Transnational expertise and the expansion of the international tax regime: imposing ‘acceptable’ standards

Abstract

Global economic governance outcomes in areas such as corporate taxation may be influenced by transnational policy communities acting at national and transnational levels. Yet, while transnational tax policy processes are increasingly analyzed through the politics of expertise, national preferences have usually been derived from domestic interest group preferences. We know little about how technical expertise interacts with interest group politics at national level, an important deficit given the sovereignty-preserving, decentralized way in which transnational tax norms become hard law. This article examines the drivers of expansion of the UK’s bilateral tax treaty network in the 1970s, which cannot be explained solely through monolithic interest group politics. Evidence from the British national archives demonstrates how tax experts in the civil service and the private sector, members of a transnational policy community, used tax treaties to impose OECD standards for taxing British firms on host countries, at times overruling the preferences of other political, bureaucratic and business actors. Expertise politics and business power may shape the development of norms and focal points within a transnational policy community, but it is often their interaction at domestic level that determines the implementation of transnational norms as hard law.

Keywords:

• Transnational policy communities
• developing countries
• foreign direct investment
• taxation
• United Kingdom
• bilateral tax treaties

Introduction

The political debate around the international tax regime that has emerged in recent years focuses on a popular perception that it is far too vulnerable to tax avoidance and evasion (Buttner & Thiemann, 2017; Christians, 2010c; Eccleston & Smith, 2016; Hakelberg, 2016; OECD, 2013; Rixen, 2008a; Seabrooke & Wigan, 2016), which occur through Global Wealth Chains (Seabrooke & Wigan, 2017). Two views in IPE scholarship explain how this situation arose. In the first, state-centric, view, the problem is one of path-dependence. States became locked in to the sovereignty-preserving, self-enforcing design of a regime designed to prevent double taxation, which made it harder for them to deal with the prisoners’ dilemma of tax competition that enabled tax avoidance and evasion (Genschel & Rixen, 2015; Picciotto, 1992; Rixen, 2011; Sharman, 2006). In the second, transnational view, a community of tax negotiators and tax professionals shielded from political attention elaborated a complex set of technical standards aimed at minimizing double taxation, then defended the integrity of these standards when outsiders proposed radical reforms to resolve the avoidance and evasion problems (Buttner & Thiemann, 2017; Grinberg, 2017; Picciotto, 2015; Seabrooke & Wigan, 2016; Ylönen & Teivainen, 2017).

These accounts are not mutually exclusive, and combining them offers a missing piece in the puzzle: the role of expertise in national preference formation. Scholarship examining the national political economy of international tax rules is limited to only a few examples (Barthel & Neumayer, 2012; Eden, Dacin, & Wan, 2001; Sadiq, 2012),¹ yet the international tax norms that are the focus of transnational accounts only become tax law when they are adopted by national governments, as the hard law of bilateral tax treaties and national tax codes (Christians, 2007). In the state-centric view, in contrast, national preferences are a function of the aggregate welfare implications and interest group politics concerned with the tax-driven effects of investment promotion and revenue raising (Dietsch & Rixen, 2016; Hakelberg, 2016; Rixen, 2011). Corporate capital has a clear interest in the elimination of double taxation, while there is no organized lobby against it. As a result, states are expected to have a first-order preference for stimulating trade and investment by concluding tax treaties that eliminate double taxation, and a second-order one for sacrificing as little tax revenue as possible when doing so (Chisik & Davies, 2004; Rixen, 2011; Rixen & Schwarz, 2009).

The UK is an archetypal case of tax treaty diffusion, having been an active participant in the multilateral level of the regime since its inception (Avery Jones, 2011) and concluding more bilateral tax treaties than any other state. It was in negotiations with 38 states outside the OECD during the 1970s. Yet some of its negotiating decisions during this time, notably to walk away from talks with Brazil, were taken against strong political pressure generated by industry lobby groups. A closer look at the process of preference formation shows that it was dominated by members of a transnational policy community (Tsingou, 2014), whose members had shaped a focal point, the OECD model convention, over decades. They used their roles in the domestic setting to translate the standards embodied by the OECD model into a wide network of bilateral tax treaties. Political pressure on treaty negotiators came from actors in government and business, not all of whom took for granted the aim of adherence to the OECD model. This led to heterogeneous preferences and influencing capabilities among business representatives and civil servants, even where they may have shared the same ultimate goals.

A further contribution of this article is to question the narrative that the international tax regime spread to developing countries because of their desire to stimulate investment flows by eliminating double taxation (Baistrocchi, 2008; Barthel & Neumayer, 2012).² In fact, it was often the UK that initiated tax treaty negotiations with developing countries, to place constraints on their ability to tax British investors. Negotiators from the Inland Revenue argued that there was rarely any double taxation problem to resolve, only a concern that developing countries might tax British investors in ways that were not ‘acceptable’, and that those firms risked ‘losing out’ to their competitors in the absence of a treaty. The precise benefits sought by actors in the UK depended on their familiarity with the norms formulated by the transnational community. While non-experts were motivated by the short-run costs to British firms associated with particular countries’ tax systems, members of the transnational policy community were focused on exporting tax standards they had formulated at the OECD. These, too, shielded their multinational firms from what they deemed to be unacceptable features of their tax systems, but community members’ longer-term outlook prioritized adherence to this focal point even if that meant disadvantaging British businesses in the short-term.

The article begins by outlining the development of the international tax regime from a state-centred perspective, focusing in particular on the relationships between OECD members and non-member developing countries. It then moves on to discussing the transnational view. Members of a transnational policy community take for granted certain focal points, which may lead them to different preferences in the domestic context than other actors, even if they share the same end goals. If community members have sufficient instrumental power, the national preferences arrived at by governments may differ from those that might be arrived at simply through analysing the presumed interests of domestic stakeholder groups. To demonstrate this, a detailed analysis of civil service documents from the 1970s demonstrates a business lobby group ‘speaking with two voices’: tax experts who were part of a transnational policy community and were brought into the confidence of the Inland Revenue, and others who were not. This polarization, which could also be found within the civil service, was most evident in the politicization of negotiations with Brazil. Non-expert business lobbyists used instrumental power to convince other government departments to lobby for an agreement, while tax community members in the Inland Revenue and in the same business lobby group worked together successfully to counteract this pressure. Their concern was that any possible agreement would have run counter to the norms embodied by the OECD model convention, the regime’s most powerful focal point.

