Why taxing global companies is hard

The sun shines on Ardbeg distillery warehouse

Half of scotch whisky is sold in China and North America.  Should the profits of whisky companies be taxed in Scotland, where the whisky is made, or in China and North America, where half of it is sold?  Where exactly is the profit on these sales made?

Now ask yourself the same question about the profits made by Google, which sells advertisements in Europe on websites built in in the US using technology designed in the US.

Companies are generally taxed on their profits, which are defined as revenues minus costs.* That’s reasonably clear in the old world, in which revenues and costs happened mainly in the same place.  But with global companies, revenues and costs can be in different places. If Google’s costs are mainly in the US and some of its sales are in Europe, where are the profits on those sales made, and where should they be taxed?

The problem isn’t only lack of data or lack of transparency. Even if we had full information on a company’s activities in each country (which we should) that doesn’t answer the underlying question: where is this profit being made?  Profit is the difference between two much bigger, geographically dispersed flows (revenues and costs), and it isn’t straightforward to assign a geography to it.

I think we can all agree that, whatever else we might think, the profit is not being made in the Bahamas or the Cayman Islands.  The lack of clarity about where profits should be taxed should not be an excuse for companies to be able to set up artificial structures that channel their reported profits to low-tax jurisdictions.

How we answer this question has big implications for the global distribution of taxes, and hence of the provision of government services. If we tax profits in proportion to sales, then taxes will be mainly due in countries with wealthy consumers.  If we tax profits in proportion to employment, then taxes will be mainly due in countries with cheaper workers.  If we tax profits in proportion to costs, then much will depend on how we treat the value of intellectual property, natural resources, brand values, the cost of capital, and so on.  It will also make a big difference if we compare costs at market exchange rates or at purchasing power parity.

While most countries have a shared interest in getting this sorted out, so that global companies are paying tax on their profits somewhere, the interests of countries diverge when it comes to deciding how to sort it out.

One thing is clear. It cannot be for OECD countries alone to decide the answer to this.  Developing countries need more than technical assistance to help them administer an international tax system designed by the OECD: they need to be part of agreeing the rules of the game.

This is, of course, a rather simplistic, small fraction of a much more nuanced and complex debate. But judging by the commentary on Google’s tax affairs, there isn’t a widespread appreciation of the basic conceptual difficulty of how national governments should go about taxing the profits of global companies.

[* Note: as well as paying taxes on profits, companies also contribute to tax through the income taxes paid by their employees and shareholders, and through the sales or value-added tax paid on their sales.]

Update: here is John Gapper of the FT making the point that this may be the beginning of a big international battle on how to tax profits.

12 thoughts on “Why taxing global companies is hard”

  1. Pingback: Why taxing global companies is hard – Blog – Owen abroad | Public Sector Blogs

  2. Thank you Owen – this sets out the problem very clearly. I worked in the tax department of an accountancy firm a long time ago, so maybe my thinking is too influenced by that experience. But it seems to me like a ‘second best’ problem. The first best system of company taxation, at least according to the OECD and the framework of tax treaties, is to tax profits. But that is getting harder and harder because of (1) the spread of tax havens and (2) asymmetries of information allowing companies to shift profits to wherever suits them. The ongoing erosion of the tax base makes this system untenable.

    Unfortunately, as with other second best problems, the theory does not tell us which of the alternatives is most effective. Revenue is the hardest to game, but has severe distributional consequences: Saudi Aramco could become the largest U.S. taxpayer. My personal preference would be to retain the profit basis with three modifications: mandatory country-by-country reporting of revenue, profit and tax; tweaking IFRS to require disclosure of taxable profit as well as accounting profit; and an alternative minimum tax of (say) 5% on revenue if the profit tax fell below that figure. Now back to the experts.

