Fuel for Thought: The what, why and how of motoring taxation

7/31/2019 Fuel for Thought: The what, why and how of motoring taxation

1/120

Fuel for ThoughtThe what, why and

how of motoring taxation

Paul Johnson,Andrew Leicesterand George Stoye

Institute for Fiscal Studies

May 2012


2/120

b

The Royal Automobile Club Foundation or Motoring is a transport policyand research organisation which explores the economic, mobility, saety andenvironmental issues relating to roads and their users. The Foundation publishesindependent and authoritative research with which it promotes inormed debateand advocates policy in the interest o the responsible motorist.

RAC Foundation8991 Pall Mall

LondonSW1Y 5HS

Tel no: 020 7747 3445www.racoundation.org

Registered Charity No. 1002705

May 2012 Copyright Royal Automobile Club Foundation or Motoring

This report has been prepared or the RAC Foundation by the Institute orFiscal Studies. The report content is the view o the authors and does notnecessarily represent the views o the RAC Foundation.
http://www.racfoundation.org/http://www.racfoundation.org/


3/120

c

Fuel for Thought

The what, why andhow of motoring taxation

Paul Johnson,Andrew Leicesterand George Stoye

Institute for Fiscal Studies

May 2012


4/120

AcknowledgementsThe authors are grateul to the RAC Foundation or unding this work, and orco-unding rom the ESRC Centre or the Microeconomic Analysis o PublicPolicy (RES-544-28-5001). They would like to thank David Bayliss, StephenGlaister, Peter Levell, Peter Mackie and Tom Worsley or useul commentson early drats, and Ben Kennington or meticulous copy-editing. Data romthe UK Living Costs and Food Survey are collected by the Oce or NationalStatistics and distributed by the Economic and Social Data Service. Datarom the National Travel Survey are collected by the National Centre or SocialResearch and distributed by the Economic and Social Data Services. Crowncopyright material is reproduced with the permission o the Controller o HMSOand the Queens Printer or Scotland.

Contact details

[email protected]; [email protected] (corresponding author);[email protected]

DisclaimerThe contents o this report represent the views o the authors alone, and notthose o the Institute or Fiscal Studies or the RAC Foundation. Any errors arethe sole responsibility o the authors.

i


5/120

Contents

RAC Foundation Viewpoint iv

Executive Summary ix

Introduction 1

Economic Principles o Motoring Taxation 5

2.1 Revenue raising 6

2.2 Distributional eects 7

2.3 External costs o road use 8

2.4 The role o non-tax instruments 13

2.5 Other relevant issues 14

The Structure o UK Motoring Taxation 17

3.1 Fuel duties 18

3.1.1 Current rates o duty 18

3.1.2 Historical duty rates 193.1.3 Duty dierentiation 22

3.1.4 International comparisons 26

3.2 Vehicle Excise Duty 29

3.2.1 Current and historic rates 29

3.2.2 The eect o VED on emissions 31

3.2.3 International comparisons 33

3.3 Company car and uel taxation 33

3.3.1 Company car taxation 34

3.3.2 The taxation o employer-provided uel 37

3.4 London Congestion Charge 38

3.5 Distributional aspects o motoring taxes 39

Motoring Tax Revenues 45

4.1 Current and historical revenues 46

4.2 Revenues in the uthure 49

1

2

3

4

ii


6/120

4.2.1 Short-term revenue projections 49

4.2.2 Raising the share o green taxes in total receipts 50

4.2.3 Long-term projections 53

Do UK Motoring Taxes Accord with the Economic Principles? 595.1 Taxes and the external costs o motoring 60

5.1.1 Fuel taxes and marginal externalities 60

5.1.2 Duty dierentiation between petrol and diesel 62

5.1.3 The role o Vehicle Excise Duty 63

5.2 Other economic principles 66

5.2.1 Simplicity 66

5.2.2 Consistency 66

Policy Options or the Future 69

6.1 Reorming the current set o motoring taxes 70

6.1.1 Are uel taxes too high? 70

6.1.2 A air uel stabiliser 73

6.1.3 Reorms to other existing motoring taxes 78

6.1.4 Multi-part instruments 78

6.1.5 Building congestion-related payments into existing instruments 79

6.2 Introducing new instruments 81

Conclusions 87

Reerences 91

Appendix 95

5

6

7

8

iii


7/120

RAC Foundation ViewpointAt each Budget, motoring taxation catches the headlines. With good reason:the average household spends just over 50 a week on motoring. Receipts

rom uel duty and Vehicle Excise Duty (VED) are one o the biggest sources ounds or the Treasury, contributing about 38 billion a year, some 7% o all theExchequers income. The government uses more than one way o classiyingtaxes, but motoring taxes account or 85% o all those classied as greentaxes. Less than one third o this money is spent on national and local roads.

Nobody likes paying taxes. Yet, like death, they are a act o lie. The best onecan hope or is that they are at least equitable, understandable and justiable.We asked the experts at the respected and independent Institute or FiscalStudies to investigate how motoring taxes are being determined, and whetherthey meet these three criteria.

Their answer is worrying, but probably o little surprise to even casualobservers. Whilst there are at least two quite dierent undamental principlesby which these important taxes might be set to both o which governmentappeals rom time to time what happens in practice bears little relation toeither o them.

This report relates a comprehensive history o policy announcements onmotoring taxation. The clear inerence is that there has been no coherent, long-term policy.

Rates o uel duty and VED are changed requently and, it seems, proposalsare revised even more so. The approaches taken by successive governmentshave oten smacked o policy made on the hoo, with short-term politicalexpediency dominating any willingness to establish a air system which istransparent and appropriate, to the needs o both the individual and the state.

Against this backdrop, a new diculty has arisen or government. Despite

a projected growth in trac the Department or Transports January 2012estimate1 is or 44% more trac by 2035 the IFS, using the governmentsown gures, notes that revenue rom motoring taxation is set to drop by13 billion a year by 2029 (to 25 billion, rom 38 billion in 2010). This issimply due to the improvement in the uel eciency o vehicles, as existingtechnologies are rened and new ones are adopted in response to thegovernments climate change targets or greenhouse gas reduction.

Will the government stand by while motoring tax revenues decline? The IFSreport estimates that to preserve the current level o uel duty revenue, the rate

1 Department or Transport (2012). Road Transport Forecasts 2011. Retrieved 13 April 2012 rom http://assets.dt.gov.uk/publications/road-transport-orecasts-2011/road-transport-orecasts-2011-results.pd.

iv
http://assets.dft.gov.uk/publications/road-transport-forecasts-2011/road-transport-forecasts-2011-results.pdfhttp://assets.dft.gov.uk/publications/road-transport-forecasts-2011/road-transport-forecasts-2011-results.pdfhttp://assets.dft.gov.uk/publications/road-transport-forecasts-2011/road-transport-forecasts-2011-results.pdfhttp://assets.dft.gov.uk/publications/road-transport-forecasts-2011/road-transport-forecasts-2011-results.pdf


8/120

per litre would have to be increased by over 50% above the present 57.95p/litre (plus VAT, as at March 2012). Would that be politically acceptable? Or evenjustiable? The RAC Foundation argues that in the longer term, governmentmust redesign the system o motoring taxation to be airer and moretransparent across the spectrum o users o dierent technologies, and to raise

a total revenue that is politically acceptable given other sources o revenue tothe Exchequer.

The IFS report shows that motoring taxation is, on average, mildly progressive.For the 10% o households with the lowest incomes, uel duty and VEDaccount or 3.6% o total expenditure, whereas they account or at least 5.9%in the case o higher-income households. But there is no such thing as anaverage household; either a household runs at least one car or it runs none.In the lowest income band, about hal o all households do run at least one car.Amongst households that incur the considerable cost o having a car, currentmotoring taxation is regressive. The cost o uel and VED represents 8.1% othe budget o the poorest 10% o car owners, but only 5.8% o the 10% withthe highest incomes.

Some will insist that motorists are taxed heavily simply because they can be,and the government needs the money. Many people nd that use o the caris unavoidable, and the movement o goods an essential activity. In practicethere is little alternative to paying up. I a government insists that a particularsum must be raised by indirect taxation, one principle is that a good structureis one which taxes the less responsive sectors relatively highly and the moreresponsive sectors lowly. This yields the required total revenue whilst causingthe least distortion to the quantities people buy; and that does the leastdamage to the economy and total personal welare. As the IFS notes, despiteoccasional pronouncements to the eect that the government needs themoney or cannot aord a motoring tax reduction, government does not reallyattempt to sustain this line o argument.

