John Prebble[*]

10.1 Introduction and Overview

10.1.1 Summary

This chapter develops "Does the Use of General Anti-avoidance Rules to Combat Tax Avoidance Breach Principles of the Rule of Law? A Comparative Study".[1] That chapter concluded with a list of issues for attention in respect of the drafting and operation of sections BG1 and GA1 of the Income Tax Act 2007, known together as the general anti-avoidance rule.

The chapter paper considers and makes recommendations in respect of three issues:

(1) Should the Income Tax Act be amended to make it clear that sections BG1 and GA1 override double tax treaties that New Zealand enters?

(2) The Tax Administration Act 1994 empowers the Commissioner to issue private binding rulings in respect of all manner of questions that arise under the Income Tax Act and other tax legislation. Should that power be removed in respect of applications for rulings on the operation of the general anti-avoidance rule?

(3) Should the Commissioner issue interpretation statements on the policy and operation of the general anti-avoidance rule?

10.1.2 The nature of tax law

As explained in Chapter 4 of Learning from the Past, Adapting for the Future: Regulatory Reform in New Zealand, tax law is highly formalistic. This formal nature of tax law means that it is vulnerable to avoidance techniques. Broadly speaking, tax avoidance entails employing transactions that have the same economic substance as a taxable transaction, but a different tax result. For example, an individual may be taxed at a certain rate if he or she carries on business as a sole trader. By organising as a company the taxpayer may enjoy a lower rate of tax. Such a strategy does not necessarily amount to avoidance. The point made here is that a change from sole tradership to trading as a corporation entails no essential economic change. Nevertheless the tax result may vary considerably.

Tax law lends itself to avoidance because of the phenomenon of ectopia. "Ectopia" refers to the gap between the law on the one hand and the substance to which it relates on the other. In respect of tax law, and in respect of avoidance, the most important substance comprises business profits. There is, however, a significant gap between the law as drafted and the economic nature of business profits. This gap has been explained in a number of articles.[2] The result of ectopia is that people are able, as it were, to slip into the gap between tax law on one hand and the substance to which it relates on the other, by giving that substance a legal description that is different from the apparent nature of the transaction in question. Since tax law operates on the legal description of transactions rather than according to economic substance, and since a transaction's legal description can depart markedly from its economic substance, taxpayers often contrive to reduce or to eliminate tax by dint of the manner in which they frame their transactions.

10.1.3 Legislative response: a general anti-avoidance rule

Some legislatures have responded to the problems described by enacting general anti-avoidance rules. In New Zealand, the general anti-avoidance rule is section BG1 of the Income Tax Act 2007. In effect, section BG1 says that if an arrangement has the effect of avoiding tax then the arrangement is void against the Commissioner of Inland Revenue for tax purposes even though the arrangement complies with the rest of the Income Tax Act. By "void against the Commissioner for tax purposes" section BG1 means that the arrangement itself remains on foot and binding between the parties, but when it comes to income tax the parties cannot rely on the arrangement to determine their tax liability. Broadly speaking the law determines that liability according to the substance of the transaction.

When does the general anti-avoidance rule apply? The answer to this question is necessarily imprecise. It is imprecise because the general anti-avoidance rule is to some extent a contradiction in terms. That is, on one hand the taxpayer has complied with the law. On the other hand, the transaction has achieved a reduction in tax. Everybody agrees that the general anti-avoidance rule does not apply simply because there is a reduction in tax. As McCarthy P explained, a literal interpretation would:[3]

… result in the avoidance of transactions which were obviously not aimed at by the section. So the Courts have had to place glosses on the statutory language in order that the bounds might be held reasonably fairly between the Inland Revenue authorities and taxpayers.

