The social construction of multinational firms’ tax behaviour

This week I am trying to help the undergraduate students that I teach to understand a range of scholarship that uses social constructivism to study the political economy. When you go back to first principles to teach something you often see some of your own work in a new light, and that is what I am going to write about today. I’m briefly going to sketch out some social constructivist thoughts about multinational firms’ tax behaviour.

For a side project I’ve been reading some of the academic literature on tax compliance. There’s a great summary of the different theoretical perspectives on tax compliance in this working paper by Odd-Helge Fjeldstad and colleagues. The authors divide them into five:

  1. Economic deterrance. This is the classic rational choice framework, in which taxpayers weigh up the incentives, essentially the tax rate versus the penalty and the probability of being caught.
  2. Fiscal exchange. This is where the notion of ‘quasi-voluntary compliance’ comes in. If taxpayers think they will receive something in return – public services, in other words – they are more likely to pay. Although the evidence for this, according to the ICTD paper, is weak.
  3. Social influences. The final three positions, which I am going to appropriate into the social constructivist field, have stronger evidence to support them than fiscal exchange. The first is that tax compliance relates to perceived social norms: if they think their neighbours are honest, people are more likely to be honest themselves.
  4. Comparative treatment. From this perspective, people are more likely to comply with tax rules if they think the system treats them fairly relative to their compatriots.
  5. Political legitimacy. The final constructivist view is that if people trust the government and the tax authority, they are more likely to comply.

There are loads of empirical studies looking into the determinants of tax compliance behaviour among individuals and small businesses. It’s fascinating, but I think I found a big gap: I didn’t find a single study explaining the attitude to compliance among large businesses (or, to clarify, I found studies about staff compliance with management decisions, but not about compliance with externally-derived rules and regulations).

What do I mean by compliance? Trying to define this could easily veer into a discussion about tax morality, but that’s not where I want to go. We know that multinational businesses have a degree of choice over how aggressive their tax position is, and I want to ask, empirically, “what determines this?” Most of the existing work examining multinational firms’ decisions about their tax affairs tends, I think, to be situated in the rational choice, economic deterrance framework. According to that view, the first theoretical position listed above just needs to be adjusted to reflect the additional risks faced by businesses, in particular the reputation risk attached to aggressive tax planning (here is a paper that tries to do this empirically). Current discourse suggests that this risk is increasing, and businesses are revising their positions in response.

John Maynard Keynes

It all starts with Keynes…

This is where my undergraduate class comes in. We could argue that large business decision-making is more likely to be rational than that by small companies and individuals, because a successful large organisation has to create systems that obtain the right information and use it effectively. But a growing trend of political economy work, especially since the financial crisis, has focused very much on the fact that behaviour in financial markets (often by actors who are part of very large companies) can’t be explained solely through rational choices. My students are reading Andre Broome’s excellent introduction to constructivist international political economy, which refers to work in this vein on capital account liberalisation, european monetary union, and behaviour within financial markets.

Economic constructivist thought is often traced back to John Maynard Keynes’ analysis of how actors in financial markets deal with their inherent uncertainty. According to another useful book chapter by Rawi Abdelal and colleagues [pdf]:

Keynes lists “three techniques” economic agents have devised for dealing with this situation, all of which are inherently constructivist. First, “we assume that the present is a much more serviceable guide to the future than a candid examination of the past would show it to have been hitherto.” Second, “we assume that the existing state of opinion … is based on a correct summing up of future prospects.” Third, “knowing that our own judgment is worthless, we endeavor to fall back on the judgment of the rest of the world … that is, we endeavor to conform with the behavior of the majority or average…to copy the others … [to follow] … a conventional judgment.” In short, Keynes’ macro-economy rests upon conventions, that is, shared ideas about how the economy should work.

As the works cited by Broome show, this insight has been applied beyond financial markets to many other processes involving decision-makers and market actors. So why not multinational taxation? If we decide to do so, we can come back round to the theoretical perspectives I set out at the start. Perhaps the calculation about how much uncertainty to build into a company’s tax position is subject to the kinds of conventions identified by Keynes. And perhaps, beyond just uncertainty, the social conventions that influence business decision-making about taxation incorporate perceptions of social norms among decision-makers’ peers, as well as notions of comparative treatment and political legitimacy.

