Does Rigidity of Prices Hide Collusion?

Jiménez, Juan Luis; Perdiguero, Jordi

doi:10.1007/s11151-012-9337-9

Cited by 23 publications

(15 citation statements)

References 35 publications

(32 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Simple screens analyze strategic variables as prices and market shares to determine whether firms depart from comp etitive b ehavior. While there are many applications of simple screens to various regular markets (see Abrantes-Metz et al, 2006;Esposito and Ferrero, 2006;Harrington, 2008;Hueschelrath and Veith, 2011;Jimenez and Perdiguero, 2012;Abrantes-Metz et al, 2012;Froeb et al, 2014 , for some examples), applications to bid-rigging cases are rather rare (see Feinstein et al, 1985;Imhof et al, 2018;Imhof, 2017b , for exceptions).…”

Section: Screensmentioning

confidence: 99%

“…This literature differs from the majority of studies on detecting bid-rigging cartels that use econometric tests typically not only relying on bidding information, but also on proxies for the costs of the firms (see Porter and Zona, 1993;1999;Pesendorfer, 2000;Bajari and Ye, 2003;Jakobsson, 2007;Aryal and Gabrielli, 2013;Chotibhongs and Arditi, 2012a;2012b;Imhof, 2017a ). Finally, our paper is also related to studies on screens in markets not characterized by an auction process (see Abrantes-Metz et al, 2006;Esposito and Ferrero, 2006;Hueschelrath and Veith, 2011;Jimenez and Perdiguero, 2012;Abrantes-Metz et al, 2012 ).…”

Section: Introductionmentioning

confidence: 96%

“…They are thus of interest for competition agencies in order to detect cartels and to enforce competition laws. In the light of the vast variety of screens proposed in the literature (see Harrington, 2008;Jimenez and Perdiguero, 2012;OECD, 2013;Froeb et al, 2014 ), the question arises which detection metho d some comp etition agency should cho ose in practice. However, very few papers, if any, systematically investigate the performance of the screens based on statistical methods.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Machine learning with screens for detecting bid-rigging cartels

Huber

Imhof

2019

International Journal of Industrial Organization

View full text Add to dashboard Cite

We combine machine learning techniques with statistical screens computed from the distribution of bids in tenders within the Swiss construction sector to predict collusion through bid-rigging cartels. We assess the out of sample performance of this approach and find it to correctly classify more than 84% of the total of bidding processes as collusive or noncollusive. We also discuss tradeoffs in reducing false positive vs. false negative predictions and find that false negative predictions increase much faster in reducing false positive predictions. Finally, we discuss policy implications of our method for competition agencies aiming at detecting bid-rigging cartels.

show abstract

Section: Screensmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 96%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning with screens for detecting bid-rigging cartels

Huber

Imhof

2019

International Journal of Industrial Organization

View full text Add to dashboard Cite

show abstract

“…Our paper is related to an expanding literature using screening methods for detecting cartels (see Abrantes-Metz et al, 2006;Harrington, 2007;Bolotova et al, 2008;Abrantes-Metz et al, 2012;Jimenez and Perdiguero, 2012;OECD, 2014;Froeb et al, 2014). More specifically, it considers statistical screens for flagging bid-rigging cartels as Imhof et al (2018) and Imhof (2019).…”

Section: Introductionmentioning

confidence: 99%

Deep learning for detecting bid rigging: Flagging cartel participants based on convolutional neural networks

Huber¹,

Imhof²

2021

Preprint

View full text Add to dashboard Cite

Adding to the literature on the data-driven detection of bid-rigging cartels, we propose a novel approach based on deep learning (a subfield of artificial intelligence) that flags cartel participants based on their pairwise bidding interactions with other firms. More concisely, we combine a so-called convolutional neural network for image recognition with graphs that in a pairwise manner plot the normalized bid values of some reference firm against the normalized bids of any other firms participating in the same tenders as the reference firm. Based on Japanese and Swiss procurement data, we construct such graphs for both collusive and competitive episodes (i.e when a bid-rigging cartel is or is not active) and use a subset of graphs to train the neural network such that it learns distinguishing collusive from competitive bidding patterns. We use the remaining graphs to test the neural network's out-of-sample performance in correctly classifying collusive and competitive bidding interactions. We obtain a very decent average accuracy of around 90% or slightly higher when either applying the method within Japanese, Swiss, or mixed data (in which Swiss and Japanese graphs are pooled).When using data from one country for training to test the trained model's performance in the other country (i.e. transnationally), predictive performance decreases (likely due to institutional differences in procurement procedures across countries), but often remains satisfactorily high. All in all, the generally quite high accuracy of the convolutional neural network despite being trained in a rather small sample of a few 100 graphs points to a large potential of deep learning approaches for flagging and fighting bid-rigging cartels.

show abstract

“… 14. For papers that show the relation between price rigidity and collusion in the petrol market, see Abrantes-Metz et al (2006) and Jiménez and Perdiguero (2012). …”

mentioning

confidence: 99%

Fuel prices at petrol stations in touristic cities

2019

Self Cite

View full text Add to dashboard Cite

While it may seem obvious from common sense that retail fuel prices should be higher in touristic rather than in nontouristic cities, the related empirical literature has failed either to clearly support this assumption or to qualify and quantify the effect of tourism on retail fuel prices. Using a self-created data set that includes prices across a large sample of petrol stations located in mainland Spain and the Balearic Islands, we seek to evaluate the effects of tourism on local destination retail petrol prices. The estimations prove three main effects: Petrol prices are higher in cities with high levels of tourism, nearby beaches, and/or with the socioeconomic influence of a national park. In fact, petrol stations add a margin of around 6.6% and 6.4% for unleaded gasoline 95 and diesel, respectively, in touristic cities. Second, the more touristic the city, the higher these prices. Third, in the case of touristic municipalities, the percentage of travelers from abroad and being within the area of influence of a national park are two of the main drivers that explain this overpricing.

show abstract

Does Rigidity of Prices Hide Collusion?

Cited by 23 publications

References 35 publications

Machine learning with screens for detecting bid-rigging cartels

Machine learning with screens for detecting bid-rigging cartels

Deep learning for detecting bid rigging: Flagging cartel participants based on convolutional neural networks

Fuel prices at petrol stations in touristic cities

Contact Info

Product

Resources

About