32816 Africa Region Human Development Working Paper Series Managing Public­Private Partnership Lessons from Literacy Education in Senegal Bjorn Harald Nordtveit A study prepared for the World Bank Africa Region The World Bank Washington, D.C. Copyright © May 2004 Human Development Sector Africa Region The World Bank The views expressed herein are those of the author and do not necessarily reflect the opinions or policies of the World Bank or any of its affiliated organizations Cover design by Word Express Interior design by Word Design, Inc. Cover photo: Bjorn Harald Nordtveit ii Table of Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .viii Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6 Defining outsourcing as used in the Senegalese partnership approach . . . . . . . . . . . . .7 Testing outsourcing: Outcomes of the Senegal project for women's literacy (PAPF) . . .8 The case for outsourcing in the Senegalese context . . . . . . . . . . . . . . . . . . . . . . . . . . .9 I The Process of Outsourcing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 How does the selection and implementation process function? . . . . . . . . . . . . . . . . .13 How is the process set up? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 How much time does it take to select and implement a subproject? . . . . . . . . . . .17 Who are the stakeholders? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 The beneficiaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 The private providers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 The providers' organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 The contract-managing agency (AGETIP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 The Project Coordination Unit (PCU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 The Department for Literacy and Basic Education (DAEB) . . . . . . . . . . . . . . . . .22 Selection committees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 National Committee for Literacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 How to ensure transparency in selection of providers . . . . . . . . . . . . . . . . . . . . . . . .23 How to set up a good monitoring and evaluation system . . . . . . . . . . . . . . . . . . . . .25 Why is the interaction between villagers and providers so important? . . . . . . . . . . . .28 How does decentralization affect the program? . . . . . . . . . . . . . . . . . . . . . . . . . . . .31 Conclusion and recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31 II The Courses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 What are the characteristics of "good" courses? . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Literacy skills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 Basic functional skills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34 The integrated approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 iii III Costs and Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43 General Conclusion and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 Annex 1: Time Frame for Subproject Implementation . . . . . . . . . . . . . . . . . . . . . . . . .49 Annex 2: What Are the Reasons for Investing in Adult and Non-Formal Education? . .51 Annex 3: The Public­Private Partnership Criteria Checklist . . . . . . . . . . . . . . . . . . . . .55 Annex 4: Summary of Senegal's Procedures Manual . . . . . . . . . . . . . . . . . . . . . . . . . .57 Annex 5: The Gambia's Procedures Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 iv Foreword O ne of the Education For All (EFA) Literacy efforts up to now have often proved targets is to halve the rate of adult inadequate, at both national and international illiteracy by 2015. The importance of levels. Therefore, it is important to develop reaching this goal is widely acknowl- more effective and efficient ways of delivering edged. Accelerated expansion of primary edu- literacy programs. This includes efforts to link cation will help. However, the literacy target literacy programs more effectively with strate- cannot be reached without a sharply increased gies of poverty reduction and with programs in provision of basic education to adults and out- agriculture, health, HIV/AIDS prevention, con- of-school youth who have either not had access flict resolution and other social concerns. Fur- to primary education, or who have dropped thermore, diverse and meaningful literacy out before gaining basic literacy. This is partic- requires flexible modes of acquisition and ularly the case in Sub-Saharan Africa where delivery, using appropriate materials and lan- high illiteracy and low basic education attain- guages, focusing on relevant purposes, and ment represent a serious constraint on a coun- generating interesting, culturally relevant and tries' ability to reach national economic, social gender-sensitive materials at the local level. and political development goals. Finally, strong community ownership of the The launching of The United Nations' Liter- purposes and processes of literacy is indispen- acy Decade (2003-2012) illustrates well the sable to ensuring that programs are flexible importance given to literacy and basic educa- and responsive to the demands of adult learn- tion at the international level. The Decade will ers, and to ensure sustainability. focus on the needs of adults, with the goal that Senegal has pioneered an approach to adult people everywhere should be able to use litera- functional literacy education that responds to cy to communicate within their own communi- many of the above concerns. In this model, the ty and in the wider society. Earlier efforts have government outsources the actual provision of failed to reach the poorest and most marginal- services to civil society and local social entre- ized groups of people, including the majority preneurs. The model is known as "faire-faire" of rural women in Africa. Therefore, the (getting things done). Its originality lies in the Decade will particularly address the needs of use of many small providers who usually are such populations. village-based, each implementing a small liter- v vi Managing Public­Private Partnership acy project--rather than signing contracts with model used in Senegal can be adapted to other big provider organizations to implement large- countries and be flexibly adapted to different scale projects. The model has influenced policy local contexts and needs. An added advantage on adult literacy education in several other may be the potential for quick expansion of countries, and there is wider international delivery. interest in it beyond the West African region. The Norwegian Education Trust Fund has Therefore, the purpose of the present study is funded the research of this study and its pub- to explore what can be learned from the Sene- lishing costs. The assistance of the Norwegian galese experience. Among questions asked are: government, through this fund, to stimulate How well is it working? What organizational interest for adult and non-formal basic educa- mechanisms are used? How have procedures tion for vulnerable groups, is gratefully evolved in response to issues arising during acknowledged. implementation? What is the cost? The study concentrates on the procedures used and on Birger J. Fredriksen the practical management lessons from imple- Senior Education Adviser mentation. It argues that the partnership Africa Region, World Bank Acknowledgements Particular thanks are due to Jon Lauglo, the task team leader for this study, whose com- ments, suggestions, and edits greatly improved this text. Thanks are also due to Helen Abadzi, Aya Aoki, John Comings, Luis Crouch, Peter Easton, Amadou Wade Diagne, Linda English, Birger Fredriksen, Henner Hildebrand, Steven Klees, Alain Mingat, Mamadou Ndoye, Tonia Marek, Meskerem Mulatu, Susan Opper, John Oxenham, Maman Sidikou, Alan Rogers, Serge Theunynck, and Adriaan Verspoor, all of whom contributed advice and critiqued this paper in the course of its preparation. Special thanks are due to the local team that facilitated the research in Senegal, particularly to Alassane Ndiaye and staff in the Department of Adult and Basic Education, as well as Lamine Sarr and Bineta Sall and staff in the Project Coordination Unit, both for their help with logistics and for their comments and advice on the text. vii Acronyms AGETIP Agence d'Exécution des Travaux d'Intérêt Public (Contract-Managing Agency) CIDA Canada International Development Agency CBO Community-Based Organization CDF Comprehensive Development Framework CNEA Comité National de l'Eliminiation de l'Analaphabétisation (National Committee to Eliminate Illiteracy CNOAS Coordination Nationale des Opérateurs en Alphabétisation au Sénégal (Providers' Association in Senegal) DAEB Direction de l'Alphabétisation et de l'Education de Base (Government Agency for Literacy in Senegal) ICR Implementation Completion Report IDA International Development Association GTZ Deutsche Gesellschaft für Technische Zusammenarbeit (German Technical Co- operation) NGO Non-governmental Organization NPV Net Present Value PADEN Projet Alphabétisation des Elus et Notables Locaux, Literacy project targeted at building capacity of local officials PAPA Programme d'Appui au Plan d'Actions (Literacy Support Program) PAPF Projet d'Alphabétisation Priorité Femmes (Women's Literacy Project) PCU Project Coordination Unit (also designed as CCS: Cellule de Coordination et de Suivi du Projet) PPP Public-Private Partnership PRA Participatory Rural Appraisal SAR Staff Appraisal Report UNESCO United Nations Educational, Scientific and Cultural Organization WDR World Development Report viii Executive Summary T his study describes a model for public- In all, over 300 separate subprojects were private partnership used in a women's financed--most of which used learning literacy project, or Projet d'Alphabéti- approaches, language instruction, and a sched- sation Priorité Femme (PAPF), in Sene- ule that was decided locally and with the par- gal. The focus is on practical experience, and ticipation of the beneficiaries themselves. Most lessons learned with regard to outsourcing lit- courses worked in close cooperation with local eracy courses to the private sector. The govern- women's associations, which facilitated the ment's literacy department, the Departement organization of the courses (see Figure 1). d'Alphabétisation et de l'Education de Base Some of these associations later became litera- (DAEB), maintained responsibility for policy cy providers in their own right. The private formulation, monitoring and evaluation; and sector was used not only for the actual provi- private providers received financing from the sion of courses, but also for management of government to implement local literacy cours- funds through a contract-managing agency, es. The providers consisted mainly of local Agence d'Exécution des Travaux d'Intérêt community-based organizations (CBOs). Public (AGETIP). During a period of five years (1996­2001), As noted in Figure 1, the main stakeholders the Women's Literacy Project enrolled almost in the PAPF partnership were the beneficiaries, 200,000 participants in local-language literacy the provider organizations, the department courses.1 Within this project, each provider responsible for adult literacy education implemented a small project, or "subproject", (DAEB), and the contract-managing agency covering between 10 and 20 courses in a cor- (AGETIP). The providers bid for public funds responding number of villages. Each course to finance their projects. "Outsourcing" in this enrolled roughly 30 learners, mostly women. A text is used to describe the process, by bidding, course instructor, whose salary was paid by the through which public funds are made available provider, conducted the classes. The partici- to private providers for implementation of lit- pants not only learned to read and write, but eracy services.2 In the PAPF, a "procedures also received instruction in health and hygiene, manual" explained the bidding and selection and in income-generating activities. process, and the requirements of the successful 1 2 Managing Public­Private Partnership Figure 1: Subproject set up The Literacy Department (DAEB ) is in charge of monitoring and evaluation Providers implement a Each course enrolls 30 "subproject" covering learners in coordination 10-20 courses with a local women's The contract-managing association agency (AGETIP) oversees financial issues bidder. The procedures manual required each providing management and accounting courses provider, in cooperation with villagers, to pre- for its leaders. The women's associations often pare a literacy subproject proposal. The proce- encouraged women's participation in local dures manual defined a standard unit price decision-making, including village manage- that the successful bidder received for each ment and maintenance. beneficiary enrolled in literacy classes. Hence, The provider organizations responsible for the bidding was not a competition for the low- implementing subprojects included communi- est cost, but instead, for the highest quality. ty-based organizations, religious associations, The subproject proposals were checked for and larger non-governmental organizations. accuracy at the local level by government staff. The PAPF aimed at creating additional They were then sent to a selection committee providers through giving new associations and that analyzed all the proposals and decided organizations access to funding. The selection which ones should be financed. The proposals criteria reflected this aim. If the selection crite- approved for financing were sent to the con- ria had been stricter, the program would have tract-managing agency (AGETIP), which strengthened the existing providers instead of established contracts, paid the providers, and facilitating the creation of new ones.3 In the managed and verified the contracts. After PAPF case, stricter selection criteria may have obtaining the first installment of the funds, the yielded better results, but would have created providers began implementing the subproject. less grassroots activism and fewer opportuni- The main stakeholders of the Women's Lit- ties for program expansion. eracy Project (PAPF) were the learners. They Many providers belonged to an umbrella were usually members of a women's organiza- association, the Coordination Nationale des tion. A longitudinal study of the outcomes of Opérateurs en Alphabétisation au Sénégal PAPF indicated a higher success rate in com- (CNOAS), which had as its aim to coordinate munities where the subproject was based on an the literacy providers' activities. This provider existing women's association, and where the organization disseminated information for the villagers had set up a local management com- PAPF, and was also used to ensure transparen- mittee at the start of the implementation phase cy in the bidding and selection process. How- (DAEB, 2001). If such associations and com- ever, CNOAS did not play a key role in the mittees did not exist in the target zone, the PAPF partnership. The reduced level of inter- PAPF required that the providers facilitated vention of CNOAS may partly be due to its their creation. Where such an association lack of permanent salaried staff (the staff already existed, the subprojects improved it by responsibilities were held by elected providers). Executive Summary 3 It is hoped that the association will be involved PAPF. The (non-integrated) courses had been in setting ethical standards of conduct for the organized as a one-time event, i.e., the providers and that it will arbitrate in dispute providers implemented a course without set- cases. ting up any permanent literacy or reading cen- The government's literacy department ter in the villages. With the integrated model, (DAEB) was responsible for data collection however, the providers created permanent lit- and technical evaluation of the subprojects. eracy centers in the beneficiary villages; and The technical evaluation had as its objective to tried to make literacy activities more perma- ensure that the lessons corresponded to the nent through the use of a relais, a local substi- participants' requirements, and that the tute for the provider-financed instructor. The provider had hired good instructors for the relais was supposed to continue to organize courses who were using adequate teaching learning activities in the literacy center when methods to conduct the classes. The contract- the provider's support came to an end. managing agency (AGETIP) was responsible The integrated program was among the for financial monitoring, i.e., to check that the most expensive of its kind in Senegal, and had providers spent funds according to the agree- a base unit cost of about $504 per person for a ment between the provider and the govern- course intended to cover over 450 hours, nor- ment. However, both AGETIP and the DAEB mally completed in less than 18 months.5 If the had problems in implementing the monitoring program costs at government and outsourcing and evaluation work. The Project Coordina- agency levels were added, the unit cost came to tion Unit (PCU), a separate unit set up to over- $58, with an initial investment cost in human see project implementation for the PAPF, did capacity building and materials of $12. The most of the monitoring. It functioned as an public cost per enrollee can thus be estimated independent unit that coordinated with DAEB, at $70, whereas the private costs incurred by and reported to the World Bank and to the the learner usually did not exceed $26 (oppor- Ministry of Education's cabinet. However, tunity + direct costs).6 This investment was inefficient monitoring and evaluation led to designed to provide a package of advantages lack of accurate information on the learning both for the participant and for the communi- outcomes and impact of the literacy courses. ty. The participant had, in many cases, direct Data reported were often contradictory and economic benefits inasmuch as he or she did not establish a clear overview of the impact learned some income-generating activities, of the program. Also, rules preventing inade- how to keep track of income, establish a bank quate behavior and misconduct were not account, and become eligible for micro-credit enforced sufficiently well. Poor monitoring and programs. Intended non-monetary benefits slack enforcement of rules led to reduced included better health. The providers were accountability of certain providers, most likely expected to coordinate with vaccination initia- resulting in substandard results in many sub- tives and with other local public health pro- projects. grams. In addition, some of the learning mate- The PAPF financed different types of litera- rials dealt with health issues. The community cy courses. Most courses separated literacy and may have benefited from the programs insofar post-literacy activities, i.e., post-literacy sub- as their markets became more efficient, health- projects were selected and financed after the related expenses were reduced, and literate literacy project had been implemented. A new people became more resourceful participants in type of subproject, which integrated literacy local development initiatives and in the civil and post-literacy activities, was found to yield society. However, the degree of impact which better results and was increasingly used by the literacy training had on each of these benefits 4 Managing Public­Private Partnership is difficult to gauge empirically and depends on projects. Adequate decentralization of the quality and focus of each subproject. monitoring and evaluation is important. The return on investment, in comparison to 6. A system of sanctions is needed to deal primary education, is believed to be quicker for with weak implementation and with adult education than for education of children, embezzling of funds. Sanctions must be since the newly acquired knowledge can be put rigorously enforced. into use immediately. In the PAPF, as in other 7. The procedures manual should be as cases where use is made of market mecha- practical as possible, and clearly explain nisms, the use of competitive bidding was how funds are obtained. Methods for intended to improve cost-effectiveness. How- interaction with the villagers, e.g., ever, the benefit of competition was con- through Participatory Rural Appraisal strained by the fact that services were pur- (PRA) techniques, should be described. chased by the government on behalf of the 8. Likewise, the contract between the beneficiaries, not by the beneficiaries directly. providers and the government should be Hence, new projects in Senegal are currently as clear as possible, and provide concrete seeking to involve the villagers in the process of guidelines for the desired input (i.e., char- selecting providers to better ensure that the acteristics of beneficiaries) and output services purchased are appropriate to the ben- (i.e., learning impact) of the partnership. eficiaries' specific situation. 9. A technical (not political) committee The lessons learned through the Women's should select providers. The committee Literacy Project (PAPF) are institutional, eco- should include representatives from civil nomic and literacy-specific in nature. Key find- society to ensure fairness and trans- ings from the analysis of the outcomes and ref- parency. erences to sections of the present study are as follows: Section II: The Courses Section I: The Process of Outsourcing 10. Use of an "integrated" type of program, combining literacy teaching with func- 1. The partnership approach requires strong tional skills demanded by the benefici- political support. aries, boosts the participants' enthusi- 2. Systematic involvement of beneficiaries in asm to participate in the courses; planning and implementation is needed to literacy training may successfully be ensure that the program is driven by coupled with income-generating and grassroots demand. other poverty-alleviating activities. 3. An umbrella association of providers is 11. Teaching could in many cases benefit important for information dissemination, from more use of peer tutoring and and as a safeguard for fair selection.7 group-work methods to ensure better 4. The use of an independent contract-man- and more rapid acquisition of reading aging agency can ensure rapid set up of and writing skills. contracts, and timely transfer of funds. 12. Creation of women's associations and 5. Monitoring and evaluation are key to training of the associations' leaders in high quality project performance. A good management and accounting will, in monitoring and evaluation system should many cases, have beneficial effects on be set up before implementation of sub- the village's social structure, facilitating Executive Summary 5 women's participation in decision-mak- 16. Another important factor for cost-effec- ing, including village management. tiveness is to ensure that the enrollees 13. Use of a local substitute instructor who correspond to the target beneficiary has been specifically trained for contin- characteristics (e.g., that priority is uing literacy activities after the provider given to enrollment of illiterate or semi- finishes its intervention can be a means literate people--if that is an explicit tar- to preserve continuity after the "out- get criterion). sourced course" is completed. 17. Clear indicators for success should be established; a clear and fair monitoring and evaluation system is needed; and it Section III: The Costs is essential to check that the targets in program enrollment, retention and suc- 14. Often, the most expensive cost factors cess are upheld. in a literacy project are dropout and 18. Waste could be reduced through ensur- failure to pass the final exam (i.e., fail- ing that the defined indicators corre- ure to learn to read and write). Extra spond to the program goals, and that investments can, in many cases, be well the tests administered to the beneficiar- justified if they improve retention and ies include indicators of these goals. completion rates. 19. To avoid misuse of funds, one should 15. A minimum success rate target could be ensure that the penalty for fraud is indicated in the procedures manual and greater than the cost to implement the in the subsequent contracts between the program (through use of fines and other providers and the contract-managing penalties). agency, as a means to ensure adequate completion. Introduction T his study analyzes the use of outsourc- and adults who have missed out on primary ing as a method for implementation of schooling. With the achievement of universal a large-scale literacy program in Sene- primary education--the target year is 2015-- gal. The focus is on implementation the objectives of adult education provisions in experience and on practical lessons learned Sub-Saharan Africa may shift toward serving from that experience. The first part of the those who already have acquired basic literacy report defines outsourcing and explains when and numeracy skills in school.8 In a situation and how it can be used. The second part is of shifting target groups, many countries may more specifically literacy-oriented and analyzes not wish to invest in a top-heavy implementa- the impact at the local level of the program in tion system that relies on public providers, and question. The third part examines the costs may wish to use a more flexible approach to involved. literacy education. One such approach is a Targets for improved adult literacy were public-private partnership model known as agreed at the World Education Forum in Dakar faire-faire in French ("getting things done"), in 2000. The case for adult literacy programs which is used to implement women's literacy has also been further examined in several courses in Senegal. Some of these literacy recent World Bank publications that review courses were organized under the IDA-funded findings on the cost and impact of such educa- Women's Literacy Project or Programme d'Al- tion (Abadzi, 2003; Oxenham et al, 2002, phabétisation Priorité Femmes (PAPF). The Carr-Hill et al, 2001, Lauglo, 2001). Though present study describes how this model worked more research is needed for better documenta- in the case of PAPF, what the problems were, tion of impact, the present study assumes that and how these problems were addressed.9 the case for adult literacy and related educa- PAPF was one of several projects in Senegal. tion rests on a strong enough international After independence in 1960, Senegal achieved consensus to justify more attention on issues of limited results in implementing youth and adult implementation. literacy courses. In 1995, a new implementa- In many countries, primary education is still tion strategy was tested to determine whether far from universal, and there is therefore a private providers (including NGOs, CBOs, and need to offer basic education to adolescents religious associations) could deliver literacy 6 Introduction 7 education more efficiently. The strategy, based between the private and the public sectors. PPP on government outsourcing of literacy services, in education can be described as a "partnership became widely used from 1996, and was close- between the public and private sector for the ly linked to Senegal's goal to accelerate devel- purpose of delivering a project or a service" opment of basic education. The Canadian and "come in a variety of different legal or development agency CIDA, together with IDA contractual forms, but at the heart of every and GTZ, co-financed the implementation of successful project is the concept that better the strategy through the projects PAPA,10 value for money be achieved through the PAPF,11 and PADEN,12 respectively. The IDA- exploitation of private sector competencies and financed PAPF was implemented between 1996 the allocation of risk to the party best able to and 2001. One of its main goals was to test the manage it" (IFC, 2003). The particular part- feasibility of outsourcing, and to check whether nership model used in Senegal is based on out- the process could be used and expanded nation- sourcing of public funds to the private sector wide. Because of its success in providing rapid- for implementing literacy courses. Risk is ly expanding literacy services to a large popula- shared between the public and the private sec- tion, the project became a model for literacy tors.13 programs in many neighboring countries, such The following questions are posed: as Burkina Faso, Chad, Côte d'Ivoire, The Gambia, and Guinea. Likewise, other govern- · What is outsourcing, and when should it ment sectors in Senegal and elsewhere used be used? (See Section I on process issues); PAPF as a model and adopted the "faire-faire" · How should a successful outsourced liter- type of outsourcing as a method for implemen- acy program be set up? (See Section II on tation of social services. Several new literacy course-specific and pedagogic issues); projects in Senegal (of which the major ones are · What are the costs? (See Section III on financed by IDA and CIDA) continue to finance cost issues); literacy courses by using outsourcing. The Dakar framework for Education For All Many of the advantages and disadvantages commits countries to a 50% reduction in levels in outsourcing of adult and non-formal edu- of adult illiteracy by 2015, specifically focusing cation are the same as for delivery of other on women. Also, literacy programs are dis- social services, such as preschool programs; cussed in the Millennium Development Goals social projects targeting street children; in conjunction with development among adults, refugees and other vulnerable populations; as a strategy to reduce poverty. The national social funds and poverty-alleviation pro- commitment toward adolescent and adult illit- grams. Within a public-private partnership eracy have generated further interest in the out- framework, "outsourcing" (or "contracting sourcing approach, especially in assessing the out") refers to the process through which extent to which it facilitates fast expansion of public funds are made available to private provision and genuine partnerships between providers for implementation of services.14 In government and non-government providers. education, outsourcing has, in some coun- tries, been used to implement services deemed peripheral to education itself, such as school Defining outsourcing as used in the transportation, meals, cleaning, and mainte- Senegalese partnership approach nance. Outsourcing methods have also been used to build infrastructure (construction of Public-private partnership (PPP) refers to schools, hospitals) and to set up support pro- modes of cooperation (contractual or not) grams (various training, computerized infor- 8 Managing Public­Private Partnership mation systems). More recently these meth- Testing outsourcing: Outcomes of the ods are being used to provide social services, Senegal project for women's literacy including health care and other public servic- (PAPF) es. In Senegal, outsourcing of the provision of literacy courses is done through a bidding Senegal's Women's Literacy Project (PAPF) tried process, where the winning group is contract- out a new division of responsibility between the ed to set up and implement a subproject with- public and the private sector in literacy provi- in the larger project or program. sion. The government responsibilities included Governments and donor organizations have policy, definition of administrative mecha- contracted with private providers to carry out nisms, guidance, monitoring and evaluation. services for three main reasons: (i) the govern- For its part, the private sector was given ment wishes to concentrate on certain key responsibility for execution of the project. In tasks, and to use private providers for tasks the PAPF, private sector execution applied both where private provision has a comparative to the provision of literacy services and to the advantage; (ii) in a number of countries, pub- management of funds. This approach was given lic social services are inefficient; (iii) many the name "faire-faire" by the then Minister of donors view private provider-implemented Literacy and National Languages. programs as less corrupt than traditional pub- During a five-year period (covering 1996- lic delivery. 