Project assistance completion report : rural sector grant project no. 633-0077 (1980-1988)

The report focuses mainly on Phase II of the project, which began in 1983 and which funded subprojects (SP's) designed to increase rural income and employment. The project suffered from an extraordinarily complex, poorly integrated design; lack of involvement by the Government of Botswana (GOB); and, most seriously, a 6-year drought (which began in 1982). When the communal area coordinator left the project prematurely in 1987, progress slowed considerably, indicating that the managerial and TA requirements of the project had been underestimated throughout. Despite these problems, there were some notable achievements. (1) The SP to train a cadre of rural industrial officers (RIO's) was enthusiastically supported by the GOB and is probably the project's greatest success. Currently, the cadre deals directly with developing opportunities for rural income generation. However, the amount of work done by the RIO's was not as extensive as hoped for due to the amount of time required to implement the GOB's financial assistance programs at each local level. Utilization of the Productive Employment Development Fund established under the RIO component remained low and few of the projected number of jobs were created. (2) Some of the senior district officers trained under the district institutional development SP have moved into other positions, but those who resumed their posts appear to have improved job performance and are generally regarded as outstanding officers. (3) The SP to establish horticultural estates was probably the least successful project component. The drought was a major constraint, as was the number and complexity of necessary interventions (e.g., construction, agricultural extension, dam building and irrigation, transport and marketing, etc.). A major lesson learned is that project design should be preceded by complete baseline studies and economic analyses. In this project it appears that detailed implementation plans were drawn up before the necessary analyses were even considered.