Independent Evaluation - Home > Search

Implementation Completion Report (ICR) Review - Social Investment Fund


  
1. Project Data:   
ES Date Posted:
09/29/2003   
PROJ ID:
P039292
Appraisal
Actual
Project Name:
Social Investment Fund
Project Costs(US $M)
 11.67  14.05
Country:
Belize
Loan/Credit (US $M)
 7  8.4
Sector, Major Sect.:
General education sector, Health, Other social services, Roads & highways, General water/sanitation/flood protection sector,
Education; Health and other social services; Health and other social services; Transportation; Water sanitation and flood protection
Cofinancing (US $M)
 0  2.1
L/C Number:
L4142; LP288      
   
Board Approval (FY)
  97
Partners involved
IDB, DFID, PAHO 
Closing Date
12/31/2001 03/31/2003
         
Prepared by: Reviewed by: Group Manager: Group:  
Soniya Carvalho
Howard Nial White Alain A. Barbu OEDST

2. Project Objectives and Components:

a. Objectives
The overall objective of the project is to improve the access of the poor in relatively disadvantaged communities to basic social and economic infrastructure and services, largely by supporting small-scale projects, which are identified and implemented with a high degree of community participation. Achieving this objective would contribute to the country's poverty reduction goal. The project would support the government of Belize in: (i) establishing an efficient, transparent, demand-driven participatory mechanism for financing basic services and infrastructure to targeted groups; and (ii) channeling resources to community projects involving social assistance, basic infrastructure and small-scale productive activities (PAD Page 1).

b. Components
COMMUNITY PROJECTS - would be sponsored by public or private organizations able to design community projects to meet the Social Investment Fund's (SIF) established criteria. Community project generation would be demand-driven, but would be combined with promotional efforts to develop institutional capacity as needed, to target the poorest communities. Communities would make a 5-10% cash or in-kind contribution, depending on the type of community projects.

MICROENTERPRISE CREDIT (no Bank financing) - would provide credits to community-based microenterprises. Management of credits and training to community groups would be sub-contracted to a NGO or a financial intermediary with demonstrated experience [financed by European Union].
INSTITUTIONAL SUPPORT - would help develop a viable, autonomous institution to manage the Fund and to finance the administrative and operational costs of the organization for a four year period.

c. Comments on Project Cost, Financing and Dates
The original loan amount was 7 million and a supplement was approved in FY01 for 1.4 million to "compensate for the amount of loan that was diverted, with Bank agreement, for urgent repair and rehabilitation activities" following Hurricane Keith (ICR page 2). The project closed 15 months later than originally planned. Cofinancing was mobilized from IDB, DFID, and PAHO although there was none at appraisal.


3. Achievement of Relevant Objectives:

The project channeled resources to community subprojects and expanded the population's access to small-scale infrastructure. However, the project's poverty targeting, community participation, and social assistance-support objectives were only partially met, and there are issues about the project's efficiency.
The project under-spent the target amount in the poorest Districts, and poverty targeting for communities "showed a substantial bias in favor of medium-income communities..." even considering only the resources devoted to non-emergency activities (ICR page 9). Table 2 on ICR page 8 shows that Medium Poverty Areas received 186 percent of the target amount, Low Poverty Areas 71 percent of the target amount, and High Poverty Areas only 60 percent of the target amount. "Had the SIF staff devoted more time and effort to preparing the High Poverty communities to develop and submit subproject proposals, this leakage of SIF resources to the less-poor might have been avoided" (ICR page 9).
Subprojects were implemented with varying degrees of community-involvement as against the project objective of "supporting small-scale projects, which are identified and implemented with a high degree of community participation". "Project start-up delays put pressure on the SIF team to get subprojects on-line quickly. They succeeded in this respect, but this lack of time may have contributed to insufficient effort being dedicated to promoting substantial, broad-based, and up-front community participation in the subproject identification process" (ICR page 6).
The project financed relatively few social services and training subprojects. "The SIF staff had a strong engineering background and, initially, tended to promote physical output against "softer" subprojects and activities. These tendencies were reinforced by the need to show quick, visible results after a slow project start, and by the need to respond to hurricane damage in late 2000 and again in 2001" (ICR page 10). "..the portfolio of SIF subprojects might have looked somewhat different had the SIF used a more proactive participatory needs assessment process up-front, ensuring a broader representation of community members and specific consultations with women" (ICR page 6).
The number of subprojects also fell short of targets--although the project planned to implement over 500 subprojects, it was only able to implement 187 even with the supplement. The main reasons given by the ICR are underestimation of costs at appraisal, lower than expected demand for social assistance programs, infrastructure designs to better tolerate future hurricanes, and price increases as a result of the difficulty in getting inputs after the hurricane. An unanswered question is whether the kind and quality of the infrastructure built justified the average subproject cost which for water facilities and schools was $110, 000 (compared with the overall average cost of $20,000 estimated at appraisal).
The extent to which the project succeeded in creating an "efficient mechanism" for financing basic services and infrastructure is also unclear. "Over the life of the project, administration costs as a percentage of subproject investments were about 21 percent..." (compared with the 15 percent estimated at appraisal). "The relatively high cost of administration can be explained by the relatively low volume of investment expenditure, which did not enable the SIF to achieve the economy of scale of most other funds and, thus, resulted in a higher than average overhead rate" (ICR page 13).

4. Significant Outcomes/Impacts:

The ICR contains limited data on outcomes/impact--a proposed impact evaluation was much delayed and not available at the time of the ICR.
  • SIF constructed small-scale infrastructure and drew attention to the concepts of poverty targeting and community participation.
  • SIF showed flexibility and played a critical role in the aftermath of Hurricane Keith, as it was "among the very few national agencies able to respond quickly and well to the communities' emergency repair needs".

