Externalities in international tax enforcement: Theory and evidence

Thomas Tørsløv, Ludvig Wier, Gabriel Zucman 21 April 2020

a

A

Multinational firms can avoid taxes by shifting profits from high-tax countries to low-tax countries. A number of studies suggest that this profit shifting causes substantial losses of tax revenue (Bolwijn et al. 2018, Clausing 2020, Crivelli et al. 2015, Tørsløv et al. 2018). In principle, tax authorities in high-tax countries can attempt to reduce profit shifting by increasing the monitoring of intra-group transactions and enforcing more strongly the rules governing the pricing of these transactions. Moreover, many countries have general anti-avoidance provisions, according to which transactions that are undertaken with the primary goal of avoiding taxes are illegal. This legal framework is a potentially powerful tool to curb profit shifting to sparsely populated, zero-tax countries such as Bermuda, where little real activity takes place. Why, despite the sizable revenue losses involved, does profit shifting nonetheless persist? 

Our new paper (Tørsløv et al. 2020) provides a novel answer to this question by studying the incentives faced by tax authorities. We show theoretically that the fiscal authorities of high-tax countries can lack the incentives to combat profit shifting to tax havens. Instead, they have incentives to focus their enforcement efforts on relocating profits booked by multinationals in other high-tax countries. This crowds out the enforcement on transactions that shift profits to tax havens and does not increase the global tax payments of multinational companies. The incentive problem can help explain why profit shifting to low-tax countries persists, even when the legal framework (such as general anti-avoidance provisions) to curb it exists. 

The peculiarities of international tax law

To ensure profits are taxed according to the prevailing internationally agreed rules, tax authorities in high-tax countries routinely audit companies. This work is usually carried out by dedicated transfer pricing units operating inside tax authorities. Concretely, transfer pricing units can ask for transfer pricing documentation, in the form of detailed reports prepared by firms to justify their internal transactions. These reports are usually long and hence time-consuming to audit. With limited resources available, the tax authorities must prioritise which companies are asked to deliver transfer pricing documentation. 

Unlike other forms of tax enforcement, the enforcement of transfer prices is asymmetrical – it can only raise tax revenue. In the context of individual income tax audits, auditors investigate both over- and under-payment of taxes and hence audits can result in a reduction in taxes paid. This is not the case when it comes to transfer price enforcement, since each country tries to increase its own tax base. France, for example, audits French companies with the aim to correct transactions that are disadvantageous to France but ignores any findings that would result in a lowering of the French tax base. The same goes for other countries. 

As stated by the OECD (2010: 2), “transfer pricing is not an exact science.” In some cases (e.g. service payments such as royalties, purchases of intellectual property such as brands) the correct arm’s-length price is not conceptually clear (Devereux and Vella 2017). As a result, there is uncertainty involved in setting arm’s-length prices, which implies that firms will at times be at odds with tax authorities even when they do not voluntarily engage in tax-minimisation behaviour. 

The incentives facing a tax authority

Current tax law incentivises high-tax countries to focus their efforts on other high-tax countries. To understand the logic of the argument, take the case of Denmark, a country where the corporate tax rate is 22%. One euro of profit relocated to Denmark by the Danish tax authority is worth the same to Denmark whether it comes from Sweden, where the corporate tax rate is also 22%, or from Bermuda, where the corporate tax rate is 0%. That is, the Danish tax authority does not internalise the externality of reducing the corporate tax base in the partner country. It may, on the other hand, be easier for the Danish tax authority to relocate one euro booked by a multinational company in Sweden, for two reasons. First, it is more likely to succeed, because firms are unlikely to spend much resources opposing this transfer price correction. For them, whether profits are booked in Denmark or Sweden makes little difference to their global tax bill, since the tax rates in Denmark and Sweden are the same. Second, if there is a dispute between Denmark and Sweden, it is likely to be settled relatively quickly through the dispute resolution agreements in force among OECD countries and EU countries. The correction of transactions between Denmark and Sweden crowds out the correction of transactions between Denmark and low-tax countries. Such corrections are harder to make as firms spend more legal resources to defend their transfer pricing optimisation and take more time (due to a lack of cooperation with some tax havens). In this paper, we formalise this argument and make precise the conditions under which it is optimal for high-tax countries to focus their enforcement resources on relocating profits booked in other high-tax countries.

