Ahead of His Time

• Author: Timothy J. Brennan
• DOI: 10.1177/0003603X15625108
• Date: 01 March 2016
• Published date: 01 March 2016

Article
The Antitrust Bulletin
2016, Vol. 61(1) 109-120
Ahead of His Time: The
ª The Author(s) 2016
Reprints and permission:
Singular Contributions
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0003603X15625108
of Richard Markovits
abx.sagepub.com
Timothy J. Brennan*
Abstract
Prof. Markovits has been a prescient commenter on antitrust. His early and insightful characterization
of the harm from merger as the loss to consumers from being able to trade off their two top choices
can be seen in the use of diversion ratios in unilateral effects merger cases. So too does the change in
assessing unilateral effects mergers from market concentration to direct measurement of competitive
effects match his skepticism of market definition. I share Prof. Markovits’s view that prior dominance
should not be a prerequisite for monopolization cases, at least those involving exclusionary practices. I
am skeptical that intent should play a role in monopolization, other than as a source of expert evidence
that a practice will be harmful. Whether antitrust should be about maximizing consumer welfare or
punishing powerful actors with bad intent may be a fundamental divide in perspective in the antitrust
community.
Keywords
intent, exclusion, merger harm, market definition
I am honored to have been invited to contribute to this symposium on the contributions of Richard
Markovits to the analysis of antitrust law and policy, reflected in his recently published two-volume
Economics and the Interpretation and Application of U.S. and E.U. Antitrust Law.1 In important ways,
particular in his analysis of mergers, Prof. Markovits, has been a couple of decades ahead of the
received wisdom. I will begin my comment by highlighting these contributions. They have become
part of the conventional wisdom in recent years, more so than Prof. Markovits’s portrayal of his work
1. RICHARD S. MARKOVITS, ECONOMICS AND THE INTERPRETATION AND APPLICATION OF U.S. AND E.U. ANTITRUST LAW, VOL. I: BASIC
CONCEPTS AND ECONOMICS-BASED LEGAL ANALYSES OF OLIGOPOLISTIC AND PREDATORY CONDUCT (hereinafter MARKOVITS I);
RICHARD S. MARKOVITS, ECONOMICS AND THE INTERPRETATION AND APPLICATION OF U.S. AND E.U. ANTITRUST LAW, VOL. II:
ECONOMICS-BASED LEGAL ANALYSES OF MERGERS, VERTICAL PRACTICES, AND JOINT VENTURES (hereinafter MARKOVITS II) (2014).
* School of Public Policy, University of Maryland Baltimore County, Baltimore MD, USA; and Resources for the Future,
Washington DC, USA
Corresponding Author:
Timothy J. Brennan, School of Public Policy, University of Maryland, Baltimore County, Baltimore, MD 21250, USA.
Email: brennan@umbc.edu

