2000 Project Evaluation Results
Prepared by the UNCDF Evaluation Unit

---

Bangladesh

---

I. Basic Project Data

Project Number:	BGD/91/CO6
Type of Evaluation:	Final
Project Title:	Integrated Aquaculture (Duckweed)
UN Cooperating Organization:	UN Office for Project Services (UNOPS)
Government Executing Agency:	Ministries of Fisheries & Livestock
National Implementing Agency:	PRISM (Bangladeshi NGO)
Sector:	Fisheries (440) Crops (420)
Sub-sector:	Aquaculture/Crop Improvement & Management
Date Project Approved:	September 12, 1983
Date Project Began:	September, 1993
Date Project Ended:	September, 1998
Date Project Evaluated:	November 7 - December 15, 2000
UNCDF Budget:	US$     1,982,715
Gov't. Budget:	Tk. 1,527,000
Total Budget:	US $ 2,022,377
Actual UNCDF Expenditures:	$1,954,671

 

II. Background

In recent years, the viability and profitability of fish farming in Bangladesh was demonstrated in a number of projects; such as the DANIDA-funded project involving 1,000 small producers in Mymensingh. Given the high population pressure, the rising demand for fish and the decline of the natural fish stock due to environmental degradation of fish habitats, fish farming is an important option for Bangladesh. The high price of fish feed in Bangladesh highlighted the need for alternative low-cost fish feed. The main rationale for this project was to demonstrate the viability of local duckweed (lamnaceae) cultivation for use as a fish feed substitute, given its high protein content. It was hoped that the use of duckweed as a fish feed substitute would promote fish farming, ultimately increasing the incomes and improving the diets of fish farmers and their families.

Since 1989, PRISM, a Bangladeshi NGO, has conducted pilot duckweed and fish production in Mirzapore on land donated by Kumudini Welfare Trust (a Bangladeshi social welfare organisation previously supported by UNCDF). PRISM started cultivation of three native species of duckweed in village ponds using inputs of inorganic fertilizer and water, while at the same time cultivating a poly culture of carp and monoculture of tilapia. The yields from fish production were found to be consistently high. On the success of this initial pilot, PRISM established a second project site in Manikgonj district (Shibaloy Thana) and set up phase II of the project for which UNCDF funding was requested.

Project activities started in September 1993 and ended in September 1998. The institutional arrangement was for PRISM to provide the credit and inputs needed to run fish farming enterprises in return for a percentage share of the profits. Originally, the enterprises were designed to run as joint stock companies involving both the landless and land-holding farmers. Following the recommendations of the 1995 mid-term evaluation, and in response to field conditions where joint stock companies were not viable as a structure for producer groups, the enterprises are established as informal groups, most of which are family-based enterprises.

III. The Project

Development Objective

The development objective was to contribute to improving incomes and nutritional levels of landless and marginal farmers by providing them access to the technology and credit financing needed to enter into high-return fish farming.

Immediate Objectives

The immediate objectives were the establishment of small-scale enterprises involved in intensive fish-farming and duckweed production. At the end of the project period, 1,000 farmers, organized in small-scale enterprises, were to have doubled their incomes based on a sustained fish and duckweed production of 5 metric tons per hectare and 200 metric tons per hectare, respectively.

The expected project outputs were to have:

1. 100 duckweed cum fish production units in operation on a self-sustained basis producing 500 tons of fish and 20,000 tons of duckweed yearly.
   
2. Credit and technical assistance provided to 1,000 pond operators to sustain production with only casual or no assistance.
   
3. Attained yields of 5 metric tons of fish and 200 metric tons of duckweed per hectare.
   
4. Enrolment of 750 landless and marginal farmers in the programme.
   
5. Enrolment of 375 women in the programme.

IV. Findings

Results

The project met its immediate objective of promoting carp polyculture in the two project areas of Mirzapore in Tangail district and Shibaloy Thana in Manikgonj district. The project data shows that between 1993 and 2000, 846 group members were mobilized and organized into 157 small-scale enterprises of which 112 are informal groups and 45 are joint stock companies. This follows closely the project target of organizing 1,000 farmers into 100 small-scale enterprises by the end of the project period. Most of the groups are currently family enterprises and the participation of women is, in most cases, not as independent managers, but as family members, though the women were found to be familiar with the fish culture technology.

The average fish production per hectare per year is close to 7 metric tons, which is higher than the expected yield of 5 metric tons per hectare per year. For the same period, however, the total fish production was 1,845.32 metric tons or a yearly average of 263.6 metric tons against the yearly target of 500 metric tons and total duckweed production was 6,191.17 metric tons or a yearly average of 884.45 metric tons against the target of 20,000 metric tons per year. The reason for this is related to the limited success in meeting the other immediate objective of promoting duckweed as the sole fish feed; the main reason being that duckweed can only be grown under natural conditions for 4 months of the year. As such, duckweed can be cultivated during the dry seasons only with irrigation, requiring the availability of bodies of water and imposing additional costs, affecting the profitability of duckweed cultivation.

