More roads and road pricing -
the way to go?
IoD Policy Paper
_________________________________________________________________
GraemeLeach
This policy paper was written by Graeme Leach, ChiefEconomist. It was produced by Lucy Chard.
August 2001
ISBN 1901580 66 0
Copyright Institute of Directors 2001.
Published by the Institute of Directors
116 Pall Mall, London SW1Y 5ED.
COPIES AVAILABLE FROM:
Director Publications
Publications Department
116 Pall Mall
London
SW1Y 5ED
Tel: 020 7766 8766
Fax: 020 7766 8787
Price 5.00
Contents
Summary 5
1 Introduction 8
2 Transport trends 10
3 The drivers of road traffic growth 13
4 Congestion the theory 16
5 What is theexternality cost of roads in the UK? 17
6 Road pricing 21
7 More roads? 26
8 Technologyissues 29
Summary
"The way we run our road system in Britain is the lastremnant of the Stalinist state. We ration road space the same way the formerSoviet Union rationed bread, by queuing. It is incredibly inefficient."
Professor David Begg,Chairman of the Commission for Integrated Transport (The Times 11th July 2001)
Congestion on the roads is not a new phenomenon chariots filled the streets of ancient Rome but this does not mean that weshould accept the status quo. In this paper the IoD argues for the introductionof widespread road pricing together with the construction of extra capacity onthe road network.
The IoD argues that the solution to road congestion involvesboth demand-side (road pricing) and supply-side (new road capacity)elements. Between 1952 and 1999 thenumber of total vehicle kilometres travelled by cars and vans increased by1300%, whereas total road lengths rose by just 25% over the same period!
The road network in the UK desperately needs apricing system which will (1) Reduce congestion (2) Finance investment in newcapacity (3) Internalise the marginal social cost (accidents, road damage,pollution and congestion) of road use to increase economic efficiency.
The introduction of road pricing should be revenueneutral, offset by reductions in fuel duty.
Even under the Governments best case scenario, roadcongestion is forecast at current levels in 2010. In other words, roadcongestion will be no less than today and could be substantially worse although the Government envisages improvement in the worst black spots.Successful implementation of the 10 Year Transport Plan is expected to resultin a 17% rise in total road traffic over the 2000-2010 period, with congestionlittle changed from current levels. DTLR forecasts show total road traffic inthe absence of the Transport Plan increasing by 22% over the 2000-2010 period.Without the plan DTLR forecasts that congestion, already very high in manyurban areas, will rise by 15% across the network and by 28% on inter-urbantrunk routes.
The IoD is concerned about the road trafficimplications of the DTLR forecasts, in two key respects (1) The forecasts mayunderestimate the demand elasticity of road use with respect to economic growth(2) By aiming for a 50% increase in rail passenger traffic (as part of the 10Year Transport Plan, 2000-2010), the forecasts may overestimate the potentialfor moving people off the road and onto the railways. Over the coming decade itis very difficult to believe that the railways could provide the flexiblemobility or potential passenger capacity to effectively counter congestion onthe roads. Even if rail passenger traffic was to rise by 50%, the increase wouldaccount for less than 3% of passenger traffic on the roads in 2010.
Road pricing and road capacity are linked. Roadpricing could provide far more effective signals than at present, as to whetherto invest in greater capacity. Revenues from road pricing might also help tonarrow the current imbalance whereby only 14% (1999/2000 figures) of road usertaxation is actually spent on roads! Political pressures associated with roadpricing are more likely to ensure a greater degree of hypothecation of revenues,than under the current fuel duty regime.
Road pricing schemes (urban and inter-urban) arecomplex, but they are also both technically feasible and inevitable. However,the IoD recognises that there are significant political and practical obstaclesto overcome. The heavy burden of taxation on motorists already, means that inorder to be politically acceptable, any switch towards road pricing would needto be revenue neutral. This is acceptable given that rough orders of magnitudesuggest that current externality costs associated with road use, equate withtotal road revenues, less expenditure on roads.
Road pricing is far less of a blunt instrument thanfuel duties. At the present time the cost of motoring does not reflect thedifferences between (1) Rural and urban use (2) Peak and off-peak times. Roadpricing would reduce the cost of rural road use and increase that in urbanareas. At present, under the fuel duty regime, congested roads are in effectunder-priced whereas many rural roads are over-priced.
Estimates of the demand elasticity of road trafficwith respect to fuel price, suggest inelastic demand for road traffic withrespect to road prices. Other difficulties might be that some of the newcapacity will become filled by people who are now being deterred by congestion,but this is a contentious issue.
Greater private sector capital needs to be introducedinto the road system (1) Leveraged in as part of public-private partnershipdeals (2) Introduced as a result of privatisation of parts of the network.Privatisation may be crucial to generating sufficient up-front capital toinvest in improvements in other parts of the transport infrastructure. Giventhe potential for exploiting a monopoly position, any privatisation would needto be subject to a new regulator RoadReg.
Due to the market failure associated withexternalities, road pricing offers the potential for more efficient economicoutcomes. There is also the potential for a potential Double-Dividend wherebyroad pricing revenues could be used to offset the distortionary impact oftaxation in other parts of the economy. Obvious examples include thepossibility of reducing taxes on savings and capital.
From a business perspective the status quo will verysoon become unacceptable and more radical solutions will be demanded, incontrast to the rationing by congestion currently on offer. Despite theimportant issue of market failure associated with externality costs, the roadnetwork displays all the characteristics of an economic activity which has beenowned, operated and starved of investment by the public sector, over manydecades.
