LAUSR

Abstract This study takes the 39-day Glasgow Subway closure in July 2016 as a natural experiment to evaluate the effect of subway closure on bike-sharing trips. We find that bike-sharing trips increased by 20.7% for incoming trips and 20.1% for outgoing trips on average for each bike station in the proximity of subway station during the subway suspension. Some of this change persisted, with 12.4% of the increased bike-sharing trips remaining after the resumption of the subway service. Our findings suggest that first, subway and bike-sharing trips are substitutes; second, this temporary service disruption was not enough to break commuters’ long-term habits, and third, the diversion factors implied by our results are much lower than the recommended values for UK cities.