The report discusses in detail the second line o argument as to how motoringtaxes might be structured: rather than simply raising revenue, taxes should

refect the various economic and social impacts that the use o vehicles hason other road users and the community at large. These impacts include traccongestion, accidents, visual intrusion, noise and air pollution, and greenhousegas emissions.2

Revenue raising and paying the external costs are quite distinct principles.The amount o tax gathered under a revenue-raising regime might bear noparticular relation to the yield that refects external economic, environmentaland social impacts.

2 This is similar to the polluter pays principle that guides charges and taxation in other sectors, includingenvironment and waste.

v


9/120

So what o the external costs justication? By this ormula is tax too high ortoo low? The IFS notes that it is sensible to include within conventional uelduty a charge that refects the carbon price o burning a litre o uel: i carbon isthe problem then tax (price) it directly. This suggests a contribution to the dutyrate o around 14p/litre at current ocial carbon prices.

The RAC Foundation agrees with this approach to carbon taxation. Butwe note that other sectors burning hydrocarbons and producing carbondioxide pay lower rates o duty than road users: agriculture, railways and buscompanies. Most notably, private householders pay no duty on domestic uels,and a reduced rate o VAT. 25% o all greenhouse gas emissions come romburning these uels, as against 19% rom roads.3

The advent o the pure electric car charged up rom the domestic electricitysupply creates a urther problem. Drivers will pay the carbon charge implicit inelectricity due to the generating companies having to pay or carbon credits.But the price o credits is currently much lower than that used by the UKgovernment to make policy decisions. From the perspective o the purchaser,at current showroom prices ater government grants, electric vehicles can onlycompete with conventional vehicles because they are excused duty on theiruel equivalent to that imposed on petrol and diesel. It is uncertain how longgovernment will be willing to preserve this inconsistency.

On top o the 14p or carbon you can reasonably add to uel duty elementscorresponding to external costs such as air pollution. But then you havea problem: by ar the most costly damage inficted by road users is traccongestion. This is not directly related to the amount o uel used, and hence itis hard to justiy actoring it into uel duty. Trac congestion varies enormouslyby time and place. Thus no single, universal rate o uel duty can be regardedas appropriate or all. The IFS estimates that at the current rate o duty, abouthal o all vehicle miles are taxed too highly many o them on uncongestedrural roads; one quarter are taxed about right; and the remainder are chargedtoo little in some congested urban locations, ar too little.

The IFS is driven to the conclusion that i the paying the external costsjustication is to be sustained then a signicant portion o current uel dutymust be replaced by some orm o pay-as-you-go charge based on distancedriven in congested conditions. Such a charge moderates demand, andthereore reduces congestion. There is a net overall gain, and the economicvalue o the road is increased. The amount o revenue is determined by thecharacteristics o the road and the level o demand or use o it. But this begsthe question: who gets the benet o the use o the revenue?

3 Committee on Climate Change (2012). The 2050 target achieving an 80% reduction including emissionsrom international aviation and shipping. Retrieved 13 April 2012 romhttp://hmccc.s3.amazonaws.com/IA&S/CCC_IAS_Tech-Rep_2050Target_April2012.pd.

vi
http://hmccc.s3.amazonaws.com/IA&S/CCC_IAS_Tech-Rep_2050Target_April2012.pdfhttp://hmccc.s3.amazonaws.com/IA&S/CCC_IAS_Tech-Rep_2050Target_April2012.pdf


10/120

Road users cause congestion to each other, and most o the disbenets aresuered by drivers themselves, not the population at large. So it might beargued that, in airness, the revenues should be used to compensate at leastto some degree those paying the taxes. This could be achieved throughimprovement o the road (or the general transport network). I implemented,

road users as a group could be made better o. But i (as is currently the case)most o the revenue is swallowed up by general government expenditure, thenmost motorists paying the tax will be made worse o.

A urther consideration is that i congestion is severe, and hence the revenuein a particular location rom a correctly set pay-as-you-go charge is high,then this is a clear signal that there is not enough road capacity. The questionarises: should the revenue rom the tax be used to und an expansion ocapacity? This might come about either through more active managemento the existing asset or by physical enhancement. In other words, there is acase or ploughing the tax revenue back into the road system to alleviate thespecic congestion generating the revenue, rather than using it or generalexpenditure. This is essentially what happens in our other utilities: the chargesor water and power are dedicated to paying operating costs, and the costso providing and nancing sucient inrastructure, to ensure delivery o anadequate level o service.

Overall, one cannot help but think that the current levels o motoring taxationhave little to do with either sound revenue-raising principles or external costarguments. Rather, they are an accident o history by which the governmentraises as much as it can get away with.

The IFS report points out the undamental diculties posed by the notion oa air uel price stabiliser. It is not possible to make a credible promise toprotect consumers rom the consequences o variation in the global US dollarprice o crude oil. As the report notes, that is essentially the conclusion whichthe government has itsel reached. What government can and should do is toset out the principles by which it intends to set tax rates, and thereby establisha greater degree o stability.

vii


11/120

Concluding remarksIt is easy to see why the Treasury likes motoring taxation, especially uel duty.It is simple to collect, is almost impossible to evade, and brings in signicantrevenue: some 7% o all tax receipts. It is also a dependable source o cash

with less short-term fuctuation in revenue than some other taxes. The demandor road uel is relatively inelastic: most people, most o the time, have to usetheir cars to go about their daily business. I the cost o uel rises at the pumps,motorists might be able to make marginal changes to their driving behaviour,but they are more likely to cut back spending elsewhere rather than risk endingup in a position where their tanks run dry or they lose mobility. To all intentsand purposes, the greater part o transport spending is not discretionary.

But what is the point o uel duty? Is it an environmental tax? Is it aboutchanging behaviour? Is it the bill drivers must pay to account or the damagethey impose on society? Or is it simply a revenue raiser?

The conclusion must be that there is no clear answer.

That is why the RAC Foundation urges the government to do the ollowing:

1. admit that uel duty is a requently altered tax without adequate rationale;2. recognise the acute impact that high uel prices have on car-owning

households, particularly those on low incomes;3. recognise the dramatic all in uel duty revenue that is inevitable as the

decarbonisation o road transport proceeds apace; and4. start a dialogue on what a more transparent system o motoring taxation

might look like, which should include a consideration o alternatives to uelduty and VED.

There are 34 million drivers in the UK. Historically they are not a militant group.Yet there have been sporadic protests about the price o uel and the level ouel duty witness the blockade o oil reneries and storage acilities in 2000that brought London to within a ew hours o running out o ood and was

arguably the biggest threat to the premiership o Tony Blair.

When the economic climate is benign, people are condent about their utures,and the cost o running a car (not least the price o uel) is a manageable parto household budgets, then motoring taxation might be an irritant to manybut not a source o mass discontent. But today everything is rather dierentand each time drivers visit the pumps their nances are dealt another severeblow in large part because o a tax system they increasingly resent and donot understand the reasons or. It is time or policymakers to recognise theseconcerns and address them.

viii


12/120

Executive SummaryPrivate transport costs make up some 13% o the average household budget,and it is estimated that in 2011/12 motoring-related taxes (uel duties and

Vehicle Excise Duty, VED) will raise more than 38 billion, or almost 7% ototal revenues. The benets o a well-designed system o motoring taxes arethereore potentially substantial. This report provides a detailed examinationo the current system o motoring-related taxes in the UK, assesses how wellthese taxes match up to the economic principles underlying good tax policy,and makes a number o suggestions or reorm.

The key insights are that motoring taxes are justied primarily by the externalcosts (externalities) generated by road use. These costs are not taken intoaccount in private choices o how much to drive. Without policy action, thisleads to excess motoring rom the perspective o society as a whole. Taxeson motoring can increase the private costs internalise the externality andcorrect this problem. Ideally, these corrective taxes should be targeted on theexternalities directly (or on a close approximation to them), and set at a levelequal to the marginal external cost at the socially optimal level o demand.

However, we conclude that the current tax system does a poor job at targetingthese external costs. In particular, uel taxes are completely unable to capturevariation in external costs by time and location most notably the costs ocongestion. A system o road pricing or congestion charging which is able totake such variation into account more accurately would be preerable. Thisconclusion is not one which is new or surprising, and policymakers have shownlittle appetite to act on it beore. However, as our ndings here make clear, realimpetus or reorm may come rom scal considerations. Without action, thereis likely to be a long-term erosion o the motoring tax base. Road use, though,is expected to continue to increase. Road pricing not only targets the externalcosts o motoring more precisely, generating the potential or signicantwelare gains, but also provides a more robust revenue source.

Motoring taxes are already shrinking as a proportion o total receipts, and areexpected to amount to just 6% o revenues by 2016/17. This would be thelowest share since 1954. The government has pledged to increase the share ototal revenues coming rom environmental taxes, a target which is made moredicult by the alling share o motoring taxes. On present orecasts, the pledgewill be hit with only 1 billion to spare by the end o the Parliament less thanthe cost o reezing uel duty rates or a single year. I motoring taxes remainedat their current share o receipts, the pledge would be hit with more than4 billion to spare. That said, there is no compelling environmental or economicreason or such a pledge to be made in the rst place.