2001), about 200,000 people, of whom 87 % Using private providers to implement social were women, enrolled in literacy training in the services aims to improve the cost-benefit of local languages of Wolof, Pulaar, Sereer, Joola, delivery but without sacrificing the strong Mandinka, and Sonike. Over 180,000 literacy equity concerns of these services. The objec- booklets were procured; and about 140 titles of tive is to achieve faster implementation, high- novels and technical books on health, agricul- er disbursement rates, and a stronger impact tural, technical training, culture, religion, litera- than traditional, government-executed pro- ture, and traditional medicine, were published grams. Outsourcing, however, is no panacea; in the local languages. To contribute to the cre- some of the weaknesses of government-imple- ation of a literate environment in national lan- mented programs also apply to provisions guages, five quarterly newspapers, one in each outsourced to private providers: "It is often region where the project was implemented, were stated that privatization or `NGOization' produced and circulated in six of the local lan- would reduce corruption but this is seldom guages. To further improve the acquired skills, rigorously evaluated. Private providers and the project provided the target villages with NGOs can also siphon off or waste funds and books, posters and other learning materials. perform poorly in terms of service delivery" The dropout rate from PAPF in the regions (Azfar & Zinnes, 2003, p.13). This critique covered under the project is estimated at about can be applied to many outsourced programs, 15% by the government (DAEB, 2002), consid- including those in Senegal. Also, non-formal erably less than previous literacy projects in the education or literacy education services are country.15 From an international perspective, difficult to specify for delivery as per con- such low dropout is an outstanding achieve- tract. This difficulty, and the problems of ment. The low dropout is probably due to the monitoring actual delivery, may lead some participation of the villagers in establishing and providers to shirk on supply during the exe- defining the courses, and by involvement of cution of the program. Such shirking is hard local women's organizations in the project. to detect (see Section II.4 on monitoring and About 312 literacy subprojects were evaluation). financed out of 960 requests, and 40 post-lit- Introduction 9 eracy subprojects out of 89 requests (the later improved the literacy sector in two main areas: "integrated" subprojects included both litera- It improved the courses and made them fit the cy and post-literacy activities). The partici- local context, and it enabled quick expansion pants were trained in literacy and numeracy, of literacy provision. By making the courses fit and also received basic skills training in better to the local context, the local demand health and nutrition, as well as in certain for literacy increased. Such development in income-generating activities. In addition, Senegal is similar to the course taken in other some beneficiaries received training in busi- countries where "public service providers have ness entrepreneurship and management and dramatically improved the delivery of their administration. The management training services, and have often reduced their costs was particularly structured to enhance the too, by using small and even micro-scale local functioning of the local women's organiza- community or private enterprises to fill the gap tions. In areas where such organizations did between themselves and their clients" (Harper, not exist, the providers encouraged their cre- 2000, p.22). Since the providers are based at ation. The women's organizations normally the community level and can interact with vil- followed up and monitored participants' lagers to provide services that are based on attendance in the courses. local demand, there is reason to believe that outsourcing leads to services that are user- friendly. Figure 2 uses the Senegalese model to The case for outsourcing in the illustrate the shift from public implementation Senegalese context to private implementation of public services (management of funds is not shown). In Senegal, both private and public project As noted in Figure 2, the outsourcing model stakeholders claim that the partnership method as it was set up in Senegal is characterized by a Figure 2: Literacy: Centralized Government Model and Outsourcing Model in Senegal (Financial mechanisms are not shown) (a) Before outsourcing (centralized government model) Central government Local admin. Population generates implements receives public implementation plan plan service (b) After outsourcing (participatory model) Beneficiaries formulate and Central Local admin. receive social government supervises services generates execution of Providers general policy policy generate and implement projects with the local population 10 Managing Public­Private Partnership feedback loop back to the central government. The PAPF project had as its objective to acti- The operation of the feedback system was not vate and create civil society, both at the always successful, and often relied on supervi- provider side (i.e., it aimed at creating more sion from the project coordination unit. Feed- providers) and at the beneficiary side (i.e., it back opportunities also exist in some govern- aimed at creating local women's associations ment-executed programs using formative that could manage their own literacy activi- evaluations, ombudsmen, and organizational ties). Instead of undermining the democratiza- means of aggregating views "from below" for tion process, this project has probably stimu- articulation "high up." However, PAPF lated it, giving local women a "voice" in the required direct participation of the villagers in local development process. planning the subprojects, and in that way ren- Some of the provider associations have dered the providers more accountable to the developed an acute sense of marketing their beneficiaries. The 2003 World Development services, and can be compared to small busi- Report considers such "short way" of account- nesses. Community-based organizations ability (to beneficiaries), to be generally more (CBOs) are often for-profit. In Senegal, the dis- efficient than the "long way" of accountability tinction between non-profit and for-profit is to the Government. not always clear. Often non-profits will still try Outsourcing has at times been criticized for to maximize profits so as to be able to pay undermining democratic decision-making and their staff members a decent salary. responsiveness to beneficiaries, since some PAPF providers generally fall into three cat- providers may adjust their approach to coin- egories as follows: cide with the donors' funding criteria (Cannon, in Pearce, 2000). Most donors and govern- · International or national non-governmen- ments, however, are interested in developing tal organizations (NGOs); programs that are well received by the popula- · Local development organizations (often tion and that respond to genuine needs, not to language associations, religious non-prof- a set of arbitrary "funding criteria." They its, poverty-alleviation associations, therefore have an interest in building feedback etc.);16 mechanisms that can help steer the project on · Local community-based organizations a path that is desired by the target population, (many of which are for-profit women's while maintaining cost-effectiveness. The associations that have previously received process, at least as it was set up in Senegal, is literacy training and that have now believed to enhance democracy, rather than become providers in their own right). undermining the democratization process. If anything, the process will make providers It can be argued that the government too, (including non-governmental organizations could implement public services based on par- with their own "agenda") more responsive to ticipatory approaches, by decentralizing the grassroots needs. decision-making responsibilities to the lowest The women's associations and providers in possible operational level of government, Senegal are good examples of civil society as an whether elected or appointed. However, as "intellectual space where people in a myriad of noted by Harper (2000), in many countries, different groups and associations can freely public administration seems remote from the debate and discuss how to build the kind of general population and the level of people's world in which they want to live" (Howell & trust in that administration is low. In Senegal, Pearce, 2001, p. 2). The private providers rep- the colonial legacy was an elite-oriented resent different fractions of such a civil society. administration that has historically not been at Introduction 11 ease with participatory decision-making and which aimed at encouraging the creation of styles of program implementation. Parents' new provider associations since it gave new associations have contributed to a closer con- associations easy access to funding. The new nection between the school administration and providers had many different origins: the community in primary education.17 In non- formal education, private providers (NGOs · Many provider associations divided into and others) appear to have taken the role of two or more associations. In most cases, such intermediaries, bridging the gap between some of the core technical personnel in the administration and the population.18 Thus, the initial organization set up a new asso- one aspect of the case for outsourcing in Sene- ciation. This phenomenon of division is, gal is based on the client-oriented approach of according to PAPF personnel, the most the providers, which would have been difficult important reason for the growth in the to achieve through public delivery. number of provider organizations. Outsourcing, as practiced through PAPF, was · Retired teachers and civil servants, seeing a quick process of obtaining funds; it took less that there was a market in literacy, estab- than one year from proposal to the financing of lished provider associations. the subprojects. Such short delays were difficult · Associations and organizations that had to achieve through government mechanisms. earlier been interested in other issues, Also, efficiency was promoted by the providers' (e.g., sports or diverse youth activities), competition for financing (although this way of became interested in providing literacy efficiency-maximization had its limits; see Sec- classes. In many cases, they kept literacy tion I.2.vii on selection). In its Toolkit for Pri- activities separate from these other activi- vate Public Partnership in Education, the Inter- ties (although the PAPF strongly promot- national Finance Corporation states: "Two ed integration of such other activities with obvious conditions need to be met for success-- literacy). sufficient public funding and the desire by both · Finally, as noted above, the subprojects public and private providers to try the PPP aimed to provide support to existing route" (IFC, 2003). In Senegal, these conditions women's associations. In villages where were met, since the government, together with such women's organizations did not exist, funding agencies such as IDA and CIDA, pro- the providers helped the women to organ- vided sufficient funding to implement high ize themselves. Many such women's quality courses, and the partnership-outsourc- organizations later became providers in ing approach was constructed in collaboration their own right. with civil society organizations, ensuring that the approach was "owned" by both the public The multiplication of provider associations and private sectors. was reflected in the number of subproject pro- Finally, another aspect of the case for using posals submitted for funding. In 1995, with the outsourcing has been the capacity of this pilot phase of the project, only 77 proposals approach to achieve rapid growth of provision. were submitted, of which 22 were selected. In Unlike the many cases where development 1996, 135 proposals were submitted, and 54 projects suffer from slow implementation and selected. Two years later, 189 subprojects were disbursement of funds, the expansion of the submitted; 68 were selected. In 2000, two PAPF project was constrained by available selection procedures were initiated (one for the funding during the project period, not by slow "normal" program and one for the new, "inte- disbursement. The rapid growth was due, grated" program; see Section II for description among other factors, to the selection criteria, of the integrated program). During this year, a 12 Managing Public­Private Partnership total of 368 subproject proposals were submit- costly) "integrated" courses. A comparison of ted, and 95 were selected (all data are from results between 1997 and 2000 shows the fol- DAEB, 2001). lowing considerable improvements, in terms of The multiplication of provider organizations the percent of learners who completed the does not seem to have adversely affected the courses and passed performance tests.19 quality of the learning, since assessments by Improved adult literacy rates, and the capac- DAEB show consistent progress in the learning ity for fast expension, while at the same time outcome of the courses. This may partially be improving quality, are attractions of the out- a result of the more extensive (and also more sourcing approach in Senagal. Table 1: Success rates PAPF20 Percent of learners who had the following characteristics at the end of the courses: 1997 2000 Fluency in reading 57% 75% Ability to write a short text 28% 63% Ability to resolve problems using arithmetic 6% 44% Source: DAEB, 2001 SECTION I The process of outsourcing T he PAPF outsourcing system tried to to fit this kit, but it contains all specified activ- involve the community stakeholders ities as shown in Table 2. during the formulation and delivery In Senegal, the process was cyclical, i.e., process. A key aspect of the process stages three through five in Table 2 were was the rapid selection and quick transfer of repeated every year. Instead of choosing one or funds to the providers, which aimed to mini- two large (international) providers, the Sene- mize the time-span between formulation and galese government opted for many small implementation of subprojects. The establish- providers, each catering to a limited number of ment of contracts and transfer of funds were beneficiaries. This approach stimulated grass- handled by a parastatal contract-management roots development within small-scale associa- agency. tions and organizations, thus strengthening Crucial issues arising from outsourcing are civil society. The selection and delivery monitoring and evaluation, how to ensure an processes (steps one and two) were reviewed impartial and fair selection process, and the every year through participatory mechanisms question of decentralization. Monitoring and to ensure that the performance of both evaluation are particularly important in order processes and implementation were optimal. to prevent substandard results in the execution The contracts and funding packages were stan- of the contracted services. Inevitably projects dardized to limit time-consuming negotiations that are based on outsourcing have had sub- with the providers. standard results because of the lack of proper monitoring and evaluation. This problem is not unique to outsourcing. With programs that How does the selection and are conceived, implemented and evaluated by implementation process function? the government, problems can arise if quality controls are not instituted "outside" the gov- How is the process set up? erning bureaucracy. Setting up a public-private partnership The tasks of the providers are defined in the approach has five stages in the IFC toolkit on procedures manual, which is the legal docu- education. The PAPF project was not designed ment informing any interested party about the 13 14 Managing Public­Private Partnership Table 2: Setting up a public-private partnership (adapted from IFC toolkit) Stages (IFC toolkit) Activities (IFC toolkit) Application in Senegal Stage One: 1. Identify candidates' services or projects 1. The government identified literacy as a Assessing service delivery for delivery through partnerships; priority area, and established the position options, including public­ 2. Evaluate which new or existing of a delegate minister to deal with the private partnerships services/projects might be suitable for issue; partnerships based on the government's 2. The delegate minister defined the own policies and criteria; partnership approach through outsourc- ing as the preferred method for service delivery; Stage Two: Preparing for 3. Define the project and services to be 3. A program coordination unit was set delivery through public­ delivered; up; private partnerships 4. Select the management team; 4. A colloquium gathering of 5. Select the preferred method for representatives from the Government choosing the private partner; and civil society (in Kolda) defined 6. Obtain the necessary approvals; service delivery conditions; 7. Define and determine the evaluation 5. A selection procedure was defined by criteria; all stakeholders and written in a 8. Develop and deliver the public procedures manual; marketing or involvement strategy; 6. The procedures manual contained technical evaluation criteria for selection of subprojects; 7. Approval was obtained from the Gov- ernment and IDA for the selection proce- dure and implementing method; 8. A public marketing strategy was defined, based on awareness-raising in rural areas and distribution of the proce- dures manual to potential providers; Stage Three: Selecting 9. Issue calls for proposals; 9. The call of proposals was cyclical; the partner 10. Evaluate the proposals; once a year there was a bidding process; 11. Select the preferred proposal and 10. A technical committee with civil proponent; society observers evaluated the bids; 11. The preferred proposals were select- ed for implementation; Stage Four: Negotiating 12. Commence negotiations between the 12. Unit prices were standardized, and and entering into contract two parties; there were standard components to be 13. Prepare the draft contract; included in the proposal, so negotiations 14. Reach agreement and prepare a were kept at a minimum; Memorandum of Understanding (MOU); 13 ­ 15. The contracts were standardized, 15. Negotiate and sign the final contract; and were issued by the contract- managing agency (there was limited room for individual negotiation); Stage Five: Implementing 16. Commence the implementation phase; 16. The provider implemented the and Monitoring 17. Regularly monitor performance; subprojects; the contract 18. Ensure compliance with the contract. 17. The government and the contract- managing agency ensured ongoing moni- toring of performance; 18. A final evaluation (exam of learning outcomes) ensured compliance with the contract. The Process of Outsourcing 15 way to obtain funds (where to apply, when, financed. The composition of the selection how, and what is required after receipt of committee included literacy professionals from funds). In principle, the procedures manual is the government as well as representatives from the public sector's tool to state exactly what the providers to ensure transparency and fair- the government wants in terms of how a proj- ness in the selection. The proposals approved ect is to be implemented. Hence, the manual for financing were sent to a parastatal con- needs frequent updates, since the government tract-managing, or "outsourcing" agency policies vary over time. In the description (AGETIP--Agence d'Exécution des Travaux below, the PAPF procedures manual 2000- d'Intérêt Publique), which established con- 2002 will be the main reference.21 The proce- tracts, paid the providers, and managed and dures manual required each provider, in coop- verified the contracts. After obtaining the first eration with villagers, to prepare a literacy installment of the funds, the successful subproject proposal. providers began implementing the subproject. In the case of the PAPF, the private providers During implementation, the providers were were paid to do the following work: responsible for regular follow up and monitor- ing; the government was responsible for over- · In discussion with villagers, establish a all technical supervision; and the contract- service delivery program for non-formal managing agency was responsible for financial education (i.e., a "subproject"); supervision. In addition, a providers' associa- · Recruit, pay, provide initial training, and tion was created to assist in quality assurance in-service training of staff; of implementation, to represent the providers' · Upgrade or construct education facilities; interests, and to help in case of conflict · Procure equipment; between a provider and the other parties. · Implement the subproject (which, in the The selection and implementation process is integrated program usually included post- displayed in Figure 3. According to ideals for literacy and income-generating activities, participation and feedback, it was the benefici- as well as training of the local women's aries who would tell the government what they association's leaders); wanted the providers to do (within basic poli- · (In the integrated program): Initiate a cy guidelines as displayed in the procedures process to achieve sustainability of learn- manual). Hence, the villagers are intentionally ing activities through the training of a placed at a "high" level in this graph. relais, and through the set up of a perma- The subproject had a defined cycle, which is nent learning and resource center in the described in detail in the procedures manual. village; The cycle consisted of the following actions · (In the integrated program): Coordinate (adopted from the procedures manual, literacy activities with other organizations 2000­2002): implementing local development initia- Step 0--Information: Government represen- tives; tatives and the providers' association distrib- · Monitor and evaluate the progress of uted the procedures manual and other relevant service implementation. information to interested providers, and indi- cated the target areas of program implementa- The subproject proposals were checked for tion. Also, information about the selection of accuracy at the local level by government providers was published in local and national staff.22 They were then sent to a selection com- newspapers. The selection of target areas was mittee that provided a technical analysis of all done by local authorities and was based on proposals and decided which ones should be needs. The literacy courses aimed to correct 16 Managing Public­Private Partnership Figure 3: Outsourcing in Senegal Th e providers adapt the subprojects' implementation and curriculum to needs identified in the vi ll ages. Village 1 Village 2 Village 3 Village x Foll ow-up on literacy sector Provid ers (NGO , Providers' Union For-profit, other) Each provider presents a project proposal to the local authorities who verify its accuracy Poli cy guidance, quality control and training Ministry of Prefecture and Education Regional Levels The regional li teracy committee, after pre-selection. sends the proposals to the selection Contracting and financial control committee Contract- Selection Managing Agency Committee A fter (technical) review, a certain number of projects are selected for financing and sent to the C ontract-Managing Agency disparities, so generally the poorest (and most information of the future participants in the illiterate) areas were selected. literacy courses. Step 1--Assessment of needs: The providers Step 3--Check of consistency: The provider prepared subprojects based upon the require- sent the subproject proposal to the inspector of ments in the procedures manual, and selected a education at the department level, who zone of intervention (within the target areas checked the consistency and accuracy of the defined by the government). The subproject subproject proposal with an on-site inspection typically consisted of 20 literacy courses in as in at least 20% of randomly selected interven- many villages.23 The providers and the benefi- tion areas (20% was considered a representa- ciaries assessed the needs of the village by using tive number of sites; it would be too costly to participatory research methods. check all the sites).24 The inspector also Step 2--Subproject proposal: The provider checked whether the proposal was in conform- wrote a subproject proposal, which included ity with the local development plan. the assessment of needs from each village Step 4a--Preselection: All proposals that where the subproject would be implemented. were deemed eligible were sent for technical It also included the name, age and literacy analysis. The selection process had two stages: The Process of Outsourcing 17 The first stage consisted of preselection at the How much time does it take to select regional level, where a committee analyzed the and implement a subproject? capacity of the providers to implement the sub- project. The committee also examined the When setting up an outsourcing project, it is extent of the providers' familiarity with the important to be aware of the time require- intervention area. Any proposal showing that ments of selection and implementation of sub- the providers lacked the capacity to implement projects. Also during the first years, some time the subproject (e.g., lack of adequate instruc- must be allocated for setting up, learning and tors), or had insufficient familiarity with the trial of the process. In Senegal, the system now geographical area, was rejected. appears to be running smoothly. Rigorous Step 4b--Selection: The proposals that were scheduling of activities is needed to avoid any not rejected were sent to the selection commit- situation where potential providers cannot pre- tee for final selection. This committee consist- pare the subproject proposal on time or cannot ed of technical staff members (as opposed to begin implementation immediately upon elected politicians) from the Ministry of Edu- receipt of funds because they received the cation, AGETIP, and the Ministry of Family, funds during the agricultural (rainy) season. Social Action and National Solidarity. Also, Most subprojects are implemented exclusively the committee had two observers representing in the dry season, typically from November to the providers and the beneficiaries. After selec- May. This is the time when the population can tion, the successful subprojects were approved make themselves available for learning activi- by a politically appointed body (usually the ties; in the rainy season they are too busy with Approval Committee), which verified that the agricultural activities (see detailed description right procedure had been followed during the of the timeframe in Annex 1). selection process. Finally, the subprojects were Changes in the agriculture production meth- sent to the outsourcing agency (AGETIP) to ods (especially in North Senegal) and the evolu- establish the contract.25 tion of PAPF's literacy program toward a more Step 5--Establish the contract: Upon integrated poverty-reduction program (see Sec- approval, AGETIP proceeded to contract with tion II on Courses) changed the necessity of the provider and to pay the first installment of implementing the subprojects exclusively during the contract into an account that was opened the dry season. Increasingly, subprojects were by the provider specifically for the subproject. adopting a more cultural calendar, and started Step 6--Implementation: The provider remo- courses after the Ramadan.26 Although literacy bilized the beneficiaries, recruited the necessary activities are currently implemented during the staff, procured the necessary equipment, and rainy season, they are usually not attended by all was responsible for implementing the subproject the participants. Often, the provider-paid course in accordance with the project proposal and the instructor is absent, and the classes are conduct- procedures manual's dispositions for reporting. ed by the relais (the local substitute for the Step 7--Monitoring: The financial monitor- instructor). The classes often focus on repetition ing was done by the outsourcing agency of earlier lessons and special tutoring for weak (AGETIP) that checked that implementation students during this period. Other countries that accorded with the contract. The Ministry of have adopted the outsourcing mode (e.g., Burk- Education did the substance monitoring, i.e., it ina Faso, Chad, Côte