Today I’m talking to Erik Verhoef from the Vrije Universiteit about the project IP-SUNTAN. He doesn’t waste any time. He immediately launches into the discussion by telling me that IP-SUNTAN is a collaborative project between three universities and three urban urbans areas based in Vienna, Austria; Amsterdam, the Netherlands; and Stockholm, Sweden. He says with crystal clarity that the main objective of the project is to find innovative methods to affect mobility behaviour. I ask, “why?”
Before my question can even cross the room, Erik sharply provides a number of reasons why we have to change people’s mobility behaviour. He points out that cities all across Europe (and the world at large) are facing ever increasing congestion levels, increasing concerns over environmental impacts, and dilemmas over limited space (such as parking). He ties all of these issues to the overconsumption of mobility goods and mobility related goods. I reflect on how everything Erik has said so far is true, and I’m curious to hear what brand new solution he has to offer.
Erik responds in an unexpected way to my statement about brand new solutions: he points out that, funnily enough, economic theories have presented the theoretical answer to this problem a long time ago, in fact Erik thinks we’ve had the theoretical answers since the 1920s! Woah, I gasp and say, “then why haven’t we fixed this yet?” Erik asks me to bear with him for a moment. Firstly, he wants to more thoroughly explain the concept before diving into this question. I agree with him as I don’t want to get too confused. He says, “we’ve had the idea of implementing road pricing in order to internalise the societal cost of transport for some time, we’ve only seen very limited application of that in practise.” I realise I need a little more detail, but Erik is already ahead of me.
Erik explains that road pricing works by getting the motorist to pay a toll or a charge per kilometre as this would reflect pricing based on their actual mobility behaviour. This is of course is in stark contrast to how we pay for our roads in most countries today: currently, everyone tends to either pay a fixed annual fee or taxes on a car purchase (sometimes a mixture of both). Erik points out that this is not the most rational way to account for road expenditure. Instead, Erik argues, prices should reflect the societal costs of our transportation. So, now that I understand this concept better, I want to know why we’ve been using a less rational system for nearly a century.
Erik points out that although this is all good in theory, there is a very limited acceptable range of options on price policy. Erik bluntly says, “you don’t have to be a political genius to know that if you propose a new tax, namely a road tax, people won’t be applauding you.” Now I understand in crystal clarity why we haven’t changed our methods in almost a hundred years: no one likes new taxes. Erik, continues, “therefore we have to find ways to use existing taxes and existing price instruments and use them in a more intelligent way.” Now Erik really has my curiosity as I’m not convinced increasing taxes would change people’s behaviour anyway. So, I’m pretty excited to hear what his proposed solution is.
Tradable Peak Permits. Erik says this phrase with very deliberate clarity, I know by the manner in which he says this that this is the big idea of the project. He tells me that it’s the best compromise option as it’s a mid-point between taxes on the one hand and subsidies on the other hand. He points out that this system is designed to be budget neutral, and therefore much more societally and politically acceptable.
You don’t have to be a political genius to know that if you propose a new tax, namely a road tax, people won’t be applauding you.
This all sounds great, and I’m intrigued to go into this concept a little deeper. Erik seems excited to explain it. He tells me, the first time he wrote about this was back in 1997, but back then he couldn’t see how hundreds, possibly thousands, of mobility permits could be traded by motorists. Nowadays, of course things are very different thanks to technological developments: simply put, in 1997 the internet wasn’t what it is today, and we didn’t have the digital technologies which allow us to track the behaviour of individuals or the behaviour of vehicles. Erik adds, “it’s not rocket science anymore, you can implement it in many ways.”
Thankfully, Erik isn’t going to sketch out all the ways we can implement mobility permits. Instead he opts to talk me through a specific one: the one he is proposing would be based on an existing financial instrument being used in the Netherlands: peak avoidance measures. Basically, citizens are rewarded for avoiding the peak travel times. The process begins by measuring how a person typically tends to travel in a week. For example, a person might hit the road every weekday morning during peak hours. If that person then manages to reduce their peak time usage to four days a week instead of five, then they will receive one reward per week. This person could do even better; if they can reduce it by another day, they get another reward. Okay, I now understand how peak avoidance measures work, but what do they have to do with tradable permits?
