University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
MATTHEW COLLETT[*]
The purpose of this paper is to influence the way we think about the fiscal residence of multinational enterprises (‘MNEs’), and to propose a new test of corporate residence that might be adopted by countries which base their tax treaties on the Organisation for Economic Co-operation and Development Model Convention with respect to Taxes on Income and on Capital (‘OECD Model Tax Convention’).[1]
Internationally, tax authorities are likely to continue using residence as a basis for corporate income taxation because they assume that companies, like individuals, benefit from the use of a nation’s economic infrastructure. However, the existing rules, which are used to determine a company’s residence for taxation purposes, no longer function in some cases. Those rules were developed many years ago and were based on the assumption that most companies were incorporated in, and centrally managed and controlled from, the one state.
Advanced technology and the communications revolution are rendering many tax concepts based on physical presence, including central management and control and place of efective management, less appropriate and effective. This is because today company directors who are dispersed throughout the world are able to meet wherever they choose or may confer via videoconferencing without leaving their homes. Companies can be incorporated in places where they do not conduct business.[2]
This paper will incorporate a detailed case study to demonstrate that MNEs can manipulate the traditional residence rules to the point where residence becomes purely a matter of convenience. As a result, the tax base of many countries including Australia could be threatened. Accordingly, it is important to explore alternatives to the current test of corporate residence and the tiebreaker test for companies when developing tax treaties.
This paper will approach the problem of company residence by comparing companies to individuals. It is recognised that it is easier to link the right to tax individuals on the basis of residence than it is to make that link in the case of companies, due to the services and infrastructure the state provides for the benefit of individuals. However, by first considering the factors which tie an individual to a particular state, it will be shown that, by analogy, some of those factors apply to companies. It will be argued that, like an individual, one may consider that a MNE resides in the place where it has its closest economic, political, cultural and legal links. If this fails, a MNE’s residence may be where its greatest level of activity occurs, measured by an analysis of where the MNE’s most significant functions, assets and risks are located and managed.
Finally, an alternative ‘tiebreaker test’ will be put forward which – while not purporting to solve all the problems of corporate residence – is better aligned with commercial realities and is less capable of manipulation. It must be stressed, however, that this is intended to be an exploratory exercise rather than one which will yield a definitive outcome.
In this Part I will provide a brief outline of the international tax system and some of its key concepts.
All nations seek to obtain their fair share of tax[3] from the activities of MNEs within their jurisdictions. For example, Australia’s domestic law seeks to capture, for taxation purposes, the Australian source income of non-residents and the worldwide income of residents. Also, countries look to the international tax system to provide some means of allocating taxing rights over the profits of business activities which fall between jurisdictions and to avoid double taxation caused by residence-residence and residence-source conflicts.
While double taxation may be relieved unilaterally, many countries negotiate bilateral agreements (‘tax treaties’) as a means to avoid double taxation and to provide a framework for combating fiscal evasion between the contracting states. Amongst other considerations, tax treaties provide reciprocity and a measure of certainty for persons wishing to invest in the other treaty partner country. Although there are other model treaties (notably the United Nations Model Double Taxation Convention between Developed and Developing Countries (1999)), many countries follow the OECD Model Tax Convention when negotiating bilateral tax treaties.
According to the International Tax Glossary, the residence principle of taxation holds that ‘residents of a country are subject to tax on their worldwide income and non-residents are only subject to tax on domestic-source income’.[4] By contrast, under the source principle, a country seeks to tax ‘income arising within its jurisdiction regardless of the residence of the taxpayer, ie residents and non-residents alike are taxed on income derived from the country’.[5] A leading Canadian authority has noted that ‘[t]he source of income is not purely a matter of geography. The source of income may be what produces it, where it comes from, or who pays it.’[6]
Nations have mostly relied on a mixture of source and residence taxation and foreign tax credit systems or exemptions (domestically) and tax treaty ‘tiebreaker tests’ in double tax agreements (in the case of residence-residence conflict). These mechanisms serve as a prelude to source and residence taxation under tax treaties. Many countries agree to limit source taxation of business profits accruing to non-residents where there is no ‘permanent establishment’ (that is, a fixed place of business such as a mine or factory) to which those profits are attributable.
A commonly accepted theoretical basis for residence taxation is that residents enjoy the benefits of the social, economic, physical and legal infrastructure, which is paid for by tax revenue. On the other hand, the basis of source taxation may be that an entity makes use of the local infrastructure for the generation of income, profits or gains.[7]
At present, most countries use a mixture of source and residence taxation to tax both individuals and companies. There are considerable variations as to the composition of this ‘mix’. However, it is quite possible to tax companies solely according to where their income is sourced. This is evidenced by the Mexico draft (1943) and London draft (1946) of the League of Nations Model Bilateral Convention for the Prevention of the Double Taxation of Income (‘MBC’).[8] In these MBCs, even though a fiscal domicile was ascribed to individuals, no attempt was made to allocate a place of residence to companies. Further, these MBCs did not contain a tiebreaker test for companies. Taxing rights over company profits under these early treaties was allocated purely on the basis of source.[9] Given the widespread international practice of taxing the foreign source income of residents[10] and concerns about equity and erosion of the national tax base, it may be difficult to persuade the majority of countries to abandon the residence principle for companies.
