Long-term Effects of Early Life Maize Yield on Maize Productivity and Efficiency in Rural Malawi

The paper assesses the effects of maize yields just prior to birth (in utero), in the first and the second years of life on adult life productivity and efficiency of maize farmers born between 1984 and 1995 in rural Malawi. To ensure that early life maize yields are not confounded by omitted local chacteristics, they are transformed into relative maize yields by using a cumulative gamma distribution. I find that maize yield just prior to birth significantly increases maize output in a farmer's adult life. However, relative maize yields in the first and second years of life have no long-term effects on maize production. Furthermore, there is no long-term impact of early life maize yields on the technical efficiency of maize production. These findings survive a number of robustness checks including alternative definitions of early life maize yields, controlling for migration and allowing for serial correlation. Furthermore, the results are not driven by sample selection originating from survival induced by maize yields in early life. Thus, low maize productivity in early-life begets low maize productivity in adult life. The paper finds that the impact of inputs under the farm input subsidy programme (FISP) on maize productivity is almost of the same order of magnitude as the long-term impact of maize yield in utero.