Recent papers by Dixit (1980) and Eaton and Lipsey (1981) have explored the strategic use of committed product-specific capital by an incumbent firm in order to influence a post-entry equilibrium, and in some cases make unattractive. By committing or sinking part of the production costs prior to entry, the incumbent creates a credible shift in its own reaction function if should occur. In this way the payoffs in the post-entry equilibrium can be influenced to the incumbent's advantage. The entry thus consists of two stages: the first, in which the incumbent is able to strategically commit itself prior to entry; and the second, or post-entry period, in which a duopoly equilibrium is realized. The more costs that are put up front by the incumbent, the greater is the post-entry output it will wish to sell for a given output of the entrant, and the lower will be the resulting equilibrium price. If the set-up or fixed costs of are sufficiently high, may be rendered unprofitable. This paper extends the idea in a natural way. All the production costs of a quantity of output can be sunk, or committed, by stockpiling inventory of the product. The stock of inventory held by the incumbent firm represents a credible shift in its reaction function in response to entry, in the following way. The inventory is a stock that can be sold off over time in a way consistent with intertemporal profit maximization, constrained by the presence of the entrant. Moreover, the threat to sell off the inventory in response to is credible, since, given that has occurred, the opportunity cost of the inventory is zero. It seems reasonable to require that any strategic asymmetry between incumbent and entrant should be restricted to commitments made by the former prior to entry. Conditional on such commitments, the post-entry game should give neither firm a strategic advantage. This requirement is the same as the game-theoretic concept of perfectness: 1 can be credibly deterred because the entrant knows that, were to take place, it would be in the incumbent's own interest to follow the anticipated sales programme of inventory. In order to demonstrate that may be deterred by holding a stock of inventory that is sufficiently large (that deterrence can be made the perfect equilibrium of the complete game), the properties of the equilibrium with must be derived. The post-entry equilibrium is modelled here as a pair of Nash strategies in sales programmes. If the sales programme anticipated by the entrant in the post-entry equilibrium implies a negative present value, will be deterred. Thus, successful deterrence implies that the equilibrium with is never realized. The incumbent can then continue to receive monopoly profits, less the holding cost of the inventory, which serves as a credible deterrent to but is never sold.