The accident rate on the project roads was an indicator for the achievement of the project outcome. However, data were unavailable and improvements in road safety could not be assessed quantitatively.
Despite fielding 13 missions, ADB failed to learn of the government’s efforts to undertake component 3B, indicating a possible lack of communication, which needs to be improved.
Providing fiscal resources was clearly urgent to ensure uninterrupted service and to reconstruct houses after the political disturbances. However, providing funds for undamaged but old water supply and sanitation systems was not as urgent and should have been completed as a separate project.
According to the PCR (para. 9) national contractors lacked experience in environment management and engaged national environment consulting firm to prepare site specific EMPs. The lapses that occurred in the initial pre-construction stage were addressed with ADB’s support and advice.
This is to identify a plausible causal relationship between investments and actions taken by the program and a change in the performance indicators. This was not the case in the performance indicators under output 1. Even under the most optimistic scenario, it is difficult to envisage that the financing of the 2016 national budget could have impacted the contribution of medium- and small-medium-sized enterprises sector to GDP, labor productivity in agriculture, and the volume of nonresource exports by enterprises in the same year.
ADB’s Project Administration Instructions stated that the use of conditions for withdrawal should be limited to the extent possible. In this program, four disparate conditions were included as conditions for disbursement without discussion or justification in the RRP.
The DMF was not well designed and had deficiencies in the measurement of performance indicators for the project impact and outcome. Target dates should have been indicated; and the five outputs of the project involved the implementation of three components. These should have been integrated more closely with the outputs of the DMF.
Procurement of IT systems—including hardware, software, and related communications equipment— requires the involvement of specialized personnel. However, this is not a full-time job, and on-call consultant support could have been considered for the executing agency.
The bid price for the GFMIS application and main servers should have been better estimated. The bid was 43% above the price.
Although the supporting TA addressed issues related to budget preparation and execution, continued support for the entire PFMR process is needed.
PFMR is multifaceted and requires sufficient time to implement. Supporting TA and project implementation should go hand in hand. The PMO was established to manage the procurement process with TA support. However, central treasury staff had to carry out both project-specific tasks and their day-to-day functions. This led to competing priorities between their own jobs and project requirements, and delays in dealing with project issues.
Knowledge of procurement procedures is a key aspect of most ADB funded projects and a lack of understanding of the two-stage process should have been rectified earlier. Further training should have been provided to the executing agency, especially since it had limited exposure to ADB procedures and to the prices of the goods and services to be procured under the project.
The number of projects and subprojects to be implemented under any assistance package must consider the absorptive capacity of local contractors, availability of construction materials, and local labor. Close attention must be given to the quality of the facilities and infrastructures built. Utilizing local contractors and labor have their trade-offs, such as implementation delays and deficiency in financial resources, but the quality of the completed facilities is nonnegotiable. This would require closer supervision, frequent monitoring, and faster processing of payments.