Skip to main content
SpringerLink
Log in

Hidden collusion by decentralization: firm organization and antitrust policy

  • Published:
Journal of Economics Aims and scope Submit manuscript

Abstract

This paper develops a theory of the centralization of firms engaged in multi-market collusive agreements. A centralized organization (called the unitary or U-form) allows price coordination across several markets, whereas with decentralized (the multidivisional or M-form) firms the probability that the antitrust authority will find evidence of collusion on one market while investigating the other is lower. We show that the firm’s choice of internal structure depends to a large extent on product substitutability and the instruments used by the antitrust authority.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Notes

  1. Case COMP/36.545/F3.

  2. Cases COMP/E-1/36.604 and COMP/E-1/36.756.

  3. Case COMP/E-1/36.756.

  4. Case COMP/E-2/37.533.

  5. Particularly if markets are asymmetric (Bernheim and Whinston 1990), firms’ objective functions are concave (Spagnolo 1999), or demand is random (Matsushima 2001).

  6. Connor (1997) also points out that ADM is a very centralized firm: “For a company of its size and diversity, ADM is managed by a remarkably small number of managers”.

  7. Rasch and Wambach (2009) analyze a spatial model with two firms and three products. The decentralized organization of the two-product firm increases its incentives to deviate, but reduces the incentives to cheat for its rival (which produces only one product). The overall effect on collusion depends on transport costs. The two-product firm generally prefers decentralization in order to have greater bargaining power and a larger market share.

  8. Thomas and Willig (2006) consider another drawback of two linked collusive agreements: firm payoffs can be lower if information is perfect in one market but imperfect in the other. Under imperfect information, temporary price wars have to be used to discipline firms (as in Green and Porter 1984), but this extends to the other market if the collusive agreements are linked. Thomas and Willig (2006) show that the two collusive agreements may be isolated from each other to prevent the contagion risk from price wars. In our models, they may be separated to avoid the contagion risk from antitrust authority investigations.

  9. See Garoupa (2007) and Baccara and Bar-Isaac (2008).

  10. Aubert et al. (2006) consider the impact of leniency programs and reward programs for informants. The provision of rewards by the antitrust authority to self-reporting employees drives colluding firms to increase employee compensation. Firms also have an incentive to reduce turn-over and restructuring in order to avoid such compensation for new employees. As such, antitrust policy may increase inertia in the internal organization of the firm.

  11. Harrington (2006) presents some information about cartel structure and notes that it differs between cartels.

  12. In other words, we consider firms which compete with homogeneous products under markets with two substitutes.

  13. The general demand functions are:

    $$\begin{aligned} Q^{i}\left( p^{i},p^{j}\right) =\max \left\{ 0,\min \left\{ a-bp^{i}+d\left( p^{j}-p^{i}\right) ,\frac{\frac{a}{b}-p^{i}}{{\frac{b+d}{b(b+2d)}}}\right\} \right\} ,\forall (i,j)\in (A,B),i\ne j \end{aligned}$$

    Belleflamme et al. (2000) describe how these demand functions can be derived from a quadratic utility function. The demand for each product has a kink at a critical price at which consumers stop buying one of the products and demand a quantity \(\left( \frac{a}{b}-p\right) /\left( \frac{b+d}{b\left( b+2d\right) }\right) \) of the other, where \(p\) refers to the product for which the quantity is positive (see Vives 1999). As the values of \(a\) and \(b\) do not play a major role, we assign particular values to them: \(a=10\) and \(b=1\).

  14. In our paper, firms decide on their organizational structure before negotiating any collusive agreement. The chronology is actually arguably less clear. We can reinterpret our results by considering two stages. There is an informal first meeting attended by the CEOs of both firms which cannot be observed by the antitrust authority. In this meeting, the CEOs decide first whether to form cartels and second the organizational structure of their firms. This agreement is not binding, as firms can consequently decide on another organizational structure, but this will not happen since the organizational structure must be a Nash equilibrium of the non-cooperative game. After this first meeting, the CEOs choose the organizational form. Before competition takes place, the CEO (under the U-form) or managers (under the M-form) meet to discuss the collusive agreement. The choice of the organizational structure is then made after the first meeting between the CEOs but before managers are involved. This alternative interpretation of the model is closer to the real-life examples we cite in this paper, but requires a more complex model which in the end yields the same results as those presented here.

  15. This assumption is used to remove the well-known problem of equilibrium selection in coordination-game theory. Alternatively, we could assume that the organization’s choice is cooperative or that firms choose their organizational design sequentially.

  16. Choi and Gerlach (2009) mainly analyze the case where \(\mu _{1}=1\) and \(\mu _{2}=0\). They also consider an extension in which the antitrust authority can increase its investigative capacity leading to \(\mu _{2}=1\). The parameters \(\mu _{3}\) and \(\mu _{4}\) are irrelevant in their framework.

  17. We disregard the asymmetric case as it makes the model and the calculations more complex. Considering a symmetric equilibrium when firms are symmetric is natural. When they are asymmetric, firms face multiple equilibria and the selection criterion is less clear. The results depend on this criterion, which is not always the same across different economic analyses. Moreover, the no-deviation constraints are different and bind for different values of \(\delta \). If one constraint is violated then firms have an incentive to change the agreement by modifying the selection criterion, all of which yields a greater number of possible cases.