    1. Thanks Rupert. I don’t see this quite the same way as you. Even without tax havens and asymmetric information, there would still not be a good answer to the question: where is the profit made?  If an iPhone is designed in Cupertino, assembled in China from parts made in Malaysia and Taiwan, uses software written in the US and India, sold to a British consumer in a store on Regent Street in London and supported thereafter from a call centre in Dublin, where is the profit made?  The lack of a clear answer to this is NOT because of tax havens or because we don’t have all the information. Even if I had perfect information, I still couldn’t tell you where the profit is made.  And that means that it is not as simple as you make it sound to “retain the profit basis”.

  3. Agreed – there isn’t a clear answer to the question “where is the profit made”. The implicit assumption in much of the commentary on this is that multinationals have the same net profit margin in every country they operate, and should thus the tax base will be proportional to their local revenue (even if the rate varies). That is an unrealistic assumption. But as you make clear, global supply chains are so complex that we can’t determine where the profit is made, no matter how sophisticated the accounting rules. That is why I argue we need country by country reporting: it gives us more data on where companies book their costs, and if the global distribution of profits turns out to be inversely correlated with the applicable tax rate, that strengthens the case for an alternative minimum tax on revenues.

    1. Very useful discussion. From my perspective (I am not a tax or accounting professional) it appears the problem of identifying where profits are made is subsidiary to if genuine profits are made. The Google UK settlement relates to profit made from intra group transactions.This suggests to me that there is scope for identifying group corporate structures that include entities which are heavily associated with a single flow (profit, revenue, cost). Companies which achieve minimum thresholds for each might attract advantageous corporate tax rates against companies that do not.

      If the process of incorporation involved an assertion about corporate purpose and the evidence from financial statements aligned with such purpose there would be a formal basis for different tax rates which might act as a disincentive towards the creation of artificial businesses.

  4. Meh.


    While Owen’s right in his point the answer is simple.


    Don’t try to tax corporations. Just tax shareholders on their incomes from their investments.

    1. Tim – That was my first instinct too: tax the income in the hands of the shareholders.

      The fly in that ointment is International equity.  A company owned by Europeans but working mainly in Mozambique would have its profits taxed in Europe, even though most of the value added was in Mozambique.

      That would be fine if we had an effective  mechanism for international redistribution, but we don’t.  So I think it is reasonable for the Government of Mozambique to want to tax companies on profits that they make there.


      1. Two answers to that.



        Profits from business (OK, this is about entrepreneurial profits but a foreign company in an economy can, with a squint, be regarded as an entrepreneur) are trivial as compared with the consumer surplus. 3% or so, the consumer getting 97% of the value created. It’s the value of being able to use the things made which is important, not the profit skim off the top of that value.


        So, in the scheme of things “the Government of Mozambique to want to tax companies on profits that they make there” is irrelevant.


        Bluntly, do we care that some foreigner is making 4% on their capital or that farmers have fertiliser to use?


        Resource rents and so on of course are different. Oil, copper, minerals in general, should be taxed until the pips squeak. But given that they are not, by definition, mobile, this is not distortionary.


    2. Hi Owen,

      Nice blog post cutting through the noise to the core question at the heart of the international corporate tax debates.

      Your whisky example though is probably too simple, as it looks like a straightforward matter of production-for-export, rather than multinational corporations operating across borders. If Walmart sets up a distillery in Scotland to produce whisky to sell through its stores in China – that would be more like it. Then the question would be how much of Walmart’s whisky profits should be taxed in Scotland, China; or in the US where its finance, strategy, and super-market know-how comes from.

      i.e as well as different physical stops in the value chain we have to consider where and how to tax the profit associated with what the multinational brings in terms of technology, brand, know-how and the ability to raise finance  – this is different from ‘costs’ which you put it under (under the current system such passive income streams IP royalties, interest & dividends are taxed in the ‘residence’ country towards which they flow). This stuff is often perceived as being, by definition, a means of tax avoidance or profit shifting, but to look at it that way seems back-to-front: afterall what we want multinationals for in the first place is to enable international flow of technology, know-how and capital which raises productivity of the local assets & activities.

