in achieving antitrust goals, to attain efliciencies in production and distribution.

Antitiiist law is composed of two very different theories of how competition
may be injured — how. through misdirection of re.sources, the output of what
consumers want most is restricted. The first theory holds that competiti(ni may
be injured, and resources misallocated so as to reduce the real wealth of the
community, by the elimination of competition among consenting rivals. This is
the theoi-j' uixin which the law against cartels and horizontal mergers is based.
Though sometimes ditticult to apply in merger cases because of inability to pre-
dict whether the output-restricting effect of the merger, occasioned i>y the fewer
number of competitiors. will be outv.-eighed by the output-expanding effect, occa-
sioned by more eflicient production and distribution, it is appropriate to stress
the very real danger that competition can be injured liy horizontal merger in
the same manner as it can l.e by cartels when llie merging companies achieve con-
trol of high proportions of sales in a market.

A second theory of how competition may be injured, the (me relevant in the
assessment of vertical arrangements, holds that any substantial adverse effei-t
upon competitors' access to customers or suppliers can be harmful to competition.
The foreclosing of rivals from those sources of supjily or markets which absent
the vertical arrangement might conceivably be available lo them, this theory
holds, will make for fewer actual or potential market larticipants thereby mak-



287

iug competition less effective than would be the case if the vertical arrangements
were proscribed. Thus, when a "dominant"' firm acquires ownership of retail out-
lets (as in the Broun Shoe case, or as in the recent action against Hart, Shaffner
and Marx) or, similarly, when such a firm makes a long-term exclusive dealing
contract with its outlets (the Anchor Serum and Dictograph cases provide ex-
amples) an "exclusionary tactics" theory is applied. The theory seems to be that
"equally or better qualified"' competitors are "arbitrarily" prevented from access
to the "better outlets" thus maintaining the market position of the seller who
integrates by merger or contract ; or in its more extreme version, that vertical
integration is a means of leveraging the integrating firm into new or greater
monopoly.

"The theory of exclusionary tactics underlying the law" as Professor Bork and
I have elsewhere stressed, "appears to be that firm X which already has ten
percent of the market can sign up more than ten percent of the retailers perhaps
twenty jiercent and by thus 'foreclosing' rivals from retail outlets obtain a larger
share of the market. But one must then ask why so many retailers are willing to
limit themselves to selling X's product. Why do not ninety percent of them turn
to X's rivals? Because X has greater market acceptance? But then X's share of
the market would grow for that reason and the requirements contract would
have nothing to do with it. Because X offers them some extra inducement? But
that sounds like competition. It is equivalent to a price cut and surely X"s com-
petitors can be relied upon to meet competition.

"The theory of exclusionai*y pi^actices. here exemplified in the use of require-
ments contracts, is in need of one or two additional assumptions to be theoretically
plausible. One is the assumption that there are practices by which a competitor
can imix>se greater costs uix)n his rivals than upon himself. That would mean
that X could somehow make it more expensive for his rivals to sign retailers to
requirements contracts than it is for X to do so. It would be as though X could
offer a retailer a one dollar price reduction and it would cost any rival two dol-
lars to match the offer. . . .'"

"Thp other assumption upon which the theory of exclusionary practices might
re.st is that there arc imperfections in or difficulties of access to the capital market
that enable X to offer a one dollar inducement {it has a bankroll) and prevent its
rivals from responding (thej' have no bankroll and, though the offering of the
inducement is a responsible business tactic, for some reason they cannot borrow
the money). But it is yet to be demonstrated that imperfections of this type exist
in the capital market."

The exclusionary practices theory — foreclosure — has been applied with vary-
ing but increasing rigor in a wide variety of contexts under Sections 3 and 7 of
the Clayton Act, the Robinson-Patman Act, the Federal Trade Commission Act,
the Millard-Tydings and McGuire Acts, as well as under the Sherman Act. The
forms of vertical integrations embraced include, in addition to merging with
suppliers or outlets, exclusive dealing contracts, requirement contracts, territorial
allocations, franchise arrangements, non-collusive resale price maintenance, dis-
criminatory pricing contracts, tie-in sales, and full-line-forcing contracts among
others.

