• Specific Year
    Any

Taylor, C John --- "Twilight of the Neanderthals, or Are Bilateral Double Taxation Treaty Networks Sustainable?" [2010] MelbULawRw 8; (2010) 34(1) Melbourne University Law Review 268

TWILIGHT OF THE NEANDERTHALS, OR ARE BILATERAL DOUBLE TAXATION TREATY NETWORKS SUSTAINABLE?

C JOHN TAYLOR[*]

[The networks of bilateral double taxation treaties that have developed since the Second World War are regarded by many as being very successful. Double taxation treaties had their origins in treaties entered into by continental European countries prior to the First World War. Reflecting these origins, bilateral double taxation treaties have adopted a schedular approach with different rules for different types of income. When relieving residence–source double taxation using the credit method, double taxation treaties adopt a measured approach under which tax paid in the source country is relevant to the determination of the nature and extent of the relief granted by the residence country. Critics of bilateral treaties have argued that their schedular structure does not represent a principled approach to the problem and that their measured approach means that relief they provide through a credit method becomes highly complex. Through an examination of Australian practice of treaty negotiation and administration this article identifies several practical reasons for the success of bilateral double taxation treaties despite their allegedly less principled and arguably complex approach. The article then examines threats to the sustainability of bilateral double taxation treaty networks. Having regard to these threats to the sustainability of bilateral double taxation treaty networks the article then discusses possible responses to these threats, including unilateral adoption of a composite tax system providing notional relief and the multilateral adoption in trading blocs of such a tax system.]

I INTRODUCTION

The title to this article possibly needs some explanation. My understanding of current research is that homo sapiens neanderthalis was a highly successful large-brained human well suited to the environment they evolved in and which dominated Eurasia for nearly 200 000 years. Then approximately 45 000 years ago their range became significantly restricted as they faced pressures from a deteriorating climate and competition from anatomically modern humans. By 30 000 years ago they seem to have disappeared entirely. There are various theories as to why they disappeared (some of which involve our species, homo sapiens sapiens, in what should probably be called genocide) but among them is that they were not able to adapt quickly enough to the combination of climatic change and competition from our ancestors.[1]

Similarly, the networks of bilateral double taxation treaties that have developed since the Second World War are regarded by many as being very successful. The form of these treaties can be traced back to treaties entered into by continental European countries prior to the First World War and to drafts developed by the League of Nations between the First and Second World Wars. Critics of bilateral treaties have argued that their schedular structure does not represent a principled approach to the problem of international double taxation and that their measured approach means that the relief they provide through a credit method becomes highly complex.[2] This article argues that networks of bilateral double taxation treaties currently face a series of threats to their sustainability and that many of these threats are products of their schedular structure and measured approach to relief and to their inability to adapt rapidly enough to changed circumstances.

The article begins by tracing the origins of modern double taxation treaties and argues that many of their significant structural features are products of those origins. The article contrasts the structural features of modern double taxation treaties with the system of Dominion Income Tax Relief that operated within the British Empire, from 1920 until 1946 in the case of Australia, and until later dates in some instances. The article observes that the global approach to relief used in the system of Dominion Income Tax Relief would appear to be a more logical and principled solution to the problem of international juridical double taxation than the schedular approach adopted in modern double taxation treaties.

The article then asks why the system of bilateral double taxation treaty networks flourished after the Second World War while the Dominion Income Tax Relief system disappeared. Through an examination of the Australian practice of treaty negotiation and administration, this article identifies some practical reasons for the success of bilateral double taxation treaties despite their allegedly less principled and arguably more complex approach. This section of the article also examines the history of the administration of the system of Dominion Income Tax Relief and identifies reasons for its demise.

The article then briefly examines threats to the sustainability of bilateral double taxation treaty networks: (a) electronic commerce; (b) financial engineering; (c) tax competition; (d) Directives by the European Union Commission and the interpretation of the Treaty on European Union (‘EU Treaty’)[3] by the European Court of Justice (‘ECJ’); (e) transfer pricing problems; and (f) the lengthy time periods needed to modify an entire double taxation treaty network. Having regard to these threats to the sustainability of bilateral double taxation treaty networks the article then discusses one possible response to these threats; namely the multilateral adoption in trading blocs of a composite tax system providing notional relief (an updated and modified version of the system of Dominion Income Tax Relief).

II THE ORIGINS AND STRUCTURAL FEATURES OF MODERN DOUBLE TAXATION TREATIES

Although double taxation treaties in some form can be traced back to earlier periods (ancient Greek city states, for example, had agreements to prevent some forms of double taxation between them) the origins of modern double taxation treaties can realistically be found in treaties negotiated, largely, but not entirely, between continental European states in the late 19th and early 20th centuries[4] and in work done by the League of Nations between the end of the First World War and the beginning of the Second World War.[5]

The current structure of modern double taxation treaties reflects these origins. One of the first things that a reader familiar with the tax laws of a system — such as the United States’ system, which uses a ‘global’ approach to the taxation of income[6] — notices when reading double taxation treaties for the first time is that treaties are based on classifying income into different categories and on being able to give distinct tax treatment for computational purposes to each of these different categories of income. This structure is reflective of the fact that, in the 1920s when the League of Nations was developing its model treaties, the income tax systems of most continental European countries were schedular systems that at least included, and in many instances were confined to, impersonal taxes. Impersonal taxes were taxes on distinct categories of income (such as land, business profits and so on) sourced within the jurisdiction.

At the time of the Report on Double Taxation Submitted to the Financial Committee by Professors Bruins, Einaudi, Seligman and Sir Josiah Stamp (‘Report of the Four Economists’) for the League of Nations in 1923, only Britain, the United States, Germany and the Netherlands (partly) were regarded as having personal income taxes.[7] The Double Taxation and Tax Evasion: Report and Resolutions Submitted by the Technical Experts to the Financial Committee of the League of Nations (‘1925 Technical Experts Report’) noted that ‘[m]ost of the other nations of Europe and America seem to be moving slowly but definitely in the [direction of levying a personal tax].’[8] For example, Italy had introduced a graduated income tax which was to come into force in 1925.[9] Nonetheless the 1925 Technical Experts Report observed that previous treaties had been concluded by states for which impersonal taxes were ‘the more important — if not almost the only — system of taxation in force.’[10] As will be discussed in more detail below, these treaties dealt with the problem of international double taxation by assigning exclusive rights to tax particular categories of income to one or another of the contracting states. This approach is understandable given that all of these countries principally relied on schedular impersonal taxes levied on a source basis. The allocation of taxing rights on a class of income basis effectively amounted to a series of source rules, thus resolving potential conflicts of source rules which were the most likely cause of international double taxation between countries using source-based schedular impersonal taxes.

Britain, which taxed income on a worldwide basis and determined corporate residency on the basis of central management and control, was isolated within the League of Nations Committee of Technical Experts in favouring taxation by the country of residence only, as others have observed.[11] The United States, which also taxed residents on a worldwide basis under a global concept of income, was never a member of the League of Nations, but from 1926 onwards sent representatives as ‘observers’ to the Committee of Technical Experts.[12]

The following two sections will examine the history of the development of the structural features of modern double taxation treaties in more detail.

A The League of Nations Committees and the Structure of Modern Double Taxation Treaties

In 1921 the Economic and Financial Committee of the League of Nations, in response to submissions by the International Chamber of Commerce, commissioned four economists, Professors Bruins (the Netherlands), Einaundi (Italy), Seligman (the United States) and Sir Josiah Stamp (the United Kingdom) to report on the problem of double taxation. The Report of the Four Economists argued that competence to tax a person should correspond with the person’s economic allegiance,[13] the most important determinants of which were the origin (or source of the wealth) and the domicile (or residence) of the consumer of the wealth.[14] In the view of the economists, the importance of these factors varied between different types of income, with origin being more important in the case of income from land, commercial, industrial and agricultural enterprises and, so far as a tax on capital was concerned, mortgages.[15] Conversely the economists saw domicile as being the more important factor in the case of moveable property, transferable securities, general credits and personal earnings, and, in so far as a tax on income was concerned, mortgages.[16] The economists envisaged that a global personal tax should be exclusively levied on a residence basis and that allocation of taxing rights in the case of impersonal taxes should be based on the concept of economic allegiance.

The Report of the Four Economists then identified four possible ways in which double taxation could be prevented.[17] The first was an unlimited foreign tax credit as had originally been applied in the United States. The second was an exemption by the country of origin of all income of non-residents. The third was a system, like the system of Dominion Income Tax Relief operated by Britain, which divided the tax and allocated the relief given between the two states. The fourth was the division of sources of income between the two countries so that each country taxed the sources of income assigned to it. Advantages and disadvantages of each of these methods were noted by the four economists.[18] The relative contributions of the economists to the Report of the Four Economists appear to have been, in descending order: Seligman, Stamp, Bruins and Einaudi.[19]

As Graetz and O’Hear have rightly observed, the principal authors of the Report of the Four Economists were thus from capital exporting creditor nations and their report reflected the interests of those nations.[20]

Prior to the release of the Report of the Four Economists, the Secretary-General of the League of Nations, following a recommendation of the Financial Committee, requested a number of European countries to nominate a technical official to be part of a committee to examine double taxation and fiscal evasion.[21] Seven European countries nominated representatives and the committee held five meetings from June 1923 to February 1925.[22] The Committee of Technical Experts took into account the Report of the Four Economists and the terms of double taxation treaties previously entered into, particularly the most recent treaty between Italy and Czechoslovakia concluded in 1924.[23] The Committee characterised the previous work of the four economists as an examination of the topic from ‘the theoretical and scientific point of view’ and regarded itself as being ‘asked to consider the administrative and technical aspects of the question.’[24]

The 1925 Technical Experts Report noted that the double taxation treaties concluded up to that time had all used the fourth method of relief identified in the Report of the Four Economists, namely, assignment of taxing rights to different classes of income according to the country of origin.[25] The 1925 Technical Experts Report examined the 1924 Italy–Czechoslovakia Treaty in some detail and noted it applied the approach of ‘classification and assignment’ of taxing rights to income not only to impersonal taxes but also, ‘within narrower limits’, to personal taxes.[26] Under the treaty, income from immoveable property, income from mortgages, income from an industry or business (other than one carried on by a joint-stock company), and earned income were taxed on an origin basis, with all other types of income that were subject to personal taxation being taxed on a domicile or residence basis.[27] The 1925 Technical Experts Report noted that this approach meant that personal taxes were subject to ‘lighter taxation’ by the country of domicile as the assignment of exclusive taxing rights to some classes of income to the source country narrowed the residence country’s tax base.[28] The 1925 Technical Experts Report explained the application of the assignment approach to personal taxes in the 1924 Italy–Czechoslovakia Treaty as follows: ‘This is, no doubt, intentional, because the treaty was to be concluded between States in which the “real” taxation of income was the more important — if not almost the only — system of taxation in force.’[29]

The assignment of taxing rights in the 1924 Italy–Czechoslovakia Treaty bears some similarity to the approach taken in modern double taxation treaties following the Organisation for Economic Co-Operation and Development Model (‘OECD Model’). A key difference, however, is that the assignment of rights in relation to each particular category of income to either the source or residence country is always exclusive. By contrast, in relation to several categories of income, the OECD Model limits but does not eliminate the source country’s taxing rights, and allows the residence country to tax that income if it allows a foreign tax credit. While the approach taken in the 1924 Italy–Czechoslovakia Treaty might have been suitable for two countries that regarded impersonal taxes as more important than personal taxes, the reduction in the tax base of the personal tax, with consequent effects on the applicable marginal rates that it involved, would not have been acceptable to countries such as the United Kingdom and the United States, whose income taxes were confined to personal taxes levied on a worldwide basis.

The 1925 Technical Experts Report recognised difficulties with the approach in the 1924 Italy–Czechoslovakia Treaty and claimed to ‘have discovered the solution in refraining from suggesting any one single system as applicable to every form of taxation.’[30] Hence the 1925 Technical Experts Report made different suggestions in relation to impersonal and personal taxes.[31] It generally gave the country of origin the exclusive right to levy most impersonal taxes and gave the country of domicile the right to levy personal taxes.[32]

The resolutions of the Committee of Technical Experts also allowed the source country to levy personal taxes on particular sources arising within that country including income from immoveable property and income from agricultural, industrial or commercial establishments other than dividends from shares.[33] Where the source country did levy personal taxes on a source basis the resolutions recommended that ‘bilateral conventions should, if possible, be entered into … with a view to avoiding any double [taxation]’.[34] While the resolutions stated that the ‘precise method of avoiding double taxation must be a matter to be worked out in detail between the States concerned’, the experts indicated two methods which they considered might be of assistance to states entering into such conventions.[35] The two alternatives were either: (a) a foreign tax credit with an overall limitation, with the credit being calculated either by reference to the tax payable on the foreign income under the residence country’s rates or by reference to the foreign tax actually paid; or (b) that the total income be apportioned between the source and residence country but with the rate applied by the residence country being determined by reference to the taxpayer’s worldwide income[36] (in some respects this appears to be a variation on the approach applied by the United Kingdom under its Dominion Income Tax Relief system, discussed below).

The use of different models for impersonal and personal taxes might have been workable where both contracting states only or predominantly had either impersonal taxes or personal taxes — but what of the position where one state relied primarily or exclusively on impersonal taxes while the other state only imposed a personal tax? The 1925 Technical Experts Report recognised that particular problems would arise in the latter situation and included a resolution to the effect that either the origin state would exempt income or that the residence state would employ one of the methods described in the resolutions relating to personal tax.[37] The 1925 Technical Experts Report noted that here ‘the decisive factor will be the relative importance in each particular case of the interests involved’ and noted that progressive rates applied to personal taxes were likely to mean that origin countries made smaller sacrifices.[38] The proposed solutions (other than exemption by origin country), however, worked on the basis that the residence country would only have the limited rights discussed earlier to apply a personal tax to the domestic source income of non-residents. In reality, both the United Kingdom and the United States applied their personal taxes on both a residence and a source basis. It appears that the United States in particular was concerned that the approach adopted in the 1925 Technical Experts Report would mean that it would be unable to tax or would at least be unduly restricted in its taxation of non-residents on United States sourced income.[39]

The second report of the Committee of Technical Experts in 1927[40] reflects the influence of the United States observers who joined it in 1926. The United States’ objective was to use double taxation treaties as a means of lowering, but not eliminating, source taxation, thus leaving the residence country a residual right to tax after allowing a foreign tax credit.[41]

This report produced four draft conventions, but only the first of these drafts, Draft of a Bilateral Convention for the Prevention of Double Taxation (‘Draft I’ or ‘Draft IA’), will be discussed in this article.[42] Draft I gave the source country the right to levy impersonal taxes and effectively assigned source rules for each category of income, and gave the residence country the right to levy personal taxes. The latter right was subject to giving a foreign tax credit calculated as the lesser of the tax that would be payable in the residence country on the income taxable in the source country or the actual tax paid in the source country for personal and impersonal taxes imposed by the source country. Draft I provided that the total credit given would not exceed a specified percentage of the total personal tax leviable in the residence country.[43] As Avery Jones has observed, the current schedules in the OECD Model essentially correspond with those identified in Draft I of 1927.[44]

As Picciotto notes, Draft I ‘suited only states which accepted the distinction between personal and real taxes.’[45]

The 1927 Technical Experts Report had envisaged that their model could also be used where the residence country had a personal tax while the source country levied schedular impersonal taxes.[46] It had also stated that the text of Draft I ‘could be abridged’ where the two contracting states had ‘sufficiently similar’ tax systems.[47]

During the 1928 League of Nations General Meeting of Government Experts,[48] two further alternatives to Draft I were developed, neither of which made a distinction between personal and impersonal taxes. Convention No 1B, which was designed for two countries each using a global personal tax,[49] allowed taxation at source of income from immoveable property and the business profits of a permanent establishment but otherwise gave exclusive taxing rights to the state of fiscal domicile (defined as the taxpayer’s ‘normal residence’, or ‘permanent home’).[50] The residence country was required to grant a foreign tax credit for any taxes levied by the source country.[51]

By contrast, Convention No IC, which was designed for use in the situation where one country used a global personal tax while the other used a schedular system that included impersonal taxes,[52] gave the source country the exclusive right to tax income from immoveable property and the business profits of a permanent establishment but also permitted the source country to tax income from loans and shares.[53] It required either that the residence country provide a foreign tax credit for the source tax or that the source country provide a refund to prevent double taxation.[54] Avery Jones observes that the current OECD schedules can be traced back to the 1927 and 1928 drafts.[55]

It is notable that, while Convention No 1B and Convention No 1C sought to limit source country taxing rights, they did so by limiting the classes of income that the source country was permitted to tax rather than by imposing caps on the tax rates that the source country could apply to income generally or in particular categories.

