Taxation without representation is tyranny. —James Otis
Modern fiscal systems emerged in Western Europe and North America during the half century that followed the American and French Revolutions. Although modern income and turnover taxes did not yet exist, by the middle of the nineteenth century the basic legal framework for raising these taxes had been established and with it the foundation for the spectacular increase in tax revenue that would occur almost a century later, during and after World War I.
In general, the basic legal framework calls for taxation according to the rule of law. The fundamentals of this framework are that (1) a tax can be levied only if a statute lawfully enacted so provides, (2) a tax must be applied impartially, and (3) revenue raised by a tax can be used only for lawful public purposes, not for the prince’s private ends. The rule of law contemplates that these principles will be enforced by independent courts.
The role of the courts is often referred to in this chapter. In some developing and transition countries, however, the judicial system does not, for various reasons, effectively fulfill its role. This is a substantial impediment to the rule of law in tax matters. A discussion of the ramifications, although important, is beyond the scope of this chapter.
In addition to these very general principles, the power to make tax laws is subject to several types of legal limitations. Their sources include (1) constitutional or other basic legal principles underlying an organized society, (2) international agreements, (3) interpretation of the tax laws by the courts, (4) the general framework of civil law and public law, and (5) the political structure of the country as a centralized or a federal state.
Tax laws must be drafted in the context of this legal framework, as it applies in the particular country in question. This chapter reviews the principles underlying this framework in general terms and on a comparative basis. Of course, where a particular country is concerned, further study will be needed to determine specifically how these principles are applied in that country.
I. Legal Foundation; Power to Make Tax Laws
The first principle is that any tax must have a firm basis in law. Much of the history of Western political movements has been based on opposition to arbitrary taxation. Parliamentary government in Britain evolved largely to constrain the monarch’s ability to raise revenue. During the seventeenth century, the House of Commons, the elected lower house, was recognized as having the exclusive right to initiate revenue laws.1 The American Revolution began as a protest against Britain applying taxes to the American colonies without the consent of their elected legislatures. As democratic government spread, legislative branches became the seat of power of the purse.
In light of this history, in most countries there is a basic constitutional principle that any act of taxation must have a legal basis. This principle means that no tax can be levied except under authority of a law.2 In many countries, this principle is written into the constitution.3 In others, the principle is not directly stated in the constitution, but is derived from another constitutional rule, as in Switzerland, where the principle of the legality of taxation is derived from the principle of equality of taxation.4 In Germany, the legal basis for taxation rests on the combination of two other constitutional provisions: the provision guaranteeing personal freedom, which cannot be restricted except by law,5 and the provision requiring a legal basis for any act of administration, including any administrative act of tax assessment and collection.6
Constitutions differ in the extent to which they allow the legislature to delegate tax law making authority. At one extreme, the principle of legality can mean that no delegation is permissible; at the other extreme, it can require only that taxes have a legal basis under the constitution, and if the constitution permits delegation of legislative power generally, then delegation is also permitted in matters of taxation. An intermediate position places limits on delegation, holding that for a tax to have a firm basis in law, its essential elements must be provided in an enabling law. Such elements would include, among others, definitions of taxpayer, taxable event or object of taxation, and tax base; tax rates; and basic rules for administration. This does not mean that all the details must be included in the law. As discussed below,7 implementing regulations can be issued by the executive branch of government in accordance with the framework of administrative law. In some cases, the law may take the form of a decree by the executive branch, if permitted under the constitution.