For instance, taxpayers can reduce their tax by investing in tax-preferred investments, such as Kiwisaver.[4] Nobody suggests that such a transaction amounts to avoidance. Examples of when transactions do amount to avoidance are included in Chapter 4 of Learning from the Past, Adapting for the Future: Regulatory Reform in New Zealand.[5]

In order to respond to the many possible ways in which people can arrange their taxation and their business, it is necessary to have a rule of very broad scope. That is the general anti-avoidance rule. A consequence is that it is not possible to define exactly when the rule applies. The authors of "Does the Use of General Anti-Avoidance Rules to Combat Tax Avoidance Breach Principles of the Rule of Law? A Comparative Study" concluded that the imprecision of the general anti-avoidance rule does indeed amount to a breach of the rule of law.[6]

10.1.4 Justification of breach of the rule of law?

The conclusion that general anti-avoidance rules by their nature amount to breaches of the rule of law is not the end of the argument. Some philosophers argue that breaches of the rule of law may be justified occasionally if the circumstances or imperatives of society justify it.[7] The authors of "Does the Use of General Anti-Avoidance Rules to Combat Tax Avoidance Breach Principles of the Rule of Law? A Comparative Study" argued that two factors justify the breach of the rule of law that is occasioned by a general anti-avoidance rule. The first is the nature of tax law, which has been described. The second is the nature of certain tax practice. There is a well-funded industry that has planning to reduce the impost of taxation as its reason for existence. It is difficult for any government to counter this industry; the general anti-avoidance rule is necessary to arm the government to respond to creative tax planning.[8] The point just made is not meant to imply that all tax planning is evil as contrary to the interests of society; far from it. Tax planners advise their clients about pitfalls that the complexity of tax law places in their path. Many of the choices that tax law offers to people as they plan trading or investment structures are unexceptionable. The fact remains, however, that some tax planners see it as their role, or as part of their role, to help clients to avoid tax by contriving artificial transactions and structures that operate by altering the legal nature of those transactions and structures.

10.1.5 Matters arising from Stage One of the Regulatory Reform Project: Issues to be addressed

If one has decided, as the authors of "Does the Use of General Anti-Avoidance Rules to Combat Tax Avoidance Breach Principles of the Rule of Law? A Comparative Study" decided, that a general anti-avoidance rule is justified, a second question arises. That is how, if at all, should New Zealand's general anti-avoidance rule be improved in order better to regulate the fiscal system? This improvement could relate either to the terms of the law itself or to the manner in which the general anti-avoidance rule is administered.

At Stage One of this project the authors identified a number of issues for improvement that may justify attention. This present chapter considers three of those issues:

(1) General anti-avoidance rules and double tax treaties.

(2) Private binding rulings on tax avoidance.

(3) Interpretation statements by the Commissioner of Inland Revenue on avoidance questions and on the applicability of section BG1 of the Income Tax Act 2007.

These three questions will be considered below. In addition to the three questions just identified the authors identified a number of matters that would repay additional work. These matters are:

10.1.6 An enumerated anti-avoidance rule

Should New Zealand adopt an enumerated general anti-avoidance rule? This question relates to the nature of the New Zealand rule. The New Zealand rule is short and generalised. Some general anti-avoidance rules are much more elaborate, notably the rule in Australia.[9]

10.1.7 Publication of names of avoiders

Should the names of tax avoiders be published? It used to be that the names of tax evaders were published in the press once or twice a year.[10] Tax evaders are people who have engaged in illegal activity in order to reduce tax, perhaps simply not reporting part of their income. Avoidance appears to be becoming an increasing problem.[11] Should Parliament adopt, as a deterrent, a procedure whereby taxpayers who have been found to have engaged in transactions that are void have their names published in the press?

10.1.8 Anti-avoidance and transfer pricing

Are anti-avoidance rules in respect of transfer pricing adequate? Transfer pricing refers to the practice of setting prices when goods or services are traded between branches or companies that belong to the same global group. Take the example of oil. When a New Zealand-based company imports oil or petroleum typically the company will buy the product from another subsidiary in the same group. From the perspective of the group as a whole, and in the absence of tax considerations, it may not be important at what level to fix the intra-group price for this transaction. The reason is that if one company within the group makes an excessive profit at the expense of another company within the group it does not particularly matter to the ultimate owners of those companies because the owners are the same. They are the people who own shares in the overall holding company of the group in question.

