pith. sign in

arxiv: 2606.30470 · v1 · pith:YHD4VKCPnew · submitted 2026-06-29 · 💰 econ.GN · q-fin.EC

Swimming in Dark Water: When Cartels Mimic Competition

David Imhof , Thierry Madi\`es , Martin Huber This is my paper

Pith reviewed 2026-06-30 03:08 UTC · model grok-4.3

classification 💰 econ.GN q-fin.EC
keywords bid-riggingcartelsoverchargesmachine learningprocurementcollusion detectionroad construction
0
0 comments X

The pith

A Swiss road cartel allocated contracts by observable costs to mimic competition and evade detection while overcharging at least 45%.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper examines a bid-rigging cartel active in Ticino road construction from 1999 to 2005. Using detailed internal records, the authors reconstruct how members coordinated bids and allocated contracts through a formal convention without side payments. They implemented a cost-based allocation mechanism that closely matched the first-best collusive outcome. Regression and machine-learning analyses show that observable cost proxies predict winning bids and rankings, indicating that members deliberately mimicked competitive bidding to avoid standard detection. Double machine learning estimates average overcharges of at least 45 percent, and potentially higher.

Core claim

The cartel implemented a cost-based allocation mechanism that closely approximated the first-best collusive outcome. Observable cost proxies systematically predict both winning bids and bid rankings, suggesting members strategically mimicked competitive bidding behavior to evade econometric detection. Double machine learning estimates average overcharges of at least 45%, and potentially substantially higher.

What carries the argument

the 'convention' agreement that coordinated bids and allocated contracts using observable cost proxies without side payments

Load-bearing premise

The documentary evidence fully and accurately captures the cartel's internal coordination, and the cost proxies are exogenous and sufficient to identify the allocation mechanism without bias.

What would settle it

Observing that cost proxies fail to predict bid winners and rankings in a comparable non-cartel procurement market, or finding that overcharge estimates fall substantially below 45% with different cost measures or specifications.

read the original abstract

This paper analyzes the internal organization and economic effects of a bid-rigging cartel in the road construction sector of the Swiss canton of Ticino, active from 1999 to 2005. Using exceptionally rich documentary evidence, we reconstruct how cartel members coordinated bids and allocated contracts under a formal agreement known as the 'convention'. We show that, despite the absence of side payments, the cartel implemented a cost-based allocation mechanism that closely approximated the first-best collusive outcome. Regression and machine-learning analyses indicate that observable cost proxies systematically predict both winning bids and bid rankings. The evidence further suggests that cartel members strategically mimicked competitive bidding behavior, allowing them to evade standard econometric detection methods. Using double machine learning, we estimate average overcharges of at least 45\%, and potentially substantially higher, highlighting the significant financial harm caused by this sophisticated form of collusion.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. This paper analyzes a bid-rigging cartel in the Swiss canton of Ticino's road construction sector (1999-2005). Using rich documentary evidence, it reconstructs the cartel's formal 'convention' for coordinating bids and allocating contracts via a cost-based mechanism without side payments that approximates the first-best collusive outcome. Regression and machine-learning analyses show that observable cost proxies systematically predict winning bids and bid rankings. The paper argues that cartel members strategically mimicked competitive bidding to evade standard econometric detection. Double machine learning yields average overcharge estimates of at least 45% (potentially higher).

Significance. If the identification holds, the paper provides concrete evidence of sophisticated collusion that achieves efficient allocation while avoiding detection, with large welfare costs. This advances understanding of cartel internal organization and the limits of existing detection methods in procurement markets, with direct implications for antitrust enforcement and econometric practice.

major comments (2)
  1. [Regression and machine-learning analyses; double machine learning estimation] The central claim that cartel members mimicked competition (and the resulting 45%+ overcharge estimate) rests on the regression/ML finding that cost proxies predict bids and rankings. This interpretation requires that the proxies are exogenous to the bidding process and that the documentary record fully captures the allocation rule. If proxies correlate with unobserved cost shifters or if documents omit side arrangements, the observed correlations cannot distinguish mimicking from ordinary competition, leaving the DML overcharge unidentified. A formal discussion of exogeneity threats and robustness to alternative proxy constructions is needed.
  2. [Reconstruction of the convention] The reconstruction of the cost-based allocation mechanism as approximating the first-best collusive outcome without side payments is load-bearing for the claim of sophisticated internal organization. The paper should clarify how the documentary evidence rules out selective reporting or unrecorded transfers that could alter the efficiency assessment.
minor comments (2)
  1. [Empirical analysis] Clarify the exact set of cost proxies used in the regressions and ML models, including any construction details or data sources.
  2. [Double machine learning estimation] Provide more detail on the double machine learning implementation, including the choice of nuisance estimators and cross-fitting procedure.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major comment below, indicating where we will revise the manuscript to strengthen the analysis.

