For whom the road tolls?


The future for road congestion charging in the UK

In the next 30 years, the UK’s car population is likely to increase by around 50%. With our roads already heavily congested, we need to find ways to cope with this large increase. One solution is charging for the use of congested roads. Many trial schemes are already in place within Europe and London is set to introduce congestion charges in February 2003. In this article Peter Hills and Phil Blythe look at the options and technology available for implementing road use pricing.

For those who regard today’s traffic levels as already too high, the prospect of a further 50% increase is unthinkable. And yet, if real disposable incomes go on rising over the next 25–30 years as they have since the mid-1970s, then the car population is almost bound to go on rising, until some ‘saturation’ level of ownership is reached by about 2030. By then, today’s 23 million cars registered in the UK could have grown to 33–36 million. If traffic is not to grow commensurately, the nettle of the hitherto radical policy of road-use pricing (including congestion charging in urban areas and the tolling of congested motorways) will have to be grasped. The legislation for this is in place, through the Transport Act 2000, and much experimentation is afoot, but the political will at the national level is still lacking. This brief review of the future for congestion charging and of emerging technologies, for vehicle/roadside communication and automated toll-collection systems, concludes that the lack of appropriate technology will not be the constraint in implementing road-user charging in the near future in the UK.


The milestone report ‘Traffic in towns’, better known as the Buchanan Report (HMSO, 1963), predicted that the 12 cars owned per 100 people in 1962 would rise to 38 cars per 100 people in 1995. They were right; and this heroic prediction underscores all the problems to which the relentless growth of car ownership has given rise over those three decades. The Buchanan Report was wrong, however, in assuming little further growth beyond that. It predicted that a ‘saturation’ level in ownership would be at only 40–45 cars per 100 people by about 2010. The truth is that saturation ownership levels are already being reached in several countries (USA, Italy and Luxembourg, for example are at 60–65 cars per 100 people). This and other evidence points to the UK being still only two-thirds of the way to saturation level in car-ownership. If real disposable incomes go on rising, the overall size of the car population in the UK is likely to rise, from 23 million this year to an eventual plateau of 33 million to 36 million some time before 2040.

However, the congestion costs will be disproportionately high. The CBI has estimated the overall cost of traffic congestion, to UK plc, as being in excess of £21 000M per year. This is more than all the other ‘external’ costs of road traffic that fall on society (accidents, noise, pollution, CO2 output, etc.) put together. The potential economic returns and social benefits of reducing congestion, therefore, are huge.

The case for traffic restraint

One of the direct consequences of rising car ownership and use is the steady year-on-year decline in patronage for most forms of public transport. Across the country, outside of Greater London, bus use has been declining relentlessly since the early 1960s, at between –2.0% and –4.0% per year, as passengers opt for the strongly perceived advantages of going by car.

The Tory government of the early 1980s regarded the inability of public transport operators to retain their market share as being due to lack of competition and to excessive regulation, but, in the English metropolitan counties, the privatised bus operators have lost more than 40% of their person kilometres over the 17 years since deregulation. Ironically, Greater London, the only place in Great Britain exempt from the 1985 Act, has managed to hold on to its patronage and, in recent years even to increase it. London, of course, is unique in its size, its chronic traffic congestion and its comprehensive railway network, encouraging very high levels of public transport use, especially for journeys to work in the centre.

All of this suggests that the scope for public transport to compete successfully against cars, so as to curb the future growth of traffic, is severely limited. It would require authorities with powers similar to those of Transport for London (TfL) to be set up in all metropolitan areas or regions in Britain, to enforce quality contracts with private operators, to plan new public transport investment and to administer increasing volumes of public subsidy in order to stabilise the market share.