Here is one example, a quote from a survey of corporate tax directors [pdf] conducted by Judith Freedman and colleagues back in 2007 (and which it would be interesting to repeat now). There are remarks here that might support the fiscal exchange, social influence and political legitimacy theories:

One respondent said that his firm’s CSR policy does not extend to paying more tax than is due under the law; they are not interested in ‘making donations to Government’. Others echoed this view, arguing that they could spend their tax savings more wisely than the Government could. At least two firms suggested that there would be a greater social aspect to taxpaying if the amounts collected were earmarked for particular public services, rather than going into general revenue.

Social constructivists devote a lot of time to identifying and analysing the different groups among which social conventions form. There will be overlapping communities within and across companies, including for example the Davos elite, the tax function, social classes, nationality, and so on. I’ve written previously about a paper that touches on transfer pricing practitioners as a social group.

Corporate decisionmakers will have different information depending on their membership of different communities, but they may also be influenced by different social norms within their communities. This morning, for example, Chris Lenon suggests that corporate tax advisers are frustrated by what they see as a lack of rationality on the part of their boards, reporting “a gap between [tax advisers’] perception of the direction of travel in this debate and the denial of this at Board level because of the impact on earnings.”

I understand the normative case against using social norms to make up for deficiencies in the law, as well as the populist argument to the contrary. But I wonder if that debate might be seen in a different light if we begin from a constructivist ontology, in which, like it or not, social norms of one kind or another have always conditioned corporate tax behaviour.

What ‘tax responsibility’ might look like in the real world

In my conversations with tax executives from multinational businesses, I’m used to hearing them talk about the decision any business makes about how aggressive its tax positioning is going to be. The position advanced by many campaigners is that a socially responsible business will make a conscious decision to be at the less aggressive end of this spectrum. The savvier campaigners try to make a business case for this, as well as an ethical one.

The tricky part comes is in clarifying what ‘responsible’ looks like. I’ve never met a tax executive who will admit that her company chooses to take a more aggressive position than others, and yet I’m sure some do. All companies are able to put out vague statements about compliance with this and cooperation with that, but I’ve never read one where I could conclude from it that particular examples of tax planning schemes would be ruled out. So if campaigners want to paint a picture of a responsible business that would require behaviour change from some companies, they need to find some concrete statements at policy level that act as a litmus test.

(Yes, I understand that many in the tax profession think this discussion is a poor substitute for fixing the law to eliminate companies’ discretion and tax enforcement by the mob…I’m not getting into that today!)

On this subject, a great paper has popped up in the journal ‘Management Accounting Research’, by a Danish researcher called Christian Plesner Rossing. It’s a study of an unnamed European multinational, referred to as ‘Global’, which appears to have formulated a tax policy that consciously changes it towards a more risk-averse position. The paper also discusses the role the policy plays in helping the tax department do its job, defined largely as risk management. Transfer pricing is a tricky area, in part because it quickly becomes so complex that senior executives struggle to supervise the transfer pricing manager effectively, and in part because the process of adjusting internal transactions can have a negative impact on the performance metrics of different parts of the group.

I hoped that the clear statement of intent to be ‘conservative’, rather than to be ‘somewhere in the middle of the spectrum’ as most companies tend to say, might lead to some clear indicators of what conservative looks like. Let’s see.

Reputation risk

According to the new policy document, the company’s aim in formulating the policy was as follows:

‘[Global] does not want its tax affairs to appear in the public domain. [Global] will manage its compliance affairs to minimize the risk of any public comment’.

Now I’m sure any company would want to minimise this risk, but this company seems to have prioritised this, as explained by the transfer pricing manager:

We will avoid [transfer pricing] penalties and we will not be on the front page – that is one hundred per cent sure.

Wanting to be one hundred percent sure would surely mean adopting a very conservative approach. More conservative than, say, Vodafone, whose tax code of conduct states:

It is not appropriate for the details of the Group’s tax affairs to appear in the public domain. Vodafone will however only enter into transactions which would be fully justifiable should they become public.

Adjustment risk

The ‘Global’ tax strategy document goes on to say:

All positions taken in the tax returns must be supportable and, on the balance of probability, be more likely than not to be agreed by the appropriate tax authority…The tax department will aim to have no adjustments to the tax returns…Group Tax will take a conservative position in respect of the tax charge in the accounts’.

