Review of Industrial Organization 12: 219–230, 1997.
1997 Kluwer Academic Publishers. Printed in the Netherlands.
Speciﬁcation and Testing the Proﬁt-Concentration
Relationship in Australian Manufacturing
MITA BHATTACHARYA and HARRY BLOCH
Department of Economics, University of Tasmania, Hobart, Australia 7001
Abstract. Research in the mainstream of industrial organization has tested the relationship between
proﬁt rates (an index of performance) and concentration (an index of structure) including other
variables (e.g., capital intensity, advertising intensity, growth, measures for barriers to entry, import
and export intensity). Speciﬁcation of this relationship is often largelyad hoc and its testing is subject
to a number of statistical criticisms. Major criticisms that require attention are: i) omission of the
relevant explanatory variables, ii) simultaneous causality among variables, and iii) measurement error
in the variables. This paper derives a proﬁt-concentration relationship from a well known oligopoly
model. Empirical analysis is carried out against a sample of Australian manufacturing industries for
1984–85. The resulting estimates suggest the importance of dealing with each of the speciﬁcation
and testing issues in explaining the proﬁt-concentration relationship.
Keywords: Oligopoly, price-cost margin, concentration, proﬁtability.
JEL Classiﬁcation: D43, L13, L60.
The structure-conduct-performance (SCP) paradigm has played an important role
in industrial organization research since the pioneering work of Mason (1939). The
theory predicts that market structure determines the conduct (strategy) of ﬁrms and
this in turn determines industry performance. Empirical research during the last
few decades has provided useful insights into the relationship between proﬁtability
(an index of performance) and various industry structural characteristics (see, for
example the surveys in Scherer and Ross (1990, Ch. 11) and Schmalensee (1989).
The relationship between proﬁt and concentration is the most widely examined
hypothesis in the SCP paradigm. Almost all studies ﬁnd a positive (although some-
times weak) association between these two variables. However, in Australia, the
hypothesis of a signiﬁcant positive association between proﬁts and concentration
is rejected in most studies. This paper develops a proﬁt-concentration relation-
The authors are grateful to John Stanton, Glenn Otto and anonymous referees of this journal for
useful suggestions. Helpful comments from participants of 1995 Industry Economics Conference at
the University of Melbourne and 1995 Ph.D. Conference in Economics and Business at the University
of Western Australia are also appreciated.