When tax factors are added to the calculus, manipulation of transfer prices becomes relevant. Take again, for instance, oil imports. If the New Zealand subsidiary of a multinational petroleum group pays an excessive price to its foreign supplier, the result will be to increase the deductions of the New Zealand subsidiary for tax purposes. Increasing tax deductions causes a reduction in taxable profit. A reduction in taxable profit means a reduction in tax. For this sort of reason it is important to any country that prices fixed in transfer pricing transactions should be fixed at arm's length rates. Nevertheless, it is often very difficult to discover whether this has been done. The result is that tax avoidance by manipulation of transfer pricing can reduce the taxable income of companies in New Zealand that are members of multi-national groups. Issues of transfer pricing are becoming increasingly important, to the extent that it is arguable that they now constitute the most important issue in international taxation.[12] A question that should be addressed is whether New Zealand's anti-avoidance rules that target price manipulation by transfer pricing are sufficiently robust.

10.1.9 Is tax avoidance an area where responsive regulation might prove effective?

In this context "responsive regulation" refers to a practice of regulating citizens or industries by allowing a fair measure of trust, subject to a prior decision that one company or another is worthy of that trust. If an actor in an industry, in this case a taxpayer, is the beneficiary of responsive regulation practices, that actor will be troubled much less frequently by the actions of the state, in this case auditing for tax purposes.

10.1.10 Use of the tax system to promote economic policy

In addition to the questions described above, there are two areas where more general problems of tax avoidance come about. These areas comprise the difficulties that arise when the tax system is used to promote government social programmes or government economic goals. An example of a social programme is Working for Families.[13] This programme is delivered through the New Zealand tax system even though essentially it is a subsidy to poorer and middle income families with children.

An example of tax being used to promote economic goals is that mentioned already, Kiwisaver. Another comprises certain incentives that find a place in policy on research and development.[14] The problem with using a tax system and a tax statute to promote social and economic goals is that rules that are necessary for this purpose create loopholes in the statute. It is sometimes possible for taxpayers to exploit these loopholes for avoidance purposes.

10.2 The General Anti-avoidance Rule and Double Tax Agreements

10.2.1 Double taxation

Double taxation is a phenomenon that occurs internationally. Most jurisdictions tax income on the basis both of source and of residence.[15] The result is that if a resident of one country derives income that has its source in another country that income may bear tax in both jurisdictions. Double taxation is a serious consequence, because income tax systems do not automatically allow a credit or a deduction for tax on the same income that is levied by another jurisdiction.

There are many examples of double taxation. Two will suffice. Suppose a resident of New Zealand deposited money in an interest-bearing account in Arcadia. The bank would pay interest to the New Zealand resident. Arcadia would tax that outward-flowing interest. When the interest reaches the New Zealand resident, New Zealand will tax the interest a second time.

Another example could be an American oil-drilling company. Suppose that a company that has a licence to drill for oil in New Zealand engages the American oil-drilling company to carry out the drilling. The licence-holder will pay fees to the drilling company. New Zealand will tax the fees because the fees arise in New Zealand. That is, the fees have a New Zealand source. In addition the United States will tax the fees because the oil-drilling company is resident in, and incorporated in, an American state.[16]

The two examples are relatively simple. Double taxation can be much more complex. For instance, a few jurisdictions, notably the United States and the Philippines, tax income on the basis not only of source and of residence but also on the basis of citizenship, potentially leading to treble taxation. Another problem is that countries may give different answers to the question of whether someone is "resident" for tax purposes. The result may be that two countries will both consider that a particular taxpayer is resident within their jurisdictions. If that taxpayer earns income that has a source in a third jurisdiction there may be three countries that levy tax on the income.