Thankfully, Erik is as swift at answering questions as I am at generating them. He expands by saying,
“for tradable mobility permits you should imagine we’ll use the same digital infrastructure.” Erik elaborates by saying, we can track the behaviour of vehicles and individuals (using GPS) and we have a means of communicating with these individuals through the internet. The most promising thing Erik says is that apparently, in the Netherlands at least, this technology is well tested and has been used a lot already for peak avoidance measures. So, naturally this technology is ready for an experiment incorporating tradability.
Erik tells me that the current project is about moving the focus away from rewarding individuals to letting them trade instead. So, individuals are immediately given permits that allow them to use the road a certain number of times a week. So, for instance in a five-day working week, you can perhaps get four peak permits. This of course means that if you can avoid one trip in the peak hours, then you’ll be absolutely fine. This immediately creates an incentive for you to use public transport or to cycle one day a week. However, the system actually gets much better; if you cycle two days a week, you actually end up having a spare permit. So, what are you going to do with it?
Erik then explains that if you do end up having a spare permit, there will most probably be someone who needs it more than you do. So, they’ll end up buying the permit from you. This of course strengthens the incentive to save up as many permits as feasible. Erik says, “this is so much more acceptable that road pricing because it means that instead of paying four days a week, you can actually be financially balanced if you choose to alter your behaviour a little.” Most importantly this system leads to balance on the aggregate level.
Erik explains that with tolling there is a net financial loss from the road user to the government, whilst with rewarding there is a net financial loss from the government to the road user. The second option might be effective for congestion, but it isn’t financially sustainable. Afterall, governments have finite funds. Using tradability, the financial flow is between road users themselves. This of course makes the whole system attractive to road users because it is financially sustainable, and the government doesn’t have to subsidise anyone. Great, I think to myself, it’s a kind of marketplace of miles. Erik responds, “You could say that!”
Nevertheless, Erik is quick to correct me: it isn’t just a marketplace of miles but also of time because this system deals with the time of day (peak and off-peak hours) and where you are in the road network also affects the market. This all sets off a spark in my mind; I say, “isn’t this a bit like carbon tax trading?” Erik responds affirmatively, he tells me that we’ve indeed seen this concept of tradability for carbon, and environmental pollution, so in that sense it’s not new, but no one has ever thought to apply it to road transport. The most interesting aspect is that these instruments are usually directed at firms whilst in this case the instrument is directly targeting individuals. Erik explains that the reason we can now use these instruments on individuals is due to technological innovations which make these kinds of transactions feasible on a large scale.
This neatly brings me to my next set of questions. Trading in my mind is quite a complex thing, so I assume that there would be some kind of mobile application which would automatically find buyers and sellers of permits. I ask Erik if this is the case. He says, “exactly, we’ve actually already tested tradability in this way; it was the first experiment of this type we’ve done.” I tell Erik I’m really curious to find out more.
Erik tells me that they ran the experiment with one of the participating cities in the project. However, to reduce the level of sheer chaos that an experiment could cause, they decided to examine virtual mobility behaviour instead of real mobility behaviour. Erik assures me that regardless of its realness, it allowed the project to see if people could use a trading system as they are meant to and if they had problems comprehending it. Whilst thinking about the testing in an actual city, I realise I haven’t considered something really important. Until this point, I haven’t really questioned why the project had chosen Vienna, Amsterdam, and Stockholm. Why do these three cities make good test cases? Erik is quite happy to take a little step detour.
Erik explains that the cities were chosen because they were actually similar enough to implement the same kind of financial instrument, but they were different enough to have their own distinct geographical features. For example, Stockholm has more spacious surroundings whilst Amsterdam is a city which is in tight network of cities relatively close to each other, and finally Vienna has some accessibility issues as it is surrounded by mountains, therefore there are different levels of congestion for each city. Erik also emphasises that the project requires a strong academic consortium: all three of these cities had strong research groups in transport economics. Erik says, “if you do this type of experiment as a part of a project, it is very convenient to have the city be the same city where the university is located.”
Now we zoom back into the example Erik wanted to take me into. He tells me that in Rotterdam they thoroughly tested the concept of trading. In this experiment they recruited participants from an earlier peak avoidance experiment, so the participants were representative of the type of people you would find in experiments on the roads at peak times. These participants were then put into a serious gaming environment. I must confess, the first thought that comes into my mind is paintball, so I decide it’s a good idea to clarify what this means, “what is a serious gaming environment?”