The influence of Draft IA can be seen in several treaties entered into by continental European countries, where both countries had both personal and impersonal taxes, prior to the Second World War.[56] Here it is important to bear in mind that, where a continental European country levied both personal and impersonal taxes, the personal tax operated as a residual tax which excluded anything taxed by the impersonal tax. Hence the League of Nations Draft IA and treaties based on it amounted to a compromise under which the source country had the right to impose impersonal taxes while the residence country was able to tax its residents (but not non-residents) through the personal tax on any items to the extent that they were not covered by the impersonal taxes imposed by the source country. Treaties entered into by continental European countries after 1928 but prior to the Second World War, where one country distinguished between personal and impersonal taxes while the other did not, continued to adopt a schedular approach to the allocation of taxing rights under which rights to tax particular classes of income, including the personal tax, were exclusively allocated to either the source or residence state.

Hence, these treaties do not contain provisions requiring the residence state to either exempt foreign source income or grant a credit for foreign taxes. Although these treaties adopt a schedular approach to the allocation of taxing rights, they do not distinguish between personal and impersonal taxes.[57] A similar approach — that is, a schedular allocation of exclusive taxing rights without either an exemption or a foreign tax credit requirement — was also taken in treaties in this period and earlier between continental European countries where both countries either exclusively or primarily levied a personal tax.[58]

B The United States, the United Kingdom, and the Significance of Their Early Treaties

Significantly, both the United Kingdom and, to a lesser extent, the United States did not enter into any double taxation treaties until a relatively late stage. The United States’ first comprehensive double taxation treaty relating to income tax was with France in 1932,[59] followed by treaties with Sweden in 1939,[60] France in 1939,[61] and Canada in 1942.[62] The United Kingdom’s first comprehensive double taxation treaty was with the United States in 1945,[63] although the United Kingdom had previously entered into a very differently structured treaty with Ireland in 1926[64] and more limited treaties with Spain[65] and Switzerland.[66] The United Kingdom had developed a model double taxation treaty based on the League of Nations Convention No 1B, but was unsuccessful in its attempts in the 1930s to negotiate treaties based on it with continental European countries. The attitude of the British Treasury throughout the 1930s was opposed to entering into double taxation treaties. It took the view that, ‘in practice interest payments … were not normally taxed at source’ and that relieving double taxation on business profits and dividends would only encourage British firms to set up business establishments in foreign countries and British investors to acquire foreign shares.[67]

The late entry of these major capital exporters, each of which taxed on both a residence (citizenship in the case of the United States) and source basis,[68] into comprehensive double taxation treaties is significant as both of them taxed income using what was really a global system based on personal taxes. Although the United Kingdom system divided income into schedules, this was done entirely for computational purposes, with the liability for United Kingdom taxation being personal rather than impersonal. By the time these countries began negotiating double taxation treaties, many of the formal and structural characteristics that were expected in a double taxation treaty had already been settled in treaties entered into by continental European states and in drafts prepared by the League of Nations Technical Experts. The early treaties of the United States and the United Kingdom built on and developed these foundations and were entered into in a world where impersonal taxes still formed a significant element of the tax systems of several continental European states.

Treaties entered into between a country using a global system of personal taxes with a worldwide jurisdictional base and a country using a schedular system of impersonal taxes with a source jurisdictional base inevitably focused on points of intersection between the two systems. As a global system taxed everything, it intersected with a schedular system at each schedular taxing point. Income from dividends, income from immoveable property, income from services, and so on were all income so far as a global system was concerned.

By contrast, schedular systems initially did not have a comprehensive concept of income. Under these systems it was initially only meaningful to ask whether an item fitted within a particular schedule — there was no overarching concept of income. When personal taxes were introduced into schedular systems based on impersonal taxes, they formed a residual category that taxed items not dealt with under the specific schedules. The primary bases of taxation in these systems remained the impersonal, source-based taxes on particular categories of income.[69]

The global system could accommodate each of the categories in a schedular system, whereas none of the schedules in a schedular system (including the residual or complementary personal tax) considered in isolation comprehended the concept of income in a global system. Hence, the obvious solution for preventing international juridical double taxation between two countries using these different systems was to adopt a structure which mirrored the structure of a schedular system in assigning rights to tax. As noted above, earlier treaties between countries using global systems and countries using a schedular system took this approach and, in some instances, went so far as to dissect an impersonal source-based tax out of a comprehensive income tax so as to characterise the dissected portion as an impersonal tax.[70]

The next key development used this model as a basis. In treaties in the late 1920s and 1930s between continental European countries with a largely global approach to income taxation, the distinction between personal and impersonal taxes was abandoned. As noted above, however, these treaties differed from modern double taxation treaties in that they assigned exclusive taxing rights on a schedular basis to either the source or residence country and did not require the residence country to grant either a foreign tax credit or an exemption. Assigning taxing rights on a schedular basis, as had naturally been the practice in treaties between countries using schedular systems dominated by impersonal taxes, meant that conflicts of source rules (the major type of international double taxation possible under a source basis schedular system) could be avoided while including a definition of fiscal domicile or residence was aimed at avoiding double taxation caused by assigning the exclusive right to tax certain types of income to the residence country.[71] The United States’ first treaty with France in 1932 in general follows this pattern, although it does not in its terms give an exclusive right to the source country to tax industrial and commercial profits of a permanent establishment,[72] and was primarily concerned with particular features of the French tax system as it applied to United States parents of subsidiaries in France and to French permanent establishments of United States companies.[73]

It is not until the United States–Sweden Treaty 1939, however, that two further key structural features of modern double taxation treaties emerge. As noted above, United States policy in the League of Nations committees had been to limit but not eliminate source-based taxation while preserving the right of the residence country to tax, through the application of the foreign tax credit, the difference between the source country tax rate (as reduced by treaty) and the residence country tax rate. This approach meant that the allocation of taxing rights could, in some instances, be permissive and not exclusive. At the same time the United States also wished to retain its right to tax non-residents on United States sourced income. To do so required that the allocation of permissive taxing rights be combined with the elimination of the distinction between personal and impersonal taxes. Requiring the residence country to grant a foreign tax credit was appropriate, given that residence–source conflicts were only possible where the residence state levied a personal tax on a global basis, which had long been United States domestic law.

This combination of limited permissive taxation at source of some classes of income and the use of a foreign tax credit can be seen for the first time in United States–Sweden Treaty 1939. Nonetheless, this Treaty still permits greater source taxation than is common in modern double taxation treaties. In particular, it gives the source country the exclusive right to tax the industrial and commercial profits of a permanent establishment.[74]

A permissive, as distinct from exclusive, right to tax the industrial and commercial profits of a permanent establishment is found for the first time in United States–France Treaty 1939 and then in United States–Canada Treaty 1942. Under the former Treaty, the United States was required to grant a foreign tax credit but in the case of France the type of relief from international double taxation depended on whether the relevant French tax was a schedular tax or not.[75] The latter treaty did not require the residence country to grant a foreign tax credit and proposed that double taxation be resolved through a mutual agreement procedure.[76]

The combination of permissive source taxation of industrial and commercial profits of a permanent establishment, limited source taxation of dividends and some royalties, and the reciprocal granting of a foreign tax credit by each country in its capacity as a residence country first appears in the double taxation treaty between the United States and the United Kingdom in 1945.[77] This treaty clearly had a significant influence on the subsequent treaty practice of these two major capital exporters who, as discussed below, shortly afterwards rapidly began to develop significant bilateral double taxation treaty networks.[78]

The granting of permissive, as distinct from exclusive, taxing rights to the source country in relation to particular classes of income thus can really be regarded as a second structural feature of modern double taxation treaties. As noted above, prior to the development of this idea, allocation of taxing rights had been on an exclusive class of income by class of income basis. Hence in modern double taxation treaties we now typically see some types of income being taxed exclusively on a source basis, some being taxed exclusively on a residence basis, and others where both the source country and the residence country are permitted to tax provided that the residence country grants a foreign tax credit for the source country tax paid.

If the world had been confined to countries taxing on a territorial basis then double taxation treaties would have only needed to contain source rules. If the world had been confined to countries taxing on a residence basis then double taxation treaties would have only needed to contain dual residency tie breakers. As it was, the world contained some countries that taxed on a territorial basis and some that taxed on a worldwide basis. Hence, some compromise between the taxing rights of the two sets of countries was probably inevitable especially when countries, such as the United States which taxed on a global basis, entered into treaties with countries which taxed at least some income on a source basis only.

Given, as argued earlier, that the intersection between schedular and global systems almost required the adoption of a schedular structure in treaties, it was natural for model and actual treaties between territorial and worldwide systems to implement compromises on a class of income by class of income basis. Hence, in League of Nations draft Convention IC, in United States–Sweden Treaty 1939 and in modern double taxation treaties, we typically see reductions in the source country’s taxing rights varying from one class of income to another, and the residence country being required to provide relief either in the form of an exemption or a foreign tax credit.[79] Importantly, one of the compromises made in this period was to allow source countries full taxing rights in relation to business profits where a non-resident conducted business through a permanent establishment in the source country[80] and where a locally incorporated company was controlled by non-residents. The other side of the compromise was for source countries to reduce their taxes on investment income while allowing residence countries the right to tax that income after giving a foreign tax credit.[81] These structural features of modern double taxation treaties effectively require the identification of the source of different classes of income and the determination of the jurisdiction in which a particular taxpayer resides.

A related and third structural feature of modern double taxation treaties operating in conjunction with a foreign tax credit is the measured nature of the relief provided under that combination. The use of measured relief can be traced back to the origins of the United States foreign tax credit system. When the United States first introduced a foreign tax credit in 1918, it did not impose any limitations on the amount of the credit because United States income tax rates were relatively high at the time the foreign tax credit system was introduced. In 1921, the United States introduced an overall limitation under which the credit was a proportion of the taxpayer’s overall United States tax liability determined by reference to the proportion of the taxpayer’s global income derived from foreign sources.[82]

The credit was confined to United States citizens, residents and domestic corporations, required the determination of foreign taxes actually paid, required that the foreign tax be an income tax, extended to underlying foreign corporate tax paid by a subsidiary of which a United States corporation held majority ownership, and depended on the foreign country either crediting or exempting United States source income against its own income tax.[83]

This form of credit and limitation, in contrast to a limitation based on rates as under the system of Dominion Income Tax Relief, inevitably leads to a measured approach to relief as it requires that determinations be made of: (a) foreign tax paid; and (b) domestic tax otherwise payable. The reduction, but not elimination of source country taxation, was on the basis that the actual residual amount of residence country tax would be collected via the use of the foreign tax credit. In actual practice, the first use of a foreign tax credit in double taxation treaties was in treaties involving the United States, which naturally applied its own measured approach to relief in the treaties that it entered into.

When a foreign tax credit system with an overall limitation is combined with a double taxation treaty structure in which the source country bears a portion of the relief by reducing its taxes, or (in the case of business profits) limiting its tax rights, in relation to particular classes of income, the operation of the system becomes inherently complex. Over time the inherent complexities of a measured approach to relief have intensified through the development of more complex forms of limitation such as the United States’ ‘basket system’.[84]

Arguably, basket systems of limitation are, to a large extent, products of the schedular structure of double taxation treaties, as the perceived need for basket systems only arises when different categories of income are taxed differently.[85]

A fourth structural feature of modern double taxation treaties that can also be traced back to work of the League of Nations is the approach that they take to transfer pricing. The League of Nations commissioned the Carroll Reports, which were reports on the taxation of foreign and national enterprises compiled by Mitchell B Carroll on the basis of national reports provided by 27 countries.[86] The Carroll Reports identified three different, but not necessarily mutually exclusive, approaches that countries took to the problem of transfer pricing: (a) the requirement that branches and subsidiaries of foreign companies maintain separate accounts; (b) the use of empirical methods based on the assumption that the local operation made the same percentage profit as the business as a whole or as other similar businesses and applying that percentage to identifiable factors such as turnover or capital employed; and (c) the apportionment of part of the entire profit of the business to the jurisdiction.[87]

The majority of reporting states adopted the first approach, often supplemented by the ability to rely on the second either as a check on the veracity of the separate accounting or as default position where separate accounts were either not available or were not satisfactory. Only a few countries used the third approach, although, significantly, France adopted it in taxing income from securities on the basis of the proportion that the local affiliate’s assets represented of the assets of the global enterprise. The problem was principally seen as one of determining the profits of a branch and the report characterised transfer pricing between parent and subsidiary companies and between subsidiary companies as the ‘diversion of income’. A local branch was, after all, merely part of a transnational business, which, as a matter of corporate or business law, was not obliged to maintain separate accounts for each of its branch operations. Subsidiaries, by contrast, already maintained separate accounts. In the case of branches it was necessary for tax law to prescribe additional requirements on the enterprise so that rights to tax global profits of an enterprise could be determined.[88]

The Carroll Reports formed the basis of the 1933 League of Nations Draft Convention Adopted for the Allocation of Business Income between States for the Purpose of Taxation (‘1933 Draft Convention’).[89] In relation to the taxation of branches, the 1933 Draft Convention built on the foundations of the League’s 1928 Draft I, draft Convention IB and draft Convention IC, by providing that business profits of an enterprise domiciled in one state could only be taxed in another if they were attributable to a ‘permanent establishment’ carried on in the second state.[90] The profits of a permanent establishment were to be ascertained by reference to separate accounts that were established and maintained, adjusted ‘to re-establish the prices or remunerations entered in the books at the value which would prevail between independent persons dealing at arm’s length.’[91] Where accounts could not be produced or could not be satisfactorily adjusted, profit could be determined as a percentage of turnover by reference to the profits of comparable enterprises. An apportionment approach based on appropriate factors was also permitted where neither of the foregoing methods produced an appropriate result.[92]

By contrast, in the case of parent and subsidiary transactions or transactions between subsidiaries, the 1933 Draft Convention only permitted adjustment of their accounts where ‘both enterprises are owned and controlled by the same interests’ to take account of the diversion of income or expenses between them.[93]

In other words, in these cases reliance was only to be placed on the separate accounts as adjusted under the arm’s length standard backed up by empirical methods, but not by an apportionment approach. As Picciotto has observed, arts 7 and 9 of the OECD Model can be traced to the 1933 Draft Convention in their reliance on the arm’s length standard, and in permitting the use of apportionment in the case of branches (though, under the OECD Model, this is only permitted where it has been customary in the jurisdiction in question) but not in the case of subsidiaries.[94]

The difference in approach can be traced back to thinking that sees the separate legal entity doctrine as reflecting economic as well as legal reality. Importantly, for the purposes of this article, these features of the 1933 Draft Convention found their way into actual double taxation treaties of the 1930s[95] and, by the time of the double taxation treaty between Australia and the United Kingdom in 1946,[96] were regarded by the United Kingdom — which was then fast developing the world’s largest bilateral double taxation treaty network — as an expected feature of international double taxation treaty practice.[97]

C Summary

In summary, the key structural features of modern double taxation treaties are:

• They have a schedular structure that requires the classification of income into categories for computational purposes.