The state-centred view of the international tax regime

The development of the international tax regime is often understood through an ‘open economy politics’ (Lake, 2009) type lens. Governments aim to maximize national welfare while maintaining the support of three domestic interest groups: labor, individual capital and corporate capital (Rixen, 2011, p. 200). In this view, business capital usually possesses the greatest instrumental power because it has a clear and shared preference, and it prefers the alleviation of double taxation (Rixen, 2010, pp. 595–6). Writing about the determinants of great powers’ international tax policy, Lukas Hakelberg (2016, p. 513) points to ‘domestic constraints’ that shape a government’s incentive to shift the tax burden between labor, consumption and capital, as well as the impact on domestic industries’ competitiveness.

In the story advanced in most detail by Thomas Rixen (2008b, 2010, 2011), these incentives at the domestic level create a strong preference for cooperation between states to eliminate double taxation, which at interstate level produces a coordination game that can be resolved through a sovereignty-preserving regime without multilateral enforcement (see also Radaelli, 1998). The result is a global network of over 3000 bilateral tax treaties, often called ‘double taxation agreements’, almost all of which are based directly or indirectly on an OECD model convention (OECD, 2017). The OECD model reflects the interests of capital exporting countries, because it imposes a solution to the double taxation problem that imposes a larger share of the costs onto the capital importing country (Brooks & Krever, 2015; Dagan, 2000; Genschel & Rixen, 2015; Irish, 1974; Paolini, Pistone, Pulina, & Zagler, 2016; Thuronyi, 2010). Capital-importing developing countries nonetheless seek tax treaties with OECD states as the price of attracting inward investment and because they can ameliorate some of the bias through bilateral negotiations (Barthel & Neumayer, 2012; Chisik & Davies, 2004; Hearson, 2018; Rixen & Schwarz, 2009).³

Because they primarily constrain states’ ability to tax inward investment, tax treaties are potential instruments of tax competition, offering inward investors a more credible commitment to a lower effective tax rate in the future than domestic law alone would provide (Baistrocchi, 2008; Barthel & Neumayer, 2012). Historically, this was especially the case where a treaty provided a matching credit against home country tax for tax exempted by the host country as an investment incentive, even though the tax had not been paid (Hearson, 2017). Also known as ‘tax sparing’, such clauses ensured that the benefit of any reduced taxation in the host country accrued directly to the investor, which had no tax liability in its home country on the income concerned. Investors able to take advantage of incentives in this way would have significantly lower tax costs than those which were not, and there is evidence such clauses significantly affected investment flows (Azémar, Desbordes, & Mucchielli, 2007).

Rather than the revenue-maximizing motivation posited by critical legal scholars such as Dagan (2000), capital exporting countries may therefore be motivated by competitive pressure to conclude a wide network of tax treaties, in order to enhance the competitive position of their outward-investing multinationals. Tax is a business cost, and so reducing it allows firms to outcompete those that pay a higher rate. The form of that pressure is quite subtle, however, and requires an understanding of the interaction between home and host tax systems in each case. More generally, we can see that the incorporation of developing countries into an international tax regime developed within the League of Nations and OECD acts primarily to constrain their ability to tax foreign investors other than in ways that its architects deemed appropriate.

The transnational view

At the turn of the 1970s, cooperation around double taxation rested on a set of norms developed by experts operating since the 1920s in a ‘quiet politics’ scenario (Culpepper, 2010). As Genschel and Rixen (2015, p. 163) describe, the low political salience of international taxation from the 1930s to the 1960s had ‘allowed the experts to craft a compromise solution without major intervention from their political principals.’ This compromise was set out in the OECD model convention, first agreed in its modern form in 1963, and eventually published in 1977 (Owens & Bennett, 2008). Today, the OECD model ‘represents the general consensus on international taxation’ (Rixen, 2011, p. 207). It is a ‘focal point…defined as social conventions that are followed ‘automatically’ because they have become self-evident’ (Rixen, 2010, p. 201). Embodied within the OECD model are a set of norms, most notably the ‘arm’s length principle’ for allocating taxable profits between jurisdictions (Buttner & Thiemann, 2017; Ylönen & Teivainen, 2017) which set the parameters of policy in the area of double taxation (Avi-Yonah, 2007). As Genschel and Rixen (2015, p. 163) set out:

The OECD Model Convention was embedded in a broad epistemic consensus on ‘how to do double tax relief properly’, which in turn reinforced its status as the self-evident reference point in matters of double tax relief once cross-border investments and capital mobility started to increase in the 1970s.

While state preferences no doubt influenced the shape of this consensus, tax historians commonly regard it as having formed among a transnational group of technical experts (Graetz & O’Hear, 1997; Jogarajan, 2017; Picciotto, 1992). The OECD model’s ‘direct parents were…senior tax officials from European countries’ (Owens & Bennett, 2008) and its lineage begins with the League of Nations’ Committee of Technical Experts on Double Taxation and Tax Evasion (Mcintyre, Bird, & Fox, 2005; Rixen, 2008a; Vogel, 1986). The preface to that committee’s report stresses that, ‘although the members of the Committee are nominated by their respective Governments, they only speak in their capacity as experts, i.e. in their own name’ (League of Nations, 1927, p. 6).

It is already very common to describe this transnational group of experts as an ‘epistemic community’. In the tax law literature, for example, Alison Christians (2010b, p. 22) describes how OECD staff, civil servants representing national governments, and other professional stakeholders ‘form an intertwined epistemic community that holds an important and influential position in the law-making order.’ These individuals ‘diagnose and prescribe tax policy reforms that are informed by, and that play out within, national legal regimes.’ Diane Ring (2006, p. 148, see also 2010, p. 681), similarly suggests that international tax negotiations are best understood as ‘epistemically informed bargaining’, in which an epistemic community ‘served as a driving force in the double taxation problem, both in terms of providing a forum for discussion and providing a base of expertise to structure the debate.’ Jason Sharman argues that ‘[t]ax administrators are enmeshed in a trans-national epistemic community.’ Rixen (2008a, p. 13), commenting on Webb (2006), regards the community as being ‘comprised of tax bureaucrats and business association representatives,’ having ‘succeeded in excluding civil society from international tax matters by defining the issues as being ‘purely technical’ in nature.’