Because a state must have revenue to survive, the constitution usually allocates, either explicitly or implicitly, some tax-levying authority to the central government, but the power to enact particular types of tax laws may be limited.8 Such limitations can create serious problems for tax policy. For example, in the United States, the legislative powers of the Federal Government are limited to those specified in the Constitution. The Constitution provides specifically that the Congress has the “power to lay and collect taxes, duties, imposts and excises” through an act of Congress.9 The procedure for enactment sets forth a special requirement for tax legislation: such legislation must originate in the House of Representatives.10 Otherwise, the same procedure must be followed for tax laws as with any other laws. There is, however, a specific limitation on direct taxes, requiring these to be apportioned on the basis of population. This provision was held not to authorize enactment of an individual income tax.11 When this was corrected by constitutional amendment, the Supreme Court read the amendment relatively narrowly, taking to itself the decision as to whether a statute taxed “income” within the meaning of the amendment.12
Although there is no written constitution in the United Kingdom, British tax law also respects the principle of legality on the basis of the prescription of “no taxation without representation” that was introduced in the Magna Carta in 1215. This principle was reiterated in 1628 in the Petition of Rights, which states that “no man be compelled to make or yield one gift, loan, benevolence, tax or such like charge, without common consent by act of Parliament.” This principle is one of the cornerstones of Western democracies, in that the consent to be given by the representatives of the taxpayers in parliament is considered to be a democratic guarantee against arbitrary taxation by the government.
From the principle of legality, some countries have derived the principle of annuality,13 according to which a tax law can only have effect for one budgetary year. This does not mean that all tax laws have to be voted by parliament every year, but that parliament must annually consent to the government’s levying taxes in accordance with existing statutes for the next budgetary year. In most countries, this principle is accepted as a principle of budgetary law, rather than of tax law, and its specific operation will depend on the constitutional provisions and other laws governing the process for adopting the annual budget.
The general principle of the legality of taxation has in some countries given rise to another principle that the tax administration may not conclude an agreement on tax liability with the taxpayer.14 This is because when the statute says that tax is due, it must be strictly applied, and it is not within the power of the tax administration to agree to reduce the amount of tax. In some countries, the prohibition of such agreements is based on the idea of the tax law as being of public order.15 This means that the tax law has a special status as a statute that is essential to an organized society, similar to that of criminal law, on which agreement between the police authorities and the criminal is not possible either.16 This principle also plays an important role in the interpretation of tax laws by the courts.
II. General Principles and Limitations on Power to Make Tax Laws
A. Principle of Equality
The principle of equal treatment under the law applies not only to taxation, but to all laws. It can be viewed as an application of the concept of legality, under which the law must be applied without exception to all those in the same circumstances.17 It has two meanings, one essentially procedural and one substantive. The procedural meaning is that the law must be applied completely and impartially, regardless of the status of the person involved. This means that no one may receive either preferential or discriminatory treatment in the application of the law or may be denied procedural rights to challenge application of the law to him or her.
The substantive meaning of the principle of equal treatment starts from the position that persons in equal circumstances should be treated equally. Without clarification, this principle does not mean very much, because it admits that people who are not in the same circumstances can be treated differently. Therefore, the question becomes whether laws are prohibited from using certain criteria to discriminate among persons. While the list of prohibited criteria differs among various jurisdictions, they usually include ethnicity, religion, and gender. The exact application of this prohibition against discrimination in a particular country will depend on (1) whether the courts are competent to strike down legislation as unconstitutional and (2) what kind of discrimination is prohibited under the constitution.18 The principle also requires that both the purpose of the unequal treatment and the means to effect it have a rational basis. For example, treating higher-income taxpayers differently by applying graduated rates satisfies both tests; it is rational both to conclude that a taxpayer’s ability to pay increases with his or her income and to enact graduated rates as an implementing technique. While some approaches to tax legislation are clearly rational, many distinctions that tax laws draw are difficult to evaluate. Whether they are seen to violate the principle of equality depends on the level of scrutiny to which the rule is subjected.
