Get access



  • *Evans is a Managing Director of LECG, LLC and Executive Director of the Jevons Institute for Competion Law and Economics at University College London where he is also a visiting professor. Salinger is an economics professor at the Boston University School of Management. We would like to thank Chris Nosko for exceptional research support and Howard Chang, Gregroy Crawford, Pascal Courty, Mark Frankena, Lubomira Ivanova, Anne Layne-Farrar, Albert Nichols, Emmanuel Petrakis, Bernard Reddy, William Rogerson, Mark Schankerman and Richard Schmalensee, as well as the Editor and referees for comments and suggestions. An earlier version of this paper was presented at the Industral Organization Society meetings in Chicago in April, 2004 and the CEPR Conference on Applied Industrial Organization in Hydra, also in April, 2004. Microsoft Corporation provided research funding for which the authors are grateful. The opinions expressed are those of the authors alone and we retain responsibility for all errors.


We model competitive bundling and tying, allowing for marginal cost savings from bundling, fixed costs of product offerings, and variation in customer preferences. Pure bundling can arise either because few people demand only one component or because, with high fixed costs, a single product efficiently satisfies customers with diverse tastes. We conclude by analyzing empirically the bundling of pain relievers with decongestants. The discount for the bundled product is large. We argue that our model provides a simpler, more compelling explanation for the size of the discount than the demand-centered approach to bundling by a monopolist.