A tax on the rich that delivers few votes and less revenue

Gordon Brown's 50 per cent top tax rate would probably have failed as high earners emigrated or used loopholes

So it looks as though Tony Blair has won his argument with Gordon Brown over the top rate of income tax. We will have to wait a few weeks more for confirmation, but it seems that the plan supported by the shadow Chancellor, to increase the top rate of tax from 40 per cent to 50 per cent for incomes of more than pounds 100,000, will not now go ahead. Mr Blair apparently felt that it gave the wrong signal about Labour's tax intentions.

But there is an even more powerfulreasonthan Mr Blair's political judgement for opposing the idea. It is the possibility that, far from increasing tax revenue, it might actually cut it. Most people would assume that if you put tax rates up, people pay more tax. For most taxes and most people that is probably true. But for the particular group of people who might be hit, this may not be right. To see why, here are some very rough calculations which show on various assumptions the extent to which any additional revenue would be offset by other declines.

The starting point is the Inland Revenue calculation that implementing this top rate of tax on all earnings over pounds 100,000 would raise an additional pounds 1.1bn in revenue. That may sound a lot, though it is actually only a third of 1 per cent of public spending. It seems astonishing, but that calculation is based on the assumption that there would be no change at all in people's behaviour as a result of the tax increase. That is unrealistic, for there are bound to be some changes, and these changes are going to cut revenue. So that pounds 1.1bn is an absolute maximum. The real total will be lower.

Start, then, with the pounds 1.1bn. The people who might pay can avoid it in one of two (legal) ways. They can either leave the UK tax net altogether, or they can change their tax arrangements to cut nominal income. The first question is: how many people might leave?

There are about 120,000 people who have a taxable income of more than pounds 100,000. Not many of those either have the opportunity or would want to go to the upheaval of moving abroad just because of income tax. The main group of people who emigrate for tax reasons are those selling businesses and retiring, and they are avoiding inheritance and capital gains tax rather than income tax. But some people might take the rise in income tax as a signal that other taxes might also be increased and act accordingly. Others might be trying to reach a decision and this would be the thing which pushed them over.

So let us assume that, within a couple of years, 2 per cent of those high earners, 2,400 people, would leave. That may be too high or too low, but it is at least a figure. For those people, the Exchequer loses all tax: not just the income tax that they would have paid, but their capital gains tax, the VAT on their purchases, what they spend on petrol, the council tax, and so on. Let us assume, too, that these tend to be richer than the average of the 120,000: that they have been earning pounds 200,000 and paying a total of, say, pounds 100,000 a year in all forms of taxation. That knocks pounds 240m off revenues.

Next, look at the people who stay in the country, but change their habits. There are two possibilities here. One is simply to earn less. Some people who were going to retire soon might decide to do so three or four years earlier than they planned. Others might trade money for lifestyle: leave the high-pressure job in the City and work for a charity. Let us assume that another 2 per cent of the 120,000 opt for a change of lifestyle and that their tax payments fall by, say, pounds 50m.

The other change of habit is less radical, for it involves using the various available tax loopholes more thoroughly. Most high-earners are probably already using their full pension allowances, but not all will be. Assume that this prods most of those laggards into action, and that, at the margin, some high earners opt for other non-salaried benefits instead of more cash. A whole industry exists developing such reward packages, but at 40 per cent many people prefer the convenience of cash. Assume, too, that a rather higher proportion of taxpayers buys into tax-favoured investments such as enterprise investment trusts. If, on average, each of the 120,000 people managed on average to clip pounds 5,000 off their tax bill in this way, that would be a further cut of pounds 60m in revenues.

Add this up and you can see that a third of the additional revenue raised by the tax increase would very plausibly slide away. This direct loss might be much more: it is certainly hard to see it being less.

On top of this are the second-order effects: the impact on people who would not pay the tax, but fear that they might at some future stage find themselves doing so. People on the present top rate of 40 per cent pay pounds 30bn of the total pounds 70bn income tax revenue. Any increase in income tax will focus attention on the various opportunities everyone has to cut their bill - typically by making sure that they take up their full pension allowances. If higher-rate payers chipped just 1 per cent off their tax bill by exploiting these schemes, that would be another pounds 300m off revenues. Suddenly two-thirds of the expected gain from the tax increase is lost.

So, on these pretty cautious assumptions, after a couple of years, when people had had time to adjust their behaviour, the additional revenue would not be pounds 1.1bn, but more like pounds 300m to pounds 350m. It might well be less. If one makes more radical assumptions about likely changes in people's habits, you could find the government actually losing revenue.

No one can prove this, because no one can predict how behaviour will change. But we do know from phenomena, such as the unpredicted surge in cross-Channel booze imports, that once people get it into their heads that they can avoid a tax, they will race to do so. If they think tax rates are being increased for largely symbolic reasons, then the impetus to avoid those higher rates is all the greater.

All the talk of a higher top tax rate will already have done some damage. Anyone who moves in these circles will know people who have moved out of the UK in the past year or so in case Labour puts up taxes. If Mr Blair has now imposed his will on his colleagues, then it will be because he recognises that higher tax rates are the wrong symbol. But he and his friends should take comfort from the fact that what makes electoral sense also makes practical sense. Holding rates at a relatively low level actually underpins the government's future revenue base, not the reverse.