mlex
A B
E X T R A
The settlement myth
Stephen Kinsella argues that Article 9 is in danger of
being over-used, that the procedure doesn’t provide a
“settlement” mechanism, and that recent enthusiasm
for it is likely only to shift workload from the
European Commission to the courts
Stephen Kinsella
Stephen Kinsella OBE is partner and head of the European antitrust group at
Sidley Austin, based in the Brussels office. He has practiced European competition
and regulatory law for almost 30 years. Stephen has extensive experience in the
application of EU competition law, including merger control, cartel investigations,
litigation and antitrust counseling on a broad variety of commercial practices.
MLex’s online market intelligence services have become indispensable primary resources
for anyone requiring reliable, comprehensive, real-time intelligence, commentary and analysis
about the impact of regulation on businesses around the world.
MLex customer services +44 (203) 402 7000
[email protected] – www.mlex.com
AB EXTRA – ANTITRUST SETTLEMENTS
THE SETTLEMENT MYTH
Stephen Kinsella argues that Article 9 is in danger of being over-used, that the
procedure doesn’t provide a “settlement” mechanism, and that recent enthusiasm
for it is likely only to shift workload from the European Commission to the courts
I
t is increasingly common to read in MLex and other
publications that a European Commission antitrust
investigation, particularly one involving high-tech
markets and abuses of dominance, might end in some form
of “settlement” between the commission and the company
concerned. There are many good reasons why journalists might
use terminology that they think will make the position clearer
for their readers. However one can even find direct examples
of commission officials using the word “settlement” to refer
to the procedure for accepting commitments under Article
9 of Commission Regulation 1 of 2003. Indeed in a press
conference on Nov. 14, in response to questions regarding the
Google case, Commissioner Almunia was quoted as saying:
At the end of the process, we or I will have to make
a decision along with my colleagues in the commission
on whether a sufficient compromise has been made and
can be made legally binding [my emphasis].
I would like in this brief comment to make three points: that
Article 9 is in danger of being over-used; most importantly,
that it does not provide a “settlement” mechanism; and that the
recent enthusiasm for Article 9 is likely only to shift workload
from the commission to the courts.
Excessive use?
If we go back to the time that Regulation 1 was drafted, there was
apparently very little discussion of the likely scope and impact
of Article 9. The commission was understandably focused
more on Article 101 cases than Article 102, and specifically
on the consequences of moving away from a notification
system to one where companies would have to make their own
assessments. As a result, it was likely that there would be even
more need for precedents to guide that assessment process, and
in its White Paper on Modernisation, the commission gave the
following assurance to those concerned at its withdrawal from
handling notifications:
In a directly applicable exception system, commission
policy on competition would continue to be reflected in
prohibition decisions in individual cases, and these would
be of great importance as precedents. As the commission
would be concentrating its attention on the most serious
restrictions, the number of individual prohibition
decisions can be expected to increase substantially.
The overall impression was that Article 9 might be used
rather exceptionally. Its take-up was initially slow and only
five commitment decisions in application of Article 102 were
adopted in the first five years, balanced by an equal number
of prohibition decisions. However in recent years the pace
has increased dramatically, to the extent that of the 15 cases
under Article 102 to have been closed by a decision since 2009,
13 of them have been a commitment decision and only two
have been prohibition decisions. In other words, some of the
most important cases, centering on companies with enormous
market power who the commission has concluded abuse that
power to the detriment of the EU market and consumers, are
being closed through a process that is not subject to the full
range of those checks and balances which evolved to prevent
misjudgment and abuse of administrative powers.
Lately a number of commentators have begun to question
whether the commitments procedure is being used
excessively,1 to the detriment of the “normal” Article 7
procedure, which establishes infringement and helps develop
precedent. The statistics seem to support that concern. Since
I first proposed this short article, Professor Richard Whish
has expressed his own reservations that it is being invoked
too readily to close difficult cases, with a knock-on impact
on both deterrence and the development of jurisprudence.
Other eminent figures including the chair of the OECD
Competition Law and Policy Committee, Frederic Jenny, have
echoed those concerns.
6 January 2014
mlex AB EXTRA
1
AB EXTRA – ANTITRUST SETTLEMENTS
Wrongful use?
