation and stress on the meaning people make of what they do, treats actions differently than those schools of thought which look at human behaviour whithin frameworks established in the natural sciences. Here people, like plants or inanimate matter, are assumed to act in accordance with laws which govern their behaviour. Actions can be explained when they are seen to follow such laws and the laws are confirmed by statistically verifiable evidence. Much evaluation work has followed this latter school, testing certain a priori assumptions, often with quantitative measures, at some distance from the object being observed. 2 - Qualitative evaluation resists predetermined hypotheses and seeks both openness and proximity to the particularities of a given situation. The qualitative evaluator needs to know that situation before he or she can describe how people act within it. Since good interpretation depends on understanding the meanings people assign to their own behavior and give to what they experience, the concept of verstehen, or unders.anding, has been a key idea underlying qualitative evaluation. The researcher needs to participate in the social world of those he studies deeply enough to understand their situation from an "insider's" perspective. "The tradition of verstehen or understanding places emphasis on the human capacity to know and understand others through sympathetic introspection and reflection from detailed description and observation."1/ (See Annex for more detailed description of methodology and its potential for replicability). This report is an operational summary of four major reports prepared after each of my stays of roughly three to four months duration which alternated between the two cities under review, La Paz, Bolivia and Guayaquil, Ecuador.2/ The second phase of this research, commencing in October, 1983, is devoted to two activities, replicating this evaluation approach in two additional countries, Brazil and Thailand, each of which has expressed interest in it, using local personnel, and producing a monograph which will expand on the findings reported here and on the methodology, including its replicability, with greater description of the human dimension from which both findings and method evolve. This report does not exhaustively evaluate the two urban projects reviewed. Rather, it highlights aspects of these projects which lend themselves to action, directly through implementation or indirectly throuigh policy formation. The perspective from the barrio followed by this research is not proposed in order to exclude other perspectives but to provide one essential ingredient of decision making. Finally, while participant-observer, qualitative evaluation provides insights for contextual differentiation by project site, it also serves as a basis for inductive generalizations about new development programs. 1/ Michael Quinn Patton, Qualitative Evaluation Methods, Sage Publications, Beverly Hills, 1980, p.45. 2/ Copies of these reports are available from the office of Neil Boyle. (LCPUR), World Bank,-A-835, (Telephone 477-2333). 3- II. THE PROJECTS The Projects which were the object of this study were far more alike than the cities in which they took place. Both the Bolivia and the Ecuador projects (Loans 1489-BO for US$17.0 million and 1776-EC for US$31 million respectively) were designed to support a variety of ways to better conditions for the urban poor: low-cost expandable-core housing, slum upgrading, and credit for artisans and small-scale enterprises. There were differences as well: in La Paz the Bank loan also financed a number of retail markets and a health outreach service for mothers and infants; in Guayaquil, the Bank project included a large number of home improvement loans in poor areas (see Table). In both cities, however, the projects were supporting what were at the time innovative urban development concepts, and the effort was as much to orient and develop institutions in these new approaches as to create particular tangible products. While a total of eight implementing agencies participated in these two projects, only two were intimately involved with this participant-observer evaluation throughout the duration of their execution, the Municipality of La Paz (RAM) and the Ecuadorian Housing Bank (BEV.). While other agencies were interested in the findings of this study, particularly the Municipality of Guayaquil regarding its arduous experience in attempting to upgrade Guasmo Norte (see section .III.B,. below), HAM and BEV were the host agencies for my two places of residence and as such were naturally most attentive to this research. A. The Setting It would be difficult to pick two major cities of Latin America more dissimilar than La Paz, Bolivia and Guayaquil, Ecuador. The former is a picturesque capital city located in the high altiplano of the Andes, the latter an unpreposessing tropical port city bordered on most sides by marshes and swampy estuaries. The importance of La Paz is largely political, as the administrative center of the Government of Bolivia. Guayaquil's stature is largely due to its economic importance as the commercial and financial center for Ecuador. Both cities are the major urban places in their respective countries, La Paz at roughly 800,000, two thirds the size of Guayaquil (estimated at 1,200,000). What La Paz and Guayaquil have in common is shared by most major urban places of the developing world. Both cities have been growing rapidly, at rates of 3.5% (La Paz) and 5% (Guayaquil) annually over the last twenty years. Much of each city's population, roughly a third, is made up of in7-migrants, many of whom, especially in La Paz, retain close ties to family members who have remained in the countryside. The influx of migrants was higher than normal from late 1982 to mid-1983 due to the very adverse climatic conditions in central and eastern South America which caused unprecedented drought in Bolivia and floods in Ecuador. These desperate newcomers have joined tae large numbers of persons in both cities who live in poorly-serviced, unsanitary accomodations. Rough estimates place slightly over a third of La Paz's population and at least half of.the inhabitants of Guayaquil in houses lacking water and/or sewerage connections. -4- TABLE: PROJECT COMPONENTS, LA PAZ AND GUAYAQUIL Implementing Credit Cost Agency Upgrading New Housig Progia Health US$mill: La Paz* (1489-BO) Municipality about 7,000 Artisan and 8.2 (HAM) families Home Improvements CONAVI 2a000 7.6 residential (and 6 hectares industrial sites) BANVI 400 sites 2.3 BISA Artisan (and 5.2 10 hectares industrial s ites) Ministry of child-care 1.2 Health program SUB-TOTAL 2405 Guayaquil** Municipality about 3,400 8.4 (1776-EC) families about 3,200 6.9 sites about 6,400 8.3 home improvement. loans, upgrading and site and service BEV 504 units 1.6 3,000 home 3.9 improvements loans in existing low- income settlements Banco del about 6,400 9.0 Pacifico loans to 5,200 artisans SUB-TOTAL 38.6 TOTAL 62.9 * As amended, February 5, 1981. ** From Staff Appraisal Report (2415-EC), November 9, 1979. Total figure does not include contingencies (US$10.8 million) or costs of technical assistance, projects units and other amounts difficult to allocate by program (US$2.2 million). -5- The people of La Paz and Guayaquil also share another common characteristic; most earn little income. During the time of this study, from February 1982 to August 1983, these people have become evel. poorer. When I began this study in La Paz in the first quarter of 1982, median monthly household income was estimated at approximately 9,200 pesos.3/ That had risen, in nominal terms, to 16,000 pesos by mid-1983, an increase of 74%, but the consumer price index of Bolivia had gone up 300'O over the same period. Median income in Guayaquil rose, in nominal terms, from 7,600 sucres in early 1982 to 10,700 sucres in mid-1983, an increase of £1%, but the consumer price index of Ecuador went up 60% in the same period. The increase in hardship experienced by below-median income households in the 1982-83 period was apparent from living with the poor in both La Paz and Guayaquil. The worsening trade and foreign exchange situations, coupled with the already mentioned natural disasters or drought and flooding, caused a lowered demand for workers in industry and an increased supply of labor as people moved to the cities from non-producing agricultural areas. The great majority of my neighbors in poor areas of La Paz and Guayaquil suffered a noticeable decline in living standards over the 18 months of this study.. To cite a few cases: artisans faced shrinking markets and found it increasingly difficult to replenish their inventories; people ate meat less frequently; the number of persons, especially young adults, complaining about the difficulty of finding a job increased considerably during this period. Adverse economic conditions have particularly severe.repercussions among the poor, for whom any loss is major. B. Findings Learning in these eighteen months has been a cumulative process resulting from research at a number of project sites.in both La Paz and Guayaquil. A chronological summary of the research conducted, by time and place, is provided in the following chart. Research was facilitated by the fact that various project components were affecting people in the same general geographical area of each city. My places of residence, whether the upgrading area in La Paz or the new low-cost housing area in Guayaquil, were also the residential sites for artisans who borrowed credit from another project component. In Guayaquil, the new housing site was contiguous to the slum that was to be improved as part of the upgrading component, and neighbors had friends and relatives in both areas. Thus, proximity to beneficiaries of one component via place of residence also provided access to beneficiaries of other components given their nearby location. Interviews and surveys were drawn from and served to reinforce observations in all project components. The summary observations which follow are broken down by program area, upgrading, credit, and new housing. 