For the purposes of this paper, the following broad definition of a ‘MNE’, adopted by the OECD in its Guidelines on Multinational Enterprises (1976), will be used:
[MNEs] usually comprise companies or other entities whose ownership is private, state or mixed, established in different countries and so linked that one or more of them may be able to exercise a significant influence over the activities of others, and, in particular, to share knowledge and resources with the others.[11]
For the sake of simplicity, the paper will focus on single companies which operate in several jurisdictions rather than company groups. Thus, I will look at MNEs and their representative offices, agencies and branches but subsidiaries and associated enterprises will not be considered.
C Central Management and Control
Along with the place of incorporation, the location of central management and control (‘CMC’) is the most important company residence test in common law jurisdictions. The place where a company has its CMC is a question of fact. It is generally the place where the directors meet to carry out the company’s business and to determine corporate policies, especially those relating to the company’s financial and business affairs.[12] The most significant factors used by Australian and United Kingdom courts to determine the place of CMC[13] include:[14]
• if the board is functioning in accordance with its responsibilities, the place where the board holds its meetings or otherwise exercises its decision-making power;[15]
• if the board is not functioning in accordance with its responsibilities, where the policy of the company is determined by the persons performing the functions which should be performed by the board;[16] and
• where the majority of directors reside and work.[17]
Some other factors, which may be relevant (but not necessarily decisive), include:
• where the dividends are declared;[18]
• where the general meetings are held;[19]
• where the company operates its bank account;[20] and
• where the company’s books and corporate seal are kept.[21]
By the early 20th century, the United Kingdom courts had recognised that a company may reside in more than one place.[22] However, as Lord Radcliffe explicitly recognised in Unit Construction Co Ltd v Bullock,[23] the courts merely determine whether a company is resident in the jurisdiction – that is, they are not called upon to resolve competing claims between jurisdictions. In other words, the CMC test is not a tiebreaker test.
Using the central management and control test in the modern world to determine residence is becoming more difficult because the place of CMC is frequently divided between two or more jurisdictions, leading to multiple residence. Moreover, in some cases it may be impossible to precisely locate the location of CMC. Even in the early cases, which were relatively uncomplicated by modern standards, the courts experienced difficulty in determining the location of CMC. For example, those cases involved deciding between two competing jurisdictions, and generally it was fairly clear where the controlling mind was located.[24]
In Esquire Nominees Ltd v Federal Commissioner of Taxation,[25] an Australian accountancy firm set up a company in Norfolk Island. The company’s directors all resided on Norfolk Island and all company and directors’ meetings were held there. Justice Gibbs (who was upheld on this point on appeal) found that the company was a resident of Norfolk Island. He held that while the directors, in fact, complied with the wishes of the accounting firm, the firm ‘had power to exert influence, and perhaps strong influence, on the appellant, but that is all.’[26] This case supports the contention that the central management and control test is easily manipulated to produce artificial results.
Modern methods of transportation and communication, such as videoconferencing and air travel, mean that company directors may be highly mobile and can hold board meetings and make important decisions without leaving their respective home jurisdictions. Other modern technologies are also likely to have a significant impact. According to Jai Mavani and Lubna Kably, ‘[w]ith the enactment of laws in several countries recognising electronic signatures, it would not be long before most countries give legal sanctity to resolutions passed via technological medium.’[27]
This test is used by a number of countries (including Germany and The Netherlands) to determine the residence of non-individuals. The term is often used interchangeably with ‘place of effective management’, which is discussed below. Klaus Vogel states that German case law provides that the place of management is where the management’s important policies are actually made.[28] This is the place where management directives are given, and not the place where they take effect. If these criteria cannot determine the place of management, then the place where the top manager resides will be used to determine residence – a somewhat arbitrary criterion if most or all of the activities of the company are elsewhere.
Tiebreaker rules have been developed to deal with the problem of double taxation arising from a ‘residence-residence’ conflict. This occurs where both contracting states treat a person as a resident for tax purposes under their domestic law (with the result that the person is fully liable to pay tax in both states). These tax treaties contain principles which can be utilised to break a deadlock and determine the state of residence for the purposes of the tax treaty.[29]
The corporate tiebreaker test used in the OECD Model Tax Convention is the place of effective management (‘POEM’).[30] The Commentary on article 4 states:
The place of effective management is the place where key management and commercial decisions that are necessary for the control of the enterprise’s business are in substance made. The place of effective management will ordinarily be where the most senior person or group of persons (for example a board of directors) makes its decisions, the place where the actions to be taken by the enterprise as a whole are to be determined; however, no definitive rule can be given and all relevant facts and circumstances must be examined to determine the place of effective management. An enterprise may have more than one place of management, but it can have only one place of effective management at any one time.[31]
‘Place of effective management’ is not further defined in the OECD Model Tax Convention Commentary. International commentators have attempted to explain it by reference to domestic law concepts, such as the location of CMC and ‘place of management’.[32] However, there appear to be differences between POEM and CMC in some important respects. This is because the place of effective management test focuses on where the high level decision making of a company occurs. This means that if a board of directors meet in State A, but the final directing power rests with other persons who meet in State B, CMC may be in State A and the POEM may be in State B.
