|1. Project Data:
ES Date Posted:
Project Costs(US $M)
Loan/Credit (US $M)
Sector, Major Sect.:
|Irrigation & drainage, Central government administration, Sub-national government administration, Other social services,
Agriculture fishing and forestry; Law and justice and public administration; Law and justice and public administration; Health and other social services
Cofinancing (US $M)
Board Approval (FY)
|Christopher D. Gerrard
||Peter Nigel Freeman
||Alain A. Barbu
|2. Project Objectives and Components:|
a. ObjectivesTo accelerate the Government's program for land privatization and restructuring of representative state and collective farms in a systematic manner and to provide models which could serve as a basis for wider geographic replicability. This objective would be achieved through the provision of:
(1) Essential support services in support of privatized agriculture: land registration and titling, farm information and advisory services, credit services, and rehabilitation of critical irrigation and drainage infrastructure;
(2) An enabling environment to build linkages between key institutions: Ministry of Agriculture, State Land Committee, State Irrigation Committee, Agrarian Reform Commission, and agricultural banking institutions;
(3) Community-based social services and formation of village groups in support of land privatization and farm restructuring.
b. ComponentsThe project had four components:
(a) Farm privatization support services (US$ 18.57 million at appraisal, US$ 11.46 actual).
(b) Rehabilitation of main irrigation water supply and drainage works in project farms (US$ 4.20 million at appraisal, US$ 8.33 million actual)
(c) Community Development Program (US$ 0.98 million at appraisal, US$ 0.55 actual)
(d) Project Management and Implementation (US$ 1.92 million at appraisal, US$ 3.46 actual)
The farm privatization support services component also had four subcomponents:
(i) Land registration services (US$ 2.32 million at appraisal, actual not available in ICR)
(ii) Farm information and advisory services (US$ 2.81 million, actual not available)
(iii) Water user's associations (US$ 1.46 million, actual not available)
(iv) Privatized farm credit services (US$ 13.44 million, actual not available).
c. Comments on Project Cost, Financing and DatesThe principal reason for the reduction in project costs was the appreciation of the US dollar during the project period, as all IDA credits and IFAD cofinancing were disbursed.
The project was extended for an additional 18 months to allow completion of additional irrigation and drainage rehabilitation works, which was financed by reallocating US$ 1.5 million of IFAD funds, in six additional villages in the same raions as the pilot farms.
It is difficult to reconcile the narrative description of the reallocation of project funds among components in section 5.4 of the ICR with Annex 2 on project costs and financing. How can one reallocate US$ 1.5 million from a component (the community development component) that only cost US$ 0.98 at appraisal? According to Annex 2, IDA funds were not spent as planned. What explains the significant reduction in expenditures on the farm privatization support services and the increase in expenditures on project management? How can it be that the rural credit component fully disbursed (US$ 13.44 million at appraisal), when the total expenditures on all four subcomponents of farm privatization support services were only US$ 11.46 million?
|3. Achievement of Relevant Objectives:|
The project successfully accelerated the land privatization and restructuring of state and collective farms in a systematic manner. Building on the success of the land registration and titling on the six pilot farms (which was completed one year into project implementation), land registration and titling was rolled out nationwide. By project closing, the privatization of former state and collective farms was 99.2 percent complete. The perceived fairness and acceptability of the process that was piloted and tested at the farm level before rolling it out nationwide was key to this success. On a given state or collective farm, land was first divided into parcels estimated as nearly equal taking into account area and quality. Then parcels were chosen by farmers, whose place in the selection queue was determined by lottery. This assured that those responsible for the preliminary division would do their best to ensure evenness in value, not knowing which they themselves might receive.
Agricultural production in Azerbaijan increased by about 50 percent between 1997 and 2002, after falling between 1991 and 1996 (when land titling began). Equally significant, farmers have undertaken significant shifts in the mix of crops produced (such as from cotton to sugar beets) as they have identified better income-earning opportunities. Winter wheat production has been introduced. Tea production has fallen. Agricultural productivity on the six pilot farms increased 250 to 300 percent between 1997 and 2001.
The independence of the national-level project management unit (PMU) from any single agency, and close cooperation among central ministries and agencies involved in the privatization process were also keys to success. In 2000, the Government renamed the PMU as the Agency for the Support of Development of the Private Sector, and charged it with management of additional World Bank and IFAD initiatives in the sector.
|4. Significant Outcomes/Impacts:|
(1) Participants at the stakeholder workshop in May 2004 cited the success of the farm information and advisory services (FIAS) component more frequently than any other as key to the overall success of the project. The FIAS unit in the Ministry of Agriculture produced and distributed booklets describing legal and management issues facing private farmers, access to credit, appropriate crop production methods and related topics, provided training and seminars for newly-private farmers, and undertook testing and demonstration of new on-farm methods. These activities completely changed the rural world view, among other things, by conveying the message that the government's rural policy was public and up for discussion, and that rural residents were to participate meaningfully in the decisions affecting them.
(2) Rehabilitation or new construction of irrigation and drainage works were completed on three pilot farms in 1999, two more in 2000, and the sixth in 2001. These activities were expanded to six additional villages in 2001 (by means of reallocated IFAD funds) and completed during the 18 month extension of the project. A subsequent irrigation project is currently being implemented.
