The Regulation of Surface Freight Transportation: the Welfare Effects Revisited

This paper reexamines a much-studied topic, the effects of surface freight regulation. It demonstrates that several studies use invalid methods to estimate the welfare costs of rate regulation, develops a correct procedure, and provides esti­ mates of the welfare effects using data and modal market share relationships estimated by Boyer. The paper also analyzes some implications of the common assumption that the demand for total freight shipments by all modes is perfectly inelastic. N UMEROUS studies have addressed the prob­ lem of estimating the costs of resource misal­ location due to the regulation of surface freight transportation. Regulation creates inefficiency to the extent that it does not systematically relate prices to marginal costs. The theoretically correct method for estimating this inefficiency is to mea­ sure the loss in total surplus arising from the regulator's pricing policy. Most of the early studies of the effects of regula­ tion in this industry did not use this method, instead using the "comparative cost" approach.1 The latter assumes that modes are perfect sub­ stitutes. The cost of regulation is then estimated as the difference in shipping costs between the mode by which shipments are transported and the lowest cost mode. More recent studies have used estimates of cross elasticities of demand to take account of the fact that most shippers do not regard transportation modes as perfect substitutes.2 Unfortunately, each of these studies adopts an approach that is not in general theoretically correct and that is likely to produce a biased estimate. Section I of this paper examines the methods for estimating welfare losses, and sections II and III use the data and economet­ ric results in one paper, Boyer (1977), to calculate the magnitude of the bias from using incorrect methods. As is customary in these studies, we are ignoring other costs of regulation. Received for publication November 21, 1981. Revision accepted for publication March 23, 1983. *Northwestern University and California Institute of Tech­ nology, respectively. The work reported here was supported financially by the Caltech Environmental Quality Laboratory. The authors would like to thank F. M. Scherer and E. Shin for helpful comments. 1 See, for example, Peck (1965) and Harbeson (1969). 2 See, for example, Moore (1975), Boyer (1977), Levin (1978), and Friedlaender and Spady (1980). [ 80 ] Replaces California Institute of Technology Social Science Working Paper# 3 ? 0I. Welfare Measurement in Interrelated