Jump to ContentJump to Main Navigation
The Hypocritical HegemonHow the United States Shapes Global Rules against Tax Evasion and Avoidance$

Lukas Hakelberg

Print publication date: 2020

Print ISBN-13: 9781501748011

Published to Cornell Scholarship Online: September 2020

DOI: 10.7591/cornell/9781501748011.001.0001

Show Summary Details
Page of

PRINTED FROM Cornell University Press SCHOLARSHIP ONLINE (www.cornell.universitypressscholarship.com). (c) Copyright University of Cornell University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in Cornell for personal use. date: 21 September 2021

The BEPS Project

The BEPS Project

Long Live Arm’s Length

(p.106) 6 The BEPS Project
The Hypocritical Hegemon

Lukas Hakelberg

Cornell University Press

Abstract and Keywords

This chapter reveals the US government's struggles in assuming its usual leadership position in the Organisation for Economic Co-operation and Development (OECD) initiatives. This happened for entirely domestic reasons: the Obama administration's inability to implement its preferred solution to BEPS (base erosion and profit shifting)—a tightening of controlled foreign company (CFC) rules—in the face of opposition by US multinationals, paired with the administration's strong political commitment to tax fairness, which prevented the administration from abandoning the initiative altogether. The administration's lack of purpose initially opened agenda space for other governments. Between the release of a first set of discussion drafts and the final BEPS reports, however, the United States fought a successful rearguard battle, retrenching attempts at expanding the taxing rights of source countries and essentially preserving the status quo. This success occurred despite the inclusion of the Group of 20 (G20) emerging economies, which could be expected to shift the power balance away from the United States, and in accordance with the preferences of US multinationals. The diffusion of unilateral initiatives by source countries, which are still subject to political conflict, confirms their frustration with the outcome of the BEPS project.

Keywords:   BEPS project, Obama administration, US multinationals, tax fairness, BEPS reports, taxing rights, source countries, base erosion, profit shifting

Numerous commentators suggest that the base erosion and profit shifting (BEPS) project, which was launched by the Organisation for Economic Cooperation and Development (OECD) in 2013 in response to a request from the Group of 20 (G20), was the most far-reaching attempt at rewriting international tax rules since the establishment of the international tax system in the 1920s. Yet commentators’ superlatives refer much more to the project’s ambition than to its final outcome. Scholars acknowledge the OECD’s ability to produce new rules and recommendations on virtually all aspects of international taxation within just two years, particularly since discussions on individual rules had regularly taken up to a decade before the BEPS project (Ault 2013; Grinberg 2015). Scholars are less sure about what impact the final reports on fifteen action items—ranging from transfer pricing and the definition of permanent establishments, to new requirements for country-by-country reporting (CbCR)—will have on the way multinational firms are currently being taxed. In fact, the BEPS recommendations largely preserve the cornerstones of the international tax system, including the arm’s-length standard (ALS), separate entity accounting, and the benefits principle (Büttner and Thiemann 2017; Picciotto 2015). Still, some scholars suggest that certain elements in the BEPS reports—like CbCR—represent significant steps in the direction of unitary taxation and formulary apportionment (UT+FA) (Avi-Yonah and Xu 2016; Seabrooke and Wigan 2016). Accordingly, the buzzword informing the current academic debate on BEPS has been “creative ambiguity” (Büttner and Thiemann 2017; Grinberg 2015; Picciotto 2015).

(p.107) Why the authors of the BEPS reports had to resort to ambiguous and sometimes contradictory language to accommodate diverging interests also remains subject to debate. Most analysts suggest that the increased political salience of international tax, resulting from tax avoidance scandals in the aftermath of the financial crisis, put the established expert consensus under pressure. Whereas some argue that this pressure enabled nongovernmental organizations (NGOs) to insert alternative expertise into the discussion (Seabrooke and Wigan 2016), others claim that pressure from their political masters caused national regulators to emphasize national interest instead of technical soundness during negotiations (Grinberg 2015). As a result, conflicts over the allocation of taxing rights emerged between source and residence countries, pitting several large EU members and the emerging economies in the G20 against the United States (Grinberg 2016a).

Against this background, this chapter will clarify that the US government struggled, indeed, to assume its usual leadership position in OECD initiatives. Yet this happened for entirely domestic reasons: the Obama administration’s inability to implement its preferred solution to BEPS—a tightening of controlled foreign company (CFC) rules—in the face of opposition by US multinationals, paired with the administration’s strong political commitment to tax fairness, which prevented the administration from abandoning the initiative altogether. The administration’s lack of purpose initially opened agenda space for other governments. Between the release of a first set of discussion drafts and the final BEPS reports, however, the United States fought a successful rearguard battle, retrenching attempts at expanding the taxing rights of source countries and essentially preserving the status quo. This success occurred despite the inclusion of G20 emerging economies, which could be expected to shift the power balance away from the United States, and in accordance with the preferences of US multinationals. The diffusion of unilateral initiatives by source countries, which are still subject to political conflict, confirms their frustration with the outcome of the BEPS project.

Points of Departure: Limiting Taxation at Source Through Transfer Pricing

Whereas the US government has enforced international cooperation against tax evasion, the United States has not followed through with proposed domestic and international measures against tax avoidance. The reason is opposition from US multinationals, defending their tax-planning practices, and an underlying dilemma faced by developed countries organized in the OECD. In general, these countries host the headquarters and intellectual property (IP) of multinational (p.108) corporations. These countries are thus interested in an international tax system that emphasizes residence taxation and allows “their” multinationals to repatriate profits from emerging and developing countries where production and sales take place. To this end, developed countries have created OECD transfer pricing guidelines that link taxable profits to added value and added value to the location of IP. Based on these rules, the Chinese subsidiary producing and selling cars on behalf of a German manufacturer pays license fees to the parent company for the use of its IP. This payment reduces the taxable profit in China and increases it in Germany, as license fees are deemed passive income and as such taxable at residence. If the manufacturer manages to locate its IP in a corporate tax haven, however, the same rules also enable it to shift profits there instead of repatriating them. Developed countries can thus either choose to curb profit shifting and risk more source-based taxation or insist on residence-based taxation and risk more tax avoidance. In any case, these countries lose part of their tax base to foreign governments. In contrast to more tax avoidance, however, more source-based taxation would not only reduce the tax revenue of residence countries but also increase the effective tax burden on multinationals headquartered there. In order to reduce the foreign tax burden for their multinationals, OECD governments have thus generally given priority to limiting source-based taxation (Avi-Yonah 2000; Dharmapala 2014).1

With the advent of a digital economy dominated by US corporations and the consolidation of the common market, however, this OECD consensus was put into question. In fact, large EU member states grew increasingly concerned at the ability of US multinationals to channel profits out of the common market untaxed.2 With the complicity of several small EU member states, these companies had set up tax-planning schemes like the “Double Irish with a Dutch Sandwich” to minimize the taxable profits of their subsidiaries in large EU member states. These companies achieved this result through cost-sharing arrangements allowing them to transfer the rights to the foreign use of their IP from the United States to subsidiaries in Ireland, Luxembourg, or the Netherlands. These subsidiaries were granted special deals minimizing tax payments to the respective government and then started collecting license fees for the use of their parent company’s IP from their sister subsidiaries in the rest of the EU. These payments reduced taxable profits in large and high-tax member states and increased them in small and low-tax member states. Because of the loopholes in check-the-box rules discussed in chapter 3, these schemes also enabled US multinationals to avoid being taxed on their foreign profits in the United States (Avi-Yonah 2000; Dharmapala 2014; Pinkernell 2014).

