New Equity Inputs to Prioritize Bikeshare Infrastructure Allocation: Learning From the COVID-19 Period

Abstract

From “pop-up” road closures to decreased transit frequencies, the COVID-19 pandemic has changed the overall supply of transport options. Even in the absence of a change in bikeshare supply, the pandemic provides a “natural experiment” under which we can assess changes in bikeshare use across diverse communities in response to transportation system changes. The pandemic offers a unique moment to particularly measure changes in use for low socioeconomic status (SES) populations as historically limited deployments of bikeshare in low-income neighborhoods limit evaluation of key metrics for this population. For low SES users to realize greater accessibility through bikeshare, they may need to take relatively longer trips, given the sparse nature of the network in low-income areas and the existing inequitable geography of opportunities in urban environments in the United States. As such, we measure the effect of the COVID-19 pandemic on average daily bikeshare trip durations in Philadelphia, PA—the major city with the highest poverty rate in the United States. Through an interrupted time series approach, we find that the effect of the pandemic on trip duration for all bikeshare users is substantial (approximately 7–12 min increase), positive, and similar across diverse geographic areas. Importantly, these findings are persistent and statistically significant even when fitting models only on data from predominantly low SES areas of Philadelphia. This change pattern suggests first that low SES users exhibit roughly equal propensity as the general population to take longer trips, and second that bikeshare can provide a resilient, equitable travel mode.