«Björn Hårsman Inregia 2001-11-11 To be presented at the IMPRINT-EUROPE seminar in Brussels November 21 and 22, 2001 Contents 1. Introduction 2 2. ...»
Urban road pricing acceptance
To be presented at the IMPRINT-EUROPE seminar in Brussels
November 21 and 22, 2001
1. Introduction 2
2. Lessons from economic theory 3
3. Surveys and interviews 8
4. Experience with road pricing 11
5. Conclusions 15
At noon on Thursday February 1, 1990, a cordon toll system opened in the city of Oslo. At the same time a road tunnel, financed by the expected toll revenues, was opened under central Oslo. With one stroke 50 000 cars disappeared from the inner part of the city, and within a year, the number of accidents had dropped by almost 20 percent. The initial success of the tollgate system has continued. The main aim of the system was to finance road investments and the road network has been extended more or less according to the original plan. And the fraction of people in favour of the toll system has increased substantially.
The idea of relating charging directly to the use of roads is over 300 years old. In the UK and the US many turnpikes were built as private or public toll roads from the late 17th to the middle of the 19th century. However, with the exception of Singapore, the Oslo toll system and similar toll rings in Bergen and Trondheim are the first compre- hensive road pricing systems implemented at the urban level. The examples set by the three Norwegian cities has certainly inspired and intensified the discussion about ur- ban road pricing among both politicians and experts. However, as yet no other cities have followed in their footsteps. What is hindering acceptance of road pricing?
This paper reviews and summarises what we do and do not know about the accep- tance of urban road pricing. The term road pricing is used in this paper to denote the entire spectrum of charging systems and goals. Road pricing may have flat or vari- able toll rates, conventional or electronic collection methods. Revenues may be re- funded to companies or governments, and may or may not be earmarked for road or other transport investments. The discussion is structured according to the four main sources providing us with knowledge concerning acceptability. The following section reviews what we can learn from economic theory. The third section describes the re- sults of various surveys and interviews. In the fourth section, we draw conclusions from cities that have discussed, planned or implemented road pricing schemes from the point of view of acceptance. A summary is provided in section five.
Most examples presented in this paper, particularly in section four, are based on ex- perience from the PRIMA project, a DG TREN-supported collaboration among 7 cities to investigate key factors in building road pricing acceptance. Cities studied included Stockholm, Oslo, Trondheim, Lyon, Marseille, Zürich, Bern, Rotterdam and Barcelona.
2. Lessons from economic theory Public acceptance is by no means a clear–cut concept. Assume some individuals perceive a policy measure acceptable but others do not. What criteria should then be used to decide if it is acceptable or not for the collective? It is not acceptable if the agreed upon decision criteria states that a measure should only be implemented if no one will lose and at least one gain. If instead the majority rule is applied the measure will be regarded as acceptable as soon as the fraction supporting it exceeds 50 percent.
What if the people and their political representatives have different views? If less than 50 percent of the affected population accepts a new measure, it might still be deemed as acceptable by the politicians in charge or rather by a majority of them. The Oslo toll system is a good example of such an ambiguity. In other cases a policy measure might be accepted by a majority of the citizens but considered to be unacceptable by the political level. Lowering gasoline taxes is an example.
There is also a time aspect related to acceptance. Changing values, new knowledge or a new “state of the world” may make a formerly unacceptable policy become acceptable and vice verse. In the Stockholm region a cordon toll system was planned and decided in the beginning of the 1990´s. The system was rejected in 1997 because of a changed political situation.
These ambiguities notwithstanding, economic theory provides us with some insight into what kind of individual acceptability reactions that can be expected. Economic theory assumes that individual road users are rational and base their trip decisions upon a comparison of the benefits and costs that they will receive from using the road. The costs considered by them normally do not include the congestion cost their travelling impose on other road users or the negative environmental effects their driving impose on a smaller or larger fractions of the population. Since these cost components are not considered, some trips will add more costs than benefits to the society and as a result the road system will be overly or inefficiently used.
The disregard of these “negative external effects” and the resulting inefficiency constitute the basic theoretical rational for road pricing. By introducing a charging mechanism directly related to the use of a road it is in principle possible to force all travellers to also consider the external costs. If such a charging system is implemented there will be both winners and losers. According to the fundamental economic welfare criteria, the Kaldor-Hicks criteria, such a change will increase the overall social welfare if the winners are able to compensate the losers and still be on the winning side. If each traveller has to pay a charge that equals the gap between the social and private cost of the trip it will be possible for the winners to do that.
Road pricing also has another type of impact. Except for the short-term influence on transportation flows other markets will be affected. In particular, the introduction of road pricing may induce households to change their location patterns, their workplaces, and their travel decisions as regards retail services, recreation etc. Location 3 adjustments will affect the markets for labour, land, retail services etc., as well as travel behaviour. Firms will also be affected in a similar way. In general, the interaction costs will be higher. This means that the road pricing system will make the location of some firms non-optimal, and that may in the end force some of them to relocate. In any case, the impact will include both that firms will have to pay a toll and that their locations may not be optimal more.
As shown by e g Rietveld and Verhoef (1998) and Johansson and Forslund (2000) the net benefits of travellers affected by a pricing scheme closing the gap between private and social costs can be assessed by dividing them into three groups. Those who stay and pay will on average become losers. They lose more than they gain because their toll payment exceeds the value of time they save (it has to in order to close the gap).