More pressing in the mind o the Chancellor ought to be long-term projectionsrom the Oce or Budget Responsibility (OBR) which show that, by 2029/30,

ix


13/120

uel duty will make up only 1.1% o GDP compared to 1.7% today. For VED,the gures are 0.1% and 0.3% respectively. The total decline in motoring taxesis equivalent to 13.2 billion a year in todays terms. This is roughly the revenuegenerated by increasing the basic rate o income tax rom 20p to 23.4p,increasing VAT rom 20% to 22.7%, or increasing uel duty by more than 50%.

The OBR orecasts or VED may be unduly pessimistic, as they assume nouture changes to the carbon thresholds on which VED liabilities are assessed which, given the greening o the feet, seems unlikely. However, the orecastsor uel duty may be optimistic. I oil prices remain high, there will undoubtedlybe resumed pressure on the Chancellor to cancel uture uel duty increases, ashappened in the 2011 Autumn Statement, whereas the OBR gures assumeduty rises in line with infation each year.

There are a number o externalities associated with motoring. Some, such ascarbon dioxide emissions, relate very strongly to uel consumption, makinga tax on uel an appropriate instrument. At current carbon values, burning alitre o uel generates a marginal externality o around 14p. Others includingnoise costs, accidents and, above all, congestion are not at all related touel use. Instead, they vary by time and location o driving. Department orTransport (DT) estimates suggest that the marginal external costs associatedwith around hal o the kilometres driven in the UK are very low less than 5pper kilometre. However, driving in the most congested areas o the countryis associated with extremely high marginal external costs o almost 2.50per kilometre. These gures explain why a road pricing system which moreprecisely captured this variation could have signicant welare gains, estimatedby the Eddington Review in 2006 at up to 25 billion per year in the long run.

The issues surrounding economic eciency and revenue erosion arehighlighted in particular by the potential or increased electrication o thevehicle feet. Estimates rom the Committee on Climate Change suggest thatby 2030, 60% o new cars sold will be electric. In the current motoring

x


14/120

tax system, electric cars generate no revenue: they consume no hydrocarbonuel and are exempt rom VED. These vehicles still, though, generatecongestion externalities which are at present untaxed.

There are, o course, substantial barriers to road pricing. Public opinion is likely

to be hostile. By making it clear that road pricing would be largely or whollyoset by reductions in taxes on uel, opposition may well be more muted, andperceptions o road pricing have been seen to change or the better in placeswhere it has been introduced, both in the UK and abroad. A complex pricingstructure may be costly to implement and run. Simpler systems targeted on themost congested areas could capture a large raction o the potential benets.

There would also be winners and losers rom a move towards road pricing.Understanding who would be aected, and how, would be crucial or goodpolicymaking. That is why the government should start preparing the groundor this kind o reorm as soon as possible and allow or a proper evidencebase to be ormed. Failure to act is not a neutral choice: i this course is taken,the government will have to replace signicant amounts o lost revenue andthat will also create winners and losers, whilst leaving a motoring tax structurethat is ever more uncoupled rom the external costs generated.

Excise taxes on uel are by ar the largest source o motoring tax revenue,making up 85% o motoring-related receipts. The majority o uel is taxedat 57.95 pence per litre, with 20% VAT charged on top. The history o ueltaxes is dominated by the duty escalator which ran rom 1993 until 1999,and saw real duty rates increase by more than 70%. In the decade ollowingthe end o the escalator, by contrast, there were no real-terms increases induty rates. Indeed, the most notable eature o recent uel tax policy hasbeen the large amount o uncertainty about what rates will be set, with dutyincreases repeatedly announced, then delayed, then abandoned altogetherunder successive Chancellors. This is not a recipe or good policymaking, nordoes it provide reliable long-term signals to road users or engender certaintyin the public nances. The idea o a air uel stabiliser tying the rate oduty to the underlying pre-tax uel price has gained some traction recently.

However, such a policy would be very dicult to implement or two principalreasons. Firstly, it is extremely hard to identiy the trend in pre-tax prices,and thus the appropriate tax rate to set. For instance, i long-term increasesin uel prices were to be mistaken or short-term shocks, then a stabiliserwould lead either to a permanent erosion o uel duty revenue, or else sharpadjustments to duty rates once the mistake had been realised underminingits power as a stabiliser. Secondly, increases in oil prices do not lead to anysustained increase in tax receipts, and so the policy could only urther increaseuncertainty about the public nances.

In the absence o a move towards road pricing, is the current level o uel dutyjustied by the external costs o motoring? This is extremely dicult to answersatisactorily. DT estimates suggest that a kilometre driven on the current

xi


15/120

network generates, on average, a marginal external cost o 19.5p, substantiallyhigher than current uel duty rates. However, at optimal levels o road use(which is the appropriate measure or judging the correct tax) congestionexternalities would be much lower than at current levels. It may also be moreappropriate to weight the low externality kilometres more heavily to determine

the correct uel duty rate i those driving in the most congested areas areleast responsive to prices. Estimating the appropriate uel tax rate wouldrequire detailed modelling which is beyond the scope o this report. However,the point that a single rate o duty cannot adequately capture the variation inexternalities remains the crucial insight.

The other major motoring-related tax VED accounts or about 16% omotoring tax receipts. The most signicant reorm to VED came in 2001, whenthe tax was based or the rst time on the uel eciency o the vehicle. Thetax now provides direct incentives to buy more ecient cars. In recent years,the average eciency o new cars has increased markedly, though the extentto which this is attributable to VED reorms, rather than resulting rom tighterregulation and higher uel costs, is very hard to know. What is clear is thatthe incentives within the VED system to buy more ecient cars have becomear stronger. In 2001, the dierence in lietime VED payments (over 15 years)between the most and least polluting cars amounted to 2,400, whilst by2011 it had risen to 7,020. The introduction in 2010 o a showroom tax, ahigher VED rate or polluting cars in the rst year, helps to explain this trend.This reorm may be justied i people are unable to properly account or uturerunning costs in making purchase decisions, though the evidence on this is notully conclusive. More UK-specic research o both issues would be helpul orpolicymakers.

Incentives to buy more ecient cars have also been built into the system ocompany car and uel taxes. Fleet sales account or around hal o new carpurchases, so the tax treatment o company cars is potentially very signicant.The oddest eature o the company car tax system is that it penalises dieselvehicles more than petrol vehicles, whereas both uel duty and VED treat petroland diesel the same. From an environmental perspective there may in act be

a reason to tax diesel more heavily, since a litre o diesel emits slightly morecarbon than a litre o petrol. However, it would seem the most sensible policyto treat the two uels consistently across the set o motoring taxes.

There is no compelling rationale to tax motoring more heavily than other ormso consumption on the basis o distributional concerns. Motoring taxes appearat present to be slightly progressive, having a little less impact on poorerhouseholds than on richer households. This is related to vehicle ownership:41% o the poorest tenth o households own a car, compared to 96% o therichest tenth. There is also some evidence that richer households own larger

and more-polluting cars, which are more heavily taxed. A uller analysis o thechanging distributional eect o motoring taxes, and how this might look in theuture, would be a useul contribution to the debate surrounding tax reorm.

xii


16/120

1. Introduction

Fuel for Thought The what, why and how of motoring taxation

Motoring is an important part o daily lie. More

than three quarters o UK households own atleast one car.1 In 2010, the average household

spent 52.80 per week on private transport,

representing around 13% o total expenditure.2

In the same year, almost 496 billion vehicle

kilometres were driven on British roads. Whilst

recent years have seen a small all in road use,

the long-term upward trend is clear (see Figure

1.1). Moreover, the most recent Department or

Transport (DT) orecasts predict that road use will

continue to grow, with total road trac in England

predicted to be 43% higher by 2035 than it was

in 2003 (table TRA9905 rom DT, 2011).

1 Authors estimate rom the Oce or National Statistics (ONS) 2009 Living Costsand Food Survey.2 ONS (2011) estimates based on 2010 Living Costs and Food Survey.

1


17/120

Introduction

Figure 1.1: Total vehicle kilometres driven on British roads, 19502010

Totalveh

iclekilometres(billion)

600

500

400

300

200

100

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

2010

0

Source: DT (2011: 1)Note: Figures converted to kilometres using 1 mile = 1.609 km.

Motoring taxes also represent an important source o revenue or theExchequer. In 2011/12, total receipts rom the major motoring-related taxes excise duties on vehicle uel (including associated VAT) and Vehicle ExciseDuty (VED) are expected to be 38.3 billion, or 6.7% o total revenues.3The benets o a well-targeted and eective system o motoring taxes couldthereore be substantial.