10.2.2 Double taxation and national statutes

As mentioned, in principle and as it were by default, countries of residence do not allow a deduction or credit for tax that is levied by other countries at source. However, most countries vary this default position considerably. For example, in the case of business taxation, a country of residence will usually, in its income tax statute, grant credit to taxpayers who have suffered tax at source. Thus, for example, the United States would ordinarily grant a credit to the oil-drilling company mentioned above in respect of tax imposed on drilling profits in New Zealand. Such a credit is known as "unilateral relief".[17] For other countries, unilateral relief may constitute exempting foreign source income from taxation.

10.2.3 Double tax treaties

As far as double taxation is concerned, double tax treaties usually operate in broadly the same manner as national income tax legislation. That is, double tax treaties provide that jurisdictions of residence will grant credit for tax suffered at source, or sometimes wholly exempt that income from taxation.[18] Double tax treaties, as their name implies, are between states that consider or expect that their residents may suffer double taxation by virtue of the operation of the tax systems of the two jurisdictions. Why is it that double tax treaties are necessary when unilateral treaty relief is generally offered in most domestic tax law? There are two reasons.

First, double taxation treaty relief is often more generous than unilateral relief. In order to promote trade between their countries, two states may agree that they will grant relief to residents of the other country that is more generous than the relief that they grant in their national legislation.

Second, double tax treaties follow very uniform templates. As a result, people from all countries who are knowledgeable about international trade and investment are likely to be reasonably familiar with the broad terms of double tax agreements. Consequently, a foreigner dealing with a New Zealander can obtain a good understanding of New Zealand double tax relief by reading the treaty between the country of that foreigner and New Zealand. The proposed investor or trader will have more confidence in what is found in a treaty than in the domestic legislation of the other country, which will often be difficult for foreigners to understand.

Double tax agreements operate by overriding the provisions of national taxing statutes. For example, section BH1(4) of the New Zealand Income Tax Act 2007 says that a double tax agreement overrides the Income Tax Act in relation to income tax. Section BH1(4) means that where double tax relief is more generous than unilateral relief, or where there is no unilateral relief, relief under a double tax agreement will prevail.

10.2.4 Relationship between double tax agreements and general anti-avoidance rules

Bearing in mind that a double tax agreement overrides the terms of domestic tax legislation, the question arises as to whether a double tax agreement overrides the provisions of a general anti-avoidance rule. Suppose a trader or investor is able to exploit the terms of a double tax agreement in order to reduce taxation, does the double taxation agreement override the general anti-avoidance rule, as it were switching that rule off, or does the general anti-avoidance rule still apply? One argument is that if a taxpayer exploits a double tax treaty in order to avoid tax the transaction is immune from section BG1 of the Income Tax Act, the general anti-avoidance rule, because section BH1(4) ousts the operation of the general anti-avoidance rule.

10.2.5 Example of using a double tax agreement to minimise taxation: conduit companies

A "conduit company" is a company that is inserted between, for example, a borrower and a lender in order to obtain benefits under a double tax treaty. The benefit may be, for instance, a reduction in tax on interest.

Where interest is paid by a borrower in a source country to a lender in a residence country the ordinary tax result is that the jurisdiction of the borrower imposes withholding tax at a flat rate on the outward flowing interest. "Withholding tax" implies that debtors/payers must deduct tax from interest that they pay and must pay only the net interest to the lender in the residence country. Double tax treaties often reduce or eliminate withholding tax on outward flowing interest.[19] The problem with withholding tax on interest that flows internationally is that although in principle the borrower pays tax on behalf of the lender, in practice lenders require borrowers to pay interest in gross. If there is any tax the lender requires the borrower to bear it is the cost of withholding tax. Tax planning using conduit companies is an example of employing double tax agreements to minimise withholding tax on interest.

10.2.6 Northern Indiana Public Service Co v Commissioner of Inland Revenue

The Northern Indiana case involved a borrowing by a United States company on the Eurobond market.[20] The Eurobond market involved Europeans lending money internationally, typically to American borrowers. The problem that Northern Indiana faced as the borrower was the interest paid to European lenders it would have to bear and the United States withholding tax on interest that it was obliged to charge those lenders. The lenders required that their interest should be paid in full. Thus the cost of the tax, although in principle an impost on the lender, was visited upon Northern Indiana. At least, Northern Indiana would have suffered the tax had it not engaged in planning by establishing a conduit company.