As a result, US-owned coffee chains, book retailers, and computer firms enjoy a massive competitive advantage in the common market relative to their local (p.109) competitors, which lack access to the same tax-planning techniques. Large EU member states could not implement countermeasures through European cooperation, as the unanimity requirement in tax matters enabled the small capital-importing member states to block the passage of meaningful anti–tax avoidance directives in the Council of the European Union. Large EU member states were also unable to implement unilateral defense measures or issue credible sanction threats because common market legislation and jurisprudence from the European Court of Justice (ECJ) prevent member states from limiting market access for other member states. The ECJ’s Cadbury Schweppes ruling, for instance, bars large member states from applying CFC rules to subsidiaries of resident groups incorporated within the EU. As a result, European multinationals can shift profits to their subsidiaries in low-tax EU countries without having to fear that the tax authorities in their country of residence include these profits as deemed passive income in the headquarters’ tax base. As a result of ECJ jurisprudence and the common market legislation on which it is based, tax competition has thus been more intense inside the European Union than in the rest of the world (Genschel, Kemmerling, and Seils 2011; Hakelberg 2015b).

Setting the Agenda: Starbucks and the Inclusion of Emerging Economies

Against this background, tax experts and administrators in the EU have been looking for remedies to tax avoidance in the common market at least since the early 2000s. In 2001, the European Commission first presented its idea for a common consolidated corporate tax base (CCCTB) to the Council of Ministers. The concept foresees the consolidation of the earnings and losses reported by a group’s EU subsidiaries at its European headquarters. Instead of having each member state tax the profit reported by a group’s local subsidiary, the group’s consolidated profit is to be apportioned to member states based on local workforce, payroll, sales, and fixed assets. Member states can then apply their respective tax rates to their share of the consolidated profit. This application of unitary taxation and formulary apportionment (UT+FA) at the EU level should prevent multinationals from shifting profits from high-tax to low-tax member states, for instance, through license fee payments between sister subsidiaries. After all, the shifted profit would be included in the group’s consolidated result in whichever member state the subsidiary receiving the payment is located (European Commission 2001).

In parallel, corporate tax lawyers and officials in the EU engaged in an intense debate over the definition of Internet servers as permanent establishments (PE). (p.110) The contentious issue was whether transactions processed via a given server—for instance in the context of online shopping—could be taxed by the country in which the server is located (Pinkernell 2014). Given the redistributive consequences of these proposals, however, member states have since failed to agree on the CCCTB, whereas the United States—home to virtually all Internet giants—insisted on limiting source countries’ right to tax Internet transactions at the OECD.3 As a result of this deadlock, expert officials in large EU member states had a hard time raising interest for the issue of tax avoidance with their political masters. As a German tax official explained in 2015:

All of these issues have been discussed in the OECD’s tax committee for at least fifteen years. They never became more than printed paper. Not because they didn’t make sense but because there was no political backing. There was no tailwind. So how did it reach the agenda? I believe that politics is not really projectable but there are opportunities and time slots. When I first told the minister about what we had been discussing among experts, he replied that was a nice topic, but he wouldn’t fight a lonely battle against Google, Apple, or whoever. And that was it for the moment. That must have been around March/April 2012. And then—I still remember like it was today—just before the G20 summit in Los Cabos, in November 2012, George Osborne [then the UK’s minister of finance] expressed his outrage over tax avoidance by Starbucks. Suddenly our minister had this catchy example and my colleagues mailed me from Los Cabos, asking how one could integrate the tax avoidance issue into the final communiqué.4

Hence, the United Kingdom and Germany responded to public outrage over Starbucks’ tax avoidance in the common market by involving the G20 in the issue. Compared with the CCCTB, this seemed to be a feasible and system-preserving way of increasing the pressure on multinationals.

The British-German G20 initiative also resonated with the Obama administration. The US government was less concerned with US multinationals avoiding taxes in Europe but criticized US corporations for hoarding their foreign profits in tax havens to defer tax payments in the United States. As discussed in chapter 5, President Obama and Treasury Secretary Geithner had presented a strategy for “leveling the playing field for US taxpayers” shortly after entering office. This strategy foresaw the tightening of CFC legislation through reforms of check-the-box rules and other provisions enabling tax deferral. Just as the Clinton administration did (see chapter 3), however, Treasury faced massive opposition from multinationals against these proposals. Again, business argued that an amendment (p.111) of check-the-box rules would lead primarily to higher taxation of US corporations in EU countries, as a disjunction of their hybrid subsidiaries in Ireland, Luxembourg, or the Netherlands endangered the schemes set up to channel profits out of the common market. According to a tax expert with the congressional staff, many lawmakers were impressed by that argument, thinking “it [was] better for US companies to make money than for Europe to make money as a result of US tax reform.”5 So instead of repealing check-the-box rules, in 2010 the Democratic Congress extended the provisions that had turned them from regulation into legislation under George W. Bush. Even Carl Levin voted in favor (Drawbaugh and Sullivan 2013). As a result, the reform of check-the-box rules disappeared from the US Treasury’s Green Book of revenue proposals for fiscal year 2011 (cf. US Treasury 2010, 2009).

Likewise, reforms of deferral and foreign tax credit rules did not make it beyond consultation phase. In the Green Book for fiscal year 2010, Treasury had proposed to disallow the deduction of “expenses from overseas investments while deferring US tax on the income from the investment.” Moreover, the department sought to end corporations’ ability to receive foreign tax credits for expenses that are either artificially separated from foreign profits through a hybrid entity or based on investments in high-tax countries made only to shelter profits in low-tax countries from US taxation through “cross-crediting” (US Treasury 2009, 29–31). The measures were supposed to motivate US multinationals to repatriate their foreign profits and limit eligible credits against the corresponding tax bill. Unsurprisingly, however, business lobbyists rallied against the measures, arguing that the result of repatriation and limited credits—taxation of foreign profits at 35 percent—would put US multinationals at a competitive disadvantage relative to corporations from most other OECD countries, which exempted foreign profits from taxation (Javers 2009; Leone 2009; Montgomery and Wilson 2009). Senior Democratic tax writers in Congress heard their arguments. Max Baucus, chairman of the Senate Finance Committee, commented on the Obama-Geithner initiative, saying “further study is needed to assess the impact of this plan on US business” (Calmes and Andrews 2009). Richard Neal, chairman of the Subcommittee on Select Revenue Measures of the House Ways and Means Committee, told reporters he had personally lobbied the president to abandon the characterization of tax deferral on foreign profits as tax avoidance (Cohn 2009). As a result of the chairmen’s lack of interest in measures interfering with corporate tax planning, no corresponding legislation made it beyond their committees. The proposals for reforms of deferral and foreign tax credit rules were thus still included in Treasury’s Green Book of revenue proposals when in 2013 Democrats also lost their Senate majority (US Treasury 2013).

(p.112) Although the United States and the EU G5 were thus looking at the issue from quite different angles, tax avoidance remained a problem for both, given the increased public awareness and their inability to implement countermeasures domestically. Accordingly, the G20 leaders declared in Los Cabos, “we reiterate the need to prevent base erosion and profit shifting and we will follow with attention the ongoing work of the OECD in this area” (G20 Leaders 2012). In response, the organization presented a report and an action plan on BEPS at the beginning of 2013. The documents summarized the OECD’s past efforts against tax avoidance, identified fifteen pressure areas (actions) in which reforms were needed, and proposed the BEPS project as an “effective and inclusive process” for their elaboration (OECD 2013a, 26; 2013b). As part of the OECD’s inclusiveness agenda, the organization proposed to integrate non-OECD G20 members into its tax policy committees during the project. Along with the entire BEPS action plan, G20 leaders endorsed this suggestion at their 2013 summit in St. Petersburg (G20 Leaders 2013). From the perspective of the OECD, opening deliberations to emerging economies should prevent the emergence of an alternative venue and thus secure the organization’s position as the central forum for decisions on international tax policy. As a German tax official explained in an article:

The BEPS project has strengthened the OECD’s leading role in international tax policy. From the German perspective, this is a strategic success, since principles developed by the OECD tend to reflect the interests of an industrialized country like Germany. These standards will evolve to take the interests of emerging and developing countries into account. But at the same time, they provide a chance for continued unification of international tax standards, which is in the particular interest of Germany with its globally connected economy.

