How Gothenburg learned to love congestion pricing

A small but increasing number of cities have implemented congestion schemes and charges, including London, Milan, Singapore and Stockholm. What can we learn from Sweden’s second-largest city?
Gothenburg's pricing scheme isn't actually aimed at reducing congestion, but rather at raising revenue for large-scale infrastructure projects such as a major rail tunnel. (J-O Yxell)

In 2013, Sweden’s second-largest city, Gothenburg, introduced congestion charges for cars. At that time, attitudes among commuters about the idea were very negative, even prompting some political turmoil. As time has passed, however, city residents’ views of the programme have become more favourable, among both car commuters and others.

Four years on, what can we learn from Gothenburg’s initiative? Research by Mistra Urban Futures and Chalmers University of Technology has made several notable findings. For instance, women who commuted by car before the congestion charge were twice as likely to change to another mode of transport following the introduction of the scheme, after controlling for relevant factors. Likewise, the relative accessibility of different transport modes had an important effect on which adaptation strategies people chose.

Perhaps most important, the researchers found no significant changes in travel satisfaction in any of the studied groups. The congestion charge may be unpopular for various reasons, but it does not seem to have any larger effects on everyday lives. Instead, people simply learn to adapt to it.

The majority of the world’s cities have left traffic volumes essentially unrestricted, despite the fact that congestion is almost universally acknowledged as a problem. Nonetheless, a perceived lack of public support for strategies to limit congestion has remained a key reason for why city authorities do not take significant steps on this issue.

[See: Beijingers skeptical of plans for a congestion charge]

The debate over what regulators can do in this regard is not new. “Road pricing” policies have been discussed in economic literature for almost a century, fuelled by the question of how to provide a cost-efficient means to alleviate congestion in urban transport. Today, however, only a small — albeit increasing — number of cities have implemented congestion schemes and charges, including London, Milan, Singapore and Stockholm.

Because of this fairly small sample size, the implementation of Gothenburg’s scheme was keenly followed by researchers. They were interested particularly in studying changes in travel satisfaction and attitudes for those affected (or unaffected) by the new policy. They also looked at adaptation strategies among different socio-demographic groups.

Infrastructure revenue

So, how did it all work — and why was the decision made in the first place to imposing congestion pricing on Gothenburg’s streets?

The Gothenburg scheme is technically a tax. Its primary purpose is to contribute to the financing of large-scale infrastructure investments, such as a railway tunnel under the city. Calling the new levy a fee or toll would not have been compatible with Swedish law, as fees can be used only to build new infrastructure.

[See: Why Helsinki’s innovative on-demand bus service failed]

Of course, the aim also was to reduce traffic during peak hours. Yet Gothenburg is a medium-sized city with some 500,000 inhabitants, and traffic congestion was never a significant issue. Further, public transport use is low, compared to larger cities, including Stockholm.

The key reason for introducing the congestion charges, then, was not actually congestion. Rather, it was connected to an opportunity to receive governmental grants for infrastructure. But this opportunity was on offer only on the condition of local co-funding — in other words, the revenue from the congestion charge.

This was something that created a basis for the necessary political support. With the revenue and the national funding, it would finally be possible to finance a large package of infrastructure investments, including significant improvements of the railway through the city.

The Gothenburg tax is connected to the registration numbers of the vehicles passing through any one of 37 automatic control stations, which automatically register vehicles’ license plates. (Jan Riise)

Under the scheme, rivers are charged around USD 1 to enter the city centre during low-traffic periods, a pricing structure that climbs to almost USD 3 at peak hours. Between 6:30 pm and 6:00 am, there is no charge, while weekends, public holidays and the month of July are also free.

These prices are far below many other such programmes. Oslo introduced congestion charges in October, with less differentiation between peak and off-peak hours — about USD 5.50 to 6.50 per day. Electric and hybrid cars are excluded from the toll. In London, driving within the congestion zone costs USD 15 per day, Monday to Friday. Singapore has a technologically advanced system where congestion charges vary according to the actual congestion and time of day, but a typical charge is about USD 1.50. Finally, in Milan’s historical center on weekdays between 7:30 a. m. and 7:30 p. m., drivers have to pay about USD 6 to enter.

[See: How a city in Spain got rid of its cars]

In Gothenburg, the scheme applies to the cost to travel by car during daytime within the limits of the central part of the city — not only “downtown” but including relatively large residential areas. The tax is connected to the registration numbers of the vehicles passing through any one of 37 automatic control stations, which automatically register vehicles’ license plates.

There are also limits on the charge. Drivers can be charged only a maximum amount per hour and per day, to alleviate the burden on those travelling frequently or who happen to pass multiple control stations on the same journey.

And what have the results been? The total traffic volume during charged hours decreased by around 10 percent in the first year, according to the Swedish Transport Administration. Car traffic volumes since then have increased, but not as much as the initial decrease. The same pattern has been seen in Stockholm. Although different schemes are not directly comparable, this reduction is smaller than for some previously studied cities.

Positive shift

More interesting for researchers, however, have been the qualitative results. Before the scheme’s implementation, a survey was distributed to 3,500 car owners in the Gothenburg region, followed by a follow-up survey to the same respondents a year later. Respondents were asked about commuting habits, attitudes toward the congestion charge, the environment, “automobility” (use of their private vehicles) and public transport. They also were asked about their satisfaction with travel, along with sociodemographic and geographical variables.

One key finding from this data was that women who commuted by car before the scheme’s implementation were twice as likely as men to change to another mode of transport following the policy’s introduction, even after controlling for other relevant factors.

The finding appears to confirm previous research showing that women are more likely to travel by public transport. In Gothenburg, too, women appear to be more prone to shifting their mode of transportation than men. Other sociodemographic variables such as age, income and education did not have any significant effect on this behavioural change.

[See: Moscow makes over its streets with people in mind]

Researchers also discovered that introducing a congestion charge does not seem to affect satisfaction with travel among citizens affected by the scheme. Indeed, during the follow-up study, all groups reported a relatively more positive view of the pricing programme. This was a finding across the board, regardless of whether people adapted by shifting modes of transport or reducing their commuting, or if they continued with the same commuting behaviour. The effect was more pronounced among those adapting their commuting behaviour.

Overall, researchers found Gothenburg respondents moving from 2.40 to 3.03 on a 7-point scale, where 7 indicated higher popularity. Car commuters were, as expected, slightly more negative than people going to work by other means of transport, moving from 2.24 to 2.56 during the study period, compared with 2.79 to 3.24 for the latter group.

The perceived “fairness” of the congestion charge tax remains negative. This figure increased only slightly from 2.57 before to 2.78 a year later, with the same differences regarding car commuters and others.

The findings suggest that although a congestion charge scheme itself may be unpopular for several reasons, its experienced effects need not be negative, as people adapt. Attitudes toward the Gothenburg scheme were very negative before its introduction, but households that changed travel behaviour and those who chose to pay the new charge both grew less sceptical over the course of the study period.

Back to top

More from Citiscope

Latest Commentary

Jonas Nässén

Jonas Nässén is a senior researcher in energy and environment at Chalmers University of Technology doing interdisciplinary research related to societal transitions toward long-term climate targets, partly within the framework of Mistra Urban Futures in Gothenburg, Sweden.

Jan Riise

Jan Riise is manager for engagement at Mistra Urban Futures, a trans-disciplinary research centre for sustainable urban development located at Chalmers University of Technology in Gothenburg, Sweden.