Gothenburg to implement congestion charging

Gothenburg, which is line to become Sweden's second major city to implement congestion charging, will not enjoy the pre-deployment trials and referendum which Stockholm did. But, says the STA's Eva Söderberg, this is less of an issue than might be imagined

Sweden's capital city Stockholm was afforded the unique opportunity of a full-blown technology and user trial before its congestion charging scheme went live on 1 August 2007. The trial was pivotal in terms of public opinion and acceptance - and, ultimately, in helping to secure a majority 'Yes' vote when the local population was asked whether it agreed or not to implementation.

The 2124 Swedish Transport Agency (STA) is now working with the City of Gothenburg towards the 2013 start-up of a congestion scheme there which forms part of a much wider transport network improvement programme (the West Sweden Infrastructure Programme, which will take place over the next 25 years). This time there will be no trial with a following public referendum on whether the scheme goes ahead, says Eva Söderberg, the communications officer within the STA responsible for the Stockholm scheme.

"The country's senior politicians feel that Stockholm has already provided the answer - that although Gothenburg faces the same issues of public acceptance that Stockholm did prior to go-live, they don't need another costly trial in order to 'dare' to go ahead," she says.

"Worldwide, there has been a lot of admiration of Swedish politicians' courage in going ahead with Stockholm but the reality is that the trial was a real differentiator. Never before in Swedish history have we seen such a swift change in public opinion - people were very against charging before it.

"The trial was especially important given the nature of congestion charging. With most laws which significantly change people's lives, it tends to take some time after their introduction before their effects are really felt. But with charging the effect is immediate." The vote on the Stockholm scheme was organised at the county level and some 17 of the 26 around the city decided to hold one. As a result, the question asked of voters was not always the same.

"Put the vote up to some form of scientific scrutiny and it would probably fail," says Söderberg, "but the most important thing to remember is that there was a trial and most people could vote. Most people, therefore, had the opportunity to change their minds if they wanted to."

Part of a whole

There are several other issues which have guided the decision to deny Gothenburg a vote. One is the length of time left to organise one, which is seen as too short. Another is that the congestion charging scheme is part of something much larger and is a deal between all the political parties; take away congestion charging and there would be a huge revenue hole and the political deal regarding the other projects would fail. Söderberg notes a further reason for local opposition in Gothenburg: that there is nowhere near the congestion problem there that there was in Stockholm, to the extent that to call the scheme a congestion charge is perhaps somewhat misleading. Nevertheless, she denies that not having a vote is a retrograde step.

"There was a lot of work done for Stockholm which means that we know better what to focus on in the new scheme. With Gothenburg we're going to concentrate on selling the wider infrastructure improvements, including those to railways, highways and public transport."

Multi-million initiative

The STA spent some SKr60 million (about e6 million) on public information in the 12 months up to the Stockholm go-live. The nature of the scheme dictated some of the decisions made. For example, the Stockholm scheme only charges domestic drivers, not those from abroad. That was a major reason why the transponder-based Dedicated Short-Range Communications (DSRC) technology used during the trial was abandoned in favour of an Automatic Number Plate Recognition (ANPR)-based solution; the integrity of the national vehicle register combined with the accuracy of ANPR were together seen to make DSRC surplus to requirements. It also made it easier to identify who to target with information. Two letters were sent to all vehicle drivers in the whole of Sweden, according to Söderberg.

"This was perhaps our best communications channel as it was read by a very large number of recipients. There was a very active media campaign and meetings were held in public areas such as shopping malls and at major transport nodes where both the technology and the concept were demonstrated and experts were on hand to answer the public's questions.

"We also invested hugely in public services. In fact, we overestimated the need in this respect but we wanted to make sure that when people dialled one of our help-lines they got immediate answers. We didn't want people to be criticising administrative failings; we wanted them instead to be concentrating on the scheme itself." Informational responsibility was divided between the STA, which handled the issue of 'how' (how to pay, how the system works and so on), and the City of Stockholm, which looked after the 'why' (why congestion charging was needed and its consequences).

"Focusing on our own areas of responsibility allowed answers to be given which were not held to be coloured politically and helped to engender trust," Söderberg says. "We've also been very open about where revenues have been spent. That dispels notions of hidden agendas. Perceptions of honesty are very important - that even extends to admitting so if you don't have an answer to a question.

"It's notable that well-informed people tend to be more trustworthy of the technology they use. We wanted people to feel safe within the system and a main focus of the Gothenburg programme will be to leave people with no questions left to ask." In total, the Stockholm trial cost SKr 4.2 billion (e0.42 billion). This includes the attendant improvements to public transport and the setting-up of park-and-ride schemes around the city; the actual 'core' scheme cost SKr 1.7 billion (e0.17 billion). This was a direct spend by the Swedish Government.

"We were criticised for cost but there were huge time pressures. There was also a huge level of built-in redundancy. 62 IBM, which manages the scheme, has said that if it knew then what it knows now the trial would have cost half as much," Söderberg continues.

The move to Gothenburg

"The Gothenburg scheme will cost less, not least because it can take advantage of what we've already done in Stockholm. We've made lots of software improvements and other changes to the system. For instance, when we first started people had to pay within five days and would not be invoiced. We also opened lots of different channels of payment through retail outlets, the internet and so on. Now, users pay monthly and receive an invoice. It's a much simpler system which, again, increases transparency and is more cost-efficient.

"Also, we'll be starting our information campaign for Gothenburg just six months out from go-live, ramping up significantly in the last two to three months. With Stockholm, we started 12 months out, and also ramped up towards the end, but back then congestion charging was an unknown quantity in Sweden. This time, we reckon on being able to go with the shorter run-up. You have to strike a balance between starting too early, and presenting people with an issue which in terms of time is too remote to deal with, and starting in time to allow people to take an issue on board and gain an informed opinion of what's happening." The information campaign for Stockholm is ongoing, and Söderberg expresses surprise at just how high a level of ongoing interest there continues to be from the media.

"Every change to the system means we have to intensify our media effort. Our website is visited very regularly and our customer service phone lines continue to receive a lot of calls. Call-handling capacity has obviously been scaled back since the initial launch but there are still 70 staff employed in our customer service centre. They also handle queries on other issues and help with manual sorting of ANPR images which can't be sorted automatically, so there's an efficient use of resources."

Overall levels of satisfaction with the public information scheme are high. Söderberg says that opinions range from those who feel that there is too much information available to those who still feel that they don't get enough. There remains the issue of how to adequately inform occasional users but she notes that the monthly invoices do a good job of this: "Really, people don't have to do anything until they receive an invoice and in any case the invoice itself holds a lot of information on how to contact us.

"We see a lot of use of our website and users can access their accounts using a PIN. That service isn't actually used much, which suggests the invoice is sufficient for most people."

Looking back over the course of the information campaign, Söderberg says that it is difficult to see where anything would have been done differently.

"The cost of the trial is the only thing that in retrospect could have been done differently - but that's with hindsight and has much to do with the fact that we now have a much simplified system.

"From an operational perspective, I'd also push the politicians to be more specific about where the revenues go. One way to have a system which is less politically sensitive is by proving to people that there are long-term benefits from the charge which affect them directly and that it isn't just a short-term grab which is useful politically. Again, it's all about transparency and it's one reason why we now publish monthly statistics on a variety of different performance measures.

"Congestion charging also needs to be put into its wider context and it's important to highlight the sustainability element. Charging isn't a single solution to a huge problem and it shouldn't be presented as such. You have to be honest about that. And transparency extends to respecting the complexity of the problem and then being sure that the proposal you're offering is the right one."