The principle of equality has been applied in different ways by the courts of different countries to limit the power of the legislator. In France, the principle of equality before the law has been held to prohibit the denial of procedural rights to some citizens but not to others.19 The Constitutional Court has also struck down distinctions drawn by the legislator on the basis that they did not rationally carry out a purpose of the statute in the public interest.20 In Germany, the Constitutional Court has interpreted the constitutional guarantee of equality as calling for equal taxation of similarly situated persons. It has found, for example, the de facto unequal taxation of interest income (due to the absence of withholding) to be constitutionally impermissible, thereby requiring the legislature to enact measures to lead to more comprehensive taxation.21 In Slovenia, the Constitutional Court has found a provision of the income tax law in violation of article 14 of the Constitution, which provides, “[a]ll are equal before the law.”22 The provision in question included reimbursed expenses of independent contractors in the tax base, thereby treating this class of persons unequally compared with employees. In Belgium, the principle of equality was held to prohibit taxing companies providing professional services (lawyers, accountants, tax consultants, physicians) at the maximum rate of the progressive rate scale of the corporate income tax, thereby excluding these companies from the lower brackets, while all other companies could benefit from these lower rates. It was held that the circumstance that a company was engaging in professional services was irrelevant as a criterion to determine the tax rate applicable under the corporate income tax.23 It should be noted, however, that although the Belgian principle of equality is similar to the equal protection clause in section 1 of the Fourteenth Amendment to the U.S. Constitution, the U.S. Internal Revenue Code (IRC) contains a specific provision denying the application of the lower corporate income tax brackets to personal services companies.24 In the United States, this distinction is not considered a violation of the equality principle. The U.S. courts have generally been reluctant to strike down tax laws on the basis that they fail to provide equal treatment to equals.25
B. Principle of Fair Play or Public Trust in Tax Administration
The principle of fair play or public trust means that the taxation authority must not be allowed an unfair advantage in its dealings with taxpayers. Application of this principle suggests that (1) the authority must notify a taxpayer of any action the authority may take relating to that taxpayer, (2) during litigation, a taxpayer must be afforded all the rights of process allowed the authority, and (3) the authority must be bound by its interpretation of the law as applied to a taxpayer’s particular situation. In most countries, these rules of fair play are part of the general administrative law. However, exceptions to these rules can be made when fair play does not suffer as a result. For example, an authority may take action without notice if it reasonably suspects that the taxpayer would destroy evidence or flee the jurisdiction.
This principle is somewhat contrary to the principle of public order, according to which the tax statute must be strictly applied under all circumstances.26 Thus, the principle of fair play would hold that a taxpayer can rely on the statements of the tax administration if the taxpayer has given to the tax administration a full and fair representation of all the facts. The taxpayer can invoke the interpretation of the law by the tax administration even when such interpretation is erroneous. On the other hand, the principle of public order would suggest that if the tax administration erroneously applies the tax law, it is entitled to correct this application, even if this were disadvantageous to a taxpayer acting in good faith. Since both principles are usually applied simultaneously, there are sometimes contradictory decisions in the courts. One way that courts strike a balance is by holding that a taxpayer is not entitled to the tax treatment following from the administration’s erroneous interpretation, but that the taxpayer is not liable for penalties if he or she followed the administration’s interpretation in good faith.27
The principle has in some cases been codified. In the United States, penalties are abated where the taxpayer relied on erroneous written advice furnished by an employee of the Internal Revenue Service.28 In France, taxpayers can rely on a favorable administrative interpretation of tax statutes and regulations in contesting an assessment of deficiency in tax, even if the interpretation is contrary to law.29
The principle of public trust in the tax administration has also been used as a basis for preliminary rulings that can be issued by the tax administration on the application of the tax laws.30
C. Principles of Proportionality and Ability to Pay
The principle that tax liability should be based on the taxpayer’s ability to pay is accepted in most countries as one of the bases of a socially just tax system. The principle of ability to pay is, for example, opposed to head or poll taxes, against which the British revolted in 1990.31 Although it is used as a general principle for legislators in the design of the tax system, it is not included in the constitution of most countries and therefore cannot be enforced before the courts to limit the taxing power of the government.