• They typically share the burden of relieving international double taxation by either assigning exclusive rights to tax particular categories of income to one jurisdiction, or reducing the source jurisdiction’s taxing rights and allowing the residence jurisdiction residual taxing rights.

• They require the identification of the geographic source of each category of income.

• They require that the jurisdiction in which a taxpayer resides be determined.

• When they use a foreign tax credit method of reducing international double taxation they use a measured approach to providing relief.

• They use the arm’s length approach to resolving transfer pricing issues and only permit any form of formulary apportionment at all in the case of branches.

This article has argued that each of these features reflects the origins of modern double taxation treaties in treaties entered into by continental European states in the late 19th and early 20th centuries, and the work of committees of the League of Nations, which were significantly influenced first by countries that used schedular impersonal taxes and later by the United States’ advocacy of the foreign tax credit. The article has further argued that the schedular structure of double taxation treaties has contributed to the development of more complex limitations (such as the United States basket system) in foreign tax credit systems.

III THE SYSTEM OF DOMINION INCOME TAX RELIEF

A Basic Outline of the System of Dominion Income Tax Relief

The system of Dominion Income Tax Relief was introduced in the United Kingdom as s 27 of the Finance Act 1920 (UK) 10 & 11 Geo 5, c 18 and, in general, continued to apply in the case of a particular dominion or colony until the United Kingdom entered into a double taxation treaty with the dominion or colony concerned. The system implemented recommendations by the 1920 United Kingdom Report of the Royal Commission on the Income Tax.[98] Under the system of Dominion Income Tax Relief, the United Kingdom provided a credit for the lesser of the rate of dominion tax or one half of the relevant United Kingdom rate.

The aim of the system of Dominion Income Tax Relief was that the total tax paid by a United Kingdom resident on dominion sourced income should not exceed the greater of the United Kingdom or the dominion rate. In other words the total relief provided was required to equal the lesser of the relevant United Kingdom rate or the relevant dominion rate. Where the relief provided by the United Kingdom did not produce this result the expectation was that the relevant dominion would provide a rebate of tax to make up the difference.[99]

B Features Distinguishing Dominion Income Tax Relief from Modern Double Taxation Treaties

Several features of the system of Dominion Income Tax Relief which distinguish it from the form of relief provided under double taxation treaties and under foreign tax credit systems may be noted.

First, unlike double taxation treaties, the system was based on a global concept of income. Relief was provided for all types of income and the extent of the relief provided did not vary according to the type of income involved. Hence, classification of income into classes was not necessary.

Second, under the system, from the United Kingdom perspective, the sole question was whether tax had been paid on the same income in both the United Kingdom and in a Dominion. Thus, questions of residence and source were irrelevant. Under the system it was possible for a resident of a dominion to claim relief from United Kingdom tax on income sourced within the relevant dominion.[100]

Third, the view of the United Kingdom tax administration was that only a comparison of rates was necessary and that average rates could be used for this purpose. For this reason, in the view of the United Kingdom tax administration, it was not necessary that there be any arithmetic correspondence between the amount of dominion tax paid and the amount of United Kingdom tax paid. From the United Kingdom perspective at least, differences in base and income years and changes in rates did not affect the operation of the system.[101] Hence, while the system required a comparison of actual rates, there were notional elements to the relief provided under the system.

Fourth, the overall aim of the system was that relief would be provided at the lower of the two rates. This was achieved by the United Kingdom imposing an upper limit for its relief equal to half the applicable United Kingdom rate and expecting the relevant dominion to provide such further relief as was necessary to fulfil the overall aim of the system. Hence the system envisaged the same sharing of relief between the two jurisdictions for all types of income. Unlike a system of foreign tax credits operating in conjunction with a modern double taxation treaty, the system was not based on contributions to relief by each jurisdiction varying for different types of income.

Fifth, while the system provided unilateral relief by the United Kingdom it also envisaged that reciprocal relief would be provided by dominions.[102]

At the same time, colonies, as distinct from dominions, were all required to provide reciprocal relief. Hence the system had unilateral, bilateral and multilateral elements.

C Summary

In summary the key structural features of the system of Dominion Income Tax Relief were:

• It was a global system applying to all types of income.

• Residence of the taxpayer and the geographic source of income were irrelevant to determining whether or not relief was available.

• Geographic source was used for computational purposes as the foreign rate of tax applied depended on geographic source.

• It was a system based on a comparison of rates not on comparing actual tax paid.

• It aimed to ensure that the total tax payable was equal to the highest of the two countries rates.

• It anticipated that the two countries would share the burden of providing relief by limiting the amount of credit given by the United Kingdom and expecting the Dominion to provide additional relief so that the total relief was at a rate equal to the lower of the two countries rates.

• It had unilateral, bilateral and multilateral elements.

IV BILATERAL DOUBLE TAXATION TREATIES AND DOMINION INCOME TAX RELIEF: EVIDENCE FROM AUSTRALIAN PRACTICE

A Expansion in Double Taxation Treaties

The end of the Second World War saw an explosion in the number of bilateral double taxation treaties. In part, the expansion in the number of bilateral double taxation treaties after the Second World War was due to a program by the United States of offering to enter into double taxation treaties with other jurisdictions.[103] By this stage the impersonal schedular tax systems — systems that had given rise to the schedular categories in double taxation treaties — were becoming less common throughout Europe.

Another reason for the expansion of the number of double taxation treaties in this period was that the United Kingdom, following its double taxation treaty with the United States in 1945, began inviting its dominions and colonies and selected European countries to enter into double taxation treaties. Indeed, between the end of the Second World War and 1950, the United Kingdom entered into comprehensive double taxation treaties with no fewer than 47 new treaty partners,[104] which meant it had, at the time, the largest double taxation treaty network in the world.

In some senses, the development of double taxation treaty networks in this period was a product of the end of colonialism. While some countries such as the United Kingdom and the Netherlands entered into double taxation treaties that extended to at least some of their colonies, decolonisation and independence movements were eventually to mean that the dominant form of double taxation treaty was a bilateral treaty between independent nations alone. One reason for the success of bilateral double taxation treaties in this period was therefore that they suited the geopolitical conditions that prevailed between the end of empires and the emergence of trading blocs.

B Reasons for the Demise of the Dominion Income Tax Relief System

The end of colonialism alone cannot explain the demise of the system of Dominion Income Tax Relief in this period. By 1945 both the United Kingdom and the dominions wanted the system replaced. All these countries only had personal income taxes founded on a concept of income that did not differentiate between categories of income except for computational purposes and that already had a comprehensive (and some would say principled) system of double taxation relief. This begs the question of why these countries should prefer to enter into a series of bilateral double taxation treaties based on a division of income into schedular categories that were different from the computational categories that any of them used.

In part, the answer lies in defects in the administration of the system of Dominion Income Tax Relief. The system as administered did not have sufficiently detailed rules agreed between all parties.[105] For example, in the context of the United Kingdom–Australia relationship, classification problems arose in 1939 when Australia began changing its corporate–shareholder taxation system from an imputation system to a classical system. Differing attitudes to economic double taxation in the two jurisdictions led to international juridical double taxation.[106]

In addition, to remove international juridical double taxation the system required, in some cases, that relief be given by both the United Kingdom and then by the dominion in question. In practice this meant that lengthy delays were experienced by taxpayers in obtaining complete relief. The system of Dominion Income Tax Relief would have been more workable if the dominions, having regard to the extent of relief to be provided by the United Kingdom, had set a rate of taxation on income flowing to the United Kingdom which meant that the total tax borne did not exceed the higher of the United Kingdom and dominion rates.

Perhaps most importantly, however, the contributions of the United Kingdom and the dominion in question to providing relief were essentially proportionate, and were based on rates applied to a global concept of income that did not differentiate between categories of income. By contrast, under the double taxation treaties that replaced Dominion Income Tax Relief, the division of income into categories meant that the contributions of the United Kingdom and of each dominion to providing relief could be varied from one category of income to another. This meant that compromises in relation to one category of income could be balanced by countervailing compromises on another category and that compromises could be subtly varied according to the economic and political characteristics of the respective treaty partners.

Moreover, the notional and partial relief provided by the United Kingdom applied unilaterally with the only variation between its application to dominions arising in respect of participating and non-participating Dominions. Where a dominion elected to participate, the expectation was that its participation would take a form which ensured that the total tax payable did not exceed the higher of the United Kingdom and dominion rates. Thus, a system of bilateral double taxation treaties was both economically and politically more flexible than the system of Dominion Income Tax Relief when applied between countries, such as the United Kingdom and the dominions, whose economic and trading positions varied considerably.

The combination of limiting, but not necessarily eliminating, source country taxation with the granting of a foreign tax credit by the residence country — allowing as it did for a flexible sharing of the burden of relieving international double taxation which could be subtly varied on a treaty by treaty basis — was arguably a significant contributor to the increased popularity of bilateral double taxation treaties after the Second World War.

C Australian Double Taxation Treaty Practice

Two examples from Australian double taxation treaty practice in the period from 1946 onwards are illustrative of these advantages of the schedular structure of double taxation treaties.

At the time of the negotiation of Australia’s first double taxation agreement with the United Kingdom in 1945–46, no developed country in the world could have been more committed to source-based taxation than Australia. By contrast, the United Kingdom had championed residence-based taxation in international fora since at least the 1920s. When the United Kingdom initially proposed a comprehensive double taxation treaty to Australia, it sought zero source country taxation on all dividends, interest and royalties.

The main concern at the time, however, was the high effective tax rates which applied under the system of Dominion Income Tax Relief (as administered in practice by the United Kingdom) on dividends flowing from Australian companies to United Kingdom shareholders. Australia wanted the United Kingdom to give a full credit for at least the Australian tax on the dividends if not for the underlying corporate tax. At the same time, Australia wanted to limit any reductions that it agreed to in source taxation of dividends, interest and royalties. The principal concern for the United Kingdom was the high effective rates of tax payable under the system of Dominion Income Tax Relief on dividends paid by Australian subsidiaries wholly-owned by United Kingdom parents.

The treaty that was ultimately produced through somewhat testy if not acrimonious negotiations and through the intervention of politicians represented a compromise under which both sides could claim to have achieved at least some, but not all, of their significant aims. Australia agreed to not tax dividends paid by Australian resident subsidiaries that were wholly-owned by United Kingdom parents and agreed to halve, but not eliminate, its tax on other dividends flowing to United Kingdom shareholders. For its part, the United Kingdom agreed to give full credit for Australian tax levied at source on income received by United Kingdom residents. This included a credit for underlying corporate tax on dividends irrespective of the level of shareholding in question. Australia retained full rights to tax interest, film royalties, mineral and industrial royalties and rents at source but agreed that literary royalties would only be taxed on a residence basis. Hence, the treaty represented a series of offsetting compromises which were only really possible under a schedular approach to computation.[107]

A further example can be seen in the negotiation of the second double taxation treaty between Australia and the United Kingdom in 1966–67. This treaty[108] had its origins in an offer by the United Kingdom to revise the 1946 United Kingdom–Australia Treaty following the introduction of a corporations tax in the United Kingdom in 1965. Prior to 1965, the United Kingdom had applied what was, in effect, a form of imputation system to corporate shareholder taxation and had, from 1945 onwards, allowed foreign tax credits for underlying corporate tax irrespective of the level and character of shareholding involved.

In this period the United Kingdom had also pursued a policy of encouraging offshore investment by United Kingdom residents. This policy was reversed in 1965 and the aim became to encourage investment in the United Kingdom by United Kingdom residents. In addition, as the corporations tax adopted the separate entity theory of corporate-shareholder taxation, underlying foreign corporate tax was seen as a distinct tax on a distinct taxpayer from any foreign tax on dividends. Giving foreign tax credits for underlying corporate tax irrespective of the level and character of shareholding was seen as being inconsistent with this paradigm and, as a consequence, underlying foreign tax credits were then limited in United Kingdom domestic legislation to Commonwealth resident shareholders holding 10 per cent or more and to other residents holding 25 per cent or more. Following these changes in its domestic law, the United Kingdom sought to revise provisions in its double taxation treaties (including the treaty with Australia) which provided for a credit for foreign underlying tax irrespective of the level of shareholding.[109]

When Australia was asked by the United Kingdom to revise the 1946 treaty on this basis, the Australian reply was that it would prefer to negotiate a new treaty. There were several features of the 1946 treaty that Australia was no longer satisfied with by 1966. Prominent among these was the exemption of dividends paid by wholly-owned subsidiaries from tax at source. In the interim, Australia had successfully argued for what was in practice a 15 per cent tax rate on all dividends in its treaties with the United States in 1953,[110] Canada in 1957[111] and New Zealand in 1960.[112] Furthermore, Australian policy by 1966 was to encourage Australian equity participation in direct investment by foreigners in Australia.

The draft OECD Model Double Taxation Convention was published in 1963, and, in 1966, the United Kingdom was a member of the OECD but Australia was not. Prior to the opening of negotiations, the United Kingdom sent to Australia a draft treaty that was derived from the 1963 OECD Model but, in some respects, gave greater taxing rights to the residence country. At this time, Australia still maintained a policy of high levels of source taxation. Australia was conscious that this was the first double taxation treaty it had entered into since the publication of the 1963 draft OECD Model; it would thus ‘stand as something of a precedent’. Australia wanted to preserve its high levels of source taxation as much as possible. Australia also wanted to avoid including a non-discrimination article in the treaty as it was concerned that several provisions in its municipal law would infringe it. Australia wanted relief from the United Kingdom on dividends paid by United Kingdom incorporated companies to Australian residents in situations where all or a very large proportion of the income of the United Kingdom company was derived from Australian sources (referred to in the correspondence and discussion of the time as ‘the New Broken Hill Situation’).[113]

In the negotiations that followed, Australia was able to obtain many concessions through use of its consent to limiting the United Kingdom’s credit for underlying Australian corporate tax, its agreement to the United Kingdom’s proposals to tax shipping and aircraft on a residency basis, and arguments based on the treaty that the United Kingdom had recently concluded with New Zealand. These concessions included, at variance with the OECD Model: a uniform 15 per cent rate of withholding tax on dividends together with higher rates of withholding tax on interest and royalties than were normal under United Kingdom treaty practice; the omission of the non-discrimination article; and the restriction of the other income article to third country tax. Australia was able to successfully maintain these high rates of withholding tax and its opposition to a non-discrimination article in nearly all its subsequent treaties for over thirty years.[114]

Australia, based in part on concessions made by the United Kingdom to Canada in a protocol to their treaty, was also able to obtain relief from United Kingdom taxation of dividends paid to Australian residents in the New Broken Hill situation.

Other examples from the history of Australian treaty practice could be given,[115]

but the two above are sufficient to illustrate the point that the structure of modern double taxation treaties and their bilateral nature enabled offsetting compromises to be made by both parties to the relevant negotiation and enabled compromises to be adjusted according to the economic and political nature of the relationship in question.

V THREATS TO THE SUSTAINABILITY OF BILATERAL DOUBLE TAXATION TREATY NETWORKS

Part IV argued that, while larger political and economic factors contributed significantly to the extraordinary spread of bilateral double taxation treaties after the Second World War, another significant contributing factor was the offsetting compromises that the schedular structure of bilateral treaties permitted. When combined with a foreign tax credit by the residence country, these offsetting compromises allowed a flexible sharing of the burden of relieving international double taxation.

