Agronomic research to increase yields has a very high rate of return and policymakers and development stakeholders are interested in the distribution of the benefits of these yield increases: Who gains and who loses? Do producers or consumers benefit more from an increase in yield? Are there poverty-reducing effects of yield increases among the various typologies of farm households?
In a recent IFPRI brownbag seminar, held on October 23, 2019, Nicholas Minot, Deputy Division Director, Markets, Trade, and Institutions Division, IFPRI, Washington DC, presented a method to address these questions. Expanding on an ex-ante approach developed by Deaton, Minot and his colleague Rachel Huang used micro-simulation to assess the distributional welfare effects of a change in the yield of a crop. The researchers used this model to estimate the impact of a hypothetical 10 percent increase in cassava yields on income and poverty rates among different types of Nigerian households defined by region, farm size, net position in cassava, income category, and sex of head of household. The model was simulated using data from the 2012-13 Nigeria General Household Survey, which covered 4,802 households.
The research team chose cassava as Africa South of the Sahara is home to 8 of the top 10 cassava producers, with Nigeria being the largest cassava producer in the world. In 2011, the government of Nigeria spent US$ 400 million on agricultural research, and 10 percent of its researchers were working on cassava. In addition, the IITA works on breeding high yielding disease and climate resilient cassava varieties.
The ex-ante model simulations show that a 10 percent increase in cassava yield would generate large aggregate benefits estimated at US$ 219 million per year. Because cassava is non-tradable in Nigeria, the yield increase results in lower cassava prices. This hurts net sellers and benefits net buyers of cassava. Farmers retain 31 percent of overall benefits, and consumers receive the remaining 69 percent. Despite transferring benefits to consumers, the technology remains pro-poor as most cassava consumers are rural households and urban poor. The new technology reduces the national poverty rate by 0.2 percent and lifts 385 thousand people out of poverty. Larger farms lose from price reductions because many are net sellers.
Minot also highlighted broader implications of the study. Because most of the benefits of the improved technology are transferred to consumers (including many rural consumers), the cassava consumption patterns are at least as important as grower characteristics and adoption patterns in determining the distributional effect. Furthermore, to estimate the impact of yield increases, it is not enough to look at the composition of growers. It is also necessary to estimate the impact on prices and the effects of prices on farmers and consumers. Fortunately, this is not too difficult with the increased availability of household survey data.
Minot finalized his presentation by discussing the limitations of the ex-ante simulations. The model is based on the assumptions that the new yield-increasing technology does not affect wages, land rents, and other input prices. The model also assumes that new technology does not affect other crops or exchange rates.
The seminar concluded with a discussion and a feedback round. Participants found the model useful for priority setting and reflected on the investment that would be necessary to achieve a 10 percent yield gain particularly in Malawi. One participant noted that the model assumes complete adoption of new technology, which is typically a challenge in reality. Minot responded that the method can be used to simulate partial adoption, and the paper includes an analysis of the sensitivity of the results to partial adoption. Participants also discussed the potential of calibrating the model to analyze heterogeneous production/yield issues in Malawi.
The presentation is available as SlideShare below.
Click here to download the presentation as pdf document. (1.06 MB)
Further reading
Deaton, A. 1989. Rice prices and income distribution in Thailand: A non-parametric analysis. Economic Journal 99 (Supplement): 1-37. Article available here. DOI: 10.2307/2234068
Nweke, F.I., D.S.C. Spencer, and J. Lynam. 2002. The Cassava Transformation: Africa’s Best Kept Secret. East Lansing: Michigan State University Press. Article available here.
Takeshima, H. 2011. “Distribution of welfare gains from GM cassava in Uganda across different population groups and marketing margins.” African Journal of Agricultural and Resource Economics 6 (1): 1-20. Article available here.
