22974 WORLD BANK OPERATIONS EVALUATION DEPARTMENT SPRING 1999 NUMBER 1S0 Capacity Building in the Agricultural Sector in Africa C APAClfY BUILDING HAS RISEN HIGH ON THE AFRICAN development agenda. The Bank, with other donors, is seeking appropriate means to work with African govern- ments to enhance their capacity to implement development pro- grams throughout the continent. Given the rural nature of most African economies and the concentration of the poor in rural areas, there is a pressing need to increase capacity to promote agricultural development. The studies discussed here explore the lessons to be learned in agricultural sector capacity building from the implementation of four agricultural projects with significant capacity-building components. The projects were implemented in Kenya, Ghana, Guinea-Bissau, and Malawi-a group of countries that is sufficiently diverse to encompass a significant segment of the range of African experience. Four Projects: An Overview Capacity Building and Development The four cases demonstrate the difficulties In three cases, outcomes of the projects faced in designing and implementing inter- were rated as marginally satisfactory; in ventions to strengthen the capacity of gov- the fourth case, Malawi, as unsatisfactory. ernments to support agriculture. They The institutional development ratings were provide powerful evidence of the need to all moderate or modest. Sustainability reexamine established modalities for exter- was rated as likely in Kenya and Ghana, nal assistance, using a realistic evaluation uncertain for Malawi, and unlikely in of existing efforts. Taken together, the find- Guinea-Bissau. At best, this is a picture of ings suggest that the Bank may need to meicepromnc.Tehiheed adjust its approach toward building capac- attention given to capacity building in ity in the agricultural sector. This overview Africa reflects a concern that development identifies some of the lessons that can be is hindered by widespread weakness in the drawn and notes areas where the evidence institutions responsible for implementing is inconclusive or contradictory. development programs. The broader 2 World Bank Operations Evaluation Department 0 environment of government capacity, which formed the can projects included this objective, and the results offer context for the implementation of these agricultural insights on such efforts, as well as raising questions projects, was negative. Poor economic performance had regarding the Bank's capacity to provide support for such resulted in declining budgetary resources that were an enterprise. spread thinly over government programs. Government The Bank has also supported capacity building on employment policies had led to overexpansion of gov- behalf of the continued provision of government services ernment payrolls, with a resulting decline in the real that are appropriate in a liberalized environment. This sup- wages of public servants and shrinkage of the funds to port has focused on the agricultural extension services and cover the costs necessary for public servants to perform the research required to provide the requisite knowledge. their tasks. Poor incentives led to declining perfor- mance. Donor efforts to insulate "their" projects from The 1'rojects the general decline in administrative capacity have had The four projects contain all three elements of this strat- only limited success, and have deepened the deteriora- egy in varying combinations: that is, promoting policy tion by further distorting the incentive system. reform, building policymaking capacity, and building Although some success has been achieved in imple- capacity for extension and research. menting structural adjustment programs, with a conse- For the institution-building elements of this strategy, quent reduction in government activities to a more the Bank used three primary instruments: manageable size and liberalization of economic policies that improved resource allocation and producer incen- * The design and implementation of organizational tives, there has been less success in reviving the capacity structures (such as planning units, extension and of public institutions. research services, and agricultural credit agencies) Capacity can be defined as the people, institutions, * The provision of short- or long-term technical and practices that enable countries to achieve their assistance development goals. It encompasses human skills and * Training. institutional and organizational structures, procedures, and systems. Although capacity is a key factor in develop- The Framework for Capacity Building ment, it is intangible. Because it is difficult to quantify and measure, it is correspondingly difficult to monitor and The Length and Complexity of the Capacity-Building evaluate. The four African cases indicate that the difficulty Process in defining capacity-building goals in monitorable terms Capacity building is a lengthy process, particularly (and the consequent absence of explicit performance where (as in Guinea-Bissau) initial capacity is very criteria) has led to capacity-building components receiving