One solution might have been to borrow from a Dutch company that would aggregate loans from throughout Europe. The advantage would be that the United States-Netherlands Income Tax Treaty reduced withholding tax on outward-flowing interest to zero.[21] Thus Northern Indiana could have paid interest to European lenders or, at least to a Dutch lender, without withholding any tax. The problem with this solution is that when the Dutch lender received the interest it would be obliged to pay tax at Dutch rates on that interest.

A more refined solution, which was adopted by Northern Indiana, was to establish a company in the Netherlands Antilles. The advantage of employing the Netherlands Antilles was that the Antilles counted as part of the Netherlands for purposes of the United States-Netherlands Income Tax Treaty. Further advantages were that the Antilles levied no tax on interest that flowed inwards or outwards. To take advantage of these benefits, Northern Indiana established a subsidiary in the Antilles that we may call "Finance".

Finance's method of operation was to borrow from European lenders, issuing bonds to them that carried 17.5 per cent interest. Finance then re-lent the borrowings to Northern Indiana at 18.5 per cent interest.

Interest flowed from Northern Indiana at 18.5 per cent. In calculating its profits for purposes of United States taxation Northern Indiana claimed a deduction of the 18.5 per cent interest. However, because of the United States-Netherlands Income Tax Treaty, as mentioned, Northern Indiana did not have to pay withholding tax on the outward flowing interest.

The result for Northern Indiana was that its cost of borrowing was in effect 17.5 per cent, because Finance was its subsidiary and Northern Indiana took advantage of the interest rate that Finance paid. The real significance of the transaction was that, without the conduit company, Northern Indiana would have incurred the additional cost of United States withholding tax on the outward-flowing interest. The Eurobond lenders would not have agreed to lend if they had had to bear that interest themselves. The terms of the loan required that Northern Indiana should bear the interest. That is, the overall result was that the tax that the United States Government would ordinarily have charged on the outward flowing interest remained as value within the corporate structure of Northern Indiana.

Finance is described as a "conduit company" because its function was to allow loans to flow through Finance to Northern Indiana and to allow interest to flow back to European lenders. From an economic and substantive perspective, Finance had no role. Its only importance was to reduce, indeed to eliminate, the tax that would otherwise have been borne by Northern Indiana.

The United States Internal Revenue Service challenged the tax result that Northern Indiana claimed. However the courts upheld the taxpayer and Northern Indiana won.[22]

10.2.7 Conduit structures and the New Zealand general anti-avoidance rule

In the context of the New Zealand Law Foundation Regulatory Reform Project the question arises, would section BG1 of the New Zealand Income Tax Act 2007 strike down a transaction like the Northern Indiana conduit case in the corresponding case of a New Zealand borrower contracting loans via a conduit company that had been established in the jurisdiction of a treaty partner? The answer to this question is uncertain. Scholars have taken opposite points of view.[23]

10.2.8 Legislative reform

As mentioned, the issue in respect of the case of conduit companies is whether the New Zealand Commissioner of Inland Revenue could apply the general anti-avoidance rule to frustrate tax planning. If yes, the Commissioner could collect withholding tax from the New Zealand borrower. If no, the Commissioner would forego the tax. This example is only one of many possible international tax planning schemes. There is uncertainty as to the outcome. As a result, some countries have legislated to provide that double tax treaties are subject to their general anti-avoidance rules.[24] The author submits that New Zealand should do the same.

Legislation to make treaties subject to general anti-avoidance rules would not be contrary to treaty policy. Tax treaties have several policies apart from minimising double taxation. One of these policies is to frustrate avoidance. Further, in principle double tax treaties follow the general rule of treaty interpretation; that is, treaties should be interpreted purposefully and substantively. The purpose of any double tax treaty is to confer benefits on the residents of the states that are parties to the treaty. In an example like Northern Indiana a resident of, say Spain was able to take advantage of the United States-Netherlands Income Tax Treaty by virtue of the establishment of Finance as a conduit company in the Netherlands Antilles.

