The questionable credentials of green taxes

Andrew Dilnot and Laura Blow on the effects of raising fuel duties

Every government needs ways of raising money that we either don't notice, or think are justified. The last government took the art of well- disguised tax increases to new heights in 1993. In that year Norman Lamont and Kenneth Clarke raised taxes by the equivalent of 7p on the basic rate of income tax and nearly got away with it. VAT on fuel was their downfall; it was only a small part of the money, but it attracted the spotlight, and lost them the argument.

What should New Labour do? A popular modern defence of tax increases is the environment. And one reason for the popularity is that this is potentially a very good argument. If the consumption of some goods imposes costs on people other than those consuming them, it is perfectly sensible to seek to charge for these "externalities", so that we all face the full costs of our consumption. Motoring is an example of just such a good. Driving around brings benefits to those driving and being driven, but costs to others - congestion of the roads, damage to the roads, local air pollution, noise, accidents, and emission of the global warming gas, carbon dioxide. If we could find well targeted ways of taxing these "bads", we might well want to do so.

Kenneth Clarke was well aware of this, and introduced a policy of increasing the tax on road fuels (petrol and diesel), ultimately by at least 5 per cent a year more than inflation. With similar arguments in mind he also announced minimum real increases in tobacco of 3 per cent a year.

The Labour Government has gone further still, and moved to minimum real increases in road fuel duties of 6 per cent a year and for tobacco of 5 per cent a year. These are large changes to big taxes. Fuel duties, even excluding VAT, already raise one quarter as much as income tax, and tobacco half of what fuel raises. The increases in these taxes are an important source of the growth in government revenue that will continue throughout the Parliament. Rapid declines in fuel consumption could reduce the revenue gains, but seem unlikely given the relative insensitivity of consumption to price. The fuel increase alone is equivalent to an increase of around pounds 7bn per annum by the end of the Parliament, the tobacco to pounds 2.5bn, a combined effect equal to 5p on the basic rate of income tax. These tax increases should help to make it easier for the Government to stick to their pledge that neither basic nor higher rates of income tax will rise.

Given the scale of these changes, we need to be convinced that there is a strong case for them, and the natural place to start is with the environmental arguments. There can be no doubt that congestion costs are very large, and quite possibly the largest external cost of motoring. Cambridge University economist David Newbery estimates the cost at some pounds 20bn. But taxes on road fuel are ill designed to tackle congestion, which requires an approach which can vary charges by time and place. Fuel duties are no better targeted on road damage, which largely depends on axle load and the type of road surface.

Local air pollution is clearly a cause for concern, and is receiving increasing amounts of attention. But once again, a tax which is simply a function of fuel consumption is a poor instrument to tackle this problem. Emissions per litre of fuel consumed of pollutants such as black smoke, carbon monoxide, nitrogen oxides and volatile organic compounds vary substantially across different vehicles and different types of fuel. And the extent of local air pollution is strongly affected by time, place, weather conditions, and existing concentrations of pollutants. There is one externality created by motoring which is well targeted by a tax on road fuels, which is the emission of carbon dioxide, the main global warming gas. Emissions bear a straightforward relationship to fuel use, there is not at present any effective technology for filtering out the emission of the gas, and the time and place of emission are largely irrelevant.

And yet motoring is responsible for only 20 per cent of UK emissions of carbon dioxide. Increased VAT on domestic fuel has been ruled out, and Brussel's proposal for a broadly based carbon tax has been greeted coldly by both Conservative and Labour parties. It seems somewhat odd to argue for very large increases in tax on the source of one fifth of carbon dioxide emissions while seeking to avoid increases on the remainder.

The strongest argument used against the imposition of VAT on domestic fuel was that it would hit those on low incomes. A compensation package of benefit increases was eventually proposed, but too late to avoid political embarrassment and defeat in the Commons. The distributional impact of increasing fuel duties is very different to that from VAT on fuel, not least since the poorest households are unlikely to have cars, and even those few who do will tend to drive them relatively little.

The chart on the left shows the proportion of total expenditure which goes on road fuels for all households as total spending rises. At low levels of total spending the fuel share is very low, climbing sharply, levelling off, and then falling for those with higher spending and income. And this is the pattern of losses created by raising fuel duties.

If we look, in the right hand chart, only at those households with cars, we see a very different pattern, with the share of road fuels in total spending falling fairly steadily as total spending rises. This is not much of a surprise, but points to a group about whom we might be concerned. Poor households in rural areas may rely far more on cars than their urban counterparts, because of the lack of public transport, and their need to travel greater distances anyway. Recent work at the Institute for Fiscal Studies has shown that poor car users in rural areas are the group hardest hit by increases in fuel duties.

None of this implies either that we should not think of using taxes to tackle environmental problems, or that the inevitable distributional problems should rule out change. Taxes can be an effective instrument in environmental policy, and distributional problems caused by tax changes can be compensated for. But good environmental taxes need to be targeted effectively and clearly on specific problems, and we need to be aware of distributional issues well in advance. The planned increases in road fuel duties will raise a lot of money for the Government, which is, quite rightly, the primary purpose of taxation. Whether they are the best ways of raising money or of tackling environmental problems, is far less clear.

Laura Blow and Ian Crawford: `The Distributional Effects of Taxes on Private Motoring', Institute for Fiscal Studies, 7 Ridgmount Street, London WC1E 7AE, Price pounds 10