Detecting Large-Scale Collusion in Procurement Auctions

Kawai, Kei; Nakabayashi, Jun

doi:10.2139/ssrn.2467175

Cited by 40 publications

(34 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More granular data required for these tests for collusion can be collected once a key subset of firms is identified, at significantly lower cost. Other data-driven studies develop screens for particular exotic auction formats such as average-price auctions or multi-round auctions [32] -our approach can also complement these context-specific screens.…”

Section: Discussionmentioning

confidence: 99%

A network approach to cartel detection in public auction markets

Wachs

Kertész²

2019

Sci Rep

View full text Add to dashboard Cite

Competing firms can increase profits by setting prices collectively, imposing significant costs on consumers. Such groups of firms are known as cartels and because this behavior is illegal, their operations are secretive and difficult to detect. Cartels feel a significant internal obstacle: members feel short-run incentives to cheat. Here we present a network-based framework to detect potential cartels in bidding markets based on the idea that the chance a group of firms can overcome this obstacle and sustain cooperation depends on the patterns of its interactions. We create a network of firms based on their co-bidding behavior, detect interacting groups, and measure their cohesion and exclusivity, two group-level features of their collective behavior. Applied to a market for school milk, our method detects a known cartel and calculates that it has high cohesion and exclusivity. In a comprehensive set of nearly 150,000 public contracts awarded by the Republic of Georgia from 2011 to 2016, detected groups with high cohesion and exclusivity are significantly more likely to display traditional markers of cartel behavior. We replicate this relationship between group topology and the emergence of cooperation in a simulation model. Our method presents a scalable, unsupervised method to find groups of firms in bidding markets ideally positioned to form lasting cartels.

show abstract

Section: Discussionmentioning

confidence: 99%

A network approach to cartel detection in public auction markets

Wachs

Kertész²

2019

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Very few previous studies have used an RDD approach to exploit the quasi-experimental variation induced by an auction. Among the few, Kawai and Nakabayashi (2015) use this approach to detect collusion in consecutive procurement auctions, when bids at the first auction fail to meet a secret reserve price. The authors focus on failed auctions where the bids of the lowest and the second lowest bidders are very close to each other, and test whether they preserve their order at the second auction.…”

Section: Fx Marketsmentioning

confidence: 99%

An Auction-Based Test of Private Information in an Interdealer FX Market

Bonaldi¹,

Villamizar‐Villegas²

2018

View full text Add to dashboard Cite

There are several financial markets where dealers trade a large share of total volume, while also having access to periodic auctions of the same asset conducted by a third party. For such a market, we derive a test of private information about the value of the asset that combines data on both bidding behavior and market trades. Our approach is to test for private versus common values, as defined in auction theory. We use changes in trading prices of extreme bidders before and after the auction to test the null hypothesis of private values (no private information) against the alternative of common values (private information). Additionally, we use a regression discontinuity design where we compare the behavior of dealers bidding right below and right above the auction's cutoff price to control for inventory effects, understood here as decreasing marginal valuations as functions of inventory. Our case study are foreign exchange auctions conducted by the Central Bank of Colombia during the period 2008-2014, and the corresponding interdealer market for the Colombian peso against the US dollar. Overall, the data does not reject the null hypothesis of private values. Specifically, information about other bidders' valuations has no significant effect on trading prices, not even shortly after the auction takes place.

show abstract

“…The first is that although all auctions include a reserve price, these reserve prices are not set to extract greater surplus for the city along the lines of Myerson (1981) or Riley and Samuelson (1981). Rather, consistent with recorded practice, reserve prices are engineering estimates (Ohashi, 2009, Tanno and Hirai, 2012, Kawai and Nakabayashi, 2014 This normalization lets us take the comparative statics of Propositions 3, 4 and 5 to the data, even though there is heterogeneity in minimum prices. As a robustness test, we also study the distribution of log-winning-bids using reserve prices as a control variable (see Table OA.7).…”

Section: Data and Empirical Strategymentioning

confidence: 94%

“…Porter (1983), along with Ellison (1994) (see also Ishii, 2008) use patterns of price wars of the sort predicted by repeated game models of oligopoly behavior (Green andPorter, 1984, Rotemberg andSaloner, 1986) to identify collusion. In a multi-stage auction context, Kawai and Nakabayashi (2014) argue that excess switching of second and third bidder across bidding rounds, compared to first and second bidders, is a smoking gun for collusion. We propose a test of collusion exploiting changes in the cartel's ability to implement effective punishments.…”

Section: An Important Observation Frommentioning

confidence: 99%

Collusion in Auctions with Constrained Bids: Theory and Evidence from Public Procurement

Chassang¹,

Ortner²

2019

Journal of Political Economy

View full text Add to dashboard Cite

We study the mechanics of cartel enforcement and its interaction with bidding constraints in the context of repeated procurement auctions. Under collusion, bidding constraints weaken cartels by limiting the scope for punishment. This yields a test of collusive behavior exploiting the counter-intuitive prediction that introducing minimum prices can lower the winning bid distribution. The model's predictions are borne out in Japanese procurement data, where we find evidence that minimum prices weakened collusion. A robust design insight is that setting a minimum price at the bottom of the observed winning bid distribution necessarily improves over a minimum price of zero.

show abstract

Detecting Large-Scale Collusion in Procurement Auctions

Cited by 40 publications

References 16 publications

A network approach to cartel detection in public auction markets

A network approach to cartel detection in public auction markets

An Auction-Based Test of Private Information in an Interdealer FX Market

Collusion in Auctions with Constrained Bids: Theory and Evidence from Public Procurement

Contact Info

Product

Resources

About