Road Pricing Context
Implementation and Evaluation
Dutch National Case
Urban Road User Charging Online Knowledge Base
What Is Known About The Theme?
How is Transferability defined?
In general academic literature, a distinction tends to be drawn between the concepts of generalisability and transferability. Generalisability is normally defined as the extent to which data obtained from a particular population, under unique study conditions, at a particular point in time and space can be applied more widely to other populations, conditions, times and spaces. So, in the case of road pricing, assuming that the results of a limited trial provide data that can be used to predict impacts elsewhere, be it expanding the trial to the full city / region or to another city or country, would be classed as generalisation. Transferability is normally seen as a subset of generalisability, which applies to more specific occasions where we may be attempting to generalise research results to a context in which they were not studied. This is most commonly found in the desire to generalise results from one location to another (e.g. from one city to another), but it could potentially apply to any critical dimension that limits the meaning of research findings. The main point is that transferability analysis focuses on identifying contextual differences and dealing with their impacts.
What Approaches to Transferability have been adopted in previous studies?
Transferability has frequently been addressed as a theme within individual transport research projects and has sometimes been a significant focus of the work undertaken. However, there have also been research projects which have investigated the potential for implementing road pricing schemes that have not considered transferability in any significant or formal manner. Therefore, experience in this area is somewhat variable.
The most significant work on transferability within transport policy may be that carried out by the TRANSPLUS project (TRANSPLUS, 2003), which built on work carried out by Rose (2001) to produce a series of guidelines for how cities might best learn from previous experience. The concept of transferability was subdivided into three typologies:
• territorial transfers, described as “horizontal” and covering the full range of situations where experience may be transferred between different locations
• transfers between institutions at different levels, described as “vertical” and covering issues relating to the scaling up or scaling down of policies
• transplanting of institutions and related competence instruments, covering situations where changes of institutional structures may be required.
While the first of these areas may be the one most commonly associated with road pricing schemes, the role of institutions and institutional structures in the successful implementation of urban road pricing has also been seen as important. For example, in the AFFORD project, it was concluded that urban road pricing schemes are likely to be easier to implement in cities where decisions about the transport system are taken primarily at the urban regional level rather than rather than based on negotiations between smaller urban districts (Milne et al, 2001). The subsequent experience of the political processes in London and Edinburgh may tend to support this idea.
The practical consideration of transferability within the TRANSPLUS project focuses primarily on mechanisms for fostering the transfer of desirable innovations, so turning transferability into an explicit objective of transport policy rather than a technical feature to be studied and understood. It identifies four levels of activity in the transfer of innovations from one location to another: collaboration, networking, dissemination and osmosis. This appears to assume implicitly that territorial transfers dominate and underlie most institutional issues. It proposes knowledge sharing, through mechanisms such as benchmarking, as potential approaches for increasing the transfer of desirable innovations.
The DANTE project, which was primarily based on interviews with decision makers and implementers in nine case study cities, considered transferability as part of assessing the potential for implementing travel reduction strategies. Most of this work involved identifying and categorising barriers and considering the extent to which they could be overcome. Five explicit categories of barriers were defined:
• resource barriers
• institutional / policy barriers
• social / cultural barriers
• legal barriers
• side effects.
Pricing and taxation measures were considered to have the greatest probability of barriers occurring but, interestingly, one of the lowest levels of seriousness in terms of issues that may not be able to be overcome. In contrast, barriers to land use planning measures, which were also considered to have a high probability of occurring, were considered the most difficult to overcome. The approach to transferability involved simply creating a checklist to help identify the most appropriate choice of measures and to assess the barriers that may arise. As in the case of TRANSPLUS, transferability is seen as being a high level policy objective, rather than an outcome.
In the UK, The Commission for Integrated Transport commissioned a report focussing on transferability issues in European best practice, affecting the delivery of integrated transport (W S Atkins, 2001). However, this comprises little more than a very general consideration of barriers and an assessment of the extent to which they may prevent certain perceived transport policy success stories being repeated in the UK.
Most of the key research studies about urban road pricing involving particular cities (eg TRANSPRICE, CUPID / PRoGR€SS, EUROPRICE) have provided little more than an informal commentary on transferability issues. Consistent with the above discussion of the importance of the theme, these studies have tended to focus very much on the individual perspectives of cities and have been considerably weaker on generalisation issues. However, some more academic studies (eg IMPRINT-EUROPE, (Kendzia and Korver, 2003)) have given explicit consideration to barriers, which suggests that they may have considered transferability issues implicitly when comparing different locations / contexts.
An exception here is the SPECTRUM project (Timms et al, 2005), within which transferability was a significant theme. This work, drawing upon Richard Rose and TRANSPLUS, emphasised the nature of transferability as a learning process. In particular, it distinguished between two approaches to studying transferability:
• an ex-ante or prospective assessment of transferability, which attempts to anticipate under what circumstances, and to what extent, a programme or policy that “works there” will also “work here” (Timms et al, 2005)
• an ex-post or retrospective assessment of transferability, involving an historical account of how policies were identified in exporter jurisdictions as worthy of emulation, and how they were applied in importer jurisdictions (Timms et al, 2005).
While it may be inevitable that current studies about urban road pricing schemes will tend to focus on the ex-ante perspective, due to the general lack of operational schemes and the future-orientated nature of such projects, the ex-post approach should not be forgotten and may provide valuable insights as opportunities emerge.