In any other sector, faced with a rapidly rising demand, the most obvious response usually is to expand supply to keep pace. This would be possible in the case of road traffic, except in the larger urban areas, but is no longer regarded as an acceptable solution in isolation. On interurban trunk roads and motorways, capacity can and will be provided mainly by widening the congested sections of the network rather than building new routes in green fields. However, in suburban areas, where car ownership and use are rising fastest, the opposition to new road building is considerable. Often, the argument revolves around the question of ‘induced’ traffic, i.e. that building new roads, by adding capacity to the network, creates more traffic than before. Although the Standing Advisory Committee on Trunk Road Assessment confirmed in their report (SACTRA, 1994) that this was a real phenomenon and should be accounted for in highway planning and evaluation, it is usually only at the level of a side-effect. Indeed, except when the amount of induced traffic is sufficiently large to cause congestion on the new road, it will generally contribute to the economic benefit of the scheme. The National Road Traffic Forecast (NRTF, 1997) confirmed this, by demonstrating that, in the long term, even if no more road capacity were built, future traffic would be ‘suppressed’ by only 10–12% due to increasing congestion. The economic cost of this mounting congestion, of course, would be huge.

The third approach, often canvassed, is one that invokes planning policies aimed deliberately at reducing the need for travel and the perceived dependence on cars to provide it. Advocates of this approach argue that residential densities should be increased sharply; new developments concentrated on ‘brownfield’ rather than ‘greenfield’ sites and in corridors amply served by public transport. Much tighter parking standards and persistent campaigns to change people’s lifestyles in favour of bicycling, walking and (if necessary) car-sharing are all part of this approach. Unfortunately, apart from a few well-publicised successes, so much of this goes against the direction of current market forces in planning and development that it is hard to see how this will be effective in curbing overall traffic growth, even in the long term.

The remarkable features of the 1998 Transport White Paper, and of the 2000 Transport Act to which it gave rise, are threefold: first, it recognised for the first time, at government level, that traffic restraint policies are inevitable (‘demand management’ is the euphemism employed). Secondly, no single approach to the problem of future traffic growth will be effective on its own, hence the emphasis on ‘integration’ and multi-modal studies. Lastly, it introduced the possibility of including congestion pricing as the one part of the ‘integrated package’ of policies that has the necessary potency to curb traffic growth.

Paving the way to implementation

The easiest way, administratively, to raise the overall ‘price’ of car use is for government to pile on the fuel duty and VAT onto the price of fuel at the pumps. This is more efficient and fairer than simply raising the annual vehicle excise duty (VED), as the amount of fuel taxes paid will (at least) reflect the distance travelled. The Royal Commission on Environmental Pollution (RCEP) in 1993 – having largely dismissed the arguments for congestion charging – proposed the doubling of retail fuel prices in real terms within ten years.

On the strength of this advice, the (then) Conservative government introduced the ill-starred fuel-price ‘escalator’, at +5% p.a. in real terms, ‘for the foreseeable future.’ In the run-up to the 1997 election, New Labour promised to raise this to +6% p.a. By autumn 2000, when the whole policy unravelled in the face of blockades and civil protests, the retail price per litre of petrol in the UK was the highest in the EU.

Even without the spontaneous revolt, the arguments against continuing the escalator were compelling. As, although fuel tax is paid proportionately to the use made of the road-system, it makes no distinction between that use being on a rural road at night and on a city centre street at the peak of the morning journey-to-work rush. In short, fuel taxes are a very poor proxy for the marginal costs of congestion.

Nevertheless, the political set-back of the successful fuel tax protest posed a problem for the government in dealing with the increasing inflows of foreign HGV hauliers, filling up at Irish and continental Channel ports with their (now cheaper) diesel and using the British road network ‘for free.’ This led the Treasury in November 2001 to publish a consultation paper on distance-based charging for all HGVs, British and foreign alike, to ensure fair competition in haulage and shift to an efficient direct charging regime ‘at the point of use’. The precedent for this already exists: Switzerland has had a nationwide charging scheme for HGVs since 2000 and Germany plans to do so in 2003. Implementation is relatively straightforward, as most HGVs are equipped with GPS location-finding equipment and many already have local cellular radio communications.

The Chancellor confirmed plans for distance-based HGV charging in his April 2002 budget, with the backing of the Freight Transport Association (FTA) and of the CBI. It remains to be seen how quickly such a scheme can be ‘rolled out’ to include cars and other vehicles using our congested motorways. The Commission for Integrated Transport (CfIT) sees this as the logical corollary to developing congestion charging in the urban ‘hot spots’.