The objectives of the United States-Netherlands Income Tax Treaty certainly did not include conferring benefits on residents of Spain. Thus, had a general anti-avoidance rule been applied in the Northern Indiana case there would have been no reason to suggest that there was a departure from the policy of the treaty.

There is one reservation to the conclusion in the last paragraph. This is that while the Official Commentary to the OECD Model Tax Convention on Income and Capital provides that double tax treaties do have an objective of frustrating tax avoidance, this objective is carried out by frustrating only transactions that have tax avoidance as a primary goal.[25] One argument is that the New Zealand general anti-avoidance rule, section BG1 of the Income Tax Act 2007, is in broader terms than this objective of the Commentary to the OECD Model Convention on Income and Capital. If so, the difference may matter because double tax agreements are interpreted partially by reference to the Official Commentary. Nevertheless treaty overrides, while uncommon, do occur. All nations are, at least to some extent, opposed to tax avoidance. It is hard to think that a treaty partner would object if New Zealand were to insist, by statute, that its section BG1 overrides attempts to avoid tax by exploiting or abusing double tax treaties.

10.3 Private Binding Rulings

10.3.1 Example: Challenge Corp Ltd v Commissioner of Inland Revenue[26]

The Challenge case involved a non-binding ruling by the Commissioner of Inland Revenue. Challenge was a profitable company. It wanted to take advantage of the loss consolidation provisions of the Income Tax Act 1976. Those provisions said that if the same persons held the shares in the same proportions in two companies at the end of the income year, then the losses of one company could be set off against the profits of the other. The provisions were in the Income Tax Act 1976 section 191(4) and (5). Challenge decided to take advantage of the rules and bought all the shares in a company called Perth. Perth was essentially defunct but had accumulated losses. There was in section 191(5)(C)(i), a specific anti-avoidance rule. This rule said that in determining whether companies could treat themselves as groups and therefore set profits and losses off against each other the Commissioner should disregard any alterations of shareholding that were of a temporary nature and that affected the incidence of income tax. As far as Challenge was concerned the purchase of the Perth shares was permanent. There was no point in Challenge wanting to resell the shares to anyone else because once Perth's losses had been set off against the profits of Challenge there would be no value left in Perth. Consequently, the rule in section 191(5)(C)(i) did not apply.

Challenge was concerned to ensure that if it went ahead with the Perth transaction the company would in fact be permitted to set losses off against profits. Challenge therefore inquired with the Commissioner. The Commissioner confirmed Challenge's interpretation. However, later the Commissioner changed his mind. He appears to have decided that the transaction with Perth was contrary to the spirit of the legislation. He therefore disallowed the attempt by Challenge to deduct the losses of Perth in consolidating the taxable income of Challenge. Challenge objected and won before the High Court and Court of Appeal, but the Privy Council upheld the Commissioner, holding that the general anti-avoidance rule prevented Challenge from using the loss consolidation provisions, even though Challenge satisfied the specific anti-avoidance rule in section 191(5)(C)(i), that is, even though the transfer of shares was permanent and not temporary.

10.3.2 Proposals for an advance rulings regime

Cases like Challenge were relatively common in the 1980s. This state of affairs led to a demand for provisions to be added to the income tax legislation to authorise the Commissioner to give rulings that would be binding. In 1986 the book Advance Rulings on Tax Liability was published.[27] This book surveyed several jurisdictions that allowed binding advance rulings, made recommendations for an advance rulings regime, and included a draft Bill. Eventually a regime similar to that proposed in the draft Bill was enacted in 1995 and added to the Tax Administration Act 1994 as Part VA of the Act.[28]

The rulings regime in Part VA of the Tax Administration Act 1994 covers a number of categories of rulings. For example, there is a category called a "product ruling". A product ruling is where the Commissioner rules on the tax consequences, of, for example, a particular investment that is to be offered to the public. This present chapter, however, is concerned with private rulings. Private rulings are rulings in respect of an individual taxpayer who proposes a particular transaction or structure.

