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 Are The Key Features And Examples Of Scheme Design?

General considerations

Scheme objectives and constraints, as outlined in the previous chapter, clearly have a strong bearing on the details selected for scheme design. A reduction in congestion has been the main objective in the examples we have seen of large scale implementations in Singapore, London and Stockholm. The raising of net revenues for use on new transport infrastructure or public transport is often a strong second objective, and indeed the first objective in the examples of the Norwegian toll rings.

Elements of scheme design have strong links to objectives, as well as acceptability. This can be exemplified by the Edinburgh case, where it was clear that the control of congestion and environmental improvement were important consequences of URUC, particularly from the point of view of public acceptability of the measures (Catling, 2001). For example, if car users were charged for outbound journeys during the morning peak, there could well be a problem of public acceptability, even though this might make total sense from the objective of revenue- raising. Similarly, there might be acceptability problems if there were the same charge imposed on an inter-peak journey away from the city centre as for a commuter trip to the heart of the city during the morning peak, even though this might make perfect sense for revenue-raising.

There is a strong link between scheme design and the cost of operation. Hence it is advisable to consider some business modelling early on in the scheme design process. This would allow the various scheme options to be assessed in line with the objectives that have been set and assist with selecting the best solution. It would also assist with the selection of an appropriate range of charges and exemptions.

Key characteristics of road pricing schemes include type of charging regime, the level and structure of the prices, where and when the prices are levied, and the extent to which prices and/or exemptions are allowed to vary by class of users (DfT, 2008).

Type of charging regime

URUC schemes can broadly be classified into four types:

• Point based charges  (e.g. tolls to cross a bridge or to enter a section of motorway)
• Cordon based pricing: A charge is levied for crossing a cordon, and may vary with time of day, direction of travel, vehicle type and location on the cordon. There could be a number of cordons with different prices.
• Area licence based pricing: A charge is levied for driving within an area during a period of time. The price may vary with time and vehicle type.
• Distance or time based pricing: Price is based upon the distance or time a vehicle travels along a congested route or in a specified area, and may vary with time, vehicle type and location.

Point based charges are reasonably commonplace, but they are generally limited to specific small locations and not spread across the network. In modern times, estuary crossings in the UK such as the Forth Bridge, Severn Crossing and the Dartford Tunnel all involve some form of point based charges.

Cordons are simply combinations of point-based charges located to form a continuous boundary around an area. Cordon schemes are implemented in Singapore, Stockholm and several Norwegian cities, and are the most common form of electronic road pricing. Their main advantages are their flexibility in having variation in charges by time of day and vehicle type, and that each individual trip made into the area can be subject to a charge.

Two concentric cordons, like the inner and outer toll ring proposed for Edinburgh (Saunders, 2005) are an example on how simple cordon based pricing can be expanded to a multi-zonal system. Conceptually, more than two cordons could encircle a city centre, and charges could be varied according to a range of factors.

It is possible to combine point based charges with a cordon or area pricing system such as in Singapore where there are charges for expressways as well as for crossing into the Central Business District.

Area licensing or entry permit charges on the other hand, typically have charges that are levied on a daily basis. There is thus no incentive to limit the number of daily trips once the licence has been purchased. Their main attraction is that they are simple to understand and straightforward to implement (ITS United Kingdom, 2007). The London Congestion Charge is the most well known example of this type of charging regime. Users pay a daily charge to enter or be within the charging zone, and they can enter and exit as many times they like during the day. The charge is operational between the hours of 0700 and 1800 Monday through Friday.

Multi cordon or zonal based charging, like the scheme implemented in Trondheim in 1998, could be viewed as a stepping stone towards distance based charging. However, the Trondheim scheme had charging for crossing a zone boundary in one direction only, and charging was limited to a maximum of one per hour. Obviously, in such a scheme the charge could be levied to every vehicle crossing the boundary, both inwards and outwards. The charge could be varied depending upon factors such as time, location and vehicle type. Neighbouring urban sub regions would be encircled by cordons, and trips taking place within a cordon would be uncharged. There are currently no examples of full scale implementations of distance based charges in urban areas. The technology needed for this type of charging regime is however already well advanced, see Chapter 4.

Parking controls and environmental zones are alternative options for obtaining car restraint and environmental protection of central urban areas. For instance, on- or off-street parking in Central London can cost up to € 5 per hour, and trip end restraint of this scale can represent a far greater disincentive to car usage than the € 10 daily Congestion Charge. A number of cities have introduced environmental zones for motorised traffic in order to meet EU targets for reductions in emissions. (Presentations from a seminar Mobility Management by Environmental Zones and Urban Charging, 29.-30. September 2008 in Berlin of the German Institute of Urban Affairs in co-operation with CCRE/CEMR and the CURACAO Project to be found at: http://www.difu.de/extranet/seminare/berichte/08Umweltzonen/) A further discussion of these policy measures is outside the scope of this report.

