Review of Industrial Organization (2005) 27:107–124 © Springer 2005
The Collusive Equilibrium in a Quantity-Setting
Supergame: An Application to Taiwan’s
Department of Finance, National Kaohsiung University of Applied Sciences, 415 Chien
Kung Road, Kaohsiung 807, Taiwan
Abstract. This paper examines the stability of the collusive equilibrium in Friedman’s
repeated game through an empirical study of Taiwan’s ﬂour market. We calculate the pay-
off streams following a deviation or adherence for each ﬁrm. The evidence shows that the
speciﬁed punishment path is credible, and could sustain the collusive allocation.
Key words: collusive equilibrium, excess capacity, repeated game.
JEL Classiﬁcations: L13.
The “New Empirical Industrial Organization” school has long been con-
cerned with measuring the degree of market power in markets and under-
standing its fundamental determinants. A signiﬁcant portion of the literature
employs an empirical model based on the theory of conjectural variation to
estimate the conduct parameter.
This parameter is utilized to measure the
degree of competition in a quite general way, nesting the perfectly competi-
tive, monopolistic, and Cournot models.
Corts (1999) examines this conjectural variation method and argues
that the static parameter does not accurately measure market power in a
dynamic oligopoly model. As for Corts’ argument, there still exist some
possibilities for the estimated conjectural variation accurately to measure
the market power. For instance, if the collusion regime is sustained in
equilibrium in every period, then the estimated parameter could correctly
measure the market power. However, Bresnahan (1989) reviews some car-
tel cases and argues that the cartel should be expected to fall apart and
See Bresnahan (1989) for the survey of the literature.