UC Berkeley

This dissertation consists of two essays on agricultural trade in sub-Saharan Africa. The 42 countries of continental sub-Saharan Africa include 21 of the 24 poorest countries in the world. Unlike industrialized countries where structural transformation and income growth have led to declines in the share of agriculture in overall output and consumption, nearly two-thirds of the labor force in sub-Saharan Africa still works in agriculture and nearly half of consumer expenditure is on food. Agricultural products are produced by tens of millions of farmers and consumed by hundreds of millions of consumers across Africa. In this dissertation, I show that the costs of trade between producers and consumers in different locations are very high, I explore the consequences of these high trade costs, and I evaluate the effects of a type of trade policy that has been used to insulate markets in particular countries from high and volatile prices elsewhere. My findings can be used to improve the design and understand the impact of infrastructure investment, trade liberalization, agricultural technology adoption, and price stabilization initiatives in Africa and elsewhere in the developing world.

In the first chapter, I estimate and solve a dynamic model of agricultural storage and trade in sub-Saharan Africa using a new intra-national dataset of monthly prices and production of the 6 major staple grains from 2003 to 2013 and a new approach to identify cost parameters when trade and storage are unobserved. The model includes monthly storage in each of 230 large hub markets in all 42 countries of continental sub-Saharan Africa, monthly trade between them, as well as monthly trade with the world market through 30 ports. I find median intra-national trade costs over 5 times higher than elsewhere in the world along with significant extra costs for trade across borders and with the world market. I then simulate a counterfactual in which trade costs for staple grains are lowered to match an international benchmark. Lowering trade costs results in a 46% drop in the average food price index, a 42% loss of net agricultural revenues, and a welfare gain equivalent to 2.2% of GDP. I show that 86% of this welfare gain can be achieved by lowering trade costs through ports and along key links representing just 18% of the trade network, supporting a corridor-based approach for infrastructure investment and trade policy. In an extension, I find that the effects of agricultural technology adoption depend crucially on trade costs, with technology adoption increasing farmer incomes only when trade costs are low. Compared to my dynamic monthly model with storage, a static annual model of agricultural trade underestimates trade costs by 23% and welfare effects by 33% by failing to correctly identify when trade occurs.

In the second chapter, I investigate the empirical effects of temporary export restrictions, which have been widely used by many countries in sub-Saharan Africa and elsewhere in recent years in an attempt to stabilize domestic prices of staple grains. I use monthly, market-level price data from a 10-year period during which 13 short-term export bans on maize were implemented by 5 countries in East and Southern Africa. I find no statistically significant effect of export bans on the price gaps between pairs of affected cross-border markets. My results for price gaps match those from a simulation of the model developed in the first chapter in which export bans are not implemented. However, prices and price volatility in the implementing country are significantly higher during export ban periods in the data than in the model simulation with no bans. Export bans in the region are imperfectly enforced, divert trade into the informal sector, and appear to destabilize domestic markets rather than stabilizing them.