d'Ivoire, and Gambia) checked that the courses were adequate. At the have adapted the process to fit their own selec- end of the subproject lifecycle, the ministry tion system and climatic/agricultural seasons. made a final evaluation of the subproject, upon However, most of them have the same core which it was closed by AGETIP. strategy, as shown below:27 18 Managing Public­Private Partnership Table 3: Implementation schedule Dry season Rainy season Dry season Information, participatory Submit subproject proposal, Recruit and train village instructors, assessment of needs in the selection, establishment of contract, procure equipment, begin communities, setting up transfer first funds subproject implementation course design The design shown in Table 3 reduces the The beneficiaries time needed from the submission of proposal to the first transfer of funds--to at most four The beneficiaries were mainly groups of illiter- months. This rapidity avoids frustration ate and poor women living in rural areas. among providers (e.g., uncertainty about Often, but not always, they represented the whether their application will be successful poorest segment of society. They contributed and about start up time). Delays can seriously financially to the course (typically 2,500 FCFA disrupt the implementation phase and cause or about US $3 for the duration of the course). dropout and dissatisfaction among beneficiar- They frequently organized themselves into ies and providers alike. The importance of associations with the intention of improving upholding timelines was the reason why PAPF their household income. The providers sup- used a parastatal agency (AGETIP) for con- ported these associations, and helped the bene- tracting and for paying the providers. AGETIP ficiaries to create them in villages where they was not constrained by the usual government did not exist prior to the literacy course. decision cycles (which were much more time- The beneficiaries and the providers defined consuming). the basic skills components of the course. Lit- eracy and numeracy were mandatory subjects; basic functional skills courses were optional. Who are the stakeholders? Furthermore, the beneficiaries set up a local management committee to ensure correct As noted in Figure 3, the main stakeholders of implementation of the subproject in their vil- the process were the beneficiaries and the pri- lage. This committee was responsible for keep- vate providers that ensured the implementa- ing a registry of participants and an inventory tion of the service. The government was repre- list for the center. Finally, the beneficiaries sented by two agencies, the Project decided on a course schedule, the language of Coordination Unit (PCU)28 for the PAPF and instruction to be used in the course, and some- the DAEB (Department for Literacy and Basic times they helped the provider choose a course Education). Many other government and aux- instructor. iliary agencies participated in the outsourcing and control process, such as the departmental The private providers and regional education bureaus, the preselec- tion and selection committees, the contract- One of the aims of the PAPF was to stimulate managing agency (AGETIP), and the CNEA the creation of new providers. Consequently, (National Committee for the Elimination of the procedures manual had certain measures Illiteracy). to encourage new providers to apply for The Process of Outsourcing 19 funds (e.g., the financing included institution- The interaction between the providers and al support for the provider organization, and the beneficiaries usually followed the following human resource building). This policy proved development: successful, since the number of requests for financing increased from 77 to 368 over only 1. A first meeting with chief of village/ five years (478% increase). Many of the new administrative committee providers were grassroots organizations such 2. Interact with local women's association/ as local women's groups that had previously undertake participatory needs assessment received literacy training from another 3. Create a project management committee provider. Other providers were created from 4. Develop subproject proposal NGOs or associations that split into two 5. Write a contract with the management provider groups. committee The providers did participatory needs 6. Implement the subproject assessments in the villages where they pro- 7. Monitor and evaluate the subproject posed to intervene, and also wrote the subpro- ject proposal. In each village, they signed an The initial assessment, the implementation, agreement with the village's management com- monitoring, and evaluation were--at least the- mittees, so as to clarify their own responsibili- oretically--done in close cooperation with the ties and those of the beneficiaries. The community and the local management com- providers that received financing for the sub- mittee. This was necessary to avoid subprojects project recruited one instructor for each village that were excessively provider-driven, and to of intervention who was responsible for proj- ensure village ownership of the subproject. ect implementation in that village. The charac- What are the core characteristics of a "good teristics of a typical provider in the PAPF are provider?" A longitudinal study of the PAPF illustrated in Box 1. showed a significant correlation between the Box 1: Characteristics of a typical provider in Senegal: · Is a local organization; · Mainly provides literacy courses, but may also occasionally be involved in providing micro-credit schemes, income-generation and health projects; · Has between one and six years of experience in project management; · Has a motorcycle and an office with a computer and a telephone; · Has an administrative staff (86% of these are male) composed of a manager, an accountant, and two supervisors who are responsible for monitoring the village activities; · Has one course instructor in each subproject village, i.e., each provider has about 20 instructors, of which 60% are women (93% have lower secondary school and 27% have higher secondary school diploma) and 40% are men (44% have lower secondary school and 10% have higher sec- ondary school diploma); · Many (28%) of the instructors are young and without experience, and 37% have only one to three years of experience; · These instructors are trained in adult education methods for an average of 19 days before taking up their position in the subproject village; · Each subproject provides literacy courses for 300­600 beneficiaries. Source: Study of 15 providers in longitudinal study, DAEB, 2001. 20 Managing Public­Private Partnership existence of a local management committee at tion about the annual bidding process. Neither the onset of the subproject and its learning suc- has there been much complaint about the cess (DAEB, 2002). Using the numbers of this impartiality of the selection. same longitudinal study, no significant correla- It is debatable whether the providers' tion could be established between teacher organization is merely an interest group for experience and learning achievement, or providers themselves or whether it could also between provider type (village association, to some extent act as a body concerned with NGO, religious group) and learning achieve- the quality of services rendered. In a partner- ment. It seems therefore that one core feature ship approach, it is necessary to have a body of the program was the involvement of the that not only can interact with the government local management committee and the local on behalf of the providers, but also function as women's association in the subproject. A an arbitrator in case of charges of unethical "good provider" would thus be the provider behavior against individual providers. Moore that involved the beneficiaries' organizations in & Stewart (2000), in their analysis of NGOs, the subproject formulation and execution. In describe the functions of the providers' associ- other words, the beneficiaries' input seems to ation in terms of corporate governance, and be as important, or even more important than argue for the set up of an association that the providers' input for meaningful learning to ensures control of standards for ethical con- take place. duct and accountability. According to Moore & Stewart (2000, p. 88) the implementation of The providers' organization such norms has the following results: The providers' organization (CNOAS) was 1. "By providing clear standards and prac- created with assistance from PAPF and the tices of accountability and transparency, CIDA-financed PAPA project when the Sene- [the norms] take much of the sting out of galese government adopted its outsourcing pol- the charge of non-accountability [of icy in the beginning of the 1990s. CNOAS had NGOs], and much of the force out of the as its aim to represent the providers' interests argument that government must step in to and to provide training to new providers in regulate NGOs because [no one] else is various areas including financial and human doing the job. resource management. In addition, it was 2. They ease the problem of introducing hoped that it could be of help in ensuring need- `institutionalized suspicion' mechanisms ed control in the evolution of the sector. This into NGOs that have out-grown their control function included the establishment of management and leadership capacities. professional ethics for the provider group. There is now an objective argument for CNOAS has had limited success in achieving doing the right thing: `unless we do it, we these aims, and has instead functioned as an shall lose our membership of and recogni- information-providing agency. During the tion by the NGO association.' selection process, it represented the providers, 3. They provide donors with some kind of since two observers from CNOAS participated quality rating that can be traded off during selection sessions. The two main against more expensive, detailed, intru- achievements of the CNOAS were to provide sive inspections or output evaluations. .... information and to ensure impartiality in the If membership of reputable NGO associa- selection process. These tasks seem to have tions becomes the norm, then the reputa- been well performed, since there have not been tion (and financial health) of the NGO many complaints about the lack of informa- sector as a whole can only improve. The Process of Outsourcing 21 4. They require the creation of collective The agency had four main functions in the organizations for self-regulation that will PAPF: (a) to establish the contracts with the have an incentive to provide the collective providers; (b) to train them in financial man- service that their members cannot effi- agement and bookkeeping; (c) to pay them for ciently provide for themselves. [...] NGO their services; and (d) to verify that the funds Associations could provide: staff training, were spent correctly. Theoretically, AGETIP shared management consultancy services, was supposed to have local controllers who insurance, or information about potential could check regularly the implementation sta- funding sources." tus of each project. In practice, the organiza- tion had problems setting up a viable monitor- Is it necessary to actively support the forma- ing system. One main constraint was that the tion of an independent providers' association agency was Dakar-based and did not have when a program is being established? Moore offices throughout the country. The set up of & Stewart insist that the provider association local offices is now being planned. must be financed solely by the providers them- The main success of private contract-man- selves, and by the services it provides to its agement was the rapidity of processing con- members (Moore & Stewart in Pearce, 2000). tracts and of transferring payments to the However, as the Senegalese case has shown, providers. This achievement is significant, such associations may need initial start up cap- since for other projects the use of government ital for institution and capacity building. Per- procedures invariably has caused funding to formance of the four functions described by become "stuck" somewhere and caused Moore & Stewart has not been successfully delays to implementation. However, different achieved in Senegal, probably because the contract-managing agencies are not equally providers' organization has been staffed with efficient. The choice of a contract-managing non-salaried (elected) providers whose main agency is therefore an important decision, job it is to oversee their own association. and the government and/or donor agency need to monitor the agency's performance The contract-managing agency closely. (AGETIP) The Project Coordination Unit The contract-managing agency, the Agence (PCU) d'Exécution des Travaux d'Intérêt Public (Agency for Implementation of Public Works-- The PCU was the government institution with AGETIP), is mainly involved in contract-man- main responsibilities for implementation of aging of public construction works, such as the PAPF. It had financial, managerial and building roads. With the PAPF, it became technical responsibilities, oversaw the overall involved in the social sector and allocated two project activities, ensured planning and mon- controllers to check the local implementation itoring, and was responsible for reporting to of subprojects. These two staff members, with the government and the external financing some administrative support, had the core agency. It also arranged training at provider responsibilities for the management of the and government levels. The PCU had a cen- project from AGETIP's side. But this was not tral role in the project implementation (see sufficient to keep up with the rapid expansion Figure 4). of the project, and soon lead to infrequent on- The PCU only dealt with PAPF issues. In site inspections of the implementation status of 2003, during a period of transition to a more subprojects. integrated outsourcing program (i.e., main- 22 Managing Public­Private Partnership Figure 4: Coordination role of the PCU DAEB/Gvmt. World Bank PCU Providers AGETIP streaming of the donor aid to the sector), it The Department for Literacy and maintained its coordination role. A merger Basic Education (DAEB) with DAEB is planned for 2004; the specifics of this merger, however, are not clear. The PCU The DAEB was at first not too involved in the had a small number of core staff (the most PAPF, except for evaluation and statistics pur- important being the director, who had mainly poses. Increasingly, the department became a political role; and the vice director, who had involved in the monitoring of the subprojects technical responsibilities and organized moni- and in related policy issues. Its responsibilities toring trips, seminars, training sessions, work- were both of a political and technical nature: shops, etc.). The project office kept an exten- sive database on the project. · Establish a national literacy plan; · Propose geographical target zones and The main information flow went through target beneficiaries for each literacy pro- the PCU. The contract-managing agency gram; (AGETIP) and the government (DAEB) also · Carry out cost-efficiency studies for vari- had a monitoring/evaluation role in the proj- ous subprojects; ect. The World Bank had a financial agreement · Update statistics; and link with AGETIP for processing pay- · Evaluate curricula; select curricula that ments. The DAEB was sometimes interacting should be used by subprojects; directly with AGETIP, but the feedback loops · Monitor and evaluate subprojects. from providers and beneficiaries usually went through the PCU (and sometimes through At present, the government has tried to local government). There was less feedback merge the PCU fully into DAEB, so as to from Providers to AGETIP and DAEB. ensure a better coordination of literacy activi- The Process of Outsourcing 23 ties. Such a merger, however, has proved diffi- and gave it ratings according to a well-defined cult since the government currently lacks the matrix (see Box 2 on eligibility for funding for ability to perform proper monitoring and eval- a subproject). In the selection process the ben- uation of the program. eficiaries were present only through the pre- liminary project formulation, which was based Selection committees on participatory approaches. A more active role of the beneficiaries could improve the As noted above, the providers submitted the quality of service delivery. Such a method is subproject proposals to the public education planned in The Gambia (see Figure 6). service at the department level. The head of education in the department (Inspecteur de National Committee for Literacy l'Academie) checked the accuracy of the pro- posal (e.g., the legal existence of the provider, The National Committee for Literacy (CNEA) accuracy of information about beneficiaries). was composed of representatives from differ- The departmental head of education was also, ent ministries that were working with develop- at the implementation stage, involved in the ment and literacy-related issues. Civil society monitoring and evaluation of the subproject. was represented through the providers' associ- As opposed to the educational service at the ation(s) and through organization(s) represent- department level, the selection committees ing the beneficiaries. The role of the committee were not permanent institutions but convened was mainly political. It took part in policy for- once a year to decide which projects would be mulation on literacy-related issues, mobiliza- financed. Preselection committees were operat- tion of funds, coordination with donors and ing in each of the five target regions for project other interested parties, and in an annual implementation. Each preselection committee review of the results. was composed of the regional director of liter- acy, the regional director of community devel- opment, the head of the regional council, the How to ensure transparency in head of the regional service for local develop- selection of providers ment and two observers who were representa- tives of providers. The concept of public-private partnerships has The (main) selection committee was com- been frequently used to describe the outsourc- posed of technical, non-political staff, repre- ing system as it is practiced in Senegal. The senting the Ministry of Education, the Ministry word "partnership" refers to a situation of of Family and Social Action, and AGETIP. joint action for shared purposes based on Additionally, it had two observers who repre- strong mutual trust. This appears to have been sented the providers (selected by the CNOAS), largely achieved in the Senegalese PAPF. The but who did not have the right to vote. The sector was characterized by the absence of two observers were supposed to ensure fairness complaints from providers about unfair treat- and transparency in the selection process. The ment or about serious problems in the transfer committee's objective was to make the selec- of funds. Also, few complaints were received tion as technical and apolitical as possible. If about services promised but not delivered. the committee were composed of politicians, Very few reports from monitoring or evalua- one might suspect that the selection of tors indicated serious faults in the implementa- providers would become politically biased. tion process. The low number of complaints During selection, members of the committee gives support to the view that the mechanisms individually read each subproject proposal, of the outsourcing approach of the PAPF were 24 Managing Public­Private Partnership Box 2: Criteria for eligibility for a subproject The subproject proposal was analyzed and evaluated by using the following criteria and points. (In order to be eligible, the subproject proposal must score at least 60% of the maximum points indicat- ed below, i.e. it must score at least 114 out of 190 points): (1) Capacity of the provider (50 points) · Legal existence; · Prior experience in at least one education or development project; · Have necessary human resources to implement the proposed subproject; (2) Knowledge of the implementation sites (40 points) · Implementation zone corresponds to target zone for the PAPF; · The geographical, social, economical, and cultural aspects of each site are specified; · A list of participants is provided indicating the names of the beneficiaries (at least 75% of them should be women in the 15 ­ 39 age group); (3) The subproject program (50 points) · The indicators for progress correspond to PAPF's stated goals and are clearly formulated (course type and time, learning methods, etc.); · Responsibilities for the implementation of the activities are well defined; · The beneficiaries have set up a management committee; · Questions of sustainability of the activities are addressed; (4) The management system (40 points) · The management system includes a subproject implementation plan; an administrative and tech- nical monitoring system; and an evaluation plan; · The budget is well detailed and does not exceed the specified limits; · The contribution of the beneficiaries is agreed upon and does not exceed the specified limits (2,500 FCFA); (5) Institutional issues (10 points) · The provider has established a plan for the utilization of institutional support for the provider and the beneficiaries (should not exceed 10 % of total costs). basically sound, i.e., the project achieved a rea- tion of the proposal, the reason why it was sonably transparent and fair selection process rejected. and quick transfer of funds. The manual clearly stated what the proce- The partnership was sound for several rea- dures were for all stages of implementation. sons: Most project information was easily The provider knew which reporting and moni- obtained. This was particularly true for the toring arrangements were required, and how selection process, since the criteria for eligibili- the beneficiaries would be assessed at the end ty were clearly stated in the procedures manu- of the project. Moreover, the procedures man- al (which was freely disseminated). Also, there ual indicated the possibility of sanctions for was a procedure for appeal regarding rejected faulty implementation. Such rules enhanced subproject proposals. The manual's clearly the impression of rational decisions; the stated information made the providers aware provider should never have the feeling of being of what their duties were, and in case of rejec- the victim of arbitrary controls or of being The Process of Outsourcing 25 sanctioned for unclear or personal reasons. In · The procedures manual was freely distrib- reality, however, sanctions were never used. uted to all interested parties; The lack of sanctions for faulty implementa- · Prospective providers could appeal rejec- tion therefore ran the risk of rewarding fraud- tion of proposals; ulent behavior. In a few cases, dishonest · The government appointed a "non-politi- providers were reselected to implement new cal" selection committee, which selected subprojects after embezzling funds from previ- projects on their technical merit, not on ous subprojects.29 the political affiliation of the provider. One additional factor of transparency was linked to the composition of the selection com- mittee(s), which aimed to be as "non-political" How to set up a good monitoring and as possible (even though the committee was evaluation system politically appointed). Selection was to be based on technical criteria judged to be appro- The 2004 World Development Report, writing priate by literacy professionals. Had politicians on social service provision for the poor in gen- participated directly in the selection, judgment eral, points to the need to strengthen account- based on technical criteria would have been ability in three key relationships in the service more difficult since there would then have been delivery chain: "between poor people and pressure to use the selection as a political tool. providers, between poor people and policy- Such political interference occurred at least makers, and between policymakers and once, during national elections, when pressure providers" (WDR, 2004, p.1). Good evalua- was put on the selection committee to change tion and monitoring can improve accountabil- the procedure into a more political process ity. In a recent review of literacy programs in a instead of purely a technical one. As a result, number of countries, the World Bank found the implementation decree (which confirms the flaws in the monitoring and evaluation designs financing proposal from the selection commit- in nearly all the programs covered: "Overall, tee) was not signed on time and the funding effectiveness indicators are sparse, and it is was delayed, resulting that year in a higher unclear how many participants entered a pro- dropout of learners than usual (since the liter- gram and how many emerged literate. [...] acy courses began too late to be finished by the Implementing agencies and NGOs tend to have onset of the rainy season). This political inter- weak and possibly unreliable record-keeping" ference also led to the reselection of some (Abadzi, 2003, p.10). Consequently, monitor- providers who previously had shown unsatis- ing and evaluation were problematic issues in factory results. This pressure on the selection the PAPF. process was temporary and has apparently not Two basic functions of monitoring and eval- been experienced since then. uation are necessary: (i) to hold providers to In conclusion, the following issues have been their contract and to make sure that only good of particular importance in building partner- providers are retained; incompetent or corrupt ships and to enhance the transparency and providers will not have their contracts renewed validity of the outsourcing approach used in and (ii) evaluation of the program as a whole Senegal: (e.g., PAPF) in terms of overall outcomes and impact. PAPF had originally foreseen several · The "rules of the game" were clearly stat- types of control, and at several levels. The first ed in the procedures manual, in particular four types described below belong to function the criteria for selection were well (i) above since they evaluate specific subpro- explained; jects; the evaluation by the providers' organi- 26 Managing Public­Private Partnership zation (point 5) was supposed to perform both have a car, sufficient funds to cover gaso- function (i) and function (ii); and the longitu- line expenses, etc.). dinal studies mentioned in point 6 belong to function (ii). 