Now, an aim of having no adjustments, if the tax department is assessed in this way, would surely preclude taking more aggressive transfer pricing positions that come with a risk – all be it a manageable one – of adjustment.

In contrast, Vodafone’s policy, which also invokes ‘more likely than not’, goes on to add the caveat that

there are instances in which a filing position will not meet the more likely than not standard but would still be tenable…[such as] Where there are current uncertainties or opportunities created by recognised errors in law not yet corrected.

I wonder how these two apparently different positions would apply in a developing country where the law and enforcement are weak. It’s clear that firms take liberties in countries without effective transfer pricing enforcement that would not pass the ‘more likely than not’ test, and would lead to adjustments, if they were properly investigated. That explains why, just a couple of years after Kenya began transfer pricing audits in earnest, it chalked up ten major adjustments of multinational firms.


Finally, the policy adds:

All transactions must have a business purpose. All decisions in [Global] must be based on the underlying business and not on internal tax sub optimization. The Group will not undertake nor accept purely tax driven transactions.

This sounds welcome, but hardly radical. Most companies seem to feel comfortable saying this, and my guess is that, while there are plenty of examples of purely tax-driven artificial transactions, most tax planning in a multinational firm also has some kind of underlying substance. Vodafone’s policy says the same thing, and gives quite some detail on how it defines artificiality.


We might possibly conclude two ways to start to define a more ‘responsible’ policy on the basis of this rather quick analysis:

  1. Less risk (vis a vis both the media and tax authorities) is entertained, which means that more conservative positions are taken. (It’s interesting that ActionAid’s tax responsibility guide doesn’t say this, it just states that the level of appetite for risk should be articulated).
  2. Risk is assessed in the same way in countries with weaker legislation and enforcement as it is in countries where it is stronger, so that deficiencies in these areas do not lead to a more aggressive position being taken because it is less likely to be challenged.

Base Erosion and Profit Shifting? It takes one to know one

Last week I was at a research workshop for PhD students at the International Bureau for Fiscal Documentation in Amsterdam. It was very interesting to be in the Netherlands in a rather introspective week about the country’s tax treaties.

The Netherlands is quite sensitive about the accusation that it’s a tax haven, as I found out when the case study of SABMiller I worked on for ActionAid made front page news in the Netherlands. So the OECD’s fingering of the Netherlands in its BEPS report earlier this year has stimulated some interesting debate there, which culminated last week with rival reports published by the Centre for Research into Multinational Corporations (SOMO) and Holland Financial Centre. The combined conclusions are summarised by as follows:

Dutch tax treaties cost developing lands €700m, but earn us €3bn

That’s the Netherlands, but what about the UK? During the IBFD workshop, a Dutch professor, possibly feeling a little patriotic, got rather angry about the BEPS project. Will it really tackle the problem, he asked? He pointed out that it’s being spearheaded by the UK, a country that has just introduced a ‘patent box’ to lure in intellectual property with a ten per cent tax rate. “Talk about base erosion!”

So I thought this little footnote to the Financial Times’ piece on Cadbury’s tax arrangements was interesting:

Dozens of big and medium-sized UK companies are rushing to set up offices in tax havens such as Jersey, Malta, Ireland, the Netherlands, Luxembourg and Switzerland to take advantage of a policy introduced by the British government in April….which creates an “ultra competitive” 5.75 per cent tax rate – a quarter of the full rate – for subsidiaries in tax havens that provide finance for other parts of a multinational group […]

Action Aid, a charity, warns that the rules, which encourage companies to borrow at home and abroad, could cost developing countries up to £4bn of tax revenues. The Treasury rejects the estimate and says it has never acted as “the world’s tax policeman”. But the concern is justified, says a multinational executive. “It is an open invitation to strip everyone else’s tax base”.

I thought it was just campaigners who thought this, but it seems business, too, agrees that the UK is simultaneously positioning itself as leading the charge to tighten up international tax rules, and passing laws that exacerbate (or even exploit?) their weaknesses.

Ben Saunders post on the Fair Tax mark is interesting, I hope it leads to a good debate on methodology

Clamping down on Google’s tax avoidance: don’t hold your breath

Image representing Eric Schmidt as depicted in...