3/ Income estimates are from census data (1976) for La Paz and household surveys conducted in Guayaquil by the Instituto Nacional de Estadis.tica (1975), adjusted for changes in the value of currency in the intervening time period and in the pattern of income earned, as revealed by .informal surveys by the author. -6- Place Time Project Component Method 1. La Paz, . Feb-June, 1982 upgrading participant-observation Bolivia (8 de Diciembre) sample surveys interviews artisan credit sample survey interviews new housing sample survey (Rio Seco) interviews (before situation) 2. Guayaquil, July-December, 1982 low-cost housing participant-observation Ecuador (Floresta) sample surveys (before and during interviews project) sample surveys artisan credit interviews upgrading (ex-post sample surveys in Guasmo Norte, interviews where project was rejected, and ex-ante in Guasmo Fertisa) home improvement sample surveys credit interviews 3. La Paz, February-May, 1983 upgrading participant-observatior Bolivia (8 de Diciembre) interviews sample surveys 4. Guayaquil, June-August, 1983 low-cost housing participant-observation Ecuador (Floresta) sample surveys interviews III. UPGRADING Four salient issues emerge-from living in and observing upgrading projects in La Paz and Guayaquil: the importance of communication between implementors and beneficiaries, the catalytic effects of upgrading, the integral nature of a community, and the intangible benefits of upgrading. A. Communication. Effective communication between implementors and beneficiaries during all phases of project implementation and reaching through local leaders directly to the affected general population is vital to project suczess. Local leadership plays an important role as an intermediary and communications link between implementors and beneficiaries. Where this leadership is progressive and appreciates the advantages which will come to an area with upgrading, as was the case in 8 de Diciembre and Guasmo Fertisa5/, in La Paz and Guayaquil, respectively, it can greatly aid project promotion in the community. Where the leadership does not favor the project, fearing that it will undercut its own authority, as was.the case in Guasmo Norte and San Antonio, in Guayaquil and La Paz, respectively, it may completely undermine any chance for project implementation or at least delay it for a considerable time. Even if the local leadership is well-disposed to an improvement project, and certainly where it is not so disposed, effective implementation should be carried out by promotion at all levels of a community. One should not assume, as many development agency personnel do, that dealing with only the local leaders is enough. Active involvement of the beneficiary population through the formation of block groups for both discussion and work purposes is key to project success. Where local leaders might resist acceptance of a project to retain power, they will perforce support it if they see it is desired by the people they represent. HAM-BIRF, the executing unit of the Municipality of La Paz, finally won acceptance of an upgrading project in San Antonio, after three years of intransigence by local leadership, by going around these leaders to more decentralized points of authority and establishing work groups at the block level for promotion and execution of project activity. The Guayaquil Municipality, which never circumvented the local leadership in Guasmo Norte nor formed any grassroots groupings of beneficiaries, ceased all efforts at upgrading this area after a similar three-year period, convinced that the people living in the area did not want the project. The problem, however, lay with the leaders who misrepresented the project to their followers due to their wish to obstruct any Governmental undertaking, which they saw both as jeopardizing their own hold on the population and as inimical to the interests of the radical political party6/ most of them followed. 5/ Project not carried out there due to Bank loan restructuring, in mid-1983. 6/ Movimiento Popular Democratico, a party of the far left very critical of Government. IBRO 1752CR NOVEMBER 1983 aOLIVIA City of La Paz URBAN PROJECT SITES Rio SECO- VILL ANOO A lL RA L ' ' lIBRAZIL | *\ BRAZ ,z8 DE DICIEMBRE a 0 L ICONAVI Now Housing LA PAZ L 16 A Seiecd Roads Rivers -s - Intornational Soundaries / ILOMETERS (HlLEX ,/E @-, A tAGUAY -,- CHILE N 'A BRrm N4 NA ,e_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ A further reason to disseminate the project widely and enlist rank and file support is to enhance the chances for active participation in complementary project improvements not directly supported by Bank loans but very much part of what the community perceives to be upgrading. The residents of 8 de Diciembre. desired two improvements not included in the project, paving the roads and constructing a community center, but they have failed to act on their own to obtain these improvements, at least partly due to HAM-BIRF's complete inattention to community organization. People need to be mobilized to act on their oxn behalf in the solution of common needs. Communication should not be a one time thing, but a continual two-way process throughout the life of a project. When there are delays in project execution, as there were in both projects and generally will be, communication about the nature of those delays and realistic projections as to their duration will do much to allay community suspicions born of long-standing mistrust of government authorities. The people undergoing area improvement are concerned about project delays; they will listen to and appreciate explanations of causes for delay. Similarly, the project implementing unit should listen to the area residents as they proceed with project execution. Were such listening to occur, matters of equity (such as just compensation for property lost as a result -of area improvements) or of a technical nature (such as the positioning of sewerage lines at heights which made access by households difficult) would be better attended to by executing staffs. To enhance this dialogue thought should be given to the establishment of a group of experts paid for by and reporting to the community being upgraded to serve as a counterweight and equalizer between community and executing agency. B. Catalytic Effects Upgrading was a major incentive for economic development;. physical improvements in a neighborhood, through the provision of infrastructure, gave the people moti°vation to improve their houses. These improvements were done largely with paid labor. Often added space was used for rental. With road widening and increased access a number of small commercial establishments and workshops opened up. The income and employment generated by these changes, all brought on by the upgrading, through not measured were clearly significant. In the area of 8 de Diciembre in La Paz, which received water and sewerage systems and street widening, two thirds of the families made major improvements to their houses in the first three years of project life. Three other similar low-income areas of La Paz7/ which did not experience upgrading were selected as control areas. Over the same time period the percentages of houses being improved in these areas were 16, 12, and 8 (an average of 12%). -The differences between a little over one tenth in these three communities and two thirds of households in 8 de Diciembre making improvements is due to the presence of the upgrading project in the latter area. People with land tenure and household connections to city water supply and sewerage systems invest in their houses, even during times 7/ Kantutani, San Antonio and Tacagua. of economic stress, such as Bolivia has experienced over the last two years. This same phenomenon has been reported for other cities of the developing world;8/ it was dramatically evident in La Paz. Improvements in housing as a result of upgrading or Qf home improvement loans are generally done, at least in part, by paid labor. About 90 percent of the houses undergoing improvement with project funds generated paid employment. Literal self-help construction, where home owners are the exclusive workers on home improvements, is extremely rare among project beneficiaries in low-income urban areas. In addition, many home improvements were additions for rental purposes which will generate additional income for the homeowners and increase the housing stock for the city. C. The Community Upgrading needs to be approached as a holistic endeavor. The community living in the area undergoing improvement includes all residents of the area, only some of whom may be direct beneficiaries of the project. The upgrading project will generally directly address only a fraction of the needs of an area. The presence of the project, however, will increase people's expectations regarding all needed improvements. The project may also cause resentment and discontent among families who live alongside the improvement area if these families lack and desire the improvements and no provision is made for an extension