While the international network of bilateral double taxation treaties has been extremely successful, several developments in recent years raise questions about the sustainability of double taxation treaty networks in the future. The following section examines some contemporary threats to the sustainability of double taxation treaty networks arising from: (a) electronic commerce; (b) financial engineering; (c) tax competition; (d) Directives of the European Union Council and the interpretation of the EU Treaty by the ECJ; (e) transfer pricing problems; and (f) the lengthy time periods needed to modify an entire double taxation treaty network. This section argues that the reason why many of these developments threaten the sustainability of bilateral double taxation treaty networks is because of the combination of the schedular structure with foreign tax credit relief, which, ironically, contributed to the success of bilateral treaties in the first place.

A Challenges from Electronic Commerce

First, the emergence of electronic commerce has raised several challenges for its taxation under double taxation treaties. The challenges for tax systems generally posed by electronic commerce have been summarised by Pastukhov as follows:

The Internet, a means of communication unlike any before it, does not fit into the current regime of international taxation. First, it allows businesses to operate without creating a permanent establishment in any country. Second, it allows businesses to relocate their taxable activities around the world at low cost and without interruption in response to changes in legal and economic environment. Third, it complicates the attribution of income and expenses to a particular part of the transaction. Fourth, the Internet clouds the distinction between providing services and transferring property. Finally, as a result of all the foregoing, it complicates the administration and collection of taxes. In sum, electronic commerce creates opportunities and incentives for businesses to avoid taxation under the current international taxation regime.[116]

These features of electronic commerce pose some problems of particular relevance to double taxation treaties. The first is whether transactions conducted through a website or a server constitute a permanent establishment and if so how income can be attributed to it.[117] The second is classification issues (for example, whether the transaction represents a supply of property or a provision of services).[118] The third is applying source rules in treaties to electronic commerce transactions once classified.[119]

While the OECD has done considerable work on these issues,[120] in the treaty context, the approaches it has recommended have not been universally accepted.[121]

Moreover, while perhaps clarifying the position, those approaches still leave considerable scope for the erosion of source-based taxation through electronic commerce.[122] While these problems are not necessarily unique to the tax treaty context it is notable that in that context the first two can be seen as being products of the limiting of source taxing rights to business profits in treaties through the use of the permanent establishment concept and of the schedular categorisation of income into classes in treaties.

B Challenges from Financial Engineering

Financial engineering has blurred the borderline between debt and equity instruments resulting in different approaches to classification in different jurisdictions and creating classification problems for the schedular approach adopted in double taxation treaties. In a seminal 1994 article, Professor Alvin Warren Jr of Harvard University drew attention to the equivalences, such as ‘put–call parity’, observed in financial theory that give rise to the ability to create synthetic equity and synthetic debt.[123] Synthetic equity, for example, has at least certain formal characteristics of debt but has economic characteristics equivalent to those of an equity instrument.[124]

Jurisdictions such as Australia might endeavour to develop domestic rules that aim to give tax treatment to financial instruments that reflect their economic rather than their formal characteristics or at least may endeavour to develop domestic rules which attempt to neutralise differences that otherwise exist in the tax treatment of certain and contingent rights and obligations. The first approach does not sit easily with the essentially formal approaches to the classification of returns on investment into the schedular categories of double taxation treaties. So far as the latter approach involves some form of compensatory tax on gains on assets where the gain is taxed on a realisation basis,[125] the classification of the compensatory tax for the purposes of the schedular categories in double taxation treaties would appear to be very uncertain.

The former difficulty might be able to be solved by a substance rather than form approach to categorisation of income items into schedules for double taxation treaty purposes but this would require a uniformity of approach to the judicial interpretation of treaty obligations by the courts of all contracting states. This seems a most utopian hope indeed. The latter difficulty might be able to be solved by the development of a universally agreed categorisation of such a tax or by the creation of a new category for double taxation treaty purposes.

While commentators cannot agree on the optimal form of such a compensatory tax, and, indeed, on whether such a tax is an appropriate way to deal with problems raised by financial contract innovation, developing an international consensus on the categorisation of such a tax appears to be a long way off. In the interim, the continuing development of synthetic financial products enables differences in the tax treatment of instruments between countries to be exploited and enables treaty rates of withholding tax, and in some instances foreign tax credits, to be obtained on instruments that have the economic characteristics of another type of return on investment. This undermines any substantive rationale that ever existed for the application of differential rates of tax to interest and dividends in double taxation treaties.

C Challenges from Tax Competition

There is evidence that reduction in the non-tax barriers to free movement of capital across national boundaries has increased the importance of tax as a consideration in the locational decisions of both portfolio and non-portfolio investors.[126] This has led to increased tax competition between jurisdictions and has been a contributing factor to a lowering, and in some cases an elimination, of corporate taxes and of withholding tax rates both within and outside double taxation treaties. To the extent that withholding taxes are eliminated entirely, the functions of double taxation treaties as mechanisms for the sharing of the burden of relieving international double taxation in a manner which permits both source and residence taxation of the one category of income, are reduced. If the above argument is correct, then this development undermines one of the features of double taxation treaties that led to the expansion of double taxation treaty networks after the Second World War.

At the same time — to the extent that withholding taxes are eliminated and exemptions for foreign source income are provided through double taxation treaties rather than unilaterally — these developments can be seen as giving a more complete expression to the post-First World War compromise of giving the primary right to tax business profits to the source country with the primary right to tax investment income falling to the residence country.

D Challenges from the Development of Trading Blocs

A fourth challenge to the sustainability of bilateral double taxation treaty networks comes from the development of multi-jurisdictional trading blocs, particularly the European Union (‘EU’), the most integrated of the trading blocs currently in existence in the world. If we take the EU as the paradigm case of the extremity of this threat to the sustainability of bilateral double taxation treaty networks, then the networks can be seen as being threatened by both positive integration measures adopted by the trading bloc (in the EU’s case, largely through Directives by the European Council) and by negative integration measures requiring (in the case law of the ECJ) consistency between the fiscal laws of member states and the law of the trading bloc (here, EU law).

There is an extensive literature on the effects of ECJ jurisprudence on the fiscal sovereignty of member states of the EU. In essence, this jurisprudence is concerned with whether fiscal laws of a member state infringe one of the four freedoms guaranteed under the EU Treaty and, if one or more freedoms is infringed, whether that infringement can be justified. Where a double taxation treaty is involved, the ECJ acknowledges that the power to choose the criteria and allocation of taxing jurisdiction lies within the competence of member states, but at the same time requires that the exercise of fiscal competence by member states comply with EU law. While EU law has not been interpreted as a right to most favoured nation treatment, ECJ cases have held in some instances that a member state was required unilaterally to extend benefits contained in one of its treaties to nationals of other member states irrespective of whether or not a bilateral treaty was in existence.[127] Very little of the jurisprudence concerns the application of the fiscal laws of a member state to a national of a non-member state.[128]

Of the four freedoms guaranteed in the EU Treaty, only the ‘free movement of capital’ may be invoked by a national of a non-member state to challenge the validity of a law of a member state as being inconsistent with the EU Treaty. Hence, determining whether a national of a non-member state is protected by the EU Treaty can involve determining whether the law of the member state in question infringes the free movement of capital as distinct from the freedom of establishment.

Two approaches to drawing this distinction can be found in the ECJ jurisprudence and in the writings of commentators. One view is that the distinction turns on whether the relevant investment is portfolio or non-portfolio. If the former then the restriction will be an infringement of the free movement of capital, if the latter then the restriction will be an infringement of the freedom of establishment. Another, and arguably the better view, is that the distinction turns on whether the rule in question itself applies both to portfolio and non-portfolio investors or whether its application is confined to non-portfolio investors. In the former case a restriction will be an infringement of the free movement of capital while in the latter case the restriction will be an infringement of the freedom of establishment.[129]

The views of commentators differ on whether decisions of the ECJ threaten the sustainability of bilateral treaties between EU member states and non-member states. Arguably the greatest threat is to limitation of benefits articles in United States treaties with individual EU member states. Influential United States commentators such as Michael Graetz and Alvin Warren Jr[130] and Ruth Mason[131] are concerned that decisions of the ECJ in the non-tax Open Skies cases mean that these articles could well infringe the EU Treaty.[132]

More fundamentally, Graetz and Warren argue that the core issue is whether an EU member state will be permitted to enter into a bilateral treaty with a non-member state that grants benefits to nationals of the non-member state that it does not grant to nationals of member states. They note that the ECJ cases on bilateral treaties come to conflicting conclusions on this issue, but consider that the Court’s language suggests that it will follow a fact specific approach and determine whether the resident and non-resident are ‘in similar circumstances’.[133]

It must be noted that some commentators from within the EU have criticised these views and do not regard the decisions of the ECJ as threatening the sustainability of bilateral double taxation treaties between EU member states and non-member states.[134]

Nonetheless it is also notable that there is considerable disagreement among European commentators on the effect of ECJ decisions in general and on their effect on double taxation treaties in particular. Given this divergence in views it can only be said at present that the effect of ECJ decisions on the sustainability of double taxation treaties between EU member states and non-member states is uncertain.

Moreover, whether or not the views of important United States commentators such as Graetz and Warren and Mason represent correct interpretations of the current state of ECJ decisions is, in some respects, beside the point. How commentators and policy makers in non-member states perceive the effect of ECJ decisions may well prove to be very significant in determining whether countries like the United States continue to believe that they can confidently enter into double taxation treaties with EU member states. The possibility that limitation of benefit articles in United States treaties might infringe the EU Treaty is clearly a major concern for these commentators. While this concern would only be shared by other non-member states that include limitation of benefit articles in their treaties as a matter of policy (for example, Japan), if such concerns ultimately led to the dismantling of bilateral double taxation treaty networks between EU member states and important economies such as the United States and Japan, the effect on bilateral treaty networks as a whole might well be devastating.

Possibly the more significant threat from the EU to the sustainability of double taxation treaties comes from positive integration measures introduced via European Council Directives. Important Council Directives that raise questions about the sustainability of bilateral double taxation treaty networks include the Parent–Subsidiary Directive,[135] the Interest and Royalties Directive,[136]

the Savings Interest Directive,[137]

the Assessment Assistance Directive,[138]

the Recovery Directive,[139]

and the Arbitration Convention.[140]

These Directives are multilateral instruments which are effectively performing similar functions to some of those traditionally performed by double taxation treaties. These include: limiting or eliminating certain source country taxing rights; relieving international double taxation through the use of an exemption or foreign tax credit method; allocating profits through the use of arm’s length and separate accounting principles; facilitating the exchange of information; assisting in collection; and providing a dispute resolution procedure. So far as these multilateral instruments perform these functions, they render equivalent provisions in double taxation treaties between EU member states otiose.

The combined effect of these Directives and the responses of member states to ECJ decisions holding that domestic law provisions had infringed the EU Treaty is producing a more harmonised system of income taxation within the EU. This harmonised system is characterised by low or zero rates of withholding tax, the use of either partial exemption or notional relief systems of corporate–shareholder taxation, and the use of the exemption system rather than a foreign tax credit system of relieving international double taxation of business profits and non-portfolio dividends.

While the responses are not entirely uniform, the overall direction is clear: the division of taxing rights on the basis of schedular categories is becoming less important as is the use of foreign tax credits and hence measured relief, at least in the case of non-portfolio investment. While double taxation treaties entered into by EU member states still retain the schedular structure of the OECD Model, and while treaties between EU member states and non-member states, for example the United States, might continue to contain foreign tax credit provisions, the reality is that differential treatment on a schedular basis of investment income derived by non-residents is increasingly disappearing from the domestic law of EU member states, as are foreign tax credit provisions. In these situations, double taxation treaties will continue to set limits on the taxing powers of a particular EU member state in a given bilateral relationship, but the constraints of EU law are likely to mean that the domestic law of an EU member state will not in fact approach those limits.

At a more general level, as Panayi has observed,[141]

there is a fundamental inconsistency between the very existence of bilateral treaties and efforts to establish multilateral communities within trading blocs. To the extent that trading blocs become more integrated, bilateral treaties between member states within the trading bloc are likely to be increasingly seen as both inconsistent with the multilateral obligations of member states and redundant.

E Challenges from Transfer Pricing

Difficulties associated with determining arm’s length prices, particularly in relation to intangibles and to the benefits of synergy, make application and operation of the arm’s length standard increasingly problematic.[142] As Picciotto pointed out in 1992:

the problems involved in establishing arm’s length prices for specific intrafirm transactions are at their most intractable when it comes to intangibles. Intangibles are likely to be distinct if not unique, throwing into doubt the validity of comparisons with similar items, if any are available on the market. Pricing based on costs is extremely difficult, since they are essentially joint cost factors with uncertain outcomes, so that either fixing a price based on cost, or allocating cost contributions in relation to anticipated benefit, is likely to be arbitrary.

Above all, intangibles most clearly raise the problem of the allocation of profits from synergy: the additional profits generated for an integrated firm by reason of its features as an organization, and not attributable specifically to any of its parts. Like joint costs, synergy profits cut to the heart of the separate enterprise approach to the taxation of international company groups, since such profits by definition cannot be attributed to one particular affiliate.[143]

Furthermore the combination of electronic commerce and transfer pricing compounds problems of classification: for example, is there a distribution of a product or the provision of a service? This in turn raises questions as to what transfer pricing methodology is appropriate in the circumstances.[144]

Given that many jurisdictions adopt the recommendations of the OECD Transfer Pricing Guidelines[145] and apply an arm’s length standard in their domestic law, the problems relating to the standard are not confined to the operation of double taxation treaties. Nonetheless, as noted above, the use of the arm’s length approach, developed out of the Carroll Reports and incorporated from the beginning into the business profits and associated enterprises articles of double taxation treaties, can be regarded as one of the distinguishing features of a modern double taxation treaty. Moreover, arguably, some jurisdictions, such as Australia, currently use the provisions of art 9 of the OECD Model in their treaties as a supplement to ameliorate inadequacies in their domestic legislation.[146]

Where a treaty does apply, the extensive reliance placed by jurisdictions (such as Australia) on the mutual agreement procedure to resolve transfer pricing issues can be seen not so much as an indication of the value of treaties, but rather as condemnation of the arm’s length standard, and as evidence that transfer pricing issues are often more concerned with inter-jurisdiction allocation of income rather than with tax avoidance.

F Time Associated with Re-Negotiating All Treaties in a Network

A final problem with bilateral double taxation treaty networks, which threatens their sustainability, is the relative slowness with which they respond to changes in tax policy and economic conditions.[147]

This is a product of the very bilateral nature of double taxation treaties. Each treaty has to be negotiated and re-negotiated separately with variations being made according to economic and political considerations relevant to the particular bilateral relationship.

While it was argued above that the ability to subtly adjust treaties according to the nature of that relationship was one of the reasons for the increase in their popularity after the Second World War, that strength is also a weakness when it adversely affects the speed with which a country can adjust its bilateral treaties. Hence, even when, as is the case with electronic commerce, changes are made to the OECD Model or its attendant commentaries to combat problems identified by the OECD with the operation of double taxation treaties, it may take several years for those changes to be reflected in the treaties that a particular country has in force.

In this respect bilateral double taxation treaty networks resemble the Neanderthals: successful for a long period; adapted well to conditions that prevailed at a particular time; but not able to respond sufficiently quickly to changing circumstances.

G Summary

None of these challenges in isolation is likely to be a critical threat to the sustainability of bilateral double taxation treaty networks, but it is strongly arguable that their effect in combination will be to make that sustainability increasingly problematic. It is notable how many of these challenges are products of the schedular structure of double taxation treaties and the measured approaches to relieving international double taxation and transfer pricing that have long been associated with modern double taxation treaties.

VI A POSSIBLE RESPONSE

What responses, then, are possible to deal with the combined effect of these threats to the sustainability of bilateral double taxation treaty networks? While various options have been suggested by commentators,[148]

this article will focus on one possibility that emerges from the foregoing analysis of the structural features of double taxation treaties and of the threats to their sustainability.