The difficulties of introducing road pricing andprivate capital, together with the time lag in the planning and construction ofnew road infrastructure, mean that even if radical solutions were accepted, thebenefits might not be felt nationally for a decade. There are no quick and easysolutions to road congestion.
1 Introduction
"If we were to shift the tax burden away from vehicle exciseduty and fuel duty towards congestion charges it would create a big reductionin congestion without the motorist having to pay any more in aggregate".
Professor David Begg,Chairman of the Commission for Integrated Transport (The Times 11th July 2001)
This paper focuses on howthe road transport system in the UK might be changed in order to create a trulyworld class road system. By necessity the paper focuses primarily on car use,given the overwhelming dominance of cars as the primary mode of road traffic.Some might question the focus on the car, given the clear intention of theGovernment to get more people off the road and onto the railways the numberof passenger kilometres on surface rail rose from 33 to 38 billion during the1990s, an increase of 15%.
The Government has set atarget (Transport 2010 The 10 Year Plan,DETR, July 2000) of a 50% increase in rail use measured by passengerkilometres by 2010, but in the post-Hatfield rail environment this target willbe tough to achieve, given the financial restraints on Railtrack and consumerdispleasure at the quality and timeliness of the rail system[1].
But even if thesedifficulties prove to be ephemeral, and the 50% increase is achieved a verybig if this will still only move an extra 19 billion passenger kilometresonto the railways.
A 50% increase in railpassenger kilometres is equivalent to only 3% of car passenger kilometres in1999[2].If one makes the assumption that car passenger kilometres will rise over thenext ten years by the same rate as over the past decade, then the total shouldreach 663 billion by 2010 an increase of 42 billion. A 19 billion passengerkilometres switch would be equivalent to just 2.9% of car passenger traffic in2010. Whilst this might be welcomed, one can see that tackling the problems ofroad congestion in the UK requires far more direct action.
Speed-flow theory
Road transport congestioncan be illustrated by reference to the speed-flow curve in Figure 1.1. Thecurve shows the movement of vehicles along a particular road and how motoristsinteract and impose delays and costs on each other. In a free-flow situation(point A) there is no interaction between vehicles and speeds are high. Asextra vehicles join the road the average speed is reduced, but an increasedflow is achieved up until point B the flow depends on the number of vehiclesjoining the road and their speed. Clearly the optimal individual solution is atpoint A, whereas the collective point of maximum efficiency is at B. However, ifmotorists continue to join the road speeds are reduced and so is the flow ofvehicles towards C.
Figure 1.1
The Speed-flow curve
|
Average
speed
|
||||
|
||||
Flow (vehicles per hour)
At point C the motorist isin a bottleneck or stop-and-start situation which imposes considerable socialcosts on other users time delays and pollution etc. The aim of this report isto analyse how the congested elements of the UK road system might reach pointB, as compared with a current situation at point C.
Central to this argument isidentifying the means by which traffic growth can be maintained whilst at thesame time providing an adequate infrastructure. We concentrate on the roadnetwork owing to the difficulty of transferring billions of passengerkilometres from road to rail. Current policy has neither stopped traffic growthnor provided an adequate infrastructure. In other words, the solution to ourtransport difficulties has both a demand-side and supply-side element.
2 Transport trends
Over the past fifty yearspassenger transport mode has changed dramatically and we have seen an enormousincrease in road traffic by car. In 1952 only 27% of passenger kilometrestravelled were by car, van & taxi.By 1999 this proportion had soared to 85% (Table 1.1, Transport Trends 2001 Edition, DETR).
Between 1952 and 1999 thenumber of passenger kilometres travelled by car, van & taxi rose from 58billion to 621 billion. Over the same period the number of passenger kilometrestravelled by rail rose from 38 billion to 42 billion, whilst that for buses andcoaches fell from 92 billion to 45 billion.
Figure 2.1
Road traffic by type ofvehicle shows that between 1952 and 1999 the number of vehicle kilometres bycars and taxis increased from 30.6 billion to 380.1 billion, by motor cyclesfrom 4.2 to 4.6 billion, by buses and coaches from 4.1 to 5.0 billion, by lightvans from 7.8 to 49.2 billion and by goods vehicles from 11.2 to 28.1 billion.As a result, cars and taxis account for 81% of motor vehicle traffic in 1999.
Passenger traffic (billionsof passenger kilometres) in cars, vans and taxis only increased by 7% in the1990s, but this figure is slightly misleading as an indicator of growth. If onetakes the 15-year period 1985-1999 then the growth in traffic was 41%.
Motor vehicle traffic byroad class shows that over the 1985-1999 period motorway traffic (billionvehicle kilometres) rose by 120%, trunk road (built-up) traffic by 9%,principal (built-up) traffic by 22%, trunk road (non built-up) by 60%,principal (non built-up) by 50% and minor road traffic by 43%. Overall roadtraffic rose by just over 50% between 1985 and 1999.
Between 1959 (firstmotorway completed) and 1999 the road length for motorways has increased from13 km to 3,358 km in 1999. Trunk road lengths have declined from 13,274 km in1952 to 12,150 in 1999. Principal roads have increased from 31,484 km in 1952to 36,045 in 1999. Other B, C and unclassified roads increased from 252,756 kmin 1952 to 320,361 km in 1999.
