Transport infrastructure is conceived as an important determinant of regional economic development. While this prediction especially holds from a theoretical perspective on the premises of endogenous growth theories, from an empirical perspective it is not that easy to verify this causal link, though. Measuring whether transport infrastructure indeed fosters regional development or if it is merely an endogenous reflection of the higher transportation demand in prospering regions is a challenging task. This paper analyzes the long-run effects of port facilities on regional income levels in Germany. Since it is very likely that results from standard least square estimations suffer from endogeneity problems, we base the identification on exogenous long-run instruments. In particular, port facilities built before the industrial revolution provide an adequate instrument for current port infrastructure since they are exogenous to recent economic development but the natural requirements of a port site form an experimental character of their locations. Results from German district-level data for 1991–2008 hint at a positive correlation between port locations and gross domestic product per capita, but do not provide evidence for a causal relationship regarding the instrumental variable approach. Further robustness checks with population distribution as outcome hint at positive long-run persistence effects of port infrastructure.