The problems with measuring tax systems

The following blog by TJN’s Nicholas Shaxson (currently on a book writing sabbatical) was originally posted on the SPERI blog and is re-posted here with permission.

In debates about tax policy we need to de-emphasise the role of economics and measurement and rekindle the politics

In the past few years there have been several efforts to understand and even measure ‘spillovers’ – that is, how one state’s tax or legal system can transmit damage to other states’ tax or legal systems.  Perhaps the best known of these efforts is the Tax Justice Network’s Financial Secrecy Index (FSI), which attributes a secrecy score to every country measured, then combines that score mathematically with a size weighting, to create a ranking of the world’s most important secrecy havens.  The index has been extremely effective in drawing attention to the issues, and in uncovering a lot of new data and analysis and understanding of the offshore phenomenon which lies at the heart of financial globalisation.

The FSI deals with secrecy: there is a clear need for something similar in the area of tax.  Corporate tax loopholes in one country, for instance, can have similarly damaging effects on the corporate tax systems of other countries, by encouraging multinationals to shift profits to the lower-tax jurisdiction, depriving the higher-tax jurisdiction of revenues.  People are, rightly, rather angry about this.

A little work has already been done in this area.  The IMF published a paper in 2014 entitled ‘spillovers in international corporate taxation,’ a first stab at measuring the scale of the phenomenon.  Coming at it from a different angle, Oxfam recently published a report on ‘the world’s worst corporate tax havens’ whose methodology produced a ranking a bit like Financial Secrecy Index.  The corporate tax havens of Ireland and the Netherlands recently published ‘spillover analyses’ of their own tax systems which, surprisingly or unsurprisingly, depending on how cynical you’re feeling, largely absolved themselves of blame.

This is an area where much more work needs to be done.  Now Andrew Baker and Richard Murphy offer a new, broad framework for thinking about how one might go about it.

Their blog Reframing Tax Spillovers, and the associated paper for the APPG, rightly highlights the flaws in these other projects, and offers something more comprehensive and useful.

Crucially, Andrew and Richard recognise that there are different dimensions of spillover: not just from one country or state to another, but also between different taxes in the domestic economy.  For instance, the corporate income tax was originally set up in order to defend the ordinary income tax: if you have no corporate tax then rich folk simply convert their ordinary income into corporate income and escape the income tax.  As they put it: ‘Taxpayers will try to divert part of the income that should be subject to this tax to another tax or location, or both.’  This happens all the time – and of course there are spillovers that cross both tax boundaries and national borders.  This early-stage concept is highly welcome, and I can imagine it flowering into something big and useful.

Yet there is another generic issue that also needs highlighting: the conservative bias in measurement itself.

Much has been said about neoliberalism – the disenchantment of politics by economics, as Will Davies has put it – in a sense, the effort to shoe-horn as many aspects as possible of life, the universe and everything into the price system.

Much has been written about how neoliberalism, neoclassical economics and the economics faculty at Chicago University have injected a conservative bias into economics.  But there’s an even deeper problem than this: in the area of tax, the very act of measurement is likely to impart a conservative bias.

This is for a pretty simple reason.  Take a corporate tax cut, for instance. Leave aside the question of tax spillovers to other countries, and start by asking: ‘does this tax cut help my own country?’  What does it look like from a purely selfish national perspective?

Many studies have done this.  Does the corporate tax cut foster new corporate investment, or bring in Foreign Direct Investment (FDI)?  Oceans of work have been done here, and plenty of the political discourse in Britain, and in many other countries, leaves the matter at that. If it attracts FDI then that tax cut is ‘competitive’ – so let’s do it. After all, who could oppose a ‘competitive’ tax system?

But of course the story doesn’t end there.  FDI is a means to an end, not an end in itself.  If you have to spend a lot of treasure to attract that FDI, the cost may not be worth it.  Britain’s corporate tax rate cuts since 2010 are forecast to cost nearly £15 billion a year in lost tax revenues by 2021 – which is well over a third of the education budget.  It’s hard to see that this equation makes for a ‘competitive’ tax system – whatever ‘competitive’ might mean in this context.  I would argue that pretty much all of that academic work just measuring elasticities is, for this reason, pretty meaningless from a policy perspective.

If one could do a good cost-benefit analysis – as in ‘here are the benefits of a given tax cut, weighed against the costs’ – then one might be able to draw a better conclusion about the merits or disadvantages.

But this is where the conservative bias comes in. It’s relatively much easier to measure the ‘benefits’ side of a tax cut – FDI responses, elasticities and so on – than it is to measure the costs.

That is, it’s relatively easier to measure things like investment responses, elasticities, which in isolation tend to favour tax-cutting, than to measure the other side of the ledger where the damage of tax cuts shows up: such as by reducing the long-term benefits of (tax-financed) infrastructure or education, which might play out over decades; or confidence in the overall tax system and democracy itself, as corporate tax rates fall far below personal income tax rates, or the effects of higher inequality exacerbated by corporate income tax cuts, and so on.

As the US public finance expert Robert G. Lynch put it to me recently: ‘It is much harder to measure the damage from reductions in public investment due to tax cuts than it is to measure the benefits from tax cuts.’

Researchers often stick to what they can measure, and to the extent that they do acknowledge the other, harder to measure side of the equation, it tends to be in more narrative form.  So even when there’s a suitably nuanced report, policy-makers can cherry-pick out the numbers and throw away the narrative as so much fluff.  This happens all the time.  And that’s before we even get to the lobbying and role of private finance sponsoring some academic research.

There’s no obvious way around this: more narrative emphasis on the hard-to-quantify stuff will just get airbrushed out of the way of tax cuts.  What is necessary is to de-emphasise of the role of economics and measurement in these debates, and to rekindle the politics.  Civil society has been doing a decent job here: but academia and policy-makers have too often been in the thrall of the economists.

What is needed, at the end of the day, is to pursue the disenchantment of economics by politics.