This report, however, suggests that we are some way rom this ideal. Weprovide a detailed overview o the current system o motoring taxation and howit has changed over time. We compare the main eatures o the current system

to the economic principles which should underlie good policy in this area, and

3 Authors calculation rom Oce or Budget Responsibility (OBR) November 2011 gures.

22


18/120


suggest options or reorm. In addition, we highlight the issue o the long-termsustainability o revenues under the current system, which (together with soundeconomic principles) lead almost inexorably to the conclusion that, in the end,some orm o pricing or road use will be necessary.

The report is organised as ollows. Section 2 sets out the economic principleso road taxation, ocusing on the external costs associated with road use,which orm the essential rationale or intervention. Section 3 describesthe current and historical structure o motoring taxes in the UK, with someassessment o how the UKs situation compares to international experience.Section 4 analyses motoring tax revenues and looks ahead to what mighthappen to receipts in the uture, both short- and long-term. Section 5discusses how the current system compares to the economic principlesupon which an ideal system would be based, beore Section 6 outlines policyoptions or the uture, both in the context o the current set o motoring taxesand considering what might be gained rom introducing new tax instruments.Section 7 concludes.

3


19/120

Introduction 4


20/120

2. Economic Principles

of Motoring Taxation


This section provides a non-technical discussion

o the main economic principles relevant to howmotoring is taxed. We argue that whilst issues

such as the need to raise revenues eciently,

and distributional concerns, are relevant, the

main reasons or treating the consumption o

motoring dierently to other consumption in the

tax system are the external costs (externalities)

associated with road use. A well-designed set

o motoring-related taxes should, in principle, be

grounded in evidence about the scale o these

costs, with tax instruments being appropriately

targeted on them.

5


21/120

Economic Principles of Motoring Taxation

Revenue raising

As noted above, motoring-related taxes are a signicant source o revenuesor the Exchequer. We discuss revenues in more depth in Section 4. In general,we could consider motoring taxes as part o the wider tax system and assesswhat a ully optimal system o taxes would look like. Such an approach wouldtypically be based on nding the most ecient way to raise a set amounto revenue. Some early, simple results include the Ramsey (1927) inverseelasticity rule, which states that when the only sources o revenue are taxes oncommodities and given a market in which there is perect competition taxesshould be higher on goods with less price-elastic demand. Modications to this

simplied world generate other insights or example, the Corlett and Hague(1954) nding that taxes should be lower on goods which are complementsto working and substitutes or leisure, in order to reduce the labour supplydistortions o taxation. O course, in reality, the world is much more complex.There are many dierent sources o taxation income, consumption,transactions o particular goods, corporate prots and so on and a large seto imperect markets in which people make decisions while aced with diverseincentives and incomplete inormation.

To the extent that the elasticity o demand or motoring is relatively low (i.e. the

demand is inelastic), and motoring is associated more strongly with leisure thanwith work, there may be a case or taxing motoring more heavily than otherorms o consumption.4 Parry and Small (2005), or example, in their work onoptimal taxes on vehicle uel (under the assumption that this is the only ormo available motoring tax) estimate that or the UK, the optimal Ramsey tax(the revenue-raising component) o uel duty is about 22% o the tax rate thatwould be justied by external motoring costs. We discuss the Parry and Smallpaper in more depth in Section 6.

4 In the UK, the general consumption tax is the Value Added Tax (VAT). This is applied to the purchase

o new cars, and to vehicle uel (and also on top o uel duty), as well as to the costs o vehiclemaintenance and servicing. For second-hand cars, VAT is payable under the margin scheme, on theexcess o the resale price over the acquisition price by second-hand dealers(see www.hmrc.gov.uk/vat/start/schemes/margin.htm).

2.1

6
http://www.hmrc.gov.uk/vat/start/schemes/margin.htmhttp://www.hmrc.gov.uk/vat/start/schemes/margin.htm


22/120


As discussed in Mirrlees et al. (2011), however, it seems hard to build astrong case or dierential treatment o various orms o consumption inthe tax system through arguments about their price elasticities or theircomplementarity to work.5 There are substantial inormational requirementsto know how price responsiveness varies across consumption goods. In the

case o motoring, it is likely that some trips will be very price elastic, and othersvery price inelastic; some will be leisure trips, whilst others are associatedwith work. It is hard to see how these dierences could be easily built into anecient revenue-raising design.

One aspect o the revenues rom motoring taxes which has received someattention is the idea o the double dividend (Pearce, 1991; Bovenberg &Goulder, 2002). Taxes on externality-generating activities like motoring (seesection 2.3) correct a distortion and thus lead to society being better o thisis the rst dividend. They also generate revenue which can be used to reduceother taxes, such as income taxes, which distort decisions over whetherand how much to work the second dividend. Perhaps not surprisingly, thisis a seductive proposition or policymakers, and is also implicit in the callsor a green tax shit (see, or example, the work done by the Green FiscalCommission in the UK, which advocated that 20% o revenues come romenvironmental taxes by 2020).6 However, there is considerable debate aboutwhether the double dividend idea really holds up (see Fullerton et al., 2010, ora summary). Instituting higher taxes on environmentally damaging activities,including motoring, raises costs and prices within the economy, which, in turn,reduces real wages and reduces the incentive to work (Bovenberg & de Mooij,1994). Thereore it is not clear-cut that, even i the revenues rom motoring taxesare used to reduce taxes elsewhere, the incentives to work are made stronger.

Distributional eects

Governments might be concerned about policies which have a relativelygreater impact on the poor. Distributional concerns mean that considerationwill be given to weighting overall social welare more heavily towards poorer

people. Thus there may be a case or taxing motoring more heavily than otherconsumption i richer households make greater use o private motoring thanpoorer households. We provide some evidence on the distributional eects omotoring taxes in section 3.5. The general picture is that in the UK, the burdeno motoring taxes alls slightly more heavily on households in the middle andupper parts o the income distribution, and slightly less heavily on those at thevery bottom and very top. This is intuitive: poorer households are less likelyto own cars, but as car ownership quickly rises and fattens o with income,those at the very top are hit less hard by taxes on uel or vehicle ownershipthan those in the middle o the distribution.

5 See, in particular, Taxing Goods and Services, Chapter 6 o the editorial volume o Mirrlees et al.(2011), www.is.org.uk/mirrleesreview/design/ch6.pd.6 www.greenscalcommission.org.uk

2.2

7
http://www.ifs.org.uk/mirrleesreview/design/ch6.pdfhttp://www.greenfiscalcommission.org.uk/http://www.greenfiscalcommission.org.uk/http://www.ifs.org.uk/mirrleesreview/design/ch6.pdf


23/120


In general, though, we should be cautious o using distributional arguments ordierential consumption taxes. Motoring taxes are just one component o scalpolicy. To the extent that we are concerned with distributional eects, theyshould ocus on the tax and benet systemas a whole rather than consideringeach part o it individually.

External costs o road use

The most compelling economic argument that justies higher taxes onmotoring than on other consumption is the principle o externalities. Put simply,private decisions about whether and how much to drive impose costs on otherpeople which are not taken into account in those decisions. This leads to roaduse that exceeds the socially optimal level. Motoring taxes can internalisethese external costs and help to reduce road use to the ecient outcome.Figure 2.1 illustrates the argument graphically.

Figure 2.1: The external costs o motoring

Marginal social cost

Marginal private

Demand

Road use

Costs

1 0

Source: Authors own

Imagine a motorist deciding how much to use the roads in their local area. Thelower the cost, the greater the demand, as illustrated by the downward-slopingdemand curve. Greater road use leads to greater private costs o motoring more time spent in the vehicle, the costs o uel and so on. The private costs

o each additional kilometre driven are also likely to rise: or example, as roaduse increases, overall speed will all as congestion rises. Thus the marginalprivate cost (MPC) curve slopes upwards. However, the costs to society o the

2.3

8


24/120


road use include not only these private uel and time costs, but also the costsimposed on others. These externalities include the costs o carbon and othergreenhouse gas emissions which aect global climate change, the noise costsimposed on pedestrians and local residents, the risk o accidents, the damagecaused to the road, and the act that the presence o the motorist increases

congestion which imposes additional time costs on other drivers. The marginalsocial cost (MSC) curve thereore lies above the private costs, with thedierence between the two representing the marginal externality rom eachadditional kilometre driven. At low levels o road use, costs such as congestionare relatively small. At higher levels they increase rapidly, so we might expectthe marginal externality to increase with total road use.

Since people make private motoring choices based on the private costs whichthey incur, the total level o road use without intervention would be given by q

0,

above the socially optimal level q1

. By imposing a tax t on road use equal tothe size o the marginal external cost at q

1, the private cost curve shits up (the

externality being internalised into private decisions) to give a new equilibrium atthe social optimum. This would generate social benets equal to the shaded area.