In its original formulation, an epistemic community is ‘a network of professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue’ (Haas, 1992). It is easy to see the attraction of this concept for scholars of the international tax regime. Tax is a technically complex area in which a transnational community claims a monopoly on legitimate expert knowledge, propounding a policy project through the process of international standard formation that takes place in arcane committees of the OECD, but ultimately – mostly via bilateral treaties – takes the form of national tax laws (Buttner & Thiemann, 2017; Christians, 2010b; Picciotto, 2015; Ring, 2010).

Establishing the causal links between national and international settings is, however, a challenge for the epistemic communities literature, which has tended to focus on demonstrating the existence of particular communities, rather than on understanding how and in what circumstances they are able to influence – or indeed may be influenced by – national policies (Antoniades, 2003; Davis Cross, 2012). Haas himself suggested their influence came mainly in times of uncertainty and crisis for policymakers, which is unhelpful for the century-long incremental development of the international tax regime. More useful is Haas’ notion that policy influence comes in part through ‘infiltration’ of government bureaucracies by community members, but this still characterizes the community as an exogenous influence on national bureaucracies. The concept of an epistemic community is thus ill-suited for situations in which bureaucrats themselves form part of the community, where ‘the decision makers whom members of an epistemic community advise turn out to be themselves’ (Jacobsen, 1995, p. 302). One possibility is to consider the internalization of community ideas by bureaucrats through processes of ‘socialisation’, but such ideational change is empirically rather difficult to measure, and likely to depend on characteristics of the national setting, the issue area and any international organizations concerned (Beyers, 2010; Johnston, 2005, 2008; Zürn & Checkel, 2005).

A more appropriate theoretical concept is that of a transnational policy community, which ‘refers to a group of officials, whether public or private, that exhibits particular characteristics’ including similarities in education and career development, a strong sense of affinity to each other, and a set of interests ‘defined and articulated in terms of widely accepted principles’ (Tsingou, 2014, p. 233). Such communities use club settings, in which ‘Members place a limit to the range of actors involved in the making of policy and define what type of actor is relevant’ (Tsingou, 2014, p. 231).

To characterize the international tax community in this way, we must establish the ‘widely accepted principles’ on which its interests rest, the mutual affinity and common characteristics of its members, and club membership rules. The departure point for such a description is the community’s original aim of alleviating double taxation in order to promote trade and investment. The OECD Model Convention states its own main purpose as being ‘the application by all countries of common solutions to identical cases of double taxation’, it being ‘scarcely necessary to stress the importance of removing the obstacles that double taxation presents to the development of economic relations between countries’ (OECD, 2017, p. 9). To achieve this, the community settled on an agreed equilibrium point amongst multiple stable equilibria – the OECD Model (Rixen, 2008b) – then promoted adherence to it. States now take care of the ‘heroic’ double taxation that motivated the original League of Nations work through their national tax laws (Dagan, 2000), and the words ‘double taxation’ have been removed from the Model’s title. It has instead come to embody a consensus view of how to tax cross-border income and capital that transcends the original double taxation problem and provides the international tax community with a compelling ongoing claim to authority.

One of the participants in the early League of Nations work, Edwin Seligman (1928, pp. 143–144) observed that, while at first the technical experts’ ‘concern was primarily to enter into some arrangement which would be politically agreeable to their respective countries’:

when they learned to know each other more intimately; and especially in proportion as they were subjected to the indefinable but friendly atmosphere of the League of Nations, their whole attitude changed. Suspicion was converted into confidence; doubt was resolved by the feeling of certainty of accomplishment; and aloofness gave way to warm personal friendship which contributed materially to smoothing out the difficulties.

According to Picciotto (1992, p. 37), ‘perhaps the most important outcome of the inter-war years was to begin to create a community of international tax specialists…a community within which ideas and perspectives as well as economic advantage could be traded.’

Today, the burden of participating in numerous international meetings is a common complaint overheard by the author among government officials and business representatives during coffee breaks at such meetings, but it is clear that close social relationships develop as a result. One staff member of an organization that frequently hosts international tax meetings observed during one such coffee break, ‘these people are friends, they stay at each other’s houses.’⁴ According to a former treaty negotiator from an OECD country, participation in OECD meetings ‘was very much a club, people didn’t want to lose that gig, a really clubby arrangement.’⁵ Elements of this ‘clubbiness’ observed at international meetings include delegates’ habitual reference to each other in formal discussions by first name, and the clearly warm nature of informal discussions between longstanding members – regardless of their professional affiliation – during breaks and over dinner.

In addition to this sociological proximity, the community is notable because its membership incorporates experts from government and business, a logical consequence of a professional environment characterized by ‘revolving doors’ (Seabrooke & Tsingou, 2009). This process began right at the start, with Thomas Adams, the US-appointed member of the League committee, who chaired a committee for the US Chambers of Commerce as well as participating in the International Chambers of Commerce’s work; his successor, Mitchell Carroll, was a lawyer advising multinational firms on their tax affairs, as well as working on behalf of the US at the League (Carroll, 1978; Graetz & O’Hear, 1997, p. 1070). Today, a large proportion of tax advisers in the private sector have experience working in national revenue authorities, while the senior management of the OECD secretariat’s Centre for Tax Policy and Administration combines officials recruited from the private and public sectors.

In the UK, for example, ‘the corporate tax reform policy community has a tightly integrated and fairly constant membership’ leading to ‘an almost astonishing assimilation of professional expertise to the legislative function, born no doubt of many a congenial meeting over coffee and biscuits in Whitehall’ (Snape, 2015, p. 89). The UK government used secondees from Deloitte to help develop reforms to its laws surrounding taxation of multinational companies, who subsequently returned to the firm to advise private clients (Lawrence, 2012). Its revenue authority has a governing board drawn primarily from the private sector, while the UK branch of the International Fiscal Association (IFA), ‘the leading non-governmental international organization dealing with tax matters’, counts the UK Treasury’s most senior representative on OECD tax committees among its branch officers (IFA UK Branch, no date). Founded by Mitchell Carroll, the IFA’s ‘membership consists of high level representatives from both the private and the public sectors’, and its 2000-strong annual conference is the largest in a packed annual schedule of multistakeholder international tax gatherings (IFA, no date).

