A note o the electronic road pricing in Singapore

Road & Transport Research, Jun 2000 by Tay, R

Traffic congestion has been a major problem for many cities around the world and the conventional solutions adopted by traffic planners tend to ignore the role played by pricing as an alternative to investment, leading to expensive solutions that might not be efficient or economical (Lam 1988). This is particularly true in western countries such as the United States where urban roads are grossly overbuilt and inefficiently priced (Keeler and Small 1977). As traffic congestion becomes increasingly serious in some Australian cities, the article by Luk, in an earlier issue of this journal, provides a useful review on one of the few attempts at road pricing in the world. Even though there are several reasons for implementing road pricing, the publicly stated objective of the Singapore Electronic Road Pricing (ERP) scheme has always been to reduce traffic congestion, which implies that the ERP scheme in Singapore is in fact a congestion pricing scheme. Unfortunately, the economic efficiency and welfare implications of the ERP scheme in Singapore were beyond the scope of Luk (1999). This important topic nonetheless has been examined previously and will be highlighted here.

Transport economists have long emphasised the need to use congestion pricing as a tool to alleviate traffic congestion. The objective nevertheless is not to achieve free flow' traffic, as many transport planners desire, but to reduce the inefficiency created by over consumption of undervalued trips. Traffic congestion is a classic example of what economists call consumption externality and its solution, congestion pricing, is an example of the Pigouvian tax. As early as 1920, Pigou argued that motorists entering a crowded road network not only incurred a cost to themselves but also raised the travel cost of existing users. The latter cost, known as externality or external cost, increases the marginal social cost of an additional trip above its average cost. In making their decisions, motorists consider only their own private cost and will continue to take additional trips until the marginal private cost of trips equals the marginal benefit. This equilibrium is shown in Figure 1 at X trips and is not the socially optimal level.

In economic theory, resources are allocated efficiently when the cost of the last unit of a good or service, known as marginal cost, is equal to the marginal benefit or the benefit derived from its consumption. As shown in Figure 1, the socially optimal consumption is Y trips. Therefore, without government intervention, an excessive number of trips are taken which leads to an inefficient allocation of resources. To correct this inefficiency and induce motorists to make only Y trips, Pigou (1920) suggested that motorists using congested roads be charged an amount equivalent to the difference between the average cost and the marginal cost, a concept that was subsequently refined by Vickrey (1967) and Walters (1968).

Despite being straightforward conceptually, the implementation of road pricing has been hampered by both technological and political constraints. McCarthy and Tay (1993b) provides a brief summary of attempts in ERP around the world. One of the most widely cited cases is the Singapore Area Licensing Scheme that has recently been converted to the ERP scheme. Although considered by many traffic planners, including Luk (1999), as a successful model in reducing urban traffic congestion, the Singapore ERP scheme received much less support from transport economists.

The problem lies in the different interpretation of what an optimal level of traffic is. The conventional wisdom of many traffic engineers and planners is to use tolls as a means of reducing traffic flow to an acceptable level prescribed by engineering design (for example, average speed in the case of Singapore's ERP), irrespective of the cost and benefit associated with the number of trips. Unfortunately, this level, depicted by Z in Figure 1, is usually lower than the socially optimal level. As a result, the road system is under-utilised and resources are allocated inefficiently. Wilson (1988a, 1988b) and McCarthy and Tay (1993a,1993b) found that the congestion toll in Singapore was set too high, resulting in a net loss of welfare to motorists.

Therefore, using price elasticity as a guide to set tolls to achieve the desired outcome prescribed by engineering considerations (travel speed), as proposed by Luk (1999), will more than likely create a negative impact. In order to ensure an improvement in efficiency and an increase in welfare, congestion tolls should be set using estimates of the congestion externality. It is heartening to see that the Singapore government has reduced the congestion toll but it has also increased the number of time periods and enlarged the areas affected by the ERP scheme. More research should be conducted to estimate the congestion externalities on these sections of the road network during the time periods when the scheme is in operation to assist policy makers in setting the correct congestion price.