Figure2.2
These statistics reveal afascinating picture. Between 1952 and 1999 the number of passenger kilometrestravelled increased by 970% whereas total road lengths rose by just 25% overthe same period. Over the past 25 years road length has increased by 0.5% perannum. If one looks at road traffic by type of vehicle, thereby measuringbillion vehicle kilometres, the increase is even greater, with a 1300% increaseover the past 50 years in the number of vehicle kilometres for cars and vans.Total motor vehicle kilometres rose by 691% over the same period.
Even allowing for the factthat many of the new roads were motorways, the divergence between the growth invehicle kilometres and road lengths is huge. Motorways permit greater trafficnumbers per kilometre as a result of having more lanes and faster speeds.
The end result is notsurprising. The proportion of road links that are congested is some two and ahalf times greater than in Italy, three times greater than in Germany and fivetimes greater than in France (The roadmore travelled needs expanding, Martin Wolf, Financial Times, 15thNovember 1999).
Figure 2.3
3 Thedrivers of road traffic growth
Greater road traffic canonly come from two sources, more vehicles and/or more vehicle useage. Becauseof the overwhelming dominance of the car in overall road traffic, we willconcentrate on this source. Success in alleviating traffic congestion willdepend on policies aimed at car users 81% of total traffic.
Over the past fifty yearsthe proportion of households with one or more cars has risen from 14% to 71%.Moreover, 22% of households now possess two cars and 5% of households possessthree or more cars. This growth in car penetration is all the more impressivewhen one considers that the number of households has risen enormously as well there are now around 25 million cars in the UK[3].
Reasons for the growth incar ownership and use are not difficult to find. Glaister (Practical Road Pricing, S. Glaister, SMF Memorandum, No. 32, March1998 and Predict, but dont provide?S. Glaister, New Economy, Autumn 1999) highlights the following influences:
Improved standards of living allowing a majority ofhouseholds to afford the improvements in the quality of life that mobility and especially the private car offer. There is a strong correlation acrossthe globe between GDP per capita and mobility (passenger kilometres).
Falls in the relative cost of buying and running acar as a proportion of income at the same time that the relative cost ofpublic transport has risen.
Changing household structure falling average size and growth in the number of individual households.
Growth in female labour force participation and carlicence holding. As an illustration, 69% of females in the 20-29 cohort nowhold a driving licence, compared with 54% in this age group fifteen years ago.
Post-war sub-urbanisation and population outflow fromthe major conurbations. For example,population outflow from Greater London averaged 0.5% per annum in the 1980s,although population in London has recently started to increase.
Other factors encouraginggreater car ownership might include growth in out-of-town shopping centres,together with fears over personal safety and that of children walking/cyclingto school.
Figure 3.1
De-coupling demand?
A reading of the transporteconomics literature confirms that the size of the income effect GDP and realincome growth on traffic generation is large, and can be the most importantsingle factor (see: Transport and theEconomy, SACTRA, DETR, August 1999).
Can road use be de-coupledfrom economic growth? National Road Traffic forecasts by DETR in 1989 brokewith previous forecasts by showing that traffic intensity would turn and startdecreasing around the turn of the century. By decreasing intensity it was meantthat car traffic growth would be less than GDP growth. This broke with aprevious trend throughout the 1960s, 1970s and 1980s for increasing intensity.
In essence, as carownership approached saturation, its elasticity with respect to income wasforecast to fall below one, and this meant that intensity as measured, wouldfall. However, it should also be noted that even with intensity forecast todecline, the actual traffic level and the implied level of congestion wouldcontinue to rise.
During the 1990s the rateof growth in road traffic by type of vehicle did slow. Over the 1990-1999period road traffic by car rose by 13%. Over the 1980-1989 period the growthwas 54%. Nevertheless, we should be cautious in interpreting this slowdown,since we risk mistaking the symptom of the disease for the cure.
The Governments latestforecasts (Transport 2010 The 10 YearPlan, DETR, July 2000) show road traffic and congestion forecasts with andwithout the implementation of the 10 Year Plan. Without the introduction of theplan, DETR forecast total road traffic would increase by 22% over the 2000-2010period. Successful introduction of the plan results in a 17% increase in totalmotor traffic over the same period. The relatively small differential inoverall traffic growth is in part attributable to the mitigating effects ofcongestion on traffic growth, in the absence of the plan. Without the plan,DETR forecast that congestion, already high in urban areas, would rise by 15%across the network and by 28% on the inter-urban trunk network.
By 2025 road traffic levelsare forecast to be between 36% and 57% higher than in 1997, depending onwhether low or high variant forecasts are used! It is simply ludicrous toentertain the idea that this volume of passenger kilometres could somehow beshifted from road to rail or other modes of transport.
The flexibility provided bycar use is a very powerful constraint on the use of public transport. Thereisnt a train to take you to the supermarket or school. In addition to thestrong influence of personal mobility permitted by the car, other strong forcespoint towards growth in car use:
The number of single person households is set to risesharply over the coming decades.
Continued growth in real incomes.
Recent falls and reductions in the rate of growth in UK car prices increase the relative attraction of car ownership.
Professor S. Glaister(Glaister, Autumn 1999, op cit) has stated that,
currentpolicies will not be enough to prevent the increase of road traffic due toroutine economic growth and to this we must add growth due to improvements incar quality and reductions in manufacturing costs and demographic factors.
4 Congestion the theory
Trafficcongestion is a good example of market failure and so the idea of imposing acharge for congestion is not new way back in 1964 the then Ministry ofTransport published the Smeed Report which advocated road pricing. As roadspace is a valuable and scarce resource it is natural that economists shouldargue that it should be rationed by price.
