Evaluations are clear on one thing: Sticks beat carrots for modal shift

Local Transport Today, May 12th  2017

Steve Melia


‘Smarter choices’ have been in the news again recently; the Behavioural Insight Team reported that several ‘low-cost behaviour change interventions’ designed to reduce commuting to Heathrow made no measurable difference (LTT 715). Their comments on evaluation methods (don’t trust self-reported surveys) have re-opened a long-running controversy: how do we know any of these measures actually works? ‘

Smarter choices’ is a catch-all phrase, encompassing anything from a nice email extolling the virtues of car-sharing to closing down the staff car park. The Heathrow trial was at the softest end of that spectrum; we have recently evaluated another intervention at the hardest end of the spectrum, where a university increased parking costs for some and removed the right to park from others. These measures reduced car travel to the campus and also licence-holding, car ownership and car travel for other purposes. These findings have implications for the debates on ‘smarter choices’ and on the changing travel patterns of young adults. They also have potential implications for transport modelling and urban travel demand management.

Over the past ten years the University of the West of England (UWE) has tried in different ways to constrain car travel to its main campus in the northern suburbs of Bristol. The first priority in the early years was to improve the bus services, which were very poor. UWE initially contracted new bus services, which increased patronage but seemed to make little difference to rates of driving to the campus; the early evaluations suffered from all the usual problems of voluntary self-reported surveys so it is difficult to be entirely sure about that.

The roads around the main campus are heavily congested. To expand its activities there UWE had to convince the planning authority that expansion would not generate more traffic. In 2013 it followed some other universities in removing the right to park from newly-starting undergraduates who lived within an Exclusion Zone, covering most parts of Bristol where students live. The cost of parking permits for staff, postgraduates and undergraduates outside the Exclusion Zone were all increased at the same time. As the last cohorts of undergraduates with the right to park entered their final years we set out to measure what difference all of these changes had made.

In 2010 I had conducted a pilot study, which found that 50% of undergraduates were still driving to the campus despite the big improvements in bus services and (fairly small) parking charges at that time. This finding was very different from the ‘official’ UWE travel surveys, which were showing dramatic falls in driving. Although the samples were not exactly comparable the main difference was survey method. My survey was conducted on paper at the beginning of lectures, whereas the official surveys were completed on-line in response to email invitations. The in-lecture method achieved response rates close to 100%, whereas the official surveys were vulnerable to ‘self-selection bias’ – a perennial problem for smarter choices evaluations. The travel planner came to believe, although he could not prove, that the policy changes and publicity around them were motivating more non-drivers, particularly cyclists, to complete ‘after the event’ travel surveys.

From 2012 onwards UWE began conducting cordon counts to address that problem (with a small self-reported element for people who park off-campus and walk in). These counts showed 40% single occupancy driving in 2012, falling to 25% in 2016 with the biggest fall, as expected, in 2013. Buses and walking were the main beneficiaries.

Over the past two years Ben Clark and I have used in-lecture questionnaires again to compare the last cohort of students with the right to park and the first cohort where only a small minority are allowed to park. The second group were less likely to drive to campus, although the fall (from 33% to 24%) was not as big as we had expected because the other measures had already reduced driving, particularly amongst the majority of students who live in Bristol. We also found lower car availability (by 16 percentage points) and licence-holding (by 10 percentage points) amongst the cohort who started after the policy change. As a result, they were also less likely to drive elsewhere for other purposes. The policy change created a gender difference for the first time; men were more likely to park off-campus or try to evade the controls on-campus whereas women were more likely to change their mode of travel.

Overspill parking has created some tensions with residents around the campus, although the problem remains small compared to the big changes in travel behaviour. The scale of off-campus parking did not change significantly between 2015 and 2016 despite the closure of one car park and tighter enforcement on the campus. UWE has responded to residents’ concerns by supporting, and in one case financing, the extension of parking controls on surrounding streets. Two other concerns expressed before the policy change proved unfounded; it did not reduce applications to study at UWE, nor did it move more students outside the Exclusion Zone.

These findings support a growing body of evidence that, like it or not, the ‘sticks’ are more effective than the ‘carrots’ when it comes to modal shift. That doesn’t mean we don’t need the carrots; if driving is constrained by policy or unavoidable circumstances, people are entitled to expect improved alternatives. The findings also support the scepticism of the Behavioural Insights Team about some evaluations of smarter choices programmes, although the failure of the Heathrow experiment, which relied on responses to letters or emails, should cause no great surprise.

Our findings about licence-holding and car ownership were less expected and have wide-ranging implications. In transport models, and conventional transport wisdom, licence-holding is generally treated as an external factor, uninfluenced by demand management tools such as parking restraint. Clearly more research is needed on this but if that assumption proves to be wrong then the long-term potential for travel demand management, particularly in urban areas with growing numbers of young adults, may be much greater than anyone looking at the short-term evidence would imagine. Link to Project Report: Melia, S. and Clark, B. (2016) Evaluation of the change in parking policy on Frenchay campus .

Project Report. University of the West of England. Available from: http://eprints.uwe.ac.uk/30990 


Dr Steve Melia is Senior Lecturer in Transport and Planning at the University of the West of England and author of Urban Transport Without the Hot Air.