weak. Improvement may require commitment beyond the too little attention in Bank projects. customary time limits of a World Bank project. Where a Activities in each of these categories were to be successor project is justified, there is a need to program found in all four projects. But capacity building, for continuity, so that early gains are not eroded by hia- broadly defined, also includes the effect of government tus or shifting priorities. policies on the capacity of institutions and actors outside The human components of capacity-building activities government-for example, policy reforms that enhance are inherently unpredictable, which gives rise to the need the capacity of the private sector, the other main compo- for careful supervision, combined with the flexibility to nent of Bank strategy, can also be described as capacity adjust in accordance with the insights gained during building. It is probably here that Bank-supported poli- project implementation. This suggests that capacity-build- cies have had the most positive impact on capacity. ing efforts would benefit from continuity in Bank support, which has not been easy to achieve. The Bank's new lend- The Bank Approach ing instruments give more flexibility here. From appraisal Following the disappointing results of initial efforts at to completion, two of the projects each had a total of four agricultural development in Africa, the Bank concen- task managers, one had five, and Guinea-Bissau had six. trated its attention in the 1980s on improving the policy If support extended over two project cycles, the number of environment, in agriculture as well as in other sectors. task managers would presumably rise even higher. In addition to promoting adjustment policies to improve incentives, the Bank has sought to enhance the capacity Sustainability of governments to formulate agricultural policies and The essence of capacity building is sustainability-the programs by building analytical and policy-formulation creation of institutions and practices that continue to capacity in ministries of agriculture. All four of the Afri- perform after a project is completed. Precis 3 Many of the barriers to sustainability have the same term policy-reform tactics. It follows that the advisability root cause: the inadequacy of local resources to support of combining policy reform conditionality and capacity- project activities after donor funds have been drawn building objectives as project components should be down. Special incentives to project staff, explicit or assessed rigorously in project design. implicit, are discontinued, and even where resources are allocated from the government budget to fund staff, Capacity-Building Instruments money is not available to complete unfinished capital works (Ghana and Guinea-Bissau) or to provide appro- Puiblic Servicc Incentitcs and Capacity Building priate funding for the non-staff, recurrent items that had It is now widely recognized that the underpayment of been carried by the project budget. (The Malawi study, public servants is a source of capacity weakness for example, highlights the difficulties of maintaining an throughout most of Africa and a serious impediment to effective, personnel-intensive extension strategy in a the effectiveness of capacity-building interventions. very poor country.) Indeed, sustainable capacity building may not be pos- The failure to budget funds to sustain a project after sible without reform in civil service incentive structures. completion can be taken to mean that the government In the Kenya project, however, progress in capacity awards less priority to the project than was expected when building was possible even without systemic reform. the project was appraised (it should not have been imple- (It should be noted that the erosion in the real value of mented) or that project design was not based on a realistic Kenyan public sector salaries has not reached the appreciation of the scarcity of budgetary resources (it extremes seen in many other African countries, such as should have been designed to operate at lower cost). For Guinea-Bissau.) Nevertheless, the main weight of the Bank, which mounts ongoing macroeconomic and analysis suggests that such progress is likely to be public expenditure reviews, it should be possible to severely constrained in the absence of more general subject sectoral projects to tougher appraisal standards reforms. But even in the Kenya case, the positive contri- with regard to their financial sustainability, in light of bution of technical assistance in a generally deteriorat- known budgetary constraints. ing civil service environment seems to have been a case The sustainability of donor-financed projects is of the project providing, rather than building, capacity. related to two other issues-the lack of an effective That is, long-term technical assistance personnel under- system to prioritize donor-funded projects in light of took line tasks for extended periods, and in so doing future resource constraints and the tendency of donor provided more continuity than did the national staff. projects to be designed to unrealistic