Roadpricing internalises external costs and replaces a queuing mechanism with theprice mechanism.
Thetextbook economic theory of road pricing is discussed below and is notcontroversial. The analysis is taken directly from a recent DETR study (Internalising External Costs, pages152-153, in Transport and the Economy,SACTRA, DETR, 1999).
Economictheory postulates that the external costs (road damage[4],accidents, environmental costs[5]and congestion) arising from road use provide a rationale for traffic chargesas a result of the misalignment of marginal benefit with marginal social cost.
Byusing the road car owners impose a social cost externality on other roadusers with the result that the social costs are higher than the private costalone. This means that marginal social costs (the costs imposed on others by anextra trip) are above marginal private costs (fuel, car wear and tear etc). Ifroad users only take account of the private costs of car use then they fail toaccount for the external costs on others (wear and tear on the roads,congestion, pollution etc). This means that car use (measured in vehiclekilometres) will be too high, because drivers do not face the social cost oftheir decisions road use is too high when roads are zero priced.
Economictheory suggests that a consumer will use a road up to the point where themarginal benefit of car use equals the marginal private cost. The key pointwith regard to congestion is that a road user imposes a cost on others, but ifthe road in question is free to use, then it will be consumed beyond the pointof maximum efficiency the basic principle of road pricing is that usersshould pay the costs they impose on others.
5 What is the externalitycost of roads in the UK?
Estimating the cost ofexternalities the social costs of road traffic is a very difficult exercisesince our knowledge of the costing of many of the items is immature and oftensubject to huge variations caused by the context in which transport servicesoperate the level of traffic, time of day and route network.
A literature review by TheSmith Group (Clearing the way practicalsolutions for urban road user charging, The Smith Group, November 1999),estimated that a central estimate for the external costs of trafficexternalities in the UK was of the order of 30 billion per annum, with a rangefrom 25 billion to 61.5 billion.
Table 5.1 Aggregate social costs of trafficexternalities in the UK
( bns, 1998)
|
Source |
Smith Group Central estimate (bn) |
% of total |
|
Congestion |
20 |
67 |
|
Air pollution |
3.5 |
12 |
|
Accidents |
2.4 |
8 |
|
Noise |
2.2 |
7 |
|
Road damage |
1.7 |
6 |
|
Global warming |
0.2 |
- |
|
|
30bn |
100% |
(Source:The Smith Group)
DETR has published aliterature review of the external costs of road transport excluding the costsof road congestion. The range in estimated costs from 8 billion to 39 billionis shown in Table 5.2 [6].
Table 5.2 Annual external costs of UK road transport(bn)
|
Cost |
Pearce (1991) |
Mauch & Rothengatter (1991) |
RCEP (1994-95) |
Maddison (1993) |
|
Accidents |
4.4-7.5 |
13.3 |
5.4 |
2.9-9.4 |
|
Noise |
0.6 |
3.4 |
- |
2.6-3.1 |
|
Air pollution |
2.4 |
6.2 |
4.6-12.9 |
19.7 |
|
Climate change |
0.4 |
4.1 |
- |
0.1 |
|
Totals |
8.1-10.9 |
27 |
10.0-18.3 |
25.3-39.3 |
(Source: cited in SACTRA/DETR, 1999)
Other estimates arediscussed below, and range from 7 billion to 19 billion per annum. Researchundertaken by NERA (National Economic Research Associates, 1997) estimated thetotal time cost of congestion to road users to be 7 billion, split between acost to business of 2.5 billion and the cost to private motorists, van driversand bus passengers of 4.5 billion (Transport chapter, in Applied Economics, Eighth Edition, A Griffiths & S Hall, 1999).
Table 5.3 The marginal time costs of congestion inGreat Britain (1993)
|
|
Urban Central peak |
Urban Central off-peak |
Non Central peak |
Non Central off-peak |
Small Town peak |
Small Town off-peak |
Other trunk and principal roads |
Motorway |
|
Marginal congestion cost per passenger kilometres driven - pence |
44.74 |
35.95 |
19.51 |
10.75 |
8.47 |
5.17 |
0.32 |
0.23 |
(Source:Table 9.2, p. 132, Newbery, 2000)
Newbery (Pricing and congestion economic principlesrelevant to pricing roads, in Readings in Microeconomics, edited by T.Jenkinson, OUP, 2000) has estimated that the marginal time cost of congestionwas around 19 billion in 1993. If additional external costs such as roaddamage were also included then external costs amounted to 20.7 billion.
Figure 5.3 illustrates thelarge differences in congestion costs between peak and off-peak periods andbetween urban and rural road users.
Figure 5.4 shows estimatesof the marginal external costs of passenger transport (in pence per passengerkilometer) in the UK (Peirson and Vickerman, 1997, published in Transport an economic and managementperspective, DA Hensher and AM Brewer, OUP, 2001).
Table 5.4 The marginalexternal costs of car passenger transport in the UK (pence per passenger km)
|
Mode |
Global warming |
Air pollution |
Noise pollution |
Congestion |
Accidents |
Total MEC |
LRMC |
Efficient price |
Current price |
|
London car peak |
0.03 |
1.67 |
0.39 |
15.08 |
1.5 |
18.43 |
7.12 |
25.55 |
11.28 |
|
London car off-peak |
0.02 |
1.25 |
0.39 |
1.65 |
1.5 |
4.81 |
6.54 |
11.35 |
10.04 |
|
Inter-urban car |
0.02 |
0.35 |
0.08 |
0.85 |
0.15 |
1.45 |
5.15 |
6.6 |
7.78 |
(Source: Peirson andVickermann, 1997, Table 1:281)
The Peirson and Vickermanstudy is interesting because it estimates efficient prices based on long runmarginal cost pricing and current prices charged for a range of passengertransport modes. On average, current prices for interurban travel in the UK aresimilar to the efficient price. Car external costs are slightly lower thanestimates of current costs including current taxes.
