Rating the composition: Deconstructing the demand-side effects on transit use changes in California

ABSTRACT

Transit use in the U.S. has been sliding since 2014, well before the onset of the COVID-19 pandemic. The largest state, California, was also losing transit riders despite substantial public investment and increased service in the pre-pandemic period. This downturn prompted concern among transit managers and planners interested in service-side interventions to reverse the decline. However, relatively little is known about changes in the demand for public transit and how shifts in demand-side factors have affected patronage. Drawing on California data from the 2009 and 2017 National Household Travel Surveys, we quantify demand-side changes as a function of two factors—changes in ridership rates of various classes of transit riders (“rate effects”) and changes in the composition of those rider classes (“composition effects”). Statewide, we find that while shifts in the population composition were in some cases associated with lower levels of ridership, the largest declines in transit patronage were associated with falling ridership rates. Specifically, those with limited automobile access and Hispanic travelers rode transit far less frequently in 2017 compared to 2009. Transit ridership rates and rider composition in the San Francisco Bay Area were relatively stable during the study period, while both rate and compositional changes in the Los Angeles area were associated with much lower levels of total ridership. Overall, our findings demonstrate the important role of demand-side factors in understanding aggregate transit use, and suggest that planners and managers may have limited policy tools at their disposal when seeking to bolster ridership levels.