Automobile manufacturers have long had a policy of introducing annual style changes into their new model cars. Textbook publishers have a policy of periodically bringing out revised editions of their popular books. Recently, IBM changed the operating system used in its personal computers causing a reduction in the compatibility between its old and new machines. If one were to ask the typical man in the street to explain what is happening in these markets, he would not hesitate before giving the following answer. In economists' terminology, he would say that for the above firms the incentive to introduce new products that make old units obsolete is" too high"; ie, the firms have an incentive to practice planned obsolescence. Of course, just as is true for the man in the street, economists have also long observed and commented on the speed with which new products are introduced. For example, Galbraith [1958] suggests that the annual model changes of automobile manufacturers are socially wasteful, while Fisher, Griliches, and Kaysen [1962] provide empirical estimates of the costs of these changes. There have also been numerous theoretical attempts to model the idea of planned obsolescence (or the related idea of killing off the market for secondhand goods), but most of these models have in fact considered the choice a monopolist faces concerning how much durability to build initially into his product, rather than the monopolist's incentive to introduce new versions of his product over time.'In this paper I consider planned obsolescence from the standpoint of a firm that can change the nature of what it sells, and show that the man in the street is in fact correct. In many settings the incentive for a durable goods monopolist to introduce new products that make old units obsolete will be above that which is socially optimal.