Excerpt From:
The Role, If Any, of Economic Analysis in Antitrust Litigation
(Southwestern University Law Review, 1981)
By JAMES V. DeLONG

NOTE: This article, written in 1980, was rendered pretty much obsolete by the antitrust revolution of the 1970's and 1980's. However, in the light of the recent suits against Microsoft and Intel, the excerpt quoted below remains relevant. Given the government's current, disturbing proclivity to revivify the antitrust doctrines of thirty years ago, the rest of the piece may soon become so. To obtain a copy of the entire paper, email the Regulatory Policy Center.

PURPOSES OF THE ANTITRUST LAWS

In discussing possible future applications of economic concepts one must begin at the beginning. Granting that the antitrust laws do have economic purposes, what are these purposes? This seems like an odd question on the ninetieth anniversary of the Sherman Act, but it is nonetheless a real one. The short answer, the absolutely obvious answer, is of course "to preserve competition," and in the historical context of the Sherman Act this answer was accurate enough. The builders of the agglomerations of the late nineteenth century had elimination of competition as their precise and announced purpose. Their goal was to merge or eliminate all rivals and make monopoly profits, and at least temporarily some of the combinations of this period achieved eighty or ninety percent market shares.<114> Everyone can agree that the need to preserve competition in the sense of preventing such monopolies is an admirable goal, even if there is no consensus on the market configuration that should exist.

Defining "competition" or "a competitive market" in an affirmative sense -- saying what it is rather than what it is not -- is a much more difficult problem. As Professor Robert Bork points out in a very valuable part of his book, the term "competition" has been used by courts and commentators in many different ways, often with little awareness of the shifts in meaning or the impact of the shifts on the analysis.<115> Here are the key definitions:

1. A maximization of rivalry to maintain a kind of Hobbesian war of all against all. In extreme form, this concept is the reductio ad absurdum of the antitrust laws, since two partners establishing their rates are, in a literal sense, restraining trade.<116> Although the idea has never been taken seriously as a policy prescription, it has an impact because it transforms readily into the belief that maximization of rivalry equals maximization of rivals, and thus any arrangement that threatens to decrease the number of rivals is anticompetitive. This is the line of reasoning that leads to the conclusion that efficiency is part of the offense.

2. A variant-Hobbesian view which allows independent entities to merge and integrate their operations completely, but requires total rivalry among all entities which remain separate.<117> This approach is not as obviously unworkable as the first one, but it too, applied in its rigorous sense of prohibiting cooperation between rivals, would lead to strange results. It would mean no trade associations, for example, and no agreement to standardize threads for nuts and bolts.

As a paradigm, though, it is important. It furnishes the bedrock concept of antitrust law which makes the entity the unit of analysis. The internal workings of an entity, no matter how large, are immune from scrutiny; the relations between two entities, no matter how small, are not. As was discussed in connection with Topco,<117a> this tends to limit the repertoire of responses to market situations, promoting internal expansion and eliminating the alternative of partial integrations,

3. An amorphous state of the market in which no seller can alter the price by varying his output and no buyer can alter it by varying his purchases. This is the formal economist's definition of pure competition. It is a useful analytic tool, but because the existence of such a market depends upon a number of unattainable preconditions it is not a good statement of public policy goals.<118> This concept has been influential because it tends to be bastardized into the rivalry or fragmentation definitions. It also furnishes the ideal of the "competitive price" -- the point at which the price equals the marginal cost of the least efficient producer.

4. An amorphous fragmentation of industries and markets. This is the Brown Shoe<119> definition, and seems related to both maximization of rivalry and perfect competition models, except that it is even more vague. For the most part, this approach is associated with the populist goals of antitrust law more than the economic.

5. An antibondage view which emphasizes the importance of maintaining the freedom of action of the small businessman by making restraints imposed upon him, or, to shift perspective, restraints he imposes on himself to procure compensating benefits, illegal.<120>

6. A guarantee of equal opportunity for the smaller businesses in an industry dominated by a large firm.<121> This definition has not been accepted by the courts, but it represents the theoretical core of the plaintiffs' cases in some recent suits involving dominant-firm conduct. The larger firm must permit or volunteer sufficient partial integration of its own activities with those of the plaintiff to enable the latter to operate as efficiently as the dominant firm.<122>

7. A market sufficiently deconcentrated to prevent sellers from colluding tacitly.<123>

8. A conduct analogue of the structural definition given immediately above: an absence of practices which make tacit collusion easier.<124>

9. A term of art designating a state of affairs in which consumer welfare cannot be increased through judicial decree.<125> This means a situation in which any economic gains from a particular judicial action -- increasing the number of rivals, for example -- would be more than offset by losses in other areas, such as the area of productive efficiency.

Obviously, these definitions are not synonymous. Some are descriptions of an economic structure, some are specifications of conduct. Aside from this, they are inconsistent, and some are almost direct opposites of others.

Notes:

<114> See A. Chandler, The Visible Hand (1977). Some mergers created large economies in production and distribution and were successful for that reason, but, in the eyes of contemporaries, it was the suppression of competition that counted. There were many failures of combinations that did not increase efficiencies before people began to understand that merger or cartel formation was not alchemy. Id. At 315-44.

<115> R. Bork, The Antitrust Paradox (1978), at 57-61. Bork sets forth five different meanings. In the instant list, the first two are contained in Bork's first definition. Numbers three, four, five, and nine are also Bork's. Numbers six, seven, and eight are drawn from other sources, as noted

<116> See, e.g., Broadcast Music, Inc. v. CBS, Inc., 441 U.S. 19 (1979).

<117> See, e.g., United States v. Citizens & S. Nat'l Bank, 422 U.S. 86, 116-17 (1975).

<117a>United States v. Topco Associates, 405 U.S. 596 (1972).

<118> R. Bork, The Antitrust Paradox (1978), at 59-60.

<119> Brown Shoe v. United States, 370 U.S. 294 (1962).

<120> See, e.g., Perma Life Mufflers, Inc. v. International Parts Corp., 392 U.S. 134 (1968). Contra, Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 53 n.21 (1977).

<121> See Sims, Antitrust Law is No Business Equal Opportunity Act, Legal Times of Washington, March 10, 1980, at 11, col. 1.

<122> See, e.g., Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263 (2d Cir. 1979), cert. denied, 444 U.S. 1093 (1980).

<123> This could be stated more generally as a desire to protect "workable competition" as that concept was developed in the 1940's and 1950's. See F. Scherer, Industrial Market Structure and Economic Performance (2d ed. 1980), at 41-44. However, the statement in the text represents the crux of contemporary concern.

<124> See Antitrust Division Memorandum on Identification and Challenge of Parallel Pricing Practices in Concentrated Industries, 874 Antitrust & Trade Reg. Rep. (BNA) F-1 (July 27, 1978). Both this definition and the structural equivalent could be encompassed in the two "rivalry" definitions; however, they represent a response to a specialized problem that seems worth separate identification.