Investigating observed and unobserved variation in the probability of ‘not travel’ as a behavioural response to restrictive policies

• Published in: Transportation research. Part D, Transport and environment Transport and environment, 2019-12, Vol.77, p.63-76
• Main Authors: Flügel, Stefan, Fearnley, Nils, Killi, Marit
• Format: Article
• Language: English
• Subjects: Diversion; Monte-Carlo simulations; Passenger demand; Restrictive policy; Suppressed demand; Trip cancellation
• ISSN: 1361-9209; 1879-2340
• DOI: 10.1016/j.trd.2019.10.008

•With mixed logit models, we investigate ‘not travel’ as a behavioral response.•The data includes close to 17,000 stated mode choices.•Car users are more prone to react with ‘not travel’ compared to PT-users.•The probability for ‘not travel’ is lower for work trips.•Travel distance has a significantly positive effect. Besides technological improvements, restrictive car policies are likely to be the most effective measures for reducing greenhouse gas emissions from local passenger transport. Restrictive policies may lead some individuals to choose to not travel to otherwise useful or enjoyable activities. This paper therefore explores what factors drive the probability of ‘not travel’ as a behavioural response to restrictive policies. Using stated choice data among car owners in the 10 largest cities in Norway, we investigate observed and unobserved taste variation for ‘not travel’ given different (hypothetical) policies. The empirical evidence suggests that the likelihood of ‘not travel’ (a) is lower for work-related trips; (b) is higher where respondents state they have no decent alternatives; (c) increases with trip distance; and (d) increases with the intensity of the policy. We perform Monte-Carlo simulations illustrating different predicted choice behaviour for car users and public transport users under different types of stylized policies (travel time changes versus travel cost changes).