It needs to be pointed out that while the place of effective management test may give a certain outcome, this outcome can be manipulated. If a company flies its board to a remote island to make its most important decisions, then that may be the POEM for that financial year, even though no other activity occurs there.
It has been widely accepted that individuals should be taxed on a residence basis because of the social, familial and economic ties to their place of residence. Multinational enterprises are not human and they do not have social or family ties although their decision-makers, owners and workers do. Nonetheless, companies, including MNEs, arguably do have features which link them to particular jurisdictions. They also consume public goods, benefit from the use of public infrastructure and have impacts on the environment – while the environmental impacts are not ‘paid for’ by tax, tax can be seen as a compensating mechanism.
In this section, I will compare companies to individuals using the framework of article 4(2) of the OECD Model Tax Convention, the residence tiebreaker provision.[33] Article 4(2) lists the factors which will determine the residence status of an individual who is a resident of both contracting states to a tax treaty.
The OECD Model Tax Convention is a template tax treaty. Its main purpose is to relieve international (juridical) double taxation, which can be generally defined as ‘the imposition of similar taxes by two (or more) States on the same taxpayer in respect of the same subject matter and for identical periods’.[34]
Article 4 of the OECD Model Tax Convention defines ‘resident’ by reference to the domestic law of the two contracting states, thus creating the possibility that a company will be a resident of both states. However, articles 4(2) and 4(3) contain ‘tiebreaker’ clauses which allocate residence for the purposes of the Convention to one of the two states.[35]
In the case of an individual, the tiebreaker rules look at various indicia of personal attachment to a state with a view to determining which state ‘it is felt to be natural that the right to tax devolves.’[36] In the case of a company, the tiebreaker test is based on the place of effective management.
If Australia regards a company as its resident because the company was incorporated in Australia, and Country B considers the company to be its resident on the basis that the company’s administrative or practical management occurs in Country B, then the tiebreaker rules will operate to determine the company’s residence status. Similarly, in the case of an individual, Australia may claim the individual as a resident for tax purposes on the basis that his or her domicile is in Australia and his or her permanent abode is not outside Australia, whereas Country B may regard registration or the grant of a permanent visa as the criterion for residence.[37]
The first limb of article 4(2)(a) sets out the first individual tiebreaker test as follows: ‘he shall be deemed to be a resident only of the State in which he has a permanent home available to him’.[38] The OECD Model Tax Convention Commentary on article 4 states that a permanent home is located in the place where an individual owns or possesses a home which is arranged and retained for permanent use as opposed to a stay of short duration.[39]
The closest equivalent to a ‘permanent home’ for a MNE might be its headquarters. However, the headquarters may become less important, as more and more MNEs move from hierarchical structures to different business models. For example, Rio Tinto claims that it now has largely autonomous business centres scattered around the world.[40] Another possibility, according to Peter Muchlinski, is that ‘the firm itself could be reorganized into smaller, self-standing units of decision-makers who will come together in a mix that fits the business tasks the firm faces, without creating permanent organizational structures.’[41]
Given that the headquarters may take on different roles and functions it is unlikely that the concept of ‘permanent home’ will be useful in determining where a MNE resides.
The second limb of article 4(2)(a) states that ‘if he has a permanent home available to him in both States, he shall be deemed to be a resident only of the State with which his personal and economic relations are closer (centre of vital interests)’.[42] The OECD Model Tax Convention Commentary on this article lists several factors which may be considered in deciding the ‘centre of vital interests’. They are: an individual’s family and social relations; occupations; political, cultural or other activities; place of business; place where the individual administers his or her property; and location of possessions.
The benefit of residence in the state where an individual’s vital interests are located is that that state’s laws, police, courts, hospitals, roads, air and water make the greatest contribution to his or her daily life, family life, and economic existence.
While a company does not have personal relations, it does have other economic, political, cultural and legal links.[43] For example, the shareholders, directors and lenders of a company – all of whom are necessary for the company’s continued existence – take a close interest in the company’s welfare. We can, therefore, draw an analogy between where a person’s parents and siblings live and where a company’s shareholders or directors live.
