Three Essays in Agricultural Economics

Abstract

The three essays that compose this dissertation are on the topic of risk in agriculture in developing countries, where farmers are completely exposed to these risks. The first two chapters focus on price risk and the third focuses on the risk derived from atypical rainfall levels. In the first essay, I address the impacts of a temporary price support policy (a policy that stabilizes and increases prices), in Peru, on the welfare of farmers it aimed to benefit. I exploit the geographical discontinuity of the policy to casually estimate its impacts using differences-in-differences. I find that although the average potato farmer did not significantly change their total consumption due to the policy, the groups of net consumers, farmers with larger shares of potato on total farm sales, and larger farmers increase their consumption due to the price support. Moreover, I find these positive impacts vanish or become negative one year after the policy implementation, suggesting that price support policies are only effective in the short run and for farmers with certain characteristics. In the second essay, I study the production behavior of farmers when introducing a price insurance, i.e., an insurance that stabilizes prices and increases low prices. To assess the causal and pure impact of price risk and price insurance on production behavior, I run an artefactual lab-in-the-field experiment with farmers in Peru. In the experiment, farmers played three games in random order: a baseline game where only price risk exists and it is introduced randomly; a second game where both price risk and an actuarially fair mandatory price insurance exist; and a third game where price risk and the price insurance exist, but the insurance is voluntary and random discounts (of 0%, 50%, or 100%) on the premium are offered. My results suggest that, on average, (i) price risk does not significantly change production relative to price certainty and (ii) neither does the provision of compulsory insurance against price risk, but (iii) the introduction of voluntary insurance causes the average producer on the market to produce more in situations of price risk than in situations of price certainty. Additionally, I find that even in the absence of premium discounts, the insurance against price risk would have a large (i.e., 70-percent) take-up rate. In the third essay, I study how rainfall shocks (i.e., rainfall deviations from the mean) affect rural labor and child health, and I assess the role of mothers’ time use on child nutrition, among rural families in Uganda. My results show that less rainfall in the last month decreases mothers’ timeshare in housework, increases her share in other household-related activities (e.g., fetching water), and does not change her timeshare at the household farm. Less rainfall in the last month also decreases child nutrition. Nonetheless, using mediation analysis, I find that none of the mother’s time-use variables appears to be a mediating factor between rainfall variability and child nutrition. These results suggest that mothers adjust their time-use due to rainfall variability with the objective of preserving their children’s nutrition levels.