The project was successful in terms of meeting the development objectives of contributing to improving incomes and nutritional levels of landless and marginal farmers. Over 93 percent of the sampled beneficiaries mentioned eating more fish and an improvement in the nutritional status of their households as a result of involvement in the project. However, the project has mainly succeeded in reaching the better-off groups amongst the poor and the middle farmers, with the average size of landholding of the sampled beneficiaries being 2-3 acres.

Project Design

The original project design of organising the producer groups (pond operators) into Joint Stock Companies of pond owners and the landless was not viable due to the high overhead costs of registration and tax liabilities of companies. Additional problems were encountered in terms of management, due to lack of group cohesion and conflict between members, as well as low levels of literacy and bookkeeping skills. These findings were corroborated by this evaluation.

To be successful, collective enterprises need to be homogenous in composition and members need to be highly motivated to avoid conflict and co-option of benefits by better-off and more educated members. The current strategy of focussing on family-based enterprises or close-knit and kinship groups is more viable as they minimise conflicts, are easier to manage, and as informal groups, avoid registration costs and tax liabilities. However, given the skewed structure of land and pond ownership in the country, the poorest tend to be excluded from this type of investment opportunity. It is felt that a greater effort should be made to include the poorest in project activities through more innovative programmes; such as, leasing of ponds, access to government khas lands, subsidising the excavation costs of ponds with food grains available from government programmes (e.g. Food for Work and Vulnerable Group Development), and diversification of activities.

Impact

The socio-economic impact on beneficiaries was positive: fish consumption rose; the price of fish decreased by 20-40%, and family incomes increased. There were, however, yearly variations in earnings due to risks associated with floods. The incomes were mostly used for additional investments in the crop sector, housing improvements, purchase of clothing, household effects and education of children of the beneficiaries. Sixty-two percent of the sampled beneficiaries had been able to build up assets since joining the project, and in forty-eight percent of cases, these were either wholly, or partly, financed from dividends earned from the project.

The project is sustainable in terms of the technology learnt and continuation of activities by the groups/enterprises. Ninety-seven percent of the sampled beneficiaries were of the opinion that they would continue with the new technology in fish farming even without the help of PRISM.

Following the end of the project, PRISM is experiencing scarcity of operational funds due to:

1. high fixed costs incurred in the formation of joint stock companies during the initial years,
   
2. lack of profitability of most of these companies,
   
3. poor recovery rates on loans, and
   
4. outstanding new loans, for which repayment has not yet started.

V. Recommendations

The main recommendation from the evaluation is for PRISM to continue activities in the project areas and to replicate to other areas, but based on the lessons learned. The core group of activities of the project such as group formation, provision of credit, technology and other extension advice should be continued with the caveats mentioned; particularly the problem with commercial cultivation of duckweed and its promotion as sole fish feed, and the exclusion of the poorest.

A two-pronged strategy is needed to further the goals of increasing agricultural investments and reduce rural poverty. Replication and continuation of project activities require greater efforts to include the poorest, better preparedness of groups in terms of basic literacy and numeracy skills, and better motivation and social mobilization to promote group cohesion.

The involvement of women in project activities should be better designed to enable women to take on greater managerial responsibilities. Suggested strategies include targeted women group formation, inclusion of women-headed households, and improving their access to productive inputs, including government khas lands.

Debt liabilities of some joint stock companies formed at the initial stage of the project should be written off, given that losses were incurred because of faulty project design, for which group members cannot be held accountable.

VI. Lessons Learned

The project highlights some important lessons in terms of a poverty alleviation strategy for Bangladesh:

• Investment opportunities in fish farming can be profitably exploited, leading to increased incomes for farmers and improvement in their socio-economic condition.
• The potential for addressing the needs of the poorest in the farm sector is limited by the skewed distribution of land and pond ownership. The focus therefore has to be the non-farm sector for this landless category, or a multiple set of strategies combining different farm and off-farm activities.
• Project design must ensure adequate research and careful analysis of all aspects. Project performance was affected by faulty design and inadequate analysis of important issues; such as - site selection (sandy soil and frequent floods in Shibaloy), viability and financial implications of round the year duckweed production, the financial implications of forming joint stock companies, the need to prepare the landless and collective groups with literacy and numeracy skills, adequate motivation, and learning from the experience of other NGOs in group formation.

VII. Evaluation Team

The evaluation was carried out by:

• Dr. Nasreen Khundker, Aquaculture Specialist (Team Leader)
  
• Mr. Monir H. Chowdhury, and
  
• EmTek Consulting Ltd., Dhaka, Bangladesh.