To my mind, more troubling than the incidence of use is whether
Article 9 is being used inappropriately, even unlawfully. Let’s
start with the text of Article 9 (see box):
Article 9 – Commitments
1.Where the Commission intends to adopt a
decision requiring that an infringement be brought
to an end and the undertakings concerned offer
commitments to meet the concerns expressed
to them by the Commission in its Preliminary
Assessment, the Commission may by decision
make those commitments binding on the
undertakings. Such a decision may be adopted for
a specified period and shall conclude that there are
no longer grounds for action by the Commission.
It is worth breaking that wording down a little. At face value it
appears clear that an Article 9 decision is only possible where
the commission has prima facie evidence of breach of the law
and had already formed a preliminary intention to proceed
to an infringement decision. In other words, the case team,
often after a fairly detailed examination of the evidence, has
reached the conclusion that the competition rules are being
infringed and that significant harm to the European market
and consumers has occurred. The commitments have to be
offered by the undertaking concerned, and they must meet
the objections that will previously have been spelled out in a
Preliminary Assessment. The decision may, but does not need
to, be limited in time and it will conclude that there are no
longer grounds for action but it most certainly cannot conclude
that there were never grounds for action.
Only last month, Alexander Italianer, Director General
for Competition, in a speech to the CRA conference in
Brussels, listed the three criteria that need to be satisfied for
commitments under Article 9 when he said: “They should be
quick. They should be offered promptly, at the first opportunity,
not at the end of the procedure. They should be sufficient.
The commitments should effectively address the competition
concerns that we have uncovered. They should be sensible.
Good commitments are not more complicated than necessary.
They should not be difficult to implement and to monitor.”
Why would the commission allow a company to escape with
commitments when there is no doubt in its own mind that it
could have taken a decision, imposed sanctions and possibly
fines? We read that the principal justification is that it will save
time and resources that can then be allocated to other cases.
For the commission, Article 9 is said to be simply there to
provide a quicker and more efficient solution. Looked at from
the point of view of the company concerned, it escapes the
proceedings without a formal finding of liability and is free to
get on with running its business without the shadow of years
of confrontation and uncertainty.
However, in terms of substance, it ought to be clear that the
commission is not entitled to accept any offer that fails to
“meet” the concerns identified in its Preliminary Assessment.
And because we are entitled to assume that it will have been
thorough in its (unpublished) Preliminary Assessment, we
can rely on that assessment having pointed out all the serious
concerns which the case team harboured following its initial
investigation. Article 9 is not to be confused with the so-called
“settlement” route open to cartelists who strike a bargain for
a relatively small discount from their fine. Under Article 9, the
undertaking concerned is obliged to amend its behaviour so as
to put its compliance beyond question. The Oxford English
Dictionary provides a definition of “meet,” meaning “to
satisfy; to satisfy the requirements of; to be able or sufficient
to discharge.” Applying that definition, Article 9 does not allow
the commission and the defendant to “meet halfway.” The
dominant company cannot expect to escape with anything less,
in terms of behavioural change, than would have resulted from
an infringement decision.
There are a number of explanations for the confusion which
seems to exist regarding Article 9 and “settlement.” Lax
terminology by officials, commentators and the press is part of
it. That may have been compounded by misinterpretation of the
very little case law on the subject. In particular, the Alrosa case2
is sometimes cited as giving the commission almost unfettered
leeway in relation to use of Article 9. But that is a misreading of
the judgment. In Alrosa, the court, rightly, confirmed that the
commission has great discretion whether to accept commitments
rather than proceed by way of a prohibition under the “normal”
(and most would say “preferred”) route of Article 7. But there
is no suggestion in the Alrosa case that Article 9 gives the
commission carte blanche to shrink from its duty vigorously
to enforce the competition rules where it has established a
prima facie infringement. Indeed, the Alrosa case was more
concerned with an unsuccessful attempt to contest the right of
the commission to accept under Article 9 a remedy that it could
arguably not have achieved under Article 7. The court pointed
out the real nature of the trade-off that is reached in Article 9:
Undertakings which offer commitments on the basis
of Article 9 of Regulation No 1/2003 consciously
accept that the concessions they make may go beyond
what the Commission could itself impose on them in
a decision adopted under Article 7 of the regulation
after a thorough examination. On the other hand, the
closure of the infringement proceedings brought against
those undertakings allows them to avoid a finding of an
infringement of competition law and a possible fine.