standards. Low salaries have a number of obvious negative consequences. It may be impossible to attract appropri- Policy Reforimz ately skilled project staff. The decline in the real value During the implementation of the four projects, the major of civil service salaries can result in grotesque differen- thrust of Bank strategy in Africa was to promote structural tials between local staff and foreign advisers, which adjustment. In the agricultural sectors this typically dampens the potential for fruitful cooperation. And inad- involved trade and price liberalization, privatization of equate incentives are likely to be associated with a high input supplies, and elimination of subsidies. Such reforms rate of turnover of the most able staff, making it diffi- were being promoted by the Bank in all four countries. cult to sustain improvements in organizational capacity. In the pursuit of policy reform, the Bank uses Some of the tactics adopted by the Bank to handle this conditionalities. This gives rise to divergent views problem are seen in these cases. The Bank cannot pay civil regarding the balance to be struck among different servants, but it can pay local consultants, allowing staff to objectives within the agricultural sector. In the pursuit of shift from line civil service posts into consultant status. policy reform, it may appear necessary to delay project Although this point is not explored in the project apprais- appraisal or interrupt implementation to bring pressure als, one function of the training opportunities provided to bear in support of the desired changes. But this may through the projects is to give additional incentives to staff. break the continuity required for effective institution However, such project-based means for boosting civil building. This issue has led to differences between OED service incentives are flawed. The incentives are unlikely and the country teams regarding where the balance to be sustainable after project completion, so that gains should be struck (as in Ghana and Guinea-Bissau). from the project are likely to be lost. Introducing special While Bank Country Assistance Strategies should be project-based incentives is likely to further distort an adjusted to the changing needs of specific countries, longer- already ineffective structure of public service incentives. term institution-building interventions should be insulated Fundamentally, project-level interventions are an from the fluctuations and uncertainties resulting from short- inappropriate approach to a systemic problem. If the 4 Precis ED OED PARTNERSHIPS AND KNOWLEDGE GROUP Editor-in-Chief: Elizabeth Campbell-Page Series Editor: Caroline McEuen cause of weak government capacity is the repression of Graphic Design: Kathy Strauss, Lunn Lestina professional salary levels in the public service in econo- Assistant: uic ureishi-Hu mies with increasingly competitive labor markets, then I' y Q q that distortion needs to be tackled through the reform of PRECIS ORDER FORM government pay structures. For the Bank, two conclusions follow. First: the Bank * Precis are available at no charge by contacting the needs to press ahead in promoting systemic civil service OED Help Desk: tel: 1-202/458-4497, or fax the form reform. This is known to be difficult, and the results of below to: 1-202/522-3125. You can also e-mail your such efforts have been modest. But the overall weakness request to: OED Help Desk@worldbank.org of civil service capacity is becoming as widespread a bar- * Mail this form to: rier to the successful implementation of development pro- Precis-OEDPK, World Bank grams as the poor macroeconomic policy environment 1818 H St. was a decade ago. Washington, D.C. 20433, USA Second: realism is needed in project design regarding attn: Editor-in-Chief the implications of recurrent resource constraints. Donors need to be more aware of possible contradictions in their Quantity Title Precis # own interventions between global appeals to the virtues of a "lean and well-paid" public service and project- and sector- level interventions that become an important vehicle for the expansion of pressure on recurrent budgets and for the cre- ation of nonsustainable government activities. * Please add my name to the Precis mailing list. O Name_ Technical Assistance The most negative conclusions of these evaluations Address_ relate to the ineffectiveness of the technical assistance City State supplied (the notable exceptions were in the Kenyan project). The evidence suggests that the large numbers of Postal Code Country technical assistance personnel included in projects reflected initiatives from the Bank during project design, Telephone sometimes against the stated reluctance of the recipients. The impression is gained that the hosts were willing to * For large orders, please contact the World Bank receive foreign experts as a means to gain access to the InfoShop at tel: 1-202/458-5454, fax: 1-202/522-1500, or other, more acceptable, components of the project (for through e-mail at books@worldbank.org example, equipment and training). *- This and other OED publications are on the Internet, at