The analysis of urban transferability issues in SPECTRUM was undertaken by comparing predictive case studies carried out during the project with results from CUPID / PRoGR€SS, via construction of a logical framework, that attempts to encapsulate the features of instruments tested and the characteristics of cities. However, in practice, the scope of the exercise was quite limited. While it is encouraging to note that the results from SPECTRUM tend to be consistent with the recommendations from the PRoGR€SS cities, the analysis is ultimately rather vague and unable to lead to many powerful conclusions.
The METEOR project (METEOR, 2006) has addressed transferability issues for the full range of policy measures, including road pricing, being considered by CIVITAS cities. Similar to TRANSPLUS and DANTE, the focus has been on attempting to improve transferability and overcome barriers. The subtext within METEOR appears to be an assumption that many transferability problems originate from a failure to understand and follow best practice, so the emphasis is on defining logical steps and guidelines for decision-makers to follow. It is suggested that any transferability process must go through four phases:
• a demonstration phase where best practice is identified in the originator city;
• a transferability phase where the compatibility of that best practice with the needs of the receptor city is appraised
• an assessment phase where the specific barriers amenable to change and factors of success are identified in the receptor city; and
• an implementation phase where the good practice is implemented in the receptor city.
Linked to these phases, ten logical steps are defined for actions in receptor cities:
(i) diagnosis of the problems requiring action;
(ii) characterisation of the city context;
(iii) analysis of the city context and of implications these may have for problems identified in (i);
(iv) search for similar contexts;
(v) selection of examples of sources from which to transfer;
(vi) identification of measures with potential to transfer;
(vii) packaging and dimensioning of measures to be transferred;
(viii) ex-ante assessment of measures;
(ix) identification of need for adjustments; and
(x) implementation of measures.
METEOR goes on to consider issues of packaging for specific policy measures, identifying public transport measures and clean fuel technologies as areas that have the potential to enhance the ability of road pricing to meet its objectives. A detailed assessment of drivers of and barriers to policy implementation is also carried out, during which road pricing is presented as particularly sensitive to political, institutional, technological and acceptability related contexts. The relationship between transferability of policy measures and a range of other contextual variables is considered, including city size, degree of urban sprawl, level of public support, the quality and density of public transport networks etc. The emphasis of the METEOR approach is on identifying relevant contextual variables, understanding the nature and strength of the relationship with transferability and, if possible, making contextual adjustments to improve the transferability relationship. However, all the discussion is very general and little evidence is provided to demonstrate how this may lead to positive outcomes in practice.
Thus, while a number of studies have developed approaches for addressing transferability and have considered it in the context of urban road pricing, this is an area that is still in its infancy and there is currently no single accepted methodology. Those approaches that have been tried have tended to focus on attempts to improve transferability in practice rather than on increasing academic understanding of the processes at work. Even so, what has been achieved to date appears to consist more of tentative ideas than hard evidence and purposeful actions.
What Evidence Exists About Transferability from Real Urban Road Pricing Applications?
Beyond the environment of research projects, a number of common observations have been made about the transferability of evidence from real urban road pricing applications. In summary:
• Singapore: It is widely understood that the city-state and island geography of Singapore has made adopting road pricing much simpler than in other locations, where competition from neighbouring cities may be a significant factor. In addition, the unique cultural setting is regularly cited as having reduced stakeholder resistance compared to what might be expected in Europe.
• Norwegian Toll Rings: The longstanding history of toll payments for using transport infrastructure in Norway is widely considered to have helped achieve acceptance for the principle of urban toll rings. In addition, the sole objective of raising revenue for transport infrastructure projects, while not attempting to reduce travel, has allowed policy-makers to sidestep many of the controversial issues relating to demand management and to justify the tolls on the basis of new infrastructure from which travellers benefit. So, acceptability experience may not be so easily transferable, although the Norwegian cities have been facing a unique acceptability problem when the initial infrastructure funding justification for pricing has run out.
• London: It is generally accepted in the UK that London’s size, the scale of its public transport network and the very high proportion of journeys to central London by rail (around 70% prior to the implementation of congestion charging) make it atypical. On the one hand congestion is seen to be more serious there, making the case for charging stronger. On the other, it was much easier to accommodate the diverted journeys on bus and rail services. The dominance of central London’s economy will also have diluted the threat to economic activity (May, 2006).
• Durham: At the other end of the spectrum, the single road scheme in Durham is also atypical, though there is some interest in replicating it in rural tourist attractions (May, 2006).
• Stockholm: Stockholm is more typical, in terms of size, of many European cities. However, its topography makes it uniquely suited to cordon charging, since it limits the number of charging points.
Of course, just because a road pricing application has certain features that are atypical does not mean that little can be learned from it or that experience in other areas cannot be generalised. The comprehensive monitoring of impacts that has been carried out in London and Stockholm has provided probably the best real evidence to date about many aspects of urban road pricing, including effects on travel behaviour and network conditions, impacts on economic activity within and beyond the charged area, and the logistical effort and costs involved in operating the system (TfL, 2008; Stockholmsförsöket, 2006a).
General conclusions that may be drawn from the small number of urban road pricing schemes that have been implemented include the fact that none of them has, so far, been viewed widely as a failure because they have all gone a significant way towards achieving their principal objectives. Given that experience covers a range of situations and objectives, even such a limited sample suggests that pricing policies can be a useful tool and have the potential to be applied more widely. Moreover, the discussion of comparisons between predicted outcomes and actual experience in Chapter 6 suggests that predictive models are able to reflect quite reliably the resulting changes in behaviour and in traffic conditions. The ability to transfer such models will depend on the consistency of the underlying behavioural response relationships (eg values of time and elasticities).
No information on this theme is currently available from the case studies