of project benefits to them. While an upgrading project cannot and should not address all the needs of all the people in and around an area selected for improvement, it behooves.planners and administrators to be aware of the integral nature of the environment they are to change. Where the project does not directly address particular needs, arrangements may be made to cover these either by encouraging the community itself to do so with its own labor and perhaps counterpart funds and planning for the timely involvement of other agencies and for extending project benefits to adjacent areas over time. By April 1983, over three years after project inception, about one third of the households in the sub-district of 8 de Diciembre had not received all of the benefits of the upgrading of the area, as may be seen in the following table: Recipients of Benefits of Upgrading 8 de Diciembre, La Paz Group Benefitted _ Not benefitted % homeowners 180 90 20 10 renters 85 42 115 58 TOTAL 265 66.3 135 33.8 8/ Paul Strassman, The Transformation of Urban HoujinE; the Experience of Upgrading in Cartagena, The Johns Hopkins University Press, Baltimore, 1982, p.7,153. Difficulties were encountered in the provision of water and sewerage lines for roughly 20 households located at the edges of rivers. The major disadvantaged group, however, consisted of renters; there were just as many renters as homeowners (200 each), but 58% of the renters had not been provided access to water and/or sewerage connections. Inasmuch as the public water taps used previously by all residents had been removed with the installation of water lines, these renters were now worse off than before, having to either request or purchase water from a nearby owner or walk a greater distance to an adjoining uaimproved area where standpipes were the only source of water. On the basis of this first upgrading project in La Paz, HAM-BIRF has obliged all property owners in subsequent upgrading areas to make household connections to the facilities provided by the project, water and/or sewerage, whether the property is owner-occupied or rented. Whatever increases in rent result from providing renters connections to infrastructure will be offset, though admittedly not directly compensated for, by the improvement in sanitation due to all residents utilizing water and sewerage systems. Other aspects of community upgrading vary in importance according to one's perspective. While living in 8 de Diciembre, I became convinced of the importance of the fact that the project was not extended to roughly 400 households who lived in two unserviced areas adjacent to the project .site. I also shared the feelings of most of my neighbors and others in the community that certain improvements not included in the upgrading project should be made for the area to really be considered "improved". These included channelling a filthv and unruly river which bordered the area, paving roads, and providing street lights. Each of these betterments has a cost, which may be difficult to recover. More important, the public financing needed for these improvements was simply unavailable in Bolivia given the very difficult economic crisis through which the country is passing. From the perspective of Washington or even of the executing agency's office, it seems obvious that the project was neVer intended to channel the nearby river or respond to the petitions from neighboring areas. But the view from the barrio is broader than that from above, or outside. If improvements not directly financed by the project are encouraged by writing them into the project agreement as Government counterpart obligations or by organizing the community to assist them to make these improveme-tts themselves, a more comprehensive and appreciable upgrading process could result. If provision can be made for extending project benefits over time to contiguous areas needing and willing to pay for them, a desirable spread effect will occur and friction between haves and have-nots will be avoided. D. Intangible Benefits Improvements generated by urban upgrading transcend the purely physical or economic. While this kind of summary, final report is not the appropriate type of document for elaborating this point, and precise measurement of non-physical non-economic matters is often impossible, the importance of such issues is not diminished because they cannot be quantified. -12- Clearly, people in the upgrading area of 8 de Diciembre felt better about themselves because of the improvements taking place in their community. Where one lives has an impact on how one is perceived by others as well as oneself. As the people of 8 de Diciembre received water and sewerage and their streets were widened to allow vehicular passage, people felt they were no longer in a marginal, poor part of the city. Whereas before they had worn their work clothes to work, now they wore better clothes as they left their homes and changed at their place of work. Where they may have been ashamed of inviting more affluent persons to their neighborhood, now they do so with pride. Most important, at least some residents have seen what improvements can be made with concerted effort and now want to improve other areas of their life, such as organizing by-block to pave the roads and forming a consumer cooperative to lower the prices of groceries. Despite a rise in land values in 8 de Diciembre of over 250% in real terms over the last three years, not one homeowning family has moved out of the area; a stable and permanent community has been created. The development process, in one small area of La Paz, has begun and -- with or without active promotion and despite very real and large obstacles of a macro-economic nature -- it will go on. IV. CREDIT PROGRAMS Of all the major types of programs I observed in this research - upgrading, credit, and new housing -- the credit programs were the most problem-free. This was to be expected; credit represents the least intervention of these three types of urban development. There is no major change in an existing low-income community as in upgrading; nor is there an attempt to create an entirely new urban environment as in new housing projects. Rather, credit programs have the humble and laudable goal of providing a scarce resource, capital, to support and further the energies of an already motivated and directed individual or enterprise. The one salient critical finding which come out of close observation of artisan and home improvement credit programs in both La Paz and Guayaquil was the importance of the lending agency knowing the world of the borrower, or client. This often called for extension workers and the actual location of offices in the poor neighborhoods of the city. For a lending institution to reach and respond to the needs of the poor it must understand and deal effectively in the environment where the poor live. Not being an artisan or a home improver, I was not literally a participant-observer in credit assistance. Not being an institutional evaluator, at this time, I was not in a position to delve very much into the inner workings and decision-making of the banks and municipal agency which ran the credit programs under review. Instead, I employed intensive interview techniques and, together with two or three research assistants in each city, sought out samples of artisans in both cities and home improvers in Guayaquil, selected at random but stratified by location to be representative. The objective of these surveys was to determine how helpful the credit had been, what changes it had brought about, what kind of persons had taken advantage of the programs, and what issues might be addressed to better these programs in the future. Due to inherent limitations of the research methodology, this evaluation does not assess the cost effectiveness of credit programs as a policy of development assistance. -13- A. Artisan Credit Evaluation of artisan credit programs assumed a more tentative and rudimentary character in La Paz than in Guayaquil, the former place serving as a testing ground for the latter. Sample sizes were small in both places. Survey findings confirmed what is generally expected from artisan credit programs. Artisans have increased both earnings and the number of employees. The lifetimes of loans received varied from six months to four years. Based on very general estimates, by the artisans themselves, earnings went up by two fifths to three fourths while they had loans. Corresponding average increases in employment were from 25 to 51 percent, for loans of six months and four years, respectively. This program provided a major economic boost at the level of the micro-enterprise (normally between two and four workers) which was otherwise not available to them, given the scarcity of capital accessible at even near market rates to artisans in either of the two cities studied, Beyond this general verification of expected benefits, qualitative evaluation is not particularly suited to the assessment of artisan credit programs, at the level of the benefi4iary. Quantitative and institutional evaluation are needed to address the key issues here. However, in the areas of technical assistance and agency-borrower relations, the qualitative approach is appropriate. Technical assistance is highly valued by artisans and apparently helps their productivity, yet its delivery was incomplete in degree and kind in both La Paz and Guayaquil. Clear majorities of artisans in each city wanted technical assistance, especially directed to the production and marketing of their particular industrial activity (most were