PACR of a project (9/82-3/91) to improve and maintain rural roads that serve potentially high food production areas in Mauritania”s Guidimaka and Gorgol regions. The project was revised in 1985 because assumptions made in the design document were found to be invalid. The UN Sudano-Sahelian Office (UNSO), under which road construction activities in the regions had been initiated, had been working with inadequate equipment, inexperienced construction staff, and within unacceptable design standards. Under the revision, A.I.D. rehabilitated 70 km of road from M”Bout to Selibaby (originally worked on by the UN project); undertook extensive training of construction brigade personnel; instituted work plans; and monitored project performance more strictly. As a result, the program achieved positive results. The project built a total of 233 km of new road, built 164 drainage structures between the regional capitals of Kaedi and Selibaby, and rehabilitated the main road between Boghe and Kaedi. The impact of the new road system on economic development in the Guidimaka and Gorgol regions has been significant; roads have facilitated access to markets and provided a means of moving social services and agricultural inputs into the high-potential food production areas. Moreover, the project has become a model for road construction in other African nations. A 35-member maintenance brigade received training in secondary gravel roads maintenance and in the operation and maintenance of road construction equipment. However, only 5 of a planned 24 people from the Department of Public Works were trained in construction planning and implementation, and in maintenance, although 4 others from a different government department received road maintenance training under another project. The project purchased equipment and visual aids for a government training center for equipment operators and maintenance mechanics. Project managers faced numerous difficulties during implementation. Tax exoneration from the Mauritanian government was not secured until after the project began and this caused procurement delays and work cessation. Unstable drainage structures required extensive repairs and additional funding. Technical expertise needs had been underestimated and the project had to add a direct hire engineer, a project development officer, and a project officer to the implementation team. Finally, A.I.D.”s system for payment of monthly invoices received from the Mauritanian government was inefficient, requiring almost 45 days to process vouchers. The following lessons were learned. (1) Invalid project design assumptions made it impossible for the project to fulfill the original output targets. (2) Periodic meetings with the Mauritanian government and other donors in the transportation field greatly improved the project”s outcome. (3) The importation of project commodities and the necessary tax exoneration for the items should have been negotiated and agreed upon in writing with all related host government agencies before the grant was signed; a great deal of time and effort was lost by the Mission in trying to obtain tax exonerations after the fact. (4) Although the Mauritanian government has exhibited sincere concern for road maintenance, it has not yet developed an institutionalized system to provide for this. A host government maintenance policy should have been included in the original agreement to guarantee a budget for long-term recurrent costs.