read point-by-point responses

  1. Referee: [Regression and machine-learning analyses; double machine learning estimation] The central claim that cartel members mimicked competition (and the resulting 45%+ overcharge estimate) rests on the regression/ML finding that cost proxies predict bids and rankings. This interpretation requires that the proxies are exogenous to the bidding process and that the documentary record fully captures the allocation rule. If proxies correlate with unobserved cost shifters or if documents omit side arrangements, the observed correlations cannot distinguish mimicking from ordinary competition, leaving the DML overcharge unidentified. A formal discussion of exogeneity threats and robustness to alternative proxy constructions is needed.

    Authors: We agree that a formal discussion of exogeneity is warranted to support the interpretation. The proxies (project size, location, and material needs) are pre-determined and observable before bidding, and the convention documents show they were used explicitly for allocation. In the revision we will add a dedicated subsection on exogeneity threats, including potential unobserved shifters, and report robustness checks with alternative proxy constructions (e.g., different variable subsets and functional forms). This will clarify the basis for the DML overcharge estimates. revision: yes

  2. Referee: [Reconstruction of the convention] The reconstruction of the cost-based allocation mechanism as approximating the first-best collusive outcome without side payments is load-bearing for the claim of sophisticated internal organization. The paper should clarify how the documentary evidence rules out selective reporting or unrecorded transfers that could alter the efficiency assessment.

    Authors: The evidence consists of the complete set of seized cartel records, including the full convention text, meeting minutes, and allocation logs for all contracts in the period. These show no side payments or deviations. We will add a clarifying paragraph explaining the comprehensiveness of the seized materials, their cross-verification against public bid data, and why selective reporting or unrecorded transfers are inconsistent with the documented patterns and the absence of any such arrangements in the records. revision: yes

Circularity Check

0 steps flagged

No significant circularity; estimates rest on external data and DML

full rationale

The paper reconstructs cartel behavior from documentary evidence and applies regression, ML, and double machine learning to observable cost proxies and bids. No step reduces a claimed prediction or overcharge estimate to a fitted parameter by construction, nor relies on self-citation chains or imported uniqueness theorems. The central results are identified from data under stated exogeneity assumptions rather than being tautological with inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; the analysis is empirical based on documentary evidence and statistical methods.

pith-pipeline@v0.9.1-grok · 5676 in / 1053 out tokens · 59166 ms · 2026-06-30T03:08:01.465346+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

134 extracted references · 2 canonical work pages

  1. [1]

    Journal of Public Procurement , year=

    Cartel detection in public procurement--Evaluation of five econometric methods , author=. Journal of Public Procurement , year=

  2. [2]

    Chernozhukov and C

    V. Chernozhukov and C. Hansen and M. Spindler , Date-Added =

  3. [3]

    Fighting bid rigging in public procurement: Report on implementing the OECD Recommendation , Year =

    OECD , Date-Added =. Fighting bid rigging in public procurement: Report on implementing the OECD Recommendation , Year =

  4. [4]

    Skrzypacz and H

    A. Skrzypacz and H. Hopenhayn , Date-Added =. Tacit Collusion in Repeated Auctions , Year =. Journal of Economic Behavior and Organization , Number =

  5. [5]

    Optimal Collusion with Private Information , Year =

    Susan Athey and Kyle Bagwell , Date-Added =. Optimal Collusion with Private Information , Year =. RAND Journal of Economics , Number =

  6. [6]

    Aoyagi , Date-Added =

    M. Aoyagi , Date-Added =. Efficient Collusion in Repeated Auctions with Communication , Year =. Journal of Economic Theory , Number =

  7. [7]

    Hendricks and R

    K. Hendricks and R. P. McAfee and M. Williams , Booktitle =. Auctions and Bid Rigging , Volume =

  8. [8]

    Tibshirani , Date-Added =

    R. Tibshirani , Date-Added =. Regression Shrinkage and Selection via The Lasso , Volume =. Journal of the Royal Statistical Society, Series B , Number =