However, looking at some of the more recent concluded and
ongoing cases, it is hard to avoid the suspicion that Article 9 is
6 January 2014
mlex AB EXTRA
2
AB EXTRA – ANTITRUST SETTLEMENTS
being misused, either to obtain unjustified results in cases that
Challenges to unsatisfactory outcomes in an Article 9 case can
were not rock solid or, increasingly, to strike a private bargain
also lead to the unedifying spectacle of the commission and
allowing companies to get away with unlawful behaviour. That
the dominant company appearing as co-defendants in a case
impression is heightened by contrasting the language used in the
brought by European customers and competitors harmed by
preliminary paragraphs of certain decisions, which demonstrate
abusive practices. But if the commission wishes to test the
a surprising variety in their account of the commission’s
limits of its discretion under Article 9, it may prove necessary
position. Therefore in the Distrigas decision in 2007, it stated
for the courts to intervene. Many of us had been watching with
that “according to these Preliminary Assessments, Distrigas
interest the Hynix appeal,3 in which Hynix was challenging the
is dominant” and that its contracts “limit the scope for other
commission decision to accept commitments from Rambus in
gas suppliers to conclude contracts with customers and so
an alleged “patent ambush” case. There had been hopes that any
foreclose their access to the market.”
court ruling might have brought some
In 2012 in Thomson Reuters, we read
clarity, but the appeal was withdrawn
that “according to the Preliminary
by Hynix in July 2013 following reports
The failure to make
Assessment, Thomson Reuters is
of a settlement between the companies
a reasoned decision
dominant.” Yet in other decisions
concerned.
the wording is far more equivocal.
in cases of such
One would normally expect that the
More generally, and as spelled out by
importance deprives
Preliminary Assessment, which is kept
Professor Whish and others, the failure
confidential, will have expressed clear
to make a reasoned decision in cases
others of much-needed
views on dominance and abuse. But in
of such importance merely serves to
precedents
the 2011 IBM decision, the preamble
deprive the courts of much-needed
confirms only that “in a Preliminary
precedents. If, for instance, a major
Assessment ... the commission formally expressed preliminary
issue involving appropriate use of standard essential patents is
concerns,” while in Rio Tinto in 2012, the language appears
resolved by commitments without admission of wrongdoing,
even weaker, with a reference to having reached a “provisional
where is the guidance for the industry to follow, whether
conclusion ... that [certain practices] raised concerns [and]
licensor or licensee? And it is at odds with the oft-spoken aim
might produce negative effects.” It must be apparent to any
of encouraging recovery of damages that the commission
objective observer that something is amiss here. It almost
should resort to non-findings that make even heavier the
looks as if there are some cases where in addition to letting the
already weighty burden faced by plaintiffs in civil cases.
company draft the commitments, the commission has handed
over control of the operative part of the decision.
Sooner or later this question will come before the court, which
will be asked to rule on the proper bounds of Article 9 and
A false economy?
whether it gives to the commission an unrestrained discretion
to compromise difficult cases and conduct private negotiations
I am aware of a number of complainants who have been
with powerful companies. I sincerely hope the court will give
dissatisfied with Article 9 decisions that they felt failed
the right answer and put Article 9 back in its box. n
to deliver on earlier assurances or which, without proper
explanation, stepped back from the preliminary conclusions.
That itself could call into question the procedural efficiency
justification for Article 9 if the commission first finds itself
Stephen Kinsella OBE is partner and head of the European antitrust
having to adopt reasoned decisions to reject complaints (more
group at Sidley Austin, based in the Brussels office. The views expressed
than 20 for instance in the Google case) and then faces appeals
in this article are personal to the author and do not necessarily reflect the
both to those rejection decisions and to the Article 9 decision.
view of Sidley or any of its clients. Stephen Kinsella has represented a
Is such a predictable transfer of work from one EU institution
number of parties in antitrust investigations where Article 9 settlements
to another self-evidently efficient?
have been accepted or proposed.
Footnotes
1
See, for example
https://www.competitionpolicyinternational.com/the-europeancommission-s-practice-under-article-9-regulation-1-2003-acommitment-a-day-keeps-the-court-away/
2
Alrosa: Case C-441/07 P, judgment of 29 June 2010
3
Hynix: Case T-148/10 of 25 March 2010
6 January 2014
mlex AB EXTRA
3