Multinational enterprises depend on the existence of company laws,[44] in particular those laws which permit the formation of a limited liability company and the conduct of mergers and acquisitions. They also depend on the existence of intellectual property laws,[45] which protect the MNE’s innovations and brands, and contract laws, which, for example, enable that intellectual property to be licensed. This legal infrastructure, which varies between states, impacts on the company’s economic existence. Without the limited liability framework, investors may be less willing to invest. Unless companies can merge, they may not achieve the critical mass necessary for international operations. Further, without the benefit of intellectual property laws, innovation would be unprotected from exploitation, reducing the incentive for a MNE to invest in research and development.
Multinational enterprises also have political and cultural ties. They may donate to political parties and lobby governments for policy change or assistance – including financial assistance in the form of subsidies or the removal of trade barriers to the company’s products. Companies may also donate to various cultural and sporting events. (The way the company sees itself may be useful in determining residency – does the company identify itself as Australian, American, British, or Japanese in its advertising? Does it mainly sponsor activities or causes in a particular country?) Arguably, it would be rational for MNEs to make the largest contributions in countries from which they receive the greatest protection and benefit. Multinational enterprises also have environmental ties with a country due to the natural resources they consume and the use they make of the land – consider for instance the environmental impact caused by BHP’s Ok Tedi mine.
The location of an individual’s principal assets – such as investments in realty (including the family home), securities, bank deposits and insurance policies – can help to determine residence. Likewise, a corporation may be deemed to be a resident in the place where its principal assets are situated. This may be the place where its patents and trademarks are (noting that intangibles contribute to a high percentage of a product’s value), where its profits are repatriated to, or where its real estate is located. Even if the major assets are not all located in the one jurisdiction, we may nonetheless be able to look at where there is a preponderance of assets.
Broadly speaking, therefore, companies may be considered to obtain the greatest benefit from, and be deemed to have the strongest ties to, those places where they use the facilities and the legal and economic infrastructure to the greatest extent. However, ‘nexus’ raises major problems of quantification and comparison. How do you measure an entity’s consumption of public goods in a particular jurisdiction, let alone undertake meaningful comparisons with competing jurisdictions? This is a daunting task at first sight. For example, how do you measure an entity’s use of the legal system to protect its patents? This use might range from registering their patents to seeking legal advice on possible infringements and taking court action.
Article 4(2)(b) of the OECD Model Tax Convention states that if the ‘centre of vital interests’ and ‘permanent home’ tests do not resolve the issue of residency, then the individual ‘shall be deemed to be a resident only of the State in which he has an habitual abode’.[46]
Looking at the OECD Model Tax Convention Commentary on the article, ‘habitual abode’ seems to be established by looking at the individual’s pattern of activity over a long period of time. For example, one may look at the individual’s stays in each contracting state.[47] The equivalent of ‘habitual abode’ for a company would be: which country does the company interact with on a regular basis?
What is called for here is a dominant presence in one jurisdiction. I would argue that this can be shown by a greater level of activity in one state over another – measured by an analysis of where the MNE’s most significant functions, assets and risks are located and managed. Further, there appears to be nothing that gives us less reason to assume that a company’s past habits represent its current disposition as to its centre of gravity – accordingly, we look at several years of income rather than only one year of income.
Functions, assets and risks (‘FAR’) is a term used in transfer-pricing methodology to help identify the most economically significant activities of the corporation. The purpose of this is to allocate profit appropriately. The type of functional analysis to be looked at is that which is set out in a number of Australian Taxation Office rulings.[48]
The following observations need to be made about using transfer-pricing methodology and functions, assets and risks analysis.
1 Functional analysis is factually based and is aimed at determining what the entity does, the origin and use of the information it holds, where it generates costs and value and how this might differ from other similar enterprises.[49]
2 It is not enough to merely list the organisation’s FAR. The vital part of the process is using this information to ascertain which are the most important FAR to the value added by the business activities of the enterprise.[50]
3 The focus of transfer-pricing methodology is to identify the source of profits. However, some FAR of an organisation are arguably also indicative of where the corporation really resides.
For example, several commentators agree that the treasury function of a MNE is integral to its continued existence. According to Paula Eastwood and Jerry Huynh, the treasury function is ‘integrated firmly in the value chain of the organization – central to its operations, planning and direction.’[51] Kirt Butler states that the treasury division of a MNE carries out the following functions: determining overall financial goals and strategies; managing and financing the corporation’s domestic and international trade; consolidating and managing financial flows; and identifying, measuring, and managing the firm’s overall exposure to risk.[52] These ‘higher level’ head-office type corporate functions can tell us where the ‘brain’ of the corporation is and, therefore, where it really resides.
The importance of the MNE’s principal assets in determining fiscal residence has been discussed above. Just as an individual’s principal assets help to determine where he or she really resides, the main assets of a corporation are also closely linked to a company’s fiscal residence, because the state protects those assets through its laws and enforcement agencies. An organisation’s main assets may include its intellectual property, key staff, bank accounts, and any tangible assets it holds, such as real estate. Again, we may look to where these assets are predominantly held.