The defects in the foreclosure theory are equally applicable to all the forms
in which it is manifested, and not met in the cases in which it is applied. The
Antitrust Division, the Federal Trade Commission and the Courts have not
faced up to demonstrating how any of these forms of vertical business relation-
ships can impose higher costs upon rivals.

I do not state that there can be no vertical arrangement which will be injur-
ious to competition, nor that all practices currenly judged exclusionary should
be per se lawful. I do urge strongly, however, that standards of assessments be
realigned with the basic goal of antitrust — providing society with the maximum
output that can be achieved with the resources at is command. Competition serves
this end. Protecting competitors from more efficient rivals (aggressive or not) is
not protecting competition. Both the prosecuting agencies and the courts increas-
ingly treat vertical contracts either as conclusively illegal or so presumably illegal
that relevant and appropriate aspects of efficiency are ignored even when occas-
ionally such economic evidence is presented or admitted.

Specifically it is recommended that no vertical arrangement be treated as per
se illegal. PresmnaMy legal would be a far better rule. And in applying the es-
sentially implausible "incipient monopoly" hypothesis contained in the Clayton
and Robinson-Patman Acts (and increasingly in the Sherman Act by judicial
adoption) by which certain business practices, including a variety of vertical



288

arrangements, are to be illegal where their effect ma^ be suhstantially to lessen
competition or tend to create a monopoly in any line of commerce, it is recom-
mended that both prosecution agencies and courts apply reasoned explanation
of how this result is supposed to be achieved. If any such practice may have the
tendency which the si)ecific language of theses tatutes calls for, then it should be
prerequisite to finding illegality that it be explained how and why the activities
complained of are more rather than less likely to restrict rather than to expand
the output of the goods and services involved.



Working Paper for the Task Force on Productivity and Competition :
Advertising and Product Differentiation

(By Richard Posner)

"Product differentiation" is the phenomenon of purchasers' distinguishing
among different sellers or brands of the same product. Some consumers prefer
Bayer to other brands of aspirin ; some construction companies prefer Euclid to
other manufacturers of earth-moving equipment. Product differentiation, which
manifests itself in consumer loyalty, is associated with product differences, trade-
marks, differential reputations of sellers for reliability, promptness, answer-
ability, etc. — and with advertising. Of late, product differentiation of consumer
goods has moved to the center stage of antitrust concern :

(1) The attack on franchising in the Schtvinn case was premised in major part,
especially at the Supreme Court level, on the argument that franchising contrib-
utes to product differentiation.

(2) A recent proposal by Donald Turner to limit the advertising expenditures
of firms convicted of antitrust violations is bottomed on the idea that advertising
contributes to market power.

(3) In the Pabst case and again in the Merger Guidelines the Justice Depart-
ment declared that in a case involving an advertised product (like beer) , members
of the industry who do not sell in a local market will not be considered a part of
that market. Even if transportation costs do not preclude their selling there, they
would have to overcome the allegiance of the consumer to the established brands.

(4) The Clorox decision holds that a merger that enhances the ability of the
resulting firm to advertise violates antitrust principles, on the theory that
product differentiation serves to entrench the dominant firms in an oligopoly.

(5) The Supreme Court held in Clorox and the Department intimates in the
Guidelines that economies of scale in advertising or promotion will never be ac-
cepted as a justification for an otherwise unlawful merger, although production
economies just might.

These examples could be multiplied, but they will sufiice to show the pervasive
concern of the antitrust agencies with advertising and product differentiation.

