I wrote this article in August 2008 for the Guardian's Comment is Free. I still think it relevant today. Its objectivity is still open to real doubt. I apologise in advance for the fact that some of the links are now out of date: they all worked when it was written:
When the media talk about tax it always seeks comment from the Institute for Fiscal Studies (IFS). They usually make the link to the organisation by describing it as "independent" or "non-partisan". But is it free from bias? I used to buy this claim, but I have been reading recent output from the IFS as part of its Mirrlees review. This review, headed by Sir James Mirrlees, a Nobel prize-winning economist, is, according to the IFS, meant to "identify the characteristics of a good tax system for any open developed economy in the 21st century".
I presumed when approaching the material that I would find reports whose objectivity accorded with the IFS aspiration to be "seen to be free of bias". I have been profoundly disappointed.
A range of examples illustrates the point. In June the IFS published a paper on the future of corporation tax, charged on company profits. The press statement said:
Corporation tax should be reformed or replaced by a higher VAT rate … to reduce disincentives to invest in the UK, according to two studies commissioned by the Mirrlees review of the British tax system.
The institute has suggested some compensation for the VAT charge through reduced national insurance contribution charges, but let's be clear, corporation tax is expected to raise £52bn for the Treasury this year and VAT £84bn. VAT would have to increase to about 28.3% from the existing 17.5% to recover the loss of tax on corporate profits in my calculation. Ignoring for a moment the inflationary impact, this move is massively regressive. The less well off pay a higher proportion of their income in VAT than do the wealthy because they must spend all they earn to live. Worse, since corporation tax is in effect a tax on profits of companies available for distribution to its owners, the vast majority of whom, by value, are in the top 10% of income and wealth owners, almost all the benefits of this change go to that group.
Isolated recommendations do not make an issue though. It is a pattern of similar recommendations that do create a cause for concern. Other points made in the papers published as part of the review are similarly regressive. For example, the review recommends that VAT be charged at the standard rate (currently 17.5%, but maybe 28% if the IFS got its way on corporation tax) on food, children's clothing, books and other currently zero-rated items. The IFS claims, on the basis of economics I have shown to be without foundation, that this would "interfere less with people's spending decisions" as if there is a real choice between food and luxury items. It also suggests some compensation through increased benefits, but it is quite clear in the example given that only the lowest 30% of households would benefit (those enjoying incomes of £16,000 or less a year). That would leave those on middle income decidedly worse off, to provide £11bn for what the IFS calls "further desirable tax reductions". Since the rich almost invariably benefit most from tax reductions because they do pay most tax in absolute terms, the direction of redistribution is likely to be from the poorest to the richest.
That is also the case with regard to inheritance tax, where the thinktank says (pdf) that given "inheritance tax currently raises less than £4bn a year, consideration could be given to abolishing it altogether". It goes on to say it "do[es] not advocate the introduction of a regular wealth tax". All it does is suggest that capital gains tax might be charged on death but that its payment be deferred until an inherited asset is sold, but again the consequence is perverse. The wealthy can always afford to defer the sale of assets. Those of lesser means inheriting assets may well want to realise them to settle their own liabilities. It is, once again, the less well off who are likely to pay most as a result of this recommendation.
And finally, the IFS makes the extraordinary comment (pdf) that:
To discourage investors from hiding their wealth in foreign tax havens, the authors recommend exempting interest income from personal tax, and allowing shareholders to deduct an imputed normal return on the basis of their shares before imposing tax on dividends and capital gains.
I've done the sums: this would cost the government at least £10bn a year in lost tax revenue, and the saving would (unsurprisingly) go mostly to higher rate taxpayers, some of whom would benefit directly from a change in policy motivated by their previous tax evasion.
But the real question is this: is this thinktank really free of bias? Can it be, when it suggests abolishing tax on all corporate profits, on interest, on wealth and on half of all dividends (as is likely) whilst at the same time suggesting VAT be charged on all food and promoting a massive increase in VAT in general, when that tax is known to be heavily regressive, whilst also proposing tax cuts that would deny the government the revenue to compensate the losers from these changes? I don't think so.
I think the IFS has a bias towards the neoliberal view that suggests that labour should be heavily taxed whilst capital is left virtually tax free. It is one to which successive UK governments have subscribed. As another of the reports (pdf) published by the review says "Whether this move to the right [in taxation attitudes] will persist and to what extent it is now a fact of political life is hard to say." It seems to me that either the IFS wants to make sure it does or that the whole report is just an indication of a lack of political will on the part of the UK to cooperate internationally to ensure capital is taxed.
Whatever the motive, the IFS's claim to be unbiased appears to me shaky. Some of its proposals are very dangerous indeed and about as far removed from the characteristics of a good tax system for any open developed economy in the 21st century as it is possible to be.