  9. [9]

    van der Laan and E

    M. van der Laan and E. C. Polley and A. E. Hubbard , Date-Added =. Super Learner , Volume =. Statistical Applications of Genetics and Molecular Biology , Keywords =

  10. [10]

    B. D. Ripley , Date-Added =. Pattern Recognition and Neural Networks , Year =

  11. [11]

    A Logical Calculus of Ideas Immanent in Nervous Activity , Volume =

    Warren McCulloch and Walter Pitts , Date-Added =. A Logical Calculus of Ideas Immanent in Nervous Activity , Volume =. Bulletin of Mathematical Biophysics , Number =

  12. [12]

    Spindler and V

    M. Spindler and V. Chernozhukov and C. Hansen , Date-Added =. R: Highdimensional metrics , Url =. 2016 , Bdsk-Url-1 =

    2016

  13. [13]

    Random Decision Forests , Year =

    Tin Kam Ho , Date-Added =. Random Decision Forests , Year =. Proceedings of the 3rd International Conference on Document Analysis and Recognition Montreal , Pages =

  14. [14]

    Spindler, V

    M. Spindler, V. Chernozhukov, C. Hansen , Date-Added =. R: Highdimensional metrics , Url =

  15. [15]

    Random Forests , Volume =

    Leo Breiman , Date-Added =. Random Forests , Volume =. Machine Learning , Number =

  16. [16]

    Bagging predictors , Volume =

    Leo Breiman , Date-Added =. Bagging predictors , Volume =. Machine Learning , Number =

  17. [17]

    Working Papers SES 484, Faculty of Economics and Social Sciences, University of Fribourg (Switzerland) , Title =

    David Imhof , Date-Added =. Working Papers SES 484, Faculty of Economics and Social Sciences, University of Fribourg (Switzerland) , Title =

  18. [18]

    Working Papers SES 483, Faculty of Economics and Social Sciences, University of Fribourg (Switzerland) , Title =

    David Imhof , Date-Added =. Working Papers SES 483, Faculty of Economics and Social Sciences, University of Fribourg (Switzerland) , Title =

  19. [19]

    J. E. Harrington , Booktitle =. Behavioral Screening and the Detection of Cartels , Year =

  20. [20]

    Antitrust Chronical , Title =

    Carlos Emmanuel Joppert Ragazzo , Date-Added =. Antitrust Chronical , Title =

  21. [21]

    Government at a Glance , Year =

    OECD , Date-Added =. Government at a Glance , Year =

  22. [22]

    Antitrust Chronical , Title =

    Carlos Mena-Labarthe , Date-Added =. Antitrust Chronical , Title =

  23. [23]

    Competition Policy International , Title =

    Ernesto Estrada and Samuel Vasquez , Date-Added =. Competition Policy International , Title =

  24. [24]

    Froeb and David S

    Luke M. Froeb and David S. Sibley and Michael J. Doane and Brijesh P. Pinto , Date-Added =. The Oxford Handbook of International Antitrust Economics , Title =

  25. [25]

    Abrantes-Metz and Luke M

    Rosa M. Abrantes-Metz and Luke M. Froeb and John F. Geweke and Christopher T. Taylor , Journal =. A Variance Screen for Collusion , Volume =

  26. [26]

    Abrantes-Metz and Michael Kraten and Albert D

    Rosa M. Abrantes-Metz and Michael Kraten and Albert D. Metz and Gim Seow , Journal =. Libor Manipulation , Volume =

  27. [27]

    Aryal and M

    G. Aryal and M. F. Gabrielli , Journal =. Testing for collusion in asymmetric first-price auctions , Volume =

  28. [28]

    A Study of the Internal Organization of a Bidding Cartel , Volume =

    John Asker , Journal =. A Study of the Internal Organization of a Bidding Cartel , Volume =

  29. [29]

    Collusion and Price Rigid , Volume =

    Susan Athey and Kyle Bagwell and Chris Sanchirico , Journal =. Collusion and Price Rigid , Volume =

  30. [30]

    Bagwell and R

    K. Bagwell and R. W. Staiger , Journal =. Collusion over the Business Cycle , Volume =

  31. [31]

    Barriers to New Competition , Year =

    Bain , Editor =. Barriers to New Competition , Year =

  32. [32]

    Bajari and L

    P. Bajari and L. Ye , Journal =. Deciding between Competition and Collusion , Volume =

  33. [33]