Eric Schmidt. Image by Charles Haynes via CrunchBase

This is a post I wrote for the LSE Policy & Politics blog.

Google’s executive chairman Eric Schmidt will stand up to give a talk at the LSE this evening after a week of unprecedented criticism of the search giant. I wonder if he still feels the same way today as he did last October, when he told journalists:

“I am very proud of the [tax] structure that we set up. We did it based on the incentives that the governments offered us to operate…It’s called capitalism. We are proudly capitalistic. I’m not confused about this.

Last Thursday Matt Brittin, the company’s vice president, was told “I think you do evil” by the chair of parliament’s Public Accounts Committee when he made a second appearance to defend what he had openly admitted were the company’s tax avoidance activities.

Coverage of a pre-G8 summit meeting for business leaders at 10 Downing Street on Mondayattended by Schmidt, focused on the tax avoidance questions, and whether the company’s tax practices had been discussed. David Cameron and Nick Clegg did indeed claim to have challenged Schmidt for his aggressive tax position. Just yesterday, Ed Miliband took the fight to the belly of the beast, throwing the company’s own words back at it. He said:

“I can’t be the only person here who feels disappointed that such a great company as Google, with such great founding principles, will be reduced to arguing that when it employs thousands of people in Britain, makes billions of pounds of revenue in Britain, it’s fair that it should pay just a fraction of one per cent of that in tax.”

The political consensus appears to be that the international tax system is broken and needs to be fixed by governments – to change those incentives identified by Schmidt – but also that, in Miliband’s words, responsible companies should “do more than obey the letter of the law.”

Here’s the problem with this analysis: the reason that Google (and Amazon, Apple and Starbucks for that matter) pay so little tax in the UK is not simply a matter of exploiting weaknesses in the system. It’s also because that’s how the system is designed. We’ve built a set of global rules that define a company’s tax liability not by how much stuff it sells in Britain, but by where it locates the business functions that add most value to the things it sells.

And the reason we did this is because we thought it suited us. On the global stage, Britain is a reasonably sized, but not huge, consumer market. But British businesses have a lot of customers overseas. So we figured that we would be better off with a system that’s biased towards – in the language of international tax – a multinational company’s place of residence, rather than the source of its revenue. It’s that bias that allows US companies to sell billions in Britain while incurring a relatively small tax liability here – a liability that they can reduce further through tax avoidance.

What our political leaders rarely mention is that the difficulties that e-commerce would create for our tax system were foreseen as early as 1997, and supposedly resolved as early as 2001 (pdf). The crucial agreement here was that the sale of a product over the internet would be treated just like the sale of its physical equivalent. If I stand on the other side of the UK-French border, stretch my arms across it and sell you a book, I’m running a business in France and my profits are taxed in France, even though you, my customer, are in the UK. The same would be true no matter how many books I sold. Under the rules agreed in 2001, exactly the same rule applies if I sit in an office in Luxembourg and transfer a book to your Amazon Kindle, no matter how far away I am.

At a meeting debating international tax rules last year, the UK and US opposed efforts by India and other developing countries to change precisely this kind of rule. They wanted to be able to tax the profits of foreign companies that sell services into their huge and growing markets without needing a physical presence. The UK said no, outright, because it would be a “fundamental change in the balance of source and residence taxation.” The OECD, which is in charge of the current review of international tax rules, says it isn’t aiming to change this balance.

The system as it stands suits countries like the US and UK, which are home to large multinationals that sell services abroad without needing a big physical presence.  Many companies selling services into developing countries are either British or providing the services from Britain, and our government doesn’t want to surrender the right to tax these companies. A side-effect of this decision is that it also means getting less tax from the foreign companies with the biggest consumer presence and visibility here.

David Cameron says he’s going to sort the system out, and no doubt there will be some changes that tighten up some of the areas most exploited for tax avoidance. But untangling the costs and benefits of international tax reform is complex, and it’s unlikely that the outcome will put an end to all of the counter-intuitive tax results.

Putting a price on the reputation risk from tax avoidance

What are the reputational consequences of perceived corporate tax avoidance? That’s the question that introduces today’s “Tax and Reputation Forum,” organised by the Oxford Centre for Business Taxation and friends. (It’s at King’s College London, so after the High Court the other week, I’m beginning to think that Aldwych is the centre of tax news!)

