Private interest support for efficiency enhancing antitrust policies.
Buchanan, James M. ; Lee, Dwight R.
Recent discussions of antitrust based on the private interest
theory of government conclude that the real, as opposed to the stated,
purpose of antitrust legislation is to protect politically influential
industries against competition. Yet several prominent antitrust scholars
who accept the private interest theory of government in general see
antitrust legislation serving the public interest by increasing
competition. We argue that the private interest theory of government is
consistent with the view that antitrust legislation promotes
competition. Indeed antitrust legislation may be supported by organized
interest groups because such legislation increases the competition they
face.
I. INTRODUCTION
The traditional justification for the enactment and enforcement of
antitrust legislation is that it serves the public interest by
preventing businesses from engaging in practices that reduce
competition. Over the last decade some economists have become skeptical
of this justification. A small, but growing number of economists see the
public interest justification for antitrust enforcement as a pretense
for using government to promote the private interests of politically
organized groups.[1]
Within the private interest theory of government model, it is
natural to see antitrust activity as motivated by the desire to reduce,
rather than increase, competition. Certainly a private interest is
better served by obtaining an anticompetitive increase in the price of
the product being sold, than by obtaining a competitive reduction in the
price of one of the many products being bought.
Yet major contributors to, and strong supporters of, the private
interest theory of government are reluctant to embrace the private
interest explanation of, at least, the origins of antitrust policy.
George Stigler, for example, has said, "So far as I can tell,
it's [the Sherman Act] a public interest law ... I like the Sherman
Act.''2 Bork [1978, 61] has argued that the "legislative
history of the Sherman Act ..., displays the clear and exclusive policy
intention of promoting consumers welfare." And according to Posner
[1976, 23], "the framers of the Sherman Act appear to have been
concerned mainly with the price and output consequences of monopolies
and cartels."
With the view prevalent, even among those sympathetic with the
private interest theory of government, that antitrust legislation owes
much to the desire for a more competitive economy, it is not surprising
that antitrust policy has been resistant to private interest
explanations.3 Our purpose is to weaken this resistance by arguing that
the private interest theory of government is not necessarily
inconsistent with the view that support for antitrust activity is
motivated by the desire to increase the competitive efficiency of the
general economy. Indeed, we shall argue that the private interest theory
of government suggests that organized interests may favor antitrust
actions as a means of limiting their own anticompetitive practices.
It should be emphasized at the outset that we will not attempt to
establish that antitrust enforcement does increase economic efficiency.
Our more modest objective is to reconcile the seemingly inconsistent
views of those who both accept a private interest theory of government
and believe that economic efficiency is served by antitrust policy. The
difficulty of addressing the issue of the effect antitrust enforcement
has on the overall efficiency of the economy is reflected in
Stigler's [1982, 44] admission, "It would be gratifying to me
if I could report that our profession's changing view [toward
support of antitrust policy] was based upon the systematic study by
economists of the effects of the policy, in short, that hard evidence
carried the day. Unfortunately, there have been no persuasive studies of
the effects of the Sherman and Clayton Acts throughout this
century.''4
In the next section we present a simple model of a coalition of
industries, each of which is interested in achieving its own protection
in political competition with other industries. In section III, a
conflict among the individual members of the coalition, which is central
to our argument, is discussed. The possibility of moderating this
conflict with effective antitrust policy is considered in section IV.