The second group includes those who change their travelling pattern –to another mode, route or time of day. They are obviously made worse off since they would not otherwise have changed their behaviour.
Those who use public transport and continue to do so risk facing a more crowded environment. This reduction in quality implies that they become worse off provided that faster or more frequent public transport (made possible by less congestion) does not outweigh the crowding effect.
It no doubt seems strange that all three groups will become worse off if congestion tolls are introduced. The explanation is that the toll collector is the main gainer of the pricing scheme. If all effects are added up the net benefit will be positive as expected.
A politician believing in the idea of road pricing and also interested in being reelected would immediately realise that the allocation of toll revenues to different uses is a strategic issue. It seems evident that everyone would gain if the revenues were in some way channelled back to the three groups of travellers. As long as we assume that all travellers are identical as regards their appreciation of a given reduction in travel time this means that it should be relatively easy to “sell” road pricing to the public. However, public acceptability will be more difficult to obtain as soon as we relax this simplifying assumption and instead assume that people differ in their valuation of travel time reductions. Eliasson (1998) has shown that this kind of heterogeneity makes it is very difficult to design a congestion pricing scheme that fulfils the two following requirements.
• More than half of those affected should be better off with than without the pricing system
• Aggregate welfare should increase for those involved Designing and implementing a congestion pricing scheme, which is optimal for the average individual and returning the revenues to the losers is not enough to fulfil the two requirements. A majority of the users may still be worse off. This is further explained by the more general problem indicated by figure 1. The figure indicates that a skewed distribution of the benefits resulting from a certain policy measure can cause a large variation in net benefits if all of them are charged the same amount. The curve in the figure shows the distribution of benefits ordered from the highest to the lowest benefit and the charge is indicated by the line A. The fraction of losers is larger than the fraction of winners in spite of the fact that the total gain is positive.
Figure 1. Distribution benefits versus charge As a consequence we can expect that politicians who think that people may have rather different values will find it more difficult to argue for road pricing.
It will be even more troublesome for those who care about equity and who believe that the value of time tends to be higher the higher the income level.
Rietveld and Verhoef (1998) note that political acceptability is not only related to the trade-off between total welfare gains and equity aspects; one should also consider the attitudes concerning the severity of environmental problems caused by road traffic.
Many people participate in elections even though the probability that their vote will influence the election outcome is exceedingly small. Using this observation, Rietveld and Verhoef argue that people judge political parties both according to their personal interest and according to social values such as caring for the environment. Their conclusion seems to be that there is some scope for environmentally friendly policies such as road pricing even if such a policy does not gain support from a majority of the citizens.
Uncertainty is another aspect with bearing upon the acceptability issue. If we leave the deterministic world of the economic textbook it seems clear that any classification of a policy as “efficient” or “fair” or “environmentally friendly” is bound to be uncertain. Keeping things simple such a categorisation can be illustrated as in figure 2.
Figure 2. Schematic classification system for transport policy measures The figure indicates that any (transport) policy proposal can be categorised according to the expected effect upon the objectives and according to uncertainty, that is according to the risk that the expected effects will turn out to be wrong.
It seems reasonable 5 to believe that political decision-makers aiming to achieve an efficient and sustainable transport system would tend to prefer measures having large positive effects to those expected to have small positive effects provided that the related risks were of the same size. And that they would prefer low risk to high risk measures provided the expected outcome was the same. Some of them would perhaps be inspired by the theory of corporate finance, or by their own experiences of owning shares and think in terms of portfolios or packages of measures rather than in terms of single measures.
They would ask their experts to compose and characterise a number of policy measure portfolios in terms of expected outcome and risk. Then they would think about the proper trade-off to make between expected outcome and risk. For this reason it should not come as a surprise if politicians now and then reject policy proposals advocated by deterministically inclined economists.
In the literature on acceptability, e g in deliverable 3 from the CUPID project, it is often stated that social acceptance of road pricing depends on both the “perceived” and the “real” benefits. It seems more fruitful to assume that ex ante assessments of benefits and costs almost always are associated with uncertainty either they are made by ordinary travellers, experts or politicians. Doing that it becomes important to try to find ways to handle the uncertainty in a rational way.
In most cities, major policy decision concerning the transport system will involve politicians representing different layers of government. As a result it will be difficult for the national government to impose an urban road pricing schemes without approval from the regional and local politicians concerned. National decisions of that kind may have been possible when centralised decision making was more common but other kinds of decision mechanisms are needed in an environment characterised by increasing decentralisation and by growing demands for influence from local and regional politicians.
Likewise it is not possible for politicians at the local and regional level to introduce road pricing without approval from the national government. In most countries the urban areas lack the legal authority to introduce road pricing autonomously. The often held opinion that the cities first have to deliver a concrete proposal for road pricing before changing legal backgrounds at national level is not convincing. The incentive for cities to invest in planning road pricing systems is very low if they do not know when (if ever) national legislation will allow them to introduce such measures. The necessity of political negotiations across layers of government and between different parties as well as the need to consider voter reactions mean that a long process of actions and interactions is needed before implementing an urban road pricing system.