Before analyzing the agencies' concern, a word about the role of product
differentiation and advertising in a complex modern economy. No one objects,
surely, when a seller improves his product or earns a reputation for reliability
that distinguishes him from his rivals. These are important and salutary forms
of competition, no less worthy than price competition among sellers of undif-
ferentiated products. Concern creeps in only when it is suspected that product
differentiation is created or entrenched by advertisiing. We shall discuss in a
moment the criticisms of advertising ; here let us just remind that advertising
plays an indisi>ensable social role. A modern economy requires the generation
of a vaist amount of information on the identity and location of sellers, on types
of and changes in product and on prices and other terms of sale. It is not
surprising, therefore — ^and certainly not to be deplored — that there is a vast
amount of advertising : a practical alternative, not involving economic stagna-
tion, is not immediately evident. Nor is it surprising or plorable that there is a
good deal of repetitiousness in advei-tisiug, for changes in the identity of buyers
and sellers, the constant influx of new consumers, forgetfulness, and frequent
changes in products, make it imperative that the advertiser repeat his message
over and over again. Nor, finally, is it sairprising, or in a democratic and
egalitarian isociety deplorable, that advertising is frequently vulgar by the
standards of intellectuals: intellectuals ai"e a small minority of the consuming
population, and it would be cultural tyranny were their tastes to dominate
advertL^ing directed at the mass buying public.



289

With this as backgr'ound, let us coni^ider the ar^iments of those who believe
that advertising may be inimical to the objectives of the antitrust laws :

( 1 ) Advertising, except so much as is necessary to provide the consumer with
"essential" information, is said to waste resources; anything that may lead to
an increase in advertising beyond some minor useful level is, therefore, un-
desirable. This assumes, however, that we can distinguish "information" from
"l)ersuasive" advertising, and draw a line above which advertising contributes
less to the consumer's ability to make rapid and satisfactory choices than it costs
to advertise. And it also assumes that consumers are so foolish as to be willing to
pay more for advertising (in a higher price for advertised brands) than its
value to them in helping them make choices. Both assumptions are highly
dubious on their face.

(2) Advertising is said to distort consumer choice, to make the consumer buy
many things he doesn't really want. This "brainwashing" theory would be more
plausible if there were a monopoly on advertising. In fact, advertisers compete
for the consumer's patronage. One would exi>ect the best products to win out in
competition among advertisers, just as the market in ideas, a market also
characterized by inflated claims, is assumed to lead to the adoption of the best
ideas. Why individuals can be trusted to make intelligent political choices, but
not intelligent product choices, is not explained.

(3) Advertising is said to be an important factor in the diminished rivalry that
is thought to characterize many oligopolistic markets. The reasoning is that
advertising creates brand loyalties that rival sellers find very difficult to erode
and that this is a source of formidable barriers to new entry into concentrated
markets. Thus, the argument runs, if it costs Proctor & Gamble, Colgate-
Palmolive, and Lever Brothers 2 cents to sell a bar of soap, because they got
there first, it would cost a new entrant (say) 2.2 cents. The established firms,
therefore, can price up to 2.19 cents (if they collude, tacitly or expressly) without
attracting new entry, and thereby realize a monopoly profit of 19 cents.

Apart from the question whether collusion by oligopolists is as routinely com-
monplace as the argument assumes, no proof has yet been offered that it is easier
for the first advertiser to win a consumer's patronage than it is for a second
advertiser to iShift it to him. The fact that the soap companies are constantly
bringing out new brands suggests a taste for novelty on the part of the consumer
that does not square with the theory of the first adverti-ser's advantage. To be
sure, advertising has cumulative effects and that will give an established firm
an advantage over a newcomer. But how is that different from the advantage an
established Ann has by virtue of an experienced organization, customer contacts,,
a plant that has been paid for, etc.? ^Moreover, a new entrant will often have
access to a ready fund of accumulated consumer goodwill : by distributing throiigh
and under the trademarks of one of the established retail chains such as Sears
Roebuck.

The theory that advertising creates barriers to entry is said to be confirmed
by, but receives at most partial support from, statistical studies showing a cor-
relation between the amount of money an industry .spends on advertising and the
industry's profit rate. The studies are far from conclusive, in part because fol-
lowing normal accounting practice they treat advertising expenditures as current
expenses of the year in which incurred, rather than as a capital investment whose
effects persist into subsequent years. The rate of return in industries that
advertise heavily tends, in consequence, to be overstated.