(Fehling 2015, 822)6

Also, the German representative in the OECD’s fiscal affairs committee expected that “the inclusion of all G20 members in the discussion as opposed to a pure OECD discussion [would lead] to a stronger regard for the interests of source countries” (Kreienbaum 2014, 637). Along with the US government, which wanted to strengthen residence taxation through tighter CFC rules, bargaining over BEPS thus involved two country groups with at least partial preferences for increased source taxation. Large EU member states wanted to prevent US multinationals from channeling profits out of the common market untaxed, but still defended the arm’s-length standard as the international tax system’s underlying principle. Emerging economies participating as observers without voting rights aimed at a more fundamental redistribution of taxing rights toward source countries (Piltz 2015).

(p.113) Toward New Rules? International Bargaining over BEPS

The Obama Administration’s Lack of Purpose

When entering negotiations over BEPS, the Obama administration faced a dilemma. The administration’s aim was to use the project to finally pressure business and legislators at home into the adoption of tighter CFC rules. Robert Stack, then the Treasury’s international tax counsel, urged US multinationals to end the deferral of tax payments on foreign profits to prevent source countries from claiming a larger share of this supposedly “stateless income” (Stewart 2014). By enhancing Treasury’s ability to tax the passive income US multinationals were hoarding in corporate tax havens, the United States would have enforced the residence principle and strengthened its dysfunctional worldwide taxation system (Grinberg 2015; Stack 2015). Yet US multinationals still had no interest in paying tax on the billions of foreign profits hitherto stashed away offshore. Instead, they advocated for a repatriation tax holiday, providing for a tax-free return of foreign profits to the United States, and a switch to a territorial tax system, exempting future foreign profits from US taxes (National Foreign Trade Council 2015; Silicon Valley Tax Directors Group 2015).7 The Republican chairmen of the Senate Finance and the House Ways and Means Committee, who were the main targets of business lobbying, soon adopted this approach. Accordingly, Congress continued to oppose government proposals amounting to a repair of the worldwide system (Camp 2015; Camp and Hatch 2014). As the Obama administration was unable to get its preferred approach through Congress, the administration could not go first and forge international consensus around its domestic regulatory model. In the absence of a US template for an emerging international standard— as FATCA had been for the multilateral AEI (see chapter 5)—other countries used the opportunity to fill the agenda space (Grinberg 2015; Herzfeld 2015a).

For large EU member states, strengthening CFC rules made little sense for two reasons. First, the ECJ’s Cadbury Schweppes ruling prevented them from applying such rules to subsidiaries in low-tax countries inside the EU. Hence, CFC rules could not prevent resident multinationals from shifting profits to Ireland or Luxembourg (ECJ 2006; Ruf and Weichenrieder 2013). Second, the main concern of large EU member states was the ability of US multinationals to channel profits out of the common market untaxed. Yet stronger CFC rules merely enable the residence country—in this case the United States—to include profits booked in corporate tax havens in a resident company’s tax base. Stronger CFC rules do not enable EU countries to tax nonresident multinationals. Therefore, large EU member states chose a different and somewhat contradictory approach to the BEPS project. While these countries remained committed to the OECD’s existing (p.114) international tax system based on separate entity accounting and the arm’s-length standard, they wanted to tweak the rules so as to allow tax examiners greater discretion in the assessment of intra-firm transactions and the identification of PE status. These measures were supposed to give European tax authorities a stake in the taxation of US multinationals and send a signal to emerging economies that their source country interests could be taken into account within the international tax system developed by the OECD (Grinberg 2015; Kreienbaum 2014).

As nonmembers, emerging economies did not always feel bound by the organization’s standards when determining the taxable income of local subsidiaries of European firms. As this possibility could lead to unexpected increases in the firms’ tax burden, the second goal of large EU member states in the BEPS project was to improve the acceptance of OECD rules among emerging economies. A senior German tax official described this reasoning as follows:

For instance, we don’t have a double-tax agreement with Brazil. Nothing but trouble, because Brazil says, “We make the rules and profit splits as we think is right and if you don’t want that you don’t do business in Brazil.” Then the German firm tells us, “The Brazilians deceived us, now you have to do the offset so that we’re not taxed twice.” And when we reply that wouldn’t be in line with OECD principles, they tell us that Brazil isn’t even an OECD member. And that is correct, too. … ​So if I want the OECD to set global standards, I need to convince nonmembers that the rules developed there are also good for emerging economies. And I believe integrating China in the OECD process is better than China developing its own standards. But that means that the club of industrialized countries needs to depart to a certain degree from its usual reasoning. You will not convince emerging economies by saying “this is the rule, take it or leave it.”8

In many respects, the initial framing of the BEPS project corresponded to the European priorities. In its first BEPS report, the OECD announced that “the main purpose of [the BEPS action] plan would be to provide countries with instruments, domestic and international, aiming at better aligning rights to tax with real economic activity” (OECD 2013b, 8). At their St. Petersburg summit, G20 leaders confirmed this objective, decreeing that “existing international tax rules on tax treaties, permanent establishment, and transfer pricing will be examined to ensure that profits are taxed where economic activities occur and value is created” (G20 Leaders 2013, 4). Although the emphasis on the link between taxation and real economic activity acknowledged the interests of source countries, the OECD excluded a fundamental switch to UT+FA from the outset. In the BEPS action plan, the organization underlined “consensus among governments that (p.115) moving to a system of formulary apportionment of profits is not a viable way forward,” and argued that “it [was] also unclear that the behavioural changes companies might adopt in response to the use of a formula would lead to investment decisions that are more efficient and tax neutral than under the separate entity approach” (OECD 2013a, 14). Given the short time frame and the familiarity of all involved actors with the existing system, solutions to BEPS should rather be found within the bounds of established principles (Ault 2013).9 As the OECD explained in its 2013 action plan:

The importance of concerted action and the practical difficulties associated with agreeing to and implementing the details of a new system consistently across all countries mean that, rather than seeking to replace the current transfer pricing system, the best course is to directly address the flaws in the current system, in particular with respect to returns related to intangible assets, risk and over-capitalisation. Nevertheless, special measures, either within or beyond the arm’s length principle, may be required with respect to intangible assets, risk and over-capitalisation to address these flaws.