      Ultimately of course, corporations don’t really pay tax – one way or another the money has to come out of the pocket of a human being – be it an investor, worker or consumer. There are good reasons to tax corporations (because they are big and easy to find, and arguably as a backstop to prevent people incorporating to avoid tax), but it seems to me that part of why the international tax debate has become so intractable is that we have loaded the ‘hard question’ which you outline, with impossible expectations.

      On one hand we want the international tax system to be progressive and tax the rich investors, and on the other hand we want it to generate these tax revenues in the countries where the customers or employees are. But we want it to do this without taking money out of the pockets of customers or employees (i.e. by raising the cost of capital, driving out investment). There has been, I think, a lot of wishful thinking about the extent to which it is possible to do this.

      I think there are three separate questions in what you outline:

      Firstly – the question about ‘where to tax’ in relation to the extractive industry can be separated out, here you are mainly trying to tax economic rents on natural resources – and it is clear that these revenues should be local.

      Secondly – is your hard question about where to tax entrepreneurial profits. Tim Worstall’s answer has a logical simplicity, but feels inequitable internationally. I think for other the reasons, and for corporations to retain their license to operate, there does need to be some division of corporate tax between countries (but this cannot be an exercise in wishing-for-the-impossible).

      Thirdly is the question of tax havens. It seems like a lot (most?) of this is being driven by the US’s worldwide tax system, with taxes deferred rather than avoided altogether. But the problem often seems to loom larger than it really is. Again there seems to be a lot of wishful thinking here that the US won’t mind if Europe starts taxing what US shareholders and taxpayers believe is rightfully theirs. Perceptions that there is a lot more to play for here in terms of profits-untaxed-anywhere I think clouds clear debate over the second question, and  I don’t think transparency solves this.

      Have some thoughts on ways forward, but this reply is long enough already!

    3. For some reason the Google story has created a huge amount of popular interest in something that academics, tax professionals and NGOs have been wrestling with for years.

      I’m glad you began by repeating a fundamental that many overlook, ie what we mean by profits. Even in a domestic context, allocations of costs and profits is not necessarily straightforward. No customers, no sales; no product, no customers; no manufacturing, no product. Mix all that into an international context and we should perhaps expect no easy answers?

      But not only do we need to look at revenues and costs, we also need to look at timing and recognition, as well as the impacts of international treaties dating back to the League of Nations. The infamous Permanent Establishment is possibly the clearest example of how what worked in 1929 does not work now. So I think we do have to accept the work under BEPS to remedy some of that.

      I believe everyone agrees with Maya on extractives. But how that fits with taxes on profits, or even Formulary Apportionment seems conceptually tricky.

      I think the evidence is that CT does stick at stages in the chain. The share is not agreed but from an enforcement aspect, ending a Tax on companies seems to me to offer huge avoidance potential. It is also highly efficient and allows Govts to use tax policy in pursuit of wider goals. End CT and all that is trickier.

      If the US had a lower tax rate, would US companies remit more profit home? I’m not sure if we have the evidence for that. Deferral seems to have gone on for decades and surely  Tax Havens now are host to $tn, on a permanent basis? Or will there be another US amnesty?

      Too many other aspects to comment on but others may take it on.

    4. Pingback: Links I Liked | From Poverty to Power

    5. Thank you for your post Owen. I am curious about what has been accomplished on this matter since 2016. Where profits are taxed has important consequences on where wealth accumulates. This is also especially important in an exceedingly globalized world. It also means that a company must know where everything happens in their value chain, which isn’t always the case. Additionally, where the tax is placed has to do with incentives. For example, taxes on successful businesses are often very valuable for jurisdictions, so they attempt to attract these businesses. However, companies want fewer taxes, so they go to the place where they have to pay less taxes. This creates a balancing act for where companies choose to locate. It seems that we should have figured this issue out already, and I have always assumed that companies are taxed just according to the total dollar amount in profit that they make, and in the country they are registered. But obviously it is not that simple.

Leave a Reply

Your email address will not be published. Required fields are marked *