tailors, shoemakers, woodworkers and seamstresses). Yet most artisans in both cities received no technical assistance at all. The minority in Guayaquil that did receive technical assistance reported earning profits which were roughly 40% more than artisans who did not receive technical assistance. The training institution in Ecuador that was to provide this service, CENAPIA, had reached less than half of the borrowers who wanted technical assistance. Courses provided had not been technical enough and had been accompanied by almost no follow-up visits. Problems of coordination between CENAPIA and the Banco del Pacifico were, in mid-1982, at the time of this study, insurmountable. Attention needs to be given to the administration of both credit and technical assistance by the same general authority, much as has been done by the Bedford-Stuyvesant Restoration Corporation for a low income area of New York City. A successful artisan credit program requires strong commitment from the implementing agency. The artisan is a fledgling entrepreneur without much education or training in business practices. An agency that lends credit to artisans and does not visit them at least once every six months, as was the case with BISA in La Paz, has difficulty maintaining good relations with clients and understanding their evolving credit needs. The world of the artisan in a city of a developing nation is far different from the world of the formal banking institution in that same city. With this in mind, the Banco del Pacifico had established a special unit of urban extension workers who became familiar with particular low-income areas of the city and served to both promote and collect credit, a method -14- of operations that artisans unanimously reported to be most effective, which lowered the rate of arrears from approximately 20 to under 3 percent0 This system, together with well-designed and coordinated technical assistance, could serve as a good model for similar artisan credit programs elsewhere. Credit appears to significantly improve the performance of artisan bdsinesses. The most effective loan operations require a thorough understanding of the clientele population accompanied by technical assistance geared to the needs of artisans. B. Home Improvement The only major home improvement credit program in these two urban projects9/ was one administered by the Banco Ecuatoriano de la Vivienda (BEV) targeted at low-income househol.ds in areas of Guayaquil settled by poor persons. Loans were for 10 for 15 years. Loan limits were US$1,000 for property not secured by mortgages and $1,800 for secured property. The sample survey represented M5% of the 300 borrowers holding loans for sufficient time to have ;_impleted their improvements (from three to ten months). Three salient general obeervations emerge from the survey: 1. As with upgrading and irtisan credit, home improvement loans catalyze the mobilization of other resources, both material and human., Bank-supported lbans paid for an average of slightly less than 30% of the total outlay for improvements. The other monies used for improvements were largely (about 80%) generated from personal savings. Two thirds of borrowed funds were used to pay costs of labor by the 90% of the sample that used paid workers in the construction on their homes. 2. Again, as with artisan credit (and upgrading) home imprbvement credit needs to be actively promoted if it is to both reach the poor and be executed on a cost-recoverable basis, i.e. keep arrears to a manageable minimum. This home improvement loan program reached people having average household monthly incomes of about one and a half times what they had reported to BEV upon application (generally less than a year earlier than the survey), and well over (112%) the upper income limit for program eligibility (after discounting for intervening changes in the consumer price index). Not surprisingly, the better off persons in lower income areas designated for this program were the first to seek and obtain credit; they were familiar with banking and knew how to fill out forms so as to be eligible. Of additional concern, roughly 30% of the borrowers were in arrears in late 1982,. As noted in the case.of artisan credit, there is a real need for effective outreach into lower-income areas to properly administer credit programs destined for the poor. This clientele is unfamiliar with banking practices and needs to understand them before-they enter into binding agreements with what appears to them to be (and has, often been) alien entities. Similarly, to reduce the rate of default on loans and assure timely payments, good and thorough collection procedures are recommended. 9/ In La Paz, HAM-BIRF administered a loan program as part of its upgrading which had limits of $300 and was generally used to rebuild walls torn down to make way for road widening. 3. Reacal is an end-use of a home improvement loan program which is to be encouraged. Close to 30% of the borrowers of home improvement credit reported that they intended to rent part or all of the additional space they were constructing with loan funds. These aspiring landlord households earned roughly two thirds as much as those sampled who had no intention of renting. Renting is of course an increment to income and as such increasingly important in urban economies with growing unemployment and declining purchasing power of money earned. In addition, encouraging a small number of rental units in owner-occupied dwellings is one way to maintain sanitation standards and living conditions generally. One or two tenants in an owner-occupied residence receive far better services and return on their monthly "investment" in shelter (their rent payment) than rentets in tenements or absentee landlord-owned premises. The former, more personalized situation allows renters to seek desired services from their landlors who, being co-residents and neighbors as well, are apt to be more responsive to the tenant's demands than an absentee landord. As long as the supply of land and housing is limited, some rental is inevitable; one way to improve conditions for tenants is to encourage dispersal of rental units in owner-occupied and, hence, better managed dwellings. V. NEW HOUSING In retrospect, I find the new housing programs observed to be the most difficult to assess. From a number of surveys of residents of new housing projects and of people selected for future occupancy of such housing and from four months of residency in one such area (Floresta in Guayaquil), I find new low-cost housing somewhat an anomaly. Clearly, housing is a most desirable commodity for lower income as for other persons. Home ownership is particularly valued, not Just by middle-class families but by those earning below-median incomes as well. Yet housing is an expensive commodity rarely affordable much below the median. The attempt to put new, completed or semi-completed residential units together with low-income persons appears to produce distortions. The low incope resident of this housing cannot afford the kind of additions and improvements that he or she wants and needs for minimal comfort, and the house itself becomes increasingly marginal on a scale of priorities for families confronting an increasingly constrained job market and the rapidly falling purchasing power of their already meager incomes. New, completed housing units of an acceptable size are not affordable by families much below the median income levels of La Paz and Guayaquil. Even at about the fortieth percentile families can afford a small core but not credit to increase house size to adequately meet their needs. Middle income persons who can afford such housing may be legitimate targets of development programs inasmuch as the formal housing market -is priced beyond their reach. However, if the objective is to improve the .living environment of the urban poor and recover project costs, emphasis must be on the betterment of existing neighborhoods through upgrading, including the extensive provision of credit for home improvement, particularly directed to the formation of rental accomodations, and minimal sites and services solutions for families willing to live in such rudimentary housing. A. The Before Situation In both La Paz and Guayaquil the primary reason people wanted to move to the new, Bank-supported housing was to become homeowners. A secondary but still important reason for this move was to have increased space and comfort, if not immediately, then through home expansion. The overriding desire was to leave rental units, many of which, especially in La Paz, were small and overcrowded and to gain access to an asset (via a 20-year mortgage) which represented the first major capital acquisition for these below-median families. B. The House The feature which first struck the eye of an observer of both the houses of the CONAVI-built Rio Seco and the BEV-built Floresta projects, in La Paz and Guayaquil, respectively, was their small size, Houses in La Paz were one room plus bathroom structures of 18 square meters; those of Caayaquil had a common living-room/kitchen/dining area plus two bedrooms and a bathroom, all in 27 square meters. Inasmuch as people did not move into the houses in Rio Seco during the time of this research, my assistants and I could only interview people as they visited the new site. As foreseen in the project design, almost all said they. planned to increase the size of the house. In Floresta, the low-cost housing project of about 2,700 units where I lived, over half (57%) of the households had added to the size of their houses after