How Well Does Distance Proxy for Freight Costs? Evidence from U.S. Railroads in 1914

Abstract

Researchers often invoke distance as a proxy for freight costs when direct measures are unavailable, but evidence on the historical strength and nature of this relationship is generally absent. In this note, I provide evidence on the historical relationship between railroad freight rates and route distance. Using data on point-to-point railroad freight rates from a newly discovered historical compendium, I show that freight rates are approximately linear in distance but that the relationship between rates and distance is significantly different for short- and long-haul routes, with short-haul freight rates being relatively flat in distance up to a few hundred miles. I provide a point estimate on long-haul rates per ton-mile for Class 1 L.C.L. merchandise and discuss implications for research interpreting variation in distance as variation in prices.