4. Monitoring and evaluation by the providers, who were required by contract 1. Financial monitoring by the outsourcing to recruit two supervisors for regular agency: Regular checks that funds are monitoring of the progress of the project. being disbursed according to the plan, In PAPF, the providers' monitoring of and that the services paid for have been their own sites seems to have been, in gen- delivered (i.e., check that the beneficiaries eral, successful. participating matched the number paid for). However, in the case of PAPF, the 5. Monitoring of the provider sector as a contract-managing agency failed to whole by the providers' organization: achieve a good monitoring and evaluation Data gathering and interaction with system. Its performance was affected by providers in order to identify and prevent work overload due to rapid expansion. cases of corrupt practice among Further, the agency lacked the infrastruc- providers. The providers' organization ture and equipment needed to perform did not carry out this function. appropriate management, supervision and monitoring of the financial aspects of 6. Cost-benefit analysis and longitudinal the subprojects. studies by the Ministry of Education: these are activities aiming to identify fac- 2. Quality monitoring by the Ministry of tors of success and failure in the Education: Regular checks in order to approaches used by the providers. In ascertain that the quality of the purchased PAPF, outsourcing was seen as an evolv- service (i.e., literacy courses) is adequate. ing process, where the providers would The quality monitoring by the DAEB and improve their subprojects year by year, PCU did not always cover a sufficient and where the possibility of reselection number of courses. would be an incentive for providers and help build trust. However, the govern- 3. Assessment of learning outcomes by Min- ment did not set up studies to assess the istry of Education (Final Exam): A final impact of the subprojects early enough to assessment of skills acquired by each par- gather base line data though the intent ticipant was to be carried out. Such had been to do so both in "experimental" assessment is essential, since it shows and "control" villages. After implementa- whether the "product" corresponds to the tion of the literacy course, both experi- contract (especially the level and quality mental and control villages were to of learning, as well as the dropout rate). receive a post-test, which would make In the case of PAPF, local education offi- possible an analysis of the impact of the cers conducted the exam. However, literacy courses. Such a design would because of workloads it was often done have led to an estimation of economic and considerably after the course was com- other benefits of the PAPF, and made it pleted, when only a few learners were possible to assess its cost-effectiveness.30 available to participate. Also, the local education offices lacked funds for carry- At all levels, the project suffered from lack of ing out the assessments (e.g., they did not data, or from contradictory data. Better capac- The Process of Outsourcing 27 ity and equipment would have been needed to with the villagers the establishment of trust is furnish adequate data, especially given the particularly important, since this is the only unexpected rapid development of provider way to ascertain whether the service delivery is organizations, which multiplied more than sev- progressing in a satisfactory way. If the benefi- enfold during the six years of project imple- ciaries feel that there are serious deficiencies in mentation. the subproject, they must be able to communi- It is important to be realistic in the planning cate their dissatisfaction to the local adminis- of needed monitoring and evaluation. If the tration. In such cases, the cause for discontent state sector (or the outsourcing agency) lacks can be investigated, and errors from the capacity to carry out the needed evaluation provider's side can be corrected (or else the and monitoring tasks, these functions must provider must be replaced!). It is important themselves be outsourced to a third party. The that prior to project implementation, the local use of a third party may have the added bene- administration makes the beneficiaries aware fit of greater evaluative impartiality than what of both their rights and obligations vis-à-vis is achievable by government agencies responsi- the provider, so as to prevent unjustified com- ble for implementation. A flowchart of the plaints (this was not done in the PAPF). Such monitoring and evaluation system of an out- knowledge will enable the beneficiaries to sourced project as it was intended to function know if the implementation corresponds to the in the PAPF can be conceptualized as shown in contract, and whom they can contact in the figure 5. administration in order to complain if the As illustrated in Figure 5, the main evalua- course fails to meet the contracted require- tion and monitoring interaction occurred at ments. In PAPF, in a few cases, the providers the local level, where the providers were mon- clearly failed to implement their subproject itoring their own subproject. In the local adequately. However, participants merely school administration, the officer in charge of ceased attending; in many cases, the parties literacy education was supposed to verify the involved in project management (PCU, DAEB, progress of the subprojects, by interacting with AGETIP) did not become aware of the prob- both providers and villagers. In the interaction lem until the final evaluation of the subpro- Figure 5: Monitoring and evaluation system for an outsourced project31 Beneficiaries Provider Public Sector Local level Public Sector Outsourcing Providers' Central level Agency Organization 28 Managing Public­Private Partnership jects. Such a situation could have been avoided Why is the interaction between if the beneficiaries had known where to chan- villagers and providers so important? nel complaints and concerns about the provider. In the early stages of the PAPF, some providers Another issue in need of monitoring is the prepared proposals without having had con- background of the enrollees. Literate and semi- tact with the intended beneficiaries. Many sub- literate people often wish to enroll in literacy projects were implemented by simply sending courses. They may wish to improve their liter- an instructor to the designated village, who acy skills but may also be attracted to the other would then interact with beneficiaries and start skills taught or to the course as a social event teaching. The contact between the provider for adolescent and adult women in their vil- and the villagers was distinctly weak. This lage. In the case of PAPF, this was contrary to often resulted in top-down, excessively the guidelines, but it occurred quite frequently. provider-driven projects. Since, in most cases, prior education used In keeping with revised PAPF strategies, the French as the medium of instruction, most, if procedures manual for 2000­2002 emphasized not all participants were considered illiterate in the need for institutionalized contact between their national languages at enrollment. This is the providers and the beneficiaries. The one reason why the PAPF accepted participants provider was obliged to recruit two supervisors who had previously been to primary school. to follow up activities in each of the villages Beneficiaries who had already been to school, where the subproject conducts a course. Also, not surprisingly, performed better than those the beneficiaries were involved in the concep- who had not (DAEB, 2002). Any attempt to tion phase of the subproject to ensure that the evaluate the learning gains in literacy courses course better reflected their interests. The ben- needs to take into account the skills possessed eficiaries set the schedule of classes, decided by the learners prior to the course. the language of instruction, and decided on At the central government level, financial topics for basic skills instruction. Sometimes, and technical monitoring, and evaluation are they even chose the instructor. In this way, the necessary to aid planning, ensure smooth func- PAPF sought to strengthen local ownership of tioning of the project implementation and dis- the project. courage fraud. Both monitoring and evaluation Furthermore, in the integrated program the can be outsourced when these functions cannot beneficiaries had to choose a teaching assistant otherwise be performed satisfactorily. The (relais) from the village. The intention was that technical gains from outsourcing any strategi- this person would be able to continue develop- cally important evaluative functions at the cen- ment activities beyond the term of the financed tral level need to be weighed against some loss subproject, thereby contributing to local own- to the program managers in terms of direct ership and to sustainability of activities. The knowledge of the conditions (and persons) teaching assistant was a volunteer who did not being evaluated. The monitoring and evalua- receive payment from the subproject. In many tion functions of the operator's organization, cases, this lack of incentive for the relais halted the local administration, and the providers literacy activities in the village some time after themselves (of their own subprojects), should the subproject came to an end. in most cases not be outsourced, because they Local ownership and sustainability of the are based on continuous and long-term rela- activities were ensured when they were based tionships. One core element in such relation- on existing village organizations. The involve- ships is to build trust between the parties con- ment of local women's organizations in project cerned. implementation was essential for the subpro- The Process of Outsourcing 29 ject's success (DAEB, 2002). The women's as the arrival of an outside expert can actually organizations normally followed up courses generate distrust. and monitored attendance. They encouraged In The Gambia selection process, the com- regular attendance, and in some cases intro- munities themselves decide which services they duced fines for late arrival or non-attendance. need; and they participate in the selection of In some instances, the village women estab- the provider and in the "purchase" of services. lished "hygiene police" to inspect the villagers' The 2004 World Development Report (WDR, homes and imposed fines for disorderly house- 2004, p. 6) distinguishes between the "long keeping. Women's organizations were also route" of accountability--where clients as citi- involved in income-generating activities. DAEB zens influence policymakers, and policymakers findings indicated that the chance of subpro- in turn influence providers; and the "short ject success was boosted by the beneficiaries' route" where the beneficiaries have direct organizational capacity before and during power over service delivery (as they ideally project implementation, and by the connection will, under competitive market transactions). they had with the provider organization The system as it is set up in Senegal largely per- (DAEB, 2002). In general, the closer the net- tains to "the long route." As it is set up in The working--or real partnership, the better the Gambia, it exemplifies an attempt to shorten results. The data showed a statistically signifi- the route of accountability by strengthening cant correlation between the existence of a the communities' power over providers. The local management committee (generally based WDR recommends the latter approach: "serv- on a women's organization) and the learning ice outcomes can be improved by strengthen- outcomes of the course. ing the short route--by increasing the client's Some providers outsourced participatory power over providers" (World Development planning activities and subproject formulation. Report 2004, p.6). The system in The Gambia In such cases, a consultant specializing in par- is set up as shown in figure 6. ticipatory approaches formulated the proposed Step 1: The regional authorities select the subproject and set up an agreement with the villages to benefit from the program through a villagers. This practice was generally accepted demand-based process. The communities that by PAPF management due to the fact that there will benefit from the program select one repre- were high quality providers who lacked experi- sentative from the village development com- ence in formulating a good subproject docu- mittee to represent the community in the selec- ment, and that the hiring of a participatory tion and implementation process. This villager rural appraisal (PRA) specialist indicated com- is given training about the program and selec- mitment by the provider to ensure a plan of tion procedure. good quality. On the other hand, establishing Step 2: The project is tendered for. Each trust between beneficiaries and providers was cluster of 10 selected villages is considered a especially important, as such trust is built by subproject. Each provider initiates contact direct communication between the parties with the communities to discuss a possible already in the project formulation stage. By agenda for intervention. Based on the negotiating directly with the villagers, the exchanges with each community, the provider provider could better explain the limitations of sets up a subproject proposal. the subproject and get a hands-on feeling of Step 3: Together with (technical) representa- what is going on in the village. Getting to tives from the regional education office and know the provider's staff also generates trust from the providers' organization, a representa- between the villagers and the provider, where- tive from each village organization form a selec- 30 Managing Public­Private Partnership Figure 6: An alternative system: The Gambia selection process Community 2 Community 1 Community x Step 1: Local admin. interacts with communities Step 2: Tendering for provision Step 3: Selection of Provider Step 5: Contract- Step 4: National managing agency Approval establishes contract Committee approves selection tion committee that identifies the most desir- site, duplicating each others' activities). The able provider to implement the subproject. selection focuses more on the provider's Step 4: The subproject proposal is sent to a "real" qualities, and less on the ability to national approval committee that ensures that write a good proposal. A further advantage is the correct procedures have been followed in the communities' negotiation with the the selection of the provider. providers during project formulation. The Step 5: The subproject is sent for financing. resulting selection and formulation process is The contract-managing agency immediately less "top-down" and a more client-managed proceeds with the contracting of the selected process. However, this method has the disad- provider, and transfers funds for implementa- vantage of being more complex than the prac- tion of the first phase of the subproject. tices used in Senegal (and necessitates compli- The main advantage of this procedure is cated logistics and communication that the selection of communities is more arrangements). As of spring 2004, it remained carefully considered, and one avoids the an untried approach since program imple- problem of "creaming off", i.e., that mentation is expected to start in mid-2004. providers only choose "easy" locations. Also, The Senegalese model is also being revised problems of duplication are avoided (i.e., and evolving toward a more client-based intervention of several providers in the same selection process. The Process of Outsourcing 31 How does decentralization affect · Decentralize some of the monitoring and the program? evaluation functions; · Ensure decentralized representation of the Senegal has three main administrative levels: contract-management agency. commune, department, and region. The last level is the newest one, and is part of a democ- Following these recommendations, a series ratization and decentralization process, since of modifications were undertaken in the project the "Regional Council" is elected locally and organization, resulting in the administrative has assumed many responsibilities previously process described above in Section I.2 and I.3. managed by the central level. The main changes were seen at the regional PAPF was, at its conception phase and until level, where a preselection committee was tak- 1999, a centrally executed project. During the ing over an important part of the selection mid-term evaluation, a series of studies were process, and at the department level, where the launched to analyze how the project could be person responsible for education became direct- integrated into the government's ongoing ly involved in project monitoring and evalua- decentralization process. During a workshop tion. In particular, the end-of-training evalua- in Bougainvillées on September 1-3, 1999, the tion of the beneficiaries' literacy and basic skill decentralization theme was studied from dif- knowledge became a responsibility of the local ferent angles. The main problems, as identified education division. Also, a new, "integrated" during the workshop are as follows (PAPF, subproject style was developed. The notion of Partage, November 1999): integration was both technical and geographi- cal, and took into account the local situation in · The local administration felt excluded a better way. The subprojects were to be part of because of the centralized implementa- the local development plan, and therefore need- tion of PAPF; ed to respond to local requirements and priori- · The local administration did not have ties. Such integration led to a better coordina- sufficient knowledge of the advantages tion between PAPF's subprojects and other of literacy in terms of socio-economic development projects in the area. development; · The providers were operating independ- ently from the local school administra- Conclusion and recommendations tion--there was little interaction be- tween subprojects and schools (although The processes of obtaining funds and of con- both were performing the same kind of trolling the use of funds are two core issues of activities). the outsourcing strategy. If these processes work well, the desired activities are likely to be imple- Based on these problems, a series of solu- mented (provided that the amount contracted tions were proposed, including the following: corresponds to the work to be done). Clarity and transparency are key requirements for suc- · Transfer of some financial and material cess in setting up these processes. The PAPF per- responsibilities to the local level (i.e., to formance was based on clear guidelines and a the department and regional levels); well-defined procedures manual. The main les- · Strengthen communication with the three sons learned from the project are as follows: local levels (especially by providing more information and creating awareness · The outsourcing of financial management about literacy); led to quick processing and transfer of 32 Managing Public­Private Partnership funds, facilitating rapid implementation work. That would have simplified on-site and expansion of the project. verification, and ensured a more efficient · The project's implementation manual (or financial monitoring. "manual of procedures") clearly delineat- · Monitoring and evaluation should have ed the process for obtaining funds, and been linked to enforcement sanctions proved to be an invaluable tool for the when dishonest behavior occurred. In the providers. It was prepared through a par- PAPF, dishonest behavior would be eco- ticipative process involving all the stake- nomically rewarding, since embezzling holders, including representatives from providers could keep all their gains and civil society. The first manual was tested even be re-elected for "implementation" through setting up a few pilot subpro- of new subprojects. Such rewarding of jects, and it was subsequently revised dishonest behavior is thought to be the annually. This helped make the manual a single greatest cause of substandard living document and enhanced its value as results.32 a guidebook for providers. · All monitoring and evaluation activities · The inclusion of a representative from the that could not be successfully carried out providers' association on the selection by the government and/or the contract- committee was important for ensuring managing agency should have been out- transparency in the selection process. sourced.33 Also, the providers' organization dissemi- nated information, e.g., procedures man- For a checklist on the feasibility of setting up uals, to interested providers. a public-private partnership, see Annex 3 · The providers' association CNOAS was (taken from the International Finance Corpo- not able to undertake many training and ration's toolkit on partnership in education). It monitoring activities. One reason for this is crucial that the formulation phase of new may be that CNOAS is managed by pro- projects based on outsourcing take into fessional providers whose priority is to account the role of the different parties and the strengthen their own provider operation, transmission of information between them. rather than devoting effort to sector-wide Also, new projects should plan the schedule needs. from information and awareness-raising to · Transparency and fairness in resource completed implementation of subprojects allocation were essential to build trust in inclusive of the selection process, the contract- the partnership. ing and financial management process, and the · Outsourcing requires a good monitoring monitoring and evaluation process. In setting and evaluation system. It was essential up these processes, it is important to define the that the growth of the sector (i.e., the beneficiaries' role in a comprehensive way. The financing of additional providers) World Development Report (2004) emphasizes matched the monitoring capabilities of that "Clients could play two roles in strength- the control instances. This was a problem, ening service delivery. First, for many services, especially for financial monitoring. clients can help tailor the service to their needs, Hence, a better follow up by the govern- since the actual mix [of services] cannot be ment of AGETIP's work could have specified in advance [...]. Second, clients can improved the contract-managing agency's be effective monitors of providers, since they performance. are at the point for service delivery" (WDR, · AGETIP could itself have outsourced 2003, p. 18). These two roles should be clear- parts of its supervision and monitoring ly addressed in the procedures manual. SECTION II The Courses T he definition of literacy in its simplest extension of adult literacy programs. It can be form is the ability to read a short state- more widely conceived to cater to all those ment. The definition of "functional" who have gone to school as well as for (the literacy, however, is linked to a person's usually smaller number of) those who have ability to use the skills in reading and writing. acquired literacy skills in a program especially A much-cited definition of functional literacy designed for adults. was offered by UNESCO in the early 1960s as This section will explain literacy and post- "the possession by an individual of the essen- literacy from an outsourcing point of view. The tial knowledge and skills which enable him or implementation of outsourced projects raises her to engage in all those activities required for questions such as: effective functioning in his or her group and community and whose attainments in reading, · How do the subprojects take into account writing, and arithmetic make it possible for such issues as the beneficiaries' work him or her to use these skills toward his or her schedule, and their professional occupa- own and the community's development." Lit- tions? What are the characteristics of eracy training is frequently provided in a step- "good" courses? wise progression, starting with functional liter- · Since the quality of providers varies, the acy classes and moving to supplementary quality of subprojects will also vary.34 "post-literacy" actions that may include circu- What are the characteristics of subpro- lation of newspapers locally, provision of vil- jects and providers that deliver the best lage libraries, and skills training applied to results? income-generating activities. Post-literacy sup- port can be provided in a work-related context rather than in special classes, and by other pro- What are the characteristics of fessionals as well as by literacy practitioners. "good" courses? Sometimes literacy training is also included in initial or concurrent teaching of livelihood In PAPF, the providers were autonomous in skills (see Oxenham et al., 2002). Post-literacy their choice of curriculum within guidelines adult education need not be organized as an issued by DAEB. The DAEB produced a list of 33 34 Managing Public­Private Partnership government approved booklets and curricula, in forms. Generally, the providers said that they and the providers were free to procure the teach 300 hours of literacy skills to the partici- materials they thought would best fit local pants, but they did not provide any breakdown needs as to language and subjects covered. Ide- on the time allocation according to subject ally, it was thought, the beneficiaries would (writing, reading, arithmetic). Most providers choose their curriculum; but in many cases the indicated that they teach reading and writing provider made the choice for them, since it was for an average of 160 hours and other subjects easier and cheaper to use the same instruction- for the rest of the time. Arithmetic was divided al materials for all courses run by the same into three levels: level one taught counting, the provider (each provider is implementing four arithmetic operations, and basic problem between 10 and 20 courses in as many sites). solving; level two taught use of a calculator; In a longitudinal study of 60 courses made and, level three taught geometry (surface and by DAEB, the following characteristics were volume), complex numbers and fractions. Very noticed (DAEB, 2001): few providers taught anything beyond level two. Eighty percent of the instructors only gave · Only 38% of the courses had a formally level one lessons. defined learning program, i.e., a curricu- In a review of cognitive research for develop- lum document indicating a plan for the ing countries, Abadzi (2003) notes that literacy periodic (usually monthly) progress of training for adults is often inefficient because what learning should cover; the recipients of training cannot read at the end · Half of the courses did not have a of the course, or in some cases that he or she timetable; cannot read fast enough to make the reading · Only half of the courses (53%) had ade- skill useful. To improve classes, she recommends quate instruments for follow up by the reading in small groups (reciprocal teaching) instructor (student registry, booklets for rather than reliance on the traditional method preparation of classes). where one person reads and the other repeats (the latter is the most frequently used method in The learning itself could be divided into Senegal). This means that the teacher could ben- three categories: (i) literacy skills, which efit from training in group-work techniques. include reading, writing, and basic arithmetic; Further, among the exercises Abadzi recom- (ii) basic functional skills, which include mends are the following good practices: health/hygiene, agriculture, and environment; and (iii) technical training, such as bookkeep- · Learning to count using local money and ing, management and entrepreneurial skills. transactions; discussing how to avoid being cheated. Literacy skills · Simple visual tests to determine which learners might need to sit nearer the The literacy classes taught the beneficiaries blackboard. (These could be included in basic reading and writing skills, as well as arith- one page of a textbook.) metic. Most of the themes in the instructional · Asking readers to bring in materials they booklets were linked to aspects of rural life in would like to read. settings that were familiar to the participants. A few providers (about 7% at the beginning of Basic functional skills the longitudinal study) added functional learn- ing to the literacy classes, and taught the bene- Most PAPF courses included lessons on health ficiaries some basic information on how to fill and hygiene (93%), agriculture (80%) and The Courses 35 environmental issues (67%). The weight to be nation of children and animals, practices accorded to each of these areas was to vary involving risks for HIV/AIDS transmission). according to the needs expressed by the benefi- ciaries. In principle, the course was to teach such skills for 150 hours, but the current Lon- The integrated approach gitudinal Study (DAEB, 2002) has so far pro- vided scant data on implementation. The The government tested several different types method prescribed was to use the beneficiaries' of literacy education. The PAPF program experience as a basis for discussion in order to emphasized a so-called "integrated change behavior (especially concerning vacci- approach" (see box 3) as opposed to other Box 3: The principles of the integrated approach (adapted from Sall & Diagne, 2001) 1. Political Integration Previously, the requests for financing did not take into account the local context; in particular, the local economic and social development plans. Requests under the integrated approach should take into account: · The local administration's development plan, particularly all actions that are emphasized in the program of the elected local government. · Local planned or ongoing projects, and development services that are being planned or already under implementation (rural development services, water provision, environmental plans, etc.), as well as NGO interventions (micro-credit programs, other). The political integration of the subprojects requires: · The provider's thorough knowledge of local development actions and interested parties; · Communication with these parties to investigate possibilities for future cooperation; and where appropriate, establishing formal agreements of cooperation; · The set up of subprojects with clearly outlined objectives. 2. Geographical Integration The subproject's learning program should be of assistance to the village's development plans. 3. Technical Integration The integrated approach should not be a simple integration of literacy and post-literacy activities, but a rethinking of literacy as a whole. Hence, post-literacy activities should be considered before and during literacy activities. Literacy and post-literacy activities (library, access to newspapers, production of written materials) should be conducted simultaneously so as to contribute to the creation of a literate and enabling environment. 4. Integration with Local Women's Associations In the context of poverty reduction, the subproject should focus on four issues: · Strengthening of the organizational capacities of the local women's organization; · Training to enhance the local association's performance (training of the association's leaders), and training of a person (relais) to take over the provider's role at the end of subproject imple- mentation; · Make connection between the local woman's association and other local associations, as well as local financial and administrative organizations; · Assist the local women's association in income-generating activities. 