    Baldwin and Robert C

    Laura H. Baldwin and Robert C. Marshall and Jean-Francois Richard , Journal =. Bidder Collusion at Forest Service Timber Sales , Volume =

  34. [34]

    Banerji and J

    A. Banerji and J. V. Meenakshi , Journal =. Buyer collusion and efficiency of governement intervention in wheat markets in northern India: An asymmetric structural auctions analysis. , Volume =

  35. [35]

    G. S. Becker , Journal =. Crime and Punishment: An Economic Approach , Volume =

  36. [36]

    Benoit and V

    J.-P. Benoit and V. Krishna , Journal =. Dynamic duopoly: Prices and quantities , Volume =

  37. [37]

    B. D. Bernheim and M. D. Whinston , Journal =. MultiCompet Contact and Collusive Behavior , Volume =

  38. [38]

    How Can a Cartel Be Detected? , Volume =

    Korbinian von Blanckenburg and Alexander Geist , Journal =. How Can a Cartel Be Detected? , Volume =

  39. [39]

    Connor and Douglas J

    Yuliya Bolotova and John M. Connor and Douglas J. Miller , Journal =. The Impact of Collusion on Price Behavior: Empirical Results from Two Recent Cases , Volume =

  40. [40]

    W. A. Brock and J.A. Scheinkman , Journal =. Price Setting Supergames with Capacity Constraints , Volume =

  41. [41]

    T. N. Cason and S. P. Lau and V. Mui , Journal =. Learning, teaching, and turn taking in the repeated assignment game , Volume =

  42. [42]

    Chotibhongs and D

    R. Chotibhongs and D. Arditi , Journal =. Analysis of collusive bidding behavior , Volume =

  43. [43]

    Chotibhongs and D

    R. Chotibhongs and D. Arditi , Journal =. Detection of Collusive Behavior , Volume =

  44. [44]

    Secret Price Cutting and Strategic BBuyer: An Illustration of the Divide and Conquer Tactic , Volume =

    Compte , Journal =. Secret Price Cutting and Strategic BBuyer: An Illustration of the Divide and Conquer Tactic , Volume =

  45. [45]

    Compte and F

    O. Compte and F. Jenny and P. Rey , Journal =. Capacity Constraints, Mergers and Collusion , Volume =

  46. [46]

    J. M. Connor and R. H. Lande , Date-Modified =. How high do cartels raise price? Implications for reform of the antitrust sentencing guidelines. , Volume =. Tulane Law Review, Amercian Antitrust Institute Working Paper, N. 01-04 , Owner =

  47. [47]

    Davidson and R

    C. Davidson and R. Deneckere , Journal =. Excess Capacity and Collusion , Volume =

  48. [48]

    Conglomerate Bigness as a Source of Power , Year =

    Edwards , Journal =. Conglomerate Bigness as a Source of Power , Year =

  49. [49]

    Variance Screens for Detecting Collusion: An Application to Two Cartel Cases in Italy , Year =

    Esposito and Ferrero , Journal =. Variance Screens for Detecting Collusion: An Application to Two Cartel Cases in Italy , Year =

  50. [50]

    Feinstein and Michael K

    Jonathan S. Feinstein and Michael K. Block and Frederick C. Nold , Journal =. Asymmetric Information and Collusive Behavior in Auction Markets , Volume =

  51. [51]

    Asymmetry in Procurement Auctions: Evidence from Snow Removal Contracts , Volume =

    Veronique Flambard and Isabelle Perrigne , Journal =. Asymmetry in Procurement Auctions: Evidence from Snow Removal Contracts , Volume =

  52. [52]

    A. G. Fraas and D. F. Greer , Journal =. Market Structure and Price Collusion: An Empirical Analysis , Volume =

  53. [53]

    Froeb and Robert A

    Luke M. Froeb and Robert A. Koyak and Gregory J. Werden , Journal =. What is the Effect of Bid-Rigging on Prices? , Volume =

  54. [54]

    Gilo and Y

    D. Gilo and Y. Moshe and Y. Spiegel , Journal =. Partial Cross Ownership and Tacit Collusion , Volume =

  55. [55]

    E. J. Green and R. H. Porter , Journal =. Non Cooperative Collusion under Imperfect Price Information , Volume =

  56. [56]

    Green , Owner =

    William H. Green , Owner =. Econometric Analysis , Year =

  57. [57]

    Predicting Cartels , Year =

    Paul Grout and Silvia Sonderegger , Journal =. Predicting Cartels , Year =

  58. [58]