The ability-to-pay principle is, however, constitutionally binding in some countries. For example, under the Italian Constitution, “everyone shall contribute to public expenditure in proportion to his resources.”32 The Italian Constitutional Court has held that ability to pay represents a specific application of the general principle of equality.33 The Court held, for example, that an income tax whereby the income of married people is taxed jointly violates the principle of equality and the ability to pay.34 The Spanish Constitution contains almost the same wording as the Italian.35 The German Constitutional Court held that the principle can be derived from article 3(1) of the Constitution, which states that all persons shall be equal before the law. It has concluded, for example, that a provision in the income tax that placed a limit on the deduction for required maintenance payments was unconstitutional because it failed to provide an adequate deduction and, therefore, failed to base the tax on the taxpayer’s ability to pay.36
The principle of proportionality is increasingly used by Western European courts in general and by the European Court of Justice in particular. It means that there must be some proportional relationship between the goals to be attained and the means used by the legislator.37 In the tax area, this means that taxes cannot be excessive. Even when this principle is applied to taxation, it has not prevented governments from imposing progressive taxes. In some cases, progressivity of tax rates is enshrined in the constitution.38 The principle of proportionality is generally interpreted as imposing only a marginal limitation on the taxing power of governments in the sense that they cannot impose confiscatory taxes.
In Switzerland, protection against confiscatory or excessive taxes is provided by a combination of article 22 ter of the Constitution, which guarantees private property to the citizen, and article 31, which establishes the freedom of commerce and industry. As in Switzerland, the principle of proportionality has not been enshrined in the German Constitution. It is implicitly recognized, however, by the combination of (1) the protection of personal freedom, which cannot be restricted except by law, so that each citizen is entitled to a decent subsistence minimum,39 (2) the freedom to work or to exercise a profession,40 and (3) the protection of property and inheritance.41
D. Principle of Nonretroactivity
The principle that tax statutes may not be applied retroactively can be justified on the basis that taxpayers should be able to make economic decisions with knowledge of their tax consequences and that it is unfair to provide tax consequences for an investment or other economic decision that differ from the tax treatment at the time the decision was made. Applied strictly, however, this principle would preclude any change in law, because any change, even if effective only in the future, affects the value of existing wealth. The balance is often struck by defining impermissible retroactive provisions as including only those with nominal retroactive effect, that is, those that affect a tax liability that has been fixed before the date on which the new law is passed. However, this is an arbitrary line, inasmuch as the economic effect of a tax change on existing investments does not closely correlate with the nominal retroactivity of the change.42 The arbitrariness of any definition of nominal retroactivity suggests that even if legal protection is given against nominal retroactivity, the degree of protection can never fully correspond to economic reality. Because virtually every change in tax law has an effect on existing investments, the problem of retroactivity can be dealt with only as a policy matter and not by means of a formal legal rule.
In most countries, the principle of nonretroactivity is observed not as a legally binding principle (except for a few special cases, discussed below), but as a principle of tax policy that the legislature follows as it considers appropriate. For example, in the United States, some amendments of tax law (particularly those considered to be technical corrections) are made with retroactive effect;43 by contrast, in other cases special relief is given against the application of tax changes to transactions in progress, even where the amendments are nominally prospective.44
In some countries, the principle of nonretroactivity is stated in the civil code.45 In these countries, the tax law can provide for retroactive effect, when it specifically does so in exception to the civil code. However, if there are no specific provisions in the tax statute, the civil code’s general principle of nonretroactivity will apply as the ordinary rule.46
The European Commission on Human Rights has dismissed a challenge to a retroactive tax law of the United Kingdom, holding that it did not violate the right of property under the European Convention on Human Rights.47 In this case, section 31 of the Finance Act 1978 was applied retroactively to April 6, 1976, a date that preceded even the Government’s announcement that it would legislate in this area. The provision in question denied a deduction for certain losses from tax shelters. The Government determined that retroactive application of this provision was necessary in order to deter tax-shelter promoters from devising new schemes. If anti-tax-shelter legislation were applied prospectively only, tax shelter promoters would be undeterred, because any scheme based on existing law would be valid for the period until new legislation were passed.
