This paper analyses the employment-generating impact of a tertiary road project in Nicaragua, applying a matched double-difference approach to control for initial conditions and time variant factors that simultaneously influence the placement of roads and subsequent employment growth rates. Results are promising. The author's estimates indicate an increase in hours worked per week attributable to the intervention of around 9.5–12.3 hours. Moreover, he observes tendencies of a graduation process taking place in the labour market: individuals moving out of unemployment predominately achieve employment in the agricultural sector (self-employment), whereas newly created service sector jobs primarily are taken by workers previously working in agriculture. Finally, the analysis suggests that the employment-generating effect comes through a combination of reduced travel time and better access to markets and larger, more integrated road networks.