Many journeys at peak timeswith heavy congestion are probably under priced whilst most journeys atuncongested times are probably over priced. This illustrates a key point thatfuel duties are a crude instrument at containing road traffic congestion[7].
A move away from fuel dutytowards road pricing could therefore be expected to increase the cost of urbandriving relative to that of rural road users.
Newbery has estimated thattotal road taxes were 92% of total external congestion charges in 1993. Updatedestimates (Fair and Efficient pricing andthe Finance of Roads, D Newbery, Proceedings of the Chartered Institute ofTransport, & (3)) by Newbery show this burden rose via the fuelintroduction of the fuel duty escalator in the 1990s. Total tax as a percentageof the price of unleaded petrol rose from 60% in 1990 to 82% in 1999, fallingto 76% in 2000. (Table 2.3, TransportStatistics GB, 2000). Road user taxation rose from 19 billion in 1990-1992to 36 billion in 1999-2000. (British Road Federation, April 2000).
Other estimates of congestion costs
In1988 the British Road Federation estimated that the cost of congestion inEngland was around 10 billion per annum. In 1989 the CBI estimated that thecost of congestion in England was around 15 billion per annum (in 1989prices), around two-thirds of which related to London and the South-East. As ashare of GDP the costs of congestion are almost certain to have increasedsubsequently.
According to the CBI everyhousehold in England had to spend 10 per week more than it needed to on goods andservices in order to meet the extra cost to business of road (and rail)congestion. The 12th annual report on motoring (Lex Service Group, 12th Annual Report on Motoring,2000) estimated that congestion, mainly by cars, costs the country 23 billionannually.
6 Road pricing
Newbery (2000) has statedthat,
Roadpricing is the best method of dealing with congestion and would have farreaching implications for the viability and quality of public infrastructure,and ultimately for the quality of life. The average road charge might not needto increase above current levels, for if roads were correctly priced, demandfor the use of the most congested streets would fall, and with it the efficientcharge to levy. Current road taxes are heavy and yield revenue substantiallygreater than the average cost of the current road system. In equilibrium, withefficient road pricing and an adequate road network, one would expect roadcharges to be roughly equal to road costs, so either road charges would fallbelow current tax levels, or substantial road investment would be justified,and possibly both.
Transport modelling as partof the introduction of road pricing in Central London has estimated that acharge of 5 per day for entering the cordon zone will reduce congestion by upto 15% (Congestion tolls wait for thegreen light in the cities crusade against gridlock, The Times, 9thJune 2001).
Hensher and Brewer (Transport an economic and managementperspective, DA Hensher & AM Brewer, OUP, 2001) state thatimplementation of congestion pricing involves recognition of the followingissues:
Congestion pricing would cause some motorists tochange their behaviour;
Congestion pricing would result in a net benefit tosociety;
Congestion pricing is technically feasible;
Institutional issues are complex but can be resolved;
All income groups can benefit given an appropriatedistribution of revenues[8];
Some motorists will lose;
Congestion pricing would reduce air pollution andsave energy;
The political feasibility of congestion pricing isuncertain;
An incremental approach is appropriate.
In addition to the abovebenefits derived from a reduction in congestion, two other key potentialattributes of road pricing are important to consider:
Revenue transfers to finance investment in the publictransport infrastructure (including roads). It might also be possible tointroduce privatisation of certain motorway and trunk road routes in order togenerate more considerable up-front revenues for re-investment in the transportinfrastructure.
Double-dividend economic effects (see below).
Double-dividend effectsrefer to the possibility that revenues from optimal reduction measures might beused to offset the distortionary impact of taxes elsewhere in the economy. Apopular example in the double-dividend literature is the suggestion that greentaxes can finance cuts in national insurance contributions (see. p.156, Double Dividends and Economic Growth, inSACTRA op cit).
However, we are scepticalas to whether any significant double-dividend effect is possible. If theincidence of a lower payroll tax results in a shift onto wages, then in thelong run double-dividend effects will fail to materialise.
However, this is not theonly possible route. Whilst remaining cautious, SACTRA has stated that,
a more promising possibility of reaping a double-dividend maycome through lowering taxes on capital, especially if the distortionary effectsare large relative to those from taxes on labour. If reduced capital taxationwas financed by green taxes, there is a possibility that the long run growthrate could be increased, if it is endogenous.
Wewill now consider some of the problems likely to occur when introducing asystem of road pricing. These problems fall under three headings:
Politicalforces
Inelasticdemand
Substitutioneffects and technology
Political forces
The potential forsignificant political opposition to the introduction of road charging isobvious, especially in the wake of the fuel crisis and public support for thefuel lobby in the Autumn of 2000. The degree of political opposition will alsodepend on whether the congestion charge is applied in selective urban areas oron inter-urban routes.
Despite the commitment inthe 10 Year Plan, and having won a second term of office with a landslide, theGovernment is likely to be cautious before embarking on a policy changeperceived by some as a poll tax on wheels.