O course this is a highly simplied and stylised analysis. It assumes a singletype o road and road user. But it leads to a key point: the tax should refectthe size o themarginal external costat the optimal level o road use. The aimisnot to generate total revenue which equals the total external costs: judgingwhether motoring is overtaxed by comparing total motoring revenues toestimates o the total external costs attributable to motoring is inappropriate.I marginal external costs are increasing in road use which seems plausible then total revenues will exceed the total external cost as an optimal outcome.

It is worth noting that in the presence o externalities, the case or not taxingintermediate inputs into production (a amous result rom Diamond & Mirrlees,1971) is much weaker. For example, i one were to ignore the externalities, uelused by commercial vehicles or business purposes ought not to be taxed, inorder that production be as ecient as possible. However, the environmentaland other external costs associated with motoring should be taken into

account in production decisions.

9


25/120


There are many dierent externalities associated with motoring which mightrequire multiple tax or other policy instruments. A good summary o theexternalities rom motoring is in Parry et al. (2007). They include:

Local air pollution: exhaust pipe emissions rom burning petrol and diesel,

which aect local air quality and can cause respiratory problems. Theseinclude carbon monoxide (CO), nitrogen oxides (NO

x), volatile organic

compounds (VOCs), sulphur dioxide and particulate emissions. Global pollutants: emissions o carbon dioxide (CO

2) rom burning petrol

and diesel which contribute to climate change. Congestion: as trac volumes increase in a given road space, the average

speed o all vehicles slows down, and time costs are thus imposed onother road users.7

Accidents: the costs o death and injury rom trac accidents, damage toproperty, lost productivity and reduced working time and so on.

Noise: the costs imposed on local residents arising rom the engine noise,braking and so on associated with increased trac.

Road damage: the costs imposed by one road users contribution to wearand tear on roads or other road users.

Ideally, policymakers would be able to estimate cost and demand schedulessuch as those depicted in Figure 2.1 or each dierent externality, and imposean ecient tax directly on the externality itsel. However, there are severalcomplicating actors. It may not be possible to tax the externality directly orexample, we cannot measure the exhaust pipe emissions and noise that eachmotorist generates on each journey and then send them a tax bill based onthis output. The issue, then, is whether or not we can nd some other baseor the tax which is closely related to the externality and which can be easiblymeasured and thus taxed. For example, noise costs might be related to thesize o the engine, local air pollution to the type o uel used, global pollutionto the amount o uel burnt, congestion to where and when people drive, andso on. Some proxies or the dierent externalities may be more amenable totaxation than others. In particular, uel use and vehicle purchase are existingmarket transactions, and so taxes based on uel and the vehicle are relatively

straightorward. However, monitoring where and when people drive is notcurrently routine, making geographic and time-specic taxes harder toimplement. Varying motoring taxes by location might be extremely important inseeking to adequately capture the externality or example, the marginal costso local pollution, noise and congestion are likely to be much larger in denselypopulated urban areas than rural areas. However, there are obvious practicalproblems with this: or example, charging higher taxes on uel in cities would

7 To some extent, congestion externalities can be seen as internal to the group o road users, whereasexternalities like climate change are more obviously external to society as a whole. The principle,though, is the same: assuming that an individual road user does not take into account their own impact

on congestion, they impose costs on other people. Thinking about congestion on its own, the decision todrive or an individual depends on the average cost (monetary and time) o the journey, which is less thanthe marginal cost o the journey (the relevant cost in determining the socially optimal level o road use)once the eect on other road users is accounted or.

10


26/120


lead some people living on the outskirts to drive out into rural areas to reuel.Thus or practical purposes, policymakers are oten constrained in the taxesthey can impose.

Even given an appropriate basis or taxation, another issue is the size o tax

imposed. This requires an estimate o the marginal external costs involved.Table 2.1 displays results rom three studies that have attempted to calculatethe size o the main externalities associated with motoring.

Table 2.1: Estimates o the marginal costs o road transport (p/km)

Type o cost Sansom et al., 2001(1998 prices)

DT, 2010(2002 prices)

Bayliss, 2011(2009 prices)

Low High

Congestion 9.71 11.16 13.1 4.60a

Inrastructure 0.42 0.54 0.1 0.57

Accident 0.82 1.40 1.5 0.88

Local air quality 0.34 1.70 0.4 0.57

Noise 0.02 0.78 0.1 0.50

Greenhouse gases 0.15 0.62 0.3 0.64

Total 11.46 16.20 15.5 7.76

a Note that the Bayliss congestion gure is an average rather than a marginal cost (seediscussion at the end o section 2.3).

The gures most commonly reerenced are those o Sansom et al. (2001),which estimate the size o marginal externalities or 1998. They suggestthat congestion is by ar the largest component o the external cost o eachadditional vehicle kilometre driven, accounting or around 9.7p to 11.2p rom atotal marginal externality o 11.5p to 16.2p per kilometre.

These estimates are now quite outdated and it is likely that some o the costshave changed. For example, externalities associated with greenhouse gaseswere estimated to be between 0.2p and 0.6p/km. The cost associated withemitting a tonne o carbon has risen since the late 1990s, refecting newevidence on the likely costs o climate change and new methods o assigningvalues to carbon emissions that are consistent with the various targetsthe government aces to reduce emissions in the uture. As a result, theseestimates are now too low. Burning a litre o road uel emits roughly 2.5 kg oCO

2. At the current (non-traded) value o 56 per tonne (Department o Energy

and Climate Change (DECC) estimates see Table 1 o DECC, 2011)8

or 2012,this is a cost per litre o 14p. At a (probably conservative) uel economy o

8 www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pd, Table 1

11
http://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdfhttp://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdfhttp://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdfhttp://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdf


27/120


10 km per litre (28 mpg) this suggests a marginal climate change externalityo 1.4p per kilometre. At a (probably too high) economy o 20 km it is 0.7pper kilometre. Both o these costs are larger than the high estimate ound inSansom et al., or indeed in either o the other two studies, or the marginalexternal cost associated with greenhouse gases.

It is also likely that the other externalities have changed over time. Tightervehicle emissions standards are likely to have reduced marginal costsassociated with air quality. Economic growth has probably increased themarginal congestion externality (which is largely determined by estimates o thevalue o time, which in turn depends on wage levels).

Table 2.2 shows emissions o CO2

and other pollutants rom road transport in1970, 1990 and 2009, and the proportion o the total emissions o each gasthat comes rom road transport. As can be seen, emissions rose substantiallybetween 1970 and 1990. Since then, non-CO

2emissions rom road transport

have allen markedly. In most cases, this has been accompanied by a all in theshare o these emissions rom transport, suggesting that measures such astightening emissions standards have been eective at reducing these non-CO

2

emissions more rapidly in road transport than in other sectors. CO2

emissionsrom road transport, however, rose slightly between 1990 and 2009, and nowaccount or almost 24% o total emissions, compared to 19% in 1990 and 9%in 1970.

Table 2.2: CO2

and other emissions rom road transport, 19702009

Pollutant 1970 1990 2009 Changes

tonnes % ototal

tonnes % ototal

tonnes % ototal

19701990

19902009

Carbondioxide

60.3m 8.8 109.4m 18.6 112.5m 23.8 82% 3%

Nitrogenoxides

573,300 22.3 1.068m 39.8 361,190 33.3 86% -66%

Sulphurdioxide

46,000 0.7 64,000 1.7 1,000 0.3 39% -98%

Particulates 24,000 4.8 43,000 15.4 27,000 22.7 79% -37%

Carbonmonoxide

3.327m 33.9 6.275m 69.7 1.076m 47.3 89% -23%

VOCs 570,000 30.3 997,000 36.8 87,000 10.5 75% -91%

Source: Authors calculations rom Department or Environment, Food and Rural Aairs (Dera)air pollution statistics and DECC emissions data

Another set o estimates o the marginal externalities was conducted as part oa study on road decongestion benets (DT, 2010). Estimates or the marginal

12


28/120


external costs o congestion or dierent road types in 2005 are produced byusing the National Transport Model (NTM), calculating the delay caused by amarginal vehicle, and then summing the value o lost time across all users onthe road to give the size o the marginal congestion externality. Other marginalexternalities are based on updates o the Sansom et al. estimates, and climate

change costs based on Dera (2002) guidelines. Consistent with the discussionabove, the estimated marginal congestion costs are higher than the Sansomet al. estimates, whilst the marginal air quality costs are at the lower end o theSansom gures.