---

110
The Antitrust Bulletin 61(1)
sometimes suggests. But that should not take away from the credit due Prof. Markovits for seeing some
important errors and omissions in antitrust long before others did.
I begin by reviewing Prof. Markovits’s prescient contributions to describing the harm from mergers
and follow that with some observations on his place in the recent debate about the role of market
definition in merger assessment. I then turn to monopolization, the piece of antitrust law and econom-
ics I have thought about most. It is in this area that I most notably share ‘‘outsider’’ standing with Prof.
Markovits, although our critiques have some differences—not surprisingly, for such a notoriously
complex area of competition law. I begin with a look at Prof. Markovits’s critique of the definition of
dominance and then his view that it plays an exaggerated and unnecessary role in exclusion cases, a
view with which I generally concur.
Viewing Prof. Markovits as a kindred spirit, however, does not mean agreement with everything he
has to say. I conclude by taking issue with his emphasis on the role of intent in defining illegality as a
defining characteristic of monopolization, be it exclusion or predation. That criticism is likely a dispute
I as an economist have with antitrust lawyers collectively than with Prof. Markovits specifically.
Perhaps ironically, my concerns regarding intent may have the unintended and unexpected effect of
grouping Prof. Markovits among many in the antitrust bar with whom he typically and insightfully
disagrees.
The Harm from Mergers
I first became acquainted with Prof. Markovits’s work while I was a staff economist at the Antitrust
Division. In the spring of 1979, Prof. Markovits came to give a talk on how to view the harm from
mergers. I do not recall how Prof. Markovits’s talk fit into the timeline for the issuance of the first
modern Horizontal Merger Guidelines in 1982.2 But whether it was before, after, or during the
preparation of the Guidelines, the dominant paradigm for determining the harm of a merger was to
assess the level of concentration of firms, with identical products, within a specific product and
geographic market. Whether one should measure concentration by the sum of shares of the top four
firms (the four-firm concentration ratio or ‘‘CR4’’), as had been the fashion, or the sum of the squares
of the shares of the firms (the Herfindahl-Hirschmann Index or ‘‘HHI’’), was one of the major issues
under discussion. The leading conceptual innovation, though, was to propose a procedure for defin-
ing what ‘‘relevant market’’ should mean under antitrust law, based upon the smallest range of
products and production locations over which a hypothetical monopolist could profitably institute
a ‘‘small but significant non-transitory increase in price’’3 or, as frequently abbreviated, the
‘‘SSNIP’’ test.
Into this setting, Prof. Markovits suggested not that one particular measure or method of market
definition was better than another based on some relatively arcane criterion. Rather, Prof. Markovits
came to say that the standard method of assessing mergers didn’t need to be tweaked; it needed to be
discarded. In his view, the harm from mergers was almost shockingly straightforward, based directly
on the change in options available to buyers. In his view, firms offer not identical but different
products. Different buyers have different rankings over the different products available. We can think
of this ranking as based on the net value the buyer gets from the product, the difference between the
value the consumer gets and the price. In standard economic jargon, this would be called ‘‘consumer
surplus.’’ In this setting, the supplier of the buyer’s first choice can set a price only high enough so that
the buyer’s second choice becomes appealing.
2. U.S. DEPARTMENT OF JUSTICE, HORIZONTAL MERGER GUIDELINES (rev. 1984), http://www.justice.gov/atr/hmerger/11249.pdf.
3. Id. at II.A.

---

Brennan
111
Prof. Markovits’s key and keen observation was that mergers affect a buyer only when the
suppliers of the buyer’s top two ranked firms merge. An example can illustrate the principle.
Suppose firms A and B are the merging parties. If the buyer’s top choice is supplied by C and the
second choice by D, the ability of C to charge a premium to the buyer is unaffected by the merger. If
the buyer’s top choice is C and the second choice is either A or B, the premium for C is still
unaffected by the merger, as the postmerger firm retains the incentive to supply A or B on the terms
it offered it premerger. Similarly, if the buyer’s top choice is either A or B and the second choice is
C, it cannot increase the premium it charged premerger, since the buyer could then switch to C
postmerger as easily as she could before. However, suppose the buyer’s first choice is A and second
choice is B. Now, the firm can force the buyer to compare its first choice with her third best
alternative, call that E. It can then raise the price of A (whichever was top-ranked) to the point
where the buyer would turn not to B, but to E.
One obvious problem with this is that typical markets do not seem to have buyer-specific prices, a
topic to which I will return. However, when I heard Prof. Markovits present this theory of merger
effects many years ago, it seemed to make a lot of sense. Nevertheless, the structure/conduct/perfor-
mance paradigm, refined but not rejected by the 1982 Horizontal Merger Guidelines, continued to
carry the day.
In the ensuing decades, Prof. Markovits’s conception has ended up faring very well. It gradually
became evident4 that the market definition process and HHI standard in the Horizontal Merger Guide-
lines applied primarily, if not only, to mergers where the harm was that the merger would facilitate
collusion—what we now call ‘‘coordinated effects.’’5 The relevant market had to be large enough for
the potential price increase to matter—the SSNIP test again—yet where the market was concentrated
enough—the HHI standard—so that collusion was plausible. The economic rationale for the HHI test
in this context is that the HHI is a measure of the ability of cartel participants to ascertain if variation in
demand is the result of random actions of buyers or because someone cheated on the cartel, that is, cut
the price.6
The antitrust community came to accept Prof. Markovits’s insight that facilitating collusion was not
the only, or even the main, argument against mergers. Rather, as he suggested long ago, the concern
can and often will be with the elimination of competition between the merging parties themselves—
what has come to be called ‘‘unilateral effects.’’7 Observers, including the antitrust agencies them-
selves, have noted that the paradigm pioneered in the 1980’s Horizontal Merger Guidelines does not
pertain to mergers where the effects are unilateral. For example, in the 2010 Merger Guidelines, the
U.S....