(OECD 2013a, 20)

Accordingly, amendments to the OECD’s transfer pricing guidelines summarized in actions 8 to 10 became the centerpiece of the BEPS process, at least when considering the submissions from business and civil society (cf. OECD 2017d).10

Defending Orthodox Application of the Arm’s-Length Standard

The OECD’s initial discussion draft released in December 2014 includes new guidance on how to apply the arm’s-length standard (ALS) in transfer pricing analyses. To determine the accuracy of transfer prices fixed by multinationals to value transactions between their branches, tax authorities first need to accurately delineate the transaction and then compare “the conditions of the controlled transaction with the conditions of comparable transactions between independent enterprises” (OECD 2014b, 4). The ALS is respected when the conditions and transfer prices of the controlled and uncontrolled transaction match. Yet, because of the tight integration of multinationals and the importance of intangible assets in their internal transactions, tax authorities often struggle to identify comparable transactions between unrelated firms. Therefore, the draft provides guidance on how to identify the functions and risks assumed by transacting group members based on their contracts and actual conduct. The thorough analysis of the facts and circumstances of every controlled transaction shall enable tax authorities to determine “whether actual arrangements differ from those which would (p.116) have been adopted by independent parties behaving in a commercially rational manner” (OECD 2014b, 26). A transfer price respecting the ALS should thus become identifiable also in the absence of a comparable transaction between independent firms.

Although the OECD’s draft mainly underlines the continued applicability of the ALS, the document also introduces some important qualifications: “In exceptional circumstances the transaction as accurately delineated may be interpreted as lacking the fundamental economic attributes of arrangements between unrelated parties” (OECD 2014b, 25). This is the case when an “arrangement does not enhance or protect the commercial or financial position of [involved branches],” or “the Group benefit is limited to post-tax considerations” (OECD 2014b, 25). According to the draft, such situations arise because multinationals tend to divide the activities, legal ownership, and assumption of risk related to an asset among an increasing number of separate legal entities. Because the transactions between these entities make no commercial sense in the absence of centralized control, tax authorities should be allowed to disregard them in transfer pricing analyses. The nonrecognition of a controlled transaction as described in the draft should thus enable tax authorities to attribute the income derived from the use of an asset to those group branches that actually developed or marketed it, instead of the entity formally holding ownership rights (OECD 2014b, 25–27). From the perspective of the OECD, “the non-recognition of transactions, which lack the fundamental attributes of arrangements between unrelated parties, will go far in aligning where profits are reported and where value is created” (OECD 2014b, 37). Still, the organization identified some residual BEPS risks related to information asymmetries between corporations and tax administrations.

In practice, a multinational may transfer patents or trademarks to a subsidiary in a corporate tax haven before they start to generate revenue. Since these intangibles are often unique and early in their development, tax authorities struggle to verify the assumptions the multinational made in projecting the income the patents or trademarks will generate. As a result, tax authorities cannot assess whether the transfer price the tax haven subsidiary paid for the intangible reflects the price an unrelated party would have paid or is at least economically rational. To make up for the authorities’ lack of reliable information, the draft thus proposes special measures tax examiners could apply when faced with excessive uncertainty over the value of an intangible transferred between related parties. One such measure relates to hard-to-value intangibles (HTVI) and would permit tax authorities to adjust ex ante income projections based on actual outcomes. That is, if an internally transferred patent turned out to be successful, the tax examiner could adjust the initial transfer price according to the profits eventually generated (OECD 2014b, 41). In most cases, this adjustment would result in a higher (p.117) transfer price, reduce the profits of the tax haven subsidiary, and increase the taxable profits of the multinational’s headquarters. Another measure introduces the behavior of an independent investor as a yardstick, assuming that lower risk and higher returns would lead her to invest directly in an asset instead of in a company that owns an asset but lacks the capacity to exploit it (OECD 2014b, 42). Such a minimally functional entity (MFE) should thus be disregarded for tax purposes and its profit reallocated to the parent company. To facilitate the identification of an MFE, the draft proposes a short list of easy-to-apply qualitative and quantitative indicators. Instead of performing a comparability analysis respecting the ALS, the tax examiner could simply disregard a controlled transaction based on certain attributes of the involved group branches (OECD 2014b, 44). Like the proposed guidance on nonrecognition, special measures could thus have provided tax authorities with additional discretion in the recharacterization of controlled transactions.

Accordingly, these proposals met considerable opposition from multinationals and their tax advisers. In their view, greater leeway for tax examiners increased uncertainty for taxpayers and the risk of disputes between tax authorities and corporations as well as among tax authorities of different countries. This was the case because the new guidance allowed tax examiners to replace objective contractual arrangements between related parties with subjective alternative views of the facts and circumstances of a controlled transaction. As tax authorities did not necessarily come to the same conclusions in their analyses, inconsistencies and double taxation were the likely result. In an exemplary submission to public consultations on the transfer pricing draft, the National Foreign Trade Council (NFTC), representing US multinationals like Caterpillar, eBay, Google, Microsoft, and Pfizer, expressed the following criticism:

The guidance on the identification of risk essentially mandates that business risks be allocated to the affiliate with functional control over the risk, without regard to the provisions of a written legal agreement or the observed behavior of parties acting at arm’s length. The guidance on non-recognition adopts a “commercial rationality” standard that would disregard transactions that have been actually undertaken based on a subjective determination by a tax authority that the transactions were not expected to enhance the commercial position of the parties. Taken together, these proposals would lead to tremendous uncertainty and a proliferation of disputes in all but the simplest fact patterns, are wholly disproportionate to the concerns identified, and, most importantly, are inconsistent with the arm’s length principle.

(NFTC 2015, 602)11

The big four accounting firm KPMG had essentially the same observations:

(p.118) The Discussion Draft actively encourages tax authorities to second-guess the contractual arrangements established by taxpayers by stating that they are “… ​at best …” the starting point in determining the “accurately delineated transaction.” This is recharacterization in substance, and will remove any common understanding of the relevant business arrangements. An increase in the number and size of disputes can be expected, as well as an increase in the difficulty in finding a principled solution to those disputes.

(KPMG 2015, 528)12

Therefore, multinationals and tax advisers alike requested tax authorities to give priority to written contracts between related parties in their analyses and resort to nonrecognition only when actual conduct deviates from prior legal arrangements. This approach would provide taxpayers with certainty as to their tax structures and minimize legal disputes (KPMG 2015; NFTC 2015; PwC 2015; USCIB 2015).

The same business representatives also opposed the proposed special measures. According to the NFTC, ex post adjustments of transfer prices paid for HTVI were problematic because they undermined certainty for multinationals. Therefore, adjustments should be applied only when the estimated and actual income generated by an intangible differ as a result of “events that could have been foreseen but were not taken into account in the valuation” (NFTC 2015, 614). Likewise, the Silicon Valley Tax Directors Group (SVTDG) argued that ex post adjustments should be permitted only if the taxpayer cannot show that an HTVI subject to a controlled transaction was also traded with unrelated parties under the same circumstances. Since economic projections are never completely accurate over multiyear periods, adjustments should, moreover, be prohibited after five years (SVTDG 2015, 763). As to the independent investor model and the possibility to disregard MFEs, reactions were even less forthcoming. The NFTC (2015, 614) strongly opposed the measure, as “it would substitute the judgment of tax authorities for capital allocation decisions and business judgments of MNEs.” The United States Council on International Business (USCIB 2015, 843) argued that these special measures “could result in significant realignment of taxing rights between source and residence countries and should be rejected.” The fear of a reallocation of taxing rights through the backdoor was shared by KPMG (2015), and added a political dimension to US multinationals’ technical criticism. According to a narrative popular among lobbyists in the US, the BEPS project “invit[ed] the entire world to impose higher taxes on US multinationals,”13 and special measures were one instance of this invitation. In the words of a senior lobbyist for US multinationals:

The special measures were put out there because countries wanted them out there. And the fact that they are out there doesn’t mean that the (p.119) OECD’s Centre for Tax Policy supports those. But what worries me and my members about this is when the OECD comes out with a certain model, even if it has not been approved, yet the fact that it has been published in an OECD document gives it some kind of seal of approval. So it is distracting when you see this kind of stuff coming out, because you don’t know whether someone will latch on to that independent investor model, even though it makes zero sense.14

