This report is an impact evaluation of two components of the Rural Business Development Program (RBD) in Nicaragua, specifically the components benefitting rice and plantain farmers on Nicaragua’s Pacific coast. The RBD program helped finance irrigation equipment, inputs, and extension services for plantain producers, as well as extension, inputs, and drying patios for rice producers in 2009 and 2010; the price of rice is a function of its moisture content, and by increasing access to drying patios the RBD program sought to increase the value of sales by producers. Using a unique data set collected for the evaluation, average impacts of the RBD program on participating plantain farmers were estimated using difference-in-differences, while average impacts of the rice component on beneficiaries were estimated using fixed effects regressions. Estimated program impacts were combined with administrative cost data to calculate an internal rate of return for the plantain program and for one aspect of the rice program (construction of drying patios).
Key results of the evaluation are:
-Estimated impacts suggest that the RBD rice program raised yields and revenues by 11% in the 2009 - 2010 growing season and 18% in 2010 - 2011 on average relative to the yields and revenues beneficiaries would have obtained without the program.
-The estimated ERR of the drying patio component of the RBD rice program ranges from 7% under the assumption of a five year lifespan for drying patios to 27% when assuming a 20 year lifespan.
-No ERR was estimated for the input/extension component of the rice program; part of the value of the extension/input bundles received by beneficiaries was to be paid back by each beneficiary farmer to his or her cooperative, and the sum total of payments was to serve as seed money for credit funds managed by cooperatives in future years. The reliance of future benefits of this program component on farmer repayment performance makes its long-term value uncertain; therefore I chose to focus my ERR calculation on the drying patios.
-While average impacts and the drying patio ERR are both positive, RBD rice program impacts are estimated imprecisely, and we cannot reject the null hypothesis that average impacts of the program on yields and revenues were zero.
-Estimated impacts of the RBD plantain program were large and significant for revenues and yields of first quality plantains (harvested plantains come in three different quality grades).
-The average impact of the RBD program on sales suggests that the program raised the value of plantain sales by 72% relative to what beneficiaries would have obtained without the RBD program.
-I use the estimated impact on revenues as well as administrative data on production costs and program costs to estimate the ERR of the plantain program.
-The plantain ERR ranges from -23% when assuming a lifespan of 5 years for irrigation equipment and 13% when assuming a lifespan of 20 years.
-The apparent discrepancy between large impacts on sales and the modest ERR is a reflection of the high cost per beneficiary of the program, which was around $15,062 per farmer for the cohort studied here (around $3.6 million divided by 239 beneficiaries).
Kind of data
Sample survey data [ssd]
Agriculture and Irrigation
The districts of Leon and Chinandega
Unit of analysis
450 farm households, including 300 rice producers and 150 plantain producers
Producers and sponsors
Millennium Challenge Corporation
Rice: To be considered, producers must meet several criteria, which include having sown at least 2 manzanas in rice at some point in the past (1 manzana is roughly 0.7 hectares), owning no more than 50 manzanas of agricultural land, and being at least 20 years of age. MCC chose rice cooperatives based on their ability to organize a sufficiently large number of farmers to participate in the program. In the first meeting between MCC and the cooperatives, each cooperative provided an estimated number of potential program beneficiaries. Any individual farmer within these cooperatives that met the criteria, and was willing to make the matching investment, was able to participate.
Plantain: Farmers must have at least 2 manzanas irrigated land, year-round access to water, and a maximum of 20 non-irrigated manzanas in agricultural land. Any individual farmer within these cooperatives that met the criteria, and was willing to make the matching investment, was able to participate.
Deviations from sample design
The original RBS research design consisted of geographically-based randomization, in which farmers in randomly chosen areas would be eligible to participate in the program. However, this randomization and did not take rice and banana farmers into account. In the case of these two crops, farmer cooperatives were chosen for project eligibility based on their degree of organization. Rice and banana farmers belonging to these cooperatives were allowed to participate if they met certain farmer characteristics, had legal land titles, and were willing to make the necessary matching investments. This is a very different design than one based on a geographical randomization, and calls for a different evaluation strategy.
Dates of collection
The evaluation surveyed small and medium rice and banana farmers in rural Nicaragua. The survey covered work, land, quality of life, and opinions on the economic stability of respondents and their families.
Fundación Internacional para el Desafío Económico Global