living in them for an average of approximately 15 months. As might be expected, households making improvements were higher income than those not doing so, as seen from short sample surveys done in November 1982 and again (with different respondents) in July 1983: Floresta Households with Improvements by Average Monthly Income at Two Points in Time (in Sucres) November 1982 July 1983 With Improvements 11,242 11,219 Without Improvements 8,344 9,778 Thus, families who were at about the 40th percentile on the income distribution curve for Guayaquil could not afford to increase their cramped quarters while those slightly above the median could do so. Housing in Floresta, done for the most part by a contractor, was not only small but poorly constructed. The worst problem was the lack of ventilation, a major cause of discomfort in the hot, tropical climate of Guayaquil. I am convinced that had-the architects who designed these small houses lived in them even for only two or three days time they would have made sure the windows opened more than a few inches, the ceilings were higher, more openings were created in the walls for cross-ventilation, etc. Another problem symptomatic of slipshod construction was that the walls of adjoining residential units did not always go all the way to the ceiling, leaving open spaces of a few inches through which conversation and other noises of one's neighbors could be heard. The need for house design adapted to local climatic conditions and sound supervision for quality control of construction may be obvious but, at least in the case of Floresta, Guayaquil, was not heeded. The BEV implementing unit supported by the World Bank, BEV-BIRF, has, however, been most mindful of observations made during this evaluation and has accordingly made modifications in size, ventilation, and building materials. Perhaps the major change which BEV made in its new housing program as a result of this qualitative evaluation was in the "marketing" and financing of niso-techo units. These were semi-finished structures (two walis, no internal divisions) built as a way to reach lower-income families who could not readily afford the completed units. BEV and World Bank planners assumed that the'lower income families would move into the partially-completed buildings very soon after taking possesion and finish them with inexpensive materials over time, improving the construction as income permitted. Officials were chagrined to find that five months after the time of closing on these piso-techo units, close to half (47%) of the 104 dwe-llings remained unoccupied (as compared to 20% of the completed units). In a series of lengthy interviews with a representative sample of piso-techo buyers, both in Floresta and, for those not yet moved, in their -former and continuing place of residence, it become clear that these families did not want to live in make-shift quarters but preferred to continue paying rent and accumulate sufficient capital to improve their houses at a level commensurate with or better than the completed units which were, after all, the homes of their neighbors. The high level of aspirations and the importance of peer pressure were neglected as factors influencing decision-making by project designers in BEV and the Bank who were apparently unfamiliar with, beneficiary attitudes. In the revised project, Sauces VI, where 2,000 piso-techo units are being planned, higher income families will be sought and loans provided to them to facilitate the construction o.f the dwellings. This will solve the problem of immediate occupancy but leaves unresolved the question of how to provide low income families with new housing.1/ C. The Settlement .Similar to an upgrading area, a new housing settlement is more than the sum of the individual project-financed units. A complex living environment is formed which requires the attention of the implementing agency, not only during the construction phase but during at least the. first few years after occupation. Residents need to be organized. Concerns not directly related to housing need to be addressed. Both the site and the residents will be affected by adjacent, especially more needy, 10/ BEV officials state they are targeting this new piso-techo project at families earning 8,500 sucres a month, placing them below the 40th percentile. Given the average income of buyers of Floresta piso-techo units at 8,300 sucres in mid-1982 and a rise in the consumer price index of roughly 60% in the intervening time period, BEV will have to increase the income of buyers *for these units considerably. -18- ,,;8RD 17519R NOVEMBER :,913 ECUADORl,. i City of GuJayaquil s URBAN PROJECT SITES' sqato S \tA R PC2 r @ ,\ I /I,L, !IIIIIII BE No*:usn -* Internctionscl 80undarios L b ,, N o 0.5 1.0 1.5 2.0 IV\+ '"\> KILtOME RSt ECUADOR ... : :,,,..::. ...,::.. :.:.... ... %. .:. . . .:,- |:. - - : : ....... ....... .... ... . -. G U A Y A Q U IL I',, '.:' ':,:',:', ':': .:':,:':' .... ',: : '' .:''', :, .: ' r..... .... . ...< ....7.s- 9 *~~~~... .... ..... /f;S.j. X,,''-..,,,,,..*: .......... ..... .. ...$ r. . . . . . . . . .K . . . . . :1' \: 54 M $. \ . . . . . . . . . . . . . . 19 - residential areas. While shelter projects must be kept simple for effective delivery and operation, they need to be planned with linkages to services and income-generating activities to have the desired development effects. The area of Floresta is one of three low-cost housing projects built in Guayaquil from 1979 to 1981 by BEV. The people who live in Floresta are generally pleased with the physical lay-out of the area. In particular, they appreciate the calm of the site, removed as it is from the city with far more space given to pedestrian than to vehicular traffic. This much-lauded tranquilidad contrasts most favorably with the hectic pace and loud noises of downtown Guayaquil. The people's complaints are more directed at the maintenance of the area than at its design. As was the case with upgrading in La Paz, the degree and kind of dialogue between beneficiaries and implementing agency staff could be considerably enhanced and similarly, greater attention could be given to issues that are tangential to the project. It is one thing to build a new housing settlement, another to maintain it. The fact that BEV appears to pay little attention to maintenance concerns the people of Floresta because BEV does not pass the area over to municipal authorities until an initial unspecified period of time of at least three years. The maintenance problems are inadequate garbage collection, major deterioration of road pavements, poor drainage-of common areas, and clogging of sewerage lines. Drainage and clogging could have been alleviated in the construction phase by allowing for better run-off of surfaces and installing larger mains. Now, the issue generally is keeping what is in place clean, dry and in operating order. While the extremely heavy and frequent rains from late 1982 through June of 1983 have exacerbated this maintenance problem, destroying roads throughout southern Ecuador and causing drainage problems in all parts of Guayaquil, there was reason to feel that BEV was not paying attention to this community of 2,700 families that it had created. One clear indicator of this neglect was that BEV did not send any representatives to general assemblies of Floresta residents, even when invited specifically to do so by the local elected council. BEV did not specify the price of Floresta houses in the purchase agreements, intending to wait until all the costs of the entire project could be known with certainty, and the price BEV is now demanding is more than double what was originally expected. Partly because of BEV's inattention to other concerns of the residents in its new housing projects in Guayaquil, the pricing issue is now in the hands of the elected municipal authorities and may well become an embarrassment for the central Government which has supervisory responsibility for BEV administration. 1. Community Organization Part of the problem of poor maintenance and much of the ignorance of BEV to major concerns of Floresta residents may be attributed to the very low level of organization of this population. BEV staff.haye done very little in the area of community organization. Even BEV-BIRF, 20 - which made a more concerted effort in this area than had been made previously,ll/ made the mistake of organizing residents by block previous to their moving into Floresta; once in Floresta, residents tended to associate along the pathway on which front doors opened, which typically ran betweea two adjacent blocks. By July, 1983 after an average of roughly one and a half years of occupancy, only about a third of the households in Floresta r. (non-BIRF) attended meetings with fellow residents, organized by pathways. Most of the sample residents did not attend because there were no meetings of the residents on their pathway, or the meetings that were held were felt to be a waste of time. Sound orientation by BEV staff about the procedures to follow in community meetings and clearly delineating the responsibilities of both BEV and residents, could help assure a degree of maintenance by the residents and enable them to speak with one Voice in a loud and clear manner that BEV would find difficult to ignore and might rather address before matters of dispute were taken to political arbitration. 