The genetic and economic impact of the CIMMYT wheat breeding program: a policy analysis of public wheat breeding

Doctor of PhilosophyDepartment of Agricultural EconomicsAndrew P. BarkleyPrevious studies show that there has been a deceleration in world wheat yield growth, specifically in irrigated areas, which has led some to believe that the potential for genetic gains is slowing. Some reports claim that the Centro Internacional de Mejoramiento de Maiz y Trigo (CIMMYT) breeding program "reached a plateau" in the 1980s. Such a breeding plateau would have global ramifications, since it is often poor consumers who benefit the most from yield enhancement of staple crops including wheat. CIMMYT estimates that by 2020, the developing world will need 40% more wheat than it consumes today. Because of the lack of involvement by private breeders in most low-income countries, CIMMYT, whose germplasm is used extensively in the developing world, will need to ensure that modern varieties that they release are increasing in yield to meet the rising wheat demand in the developing world. CIMMYT, a non-profit organization, distributes improved germplasm to national agricultural research systems (NARS) for worldwide utilization. CIMMYT has consistently invested a large amount of public expenditures in wheat breeding research each year for several decades. Estimates of the impact of the wheat breeding program on increasing wheat yields provides information to scientists, administrators, and policy makers regarding the efficacy and return to these investments. Quantitative estimates of yield improvements due to the wheat breeding program provide important information for future funding decisions. Wheat lines released by CIMMYT during 1962-2002 were analyzed to estimate genetic yield increases associated with the CIMMYT breeding program using test plot data from the Yaqui Valley in Mexico from 1990-2002. Using several econometric techniques including a Just-Pope production function to account for multiplicative heteroscedasticity across the different varieties, results indicate that through the release of modern varieties CIMMYT has contributed 53.77 kg/ha to yield annually in Mexico’s Yaqui Valley during 1962-2002. Estimates of the gains attributed to CIMMYT’s breeding program on a global scale equal 481.47 million (2002) USD annually from 1990-2002. CIMMYT’s average total wheat breeding cost in from 1990-2002 was roughly 13.95 million USD making the average cost-benefit ratio approximately 1:34