The event could not be better timed, and I’m hoping to see the tax profession’s criticisms of the UK parliament’s Public Accounts Committee debated directly with its Chair, Margaret Hodge. It will also be interested to observe where Treasury minister David Gauke positions himself: in front of a mostly business audience, will he be closer to his pre-Starbucks “government and business have to work together to combat the public’s misunderstanding about tax avoidance” messaging, or his bosses’ more recent “this is outrageous and we won’t stand for it” tone?

The conference blurb talks about “perceived tax avoidance”, which is understandable. You’d expect that a risk assessment for potential media coverage would be based on how people might interpret the information in the public domain, not on the underlying tax structure, which may be obscured in the accounts. But what if companies also had to contend with private information from within the tax function coming into the public domain?

Last week the Public Accounts Committee talked about evidence from “whistleblowers”, people from outside the tax function who felt that their day to day experience contradicted how the company’s affairs were described for tax purposes. That doesn’t move us beyond the issue of “perceived” versus “actual” tax avoidance.

But what about a whistleblower from inside the tax function? That appears to be what’s happened in today’s Guardian story about Marks & Spencer. The quote below, form a leaked email, is interesting in two regards. First, because it shows correspondence in a company about the value of a tax saving versus the potential cost of reputational damage from it. Second, because the email appears to have been leaked by someone within M&S:

Given that it was developed as a means to avoid UK corporation tax when it stood at 26% it now seems appropriate to reassess this. Corporation tax will be 21% by next year. Does this not render many of the advantages of having an Irish company obsolete?

From a tax management perspective there may have been advantages in avoiding the UK 26% tax rate but the process and IT overhead with the additional VAT complexity may negate these advantages. Needless to say there is also the reputational damage to M&S should it be seen to be avoiding UK tax in the current climate, as seen with recent examples such as Starbucks [and] Amazon.

I wonder if this phenomenon could really narrow the gap between the public debate on “perceived” tax avoidance, and the internal discussions about “real” avoidance.

The recent ActionAid survey [pdf] gives a nod to G4S for its “explicit reference to [internal] whistle-blowing over tax concerns.” Maybe more companies will choose to put in place a safety valve like this for employees, to try to prevent information spilling out into the public domain.

Secondments, democratic scrutiny and corporate tax

I’ve just been next door to the high court (a perk of being at the LSE!) to watch the UK Uncut Goldman Sachs judicial review. For all those who lament the quality of public debate on questions of corporate taxation, this is surely a desirable outcome: a painstaking debate through which the judiciary will decide who is right. Maybe it makes the case for putting corporate tax settlements in the public domain…but that’s another blog altogether.

What’s really reflected by the Goldman case is a debate about the integrity of the UK’s corporate tax policymaking and administration. At senior levels in both the Treasury and HMRC, civil servants mix with individuals from the top accounting firms and large businesses – on Treasury working groups, on the HMRC board, through secondments, not to mention Dave Hartnett’s acceptance of more corporate hospitality than any other civil servant. All these arrangements have been criticised by people outside the tax profession, while it seems Ministers, civil servants and tax practitioners – the ‘tax elite’ if you like – have closed ranks in defence.

All this sounds very much like what international relations theorists call an ‘epistemic community’. In the classic text on the subject, Peter Haas in 1992 defined epistemic communities based on the following criteria:

  1. a shared set of normative and principled beliefs which provide a value-based rationale for the social action of community members;
  2. shared causal beliefs, which are derived from their analysis of practices leading or contributing to a central set of problems in their domain and which then serve as the basis for elucidating the multiples linkages between possible policy actions and desired outcomes;
  3. shared notions of validity – that is, intersubjective, internally defined criteria for weighting and validating knowledge in the domain of their expertise; and
  4. a common policy enterprise – that is, a set of common practices associated with a set of problems to which their professional competence is directed, presumably out of the conviction that human welfare will be enhanced as a consequence.

I think it would be quite easy to identify all four for our ‘tax elite’ – although I’ll have to leave that for yet another blog post. The key point from the definition, however, is that an epistemic community is bound together by more than shared knowledge or profession: it also includes shared values and a worldview through which knowledge is filtered.

