Evaluating the Impact of Infrastructure Investment on Regional Labor Markets

This paper examines the effect of improvements in road transport infrastructure on the intensity of inter-municipal work commuting in the Czech Republic over the period 2011–2021. The study uses linked census data on commuting flows between more than 79,000 origin-destination municipality pairs, together with network-based estimates of travel times derived from OpenStreetMap. Using the Open-Source Routing Machine (OSRM), we estimate first-difference OLS models that relate changes in normalised commuting intensity to absolute changes in road travel time, while controlling for pre-existing commuting linkages between municipalities and a set of controls variables. We find that a one-minute reduction in travel time raises car-based commuting intensity by approximately 0.12 to 0.17 car commuters per 1,000 economically active residents. This positive effect remains consistent across both intra-regional and intra-district spatial specifications. The analysis of heterogenous effects reveals a pronounced gender asymmetry. The travel time effect is negative, sizeable, and statistically significant for female commuters in all specifications, while it is statistically indistinguishable from zero for male commuters. For women, shorter travel time thus acts as a catalyst for work commuting across all model specifications. By contrast, for male commuters, the estimated effect is statistically indistinguishable from zero in the pooled national sample, while in alternative specifications the effect remains very weak. Education-stratified results indicate that the aggregate effect is driven primarily by workers with secondary education, whereas workers with either basic or tertiary education do not appear to respond to infrastructure improvements, or more specifically to reductions in commuting time. Stratification by sector of employment shows statistically significant results particularly in the primary and secondary sectors. While shorter commuting times in agriculture and manufacturing are associated with an increased intensity of work commuting, the results for the services sector are less clear-cut and vary across model specifications. Overall, the findings suggest that the labour market returns to road infrastructure investment are highly specific. Reductions in travel time increase car-based commuting, but the magnitude of the effect is modest. Both the strength and direction of the effect depend on the population group considered and on the broader economic-geographical context. Road infrastructure may therefore operate as a potential mechanism for improving labour market access among selected sociodemographic groups, while primarily reinforcing existing commuting corridors rather than generating entirely new ones.