Got Milk? Motivation for Honesty and Functioning of Informal Markets: Evidence from India

• Departmental Seminar
• Development Workshop

Date and Location

Friday, September 25, 2015, 1:10 PM - 2:30 PM
ARE Library Conference Room, 4101 Social Sciences and Humanities

Abstract

Access to healthy diets is considered critical in combating malnutrition and the achievement of millennium development goals. Bovine milk is recognized as an important source of nutrition by FAO and for its potential to alleviate malnutrition. In many developing countries, milk is largely sourced from informal markets, which are prone to rampant cheating from adulteration of milk with water, leading to a loss of valuable nutrition. Evidence suggests that though many milkmen cheat, there is a large variation. We study in such markets whether individual milkmen vary in their motivation for honesty and whether this variation can explain why some milkmen cheat but others do not in actual informal milk markets. Our study takes place in India, where milk quality is very costly and difficult to detect. We use statistical methods to measure a milkman’s motivation for honesty in a laboratory experiment that involves rolling a die and reported its outcome. However, if some milkmen disguise their cheating behavior, statistical classification may misclassify these milkmen as honest, creating implications for the association between motivation for honesty and behavior in actual markets. Therefore, we complement our experiment with a Bluetooth die, which additionally allows us to obtain actual outcomes of the die roll. This innovation allows us to detect both dishonest as well as partially dishonest milkmen and thus obtains results at both extensive and intensive margins. We find that qualitatively both statistical or Bluetooth methods yield similar results: dishonest milkmen cheat their customers by adding substantially more water to milk than honest milkmen. However, the statistical method is unable to detect a large share of cheaters who disguise their behavior. We show that these partially dishonest milkmen cheat substantially more than honest milkmen, but statistically dishonest milkmen cheat even more than partially dishonest milkmen. As a result, statistical estimates are biased downwards by up to 20 percent. Our field setting, identification strategy, and additional experiments allow us to rule out prominent alternative explanations, such as differences in local institutions and relational contracts. Our findings can be interpreted as pointing out the importance of motivation for honesty in informal markets that are crucial for meeting human health and nutrition in developing countries.