| Short-run cost functions are estimated using plant-level data from the Portland cement industry in Pakistan. The objective of the study is to ascertain the feasibility of marginal-cost pricing in an industry that has been repeatedly accused of departures from competitive behaviour: e.g. basing-point pricing was banned by the U.S. Supreme Court in 1948, a case was filed in the Restrictive Practices Court in 1961 against cement manufacturers in the U.K., and the industry was nationalized in Pakistan in 1972. The data set used has two main advantages. First, given that the plant and not the firm is the relevant unit of production, detailed plant-level data (in particular, output and capacity) are available. Second, the industry was producing at full-capacity throughout the period under study, 1960-1991. The latter feature is of special interest because it rules out the possibility that any measured economies of scale might be capturing the effect, on unit fixed costs, of periodic idle capacity produced by demand fluctuations. The results overwhelmingly support the hypothesis of marginal costs being below average total costs, even at peak capacity utilization rates. This has important implications for public policy; in particular, it indicates that it is inappropriate to infer monopoly behaviour from any deviation of price from marginal cost. |