In order to protect his or her assets, a prudent individual will attempt to foresee and guard against loss or damage. They may, for example, take out insurance, or upgrade home security etc. These risks will probably be greatest in the individual’s home jurisdiction. Similarly, a MNE attempts to anticipate and manage risks. There are many types of risks which confront a MNE, including credit risk, interest rate risk, foreign exchange risk, commodity risk and operational risk.
It is suggested that, in attempting to identify where a company is resident, we can look at both where risk is managed and where risks are located; however, we should give greater priority to the former. A head office will frequently shift a particular risk to a particular branch. They may, for example, shift the risk of obsolescence by sending outdated equipment to a branch.
An individual’s place of birth or nationality is one of the features which will, initially at least, define his or her residence for tax purposes. Under Australian law, individuals who are born here acquire a domicile of origin. However, the place of birth becomes less relevant if the individual subsequently acquires a domicile of choice or a permanent place of abode outside Australia.
The equivalent of place of birth or nationality for a company is incorporation. Incorporation is one of the tests of company residence under Australian law. This test has the advantage of simplicity and certainty. However, it is an easy test to manipulate. A MNE can easily incorporate itself in a country where it undertakes no economic activity. As Uta Kohl has pointed out, the ability to manipulate the incorporation test is being enhanced by modern technology, which is leading to a situation where online registration anywhere in the world will be commonplace.[53]
The decision to incorporate in a particular jurisdiction does not need to have any commercial effect. Therefore, the test may not provide an indication of where a MNE's economic ties really lie. Accordingly, it should not be used as a sole test in determining residence, although its certainty and simplicity mean that it may be useful as an alternative tiebreaker test in bilateral treaties.
It should be noted, however, that incorporation has been proposed as the sole test of company residence in the context of the Review of International Taxation Arrangements (‘RITA’). In its RITA Consultation Paper, the Australian Treasury recognised that the current tests of company residency are problematic, in that certain bodies, such as offshore subsidiaries and foreign companies that are party to a dual-listed company arrangement, risk being inadvertently treated as Australian resident companies.[54] The Treasury also noted that the place of CMC could be problematic to apply.[55]
The Treasury recognised that the test of incorporation had the advantages of simplicity and certainty, while being susceptible to manipulation.[56] In doing so, it referred to the current United States problem of ‘corporate inversions’ under which a parent, incorporated in a foreign tax haven is substituted for the United States incorporated listed parent.[57] (The United States has an incorporation test of residence for companies.) However, the Board of Taxation (‘the Board’) considered that some features of Australia’s tax system such as imputation would minimise the risk because the benefits of imputation are only available to Australian resident companies.[58]
In its final report, the Board recommended the adoption of an incorporation test on the ground of simplicity.[59] The Board agreed with the Treasury that the corporate inversion problem was unlikely to occur in Australia, partly because of the Australian imputation system.[60] The Government has since announced that it has deferred consideration of changes to the domestic tests of company residence recommended by the Board pending the release of a draft taxation ruling by the Australian Taxation Office to clarify the operation of those tests.[61]
A recent United States study of corporate inversions provides an interesting counterpoint to the Board’s views.[62] The author rejects the suggestion that a proposal to exempt foreign-source dividends from resident companies would place a brake on inversions (although this may be because the article was written prior to the formal announcement of President Bush’s corporate tax plan).[63]
In this section, I will present a tiebreaker test based on the foregoing analysis and identify and discuss some of the practical, administrative and political problems such a test might raise. It is suggested that a new article 4(3) should be inserted in the OECD Model Tax Convention, using the structure of article 4(2), as follows:
Where by reason of the provisions of paragraph 1 a person other than an individual is a resident of both Contracting States, then its status shall be determined as follows:
(a) it shall be deemed to be a resident only of the State with which its economic, cultural, political and legal relations are closer (centre of vital interests);
(b) if the State in which it has its centre of vital interests cannot be determined, it shall be deemed to be a resident only of the State in which it habitually locates and manages its most significant functions, assets and risks;
(c) if its most significant functions, assets and risks are evenly divided between both States or are predominantly located in neither of them, it shall be deemed to be a resident only of the State in which it is incorporated.
Any new test of corporate residence will need to be hard to manipulate, acceptable to other national tax administrations, able to withstand scrutiny from taxpayers and administrable in a self-assessment environment.
It is suggested that the tests proposed in paras (a), (b) and (c) will prevent manipulation. In the first instance, it will be difficult for the MNE to shift all of the factors that make up a MNE’s ‘centre of vital interest’ to a jurisdiction of convenience. Purely for cultural reasons, directors, shareholders, key employees and their families are unlikely to move en masse to such a jurisdiction.