Accounts of the contemporary politicization of tax politics focus on how actors such as activist organizations who are not part of this community attempt to influence it, and the community’s resistance to such influence (Buttner & Thiemann, 2017; Picciotto, 2015; Seabrooke & Wigan, 2016). In particular, club membership is limited to those able ‘to accomplish the conversion of mental space – and particularly of linguistic stance – which is presumed by entry into this social space’ (Bourdieu, 1987). For Picciotto (2015, p. 179), in international tax, ‘law operates to defuse social conflicts and depoliticize them, shifting political and economic conflicts on to the terrain of debates over the symbolic power of texts… limiting the membership of the interpretative community and trying to ensure that they are like-minded.’ Non-governmental organizations, for example, have most successfully influenced deliberations by adopting the mantle of expertise themselves (Seabrooke & Wigan, 2016; Wigan & Baden, 2017).

Thus, multilateral interstate bargaining over taxation takes place through deliberation within a transnational policy community. Differing national preferences, as well as differences between governments and lobby groups, play a role in shaping community agreement, but once consensus has been reached, community members share the goal of advocating widespread adoption.

Bringing the national back in

The OECD model may act as a focal point within the transnational policy community, but the self-evidence of the standards it embodies does not always extend to political actors beyond that community. When domestic interest group politics exerts influence on governments, experts have to defend in the national setting a decades-old policy consensus that has been reached at transnational level. At times, as in the US in the early 1990s, domestic political actors have forced their governments to diverge from the transnational consensus (Durst & Culbertson, 2003; Radaelli, 1998). In that case, members of the policy community at the OECD ultimately shifted the focal point to accommodate the US and maintain consensus, but not without considerable complaint. A new wave of politicization since the financial crisis has further increased the attention paid by political actors to international tax rules, highlighting a divergence between public expectations and those of the international tax community, and putting pressure on governments to diverge from it (Buttner & Thiemann, 2017; Grinberg, 2017; Rixen, 2008a).

The national politics of international tax is thus influenced by, and influences, the transnational consensus. In her study of business power in corporate tax policymaking in Latin America, Fairfield (2015, p. 11) argues that ‘administrative constraints’ and ‘technical principles’ limit the exercise of structural and instrumental corporate power at national level. In international tax, even at the national level, it is the OECD Model that provides the technical language, norms, standards and guidelines that frame debate. The model certainly delimits the set of acceptable options in the minds of policymakers, to the extent that it is argued to have a ‘soft law’ status (Avi-Yonah, 2007). Grinberg (2017) argues that, ‘at least within the OECD, tax treaty negotiators feel substantially constrained to accept OECD Model Treaty provisions in their future negotiations with other sovereigns’. Sadiq (2012, p. 132) characterizes the process of international tax policymaking in Australia as follows: ‘the Australian Federal Government inherently accepts the existence of an international tax regime and adopts both the international tax policy and practice aspects embodied in that regime through its domestic rules and double tax treaties’. This is also the case outside the OECD, where OECD instruments may be referred to by courts, and where policymakers may feel constrained to follow international best practice (Baistrocchi, 2008; Brauner, 2003; Christians, 2010a; Fjeldstad & Moore, 2008). Nonetheless, OECD tax standards only become hard law when assimilated into the tax codes of individual states (Christians, 2007). Bilateral tax treaties are the main means through which this occurs.

Thus, transnational soft law formation and inter-state treaty negotiations must be considered as interrelated parts of the same process, with multiple entry points for actors seeking to influence the international tax regime. The process of tax treaty negotiation, from the initial policy considerations through to ratification, is guided in almost every country by a small team of specialist bureaucrats, many of whom also take part in the international bodies that formulate the models. For example, the British negotiators whose names appear in the files discussed below are also listed as participants in minutes of the OECD’s Committee on Fiscal Affairs or its OEEC predecessors, and its United Nations equivalent. The same can be said of their private sector counterparts.

Domestic interest group politics meets transnational expertise in the United Kingdom

The rest of this article is based on an analysis of bilateral treaty negotiations between the UK and countries outside of the OECD during the 1970s, using archived civil service documentation. The UK played a very involved role in the development of the League of Nations and OECD model conventions (Avery Jones, 2011), and the UK-US treaty of 1945 is regarded as having set the precedent for modern tax treaties (Graetz & O’Hear, 1997). It has been the most active negotiator of any country, with the widest treaty network in the world (IBFD, no date).

Despite this active role, the preferences of UK negotiators and the content of the OECD model convention have not always been identical. As noted earlier, the OECD model has a bias towards allocating taxing rights to capital exporting states, but it still permits some taxation by capital importing states of some forms of income. During the interwar period, the British expert, Sir Percy Thompson, Deputy Chair of Board of the Inland Revenue, had opposed this compromise at the League of Nations, arguing that only the home state should have the right to tax the foreign income of its multinational taxpayers (Graetz & O’Hear, 1997; Jogarajan, 2017). In the 1950s, the Inland Revenue was opposed to the creation of a fiscal committee at the OEEC, the predecessor of the OECD, but British diplomats concluded that ‘in our position in the organization it would be tactically unwise’ to try to veto it (Ellis-Rees, 1956). Over the next twenty years, the UK participated actively in the elaboration of an OECD model that reflected much, but not all, of its treaty making practice. In an exhaustive technical review of the text of the OECD model and its predecessors, Avery Jones (2011, p. 682) concludes that the UK ‘had only a meagre effect on the final result.’ The enthusiasm with which UK negotiators cleaved to the focal point of the OECD model convention thus reflects their acceptance of a compromise struck in transnational forums over several decades.

This article focuses on the 1970s, the most recent decade for which civil service records are available in the National Archives. During this period, the UK was in talks with at least 38 developing countries, concluding treaties with 16 by the end of 1980. The UK National Archives’ database of files was searched for the terms ‘double tax’ and ‘tax treaty’, yielding 2301 results. The majority of these were country-specific files originated from the Inland Revenue or the Foreign and Commonwealth Office and its predecessors. They include internal civil service correspondence, correspondence between countries, and minutes of negotiation meetings. This means that they include both the internal thinking of the UK and the positioning of the negotiating partner, supplemented on occasions by intelligence about its motivations. There are also some files relating to the UK’s general negotiating position, and quarterly ‘state of play’ reports on all the UK’s negotiations.

