Spatial Monopoly, Non-Zero Profits and Entry Deterrence: The Case of Cement

Abstract

AT least since Kaldor (1935), a number of economists have argued that indivisibilities and accompanying economies of scale in spatially extended economies can negate the zero profit result generally associated with free entry.' The essential argument is that if there are industries in which transportation costs are non-negligible and size economies call for firms that are large relative to local demand, even the most intense competition need not result in a zero profit equilibrium. Recently, Eaton and Lipsey (1978) have provided the analytical underpinnings for Kaldor's largely intuitive argument and demonstrate that pure profits can remain even with free entry of new firms and price competition. Eaton and Lipsey (1978, p. 467) suggest that their result depends critically on the.existence of the following conditions: (1) the average total cost curve is declining over some initial range; (2) customers are geographically spread out and intermingled with firms; (3) transport is costly; and (4) once the firm enters the market it has location-specific sunk costs.2 This paper offers an empirical investigation of the non-zero profit argument as applied to the U.S. cement industry. On a priori grounds, the cement appears to meet the necessary conditions for existence of positive, site associated profits. The existence, or at least the belief that the a priori conditions are satisfied has provided a basis for using the as a representative case. Scherer (1980, pp. 252-258), for instance, offers a textbook example of entry deterrence through plant location strategy that combines elements of space and the standard deterrence argument.3 Scherer then notes that, In an interview study of plant-size and location decisions in 12 industries across six nations, the author observed by far the strongest emphasis on geographic space packing as an entry-deterring strategy in the cement industry (p. 257). Scherer's example of entry deterrence via strategic plant proliferation is an extension of the Kaldor non-zero profit argument. However, the non-zero profit argument has at least one major drawback: profits in the cement have not been significantly above normal for any sustained period. The first section of this paper considers the evidence of the cement industry's approximation to the Eaton and Lipsey conditions and its historical profitability. Given the contradictory results between apparent satisfaction of the a priori conditions and the lack of supernormal profits a reformulation of the basic model is presented. An explanation offered here for the absence of supernormal profits is derivable from the property rights paradigm. The usual procedure in analyzing the reaction to new entrants in a spatially extended model has been to assign either initial locations to a set of firms or provide for a prearranged pattern of sequential entry. Both procedures implicitly assign property rights to locational rents and ignore the vital element that there will be competition to be the first plant in a given location.4 With demand for the product growing, competition to be first at a given location will insure that the timing of construction of the