To bring about the ‘pay as you go’ policies outlined above, technically, we need to introduce an efficient charging mechanism that can levy tolls and road-use charges automatically from drivers, i.e. without the need for the drivers to stop and pay or to perform any action (other than those of normal driving). Thus, charging systems should, where practicable, enable the collection of these charges at normal highway speeds and without the need for segregating vehicles into separate lanes, as with conventional toll-collection facilities. Indeed, it would be infeasible and unworkable, in many locations, to require traffic to be segregated into lanes, drivers to stop their vehicles and pay either manually to an operator or by inserting coins, banknotes or a card into a collecting machine. Building of these toll plazas such as those at the Dartford, Tyne and Mersey river crossings (and throughout Southern and Central Europe, where purpose-built toll roads are widespread) is costly and crucially requires a substantial land-area for each site. It is generally not practical to ‘retro-fit’ a toll plaza to an existing road; in urban areas, this may be unacceptable on other grounds also, e.g. the creation of additional congestion, with consequent dis-benefits, such as noise and air pollution. Moreover, purpose-built toll roads generally have a limited number of entry and exit points, whilst access to untolled roads is usually not so restricted – thus creating an additional difficulty when introducing urban road charging.

The publication of the Government’s White Paper on Integrated Transport in July 1998 caused a shift in emphasis to encourage local authorities to consider congestion charging as an option to help manage the demand for car travel. Crucially, however, the Transport Act 2000 requires Local Authorities to retain the net revenue raised from congestion charging and/or private non-residential (PNR) parking charging for a minimum of ten years – this ‘ring-fenced’ revenue has to be reinvested in local transport schemes and the support of public transport. This provision in the 2000 Act, that links the introduction of congestion charging (as a stick) to the reinvestment of net revenues in better public transport within the same jurisdiction (as a carrot), may well be the lynch-pin in cities being able to secure sufficient public acceptance for future congestion charging schemes.

The concept of direct road-user charging is not new although few trials and implementations have actually taken place. The first ‘official’ acknowledgement of the technical possibilities of direct pricing at the point of use was the Smeed Report (1964). Since then, a great deal of research has been undertaken and a number of attempts to introduce urban road-user charging have been made, most notably in Hong Kong, Singapore and Norway (Blythe et al., 2001). Motorway schemes, using electronic devices to automate existing toll-collection facilities are quite widespread and include numerous examples in the USA and in the ASECAP1 countries (Italy, France, Greece, Spain, Portugal), as well as new multi-lane tolling schemes on Toronto’s Highway 407 and the Melbourne CityLink.2

Technological options for congestion charging

Currently, several electronic technologies are used or have been considered for charging. The more important of these are briefly reviewed below.

Microwave-based digital short-range communication (DSRC) systems

These systems need road-side equipment, typically mounted on a gantry, with electronic tags in the vehicles which may be read-only, read–write or smartcard-based. Read-only tags contain a fixed identification code which, when interrogated by a roadside reading device at the charging point, conveys this identity to the roadside system. The code relates to the identity of the vehicle or the identity of the user’s account. Read-only tags operate reliably only if used for single lane operation at low speed and over a short range. However, their inflexibility, ‘dumbness’ and inability to work in a high-speed, multi-lane road situation essentially limits their application to that of automating existing toll-plazas. Read–write tags are a logical development of the read-only tag. They can receive data from the roadside and store this data directly on the tag or on a separate value-card (which may be interfaced to the tag whilst in the vehicle).

The most flexible in-vehicle units (IVUs) are transponders (smart tags) that support smartcards. They are ‘intelligent’, having the capability to handle and process many kinds of data and (potentially) to be programmed to manage a number of different applications. Such a system requires a reliable, high-speed two-way data-communications link with the roadside and more complex on-board equipment, replacing some of the processing requirements traditionally handled by the roadside equipment (Figure 2). A modular approach is adopted to the transponder’s design, facilitating ‘add-on’ peripheral equipment (e.g. smartcard readers, keyboards, displays, connections to other in-vehicle equipment). Figure 3 shows a transponder and smartcard from the ADEPT (automatic debiting and electronic payment for transport) project (Blythe 1999), a European funded project led by TORG in the early 1990s which installed trial systems in the UK, Sweden, Portugal and Greece.