It is quite likely thatpolitical constraints mean that any move towards road pricing and congestioncharging will need to be road revenue neutral, with offsetting reductions infuel duties.
A significant politicalbarrier is the problem of investment lags in that revenues from road pricingare immediate, but the financing and construction of improvements in the publictransport infrastructure will take a long time to materialise and therebyprovide a realistic alternative for car users.
Politicians local andnational will also be cautious about the level of road pricing introduced. Ifcharges are set too low, then drivers may pay the extra charge but it will havelittle or no effect on their behaviour and congestion will continue at similarlevels to before. If charges are set too high, then journey behaviour maychange, but it may also provoke a voter backlash.
However, this is not to saythat road pricing is not possible, rather that it will be difficult and maywell take a long period of time to weaken the resistance of car users.
Attitudinal evidence fromthe AA (Paying to use roads Results of the survey of world motoringorganisations, AA Motoring Policy Unit, 2000) suggests that road pricing isfeasible, providing revenues raised are ploughed back into the public transportinfrastructure and car users can see an improvement in congestion and/oralternative modes of transport. However, other attitudinal evidence points tomotorist resistance.
Figure 6.1 Whatwould you support as a way of raising money to improve public transport?
|
|
Strongly support |
Support |
Neither |
Oppose |
Strongly oppose |
|
Double cost of petrol over next 10 years |
1% |
7% |
12% |
38% |
38% |
|
Charging 2 to enter a town at peak hours |
5% |
20% |
13% |
32% |
25% |
|
Charging motorists 1 for every 50 miles they travel on motorways |
4% |
21% |
15% |
36% |
35% |
(Source: British Social Attitudes Survey, 1999)
DETR (Understanding attitudes to transport policy, in Transport trends,DETR, 2001) analysis, summarised in Figure 6.1, shows that significantproportions of the general public and motorists in particular remainsceptical as to the merits of road pricing.
How would the publicrespond to road pricing? We would argue that in order to maintain sufficientpolitical support for road pricing, two criteria need to be satisfied:
Broadly speaking road pricing will need to be revenueneutral in total, as compared with the current fuel duty regime. It isdifficult to know how far the public perceives the current fuel duty regime asbeing akin to road pricing. It is a debatable point. Sir Brian Shaw, Chairmanof the AA Motoring Policy Committee has stated (AA, 2000, op cit) that,
theUK already has road pricing some 10 pence per mile in fuel tax and 40 perquarter in standing charges paid as road tax.
Revenues from road pricing will need to be seen to beploughed back into investment in the public transport infrastructure road andrail. The above AA survey found that in 15 of the 22 motoring organisationssurveyed around the world, motorists would accept road pricing if theirgovernment promised that all the money raised would be used for additionalinvestment. However, 12 of the 15 also said that they would not trust theirgovernment to keep such a promise.
A recent survey by Lex (Lex/RAC Report on Motoring 2000,published in Financial Times, p.3, 26thJanuary 2000) revealed that around 40% of car users said they would use theircar less if public transport were improved. However, attitudes remain wedded tothe car, with motorists perceiving the cost of almost every type of publictransport to be, on average, almost twice as much as by car. 80% of motoristsbelieved it would be very difficult to live without their car. The problems onthe rail network, post Hatfield, will have only intensified this belief.
The introduction of roadpricing in Central London may result in a switch of passenger traffic requiring35,000 extra public transport seats a day, mostly at peak load periods. Thismeans that expected net revenues of around 150 million per annum will need tobe ploughed back into wider improvements in the transport network (The Times, 9th June 2001, opcit).
Inelastic demand
Owing to the limited numberof large-scale road pricing schemes world-wide, it is difficult to estimate theprice elasticity of demand for road pricing. However, economic theory andcommon sense suggest that demand will be inelastic the proportionate changein demand will be less than the proportionate change in price. This is goodfrom the perspective of revenue raising, but bad from the perspective ofalleviating congestion.
Given that the majority ofcongestion is at peak load times it may be the case that the cost of chargingfalls disproportionately on the employer and not the employee and is thenpassed on to the consumer. If this is the case, then the demand response willbe reduced.
A literature review (The effect of fuel prices on motorists,S. Glaister and D. Graham, AA Motoring Policy/UKpia, September 2000) of theelasticity of demand with effect to fuel prices supports the argument thatthere will be an inelastic response to road pricing. Glaister and Graham foundthat:
Estimates of the elasticities of car traffic withrespect to income vary from study to study. In the long run they are almostalways found to be above unity and typically fall in the range 1.1 to 1.8. Thismeans that annual GDP and disposable income growth provides a constant upwardpressure on car use.
The short-term elasticity of traffic with respect toprice is about 0.15 and long term about 0.3. Raising fuel prices is moreeffective in reducing the quantity of fuel used due to engine improvements bycar manufacturers than in reducing the volume of traffic. This means that ifGDP (and real incomes) rises by 2.5% per annum, then with a traffic elasticitywith respect to income of 1.2, traffic growth will increase by 3% per annum. Tohold the level of traffic unchanged, fuel prices would need to rise by 10% perannum (3/0.3).
Should the road pricingscheme employ variable cost pricing in order to tackle the peak load problem?Ideally the answer to this is yes. However, the issue is complicated by theissue of who pays?. The perception and response to road pricing will dependon who pays the bill employee or employer. Research suggests (p. 189, Hensher& Brewer op cit) that business users have a much lower price elasticity of demandthan individual users.