(4) Finally, it argued that oligopolists devote excessive resources to adver-
tising as an alternative mode of rivalry to price cutting, a tactic they are
said to eschew becaiise they know it \\ill lead to lower prices and profits for all
sellers in the market. However, if oligopolists spend more on advertising or
otherwise differentiating their brands than the consumer deems warranted,
attractive opportunities for lower-priced off-brand or distributor-brand sub-
stitutes will be created. Moreover, if oligopolists were able tacitly to collude to
avoid price competition, would they not al.so collude to limit selling expenses
that would equally erode their monopoly profits? Since advertising is public, an
agreement limiting it to a specified percentage of each firm's sales would be easy
to enforce.

My point is that on the basis of present kno^'\â– ledge advertising seems essentially
symmetrical with other competitive business tactics such as raising quality,
reducing production cost.s, and cutting price. It is difficult to resist the suspicion
that the hostility to advertising derives more from concern with the level of public
taste or culture than from concern with competition and efBciency.



White House Task Force Report on Antitrust Policy

(Released May 21, 1969)

Pbil C. Neal, Chairman James W. McKie

William F. Baxter Lee E. Preston

Robert H. Bork James A. Rahl

Carl H. Fnlda Geoi-ge D. Reycraft

William K. Jones Richard E. Sherwood

Dennis G. Lyons S. Paul Posner, Staff Director
Paul W. MacAvoy

The University of Chicago,

The Law School,
Chicago, III, July 5, 1968.
The President,
The White House,
Washington, B.C.

Dear Mr. President : On behalf of the Task Force on Antitrust Policy, I have
the honor to submit our Report, together with additional statements represent-
ing the separate views of several members on portions of the Report.

The Task Force was appointed in December, 1967, and was asked to report by
approximately June 30th of this year. In accordance with the broad terms of
reference given us, we have undertaken to identify the most important areas in
which antitrust policy might be strengthened by new legislative or adminis-
trative measures.

We have made a number of recommendations that we believe would, if
adopted, improve the effectiveness of the antitrust laws. In the time and with
the resources available to us it has not been possible to examine all antitrust
problems that merit attention or to conduct any significant new research. Our
recommendations are based upon available studies, a substantial body of in-
formed economic opinion, and our own background of study or experience in the
antitrust field.

Our principal recommendations deal with concentrated industries, conglomer-
ate mergers, the Robinson-Patman Act, certain aspects of patent licensing, and
the improvement of economic data relevant to antitrust matters. We have not
dealt specially with the drug industry, an item mentioned in the letter of appoint-
ment of the Task Force, but we believe that the changes recommended by us
in the patent field would have significant beneficial effects in that industry.

Although we were not asked to propose specific legislation, we concluded that
our recommendations would be most useful if subjected to the discipline of fram-
ing concrete legal principles and if submitted in that form. Accordingly our re-
port is accompanied by a number of drafts of proposed statutory provisions giving
effect to our recommendations.

I should like to take this opportunity to express my personal appreciation for
the privilege of taking part in the deliberations of this group. The Task Force
consisted of three practising lawyers, three economists, and five professors of
law, in addition to the chairman. Each member contributed substantially to the
form and substance of our recommendations and the resulting Report is very
much a joint product of the Task Force. Special commendation is due our Staff
Director, Mr. S. Paul Posner, whose exceptionally able help contributed greatly
to the work of the Task Force.

We hope the recommendations made in this Report will be useful to you and
that our proposals for new laws may find their way into legislative proposals
under this or a succeeding Administration.
Re.spectfully yours,

Phil C. Neal.

(291)



292
CONTENTS

Summary.

I. Introduction.
II. Oligopoly, or concentration in particular markets.

III. Conglomerates, or large diversified firms.

IV. The Robinson-Patman Act.
V. The Patent Laws.

VI. Problems of information.

VII. Additional recommendations.
Appendix A :

Concentrated Industries Act.

Comments to accompany Concentrated Industries Act.
Appendix B :

Merger Act.

