We are testing a new beta website for gov.scot go to new site

Costs of Congestion: Literature Based Review of Methodologies and Analytical Approaches



8.1 One of the issues surrounding congestion and any measures that are taken to alleviate it is that of the relationship between economic growth and transport. It may be argued that reducing congestion and promoting economic growth are conflicting objectives - however the relationship is complex and a growing body of research has sought to provide both evidence on the relationship and propose measures to decouple transport growth and economic growth. Following the objectives set down in the EU White Paper ( CEC, 2001), there has been increasing acknowledgement of the need to break this link, however, whilst decoupling was not the primary focus of the White Paper, it formed a headline objective for the EC Sustainable Development Strategy ( CEC, 2001a). Despite this, it should be noted that decoupling is still not wholeheartedly embraced by all policy makers. As reported in Tight et al, 2004, whilst some parties believe it is not feasible in practice, there is a train of argument that market forces should just be left to prevail. Alternatively, those with a strong concern for the sustainability agenda believe that the continued promotion of economic growth is misguided.

8.2 The starting point for considering the need for decoupling is a strong evidence base of the links between road traffic demand, income and generalized cost. Work by Graham and Glaister, 2004 provides empirically observed elasticities that indicate that if congestion is reduced, then there will be a tendency for transport demand to increase. If incomes increase then there is also a tendency for vehicle kilometres (transport demand) to increase. A summary of the evidence in their paper is as follows. In terms of responses to changes in travel time, Car trips had a short run elasticity of -0.6, (Long run = -0.29) whilst Car veh-km short run elasticity was -0.74 (long run = -0.20). With respect to changes in income, Car veh-km had a short run elasticity of 0.3 and long run elasticity of 0.73. In the case for Scotland, work by Laird (2006) on wage rages and commuting in Scotland supports these findings. It is against this empirical evidence of the link between traffic demand, income and cost that research into the potential for decoupling has been carried out.

8.3 The concept of Transport Intensity is commonly used as an indicator of the relationship between the level of transport activity and the level of economic activity, defined as the ratio of 'gross mass movement' to GDP. In practice, this is often separated into passenger and freight intensity, using passenger kilometres and tonne kilometres respectively. The indicator may be expressed as elasticity, for example showing the ratio of percentage change in passenger kilometres to the percentage change in GDP over a period. This may be viewed alongside other related measures, such as the link between transport environmental impacts (see for example Tapio, 2005) or efficiency aspects (for example technology, organisational factors, see Bannister and Stead, 2002).

8.4 The EU White Paper agenda led to research at EU level, typically to produce historical evidence of decoupling (for example Tapio, 2005) and to propose measures that might be used to achieve decoupling (for example Tight et al, 2004). Tapio (2005) explores various definitions of decoupling using the transport intensity indicator as a basis and introducing concepts such as 'weak decoupling', 'strong decoupling' and 'recessive decoupling', depending on the direction of change and size of percentage changes. The research also extends the notion of decoupling to look at the relationship between economic growth ( GDP) and road traffic emissions ( CO2), with the proposition that decoupling may also take place between economic growth and the environmental impacts of traffic.

8.5 Tapio also presents quantified decoupling evidence for the EU15 countries, based on EUROSTAT and IEA statistics for 1970 to 2001. The results for road passenger growth indicate transport volumes closely followed GDP in the 1070's, exceeded GDP growth in the 1980's and grew rather slower than GDP in the 1990's. For freight a different pattern is presented - freight traffic volumes followed GDP growth in the 1970's, fell below GDP growth in the 1980's and showed a clearly higher growth than GDP in the 1990's. A more detailed individual country analysis is also reported. This is in contrast to similar analysis for the USA (Bannister and Stead, 2002), which indicate that in the US freight sector the tonne-kilometres carried have increased at a rate well below GDP, particularly, since 1985.

8.6 The general findings support the earlier work of Tight et al (2004). This also presented a short overview of evidence of decoupling in the EU15 context, but focused more on the potential for different transport measures to contribute towards breaking the link between transport activity and economic growth, through reducing travel demand, maintaining economic growth and enhancing environmental quality. In terms of the factors that may be used to explain or influence decoupling, some historical explanations for the case of Finland are given by Tapio, including the high cost of car purchase, income changes, green urban lifestyle and impacts of technology. The role of particular transport instruments formed the core element of the research by Tight et al, however, which gathered evidence on the potential effectiveness of instruments from experts across the EU and some international bodies. This was carried out using a questionnaire and panel group meetings - whilst some of the evidence collected had a subjective element, substantial parts was based on case studies, previous work and similar quantified evidence. Thirteen of the most promising measures were studied in detail, reporting their potential impact on transport intensity, environmental load, CO2 emissions and 'possible unexpected effects' - a 'reality check' with the expert panel was also included. Quantified evidence is given based upon specific country or local experience, but with an approximation to the EU-wide level (alongside acknowledgement of the difficulties in achieving realistic figures at that scale). The prevailing outcome was that packages of instruments would hold the greatest promise for decoupling, however the seven individual instruments emerging (in no order) were: urban road pricing, hydrogen fuel cell vehicles, controlled parking zones, car sharing as part of combined mobility, high speed rail, road pricing for freight traffic and combined measures relating to traveller attitudes/traffic behaviour.

8.7 According to Bannister and Stead (2002), the basic relationships between transport and economic growth are, however, far more sophisticated interdependencies. As a result, their work starts with the proposition that transport efficiency (reflected in modes, technologies, use of resources, prices and organisational structures) should be considered alongside the more traditional measure of transport intensity. In addition, they propose that the measurement of GDP should be extended in the production of indices. Illustrative analysis is presented for EU countries, giving summary indices for the EU alongside similar measures for the USA and Canada.

8.8 In discussing the basic interdependencies, the work starts from the findings of the influential SACTRA report ( SACTRA, 1999), which was primarily concerned with understanding the link between transport and the UK economy, but also examined transport intensity. Bannister and Stead argue against the hypothesis that ultimately traffic intensity will decrease without intervention, as a result of the relationship between travel, car ownership and income. That hypothesis is based around traffic forecasts which are driven by growth in car ownership, not the distance travelled per vehicle. Car ownership forecasts are, in turn, driven by income levels and therefore related to GDP. The supposition is that car ownership levels will reach saturation level whilst income continues to growth and thereby intensity will decline in future. This is problematical - as SACTRA (1999) also states:

' the cross sectional evidence suggests that there are substantial differences in car use, which are not related to either car ownership or income'.

8.9 As Bannister and Stead therefore propose:

' Income may be less important than other factors in driving the growth in travel…a clearer understanding is required for the motivations of car use apart from the costs. This could be a fruitful area of research in different national settings'

8.10 To summarise, there is strong evidence that growth in vehicle-kilometres is a function of income and travel impedance or generalised cost as well as 'the need to travel'. Clearly transport policy that increases incomes and reduces travel impedance ( e.g. reducing congestion) has to use other measures to prevent an increase in vehicle demand ( e.g. road pricing can lock in the de-congestion benefits) or reducing the need to travel. Some of the measures needed to prevent the increase may be quite difficult to implement politically, such as road pricing. Despite this, evidence at EU level and internationally has suggested that decoupling of transport growth and economic growth has taken place historically, with differences seen between the passenger and freight sectors. Whilst the statistical relationship cannot give definitive evidence on causation, research has identified particular instruments which could be implemented to promote decoupling, seeking to maintain economic activity and achieve sustainability goals. These instruments are likely to have a more successful impact if implemented in packages. However, the underlying relationships are complex and further understanding of the demand for travel is needed before drawing firmer conclusions on the functional relationship with the economy.