Behind the 50p rate headline figures
Estimates of how much the top rate of tax will raise, or lose, have dominated finance pages. But to what extent can we rely on the headline figures?
Estimates of how much the top rate of tax will raise, or lose, have dominated finance pages. But to what extent can we rely on the headline figures?
THE PHONY WAR surrounding the potential abolition of the 50p top rate of tax has been raging on for the past couple of months.
The rate, introduced by former chancellor Alastair Darling in the 2009 Budget, was only ever meant to be temporary. However, the coalition government has said no decision will be made on its abolition until 2012 – after the yields from the first year of its implementation are confirmed by HM Revenue & Customs.
This has not stopped significant debate. The abolitionists won a significant battle last week when 20 leading economists wrote to the Financial Times claiming that the rate was a hindrance to economic growth.
However, the pro-top rate faction immediately hit back with figures from a parliamentary question in November 2010, which showed the Treasury estimates for the revenue generated by the 50p tax rate were an extra £12.6bn generated over five years.
Both sides have attempted to use figures to back up their claims. But how much reliance should we put on the various projections?
Treasury figures
The figures revealed in a parliamentary answer in November last year and publicised again last week seemed to show a significant increase in cash collected because of the tax rate. Commercial secretary to the Treasury Lord Sassoon’s answer to Lord Ashcroft’s question showed the Treasury estimated to collect £1.1bn more income tax in 2011/12 than if the 50p rate was not in place, with a further £2.8bn in 2012/13 rising to £3.2bn a year by 2015/16. Over the course of the five years, this will make an extra £12.6bn more than if the top rate was abolished. This takes into account behavioural changes, the government says.
As the table below shows, previous Budget estimates at first look very different to these. In March 2009, the Budget book (table 1.2, page 11) said that the 50p rate would bring in £1.8bn extra in 2011/12 and £2.4bn in 2012/13. These estimates were to increase in the March 2010 Budget (table A11, p140). The yield in 2011/12 was estimated to be £3.1bn and £2.7bn in 2012/13:
There is a simple explanation for why these wildly differ for 2011/12 – some of these calculations are based on the amount of cash collected, others are based on the total yield, etc.
But 2012/13 raises some more interesting elements. The June 2010 figures show some pessimism from the Treasury as the figure should actually be far greater than the March 2010 and March 2009 figures for two reasons.
First, the June 2010 figures explicitly exclude National Insurance contributions. The other two figures include them, as they estimate the total tax yield increase. Factoring in NICs should theoretically reduce the difference between the 40p and 50p rates. This is because, even if the income tax yield increases, the numbers of people paying the rate will decrease through tax avoidance, individuals leaving the country, etc – which the Treasury acknowledges through factoring in behavioural differences. As the NIC rate remains the same, this will reduce the benefit of the 50p tax rate.
Secondly, the Treasury announced limitations to the use of pensions tax relief at the same time as the 50p rate. Therefore, it should boost the collection of income tax even further. The way the Treasury publishes Budgets means the March 2010 and March 2009 figures do not include these changes. But the June 2010 figure does.
Therefore, it seems as though, if anything, the Treasury was more pessimistic on the yield in June 2010 than it was in March 2010. But really, the only thing we can take from these figures is that the Treasury is expecting some form of extra yield from the 50p rate.
Institute for Fiscal Studies
The same cannot be said for the figures being quoted by the Institute for Fiscal Studies, the respected economic think-tank. The Telegraph claimed that the IFS had said the top rate would lose the Exchequer £500m. This was based on an IFS briefing note – Can more revenue be raised by increasing income tax rates for the very rich? – which in itself was in response to the proposed introduction of a 45% tax rate.
The graph above uses a parameter called “taxable income elasticity (TIE)”, which measures how behaviour of the very rich changes when the marginal rate is increased.
The IFS estimates that, because of the behavioural changes outlined earlier, the taxable income of very rich individuals declines by 0.46% for every 0.5% the marginal tax rate increases. The 0.46 line represented on the graph does indeed show a decrease in revenue of around £500m when the 50% rate is introduced.
However, this figure is misleading. The issue of behaviour change is key to the debate and is very difficult to estimate. The Treasury’s estimate of TIE – represented by the 0.35 line – is within the margin of error of the IFS’s estimate, and shows a £1bn increase. And the following chapters of the briefing explain these TIE projections are not reliable in any case.
Slightly more reliably, the Sunday Times claimed last weekend that the IFS director, Paul Johnson, had indicated the 50p rate would be revenue neutral, raising no money when behavioural changes are taken into account.
This was given weight by the Mirrlees Review’s Dimensions of tax design publication, which said: “There does not seem a powerful case for increasing the income tax rate on the very highest earners, even on redistributive grounds-it would not generate much, if any, extra revenue to transfer to the less well off.”
The optimal effective top rate of tax would be 56.6%, it concludes – which, of course, includes National Insurance and indirect taxes. With the 50% top rate of tax (and the 20% VAT rate), this roughly currently equates to around 64%.
However, the IFS’s projections on the behaviour of the very rich are all based on data from the 1980s, when the income tax rate was changed from 60% to 40%. Stuart Adam, senior research economist at the IFS, says any estimates are “very tentative”. The institute’s best guess is slightly more pessimistic than the Treasury’s, he says. “It is certainly true that a lot has changed since the 80s and that is the reason that there cannot be any certain estimates. On the one hand, there has been more tax avoidance legislation; on the other, there is globalization, making it easier to set up abroad,” he adds.
Even an analysis of the 1980s figures cannot make conclusive judgments about how behaviour in 1980 was affected. It cannot be known whether salaries changed because of the tax, for example. To use these studies as a definitive estimate of tax lost is to ignore the studies’ conclusions themselves.
A matter of faith
The only thing we know about the estimated increase in revenue from the 50p rate is that we know very little. Without even looking at the political environment – arguably more important that the economic aspect in this debate – this is a minefield.
Despite the widespread interest in the FT letter, there wasn’t actually much in terms of economics. There was reference to the UK being “less competitive” and people moving abroad. But these are unknowns. In this respect, it is quite sensible for Chancellor George Osborne to wait until HM Revenue & Customs publishes its figures in early 2012 before making a judgment.
Even then, it will be hard to make conclusive judgments. There will still be the unknown unknowns – how many people have been put off from coming to the UK, whether it is a factor in businesses staying away, whether tax avoidance would not have taken place had the 50p rate not been in place, etc.
Until then, headline estimates will not be helpful. Indeed, the abolition of the 50p rate could well come down to politics, rather that economics. It is likely the HMRC figures will be interpreted depending on people’s original standpoint – and will be a matter of faith in entrenched economic beliefs rather than empirical evidence.