Further, if the ‘centre of vital interests’ test cannot be determined, then we can use a functions, assets and risks approach, which takes into account all of the MNE’s most economically significant FAR over time. Once again, it is unlikely that a MNE could move all of these to a tax haven, as the requisite infrastructure would not be there. The incorporation test is included as a last resort as it provides a certain outcome if the first two tests fail.
The tests proposed above might be subject to criticism as being difficult to administer. In relation to the proposed ‘centre of vital interests’ test, some factors may be objectively ascertainable – such as place of residence of shareholders and directors, political donations, cultural benefactions and location of assets – although there will still be difficult issues raised especially in relation to the value of intangibles. Admittedly, more work would need to be done to determine how to quantify and compare other factors, such as the use of the legal system and the environment.
The functions, assets and risks test, based an assessment of the MNE’s most economically significant FAR, is open to differing interpretations. Nevertheless, the test appears to be more defensible and easy to administer when one recognises that functions, assets and risks analysis is already well established in the transfer-pricing field – tax administrations and MNEs are familiar with it.
The new tests may be acceptable to other tax administrations since they closely follow the ‘centre of vital interests’ and ‘habitual abode’ tests for individuals and they are designed to align with commercial realities.
Let us assume that a large diversified MNE, MegaCoUK (‘MCU’), is incorporated online in the British Virgin Islands (‘BVI’). It conducts no substantive activities in the BVI. The firm has been reorganised and no longer has a fixed head office as such anywhere in the world – although its treasury and research and development departments are based in the United Kingdom. Three of the directors live in Australia, three live in the United Kingdom, two live in Hong Kong, and two live in Singapore. General meetings are held in the United Kingdom where most of the shareholders live. Control of voting power is concentrated in the hands of the Singaporean and Hong Kong directors. The company’s books and corporate seal are kept in the BVI.
The company’s managing director is a dual resident of the United Kingdom and Singapore. She is highly mobile and frequently makes decisions while aboard aircraft in international airspace, or while visiting the company in its various countries of operation.[64] As the directors are geographically dispersed, they normally hold board meetings via videoconference. However, this year, the managing director, the chairman and the rest of the board are planning to fly to the BVI to hold their board meeting and make the most important decisions.
The company has branches in the United Kingdom (MegaCoUK United Kingdom (‘MCUK’)) and Australia (MegaCoUK Australia (‘MCA’)). MCA and MCUK are connected through a services agreement. This provides MCA with non-exclusive rights to provide certain services to MCUK, notably selling manufactured goods in Australia using the brand name and trademarks of MCUK in the Australian market. MCU makes large donations to political parties and cultural causes, particularly in the United Kingdom, and, to a lesser degree, in Australia.
The company’s treasury function is located in the United Kingdom. MCUK is responsible for developing the general worldwide marketing strategy. Both MCUK and MCA are involved in developing and implementing the marketing strategy for Australia. MCA undertakes market research in Australia.
Both MCUK and MCA are engaged in contacting and negotiating with customers, obtaining and granting approval for contracts, creating contract documentation and entering into contracts with customers.
Intangible-based assets significantly contribute to MCA’s operations. MCUK owns the trade names, trademarks, patents and technical know-how that are used by MCA, in Australia. However, at the time of writing, MCU has commenced registering some of its intellectual property in the BVI.
While MCUK and MCA both have employees who possess special skills and knowledge, such as product technical knowledge and market information, the bulk of those employees are located in the United Kingdom. MCA has bank accounts in the United Kingdom and Australia.
MCU’s risks are centrally managed by its treasury, located in the United Kingdom. While MCU bears the risks related to international shipping costs, and shares the bad debt risk, the other risks of the operation, including insurance risk and foreign exchange risk, which is brought about because the payment system for intercompany balances is denominated in pounds sterling, are borne by MCA.
MCU makes extensive use of the transport network in the United Kingdom and Australia to transport its goods from factories in the United Kingdom to its customers in Australia. It also makes extensive use of bandwidth in both countries to market its goods to customers online. MCU has used the legal system in all three countries to undertake cross-border acquisitions (when it acquired MCUK and MCA), and to register and subsequently defend its trademarks and patents (this has mainly occurred in the United Kingdom and Australia).
Applying each of the traditional tests of company residence (leaving aside the Australia-UK tax treaty):
(1) the place of incorporation test leads to an artificial result – the BVI – which is clearly based on tax planning considerations;
(2) the control of voting power test points to Hong Kong and Singapore, where no other activities are carried out;
(3) the central management and control test is inconclusive because an examination of the most important factors – where the majority of the company’s directors reside, where the policy of the company is developed, where the board holds its meetings – does not lead to a specific location; and
(4) the place of management and POEM tests lead to an artificial result (the BVI), which is clearly based on tax planning considerations.
