Measuring the Degree of Interindustry Competition in U.S. v. Continental Can

• Author: Lawrence Wu,De-Min Wu
• Published date: 01 March 1997
• Date: 01 March 1997
• DOI: http://doi.org/10.1177/0003603X9704200106
• Subject Matter: Symposium: Economics of Antitrust Enforcement

The Antitrust Bulletin/Spring 1997

Measuring

the

degree

of

interindustry

competition

in

u.

S. v. Continental Can

BY LAWRENCE WU* and DE-MIN WU**

I.

Introduction

51

It has been more than 30 years since U.S.

v.

Continental

Can

Co.,

but in the area of antitrust, the case remains one of the Supreme

*Senior Consultant, National Economic Research Associates, Inc.,

White Plains, NY.

** Professor of Economics, Department of Economics, University

of

Kansas, Lawrence, KS.

AUTHORS' NOTE: This article is a revised version

of

research presented at

asymposium on The Economics

of

Antitrust

Enforcement which was

jointly sponsored by the Western Economic Association (WEA) Interna-

tional

and

The

Antitrust

Bulletin

and

held

at the 71st

Annual

WEA

International Conference on July 1, 1996. The insightful comments

of

Sumanth Addanki, Jonathan Baker, Phil Beutel. Carl Gwin, Rob Kneuper,

Bob

Levinson,

Rick

Liebeskind,

Bill

Lynk,

David

Monk,

Armando

Rodriguez. Chetan Sanghvi, Lou Silvia, and Mark Williams are greatly

appreciated. Much

of

this work was completed while the authors were at

the Federal Trade Commission and Tsukuba University, respectively.

©1997 by Federal Legal Publications, Inc.

52 The antitrust bulletin

Court's

most widely debated and discussed opinions.' To this day,

it is referred to as the textbook case on interindustry competition,

market definition, and the relevance of long-run competition.! The

case, which the Antitrust Division of the Department

of

Justice

brought to court in 1963, involved the acquisition of Hazel-Atlas

Glass Company, amanufacturer of glass containers, by Continental

Can Company, amanufacturer of metal can containers." The gov-

ernment challenged the merger on the grounds that the acquisition

violated section 7 of the Clayton Act. Although both companies

were among the largest in their respective industries, Continental

Can

did not produce glass containers, and Hazel-Atlas did

not

produce metal cans. The critical issue was therefore market defini-

tion: were metal cans and glass containers in the same market'i-

Market definition analysis, which attempts to identify the buy-

ers and sellers whose transactions determine the price of a particu-

lar group of products, is often acrucial first step in an antitrust

analysis. An analysis of the competitive effects

of

a merger, for

example, usually includes some study

of

market power. Market

power is often assessed by first specifying relevant product and

geographic markets.5The results of a market definition study can

United States v. Continental Can Co., 378 U.S. 441 (1964).

The

term

"industry"

is an

ambiguous

one,

as

noted

by

the

Supreme Court in its decision. [d. at 444. For instance, an industry could

be defined as a group of goods that share similar physical characteristics

and/or appearance, goods that have the same end use, or goods that are

manufactured using similar production methods or equipment. In the con-

text

of

U.S. v. Continental Can Co., the Supreme Court used the term

"industry" to refer to the similarity of production facilities and products.

[d. at 444 n.2. Groups of goods that are "dissimilar" in this respect serve,

in many instances, the same end use, and for ease of exposition, we will

continue

to use the term

"interindustry

competition"

to

describe

the

extent to which those instances are substantial enough that the prices of

the two industries restrain one another.

United States v. Continental Can Co., 217 F. Supp. 761 (1963).

Whether the merger affected potential competition with respect to

innovation was another issue, but one that is not addressed in this article.

ABA

ANTITRUST SECTION, MONOGRAPH No. 12, HORIZONTAL

MERG-

ERS:

LAW

AND POLICY 63 (1986).

Interindustry competition 53

influence

one's

conclusions about an industry dramatically, a con-

sequence well known to those familiar with the history and prac-

tice

of

antitrust. For example, in U.S.

v.

Continental

Can

Co., the

district court dismissed the government's complaint after it found

there was only one product line (containers for the beer industry)

in which the interindustry competition between metal and glass

containers was significant enough to include both industries in the

same market.v The Supreme Court subsequently reversed the dis-

trict

court's

decision (which ultimately led to the divestiture

of

Hazel-Atlas to another glass company) after it found the relevant

product market to include both glass and metal containers for all

end uses.

There have been significant innovations in market definition

analysis since U.S.

v.

Continental

Can

Co. was tried." There are,

at present, avariety of statistical and econometric methods avail-

able to the modem analyst. They include price correlation tests,"

tests

of

price uniformity and arbitrage," Granger causality tests,'?

The

district

court's

decision, supra note 3, was therefore not based

entirely on product market considerations alone. In the market defined as

"containers

for the beer industry," the district

court

dismissed the com-

plaint for the reason that it did not find sufficient evidence that the acqui-

sition led to a substantial lessening

of

competition.

Simons

&

Williams,

The Renaissance

of

Market Definition, 38

ANTITRUST

BULL.

799 (1993) describe how the analysis

of

market defini-

tion issues has

evolved

over

the decades.

See. e.g., Stigler &Sherwin, The Extent

of

the Market, 28 J. L. &

ECON.

555 (1985). Examples

of

how the methodology has been

applied

include

Cartwright,

Kamerschen,

&

Huang,

Price

Correlation

and

Granger Causality Tests

for

Market Definition, 4

REV.

INDUS.

ORGANIZA-

TION

79 (1989); and Doane &Spulber, Open Access and the Evolution

of

the U.S. Spot

Marketfor

Natural Gas, 37 J. L. &

ECON.

477 (1994).

See, e.g.,

Horowitz,

Market

Definition in

Antitrust

Analysis:

A

Regression-Based

Approach, 48 S.

ECON.

J. 1

(1981);

and

Spiller

&

Huang,

On the Extent

of

the Market: Wholesale Gasoline in the North-

eastern United States, 35 J.

INDUS.

ECON.

131 (1986).

10 See. e.g., Slade, Exogeneity Tests

of

Market Boundaries Applied

to Petroleum Products, 44 J.

INDUS.

ECON.

291 (1986); and Uri, Howell,

&

Rifkin,

On Defining

Geographic

Markets, 17

ApPLIED

ECON.

959