The modularity in the design of the automatic debiting transponder prototypes allows several different forms of payment (all of them ‘cashless’) with one device. Possession of a transponder offers users the possibility of holding a positive (or a limited negative) balance of credit-units, either directly in the transponder’s memory or alternatively on a separate smartcard interfaced to the transponder. The smartcard, being portable, can then be used for other payment purposes.

These systems are perceived by many international road administrations as the future of road-user charging, where high-volume, multi-lane roads need to be tolled without restricting traffic flow. Europe standardisation of DSRC systems nears completion and many products based upon 5.8 GHz microwave communications technology will soon emerge, though to date few commercial installations exist. Worldwide, the Singapore system and Highway 407 in Canada utilise such an approach. The key limiting factor seems to be the processing speed of the smartcard – in Singapore, each charging point has two gantries – one to start communications with the vehicle and a second (further down the road) to complete the transaction and perform enforcement measures, if necessary. The demonstration of interoperable road-user end-to-end charging and telematics systems (DIRECTS) project, just launched by the Department for Transport (DfT) using 800 or so volunteer drivers, who will have their vehicles equipped for the trial in Leeds, will finally prove an end-to-end solution for DSRC-based charging. The aim of DfT is to develop a UK national specification for interoperable payment of road-user charges, consistent with the emerging European standards.

Wide area communications-based systems

Wide-area systems are a more recent innovation in charging and tolling technology – also widely known by the term MPS (mobile positioning systems). They use two technologies adapted from other applications; namely, GPS (global positioning system), whose satellites enable suitably equipped vehicles to calculate their location accurately; and a two-way communications link (e.g. GSM) based upon cellular radio. These systems were tested in the German trials (Figure 5) in 1995–96 and Hong Kong 1998–99 (Blythe 1999). They are designed (like DSRC systems) not to disrupt the flow of multi-lane traffic on motorways. Moreover, because in urban areas ‘virtual’ toll-points can be established (and changed, as necessary), these wide-area systems will reduce the amount and environmental intrusion of roadside infrastructure required, in comparison to DSRC systems.

GPS satellite-based system

GPS satellite-based system The in-vehicle unit (IVU) contains a GPS receiver and the transponder must have a record of the locations of all charging points. At a pricing cordon, the system will deduct the appropriate charge from the credit-units stored in its account. It can use GSM to inform the central system, once a limit has been reached on the on-board account, enabling it to initiate the clearing process and allowing a range of credit-transfer options. GSM can also reload a smartcard and update the IVU with information on the charging tariff and locations of the ‘virtual’ pricing sites as well as providing an enforcement function. Such a solution lends itself to distance-based charging as well.

Third generation cellular radio technology

Proposals have been made for tolling systems based on charging for entering a radio ‘cell’, with the first trials being held on the A555 Köln–Bonn autobahn in 1996. Until recently, this option could be discounted, as phones could not offer sufficient accuracy in pin-pointing where the phone is at any given time. However, this may change with the potential locationing function that will be inherent in the third generation (3G) licences for mobile phones. The 3G companies claim an accurate location service for business phone users – perhaps down to 10–15 metres resolution – which is ample for road-user charging purposes. As the phones already have secure access and a central payment facility (as well as European interoperability), the technology needs only to provide a credible security and enforcement scheme to be considered a serious contender.

Video-based systems

Video-based systems rely on the accurate ‘reading’ of vehicles’ licence plates as the primary means of identifying, charging and enforcing vehicles in a congestion charging scheme. The big advantage of this is that it obviates the need for any invehicle equipment. Moreover, it solves the ‘occasional user’ problem, whereby those who use a particular charging scheme only rarely do not have the necessary in-vehicle charging Automatic number plate recognition (ANPR) is a variation on the automatic account identification system, which also relies on the vehicle’s number plate as its unique identifier. ANPR systems process the video images taken by a camera at the roadside or on a gantry, locate the number plate in the image and convert this into the appropriate alphabetic/numeric characters, without any human intervention. The increasing use of video cameras for road traffic monitoring has given an incentive to improve camera technology, including optical processing, to provide a wider contrast range and give clear images, even when the licence-plates are in heavy shadow or surrounded by bright headlights in direct alignment with the camera.