In countries where retroactive tax legislation is generally permitted, there are often some limitations for extreme cases. For example, the French Constitutional Court has stated that legislation may not be retroactively applied if it is penal in nature and that retroactively applied legislation generally may not affect individual cases that have already been decided by a court.48 The U.S. Constitution also prohibits retroactive criminal legislation.49 In the tax area, the U.S. Supreme Court has held that as long as the retroactive application of a statute “is rationally related to a legitimate legislative purpose,” the retroactivity is permitted by the Constitution.50
In other countries, there are broader constitutional principles limiting the permissible scope of retroactive legislation. For example, in Germany there is no general constitutional or statutory rule on nonretroactive effect of tax laws. However, the German Constitutional Court has based the principle of nonretroactivity on the concept of the “Rule of Law,”51 which includes the concepts of legal security52 and public trust.53 The German Constitutional Court distinguishes between retroactive tax laws54 and retrospective tax laws.55 A tax law is considered to have retroactive effect when it affects transactions that have been closed in the past, that is, before the law was approved and/or promulgated by the legislator. The law has a merely retrospective effect when it affects the future transactions or legal positions that have not yet been closed. The Court requires a higher standard for retroactive laws, which with a few exceptions are prohibited in principle, while merely retrospective laws are permitted. The Constitutional Court held unconstitutional an amendment to the corporate income tax law passed in 1952 that was applied to the 1951 taxable year.56 The prohibition against retroactivity under German jurisprudence is not absolute; retroactive legislation will be sustained where the taxpayer’s reliance on existing law was not reasonable, where the resulting damage for the taxpayer is almost nonexistent, where existing law was unclear or technically deficient, or in certain cases of overriding public necessity.57
Even where there is no legal prohibition on retroactive legislation, in most cases, the legislature decides to pass tax legislation on a largely prospective basis. In fact, in many cases, the political process provides taxpayers with generous protection from the effects of tax legislation for transactions in progress or investments that have been made. In some cases, however, legislatures act retroactively in order to protect tax revenue.
The following are examples: (1) The government announces that the excise tax on alcohol will be increased. The higher rate is often applied to stocks on hand (including floor stocks at the wholesale or retail level) on the date of announcement, as well as to production after that date. Otherwise, consumers would buy alcohol in large quantities to avoid the higher tax. (2) A mistake is discovered in a tax law that, if left uncorrected, could lead to a substantial revenue loss. The mistake is typically corrected with retroactive effect. Otherwise, taxpayers could take advantage of the time before the legislature passes the necessary legislation to reduce their tax liability, thus losing considerable revenue for the budget. (3) The government proposes in October 1995 changes in the individual income tax for 1996. However, the legislature does not pass the bill until May 1996. Nevertheless, the new rules can be applied for the 1996 taxable year. This is a case where the law may be considered nominally not retroactive, but merely retrospective because the law is passed before liability for 1996 is determined (i.e., December 31, 1996).58
Countries that allow retroactive tax legislation often apply a new tax law as of the date the bill was introduced in parliament. By setting an early date for the application of the tax law well before the final approval of the law by the parliament, the government prevents taxpayers from escaping the new tax provisions by rearranging their affairs during the period between the announcement of the new tax measures and the final vote in parliament. If the government announces the early date of application, the taxpayers will be warned about the new measures, so that they can take the tax consequences into account. Under such conditions, it can be accepted that the public trust of the taxpayer has not been violated.
In addition to the question of the retroactive effect of tax legislation, it is also important to consider legal restrictions on the retroactive application of delegated legislation.59 Regulations and other normative acts interpreting tax legislation are typically applied with an effective date the same as that of the law being interpreted.60 Otherwise, there would be the strange situation that the same law would be interpreted with one meaning up to a certain date and with a different meaning after that date. However, where a regulation provides a new rule of which taxpayers could not have been aware, it is often applied with prospective effect. This decision is typically left up to the body authorized to issue the normative act. For example, under section 7805 of the U.S. Internal Revenue Code, the Secretary of the Treasury decides the extent to which regulations will have retroactive effect.
