Cartel Deterrence: The Search for Evidence and Argument

• Author: Christopher Harding
• Published date: 01 June 2011
• Date: 01 June 2011
• DOI: http://doi.org/10.1177/0003603X1105600206
• Subject Matter: Article

T H E A N T I T R U S T B U L L E T I N : Vol. 56, No. 2/Summer 2011 : 345
Cartel deterrence: The search
for evidence and argument
BY CHRISTOPHER HARDING*
Discussion and argument concerning the deterrent impact of anticartel
enforcement measures have gained in significance and controversy as
the whole regulatory enterprise and rhetoric of cartel regulation has
expanded and developed. Regulators have become ever more emphatic
in the denunciation of business cartels, and deterrence has therefore
assumed a significant role in the justification of a stronger and more
determined effort of legal control. In particular, cartelists need to be
deterred into whistle blowing for leniency programs to work, and busi-
nesses need to be deterred into compliance to achieve the ultimate regu-
latory goal. Deterrence (rather than, for instance, economic
redistribution or retributive justice) then becomes a leading, perhaps the
leading objective within the legal control of cartels. Yet claims of deter-
rent effect are notoriously difficult to measure and assess. The discussion
here examines the underlying problems of penetrating and measuring
the deterrent impact of anticartel policies and sanctions and considers
the possible methodologies for assessing deterrence in this context.
I.
THE RHETORIC AND REALITY OF DETERRENCE
The deterrent effect of legal measures aimed at the regulation and
control of business cartels is now a subject of considerable interest
and debate. The following discussion investigates both the claims
within the rhetoric of legal control and attempts that may be made to
assess actual deterrent effect.
*
Department of Law and Criminology, Aberystwyth University.
© 2011 by Federal Legal Publications, Inc.

---

346 : T H E A N T I T R U S T B U L L E T I N : Vol. 56, No. 2/Summer 2011
First, it would be useful to trace briefly the emergence of a rheto-
ric of deterrence and to probe its more exact aspirations. Why has talk
about deterrence become so significant in the policy and practice of
the legal control of business cartels? During the historically earlier
stages of cartel regulation, when (at least outside the United States)
the implementation (rather than enforcement) of competition policy
and rules was largely an administrative and consensual matter,1 the
absence of strong and categorical censure meant also an absence of
deterrence talk. While undesirable forms of horizontal restrictions
(more rarely then termed “cartels,” as such)2 were negotiated away,
there was no need to deter any delinquent activity. Deterrence would
enter the picture only when official policy toward most forms of “pri-
vate” cartel hardened during the 1970s and 1980s so that determined
cartelists had to go underground and start behaving like offenders
rather than subjects of consensual regulation.3 Thus began a vicious
circle and upward spiral of enforcement:
the more secretive and
more evasive the cartels became, the greater the powers of investiga-
tion and enforcement required by regulators, provoking ever more
subterfuge on the part of companies, which in turn required yet
greater powers of investigation and sanctioning, so transforming
companies and their employees into criminal-like actors.4
In this way enforcement activity—more and more comprising a
form of policing, the use of tough sanctions, and carrot-and-stick
leniency arrangements to gain evidence—gradually moved up the
competition regulators’ agenda.5 Recently, a British commentator on
European competition law could assert that:
1
See DAVID J. GERBER, LAW AND COMPETITION IN TWENTIETH CENTURY
EUROPE (2001).
2
See CHRISTOPHER HARDING & JULIAN JOSHUA, REGULATING CARTELS IN
EUROPE 11–16 (2003) (discussing the etymology of the term “cartel”).
3
See William E. Kovacic, Competition Policy and Cartels: the Design of
Remedies, in CRIMINALIZATION OF COMPETITION LAW ENFORCEMENT 41, 51
(Katalin J. Cseres, Maarten Pieter Schinkel & Flors O.W. Vogelaar eds., 2006).
4
See id. at 51–53.
5
See id.