Epistemic communities are associated with the increasing influence that technical experts have in some areas of policymaking  (“consolidating bureaucratic power” as Haas has it). This is sometimes seen as undesirable, as it detaches policymaking from democratic accountability. And that hypothesis is my interest here.

Because the community’s self-perception is founded on ‘expertise’, any criticism of its actions – even by legislators themselves – is put down to ‘lack of understanding’. It becomes self-reinforcing. The most vocal members of the ‘tax elite’ regard most criticism as the product of commentators with a lack of understanding and/or an inherent anti-establishment, anti-corporate bias. Well yes, those elements do exist in many cases, but that doesn’t mean that there isn’t a grain of truth to anything the “tax muggles” say.

Un-PACking the secondments debate

For example, I don’t think the recent Public Accounts Committee report on the accountancy firms did a good job of analysing the issue of secondments from the Big 4 into government. Ben Saunders has quite a reasonable rant about it here. But I wonder if the tax elite would have been any more willing to accept the content of the PAC’s conclusions had they listened carefully to their witnesses, and then respectfully disagreed with them nonetheless.

In dismissing the PAC report, Mike Truman says that “[secondees’] role was to advise on the commercial implications of the proposals.” How can he be so sure? Maybe he took the word of his fellow ‘tax experts’ at KPMG on trust. I encountered a Treasury secondee from KPMG when I was lobbying on behalf of ActionAid, and I’m quite confident that he was much more implicated in the policymaking process than that, consistent with the claims on his LinkedIn page.

Also reflecting on the PAC report, Wendy Bradley argues that:

the accountancy and tax professions…think of themselves as professionals…that they are more like the barrister who can give objective advice to a party to a court case whether the person is innocent or guilty.  They certainly do not recognise themselves in the “poacher turned gamekeeper” that PAC perceives.

For me, the secondments issue is not about the Big 4 gaining inside knowledge of HMRC, but about information flow in the other direction. It turns on whether you think that there is a clear boundary between technical advice and lobbying. In the epistemic communities framework, no such boundary exists. According to Haas:

It is the political infiltration of an epistemic community into governing institutions which lays the groundwork for a broader acceptance of the community’s beliefs and ideas about the proper construction of social reality.

Richard Brooks’ new book argues that just such an infiltration has occurred in HMRC and the Treasury. If so, it’s not at all unreasonable to think that the beliefs, ideas and values that have emerged within a community that is paid for its ability to keep corporate tax bills low might diverge from those of the general public and of the legislature.

For example, here are Canadian tax professionals Brian Arnold and Larry Chapman, in an article that criticises much of the campaigning on Starbucks:

The role of tax professionals in assisting multinationals to avoid tax may require some serious soul-searching. Without our “creative” talents in manipulating legal relationships, complex legislation and tax treaties, this type of international tax avoidance would not be possible.

Wendy Bradley offers another example, the supposedly neutral value of tax “competitivity”:

Meanwhile the big glaring elephant in the room is the [government’s] commitment to a tax system which is “more competitive, simpler, greener and fairer”.  In Taxworld, a “competitive” tax rate is a lower tax rate – a rate which competes for business with other tax jurisdictions. Yes, it’s official coalition policy that, in the great multinational tax race, the UK should do its best to win the race to the bottom.

So what now?

I absolutely agree with all those bloggers who point the finger for much of this at parliament. As Wendy says:

The responsibility for the existence of tax legislation and for the quality of that legislation lies with the people who make it, with Parliament.

[…] each Budget [the Government] publish their proposals, consult on them, and then bring the legislation to Parliament along with a TIIN which tells you what the legislation does and why, how much it will raise and how much it will cost, and who will be affected.

Does Parliament ever look at them?

Do MPs ever challenge the legislation, and if they do, are any changes ever made?

That would make an interesting study, although I think we already know that on complex areas of corporate taxation the answer is no.

To help achieve a better political debate, Mike Truman and Heather Self advocate an “adopt an MP” scheme whereby tax professionals would provide “clear, dispassionate advice on the tax system”. Sounds great in theory, but the epistemic communities framework challenges the notion that such advice exists.

What I think is really needed is an open, honest exchange between the tax elite and the tax muggles. Yes it would help for the former to educate the latter on technical matters, but that should go hand in hand with discussion on the level of “normative and principled beliefs” about corporate tax. And on that, the tax profession might do well to adopt a little more humility.