Advance Rulings on Tax Liability had proposed that all rulings should be published including private rulings for single taxpayers, although in that case the ruling could be in a form that had been rendered anonymous. Parliament eventually rejected the proposal for anonymity and enacted Part VA, which provides that private rulings are secret.[29] The principal reason was that New Zealand is a small country and even if rulings were made anonymous it would be hard to hide the identity of the taxpayer in question. For example, if there were a ruling on a new aircraft lease, but only one airline happened to be acquiring aircraft at the time, people could easily guess which company the ruling applied to.

The problem of secrecy, however, is that a body of what is, in effect, secret law, develops. Rulings are not precedents in later cases, even in respect of the same taxpayer. Nevertheless the existence of a ruling that is known to taxpayers can offer advantages.

In the United States during the 1980s a practice of sharing rulings between large law firms had developed.[30] In New Zealand that practice does not appear to have become institutionalised. Nevertheless, people who are active in the area tend to discuss these matters and the nature of recent rulings becomes well known in restricted circles. Solicitors cannot circulate the rulings themselves because of client confidentiality, but the substance is often shared.

A second problem with secrecy is that since there is no public scrutiny of rulings the possibility of checking for quality is foregone. This possibility is by no means theoretical. A good deal of scholarly writing on taxation law has a theme of evaluating and criticising the work of either the Commissioner or Parliament. No doubt there would be similar criticism and evaluation of private rulings if people knew what they said.

10.3.3 General anti-avoidance rules and rulings

10.4 Proposed Commissioner's Interpretation Statement on the General Anti-avoidance Rule

10.4.1 Interpretation statements

10.4.2 Issue: Should the Commissioner publish interpretation statements on the general anti-avoidance rule?

In 1996, the author published "Commissioner's Interpretation Guidelines: some Consequences from Publishing Them".[38] (At the time what are now interpretation statements were called guidelines.)

The article of 1996 argued that despite the great pressure on the Commissioner to publish interpretation statements there are some areas of the law that are so inherently uncertain that interpretation statements are liable to do more harm than good. The article considered three examples: draft interpretation statements on form and substance; draft interpretation statements on shams; and the policy statement on section 99 of 1990. Essentially, the article argued that the task of composing interpretation statements on such inherently uncertain areas of law is not one that should be undertaken by the Commissioner. Parliament has left these areas for the courts for very good reason. They are simply not amenable to detailed legislation. While an interpretation statement does not claim to be legislation, it becomes in effect a kind of quasi-legislation. The reason is that, emanating from the Commissioner, an interpretation statement is thought to have greater authority than an opinion from someone in the private sector. This greater authority gives an interpretation statement quasi-legislative force in that the statement would appear to be a set of principles that the Commissioner will at least attempt to abide by. Because one cannot reduce the general anti-avoidance rule to a number of precise rules, or even of principles, results can be disappointing. Taxpayers, and even tax officials, may draw erroneous conclusions from the statement.

10.4.3 Comparing the statements of 1990 and 2011

Despite the matters described in the previous paragraph, when the draft of 2011 is compared with the statement of 1990 the 2011 draft does have some merits. In particular the draft of 2011 eschews two features of the 1990 policy statement. The first was that the 1990 statement laid down certain procedural requirements that the Commissioner imposed internally on tax officials when they dealt with general anti-avoidance questions.[40] Those procedural requirements were no doubt helpful, but they were not a requirement of law. Sometimes, depending on the facts of the case, going carefully through each of these requirements was not necessary. But modern tax litigation is at least as much about form and procedure as it is about substance. As a result, taxpayers from time to time tried to challenge the Commissioner's procedures on the basis of not having followed steps stipulated in the statement of 1990. These efforts by taxpayers were in the end fruitless, but they took up a good deal of everyone's time.[41]

A second feature of the statement of 1990 that is not present in the draft of 2011 is that in 1990 the statement included a good many examples of tax planning.[42] Some examples were said to be acceptable and some unacceptable.