Concluding comments are offered in section V. II. OPTIMAL OUTPUT
RESTRICTIONS FOR A
COALITION OF CARTELS
Each industry favors government policies that protect it against
competition. At the same time each industry prefers that the general
economy be as efficient, and therefore as competitive, as possible. No
industry is in a position, however, to obtain exclusive protection in
the political market for special interest privileges. Securing the
political support necessary for an industry cartel requires that the
industry enter into a coalition of other industries which seek their own
cartel restrictions. Although all industries want protection, and
therefore membership in the coalition, if the coalition is to be
effective at generating gains for its members, membership has to be less
than all inclusive.s
For the purpose of this paper the membership, and therefore, the
size, of the politically effective coalition under consideration will be
taken as given. In particular, it is assumed that, out of a total of N
industries in the economy, n industries (with n < N) have obtained
government protection allowing them to restrict output and raise price.6
The inverse demand function for each industry i in the coalition is
given by [P.sub.i ](Q.sub.i) where [Q.sub.i] is the output of industry
i, = 1,...,n. Let MCi be the average and marginal cost of production in
the ith industry. In order to simplify the analysis, we further assume
that the output restrictions in the protected industries do not affect
the consumers' surplus generated in the competitive industries.7
Among other things, this assumption rules out resource transfers from
the protected to the competitive industries as output restrictions occur
and implies that these restrictions result in increased leisure for all
employed inputs in protected industries.
It is further assumed that the economy is totally specialized in
production and totally non-specialized in consumption. All persons are
producers, with each supplying inputs to only one industry. Each person
consumes products of all industries, with each consuming an equal share
of the products of each industry. Therefore, the consumers' surplus
accruing to each industry in the coalition is directly proportional to
the number of persons in the industry. The number of persons in
coalition industry i is given by W,., with the number of persons all in
coalition industries given by
n The number of persons in all N industries in the economy is given
by W.
Those associated with each coalition industry expect to secure net
benefits from output restrictions in their own industry, but expect to
suffer losses as consumers from the output restrictions of other
industres in the coalition. The total value of the profits and
consumers' surplus realized by the coalition of industries is given
by where the first term represents total coalition profits, the second
term represents the share of the consumers' surplus generated by
the coalition industries that accrues to those in the coalition, and the
third term is the share of the consumers' surplus generated by the
competitive sector of the economy (given by CSC and independent of the
Qi) that accrues to those in the coalition.
The objective of the coalition is to restrict the Qi to those
levels, denoted by Q;, which maximize (1). This maximization necessarily
satisfies the conditions where MRi represents the marginal revenue of
industry i. The left-hand side of (2) represents the marginal loss of
profits to industry i from expanding output beyond the level called for
by the profit maximizing condition M[R.sub.i] = MC.sub.i]. The
expression Pi-M[R.sub.i.]. on the right-hand side of (2) equals the
marginal increase in the consumer surplus generated by industry i when
its output is expanded. Therefore (2) instructs each industry in the
coalition to expand output until the marginal cost in terms of lost
profits is equal to the coalition's share, We/W, of the marginal
gains in terms of increased consumers' surplus.
When the coalition includes all producers, [W.sub.c]= W, and (2)
collapses into the competitive condition Pi = MCi, i = 1,...,N. With an
all-encompassing coalition, there is no possibility of realizing
differential benefits from efficiency reducing restrictions, and so the
competitive solution is the best solution. At the other extreme, when
only one industry is cartelized, We is small relative to W and in the
limit condition (2) becomes MRi = MCi. If one industry were able to
obtain the political support necessary to cartelize without entering
into a coalition with other industries, and if that industry is very
small relative to the entire economy, then members of that industry
could ignore the loss of consumers' surplus resulting from their
own restrictions and the standard monopoly profit maximizing solution
would apply.
III. THE COALITION VS. THE CARTEL
From the perspective of the entire coalition, it is clearly
appropriate to consider the consumer surplus loss to all members of
coalition industries stemming from the output restrictions of each
industry. But from the perspective of the members of each industry, the
consumer surplus loss their own industry's restrictions imposes on
those in other industries is irrelevant. Therefore, the objective
function for members of, say, industry j is where [W.sub.j] is the
number of persons associated with industry j. The Qj that maximizes (3),
denoted by [Q.sub.j], must satisfy the condition since [P.sub.j-] MRj
> 0 when evaluated at any positive Qj. Therefore, since the
sufficient condition to the maximization of (3) requires that (4) be
decreasing in Q/, it follows that [Q.sub.j] < Q.sub.j]' The Qj
that maximizes the benefits to those associated with industry j, both as
producers and consumers, is less than the Qj that maximizes the benefits
to all those in the coalition of industries. What is true for industry j
is true for all industries in the coalition.
