Build, Move, Repeat

This dissertation examines how the built environment shapes active travel and, in turn, opportunities for everyday physical activity in increasingly sedentary societies. While planners often assume that changes to the Built Environment cause significant shifts in travel behaviour, much existing evidence is correlational and vulnerable to bias from self-selection, unobserved confounding factors, and limited data. The thesis advances causal understanding of built environment–active travel relationships by asking: (1) how and to what extent does active travel infrastructure influence travel behaviour, and (2) how and to what extent do density, destination accessibility, and land-use diversity affect walking and cycling. Four empirical studies address these questions using quasi-experimental and longitudinal designs. Two natural experiments estimate the effects of large-scale cycling network expansions in São Paulo and the introduction of cycle highways in the Netherlands, applying robust exposure measures derived from routing algorithms and geospatial analysis. A third study investigates how neighbourhood contexts moderated changes in mobility-related physical activity during COVID-19 restrictions. A fourth study uses panel data on residential movers and a random-intercept cross-lagged model to assess how relocation-related built environment changes influence mode use and attitudes. Across studies, results show measurable but heterogeneous increases in cycling and walking, highlighting the importance of context-sensitive, equity-oriented policy design and robust causal methods for transport planning.