The chain-store paradox revisited (Q1084033)

The rationality of predatory practices has been extensively debated among economists and lawyers. \textit{R. Selten} [ibid. 9, 127-159 (1978; Zbl 0374.90085)] considered a fictitious chain-store confronted with potential entrances of local competitors. In his formal analysis via an extensive game with complete and perfect information predatory behavior is precluded by the unique sequential (and perfect) equilibrium. \textit{D. M. Kreps} and \textit{R. Wilson} [J. Econ. Theory 27, 253-279 (1982; Zbl 0485.90093)] and \textit{P. Milgrom} and \textit{J. Roberts} [ibid. 27, 280-312 (1982; Zbl 0485.90013)] established in modified models with incomplete information that predation against early entrants may be rational since it creates a reputation to the effect that later potential entrants are deterred. The present paper offers a modification of Selten's model with complete but imperfect information which renders possible reputation and deterrence.