If the government protection against competition provided each
coalition industry determined within tight limits the output of that
industry, then the tension between what is best for the individual
industries and what is best for the entire coalition would be of little
consequence. Rather than be left out of the coalition of cartels, each
industry would be willing to accept the protection offered, even though
that protection fixed industry output at a level higher than the
industry would ideally like. But while government can provide an
industry with either more or less protection against competition, the
form the protection takes often leaves the protected industry
significant discretion regarding its output level.
General protections that are intended to allow a moderate level of
output restriction can create an environment which the protected
industry can exploit to achieve a further output restriction. For
example, import restrictions which are intended to reduce domestic
availability of an industry's product by reducing sales by foreign
producers may make it easier for the domestic industry to collude effectively for the purpose of reducing domestically available output
even further.8 So individual industries within the coalition of cartels
not only have the motivation to violate the coalition
"agreement" by reducing output excessively, they often will
have the ability as well. IV. ENFORCING THE COALITION AGREEMENT
WITH ANTITRUST
The members in the protected coalition are confronted with the
well-known prisoners' dilemma problem. The members are collectively
better off if all cooperate by passing up the opportunity to reduce
output below agreed upon levels. On the other hand, no matter whether
the others are expected to cooperate or not, from the perspective of
each coalition member the advantage lies in behaving noncooperatively by
reducing output below that level which is most advantageous to the
coalition.
Because of this tension between the interest of the coalition and
the apparent interests of its individual members, the net gain the
coalition can expect to realize from government protections is
influenced by the cost it incurs monitoring the behavior of the
industries in the coalition. In addition, for such monitoring to be
effective it has to be backed up with some means for imposing sanctions on those cartels which cheat. The advantages of being able to rely on
the government to support such policing are therefore twofold. First,
the cost of monitoring is reduced for the coalition if some of this cost
is assumed by the government. Second, without government support for
enforcement of the cartel agreement it would be difficult, if not
impossible, for the coalition to impose penalties on uncooperative
cartel members.
It is possible then to see why a coalition of cartels would favor
antitrust policy as a means of increasing the competitiveness of its
members even though the overriding purpose of the coalition is to obtain
government protections against competition. It is mistaken to argue that
antitrust policy is supported by special interests which desire to
reduce the competition that they face and then conclude that antitrust
policy must be supported by special interests because it reduces
competition.
We believe it is useful to think of antitrust enforcement in much
the same way economists have come to think of the owner of a firm who,
as the residual claimant, monitors the performance of the workers he
employs. It is not only the residual claimant boss who benefits from
this monitoring, but the workers as well. Although workers are better
off collectively if they all apply themselves diligently to insure the
viability of the firm, the temptation is for each to shirk and free ride
on the diligence of others. In order to overcome this temptation, to the
collective benefit, each worker is willing, indeed eager, to submit to a
boss who has an incentive to monitor the behavior of all. In fact, it is
reasonable to ask whether the boss hires the workers, or the workers
hire the boss.9
The ability of individual interest groups to exert some political
influence on their own leads us to a necessary qualification of our
thesis. We are not suggesting that antitrust policy was the creation of
identifiable industries which formed a coalition to exert unified
political pressure for protecting its members and then lobbied for
antitrust legislation as a means of monitoring its membership. An
interest group typically acts independently in its effort to obtain
political advantages. But in doing so the group invariably finds that
it, or its political representatives, has to go along with the special
interest requests of other organized groups if it is to obtain effective
support for its own requests. A coalition of special interests of the
type discussed in this paper tends to emerge from the logrolling that
characterizes the legislative process. Such a legislative coalition will
represent a loosely organized grouping, its membership will change over
time, and it will not likely act in concert to support a proposal that
in some general way advances the common interests of its members. Yet,
if the common interest of the coalition is advanced by legislation
enacted in response to other special interest activity, the coalition
can be thought of as implicitly supporting that legislative by accepting
it, and accommodating to it. It is in this sense that we argue that it
is useful to conceive of our coalition of cartels, and of its support
for antitrust legislation and enforcement.