Comments to accompany Merger Act.
Appendix C. Proposed revision of Robinson-Patman Act.
Appendix D :

Proposed patent legislation.

Comments to accompany proposed patent legislation.
Appendix B. Additional legislation.
Appendix F. Glossary.
Separate statement of Robert H. Bork.
Separate statement of Paul W. MacAvoy.
Separate statement of Richard E. Sherwood.

SUMMARY

1. We recommend specific legislaion on the su'bject of oligopolies, or highly
concentrated industries.

The purpose of such legislation would be to give enforcement axithorities and
covirts a clear mandate to use established techniques of divestiture to reduce
concentration in industries where monopoly power is shared by a few very large
firms. Up to now such measures have been employed only in the rare instances
where the monopolistic structure of an industry takes the form of a single firm
with an overwhelming share of the market. Specific legislation dealing with
entrenched oligopolies would rectify the most important deficiency in the present
antitrust laws.

Effective antitrust laws must bring about both competitive behavior and com-
petitive industry structure. In the long run, competitive strucure is the more
important since it creates conditions conducive to competitive behavior. Com-
petitive structure and behavior are both essential to the basic concern of the
antitrust laws — preservation of the self-regulating mechanism of the market,
free from the restraints of private monopoly power on the one hand and govern-
ment interventon or regulation on the other. In one important respect, the anti-
trust laws recognize the necessity for competitive market structures : the 1950
amendment to section 7 of the Clayton Act has effectively prevented many kinds
of mergers which would bring about less competitive market structures. Our
proposed remedy, which would deal with existing noncompetitive market struc-
tures, is a necessary complement to section 7.

Highly concentrated industries represent a significant segment of the Amer-
ican economy. Industries in which four or fewer firms account for more than 70%
of output produce nearly 10% of the total value of manufactured products ; indus-
tries in which four or fewer firms account for more than 50% of output produce
nearly 24%. An impressive body of economic opionion and analysis supports the
judgment that this degree of concentration precludes effective market com-
petition and interferes with the optimum use of economic resources. Pa.st exper-
ience strongly suggests that, in the absence of direct action, concentration is not
likely to decline significantl.v.

While new legal approaches might be developed to reduce concentration under
existing law — a result which should be encouraged — the history of antitrust
enforcement and judicial interpretation do not justify primarily reliance on this
possibility. For this reason, we recommend a specific legislative remedy directed
to the reduction of concentration. Our proposed Concentrated Industries Act,
wliich appears in Appendix A to the Report, establi-s^hes criteria and procedures
for the effective reduction of industrial concentration.

2. We recommend, additional legislation prohibiting mergers in irhich a very
large firm acquires one of the leading firms in a concentrated industry.



293

This legislation would supplement section 7 of the Clayton Act, which prohibits
mergers which may tend substantially to lessen competition. The primary impact
of the new legislation would be on diversification or "conglomerate" mergers.
Under section 7 of the Clayton Act, such mergers may be prevented if adverse
effects on competition can be anticipated. But the dection of such effects fre-
quently depends on factual and theoretical judgments that are highly speculative.
As a result, some mergers with potentially adverse effects on competition may
escape attack and mergers which will not harm competition will be prohibited
because the effects cannot readily be predicted. Because of the inherent limitations
of the competitive standard of section 7, the recently published Merger Guide-
lines do little to resolve these difficulties.

Our proposed legislation would prevent some possibly anticompetitive mergers
which might have gone uncliallenged because of the difficulty of applying section 7
standards, and thus would act as an effective supplement to existing policy.
In addition, the proposed legislation would have affirmative aspects in channel-
ing merger activity into directions likely to increase competition. If large firms
are prevented from acquring leading firms in concentrated industries, they will
seek other outlets for expansion which may be more likely to increase competition
and decrease concentration.

This policy of deflecting conglomerate mergers into desirable channels is
preferable to any rule that would limit mergers without regard to consideration
of market structure. Although the number of conglomerate mergers has increased
sharply in recent years, there is only a moderate tendency toward increase in