The Causal Effect of Competition on Prices and Quality: Evidence from a Field Experiment



Many social programs that seek to improve the welfare of the poor focus on the demand side of the market, for example, addressing hunger by providing poor families with money or vouchers to buy food. If supply conditions are not competitive, however, businesses may be able to raise prices and capture program benefits intended for targeted households. Busso and Galiani study a conditional cash transfer program in the Dominican Republic where such forces seemed to be at work. This program, called Solidaridad, offers poor recipients a monthly benefit in the form of a debit card that can only be used for food purchases. In the beginning of the program, few grocers were allowed to accept the debit cards and those that did charged high prices, capturing some of the benefits intended for poor families by charging them more for their purchases. Busso and Galiani, working through the Inter-American Development Bank, conducted a field experiment in this environment to study whether increasing the amount of competition among grocer stores could help to solve this problem.

Busso and Galiani's experiment randomly varied the number of additional grocers allowed to accept the Solidaridad program beneficiaries across the 72 districts in which the program operated. Districts were allowed no additional grocers, one additional grocer, two additional grocers or three additional grocers, thus introducing random variation in the degree of competition across districts. Analyzing data on consumer purchases before and after the introduction of these program changes, Busso and Galiani find that adding one participating grocer lowered prices at existing stores by 6 percent. The increase in competition was not associated with any measurable change in the quality of the goods available. These results suggest that firms previously had been exploiting their market power to extract some of the Solidaridad program benefits for themselves and that stronger competition decreased the degree to which this occurred.

Busso and Galiani's results carry lessons for the design of social policies that seem likely to go beyond the specific program they studied. If individuals are given additional resources to help them in meeting important needs but have a limited choice of suppliers, those suppliers may well capture some of the benefits intended for the needy target population. While there are many approaches for dealing with this problem, encouraging entry to increase the competitiveness of the market is one obvious mechanism that can help.