US multinationals’ opposition to the European approach of sticking with the basic principles of the established system, while expanding the leeway of tax examiners (including in source countries), left the Obama administration with a single option. Because of domestic opposition, the administration could not impose its preferred countermeasure to tax avoidance internationally. At the same time, the administration’s strong political commitment to tax fairness prevented it from abandoning the BEPS project altogether (Herzfeld 2015a). Therefore, the Obama administration entered into a rearguard battle, essentially seeking to preserve the status quo of the international tax system, including the ALS, separate entity accounting, and the benefits principle (Finley 2017b, 2017a). After the release of the transfer pricing draft, Robert Stack adopted the lobbyists’ narrative, suggesting the BEPS project was pushed forward by Europe and Australia to increase the tax burden of US multinationals (Stack 2015). Against this background, he stressed he was determined to protect the US tax base in negotiations and would make sure “that our taxpayers are being treated fairly around the world, that they have rules that are clear and administrable, and that we as the US government are not opening the door to rules that will create greater and greater tax disputes” (Parillo 2014). On the transfer pricing draft in particular, Stack affirmed, “We don’t want transfer pricing reports to become basically anti-abuse rules or CFC rules. We want them to clearly articulate the arm’s-length standard” (Parillo 2014), adding on another occasion that “there will be no free pass to recharacterize willy-nilly based on vague notions” (Sheppard 2015). The Obama administration’s focus thus shifted to preserving legal certainty for US multinationals. This was to be achieved by paring back discretionary measures in the proposed transfer pricing guidelines and through the establishment of binding arbitration for disputes between tax authorities that may arise from remaining ambiguity (Herzfeld 2015b).

In accordance with Stack’s announcement that “the US will also be heavily involved in the articulation, editing, and drafting of the final version of the transfer pricing guideline changes” (Sheppard 2015), the final report reflects the priorities of US multinationals discussed previously. The draft section justifying the need for nonrecognition with moral hazard reflected in the proliferation of (p.120) contractual arrangements and the artificial separation of functions between related parties disappeared (cf. OECD 2014b, 25–26, 2015a, 38–41). Instead, the final guidance allows nonrecognition only in “exceptional circumstances,” that is, when “the transaction viewed in its entirety lacks the commercial rationality of arrangements between unrelated parties” (OECD 2015a, 39). In contrast to the discussion draft, however, the final guidance no longer provides indicators for commercial irrationality, such as a negative pretax return for the group. It rather includes two illustrative examples for extremely irrational arrangements that considerably narrow the term’s definition (OECD 2015a, 40). Hence, the scope for recharacterization of contractual arrangements between related parties is a lot smaller in the final guidance, which is also reflected in its new emphasis on the circumstances ruling out any form of nonrecognition (OECD 2015a, 39). Directly addressing US multinationals, one tax policy analyst commented this outcome as follows: “Even though the US BEPS negotiators forced the Europeans to accept your self-serving tax-planned contracts, the problem is that you will have to follow the letter of your tax-planned contracts” (Sheppard 2016). Considering the public comments of US business, it appeared that corporations were prepared and able to do so (NFTC 2015, 811; SVTDG 2015, 724).

Similarly, the US hand in drafting the final transfer pricing guidance became evident in the almost complete disappearance of special measures. In the executive summary of the final report, the OECD proudly announced, “The goals set by the BEPS Action Plan in relation to the development of transfer pricing rules have been achieved without the need to develop special measures outside the arm’s length principle” (OECD 2015a, 12). Indeed, the independent investor model and the possibility of disregarding controlled transactions with MFEs are missing in the final guidance (cf. OECD 2015a).15 Although no longer labeled accordingly, the only remaining special measure is the ex post adjustment of controlled transactions involving HTVI. This is the only option from the discussion draft that received some positive feedback from US business during public consultations (as discussed earlier), most likely because it essentially corresponds to the commensurate-with-income provision in US international tax law (Herzfeld 2015b). Still, US corporations requested several restrictions on the use of ex post adjustments, all of which were adopted in the final transfer pricing report. These restrictions include a deadline for transfer price adjustments five years after an asset’s commercialization and a prohibition of adjustments when the information provided by taxpayers on their income projections suggests that differences from the actual outcome are the result of unforeseeable or foreseeable but unlikely events (NFTC 2015, 814; SVTDG 2015, 763; OECD 2015a, 111).

As a result, corporations can generally avoid ex post adjustments by providing tax authorities with the details of their ex ante projections. That is, their compliance (p.121) burden is once more reduced to following their self-imposed legal and economic fictions. Against this background, Andrew Hickman, head of transfer pricing at the OECD’s Centre for Tax Policy, reminded US business representatives, “‘It could have been worse’ is really the key message, so be grateful for the version you have at the moment” (Finley 2016). On the same occasion, Michael McDonald, transfer pricing specialist with the US Treasury, concluded that the differences between the discussion draft and the final report “illustrated the road not taken” (Finley 2016). In other words, the European approach of providing tax examiners with greater leeway in transfer pricing analyses had been halted by the US Treasury’s negotiators.

Preventing Enlargement of the Permanent Establishment Definition

To be sure, the rearguard battle dynamic was not limited to transfer pricing issues. The Obama administration also defused other contentious action items that threatened to enlarge the tax take of source countries. One such item was the enlargement of the permanent establishment (PE) definition in the OECD’s Model Tax Treaty, which determines when a nonresident corporation has a sufficient connection to a source country to be taxed by the latter (OECD 2014d, Art. 6). Formerly, this required a physical presence such as a local office or factory. Yet the increasing importance of the Internet for commercial transactions enables corporations to access source country markets without local representation. Therefore, large EU member states had long tried to include Internet servers in the PE definition, but were rebuffed by the United States (Pinkernell 2014). The BEPS project provided a new opportunity to lower the PE threshold. In view of this opportunity, the corresponding discussion draft included two important proposals. First, the existing exemption of certain activities, such as storing and delivery of merchandise, from the PE definition would become effective only if these activities played a preparatory or auxiliary role in a company’s business model (OECD 2014f, 15–17). Second, commissionaires selling products in their own name but on behalf of a foreign company that is the owner of the products should be included in the PE definition to avoid the artificial circumvention of PE status (OECD 2014f, 10–11).

Through these proposals, large EU member states hoped to resolve tax-planning schemes popular among e-commerce platforms and reduce the fiscal advantages they enjoy over local distributors (Sheppard 2015).16 Amazon, for instance, would no longer be able to deliver products to customers in the entire common market while billing their purchases only in Luxembourg (Grinberg 2015).17 In contrast, the US government interpreted these measures as an attack (p.122) on US multinationals and the US tax base. From the Obama administration’s perspective, the profits US-owned companies made in the common market were first and foremost taxable in the United States, where these companies’ IP had been developed. Accordingly, Robert Stack declared himself “extremely disappointed” by the PE discussion draft and urged source countries “to acknowledge the sometimes unpleasant reality that very often there’s not much value added in their jurisdictions” (Stack 2015). Against this background, Henry Louie, deputy international tax counsel with the US Treasury, threatened to enter a reservation in the OECD’s Fiscal Affairs Committee if proposed changes to the PE definition were to be added to the Model Tax Treaty (Martin 2015). To accommodate US opposition, the committee thus decided to make the adoption of the changes optional (OECD 2015c, 14). Even the signatories of the multilateral instrument (MLI) applying tax-treaty-related BEPS recommendations to existing bilateral tax treaties retain the right to reserve against the new PE definition (OECD 2017c, Art. 12 & 13). As a result, neither the US nor the small EU member states hosting the cash boxes of US multinationals currently accept the revised definition in their bilateral tax treaties (US Treasury 2016, Art. 5; OECD 2017b).18