(3) Water user's associations on the six pilot farms are distributing water, managing water infrastructure, collecting water charges, and mediating cooperation on water issues among their members. Water charges collected are covering more than 100 percent of the WUA operations, including on-farm operations and maintenance. Subsequently, the WUA model has spread throughout Azerbaijan. More than 580 WUAs, covering more than 725,000 hectares, emerged by project completion, broadly modeled on the WUAs of the six pilot farms.
|5. Significant Shortcomings (include non-compliance with safeguard features):|
(1) The actual outcome of the credit subcomponent -- representing 47 percent of project costs at appraisal -- is difficult to discern from the ICR. On the one hand, the Bank/Government implementation team made an intense effort to uncover and address the risks and constraints perceived by farmers when this subcomponent got off to a slow start with demand being lower than expected (ICR, p. 9). The credit component would likely have failed had the highest levels of government not intervened to counter the expectation of political elites that they could borrow and not repay (ICR, p. 18). By project end, the component fully disbursed, is achieving a repayment rate of over 90 percent, and has expanded into a broader geographical area, making additional villages eligible.
On the other hand, the ICR does not provide any data concerning the total expenditures of this component -- the project costs by procurement tables in Annex 2 state that expenditures on this component declined from US$ 12.80 million at appraisal to zero at completion (obviously an error). The ICR does not provide a break down of the loans by investment categories (SAR, p. 14), or by pilot farms and other farm units in the project raions. Nor does it provide any data on interest rates -- were they subsized or not? The state-owned agency (Agroprombank) that was selected as the agent for the credit program did not develop as an effective lending institution (ICR, p. 9). While it eventually functioned as anticipated, the process took longer than expected and is not shielded from being eventually forced to lend for political reasons due to its ownership status (ICR, p. 10).
(2) The NGO that was contracted to implement the community development component performed below expectations. It focused on outputs easiest to monitor and quantify -- such as building business centers and delivering classes and lectures -- rather than on searching out and encouraging fledgling economic efforts among vulnerable and commercially inexperienced village members. The community development component and the farm information and advisory services (FIAS) component implemented by the Ministry of Agriculture ended up competing against each other. The NGO's contract was eventually terminated, the remaining funds from this community development component were reallocated to the irrigation and drainage component, and the business and training centers developed by the NGO were folded at project end into the FIAS component that is being sustained by the Ministry of Agriculture.
(3) The actual quality of monitoring and evaluation is difficult to discern from the ICR. On the one hand, section 7.6 of the ICR states that M&E by the Project Management Unit was excellent. On the other hand, no baseline survey was carried out during the first year of project implementation as stated would be done in paragraph 5.15 of the SAR. Annex 3 of the ICR also points out shortcomings in M&E data collection that make it difficult to undertake a conclusive with/without project economic analysis.
|6. Ratings:||ICR||OED Review||Reason for Disagreement/Comments|
|Highly Satisfactory||Highly Satisfactory||This was a most successful farm privatization project and possibly a model for future projects in the ECA region |
|High||Substantial||The evidence presented in the ICR is not sufficient to justify a high rating. While 99.2 percent of the agricultural land was privatized, neither the credit nor community development components resulted in strong institutions in their subsectors.|
|Highly Likely||Highly Likely|
|Highly Satisfactory||Highly Satisfactory|
Quality of ICR:
|7. Lessons of Broad Applicablity:|
(1) That the process of privatizing the former state and collective farms is perceived to be fair and equitable by the participants is a key to its acceptance and success, as measured both by the speed at which the program was rolled out and by the impacts in terms of improved agricultural productivity and the changing agricultural production mix.
(2) For agricultural credit programs to be successful, it is important to address the risks and constraints that are perceived by farmers and that affect their demand for credit.
(3) NGOs do not automatically bring a grass-roots perspective to their work. In this case, the performance of the NGO in implementing the community development component was more like that of a for-profit consulting firm.
|8. Audit Recommended? Yes|
Why? This project was very successful in terms of its principal objective of privatizing former state and collective farms in a very difficult political climate, with an authoritarian government that is experiencing ongoing border disputes with its neighbors (notably Armenia). Yet the process appears to have been very open and transparent, instilled widespread ownership among the participants, and has had broad beneficial economic effects. It would be important to confirm these results, to learn lessons from this positive experience, and to fill in the gaps in the evidence that have been pointed out.
|9. Comments on Quality of ICR:|
For an intensive-learning ICR this was unsatisfactory.
As already mentioned above, it is difficult to reconcile the narrative description of the reallocation of project funds among components in section 5.4 of the ICR with Annex 2 on project costs and financing (see 2c above).
It is difficult to discern from the ICR the actual outcome of the credit subcomponent and the quality of monitoring and evaluation (see 5 above). The ICR did not provide even a minimum set of key indicators on the quality of loan portfolio supported by the credit component, including clearly defined repayment rates, non-performing loans as a percent of the total loan portfolio, and the incidence of subsidies in the program. Similar deficiencies in many ICRs for projects involving credit lines have been highlighted in OED's recent LOC study.
In addition, as outlined in the SAR, some objectives and activities of the project (such as support for central ministries and agencies) were national in scope, other activities (such as support for the land registration offices, the farm information and advisory services, the water user's associations, and agricultural credit) were regional in scope in the six raions in which the pilot state and collective farms were located, and still others (such as land surveying and mapping, and irrigation and drainage rehabilitation) were initially limited to the six pilot farms themselves. Reading the ICR for the first time, one got the impression that anything achieved beyond the six pilot farms was beyond the initial objectives of the project. The ICR should have been clearer about which objectives and activities were originally national, regional, and farm-specific in scope.