E. Other Constitutional Limitations
Depending on the provisions of a country’s constitution, various other limitations on the power to make tax laws may apply. Besides requirements for equal treatment of taxpayers already mentioned above, there may be prohibitions against the taking of private property, requirements of regional equality, prohibitions against taxing certain items or discouraging certain activities, or prohibitions against taxing an item twice. As a general principle, the constitutional provisions that limit legislative power will apply to tax legislation as to any other legislation.61
For example, in Germany, the income tax provision subjecting the aggregate income of husband and wife to a progressive rate schedule in such a manner that a married couple could pay a higher tax than if they were taxed separately was held to violate article 6/1 of the Constitution, relating to protection of marriage and family.62 Moreover, articles 1/1 and 14 of the Constitution are interpreted as allowing each citizen a decent subsistence income, so that the Government may not tax income below this minimum; as a consequence, the German Constitutional Court held that dependency exemptions under the income tax for 1983–85 were constitutionally insufficient.63
The constitutions of many countries contain provisions with respect to the freedom of speech and religion. In countries where the courts have the power to enforce constitutional provisions, these provisions are held to mean that the government may not hinder the exercise of these rights through taxation, for example, by imposing heavy taxes on churches.
In Germany and Switzerland, special taxes are levied for the financing of church activities. In Germany, the combination of article 140 of the Constitution and article 137(6) of the Weimar Constitution of 1919 allows the church to impose taxes on the members of their congregations, within the limits imposed by state law. However, articles 2/I and 4/I of the Constitution prohibit the states from granting authority to churches over nonmembers of their congregations, so that nonmembers cannot be subjected to church taxes. Since only physical persons can be members of a congregation, imposition of church tax on legal entities is prohibited in Germany. There has been a trend in recent years for people to deregister as members of a church, in order to avoid paying the church tax.
In Switzerland, cantons are entitled to impose taxes to cover the expenses of the churches; unlike in Germany, it is not the church that imposes the tax. However, article 49/6 of the Constitution provides that no person can be obliged to pay taxes for a church to which he or she does not belong. This provision is based on the freedom of thought and religion. Consequently, persons not belonging to a church are entitled to refuse to pay the tax. However, unlike in Germany, legal entities are not protected by this clause and can be subjected to taxes levied for the benefit of a church.
In many other countries (such as the United States), a church tax would be unconstitutional, because it would violate the constitutional rule of separation of church and state.
There are great differences from one country to another in the extent to which courts use constitutional grounds to strike down tax legislation. As the examples cited above suggest, the German Constitutional Court has been particularly active in testing tax legislation against constitutional principles. Inevitably, this has involved the Court in difficult-to-resolve problems and has made it an almost permanent player on the tax policy agenda. Germany furnishes an ironic contrast to the United States, where the Supreme Court has been rather reluctant to become involved in tax policy issues at the federal level, despite its activism in many other areas of the law. The Court has, however, been quite active in the area of restrictions on state tax legislation that flow from the Constitution, given their importance for the federal state. Most other countries where courts have the power to strike down unconstitutional legislation have generally shied away from invoking open-ended principles, such as equality, in the tax area, but have sometimes relied on relatively more formal criteria—particularly those involving competence to legislate—to strike down tax laws.64
F. Charters of Taxpayer Rights
Some countries have provided charters or declarations of taxpayer rights. These have taken various forms. Sometimes, they have been issued by the tax authorities. Such documents are generally declarative of existing law, without independent legal force. In other cases, there is an article of the administration law entitled “Rights of the Taxpayer,”65 or there may be a bill entitled “Taxpayer Bill of Rights,” which enacts amendments to the rules of tax procedure.66 In this event, the rules have the same legal force as other provisions of the administration law. The main effect of these charters is to prohibit arbitrary practices by the tax administration against taxpayers.
