The Causal Effect of Competition on Prices and Quality

How to cite

Busso, Matias; Galiani, Sebastian. 2014. "The Causal Effect of Competition on Prices and Quality." AEA RCT Registry. AEARCTR-0000343. https://doi.org/10.1257/rct.343-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Matias Busso* , Sebastian Galiani

* Primary investigator

Registry ID
AEARCTR-0000343
DOI
https://doi.org/10.1257/rct.343-1.0
First registered
Last updated
Status
Completed
Trial dates
2011-04-04 – 2011-12-15
Countries
Dominican Republic (Caribbean)
JEL codes
D4, L1, I3
Registry record
http://www.socialscienceregistry.org/trials/343

Abstract

This paper provides the first experimental evidence on the effect of increased competition on the prices and quality of goods. We rely on an intervention that randomized the entry of 61 retail firms (grocery stores) into 72 local markets in the context of a conditional cash transfer program that serves the poor in the Dominican Republic. Six months after the intervention, product prices in the treated districts had decreased by about 5%, while product quality and service quality had not changed. Using a theoretical model, we arrive at the conclusion that the poor segments of the population in these markets care the most about prices and much less about quality. Our results are also informative to the design of social policies. They suggest that policymakers should pay attention to supply conditions even when the policies in question will only affect the demand side of the market.

Intervention

The intervention involved bringing 61 new grocery stores (in 72 districts) into the a retail network of the Dominican Republic conditional cash transfer program. The experimental design allowed anywhere from zero to three new stores to start operating in each district.

Experimental design

The intervention consisted of an exogenous increase in the number of retailers associated with the network across districts. We use this randomized variability in market entry to evaluate the effect of an increase in competition on prices and quality. Each district was then assigned a random number in the set {0, 1, 2, 3} which defined the number of new entrant retailers that the executing agency committed to try to affiliate. Actual affiliation could, in principle, be different from the intended/randomized affiliation because of a shortage of eligible interested candidates that applied for entry into the network. Another source of non-compliance could be the failure of the executing agency to follow the intervention protocol.

Design details

Randomization method
Randomization done by computer
Randomization unit
Neighborhoods/districts
Clusters
72
Observations
440 retailers (firms) and 2160 households (consumers)
Arms
For each treatment the expected sample size was 110 retailers per treatment arm and 540 households per treatment arm
Primary outcomes
Price and quality of goods

Cited by

No citation data available for this registration yet.

Keywords