2. Tangential Aspects, Beyond Housing Similar to HAM-BTRF in La Paz which concentrated primarily on the tangible products of upgrading, infrastructure and road widening, BEV in Guayaquil focused, albeit somewhat poorly, on the main product of its project, low-cost housing. Other areas of community life in both places received too little attention from implementing agencies. This was true regarding community organization; it was also true regarding other social and economic elements of the urban landscape beyond housing and infrastructure. In Floresta, there were a number of omissions of this sort: the one bus line, which, while not funded was certainly related to the project, did not provide large enough vehicles and/or frequent enough service to satisfy demand at peak commuting hours. People had to wait in line for as much as one hour for space on the bus in the morning. Despite attempts to bring in other bus lines the existing one appeared to have secured a monopoly through informal linkages with government authorities. The communal center in Floresta was built to include a health center, day care center, library and market area, but, in fact, it remains a series of empty shells, unserviced structures going to weed. Part of the problem may have been that the other government agencies who were supposed to staff these facilities had overcommitted themselves during project design and, due to the worsening economy and their own depleted budgets, they could not fulfill their obligations. The resulting presence of this unused and deteriorating group of buildings, from the viewpoint of the residents of this low cost housing area, added to the impression of neglect and abandonment by governmental authorities and deprived these residents of iI/ The BEV-BIRF implementing unit built the 510-unit subdivision of Floresta known as BIRF, which differs from the larger area only in having 104 piso-techo units and half of the other units of brick, rather than the normal cement block or poured-concrete slab construction used in the rest of Floresta. Floresta-BIRF began to be occupied in February 1983; most of the rest of Floresta was occupied over a year earlier. 21 - socioeconomic G-uport services that become particularly important in times of economic hardship such as-Ecuador, like so much of the developing world, is passing through during this time. Urban development projects for lower income persons in developing countries must go beyond housing. Where the project is new housing, rather than upgrading, this is all the more important as an entire new environment is being created, rather than the servicing of a pre-existing one. In the former kind of project there is a far greater risk of unidimensionality; the latter one enhances the diversity and richness already present. A largely residential area, such as Floresta, presents certain inherent limitations. Yet the "extras" may be as important as the housing: a) Transportation is a vital link from home to work and to all the city has to offer; it must function reliably, conveniently and at low cost; b) Day-care centers free women for work and thus supplement increasingly strained family incomes, as such they should be integral parts of housing programs; c) Houses may serve as a commercial outlets or workshops for artisan production. Rather than be discouraged as is presently the case in BEV regulations, such usage of space and such economic activities should be encouraged by credit programs; d) Good health is basic to productivity and concentration in studies. A health center should be a functioning service for a project of 3,000 units, such as Floresta. Perhaps it could serve as a positive linkage to parts of the more needy adjacent Guasmo population as well, e) The social division might act as referral service for the un- and underemployed, if not to work opportunities to specialized state agencies which do perform this function. Given the continuing demand for construction in this kind of expandable core housing project, particularly in the piso-techo project where building is mandatory, courses in the building trade could be offered by the Housing Bank on a cost recovery baais (with payment due after an initial period of employment). The attempt here is not to be original, but to be illustrative of how a new housing project such as Floresta may and should go beyond housing to help residents realize other vital needs, often. more directly linked to income-generation than is housing per se. The cost of these measures is slight. What is required is comprehensive planning and sound management oriented to individual and community needs rather than housing production. 3. Contiguous Population Groups A final point regarding Floresta, which also ties to obs'ir-qticns made regarding 8 de Diciembre in La Paz, is the alppatrent failure of - planners to take into account the presence of popalation groups living in areas contiguous to the project area and the effect these persons have on people living within that area, on the beneficiaries. In La Paz this was largely a problem of breeding resentment and frustration in a poorer population group roughly equal in number to the population benefitted. In Guayaquil, Floresta borders an area the name of which, Guasmo, strikes dread in the hearts of many of the city"s affluenit population. Guasmo is an invasion or squatter area of over 200,000 persons which has no urban ameni'ties, save electricity. Some perspective on this awkward juxtaposition of Floresta and Guasmo may be gained from a brief history of the area. In the late 1970's as the military government in Ecuador was coming to a close and being replaced by a populist, democratic regime, the southern region of Guayaquil, where Guasmo is located, became a major staging area for diverse political.parties. This was a place of mangrove swamps largely below sea-level yet it was also the only remaining major open land area accessible to low income families in Guayaquil. As such it 'was the site of massive invasions of poor families, the first of which was Guasmo Norte, site of the unsucessful upgrading effort reported in section III. This growing squatter population clamored for land, fill and tenure, and services from the politician, local and national, each of whom promised more than the other, with far more rhetoric than concrete action. Largely as a response to this Guasmo situation, the newly-elected president ordered BEV to quickly build low-cost public housing in Guayaquil. The first site chosen was Floresta, already owned by BEV, as it, like neighboring Guasmo, was being occupied by squatters. Even as late as the time of my stay in Guayaquil, from mid-1982 to early 1983, squatters occupied BEV land adjacent to Floresta, erecting a placard as their site which read tFloresta IIIV. While I lived in Floresta, I could see several areas of increasing conflict between that area and its poorer, far larger neighbor: a) The security issue (robbery, vandalism, personal injury) was serious to Floresta residents, who were generally the victims; b) There was pressure on the urban services (water, roads and garbage collection) used by Floresta and by Guasmo residents. Trucks transported wat:er to Guasmo from main lines along commonly usied roads, thus adding to the breakdown of pavements and lowering water pressure in Floresta. Guasmo residents. deposited their garbage in overladen receptacles in Floresta, taking . advantage of the collection service which, though inadequate, was better than in Guasmo, where it was entirely absent. c) There was general malaise and fear on the part of Floresta residents, related to the security issue (a) but transcending it and due to a recognized marked inequality between the services they receive and the absence of services in Guasmo. Clearly, where a new serviced settlement is built next to a large population group lacking services, problems will arise between the two areas. In this case management of the entire issue is impeded by the unrelated dual authorities of the Municipality of Guayaquil, which has jurisdiction over Guasmo, and BEV, which planned and manages Floresta. The issue requires governmental planning and authority which coordinates activities in both areas and leads to the eventual equalization of services between both, with immediate security and preservation of the serviced area in the interim. 23- VI. CONCLUSION A. The Approach. Having looked at the diverse components of two projects in various neighborhoods of La Paz and Guayaquil utilizing different techniques, it is difficult to neatly separate findings according to particular method used. Clearly, the inform-ation about the credit programs could have been gathered without recourse to participant-observation. Even here, however, an observation such as the importance of encouraging rental in a home improvement loan program in order to help improve the living conditions of renters or an appreciation of the key ingredient of lender-borrower relations in artisan credit programs, came from lengthy discussions and an in-depth contextual understanding that the typical interviewer of survey questionnaries would not have. There were also a number of findings that I can state unequivocably could only come from the kind of participant-observer, qualitative evaluation described in this report. These include: a) The importance and significance of time, particularly delay, to project beneficiaries, in 8 de Diciembre, Rio Seco and 16 de Julio, La Paz; b) The nature of political leadership in Guasmo Norte; c) Relations between the Veneficiary population of 8 de Diciembre and the executing agency, HAM-BIRF, as viewed particularly by the former; d) Design limitations affecting the 1-iveability of low cost housing; e) The lack of community cohesion in both 8 de Diciembre and Floresta; f) The utility of