A pair of economists suggest a new way to alleviate the problem of collusion in public procurement auctions: establish price floors for the contracted work.
Collusion between firms participating in public procurement costs taxpayers around the globe billions of dollars each year.
The Organization for Economic Cooperation and Development (OECD) estimates that its member countries spend 12 percent of their GDP in public procurement, with a higher percentage in developing countries. The elimination of bid rigging, the OECD estimates, could help reduce procurement prices by 20 percent or more.
In a newly released analysis, a pair of economists suggest a new way to alleviate the problem of collusion: establish price floors for the contracted work—i.e., a minimum price below which bids are disqualified.
“This is somewhat counterintuitive since price floors reduce competitive price pressure,” says Sylvain Chassang, a professor in New York University’s Department of Economics and a co-author of the paper, which appears in the Journal of Political Economy. “But that’s exactly the point. Cartels enforce high prices and defend against entry by using the threat of price wars. Price floors make this threat less effective, which can weaken a cartel’s ability to fix prices in taxpayer-funded projects.”
“As a result, well-chosen price floors may, in fact, lower procurement prices paid by governments,” adds co-author Juan Ortner, an assistant professor of economics at Boston University. “In addition to price ceilings, or price caps, governments may want to consider committing to a price floor.”
In their work, Chassang and Ortner examined data from more than 10,000 public procurement auctions in 14 Japanese cities between 2007 and 2016. The projects auctioned consisted of small-scale construction jobs, such as repaving roads and refurbishing schools.
Over the studied time period, six of these 14 cities went from having no price floors to adopting minimum-price policies.
The remaining eight cities in the sample did not establish price floors, thus serving as a control group against which to examine the impact of price floors.
The findings are best illustrated by the below figure, which also may be downloaded via DropBox.
In October 2009, one city (Tsuchiura) introduced price floors and, contemporaneously, the distribution of winning bids shifted down. The change was significant: roughly a third of the bids dropped by approximately 8 percent. By contrast, control cities, which did not implement price floors, experienced no change in the distribution of winning bids.
This suggests that the impact of price floors on auctions is indeed causal, the researchers conclude. The authors, however, are careful to highlight that setting a poorly chosen price floor may be counter-productive.
“The devil is in the details,” Chassang and Ortner write. “A specificity of the Japanese auctions we study is that price floors are in a Goldilocks region: high enough, but not too high. They correspond to plausible, but still fairly low, bids. You don’t want to push prices up by setting very high minimum prices, but they need to be high enough so that they limit a cartel’s ability to run a price war.”