5. Significant Shortcomings (include non-compliance with safeguard features):

Pressure to implement physical outputs came at the expense of more systematic community capacity building. Some critically needed activities were not financed by SIF although resources were available for them. "Organizational strengthening allocation for up-front investment in training community groups, government agencies, and NGOs in participatory subproject identification and preparation were virtually untouched" (ICR page 10). There was a "missed opportunity for community leaders to learn participatory techniques to facilitate a community-wide process to define integrated village development plans" (ICR page 12).
  • There were coordination problems with ministries which, inter alia, reduced the project's effectiveness in engaging communities. It is "noteworthy that the original proposal that the SIF use the services of Community Development Officers of the Ministry of Human Development never materialized. With a few exceptions, those officers, although residing in the Districts, were not available to provide the facilitation and community training expected at appraisal as an integral part of SIF operations" (ICR Page 12). The use of Rural Development Officers of the Ministry of Rural Development was slightly better, but also problematic (ICR page 15).
  • The SIF tended to consult with the communities during the Community Needs Assessment exercise only after having received a specific subproject request from the Village Councils (which are the legitimate political community representatives, but do not necessarily reflect the needs of the poorest)" (ICR page 6). "While the Beneficiary Assessment conducted in August 2000 confirmed that the selected subprojects were considered priorities by the communities, the portfolio of SIF subprojects might have looked somewhat different had the SIF used a more proactive participatory needs assessment process up-front, ensuring a broader representation of community members and specific consultations with women" (ICR page 6). The Beneficiary Assessment pointed out that "..there is a need (to) focus much more on the quality of community participation and contribution in terms of decision-making processes, equal participation opportunities for men and women, information sharing, democratic communication and leadership" (ICR page 15). The ICR notes that "...the risk of not investing ample time in empowering communities to analyze and prioritize community needs, plan, and implement the subprojects financed by the SIF is that funds are spent in ways that are not considered to be the highest priority in the beneficiary communities and that communities fail to maintain these facilities once they are built" (ICR page 18), and that "...no quantitative data are available to substantiate the report's conclusion that the SIF output met the community demands" (ICR page 19).
  • Monitoring and evaluation was particularly weak. The Living Standards Measurement Survey (LSMS) which would have provided more substantive data on the impact of SIF-funded activities was delayed due to lengthy design and implementation discussions, lack of experience of local statistics staff, and hurricane Keith, and was not available at the time of the ICR. The project's MIS was a problem throughout project implementation and had yet to become fully operational at the time of the ICR. Data provided by it was sometimes of questionable reliability.
  • There were time delays in executing the project cycle. "Still, in some cases, the project cycle was slow..." (ICR page 13) and "A number of NGOs were of the opinion that the SIF's procedures were too involved, bureaucratic, and time consuming. The months that passed between proposed preparation and disbursement of sub-project funds caused the NGOs to lose credibility in the eyes of their beneficiary groups" (ICR page 10).
  • 6. Ratings:ICROED ReviewReason for Disagreement/Comments
    Outcome:
    SatisfactoryModerately UnsatisfactoryMajor weaknesses in poverty targeting, less community participation than indicated by the project's objective ("supporting small-scale projects which are identified and implemented with a high degree of community participation"), a major shortfall in delivering the expected number of community projects (187 against the appraisal figure of "over 500"), an administrative cost exceeding that projected at appraisal (about 21 percent against the 15 percent appraisal figure), and a very weak MIS (despite the project's objective of creating an "efficient, transparent" financing mechanism). See Sections 3 and 5 above.
    Institutional Dev.:
    ModestModest
    Sustainability:
    LikelyLikelyAdditional information further to the ICR provided by the Region that there was good coordination between the Social Investment Fund and the Ministry of Education with the latter meeting its obligations for both the operations and the maintenance of the schools financed through the Social Investment Fund, and that "each beneficiary community was subject to a rapid needs assessment and had to establish a user committee,..each user committee received training on O&M from SIF staff".
    Bank Performance:
    SatisfactorySatisfactorySatisfactory overall but greater attention to poverty targeting, community participation, subproject costs and demand for different types of subprojects, and monitoring and evaluation would have enhanced Bank effectiveness.
    Borrower Perf.:
    SatisfactorySatisfactorySatisfactory overall but greater attention to SIF management capacity (weaknesses in which caused substantial delays in project start-up), composition of staff skills mix (with appropriate balance between engineers and community development experts), and due priority to monitoring and evaluation (with strong MIS and impact evaluation) would have improved Borrower Performance.
    Quality of ICR:
    Satisfactory

    7. Lessons of Broad Applicablity:

    Based on the ICR, key lessons include:
    • Poverty targeting needs special attention in a demand-driven approach which, as in Belize, tends to favor communities better able to prepare viable subproject proposals who typically are the less poor.
    • The trade-off between operational speed and intensive community participation is a real one and should be explicitly acknowledged and addressed, and reflected in performance indicators.
    • It is difficult to transform a SIF from an engineering outfit to a community development organization. In Belize, a successful transformation would have required a culture change within the SIF and was not easy. The appropriate role and focus of the social fund agency should be anticipated to the extent possible from the outset and SIF staffing and institutional arrangements should reflect that.
    • Inter-agency cooperation needs considerable and explicit attention and cannot be taken for granted. There must be appropriate incentives for cooperation on both sides.

    8. Audit Recommended?  Yes

              Why?  It would contribute to OED studies on community driven development and on natural disasters. It would also help to verify the ratings.

    9. Comments on Quality of ICR:

    A good ICR overall that points to useful lessons.

    © 2012 The World Bank Group, All Rights Reserved. Terms and Conditions