Urban Congestion Charging: A Comparison between London and Singapore

Abstract

The congestion charging schemes in London and Singapore are compared and assessed in the light of guidelines set out in the Smeed Report, published by the UK Ministry of Transport in 1964, and their performance in reducing congestion and raising net revenue. The aim is to draw lessons for other towns and cities considering the introduction of congestion charging. One important result from Electronic Road Pricing in Singapore is that a per‐entry charge is more effective at reducing congestion than a per‐day charge. It is concluded that although Electronic Road Pricing in Singapore is more in line with the desirable properties outlined in the Smeed Report, both schemes are part of a wider package of transport policies, and that is probably the most important reason for their success. The main lesson for other towns and cities around the world considering the possibility of introducing congestion charging is that any such scheme ought to be accompanied by complementary measures that will provide motorists with a valid alternative to the car.

Acknowledgements

The author gratefully acknowledges support from the British Academy whilst she was at Cambridge, where most of this study was conducted, and from the Rees Jeffreys Road Fund, that supports her at Oxford, where the study was finished. Thanks are also due to Lisa Ward and Wai Wing Li, with whom she worked on some aspects of this paper, to Kian Keong Chin and Eddie Lim, from the Land Transport Authority in Singapore, to Tony Doherty from Transport for London, and to two anonymous referees for very helpful suggestions.

Notes

1. ASC cost is equal to marginal private cost. The assumption consistent with this is that road users are atomistic price takers. As the price they consider includes time costs, one could rephrase this as ‘price and speed takers’. The marginal private cost they consider (the cost considered by the marginal user when contemplating whether or not to use the road) thus includes time costs and vehicle operating costs for their trip given the level of road use (including themselves, but as they are atomistic, this of course does not matter). This is equal to the costs incurred by every user, and hence the average (social) cost.

2. This is specially important inside the European Union where as of 2004 there is still no harmonized system for electronic toll‐paying on motorways. Each country has a different system, with the result that vehicles need to be fitted with different electronic equipment for cross‐border journeys. The desirability of having the same road‐pricing technology in all urban areas in a country, however, may be outweighed by the results of a cost–benefit analysis, which may favour different options in different towns. The same argument would apply to urban areas inside the European Union.

3. There are daily credit facilities, as the charge can be paid after the trip has been completed as long as the payment takes place before the end of the day. However, there are no weekly or monthly credit facilities.

4. Information provided by TfL on request.

5. Li (1999) uses the wage rate for car owners to estimate the value of time. He then computes the theoretical congestion toll as the difference between MSC and ASC at different levels of traffic, using that value of time. He concludes that for the levels of traffic of 1990, the theoretical congestion toll computed with the value of time for car owners was around S$3, which was exactly the price of an area licence for cars at peak times.

6. In the case of RPS on the expressways, the licences also offered an unlimited number of passages through the RPS manually controlled points.

7. This fixed S$10 fee was computed by LTA by taking into account the cost of renting a temporary IU and the average ERP charges incurred by a foreign motorist each day.

8. All the data on costs in this section were kindly provided by LTA on request.

9. It may also prove convenient to wait for in‐vehicle technology until a UK‐wide system has been specified.

10. The average travel rate is the average time taken to travel a certain distance (e.g. the average time in minutes to travel 1 km). It is the inverse of speed.

11. Although TfL expected underground patronage would increase, this did not happen. Underground usage across London and specially in Central London decreased. The reasons for the decrease in passenger levels on the London Underground are probably linked to the slowdown of the economy and the decrease in tourism in London, which in turn may be linked to the war in Iraq (TfL, 2003a). In addition, the Central Line was temporarily closed for almost 3 months following a derailment at Chancery Lane station in January 2003.

12. Holland and Watson (1978, p. 15) note that “the Government had no previous experience to guide it in setting the licence fee, thus it was necessary to set the fee by judgement”.

13. They compute the congestion externality with the standard formula , where q is flow in vehicles per lane‐h and c is average cost per trip in S$. They use data from a survey conducted by the Public Works Department in Singapore on traffic speeds and volume in the RZ, and assume the value of time to be 50% of the wage rate.

14. Information provided by LTA on request.

15. Information provided by LTA on request.

16. Information provided by LTA on request.

17. Congestion Charging Seminar, organized by the Institution of Highways and Transportation, Imperial College London, UK, 19 March 2003.