| Prior studies have shown that the provision of bike paths, bike lanes, and other types of bicycle-specific infrastructure is positively correlated with bicycle commuting rates in U.S. cities. This is the first study to examine the effect of federal funding for this infrastructure on urban bicycle commuting rates. Using OLS regression, this analysis models bicycle commuting rates as a function of federal spending, population density, gas price, year, and geographical location in the 51 largest cities in the United States during 2007, 2009, and 2011. The results indicate that per capita federal infrastructure spending has a positive, statistically significantly relationship with bicycle commuting, while total federal infrastructure spending does not, and that each additional dollar of bicycle and pedestrian infrastructure spending per capita will increase a city's bicycle commuting rate by 0.047 percentage points, on average and holding other factors constant. These findings suggest that federal bicycle and pedestrian infrastructure funding is most efficient when it is allocated to cities in amounts relative to their populations, but this is not evident in the sample. The analysis also estimates a statistically significant 2-percentage point increase in bicycle commuting rates in America's largest cities in 2009, at the peak of a period of recession, suggesting that commuter mode choice is sensitive to economic hardship. |