Large sums are involved in supplying technical assis- http://www.worldbank.org/html/oed tance. Evidence from these studies suggests that, in most cases, foreign experts have often not been worth the cost. TTo order our books, please contact: Even where support is through grants (e.g., bilateral and PO Box 960 trust-fund finance), the opportunity cost should be taken Herndon, VA 20172-0960, USA into account. Tel: 20172-0 The appraisals identify some weaknesses that might Tel: 1-703/661-1580 be tackled by the Bank and the recipient governments- Fax: 1-703/661-1501 including steps to ensure appropriate language skills Internet: http://wwv.7.worldbank.org (relevant in the case of Guinea-Bissau), proper backup From the World Bank homepage, select publications. from the consulting firm supplying the experts, and more e-mail: books@worldbank.org care in selecting the experts. DISCLAIMER: OED Precis are produced by the World Bank However, the pervasiveness of the failures suggests Operations Evaluation Department, Partnerships and that there may be underlying problems that will not be Knowledge Group (OEDPK), Outreach and Dissemination susceptible to resolution through procedural improve- Unit. The views in this paper are those of the Operations ments. In Africa, technical assistance came to the fore in Evaluation staff and editors and should not be attributed to the post-independence period (for most of Africa, in the the World Bank, its affiliated organizations, or its Executive 1960s). Its main function was to ease the transition from Directors. colonial rule. The new African administrations were OED Publications 5 Recent OED Precis 179 Yemen: Forging a New Consensus (available in Arabic) 178 Municipal Development Projects: Building Institutions and Financing Local Development 177 Nongovernmental Organizations in World Bank-Supported staffed with young and inexperienced (but reasonably Projects well-paid) cadres. There were shortages of nationals 176 Meeting the Health Care Challenge in Zimbabwe with relevant professional qualifications, while there 175 Development Effectiveness, 1998: was a large pool of expatriates who had gained field Opportunities in a Volatile Environment experience in the colonial period 174 Disaster Assistance eprec nteclna eid The current situation is different on almost all counts. 173 Support for Smaller Enterprises Institutional weaknesses are no longer primarily the result 171 PCR: Uganda of a shortage of trained nationals-more typically prob- 170 PCR: Bosnia & Herzegovinia lems arise from failures to mobilize or effectively use 169 Post-Conflict Reconstruction (PCR) nationals because of weaknesses in incentive systems. Staff 168 India's Dairy Revolution have often been exposed to a succession of foreign advis- 167 Philippines: From Crisis to Opportunity ers. In such situations, technical assistance is not the solu- 166 Financial Sector Reform tion to a dysfunctional incentive system. It is unlikely that 165 A Borrower's View of the World Bank a foreign expert can operate effectively, except as an 164 Rebuilding the Mozambique Economy expensive operational substitute for a national. 163 1997 Annual Review of Development Effectiveness (ARDE) The findings of these studies also suggest another 162 Kenya's Geothermal Development Program deviation from the original technical assistance model- 161 C6te D'Ivoire: Devaluation's Benefits the quality of the available advisory staff. Because tech- 160 Feeder Roads in Brazil nical assistance typically involves recruiting staff (often 159 Transport in Thailand: Lessons for Institutional Development foreign) new to the environment, there is always a risk 158 Agriculture in Hungary of failure. However, the generally poor performance of 157 Reducing Flood Hazards and Traffic Congestion in Tunisia staff in this sample, despite their recruitment by estab- 156 Best Practices in Urban Development: Learning from Brazil lished, reputable firms, suggests that there has been a 155 Two Projects in One: Lessons from Jamaica decline in the availability of suitable talent. 154 Sustaining Rural Water Systems: The Case of Mali Greater frugality in the use of technical assistance Prdcis aussi disponible en frantais would release sufficient resources to fund more generous @ http://NIwww.worldbank.org/html/oed incentives for the professional cadres of the public ser- Precis en espanol tambien dispoinible This sg t the need foral t ern @ lhttp://www.wvorldbank.org/html/ocd vice. This suggests the need for alternative approaches OEDJ Study Series to support capacity building, including consideration of 1998 Annual Review of Development Effectiveness possible tradeoffs between the use of technical assis- tance and the funding of local staff. Any such step, Developing Towns and Cities: Lessons from Brazil and the however, must avoid the pitfalls of donor interventions Philippines distorting the public sector salary structure and lack of Evaluation and Development: The Institutional Dimension sustainability of donor-supported initiatives. (Transaction Publishers) Financial Sector Reform: A Review of World Bank Assistance