V. CONCLUSION
In the literature on the private interest theory of government,
antitrust policy has remained something of a puzzle. The private
interest theory would seem to imply that political support for antitrust
legislation is derived primarily from its ability to protect the
politically influential against competition. And there have been some
who have embraced this anticompetitive view of antitrust legislation and
supported it by pointing to instances in which antitrust enforcement has
reduced the competition faced by organized interests. But, antitrust
enforcement continues to be widely supported by economists as an
effective means of increasing competition in the economy. And indeed,
even some leading proponents of the private interest theory of
government, most notably George Stigler, have supported the view that
antitrust laws promote the public interest by enhancing competition.
In this paper we present a private interest model of antitrust
policy that is consistent with the view that, on balance, such policy
enhances efficiency by promoting competition. In our model, as in all
private interest models, organized groups use the government to promote
their narrow advantages rather than to advance broad public objectives.
Organized industry groups are politically motivated by the desire to
obtain protection against the competition they face rather than to
increase the efficiency of the general economy. But in order to succeed
in obtaining restrictions on the competition it faces, an industry must
be willing to accept restrictions on the competition other industries
face. This leads to a coalition of protected industries with each
industry preferring more protection for itself than that which is
considered optimal from the perspective of the overall coalition. The
temptation, then, is for each industry in the coalition to reduce the
competition it faces by more than is in the best interest of the other
coalition industries. The advantage to the coalition of being able to
monitor these anticompetitive practices when they go too far is obvious.
Equally obvious is the advantage to the coalition of shifting the cost
of this monitoring and enforcement to government.
In sum, the view (assumption) that antitrust legislation enhances
economic efficiency is consistent with a private interest explanation of
the political support for that legislation. Effective antitrust
legislators can serve the private interests of a coalition of protected
industries by making these industries more competitive than they would
be in the absence of antitrust laws.
REFERENCES Alchian, Armen A. and Harold Demsetz. "Production,
Information, and Economic Organization.' Amer - ican Economic
Review, December 1972, 777-95. Armentano, Dominick T. Antitrust and
Monopoly. New
York: Wiley, 1982. Baumol, William J., and Janusz A. Ordover.
'Use of
Antitrust to Subvert Competition." The Journal of
Law and Economics, May 1985, 247-65. Benson, Bruce L., M.L.
Greenhut, and Randall G.
Holcombe. 'interest Groups and the Antitrust
Paradox.' The Cato Journal, Winter 1987, 801-17. Bork, Robert
H. The Antitrust Paradox: A Policy at War
with Itself.. New York: Basic Books, Inc., 1978. Brozen, Yale,
Concentration, Mergers, and Public Policy.
New York: MacMillan, 1982. Buchanan, James M., and Dwight R. Lee.
"Cartels, Co - alitions, and Constitutional Politics.'
Constitu - tional Political Economy, Spring/Summer 1991,
139-61. DiLorenzo, Thomas J. "The Origins of Antitrust: An
Interest Group Perspective." International Review
of Law and Economics 5(1), 1985, 73-90. Faith, Roger L., Donald R.
Leavens, and Robert D.
Tollison. "Antitrust Pork Barrel." The Journal of
Law and Economics, October 1982, 329-42. Grady, Mark F.
"Toward a Positive Economic Theory
of Antitrust:. Return to the Sea of Doubt." North - western
Law School, 1990. Hazlett, Thomas. "Reason Interview with George
Stigler." Reason Magazine, January 1984, 44-48. Posner,
Richard A. Antitrust Law: An Economic Perspec - rive. Chicago:
University of Chicago Press, 1976. Stigler, George J. The Economist as
Preacher & Other Es - says. Chicago: University of Chicago Press,
1982.
. "The Origins of the Sherman Act.' The Journal
of Legal Studies, January 1985, 1-12.