Estimating the social return to transport infrastructure: A price-difference approach applied to a quasi-experiment

Using the price differences between two markets, this study proposes a method for inferring the social return to transport infrastructure (in the vein of Fogel (1964)). We apply this approach to an investment that increased the shipping capacity of a thousand-mile-long railroad in western China. The event was quasi-experimental: before the expansion, the railroad was congested in one direction, but not the other. We find that, after the investment, (1) the between-destination price differences of goods shipped in the congestion direction dropped by about 30% and (2) shipping volume increased by around 40%. In contrast, those of goods in the other direction were not affected. These estimates imply a sizable social return to this particular investment: 10% per year in the most conservative case considered.