  18. There is a third possibility if \(\theta >1\): collude in only one market and stop colluding once firms are fined (\(U_{one}\) strategy). When \(\theta =1\), this third strategy is always dominated by \(U_{seq}\): if firms benefit from colluding in only one market, then they also benefit from collusion in the other once the first cartel has been discovered and fined, as both the first and second cartels yield the same profits.

  19. Cases COMP/36.545/F3, COMP/E-1/36.604 and COMP/E-1/36.756.

  20. Case COMP/39125.

  21. Case IV/31.149.

  22. Case IV/31.865.

  23. Case COMP/39181.

  24. \(\widetilde{\Pi }_{i}\) could be negative for high values of \(d\). But these high values do not satisfy the collusion-sustainability condition.

  25. A graphical representation is provided in Sect. 5.

  26. The only advantage of this latter strategy is that the detection of one cartel does not imply the detection of the other; but this also holds for the \(M_c\) strategy when \(\mu =0\).

  27. Case COMP/36.545/F3.

  28. Case COMP/E-1/C.37.671.

  29. “Since 1993–1994, the Akzo Nobel group has been organized on the basis of a two layer structure: a “corporate centre” and directly underneath approximately 20 Business Units (“BUs”). The corporate centre co-ordinates the most important tasks with regard to general strategy of the group, that is to say finance, legal affairs and human resources. The BUs each have their own General Manager, management team and supporting services responsible for the entire operational management of the BU.” EC (Case COMP/F/38.620—hydrogen peroxide and perborate).

  30. Case COMP/E-1/36.756.

  31. EKA Chemicals AB operated a cartel in the market for hydrogen peroxide and its downstream product sodium perborate up to 1999 (Case COMP/F/38.620), but also in the sodium chlorate market up to 2000 (Case COMP/38.695). Another subsidiary (Akzo Nobel Chemicals SpA) participated in the choline chloride agreement until 1998 (Case COMP/E-2/37.533).

  32. If \(\mu >0\), the most restrictive condition depends on the degree of product differentiation.

  33. Zone 3 corresponds to the result of Cyert et al. (1995), who consider that their theory applies if there is only little product differentiation. They describe real-life examples which prove that decentralization brings about the dissolution of the collusive agreement.

  34. Although the models are different, our results here are analogous to those in Pénard (2000).

  35. In zones 1 and 2 of Fig. 1, which are the relevant ones.

  36. \(F_{2}\) now rises with \(\delta \). The major advantage of the \(Mc\) strategy over the \(U_{sim}\) strategy is to permit collusion on a second market once the first cartel has been detected; this advantage rises with \(\delta \). The advantage falls, however, with \(d\) since product substitutability affects prices under one-market collusion. Another advantage is that the second fine is paid later in the \(Mc\) strategy, which advantage is lesser as \(\delta \) rises. For high values of \(d\) this second effect dominates the first and \(F_{2}\) changes into an increasing function of \(\delta \).

  37. Details can be found in Dargaud and Jacques (2010).

References

  • Aghion P, Tirole J (1995) Some implications of growth for organizational form and ownership structure. Eur Econ Rev 39:440–455

    Article  Google Scholar 

  • Aubert C, Rey P, Kovacic W (2006) The impact of leniency and whistle-blowing programs on cartels. Int J Ind Org 24:1241–1266

    Article  Google Scholar 

  • Baccara M, Bar-Isaac H (2008) How to organize crime. Rev Econ Stud 75:1039–1067

    Article  Google Scholar 

  • Baker WE, Faulkner RR (1993) The social organization of conspiracy: illegal networks in the heavy electrical equipment industry. Am Soc Rev 58:837–860

    Article  Google Scholar 

  • Baye MR, Crocker KJ, Ju J (1996) Divisionalization, franchising, and divestiture incentives in oligopoly. Am Econ Rev 86:223–236

    Google Scholar 

  • Belleflamme P, Picard P, Thisse J-F (2000) An economic theory of regional clusters. J Urban Econ 48:158–184

    Article  Google Scholar 

  • Berkovitch E, Israel R, Spiegel Y (2010) A double moral hazard model of organizational design. J Econ Manage Strategy 19:55–85

    Article  Google Scholar 

  • Bernheim BD, Whinston MD (1990) Multimarket contact and collusive behavior. Rand J Econ 21:1–26

    Google Scholar 

  • Besanko D, Régibeau P, Rockett KE (2005) A multi-task principal-agent approach to organizational form. J Ind Econ 53:437–467

    Article  Google Scholar 

  • Besanko D, Spulber DF (1989) Antitrust enforcement under asymmetric information. Econ J 99:408–425

    Article  Google Scholar 

  • Bloom N, Sadun R, Van Reenen J (2012) The organization of firms across countries. Q J Econ 127:1663–1705

    Article  Google Scholar 

  • Chandler A Jr (1962) Strategy and structure: chapters in the history of the American industrial enterprise. MIT Press, Cambridge