Paring Back Country-by-Country Reporting

Likewise, the Obama administration—this time supported by the German government—pared back the proposal for comprehensive country-by-country reporting (CbCR) contained in BEPS action 13. The concept of CbCR was originally developed by the Tax Justice Network (TJN), a group of expert activists fighting corporate tax avoidance and financial secrecy. The concept gained prominence through subsequent endorsements by the European Parliament and the European Commission. The basic idea is to have multinationals report tax payments, profits, and activities on a country-by-country basis, thus enabling tax authorities and civil society to identify mismatches that could be the result of tax avoidance (Seabrooke and Wigan 2016). To this end, the OECD’s initial discussion draft proposed a CbCR template, requiring multinationals to annually report for each of their constituent entities pretax and posttax earnings, income and withholding taxes paid, payroll, and number of employees. To facilitate the detection of profit shifting, royalties, interest, and service fees paid among related entities would also be included (OECD 2014c, 15). As to the implementation of CbCR, the draft recommended that the parent company complete the report and then share it with each country hosting a related entity. If the parent company or its local affiliate does not promptly comply, tax authorities would receive the report from their foreign counterparts through an information exchange mechanism (OECD 2014c, 10).

(p.123) In response to these proposals, business associations from the US and Europe submitted three main points of criticism. First, they argued that many items included in the CbCR template were not normally reported on an entity basis and therefore significantly increased the compliance burden for corporations. Second, the associations warned that tax authorities might misuse information on payroll and intragroup payments for formulary apportionment purposes or to question transfer pricing analyses. Third, business feared that a direct submission of country-by-country reports to source countries would threaten the confidentiality of business secrets. Accordingly, business requested that contentious items be dropped from the CbCR template and reports be shared only among tax authorities on a treaty basis, thus enabling residence countries to withhold information from source countries when the latter do not have sufficient confidentiality safeguards in place (BDI 2014; NFTC 2014; MEDEF 2014; USCIB 2014). The United States and Germany had an open ear for these concerns. Manfred Naumann, the head of division responsible for CbCR in the German finance ministry, argued that to preserve confidentiality and prevent misuse, parent companies should report only to their country of residence, which could then decide whether to relay information to source countries based on a bilateral treaty. In any case, information should never be made public (Naumann and Groß 2014). Along the same lines, Robert Stack affirmed that the US government’s goals were to protect proprietary information, reduce the compliance burden, and ensure that information is shared only between tax administrations to preserve confidentiality (Stewart 2014).

Accordingly, the CbCR template contained in the final report on BEPS action 13 no longer obliged corporations to report payroll or intragroup royalty, interest, and service fee payments (OECD 2015d, 29). That is, all items unanimously opposed by US and European business were dropped. In addition, the final report states that “jurisdictions should have in place and enforce legal protections of the confidentiality of the reported information” and “should not propose adjustments to the income of any taxpayer on the basis of an income allocation formula based on the data from the Country-by-Country Report” (OECD 2015d, 22). Again, key business concerns over the misuse of reported data were addressed. Most important, however, the final guidance provides for CbCR from parent companies to residence countries only, whereas the annexed multilateral agreement for CbCR implementation enables governments to choose the other signatories with which they intend to share the reports (OECD 2015d, 23, 50). That is, residence countries can cherry-pick the source countries receiving information on “their” multinationals, either by not signing the multilateral agreement or by selecting among the other parties to the treaty. As a result, the original intent behind CbCR—to enable tax authorities and civil society to identify mismatches between (p.124) corporate profits, taxes paid, and economic activity—is unlikely to be realized based on the final report on BEPS action 13. Commenting on this outcome, Alex Cobham (2015a), the TJN’s current director, concluded, “the [CbCR] standard has been strangled at birth.”

The BEPS Project’s Implementation

The Proliferation of Unilateral Fixes

To become effective, the BEPS project’s final recommendations need to be transposed into national administrative practice. To this end, the OECD uses two soft law instruments, the Transfer Pricing Guidelines and the Model Tax Convention, which member states generally use as templates for their national administrative rules and bilateral tax treaties. In accordance with the Model Tax Convention, article 9 of these treaties usually obliges tax administrations to apply the arm’s-length standard in their transfer pricing analyses and ensure that both signatories consistently apply any price adjustment. In these contexts, the Transfer Pricing Guidelines are the key reference, including when disputes end in court (Genschel and Rixen 2015; Rixen 2008). Hence, the OECD’s soft law also becomes binding for nonmembers when they enter into a bilateral tax treaty with a member state or base tax treaties with other nonmembers on the OECD template. Over time, nonmembers have, indeed, developed the same propensity as OECD members to strike bilateral tax treaties following the Model Tax Convention. As a result, 3,200 such agreements are currently in place, underlining the global reach of the OECD’s tax standards (Arel-Bundock 2017). Yet the swift implementation of updates is difficult in such a system of decentralized multilateralism, as governments normally have to renegotiate their treaties one by one. To speed up the process, the OECD has begun to sponsor MLIs committing signatories to apply certain BEPS recommendations in their mutual relations. Although these agreements are open to all countries, they are voluntary and allow signatories to opt out of contentious provisions (OECD 2017b). Hence, governments can still deviate from many BEPS recommendations, as will be further discussed later.

Since the OECD’s Transfer Pricing Guidelines inform the day-to-day practice of tax administrations, a swift application of the changes introduced through the BEPS process could be expected. Yet the language on the analyses of new facts and circumstances for regular transactions and transactions involving HTVI is extensive and challenging to implement even for highly capable tax administrations in developed economies (Avi-Yonah and Xu 2016). In addition, from the perspective of some tax administrations, the guidance’s emphasis on the application of the arm’s-length standard remains at odds with the BEPS project’s promulgated (p.125) goal of aligning taxation with production and value creation. Although the scope for price adjustments has been significantly reduced in the final report, some divergence in the interpretation of the new language is likely. Accordingly, multinationals and practitioners deplore uncertainty over the exact implementation of BEPS recommendations (Büttner and Thiemann 2017). In response, the OECD continues to publish supplementary documents on the implementation of contentious sections in the revised guidelines (OECD 2017d), whereas the US government pushes for binding arbitration to ensure that the resolution of disputes between tax administrations is swift and informed by an orthodox interpretation of the new language (Finet 2015; Finley 2015). At the time of writing, twenty-five countries—hosting 70 percent of the US outward foreign direct investment (FDI) stock but excluding emerging economies—had agreed to participate (OECD 2017b; UNCTAD 2014).19 Because the legal certainty provided by binding arbitration constitutes a locational advantage from the perspective of multinationals, more countries are likely to follow (Arel-Bundock and Lechner 2017).20 At the same time, a shift in the balance of power may also enable emerging economies to resist US demands for binding arbitration, sustain incoherence in the application of BEPS recommendations, and lead to a multiplication of unresolved tax disputes in the future.

