The Scottish Government

« Previous | Contents | Next »

Listen

EXECUTIVE SUMMARY

1. The primary objective has been to conduct an international literature review on the costs of traffic congestion. This included the following sub tasks; to describe congestion within Scotland, to review definitions of congestion and how it has been measured, to describe the methods used to measure congestion costs and finally to provide an outline of the literature concerning the link between economic growth and congestion ('decoupling').

2. Limited literature exists on the locations of congestion in Scotland and this does not define congestion. The approach here was to use existing data on the impacts of congestion (delay, speed reductions and reliability problems) to describe the locations where the impacts of congestion are greatest. A broad picture emerges:

• Whilst at the national level only 11.5% of trips are affected by congestion, this figure disguises large geographic, temporal and journey purpose variations.
• Congestion impacts are largest in the cities of Glasgow, Aberdeen and Edinburgh, where up to 42% of AM peak travellers experience congestion related delay.
• The trunk road network that experiences the most congestion is that in the vicinity of these cities, plus the approaches to the Forth estuarial crossings.
• Peak hours are more congested than the off-peak. Commuting and business related trips are more affected than trips for 'other' trip purposes. No data is available on congestion impacts for freight movements.
• Congestion related delays are reported throughout Scotland, beyond Aberdeen, Glasgow and Edinburgh and their vicinity. The frequency and incidence is, however, higher in the large cities.

3. Despite frequent use of the term, congestion is often understood but not formally defined. Perceived congestion may be as important as more objective evidence in driving the need for policy measures. The definition given by the Highways Agency ( DMRB, 1997) captures the wide understanding of congestion as:

'the situation when the hourly traffic demand exceeds the maximum sustainable hourly throughput of the link.'

4. According to Goodwin 2004:

'Congestion is defined as the impedance vehicles impose on each other, due to the speed-flow relationship, in conditions where the use of a transport system approaches its capacity'.

5. These two definitions reflect the two fundamental approaches to interpreting congestion: firstly a 'traffic engineering' perspective (which underlies many measures of congestion) and secondly an economic view (related to principles behind marginal costs of congestion). At the practical level of measuring congestion, approaches are classed as travel time (or speed) based measures, volume based measures, area based measures and summary indices (or more complex model outputs). In practice, the simpler measures are more commonly applied than relatively complex measures.

• A commonly applied measure divides the 'total delay' by the 'volume of traffic' to give the 'average amount of delay' encountered by a vehicle travelling one kilometre. Delay based measures, however, disregard vehicle occupancy, values of time and other factors ( e.g. environmental impacts resulting from congestion).
• Simple measures based on speed are used particularly for a motorway context (for example, 'a congested state exists when the traffic speed is below 50km/hr').
• A more complex measure is the 'congestion reference flow' (Highways agency, 1997), based on capacity, number of lanes and other traffic related variables (junctions are considered separately to links).
• The 'level of service' indicator is a basic congestion scale running from A to F and describes operational conditions on a route or section (using variables such as speed, travel time, disruption to flows and safety). It is widely used in the USA.

6. To measure the costs of congestion, research shows three economic terms that can be used; the Marginal External Cost of Congestion, the Total Cost of Congestion and the Excess Burden of Congestion. These are summarized below.

• Marginal cost refers to the change in total transport network costs for a single additional trip (or vehicle-km). Related concepts are s hort run marginal costs (assuming capacity is kept fixed) and long run marginal costs (allowing capacity to be expanded). Marginal external costs are items of marginal cost that are not borne by the trip maker, ( e.g. for road trips they include road wear and tear, increased accident risk and environmental costs). A specific marginal external cost item is 'delay to other users', often referred to as the Marginal External Cost of Congestion.
• The Total Cost of Congestion gives the cost of congestion compared to a state of zero congestion. A frequently quoted figure is that congestion costs the UK economy £20 billion/ year (but there is no supporting evidence for this).
• The Excess Burden of Congestion compares the cost of congestion in the current traffic state to a traffic state that would be expected with optimal prices in place (optimal to maximising economic output). The Excess Burden of Congestion differs from the Total Cost of Congestion as it is highly likely (with optimal prices and an optimum level of baseline capacity) that congestion will be present on the transport network. It relates to a situation where capacity is fixed. Estimates of the Excess Burden of Congestion for the UK or at a city level have been produced and two major points emerge. Firstly, costs estimated by the Excess Burden of Congestion are substantial, but significantly less than those based on the Total Cost of Congestion approach. The second is that, in similar vein to the Total Cost of Congestion approach, there is substantial variation in the figures produced.

7. The appropriate choice of measure will vary according to the end use of the data. Where the aim has been to consider road pricing measures, the Marginal Cost of Congestion is normally calculated. To review the benefits of significant investment decisions, the Total or Excess Burden of Congestion may be calculated.

8. The Total Cost of Congestion is the easiest of the measures to calculate, but may have least policy relevance due to the cost of alleviating congestion. Calculations are based on either mathematical models (to estimate costs in the current state and the uncongested state) or actual measurements of vehicle speed to infer changes in journey time. The Excess Burden of Congestion gives a cost estimate that it is possible to address using transport policy. Unfortunately it is more complicated to calculate, requiring transport models that can estimate the impacts of road pricing. Estimating Marginal Cost and Marginal External Costs is not trivial as it is necessary to model how costs (travel time, reliability, etc.) change with an additional vehicle-km or trip. Four principal methods are based on link speed flow relationships, area speed flow curves, network assignment models and microsimulation models.

9. In terms of data requirements, all three approaches require some form of transport model (which may be static or dynamic) and estimates of the other impacts that congestion causes ( e.g. pollution, accidents, etc.). Other factors are:

• Marginal costs for each of these impacts ( i.e. for each additional trip) are also required (travel time, reliability, climate change, air pollution, noise, accidents).
• Empirical evidence suggests that the results are sensitive to the transport models used and the values used for the costs of the impacts.
• Transport models that provide estimates of junction delay in urban areas will give more robust results than those which do not, particularly as congestion costs are most significant in urban areas.
• Uncertainty in the values used for the cost of environmental impacts can significantly affect the final estimates of the costs of congestion.

10. With respect to breaking the link between transport and economic growth ('decoupling'), there is strong empirical evidence that growth in travel is related to income, the cost of travel and the 'need to travel'. The key issues are as follows:

• Where transport policy increases income and reduces cost ( e.g. by reducing congestion), other measures are needed to either prevent increased travel demand (for example road pricing to 'lock in' the benefits) or to reduce the need to travel. Some measures may be quite difficult to implement politically, such as road pricing.
• There is empirical evidence at EU level and internationally that decoupling has taken place over time, to a different extent for the passenger and freight sectors.
• Research has identified particular policies and instruments which could be used to promote decoupling whilst maintaining economic activity and achieving sustainability goals. These policies are likely to have a more successful impact if implemented together in packages.

11. The underlying relationships are, however, complex and further understanding of the demand for travel is needed before drawing firmer conclusions on the links between transport and the economy.

« Previous | Contents | Next »