Stuck in the Slow Lane: Traffic Composition and the Measurement of Labor Productivity in the U.S. Trucking Industry

Abstract

Mirroring the railroad industry of the 1940’s and 1950’s, the trucking industry today appears to be achieving impressive productivity gains. But it is easy to confuse true productivity advances in transportation industries with changes in ton-miles per unit of input that are due simply to changes in the composition of traffic, as initially happened with the mid-20th century U.S. railroads. This is due to the fact that transportation has vastly different productivities in different settings – for example, when moving long haul versus short haul traffic – and the measurement of changes in physical productivity can be overwhelmed by even subtle changes in the traffic mix. After controlling for endogenous changes in the composition of truck traffic, we find that trucking has in fact been a lagging sector of the U.S. economy over the period of our data, 1982-1997, with observed productivity changes much more likely due to changes in speed limits and the dimensions of vehicles than adoption of information technology. Our finding of a slow improvement in the physical productivity of trucking inputs does not deny the real improvements in the quality of trucking services (reliability, predictability, speed, order tracking, etc.) that have taken place in the last quarter century. But as in other service industries, true physical productivity improvements in trucking are hard to find.