www.curacaoproject.eu                      CURACAO - coordination of urban road-user charging organisational issues                   Funded by the EU

Road Pricing Context













Decision Making



Implementation and Evaluation



Case Studies






Dutch National Case









The Hague


Urban Road User Charging Online Knowledge Base

What Is The Importance Of The Theme?

What is prediction?

Prediction may be defined to include the full range of activities carried out in advance of implementation in order to aid understanding of the impacts of road pricing. In particular, it incorporates both empirical survey work and model-based studies designed to test the performance of scheme options.

A broad definition of prediction could also be taken to include the appraisal of road pricing schemes, such as the calculation of economic benefits and the assessment of wider impacts via multi-criteria approaches, through which the outputs of other empirical and model-based work are interpreted for policy-making. In this report, we have separated these areas out. Prediction is defined to include only those activities which focus on determining the impacts of urban road user charging, while appraisal, covered in Chapter 11, is defined to include activities which focus on assessing their value to society. This distinction is useful because the methodologies employed within these two sets of activities tend to be rather different, leading to separate implications and gaps in prevailing knowledge. The two themes are, however, inextricably linked since prediction information will partially be used for the purpose of scheme appraisal (as will be discussed later in this chapter).

How does it link to the objectives of road pricing?

Prediction has a tangential link to the objectives of road pricing, because it exists to aid understanding of how proposed schemes will perform against the objectives set. Figure 1.1 has shown how prediction fits into the issues considered within CURACAO, demonstrating that objectives, in conjunction with specifications of schemes (represented through the themes of scheme design, technology and Business Systems), represent the two key inputs to the prediction process.

While it would be desirable to be able to predict the performance of road pricing schemes with equal reliability for all possible objectives, in practice prevailing levels of knowledge and expertise tend to be skewed towards certain areas. Most modelling approaches currently used in transport planning are based on economic principles of behaviour, so are better equipped to predict performance against economic objectives such as efficiency and raising revenue than they are to shed light on human perspectives such as liveability. The coverage that alternative modelling approaches provide of objectives will be addressed in more detail in Section 6.3.

A particular concern regarding the relationship between objectives and prediction is the danger of optimism bias (Flyvbjerg, 2004). Experience suggests that planners may tend to assume the best case when predicting the outcomes of proposed schemes, leading to potential for underestimation of costs and overestimation of benefits. In the UK, this phenomenon has received formal government recognition (HM Treasury, 2003). In the case of road pricing, the cost of implementing appropriate technologies, collecting payments and enforcement are all areas where costs could quite conceivably be underestimated. It is also possible that predicted outcomes of schemes may be affected. In particular, reliance on modelling approaches based on economic principles of rational behaviour may naturally tend to suggest that economic policy instruments perform better than non-economic alternatives, due to the limitations of their assumptions. However, this is unlikely to be a simple situation to interpret because road pricing involves a trade-off between efficiency, in terms of modifying travel behaviour, and raising revenue, which may then be spent on providing better alternatives to the private car. Existing evidence for optimism bias in the prediction process for road pricing schemes will be assessed in Section 6.3. This issue is also likely to be important for appraisal of road pricing schemes, discussed in Chapter 11.

How important is it to decision makers?

Road pricing is being considered for widespread implementation because there is a large and longstanding body of research which suggests that it has the potential to improve the efficiency of road transport, reducing congestion and providing time savings to travellers while leading to economic and environmental benefits to society. As the number of real road pricing schemes that have been implemented is still quite small, much of the evidence to support this view comes from various forms of prediction, especially transport models. Therefore, prediction is likely to be central to the decision of an authority to formulate a road pricing policy and to play a key role in the technical process of scheme design and the political process of winning support for the policy among stakeholders. The robustness of prediction will have direct consequences for the appraisal of scheme benefits and for the identification of negative impacts that may require mitigation measures.

Using simple aggregate scoring, the UNAQ results have indicated that prediction ranks second among the themes considered, with only acceptability being rated as more important. However, these two themes are not mutually exclusive, because prediction is likely to be an important catalyst for or against acceptability. For example, predictions of improved travel conditions and enhanced environments might be expected to increase the acceptability of road pricing, while any doubts over the robustness of the predictive approaches being used to support such outcomes may tend to reduce it.

The emphasis placed on prediction in the UNAQ results may be viewed as an acknowledgement of the extent to which authorities rely on modelling approaches to help them choose the best policy options and to justify them, both as part of formal government processes and to their electorate / local stakeholders.

The role of politics in setting the background to prediction should not be overlooked. Although the aims of the prediction process, when viewed from within, are primarily technical, attempting to ensure that resources are used to best effect to improve the transport system, from the outside the role that prediction plays in decision-making is sometimes rather less clear cut. Demonstrating that a proposed scheme will broadly achieve the objectives used to justify its implementation is always likely to be a desirable prerequisite in a democratic society. But, the importance of being able to do so may vary dependent on issues such as the nature of the scheme, the ability of the prevailing approaches to provide accurate predictions and the range of other reasons that may influence the final decision on whether or not to implement. In the case of road pricing, a decision to commit resources to implementing a scheme is not irreversible to the extent that is the case when, for example, building a new road. If a road pricing scheme does not produce the desired benefits it can be modified or withdrawn. In addition, it is clear that there are known to be significant issues of uncertainty which may affect ability to produce accurate predictions. So, it follows that the main motivation for placing high emphasis on prediction (as shown by the UNAQ results), is the perception of policymakers that as much evidence as possible of the benefits of road pricing is required in order to gain sufficient acceptance from politicians and a sceptical public. The initial implementation of road pricing is essentially a short-term political risk rather than a long-term planning commitment. The differences that have been observed in approaches to prediction during the PRoGR€SS (PROGRESS,2004) project between UK and Italian cities, and the role that prediction has played in Stockholm, point to significant variations in the political decision-making culture within the EU that affect the prediction activities undertaken and attitudes towards outputs. Further discussion of these political issues are outside the scope of this chapter but it is important therefore to bear in mind the context upon which decisions are made.