Another attempt to estimate these costs is made by Bayliss (2011). He updatesthe Sansom gures to take into account changes in prices, trac volumes andtechnology over the period. However, the congestion externality is based ondividing estimates o the total congestion costs by the total distance driven.Using a rough estimate o a 20 billion total cost, this gives an estimatedexternality o just 4.6p/km, ar smaller than the Sansom et al. or DT estimates.However, it should be noted that this is thereore an estimate o theaveragecongestion externality and not the marginal externality. As noted above, thereis no reason to expect the two to be the same; indeed, we would expect themarginal externality to be larger than the average.

The role o non-tax instruments

This report ocuses on tax policy as regards motoring. However, it is clear thatnon-tax policies play a signicant role most notable amongst these being theregulations on emissions rom vehicles and vehicle uel. In the UK, standardsimposed by the European Union, commonly known as Euro standards, are therelevant instruments. They rst came into orce in 1992. The current Euro 5standards regulate diesel cars to emit no more than 0.5 g o CO per kilometreand 0.18 g o NO

x.9 As noted above, it is not always easy to use tax policy

when the external costs are hard to measure and tax directly; these dicultiesmight avour more direct regulation o this kind. The gradual tightening o Eurostandards over time has almost certainly been a huge contributing actor to the

overall reduction in non-CO2 emissions rom road transport noted in Table 2.1.In a world in which these sorts o emissions were directly taxable, however,these regulations would be inecient, since they apply across the board to allnew cars. Taxing the emissions would raise the price o running more-pollutingvehicles, and encourage those who were more able and willing to do so todrive less or buy less-polluting cars, potentially reducing the overall cost oachieving a given level o emissions reduction. Taxes also provide dynamicincentives or motorists and manuacturers, since each unit o the emissionsaces a given price, whereas regulation does not provide direct incentives to gobeyond the stipulated standards.

9 http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2007:171:0001:0016:EN:PDF

2.4

13
http://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdfhttp://www.decc.gov.uk/assets/decc/11/cutting-emissions/carbon-valuation/3136-guide-carbon-valuation-methodology.pdf


29/120


A wider point, though, is that tax policy should take account o the otherinstruments, including regulation, that are in place. In the externalities case, orexample, the presence o regulations imposes implicit prices on the pollutingactivities which may at least partly mitigate the size o the tax needed to ullyinternalise the external costs. Christiansen and Smith (2009) argue, using

a theoretical model, that in the presence o regulations combined with tax-based policies to correct externalities, tightening the regulation will haveambiguous eects on the optimal tax rate, depending on how the regulationaects the responsiveness o demand to price. I regulations are tightened,and demand responsiveness does not rise as a result, they demonstratethat the optimal tax rate alls i marginal external costs are increasing in totalconsumption. However, Parry (2009) looks at optimal vehicle uel taxes in theUSA and concludes that more stringent regulation o vehicle emissions wouldlead to ahigheroptimal uel tax. This is or two particular reasons. Firstly,when emissions standards are higher, it is harder or motorists to respondto higher uel taxes by improving the eciency o their vehicle rather thansimply driving less. When a large amount o the response to uel taxes comesthrough eciency rather than road use, the marginal impact o tax increaseson those externalities which are more closely related to distance driven thanuel consumed (such as congestion, accidents, noise and local pollution) isreduced, which reduces the optimal tax rate. Higher emissions standardsreduce this eect. Secondly, as the overall eciency o the vehicle feet rises,the distance travelled or each litre o uel consumed increases. In a world inwhich distance-related externalities can only be captured through uel taxes(assumed in the Parry analysis) rather than in a more direct ashion, this alsoraises the optimal uel tax rate.

Other relevant issues

There are a number o other economic issues that are important whenconsidering motoring taxation. We briefy discuss them here, and in Section 5consider the current system o taxes in the light o some o them.

Cross-border eects: One issue which impacts on indirect taxation morewidely is the incentive or people to buy the taxed goods more cheaply romother countries with lower taxes. For motoring, the most signicant concernis about vehicle uel. UK uel taxes and prices (certainly or diesel) are higherthan those in the Republic o Ireland, meaning that people living on the borderin Northern Ireland have an incentive to ll up in the Republic and avoidpaying UK duty. In August 2011, or example, DECC estimates suggest thata litre o diesel in Ireland cost 1.23, as compared to 1.40 in the UK. Centralestimates rom Her Majestys Revenue and Customs (HMRC, 2011a) suggestthat in 2009/10, around 70 million was lost in cross-border sales o diesel

in Ireland by UK vehicles, amounting to about 12% o the diesel market inNorthern Ireland. This loss, however, was notably smaller than in previousyears; or example, in 2003/04 the cost was estimated at 210 million, or 51%

2.5

14


30/120


o the market. This coincides with a recent narrowing o the gap between UKand Irish diesel prices.

A related concern is the issue o cabotage. Hauliers rom other countries,paying lower diesel taxes, can deliver goods on the UK mainland without

paying UK duty rates. In the 2002 Budget, the then Chancellor, Gordon Brown,announced the intention to introduce a lorry road-user charge, partly to ensure that lorry operators contribute airly and eciently towards the coststhey impose in the UK irrespective o their nationality (HM Treasury, 2002). Theintention was or such a charge to be introduced in 2005/06, but the schemewas abandoned by then Transport Secretary, Alistair Darling, in 2005. At thetime, this was justied in terms o seeking to introduce a lorry charge as part oa national system o road pricing or all road users though no such proposalswere ever subsequently introduced.

Plans or a lorry charge also eatured in the Coalition agreement published bythe Conservatives and Liberal Democrats ollowing the 2010 election, with apledge to work towards the introduction o a new system o HGV [heavygoods vehicle] road user charging to ensure a airer arrangement or UKhauliers. In January 2012, a consultation was launched on a proposed systemo lorry charging to achieve this goal, with a planned implementation date in2015 (DT, 2012). The planned charge would cover all HGVs weighing at least12 tonnes. The heaviest lorries would pay up to 10 per day to access theUK road network. UK lorries would pay on an annual or biannual basis (witha maximum annual charge o 1,000), whilst oreign hauliers could choose topay daily, weekly, monthly or annually. The intention is to oset the chargesor UK hauliers against VED liabilities (subject to limits set by EU Directives onminimum rates o annual taxation or lorries),10 leaving the vast majority payingno more in total. On this basis, the charge is estimated to raise net revenues oaround 20 million per year rom 2015/16. It should be noted that the proposedcharge would not vary according to how ar lorries drove or at what times, butwould be a simple access charge (rather like the London Congestion Charge,see Section 3).

Hypothecation: There are requently calls or revenues rom particular taxesto be linked (hypothecated) to specic parts o public spending. A recentreport rom the Environmental Audit Committee (2011) argued that increasesin uel duty receipts should be earmarked or spending on public transport,or example. In London, revenues rom the congestion charge are spent onLondon public transport, and the US Highway Trust Fund uses receipts romederal uel taxes there to pay or road building and some public transport.

There is in general no good economic rationale or hypothecation, as discussedin Advani et al. (2011). I all the spending on a particular public service such

as, or example, road building and maintenance were linked directly to

10 For urther details, see: www.lorry-ee-europe.org/data/2006-EUvignette-en.pd

15
http://www.lorry-fee-europe.org/data/2006-EUvignette-en.pdfhttp://www.lorry-fee-europe.org/data/2006-EUvignette-en.pdf


31/120


receipts rom a particular tax, such as VED or uel duty, then there is no reasonto expect an optimal pattern o spending to emerge. The marginal pound otax revenue should be spent in the most eective area, which may mean rathermore or less should be spent on service X than is raised rom tax Y in a givenyear: indeed, the US case is a good example, where the Trust Fund is topped

up year ater year with additional money. I the hypothecation is less explicit(perhaps simply promising that additional revenue rom a particular tax is tobe spent on a particular service), then so long as more than this incrementalspending was planned to be spent on that service anyway, it would beimpossible to veriy that the money had indeed been hypothecated in this way.

However, as noted by Smith (1992), it may be that hypothecation has a role inbuilding public support or particular new taxes or tax increases, without whichintroducing the measure may not be possible. I the tax is intended to correctsome market ailure, then the social costs generated by inecient publicspending ollowing hypothecation may be smaller than the social costs o nothaving introduced the tax in the rst place.

Other good practice issues or tax policy: Subject to the wider objectives oparticular taxes, in general taxes should be simple, provide taxpayers with areasonable amount o certainty over what their current and uture tax liabilitiesmight be, and be cheap to collect. In the case o motoring taxes, particularlyuel duty, we show below evidence that there is considerable uncertaintyover the path o policy, with regular changes to planned tax movements romBudget to Budget.

16


32/120

3. The Structure of UK

Motoring Taxation


This section describes the way in which

motoring is currently taxed, and how thesystem has evolved. We ocus on the principal

national motoring-related taxes: uel duties,

VED, and the taxation o company cars and

uel. We also describe briefy the London

Congestion Charge. Finally we present a

snapshot o the current distributional impact o

uel duties and VED.

