Risk, shocks, and markets: Theory and evidence from agricultural systems in low and middle income countries

This thesis explores how markets influence economic actors’ ex ante and ex post risk management strategies in Tanzania, Peru, and Ethiopia. It makes both theoretical and empirical contributions to various strands of literature with analysis across different levels of economic actors – individuals within households, households, and markets. The thesis highlights the importance of understanding actors’ heterogeneity in risk management practices and the need for a wide range of methodologies to analyze various actors and contexts. Chapter 1 provides a brief discussion on previous economic inquiry pertaining to risk management, places this thesis within the context of previous research, presents an overview of the methodologies used, and highlights the main research questions in this thesis.

Chapter 2 presents a theoretical household model for nutrition sensitive agriculture under risk. The model describes how production diversity and market access are related to dietary diversity, and empirical evidence is obtained using cross-sectional nationally representative data from Tanzania. The theoretical predictions and empirical results suggest that market access mediates the influence of production diversity on dietary diversity. These results help explain why the empirical literature has found that increased on-farm production diversity does not always translate into increased dietary diversity. From a policy standpoint, Chapter 2 implies that nutrition-sensitive agriculture interventions should target participants based on their level of market access. 

Chapter 3 examines whether intra-household decision-making regimes and labor allocations affect household investment strategies and labor productivity among spouses. Using a randomized lab-in-the-field experiment, Chapter 3 tests for differences in labor-intensive investment decision-making between men alone, women alone, and both spouses together as well as labor productivity differences between men and women. The results show that when spouses make decisions together, they can overcome some inefficiencies observed from solo decision-makers. Further, the results suggests that men put forth less effort than women in the experiment’s real-effort task, which hinders the chances of successful investment. Chapter 3 argues that future agricultural programs should take intra-household gender dynamics into account and include gender sensitization training in their implementation strategies.

Chapter 4 analyzes how marketing decisions change when households are affected by production shocks. It empirically and theoretically studies the relationship between Coffee Leaf Rust (CLR) incidence on Peruvian specialty coffee cooperative members’ farms and their propensity to sell coffee outside the cooperative. The empirical analysis uses panel data methods and shows that CLR is associated with higher rates of marketing to non-cooperative channels. Households’ risk tolerance and extent of non-coffee income amplify this relationship.  These findings underscore the need for heterogeneous institutional responses to production shocks – even within cooperatives in a small area, household responses to shocks can vary widely.

Chapter 5 studies the effect of market prices on internal conflict from a mesoeconomic perspective. The chapter tests whether changes in cereal prices in domestic retail markets across Ethiopia increase or decrease the incidence of nearby violence across different types of violent events. Using an instrumental variables empirical specification that takes advantage of exogenous international prices, the results show that increases in cereal prices raise the incidence of nearby protests and riots, conflict with local aims, and small-scale violent events. However, conflict with national aims and large-scale battles are not affected by changes in cereal prices, and there is significant heterogeneity in the results across cereal producing and non-cereal producing areas. These results suggest that different theoretical mechanisms are at play simultaneously, and food-pricing policies should take into account heterogeneity of domestic markets to promote internal stability.

Chapter 6 discusses the chapters and presents a synthesis of crosscutting themes throughout the thesis. First, there is need for economic research to identify and assess heterogeneity in economic actors’ responses to risk for future policies to maximize welfare and ensure marginalized actors are not left behind. Further, such policies should be based on research that takes into account theoretical mechanisms so that policy designers can assess whether previous empirical findings are likely to hold in the context at hand. Finally, economic researchers should be careful in the assumptions they make concerning markets in low-income countries. In many settings, assumptions on well-functioning markets, farmers’ market access, and price integration are untested and unfit, leading to research with faulty outcomes and misguided policy recommendations.