The significance of the new PE definition for the taxation of multinationals depends on the willingness of governments to adopt it in their bilateral tax treaties. Yet the Obama administration stuck with the narrower old definition in its 2016 revision of the US model tax treaty, while the Trump administration refuses to sign the MLI through which tax-treaty-related BEPS changes could be adopted (Schwarz 2016; Herzfeld 2017a). At the same time, Ireland and Luxembourg, the countries hosting the controlled entities to which US multinationals shift profits from the rest of the EU, signed the MLI but opted out of the new PE definition (OECD 2017b). That is, if France or Germany wanted to tax more of a US multinational’s activity at source by applying the new wording, they would first need to get the United States, Ireland, and Luxembourg to agree to a corresponding revision of their respective bilateral tax treaties. As a result of the power differential in such bilateral negotiations, a French or German attempt at convincing the United States to deviate from its model treaty is likely to fail. Likewise, the principle of nondiscrimination in EU law should prevent France and Germany from pressuring Ireland or Luxembourg into compliance with the new definition. The EU’s Interest and Royalties Directive would, for instance, override a withholding tax on royalty payments to these countries, as legal commentary on the United Kingdom’s diverted profits tax has recently made clear (see below) (European Union 2003; MacLennan 2016; Self 2015).21 Once more, the adoption of new PE rules at OECD level thus depends on the EU’s ability to transcend its internal (p.126) divisions. If a wider PE definition became EU policy, the outcome of bargaining with the United States would be open. If internal divisions persist, however, individual member states will remain unable to wrestle concessions from the Trump or any future administration.

Another contentious issue in post-BEPS relations between the United States and the EU is the publication of country-by-country reports. In contrast to the final BEPS guidance and the MLI for CbCR adoption, which both include strong safeguards against the publication of reports, the European Commission and several member states seek to introduce public CbCR unilaterally. Whereas the French National Assembly even passed a corresponding law—which was, however, scrapped by the Constitutional Court for discouraging free enterprise by divulging company secrets (Assemblée Nationale 2016; Conseil Constitutionnel 2016)—the Commission’s proposal is still in abeyance. First proposed as part of its “action plan for a fair and efficient tax system” (European Commission 2015b), the Commission promotes public CbCR to “enable citizens to better assess the contribution of multinational undertakings to welfare in each Member State” (European Commission 2016c, para. 9). From its perspective, “enhanced public scrutiny of corporate income taxes … ​is an essential element to further foster corporate responsibility, to contribute to the welfare through taxes, to promote fairer tax competition … ​and to restore public trust in the fairness of national tax systems” (European Commission 2016c, para. 5). To circumvent the unanimity requirement for Council decisions on taxation, the Commission proposes public CbCR as an amendment to the Accounting Directive, which already provides for public reporting in the banking and extractive industry sectors (European Commission 2016c, 4). On accounting matters, justice instead of finance ministers decide through qualified majority voting. Still, the proposal faces considerable opposition. Although the European Parliament has already adopted the amendments,22 Germany and twelve other member states continue to oppose public CbCR in the Council (European Parliament 2017; Becker 2016). In parallel, the United States threatens to stop the exchange of country-by-country reports with any country publishing the data and contemplates additional sanctions (Johnston 2016).23 In contrast to CbCR to tax authorities only, which has been adopted by the United States and the EU (European Union 2016; IRS 2016), the adoption and effectiveness of public CbCR in the EU thus remains highly uncertain.

As these examples illustrate, many governments that are primarily source countries from the US perspective are dissatisfied with the status quo–preserving recommendations of the final BEPS reports and now attempt to defend their interests through divergent implementation. In addition, unilateral measures expanding the taxation of nonresident multinationals at source have recently proliferated (Elliott and Sheppard 2016; Herzfeld 2017b). China has, for instance, (p.127) tightened exchange controls, thereby keeping local subsidiaries of foreign multinationals from paying dividends to their parent companies (Clover 2016). Most prominently, the UK government introduced a diverted profits tax in response to public outrage over tax minimization by Google and other tech firms (Houlder 2014).24 The measure enables Her Majesty’s Revenue and Customs to withhold 25 percent of profits sent offshore if it suspects a multinational of engaging in a “contrived arrangement” to avoid a taxable presence in the United Kingdom (Houlder 2015a, 2015b). The finance ministers of France, Germany, Italy, and Spain recently proposed the introduction of an EU-wide equalization tax on payments from resident customers to nonresident companies. This measure targets the digital economy in particular, which can sell online services to customers in the entire common market without having a PE in every member state (Le Maire et al. 2017). While this proposal is still at an early stage and faces considerable opposition among member states, it demonstrates that important European governments do not believe that BEPS recommendations will end tax avoidance by US multinationals in the common market. As this goal had been one of the key motivators for EU governments to enter the BEPS project, they obviously failed to defend their agenda against the US government.

Toward More Regulatory Centralization at the EU Level?

The recent political entrepreneurship of the European Commission may, however, turn the preservation of the international tax system’s status quo into a Pyrrhic victory for the United States. So far, the internal division between small capital-importing and large capital-exporting member states, paired with the unanimity requirement for Council decisions on taxation, has ensured that the EU would make little progress in fighting tax avoidance in the common market (Dehejia and Genschel 1999; Wasserfallen 2014). Since the BEPS project also failed to provide a solution to this politically increasingly salient problem, the European Commission—led by Pierre Moscovici, the French socialist commissioner for taxation and customs, and Margarete Vestager, the Danish social-liberal commissioner for competition—now seems committed to finally exit the joint decision trap in taxation by other means.25 To this end, Vestager used the Commission’s executive powers in competition policy to launch state aid investigations against Ireland, Luxembourg, and the Netherlands for the provision of selective tax advantages to Amazon, Apple, Fiat, GDF Suez, McDonald’s and Starbucks.

As a result of these investigations, the Commission has to date instructed Ireland to claw back €13 billion in corporate tax from Apple, Luxembourg to claw (p.128) back €270–€280 million from Amazon and Fiat, and the Netherlands to claw back €20–€30 million from Starbucks (European Commission 2017a). By creating uncertainty over the legality of sweetheart deals offered to multinationals, these decisions remove a locational advantage from member states making such offers and may eventually unsettle their opportunity structures in the Council. Therefore, Moscovici flanked the state aid decisions with the action plan on fair and efficient taxation, a relaunch of the common consolidated corporate tax base (CCCTB) proposal, and the proposal for public CbCR discussed earlier (European Commission 2015b, 2015a). Although none of these initiatives has yet been adopted, they perpetuate a public discourse on tax avoidance, create political demand for solutions, and may thereby pressurize member states into action.26 If this dynamic eventually enabled the EU to overcome its internal divisions and adopt the CCCTB, there would also be consequences for the US tax base.

Currently, US multinationals exploit the mismatch between the free circulation of capital in the common market and twenty-eight national tax policies to shift profits first to a low-tax member state and then to a tax haven outside the EU. This shift usually happens through royalty and dividend payments between group branches (see chapters 1 and 2). From the US perspective, such payments constitute passive income taxable at residence. If the deferral and check-the-box loopholes were closed, profits shifted out of the common market would thus increase tax revenue in the United States. If the EU adopted the CCCTB, however, profits generated in the common market would be consolidated at a multinational’s EU headquarters and then divided among member states based on local sales, production, and assets (European Commission 2015a). Accordingly, US multinationals could no longer concentrate their EU profits in a low-tax member state that does not withhold taxes on dividend payments to parent companies outside the EU. As a result, the US multinationals’ European tax bill would rise, while the amount of passive income potentially taxable in the United States would shrink.

