Mechanizing Agriculture

Economic activity in developing countries is labor-intensive, low-scale, and family run, with substantial family managerial time spent supervising hired labor. We run a randomized control trial that subsidizes access to rental equipment markets to study the impact of the adoption of mechanized practices on labor, productivity, and managerial span of control. The intervention induces greater mechanization in the upstream production stage, with labor savings concentrated in downstream, non-mechanized stages. The reduction in worker supervision needs increases the managerial span of control and allows households to increase non-agriculture income. We use the experimental elasticities to estimate a structural model where heterogeneous farmers make labor supply decisions in the family enterprise and outside of it. The consumption-equivalent welfare from the intervention is about 5.5%, and increases to 12% when accounting for shifts in labor supply. The model provides structural estimates of the marginal return to capital at 17%, and the shadow value of family labor, 14.5% below their outside option. These estimates provide the first causal effects of mechanization.