36 Managing Public­Private Partnership outsourcing programs that were more exclu- Especially in its "integrated" version, PAPF sively focused on literacy. The PAPF's inte- represented a broader approach than the more grated program mainly focused on poverty narrowly literacy-focused approach of the reduction, whereas the Canadian sponsored PAPA project. Also, those responsible for PAPF literacy program, PAPA (Projet d'Appui au saw that post-literacy activities were stressed in Plan d'Actions), was more focused on teach- PAPF but not in PAPA. Not surprisingly, PAPF ing the learners to read and write (and it saw is the most costly of the two programs. enhanced literacy skills as the best means to reduce poverty). In practice, the PAPF deliv- ered a more expensive course that provides a Conclusion wider range of activity for the learners (thus, the extra cost of PAPF as compared to PAPA, The outsourcing strategy had certain advan- reflected the extra costs of teaching more tages over traditional forms of public service basic skills and income-generating activities). delivery since it did not require standard pre- Both approaches made use of the outsourcing scriptions by government as to curriculum and approach. The characteristics of the PAPF medium of instruction. Also, guidelines about were linked to greater decentralization and skills training and program focus were less spe- more skills activities, whereas the PAPA cific than traditional government-implemented focused more narrowly on literacy activities literacy campaigns, and left more to local deci- at a lower cost.35 Currently, there is not sions. Outsourcing was structured to be enough information about the relative impact demand-based, which made policy-based cen- of these two programs to make a cost-effec- tral decisions on language or curriculum unnec- tiveness analysis comparing them. essary. Likewise, the PAPF outsourcing system The outsourcing approach is suitable for gradually began to take into account local pub- implementing both literacy-only, and integrat- lic development services, and to increasingly ed poverty reduction programs. For integrated adapt the courses to local circumstances. programs, each provider could establish a net- Other approaches than outsourcing can also work of instructors and assistants (relais) and give such advantages; they are not in any way implement a more variegated subproject than prerogatives of the PAPF. However, the out- most administrative units were previously sourcing approach made it easier to adapt to capable of organizing. Many providers used micro-circumstances, and to implement a vari- supervisors to instruct specific courses, e.g., in ety of small-scale subprojects instead of only management or specific skills training such as one standardized government program. Some dyeing. Theoretically, local administrative of the more innovative subprojects could be a units were responsible for coordinating all source of lessons for primary schooling, espe- development activities in each area, but in cially in the fields of practical skills learning, practice, these units were capable only of infor- integrated approaches, and instruction in mation generation and dissemination--not of national languages. So far, however, experi- planning integrated development program ences learned by the PAPF have not been used activities. for policy making in primary education. SECTION III Costs and Effects G overnments and funding organiza- mental agencies in responding to local and tions face several questions linked to changing needs" (Oxenham, et al, 2002, p. the financing and delivery of non-for- 37). The argument for using many locally mal adult education: Should literacy based private provider groups is linked to the programs be implemented by the state? Should view that such providers will have knowledge a mass literacy campaign be used? Who are the of local problems, and frequent contact with stakeholders? Should the program be literacy- potential stakeholders and beneficiaries.37 focused (as in the PAPA program in Senegal), Also, local provider groups communicate in or should it be an integrated poverty-allevia- local languages, whereas government deploy- tion program such as the PAPF? These ques- ment of local language teachers to the right tions raise issues of cost-effectiveness of alter- zones can be difficult, especially in areas with native approaches.36 On these questions, many different languages. On the other hand, relevant findings hardly exist yet from Senegal, state-implemented programs often benefit but some observations can be made on the from economies of scale, and will be cheaper basis of earlier studies. than smaller provider-implemented projects. In most countries, the state literacy services Advocates for state-implemented programs are not strong; and in pursuing EFA goals hold that the overhead costs of government- relating to adult and out-of-school youth, the implemented programs are lower than for pri- government needs to decide whether to devel- vate providers, and that implementation is op the public sector or to outsource service more sustainable. Table 4 summarizes some delivery. The advocates for outsourcing hold claims encountered by the author among that an outsourced program, if it is using many stakeholders in the adult literacy sector in private organizations to deliver relatively small Senegal, about the advantages and disadvan- programs, is better able to respond to the con- tages of each method. cerns of the local participants than a national The budgeting of a privately implemented state-implemented program. In a recent review program is different from a public one. A pub- of approaches and experiences in literacy and lic delivered literacy program typically focuses skills training programs, Oxenham concludes, on distribution of funds and/or learning mate- "NGOs seem to be more flexible than govern- rial to the target area, where local civil servants 37 38 Managing Public­Private Partnership Table 4: Arguments on advantages and disadvantages of public and private delivery used in the Senegalese case Public delivery Outsourced delivery through bidding Access might be more equitable; The delivery structure was set up quickly, and delivery Could be less costly (due to economies of scale; possibilities increased rapidly through the multiplication and if volunteers are mobilized to implement of providers; courses);38 It led to a fast delivery of services (i.e., fast evaluation Monitoring and control could be more successful and financing of subprojects); since these functions inhere in the administrative structures; Advantages PAPF developed a client-oriented approach, with grass- roots participation in all stages of implementation; The courses were adapted to local needs and conducted in local languages; Setting up delivery structure would have been Outsourcing had especially demanding monitoring costly and taken a long time; requirements; Slow transfer of funds through government Being profit-oriented, providers (including NGOs and channels would have caused delays; CBOs) negotiated to drive up the price; hence the unit Government would have been both implementing costs for PAPF were high; and control agent, which may lead to conflict Programs had weak sustainability (when external Disadvantages of interests and less accountability; financing ceased, most village activities ceased). A big government program may have been standardized and less client-oriented. and/or volunteers distribute the material and A: Base unit costs of provision39 implement the course. The funds or materials allocated usually depend on the target popula- The base unit cost (see box 4) was used to tion size. calculate the cost of the providers' services. A program based on outsourcing and pri- Since the unit price in contracts issued by the vate implementation frequently uses an esti- PAPF project was approximately $50, the mate of the unit cost for each enrolled person amount for financing a subproject where 300 as a basis for the financing formula of a proj- beneficiaries were enrolled in literacy educa- ect. The cost composite of the PAPF included: tion corresponded to 300 people x US $50 = US $15,000. The unit price was negotiated · Base unit cost of provision (see section A between the providers and the Project Coor- below) dination Unit and took into account both the · Cost incurred by providers and benefici- operational costs of implementing the sub- aries, including opportunity costs (see sec- project and the providers' overhead costs. tion B) Also, an amount was added to this cost in · Project operational costs (see section C) order to provide institutional support to the · Project investment costs (see section D) provider. This institutional support was included in order to encourage the creation It is important to include the cost of ineffi- of new provider groups, and to strengthen ciency in a cost analysis. A program will be existing ones. It would typically be used to financing a certain number of beneficiaries purchase non-essential office or transport who will drop out before the end of the course equipment such as computer(s), a motor- or not succeed at the final exam (see section E). bike(s), or a car. Costs and Effects 39 Box 4: Base unit costs for PAPF as a percentage of total costs Total costs: $50 Equipment 9% Training of staff 11% Training of beneficiaries 59% of which the components are · Learning materials: 24% · Salary of facilitator: 35% Monitoring 10% Overhead 2% Evaluation 1% Institutional assistance 8% Source: Diagne, Sall 2001 Calculations for PAPF unit costs are shown kind of what the beneficiaries would have above, and represent the amount of subvention earned/produced if they had not attended the or the base unit cost from PAPF to the course.42 The opportunity costs are more diffi- provider.40 The cost covered the training and cult to evaluate than the direct cost. The GDP materials deemed necessary for a literacy and per capita in 2001 was US $475 (World Bank basic skills course. This amount was set at Regional Statistics). Senegal's assessment of liv- 37,500 FCFA per enrollee. For a provider with ing conditions of 1995 cites an undated study 10 courses and a total of 300 beneficiaries, the by Niane, estimating the informal economic subvention was approximately US $15,000 per sector's43 income to 21,000­40,000 FCFA per provider subproject, or US $50 per enrollee.41 month (Net Present Value as of January 2003 The PAPF unit cost of $50 did not represent = US $46­8844). Using the lower numbers of the full cost for one participant, because it did Niane's study, one can estimate the opportuni- not take into account other aspects of the proj- ty costs of the course to a minimum of US $23 ect and participants' cost, such as monitoring (or 50% of a month's income), since the bene- and evaluation costs, or opportunity costs. The ficiaries attended approximately 450 hours of most important of these additional costs were coursework during non-core working hours45 as follows: (the course was set up by the provider in coop- eration with the attendees to avoid prime working hours). The direct costs for the bene- B: Costs covered by providers and ficiaries were relatively low (about 17%) as beneficiaries compared to the cost of foregone opportuni- ties--likely a factor hindering the enrollment Each participant contributed 2,500 FCFA (US of the very poorest. Most people interviewed $3) in cash or in-kind toward the cost of the by the author during an implementation evalu- course. Also, the costs of foregone earnings of ation (2002) pointed to opportunity cost, the beneficiaries can be taken into account in rather than direct cost, as the usual reason for calculating the social cost of the project. The dropout from the literacy course. Among the opportunity cost equals the value in cash or in- poorest, while the direct cost may have acted 40 Managing Public­Private Partnership as a barrier to entry, opportunity costs may experimentation around the relais function has have driven down attendance among those not yet led to conclusive results. enrolled. However, in either case, the effects of The PAPF project document intended that private costs on access and attendance do not "the project would finance 85% of the costs of appear to have been very strong, since the the program; the provider would cover the dropout rate in the project was surprisingly remaining 15%" (SAR, 1996, p.11). The way low (about 15%) and in many cases, the local this provider financing would function was not women's association covered the $3 entrance clearly laid out by the Staff Appraisal Report fee. The low dropout rate also indicates that (SAR) but it looked as if the providers were the participants considered the return on their perceived to behave like charitable organiza- investment (i.e., the return on direct costs and tions and that outsourcing would add extra opportunity costs) to be reasonably good. resources to their work. The government The opportunity costs discussed above were Implementation Completion Report (DAEB, relevant to "average" participants. Some par- 2002) confirmed that the providers did not ticipants had higher opportunity costs, and cover any remaining financing of the subpro- some had higher returns on investment. In jects, but that participants' fees (7% of the many cases, one or two particularly energetic provider's fee) together with the base unit cost people took the lead in setting up the course-- (93% of the fee) covered the direct costs of and thus incurred higher opportunity costs running the subprojects. than the other participants. For example, the leaders of the local women's associations often had higher opportunity costs, because they C: Project operational costs were involved in time-consuming discussions and the "hassle" of setting up the subproject in The project operational costs in PAPF included the villages. The relais (local assistant or sub- the following cost items: stitute instructors) had higher opportunity costs, since they were responsible for course · Selection, monitoring and evaluation of follow up in the rainy period and after the sub- subprojects (performed by the DAEB); project's financing of activities ceased. The · Financial management (AGETIP); high opportunity costs faced by the relais led · Operating costs for the PAPF project them to request financial incentives from the office (the Project Coordination Unit-- subproject. However, the payment of the relais PCU). was not deemed a sustainable approach, since it may have led to cessation of activities after The Staff Appraisal Report commented on subproject financing ceased. Instead, the PAPF the recurrent costs of the project in the follow- tried to institutionalize the role of the relais, by ing terms: the "recurrent cost impact is low considering it similar to the position of leader because: (i) the program and projects are pilots in the women's associations. The subprojects and limited in scope, (ii) implementation is provided specific training to women associa- being carried out by non-governmental organ- tion leaders and to the relais. Such training ses- izations, and (iii) no new government staff are sions were, in many cases, considered as incen- being recruited to execute or manage the proj- tives for the participants since it gave them the ect." The PAPF budgeted $556,000 in recur- opportunity to learn new skills, have an hon- rent costs, mainly to be covered by the govern- orific position in the village, travel, and obtain ment (SAR, 1996). This amount enabled the direct financial benefits from the per diem pay- government to follow up on project implemen- ments during travel. At this point, however, the tation, especially through the use of the PCU Costs and Effects 41 and local staff members. The amount corre- toward a successful outcome, since PAPF pri- sponds to a unit cost of US $3 per beneficiary. marily targeted illiterate people. However, lit- In addition, AGETIP received 5% of the con- erate or semi-literate people improved their tracted amounts for its services (corresponding reading and writing abilities and acquired new to a unit cost of $3). knowledge and skills, especially income-gener- ating activities and other functional skills. There is therefore a case for including them D: Investment costs (including along with other "course completers" in esti- investment in human capital) mates to the extent to which PAPF has a suc- cessful outcome.46 This approach has therefore The investment costs in PAPF included the fol- been adopted in this study. lowing cost items: For most operational purposes, the base unit cost and the operational costs (A and C above) · Transportation and communication, such are important, because they represent (i) the as cars and motorcycles for monitoring subvention to the providers; and (ii) the costs and evaluation activities; computers for of running the project. For policy considera- statistics, etc. tions, cost-benefit analysis and cost-effective- · Training in outsourcing for DAEB, ness analysis, is useful to consider the total cost AGETIP and CNOAS staff members, as of the course (i.e., unit costs based on compo- well as technical and administrative train- nents A, B, C, and D). The combined recurrent ing for the staff members of the providers. cost, opportunity cost, and capital costs over the duration of a typical course (a minimum of The budgeted investment costs (excluding 450 hours of training over a maximum of two providers' contracts) were estimated at years) come to a total unit cost of $96 per $2,214,000 (SAR, 1996). The amount corre- enrollee, of which $70 were public costs47 and sponded to investments in cars and computer $26 were private costs. When we calculate systems necessary for technical monitoring of these costs per non-dropout (or completer), the the program, as well as human capital invest- total unit cost is $113,48 of which $82 is pub- ments to enable government staff to monitor lic costs and $31 is private costs. If one con- and evaluate the program. The budgeted siders the success rate of the program, i.e., the investment costs corresponded to a unit cost of pass rate on the final examination, the unit approximately US $12 per beneficiary. price rises still higher. The government Imple- mentation Completion Report (DAEB, 2001) estimated the success rates for PAPF in 1997 E: Cost incurred by ineffectiveness and 2000 for reading, writing, and problem (dropout and failure) solving (see Table 1). Using the numbers in Table 1, an average estimate of 55% can be Some beneficiaries did not complete the course used to calculate a unit cost for "successful" or they failed the final test. The unit cost per learners.49 The unit cost now climbs to $205 literate person ("graduate") was therefore per successful course "completer," of which higher than the unit cost per enrollee. Also, $149 is public costs and $57 is private costs. some of the enrollees had some exposure to All costs are displayed in Table 5. school and may have been considered semi-lit- These estimates are subject to many uncer- erate or even literate, when they began the lit- tain assumptions, such as the opportunity eracy course. It is debatable whether the fur- costs. Dropouts frequently say they "lack ther training of such learners should count time" to participate in the course. The extra 42 Managing Public­Private Partnership Table 5: Private and public unit costs (including an estimate of opportunity costs) PROJECT UNIT COSTS in US $ Unit cost per Cost Category Enrollee "Completer" Successful (A) Basic Unit Costs 50 59 107 (C1) Government recurrent costs 3 3 6 (C2) Contract management (AGETIP) 5 6 11 (D) Investment costs 12 14 25 Public unit costs 70 82 149 Unit cost per PRIVATE UNIT COSTS in US $ Enrollee "Completer" Successful Cost Category B1 Participant's direct costs 3 4 7 B2 Opportunity costs (estimated) 23 27 49 Private unit costs 26 31 56 TOTAL COSTS in US $ 96 113 205 time is often needed for agricultural work, and a clear-cut manner. Furthermore, it did not occasionally for family obligations (tending to stipulate any bonus for successful subprojects, children, tending to sick family members). It is or any fine for unsuccessful ones. In The Gam- uncertain whether (and how) this time should bia, clauses depicting clear indicators of suc- be estimated in monetary terms. Also, as cess and failure (especially at the final exami- defined, a dropout will have dropped out nation of the beneficiaries) were included in before the course ended, and therefore will not the procedures manual to improve effective- bear the full opportunity cost of the course. ness (see Annex 4). It may be possible to This is why the opportunity costs for dropouts improve cost-effectiveness by reducing the unit described above are unlikely to be as much as cost, however, this may cause fewer providers estimated; rather, this amount should be seen to participate. Alternately, with a higher unit as a maximum opportunity cost. cost, the program can test out new literacy People enrolled in the course for a short strategies and stimulate growth of new period may have drawn some benefits from the providers, as PAPF did in Senegal. part they completed. Also some of the non-suc- A cost comparison of adult literacy educa- cessful participants (45%) may have acquired tion with primary education is useful on two sufficient skills to become fully literate later, grounds: (i) adult literacy education and pri- through self-study. However, it is reasonable mary education are both varieties of basic that the program indicators correspond to the education, and components of Education for goals of the program--and that they should All goals--though they have distinctly differ- distinguish failure and success as clearly as ent clienteles and curricula that only overlap possible. This is usually done through criteria in part; and (ii) primary education serves gen- set forth in the procedures manual, and in the erally as a point of comparison for cost analy- contract between the providers and the gov- sis of other stages in education systems. Such ernment. In the Senegalese case, the procedures comparisons do not, however, compare like manual did not outline indicators for success in with like, and therefore do not imply that the Costs and Effects 43 types compared are functionally equivalent Conclusion with each other. In the case of PAPF, the pub- lic unit cost ($70) can be compared to the Many programs justify adult literacy programs yearly unit spending for primary schooling, on equity grounds: literacy programs serve dis- which is approximately $72.50 Taking into advantaged population groups (poor people, account the repetition rate (13.6%) and com- girls and women who for socio-cultural rea- pletion rate (44.7%) in primary schooling sons have lacked access to education, ethnic and in literacy classes (dropout 15%; non- minority groups). In Senegal, the PAPF and success 50%), completion of a literacy class other programs provided access to basic skills costs about one-fifth of the completed five and literacy education for the economically years of primary education. Internationally, disadvantaged population (although not neces- the tendency is similar: nearly always, non- sarily for the poorest of the poor). There is lit- formal education is less costly than primary tle research on the monetary and non-mone- education. Oxenham et al. identifies unit tary benefits of the programs. In Ghana, costs ranging from $4-5 in the Uganda FAL, studies of literacy programs found that the to $63 per learner for FAO's Peoples Partici- program did not seem to influence the cash pation Programmes. Valerio (2003) notes that earnings of the beneficiaries, and had only a in Ghana the average unit cost per literacy small effect on their health and hygiene prac- learner is US $36 (for an 18-month program) tices (Valerio, 2003). However, the Ghana pro- and $64 per successful literacy learner, where- gram is more exclusively literacy-oriented than as the average per student unit cost51 of four the PAPF. In both countries the unit costs are years of formal schooling in the government much lower than those of primary education. system is US $216. She notes "the findings A review of international research on the suggest that literacy programs are successful impact of adult basic education (Lauglo 2001) in delivering literacy skills at a substantially concluded that there is strong evidence of the reduced cost than in the formal school sys- impact of adult education on the support par- tem. The cost per successful literacy program ents give to their children's education, on graduate, even assuming high dropout rates, improved self-confidence in public situations is between one-third to one-half of the cost of (sense of empowerment), and also some evi- four years of primary schooling" (Valerio, dence showing improved family health, and 2003; p.188). In Senegal, the public cost per better skills in market transactions. Cost-bene- successful participant in literacy courses is fit studies that have sought to estimate empiri- $149, which can be compared to two years of cally the rate of return on adult literacy cours- primary schooling, or less than one year of es have been missing, but a recent discussion successful primary schooling (counting paper for the World Bank by McMahon (2003, dropout and repetition). ch. 5) on the economics of the larger field of The favorable cost comparisons are not to lifelong education concludes that the private say that primary education in any way should and social returns to such education are likely be substituted by adult or non-formal educa- to be distinctly high. tion. As noted in the introduction, since pri- The unit costs in Senegal included public mary education is not yet universal in many costs and private costs. The unit costs to the countries, there is still a need for adolescent government were $70 per enrollee ($149 per and adult literacy education. The cost esti- "graduate"). Private costs consisted of a mon- mates mentioned above, from Senegal and else- etary fee of $3 per beneficiary, and an estimat- where, show that this type of education is not ed opportunity cost of $23. It is unlikely that especially costly. the costs borne by the participants were a 44 Managing Public­Private Partnership major barrier to course completion, since the for monitoring and evaluating results. Specifi- dropout rate was so low. Research on the rela- cally, there is a need to ensure that the distri- tive cost-effectiveness of different programs bution of enrollees corresponds to the targeted and curricula in Senegalese adult education is characteristics, that the number of dropouts be lacking. For all programs, performance indica- kept low, and that the program has a high rate tors are needed as part of an effective system of successful completion. General Conclusion and Recommendations L arge scale literacy education for adoles- projects, and may also argue that by bypassing cents and adults will be needed in many the government, the approach also avoids the countries during the next two decades, risk of public sector corruption. In this report, and even beyond, as an outreach to give a rather different argument has been made: vulnerable groups who miss out on primary that the partnership approach, as its name sug- education a chance to obtain basic education. gests, requires direction and strong involve- When universal primary education is reached, ment by the government. Government involve- the need for literacy education will taper off as ment is critical on technical levels to ensure those who attended primary school reach that a proper selection procedure is followed, adulthood. But, there will be a continued need and that the monitoring, evaluation, and feed- for non-formal education for adolescents and back systems enhance subproject effectiveness. adults who already can read and write. The The "policing" function of monitoring and outsourcing--or partnership--approach is a evaluation has often been emphasized in this flexible method for implementation that can report, but the capacity-building role of gov- meet the changing needs for non-formal educa- ernment is also important in carrying out these tion where a government does not wish to evaluative functions. In the PAPF, monitoring make heavy long-term investments in public efforts were often combined with capacity- structures for delivery of literacy courses. The building workshops to strengthen the approach can facilitate subproject implementa- providers' implementation. The government tion that meets the demand in individual loca- had more a helper's role than an enforcement tions; in fact each intervention can be tailored office's role. Both roles are necessary, however, to local requirements. Such program flexibility for successful program implementation. can generally make outsourcing a useful tool in The government's involvement is crucial new circumstances, for example in post-con- when it comes to policy formulation and flict zones, or in areas with a high incidence of implementation. Literacy projects based on HIV/AIDS. outsourcing were successfully implemented in Some proponents of outsourcing suggest Senegal largely because of the government's that outsourcing avoids government obstacles strong commitment to the approach--and to with regards to setting up and financing sub- literacy. Hence, the outsourcing approach is no 45 46 Managing Public­Private Partnership substitute for public involvement, but relies on responsibilities, financing arrangements, the government to direct the implementation indicators); process. Problems linked to financial manage- · Regularity (in monitoring and evaluation, ment and transparency in public offices need to financial transfers, selection procedures, be addressed before setting up an outsourced information dissemination); program; for outsourcing is not a quick fix for · Credibility (in data management and mismanagement of public funds.52 information, financial monitoring, fair- Some critics have suggested that the out- ness in selection, interaction with benefi- sourcing approach tends to reduce the needed ciaries and with providers). advocacy role of NGOs and other civil society organizations. Others contend that such organizations, which are not always demo- General advice for outsourcing cratic associations representative of the local projects population, are given too large a role through outsourcing; and that the approach therefore In this report, I have used the PAPF as a case is neither equitable nor democratic (e.g., if a study to illustrate the set up of an outsourced religious association is implementing a sub- project. The PAPF experience has shown that project, the members of that association may some methods for selection and implementa- be unfairly favored above others as recipients tion of subprojects worked better than others. of benefits from the activity). However, the The following list of recommendations is based growth of civil society organizations is itself on my understanding of which methods func- an objective for poverty reduction strategies-- tioned well in the PAPF project, and of what at least as envisioned in World Bank policy could have been improved, and how. Obvious- (Comprehensive Development Framework, ly, circumstances in other countries are differ- Progress Report, 2001). In Senegal, the out- ent from Senegal, and the process and methods sourcing approach led to the creation of a must be adapted to fit each case. plethora of grassroots associations eager to be involved in literacy activities. Also, the proj- ects revived many language associations I. Setting up a project (Association de Langues), local organizations that had as their mission to preserve local lan- · Design a simple selection and implemen- guages and the local culture. Regardless of the tation process based on the existing motives on behalf of the providers (service to administrative levels in the country (the the community or earning a living), the PAPF, process could consist of subproject for- together with other outsourced-based proj- mulation, verification by locally based ects, have caused civil society organizations to staff, selection, contracting, implementa- proliferate in Senegal. tion, monitoring and evaluation). The approach that has been tested by PAPF can be used to implement other services in · Use a recognized agency (NGO, or a Senegal, and other countries can benefit by parastatal agency such as AGETIP) to learning from the Senegalese experience. Cer- function as the contract-management tain general concerns are important through- agency; make sure that the agency has a out program planning and implementation: capacity for rapid contract management and for financial monitoring at local · Clarity (in selection methods, criteria for levels. selection, program goals, subproject General Conclusions and Recommendations 47 · Clearly define the role of each participant II: The Courses in the project. The participants in the sub- project could include the beneficiaries and · Make sure the selection and implementa- their organizations; the local education tion schedule fits the beneficiaries (the administration; the selection committee; schedule should not be made exclusively the outsourcing agency; the providers and to fit the administrative