With regard to the selection of charging regime, complex schemes are likely to be more expensive to implement and run and harder for travellers to use. The problem of system complexities was discussed in more detail in Section 3.2.2.

Level of charge and variations by time of day

There is a need to distinguish between charging regimes that are more for revenue collection rather than for congestion reduction, since their aim in affecting car driver behaviour are not similar. A scheme having revenue generation as the primary objective is not so much focused on travel behaviour change as a congestion charging scheme. The operator is happy if more vehicles use the facility because it means more revenue. In a congestion charging scheme, the operator will want fewer vehicles to use the facility because it will reduce congestion. Optimal charge levels and variation by time of day for these two types of objectives will be different, and this will need to be reflected in the Prediction and Appraisal phases of scheme planning.

A flat price may be more acceptable and straightforward to implement. On the other hand, letting the prices vary in accordance with time and location of congestion will improve the efficiency and benefits of the scheme. Also it is important to allow for some flexibility over time. Travel patterns will evolve and, consequently, the demand for road space will vary. A relatively fixed system may become inefficient over time.

With respect to actual levels of charges, the Central London congestion charge is probably the most expensive with a basic daily charge of €10 since July 2005. When the scheme was introduced in February 2003, the daily charge for driving or parking a vehicle within the congestion charging zone between 07:00 and 18:30 Monday to Friday was €6.25.
The 2007 permanent Stockholm congestion tax is an example of charges that are highly time differentiated. The tax per passage is €1, 1.5 or 2 depending on the time of day. The highest amount charged is during rush hours 07:30 to 08:30 and 16:00 to 17:30. A maximum amount per vehicle and day is set to €6.

With the exception of Trondheim, the Norwegian toll rings have no time differentiation during charged periods. In Oslo, the basic charge for a light vehicle is € 2.4, in force 24 hours a day and all days of the week. In Bergen the charge level is currently € 1.8 for a light vehicle and in force 24 hours a day Monday to Saturday. During 2005, the last year of operation of the Trondheim charging scheme, the basic charge (manual payment) was € 1.8 for all hours of operation (06:00 to 18:00 Monday to Friday). However, electronic tag holders benefited from discounts on the basic charge, which depended on the time of day and the amount which had been prepaid. Prices per passage for light vehicles are shown in Table 3 1. Heavy vehicles paid twice these amounts.

Table 31 Charges for the Trondheim Toll Ring

The Trondheim toll ring



Manual payment (basic charge)

€ 1.8

€ 1.8

Prepayment of NOK 500

€ 1.4

€ 1.1

Prepayment of NOK 2500

€ 1.3

€ 0.9

Prepayment of NOK 5000

€ 1.1

€ 0.7

Post payment by bank giro:



5 or fewer passages/week

€ 1.8

€ 1.4

10 or fewer passages/week

€ 1.6

€ 1.3

More than 10 passages/week

€ 1.4

€ 1.1

The scheme objective and equity factors clearly dictate whether there are discounts available for the scheme. Since the primary objective in Norway is generally to raise revenue for highway construction, providing discounts for an increased number of passages does not conflict with the scheme objective.

On the other hand, in Singapore, where the objective is congestion reduction, the electronic road pricing system sets tolls that vary with travel time but does not provide discounts for multiple crossings. One of the shortcomings of the early ALS scheme (Watson and Holland, 1978) was that a one-time payment offered multiple entries. Hence there was no incentive for the driver to plan his subsequent journeys once he had bought the licence.  At present, following the introduction of electronic toll collection, the charges are revised quarterly and differentiated by vehicle type. GPS-equipped taxis are used to monitor average speeds in the city centre and on the expressways, which are reviewed every three months.  If the speed falls below a specified threshold (20km/h in the city centre; 45km/h on expressways) charges are raised; if they rise above upper thresholds (30km/h and 60km/h respectively) they are reduced. An example of how the charge is differentiated by time period and by vehicle classes is shown in Table 3 2. The ability to employ differential pricing is a function of the technological elements of the scheme design.  Similar time-differentiated toll charges apply on SR91 in California, USA. 