Applying our proposed new test of company residence:
There are many factors which point to the United Kingdom as MCU’s ‘centre of vital interests’. Three out of ten directors, and most of the shareholders, are United Kingdom residents. Most of MCU’s patents and trademarks are currently controlled in the United Kingdom (although some of them are being moved to the BVI). Most of MCU’s donations and lobbying efforts are made in the United Kingdom. However, the company’s use of the economic and legal infrastructure in the BVI, Australia and the United Kingdom is inconclusive.
A functions, assets and risks approach indicates that many of the group’s high value functions are located in the United Kingdom. This is particularly true of the treasury, which is usually considered to be part of the CMC of a MNE. In addition, the group’s general worldwide marketing strategy and the marketing strategy for Australia are also partly developed and implemented in the United Kingdom and many of its key staff are there. The fact that MCUK (rather than MCU) has made a service agreement with MCA to carry out service functions suggests that the United Kingdom, rather than the BVI, is the state of residence. While most of the risks are borne in Australia, the treasury, located in the United Kingdom, manages MCU’s risks. It is seemingly not necessary to look to the place of incorporation – the BVI.
Residence taxation of multinational companies remains a live issue. Most countries still assert jurisdiction over those companies whom they consider owe them fiscal allegiance. However, the traditional bases of company residence tax are no longer valid in an age of electronic communications and widespread air travel.
As stressed above, this paper has been exploratory rather than definitive in nature. However, it seems likely that no single factor will be sufficient to determine where a company is resident. Moreover, just as individuals are deemed to be residents of the country where they have, among other things, their principal assets and family ties, a corporation can also be considered a resident in the place where its directors and shareholders live, its principal assets are, and where it carries out its main functions and manages its risks.
The alternative test proposed, looks in the first instance to the centre of vital interests and then to the place where most of the MNE’s highest value FAR are located or managed. It appears that this test is difficult to manipulate, administrable, potentially attractive to other tax administrations, and defensible in the face of taxpayer challenges.
[*] Analyst, International Tax and Treaties Division, Department of the Treasury. The views expressed in this article are the personal views of the author, and do not represent the views of the Department of the Treasury. The author would like to thank his current and former colleagues for their comments and suggestions.
[1] Articles of the Model Convention with respect to Taxes on Income and on Capital (2003) Organisation for Economic Cooperation and Development (‘OECD Model Tax Convention’) <http://www.oecd.org/dataoecd/52/34/1914467.pdf> at 20 November 2003.
[2] In fact, there is nothing new in this concept: see Egyptian Delta Land & Investment Co Ltd v Todd [1928] UKHL TC_14_119; [1929] AC 1 (‘Egyptian Delta’).
[3] The notion of what constitutes a ‘fair share of tax’ is open to debate, as evidenced for example by the controversy surrounding the formulary apportionment method for taxing MNEs: see, eg, Lindsay Célestin, The Formulary Approach to the Taxation of Transnational Corporations: A Realistic Alternative? (2000).
[4] Barry Larking (ed), The International Bureau of Fiscal Documentation Online International Tax Glossary (1996) <http://join.ibfd.nl/Taxglossary/> at 20 November 2003.
[5] Ibid.
[6] Robert Couzin, Corporate Residence and International Taxation (2002) 6.
[7] Justice Ian Gzell, ‘Residence and Permanent Establishments’ (Paper presented at the Fifth National Tax Retreat, Noosa, 7 August 1997) 1–2.
[8] League of Nations Fiscal Committee, The London and Mexico Model Tax Conventions – Commentary and Text, League of Nations Document C88 M88 1946, pt IIA (1946).
[9] See League of Nations, Model Bilateral Convention for the Prevention of Double Taxation of Income (Mexico Draft) (1943); League of Nations, Model Bilateral Convention for the Prevention of Double Taxation of Income (London Draft) (1946).
[10] Klaus Vogel, Klaus Vogel on Double Taxation Conventions (3rd ed, 1997) 11.
[11] Peter Muchlinski, Multinational Enterprises and the Law (1995) 13.
[12] Roger Hamilton, Robert Deutsch and John Raneri, Guidebook to Australian International Taxation (7th ed, 2001) [2.170].
[13] These are not determinative tests as such but factors a court will consider as part of a larger analysis.
[14] International Fiscal Association, ‘The Fiscal Residence of Companies’ (1987) 72a Studies on International Fiscal Law 205, 205–6.
[15] De Beers Consolidated Mines Ltd v Howe [1906] UKLawRpAC 41; (1906) AC 455.
[16] Koitaki Para Rubber Estates Ltd v Federal Commissioner of Taxation [1940] HCA 33; (1940) 64 CLR 15 (Dixon J) (‘Koitaki’). See also Malayan Shipping Co Ltd v Federal Commissioner of Taxation [1946] HCA 7; (1946) 8 ATD 75; Unit Construction Co Ltd v Bullock [1959] 3 All ER 831.
[17] Koitaki [1940] HCA 33; (1940) 64 CLR 15.
[18] John Hood and Co Ltd v Magee (1918) 7 TC 327.