Novel datasets on international tax enforcement

Our theoretical predictions are motivated and supported by novel datasets on international tax enforcement analysed for the first time in this paper. We first analyse the universe of transfer price corrections initiated by the Danish tax authority—confidential micro-data internal to the Danish administration to which we were granted access in the context of this research. We find that the vast majority of transfer price corrections (about 82%) initiated by the Danish tax authorities involve other high-tax countries. As Denmark has a moderate corporate tax rate (22% in 2015), this finding implies that the majority of transfer price corrections initiated by Denmark involve countries with higher rates and ultimately lower the taxes paid by the targeted multinationals globally. According to our estimates, the combined transfer price enforcement efforts of the Danish authorities increased the Danish tax collection by €315 million per year on average over the years 2008, 2009 and 2015, but lowered tax collection abroad by €333 million. This result is at odds with the popular perception that the enforcement activities of tax authorities increase the taxes paid by multinationals. We furthermore go through Danish court cases and find that the likelihood of multinationals appealing a transfer price correction is 50% higher whenever the transfer price correction involves a tax haven — consistent with the model prediction of multinationals putting up a fight whenever a correction disfavours their global tax bill. 

Second, we analyse data on tax disputes between tax authorities by drawing on a survey of tax authorities conducted by the audit firm EY, in which EY asked 26 major economies which countries were the main focus of their transfer price correction efforts. Consistent with our theory, the data show that the majority of high-tax countries’ enforcement efforts are directed at other high-tax countries. These corrections typically do not increase the taxes paid by multinationals, but merely reshuffle tax payments across high-tax places. From a global perspective, such corrections are welfare decreasing, since they consume resources without changing global tax payments. In effect, non-haven countries steal revenue from each other while letting tax havens flourish.  

Potential adverse effects of international efforts

Our findings have policy implications. There is an ongoing international effort to facilitate dispute resolution settlement in transfer pricing cases between high-tax countries. Our results highlight a previously overlooked adverse effect of facilitating dispute resolution: Making it easier for high-tax countries to correct each other incentivises more transfer price corrections between high-tax countries and as a result crowds out tax haven cases. The easier it is for, say, the French tax authority to relocate profits booked in Germany, the less resources it may devote to chasing the profits shifted to Bermuda—potentially increasing shifting to low-tax locales. Similarly, current efforts to implement minimum taxation at a country-by-country basis, e.g. through controlled foreign corporation (CFC) rules could further impact the incentives facing tax-authorities. Under current proposals of country-by-country minimum taxation, income declared in a tax haven may in part be subject to taxation in a high-tax country, while income declared in another high-tax country will not be. Thus, going after intentional profit shifting to tax havens may produce less tax revenue to the tax authority than going after income currently declared in a high-tax country. This could exacerbate the existing incentive for tax authorities not to go after profits booked in tax havens and thus — in turn — exacerbate profit shifting. This result changes if the country-by-country minimum tax is sufficiently high, such that the incentive for firms to shift profits disappears. 

Our results highlight an overlooked inefficiency in the current international tax system. The uncertainties involved in determining arm’s-length prices are large enough that they can soak up substantial enforcement resources, even though the related enforcement initiatives do not increase global corporate tax payments. Some argue that a fundamental corporate tax reform is needed to curb profit shifting to tax havens. A number of proposals suggest abandoning the idea of transfer pricing entirely, and using instead apportionment formulas or a destination-based cash-flow tax. This paper highlights another reason to pursue such reform — it may save public and corporate resources that currently go to wasteful and inconsequential tax enforcement.  

References

Bolwijn, R, B Casella and D Rigo (2018), “An FDI-driven approach to measuring the scale and economic impact of BEPS”, Transnational Corporations 25(2). 

Clausing, K A (2020), “Profit Shifting Before and After the Tax Cuts and Jobs Act”, Available at SSRN.

Crivelli, E, R de Mooij and M Keen (2015), “Base Erosion, Profit Shifting and Developing Countries”, IMF working paper 15/118. 

Devereux, M P and J Vella (2017), “Implications for digitalization for international corporate tax reform”, Oxford University Center for Business Taxation Working Paper 17/07.

OECD (2010), OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2010, OECD Publishing.

Tørsløv, T, L Wier and G Zucman (2020), “Externalities in International Tax Enforcement: Theory and Evidence”, NBER Working Paper No. w26899.

Tørsløv, T, L Wier and G Zucman (2018), “The Missing Profits of Nations”, NBER Working Paper No. 24701, 2016 estimates available online at missingprofits.world.

a

A

Topics:  Global governance Taxation

Tags:  incentives, tax enforcement, tax havens, multinationals, corporate taxation

Previously PhD student at the University of Copenhagen, now economist at the Ministry of Taxation, Denmark.

Postdoctoral researcher & lecturer, UC Berkeley

Professor of Economics, UC Berkeley

Events

CEPR Policy Research