Driving Up Wages: The Effects of Road Construction in Great Britain

This paper estimates the effects of road construction on individual labour market outcomes using micro-data from Great Britain between 2002 and 2008. To capture these effects, I use a measure of accessibility to employment through the road network at a very detailed geographical level. I test the effect of accessibility changes on weekly wages and hours worked. In order to tackle potential sources of bias, I use an instrumental variable which exploits the variation in employment accessibility stemming only from changes in minimum travel times between locations. I argue that, conditional on controls, small scale spatial variation in the accessibility impact of road construction can be considered to be exogenous because road schemes are aimed to improve connectivity and reduce congestion for wider and more distant areas. I further use home and work location specific individual fixed-effects to control for endogenous sorting of workers and I also restrict the sample to workers who are located very close to the projects. I find a positive impact of accessibility from work location on weekly wages and total hours worked but no effect of accessibility from home neither on wages nor hours, conditional on commuting time. These effects are not due to selection into employment as a result of road construction. I also find evidence of accessibility from home reducing commuting travel time. Increased spatial competition or agglomeration externalities are potential explanations for the findings.