agenda). the providers' organization; and the cen- tral education authorities. · Be explicit about program goals and indi- cators: include learning targets in the con- · Make sure that the beneficiaries are tract between the contract-managing involved in the subproject formulation, agency and the provider; include indica- and in all aspects of subproject implemen- tors about learning tools, e.g., learning tation, monitoring and evaluation. program, timetable, and student registry, in both the procedures manual and in the · Make sure that the local administration individual contracts; be clear about penal- informs the beneficiaries about their ties for failure to fulfill targets. rights and obligations (e.g., right to adult education classes, obligation to help the · Ensure that instructors have sufficient provider find a classroom). training in how to teach adults. The requirements for the instructor's level of · Make sure the selection committee does training and education should be included not include politicians, but only technical in the procedures manual. experts. If possible, include representa- tives of the beneficiaries in the selection · Ensure, through requirements in the pro- process. cedures manual and subsequent monitor- ing, that the providers coordinate activi- · Design realistic and effective monitoring ties with relevant development initiatives. and evaluation systems--these processes These may include local development must be outsourced if the contract-man- plans, local schools and the school admin- agement agency or the educational istration, and other ongoing development authorities do not have the ability to per- projects. form them; make sure that feedback mechanisms exist to inform those respon- · The literacy training could be coupled sible for program policy about what suc- with income-generating and other pover- ceeds and what fails (and how to fix the ty-alleviating activities. The use of peer latter!) during program implementation. tutoring and group-work methods as a supplement to "class teaching" could be a · Prevent dishonest behavior by just penal- means to ensure better and more rapid ties (such as full repayment of embezzled acquisition of reading and writing skills. funds and exclusion from participation in future bids). The providers' association · Encourage the creation of local associa- may be used, or a private company hired tions (e.g., women's associations) that can (such as a debt recovery company, or sim- give support to the literacy courses. Offer ilar), to ensure enforcement. some training to the leaders of such asso- 48 Managing Public­Private Partnership ciations (e.g., accounts, management) to up in such a way that there is no incentive strengthen their involvement in local for the provider to enroll only literate development initiatives. people in the courses (such enrollment would boost the "success" rate, make the · To promote sustainability, make use of a teaching very easy, and reduce costs con- local substitute instructor (relais) who is siderably). also given some training that would enable him or her to take a lead in con- · Although the enrollment of target benefi- tinued literacy activities after the end of ciaries should be controlled, it is strongly the subproject. recommended to allow some flexibility. In many cases, the literacy courses will attract a mix of people with different edu- III: The Costs cational backgrounds who are keen to learn more. There is nothing wrong with · Keep unit costs low by ensuring low this as long as illiterate people are given waste: the highest dropout rate to be tol- the first priority when learners are select- erated and the lowest completion rate ed. that is acceptable should be specified in the procedures manual and in the con- · Avoid waste by ensuring that the defined tracts between the providers and the con- indicators correspond to the program tract-managing agency. goals--and that the tests administered to the beneficiaries provide reliable and · Ensure, through effective monitoring, that valid data for these indicators. the beneficiaries enrolled in the subpro- jects correspond to the target beneficiar- · Use fines and other penalties to prevent ies. The evaluation methods should be set misuse of funds by the providers. ANNEX 1 Time Frame for Subproject Implementation T he schedule for selection and imple- ble for education at the regional level, in mentation of the subprojects is general- the beginning of September. ly the same every year, and consists of the following activities: 4. The preselection committee then analyzes the documents and sends them to the 1. The dissemination of information about selection committee by mid-September the outsourcing process, and the distribu- (the proposals should reach the selection tion of procedures manuals to interested committee by the end of September). parties must be done in the beginning of the calendar year, because it is necessary 5. The technical analysis is done by mid- to give the potential providers some time October, and sent for approval. The to prepare for gathering information in results should be publicized by the end of the villages. The participatory assessment October. The providers whose proposal of the beneficiaries' situation must be has been rejected have five days to resub- undertaken during the dry season--since mit the proposal for re-evaluation (the re- many villages are inaccessible during the evaluation is done by the committees for rainy season (and the villagers are also approval and technical analysis). too occupied to participate in the assess- ment at that time). 6. The contract-managing agency (AGETIP) signs the contracts with the providers 2. The providers write the subproject pro- immediately upon approval, and at the posal and submit it during the rainy sea- same time pays the first installment. The son. In Senegal, this is usually done dur- provider can therefore immediately begin ing August. The proposals are submitted procurement of equipment and the train- at the department (county) level. ing of village instructors in early Novem- ber. 3. The department checks the proposals according to the procedure, before send- 7. The second payment is transferred upon a ing the proposals to the person responsi- first report from the provider, including 49 50 Managing Public­Private Partnership (a) a description of the activities of the ing (i.e., most of the learning activities first trimester of implementation; (b) a take place in the "dry" months, from late financial report with receipts for expens- October to May). The courses should es; and (c) a request for payment for the cover at least 300 hours of functional lit- next period of project activities. The third eracy teaching (includes numeracy and and following payments are released basic arithmetic), and 150 hours of basic upon on-site inspection of the progress of skills activities. the subproject. 9. After implementation, each subproject is 8. The courses usually begin in late Novem- evaluated and the corresponding account ber or December, and should continue for is closed by AGETIP. The provider can a maximum of 18 months from contract- then begin preparing a new subproject. ANNEX 2 What are the Reasons for Investing in Adult and Non-Formal Education? T he objective of an analysis of cost-effec- extremely rare (see box 5). A recent attempt tiveness and of returns on investment in was made by McMahon (2003) to survey the education is to provide information on literature on economic returns and externali- investment alternatives to maximize ties on a wider range of provisions covered social and private net gains. Yet, empirical under the concept of "life long education." He research on the economic returns, specifically concludes that countries generally under-invest for investments in adult literacy education, is in basic and continuing education for adults. Box 5: An example of indirect and direct "returns" to adult education October, late evening in Senegal. Twenty-year old Nogoye Sene is sitting in front of me with her youngest son of three in her arms. "The NGO people proposed a literacy course to us," she said. "Before that, we had no access to schooling in the village. Sometimes, when we lacked money, we had to go to Dakar for work. We didn't know how to calculate our earnings, and to discuss our salary. We didn't earn much. The chil- dren stayed with their grandparents when we were away. Now, with the [literacy] project, we have learned to make soap and to dye clothes and we don't need to go to Dakar any more. It was difficult to afford the literacy course, but the local Women's Association helped us. Each participant had to pay 2500 FCFA." "We learned about health," she continued. "I'm responsible for the medicine box of the villages. We have vaccination and the children are healthier." When asked what else she had learned, she answered, "The most important was about health. Also I learned to write." To my question about the usage of her reading and writing skills, she responded, "I wrote a poem and I want to write more of them." Source: Author's interview with beneficiaries during evaluation of the program. 51 52 Managing Public­Private Partnership Adult education typically targets young people knowledge about health and hygiene, as well in the 15­35 age group who, it can reasonably as sensitizing them to HIV/AIDS issues. Bene- be assumed, make faster use of their new fits from education are not easily measured in knowledge than school children will. More- economic terms (see box above). For example, over, if they are relatively young, they will still the profitability for Nogoye Sene of traveling have a long benefit stream ahead of them. to Dakar and doing seasonal work there, However, he found very little information on rather than producing soap and dying clothes cost-benefit analysis on adult literacy educa- in the village is difficult to estimate, since she tion in the material he surveyed. prefers to remain in the village as opposed to Drawing on literacy surveys in Africa, it is traveling a great distance to work. It may be useful to assume that most children need five possible to estimate the rates of return by com- to six years of primary schooling to become paring the differences in earnings (applying a "literate": "A high share of the adults who market estimate of the discount rate for net have completed less than five or six years of present value calculation), and subtracting the primary schooling remain functionally illiter- social costs of the course (consisting of oppor- ate for the rest of their lives" (Bruns, Mingat & tunity costs and the direct cost of 2500 FCFA Ratakomala, 2003). This contrasts with other per participant). In addition, Nogoye Sene's countries53 where the majority of children usu- newly acquired knowledge may have some fur- ally attain fluent literacy after three or four ther direct monetary returns, since she is now years of schooling (PIRLS, 2003). The unit in a better position to discuss the price of veg- costs for non-formal education leading to liter- etables at the market and to negotiate her acy are much less than for five years of primary salary. schooling. In fact, the annual unit costs of non- Recent developments in the calculation of formal education are, in the great majority of returns in education have tried to take into non-formal education programs, lower than account effects that are not directly linked to the annual cost of the cheapest form of formal income gains, especially effects on health. schooling (which is primary school). True, in Mostly, cost-benefit research on non-monetary some cases, it was shown that the benefits effects has concentrated on issues such as accruing to the full cycle of primary schooling (i) consumer behavior, (ii) health, (iii) family are greater (Valerio, 2003; Sandiford, 1995), life, and, to a lesser extent, on (iv) asset man- but so is the expense. agement and (v) migration (Cohn & Geske, In any event, the quality of learning (and 1990). In comparing this list with the case of presumably the benefits) will vary widely Nogoye Sene, it can be noted that the literacy among providers, methods, and programs. course yielded non-monetary returns in all There are cases where, after some 18 months these domains. In addition to getting better of part-time attendance in a literacy course, skills in reading, writing and practical arith- adults on average will outperform school chil- metic, the beneficiaries learn about health and dren who have had 3-4 years of full time nutrition, and organize themselves into small schooling, in terms of basic literacy skills groups practicing income-generating activities. (often when these same adults have had no For participants in these courses, the monetary prior exposure to school, before commencing gains in education only cover a small part of their literacy course). One example is the FAL the benefits of education, since the project also program in Uganda (Okech et al., 2001). focused on health and hygiene issues. The average age of participation in literacy A recent review (Lauglo, 2001, p. 16-22) courses in Senegal is 21 years. The lessons found evidence of the impact of adult literacy include hands-on practical training, and basic education in these areas: (i) literate mothers Annexes 53 support children's education, (ii) literacy educa- same way derive direct income gains from lit- tion leads to a sense of empowerment, (iii) more eracy skills. Certain findings in Valerio's study effective communication (oral as well as writ- suggest this may apply to primary education. ten), (iv) improved family health, (v) more pro- In a longitudinal study now in progress in ductive livelihoods (There is a dearth of quanti- Senegal, the single most important reason the tative studies on the latter type of benefit). participants, both urban and rural, gave, as a One quantitative study of monetary benefits reason for enrolling in the literacy course, was on non-formal education was recently done by income improvement. Valerio (2003) on the impact of adult literacy Findings in Senegal, although not backed education in Ghana, using data from the Liv- with a solid database, are more positive than in ing Standard Measurement Study 1998/99. She Ghana. During most evaluative visits to the found scant monetary benefits and mixed PAPF project, the interviewed participants results regarding non-monetary benefits. In point to some positive impact due to participa- keeping with earlier reviews and studies (Lau- tion in the literacy courses, i.e., better knowl- glo, 2001; PIRLS, 2003) she found a clear pos- edge to health and nutrition; better knowledge itive effect of parental literacy on primary about how to organize production in small- school attendance for children. The longitudi- scale associations; more successful market nal study currently underway in Senegal did transactions; and better management of the not collect data on this question. A new impact household economy (DAEB, 2002). As noted, study, with findings expected for 2005, will the baseline study of the longitudinal study address both monetary and non-monetary currently underway found that the main rea- returns in adult education. son why the participants attend literacy cours- The Ghana program, as opposed to the es is the financial benefit they hoped for. Table PAPF, is government-delivered, and concen- 6 summarizes these impressions: trates more exclusively on providing literacy Most of the benefits shown in Table 6 above skills. One can also speculate that urban cours- will have some indirect monetary benefits, e.g., es have higher returns in direct cash terms HIV/AIDS awareness is believed to cause a because participants belong to a cash-based lower infection rate, thereby reducing both economic system, whereas rural participants family and community costs linked to the dis- often subsist from farming and will not in the ease. An added advantage is the role of litera- Table 6: Monetary and non-monetary benefits of literacy education Monetary benefits and/or preconditions of monetary benefits Non-monetary benefits Personal benefits Skills for new income-generating activities; Better health (vaccination, hygiene); Skills for calculating earnings; HIV/AIDS awareness; Access to better jobs; Happiness/entertainment (access to writing Higher cash income (?) and reading religious, entertainment and cultural literature); Individual empowerment. Externalities More efficient markets; More democratic participation/stronger Lower public health costs; capacity for collective action; Lower crime; Higher national productivity. Higher percentage of children attending school (especially girls); Health/hygiene spillover effects; Better nutrition for family. 54 Managing Public­Private Partnership cy education as a low-cost outreach service for cation is also vital to the targeted intervention especially vulnerable people. Further, since the in such areas as HIV/AIDS and reproductive learners are not children but adults, literacy health education and programs to raise aware- and non-formal education programs probably ness of civic rights and responsibilities" (World have quicker returns than formal schooling. Bank, 2001, p. 14). The overall goal of the World Bank is pover- Policy choices in education should be based ty reduction. The World Bank sees poverty as on consideration of both monetary and non- an unacceptable deprivation in human well- monetary benefits as well as concerns about being. Thus, poverty is multidimensional and equitable access to these benefits. In fact, adult not confined to cash income (World Develop- education can be seen as an equity measure ment Report, 2000/01). Moreover, "education since it provides a second chance for those who and training for youths is not only an econom- have been disadvantaged by the formal educa- ic imperative. In many countries young peo- tion system (girls in many African countries, ple's dissatisfaction and disillusionment with vulnerable ethnic groups, and the poor). The their prospects for education and work threat- importance of literacy courses to give such vul- en social cohesion and stability. Reaching this nerable groups a second chance is clear. age group through formal and non-formal edu- ANNEX 3 The Public­Private Partnership Criteria Checklist T he checklist below is appropriate for 3. Operational criteria considering whether or not to use a partnership approach (based on out- · Can the public sector organization devel- sourcing or any other public-private op appropriate operating standards for implementation mechanism). Source: IFC the project? Toolkit for PPP in Education (2003). · Are there any operational issues that can- not be realistically addressed by a private 1. Financial criteria partner? · Can the private partner be held account- · Is the project financially viable? able for appropriate performance? · Can the project function on a stand-alone basis? 4. Accountability criteria · Is it possible to define an equitable rate- setting mechanism? · Is the public willing to accept the public- private partnership approach? 2. Technical criteria · Are elected officials willing to accept a partnership approach? · Does the project have any inherent tech- · Are other stakeholders (e.g., public sector nical constraints that could be solved by workers) willing to accept a partnership the private partner? approach? · Can the public sector develop appropriate technical specifications? 5. Implementation criteria · Can appropriate mechanisms be estab- lished to monitor the performance of the · Is it possible to generate meaningful com- respective partners? petition? 55 56 Managing Public­Private Partnership · Is the project free of jurisdictional or lia- 6. Timing criteria bility issues? · Can an integral project champion be · Are the time lines adequate to develop found? operating specifications? · Can this project champion access the nec- · Is the time allocation sufficient for the essary resources? project to be delivered according to the · Can a successful transition plan be devel- performance specifications? oped? ANNEX 4 Summary of Senegal's Procedures Manual 54 Introduction 4. Strengthening the organizational capaci- ties of the beneficiaries T he government, with World Bank sup- 5. Use of providers specialized in integrated port, has implemented a functional litera- development instead of providers special- cy program since 1995. Providers execute ized only in pedagogy this project locally according to modali- ties explained in this procedures manual. The The integrated character of the intervention program has progressively been extended to is based on the following four principles: include post-literacy activities, as well as newspa- 1. Strengthening of the beneficiaries' associ- per production and printing of books in national ations languages. Despite largely successful implementa- 2. Development of endogenous human tion, there has been a lack of post-literacy train- resources (associations, leaders, members, ing sessions, lack of direct economic benefits and the relais55) from the courses, and lack of understanding of 3. Coordination of activities with local the functional and practical side of literacy pro- administrative, financial and association grams from the providers' side. Evaluation of services early implementation experiences has led to a 4. Making the beneficiaries' association revised strategy, consisting of more complete and autonomous through income-generating integrated subprojects. The present revised proce- activities and setting up a permanent dures manual will focus on the following aspects: structure for learning at the subproject 1. The functionality of the programs in implementation site. terms of improving the income of the ben- eficiaries The request for financing will ensure that 2. The formulation of subprojects that are the training plan is set up in coordination with more tailored to the beneficiaries' needs the local development plan, and that all the through more participatory needs assess- proposed activities are coordinated with other ments local development activities. The figure on the 3. Setting up a structure to ensure retention next page shows how the present procedures of literacy activities in the implementation manual sets forth these new options. sites beyond subproject financing 57 58 Managing Public­Private Partnership all the of the train- other man- and of for program the financial support and up evaluate results program the distribute monitoring of material, certification role ublicp set and and and performance of the the financial out financial management s and partners of and available providers of of evaluation echnicalT Mobilize- private successful program Monitor- program' indicators Economic- evaluation Update- lists pedagogical specialists, structures Measure- account aspects Organize- ing partners Contract- and providers' agement of the IDA, pro- for national of and the the program other periodic of from funds of providers' external of and role (i.e., disbursements the accounts financial provide audits and program annual management advice up on Financial Internal- budget) mobilization literacy Set- budget Monitor- Monitor- financial Monitor- government partners; reports Organize- technical gram Act- audits to the con- parties the for of role regulatory of interested parties procedures of all the requests subprojects other providers review with information progress subproject concerned of to and the the with Administrative Proposition- texts Annual- program partners Distribute- manual Register- funding Provide- CNEA about program Sign- tracts the in for in definition and national a other on the of zones in policy raising program development with actions role results the program target marketing), agreement literacy education a Political Participate- of establishment literacy Evaluate- warenessA- country Coordinate- partners Define- implementation Promote- (publicity Coordinate- development the Sign- financing literacy in technical ministries and responsibilities organizations and services contract-man- identity bureaus Composition and/or Concerned- Provider- Financial- partners Beneficiaries'- organizations Assistance- management Director- echnicalT- Controllers- Regional- aging responsibilities d'Exé- d'In- National and de Illiteracy) Unit) Agence ravauxT (Contract- Comité Committee Agency) (Project des Public Roles CNEA: Eliminate PCU AGETIP: I: Stakeholders 1. d'Elimination l'Analaphabétisation (National to 2. Coordination 3. cution térêt Managing Annexes 59 of the 1 II (by on benefi- evalu- the which local providers' following studies requests each proce- proceed- zone of the on with literacy criteria of criteria of the role or procedures section analysis in the on proposal whether technical account of fulfills the validity availability (see these a of member) of matrix manual deliberation into 2 rite-up echnicalT Certify statement issues: Information- implementation Agreements- ciaries Classroom- center Confirm- request and manual below) Preselection- fulfilling Individual- each subproject basis ation dures Joint- takes recommendations W- ings for the finan- (i.e., activi- contri- that proce- not of budget into prices price and the role community) the does unit by validity financial takes from that market items, cost the manual arrangements and Financial Ensure- cial income-generating ties, bution erifyV- proposal account budget overall exceed established dures in its to qual- sub- and sub- with and pro- exis- role providers sanc- circum- financial to subprojects re-evalua- has some presence subject occurs legal providers the subproject the revised or technical local of propose financing and in of to providers' this agreement the the that been the subproject for provider good of e.g., not pursuit the a a contains providers' zone of financing, revision (note: Administrative Certify- the the Control- tence Ensure- have legal tions Certify- ity relevance stances Evaluate- propositions for projects for tion when posed which errors, dispositions) alidationV- projects the with in in of use the socie- plans the as for subpro- of the proce- trans- of publica- equity civil the (c) the zones (a) presence criteria by role that concordant by from quality through records implemented through, are development target the manual, the are of technical Political Ensure- jects local and local development Guarantee- (b), selection observers ty selection of established dures parency tion selection from Liter- from la the the for for for devel- proj- and from for de Sociale of from of responsi- for and from from without Action development association l'Action official council service Ministry (Ministry Ministère Solidarity) identity literacy responsible responsible responsible (CERP) de Solidarité representative Education representative la Social, representative members rights for representative providers representative Members Observers de Composition and/or a) Regional ble Official community Official regional Official regional opment b) Representative providers' Representative ect One- Delegate Basic acy One- MFSSN: Famille, et Nationale Family National One- l'AGETIP woT- voting One the One PCU of now com- com- is role been the commit- the, has and analysis procedure preselection Recently (Selection committee decentralized Local echnicalT Stakeholders 4. mittee Note: this strengthened, selection fully 5. mittee tee) 60 Managing Public­Private Partnership the, the study rec- con- and the meth- agents same support, for to the from other subproject subproject from external the with learning learning projects in role the necessary the AGETIP in of with beneficiaries, feasibility request of if committee evaluates the external a according controls and subproject) of necessary if association participative up when echnicalT Identify- and their Conduct- using ods Set- financing Improve,- request ommendations selection Implement- conformity in tract Provider- results financial Ministry Participate- evaluations (i.e., evaluate Coordinate- development implemented locality for the con- state- to the account in proce- reports together the agency to role log the bank of financial a according up manual financial subproject statements receipts Financial Open- the Set- ments accountant appendix dures Send- and with tract-managing of sub- financ- reports proce- and the role for agree- authori- the of submit beneficiaries plan financing by local and the the for protocol technical the contracts subprojects a with manual of about indicated Administrative Prepare- requests subprojects Sign- ing Sign- ment Submit- as dures Inform- ties implementation project links the agencies benefici- consulta- program in discussions role in through the partnership professional association communities a potential and government the up Political Participate- tions with especially CNEA Promote- among aries Develop- with Set- providers' organi- internation- organiza- de and identity (Sociétés Composition and/or National- NGOs al Community-based- zations Development- tions développement) Providers Stakeholders 6. Annexes 61 on of the to activ- liter- sub- sta- for lan- skills the the internal of of evalua- for course the propose mobilize of to in evaluations information provider existence during in provide process role the place subproject providers and instruction and and persons data the the the to the the of functional impact the implementation by external in functioning centers teams request, the echnicalT Choose- dates, training, guage Identify- instructor(s) provider Identify- resource teach Participate- and on program Control- given implementation ities proposals Control- and acy project Participate- tion On- support Control- collecting tistics man- financially payment specified role as cash procedures in-kind the Financial Participate- through or by ual com- reg- of with the the liter- on role to and manage- class register requests financing local a technical and of for the reports the monitoring keeps on AGETIP up committee agreement management provider subprojects situation Administrative Set- ment Sign- the The- mittee istry Receive- requests Formulate- advice Submit- monthly the DAEB, PCU Monitor- acy of to in con- the at type they plan zones mapping in beneficiar- (manage- partnership stakehold- of skills community in role which beneficiaries local needed, learn community the a disparities local target use concentration to the committee) zones) all level management avoid subprojects Political Choose- functional wish Identify- (i.e., chooses ies) Organize- participation ment Define- reduce education with ers Define- (through to of same When/if- flict local tar- l'Ed- of girls). the (espe- de d'A- Educa- in and should organized Office program population of be (Chief identity living Nationale­ women groups (Inspection areas Inspection Office) Composition and/or Illiterate- groups get implementation cially These preferably IA- cadémie tion IDEN- Départementale ucation Departmental Education) educa- services Beneficiaries Decentralized Stakeholders 7. 