This section outlines the roles and motivations of different groups of stakeholders in the decision-making processes surrounding the UK’s tax treaties. First, it examines the preferences of tax treaty specialists in the Inland Revenue, who led negotiations, and their interlocutors in businesses. These actors saw the function of tax treaties as the dissemination of OECD tax standards beyond the OECD. Other actors in the rest of government and business did not share this point of view, and instead focused on the narrow objective of lowering British firms’ effective tax rate. This conflict is illustrated in further detail by analysing conflict between these two groups over the UK’s negotiations with Brazil.

Transnational policy community members: disseminating OECD standards

For officials inside the Inland Revenue, the major causal effect of tax treaties was not, despite their formal title, the elimination of double taxation. The reason for this was that the UK, in common with many other countries, had taken unilateral steps to prevent double taxation of its firms operating overseas, by giving them a credit against their UK tax bill for any taxes paid overseas. Recognition of this dates back at least to 1957, when an Inland Revenue civil servant wrote that with regard to one treaty, ‘the United Kingdom taxpayer gets very little benefit out of it: he will get credit for the tax paid in Colombia against the tax due on the same income in this country whether we have an agreement or not’ (Daymond, 1957). Two decades later, in 1976, a cross-department review of the UK’s approach to international double taxation, led by the Inland Revenue, made the case even more boldly: ‘in the absence of an agreement there is no question of United Kingdom investors being doubly taxed’ (Double Taxation Relief, 1976).⁶

What then was the purpose of a tax treaty for the Inland Revenue? That same note from 1957 records that, for a board of Directors in the UK, ‘the advantages of a double taxation [agreement] need no stressing’ because it ‘at once assures the directors that they will be taxed according to internationally accepted rules and they will not be subject to discrimination.’ (Daymond, 1957). These ‘intangible benefits’ are mentioned by government officials throughout the period under consideration, and according to the 1976 review they ‘include protection against fiscal discrimination, the establishment of a framework within which the two tax administrations can operate, and the expectation that an overseas authority which has negotiated a treaty will at least try to apply it reasonably’ (Double Taxation Relief, 1976).

The Deputy Chairman of the Board of Inland Revenue in 1976 was Alan Lord, who twenty years previously had represented the UK on the new OEEC committee that would eventually become the OECD’s Committee on Fiscal Affairs. According to him: ‘Above all, treaties impose acceptable standards for allocating profits to branches and subsidiaries and for dealing with transfer pricing in countries (some of them within the EEC) where such standards would otherwise be absent’ (Lord, 1967).

For the specialists, tax treaties were tools through which the UK, which had always taken a prominent role in the development of the international tax system, ensured the participation of other countries in it. This would be especially beneficial for British businesses in the case of developing countries, including those newly independent, where, as one official wrote, ‘protection against fiscal discrimination is generally worth more…because they are more likely to include deliberately discriminatory fiscal practices in their general law than are developed countries’ (Wilkinson, 1976a). A memo from as early as 1949 expresses the view that ‘The United Kingdom particularly has much to gain from the increasing adoption, particularly by under-developed countries, of sound principles of income taxation and from the conclusion on sound lines of conventions for the relief of double taxation’ (Morton, 1949).

Treaties were therefore understood as means to ensure that British firms could be competitive when they decided to invest, rather than to make investment in the treaty partner more attractive in the first place. This would mean that treaties increased investment from the UK to the treaty partner, but not usually by influencing business decisions; rather, they gave British investors a helping hand.

The effect of treaties on outward investment from the UK was not a trivial matter during the 1970s, but an important policy question. Treasury policy was to limit the impact of outward FDI on the balance of payments by encouraging it to be done out of retained earnings, investment currency or foreign currency borrowing. In 1973, at a meeting of the cross-Whitehall Tax Reform Committee handling changes to corporation tax, a Treasury official argued against measures that would prioritize overseas investment, because of the effect on the balance of payments. The concern was about foreign exchange reserves, which could be protected more through income from exports than from direct investment; furthermore, the likely shift in manufacturing abroad as a result of overseas investment would increase imports (Hopkins, 1973).

Discussing this point, the 1976 review concluded that the treaty network at that point ‘neither encourages nor discourages overseas investment in fiscal terms compared with domestic investment, except where matching credit is provided’ (Double Taxation Relief, 1976). At around this time the Inland Revenue was arguing against conceding Brazil’s demands for more comprehensive concessions in a tax treaty on the grounds that the concessions, ‘would mean that we were according outward investment a higher priority than hitherto with all that that implied for the balance of payments and the domestic economy’ (Hubbard, 1974).

The community of tax specialists who shared this analysis and these objectives was not limited to the Revenue itself, in at least one respect: it extended into the private sector. In December 1971, Alan Davies of Rio Tinto Zinc, chair of the CBI’s tax committee, wrote to Alan Lord. The letter outlined the limitations of the Revenue’s current approach to consultation, which was to solicit comments from industry by letter once negotiations were initiated. Davies cited ‘a peeved feeling on our side that some more confidence would be justified,’ and argued for more informal discussion about the progress of negotiations (Davies, 1971). The informal tone of Davies’ letter perhaps reflects a personal familiarity with the Inland Revenue officials concerned. For example, he attended meetings of the United Nations Ad Hoc Group of Experts on Tax Treaties between Developed and Developing Countries, representing the International Chambers of Commerce, as did Inland Revenue negotiators (United Nations, 1969, 1970).

The result was a system of regular quarterly meetings between the Inland Revenue and tax specialists from industry groups at which detailed information on the ‘state of play’ in negotiations was divulged, and comments sought on specific topics (Minutes of Meetings with Confederation of British Industry on double taxation, 1971–1981). The first such meeting took place in March 1972, and they continued for at least the next decade. At each meeting, the Inland Revenue participants were supplied with a status report on current and planned negotiations, which they shared verbally with the business representatives on condition that the information was not shared outside of the small, expert group. When negotiations reached a difficult point, the matters of contention would often be discussed in this forum.