It was argued above that many of the threats to bilateral treaties are related to the schedular approach combined with the measured approach to foreign tax credits that they take in order to share the burden of relieving international double taxation. One solution to these problems, therefore, might be to adopt an approach to resolving international double taxation that does not depend on a sharing of taxing rights which varies from one class of income to another, that does not use a measured system of providing relief, and that does not depend on finding that a permanent establishment is in existence before permitting source taxation of business profits.[149]

The solution could be the ‘composite income tax’ approach suggested by Harris, which adopts a global concept of income and which uses a notional system of relief.[150]

The ‘composite income tax’ approach is based on evidence by Sir Josiah Stamp to the 1920 United Kingdom Royal Commission on the Income Tax and on the system of Dominion Income Tax Relief,[151] as previously discussed in this article. Under the composite tax principle — as outlined by Harris — the source state would tax non-residents at a source rate that would reflect the level of production services, as distinct from consumption services, that the source state provided to non-resident investors. The residence state would tax its residents on their foreign source income at its residence rate which would reflect production and consumption services that the residence state provided to its residents.

This system would relieve international double taxation by giving a notional foreign tax credit based on its own source rate. It would also avoid the complexities involved when a measured approach to relief of international double taxation using a foreign tax credit is applied.[152] The adoption of the composite tax system with a notional relief based on the global concept of income in both the residence and source country would, it is submitted, represent a principled approach to international double taxation which would avoid at least some of the problems identified above.

The relationship between a composite tax system and a country’s domestic tax rules requires further explanation. A composite tax system would not prevent countries from continuing to give distinct treatment to different classes of income in their domestic law in relation to wholly domestic transactions. Nor would it prevent them, as a matter of domestic law, from requiring some degree of fixed presence within the jurisdiction before taxing business profits. It would, however, require residence states to relieve international juridical double taxation by giving a notional foreign tax credit based on their own source rate as outlined above and would require source states to apply a uniform source rate to all types of income.[153] Hence, in a composite tax system, international transactions are unaffected by whether or not the domestic law of a tax system gives distinct tax treatment to different classes of income for domestic transactions, meaning that the composite tax system does not necessarily require adjustment to that treatment.

What the composite tax system would do in this respect is remove international problems that arise due to the current differential treatment of international income flows according to the class of income concerned. As the credit given by the residence country would be notional, there would be no treaty requirement for a permanent establishment to be present before a source country could tax business profits. Hence, even where source countries as a practical matter required some degree of permanent presence in their jurisdiction before taxing business profits of non-residents, they could define the required degree of presence and adjust the definition in response to technological changes in international business operations. This would enable countries to define the requisite degree of presence in a manner which they consider would give them a source tax base for business profits that reflects the level of production services that they provided. The right to tax those profits would not be limited by a treaty obligation.

It may seem strange to advocate a return to a system which was regarded as unsatisfactory in the 1920s, 1930s and 1940s. Some of the problems that existed under the system of Dominion Income Tax Relief would be avoided by the use of notional relief and by the source country providing relief through setting a source rate for non-residents. The modern-day widespread use of personal (as opposed to impersonal) taxes which are essentially global in nature and the convergence of significant features of tax systems means that such an approach may be more workable than was thought possible in the 1920s. Nonetheless, if entered into unilaterally, such a system would still not likely be sustainable for the following reasons: (a) not being bilateral and not being schedular, it would have less scope for adjustment according to the economic and political characteristics of individual bilateral relationships; (b) if the composite tax system did not involve bilateral or multilateral treaties it would not provide a means of ensuring that relief would continue to be provided by other countries; and (c) it would not enable benefits to be provided to foreign countries for foreign relations considerations.

The problems with the application of a composite tax system identified in the previous paragraph could largely be resolved within trading blocs if the composite tax system were to be adopted via a multilateral treaty that applied throughout the trading bloc. Assuming that the countries within the trading bloc had compatible tax systems and were at an approximately equivalent level of economic development, it would be possible to set a uniform rate of source taxation within the trading bloc, or at least a uniform means for determining the rate of source taxation. Under the same assumptions, there would not be a need to make particular adjustments as the economic and political characteristics of bilateral relationships within the trading bloc would be approximately equivalent.

A multilateral treaty would provide assurance that relief would continue to be provided within the trading bloc for the duration of the treaty. Within the trading bloc, benefits could be provided for foreign relations considerations more transparently through direct financial and other contributions rather than through tax expenditures via treaties. Each country within the trading bloc could set its own progressive residence tax rate. The development of multilateral treaties within trading blocs would also eliminate the possibility of intra-trading bloc bilateral treaties, and hence eliminate the possibility of conflicts between those bilateral treaties and the law of the trading bloc. As discussed earlier, positive integration measures within the EU are arguably beginning to render at least particular provisions in bilateral double taxation treaties between EU member states otiose. The development of a multilateral treaty within such a trading bloc in some respects would be a logical extension of this process, and there have been several historical examples of multilateral tax treaties being adopted between a restricted group of countries.[154]

Bilateral treaties could continue to be negotiated on a trading bloc to trading bloc basis (or between trading blocs and non-aligned states) to ensure that relief would continue to be provided by member states of each trading bloc to member states of the other trading bloc for the duration of the treaty. The treaties could also provide for the adjustment of the source country rate applicable to intra-trading bloc transactions to take account of the different economic and political considerations relevant to relations between member states of the trading bloc and member states of the other trading bloc. These treaties, as was the case with the first bilateral treaties and as would probably need to be the case with a multilateral intra-trading bloc treaty, might need to contain source rules and residence rules,[155] but they would set overall levels of source taxation instead of limiting source taxation on schedular category by category basis. Consistent with developments in trading blocs such as the EU, such treaties would also need to contain expanded exchange of information and assistance in collection procedures.

The development of bilateral treaties between trading blocs, the development of intra-trading bloc multilateral treaties and the use of notional relief would not of itself solve transfer pricing problems. Earlier in this article it was suggested that the widespread use of the mutual agreement procedure in transfer pricing cases is evidence that transfer pricing is often more an inter-jurisdictional allocation problem than a tax avoidance problem. It is submitted that the use of the arm’s length principle, like the use of a measured approach to foreign tax credits, is reflective of a mentality which sought precision in allocation and which held a belief that profits of an enterprise could be broken up into a series of transactions each of which would have market equivalents. Experience in transfer pricing since the development of the arm’s length standard in the 1930s indicates that this assumption is often not valid, particularly in the case of unique intangibles.

The major alternative proposed to the arm’s length principle has long been some form of global apportionment similar in kind to the approach used by several States of the United States to tax multinational enterprises. The major objection raised to such proposals has in the past been that the vast differences in economic conditions and interests between countries throughout the world would make agreement impossible on the appropriate factors and the weighting of factors to be used in any formula.[156] If, however, trading blocs developed between countries at approximately equivalent stages of economic development, the prospects for agreement on the factors to be used in, and their weighting in, any formula should be easier. Developing a formula for transactions between trading blocs should also be easier, as the existence of trading blocs would reduce the number of players involved in any formula-setting process, and it should possess something closer to equality of bargaining power between participants in the formula-setting process. Another alternative approach to this problem would be to heed the message from the widespread use of the mutual agreement procedure in transfer pricing cases and include mandatory arbitration provisions as the primary transfer pricing resolution mechanism in bilateral treaties between trading groups, or between a trading group and non-member states.

It may be questionable whether multilateral treaties within trading groups and bilateral treaties between trading groups or between a trading group and non-member states would be able to respond more quickly to change than the current system of bilateral treaty networks. The degree of responsiveness to change in the former category of treaties would depend on the type of institutional mechanisms used in the trading bloc. Here there would appear to be value in having an administrative institution with power to develop and finetune regulations implementing more fundamental rules agreed on at the time of the establishment of the trading bloc and subsequently varied by a legislative body within the trading bloc. The ease with which such legislative action could be implemented would depend on the constitutional arrangements within the trading bloc,[157] but change would be implemented by one legislative action and it would not have to wait for the renegotiation of an entire network of bilateral treaties.[158]

It may be likely that the latter category of treaties would be able to respond more rapidly to change than is possible under the current system of bilateral treaties for two reasons. First, as the treaties envisaged would involve trading blocs rather than individual countries, there would necessarily be fewer treaties to change at any given time. Secondly, as outlined above, the treaties envisaged in this category would be more streamlined than current bilateral treaties as they would not aim to reduce source-country taxing rights on a class of income basis. As noted above the intra-trading bloc treaties and the inter-trading bloc treaties envisaged above would be confined to establishing source rates for both parties to the treaty and possibly setting out source rules and residence rules. This should mean that there would be fewer provisions which would require change at any given time.

If this approach were adopted the end result would bear a family resemblance to treaties developed under the current bilateral system but, like homo sapiens sapiens when compared with homo sapiens neanderthalis, would arguably be more streamlined, agile and adaptable.


[*] BA, LLB, LLM (Hons) (Syd), Cert H Ed (UNSW); Professor, School of Business Law and Taxation, Australian School of Business, The University of New South Wales.

[1] For a recent popular, non-technical account, see Stephen S Hall, ‘Last of the Neanderthals’ (2008) 214(4) National Geographic 34.

[2] See, eg, Peter Harris, ‘An Historic View of the Principle and Options for Double Tax Relief’ [1999] British Tax Review 469.

[3] Treaty on European Union, opened for signature 7 February 1992, [2009] OJ C 115/13 (entered into force 1 November 1993).

[4] Technical Experts, Double Taxation and Tax Evasion: Report and Resolutions Submitted by the Technical Experts to the Financial Committee of the League of Nations, League of Nations Doc F.212 (7 February 1925) 11 (‘1925 Technical Experts Report’) noted that the Austro-Hungarian Empire entered into treaties with several of the states forming part of the German Empire and that two treaties were concluded by the Canton of Basle-Town with Prussia and with the Grand-Duchy of Baden. The 1925 Technical Experts Report noted that, since the conclusion of the First World War, ‘treaties were concluded between certain of the Austro-Hungarian Succession States among themselves or between them and neighbouring States.’ It mentioned the 1921 treaties by the German Empire with Austria and Czechoslovakia, the 1921 treaty between Austria and Czechoslovakia, the treaties by Hungary with Roumania in 1923 and with Czechoslovakia in 1922, and the 1923 treaty by the German Reich with seven Swiss cantons: at 11–12. The 1925 Technical Experts Report made ‘special mention’ of the 1921 collective convention signed at Rome between Austria, Hungary, Italy, Poland, the Kingdom of the Serbs, Croats and Slovenes, and Roumania. This convention stated common principles to be applied by signatories in concluding double taxation treaties: at 12. The 1925 Technical Experts Report noted that several of the signatory states had entered into treaties ‘based on the same principles.’ The most recent treaty known at the time of the 1925 Technical Experts Report was the 1924 treaty between Czechoslovakia and Italy. The subsequent history of treaties leading up to the development of the Organisation for Economic Co-Operation and Development Model (‘OECD Model’) will be discussed later in this article. The 1925 Technical Experts Report is reproduced in Joint Committee on Internal Revenue Taxation, Congress of the United States, Legislative History of United States Tax Conventions (US Government Printing Office, 1962) vol 4, 4057.

[5] A similar point is made in John F Avery Jones, ‘The David R Tillinghast Lecture: Are Tax Treaties Necessary?’ (1999) 53 Tax Law Review 1, 12–17.

[6] See generally Paul R McDaniel, Hugh J Ault and James R Repetti, Introduction to United States International Taxation (Kluwer Law International, 5th revised ed, 2005).

[7] Professor Bruins et al, Report on Double Taxation Submitted to the Financial Committee by Professors Bruins, Einaudi, Seligman and Sir Josiah Stamp, League of Nations Doc E.F.S.73.F.19 (5 April 1923) 26. The Report recognised that the part of the French income tax known as the complementary tax, or impôt global, was an ‘income tax proper in its developed form’ as was ‘the similar Italian tax contemplated by the Law of 1919, the enforcement of which [had] recently been postponed’: at 45. The Report went on to state, however, that the French and Italian schedular taxes contained ‘such large elements of impersonal taxation … that they [were] rather to be assimilated to the taxes discussed in the preceding section’ of the Report (which dealt with death duties and property taxes). The Report of the Four Economists is reproduced in Joint Committee on Internal Revenue Taxation, Congress of the United States, Legislative History of United States Tax Conventions (US Government Printing Office, 1962) vol 4, 4003.

[8] 1925 Technical Experts Report, League of Nations Doc F.212, 14.

[9] Ibid 13.

[10] Ibid.

[11] See, eg, Sol Picciotto, International Business Taxation: A Study in the Internationalization of Business Regulation (Weidenfeld and Nicolson, 1992) 21.

[12] See ibid 21–2. See also Michael J Graetz and Michael M O’Hear, ‘The “Original Intent” of US International Taxation’ (1997) 46 Duke Law Journal 1021, 1074–91.

[13] 1925 Technical Experts Report, League of Nations Doc F.212, 20.

[14] Ibid 22–6, especially at 25.

[15] Ibid 39.

[16] Ibid.

[17] Ibid 41–2.

[18] Ibid 41–51.

[19] Graetz and O’Hear, above n 12, 1075 n 215.

[20] Ibid 1074–82.

[21] 1925 Technical Experts Report, League of Nations Doc No F.212, 3.

[22] Ibid. The countries and their initial representatives were: Belgium (M Clavier, Director-General of Direct Taxation); Czechoslovakia (Dr Valnicek, Head of Department at the Ministry of Finance); France (M Baudouin-Bugnet, Director-General of Direct Taxation); Great Britain (Sir Percy Thompson, Deputy-Chairman of the Board of Inland Revenue); Italy (Prof Pasquale D’Aroma, Director-General of Direct Taxation); Netherlands (Dr Sinninghe Damsté, Director-General of Direct Taxation, Customs and Excise); and Switzerland (M Blau, Director of the Federal Taxation Department). Sir Percy Thompson was subsequently appointed to sit on a commission in India and was replaced by Mr G B Canny of the Board of Inland Revenue. M Baudouin-Bugnet was subsequently appointed one of the ‘Presidents de Chambre’ at the Audit Office and was replaced by M Borduge, his successor as Director-General of Direct Taxation.

[23] Ibid 11–13. See Italy and Czechoslovakia Convention for the Prevention of Double Taxation and the Settlement of Other Questions Concerning Direct Taxation, and Final Protocol, signed 1 March [1925] LNTSer 129; 1924, 36 LNTS 229 (entered into force 19 January 1925) (‘1924 Italy–Czechoslovakia Treaty’).

[24] Ibid 9.

[25] Ibid 14.

[26] Ibid 12–13.

[27] See the discussion in ibid.

[28] Ibid 13.

[29] Ibid.

[30] Ibid 14 (emphasis in original).

[31] Ibid 15.

[32] Ibid.

[33] Ibid 32.

[34] Ibid.

[35] Ibid.

[36] Ibid 33.

[37] Ibid 20, 33.

[38] Ibid 20.

[39] See Graetz and O’Hear, above n 12, 1083–4.

[40] Committee of Technical Experts, Double Taxation and Tax Evasion: Report Presented by the Committee of Technical Experts on Double Taxation and Tax Evasion, League of Nations Doc C.216.M.85.1927.II (April 1927) (‘1927 Technical Experts Report’). This Report is reproduced in Joint Committee on Internal Revenue Taxation, Congress of the United States, Legislative History of United States Tax Conventions (US Government Printing Office, 1962) vol 4, 4111.

[41] Picciotto, above n 11, 22.

[42] 1927 Technical Experts Report, League of Nations Doc C.216.M.85.1927.II, 8. The other three drafts were Draft Bilateral Convention for the Prevention of Double Taxation in the Special Matter of Succession Duties; Draft of a Bilateral Convention on Administrative Assistance in Matters of Taxation; and Draft Bilateral Convention on Judicial Assistance in the Collection of Taxes.

