Senegal’s recent program to combat unemployment is widely regarded as a development success and a model of efficient and effective public works management. The public works and employment agency (AGETIP) has become well known in the region for managing labor-intensive public works projects with transparent, streamlined procedures, and for significantly improving the management of donor finances. Its approach has been widely replicated.
Between 1989 and 1996, the World Bank supported AGETIP with two credits. An OED audit* of the two projects highlights AGETIP’s success and notes that the Bank worked very effectively to pilot, monitor, and adjust the projects before expanding them.
The audit, which included a special independent assessment by beneficiaries, sheds new light on the sustainability of the subprojects and warns that, in spite of AGETIP’s success, grassroots participation by local communities will have to increase if benefits are to last over the long term. Beneficiaries, in their assessment, said they felt left out of the identification and planning process. Without their full involvement, the maintenance of subprojects and even the sustainability of the program may be at risk.
In the early 1980s, despite a decade of structural adjustment, Senegal continued to face serious economic problems. These were aggravated by political unrest stemming from high unemployment in urban areas. The government realized that a solution to high unemployment needed to be found rapidly, but it also knew that government agencies and public enterprises could not deliver construction and service programs quickly and efficiently. Thus, the government created the Agence d’Exécution des Travaux d’Intérêt Public (AGETIP).
AGETIP was granted special legal status, with its own charter and bylaws and with exemptions from government procurement and disbursement procedures, as well as from staffing and salary policies. The basic operating principle of the agency is delegated contract management (see box 1). Over time, AGETIP has acquired a virtual monopoly in construction management.
The World Bank supported AGETIP with an IDA credit (AGETIP I) of $20 million approved in 1989 and a second credit (AGETIP II) of $39 million in 1992. A third project, aimed at increasing the capacity of local governments, is now under consideration. The main objectives of AGETIP were to create temporary new jobs, increase the capacity of the construction industry, demonstrate the feasibility of its approach to delegated contract management, and invest in economically and socially useful subprojects.
The projects included a program of urban public works, construction, rehabilitation, and maintenance; a package of services related to work organization, management, and training; and a component for supervision and monitoring. AGETIP II also included additional pilot activities for the application of AGETIP’s experience in government procedures; the program’s expansion into rural areas; increased grass-roots participation; a feasibility study for a mutual guarantee company; and a lending instrument for municipalities.
Preparation and appraisal of AGETIP I successfully followed a learning cycle. A first project preparation facility was set up in late 1988; the project was appraised in March 1989; and a procedures manual was prepared and discussed with the government. Then, a second project preparation facility was created in May 1989, which fine-tuned the process by hiring staff and carrying out five pilot subprojects. Only after this phase was completed did full-scale implementation begin.
AGETIP I became effective in March 1990. AGETIP II followed in 1993 and expanded into rural areas, establishing three regional offices to decentralize decisionmaking. Difficulties in obtaining counterpart funds from the government and municipalities delayed startup of AGETIP I. The situation worsened in 1993-94 and IDA almost suspended disbursements in 1995. To ease the funding crisis, caused in part by a dire shortage of municipal finances, the share of municipal counterpart funds was reduced from 50 to 10 percent, and donors agreed to provide credits to fund municipal contributions. The project had hoped to boost the municipalities’ sense of ownership of the projects by increasing the municipalities’ financial contributions. But this goal was not reached.
In spite of these problems, AGETIP enjoyed strong political support and benefited from outstanding performance by both the executing agency and Bank staff. Especially noteworthy was the continuity of Bank staff working on the project.
Both projects were carried out successfully and in line with expectations. By February 1996, AGETIP had executed more than 1,250 subprojects and created almost 18,000 person/years of employment. If all subprojects are completed, AGETIP expects to have created a total of about 35,000 person/years of employment. At the time of the audit, the ratio of labor costs to the total costs of most subprojects (indicating how labor intensive they were) varied between 23 and 28 percent, above the threshold of 20 percent but below the expected level of 33 percent. Substantial capacity has been built in the construction industry. The number of engineering firms rose from about 80 in 1991 to more than 300 in 1996, and the number of registered construction enterprises soared from 350 to 1,900 during the same period.
Pilot activities had varying degrees of success. AGETIP successfully expanded into rural areas, and its procurement experience was used in a review of government practices and legislation. Training and study activities were carried out successfully. But efforts to create a joint venture with municipal credit funds to finance infrastructure projects failed.
AGETIP was established as a temporary agency. But its encouraging results stimulated an increase in activities, and the Bank and international donors eventually gave AGETIP responsibility for construction management in a large number of other projects. AGETIP also manages development projects by assuming functions that go beyond contract management. For example, it is managing a French-funded microenterprise credit fund and two separate IDA-funded projects in nutrition and female literacy.
AGETIP has become a highly visible and powerful organization. It is lean and efficient. For example, accounts are paid in an average of three days and the procurement process is completed in less than two months. About 90 percent of contracts awarded by AGETIP are completed within the specified time.
OED collaborated with the German donor agency GTZ to carry out an independent beneficiary assessment. The assessment took about four weeks in 1996, including three moderated workshops in Dakar, Saint Louis, and Diourbel. The assessment team visited 23 project sites and held 33 semi-structured interviews with beneficiaries. In all, the team listened to 159 individuals. Although the findings of this rapid assessment cannot be statistically representative, they offer a reasonable indication of AGETIP’s strengths and weaknesses.
