General Equilibrium in Vertical Market Structures: Overselling Versus Overbuying

• Pages: 149-181
• DOI: https://doi.org/10.1016/S0193-5895(07)23007-6
• Date: 16 October 2007
• Published date: 16 October 2007
• Author: Richard E. Just,Gordon C. Rausser

GENERAL EQUILIBRIUM IN

VERTICAL MARKET STRUCTURES:

OVERSELLING VERSUS

OVERBUYING

Richard E. Just and Gordon C. Rausser

ABSTRACT

The lens used by the courts and much of the antitrust literature on

predatory selling and/or buying is based on partial equilibrium

methodology. We demonstrate that such methodology is unreliable for

assessments of predatory monopoly or monopsony conduct. In contrast to

the typical two-stage dynamic analysis involving a predation period

followed by a recoupment period, we advance a general equilibrium

analysis that demonstrates the critical role of related industries and

markets. Substitutability versus complementarity of both inputs and

outputs is critical. With either monopolistic or monopsonistic market

power (but not both), neither predatory overselling nor predatory

overbuying is profitably sustainable. Two-stage predation/recoupment is

profitable only with irreversibility in production and cost functions, unlike

typical estimated forms from the production economic literature.

However, when the market structure admits both monopolistic and

monopsonistic behavior, predatory overbuying can be profitably sustain-

able while overselling cannot. Useful distinctions are drawn between

contract versus non-contract markets for input markets.

Research in Law and Economics, Volume 23, 149–181

Copyright r2007 by Elsevier Ltd.

All rights of reproduction in any form reserved

ISSN: 0193-5895/doi:10.1016/S0193-5895(07)23007-6

149

INTRODUCTION

Predatory selling has been evaluated and assessed by antitrust regulators,

the courts, and the economics profession.

1

Recently, the spotlight has turned

to alleged predatory buying.

2

The criteria for determining in output markets

whether monopolists or oligopolists are engaged in predatory actions has

been debated and various criteria have been expressed both by courts and

professional economists. In the case of monopsonists or oligopolists as

buyers in input markets, many have argued that the same criteria used to

evaluate predatory selling should also hold for predatory buying.

3

The economic literature has long focused the evaluation of predatory

conduct on the trade-off between a predator’s short run losses and the

benefits that might be achieved after its prey is harmed (Telser, 1966; Joskow

& Klevorick, 1979; Easterbrook, 1981; Elzinga & Mills, 1989, 1994; McGee,

1980; Milgrom & Roberts, 1982; Scherer, 1976; Williamson, 1977). The

short run losses suffered by the predator are viewed as an investment

incurred that is designed to discipline or eliminate its rivals. This investment

is presumed to be motivated by monopoly or monopsony rent seeking.

Accordingly, in this two-stage view, the rents or benefits accruing to

predatory actions can only be rationalized during some recoupment period

as clearly stated by Elzinga and Mills (1994, p. 560):

In simplest terms, conventional predation occurs in two stages. In the first stage the

predator prices at nonrenumerative levels to drive rivals or an entrant from the market

or to coerce rivals to cede price leadership to the predator. In the second stage the

predator flexes its monopolistic muscles by charging supracompetitive prices and

recouping the losses sustained during the initial stage.

Given the actual availability of data for the first stage, the original focus

of both economists and the courts was on the question of measuring losses

that occurred during an alleged predatory period. These losses are viewed as

a necessary investment to achieve monopoly rents. The measurement of

such losses was initially based on the cost benchmark of Areeda and Turner

(1975). This benchmark was advanced as a means to separate potential

predatory conduct from vigorous competition. As Areeda and Turner

note (1975, p. 712), ‘‘a monopolist pricing below marginal cost should be

presumed to have engaged in a predatory or exclusionary practice.’’ Given

the difficulty of measuring marginal cost, the operational Areeda and

Turner test substitutes average variable cost. Under this criterion, short run

losses are thus measured as prices unfolding over a predatory period that are

below average variable cost.

4

In essence, whenever a firm fails the cost-based

RICHARD E. JUST AND GORDON C. RAUSSER150

test of Areeda and Turner, it bears the burden of demonstrating that its

pricing was not predatory.

A complimentary test for predation has been offered by Elzinga and Mills

(1989, 1994). This test recognizes that costs are difficult to measure either as

marginal or average variable costs. They instead focus on the second stage,

introducing as their benchmark the long run competitive price in the

industry.

5

As a result, the Elzinga–Mills test allows for prices to be above

average variable cost but still, in certain circumstances, predatory. Under

the Elzinga–Mills test, an analysis of the recoupment period as well as the

predatory period is required. As argued by Elzinga and Mills (1989, p. 871),

‘‘if a predatory strategy is an economically implausible investment, as

judged by the parameters of the recoupment plan, it implies then the alleged

predator is exonerated.’’ This test can only be executed if all of the following

are determined: (1) the period of time covering the predatory period, (2) the

period of time covering the recoupment period, (3) the long run ‘‘but-for’’ or

competitive price, (4) the weighted-average cost of capital of the predator,

(5) the discount rate required to make returns during the predatory and

recoupment periods comparable, (6) a complete structural model including

demand and the supply of the firm’s rivals, and (7) the prices charged both

during the predatory and future recoupment periods. This test, as well as the

Areeda–Turner test, is implemented in a partial equilibrium framework.

Beginning in the early 1980s, the courts recognized the recoupment

standard culminating with Brooke Group Ltd. v. Brown and Williamson

Tobacco Cor (1993). In this and other Supreme Court decisions, concern has

been expressed about false positives, viz, finding a company liable for

predatory conduct when it is actually engaged in vigorous competition.

6

This ruling found that suppliers in output markets are not predatory sellers

unless the prices charged are below the seller’s cost and, additionally, the

seller has a ‘‘dangerous probability’’ of recouping its lost profits once it has

driven its competitors from the market. To be sure, the courts have

determined that ‘‘recoupment is the ultimate objective of an unlawful

predatory pricing scheme: it is the means by which a predator profits from

predation’’ (Brooke Group, 1993). In this ruling, the Supreme Court cited

several factors that must be assessed to determine whether an alleged

predator can expect to recoup its predatory losses. These factors include: (1)

the length of the predation period, (2) the extent to which the predator’s

prices are below cost, (3) the comparative financial strength of the predator

versus target firms, (4) the ‘‘incentives and will’’ of predator and prey, (5) the

size distribution of firms in the relevant market, (6) entry conditions in the

relevant market, and (7) the predator’s ability to absorb the output of target

General Equilibrium in Vertical Market Structures 151