Commodity Bundling and the Leverage of Market Power

The modern literature on commodity bundling argues that a monopolist may use bundling to extend its monopoly power into an otherwise oligopolistic market by foreclosing entrants or excluding rivals. This approach is contrary to the Chicago critique of Posner and Bork, who defend bundling as irrelevant to such monopolization, being incapable of extending the “one monopoly rent.” However, this paper shows that the Chicago position may actually understates the market structure case in favor of bundling. I find that bundling can enable entry; banning bundling can make feasible the extension of monopoly power via entry deterring. I first investigate the basic two-good model of Nalebuff (2004) where demands are independently distributed, and potential entry occurs in either market, and argue that bundling is never relevant to entry deterrence if the entrant’s product is a perfect substitute for the incumbent’s. I then show how bundling might be used as an entryenabling device, if the incumbent and the potential entrant have differentiated products in one of the two markets. Finally I relate my analysis to the Microsoft case and to the genetically modified seed industry and discuss the relevant anti-trust policy implications. 1 If the monopolist were to earn a higher profit by selling a bundle in which the bundled product is also sold at marginal cost in a competitive market, it can make the same profits by pricing its monopolized product at the bundle price minus the marginal cost of the bundled product, without practicing bundling. Since the monopoly rent is the highest return the firm could get in the monopolized market, bundling is not leading to higher profits than the monopoly rent. The monopolist makes “only one monopoly rent”, therefore bundling cannot leverage market power from one market to another.