Saving Section 2: Reframing U.S.Monopolization Law

• Published date: 01 April 2007
• Pages: 417-451
• Date: 01 April 2007
• DOI: https://doi.org/10.1016/S0573-8555(06)82016-X
• Author: Timothy J. Brennan

CHAPTER 16

Saving Section 2: Reframing U.S.

Monopolization Law

Timothy J. Brennan

Professor of Public Policyand Economics, University of Maryland Baltimore County;

Senior Fellow,Resources for the Future

E-mail address: brennan@umbc.edu

16.1. Section 2’s unfortunate distinctiveness

Antitrust policy famously divides into three categories: collusion, monopoliza-

tion, and mergers. In the United States, three very long-standing statutes initially

set out these categories, respectively Sections 1 and 2 of the Sherman Act (1890)

and Section 7 of the Clayton Act (1914).1These statutes, known in antitrust by

their section numbers, are notoriously short. Section 1 proscribes “every con-

tract, combination ...or conspiracy in restraint of trade;” Section 2 finds guilty

“every person who shall monopolize, attempt to monopolize, or ...combine or

conspire to monopolize;” and Section 7 prohibits asset acquisitions that “may

be substantially to lessen competition or to tend to create a monopoly.” The op-

erational content of these brief statutes is defined by the case law arising from

the public and private enforcement actions brought under them. Their ability to

remain a vital part of U.S. economic policy rests in their brevity and the ability

of case law to adapt them to changed circumstances and advances in knowledge.

Relatively speaking, collusion and merger law as set out in Section 1 and

Section 7 cases is uncontroversial within fairly broad margins, primarily be-

cause their focus, not necessarily exclusive, is on horizontal dealings among

competitors that suppress competition. Collusion cases under Section 1, partic-

ularly the per se violations of fixing prices and allocating markets, present the

clearest examples (Shenefield and Stelzer, 1998, pp. 43–50). Apart from liber-

tarian critiques based on freedom of contract (Armentano, 1982) and what might

be regarded as a generous view of the ability of free entry to cure all ills (Dewey,

1979), there is little dispute that horizontal collusion should be deterred. Inves-

tigations tend to focus more on factual questions such as whether there was a

conspiracy and by how much did it raise price than on theoretical analyses of

whether practices are harmful.

115 U.S.C. §§1, 2, 18.

CONTRIBUTIONS TO ECONOMIC ANALYSIS © 2007 ELSEVIER B.V.

VOLUME 282 ISSN: 0573-8555 ALL RIGHTS RESERVED

DOI: 10.1016/S0573-8555(06)82016-X

418 T.J.Brennan

There are margins of dispute, to be sure. “Rule of reason” Section 1 cases

often require delicate judgments about whether a nominally restrictive practice is

necessary to achieve greater benefits,2but those inquiries again tend to rest more

on the factual strength of an efficiency defense than on a fundamental policy

dispute.3Non-horizontal agreements, such as resale price maintenance and tie-

ins, are more problematic. For our purposes, we treat those here as variants of

monopolization.4

A similar rough consensus surrounds mergers, particular horizontal merg-

ers. The consensus on why such mergers are problematic, and how to analyze

them, constitutes the Department of Justice (DOJ) and Federal Trade Commis-

sion (FTC) Horizontal Merger Guidelines (HMGs) (DOJ, 1997). The underlying

consensus is reflected in the incorporation of these guidelines in merger analyses

taken throughout the government, e.g., in the FERC Merger Policy Statement

(FERC, 1996). This is not to suggest that merger cases are uncontroversial,

but that the root of the controversial is largely factual rather than on the prin-

ciple of how and perhaps whether cases should be brought (Eckbo and Weir,

1985;Crandall and Winston, 2003;Baker, 2003;Werden, 2004). Disputes fun-

damentally involve market delineation—the determination of who competes

with whom, and whether entry is likely to mitigate any problems created by

otherwise undue concentration.

Surrounding this question are other difficult issues involving the identifica-

tion of the mechanism of competitive harm (unilateral or coordinated effects)

(DOJ, 1997 at 2.1, 2.2),5the construction and relevance of differentiated prod-

uct simulation models (Froeb and Werden, 1996;Baker and Rubinfeld, 1999),

and other methods for estimating propensity of merged firms to reduce output

(Harris and Simons, 1989;Katz and Shapiro, 2003;O’Brien and Wickelgren,

2003). A more fundamental dispute is perhaps whether the appropriate measure

of welfare is total or just consumer welfare (Lande, 1982;Ross and Winter,

2004;Competition Bureau, 2004, §§8.5, 8.25–28, 8.31–35). But as with collu-

sion, there is relatively little conceptual (as opposed to empirical) dispute except

perhaps for mergers going beyond the market of the involved parties (Brennan,

2004b). Those, too, fall for our purposes here within the monopolization cate-

gory.

