The new global corporate tax principle: ‘value creation’
In 2013, after a period of growing public concern over the apparent scale of corporate tax abuse, a new international corporate tax norm emerged as if from nowhere. The global corporate tax base, at least insofar as it takes the form of profits arising to multinational enterprises (MNEs) was to be allocated between jurisdictions in accordance with where ‘value’ is ‘created’. This new norm of ‘value creation’ was expressly promulgated in terms in a G20 announcement, and adopted by the OECD as the guiding principle for its project to reform the international tax system: the ‘Base Erosion and Profit Shifting’ (BEPS) project. But at no stage was it explained what ‘value creation’ actually meant in this context.
Transfer pricing and the inadequacy of marginalism
An informal consensus emerged in due course to the effect that ‘value creation’ was nothing other than a synonym of ‘economic substance’, but this idea was not grounded in any formal theory of value. And nor could it have been, if prevailing economic orthodoxies are to be maintained. Modern mainstream ‘marginalist’ value theory is, in essence, a theory of how prices come into being in idealised markets. It doesn’t really have anything to say about ‘economic substance’.
As things stand, the core principle which serves to allocate between jurisdictions the corporate tax base insofar as it arises to multinational enterprises is the ‘arm’s length principle’. The arm’s length principle provides that the pricing of an intra-group transaction (‘transfer pricing’) should correspond to the pricing at which the transaction would take place between independent enterprises. In other words, the outcome of the mechanism is already meant to be the same as the outcome that marginalist value theory would deliver: essentially, a market price. And that outcome is the one that, in practice, yields huge pools of untaxed offshore profitability in the hands of MNEs. To look to modern mainstream value theory as a guiding principle underpinning the concept of ‘economic substance’, in a context where the arm’s length principle is already being applied, would therefore simply beg the question.
By way of a simplified illustration, to show how this plays out in practice, suppose an MNE avoids tax by situating its intellectual property in a subsidiary in a tax haven. This means that all its subsidiaries in countries where it does substantive business pay tax-deductible royalties to their tax haven subsidiary, and tax is thereby avoided. If the royalties are excessively high then transfer pricing regimes will bring them down, but marginalist value theory is not going to bring them down any further; in principle it should bring them down to more or less the same point. If we are confident that a shell company acting as a formal conduit for intellectual property revenues does not display ‘economic substance’, then no doubt we should reform global corporate tax norms so as to allocate the tax base away from the tax haven, but we won’t know where to allocate the tax base to instead, if we only know where ‘economic substance’ is absent, and not where ‘value creation’ is present.
To make matters worse, in the context of web-based giants such as Facebook, Amazon, and Google, the impulse to allocate profits to jurisdictions in proportions broadly commensurate with volumes of sales or numbers of users in those jurisdictions has blown the question of what ‘value creation’ means wide open. It has been modish in some technocratic circles to go beyond talking about, say, network effects, and assert that value itself is created by consumers. This proposition, which has even been advanced by the OECD in some of its output on the topic of corporate tax reform, is inferentially an outright rejection of marginalism. The activities which are said to ‘create value’ in this context – interactions on Facebook; leaving reviews on Amazon – are simply not the market interactions yielding a price between the participants that marginalism exists to describe.
The classical alternative
So absent any answers from mainstream marginalism, what does value theory in the classical tradition of Adam Smith, David Ricardo and Karl Marx have to say about this? The answer is … well it isn’t entirely clear. Ultimately, this absence of clarity in the classical tradition as it stands today stems from some simplifying modelling assumptions seemingly made in the manuscript of Volume III of Marx’s Capital, which gave rise to a problem within Marxian economics (arguably a defect in Marx’s model, but perhaps only a problem of interpretation for Marxists) known as the transformation problem. Solutions to the transformation problem abound, but the most parsimonious and mathematically satisfactory solution, which was developed by 20th century economist Piero Sraffa and his ‘neo-Ricardian’ school of followers, involves simply abandoning the core classical axiom that value derives from labour. This development led to a crisis in classical value theory over the course of the 1970s, which could have had a really interesting outcome, except that the subsequent neoclassical takeover in academic economics meant that both sides of the debate lost by dint of finding themselves to be out of a job.