8. tion 62 Managing Public­Private Partnership lit- liter- pro- of: the con- for zones on sup- groups (d) and indica- appropri- evaluate needed reviews national cost-effec- and/or (c) role a materials, as target the learning monitoring studies repertories and database documents of a program annual population structures, up program literacy projects learning providers, subprojects reports performance the the echnicalT Conceive- eracy Propose- and for Undertake- tiveness acy grams Update- (a) (b) port sultants Evaluate- ateness materials Monitor- the program Set- continuous of tors Prepare- and for of annual certain role an literacy (UNESCO, credits by for etc.), Financial Establish- budget Manage- accorded institutions ACCT has appli- literacy role the the CNEA secretariat on in up regulatory administrative that of Administrative Follow- cation texts Manage- personnel field Ensure- permanent a of the of the on status the pro- for of literacy and reports propositions proposition texts role literacy national revision Political Prepare- implementation the program Prepare- for improvement national gram Prepare- regulatory sector in- mate- and promotion environment Management Monitoring learning the studies training identity for for for Evaluation and literate a Composition and/or Senior- Unit- and Unit- rials of Unit- service for Agency (DAEB) National Stakeholders 9. Literacy Annexes 63 in (d) (h) with spe- (b) (f) studies pro- (Infor- to adapta- organi- etc. (i) (a) conception role providers, materials; formulation; IEC and program of books; financing; management contracts or (c) providers (g) materials and areas evaluations; of settings; learning editing subprojects, echnicalT -Support and/or through AGETIP the and training; of subproject (e) conducting mation­Education­ Communication) grams; translation tion cific zation of role Financial the role of provision providers, with for services, protocols PCU services Administrative Sign- cooperation beneficiaries, technical and/or of CNEA the in role Political Participate ­ on Nationale Appliqué Resource Literacy4 School identity (Ecole Economics) on Composition and/or ENEA- d'Economie National Applied National- Center Etc.- institutions Support Stakeholders 10. 64 Managing Public­Private Partnership II: Presentation of the Request 4. Efficient management The request for subvention of a subproject is Efficient management means setting up an prepared by using the template for subproject integral system of planning, budgeting, and requests in the annex of the procedures manu- monitoring and evaluation, which guarantees a al (not included in the present text). The successful implementation of the subproject. provider is invited to present any supplemen- tary information that is believed to be of use 5. Institutional support during the evaluation of the subproject. The necessary information for the request is pre- Institutional support means to allocate a part sented below. of the subvention to strengthen capacities at the provider and beneficiary levels. 1. Eligibility of the provider A provider is eligible to implement a literacy III: Eligibility Requirements for the program if: Proposal · The provider exists legally; The requests shall be made according to the · The personnel involved have the neces- model in the procedures manual (not provided sary qualifications to implement the sub- here) and shall respect the following selection project; criteria to be eligible to obtain financial sup- · The needed equipment and infrastruc- port from PAPF: tures are available. Requirement 1: Provider's eligibility 2. Knowledge about the locality and 1.1 The provider can prove that the organi- local demand for literacy zation has had a legal existence for at least 12 months prior to the request for Knowledge about the locality and local funds. demand for literacy implies that the provider 1.2 The provider has already undertaken has undertaken an in-depth participative study education training (formal or non-for- and needs assessment of the proposed imple- mal) and proves this. mentation site. The methods and the results 1.3 The provider has already managed a should be clearly indicated in the request for development project or program, funding, with precise information about the preferably a literacy program, and target zones and the indirect and direct benefi- proves this. ciaries of the program. The provider is experienced in the areas of development activities in which the 3. An appropriate and functional beneficiaries wish instruction (as learning program revealed by the needs assessment). If the provider has no experience in the consid- The learning program is considered adaptable ered development activities, cooperation and satisfactory if it clearly responds to the with specialized agents/experts needs to requests of the beneficiary population as be established. The proof of this cooper- revealed in the needs assessment. ation must be attached to the request. Annexes 65 1.4 The provider has the following person- 2.3 The provider has supplied all necessary nel (please attach CVs following the information to identify the beneficiaries model in the procedures' manual--not (type of local association, number of provided in this text): members, gender, age, prior instruction -- A coordinator with prior experience level, profession and/or occupational in management; activities). -- An account manager who has been The minimum of beneficiaries is 600, of trained in bookkeeping; which at least 75% should be women in -- One or several supervisors having the age range of 15­39 years. been previously employed as instruc- 2.4 The provider has indicated how the tors. program will affect indirect beneficiar- 1.5 The provider has or will have compe- ies--and has provided information tent instructors to implement the about the identity and location of these planned activities within the time limits beneficiaries. (development agents from the same 2.5 The occupational activities of the bene- background, who are willing to do the ficiaries are presented, including infor- job, motivated and trained for their mation about the problems the benefici- assignment). aries encounter when performing these 1.6 The provider has appointed the activities, as well as a detailed descrip- required personnel for technical, func- tion of these problems and how the lit- tional and basic skills training and has eracy course will deal with them. supplied proof of their involvement in 2.6 The expectations and requirements of the conception of the program, and of the beneficiaries have been formulated their availability to perform their in a signed agreement (with the benefi- respective tasks. ciaries) attached to the request. 1.7 The provider has a functioning head 2.7 A clear description of roles and respon- office, and if necessary, local offices. sibilities of the subproject is provided 1.8 The necessary premises (classrooms, lit- for each of its stages (conception, imple- eracy center, workshop...) to successful- mentation and evaluation). ly undertake the training, as well as ade- 2.8 The beneficiaries have established a quate equipment, are readily available management committee. or can be purchased. 1.9 It is advantageous if the provider is a Requirement 3: An appropriate and member of a providers' association. practical learning program 3.1 The problems addressed by the program Requirement 2: Provider's knowledge are clearly formulated with precise indi- about the locality and local demand cators. for literacy 3.2 The objectives of the program allow the 2.1 The implementation sites correspond to beneficiaries to learn instrumental knowl- the target zones of the program as iden- edge (i.e., literacy), to organize them- tified by local authorities. selves efficiently, to have endogenous 2.2 The provider has indicated the physical, human resources qualified to ensure local social, cultural and economic character- continuing education, to undertake activ- istics of the area in general, and has also ities that can increase local income, and provided a specific description of the to set up a documentation/reading center sites where the classes are proposed. that is functional and well used. 66 Managing Public­Private Partnership 3.3 The general and specific objectives of mechanism is set up through the selec- the program are well defined, and suc- tion and training of an endogenous cess indicators for the specific objectives relais, by connecting the local associa- are indicated. tion to local administrative and financial 3.4 The learning content is practical, i.e., it networks, and by promoting the associ- relates to the daily activities of the ben- ation's income- generating activities. eficiaries. 3.5 The proposed learning strategy leads to Requirement 4: Efficient Management self-reliance of the beneficiaries and per- 4.1 The mechanisms for administrative and manence of education activities in the technical management include: village (i.e., activities are continuing -- Operational plan for implementa- beyond subproject financing). tion of activities 3.6 The proposed activities will teach prac- -- Administrative follow up tical knowledge that can be applied to -- Technical follow up real-life situations. -- Evaluation plan 3.7- The training plan for the beneficiaries 4.2 The mechanisms for accounting and staff indicate clearly the competen- include: cies needed to exercise specific functions -- Resource mobilization plan and to perform specific activities. The -- Specification of beneficiaries' finan- need for training sessions to enable peo- cial or in-kind contributions ple to acquire these competencies, the -- Detailed budget of each activity content and duration of each training -- Indications about financial monitor- category, the place and materials to be ing used for the training, etc, should be 4.3 The requested subvention should not clearly indicated. exceed 37,000 FCFA per beneficiary; 3.8 Information on course set up is provid- each beneficiary should contribute ed, indicating the place where the toward the cost of the program with at course is conducted, the duration of the least 2,500 FCFA (see annexes of the program (which should not exceed 24 procedures manual, with a proposed months), as well as the training time budget for 20 classes with a total of 600 (the program requires a minimum of beneficiaries (not provided in this text)). 300 literacy training hours for a total of at least 450 training hours). Requirement 5: Institutional support 3.9 The training methods are indicated for 5.1 The provider organization has account- each activity. ed for the use of its part of the institu- 3.10 The learning material is indicated and tional support included in the subven- is related to the objectives and the activ- tion. ities of the program. 5.2 The provider has accounted for the use 3.11 The program ensures the creation of a of the part of the institutional support literate environment and initiates the that is intended for the beneficiaries. beneficiaries in reading newspapers and 5.3- The total amount of the institutional books. support does not exceed 10% of the 3.12 The provider has set up mechanisms for total amount, calculated from the total continuing education, which is integrat- budget for program implementation ing basic skills and literacy training. This without institutional support. Annexes 67 IV: Selection criteria for the subprojects National Languages inform the potential providers about the existence of the PAPF and The local preselection committee ensures that distribute the current procedures manual to all the request is compliant with the following interested parties. three requirements before proceeding to the analysis: (i) timely submission of the request; 2nd stage: Preparation of a subproject propos- (ii) legal existence of the provider; and (iii) the al by the provider submitted dossier includes five copies of the requests. If one or several of these three crite- The interested providers contact illiterate pop- ria are not fulfilled, the local preselection com- ulation groups likely to be interested in a liter- mittee does not proceed with the analysis. acy program, undertake a participatory needs A request is rejected if it does not conform assessment, decide with the beneficiaries on to the four criteria below. A request is eligible which functional aspects the literacy program if all four criteria are judged satisfactory: will focus, negotiate the content and imple- (a) Provider eligibility mentation procedures of the training program, (b) Knowledge about the implementation and formulate the request for financing based environment on the model presented in the annex of the pro- (c) An appropriate and practical learning cedures manual (not included in this text). program (d) Efficient management 3rd stage: Submission of the request for financ- ing of the subproject Any of these criteria are judged satisfactory if they obtain at least 60% of the maximum The provider submits five copies of the request rating (see rating sheet in the annex of the pro- to the local education office between August cedures manual--not provided in this text). 1st and 30th of each year, after having received In the selection process, the best proposals consent to implement literacy activities in the among those that are not rejected are selected selected area from the local education officer. though rating. If the annual PAPF budget The departmental officer (Inspecteur) of exceeds the amount requested by eligible sub- education verifies the following: project proposals, the proposals having obtained less than 60% of the maximum rating - The education office of the department (and that were therefore previously classified must agree to the literacy activities undertaken as `ineligible') will be selected according to the in the department by the provider in question. total score obtained during the technical analy- sis of the proposal, i.e., the projects with the - An on-site inspection of at least 20% of the highest rating will be selected first. proposed sites must verify that the information in the request is exact, and that the request is made in agreement with the beneficiaries. V: The Stages of a Subproject - The implementation sites must correspond 1: The selection of a subproject to priority areas for local development and must correspond to target areas for education. 1st stage: Information to providers The officer should formulate an opinion with- The PCU, the DAEB and the decentralized in a week of receiving the subproject propos- offices of the Ministry of Basic Education and als, i.e., between September 1st and 7th. 68 Managing Public­Private Partnership 4th stage: Preselection 7th stage: Technical analysis The departmental officer (Inspecteur) of edu- The selection committee is composed of: (i) cation sends the request with a delivery note to four members with voting power, one of which the Regional Chief Education Officer is an independent consultant selected by the (Inspecteur d'Académie) by September 8th. Ministry of Basic Education and National Lan- The local preselection committee is com- guages, one member from the Ministry of Basic posed of a representative of the Regional Edu- Education and National Languages, one from cation Council, the Regional Chief Officer of the Ministry of Family, Social Action, and of Education, head of the Office for Community National Solidarity and one from the AGETIP Development Services, and the head of the respectively, as well as three members without Office for Local Development Services. The voting power: one representing the providers, committee analyzes the requests between Sep- one representative from the PCU, and one tember 10th and 20th. from the literacy resource center. This preselection is based on the providers' The members with voting power will first capability, and on the knowledge of the local individually analyze the requests, taking into environment. If any of these two criteria are account the evaluation notes (criteria 1 and 2) not fulfilled, the request is rejected. The from the regional preselection procedure, and providers' regional organization assists in the then proceed to a joint evaluation session preselection without having a vote. The assess- before preparing the minutes from the deliber- ment of the providers' association is included ation. The technical analysis is based on an in the minutes from the preselection delibera- evaluation form (including weighting of each tion. A representative from the project coordi- selection criteria) and should be completed nation unit will provide any necessary infor- within 14 days of receiving the proposals. mation to clarify the preselection process for The observers participate in the discussions the committee. and assist during the entire selection process, and can request that their observations are 5th stage: Transmission included in the minutes from the selection process. The minutes are published by the PCU The Regional Chief Education Officer and distributed to all partners and providers by (Inspecteur d'Académie) transmits the prese- October 31st at the latest. lected requests to the PCU with transmission notes on the two aforementioned criteria 8th stage: Approval (providers' capability; knowledge of the local environment). The proposals should arrive to The approval committee, which includes the the PCU within three days of the preselection, director of the DAEB, the director of commu- or between September 21st and 24th. nity development of the Ministry of Family, Social Action, and of National Solidarity, and 6th stage: Reception the director of AGETIP, gives the final approval for financing the selected requests, The PCU sends the proposals to the selection after having ensured that a proper selection committee within three days of receipt, or procedure was utilized. between September 25th and 27th. One or two The approval committee checks the follow- selection committees can be set up every year, ing: (i) the selection procedures, and especially according to the number of requests to be ana- the strict criteria of eligibility; (ii) the process- lyzed. ing of any appeal for re-evaluation of the Annexes 69 requests; (iii) the number of classes financed as (iv) Procure the necessary learning materials compared to the number of classes the and equipment corresponding to the first providers submitted for financing;57 (iv) the installment of the financing selection committee's decisions on the need for (v) Implement activities as planned by the reformulation of any of the subproject subproject requests; and (v) the decisions on eligibility. (vi) Ensure internal supervision and evalua- When the approval committee gives its tion of the activities through the use of three endorsement, an approval decree on the selec- types of reports: (a) attendance records kept by tion is signed by the director of the DAEB. the instructor (b) the supervisor's reporting The providers that find it necessary can, at book (c) the financial manager's register that point, submit an appeal to the approval (vii) Submit to PCU and AGETIP implemen- committee, which will discuss the request for tation reports every trimester based on the appeal with the selection committee. The model in the annex of the procedure's manual request for appeal can be sent a maximum of (not included in this text) five days after the publication of the results of (ix) Request payments according to the the decisions made by the approval committee. installments shown in the annex of the proce- dures manual (not included in this text) Implementation of the subprojects (x) Receive monitoring and financial audit visits, and correct implementation and registra- 9th stage: Contracting tion methods and procedures when necessary (xi) Perform internal evaluations of the After approval, AGETIP will proceed with the training following actions: (i) sign a contract with the (xii) Participate in the lessons learned and selected providers based on a model of the con- information-sharing sessions organized by the tract provided in the annex of the procedures PAPF manual (not included in this text); and (ii) (xiii) Close the subproject according to pro- transfer the first installment of the subvention cedures set forth by PCU and AGETIP amount into a specific account opened by the provider for the subproject. Technical monitoring by the DAEB 10th stage: Implementation of the subproject The DAEB, in coordination with decentralized by the provider education agencies, ensures the technical mon- itoring of the subprojects. This monitoring The provider implements the subproject in concerns the learning and pedagogical aspects conformity with the contract signed with of the subprojects. AGETIP. In this context, the provider will: Financial monitoring by AGETIP (i) Inform and remobilize the beneficiaries with whom the implementation agreement has AGETIP ensures that the subprojects are been written implemented as planned. To ensure this moni- (ii) Check that the local management com- toring, AGETIP will recruit financial con- mittees are operational, and collect the benefi- trollers who conduct on-site inspections to ver- ciaries' financial (or in-kind) contributions to ify bookkeeping and check the learning the subproject material and equipment procured by the (iii) Proceed, if necessary, with the recruiting provider. The results of these controls will be of project personnel recorded in the controller's register. 70 Managing Public­Private Partnership Completion of the subproject AGETIP and the provider to the account that is opened specifically for the subproject. This 11th stage: Evaluation of the subproject first payment will be sufficient to cover the starting costs of the subproject. The DAEB ensures the external evaluation of The second payment will be made after the all the subprojects during a period agreed by provider submits a report that includes: (a) the PCU and the providers. an activity report for the first trimester using For the literacy aspects of the courses (read- information from the monthly reports from ing, writing, calculus), international recognized the instructors and supervisors; (b) a finan- methods of evaluation will be used; for the cial report supported by receipts; and (c) a functional aspects, specifically elaborated eval- request for payment. This second payment is uation methods will be used. The results of not contingent upon on-site verification of these evaluations will be published by the PCU. implementation but will be paid upon pres- entation of financial statements and receipts. 12th stage: Closing of the subproject The format for all requests and reports are annexed in the procedures manual (not After evaluation, the provider and AGETIP included in this text). close the subproject. The provider submits a The third payment is made after an on-site final report. AGETIP ensures that all controls verification by the AGETIP controllers, to are undertaken and establishes a closing docu- ensure that the use of the first installment of ment that is to be signed by both AGETIP and funds was made according to the implementa- the provider. tion agreement. Any funds that may be left on the provider's The subventions given to the provider that account will be used in conformity with deci- cannot be accounted for will be subtracted sions by the PCU in agreement with AGETIP. from future installments. Likewise, if any The use of any such funds will be indicated in irregularity is noted, the corresponding the final report of the subproject. amounts will be subtracted from subsequent payments, without prejudice to possible legal recourse. VI: Financial procedures All installments will be made following the same procedures. After signing the contract, the first payment is An auditor will verify the controllers' state- made in accordance with the contract between ments. ANNEX 5 The Gambia's Procedures Manual 58 Introduction implement functional literacy programs; and G overnment intervention in the litera- · Coordinate financial partners' interven- cy sector is based on the principle of tions so as to gear efforts toward the same partnership with the literacy objectives and within the Government's providers in the field (NGOs and sector plan. Community-Based Organizations) and on the integration of all efforts within the Govern- ment's action plan, which aims to: 2. Overall aim · Substantially reduce illiteracy especially To widen participation opportunities through among the female population, notably in strengthening civil society organization part- the 15-39 age bracket in order to ensure nerships in developing sustainable literacy, an immediate impact on the social, eco- numeracy, life skills and income-generating nomic and education indicators; programs. · Reduce imbalances between regions and between urban and rural areas; · Promote functional literacy and the devel- 3. Project objectives opment of alternative basic education through the medium of national lan- · Establish and test the outsourcing guages in order to improve the effective- approach ness of literacy programs; · Enable providers, through outsourcing, to · Promote a literate environment in nation- enroll 12,000 people in literacy courses al languages in order to ensure the use, · Through outsourcing, enable providers to development and sustainability of enroll 8,400 people in post-literacy activ- acquired skills; ities, including income-generating activi- · Implement a partnership policy which ties (IGA) and facilitation of access to relies on partners in the field (NGOs, micro-credit programs Community Based Organizations) to 71 72 Managing Public­Private Partnership · Ensure sustainability through the devel- for all identified training needs, including opment and production of literacy and instructor and supervisor training. post-literacy learning materials in collab- · Regional committees for the selection of oration with all stakeholders literacy providers which include a mem- ber from each village targeted (Village Development Committee) 4. Strategies · A program of capacity-building for national and regional ANFEU staff, as In the context of a partnership approach with well as for the contract-management the aim of improving the effectiveness and sus- agency's and the providers' staff tainability of current literacy activities, the Use of providers for training of 12,000 peo- strategy proposed for coordination and collab- ple in literacy and 8,400 people in post- oration between ANFEU and literacy providers literacy and income-generating activities includes: · Through work with communities and civil society organizations, a program to · Use of an independent agency for con- generate newspaper articles, short stories, tract-management and financial monitor- poems, novels, folklore and writing com- ing petitions and the development of a range · Establishment of financial and technical of literacy and post-literacy resources that monitoring and evaluation systems can be used to enhance the subprojects' · Multisectoral coordination committees at sustainability both regional and national levels · A procedures manual to determine eligi- bility criteria for literacy providers, and to 5. Implementation strategies ensure that literacy and assessment stan- dards are met during delivery Government intervention in the literacy sector · In collaboration with stakeholders, estab- is based on public-private sector partnership, lish guidelines in the procedures manual with the role of partner defined as follows: Annexes 73 Policy role Administrative role Technical role Quality Assurance Adult and Non-for- - Establish policy - Coordinate with all - Facilitate training - Monitor and evalu- mal Education Unit guidelines stakeholders of supervisors and ate courses (ANFEU)/ Depart- - Standardize train- - Approve selection instructors - Monitor and evalu- ment of State for ing procedures; inte- of providers - Approve training ate provider staff Education (DOSE) grate with formal training education Regional Offices - Contribute to policy - Select providers - Facilitate training - Monitor and evalu- formulation - Coordinate with all of supervisors and ate courses - Set up a technical stakeholders instructors selection committee (of providers National Committee - Approve draft poli- - Approve proce- for Literacy cy of ANFEU dures manual (and annual revisions of PM) Communities/ Par- - Represented in - Select and estab- - Establish a context - Evaluate provider ticipants Regional Policy lish contract with of learning (select - Monitor progress Workshops provider language, schedule, - With provider: skills...) select instructor - With provider: - Set up manage- establish mecha- ment committee nisms for sustain- - Establish links with ability, such as IGA, formal education access to credit, etc. Providers' Associa- - Represent - Coordinate - Identify additional - Establish profes- tion providers' interest providers; be an training needs for sional code of con- "information agent" providers duct for providers - Involved in arbitra- - Advise on training tion issues Providers - Coordination of - Ensure that train- - Regular quality subproject ing takes place assurance - Coordinate with - With villagers: other development select instructors actions - Recruit and train - Monthly contact instructors (and with village manage- other personnel as ment committee needed) -Data collection Contract-managing - Establish contracts - Train the providers' - Evaluation of out- agent - Payment of con- accountants sourcing approach tracts - Audits; Financial monitoring 74 Managing Public­Private Partnership The Role of Government · Integrate literacy and post-literacy pro- grams to promote functional skills and On the basis of policy directives, the govern- attitudes that can enable the learners to ment, through the Non-Formal Education participate in good governance, appreci- Unit, will guide implementation, coordinate ate environmental conservation, promote programs, monitor literacy activities and eval- positive health practices and improve uate the