17


33/120

The Structure of UK Motoring Taxation

Fuel duties

3.1.1 Current rates of duty

The overwhelming majority o road uel purchased in the UK is subject to anexcise tax, currently set at 57.95p/litre. This is the rate applied to unleadedpetrol (including ultra-low sulphur and sulphur-ree petrol), diesel and biouels(bioethanol and biodiesel).11 So-called red diesel attracts a lower rate o11.72p/litre, but is in principle sold only or use in o-road vehicles such astractors and other agricultural machinery.12 VAT, currently 20%, is applied to thecost o uelincluding duty. This means that a 1p rise in duty rates leads to a total

tax increase o 1.2p/litre. Estimates rom HMRC (2011c) suggest that a 1% risein the main rates o uel duty on petrol and diesel raises around 260 million.13Thus a 1p rise in duty rates (currently some 1.7%) would raise about 450million, equivalent to an increase in the basic rate o income tax o about 0.1p.

The decision over rates o uel duty is taken by the Chancellor in the annualBudget.14 The usual position is that duties are increased in line with priceinfation to maintain their real value. For excise duties, the measure used isthe orecast Retail Price Index (RPI) infation rate in the third quarter ollowingthe Budget. However, there is discretion to set uel duties at any rate, and

much use has been made o this discretion in recent years. We discuss thesechanges below and in Appendix A, which details the main changes in motoringtaxes at each scal event since 1989.

11 It should be noted that uels used or local bus services are eectively taxed at lower rates becauseo the Bus Service Operators Grant. As o March 2011, this provides a rebate to local bus providers o43.21p/litre o unleaded petrol or diesel.12 In 2010/11, red diesel accounted or about 5.3 billion litres (10.2%) o almost 51.7 billion litres oroad uels purchased (HMRC, 2011b).13 This is the additional revenue rom uel duty alone. It is assumed that there are no additional VAT

receipts, since higher VAT payments on the duty are oset by lower VAT payments on other goods andservices as consumers substitute their expenditure.14 Changes to duty rates are also sometimes announced as part o other scal events such as Pre-Budget Reports or Autumn Statements.

3.1

18


34/120


3.1.2 Historical duty rates

The most signicant policy announcement or uel duty came in the March1993 Budget, when Chancellor Norman Lamont implemented a uel dutyescalator. This policy introduced a deault above-infation minimum increase in

duty rates each year. The policy was designed such that the increase appliedto the average real increase in duties, allowing or some discretion to applylower increases to some types o uel (e.g. to encourage switching to lower-emissions uels).15 It is notable that the announcement o the escalator cameater two Budgets, in March 1990 and March 1991, when there were large realincreases in uel duty rates.

Initially the escalator was set at 3%, but was increased almost immediatelyin the November 1993 Budget (Ken Clarkes rst) to 5%, and raised again to6% in the July 1997 Budget (Gordon Browns rst). The escalator policy wasabandoned by Gordon Brown in the 1999 Pre-Budget Report (PBR) in the aceo increases in pre-tax uel prices. It is worth noting that this happenedbeorethe uel price protests which took place in Autumn 2000.

The result o the escalator was a huge increase in uel duty rates. In nominalterms, duty rates or the most commonly purchased petrol more than doubled,rom 23.43p/litre in March 1993 to 47.21p/litre in March 1999. In real terms (inSeptember 2011 prices), the duty rate rose rom 40.00p to 68.44p/litre over thesame period, an increase o 71%.

Following the abandonment o the escalator, uel duties were not increased inreal terms at all until 2009. Indeed, duties were requently not even increasedin line with prices, with planned upratings postponed and then abandonedaltogether in the ace o continued high pump prices. For example, the 2004and 2005 March Budgets both delayed infation adjustment to September eachyear. In both cases, these adjustments were not implemented and were thencancelled altogether in the subsequent PBRs. The March 2006 Budget againannounced that a planned infation adjustment to duties would be delayeduntil September, which was then urther delayed and only implemented in

December 2006. Together this meant that there were no cash increases in themain uel duty rates between October 2003 and December 2006.

In December 2008, the main duty rates were increased by 2p/litre. This wasdescribed in the November 2008 PBR as a move to oset the temporarycut in VAT rom 17.5% to 15% introduced by Alistair Darling as a stimulusmeasure though the rate was not subsequently reduced in January2010 when the temporary VAT cut expired. However, this 2p increase hadoriginally been announced in the March 2007 Budget to occur in April 2008,was then delayed until October 2008 in the March 2008 Budget, and was

nally implemented in December 2008 ollowing the November PBR. Thus it15 Intriguingly, the documents or the March 1993 Budget noted that developments in charging or roaduse will be taken into account in determining the appropriate level o duty.

19


35/120


should not be viewed as a temporary oset to the VAT cut, rather a delayedimplementation o a pre-announced policy.

The rst real-terms increase in the main uel duty rates ollowing theabandonment o the escalator in 1999 nally took place almost a decade

later in September 2009, when duties rose by 2p/litre (as announced in theMarch 2009 Budget), having already been infation-adjusted in April 2009.Real increases were, though, pre-announced earlier: the March 2008 Budgetpencilled in a 0.5p real increase in the main duty rates or April 2010. Perhapsunsurprisingly, plans or two years later were reormed beore they evenoccurred. In the March 2009 Budget, the planned April 2010 increase rose to1p above infation, with the same real rise announced each year until 2013.Even this announcement could not survive another years economic news:in the March 2010 Budget (the last o the previous Labour government), thepenny escalator planned or April 2010 was instead split into three smaller,staged increases in April, October and January, though the escalator policyitsel was extended by another year, to April 2014.

In his rst Budget in June 2010, George Osborne kept this policy in place.However, less than a year later in his second Budget o March 2011, urtherchanges were announced. Citing increases in the oil price, the penny escalatorwas abandoned and the main duty rates were cut by 1p/litre. The plannedinfation adjustment o duty rates was postponed rom April 2011 until January2012, with the (now) infation-only adjustment planned or April 2012 alsodelayed, until August 2012. However, these actions were conditional, underwhat was billed the air uel stabiliser. The cost o these announcements waseectively paid or by an increase rom 20% to 32% in the supplementary chargeon corporation tax or North Sea oil and gas companies. However, i oil pricesall below $75 per barrel in a sustained way, the policies will be reversed, withthe escalator re-introduced and the supplementary charge reduced to 20%. Wediscuss this stabiliser, and compare it to the air uel stabiliser proposed by theConservatives prior to the May 2010 election, in Section 6.

20


36/120


In the Autumn Statement o November 2011, yet more changes wereannounced to duty rates. The infation adjustment o 3p/litre, delayed rom April2011 to January 2012, was urther delayed until August 2012. The infationadjustment in 2012, delayed rom April until August, was cancelled altogether.Thus the pattern established under Labour, with pre-announced rises delayed

then later cancelled, re-emerged. Indeed, assuming it does actually go ahead,the delay in the April 2011 uprating until August 2012 a 16-month gap would be the longest postponement o a planned duty rise since the originalescalator was abolished. It does seem somewhat absurd that it was the 2012uprating which was cancelled (rather than being implemented a mere 4 monthsbehind schedule) instead o the 2011 uprating though the higher infation ratein 2011 meant that cancelling the 2012 adjustment was slightly less costly.

Figure 3.1 summarises these historical trends and shows the real rate o uelduty applied to the most commonly purchased petrol and diesel each monthsince January 1990. It projects duty rates ahead to March 2015, the end othe current Parliament, under the assumption that infation is as orecast inthe most recent OBR estimates (November 2011), and that duty rates areincreased as planned in August 2012, April 2013 and April 2014.

Figure 3.1: Actual and orecast real-terms duty rates on main uels

(September 2011 prices), January 1990March 2015

Dutyrate(p

ence/litre),Sept2011prices

3%escalator

5%escalator

6%escalator

1pescalator

Escalator

abandoned

Escalator

abandoned

Unleaded

Diesel75

80

70

65

60

55

50

45

40

35

30

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Source: Authors calculations based on HMRC UKtradeino data and OBR infation orecasts.

Figure 3.2 illustrates the impact o the escalator and its subsequentabandonment on the composition o pump prices, ocusing on unleadedpetrol. It shows the breakdown o the pump price into duty, VAT and pre-tax

21


37/120


components. During the escalator period, duty rose rom around 50% o thepump price to 70%, whilst total taxes (including VAT) rose rom around twothirds o the pump price to, at their peak in March 1999, almost 86%. Followingthe end o the escalator and the rise in global oil prices in the rst decade othis century, tax contributed a alling proportion o pump prices. At its lowest,

in July 2008, tax made up 57% o the unleaded pump price (which at that timewas 131.4p/litre in September 2011 prices, a level not seen again until January2011), the lowest proportion since November 1990 (when the real price was81.5p/litre).