Given the equation-heavy nature of the debate and the political importance of the labour theory of value, it is not surprising that the Marxists who still had jobs in the aftermath of this crisis – sociologists, philosophers, literary critics &c – subsequently cleaved towards heterodox readings of Marx for which the specifically quantitative content of the value theory was either irrelevant (the ‘value-form’ approach) or positively and expressly repudiated (the ‘postoperaist’ reading). So where does that leave a participant in the debate over international corporate tax norms who, perceiving marginalism to be useless in this context, turns to classical theory instead to find out where ‘value’ is ‘created’? Sraffa and his followers cannot help them since, as mentioned above, their value-theoretical approach is agnostic as to the source of value, which means Marx is the only game in town. So here is a brief guide to the various schools of Marxian value theory generally encountered today, and what they would say about the distribution of the global corporate tax base if they were adopted as the theoretical basis for the OECD’s work in this area:
A corporate tax reformer’s guide to Marxist value theory
(1) Traditional Marxism treats value as something which is produced by labour in production and which is subsequently embodied in commodities at exchange, with the consequence that vast amounts of activity which appear to be implicated in profitability – design and advertising, for example – are not treated as value creating in this framework. Marx used a vivid analogy to explain the role of certain forms of labour that appear to be implicated in profitability but are not value creating – that of a match lighting a fire. The fire’s heat is caused by the match but does not come from it, likewise the profitability in commodities may be caused by design and marketing but the value comes from those who physically manufacture and transport them. On this view, an MNE which produces commodities should see its profitability allocated for tax purposes to the jurisdictions where it has physical plant and machinery (and perhaps also, depending on the economic geography, further upstream in global value chains to the jurisdictions where its suppliers have plant and machinery). And in the case of MNEs which do not produce material commodities at all, profitability should all be allocated upstream in the global value chains to which is attached. So, for example, if you buy a dress that was advertised to you on Facebook, Facebook’s profitability in respect of the advertising fees should be allocated for tax purposes to where the dress was made (and the cotton produced, and the logistics effected &c), even though none of this took place within the Facebook group. On the face of it this would require ripping up the global corporate tax rulebook altogether, because Facebook would not necessarily have a taxable presence in the upstream jurisdictions in question – but that is equally what is involved in existing proposals to allocate profitability downstream in global value chains to where users and consumers are situated. In practice what would be required in order to allocate the tax base upstream would be states taxing MNEs where they make money, and then making fiscal transfers to states where production takes place.
(2) ‘Value-form’ Marxism takes a more philosophical approach, essentially treating the problems with which mainstream Marxism concerns itself as originating in a mistaken reading of Marx as an economist rather than as a critic of political economy. This approach is characterised by a concern with Marx’s debt to Hegel, and a foregrounding of Marx’s more abstract discussions of value such as those to be found in volume I of Capital. It began to develop in the 1960s, evolving from a renewed focus on Marx’s own writings (including manuscript sources) as opposed to the received texts that come to us via, in particular, Engels. For present purposes the key feature of this school is its focus (which it shares with marginalism) on the moment of exchange as the point at which value is brought into being. Labour still has a central role but the distinctions drawn in traditional Marxism between value-producing and non-value-producing labour fall away: any kind of wage labour is in some sense value producing provided there is a monetary exchange at the end of it. So this school of thought (though it has hugely important insights to offer in other contexts!) is as hopeless for present purposes as marginalism: like marginalism it replicates the problem of transfer pricing, leaving the offshore pools of untaxed corporate profitability in place.
(3) ‘Postoperaism’ is a more recent development. It originated in Italian activist circles in the latter part of the twentieth century and it is highly influential outside the academy, underpinning much popular critique of capitalism from the turn of the twenty-first century onwards. It foregrounds Marx’s earlier writings (particularly Grundrisse) and its focus is on features of actually existing capitalism which are said to anticipate a post-capitalist future in which value is abolished. Like value-form Marxism it departs from the idea found in traditional Marxism to the effect that the utility or desirability of things is a mere trigger for the value-crystallising moment of exchange but, in contrast to value-form Marxism, it drops the requirement that the salient labour for bringing about that utility or desirability is necessarily (from a value-theoretical perspective at least) wage labour. It views value as being produced by all of us, insofar as we contribute (through our participation in it) to the social and cultural environment of actually existing capitalism. To that extent therefore it is intriguingly similar to the aforementioned modish theories to the effect that users in the digital economy create value. Quantitatively, postoperaism has nothing to say about where value is created: it holds that, owing to the infinite replicability of the immaterial, value can no longer be treated as a quantitative relation. But its focus on the participation of consumers in value creation suggests that, if treated as the basis for corporate tax reform, it would align with allocating profitability downstream in global value chains to where consumption takes place. Which from a quantitative perspective would allocate profitability predominantly to the already wealthy jurisdictions of the Global North.
(4) Marxist feminism departs from mainstream Marxism in order to accommodate the phenomenon of unpaid labour in the sphere of social reproduction. Over the course of the 1970s there was a debate as to the role played (or not played) in the Marxist value-theoretical schema by such labour, and the conclusion is generally understood to be that it is not necessary to intervene in the value-theoretical core of the Marxist critique of capitalism in order to foreground the reproductive sphere. Nonetheless, since unpaid reproductive labour relieves capital of costs which might otherwise be borne by it, perhaps through higher wages but more likely through taxes and onward welfare provision by the state, the risks and burdens of unpaid reproductive labour are quantitatively implicated in capitalist surplus. Given the uneven geographical distribution of the fiscal resources that might relieve those risks and burdens, this would probably mean a degree of alignment between a Marxist feminist view and the traditional Marxist position as regards where profitability should be allocated to for tax purposes i.e. upstream in global value chains to the jurisdictions where material production takes place; predominantly, in other words, the less wealthy jurisdictions of the Global South.
Conclusion: value theory matters
These brief generalisations are just that, but the underlying point is a serious one: whether we like it or not, our apparent value-theoretical commitments involve taking positions on real world topics with world-spanning distributional significance. The offshore profits of the digital economy are vast reserves of wealth waiting for redistribution, and from the point of view of relieving global inequality the difference between leaving them where they are (as both marginalism and value-form Marxism might suggest), distributing them to jurisdictions where consumption takes place (as postoperaism might suggest), and distributing them to jurisdictions where material production takes place (as both traditional and feminist Marxisms might suggest), is commensurately vast.
While you are here…
If you enjoyed reading this post – would you consider supporting our work? Just click here. Thanks!
All the best, Max Steinbeis