This perspective, which has in part already become reality through the back taxes US firms have to pay to Ireland, Luxembourg, and the Netherlands, led Jack Lew, the Obama administration’s last Treasury secretary, to request an end of the state aid investigations in a letter to Commission President Jean-Claude Juncker. In his letter, Lew accused the Commission of setting “disturbing international tax policy precedents” that were “inconsistent with, and likely contrary to, the BEPS project” (Lew 2016b, 1). He insists that US multinationals’ deferral of tax payments on foreign profits in the US “does not give EU Member States the legal right to tax this income,” and concludes by urging Juncker “to reconsider pursuing these unilateral actions” (Lew 2016b, 2–3). Although the US Treasury reinforced (p.129) these requests by threatening to impose retaliatory measures on EU firms (Chee 2016; Lynch 2016), Vestager defended her directorate’s approach in her response to Lew, and has since launched further investigations leading to additional decisions against US multinationals (Vestager 2016). By preventing EU governments from effectively addressing their tax avoidance problems through the BEPS project, the United States may thus have given the EU the decisive impetus to finally harness its great power potential in international taxation. The conflict over state aid decisions could thus be the first stage in a protracted EU-US battle over taxing rights.

Theoretical Implications

The BEPS project materialized in 2012 because important G20 members simultaneously had to respond to domestic concerns over corporate tax avoidance. Whereas the conservative finance ministers in the United Kingdom and Germany responded to public outrage over revelations that multinationals like Starbucks paid little to no tax in the common market, the Obama administration was bound by its campaign commitment to restore tax fairness for the middle class, a central theme in Democratic party ideology since World War II. As a result of these constraints, governments on both sides of the Atlantic could not react to tax avoidance merely by lowering taxes on corporate profits. To address the fairness concerns that were either voiced by the electorate or rooted in the decision makers’ normative convictions, governments had to credibly commit to countermeasures. Accordingly, the G20 mandated the OECD to find solutions to the observed mismatch between taxation and value creation, to which the organization responded with the creation of the BEPS project.

Although there was agreement that something had to be done, the EU and US governments disagreed over the instruments. Whereas the EU governments in the G20 were open to making limited concessions to source countries—partly to bring US multinationals within the reach of the EU governments’ tax administrations, and partly to coopt emerging economies into the OECD process—the US envisaged a solution fully geared toward the interests of residence countries. However, business opposition prevented the Obama administration from adopting strengthened CFC rules and anti-deferral measures domestically. Therefore, the US lacked a regulatory model it could impose on other governments through the OECD process. Since the Obama administration was normatively committed to fighting tax avoidance, it could not convert the resulting lack of purpose into a complete withdrawal from the BEPS project. Hence, the only remaining strategy was to respond to business interests and defend the international tax system’s (p.130) fundamental principles against the more far-reaching reform proposals from European governments.

The Obama administration implemented this strategy through a successful rearguard action against European attempts at expanding tax administrations’ leeway in the recharacterization of controlled transactions; broadening the Model Tax Convention’s PE definition; and including internal royalty, interest, and service fee payments in country-by-country reports. Yet by preventing the international community from reforming a dysfunctional international tax system, the United States may have undermined its future ability to dominate international tax matters. Whereas FATCA provides other governments with a US-defined solution to tax evasion by individuals, the US administration’s inability to provide such a solution to corporate tax avoidance invites other governments to create their own. Therefore, the number of unilateral initiatives against corporate tax avoidance has multiplied in the aftermath of the BEPS project. Most important, it apparently motivated the European Commission to develop strategies to overcome the EU’s regulatory dispersion in matters of direct taxation. If these strategies enable the European Union to finally harness its great power potential, international tax policy may be shaped by a power duopoly rather than a single hegemon in the future.


(1.) Interview with tax adviser to the German finance ministry on June 22, 2015.

(2.) Interview with senior German tax official on March 3, 2015.

(p.153) (3.) Interviews with tax adviser to the German finance ministry on June 22, 2015, and partner in US tax law firm on April 17, 2015.

(4.) Interview with senior German tax official on March 3, 2015. Translation from German by the author.

(5.) Interview on April 21, 2015.

(6.) Translation from German by the author.

(7.) Interviews with senior tax lobbyist for US multinationals on April 23, 2015, and partner in US tax law firm on April 17, 2015.

(8.) Interview on March 3, 2015. Translation from German by the author.

(9.) The author served as senior adviser to the OECD in the elaboration of the BEPS project.

(10.) Out of a total of 9,316 pages received by the OECD during public consultations on the BEPS project, 3,014 are devoted to transfer pricing; 1,177 to CbCR; and merely 577 to CFC rules.

(11.) Submissions from the Silicon Valley Tax Directors Group (SVTDG 2015) and the United States Council on International Business (USCIB 2015) also included in the compendium of public comments convey the same points of criticism.

(12.) This statement matches the key points of criticism expressed by PwC (2015), another big four accounting firm, in its submission to the OECD.

(13.) Interview with senior lobbyist for US multinationals on April 23, 2015.

(14.) Interview on April 23, 2015.

(15.) Special measures were sometimes interpreted as CFC rules, but they do not appear in the corresponding final BEPS report (cf. OECD 2015b).

(16.) Interview with tax adviser to the German government on June 22, 2015.

(17.) The author also shows that Amazon has begun to change its tax structure in Europe so as to unambiguously subject itself to taxation at source.

(18.) That is, a source country wishing to apply the new definition to corporations residing in the US, Luxembourg, or Ireland faces the difficult task of renegotiating the corresponding bilateral tax treaties and convincing its treaty partners to abandon their general reservation in the process. At present, this result seems highly unlikely, especially when the US government is the treaty partner.

(19.) The twenty-five countries include Andorra, Australia, Austria, Belgium, Canada, Fiji, Finland, France, Germany, Greece, Ireland, Italy, Japan, Liechtenstein, Luxembourg, Malta, the Netherlands, New Zealand, Portugal, Singapore, Slovenia, Spain, Sweden, Switzerland, and the United Kingdom.

(20.) The authors show that the diffusion of bilateral tax treaties is in part driven by the competitive advantages such agreements afford corporations interested in investing abroad.

(21.) With the adoption of the Interest and Royalties Directive in 2003, withholding taxes on interest and royalty payments between related firms of different member states were abolished.

(22.) Yet the Alliance of Liberals and Democrats for Europe (ALDE) managed to introduce a safeguard clause protecting multinationals against the release of sensitive information (Amendment 82). Deputies on the left fear this will allow many companies to circumvent public reporting (cf. De Masi 2017).

(23.) According to the draft directive, public CbCR would apply to all firms with a PE in the common market that earn more than €750 million in annual revenue. In contrast to the BEPS recommendations, this provision implies that firms that are not headquartered in the EU would also be obliged to publish information. Therefore, were the United States to withhold information received from parent companies, this would not make much of a difference. Accordingly, alternative sanction mechanisms, including the application of (p.154) section 891 of the US tax code through which taxes on companies from a country that is considered to impose discriminatory taxes on US multinationals could be doubled, are currently being debated (cf. Grinberg 2016b).

(24.) Although the measure was clearly conceived to respond to public concern over tax avoidance, the measure fit in neatly with the conservative government’s tax-cut-cum-base-broadening strategy. While the diverted profits tax limits profit shifting out of the United Kingdom, the parallel introduction of a tax break for multinationals’ overseas financing activities provides an additional incentive for shifting profits toward the United Kingdom. In addition, the Tories also reduced the statutory corporate tax rate from 26 to 19 percent and introduced another tax break for research and development activity. Overall, the United Kingdom intensified international tax competition instead of limiting it (cf. Hakelberg and Rixen 2017).

(25.) The term “joint decision trap” was coined by Fritz W. Scharpf (1988) to describe deadlock in decision-making processes that are marked by a unanimity requirement and divergent interests among involved actors.

(26.) According to Gerda Falkner (2011, 12) “unsettling” and “pressurizing” are two strategies through which the Commission can change member states’ opportunity structures and thereby find an exit from the EU’s joint decision trap.