Unresolved problems with ANPR, however, still include:

  • number plates of many and different shapes and sizes

  • number plates missing, mounted incorrectly or deliberately obscured equipment to pay the charges

  • difficulties for accurate reading in poor weather, due to dirt/rain/snow

  • similarities between some letters/numbers (Os being read as Ds, for example) and

  • insufficient control of ambient light at camera positions.

To improve the overall accuracy, some vendors provide for the capture of multiple images; if ANPR determines the same plate information for all, the confidence level of the data is improved and manual interpretation may not be required. Any discrepancies are either placed in a queue for visual inspection or treated as a ‘lost revenue’ transaction.

A Government Office for London Report (GOL, 2000) reviewed the road user charging options for London (the ‘ROCOL’ report) in 1998/1999. It studied the feasibility of road-use pricing and work-place parking charging, as well as the likely impacts on business, traffic levels and users’ reactions to the charging proposals. It recommended that London should implement a video based road-user charging system, in the first instance, until the results of the DIRECTS project were available. In August 2002, the Mayor, Ken Livingstone gave the final go-ahead to proceed towards a full-scale implementation of congestion charging in central London, using ANPR. The target launch date is 17 February 2003.

How congestion charging technology may develop

These three competing families of technology for future charging systems all have different attributes, advantages and disadvantages. For many years, short-range transponder/tag based systems have been preferred, due to their simplicity of operation, potential for supporting additional services for vehicle users (Figure 2) and, most importantly, because they are easy for users to understand – you pass a point and you pay. New technologies, however, have opened up new opportunities for innovative charging schemes.

Wide area charging schemes, which rely on the in-vehicle equipment determining where the vehicle is and charging the vehicle accordingly (based on passing a pre-defined point – or on the distance travelled by the vehicle) are attractive and offer new possibilities for charging without the main disadvantage of short-range charging systems, namely the associated road-side infrastructure at every charging and enforcement point. Some infrastructure is still required for enforcement purposes (Figure 5), but this can be situated in locations where aesthetics are not a prime consideration. Effective operation and enforcement using GPS-based systems was demonstrated in the recent Hong Kong charging trials (1998–99). Moreover, the distance-based taxation of heavy goods vehicles which is being considered by the UK Treasury could probably only be efficiently implemented using some form of wide area charging (probably linked to vehicles’ digital tachographs – as in Switzerland).

Video-based charging is a very recent innovation, with London being the first large urban area to adopt such an approach. In Norway, video is used as the primary charging means in the cities of Kristiansand and Bergen: however, this is on a very small scale in comparison with London. For central London, the scheme clearly requires a very complex back-room clearing and management system, to register on a daily basis all those who wish to pay to use the charged area within the cordon and also to record and process the images of all vehicles recorded, entering the charged area – but who have not registered and paid. If the system in London is deemed a success over the next two or three years, several other UK cities seem poised to introduce a similar system.

Looking ahead, the authors believe that the likely future evolution will be a fusion of a DSRC and wide-area charging systems – which will be able to support several different charging configurations with one set of in-vehicle equipment. Most drivers seem to prefer a system that they can actually see working – through some sort of display in the vehicle.

Where might it happen? And how soon?

With the legislation in place and the technological problems resolved, which local authorities are going to be brave and take the lead? The following is a snapshot of what might happen around the UK over the next three to five years (i.e. within the ten-year plan for transport investment).

Committed projects

DIRECTS (field-trials), Leeds

This national demonstration of road user charging technical interoperability is to be hosted by Leeds City Council. Fairways Consortium was selected and awarded the contract in 2001 by the DfT. A trial of interoperability between equipment from different manufacturers will get underway in Spring 2003. The results of the project should be a full technical specification for congestion charging, using DSRC equipment, available in late 2004.