Conflict with political actors

Here I consider the preferences of non-community members, for whom tax treaties were also tools to increase the competitiveness of British firms abroad. A lack of detailed taxation knowledge, frequently lamented both by them and by the specialists, would lead to conflicts, during which the Revenue would sometimes try to persuade them that their faith in the effect of tax treaties was misplaced. ‘There can be little doubt that tax treaties are a means of stimulating trade and investment between the treaty partner countries,’ wrote the private secretary to the Treasury minister responsible for tax policy in 1976. ‘On the other hand their importance is sometimes exaggerated’ (Wilkinson, 1976b). The UK’s lead negotiator noted in 1974, referring to Brazil, that,

we should not over emphasise the importance of a DTA. It generally only affects income flowing from one country to another whereas in the short term a company will not remit much in the way of profits and will not be too bothered in the absence of an agreement (Note of Meeting, 1974), emphasis in original.

Most civil service non-specialists who engaged with tax treaty matters during the 1970s wanted British firms that were eligible for investment-promoting tax relief in developing countries to receive a corresponding credit (often referred to as ‘tax sparing’ credit) against UK tax, to ensure that they could retain the benefit of the tax relief when they repatriated their profits. As the 1976 review notes, in outlining the priorities of different departments, ‘the main cash benefit for the investor [from a tax treaty] is matching credit for pioneer reliefs’ (Double Taxation Relief, 1976). The difficulty was that this was not the Inland Revenue’s priority from tax treaties, and at times (as the case of Brazil illustrates) the two priorities even came into conflict.

The Inland Revenue sought to keep input from other departments limited and compartmentalized, and did not welcome their attempts to influence its priorities. The Treasury, Departments of Trade and Industry, and Foreign Office would each be consulted on treaties once negotiations were opened, and on specific questions concerning their content, but the Revenue would often rebuff their requests to be able to influence its priorities.

During late 1972 and 1973, an extraordinary correspondence opened up between the FCO and the Board of Trade on one hand, and the Inland Revenue on the other. The former were frustrated by their inability to influence the latter’s negotiating priorities. At a cross-Whitehall meeting in April 1972, the Revenue had merely invited them to submit ‘shopping lists’ for treaties they would like it to negotiate (Note of Meeting, 1972). ‘We have already forfeited opportunities for investment in Brazil, notably to the Germans and Japan and, as a matter of commercial policy, it is important that we should not place our traders at a disadvantage when seeking out investment opportunities in the future,’ argued one official from the Board of Trade in February 1973 (Gill, 1973). He continued that:

As you know, we have been concerned that the corporation tax system should not so limit the scope for tax sparing as to damage the UK’s ability to export to and invest in developing (and highly competitive) overseas markets. For this reason, we place great importance on the conclusion, as quickly as possible, of double tax agreements with our developing trading partners which allow for tax sparing.

The Revenue resisted this pressure, refusing even to share a list of current negotiating priorities or negotiations that were underway, because ‘a high degree of confidentiality attaches to our negotiations with particular countries’ (Smallwood, 1973). The reference to confidentiality is revealing, because this correspondence took place at the same time as the Revenue had begun quarterly meetings with tax specialists from businesses, at which exactly this information was disclosed.

‘I find the Inland Revenue’s attitude and behavior quite extraordinary,’ wrote an official in the FCO’s financial relations department, as part of correspondence that passed between these other departments. ‘I cannot imagine that any other department in Whitehall would behave in this way. Nor would we have allowed any other Department to get away with behavior like this for quite so long. I am quite clear we must call a halt now’ (Kerr, 1973a). Another lamented ‘a dispiriting and unfruitful confrontation with the Inland Revenue’ (Baillie, 1972). The problem for the FCO, in particular, was that it lacked a coherent position within itself, and the technical expertise to develop one. ‘The subject is difficult and mastering it is undoubtedly time-consuming’ mused one FCO official (Kerr, 1973b).

It was not only officials from other departments who had trouble influencing Inland Revenue officials: their own ministers faced the same problem. In general, politicians had little involvement in tax treaties at all. At the start of the 1970s, negotiators worked within enabling powers set by parliament, and would only seek ministerial guidance when making a concession that had not previously been given in negotiations. There seems to have been no political involvement in the decision with whom to negotiate, and the minister in charge, the Financial Secretary to the Treasury, did not usually have sight of a treaty until bringing it before parliament for ratification.

The technical complexity of tax treaties was inevitably a barrier to effective political scrutiny, but this must surely have been combined with the short tenure of Financial Secretaries: eleven different people occupied the position during the 1960s and 1970s, with an average tenure of two years.⁷ The longest serving, Robert Sheldon, was in post from February 1975 to April 1979. The archives demonstrate the difficulty faced by a minister trying to exert some influence over a policy area with which he was unfamiliar. During the mid-1970s, the UK had been seeking to amend its treaties to reflect changes to its corporation tax system. The civil servant who first briefed Sheldon commented that:

I got the impression that he does not realise – or did not until I pointed it out to him – that double taxation agreements also deal with other matters than dividends…. he seemed surprised when I told him we had sixty plus agreements in operation (Collins, 1975).

This lack of understanding is also apparent in the minute of the May 1976 meeting. Sheldon questioned ‘what the OECD Model was and what we would do if it turned out not to provide an advantageous pattern for the UK’ (Note of Meeting, 1976). This question illustrates that the OECD model’s status among officials as a focal point in negotiations was not shared by the Minister who supervised them, after more than a year in post.

A third category of non-specialist stakeholder was those within businesses, who were able to influence the positions of other parts of government including the FCO and DTI, but rarely to translate this into treaties. Geographic departments in the FCO, in particular, were often persuaded by businesses, which lobbied British embassies, to advocate new British tax treaties. For example, ‘UK finance houses and business interests are adamant that we are losing a significant amount of business in Spain because there is no double taxation agreement,’ wrote an official in the FCO’s Southern Europe department (Baillie, 1973). These positions fed into the central FCO departments, in particular the economists’ department and financial relations department, which as we have seen were furious that the Inland Revenue would not heed their concerns about the competitiveness of British businesses. Meanwhile, the Inland Revenue seemed content to divide and rule the geographical departments.