The problem with some of the so-called acceptable plans was that they were framed too broadly. While in some circumstances following the plan would no doubt be acceptable, this was not always so. The result was that taxpayers would adopt features of these plans that they thought would sanitise their tax planning. Objective assessment showed that this sanitisation should not have occurred, although sometimes taxpayers were successful.[43] The short point is that it is very difficult to frame a tax plan, or an element that may be incorporated in a tax plan, in a way that can be said always to be acceptable. Minor changes, or apparently minor changes, or using a tax plan in a different context, can easily turn the plan from one that is acceptable to one that is caught by the general anti-avoidance rule. For instance, even making gifts to charity can occasionally be found to be part of a transaction that is caught by the general anti-avoidance rule.[44]

While the draft of 2011 is a distinct improvement over the 1990 statement, the question arises as to just how helpful it will be. Essentially, it is a study of the judicial history and current status of the law on section BG1. Unlike an ordinary textbook, however, the draft tends to set out statements from judgments without conceptualising them and without analysing and then organising them into a coherent framework. These features are regrettably common to much scholarly writing about general anti-avoidance rules. But in order to help readers to understand how the law on avoidance operates it is necessary to analyse, conceptualise, and synthesise the law in the manner of, say, a good textbook on the law on contract. That is not a description that could be applied to the draft of 2011.

The 2011 draft may not cause as many problems as the statement of 1990. It is too imprecise for taxpayers to pretend to be relying on approval for specific schemes. The upshot, however, is that the merit of the draft from the perspective of the tax system is the same as its demerit from the perspective of the tax planner. That is, the draft does not really attempt an interpretation of section BG1 in the sense that tax planners had hoped. From the perspective of the tax system, that failure is probably a good thing. It will be difficult to use the draft, or its final form, if the form remains similar, to support arguments in respect of particular tax plans. From the perspective of the tax planner, of course, this feature of the draft is a demerit. The draft is evidence of an almost heroic optimism, but it is hard to see that the effort will achieve the aspirations of its authors or satisfy the expectations of its readers.

10.5 Conclusion

This chapter suggests possible regulatory improvements in respect of three aspects of tax policy connected with the general anti-avoidance rule: making it clear that the rule takes precedence over provisions of double tax agreements; declining to issue private binding rulings on avoidance questions; and refraining from issuing Commissioner's interpretation statements on the operation of the general anti-avoidance rule. This list evokes two observations. First, it is something of a miscellany: not, in appearance, much more than a grab-bag. Second, two of the three suggestions appear counterintuitive, in that they reduce advice available to taxpayers. Is there an explanation?

The explanation lies in the phenomenon of "ectopia", briefly summarised at [10.1.2]. An important consequence of ectopia is that tax law tends to be ad hoc and incoherent, in the sense of lacking internal consistency. Further, tax law suffers from a lack of general principles that are usable. These consequences come about because of the gap between tax law on one hand, and business profits, the most conceptually challenging subject of tax law, on the other. Legislatures try to narrow that gap, but reform is perforce ad hoc as policy makers detect first one shortcoming, then another. A striking example occurred in 2010, when the United States Congress enacted a general anti-avoidance rule, not as part of coherent tax reform, but in the form of provisions in the legislation that enacted the Obama health care reforms.[45] The three proposals in this chapter observe that pattern. Broadly speaking, the best one can do with tax reform is to make a list and work through it.

The proposals for restricting the Commissioner's rulings and interpretation statements in respect of avoidance are at first sight even more improbable than the ad hoc nature of the list of proposals itself; but the explanation is the same. By its nature an interpretation statement should work from principle to detail. Similarly, a ruling should apply the principle of relevant rules to detailed facts of a transaction or structure. Because there is a problem of being able to define avoidance only imprecisely, by way of example,[46] statements or rulings based on principle are too frequently under-inclusive or over-inclusive. Even if they are correct in their own terms, taxpayers are apt to draw too much from rulings and interpretation statements, believing that they can extrapolate and apply principles. Such is not the nature of tax law.