As sales of e-bikes increase, so does the need for reliable evaluations of which means of transport the e-bike replaces, what we call the substitution effect. A randomised controlled trial with GPS data from 98 frequent drivers in Sweden was conducted to investigate the effect of the e-bike on modal choice, the number of trips, distance, as well as perceptions of the e-bike as a substitute for the car. The results demonstrate that the treatment group increased cycling on average with 1 trip and 6.5 km per day and person, which led to a 25% increase in total cycling. The whole increase was at the expense of car use, which on average decreased by 1 trip and 14 km per person and day, a decrease in car mileage of 37%. Implications for policy and further research are discussed.