Traiining India: The Dairy Revolution The four cases provide insight about the effectiveness of Nongovernmental Organizations in World Bank-Supported long-term overseas training, generally for advanced Projects degrees. Such training often does not have the intended The World Bank's Experience with Post-Conflict Reconstruction impact, because the trainees are frequently lost to their OED Mutilingual Series intended employers, either because they seek more lucrative opportunities outside government (as with Assessing Development Effectiveness: Evaluation in the World econoist iniKen ove iovme attra ti Bank and the International Finance Corporation economists in Kenya), move ito more attractive oppor- Appreciation de l'efficacite du deueloppement: tunities in government, or are reassigned to other posts. L'evaluation a la Banque mondiale et ai la Societe financiere This observation should be treated with some care. internationale If individuals can find better jobs as a result of their Determinar la eficacia de las actividades de desarrollo: training, this is an indicator of the success of the train- La evaluaci6 en el Banco Mundial y la Corporaci6n Financiera ing, rather than the reverse. It would be perverse to Internacional propose that training be limited to the provision of Cote d'lvoire Revue de l'aide de la Banque mondiale au pays nonmarketable skills. Training needs to be coordinated Philippines: From Crisis to Opportunity with personnel and incentive policies, in light of a Filipinas: Crisis y oportunidades realistic assessment of local labor market conditions. Rebuilding the Mozambique Economy: If an officer is offered advanced training, it should be Assessment of a Development Partnership because that person is on a career track that should lead Reconstruir a Economia de Mocambique 6 World Bank Operations Evaluation Department to appropriate promotion and improved emoluments fol- Bank needs to develop country capacity-building lowing successful completion of the training. In a liber- strategies with an extended, although clearly alized labor market, the educated will expect to share in bounded, time commitment. Discrimination is the returns to an investment in human capital. required to avoid an involvement that is too brief to create sustainable capacity, as well as indefinite, Lessons open-ended commitments that generate dependency. From the many detailed points suggested by these studies (4) Capacity is not easy to define, and assessing capacity- for future Bank projects in support of African agricul- building performance will always be somewhat sub- tural development, six general issues stand out: jective. As a result, the capacity-building elements of (1)The four projects provide further evidence of the diffi- a multipurpose project can easily become the "soft" culty of achieving sustainable capacity development in segment, not subject to rigorous design, appraisal, individual ministries of governments with systemically and supervision. The mediocre to poor outcomes in weak civil services. While it may be possible to build the capacity-building components of these four temporary incentives into a project to attract a high projects suggest a need to give more explicit attention level of commitment, that commitment is unlikely to be to the design and implementation stages of such sustained after project completion unless the overall interventions. incentive system is adequate. This suggests that capac- (5) The Bank's current strategy of reducing use of long- ity building in particular segments of the government term expatriate assistance is appropriate. Such apparatus should be tackled in the context of broader assistance should not be offered where weakness of civil service reform. In the absence of such systemic capacity results from failure to use the available reform, performance expectations should be modest. national professionals effectively, but only after care- (2)As a corollary, project design should be founded on ful analysis of whether the capacity gaps are the sort realistic assessments of the domestic resources avail- that technical assistance can be expected to fill (and able to sustain project activities. This point is par- only when adequate Bank supervision is built into the ticularly important in relation to projects directed to project design). labor-intensive service provision and to training. It is (6) Sustainability is the essence of capacity building. If too easy for those designing projects that they con- sustainability is uncertain or unlikely (as in two of the sider to be of high priority to err in the direction of evaluations), it is doubtful that institutional develop- wishful thinking about the future availability of gov- ment impact can be viewed even as modest. After ernment finance. decades of technical assistance that has left so little (3) Capacity building is likely to require assistance for sustainable capacity, the application of tough-minded periods beyond the normal Bank project cycle. The evaluation criteria is an appropriate response. *This Precis was written by Edward B. Rice, based on his OED evaluation work.