Proving Competition

• Author: Benjamin King,Lee E. Preston
• Published date: 01 December 1979
• Date: 01 December 1979
• DOI: http://doi.org/10.1177/0003603X7902400404
• Subject Matter: Article

The Antitrust Bulletin/Winter 1979

PROVING

COMPETITION

by

LEE

E. PRESTON* and

BENJAMIN KING * *

How does one demonstrate, to the satisfaction of a judge,

jury and litigating parties,

that

aparticular market is effec-

tively competitive? This may seem an odd question, since the

burdern of proof in antitrust cases is on those alleging

that

competition is absent, or subject to restraint; and the bulk of

economic analysis in antitrust cases has been aimed

at

estab-

lishing the validity of such charges, i.e., "proving

monopoly,"

or

"restraint of trade."

The conventional response to such charges has not been the

presentation of a counter-analysis-that is, an analysis showing

that

competition is in fact present and effective in the

market-but

rather a point-by-point critique of whatever anal-

ysis may be

put

forward by plaintiffs. Unfortunately for many

defendants, responses of this type often prove unsatisfactory.

They tend, of necessity, to be limited to the issues, mode of

analysis and even data selected by the complaining parties.

Moreover, most markets possess some

features-franchise

agreements, patent licenses, trade association activities,

etc.-

that

constitute departures from conditions of perfect competi-

tion. And a few skips in logic can take one quickly from (1)the

incontestable presence of such features, to (2) a conclusion

that

prices and profits must be non-competitive, and finally (3) a

judgment

that

antitrust damages are due. This chain of argu-

ment is almost impossible to refute unless one can present an

alternative standard of competitive market behavior, and an

analysis

that

demonstrates a correspondence between actual

experience and this standard.

*Melvin H. Baker Professor of American Enterprise, School of

Management, State University of New

York

at

Buffalo.

* * Grant 1. Butterbaugh Professor of Quantitative Methods,

School of Business Administration, University of Washington.

©1980 by FederolLegal Publications. Inc.

787

788 THE ANTITRUST BULLETIN

We have recently had the opportunity to examine an impor-

tant

regional product market in just this context. The issue

arose in the form of rather vague charges based upon a history

of trade association activity and related market communication

among a group of sellers of an essentially fungible basic prod-

uct. Two significant product buyers charged

that

these com-

munications constituted illegal restraints upon trade, and

that

antitrust

damages were therefore due. To examine these

charges, we established a set of analytical criteria and as-

sembled data in order to describe and appraise the strength of

competitive forces in the market. Our investigation was based

upon a transaction-by-transaction record of market activity

that

we believe to be unique in the annals of antitrust analysis.

Since the matter was eventually settled without trial, the data

themselves have never become matters of public record; and

the identity of the market and of all market participants has

been disguised in this presentation.

TESTS OF EFFECTIVE COMPETITION

Under ideal circumstances, one would like to be able to

develop independent estimates of both demand and supply

functions under competitive conditions, and generate a time

series of price-quantity results

that

would serve as the

"com-

petitive norm" for a market. Then, a comparison of actual

experience with these results would serve as a test of the

presence and effectiveness of competitive forces. Unfortu-

nately, it is only rarely possible to conduct an analysis on this

basis. Market boundaries-both geographic and

product-may

be both uncertain and unstable over time; and, even within

identified market boundaries, quantitative estimates indepen-

dent of actual price and quantity levels are difficult to obtain

and subject to many limiting assumptions and qualifications.

Moreover, an analysis of this type, even if well conducted, is

difficult to present to an uninitiated audience in a courtroom

or administrative hearing.

As an alternative to the ideal theoretical approach, we

introduce three tests of effective competition suggested by the