This article appeared in Tax Journal on 13 October 2016 and is reproduced with permission. (Available as a pdf)

The art of estimating the scale of tax avoidance involves tentative extrapolations, careful assumptions and caveats. The data available is patchy, calculations are educated guestimates, and studies must first deal with the problem that the term ‘avoidance’ itself means different things to different people.

The tax gap

Perhaps the most well known estimate in the UK is HMRC’s tax gap. HMRC defines avoidance as ‘bending the rules of the tax system to gain a tax advantage that Parliament never intended … It involves operating within the letter – but not the spirit – of the law.’

What this means in practice is that the avoidance tax gap measures ‘tax under consideration’ which isn’t successfully collected. The assessment is based on a bottom-up approach that looks at avoidance schemes disclosed under DOTAS and tax risks identified through individual risk assessments of large businesses. This gives an estimate of £2.7bn lost to avoidance (for 2013/14), with £1bn relating to corporate tax.

This £1bn of corporate tax avoidance amounts to 5% of total corporation tax liabilities, or somewhere around 0.2%of overall government revenues. It is significant in terms of assessing the effectiveness of tax administration, but does not support the popular perception that corporate tax avoidance is a massive drain on public resources, nor that large businesses are running rings around the tax officials.

As David Gauke, then financial secretary to the Treasury,put it: ‘there is understandable anger when individuals or companies are perceived not to be contributing their fair share, but we can reassure the public that the proportion going unpaid is low and this government is dedicated to bringing it down further.’

Few are reassured, however. A survey by YouGov in 2015 found that two thirds of Britons believe big companies are not paying their fair share of corporation&tax; while 80% felt small businesses paid appropriate corporation tax but their larger counterparts did not.

Getting the measure of BEPS

Critically, the HMRC tax gap does not include base erosion and profot shifting which exploits gaps and mismatches in international tax rules. While such cases fall outside of HMRC’s formal ‘tax avoidance’ definition, they define the issue in the public eye through cases such as Apple, Google and Starbucks; although these companies have argued ‘we pay all taxes due under the law’, the concern is that that the law has not kept up with the reality of globalised business.

Academic and government studies confirm that BEPS is a real problem. However, there is little certainty over the scale or extent of the problem. Rough estimates are emerging. The OECD calculated that effective tax rates experienced by large multinational companies are on average 4 to 9 percentage points lower than domestic entities, putting the overall fiscal effect of BEPS at 4–10% of the global CIT tax base, or US$100–240bn. Other recent studies come up with numbers in a similar range.

Such research confirms that multinationals do pay less tax than domestic companies and that the amounts can be significant. The OECD says that measuring the scale of BEPS remains challenging because of the complexity of the issues and data limitations. It is clear though that emerging estimates do not fulfil the perception of huge lost funding for public services; nor does it say anything about whether multinationals should be paying more tax in the US or in Europe. The European Commission presents the case for change in the tax system by saying that 1 in 5 tax euros are lost to corporate tax avoidance (see, for example, www.bit.ly/2dw56nK). However, its study says that if a complete solution to the problem of base erosion and profit shifting were available and implementable, it would have an estimated positive impact of 0.2% of the total tax revenues of the member states.

Mistakes and misunderstandings

Some of the most influential numbers which shape perceptions of the scale of the problem and of potential solutions are mistakes and misunderstandings. Many of these concern the scale of multinational tax avoidance in developing countries, as I have highlighted in a policy paper ‘Can stopping ‘tax dodging’ by multinational enterprises close the gap in development finance?’ (CGD, 2015).

Often, the perception that clamping down on corporate tax avoidance would have outsized impacts on the finances of the poorest countries has been built up through simple wishful thinking. Estimates of tax avoidance in developing countries are frequently illustrated with a photograph  of a child from one of the world’s poorest countries and compared to global foreign aid. In practice, though, these estimates mainly relate to major emerging economies such as Brazil, Mexico, China and South Africa, rather than countries such as Malawi or Cambodia.

A much quoted figure is that developing countries lose three times more to tax avoidance by multinational companies than they receive in aid. The original source for this is a single sentence in a Guardian opinion piece by Angel Gurria, OECD secretary general, in 2008. In fact, Gurria was not offering an estimate of tax loss, or even talking about multinational corporations at all, but giving an estimate of the amount of capital held offshore by citizens of developing countries – gross amounts whose interest might be taxed. It is simply not, and never was, an estimate of revenue loss due to multinational tax avoidance.

Another fertile source of misunderstandings are the annual estimates of illicit financial flows (IFFs) through invoice fraud, produced by the Washington based NGOGlobal Financial Integrity. These figures (US$1trn a year) are sometimes misunderstood as an estimate of multinational tax avoidance. Kofi Annan, for example, has said that Africa loses $38bn to transfer pricing, based on this misunderstanding. Similarly, the High Level Panel on Illicit Financial Flows led by Thabo Mbeki argued that Africa loses $50bn to illicit financial flows, with the main culprit being multinational enterprises.

Not all wishful thinking relates to fiscal resources for developing countries. One figure often quoted to illustrate the problem of clever accountants and outdated tax laws is that EU countries lose €1trn a year to tax avoidance and evasion. This has been used to represent the scale of corporate ‘tax dodging’ in Europe by many MEPs and by Herman Van Rompuy and José Manuel Barroso, both former presidents of the European Council. (This also appears to be the basis for the European Commission’s statement that 1 in 5 tax euros are lost to corporate tax avoidance.) However, this figure is based on a paper byRichard Murphy which mainly focuses on tax evasion in the grey economy, such as cash-based informal businesses. It was never intended as an estimate of tax avoidance by multinational companies.

Mind the perception gap

More research is certainly needed to understand the scope and scale of BEPS, and the OECD has promised to undertake this as part of the BEPS package. However, given inflated expectations, even those organisations trusted to provide careful and technical analysis of this complex problem have taken to exaggerating their findings. The OECD BEPS Action 11 report developed a core estimate that BEPS revenue losses amount to 4–10% of CIT revenues; however, the press release (and the letter to the Financial Times from then chancellor George Osborne, EU commissioner for financial affairs Pierre Moscovici and German finance minister Wolfgang Schäuble) ran with the racier headline of ‘multinationals paying as little as 5% in corporate taxes’.

Similarly, the European Parliamentary Research Service (EPRS) commissioned a study by a group of academics which estimated that aggressive corporate tax planning costs EU countries €50–70bn a year. However, in the run up to the final negotiation of the EU Anti-Tax Avoidance Directive, the EPRS instead promoted another figure fromthe paper, €160–190bn (including €23bn from the UK), which the authors said was an overestimate. (For further detail on this figure see the panel below.)

The exaggeration of claims around tax avoidance is perhaps just one symptom of ‘post-factual democracy’, in which political narrative increasingly trumps evidence and analysis. But this risks policies being developed based on unrealistic views of potential gains and negative impacts.

Campaigners are right that the debate over shaping tax policy for the 21st century is too important to be left to the experts. But without a common evidence base, it is impossible for tax experts, campaigners, politicians and business to engage together in thinking about the challenges, options and trade-offs. While there is a clear case for anti-avoidance action, there is no way that it can deliver ‘one in five euros’ of tax revenues, as the EC suggests, or solve the problems of the poorest countries.

In other areas such health, scientific claims and political debates, fact checking initiatives have sprung up to support public understanding and the integrity of the debate. Perhaps the tax profession should take the lead in developing a public interest fact check initiative on tax.