Detecting the Negative Information Effect on Food Demand: The Case of Rice Harvested before Fukushima Daiichi Nuclear Disaster

This study aims to identify the negative information effect of food demand. To identify the effects, media coverage about the Fukushima Daiichi nuclear disaster in March 2011 was employed as a natural experiment inducing variation in food demands responses to negative information. Observations covered the period between the Japanese rice harvest season (September, 2010) and the month of the disaster (March, 2011) in order to assess the food quality and producers’ responses to the disaster. Previous studies have found that negative information decreases food demand. However, these studies take the demand system approach (i.e. AIDS model) and, therefore, find it difficult to identify consumers’ responses from the effects of suppliers’ responses (i.e. decreased motivation for producing post-disaster). In this study, we focus on the effect on the demand of rice harvested before the nuclear disaster in food supermarkets at the Tokyo metropolitan area where there was little direct damage caused by the earthquake. Using POS data from the food supermarkets, we compared sales, prices, and quantities of rice harvested in the disaster-affected areas and in the unaffected areas before and after the disaster by employing the Difference-in-Differences estimation method. The results suggest that the sales of rice harvested in Fukushima prefecture and the other disaster affected area decreased after the disaster. In addition, we examined the effects on prices and quantities separately, and found that rice prices in those disaster-affected areas experienced a surprisingly modest decrease. On the other hand, the results find the corresponding decreases in the quantity. Key Words：The Great East Japan Earthquake, Indirect Damages, Food Demand JEL Classification: Q11, Q48, Q53, Q54