---

E V I D E N C E A N D A R G U M E N T : 347
There is a very real sense today among the world’s competition authori-
ties that, if competition law is about one thing above all, it is the detection
and punishment of hard-core cartels. In the European Union, Mario
Monti, the former Commissioner for Competition, once described cartels
as “cancers on the open market economy,” and the Supreme Court in the
U.S. has referred to cartels as the “supreme evil of antitrust.”6
As cartel activity became described in these terms, yet showed no
sign of disappearing simply through the force of legal condemnation,
the deterrent method naturally entered the picture—how else to
secure respect for the prohibition? Thus at the present time, regula-
tory language is replete with the need to achieve deterrence. For
instance, in a speech in October 2009, European Union Competition
Commissioner Neelie Kroes proclaimed: “I don’t want to merely
destabilize cartels. I want to tear the ground from under them. This
requires effective deterrence across a range of competition systems.”7
And, at an Organisation for Economic Co-operation and Develop-
ment roundtable discussion on anticartel enforcement in 2005, a
national representative had opined that: “No single sanction is a suffi-
cient deterrent, but it was important that a panopoly [sic] of sanctions
is available to combat cartels.”8 The regulatory message is clear: com-
bating cartels has become a priority, and legal action to do so must
comprise effective deterrence.
Yet it remains far from clear what amounts to effective deterrence
in this context, or indeed how deterrence may be measured. A major
problem resides in the fact that, while the rhetoric is strong, it remains
inexact. What (or who) more exactly should be deterred, beyond a
broad sense of cartel activity? And when is it known whether some-
thing or somebody has been deterred, and by what? These questions
must be addressed in any enquiry into the impact and effectiveness of
anticartel measures, so as to achieve:
6
RICHARD WHISH, COMPETITION LAW 498 (5th ed. 2009).
7
Neelie Kroes, European Comm’r for Competition Policy, Remarks at
the Anti-Cartel Enforcement: Criminal and Administrative Policy Panel
Session, Speech/09/454 (Oct. 8, 2009).
8
ORG. FOR ECON. COOP. & DEV., POLICY ROUNDTABLES, CARTEL SANCTIONS
AGAINST INDIVIDUALS 105 (2005) (quoted comment was made by Canada’s
representative).

---

348 : T H E A N T I T R U S T B U L L E T I N : Vol. 56, No. 2/Summer 2011
•
a more exact idea of the object of deterrence;
•
a more exact idea of the aim of deterrent policy;
•
a more reliable sense of an absence of cartel activity as a measure
of deterrent effect (deterrent outcome); and
•
a more confident knowledge of the connection between particular
processes or instruments of deterrence and an apparently deterrent
outcome (deterrent linkage).
A. The object of deterrence
This issue is complicated by the well-known problem of agency in
relation to cartel action—the fact that cartel activity involves at one
and the same time both corporate and human individual conduct.
Immediately the question arises whether deterrent measures should
be directed at companies, the individuals who work for and give
expression to the actions of such companies, or both. This problem is
well illustrated by the enforcement puzzle: it is sometimes argued
that companies may not be deterred by fines, but executives will be
deterred by the prospect of a prison term.9 Does this imply that there
is no point in imposing financial penalties on a company and that
resources should be devoted to proving a criminal case against indi-
viduals and securing prison terms for the latter? The question of object
is also bedeviled by the problem of cartel identity. Are all cartels to be
deterred? Presumably not; rather, only “hard core” private cartels
(however defined) should be deterred.10 Moreover, is the identity of
individual cartels always clear (in terms of market, time and member-
ship)? Any meaningful measurement of deterrent effect requires an
accurate sense of the unit of measurement—how many cartels, how
many companies, how many individuals have been deterred?
9
See, e.g., id at 16; Donald I. Baker, The Use of Criminal Law Remedies to
Deter and Punish Cartels and Bid-Rigging, 69 GEO. WASH. L. REV. 693, 713 (2001);
Donald C. Klawiter, After the Deluge: The Powerful Effect of Substantial Criminal
Fines, Imprisonment, and Other Penalties in the Age of International Cartel
Enforcement, 69 GEO. WASH. L. REV. 745 (2001).
10
See ORG. FOR ECON. COOP. & DEV., FIGHTING HARD CORE CARTELS: HARM,
EFFECTIVE SANCTIONS AND LENIENCY PROGRAMMES (2002).

---

E V I D E N C E A N D A R G U M E N T : 349
B.
The aim of deterrence
The point here is to press for a more explicit description of what is
being sought or what will satisfy the deterrent objective. For, realisti-
cally, a spectrum of desirable outcomes might be listed. These could
range from the ideal, that is, complete eradication, the elimination of
all business cartels from this world, through to perhaps more feasible
outcomes, for instance, the removal of cartels from a particular mar-
ket, or the diversion of cartel activity to another geographical area or
jurisdiction with softer enforcement, or the decision of a major com-
pany not to engage in cartel conduct in the future, or for particular
executives to desist in the future. Or, in more general terms, is the aim
or emphasis to be placed on individual deterrence (actual...