Business lobbying via these departments met with limited success, partly because those other parts of government had limited influence on the Revenue, but also because one part of the private sector undermined the other, a fault line that sometimes ran within, rather than between, businesses. A memo from the CBI to the Department of Trade and Industry in 1974, covering a wide range of policy and not written by tax specialists, states that tax treaty ‘negotiations should not be left exclusively to the Inland Revenue (whose main concern is naturally the minimization of losses to the Exchequer)’ (CBI, 1974). A year later, an Inland Revenue official was ‘subjected to a 2 h intense grilling’ by CBI representatives who were not tax specialists at a cross-Whitehall consultative meeting. They had apparently suggested that future negotiations for double taxation agreements would better be dealt with by a department other than the Inland Revenue since the negotiations were currently carried out for the United Kingdom by narrow specialists who were so blinkered by the technicalities of taxation that they failed to see the full view of the picture (Minutes of a meeting with Representatives of the Confederation of British Industry, 1975).

The CBI delegation also expressed the view that the Inland Revenue’s consultation with tax experts from industry ‘was not really satisfactory since it was restricted to “taxmen” (ibid). Those ‘taxmen’ felt obliged to apologize for their colleagues’ actions in subsequent discussions with the Inland Revenue (ibid; Moran, 1975b).

To summarize the argument so far, the UK’s national preference was formed by civil servants in the Inland Revenue, informed primarily by the lobbying efforts of fellow tax specialists within business. For these two groups, all members of the international tax community, treaties were a means to disseminate ‘acceptable’ tax standards, which would protect UK businesses from what they saw as unfair taxation. This was a long-term project that required sticking closely to the focal point of the OECD Model. Their efforts were held in tension with the interests of other elements of British businesses, who lobbied elsewhere in government, successfully converting civil servants and Ministers to their cause, but largely failing to convince the Inland Revenue. For them, tax treaties were a means to secure tax sparing credits and other concessions that would reduce the effective tax rate of British firms, enhancing their competitive position. For neither group were the aims of eliminating double taxation and retaining the UK’s share of the tax base, supposedly the driving force behind the creation of the international tax regime, the main motivating factors.

The case of the UK–Brazil treaty

The UK devoted far more time and effort to negotiations with Brazil during the 1970s than almost any other developing country, and yet an agreement was never concluded. Talks in 1967 had failed, but they were taken up again from 1972, urged by British businesses keen to benefit from Brazil’s ‘economic miracle’ (Double Taxation Relief Agreement, 1974).⁸ In 1976, however, talks were suspended in spite of the fact that, as the Inland Revenue acknowledged, ‘it is British investors who will be the sufferers’ (Wilkinson, 1976b). It is hard to explain this from a simple state-centric, domestic interest group perspective, since there was strong pressure from business lobby groups for an agreement, and little interest group opposition. A transnational expertise perspective appears more suited to the case, since the British position was that it could only conclude an agreement with Brazil ‘providing for significant amelioration of aspects of their tax code that run clearly counter to OECD principles; and if they are not interested, so be it’ (Wilkinson, 1976b). This nonetheless poses the question why non-expert interest group pressure was ineffective in the UK, when several other OECD countries had accepted Brazil’s non-OECD terms.

The stalemate between the UK and Brazil concerned two unconventional demands by the latter. It insisted that the UK grant extensive ‘tax sparing’ concessions. This would mean crediting the value of a Brazilian tax exemption against the UK company’s tax bill as if it had paid full Brazilian tax. It would also mean doing the same for the reductions in taxes on cross-border payments that were built into the treaty. In the reported words of a Brazilian negotiator, ‘whilst Brazil does not want the United Kingdom to lose tax, she cannot allow the United Kingdom to collect more tax as a result of the convention’ (Note of Talks in Brazilia, 1974). Such a concession required new legislation in the UK, to which the UK eventually conceded in 1976.

The second Brazilian demand was more difficult, however. Under Brazilian domestic law, firms had to pay a withholding tax on the gross value of any royalty paid to a foreign recipient. Unusually, however, they were not then permitted to deduct the value of the royalty payments when calculating their net profits. As a result, they effectively paid tax on the payments a second time in Brazil, through income tax. While the UK’s unilateral double tax relief system gave investors a credit for taxes paid abroad, the high effective rate exceeded this credit, and so the company bore the cost, reducing its competitiveness. Brazil insisted that any double taxation agreement leave this state of affairs intact, though it was in direct contravention of the OECD model’s provisions. As noted in an Inland Revenue memo, several OECD countries had reached agreements with Brazil that permitted this practice to continue, because other concessions obtained in treaty negotiations, such as lower withholding tax rates, gave their firms a competitive advantage (Wilkinson, 1976b). This both increased the pressure on the Inland Revenue from British businesses and reduced its leverage in negotiations with Brazil. British companies ‘are undoubtedly at a competitive disadvantage as compared with companies from other countries,’ noted a background brief in August 1974 (Double Taxation Relief Agreement, 1974).

The pressure from businesses did not come directly on the Inland Revenue, but via other ministries. In October 1974, a memo from the Department of Industry to the Inland Revenue pressed the case for a treaty, citing ‘specific evidence of orders being lost by British companies apparently because of their relatively lower post-tax returns forcing them to quote higher prices in compensation.’ With no movement by December, the Department of Trade weighed in, beginning a correspondence between its Secretary of State, Peter Shore, and Chancellor of the Exchequer Denis Healey (Shore, 1974).⁹

As the pressure from business lobbyists on other government departments ratcheted up, tax specialists within British businesses reassured the Revenue that they were broadly in agreement with its view that the Brazilian terms were unacceptable (Harvey, 1976). A note of a meeting between tax specialists at the Inland Revenue and CBI records how a CBI representative was

well aware of the powerful trade and political pressures in favor of having an agreement (apparently any agreement) with Brazil which he thought could lead to an explosion in the autumn. His personal view was that the Revenue and Treasury Ministers could be under pressures from other Ministers which might lead to an agreement, in spite of the unsatisfactory features that had been discussed. Much of the pressure is based on ignorance of the effects of unilateral relief and of the likely terms of a treaty (Smallwood, 1975, emphasis in original).