extensive promotion of an upgrading project in a community, beyond local leadership, from Guasmo Norte and 8*de Diciembre; g) The human dimensions of housing and community improvement, i.e., the pride resulting from urban upgrading and home ownership. There are a number of issues which community-based participant- observer evaluation does riot address very well, if at all. These include: a) The macro economic situation of a country: This is only perceived as it filters down and effects the beneficiary group with whom one is living, many of whom have their own survival buffer strategies which cloak the naked economic data one reads in World Bank documents. Still, even here, certain human interest descriptions can be injected into the dialogue about the social and political effects of economic policies. b) The inner workings and decision making of the implementing agency. While some insight can and should be gleaneu from discussion and interviews with implementing agency staff, a good appreciation of the constraints they work under, of their finances, relations with other government agencies and key personnel decisions (to name only a few of many key issues) can only come through an evaluation focused at the level of the - 24 - institution, a most worthwhile, needed area of inquiry but not the same as the world of the beneficiary; c) Recommendations tend to suffer from the focus on the perspective of the beneficiary. This does not mean the participant- observer should not make recommendations, just that they will not be as balanced as the project manager may wish -- precisely because they will lack the aforementioned larger contextual and institutional understanding. On the other hand, their depth and relevance in relation to the beneficiary population has a value in and of itself which can readily be digested by the decision-maker as one of several kinds of information needed for project adjustment and program design; d) Where the need is for large sample surveys, control groups, and statistically, valid findings, qualitative methods must play a support role to quantitative evaluation. Both HAM-BIRF and BEV-BIRF,the key implementing agencies relating to this evaluation in La Paz and Guayaquil, respectively, have made a number of changes in their projects and in their internal organization and operations as a result of this evaluation. As mentioned in section II, HAM-BIRF is now requiring all persons contracting for upgrading to make household connections on the property, regardless of whether it is owner-occupied or rented; in addition, HAM-BIRF is undergoing restructuring to address both evaluation and communtity organization more effectively than it has in the past. BEV-BIRF is also reaching out into the community more in its home improvement loan program to better promotion and collections. .Aso, as mentioned previously, BEV-BIRF will seek to avoid the problem of delayed occupancy of piso-techo units by marketing the units to somewhat higher income families and offering loans for home build-out. Finally, both HAM-BIRF and BEV-BIRF state they will hire or contract for participant-observers to provide feedback, commencing within calendar year 1983. B. The Proiects What I learned about these two projects by looking at them from the bottom up suggests three broad conclusions which are probably relevant to aevelopment efforts generally - and probably consistently underestimated by project planners and administrators: 1) The importance of communication. The gap between haves and have-nots is made wider by the difficulty -of communication between them, exacerbated by the mistrust that the poor have of the affluent or - powerful. The disillusionment among 8 de Diciembre residents provoked by project delay, the obstructionism of local leaders in Guasmo Norte, and the resort to municipal politicians to help determine the price of houses in Floresta, all witness to the importance and neglect of communic.ation between administrators and beneficiaries. 2) The importance of community organizaton. Community organization is so often dismissed as a vague, soft term and treated (if at all) in rote, cursory fashion. But community organization needs to be built as carefully as housing and drainage systems. Existing community organizations in poor neighborhoods of developing cities are often weak and inadequately representative. In new housing developments, community clearly needs to be established. Although community organization by - 25 - government agencies is difficult and delicate, project officials should give attention to the creation of community consciousness and organization. These can serve as the basis for collective self-help improvements and the articulation of common needs.. They are the foundation for self-sustained change, which is the goal of development. 3) The importance of integrated planning. At least from the perspective of the beneficiaries, a project is but one component of what is needed to make and improve the community. While no one project can address all the myriad problems which affect an urban area, project officials can be aware of the inter-relatedness of the many aspects of urban life, and try to establish comprehensive planning and effective coordinating bodies. More generally, this research into the methodology of evaluation has demonstrated how important it is to understand the perspective of the intended beneficiaries of development projects. The research has demonstrated that qualitative, participant-observer techniques yield useful, sometimes distinctive, and relatively quick insights of operational significance for urban projects. It has also led to positive initial experiences and tentative guidelines for the replication of this approach to project evaluation. .1 27 - Annex METHODOLOGY A. The Approach The methods which have been employed in this evaluative study of two urban projects all fit into what is known as qualitative evaluation. They range along a continuum from purely experiential participant observation expressed as description to interviews of varying degrees of formality all designed and administered to bring the respondent. to express his or her opinions about the project in which he or she is a beneficiary. These interviews lead to quantification but are nevertheless qualitative in that "the fundamental principle of qualitative interviewing is to provide a framework within which respondents can express their own understandings ini their own terms"t. l/ In qualitative evaluation form follows function. There is no set or predetermined instrument through which learning takes place. Rather, the particular methods evolve-from the context of the project determined by the problems addressed. To take one example from this evaluation, when I first lived in the low-cost public housing area of Floresta, in Guayaquil, I was not sure what issues to address though general guidelines had been established. By the end of the first three months one issue among others was clearly important: there was no dialogue between the residents and representatives of the Housing Bank (BEV). Part of the reason for this may have beeni an uncommunicative stance on the part of BEV, something beyond the scope of my work; another part presumably lay with the beneficiaries among whom I was living. Upon my return, I resolved to know more about how the residents of this project related to each other and to their local leaders. While I could learn about these issues from participation in meetings and observation of interaction on my own pathway, I needed some quantitative measure of what was taking place in other pathways. Hence, an interview format was devised which, among other things, sought to elicit comments from the residents about the nature of their interaction with neighbors and their elected representatives and their opinions about the value of this interaction. In practical terms, the qualitative approach to evaluation stresses.relevance over rigor, timeliness over precision, utility over strict accuracy. Proximity to the beneficiaries is key, as is openness to them. The attempt is to learn as much as possible about how the beneficiaries are changing as a result of the project intervention, with a view to further improving the situation. A basic tenet of this approach is that development is never adequately understood on the basis of abstractions; it must be experienced to be fully understood. Since most designers, planners and supervisors of development activities do not directly experience these activities as beneficiaries, it is important that they seek to gain an understanding of that experience by having persons reporting to them whose business it is to conduct this type of learning. Whether these are anthropologists, sociologists, architects or administrators is not so important as that they attempt to relate to the 1/ Michael Quinn Patton, Qualitative Evaluation Methods, Sage Publications, Beverly Hills, 1980, p.205. ANNEX -28 beneficiaries in an open fashion which elicits maximum trust. This confidence provides the basis for the kind of participation which allows for the depth of observation needed to gain insight into project activities from the perspective of the beneficiaries. This insight can then be fed back to project planners and administrators in a useful and timely fashion so that they car., through this process, become more responsive and hence responsible "public servants". Qualitative evaluation is often complemented by quantitative analysis. I have incorporated more structured interviews which lend themselves to quantification than were envisaged at the