The Stockholm congestion pricing syndrome: How congestion charges went from unthinkable to uncontroversial

Abstract

Congestion pricing was introduced in Stockholm in 2006, first as a trial followed by a referendum, and permanently from 2007. Public attitudes went from fiercely negative before the start of the trial, to positive at the time of the referendum (where 53% voted in favour of keeping the charges) and even more positive when the charges were made permanent. Now, more than 2/3 state that they would vote in favour of keeping the charges.

How did this happen? How could such a controversial policy be introduced, survive a referendum and then settle down as an almost completely uncontroversial fact? The purpose of this paper is to describe the changes in opinion and discuss a number of explanations. The discussion rests on analyses of six surveys of public attitudes carried out 2004-2011. Among other things, we investigate to what extent behaviour and attitudes can be explained by selfinterest variables and belief in the charges’ effectiveness, and to what extent people are able to predict or remember changes in attitudes or behaviour. The findings are interesting not only because congestion pricing is a potent policy measure, but also because they illustrate how attitudes to new policies are formed and change, and may hence be applicable to other reforms, in particular environmental policies.

Analyses of the six surveys 2006-2011 reveal a number of interesting findings, for example:

- Self-reported changes in behaviour considerably underestimate actual changes: the real change in the number of charged car trips is around four times larger than the change reported by respondents. - The same is true for attitudes: self-reported changes in attitudes vastly underestimate real changes. - Self-interest seems to be influencing beliefs in the charges’ effectiveness. Consciously or subconsciously, opposition to the charges induced by selfinterest are rationalized by an expressed distrust in the effects. - Self-interest and belief in the charges’ effectiveness affect attitudes at any given point in time – but they can only explain a minor part of the change in opinion. Hence, the hypothesis that changed beliefs about effects is the main driver of the change in attitudes seems unfounded.

The swing in attitudes was perhaps the most surprising development in Stockholm. The support for the charges started from a relatively high level of 43% in 2004, but dropped to 34% in 2005 immediately before the start of the trial. In April 2006, after four months of congestion charges, support had increased to 53%, similar to the referendum outcome in September 2006. A survey in December 2007 showed another leap in the support to 65%, and subsequent surveys in 2010 and 2011 have shown similar results. The explanation most commonly put forward is that benefits turned out to be much larger than expected. Indeed, several studies, in Stockholm and many other

cities, have shown a strong link between support for congestion charges and belief in their effectiveness. But there is actually very little support for this hypothesis: the surveys reveal that this can only explain a minor part of the change in opinion. In fact, beliefs about the charges’ effects have remained surprisingly constant over time. Instead, the shift in attitudes and the debate about them can be viewed as a public and political reframing of the congestion pricing over time, using an attitude formation framework from social psychology. This interpretation simultaneously explains why it was politically rational to first ignore congestion charges, then to advocate them in spite of public resistance from large groups, and why the resistance then died down. It also illustrates the importance of legitimacy. The story can be summarised in four phases:

1. As long as the argument for congestion pricing was that it would increase the efficiency in the transport system, it failed to generate any political interest: since transport efficiency is simply not an issue that many people get enthusiastic about, the issue had virtually no political upside. Instead, when faced with the question, people associated to a superficially similar issue such as mobility restrictions or taxation. 2. This changed when congestion pricing was reinterpreted as an environmental policy, which happened in Stockholm during the mid 1990’s. While allocation efficiency in the transport sector could not arouse enthusiasm or engagement among the general public, environmental concerns definitely could. This was what was needed to get congestion pricing on the political agenda – a link to an area with where strong and emotional attitudes existed. 3. When the decision to carry out the congestion charging trial was made after the election in 2002, a fierce debate broke out. Consistent with what was said above about the necessity of emotions in politics, the arguments soon turned principal, moral and emotional, leaving little room for compromise. This might have been an inevitable development: if congestion pricing had not been elevated to a moralemotional question, it hadn’t entered the political stage in the first place. But just as inevitable, the morally supercharged arguments for congestion pricing implied (or could be perceived to imply) that all car traffic was evil and unnecessary, and should be banished. 4. When the referendum ended in a yes to the charges, the new government decided to earmark the revenues for a motorway tunnel west of Stockholm. This gave the charges democratic legitimacy; and earmarking the revenues for roads not only spoke to motorists’ selfinterest, it sent a moral signal: it’s OK to be a car driver. It indicated a reinterpretation or re-clustering of the congestion pricing issue from a morally charged anti-car measure to a technical-rational measure that was effective – it “worked” in the sense that it generated revenues and reduced congestion. This discharged some of the sentiments around the charges, and moved the debate from the moral domain to the technical-rational domain.