Minutes of the meeting and a follow-up letter from the CBI record the industry tax experts’ frustration at being unable to correct their colleagues’ ‘ignorance’ because of the confidential nature of their meetings with the Inland Revenue (Moran, 1975a). Inland Revenue memos contrast the ‘non-fiscal voices’ within the CBI with those of ‘the CBI’s Tax Committee, as a Committee of tax experts’ (Wilkinson, 1976b) and observe that, ‘the CBI will no doubt have to consider how to deal with the situation in which it is speaking with two voices’ (Smallwood, 1975).

In 1976, British negotiators were able to travel to Brasilia with their new legislative mandate on tax sparing and instructions ‘to refrain from agreeing to the unacceptable features of Brazilian law which they wish to enshrine in the treaty, but to avoid a breakdown in the talks’ (Crosland, 1976). The Brazil files stop at the turn of the 1980s, but the same debate continues. In 1992, an Inland Revenue official wrote that ‘Brazil continues to be the big prize: but it is not ripe for an immediate approach and what indications there are suggest that it will be a difficult nut to crack’ (Shepherd, 1992). The absence of a treaty with Brazil is still raised by British business lobby groups today, and the UK and Brazil still do not agree on terms (House of Commons, 2014).

The debate over the UK-Brazil tax treaty illustrates that the preferences and instrumental power of corporate capital in the UK were not monolithic, but varied depending on technical knowledge. Had the aim of the UK’s tax treaty negotiations been simply to give British firms a competitive edge by lowering their effective tax rate in Brazil, as British firms in Brazil were lobbying for via the British embassy and the Departments for Trade and Industry, an agreement would have been possible. But it would have come at the cost of implicitly endorsing Brazil’s approach to taxing royalty payments. This would have undermined the longer-term project of exporting norms embodied in the focal point of the OECD model, which motivated members of the international tax community both in the Inland Revenue and in businesses.

Conclusions

The expansion of the UK’s tax treaty network to non-OECD countries during the 1970s was part of a major growth of the international tax regime. The state-centric view of the regime’s development explains this expansion as a response to developing countries’ concerns that double taxation might undermine their efforts to attract British investment. Yet investors’ concerns about double taxation did not determine the UK’s preferences. The critical law account (Dagan, 2000; Irish, 1974), in which the UK is motivated by a desire to capture a greater share of tax revenues from British companies operating in developing countries, is also inconsistent with the evidence. Rather, officials from the public and private sector who had been complicit in the creation of OECD standards at transnational level used tax treaties as means to disseminate those standards. More than eliminating double taxation, the aim was to constrain other countries’ ability to tax inward investment. This long-term project required a disciplined focus on establishing the hegemonic position of OECD standards, even if this meant abandoning negotiations that British businesses were lobbying for, but which could only be concluded at the cost of the purity of the OECD standards.

Familiarity with this focal point determined the preferences of individual actors, but it also determined their capabilities to translate those preferences into the government position. Private sector officials who were members of the transnational tax community were influential in decisions made by the Inland Revenue. Information readily supplied to them was at the same time withheld from government officials from other departments, and from business actors without a tax specialist background. Even the government minister supervising tax officials was unable to exert influence because he lacked familiarity with transnational norms. This explanation based on bureaucratic politics and technical knowledge concurs with Poulsen’s (2015) account of the diffusion of bilateral investment treaties, but shifts the focus onto capital-exporting states.

Since the 1970s, international tax standards have become considerably more complex, but the OECD model bilateral tax treaty remains at the heart of current reforms to tackle tax competition (OECD, 2013). The challenges encountered in the process of these reforms partly reflect the conflicting preferences of different groups of states (Eccleston & Smith, 2016; Grinberg, 2016; Hakelberg, 2016), and partly reflect the politics of tax expertise within the transnational policy community (Buttner & Thiemann, 2017; Picciotto, 2015; Seabrooke & Wigan, 2016). This article has demonstrated that these two streams of scholarship need to be merged, to consider how expertise politics influences national preferences.

Recognizing the role of expertise as well as interest group politics in deriving national preferences opens up a new perspective on the development of the international tax regime, and calls for a re-examination of corporate power in international taxation. The implicit assumption of interstate bargaining models is structural corporate power in the form of market pressure on states to compete for inward investment (Arel-Bundock, 2017; Barthel & Neumayer, 2012). If bilateral tax treaty negotiations are a continuation of transnational standard-setting on which business actors exert some influence, and if businesses actors’ preferences and capabilities at national level vary on the basis of their expertise, then a more comprehensive view of the exercise of corporate power is needed.

Acknowledgments

I‘m grateful to Lucy Barnes, Jeff Chwieroth, Florence Dafe, Lukas Linsi, Jason Sharman, John Vella and Zoe Williams for their comments on drafts of this paper, as well as to the editors of RIPE and three anonymous reviewers.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Martin Hearson

Martin Hearson is a Fellow in International Political Economy in the International Relations department of the London School of Economics and Political Science. His research focuses on international tax governance and the relationship between developed and developing countries.

Notes

1 An exception is the US, where domestic political actors have repeatedly blocked participation in OECD initiatives (Hakelberg, 2016; Sharman, 2006; Webb, 2004).

2 The evidence for an investment-promoting effect in developing countries is, however, mixed (Barthel, Busse, & Neumayer, 2009; Davies, Norbäck, & Tekin-Koru, 2009; Lejour 2014; Sauvant & Sachs, 2009).

3 The evidence for the investment-promoting effect in developing countries is, however, mixed (Barthel et al., 2009; Davies et al., 2009; Lejour, 2014; Sauvant & Sachs, 2009).

4 Interview, Amsterdam, 2015.

5 Skype Interview, 2016.

6 There are exceptions illustrated in the files. For example, businesses that incurred withholding taxes on management fees abroad would find that, absent a treaty, these tax payments would not qualify for a credit against UK tax.

7 According to biographies on the UK parliament website, tenure during the period covered by this chapter was as follows: Dick Taverne, 1968–1970; Bernard Jenkin, 1970–1972; Terence Higgins, 1972–1973; John Gilbert, 1974–1975; Robert Sheldon, 1975–1979; Nigel Lawson, 1979–1981.

8 The same civil service files include a clipping from the Financial Times discussing Brazil’s “economic miracle”.

9 In 1974 the Department of Trade and Industry was split into two separate departments.