In Defense (Again) of “Silly Little Bobtailed” Coal Wagons: Reply to Peter Scott

Abstract

Peter Scott defends the classic Veblen-Kindleberger hypothesis, that sunk costs imposed inefficiencies in the British rail industry into the twentieth century, against the alternative interpretation presented in my article. Scott introduces more refined technical details of the economies of coal haulage to estimate that substantial economies-now purportedly 56 percent of operating costs-could have been reaped if only British coal had been transported in wagons; these operating economies went unrealized because of sunk costs in smallish wagons and infrastructure. That sunk costs mattered at all, in his view, was the regrettable result of fragmented ownership between the railways and the heterogenous wagonowners. Scott's comment defimitely introduces an improved basis for the estimation of operating costs. But as he acknowledges, his new estimates for rail haulage are not what lead him to such radically different conclusions.1 Instead it is two other new assumptions which generate the large cost savings he ascribes to large wagons: that the externalities were considerable and that only an institutional rearrangement of property rights would make the internalization possible. I believe that Scott overstates the magnitude of the potential benefit; moreover, his counterfactuals are quite removed from both the counterfactual I investigated and the actual history of the coal wagon. In this reply I will address Scott's misunderstanding of my operating-cost calculations, and the problems inherent in his two key assumptions, of large unrealized gains and of a radically different institutional setting. On the matter of calculating the savings to be anticipated from a system of larger wagons, Scott expresses dismay at my rather indirect and questionable assumptions. Yet surely a first approximation is to be preferred to the unquantified assertions in the earlier literature.2 At any rate, my intervention has spurred Scott to offer information that refines the engineering details of the calculation, which I accept as an improvement. He believes that I misconstrued the nature of the British rail traffic in weighting my estimates of operational savings by the proportion of coal and in freight as a whole.3 It is puzzling that Scott interprets weight[ing] the calculated operational savings by the share of traffic volume represented by coal and minerals as implying that freight traffic was indiscriminately. Most coal was certainly transported in what are now called unit trains-trains loaded with a single commodity, usually coal, other minerals, or grain-and thus was not typically mixed in with general freight. Railway costs were accumulated on a traffic-wide basis, but it was on the coal portion of the traffic that the railways of the day expected to reduce operating costs. My method ofweighting the total cost savings by coal's share in traffic most directly simulates this expectation. To generalize, as Scott does, the operational savingsfrom larger coal wagons to the whole of railway operations when they were not mixed in amid other freight introduces a strong upward bias. If Scott's projected overall savings of 30 to 35 percent is weighted by the proportion of coal in total railway