[43] See ibid 10–18.

[44] John Avery Jones, ‘Categorising Income for the OECD Model’ in Luc Hinnekens and Philippe Hinnekens (eds), A Vision of Taxes within and outside European Borders: Festschrift in Honor of Prof Dr Frans Vanistendael (Kluwer Law International, 2008) 93, 97–100.

[45] Picciotto, above n 11, 22.

[46] 1927 Technical Experts Report, League of Nations Doc C.216.M.85.1927.II, 18.

[47] Ibid.

[48] General Meeting of Government Experts, Double Taxation and Tax Evasion: Report Presented by the General Meeting of Government Experts on Double Taxation and Tax Evasion, League of Nations Doc C.562.M.178.1928.II (October 1928) (‘1928 General Meeting Report’). This Report is reproduced in Joint Committee on Internal Revenue Taxation, Congress of the United States, Legislative History of United States Tax Conventions (US Government Printing Office, 1962) vol 4, 4151.

[49] 1928 General Meeting Report, League of Nations Doc C.562.M.178.1928.II, 7.

[50] Ibid 16.

[51] Ibid 17.

[52] Ibid 7.

[53] Ibid 19.

[54] Ibid 20.

[55] Avery Jones, ‘Categorising Income for the OECD Model’, above n 44, 97–100.

[56] See, eg, Hungary and Poland Convention for the Prevention of Double Taxation in the Matter of Direct Taxes, signed 12 May [1931] LNTSer 228; 1928, 123 LNTS 47 (entered into force 21 August 1931); Roumania and Yugoslavia Convention Concerning the Abolition of Double Taxation and Regulating Administrative Assistance in the Matter of Direct Taxes, signed 30 January [1934] LNTSer 53; 1933, 146 LNTS 99 (entered into force 1 January 1934).

[57] See, eg, Hungary and Sweden Convention for the Avoidance of Double Taxation in the Matter of Direct Taxes, and Final Protocol, signed 17 June [1938] LNTSer 6; 1936, 184 LNTS 11 (entered into force 1 January 1938); France and Sweden Convention for the Avoidance of Double Taxation and for the Establishment of Rules for Reciprocal Administrative Assistance in the Case of Direct Taxes, with Protocol, Complementary Declaration and Final Protocol, signed 24 December [1938] LNTSer 3; 1936, 184 LNTS 35 (entered into force 1 January 1938); Agreement between the Kingdom of Roumania and the German Reich for the Avoidance of Double Taxation in the Matter of Direct Taxes, published 8 February 1937, 93 TNI 251-96. All of these treaties contain a provision entitling taxpayers to invoke a mutual agreement procedure to attempt to resolve any double taxation that may arise. All of these treaties also contain definitions of fiscal domicile.

[58] See, eg, Germany and Sweden Agreement for the Adjustment of Internal and External Taxation and in Particular for the Prevention of Double Taxation in the Matter of Direct Taxes, with Final Protocol, signed 25 April [1928] LNTSer 230; 1928, 81 LNTS 281 (entered into force 14 August 1928); Germany and Finland Convention Concerning Legal Protection and Assistance in Matters Relating to Taxation, and Final Protocol, signed 25 September [1936] LNTSer 219; 1935, 173 LNTS 11 (entered into force 1 January 1936); Denmark and Finland Agreement for the Prevention of Double Taxation in the Matter of Direct Taxes, and Final Protocol, signed 2 December [1938] LNTSer 106; 1937, 187 LNTS 379 (entered into force 6 April 1938).

[59] United States of America and France Convention Concerning Double Taxation, and Protocol, signed 27 April [1936] LNTSer 4; 1932, 164 LNTS 211 (entered into force 1 January 1936). As will be discussed in more detail below this treaty did not comprehensively deal with all forms of income and had some unique features arising from the French tax on securities. The United States had entered into an exchange of notes with France in 1926 which exempted navigation enterprises of one territory operating within the other territory from taxation on the profits accruing exclusively from navigation: United States of America and France Exchange of Notes Constituting an Arrangement Regarding Relief from Double Income Tax on Shipping Profits, exchanged 11 June and 8 July [1931] LNTSer 91; 1927, 114 LNTS 413 (entered into force 8 July 1927). The United States had also entered into a limited agreement with Denmark and Iceland on the taxation of shipping profits in 1922: United States of America and Denmark and Iceland Exchange of Notes Constituting an Arrangement for the Reciprocal Exemption of Shipowners from Income Taxation, exchanged 24 and 28 October and 5 December [1931] LNTSer 74; 1922, 113 LNTS 381 (entered into force 6 December 1922).

[60] United States of America and Sweden Convention for the Avoidance of Double Taxation and the Establishment of Rules of Reciprocal Administrative Assistance in the Case of Income and Other Taxes, with Protocol, signed 23 March [1940] LNTSer 2; 1939, 199 LNTS 17 (entered into force 1 January 1940) (‘United States–Sweden Treaty 1939’).

[61] Convention for the Avoidance of Double Taxation and the Establishment of Rules of Reciprocal Administrative Assistance in the Case of Income and Other Taxes, United States–France, signed 25 July 1939, TS 988 (entered into force 1 January 1945) (‘United States–France Treaty 1939’). This Convention is reproduced in Charles I Bevans, Treaties and Other International Agreements of the United States of America 1776–1949 (US Government Printing Office, 1971) vol 7, 1046.

[62] Convention, United States–Canada, signed 4 March 1942, TS 983 (entered into force 15 June 1942) (‘United States–Canada Treaty 1942’). This Convention is reproduced in Charles I Bevans, Treaties and Other International Agreements of the United States of America 1776–1949 (US Government Printing Office, 1971) vol 6, 244.

[63] Convention, United States–United Kingdom, signed 16 April 1945, TIAS 1546 (entered into force 25 July 1946). This Convention is reproduced in Charles I Bevans, Treaties and Other International Agreements of the United States of America 1776–1949 (US Government Printing Office, 1974) vol 12, 671.

[64] United Kingdom of Great Britain and Northern Ireland and Ireland Agreement in Respect of Double Income Tax, signed 14 April 1926, 933 UNTS 169 (entered into force 4 August 1926). The treaty provided for relief from double income tax to be provided in accordance with and under the provisions of s 27 of the Finance Act 1920 (UK) 10 & 11 Geo 5, c 18 at the lower of one half of the taxpayer’s appropriate rate of British income tax or the taxpayer’s appropriate rate of Irish Free State tax.

[65] Agreement between the United Kingdom and Spain Regulating the Treatment of Companies, signed 27 June 1924, [1924] TS 25, Cmd 2212 (entered into force 27 June 1924). The Agreement basically deals with agency profits.

[66] Agreement between His Majesty’s Government in the United Kingdom and the Swiss Federal Council for Reciprocal Exemption from Taxation on Profits or Gains Arising through an Agency, signed 17 October 1931, [1932] TS 24, Cmd 4142 (entered into force 28 June 1932). This Agreement provided for the reciprocal exemption of certain agency profits.

[67] Picciotto, above n 11, 25.

[68] Ibid 13; Peter A Harris, Corporate/Shareholder Income Taxation And Allocating Taxing Rights between Countries: A Comparison of Imputation Systems (IBFD Publications, 1996) 287.

[69] For example, the relative importance of the impersonal and personal taxes in France at the time of the publication of Volume 1 of Taxation of Foreign and National Enterprises: A Study of the Tax Systems and the Methods of Allocation of the Profits of Enterprises Operating in More Than One Country In France, Germany, Spain, the United Kingdom and the United States of America, League of Nations Doc C.73.M.38.1932.II.A (15 February 1932) can be seen in the French report by Jean Guillet and Roger Pierre, which devotes 19 pages to impersonal taxes and one footnote to the French personal tax: Jean Guillet and Roger Pierre, ‘Tax System and Allocation Methods in France’ in Taxation of Foreign and National Enterprises: A Study of the Tax Systems and the Methods of Allocation of the Profits of Enterprises Operating in More Than One Country In France, Germany, Spain, the United Kingdom and the United States of America, League of Nations Doc C.73.M.38.1932.II.A (15 February 1932) 57.

[70] For example, in Germany and Italy Convention for the Avoidance of Double Taxation and the Settlement of Other Questions Connected with Direct Taxes, with Final Protocol, signed 31 October [1926] LNTSer 209; 1925, 53 LNTS 245 (entered into force 15 December 1925), the German Einkommensteuer (income tax), insofar as it was levied without regard to nationality, residence or sojourn of the person liable to taxation, and the German Körperschaftsteuer (tax on corporate bodies), insofar as it was levied without regard to the head office or actual business centre of the organisation, were characterised as impersonal taxes.

[71] See, eg, Germany and Sweden Agreement for the Adjustment of Internal and External Taxation and in Particular for the Prevention of Double Taxation in the Matter of Direct Taxes, with Final Protocol, signed 25 April [1928] LNTSer 230; 1928, 81 LNTS 281 (entered into force 14 August 1928); Agreement Between the German Reich and the Republic of Finland for the Avoidance Of Double Taxation in the Matter of Direct Taxes, signed 25 September 1935, 93 TNI 251-93 (entered into force 1 January 1936); Denmark and Finland Agreement for the Prevention of Double Taxation in the Matter of Direct Taxes, and Final Protocol, signed 2 December [1938] LNTSer 106; 1937, 187 LNTS 379 (entered into force 6 April 1938).

[72] United States of America and France Convention Concerning Double Taxation, and Protocol, signed 27 April [1936] LNTSer 4; 1932, 164 LNTS 211 (entered into force 1 January 1936) art I. The Convention restricted the source country from taxing industrial and commercial profits that were not attributable to a permanent establishment in the source country. It did not contain any articles requiring the residence country to either exempt or give a foreign tax credit to income that had been taxed in the source country. Nor did it envisage that a mutual agreement procedure would be invoked to resolve any international double taxation that would arise.

[73] The particular problems dealt with in the treaty are discussed in Letter of Submittal from Henry L Stimson, Secretary of State to Herbert Hoover, President, 2 June 1932. The position had been that although the subsidiary of the United States company had been subject to various French taxes on its income and dividends, the French also taxed United States parent companies on dividends they paid on the basis that part of the earnings and profits from which the dividends were paid was French profits. Under ibid art VI, France gave up its right to impose the tax on the United States parent company’s dividends subject to procedural conditions being complied with. Similarly French law also could subject to tax a United States company doing business in France through a permanent establishment. Article V enabled a United States corporation with a French permanent establishment to elect that its dividends not be taxed under this regime, provided it paid the tax on income from securities of the profits derived from all its permanent establishments in France. Articles VII–VIII gave the residence state the exclusive right to tax income from labour or personal services and war pensions. Under art IX, the source state was given the exclusive right to tax: (a) amounts received for the use of patents, secret processes and formulas, trade marks and other analogous rights; (b) copyright royalties; and (c) private pensions and life annuities.

[74] United States–Sweden Treaty 1939, art II gave the source state the exclusive right to tax industrial and commercial profits of a permanent establishment conducted there and required that the other state exempt the profits thus taxed. The article also provided that ‘[t]he competent authorities … may lay down rules by agreement for the apportionment of industrial and commercial profits.’ Article VII gave the primary right to tax dividends to the state of the shareholder’s residence but permitted the source state to deduct up to 10 per cent of tax at source. Under art VIII, the residence state was given the primary right to tax interest but the source state retained the right to deduct tax at source. Article VI gave the source state the exclusive right to tax mineral royalties and gave the residence state the exclusive right to tax other royalties. The foreign tax credit provisions were contained in art XIV.

[75] United States–France Treaty 1939. Article 3 gave the source state the non-exclusive right to tax industrial and commercial profits of a permanent establishment. Article 14A required the United States to grant a foreign tax credit in respect of French tax paid. Under art 14B, the relief provided by France varied according to whether or not the relevant French tax was schedular or not.

[76] United States–Canada Treaty 1942. Article III gave the source state the non-exclusive right to tax industrial and commercial profits of a permanent establishment. Under art XVI, international double taxation was to be resolved by a form of mutual agreement procedure.

[77] Convention, United States–United Kingdom, signed 16 April 1945, TIAS 1546 (entered into force 25 July 1946). Article III gave the source country a non-exclusive right to tax the industrial and commercial profits of a permanent establishment. Article VI(1) limited United States source taxation of portfolio dividends to 15 per cent and limited United States taxation of non-portfolio dividends (in relation to holdings of 95 per cent or more) to five per cent. Article VI(2) exempted dividends paid by a United Kingdom company to a United States resident from United Kingdom surtax. Subject to exceptions, art VII exempted interest from source taxation. Under art IX(1) United States source taxation on mineral royalties was limited to 15 per cent and under art IX(2) mineral royalties were exempt from United Kingdom surtax. Other royalties were exempt from source taxation under art VIII. Article XIII required each country to grant a foreign tax credit in respect of tax paid in the other country including, in some circumstances, underlying corporate tax.

[78] See Picciottio, above n 11, 41; John F Avery Jones, ‘The History of the United Kingdom’s First Comprehensive Double Taxation Agreement’ [2007] British Tax Review 211; Charles H Gustafson, ‘The History of Double Tax Conventions: The United States of America Experience’ (Paper presented at the History of Double Taxation Conventions Conference, Rust, Austria, 3–5 July 2008).

[79] See, eg, United States–Sweden Treaty 1939. Note the distributive rules in arts II (industrial and commercial profits), IV (ships and aircraft), V (real property), VI (royalties), VII (dividends), VIII (interest), IX (capital gains), X (wages salaries), XI (personal services), XII (students and business apprentices), and XIII (property or the increment of property). The avoidance of double taxation was dealt with in art XIV.

[80] See, eg, ibid art II. As noted earlier, however, this article gave the source country the exclusive right to tax industrial and commercial profits.

[81] See, eg, ibid art XIV.

[82] On the history of the development of the United States foreign tax credit in this period, see Graetz and O’Hear, above n 12, 1043–57.

[83] The United States foreign tax credit system at the time of the Carroll Reports is described in detail in Joseph Weare and M L McMorris, ‘Tax System and Allocation Methods in the United States of America’ in Foreign and National Enterprises (Volume II): Studies of the Tax Systems and the Methods of Allocation of the Profits of Enterprises Operating in More Than One Country Austria, Belgium, Czechoslovakia, Free City of Danzig, Greece, Hungary, Italy, Latvia, Luxemburg, Netherlands, Roumania and Switzerland, League of Nations Doc C.425.M.217.1933.II.A (30 September 1933) 199, 223–33.

[84] For the term ‘baskets of income’, see, eg, Picciottio, above n 11, 13.

[85] For brief critical histories of the logic behind the development of the basket system of foreign tax credit limitation, see Karen V Kole, ‘The Status of United States International Taxation: Another Fine Mess We’ve Gotten Ourselves into’ (1988) 9 Northwestern Journal of International Law & Business 49; Julie A Roin, ‘The Grand Illusion: A Neutral System for the Taxation of International Transactions’ (1989) 75 Virginia Law Review 919.

[86] There were five reports in total. The first three reports were descriptions by national reporters of how the countries in question taxed enterprises operating in more than one country. Volume 4 contained the key report in which Carroll summarised and synthesised the approaches taken by the reporting countries and produced recommendations.

[87] See Picciotto, above n 11, 29.

[88] See the discussion of Carroll Reports in ibid 27–31.

[89] The 1933 Draft Convention is reproduced in Mitchell B Carroll, ‘Allocation of Business Income: The Draft Convention of the League of Nations’ (1934) 34 Columbia Law Review 473, 494–8.

[90] 1933 Draft Convention art 1.

[91] Ibid art 3.

[92] Ibid.

[93] Ibid art 5. See the discussion of the Allocation Convention in Picciotto, above n 11, 31–2.