Another possible influenceis that any new capacity which can be expected to be built, or that can befreed up by enticing some drivers off the road, will quickly become filled bypeople who are now being deterred only by congestion itself. Hensher and Brewer(2001, op cit) state that this is,
a well documentedempirical reality.
However,this view has not gone unchallenged. Commenting on a 1994 report from SACTRAentitled Trunk Roads and TrafficGeneration, the British Roads Federation (Roads Investing in Britains Prosperity, BRF, April 2000) hasstated that,
Induced traffic is a real phenomenon but its size andintensity varies widely on local circumstances.
Available evidence suggested that induced traffic wasmost likely on roads in and around urban areas, estuarial crossings andstrategic capacity enhancing inter-urban schemes, including motorway widening.It was least likely to be influential in the case of bypasses of small towns orthe upgrading or widening of existing trunk roads.
SACTRA confirmed that the main determinant of trafficgrowth is economic growth.
Induced effects also represent economic and socialbenefits for those making the journeys.
Othercommentators are also sceptical. Professor S. Glaister, in conversations withthis author, dismissed the threat as nonsense.
Substitution effects and technology
The most obvioussubstitution effect which challenges road pricing is the potential for roadusers to switch from priced to zero-priced roads. The ability to do this willvary, depending on whether it is an intra-urban or urban form of road charging.Closely related to the substitution issue is the precise nature of schemeemployed high tech electronic, low tech paper based etc and the scale ofassociated capital investment in toll booths and/or measures to ring fence acordon. None of these obstacles are insurmountable but they do complicate theintroduction of road pricing in the UK.
7 More roads?
The economic solution toroad traffic problems comprises both a demand-side and supply-side response.The demand-side response is road pricing. The supply-side response is anexpansion of the capacity of the road network, which would include new roadconstruction and expansion in the capacity of existing roads road wideningetc. It must also be remembered that road pricing and road investment arelinked road pricing provides a clear market signal regarding levels of demandand profitability as a guide to future investment decisions.
The CEBR (Centre forEconomics and Business Research) has reported that the UK has the lowest amountof road space per 100 sq. km of any major EU economy. Without a congestionproblem this would raise the issue of whether the road network has sufficientcapacity. With congestion it merely becomes another indicator of inadequateprovision. Despite periods of firm growth in road building, capital investmentin roads has certainly lagged other EU countries. This is all the moreunfortunate because there are a very large number of schemes, identified byDETR, where a cost-benefit analysis has shown very good value for money.
In2000 the CBI stated that,
25years ago the Government spent almost as much on roads and local publictransport as it raised in road taxes. Now it spends half that amount in realterms about 6 billion annually, yet it raises 36 billion from road users.Redressing that balance and changing the culture of transport will be key toa successful programme.
The CBI has called (CBI Press Release, 19thApril, 2000) for a 25 billion investment in the strategic road network overthe 2000-2010 period new roads, better use of existing roads and maintenance.
The AA has also called fora substantial programme of road investment (Buildinga transport network to rival any in Europe Investment: what needs to be done,AA Motoring Policy, 2000). The AA has stated that,
Investmentmust rise quickly from just under 6 billion in 1999 to 9 billion per annum by2003-2004 just to address the backlog of repairs and deliver a core programmeof routine maintenance and stalled improvements; and that because Britain hasinvested so little for so long, many projects will be needed investment needsto build steadily from 9 billion in 2003-2004 to 12 billion per annum by2007-2008.
In April 2000 the British RoadsFederation (BRF) called for an average 9 billion per annum expenditure onroads over the next decade. At the time this meant that HM Treasury would needto finance an additional 40 billion of road investment over the 2000-2010period. BRF figures show that only 14% of road user taxation was spent on roadsin 1999-2000. This compares with a recent peak of 30% in 1992-1993.
How has the Government responded to these demands forgreater investment?
In July 2000 DETR published Transport 2010 The 10 Year Plan, whichlaid out the Governments plans for road investment over the next decade. TheGovernment forecasts that total public and private investment and publicexpenditure (maintenance etc) on roads (strategic network, local and London)will total 59.1 billion over the 2001-2002 to 2010-2011 period.
The Government claims that incombination with investment in other parts of the transport network, the planwill permit:
Congestionto fall below current levels, particularly in large urban areas.
Bottlenecksto ease by targeted widening of 360 miles of the strategic road network.
80major trunk road schemes.
100new bypasses on trunk and local roads to reduce congestion and pollution.
130other major local road improvement schemes.
Completionof the 40 road schemes in the Highways Agency Targeted Programme ofImprovements.
Eliminationof the maintenance backlog for local roads and bridges as part of a 30 billionprogramme.
HGVlanes on congested strategic routes to provide priority for lorries and saferlanes for cars.
Much of the capex expenditure shownin the 10 Year Plan is end loaded towards the latter part of the decade. Thisis unsurprising given the planning and construction periods entailed in newroad developments, but it also means that any easing in congestion is not justaround the corner. In addition, the 10 Year Plan shows a relatively smallcontribution from private sector capital around 5bn of private investmentcompared with 27bn of public investment (Table A2, Transport 2010, DETR, July2000). The IoD would argue that a successful road pricing policy should alsoprovide the incentives for greater private sector capital investment in roads.
