According to the WHO, in less developed countries approximately 2.2 million people—most of whom are children—die annually of food and waterborne diseases. In these economies, information on the safety attributes of food is usually not available and enforcement of food safety regulations is often weak, particularly within markets for locally consumed food. Still, food safety in the developing world has long been considered a secondary concern relative to food availability. The goal of this dissertation is to contribute to a deeper understanding of some of the constraints that surround demand for safer food along food supply chains in developing countries. Consumers’ demand for safe food can be thought of as an investment in preventive health, which has been shown to be extremely low in developing countries. Hence, this dissertation contributes to the economics literature that explores the impact of health-related information on preventive health behaviors in poor countries.

This dissertation focuses on the role of food safety information in affecting people’s purchase behavior in a developing country setting. Because food safety is mostly a credence attribute that cannot be ascertained—or is too costly to ascertain—even after consumption, the provision of information has an important role to play in the reduction of information asymmetries inside the food chains. Among the several actors that are involved in food value chains, this dissertation focuses on small-scale informal intermediaries and consumers. The effect of information on these actors’ demand for safer food is assessed through the estimation of willingness to pay for food labeled as having safer characteristics, and through the analysis of the effect of different types of health-related information on the decision of whether or not to purchase food advertised as safer to eat. To achieve this, two field experiments using revealed preference methods were conducted in Kenya, where maize, the staple food, is frequently contaminated with aflatoxin, a naturally-occurring fungal toxin that is harmful to human health.

A brief introductory chapter is followed by a comparison of the advancement of food safety policy and research in developed countries with the corresponding evolution in developing countries (Chapter 2).

The framed field experiment described in Chapter 3 tests whether maize traders in informal markets are willing to pay more for higher quality and safer maize. 369 traders from different markets across Kenya participated in a second-price sealed-bid auction in which information on moisture content and aflatoxin contamination of maize auctioned was varied experimentally using labels. The results show that information on moisture content significantly affects traders’ willingness to pay and suggest the observability of moisture content is limited. Also, the effect of information does not appear to be driven by the possibility of selling drier maize to the formal sector, nor by the intention to keep the dryer maize for own family consumption. Further, the impact of providing traders with information on aflatoxin contamination is over twice as large as the effect of moisture content information. These results show that there is potential for strengthening the price-quality relationship within this context by increasing the availability of information on maize quality and safety.

Chapter 4 presents the results of a field experiment conducted among customers of small retail shops in Nairobi and smaller urban centers in eastern Kenya. Packages of maize flour were tested for aflatoxin, labeled as safe to eat when they complied with the aflatoxin regulation, and offered for sale at a 20% premium above the price of untested maize. Information messages about the health consequences of aflatoxin exposure and about local contamination prevalence were randomly varied across customers as they entered the shops. The results show that the impact of health messaging on purchase of tested maize varies significantly depending both on the specific content of the message, and on the characteristics and prior beliefs of consumers. Information on the local prevalence of aflatoxin contamination, which exceeded the vast majority of customers’ contamination priors, had the strongest impact on demand. This study demonstrates that combining information on the prevalence of a risk with its health consequences is an effective approach for encouraging preventive health behavior.