    Optimal Nonparametric Estimation of First-Price Auctions , Volume =

    Emmanuel Guerre and Isabelle Perrigne and Quang Vuong , Journal =. Optimal Nonparametric Estimation of First-Price Auctions , Volume =

  59. [59]

    Haltiwanger and J

    J. Haltiwanger and J. E. Jr. Harrington , Journal =. The Impact of Cyclical Demand Movements on Collusive Behavior , Volume =

  60. [60]

    Harrington , Chapter =

    Joseph E. Harrington , Chapter =. Handbook of Antitrust , Year =

  61. [61]

    Harrington and Joe Chen , Journal =

    Joseph E. Harrington and Joe Chen , Journal =. Cartel Pricing Dynamics with Cost Variability and Endogenous Buyer Detection , Volume =

  62. [62]

    G. A. Hay and D. Kelley , Journal =. An Empirical Survey of Price Fixing Conspiracies , Volume =

  63. [63]

    Cartel Detection In Procurement Markets , Volume =

    Hueschelrath and Veith , Journal =. Cartel Detection In Procurement Markets , Volume =

  64. [64]

    Screening for Bid Rigging, does it work? , Year =

    David Imhof and Yavuz Karagoek and Samul Rutz , Date-Modified =. Screening for Bid Rigging, does it work? , Year =. Working Paper UFC , Owner =

  65. [65]

    Collusion in Repeated Procurement Auction: a Study of Paving Market in Japan , Volume =

    Rieko Ishii , Date-Modified =. Collusion in Repeated Procurement Auction: a Study of Paving Market in Japan , Volume =. International Journal of Industrial Organization , Owner =

  66. [66]

    Final Report for DG Competition , Year =

    Ivaldi and Jullien and Rey and Seabright and Tirole , Chapter =. Final Report for DG Competition , Year =

  67. [67]

    Bid Rigging in Swedish Procurement Auctions , Year =

    Maria Jakobsson , Journal =. Bid Rigging in Swedish Procurement Auctions , Year =

  68. [68]

    Does Rigidity of PPrice Hide Collusion? , Volume =

    Juan Luis Jimenez and Jordi Perdiguero , Journal =. Does Rigidity of PPrice Hide Collusion? , Volume =

  69. [69]

    Johnson and S

    N. Johnson and S. Kotz and A. Balakrishnan , Owner =. Distributions in Statistics, Continuous Univariate Distributions , Year =

  70. [70]

    Rosenthal , Journal =

    Kevin Lang and Robert W. Rosenthal , Journal =. The Contractors' Game , Volume =

  71. [71]

    Continuity of the First Price Auction Nash Equilibrium Correspondence , Volume =

    Bernard Lebrun , Journal =. Continuity of the First Price Auction Nash Equilibrium Correspondence , Volume =

  72. [72]

    Existence of an Equilibrium in First Price Auctions , Volume =

    Bernard Lebrun , Journal =. Existence of an Equilibrium in First Price Auctions , Volume =

  73. [73]

    The Nature and Impact of Hardcore Cartels

    LondonEconomic , Date-Modified =. The Nature and Impact of Hardcore Cartels. A Report to the Danish Competition Authority , Year =. London Economic , Owner =

  74. [74]

    Screening markets for cartel detection: collusive markers in the CFD cartel-audit , Volume =

    Christian Lorenz , Journal =. Screening markets for cartel detection: collusive markers in the CFD cartel-audit , Volume =

  75. [75]

    G. J. Mailath and L. Samuelson , Editor =. Repeated GGame and Reputations: Long-Run Relationships. , Year =

  76. [76]

    Collusive Behavior and Partial Ownership of Rivals , Volume =

    Malueg , Journal =. Collusive Behavior and Partial Ownership of Rivals , Volume =

  77. [77]

    Marshall and Leslie M

    Robert C. Marshall and Leslie M. Marx , Journal =. The Vulnerability of Auctions to Bidder Collusion , Volume =

  78. [78]

    Marshall and Leslie M

    Robert C. Marshall and Leslie M. Marx , Journal =. Bidder collusion , Volume =

  79. [79]

    Equilibrium in Sealed High Bid Auctions , Volume =

    Eric Maskin and John Riley , Journal =. Equilibrium in Sealed High Bid Auctions , Volume =

  80. [80]

    Asymmetric Auctions , Volume =

    Eric Maskin and John Riley , Journal =. Asymmetric Auctions , Volume =

Showing first 80 references.