Figure 3.2: Composition o pump prices or unleaded petrol, January 1990

to September 2011

Propo

rtionofpumpprice

100%

Pre-tax

VAT

Duty

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Source: Authors calculations based on UKtradeino data

3.1.3 Duty differentiation

The UK is currently the only country within the EU which does not dierentiatethe rate o uel duty on petrol and diesel. The UK equalised petrol and dieselduty rates in November 1994; beore then diesel was taxed slightly less heavily.Indeed, in 1998, diesel began to be taxed more heavily than petrol, though thiswas partly about encouraging diesel users to purchase ultra-low sulphur diesel(ULSD), which continued to be taxed at the same or a lower rate than unleadedpetrol. Since 2001, when virtually all petrol and diesel sold has been ultra-low

sulphur or sulphur-ree, the tax treatment o petrol and diesel has been thesame. Despite this, recent years have seen a marked growth in the penetrationo diesel into the overall market or vehicle uel. According to estimates rom

22


38/120


HMRC o motor uels released or consumption, diesel (in all orms) made uparound 20% o road uels sold by volume in 1990/91, just over one third in2000/01, and 50% by 2010/11.

Duty dierentiation is not merely about petrol and diesel taxes, but also about

encouraging take-up o more environmentally riendly types o uel. The dutysystem has been used to encourage the take-up o unleaded petrol overleaded petrol, o ultra-low sulphur and sulphur-ree versions o petrol anddiesel, and (until recently) o biouels. Figure 3.3 shows how the composition opetrol and diesel sold has changed over time. There has been rapid penetrationo new uel types as they are introduced. Unleaded petrol steadily replacedleaded petrol in the early 1990s, with the petrol market then switching almostentirely to ultra-low sulphur petrol (ULSP) in the space o around two years andsulphur-ree petrol rom the late 2000s. Similar trends emerge or diesel uels.16In both markets, biouels represent a small part o overall sales, making uparound 3% o petrol sales in 2010/11 and 3.7% o diesel sales.

16 These gures are based on data rom UKtradeino which breaks down quantities o uel soldaccording to taxation category. Ater October 2007, dierential treatment o sulphur-ree and ultra-lowsulphur uels is removed, which means we can no longer identiy a breakdown between unleaded,ultra-low sulphur and sulphur-ree uels. However, according to the UK Petroleum Industry Association,certainly by 2009 almost all uels were sulphur ree (www.ukpia.com/industry_issues/uels/sulphur-ree-petrol-diesel-and-non-road-uels.aspx), so we make the assumption that non-biouel petrol and dieselwas all sulphur ree rom 2008/09 onwards.

23
http://www.ukpia.com/industry_issues/fuels/sulphur-free-petrol-diesel-and-non-road-fuels.aspxhttp://www.ukpia.com/industry_issues/fuels/sulphur-free-petrol-diesel-and-non-road-fuels.aspxhttp://www.ukpia.com/industry_issues/fuels/sulphur-free-petrol-diesel-and-non-road-fuels.aspxhttp://www.ukpia.com/industry_issues/fuels/sulphur-free-petrol-diesel-and-non-road-fuels.aspx


39/120


Figure 3.3: Composition o petrol and diesel uels sold, 1990/91 to 2010/11

Petrol

Litres(million)

Bioethanol

Leaded

Superunleaded / LRP

Unleaded

Sulphur free

Ultra-low sulphur

35000

30000

25000

20000

15000

10000

5000

0

1990/91

1992/93

1994/95

1996/97

1998/99

2000/01

2002/03

2004/05

2006/07

2008/09

2010/11

Diesel

Biodiesel

Diesel

Sulphur free

Ultra-low sulphur

30000

25000

20000

15000

10000

5000

0

1990/91

1992/93

1994/95

1996/97

1998/99

2000/01

2002/03

2004/05

2006/07

2008/09

2010/11

Litres(million)

Source: Authors calculations based on UKtradeino data

All the major shits in the use o one uel to another occur shortly ater theintroduction o a dierential rate between the new and the pre-existing most

popular uels, suggesting that, at least in part, tax policy has infuencedwhich types o uel are commonly used. For example, the March 1997 Budgetintroduced a 1p/litre dierential between the rates o ULSD and conventional

24


40/120


diesel to encourage the shit towards the cleaner ULSD. This was ollowedby an extension o this dierential to 2p/litre in the 1998 Budget, and by theollowing year ULSD had almost entirely replaced the use o conventionaldiesel. This is very similar to the situation with petrol, where a cheaper rate orULSP was introduced in the 2000 Budget, and extended in the 2001 Budget.

By 2002, ULSP had completely supplanted the use o unleaded petrol. Dutydierentiation alone, o course, does not explain the shit in uels purchased.Regulation, including more stringent EU emissions standards or vehiclesand uel which limited the amount o sulphur that could be emitted, hasundoubtedly had a substantial eect. Nevertheless, there may be a role orduty dierentiation in speeding up the process o switchover between uels.

Dierential rates or biouels were rst introduced in the April 2002 Budget. Atrst, a 20p/litre reduction in uel duty applied only to biodiesel, with this beingextended to bioethanol in January 2005. These dierential rates were guaranteedto be maintained or several years by each Budget, and were extended until2009/10 in the March 2007 Budget. However, in the March 2008 Budget, it wasannounced that the dierential would be abandoned in October 2010. This policychange resulted rom the uncertainty over the overall environmental impact ousing biouels. Although the emissions levels when using biouels are lower,there are concerns over the emissions levels incurred during theirproduction,raising the possibility that over the entire lie cycle, emissions rom biouels couldactually exceed those o their low-sulphur counterparts.17

Duty dierentiation or biouels was replaced by the introduction o theRenewable Transport Fuel Obligations (RTFO) in 2008. Initially proposedin the 2005 PBR, the RTFO requires suppliers o road transport uels toinclude a certain percentage o biouels (the obligation is 3.5% in 2010/11,rising to 5% in 2013/14) or to pay a buyout price (30p/litre in 2010/11) orthe amount undersupplied. Suppliers are issued with tradable certicatesor supplying biouels, so can achieve the target by buying surplus permitsrom other suppliers or rom the buyout scheme. Whilst uel suppliers areobligated to report on the type o biouels they supply (which helps determinehow environmentally sustainable they are), there is no penalty or supplying

less-sustainable biouels, so it is not clear that the RTFO provides the rightincentives to supply the most environmentally benecial uels.

The RTFO was in part designed to help the UK achieve its obligations underthe 2003 EU Biouels Directive, which set a target o biouel penetrationo 5.75% by the end o 2010.18 Again, concerns about the sustainability obiouels led to a rethink. The 2009 EU Renewable Energy Directive requires10% o transport uels to come rom renewable sources by 2020, with biouelsneeding to generate greenhouse gas emissions savings o at least 35% to

17 Pickett et al. (2008) note that biouels, used in conjunction with existing uels, have the potential to

signicantly reduce uture emissions. However they stress that there are a signicant number o social,economic and environmental uncertainties associated with biouels, and policy rameworks must ensurethat such issues are addressed.18 http://europa.eu/legislation_summaries/energy/renewable_energy/l21061_en.htm

25
http://europa.eu/legislation_summaries/energy/renewable_energy/l21061_en.htmhttp://europa.eu/legislation_summaries/energy/renewable_energy/l21061_en.htm


41/120


qualiy as renewable.19 It will be interesting to see how the removal o thebiouels duty incentive and the move towards the RTFO aects the movetowards biouels in uture years; we do not yet have much data dating romsince the end o the duty dierential with which to make any clear assessmento trends.

3.1.4 International comparisons

Figure 3.4 shows EU countries ranked by the pump prices o petrol anddiesel, and the proportion o price accounted or by tax. There is relatively littlevariation in pre-tax prices across countries, but much larger variation in pumpprices, driven by dierences in tax policy across EU countries.

19 http://europa.eu/legislation_summaries/energy/renewable_energy/en0009_en.htm

26
http://europa.eu/legislation_summaries/energy/renewable_energy/en0009_en.htmhttp://europa.eu/legislation_summaries/energy/renewable_energy/en0009_en.htm


42/120


Figure 3.4: Pump prices across the EU-27, August 2011

Petrol

160 Tax

Pre-tax140

120

100

80

60

40

20

0

Bulgaria

48%

Poland

51%

43%

51%

48%

49%

49%

50%

49%

50%

56%

52%

54%

48%

56%

58%

60%

55%

59%

57%

57%

59%

57%

56%

56%

61%

60%

Cyprus

Estonia

Romania

Lux

Date post:	05-Apr-2018
Category:	Documents
Upload:	patrick-mcpartlin
View:	217 times
Download:	0 times

Fuel for Thought: The what, why and how of motoring taxation

Documents