While opinions differed, the need for improved maintenance and participation by local people was a dominant concern of almost all beneficiaries. The assessment found that three groups of intended beneficiaries—municipalities, small construction firms, and consulting engineers—have been directly involved in subprojects. But the fourth and largest group of intended beneficiaries—the local people—feel left out. This has important implications for the future: since local communities do not feel they have ownership of the projects and municipalities have few if any resources for maintenance, the upkeep of assets created by subprojects remains in doubt.
Many municipal administrators felt that the subprojects were in line with the people’s needs, but other groups pointed to the lack of consultation and to projects that were not in line with priorities. (For example, one respondent commented that “neighborhood residents would have preferred a sanitation project instead of sidewalks.”) Respondents acknowledged, however, that some projects clearly originated from people’s initiatives or reflected community priorities.
Engineers and entrepreneurs noted that AGETIP could expand appraisal procedures to include
social assessments and take into account earlier studies or proposals. At present, only technical engineering studies are included in the preparation process. But some beneficiaries expressed the concern that expanding AGETIP’s mandate and activities could reduce its efficiency.
Most beneficiaries agreed that operation and maintenance needs to be improved. Beneficiaries criticized municipalities for poor maintenance but officials from municipalities pointed to the lack of resources. The assessment concluded that the real needs of the population had not been understood and that beneficiaries had not been able to participate sufficiently in project identification, resulting in little incentive to maintain assets. Those subprojects that did originate from local initiative were in fact relatively well maintained.
AGETIP has become a permanent, fee-based institution (fees fully cover operating costs) with good financial performance and management. It has created jobs, some of them permanent, and indications are that construction activity in Senegal will continue to increase in the short to medium term. AGETIP has significantly improved the management of financial resources from donors and proven itself to be an efficient and transparent management company. The agency has proved that procedures for bidding, contracting, and disbursing small public works can be made competitive, simple, and flexible. Its procurement practices seem to be leading to reforms in the government’s procurement procedures. AGETIP’s approach has been successfully replicated in several African countries, and the agency has forged networks with other social funds in Latin America.
AGETIP included a substantial monitoring and evaluation program but little is known about the impact of AGETIP subprojects beyond the number of workdays created. While information on the use of inputs and of outputs and outcome is abundant, there are few indications of the economic and social impact of subprojects or of the permanent employment created.
The sustainability of benefits created by the projects is uncertain, mostly because of operation and maintenance problems caused by low participation by local communities and lack of municipal funds. A number of subprojects, such as roads and community centers, have already deteriorated and need rehabilitation. Others do not seem to meet the priorities of the communities.
The projects’ institutional impact was substantial. They improved the management of financial resources, developed new and innovative solutions to public works contract management, and replicated regional capacity building. There are several reasons for AGETIP’s success. It was set up as a nongovernmental organization with a results-oriented management culture; it enjoys strong political support; it follows high standards of efficiency and transparency; it has a highly motivated and dedicated staff with strict ethical and professional standards; it has financial decisionmaking autonomy; and it provides training for its partners (enterprises, consultants, and municipal managers).
• To ensure sustainability, beneficiaries must be closely involved in project identification and planning. But the question of how to ensure such participation remains. Some believe that AGETIP should be granted greater responsibilities for project identification and for operation and maintenance. But experience with integrated rural development does not support this idea: transferring government responsibilities to project units does not lead to sustainable institutional development, and experience with rural development funds points to the dangers of overburdening their capacities. What is needed is capacity development at the municipal and community level.
• Careful project preparation and sequencing of actions (piloting, monitoring, and adapting before expanding) were crucial to the success of AGETIP I and II. So were the persistence and continuity of Bank staff, as well as political backing by the government.
• AGETIP’s success has important implications for its future, as well as for the future of other similar agencies in Africa. Construction management is carried out mostly by the private sector. Thus, allowing AGETIP to hold a preferential monopoly status may not be an appropriate institutional solution. A distinction may need to be made between commercial construction management and management of social projects.
Box 1: The AGETIP approach to delegated contract management
Delegated contract management works as follows: the originator of a subproject (in most cases municipalities) submits a proposal to AGETIP, which appraises the subproject and determines its eligibility. If the subproject meets AGETIP’s criteria, it enters the work program. AGETIP enters into an agreement with the originator, assuming all the rights and obligations related to the execution of the subproject. AGETIP takes no action, however, until it receives the originator’s counterpart financing (presently 10 percent).
Once it receives the money, AGETIP awards a contract for detailed engineering. It then carries out all bidding procedures, awards the contract for construction, supervises execution, and arranges payments to contractors and consultants. AGETIP also contracts specialized consultants for quality control. When the project is completed, AGETIP arranges for the transfer of project assets to the originator. AGETIP also offers frequent seminars for municipalities, contractors, and consultants to explain its procedures and mandate.
Box 2: From AGETIP to AFRICATIP
While labor-intensive projects are not new, AGETIP has been successfully replicated in several African countries. In order to benefit from each other’s experience, African AGETIPs in 1993 created a regional association, AFRICATIP (Association Régionale des Agences d’Exécution des Travaux d’Intérêt Public), as a formal networking organization for the programs. Under the aegis of specific executing agencies, AFRICATIP organizes meetings that have resulted in substantial cross-fertilization. There are now 14 member agencies, and nine potential new members attend meetings as observers. A link has also been established with social funds in Latin America, where 11 funds have formed their own network.
*Performance audit report: “Senegal—AGETIP: Public Works and Employment Projects,” by Robert van der Lugt and Thomas Kuby, forthcoming. Available to Bank executive directors and staff from the Internal Documents Unit and from regional information service centers. Précis written by Robert van der Lugt and Stefano Petrucci.