[94] Picciotto, above n 11, 31–2.

[95] For example, art II of the United States–Sweden Treaty 1939 provided that the source state could tax such of the industrial and commercial profits of a permanent establishment as were ‘allocable to’ the permanent establishment and further provided that ‘[t]he competent authorities of the two contracting States may lay down rules by agreement for the apportionment of industrial and commercial profits.’ By contrast, art III, which dealt with what we would now describe as associated enterprises, provided that

any profits which should normally have appeared in the balance sheet of the latter enterprise but which have been in this manner diverted to the former enterprise may … be incorporated in the taxable profits of the latter enterprise. In such case, consequent rectifications may be made in the accounts of the former enterprise.

[96] United Kingdom of Great Britain and Northern Ireland and Australia Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 29 October 1946, 17 UNTS 181 (entered into force 3 June 1947) (‘1946 United Kingdom–Australia Treaty’).

[97] The use of the arm’s length standard in both the industrial and commercial profits article and in the associated enterprises article of the Treaty was a cause of considerable disagreement between the United Kingdom and Australia during the negotiation of the 1946 United Kingdom–Australia Treaty as Australia wanted a savings clause inserted preserving the operation of then s 136 of the Income Tax Assessment Act 1936 (Cth). In the course of these negotiations R Willis, then Secretary of the United Kingdom Board of Inland Revenue, commented on the arm’s length principle in a letter to M J Mair, the Australian Second Commissioner of Taxation, as follows: ‘This is the League of Nations principle, and we regard it as fundamental to any double taxation agreement dealing with trading profits … It is difficult to see what other principle there could be to go on once one rejects the profits shown by the accounts’: Letter from R Willis to M J Mair, 20 June 1946, United Kingdom National Archives, IR 40/13740, 197.

[98] Royal Commission, Report of the Royal Commission on the Income Tax, Cmd 615 (1920).

[99] See ibid 16 [69]–[70]. See also Sub-Committee, ‘Report of the Sub-Committee Appointed by the Royal Commission on the Income Tax to Confer with Representatives of the Dominions on the Subject of Double Income Tax within the Empire’ in Royal Commission, Report of the Royal Commission on the Income Tax, Cmd 615 (1920) 168, 170–1. For contemporary descriptions of the operation of Dominion Income Tax Relief, see Ronald Staples, Dominion (Including the Irish Free State) Income Tax Relief: Law and Practice, over Seventy Examples; Appropriate Rate Tables (Gee & Co, 1925); R L Renfrew, The Practice of Dominion Income Tax Relief (Solicitors’ Law Stationary Society, 1934). For an account of the background to the introduction of Dominion Income Tax Relief and the negotiations with the dominions and the deliberations of the Sub-Committee, see C J Taylor, ‘“Capacity to Put up a Fight More Important Than Intimate Knowledge of Income Tax Acts and Practice”: Australia and the Development of the Dominion Income Tax Relief System of 1920’ (Paper presented at the 5th International Accounting History Conference, Banff, Canada, 9–11 August 2007).

[100] It appears likely that such relief was rarely, if ever, given in practice as each dominion had its own unilateral system for relieving international double taxation. In Australia’s case, for example, foreign source income was not taxable at all until 1930, and from 1930 until the first double taxation treaty between Australia and the United Kingdom in 1946 a broad exemption for taxed foreign source income operated.

[101] The approach of the United Kingdom Inland Revenue Department to the implementation of the scheme is set out in a circular to HM Inspectors of Taxes United Kingdom, entitled Finance Act, 1920 Section 2; Relief in Respect of Dominion Income Tax (December 1920). The printer’s copy of this circular with handwritten corrections and annotations is contained in United Kingdom National Archives file IR 40/2560.

[102] In fact only India ever fully provided reciprocal relief. Australia provided reciprocal relief at the federal level only and, as a result, was treated as not reciprocating. Proviso (b) to s 27(4) of the Finance Act 1920 (UK) 10 & 11 Geo 5, c 18 applied where a dominion was not reciprocating. The proviso could mean that less United Kingdom tax was payable than where a dominion was reciprocating but its operation involved administrative complexities and depended on extra statutory concessions. For a description of difficulties associated with the operation of the system in these circumstances see Staples, above n 99, 37–9, 65–72 and Renfrew, above n 99, 14–23.

[103] Between 1946 and 1950 the United States entered into the following comprehensive double taxation treaties: United States of America and Union of South AfricaConvention for the Avoidance of Double Taxation and for Establishing Rules of Reciprocal Administrative Assistance with Respect to Taxes on Income, signed 13 December 1946, 167 UNTS 171 (entered into force 15 July 1952); United States of America and New ZealandConvention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 16 March 1948, 127 UNTS 133 (entered into force 18 December 1951); Netherlands and United States of AmericaConvention with Respect to Taxes on Income and Certain Other Taxes (with Protocol of Exchange of Instruments of Ratification), signed 29 April 1948, 32 UNTS 167 (entered into force 1 December 1948); Denmark and United States of AmericaConvention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 6 May 1948, 26 UNTS 55 (entered into force 1 December 1948); Belgium and United States of AmericaConvention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 28 October 1948, 173 UNTS 67 (entered into force 1 January 1953); United States of America and NorwayConvention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 13 June 1949, 127 UNTS 189 (entered into force 11 December 1951); United States of America and IrelandConvention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 13 September 1949, 127 UNTS 89 (entered into force 20 December 1951); Greece and United States of America Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 20 February 1950, 196 UNTS 291 (entered into force 30 December 1953). In addition, the Convention for the Avoidance of Double Taxation and the Establishment of Rules of Reciprocal Administrative Assistance in the Case of Income and Other Taxes, United States–France, signed 25 July 1939, TS 988 (entered into force 1 January 1945) was modified by an exchange of notes in 1946, and the treaties with Canada and the Union of South Africa were modified by agreements made in 1950. The United States also offered to enter into a double taxation agreement with Australia in 1946 but Australia did not agree to commence negotiations for a treaty until 1952.

[104] See, eg, United Kingdom of Great Britain and Northern Ireland and France Agreement for the avoidance of Double Taxation with Respect to Taxes on Income, signed 19 October 1945 (entered into force 1 January 1945); Canada and United Kingdom of Great Britain and Northern Ireland Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 5 June 1946, 27 UNTS 207 (entered into force 6 November 1946); Agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and Southern Rhodesia for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 3 September 1946; United Kingdom of Great Britain and Northern Ireland and Union of South Africa Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 14 October 1946, 86 UNTS 77 (entered into force 13 February 1947); United Kingdom of Great Britain and Northern Ireland and Australia Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 29 October 1946, 17 UNTS 181 (entered into force 3 June 1947); United Kingdom of Great Britain and Northern Ireland and New Zealand Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 27 May 1947, 17 UNTS 211 (entered into force 8 August 1947); United Kingdom of Great Britain and Northern Ireland and British Guiana Arrangement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Income, published 8 August 1947 (ratified 23 September 1947); Arrangement between the United Kingdom and Cyprus for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 8 August 1947, BSP & Other 1774 p 1090–1095/1947; United Kingdom of Great Britain and Northern Ireland and Mauritius Agreement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 8 August 1947 (ratified 29 August 1947); United Kingdom of Great Britain and Northern Ireland and Northern Rhodesia Arrangement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Income, published 8 August 1947 (entered into force 6 April 1945); The Double Taxation Relief (Taxes on Income) (Seychelles) Order, United Kingdom–Seychelles, signed 8 August 1947; United Kingdom of Great Britain and Northern Ireland and Trinidad and Tobago Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 8 August 1947 (ratified 8 August 1947); United Kingdom of Great Britain and Northern Ireland and Aden Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 17 March 1948); United Kingdom of Great Britain and Northern Ireland and Antigua Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 30 January 1948); United Kingdom of Great Britain and Northern Ireland and Gambia Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 19 December 1947); United Kingdom of Great Britain and Northern Ireland and Gold Coast Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 19 December 1947); United Kingdom of Great Britain and Northern Ireland and Montserrat Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 19 December 1947); United Kingdom of Great Britain and Northern Ireland and Nigeria Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 19 December 1947); United Kingdom of Great Britain and Northern Ireland and Nyasaland Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 11 March 1948); United Kingdom of Great Britain and Northern Ireland and Palestine Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 1 April 1948); United Kingdom of Great Britain and Northern Ireland and St Christopher and Nevis Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 28 January 1948); United Kingdom of Great Britain and Northern Ireland and Sierra Leone Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 19 December 1947 (ratified 16 February 1948); Netherlands and United Kingdom of Great Britain and Northern Ireland Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 15 October 1948, 74 UNTS 3 (entered into force 18 July 1950); United Kingdom of Great Britain and Northern Ireland and Barbados Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 23 May 1949); United Kingdom of Great Britain and Northern Ireland and Jamaica Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 3 May 1949); United Kingdom of Great Britain and Northern Ireland and Malaya Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 4 March 1949); United Kingdom of Great Britain and Northern Ireland and Singapore Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 23 December 1949); United Kingdom of Great Britain and Northern Ireland and Dominica Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 14 March 1949); United Kingdom of Great Britain and Northern Ireland and Falkland Islands Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 2 June 1949); United Kingdom of Great Britain and Northern Ireland and Granada Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 2 May 1949); United Kingdom of Great Britain and Northern Ireland and St Lucia Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 13 May 1949); United Kingdom of Great Britain and Northern Ireland and St Vincent and The Grenadines Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 4 March 1949 (ratified 4 March 1949); United Kingdom of Great Britain and Northern Ireland and Sweden Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, 30 March 1949, 209 UNTS 129 (entered into force 16 September 1949); United Kingdom of Great Britain and Northern Ireland and Swaziland Arrangement for the Avoidance of Double Taxation with Respect to Taxes on Income, published 25 November 1949 (ratified 9 December 1949); United Kingdom of Great Britain and Northern Ireland and Israel Exchange of Notes Constituting an Agreement for the Continuance in Force of the Arrangement of 1947 for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income (with Annex), exchanged 10 February 1950, 86 UNTS 211 (entered into force 10 February 1950).

[105] For example, throughout the operation of the system, Australia and the United Kingdom used different approaches in calculating relevant tax rates when determining the amount of rebate allowed. Australia dissected accounts and made adjustments for differences in the corporate tax base of the two countries, processes which the United Kingdom continued to regard as unnecessary. The differences in approach can be seen in: Note by the Board of Inland Revenue on Dispatch of 2 May 1922 from the Governor-General of the Commonwealth of Australia on the Subject of Double Income Tax (October 1922); Letter from the Governor-General of the Commonwealth of Australia to the Secretary of State for the Colonies, 2 April 1924, forwarding a statement by the Commonwealth Commissioner of Taxation dated 29 January 1924; Note by the Board of Inland Revenue on dispatch of 2 April 1924 from the Governor General of the Commonwealth of Australia (1924), forwarding a statement by the Commonwealth Commissioner of Taxation on the subject of double income tax dated 29 January 1924. The first two documents are contained in Australian National Archives, Series A11804, Control Symbol 1926/317. The third document is contained in Australian National Archives, Series A461/8 Control Symbol D344/3/3 Part 1.

[106] Australia changed its treatment of non-resident holding companies in 1939 by removing the rebate for inter-corporate dividends which had previously been available to them. In 1940, Australia changed to a classical system by removing the rebate which had previously applied to dividends paid by resident companies to resident individual shareholders. The Australian view that developed as a consequence of this change was that the corporate tax was a distinct tax from the tax on shareholders. Thereafter, Australia, in providing reciprocal relief, only took into account the amount of Australian shareholder tax. At the same time it appears that, as a matter of practice, the United Kingdom did not gross up Australian source dividends for underlying Australian corporate tax for the purposes of calculating Dominion Income Tax Relief despite the relevant provisions of the Finance Act 1920 (UK) 10 & 11 Geo 5, c 18 requiring the grossing up of dividends. The end result was that, by 1946, United Kingdom companies deriving dividends from wholly-owned subsidiaries in Australia were subject to an effective rate of tax approaching 66.25 per cent. See the example of the operation of the system of Dominion Income Tax Relief in these circumstances in L S Jackson, Income Tax and Estate Duty. Double Taxation between Australia and the United Kingdom: Report of Mission to London (7 January 1946) 7–8 [20], National Archives of Australia, Series A461, Control Symbol D344/3/3 Part 2, Barcode 188492.

[107] For a discussion of the negotiation and drafting of the 1946 treaty between Australia and the United Kingdom, see C John Taylor, ‘The Negotiation and Drafting of the UK–Australia Double Taxation Treaty of 1946[2009] British Tax Review 201.

[108] See Agreement between the Government of the Commonwealth of Australia and the Government of the United Kingdom of Great Britain and Northern Ireland for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income and Capital Gains, signed 7 December 1967, [1968] ATS 9 (entered into force 8 May 1968).

[109] On the development of United Kingdom law and treaty practice in this period, see the discussion in John E Talbot and G S A Wheatcroft, Corporation Tax and Income Tax upon Company Distributions (Sweet & Maxwell, 1968) 279–80 [18-01], [18-07], 282–4 [18-08], 284–6 [18-09]–[18-11].

[110] United States of America and Australia Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 14 May 1953, 205 UNTS 253 (entered into force 14 December 1953).

[111] Agreement between the Government of the Commonwealth of Australia and the Government of Canada for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 1 October 1957, [1958] ATS 12 (entered into force 21 May 1958).

[112] Agreement between the Government of the Commonwealth of Australia and the Government of New Zealand for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, signed 12 May 1960, [1960] ATS 6 (entered into force 23 June 1960).

[113] The Australian reaction to the United Kingdom draft at the official level can be seen in Letter and accompanying memorandum from W J O’Reilly, Acting Second Commissioner of Taxation to Sir Richard Randall, Secretary of the Treasury, 16 November 1966, National Archives of Australia, Series Number A571, Control Symbol 1966/3007 Part 1 Double Tax Agreement, Australia and Britain. The reaction at cabinet level to the United Kingdom draft can be seen in William McMahon, Confidential for Cabinet Committee on Taxation Policy Submission 123: The Double Taxation Agreement with the United Kingdom (3 March 1967) National Archives of Australia, Series A5842, Control Symbol 123, Barcode 1857849.

[114] Correspondence, drafts and notes of the negotiations are contained in United Kingdom National Archives, Revision of Double Taxation Agreement Australia, IR 40/16741. The Australian strategy for the negotiations is set out in McMahon, above n 113, 1–11. For a more detailed account of the provisions of this treaty, see C J Taylor, ‘National Report for Australia’ (Paper presented at the History of Double Tax Conventions Conference, Rust, Austria, 3–5 July 2008).

[115] See the discussion of subsequent Australian treaty practice in Taylor, ‘National Report for Australia’, above n 114.

[116] Oleksandr Pastukhov, ‘International Taxation of Income Derived from Electronic Commerce: Current Problems and Possible Solutions’ (2006) 12 Boston University Journal of Science & Technology Law 310, 319 (citations omitted).

[117] The literature here is extensive. See, eg, ibid 319–20.

[118] See, for example, the discussion in ibid 321–2.

[119] See the discussion in ibid 320–1 of source rules for services.

[120] See OECD, ‘Part II — Treaty Characterisation Issues Arising from E-Commerce: (Report Adopted by the Committee on Fiscal Affairs) in 2002 Reports Related to the OECD Model Tax Convention (OECD, 2003) 33. The recommendations in this report were subsequently incorporated into the OECD’s commentary on the OECD Model. See also the discussion in Arthur J Cockfield, ‘The Rise of the OECD as Informal “World Tax Organization” through National Responses to E-Commerce Tax Challenges’ (2006) 8 Yale Journal of Law & Technology 136.

[121] See the discussion of the server/PE rule developed by the OECD in Cockfield, above n 120, 171–5.

