Incomplete markets, crop portfolios, and agricultural productivity

Using the Uganda National Panel Survey, I document that farmers face a trade-off between crop yields and risk, that most grow both high- and low-yield crops, and that poorer farmers concentrate on the low-yield, low-risk ones. I then develop an incomplete-markets model with a crop-portfolio choice, in which households trade off income returns against consumption smoothing. Wealth-poor households optimally under-invest in intermediates and shift toward safer crops. Crop portfolios, intermediate input use, and the wealth distribution are jointly determined in equilibrium. The model also reproduces the low pass-through of both income inequality and income volatility into consumption. Calibrated to rural Uganda, it implies that completing markets raises agricultural productivity by 12.8%, with 15.8% of the gain coming from eliminating the misallocation of inputs across farmers and crops. Consumption rises by 17.6% and welfare by 17.4%, with income inequality rising slightly and consumption inequality unchanged.