    Google Scholar 

  • Choi JP, Gerlach H (2009) Multi-market collusion with demand linkages and antitrust enforcement. Michigan State University, Mimeo

  • Connor JM (1997) The global lysine price-fixing conspiracy of 1992–1995. Rev Agric Econ 19:412–427

    Article  Google Scholar 

  • Creane A, Davidson C (2004) Multidivisional firms, internal competition, and the merger paradox. Can J Econ 37:951–977

    Article  Google Scholar 

  • Cyert R, Kumar P, Williams J (1995) Impact of organizational structure on oligopolistic pricing. J Econ Behav Org 26:1–15

    Article  Google Scholar 

  • Dargaud E, Jacques A (2010) Hidden collusion by decentralization: firms’ organization and antitrust policy. CEMOI Working Paper

  • Garoupa N (2007) Optimal law enforcement and criminal organization. J Econ Behav Org 63:461–474

    Article  Google Scholar 

  • Green EJ, Porter RH (1984) Noncooperative collusion under imperfect price information. Econometrica 52:87–100

    Article  Google Scholar 

  • Hammond S.D (2009) Cornerstones of an effective leniency program, address at an international competition network workshop, Sydney (2004)

  • Harrington JE Jr (2004a) Post-cartel pricing during litigation. J Ind Econ 52:517–533

    Article  Google Scholar 

  • Harrington JE Jr (2004b) Cartel pricing dynamics in the presence of an antitrust authority. Rand J Econ 35:651–673

    Article  Google Scholar 

  • Harrington JE Jr (2005) Optimal cartel pricing in the presence of an antitrust authority. Int Econ Rev 46:145–169

    Article  Google Scholar 

  • Harrington JE Jr (2006) How do cartels operate? Found Trends Microecon 2:1–105

    Article  Google Scholar 

  • Harrington JE Jr, Chen J (2006) Cartel pricing dynamics with cost variability and endogenous buyer detection. Int J Ind Org 24:1185–1212

    Article  Google Scholar 

  • Harris M, Raviv A (2002) Organization design. Manage Sci 48:852–865

    Article  Google Scholar 

  • Huck S, Konrad KA, Müller W (2004) Profitable horizontal mergers without cost advantages: the role of internal organization, information and market structure. Economica 71:575–587

    Article  Google Scholar 

  • Maskin E, Qian Y, Xu C (2000) Incentives, information, and organizational form. Rev Econ Studies 67:359–378

    Article  Google Scholar 

  • Matsushima H (2001) Multimarket contact, imperfect monitoring, and implicit collusion. J Econ Theory 98:158–178

    Article  Google Scholar 

  • Pénard T (2000) Délégation ou centralisation des décisions: l’impact sur la collusion. Rev Econ 51:5–26

    Google Scholar 

  • Puschke K (2009) Task assignment and organizational form. J Econ 96:149–168

    Article  Google Scholar 

  • Qian Y, Roland G, Xu C (2006) Coordination and experimentation in M-form and U-form organizations. J Polit Econ 114:366–402

    Article  Google Scholar 

  • Rasch A, Wambach A (2009) Internal decision-making rules and collusion. J Econ Behav Org 72:703–715

    Article  Google Scholar 

  • Souam S (2001) Optimal antitrust policy under different regimes of fines. Int J Ind Org 19:1–26

    Article  Google Scholar 

  • Spagnolo G (1999) On interdependent supergames: multimarket contact, concavity, and collusion. J Econ Theory 89:127–139

    Article  Google Scholar 

  • Spiegel Y (2009) Managerial overload and organization design. Econ Lett 105:53–55

    Article  Google Scholar 

  • Tan G, Yuan L (2003) Strategic incentives of divestitures of competing conglomerates. Int J Ind Org 21:673–697

    Article  Google Scholar 

  • Thomas CJ, Willig RD (2006) The risk of contagion from multimarket contact. Int J Ind Org 24:1157–1184

    Article  Google Scholar 

  • Vives X (1999) Oligopoly pricing: old ideas and new tools. MIT Press, Cambridge

    Google Scholar 

  • Zhou H (2005) Market structure and organizational form. South Econ J 71:705–719

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Université de Lyon 2, 69007 , Lyon, France

    Emilie Dargaud

  2. CNRS, GATE Lyon Saint-Etienne, 69130 , Ecully, France

    Emilie Dargaud

  3. CEMOI, Université de La Réunion, 15, avenue René Cassin, 97715 , Saint-Denis Messag Cedex 9, France

    Armel Jacques

Authors
  1. Emilie Dargaud
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Armel Jacques
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Armel Jacques.

Additional information

We thank the editor and anonymous referees for valuable comments and feedback. We have benefited from discussions with Ashley Roughton. We would like to thank participants in the IIOC and EARIE conferences 2013 for helpful comments and discussions. All errors remain, of course, our own.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Dargaud, E., Jacques, A. Hidden collusion by decentralization: firm organization and antitrust policy. J Econ 114, 153–176 (2015). https://doi.org/10.1007/s00712-013-0386-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00712-013-0386-9

Keywords

JEL Classification

Search

Navigation