sector's performance in terms of the standards of living through income-gener- objectives defined in the action plan. ating activities During the first phase, in addition to the · Supervise literacy activities and properly preparation of an annotated bibliography, the manage the resources provided Adult and Non-Formal Education Unit will · Monitor the total time for completion of facilitate the training of literacy instructors and literacy and post-literacy activities not supervisors as well as conceive of and develop exceeding three years from registration teaching/learning materials. · Follow monitoring and evaluation proce- The government strategy for the sector is to dures help literacy providers expand their programs · Ensure the literacy and post-literacy activ- and to coordinate donors and other partners to ities are sustainable work toward the same objectives and follow the same implementation strategies in order to Moreover, consideration will be given to avoid duplication of efforts. Partners include: those organizations that may assist and com- ANFEU/DOSE, regional offices, National Lit- plement the subprojects. This may include eracy Committee, multisectoral working extension workers (agriculture, health, envi- groups, provider associations, communities ronment, forestry, etc.), other government and providers. departments (health, social welfare, etc.), inter- national institutions and non-government organizations. The Role of the Providers Specific actions to be undertaken by each liter- The Role of the Beneficiaries acy provider include: The role of the beneficiaries includes the fol- · Sensitize and mobilize public opinion in lowing: favor of literacy · Identify and recruit instructors and super- · Will be present at regional policy work- visors in collaboration with the communi- shops ties · Will participate in the selection of · Make adult learners literate within 18 providers and instructors months through the establishment of 6 - · Will set up a management committee 10 literacy centers of 30 participants · Will establish a context of learning, each, 75% of whom should be women including the selection of language, sched- · Follow up literacy training with post-lit- ule of classes, IGA eracy activities for those participants · Together with provider, will establish who have completed the literacy compo- mechanisms for sustainability, including nent or have equivalent training/abili- post-literacy activities, IGA, etc. ties--at least 75% of the participants · Will monitor provider and assist in regu- should be women lar evaluation of provider Annexes 75 Subproject Selection and Provider Step 2. Distribute procedures manual to inter- Eligibility Criteria ested parties. Request subproject proposals (advertised in national press) for each of the The criteria for evaluating the eligibility of the clusters. Deadline: January 1st provider are as follows: Step 3. Each provider prepares a subproject · Providers must exist as a legal entity proposal and submits it at the regional level. · Providers must be office-based, have an Deadline: March 1st address, a bank account (with 3 signato- ries) and a working telephone Step 4. Each Regional Selection Committee · Providers can be NGOs (including inter- selects two providers, i.e., each cluster of 6­10 national NGOs), associations (including villages will be covered by one subproject, religious associations), CBOs and nation- which is implemented by one provider. Com- al for-profit organizations position of selection committee: 1 representa- · Providers must have at least three years tive from each village, 1 technical representa- experience in development and training tive from ANFEU/BED, 1 regional officer from · Providers must have core staff including a ANFEU/BED, 1 representative from the con- director, an accountant, an administrator tract-managing agency, 1 technical representa- and a literacy supervisor (CVs must be tive from the multisectoral working group, and attached to the subproject proposal) 2 representatives from the providers' associa- · With respect to pedagogy, the selection tion (no voting power). The selection is to be criteria for the subprojects are as follows: based on technical evaluation matrix spelled · Operator must propose to use a recog- out in the annex (not included in this docu- nized method of literacy training ment). Deadline: April 2nd · Program must contain at least 300 con- tract hours of literacy training over not Step 5. Approbation by ANFEU/DOSE and more than an 18-month period, with a information to selected providers. Deadline: minimum of 30 beneficiaries per class April 15th · Teaching/Learning materials must be acceptable, available on time, and of rea- Step 6. Issue of contracts by contract-manag- sonable cost. ing agency. Deadline: May 1st Step 7. Transfer of first installment of funds. Selection procedures and time frame Deadline: June 1st (per year) Step 8. Implement subproject Step 1. Identify 100 target villages by regional offices in collaboration with multisectoral work- ing groups and ANFEU. In each region 16 or 17 Financial aspects target villages are selected. These target villages are divided into two clusters based on criteria of The financial cycle will be divided into two geographic proximity. Each cluster contains parts: literacy and post-literacy activities. The between 6 and 10 villages. Regional education unit cost per beneficiary for the literacy com- officers visit each target village to provide infor- ponent is set at $40 (US) for the year 2004 and mation about the project and to identify focal 2005, and at $20 (US) for the post-literacy points for training. Deadline: December 1st component for years 2005 and 2006. The 76 Managing Public­Private Partnership implementation of the literacy component will be reported to the providers' association, should be made in three phases. The imple- which will consider appropriate actions to be mentation of the post-literacy component will taken (for recovery of funds or other action as be done in one single phase. needed). Each payment of funds will require at The providers will receive the first payment least an 80% success rate (of the initial enroll- upon contracting which will represent 40% of ment rate) at the given phase of participation. the total subproject cost for literacy activities. The providers' fee will be reviewed upon A second payment of 30% of the total costs completion of the literacy component, to will be paid upon completion of phase 1. A reflect enrollment in the post-literacy phase. third payment of 25% will be paid upon com- The successful provider will, upon finalization pletion of phase 2. A fourth payment of 5% of the literacy course and after positive com- will be withheld until the end of the subpro- munity evaluation, receive 50% of the total ject's literacy activities, and released upon cer- subproject costs for post-literacy activities; tification that at least 80% of the enrollees 45% will be paid upon start up of post-litera- have successfully completed the program. If cy activities, and 5% withheld under the same the success rate is lower than 80%, a fee cor- conditions as above (i.e., the success rate for responding to the unit cost ($40 US) will be the post-literacy course should be at least 80% deducted for each unsuccessful participant of the enrollees in post-literacy activities). below the 80% limit. If the success rate is Example of payments for a literacy course lower than 70%, the post-literacy element of with 300 participants: the subproject will be retendered, and the case Costs for literacy activities: US $40 x 300 participants = US $12,000 - to be paid as follows: Payment 1st installment 2nd installment 3rd installment 4th installment US $4,800 US $3,600 US $3,000 US $600 Requirements Contract set up 80% success in 80% success in 80% success in first phase second phase (of third phase (of initial enrollment) initial enrollment) Example of payments for a post-literacy course with 300 participants: Cost for post-literacy activities: US $20 x 300 participants = US $6,000 Payment 1st installment 2nd installment 3rd installment US $3,000 US $2,700 US $300 Requirements Contract set up Activities have started 80% success (of 300 participants) Annexes 77 9. Subproject monitoring and · Final evaluation of provider evaluation procedures Contract-managing agency's monitoring and The instruments required for monitoring and evaluation evaluation include: · Financial monitoring--each center will be Provider's monitoring and evaluation: visited during implementation of literacy activities · Registration documents for each center · Attendance register for each center · Supervisors' reports based on regular Closing of the subproject monitoring visits to each center The closing of the subproject will require a Providers should submit reports tied to the final evaluation and report by the provider, financial cycle; these reports should include which will include a one-page report from each bank statements. For literacy activities, three community. In addition, a final external tech- reports should be submitted accompanied by a nical evaluation and report (under the supervi- request for release of funds. These reports sion of ANFEU) will be required. After techni- should be submitted in three copies: one copy cal evaluation, a financial audit will be to BED/ANFEU at the central level, one for the undertaken by the contract-managing agency, district level and one for the contract-manag- upon which the project will be closed. Any ing agency. For post-literacy activities, two unused funds will be recuperated by the con- reports should be submitted, accompanied by a tract-managing agency and reintegrated into request for release of funds. the project budget. BED/ANFEU monitoring and evaluation: Refinancing · Integrated monitoring and evaluation (from BED) The provider is eligible for new subproject · ANFEU spot checks financing upon completion of the literacy com- ponent in an ongoing subproject. The provider Regional monitoring and evaluation cannot receive financing for more than two subprojects at any one time. · Spot checks (at least one monitoring visit Any provider with success rates below 80% to each community) during subproject implementation is not eligi- · Initial visit to identify village ble for new subproject financing before the end · Visit for final evaluation of subproject of the ongoing subproject (both literacy and post-literacy phases). Any provider with a suc- Community monitoring and evaluation cess rate lower than 70% will not be eligible for refinancing at any time. · Evaluate literacy course (required to give provider access to financing of post-liter- acy activities) Notes Executive Summary Introduction 1. Currently, several new literacy projects in 8. In some countries, a substantial popula- Senegal (of which the major ones are financed tion of illiterate and semi-literate adults will by IDA and CIDA) continue to finance literacy continue to exist in the longer term (Lauglo, courses by using outsourcing. 2001). 2. In other contexts, private companies may 9. For considerations of whether or not to choose to outsource parts of their operation to use a partnership approach in the first place, other companies, or public organizations may consult the Checklist for Partnership Projects choose to outsource to other public organiza- in Annex 3. tions. Usually, outsourcing is used when anoth- 10. Programme d'Appui au Plan d'Actions. er company has a comparative advantage for 11. Projet d'Alphabétisation Prioirité carrying out certain types of tasks, and it there- Femme. fore would be more cost-effective to hire that 12. Projet Alphabétisation des Elus et Nota- company to perform these tasks. However, bles Locaux. many other issues may enter into considera- 13. Both public and private sectors invest in tion, e.g., on a technical level, the possibility the project; also the local beneficiaries of the and cost of monitoring and evaluation; and on subprojects make a substantial investment in a political level, security and equity. it. The beneficiaries bear the highest burden of 3. In this case, experienced providers would risk. See also economic analysis in Section III. have a comparative advantage on new 14. Private firms also use outsourcing in providers, and may therefore have monopo- many areas, particularly where it is more cost- lized access to funding. effective to do so. 4. 37,500 FCFA. The conversion rate used 15. The project suffers from insufficiently in this text is US $1= 740 FCFA. streamlined data management. The DAEB has 5. The procedures manual indicated that the a different statistics system than that used by literacy training should be at least 300 hours, the management agency (AGETIP), which skills training at least 150 hours, and the pro- again differs from the one used by the project gram "should not exceed 24 months" (Manuel coordination unit. This leads to difficulties in de Procedure), 2000-2002). estimating numbers of enrollees and dropouts. 6. For comparison, it is estimated that the 16. These local development organizations unit annual cost to government of primary usually strive to obtain status as NGO. education in Senegal is approximately $72. 17. One reason why primary education has 7. Many PAPF managers also anticipated a more structured relationship between the that the providers' association could be used to local population and the administration is that organize training of providers, and to define formal education services have been imple- and monitor ethical guidelines for the sector. It mented for a longer period, and have estab- is uncertain to which extent a provider associ- lished a more structured connection between ation can be expected to perform such a role. the administration and the communities. For 78 Notes 79 less well established services, such as non-for- 25. The selection process has been increas- mal education, a structure similar to parent ingly decentralized, i.e., the regional level has associations does not exist. Private providers obtained a more decisive role in the pre-selec- under such conditions may function as a bridge tion phase. Other outsourcing programs in between administration and communities. Senegal have preferred a simpler decision 18. In a market-based outsourcing system structure, and eliminated step 4a. Most other with reliable feedback, participatory approach- countries that have adopted the outsourcing based subproject formulation and implementa- method have also opted for a simpler selection tion, and proper control mechanisms, any process, eliminating this step. NGOs which come in with a ready-made and 26. Recent administrative changes led to a contentious agenda should have a difficult time modified approval cycle in Senegal for the obtaining financing. 2003/04 school year. Most subprogram 19. These results do not control for compe- providers (PAPF, PAPA, Lutte contre Pauvrete, tencies prior to enrollment­e.g., whether the etc.) were selected in January­March 2003 for learners had had some schooling previously. a program start in March -April 2003. 20. Based on DAEB's assessment of all the 27. The integrated approach has subprojects enrollees (not including dropouts) at the end of that are active during the whole year. The cycle PAPF-financed course in 1997 and 2000. described is therefore not fully pertinent to this program. 28. The PCU was originally set up as an Section I independent bureau, but is now being integrat- 21. The PAPF procedures manual has under- ed into the DAEB. gone several revisions based both on changes 29. Source: Various interviews with AGETIP in government policies, and on improvements personnel and with former civil servants. in project administration. It has become an 30. Such impact evaluation is currently international reference. Several other countries under way--but results are not yet available. (e.g., Burkina Faso, Côte d'Ivoire and Guinea), The evaluation will assess literacy achieve- adopting slightly different procedures, have ments, as well as changes in community behav- used similarly structured manuals. ior, e.g., in health and hygiene, and HIV/AIDS 22. The selection process has been increas- awareness, as a result of literacy activities. To ingly decentralized. Currently, the subproject prevent possible placebo effects, the evaluation proposals are checked for accuracy at the pre- occurs as a "normal" classroom activity. fecture level. The selection takes place region- 31. Two-headed arrows signify interaction; ally and is subject to verification and accept- one-headed arrows denote one-way supervi- ance centrally by an approval committee. The sion/control. process described in this section pertains to the 32. Source: Author's interviews with project system before it became fully decentralized. managers (PCU and AGETIP). 23. The procedures manual for 2000-2002 33. As noted above, local monitoring should required each subproject to target at least 600 ideally be done by local government instances participants, which corresponds to 20 classes. to establish trust between the different stake- 24. This on-site inspection was not always holders. However, if the government cannot undertaken. See Section II.4 about arrange- successfully ensure such monitoring and evalu- ments for monitoring and evaluation. ation, it may be necessary to outsource these tasks to a local organization. 80 Managing Public­Private Partnership Section II 37. This is the case for the PAPF, where 33% 34. More traditionally delivered projects of the providers are local small-scale organiza- also have a large variation in implementation, tions, 43% are local CBOs, and only 18% are since the quality of teachers and local officials NGOs (of which many are small-scale Sene- to back them up varies from place to place. galese organizations), and 6% are religious However, in outsourcing, variation can be seen associations and other types of providers. both at the individual level and at the provider Source: Camara & Sow, 2001. level. Data gathered so far from the current 38. Typically, mobilization of volunteers is longitudinal study indicate that the quality of easier through government incentive schemes. the provider organization matters more than It is more difficult--albeit probably not impos- the quality of individuals, for a quality sible to base the cost of outsourcing programs provider will strive to recruit quality personnel on the use of volunteer (i.e., non-paid) instruc- and/or provide training for inexperienced staff. tors. In "problem" cases, a "good" provider will 39. 740 FCFA = US $1. The cost rates are typically find a solution; a "bad" provider may relevant to the procedures manual 2000-2002. choose to end the course. The conversion of US $ to Euro was unstable 35. PAPA had a unit cost of 22,500 FCFA, during spring 2003; FCFA is linked to Euro, whereas the unit cost of PAPF was 37,500. and has therefore been considerably strength- PAPA has currently reduced its unit costs to ened recently. During most of the implementa- 15,500 FCFA for a shorter and more intensive tion period, the unit cost of a subproject was course, which is more similar to the govern- quite stable at $50 (or, at the time, approxi- ment's intensive literacy courses, PAIS (also mately 56 Euro); currently with the weak US based on outsourcing). Neither the new PAPA dollar, it is at $66 (as of June 17, 2003). Since courses nor the PAIS have been evaluated. the cost statistics from other fields, e.g., pri- mary school, are based on 2001 and 2002 US $ levels, we have maintained the "strong" dol- Section III lar conversion levels in this text, and used a 36. The lack of impact results makes it diffi- conversion rate of 740 FCFA = $1. cult to undertake any cost-effectiveness com- 40. This is a model of subproject budgeting parison. The PAPA unit cost was set at 22,500 made by the PCU. The providers make their FCFA (and is currently reduced to 15,500 own budget, based on different (or similar) FCFA), as opposed to a 65% greater unit cost allocation of funds: the only requirement is for the PAPF (37,500 FCFA). The literacy per- that the final amount does not exceed the base formance of learners is higher for PAPF unit cost multiplied by the number of enrolled enrollees at the end of the course than for beneficiaries. PAPA; 75.2 % can fluently read a text, com- 41. The unit costs of many other projects, pared to 68.2% for the PAPA; and 85% of such as the PAPA, are lower, since these proj- PAPF learners have a theoretical mastery of ects focus more on literacy and do not include technical themes at the end of the course, com- as much learning material and training costs pared to 67.3% for PAPA learners (2000/01 on basic skills issues as does the PAPF. academic year; DAEB, 2002). The unit cost of 42. It can be argued that this cost is close to the intensive literacy program of the govern- zero, because course time will fit around pro- ment, PAIS, is 6,500 FCFA (for a 3-month ductive activity or vice versa. Evidence from course), but is reported to have minor results field visits and evaluations indicate, however, (no success results are given in government that one main reason for dropout was that the reports). opportunity cost of the course proved too high: Notes 81 the dropouts needed to concentrate their time French). The different definitions of literate on daily subsistence earnings to survive. The make it very difficult to estimate in real num- overall dropout rate, however, was low, and bers the impact of the PAPF on illiteracy in the indicates that the opportunity cost in most country. cases was not a major obstacle to participa- 47. Public costs is used to describe the cost tion--or that people were willing to pay the incurred by the government in program imple- opportunity costs because the benefits were mentation, and private costs for the cost reasonable. incurred by the beneficiaries. 43. Including subsistence farming. 48. Using the estimates in the government 44. If the numbers relate to the pre-1994 report 2001, where the dropout rate is esti- devaluation of FCFA, the Net Present Value mated at 15%. (NPV) is US $38­72. 49. Camara and Sow, in their study on costs 45. These data are based on very rough esti- of education in Senegal, used the percentage of mates of what a rural woman may earn if she 50% satisfactory outcome in the PAPF. works 10-15 hrs. per day for a month. Since 50. 2001 numbers, based on WB statistics the course is implemented in non-core working (from EFA database, 2003). hours, it has been roughly assumed that the 51. Average unit costs are for 1997/1998. possible earnings would be 50% of the total earnings potential for core working hours. The figures only purport to give a very general idea General Conclusions and of what the scale might be for foregone oppor- Recommendations tunities for the beneficiaries. 52. The World Development Report 2004 46. Also we lack information about the suggests that in some cases it is better to bypass number of people who had some preliminary the government, at least temporarily: "Coun- literacy skills when they enrolled. Some studies tries where the state is failing (often countries emphasize that semi-literate people actually in conflict) need service delivery arrangements benefit more from literacy education than illit- different from those where the state is fairly erate people do. The learning pace for semi-lit- strong [...] In Senegal--a stable democracy-- erate people is quicker than that of illiterate the reforms in education, including those that people, and the former also draw more infor- strengthen client-provider links, would go mation from basic skills classes. It might be through the Government (to strengthen the argued that it makes little sense to invest in policymaker-provider link as well). In the inefficient primary education, and then to Democratic Republic of Congo--where con- "upgrade" adult neo-literates by providing lit- flict has significantly weakened the state-- eracy training; and further that it would be a ways should be found to empower communi- better policy choice to invest more in primary ties to improve education services--even if it education so as to make the subsequent litera- means bypassing government ministries in the cy courses superfluous. The debate on national short to medium term" (WDR, 2004, p. 15). languages complicates the matter: if a person is literate in French, is that person still eligible for a literacy course in a national language, since Annex 2 he or she is illiterate in this language? In the 53. The Progress in International Literacy case of PAPF, the response has been to priori- Study (PIRLS) investigated reading achieve- tize people with no, or very little, formal ment at the fourth grade level in 35 countries. school background, and yet to also accept In most countries, a majority of the children some who were semi-literate and literate (in were literate by the fourth grade. A notable 82 Managing Public­Private Partnership exception was Morocco, the only African coun- 55. A local substitute instructor who has try participating in the study. The results from been specifically trained for continuing literacy Senegal, having a lower per-capita income than activities after the provider finishes its imple- Morocco, and a less efficient school system mentation. (and encountering the same language problems 56. The Canadian-funded PAPA project as Morocco), are likely to have even lower lit- (also based on outsourcing) funded a resource eracy scores at fourth grade than Morocco, center for literacy activities, which provides where only 23% of the fourth graders achieved different types of support for the literacy sector the lower-quarter PIRLS benchmark. (i.e., the as a whole. international average was 75%). 57. Due to last-minute financial adjustments and other issues, some of the providers may not obtain financing for all 20 classes. Annex 4 58. Draft. January 2004. 54. Edited translation by Bjorn Harald Nordtveit. References Abadzi, H. (2003). Adult Literacy: A Review gres d'Education des Adultes (2000­ of Implementation Experience. World 2002). Dakar: PAPF. Bank: Sector and Thematic Evaluations DAEB (2001). Etude Longitudinale (PAPF). Group, Operations Evaluation Depart- Dakar: DAEB ment. DAEB (2002). Etude Longitudinale (PAPF). Abadzi, H. (2003). Improving Adult Literacy Dakar: DAEB Outcomes: Lessons from Cognitive DAEB/Ministry of Education (2001). Govern- Research for Developing Countries. ment project implementation completion World Bank: Sector and Thematic Evalu- report. Dakar ations Group, Operations Evaluation Diagne A. W. & Sall B. R. (2001). Programme Department. de Formation sur la Stratégie du « Faire- Azfar, O. & Zinnes, C. (2003). Learning from Faire » en Education des Adultes. World Doing: A Methodology for Self-Evaluating Bank, unpublished document. Projects. Forum Series on the Role of Insti- Harper, M. (2000). Public Services Through tutions in Promoting Economic Growth. Private Enterprise: Micro-Privatization Washington DC: IRIS, USAID. Available for Improved Delivery. Intermediate online at USAID Website, at: http:// Technology Ed. www.usaid.gov/our_work/economic_ Howell J. & Pearce J. (2001). Civil Society & growth_and_trade/eg/forum_series/f5- Development: A Critical Exploration. sect1-zinnes.pdf Lynne Rienner Publishers. Bruns, B., Mingat, A. & Rakatomalala, R. IFC (2003). Private-Public Partnership in Edu- (2003) Achieving Universal Primary Edu- cation Toolkit. IFC (on CD-Rom). cation by 2015: A Chance for Every Lauglo, J. (2001). Engaging With Adults: The Child. World Bank. Case for Increased Support to Adult Basic Camara, M. & Saw, G. (2001). Cout et Education in Sub-Saharan Africa. Africa Financement de l'Alphabétisation au Region Human Development Working Senegal. Unpublished study. Paper Series: World Bank. Carr-Hill, R. (ed.) (2001). Adult Literacy Pro- McMahon, & Walter W. (2003). Lifelong grams in Uganda. Africa Human Devel- Learning in Developing Countries. An opment Series: World Bank. analysis with estimates of private returns CDF Secretariat. (2001). Comprehensive and externalities. Washington: World Bank. Development Framework: Meeting the Website: http://www/worldbank.org/ Promise: Early Experience and Emerging education/lifelong_learning/ Issues. World Bank. Website: http://sitere- Mingat, A., Tan, J-P., & Sosale, S. (2003). sources.worldbank.org/CDF/Resources/se Tools for Education Policy Analysis. ptember2001boardreport.pdf World Bank. Cohn, E. & Geske, T. (1990). The Economics Moore M. & Stewart S. (1998) "Corporate of Education: 3rd ed. Butterworth-Heine- Governance for NGOs?" in Pearce, J. man. (ed). (2000). Development, NGOs, and DAEB (2000). Manuel de Procedures pour Civil Society. Oxford: Oxfam GB. l'Experimentation des Programmes Inte- 83 84 Managing Public­Private Partnership Mullis, I., Martin, M., Gonzales, E., Kennedy, Sandiford, P.J, Monetnegro Cassel, M., & A. (2003) PIRLS 2001 International Sanchez, G., (1995). The Impact of Report: IEA's Study of Reading Literacy Women's Literacy on Child Health and its Achievements in Primary Schools in 35 Interaction with Access to Health Ser- Countries. Boston: International Associa- vices. Population studies, 49, 5-17. tion for the Evaluation of Educational Valerio, A. (2003). Estimating the Economic Achievement (IEA). Benefits of Adult Literacy Programs: The Oxenham, J., Diallo, A., Katahoire, A., Petko- Case of Ghana. PhD Dissertation, va-Mwangi A., & Sall, O. (2002). Skills Columbia University. and Literacy Training for Better Liveli- World Bank. (2003). World Development hoods: A Review of Approaches and Report, 2004: Making Services Work For Experience. Africa Region Human Devel- Poor People. World Bank. opment Working Paper Series: World World Bank. (2001). A Chance to Learn: Bank. Knowledge and Finance for Education in Okech, Anthony Carr-Hill, Roy A., Katahoire, Sub-Saharan Africa. Washington DC: Anne R., Kakooza, Teresa, & Ndidde, World Bank. Alice N. (2001). Adult Literacy Pro- World Bank. (1996). "Report no. 15517-SE: grammes in Uganda. An Evaluation. Staff Appraisal Report: Pilot Female Lit- Washington: World Bank in collaboration eracy Project". Washington DC: World with the Ministry of Gender, Social Devel- Bank. opment and Labour, Uganda. World Bank. (1995). "Report No. 12517-SE: PAPF (1999). Partage, Novembre 1999. Senegal An Assessment of Living Condi- Dakar: PAPF tions (In Two Volumes) Volume l: Main Pearce, J. (ed). (2000). Development, NGOs, Report". Washington DC: World Bank. and Civil Society. Oxford: Oxfam GB.