Thursday, 22 March 2012

Why the 50p tax cut is based on Osborne's own dodgy dossier

A shorter version of this was on Liberal Conspiracy

Why the 50p tax cut is based on Osborne's own dodgy dossier

Yesterday, the Government released the long-awaited HMRC report on the impact of the 50p tax. The report argues that the tax only raised around £1 billion a year – lower than the Treasury's initial forecasts of £2.6 billion a year.

This has seeped into received media wisdom, confirming the prejudices of much of the right leaning press. It was used to justify cutting the rate to 45p, and forms the bedrock for the entire edifice of the Government's Budget narrative – that they are taking 'five times more from the rich'.

But there's growing evidence that the document was not a proper basis for justifying such a massive and controversial tax give away to the richest in our society. Here is a summary of some of the report's main flaws flagged up since its release yesterday:

  1. The report rests on analysis of one year's Self Assessment returns – that of 2010/2011. It points out that, as the last Government didn't bring in the 50p tax until a year after it was announced, the super-rich essentially paid themselves their salary for 2010/2011 in the tax year of 2009/2010 instead, to avoid paying the new 50p tax (this is called 'forestalling'). HMRC estimate that between £16-18 billion in total was brought forward in this way, and say this corresponds to a yield of about £1 billion for the 50p tax after other 'behavioural effects'. Thus, there is a huge drop displayed in revenue for 2010/2011 and an inflated amount for 2009/2010. This is the main empirical basis for the claim that the wealthy 'just avoid the 50p tax' and that it brings in very little.

    But the problem with this, as many (including Faisal Islam) have pointed out, is that by its very nature forestalling can only happen once. The super rich couldn't have gone on paying themselves in tax year 2009/2010 forever. In the years after, they'd had to have stumped up the cash. In addition, only 90% of the Self Assessment returns had been received by the time HMRC had started the review. The data for 2010/2011 is thus hugely distorted, and cannot be a reliable basis for such a judgement on the 50p tax.

  2. Acknowledging this problem, the report goes on to estimate how much the tax would have brought in independent of 'forestalling' and other one-off behaviours. Again, they come up with about £1 billion. But again, this is highly questionable. It's based on high levels of other avoidance. The initial Treasury estimates already factored such avoidance in to its initial estimates in 2010. What the HMRC have done is extrapolated on the basis of an even higher Tax Income Elasticity (TIE) – the extent to which the wealthy change their behaviour, including avoidance, in the face of higher tax rates – than contained within the initial Treasury figures. It has deliberately picked a higher rate (0.45 rather than 0.35), and the 'post behavioural yield' of the 50p tax is less for this reason.

    As Paul Waugh has pointed out, this was a highly political decision. Language within the report suggests it was done at the behest of the Government. This decision, in turn, seems to have been largely based on one academic study – Brewer et al (2009). Howard Reed is among many to have cast doubt on this study, and argues that past Treasury figures have actually overestimated such avoidance (TIE) rather than underestimated it as the Government now implies. Nevertheless, it remains just one academic study – there is no proper empirical evidence presented in the report demonstrating avoidance and behaviour change on this scale over the 50p tax in the UK. The claims that rich bankers simply moved abroad after the introduction of the report has also been debunked. This, of course, is notwithstanding whether any Government should ever just accept defeat over such behaviour rather than cracking down on it!

All of which is highly wonky and technical. But the upshot is, relative to the scale of the judgement it's making, the HMRC report on the 50p tax is scarcely worth the paper it's written on (and it's a download only document!). It was hobbled from the start by George Osborne, who limited it's only empirical analysis to one highly distorted, unrepresentative year of Self Assessment returns. This is important because the report forms the basis for which the OBR scores the reduction to 45p as only costing £100 million – in reality, it is likely to be far higher, perhaps even £3 billion. If the Chancellor had wanted a proper review of the 50p tax, he could have waited and got the HMRC to analyse Self Assessment data for years in which the wealthy couldn't have 'forestalled' their income . He didn't, presumably because it would have risked returning figures that didn't suit his prejudices.

All of this is why even the OBR's boss Robert Chote has told Channel 4 that “This is a judgement based on not even a full year’s data...The costing of these sorts of changes is by no means unarguable…”.

Those on the left need to bare this in mind. Those of us opposed to this tax cut for millionaires should not allow ourselves to be presented as defending a tax which raises 'little or no money'. The dominant media frame for this debate, 'we know it raises no money but does cutting it send the wrong signals?' is based on a fundamentally flawed premise; an incomplete and effectively theoretical report. It is not about head against heart – there is good, hard headed reason to believe the 50p tax raises significant sums of money. It is up to its opponents to produce credible, empirical evidence to the contrary. Yesterday's report failed this test spectacularly, and we should say so as loudly as possible.