Review of Industrial Organization 19: 365–380, 2001.
© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
Measuring Market Power in the U.S. Aluminum
Industry: A Residual Demand Approach
Division of Business, Wayland Baptist University, Plainview, TX 79072-6998, U.S.A.
Abstract. This paper applies a reﬁned market power test to examine market power in the U.S.
aluminum industry. The study conducts a two-step analysis in which the inverse residual demand
facing the industry is ﬁrst estimated to generate the associate elasticity to infer market power, and
then the inverse partial residual demand facing the industry is estimated to produce the elasticity to
measure the potential effect of inter-industry coordination. The results suggest that only the primary
industry attains market power, and the market power can exert even without collusion between the
primary and secondary industries.
Key words: Market power, partial residual demand, primary aluminum, residual demand, secondary
The residual demand estimation has been a sound approach to the empirical eval-
uation of market performance over the recent decade. Landes and Posner (1981)
theoretically identify the relation in which a dominant ﬁrm’s market power can be
identiﬁed by its own market share adjusted with the market demand elasticity and
the fringe supply elasticity. The theory, which refers to that market power of a ﬁrm
or group of ﬁrms is determined largely by the demand curve faced by the ﬁrm or
group of ﬁrms, establishes the underpinning of residual demand estimation. Baker
and Bresnahan (1988), based on Landes and Posner’s theoretical treatment, derive
and estimate residual demand facing a single ﬁrm in an industry with differentiated
products. Baker and Bresnahan (1985) also extend the estimation to partial residual
demand facing a single ﬁrm to assess the effect of collusion or merger on market
This paper reﬁnes the residual demand analysis to estimate aggregate market
power of an industry producing homogeneous product with competitive substitutes.
That is, the analysis examines the degree of market power in the U.S. primary
This paper is a revised version of Chapter 3 of my Ph.D. Dissertation. I am indebted to David
Rosenbaum for his helpful direction. Comments on earlier versions by Matthew Cushing, Mary
McGarvey, two anonymous referees, and the editor of the Review of Industrial Organization are
gratefully acknowledged. I retain responsibility for any errors.