Monopolization cases, primarily but not exclusively brought under Section 2,

differ from both collusion and merger cases in being controversial at the more

2Broadcast Music Inc., et al., v. Columbia Broadcasting System et al., 441 U.S. 1 (1979).

3One could apply efficiency defenses even in per se cases (Clyde and Reitzes, 2004), but we do

not typically do so (Baker, 1991;Brennan, 2000).

4Exclusionary practices may be addressed under Section 1 and Section 2, and sometimes may be

found illegal under the latter yet legal under the former.U.S. et al. v. Microsoft Corp. (“Microsoft”),

87 F. Supp. 2d. 30 (D.C.D.C. 2000), direct appeal to the Supreme Court denied, 530 U.S. 1301

(2000); aff’d in part, r’vsd in part and remanded, 253 F. 3d. 34 (D.C. Circ. 2001); final judgment

entered, 2002 U.S. Dist. LEXIS 22864 (2002).

5Mergers leading to a less competitive oligopoly are included in the “unilateral” category

(Scheffman and Coleman, 2003, p. 323).

Saving Section 2: Reframing U.S. Monopolization Law 419

fundamental level of whether such cases should even be considered. Monop-

olization typically involves foreclosure or exclusion, the idea that actual or

potential rivals are precluded from competing in some market. These cases typ-

ically take the form of “abuse of dominance,”6where input or complementary

product suppliers (e.g., retailers) are threatened by a firm with market power

that its product will not be available, or only available at high prices, unless

those supplier sign exclusive contracts, accept tied products, or adopt similar

practices that keep out other entrants by “raising rivals’ costs” (RRC).7Atra-

ditional monopolization concern has been when vertical mergers or restraints

would restrict competition. A focus of some recent Section 2 cases has been the

use of discounts, particularly involving bundles of products, as a way to discour-

age buyers from turning to competitors.8

Unlike collusion and (horizontal) mergers, monopolization has been a per-

sistently controversial practice, for reasons familiar to those engaged in antitrust

policy debates over the last twenty-five years. Normally, firms cut prices without

taking into account the adverse effects on competitors’ profits. Increasing price

involves inhibition of independent action among competitors, i.e., sellers in the

same market. Vertical dealings or other relationships between firms in separate

markets should be largely irrelevant. Vertical integration and restraints have to

do with the internal organization of production; they do not expand coverage

over a market.

The “Chicago” critique pointed out that forcing buyers to do certain things,

such as accept tied goods or use only selected retailers, reduces demand for the

product, implying a justification based on cutting costs, raising quality, or im-

proving the marketing of the product. Penalizing firms, even monopolies, for

bundling, discounting, or charging low prices is to penalize them for doing what

competitive firms are supposed to do. Central to all of these concerns is that if

competition is also about beating rivals in a market, monopolization law may

protect competitors at the expense of competition, harming the public that com-

petition is nominally supposed to benefit.

An initial goal is to understand why monopolization law differs from col-

lusion and merger law in the quality and level of controversy. For example, is

62002 O.J. (C325) 65 (2002). Facey and Assaf (2002) provide a useful review of monopolization

law in the U.S., the European Union, and Canada.

7U.S. v. Dentsply (“Dentsply”) 277 F. Supp. 2d 387 (D.C. Del. 2003), r’vsd and

remanded, 399 F.3d 181 (3rd Circ. 2005); Microsoft, supra note 4; European Com-

mission, Case Comp/C-3/37.792 Microsoft, Mar. 3, 2004. http://europa.eu.int/comm/

competition/antitrust/cases/decisions/37792/en.pdf. For more on raising rivals’ costs, see

Krattenmaker and Salop (1986).

8Concord Boat Corp. v. Brunswick Corp., 207 F.3d 1039 (8th Cir. 2000), cert. denied, 531 U.S.

979 (2000) (Brunswick’s purchase share and absolute volume discounts for purchasesof motor boat

engines were found legal as above cost, with no implicit tying in the sales of different types of

engines); LePage’s, Inc. v.3M (“LePage’s”), 324 F.3d 141 (3d Cir.Pa., 2003), cert. denied 124 S.

Ct. 2932 (2004) (3M’s discounting of bundled wholesale sales of trademarked (Scotch) and store-

brand tape excluded LePages from retail outlets).