In the latest of our ESRC-sponsored Ask The Expert seminars, Professor Michael Devereux, Director of Oxford University Centre for Business Taxation, discussed whether multinationals pay enough tax.
“To ensure multinationals pay the right amount of tax, we need to reform the tax system to levy tax on activity or income that is as immobile as possible.”
Recently the media has given a considerable amount of attention to the tax affairs of large multinationals such as Google, Amazon and Starbucks. Coverage generally perceives practices of diverting profits to lower tax jurisdictions, in order to pay little or no corporation tax in the UK despite making a considerable volume of sales here, as grossly unfair. Estimates by HMRC of the ‘tax gap’ – the difference between the amount of tax that should, in theory, be collected by HMRC, against what is actually collected – show a gap of £9.5bn for large business. Surprisingly however – given the media emphasis on these businesses – the tax gap for small and medium-sized businesses is larger still, at £16.5bn. Yet in general terms such estimates are very difficult both to define and measure.
Academic estimates of avoidance by multinationals vary widely, depending on both the approach and data used. Problems include the fact that the counterfactual is very difficult to define – what would multinationals pay if they were doing things differently? Estimates of the amount of income earned in high-tax affiliates that are transferred to lower-tax entities ranges from around 5% of total income to 30% of total income, depending on the study in question. This wide variation in estimates leads to one main conclusion: we don’t really know to what extent multinationals are avoiding tax.
In the face of these difficulties, we need to redesign the tax system to ensure that large companies pay a ‘fair’ share of tax, however that is determined. But what is a ‘tax on business’? In one sense it could refer to most taxes: 90-95% of all taxes remitted to HMRC come from businesses, since businesses are good revenue collectors. In another sense, however, the phrase is meaningless, since all taxes are ultimately passed on to individuals, whether consumers, workers or shareholders. Businesses can therefore be thought to pay virtually all taxes, or pay none at all. Thought of this way, businesses are excellent tools that can be used to collect tax; but determining which individuals the tax is ultimately passed on to is key to understanding who pays tax. Are they low-paid workers or rich business owners?
The fundamental problem is the attempt to levy tax on income or activity that is highly mobile. Both economic activity and taxable income can and do move elsewhere. So what is less mobile? Natural resources – including land and property – are difficult to move. Increasing the share of tax levied on resources could be one way to ensure taxes are paid in the desired jurisdiction.
People are also relatively immobile, at least compared to money – a tax could either be levied on shareholders, or consumers. The former is problematic in practice; there would need to be a way for HMRC to tax any shares UK residents have in foreign companies – requiring substantial amounts of data to be shared between different countries’ tax authorities. A tax on consumers could work – something similar to VAT, but that was designed to tax companies’ profits in the place products and services are sold. There would be two key advantages: a tax levied on place of sale is difficult to avoid; and countries tend not to compete to lower consumption taxes, so a ‘race to the bottom’ – such as the one apparently underway for corporation tax – would be avoided.
However key challenges would remain. First, if a destination-based tax is to replace corporation tax, how similar would the ‘consumption’ element be to VAT? A common complaint about VAT is that lower income groups spend a larger proportion of their income on VAT than higher income groups, since they tend to spend more of their income and save less. In designing a new tax based on consumption one would need to be careful not to unfairly disadvantage those on lower incomes. The designing process is complicated by the fact that we know little about the progressivity of corporation tax – the tax it would replace. Which individuals currently pay for this levy on corporate profits?
Second, a destination-based tax would need to allocate tax revenue fairly between high and low income countries. It may well be the case that low income countries suffer at present from an inability to capture tax revenue from economic activity taking place in that country – but a poorly designed consumption tax would risk replicating this problem. A large proportion of the world’s rich consumers live in high income countries. Focusing tax on the place of sale rather than profit could, if poorly implemented, potentially shift tax revenues even further away from low income countries.