[19] Ibid.
[20] Egyptian Delta [1928] UKHL TC_14_119; [1929] AC 1.
[21] Koitaki [1940] HCA 33; (1940) 64 CLR 15.
[22] Swedish Central Railway Co Ltd v Thompson [1925] AC 495.
[23] [1959] 3 All ER 831, 835.
[24] See, eg, Waterloo Pastoral Co Ltd v Federal Commissioner of Taxation [1946] HCA 30; (1946) 8 ATD 165; Koitaki [1940] HCA 33; (1940) 64 CLR 15.
[25] [1973] HCA 67; (1973) 129 CLR 177.
[26] Ibid 191.
[27] Jai Mavani and Lubna Kably, Ofice Space, New Economy Style – DTAA ’s (2001) Economic Times Online <http://www.etinvest.com/ettax/news/tmay19.htm> at 20 November 2003.
[28] Vogel, above n 10, 262.
[29] Philip Baker, Double Taxation Conventions and International Tax Law (2nd ed, 1994) 126–7; Mavani and Kably, above n 27.
[30] OECD Model Tax Convention, art 4(3).
[31] OECD Model Tax Convention, Commentary on art 4, [24].
[32] OECD, The Impact of the Communications Revolution on the Application of ‘Place of Efective Management’ as a Tie Breaker Rule: A Discussion Paper from the Technical Advisory Group on Monitoring the Application of Existing Treaty Norms for the Taxation of Business Profits (2001) [15]– [16] <http://www.oecd.org/dataoecd/46/27/1923328.pdf> at 20 November 2003; Baker, above n 29, 131– 132.
[33] OECD Model Tax Convention, art 4.
[34] OECD Model Tax Convention, Introduction [1].
[35] Baker, above n 29, 126 ff.
[36] OECD, above n 32, [9] quoting the OECD Model Tax Convention, Commentary on art 4 [10].
[37] Tom Magney, Australia’s Double Taxation Agreements: A Critical Appraisal of Key Issues (1994) 9–12.
[38] OECD Model Tax Convention, art 4(2)(a).
[39] OECD Model Tax Convention, Commentary on art 4 [8].
[40] See, eg, Rio Tinto, Management Overview (2003) <http://www.riotinto.com/aboutus/mgmtOverview.aspx> at 20 November 2003.
[41] Muchlinski, above n 11, 59.
[42] OECD Model Tax Convention, art 4(2)(a).
[43] It may be possible to look at objective indicia for companies despite the ‘separate entity’ principle, which many countries have: see, eg, Salomon v Salomon & Co Ltd [1897] AC 22.
[44] Muchlinski, above n 11, 38–47.
[45] Ibid.
[46] OECD Model Tax Convention, art 4(2)(b).
[47] OECD Model Tax Convention, Commentary on art 4 [17]–[19].
[48] See especially Australian Taxation Ofice Ruling TR 94/14; Australian Taxation Ofice Ruling TR 97/20; Australian Taxation Ofice Ruling TR 98/11.
[49] Australian Taxation Ofice Ruling TR 98/11, [5.46].
[50] Australian Taxation Ofice Ruling TR 97/20, [2.39].
[51] Paula Eastwood and Jerry Huynh, E-transformation of Treasury – A Tax Overview (2001) <http://www.inforich.com.au/Articles/TM2001/eTransformation.htm> at 20 November 2003.
[52] Kirt Butler, Multinational Finance (2nd ed, 2000) 303.
[53] Uta Kohl, ‘The Horror-scope for the Taxation Office: The Internet and its Impact on “Residence”’ [1998] UNSWLawJl 53; (1998) 21 University of New South Wales Law Journal 436.
[54] Commonwealth Department of the Treasury, Review of International Taxation Arrangements: A Consultation Paper Prepared by the Commonwealth Department of the Treasury (2002) <http://www.taxboard.gov.au/content/int_tax/downloads/ita.pdf> at 20 November 2003, 53–4.
[55] Ibid 54.
[56] Ibid 54–6.
[57] Ibid.
[58] Ibid 54.
[59] Board of Taxation, International Taxation: A Report to the Treasurer – Volume 1: The Board of Taxation’s Recommendations (2003) 106 ff.
[60] Ibid 108–9. The Board did not revisit in detail the other problems arising in relation to incorporation.
[61] Treasurer, ‘Review Of International Taxation Arrangements’ (Press Release, 13 May 2003) <http://www. treasurer.gov.au/tsr/content/pressreleases/2003/032.asp> at 20 November 2003.
[62] Reuven Avi-Yonah, ‘For Haven’s Sake: Reflections on Inversion Transactions’ (2002) 27 Tax Notes International 225, 225–231.
[63] Ibid. The author argues instead that the US should move away from incorporation towards a ‘properly defined and interpreted’ ‘managed and controlled’ test of corporate residence.
[64] OECD, above n 32, [43].
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2003/42.html