Table 32 Abstract of charges on one major arterial during AM peak hours

(All charges in €,  converted at €1 = SGD 1.8)

Time Slice



Cars/Light Goods/Taxis

Heavy Goods Vehicles

 7.30am -7.35am 




 7.35am -7.55am 




 7.55am -8.00am 




 8.00am -8.05am 




 8.05am -8.25am 




 8.25am -8.30am 




 8.30am -8.35am 




 8.35am -8.55am 




 8.55am -9.00am 




 9.00am -9.05am 




 9.05am- 9.25am




 9.25am -9.30am 




Source: http:// www.onemotoring.com.sg accessed October 2008

In theory it is possible to estimate the speed flow curves using traffic engineering relationships and apply economic theory to derive optimal charges that make drivers fully cognisant of the road user charge. The methods for doing so have been documented in the literature since Walters (1961) applied this to deriving optimal charges for the Lincoln Tunnel in New York.  In principle, and by making some simplifying assumptions, an efficient charge can be calculated for any scheme. This is well illustrated for London by Santos and Fraser (2006).

In practice, for reasons of political acceptability, charges are generally derived through a trial and error process based on information gathered during the scheme development stages. In addition, the charge level would depend on the objectives of the scheme and the associated price elasticities. This is the view articulated for example in the UK Transport Select Committee (Transport Select Committee, 2006) which stated that

“If the Government led with the introduction of a national distance charge as the core component of the price structure, it would then need to work with local authorities and other stakeholders to agree the calibration of variations, and establish where and when they should apply.” (Transport Select Committee, 2006)

Once the charge has been set, it is not easy to make immediate changes if they are found to be excessive or too low, because authorities will be unwilling to admit that the charges they defended were far from correct. An example can be provided by the 1975 Singapore ALS (as described in Watson and Holland, 1978). The initial objective with a daily charge of $3 (= € 1.67) for cars was to get a total reduction of traffic for all vehicles of 30 % during the peak hours. However, there was a reduction of 44 % which invited adverse comments of an “overkill” situation (McCarthy and Tay, 1993; Phang and Toh, 1997).

Variations by vehicle type

Charges are not differentiated by vehicle type in the London and Stockholm examples. The Norwegian urban charging schemes have always charged heavy vehicles (gross weight more than 3.5 tons) twice the amount chargeable for light vehicles. An equitable way of varying charges by vehicle type to reduce congestion is to use the passenger car equivalents. Hence vehicles pay in relation to their dynamic occupation of space on the road. This was initially the basis for determination of charges in Singapore (Olszewski and Xie, 2003).


As noted earlier, Vickrey (1992) argued that all vehicles that contribute to congestion should pay. In practice, buses and public service vehicles like waste management, fire services etc are almost always exempted, as are vehicles for disabled drivers and electric vehicles. Rules for exemptions are strongly related to obtaining acceptability, and for example the London scheme has a large proportion of vehicles exempted for this reason.

Extensive exemptions have operating cost implications, and may undermine to a certain degree the objectives of a scheme. For example in the London scheme, motorcycles and mopeds are exempted. This led to a large increase in the number of these two wheel powered vehicles entering the charging area following the introduction of the scheme (Santos, 2004). On the other hand, there are issues of spatial equity that are inextricably linked to scheme design. Residents in the London charging zone receive a 90% discount (Santos, 2004) while residents in the charging area in Singapore do not receive any exemptions.

Another example is the Stockholm exemption for driving to or from the Lidingö island: Anyone who crosses by any of the three control points at the Lidingö bridge and then passes any second control point in the city within 30 minutes is exempted from the congestion tax. The reason for this is that the only connection from Lidingö municipality to the national road network runs through the city. Because of the strict rules governing a tax decision (the identification rate needs to be nearly 100%), this exemption is a major cost driver on operating cost, since affected vehicles need to be equipped with a transponder and a separate DSRC system is in operation. The permanent Stockholm charging scheme is otherwise operated by ANPR only.

An interesting case in point reflecting the public’s concern over spatial equity and its relationship to exemptions can be seen in the Edinburgh congestion charging proposals (Saunders, 2005). In Edinburgh the proposed urban congestion pricing scheme involved two cordons. The outer cordon was to be located just inside the city’s bypass in an attempt to control the increasing congestion on the edges of the city; while the inner cordon was designed to protect the World Heritage Site located in the inner city. The Scottish Executive’s guidance required that fair treatment be a high policy goal; however, the City Council included an exemption from the outer cordon charge for city residents who lived outside the outer cordon (Saunders, 2005). During the formal public review process the issue of geographic equity was raised and a recommendation was made to remove the outer cordon exemption for city residents living outside the outer cordon in order to achieve fair treatment (Saunders, 2005).  While the City Council agreed to many of the public’s recommendations, they did not remove this particular exemption. The referendum went to the public in February 2005 with the exemption still included, and the referendum failed. While the inclusion of the exemption for outer city residents was not the only reason for the cordon pricing initiative’s failure, it did exacerbate public concerns about the equity of the scheme.