Competition in Developing Countries
Friday, Jan. 6, 2017 3:15 PM – 5:15 PM
- Chair: Rema Hanna, Harvard University
Measuring and Addressing Market Power Issues in the Desludging Market in Senegal
AbstractWe test for collusion in the market for sanitation services in Dakar, Senegal by instituting real-time auctions and testing the impact of randomized differences in invitations to bid and other design elements. We ran over 3500 procurement auctions with 112 desludging operators. We find that the auctions are successful in instigating increased competition: the auctions reduce prices by 7%, and auction prices go down over time, although at a decreasing rate. Consistent with predictions of competitive auctions differences in input costs are passed through to the consumer and more invitations to the auctions result in lower prices.<br />
Participation in the auctions remains low at 18%. There is suggestive evidence that members of the largest and most organized garage collude through reducing participation when other members of the garage are invited to the same auction, but this effect is only present in the first year of the auction data. Operators who are part of the association of desludgers have lower costs and higher markups in the auctions. Members of the association pass through less of their transportation costs to the consumer and submit higher bids controlling for the garages to which they belong.
T.B.A. (To Be Arbitraged)? Extensive and Intensive Margin in Rice Trading in Madagascar
AbstractSpatial arbitrage is the driving force for the law of one price, and traders are economic agents who are specializing in arbitrage. Using trader-level biweekly survey in Antananarivo, the capital city of Madagascar, we investigate the extensive margin and the intensive margin in response to the price differences across districts. We first show that only a fraction of traders visited other districts to purchase rice, and these active traders tend to be larger than passive traders who do not other districts but only purchase rice from rice sellers visiting the city from their local districts, suggesting the importance of heterogeneity in traders. Active traders on average did not respond to the price differences both in terms of extensive and intensive margin, though larger active traders were more responsive to the price differences in extensive margin. Passive traders were more responsive to the price differences. Our randomized experiment of providing price information through SMS suggests the importance both of price information and the trade link costs in arbitrage activities. We then propose a framework to infer trade costs and trade linkage costs of active traders, finding the importance of trade linkage costs in explaining trader's location choice.
- O0 - General