outset of this study. Clearly, there are certain observations, such as the labor generated by home improvement loans or, similarly, home improvements generated by the provision of infrastructure, which lend themselves to pre-planned quantification with statistically-reliable sampling methods. Surveys may best build on the insights gained through participant- observation. It is in fact to be expected that such sequencing of qualitative and quantitative methods should be most useful in many projects; the former should help provide the framework for survey design and meaning to the numbers the latter provides. B. Practice I have outlined the key steps that a participant evaluator might follow as he or she attempts to gain an understanding of beneficiaries undergoing changes induced by a project. I cannot stress too much the extreme importance of gaining the trust of both the project implementors and the community in the early stages of the evaluation. The key at the outset is to be as open, honest and forthright about oneself and the nature of the evaluation to be undertaken as possible. Generally, there will not be obstacles from either of these two groups, especially if it is made clear at the outset that the objective of this participant-observer evaluation is simply to gain the kind of understanding which can be put to use in the betterment of the project, towards which both beneficiaries and implementors should be favorably disposed. Steps to Follow in Conducting Participant-Observer Evaluation- 1.Familiarize oneself with the background of the project: What are project objectives? Why and haw was the community being improved or people being housed chosen? How. did executing agency and beneficiary population come togethey? What is the institutional framework for project implementation? Source for this information: interviews with key representatives of funding and executing agencies and with beneficiary leaders at the time of project initiation; documents. 2. Learn general characteristics of population group being benefitted by project: History of area, permanence and prior place of residence, reasons for coming to area/housing; socio-economic profile of population (age, sex, education, employment, income,- health), being sure to sample for major internal divisions (geographical, income, owner/renter, etc.). 3. Nature of residence: One should live in a fairly-centrally located place, somewhat better than average (for upgrading area) that offers the basic comforts needed and independence of entry. If one is single it is best to live in association with a family, yet clearly 29 independent of it, so as to more easily blend 'into and become a part of community. 4. Get to know the key actors in the project well: the beneficiaries in general; their leaders, formal and informal; and the key administrators of the implementing agency. Attempt to keep the interrelationship with each of these three major groups somewhat discrete. One's manner of relating is decreasingly professional from project administrator to leader to regular beneficiary. While all know the reason for one's stay in the neighborhood, the official nature of that stay is kept most apparent to the project personnel, with whom a certain degree of personal distance must be maintained to avoid bias, real or alleged, intentional or unconscious. The leaders should also be made aware of the motive for one's presence, especially at the outset2/, less so over time. The people, however, while duly informed of one's reason for living in the community, should primarily get to view the participant-observer as a neighbor and, to varying degrees, friend. At the level of the people, the relationship is far more personal than professional. a) Cultivate a few close contacts from diverse, major segments of population. These should represent various key income groups, political factions, owners and renters (where the latter are a significant part of population), etc., one should never be overly identified with any one group, but open and accessible to all: diplomacy at the neighborhood level. b) Attempt to participate in major organizations and activities of the community, sufficiently to be appreciated and identified as a participant but not so much as to become overly committed, i.e., retain one's independence yet demonstrate involvement. 5. Inject key issues of concern into discussion after requisite confidence has been obtained: These issues will have been worked out with funding and executing institutions before entry into the community and will be refined by the participant-observer, in conjunction with the executing agency while in the project site. It is important to focus these issues on the project and bring them into conversation or encourage them when they arise spontaneously in as unobtrusive a manner as possible. The aim is always to bring others out, to serve as catalyst, and to provoke the most honest response possible. For instance, one key issue which has been shown to have affected the diffusion of project benefits and even the viability of the project itself was the nature of community leadership in the two upgrading areas observed, 8 de Diciembre and Guasmo, in La Paz and Guayaquil, - respectively. This was clearly a sensitive topic, yet it had to be addressed to find out why certain project activities did or did not take place and why they benefitted certain people but not others. Direct references to.this issue at the outset of one's stay in an area to people with whom confidence had not been established would surely have produced distorted responses. 2/ It often helps for credibility to have a letter signed by a government authority, if possible and appropriate, not the particular executing agency but a more distant, neutral body. -30- ANNEX Do not neglect the power of observation: who actually participates in community improvement work? how are houses being improved? what is the quality of work performed by executing agency? etc. Photography will help illustrate observations. 6. After considerable familiarity has been gained with the community and the way it has been affected by the project, elaborate one or more simple survey guides to help substanitiate.or determine what appear to be key findings. This should be done in consultation with implementing agency. Sample sizes should be barely representative, large enough to be taken seriously, not so large as to attract undue attention in the community or take too long in processing of data, which should be simple enough to be done manually and in little time (not more than a week). 7, Prepare report with description of actual situation, key findings, conclusions and recommendations: Prior to writing, discussion should be held with key community leaders and informants to verify thrust of report and with implementation agency to better understand issues frQm its different and important perspective. If the report is to go to funding agency, best share it with executing agency first, as a matter of courtesy, ethics.and to help with needed cooperation of latter body. Disagreements of projects implementors, if any, should be reflected in report sent to sponsors. C. Replicability One of the objectives of this evaluation was to train local researchers to see if the methods described above could be replicated in developing countries so as to provide useful feedback to host country implementing agencies on a less costly and longer term basis than that made possible by this study. Towards this end I hired three full-time local research assistants, two sociologists in La Paz, Yolanda Barriga and Luisa Fernanda Rada, and one economist in Guayaquil, Hilda Sanchez de Gavirla. Several additional part-time persons were hired in each place, including an architect, two social workers and the directors of the social and/or evaluation units of HAM-BIRF in La Paz and BEV-BIRF in Guayaquil. The local research assistants hired to assist with the participant-observer evaluation all improved considerably with experience and orientation. They were already familiar and adept at the application of questionnaires among low income populations; the transition from closed to open question interview formats was not difficult and in fact welcomed by all three. Towards the end of the study in each city, after having witnessed my own participant-observation, each of the local research assistants lived in a low-income settlement for three months. After a number of meetings with both implementing agency personnel and myself, each local researcher proved to be a most useful commentator on the area in which she was living. On the b4sis of this experience to date I feel that this methodology can be replicated in developing countries with the use of local. personnel. The numbers of persons with whom I have worked on this participant-observer evaluation has been too small to permit easy generalization. Yet, tentatively, I recommend that persons recruited for the kind of qualitative project evaluation described in this report have a range of professional skil'ls including familiarity with social survey and anthropological field methods, administrative experience in the - 31- ANNEX implementation of development projects and facility in building rapport with diverse peoples. Inasmuch as the nature of the interviews and observations of the participant-observer and particularly the insights these produce is directly related to the experience of this person, I strongly feel he or she needs to be at least a junior and preferably mid-level.professional for all concerned to obtain the maximum benefit from this kind of evaluation.