The first cordon toll to be introduced in a UK urban area was inaugurated on 1 October 2002. A £2.00 charge is now levied from drivers wishing to use the cul-de-sac road leading to the Cathedral/Castle peninsular in Durham City. It is aimed to reduce traffic substantially to the historic centre of the city – a possible model for other small historical cities? One month on, it is reported the scheme has reduced traffic by 90%.

Central London

The congestion charging scheme has now been given the go-ahead within a cordon around central London. The scheme will initially use video-based entry charging, with some mobile camera units. A standard charge of £5.00 per day will be levied to use the cordon area. Exemptions are limited however to buses, taxis and service vehicles; and residents and essential workers will receive discounts of up to 90%. The scheme is due to go live on 17 February 2003. Ken Livingstone plans to extend the boundaries of the cordon both eastwards (to include Docklands) and westwards (to include Kensington and Chelsea) and possibly to up-grade the system to an electronic one, early in his second term of office (see inset in Figure 6).

Other projects being planned


This is probably the most advanced of the other schemes. A cordon around the city centre is proposed and may be introduced as early as 2005, dependent upon the earlier completion of the new light rail system. Bristol will host a distance-based charging trial in 2003.


Edinburgh City Council has just completed a large-scale public participation exercise, inviting people living within Edinburgh and a wide area surrounding the city to choose between three options. A bare majority (51% for, 38% against) voted for a single cordon charging scheme around the central area. A minority (44% for, 45% against) voted for a double cordon scheme, where the second cordon would run just inside the bypass. Lastly, a minority (38% for, 43% against) voted for not introducing any scheme at all.

Encouraged by this outcome, the city council will probably work for the development of the single cordon scheme, in the first instance, for implementation possibly in 2005.

West Midlands

A possible charging scheme is being developed for the central Birmingham and Solihull areas and is already under discussion.

  1. Association of European Toll Road Concessionaires
  2. Refer to websites: and


Blythe, P.T. (1999) ‘Road tolling and road use pricing technology’, in J. Walker (ed.) Advances in Mobile Information Systems, Artech House, London.

Blythe, P.T., Walker, K. and Knight, P.K. (2001) ‘The technical and operational feasibility of automatic number-plate recognition as the primary means for road-user charging’, Journal of the Royal Institute of Navigation, Vol 54, No. 3, pp. 345–353 1647.

Buchanan Report (1963) ‘Traffic in towns’, UK Ministry of Transport, HMSO, London.

GOL (2000) ‘Road charging options for London’, the Stationery Office, London.

Royal Commission of Environmental Pollution on Transport (1993) ‘Final report’, HMSO, London.

Smeed Report (1964) ‘Road pricing: The economic and technical possibilities’, UK Ministry of Transport, HMSO, London.

Prof Peter Hills OBE FReng

Professor of Transport Engineering University of Newcastle Upon Tyne

Philip Blythe

Acting Director Transport Operations Research Group (Torg) University of Newcastle Upon Tyne

Peter Hills was Director of the Transport Operations Research Group at the University of Newcastle upon Tyne, before being appointed as Dean of the Faculty of Engineering in 1996. From 1985, he led TORG’s engagement in projects developing and testing microwave systems for two-way vehicle/roadside communications. He has been a member of the Standing Advisory Committee on Trunk Road Assessment (SACTRA) and a special advisor to the National Road Traffic Forecasts (NRTF) team. He was awarded the OBE, for services to transportation planning and research, in 1995.

Phil Blythe joined TORG in 1985 initially to work on EU-funded research projects on microwave tolling systems. One of the early pioneers of road-user charging schemes, he went on to manage the ADEPT I and II projects and the DISTINCT project which developed citizen smartcards. He was part of the team that studied options for road pricing in London in the early 1990s and was also part of the ROCOL team which advised the Mayor of London on the current congestion charging system. He is responsible for TORG’s role in the DIRECTS project for the DfT and he advises government agencies both here and abroad, and is a member of the e-envoys smartcard policy working group within the cabinet office.

Download Article 871KB