Erik tells me that a serious gaming environment meant that the participants had to log into the system and make a virtual mobility choice. This choice involved imagining that everyday they had to go to work with a car. They were then presented with two ways of paying for parking: firstly, they could simply use ordinary money. Secondly, they could use a tradable parking permit. They chose the concept of a parking permit because in a game environment this is easier to explain than a tradable peak permit. People were initially given a money budget and a set number of tradable parking permits. Therefore, this replicated the exact same kind of market which would be used for tradable peak permits.
Erik explains the experiment in precise detail. The participants were given two full weeks of trading experience. Their aim of course was to maximise their own budget through altering their virtual mobility behaviour. After two full weeks of trading the participants were asked about their perception of the trading instrument. It turns out that for most participants there was nothing complicated at all about the process. In fact, most found the cognitive burden to be minimal. This in part could have been down to the design of the application itself. Erik tells me it was very simple by design, “people could buy, sell or do nothing.” He tells me with excitement that the market functioned exactly as they expected it to.
Now the project is looking to move from the virtual to the real: Erik confidently tells me that they are going to implement a market for tradable peak permits. He is extremely optimistic because he believes they have already tested the two pillars which are required to make tradable peak permits possible. The first pillar being the responsiveness of people to reward mechanisms. He argues the existing peak avoidance measures implemented in the Netherlands are sufficient proof of people’s responses to incentives, and a demonstration that technology is sophisticated enough to make the system work. The second pillar being the feasibility to create a simple enough market where people could technically work in their self-interest. Again, Erik believes this was proven by the tradable parking permit experiment. Simply put, from Erik’s point of view, this is just a case of putting two things that already work together.
Erik tells me that he gets a lot of positive responses when he presents this concept. He says, “people are a lot more excited about this than when I was talking about congestion charges.” He thinks that if done correctly, this idea could benefit many cities in the world. However, before we get too carried away by the optimism, I decide now is a good a time as any to discuss anticipated problems. I look directly at Erik and ask him, “what could go wrong?”
Erik responds by saying that the whole experiment is really interesting because when they create this market, they also have to consider many of the typical problems associated with markets in general. He says gleefully, “for us economists this is a nice chance to see how a market works behind the curtain, for policy makers this a side issue, they just want it to work.” Erik explains they had to design market heuristics which helped clear the market efficiently. Just as things start getting too abstract, Erik decides it would be a good time for an example.
He explains that as permits increase and decrease in value, depending on how many are out there at any given time and how people predict how many they need in the future creates an element of speculation. So, it is important to avoid a scenario where a few players become very rich because they outsmart the system or worse still, they outsmart all the other players. Basically, this means there has to be limits to speculation, yet at the same time you need a little bit of speculation in order for the market to clear itself. I imagine this is a tricky balance, but Erik tells me they’ve put in all sorts of good safeguards into the model. I ask if these are things like price controls. Erik responds that the measures could be considered in this form.
As the interview time starts to run out, I really want to know who is going to benefit most from this peak permit market. Erik gives me a really surprising answer, “ten years ago, I would have said municipalities, however these days private firms are really concerned about the greening of their activities.” He gives me lots of ways in which firms could benefit, too many for me to keep track of. Thankfully, we decide to explore one in detail.
Erik tells me that in the Netherlands businesses provide commuting allowances. However, Erik thinks “they spend this in an incredibly stupid way.” At the moment this budget is not used to incentivise employees in a meaningful way. Erik proposes that companies use this budget and put it into a trading system, this could affect the behaviour of employees. It may even bring them closer together by the very act of trading with their colleagues. Erik stresses that there is no one size fits all solution and that different municipalities and private firms could use tradable peak permits in ways that suit them. This means that different technologies and different incentive structures can be applied depending on the context of the specific implementation attempt.
I decide to finish the interview by asking Erik, “what was the most surprising thing to come out of the research?” He doesn’t have to even pause for a moment. He says, “it was surprising how well our tradable parking permit experiment worked!” He reflects for a moment longer and says, “yes it was surprising, but when you look at it from the perspective of an economist, it’s less surprising.” I wonder what he means by this. Erik tells me an anecdote about how he met a development economist at a conference who introduced the concept of trading games to illiterate tribal communities. Apparently, they had no issues whatsoever of pursuing rational trading strategies in trading games. Erik finishes by saying, “I’m not an anthropologist so I can’t be 100% sure, but I am tempted to say that trading is a part of mankind’s nature.”