The Governments planned roadexpenditure is welcome, but in order to see it through, there will need to besubstantial political backing owing to the influence of the environmentallobby:
NIMBYISM(not in my back yard) Considerable local opposition can be generated by thoseopposed to new road developments which directly impact on their properties.However, there will also be considerable local environmental support forby-pass schemes if they improve traffic flow and take vehicles out of the towncentre. Nimbyism is a double-edged sword. At the moment, there are veryconsiderable environmental costs in town centres and other bottlenecks wherecars burn fuel going no-where. Improvements in engine technology it wouldtake 50 of todays new cars (Source: BRF, p.19, op cit) to create as muchpollution as one of a generation ago and moving traffic out of town centresoffer the possibility of significant environmental improvement. Road vehiclesare often quoted as the fastest growing source of greenhouse emissions, but infact the growth has been relatively slow. For example, AA research shows thatbetween 1992 and 1997, when car traffic increased by 10%, emissions of CO2 rose by just 3%.
SWAMPYISM(as in Swampy, a well known eco-warrior) Swampyism refers to a generalenvironmental opposition to the car and new road developments full-stop.Swampyism has been high profile, as in the case of the Newbury by-pass andTwyford Down schemes. However, creative and sensitive approaches to theenvironment are certainly possible and need not significantly damage the localenvironment, outside of the obvious impact of the actual construction.Statistics show that most levels of toxic pollutants from road transport peakedin about 1990 and have been falling since then (Transport Statistics Great Britain, 2000 Edition, DETR). Governmentfigures see Table 7.3 show that various pollutants fell dramatically duringthe 1990s. There are many potential road schemes where the estimated benefitsfar outweigh the costs, which are not being funded and developed.
Figure 7.1 Environmental improvements in roadtransport
|
Pollutant |
1990 level (000 tonnes) |
1998 level (000 tonnes) |
% reduction |
|
Nitrogen oxide |
1,305 |
823 |
-37% |
|
Carbon monoxide |
4841 |
3491 |
-28% |
|
Volatile organic compounds |
1,170 |
747 |
-36% |
|
Lead |
2.2 |
0.6 |
-73% |
|
Particulates PM10s |
67 |
42 |
-37% |
(Source: Table2.8, Transport Statistics GB, 2000)
8 Technology issues
The technology of roadpricing has developed an enormous literature which is beyond the scope of thispaper. As a result, in this section we will merely focus on the key issues.Underlying this analysis is our conclusion that whilst road pricing schemeswill often be very complex developments, they are both feasible and inevitable.
The new technology ofelectronic tolls no longer requires motorists to halt at tollbooths, thereforeit can prevent additional congestion. Drivers could be given an electronicnumber plate which signals to the recording computer the presence of thevehicle. This would be the most direct way to charge the amount specific to theroad and the time of day. The device could charge users via bank account ormonthly bill. Because there have been many objections to individual location byelectronic detectors the use of smart cards appears to be preferable. Theelectronic licence plate would be loaded with the smart card and would debitpayments. Only if the card were exhausted would the central computer monitorand bill for road use.
Clearly the introduction ofsuch schemes would take time, and over the intervening period area licencescould be sold for very congested zones, such as city centres. In the case ofSingapore, the impact of the introduction of such a scheme was an immediatereduction of 25,000 vehicles during the peak time and a rise in traffic speedof 22% (Road pricing: the case for andagainst, C Rolle, 2001).
Key issues include:
The choice between high-tech electronic or low-techpaper based systems;
The issue of failure rates and enforcement if thesystem records hundreds of millions of vehicle movements, even a 99% accuracyrate will generate millions of errors. There are therefore important issues ofwhether monitoring and invoicing or prior smart card purchase and monitoringbased systems are chosen.
In urban areas stopping traffic to collect tolls isself-defeating, because it slows traffic and would require a ridiculous numberof toll-booths. On the strategic road network toll booths are more plausible,but will also involve substantial capital investment at motorway exits etc.
Cars can now be equipped with electronic devices thatemit signals to relayed highway monitors and car owners are billed for theirhighway useage. Highway and bridge tolls are now collected electronically in,for example, Australia, California, Florida, Texas, France, Italy and Norway(p. 185, Hensher & Brewer op cit). However, the deployment of systems ofgantries and transponders in urban areas is very complex. Simple ticket basedsystems may have a significant role in urban areas.
The issue of how to link schemes between cities,without the need for multiple electronic tagging.
Notwithstanding thesedifficulties, UK driver experience of motorway tolls on the continent showsthat such road pricing schemes can be seen to be effective. In addition,examples of successful urban road pricing schemes are also available.Congestion in Trondheim, Norways third largest city centre, has beendramatically reduced by the implementation of a toll ring that collects chargesduring working hours.
In a UK context the futurefor road pricing could well be determined by the success or failure of theintroduction of road pricing in London. The current proposal is to levy a 5charge on vehicles heading into Central London between 7am and 7pm. Cameraswill record vehicles entering the zone at 112 entry points and computers willcheck vehicle registration numbers against those whose drivers have paid thecharge.
Much rests on the successor failure of the planned Central London scheme. John Dawson, Policy Directorof the AA, has stated (The Times, 11thJuly 2001) that there will be problems making the technology work. Problemscited by the AA include: a lack of capacity on the London ring road to take theadded traffic which ripples out from the charge zone; deficiencies in thedriving licence system for tracking motorists; and a tube system alreadystretched to capacity.
Professor David Begg,Chairman of the Commission for Integrated Transport has stated that,
"Ken Livingstone's congestion charge strategy is criticalfor the whole economy because if it fails in the capital, Ministers will backright off, which could set the agenda back a generation. The Government willnot achieve its targets for tackling congestion by 2010 without congestioncharges".
(The Times 11thJuly 2001)