[122] The OECD commentary now generally characterises the transmission of digital goods and services as generally resulting in business profits: OECD Committee on Fiscal Affairs, ‘Commentary on Article 12 Concerning the Taxation of Royalties’ in Model Tax Convention on Income and on Capital: Condensed Version (17 July 2008) 182, 191–2 [17.3]–[17.4]. At the same time, while the OECD commentary now regards a server as constituting a permanent establishment in certain circumstances, it rejects the proposition that a website in and of itself can constitute a permanent establishment and states that a foreign server maintained by an internet service provider will generally not constitute a permanent establishment: OECD Committee on Fiscal Affairs, ‘Commentary on Article 5 Concerning the Definition of Permanent Establishment’ in Model Tax Convention on Income and on Capital: Condensed Version (17 July 2008) 80, 98–100 [42.4]–[42.10]. Given these views, the avoidance of source-based taxation on, for example, the download of digital music would appear to be relatively easy.

[123] See Alvin C Warren Jr, ‘Commentary: Financial Contract Innovation and Income Tax Policy’ (1993) 107 Harvard Law Review 460.

[124] See the discussion of the undermining effects of put–call parity in ibid 465–73.

[125] See the discussion of possible approaches to formulaic taxation of contingent returns in ibid 477–82.

[126] See Michael J Graetz, ‘The David R Tillinghast Lecture — Taxing International Income: Inadequate Principles, Outdated Concepts, and Unsatisfactory Policies’ (2001) 54 Tax Law Review 261, 267. See also the literature noted by Graetz: at 267 n 27.

[127] For a recent review of the effect of ECJ decisions on the fundamental freedoms in the context of treaties, see Christiana H J I Panayi, Double Taxation, Tax Treaties, Treaty-Shopping and the European Community (Kluwer Law International, 2007) 163–73. For relevant ECJ case law involving the EC Treaty and bilateral treaties between member states, see, eg, Gilly v Directeur des Services Fiscaux du Bas-Rhin (C-336/96) [1998] EUECJ C-336/96; [1998] ECR I-2793; Gschwind v Finanzamt Aachen-Aussenstadt (C-391/97) [1999] EUECJ C-391/97; [1999] ECR I-5451; Compagnie de Saint-Gobain, Zweigniederlassung Deutschland v Finanzamt Aachen-Innenstadt (C-307/97) [1999] EUECJ C-307/97; [1999] ECR I-6161; F W L de Groot v Staatssecretaris van Financiën (C-385/00) [2002] ECR I-11 819; D v Inspecteur van de Belastingdienst (C-376/03) [2005] ECR I-5821; Denkavit Internationaal BV v Ministre de l’Économie, des Finances et de l’Industrie (C-170/05) [2006] EUECJ C-118/05; [2006] ECR I-11 949.

[128] Three cases that have considered the question are: Lasertec Gesellschaft für Stanzformen mbH v Finanzamt Emmendingen (C-492/04) [2007] EUECJ C-492/04; [2007] ECR I-3775; Skatteverket v A and B (C-102/05) [2007] ECR I-3871; Holböck v Finanzamt Salzburg-Land (C-157/05) [2007] EUECJ C-157/05; [2007] ECR I-4051. These cases are discussed in Axel Cordewener, Georg W Kofler and Clemens Philipp Schindler, ‘Free Movement of Capital and Third Countries: Exploring the Outer Boundaries with Lasertec, A and B and Holbock’ (2007) 47 European Taxation 371.

[129] See the discussion in Richard Lyal, ‘Free Movement of Capital and Non-Member Countries — Consequences for Direct Taxation’ in Dennis Weber (ed), The Influence of European Law on Direct Taxation Recent and Future Developments (Kluwer Law International, 2007) 17, 17–19.

[130] Michael J Graetz and Alvin C Warren Jr, ‘Income Tax Discrimination and the Political and Economic Integration of Europe’ [2006] YaleLawJl 54; (2006) 115 Yale Law Journal 1186.

[131] Ruth Mason, ‘US Tax Treaty Policy and the European Court of Justice’ (2005) 59 Tax Law Review 65.

[132] The Open Skies cases were a series of ECJ decisions on actions brought by the European Commission concerning bilateral air transportation treaties between the United States and several EU member states. The cases were: Commission of the European Communities v United Kingdom of Great Britain and Northern Ireland (C-466/98) [2002] EUECJ C-466/98; [2002] ECR I-9427; Commission of the European Communities v Kingdom of Denmark (C-467/98) [2002] EUECJ C-467/98; [2002] ECR I-9519; Commission of the European Communities v Kingdom of Sweden (C-468/98) [2002] EUECJ C-468/98; [2002] ECR I-9575; Commission of the European Communities v Republic of Finland (C-469/98) [2002] EUECJ C-469/98; [2002] ECR I-9627; Commission of the European Communities v Kingdom of Belgium (C-471/98) [2002] EUECJ C-471/98; [2002] ECR I-9681; Commission of the European Communities v Grand Duchy of Luxembourg (C-472/98) [2002] EUECJ C-472/98; [2002] ECR I-9741; Commission of the European Communities v Republic of Austria (C-475/98) [2002] EUECJ C-475/98; [2002] ECR I-9797; Commission of the European Communities v Germany (C-476/98) [2002] EUECJ C-476/98; [2002] ECR I-9855. In these decisions, the ECJ held that clauses in the agreements that enabled the United States to revoke, suspend or limit the operations of airlines that were not substantially owned by nationals of the other contracting state infringed the right to freedom of establishment in the EU Treaty. Graetz and Warren, above n 130, and Mason, above n 131, see the Open Skies cases as being problematic for limitation of benefits articles in United States double taxation treaties as the nationality requirements in the airline transport agreements had similar functions to ownership tests in limitation of benefits articles in United States double taxation treaties.

[133] Graetz and Warren, above n 130, 1248–51.

[134] See, eg, Servaas van Thiel, ‘Why the European Court of Justice Should Interpret Directly Applicable Community Law as a Right to Most-Favoured Nation Treatment and a Prohibition of Double Taxation’ in Dennis Weber (ed), The Influence of European Law on Direct Taxation Recent and Future Developments (Kluwer Law International, 2007) 75. Other commentators from within the EU do perceive threats to the sustainability of double taxation treaties due to a lack of compatibility with EU law. For example, Panayi, above n 127, 172 (emphasis in original) notes that it ‘could be argued that treaty benefits (eg credit or exemption), which reduce or eliminate double taxation only for certain undertakings or industries satisfying certain criteria, are a form of [prohibited] State aid.’ See further at 172–3.

[135] Council Directive 2003/123/EC of 22 December 2003 Amending Directive 90/435/EEC on the Common System of Taxation Applicable in the Case of Parent Companies and Subsidiaries of Different Member States, [2004] OJ L 7/41 (‘Parent–Subsidiary Directive’). In essence the Parent–Subsidiary Directive requires that: (1) cross-border dividends from an EU member state subsidiary to its EU member state parent (or a branch of the parent) be exempted from withholding tax; and (2) the EU member state of the parent company must either exempt the dividend or provide an underlying foreign tax credit for any corporate tax paid by the subsidiary in the other member state. See the discussion of the Parent–Subsidiary Directive in Ben J M Terra and Peter J Wattel, European Tax Law (Kluwer Law International, 5th ed, 2008) ch 9.

[136] Council Directive 2003/49/EC of 3 June 2003 on a Common System of Taxation Applicable to Interest and Royalty Payments Made between Associated Companies of Different Member States, [2003] OJ L 157/49 (‘Interest and Royalties Directive’). The Interest and Royalties Directive requires that a paying company in an EU member state (or its branch) must exempt from tax (whether levied on an assessment or on a withholding basis) payments of interest and royalties to a beneficial owner which is an associated company (or a branch) in another EU member state. See the discussion of the Interest and Royalty Directive in Terra and Wattel, above n 135, ch 13.

[137] Council Directive 2003/48/EC of 3 June 2003 on Taxation of Savings Income in the Form of Interest Payments, [2003] OJ L 157/38 (‘Savings Interest Directive’). The Savings Interest Directive requires ‘paying agents’ in EU member states making certain interest payments to a beneficial owner in another EU member state to either withhold tax from the payment or provide specified information to the tax authorities of the state of its establishment. That state, in turn, must exchange that information with the tax authorities of the member state of the beneficial owner of the payments. The decisive factor is that the paying agent is located within an EU member state. The source of the interest is irrelevant as is the residence of the debtor. See the discussion of the Savings Interest Directive in Terra and Wattel, above n 135, ch 14. Terra and Wattel consider that, although the Directive envisages the application of a withholding tax in the cross-border situation, the imposition of a withholding tax confined to payments to non-residents would infringe the right to the free movement of capital and payments in the EC Treaty, noting that, as secondary EC law, the Directive must be applied in conformity with primary EC law: at 638–9.

[138] Council Directive 77/799/EEC of 19 December 1977 on the Mutual Assistance by the Competent Authorities of the Member States in the Field of Direct Taxation, [1977] OJ L 336/15 (‘Assessment Assistance Directive’). The Assessment Assistance Directive is now confined to assistance in the assessment of direct taxes. The Directive requires the exchange of information, even if special investigations are required to obtain it, that may enable national administrations to effect a correct assessment of taxes. See the discussion in Terra and Wattel, above n 135, 666–80.

[139] Council Directive 2008/55/EC of 26 May 2008 on Mutual Assistance for the Recovery of Claims Relating to Certain Levies, Duties, Taxes and Other Measures (Codified Version) [2008] OJ L 150/28 (‘Recovery Directive’). The Recovery Directive covers a range of taxes, including taxes on income and capital as well as interest, administrative penalties and fines, and costs of recovery. Four types of assistance in recovery are provided for: (a) exchange of information on request; (b) notification on request; (c) recovery on request; and (d) conservatory measures upon ‘reasoned’ request. See the discussion in Terra and Wattel, above n 135, 680–2.

[140] Convention on the Elimination of Double Taxation in Connection with the Adjustment of Profits of Associated Enterprises (90/463/EEC) [1990] OJ L 225/10 (‘Arbitration Convention’).

[141] Panayi, above n 127, 232.

[142] For a recent discussion of the problems associated with the application of the ‘arm’s length’ standard to the transfer pricing of intangibles, particularly in the context of the extent to which non-traditional methods can be used in Australian domestic legislation, see Dylan Democrat Damon, ‘A Comparitive Analysis of the Legislative Framework for the Transfer Pricing of Intangible Assets in Australia and the United States’ (2008) 23 Australian Tax Forum 457.

[143] Picciotto, above n 11, 215.

[144] See OECD, E-Commerce: Transfer Pricing and Business Profits Taxation (OECD, 2005).

[145] See OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD, 2010).

[146] See the discussion on this point in Damon, above n 142, 480–91.

[147] A point, I think, first made by Richard Vann, ‘A Model Tax Treaty for the Asian-Pacific Region’ (1991) 45 Bulletin for International Fiscal Documentation 99. A similar point has also been made by John F Avery Jones, who notes that the OECD tends to change the Commentary rather than the Model and that the difficulty of changing treaties can also inhibit changes in domestic law where the features of domestic law were aimed at encouraging other countries to enter into treaties: Avery Jones, ‘The David R Tillinghast Lecture’, above n 5, 4.

[148] For example, Avery Jones, ‘The David R Tillinghast Lecture’, above n 5, 37–8 makes four suggestions for improvements to tax treaties ‘on the basis that we are locked into the present system.’ In his commentary on Jones’ lecture, Lee Burns discusses (1) a more refined model treaty; (2) a multilateral treaty; and (3) a model income tax law: Lee Burns, ‘Commentary’ (1999) 53 Tax Law Review 39.

[149] It is not suggested that the proposed solution would resolve all the problems with the sustainability of bilateral double tax treaty networks identified above. It is primarily focused on problems that arise due to the schedular structure of double tax treaties and the use of measured relief from international juridical double taxation. As discussed below it is recognised that source rules would continue to be necessary within the multilateral treaties envisaged in the proposed system. The proposed system as such is also not directly concerned with transfer pricing issues. As discussed below, however, it is submitted that the development of multilateral treaties within trading blocs should ultimately facilitate the adoption of formulary apportionment approaches to transfer pricing, which is consistent with the notional form of relief used in the composite tax system.

[150] See generally Harris, above n 68.

[151] Ibid 448.

[152] Harris gives examples showing that the composite tax system approach achieves a compromise between capital import and capital export neutrality: ibid 493–500. Harris also notes that early schedular income taxes of continental European countries and the early model tax treaties of the League of Nations had some similarities to the proposed composite tax system: at 449. One similarity was that under these systems and treaties, residence and source taxes were mutually exclusive. The division of taxing rights and the level of source-state taxation, however, was not based on an analysis of the consumption and production services provided by the source state.

[153] For a discussion of particular issues that arise under the composite income tax approach where foreign source income is derived through resident or non-resident companies, see ibid 507–10. For a discussion of the application of the composite tax system in the context of interest paid by residents to non-residents, see at 531–4.

[154] As discussed at above n 4, the 1925 Technical Experts Report made special mention of the 1921 collective convention between Austria, Hungary, Italy, Poland, the Kingdom of the Serbs, Croats and Slovenes and Roumania. As noted in Part III(B), the system of Dominion Income Tax Relief had multilateral elements to it. There have been several subsequent examples of multilateral treaties being developed within a trading bloc or between countries which were geographically proximate to each other or whose economics interests coincided. Examples of such multilateral treaties include: Denmark, Finland, Iceland, Norway and Sweden Nordic Convention on Social Assistance and Social Services, signed 14 June 1994, 1973 UNTS 354 (entered into force 1 October 1996); Agreement among the Governments of the Member States of the Caribbean Community for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income, Profits or Gains and Capital Gains and for the Encouragement of Regional Trade and Investment, signed 6 July 1994 (entered into force 1 January 1995); Regimen para Evitar la Doble Tributacion y Prevenir la Evasion Fiscal, published 4 May 2004 (entered into force 1 January 2005) (‘2004 Andean Community Income and Capital Tax Convention’). In addition, the following multilateral model double tax conventions have been developed: see, eg, Convention between the Government of _ and the Government of _ for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income (15 December 1987) (‘1987 Intra-ASEAN Model Double Taxation Convention’). Harris argues for the adoption of the composite tax system via a series of multilateral agreements within trading blocs: Harris, above n 68, 551–3.

[155] The notional relief envisaged to be granted by the residence country would only prevent residence source conflicts. Source rules would appear to be necessary to prevent more than one country taxing the same income on the basis that it was sourced within that country. Residence rules would also be necessary to prevent more than one jurisdiction taxing the worldwide income of a taxpayer on the basis that the taxpayer was a resident of that jurisdiction. Unlike the system of Dominion Income Tax Relief, confining credit relief to residents and having dual residence tiebreakers would also appear to be necessary to prevent double non-taxation.

[156] See, eg, Joann M Weiner, Using the Experience in the US States to Evaluate Issues in Implementing Formula Apportionment at the International Level (OTA Paper No 83, Office of Tax Analysis, United States Department of the Treasury, April 1999) <http://www.ustreas.gov/offices/tax-policy/library/ota83.pdf> .

[157] Clearly in the EU the requirement of unanimous votes for Council Directives on taxation has impeded the development of directives in this area. Nonetheless despite this restriction, as discussed in Part V(D) above, Directives have been able to be issued on a number of important and complex topics in cross-border taxation. Remembering that the intra-trading bloc treaty envisaged would only deal with matters such as an overall source rate or a means for setting one, source rules, residence rules, assistance in collection and arbitration, it is arguably likely that change would be required less frequently and would be easier to achieve especially if the trading bloc required something less than unanimity for the passing of tax legislation.

[158] A similar point is made by Victor Thuronyi, ‘International Tax Cooperation and a Multilateral Treaty’ (2001) 26 Brooklyn Journal of International Law 1641, 1658–9.