In 1984, the Charter of Rights and Freedoms was established in Canada, and the rights of the taxpayer are summarized in the Declaration of Taxpayer Rights.67 The tax authorities must act in accordance with the provisions of the tax law. If the action is not authorized under the tax law, it is invalid. If the action is authorized by the law, a taxpayer can challenge its constitutionality. As a result, taxpayers have sought protection under the law of privacy68 and the right against illegal search and seizure.69
In Belgium, a taxpayer’s charter70 was voted in 1986, after the power of criminal investigation in tax fraud cases was transferred from the tax administration to the public prosecutor. The main effect of the taxpayer’s charter was to prohibit tax officials from cooperating with the public prosecutor’s office in criminal investigations, thereby also discovering unreported taxable income. In addition, the reporting of instances of tax fraud by the tax administration to the prosecutor’s office became subject to a clearance by a high-ranking official of the central tax administration.
In France, the tax administration established a taxpayer’s charter (charte du contribuable) by way of administrative practice.71 In this document, the taxpayer’s rights in case of an audit were stated. In 1987, the tax laws were amended to require the tax administration to provide the taxpayer with a copy of the charter before conducting an audit and conferring legal force on the provisions of the charter.72 If the tax administration fails to communicate the taxpayer’s rights contained in the taxpayer’s charter, the audit is invalid.73
G. International Agreements
The authority of the state to legislate in tax matters may be limited by international treaties and agreements. These include (1) bilateral tax conventions, (2) multilateral treaties establishing free trade areas, (3) agreements related to the General Agreement on Tariffs and Trade (GATT) and the World Trade Organization (WTO), and (4) the Articles of Agreement of the IMF.74 Depending on their scope, bilateral tax conventions may include specific limitations on the state’s power to levy income taxes, payroll taxes, and estate and gift taxes on nonresidents. Treaties establishing free trade areas like the European Union or the North American Free Trade Agreement (NAFTA) restrict the ability to levy tariffs, frequently provide rules for indirect taxation, and may also provide income taxation rules. While typically not as important, other bilateral and multilateral treaties may also be relevant to some aspects of taxation. For example, treaties of friendship, commerce, and navigation usually have antidiscrimination clauses, which may restrict the state’s income tax treatment of nonresidents.
A special application of the nondiscrimination principle has been made in several cases before the European Court of Justice. The Treaty of European Union prohibits discrimination on the basis of nationality in the areas of free movement of workers and the freedom to provide services,75 the freedom of business establishment,76 and the free movement of capital.77 The European Court of Justice has held that even when the tax law makes distinctions that are generally considered to be relevant to such law, such as the distinction between resident and nonresident taxpayers, these distinctions violate the nondiscrimination principle if their application restricts basic freedoms.78
The same principle of nondiscrimination has been held to apply to international movements of goods, so that goods originating in a foreign country may not be subject to higher taxation than that applied to domestic goods.79 Here, the European Court of Justice has held that even though the criteria used for distinctions in the tax law were not discriminatory in themselves, because they did not specifically refer to the foreign origin of goods, any criterion resulting in de facto restrictions on the entry of foreign goods violates the nondiscrimination principle.80
Such a position on nondiscrimination clearly restricts a country’s power to make tax laws and should be kept in mind by those countries planning to enter any kind of customs union or common market organization.
The European Convention on Human Rights is an example of another international agreement that limits legislative power, including taxing power. Article 1 of the first protocol to the Convention protects the right to property, but explicitly allows states a considerable measure of discretion with respect to taxation. As a consequence, the European Commission on Human Rights has been reluctant to strike down tax legislation as violative of the Convention; this has occurred only in a case where a tax infringed on the right to religious freedom.81
The Convention also provides for procedural rules with respect to the burden of proof and the right of defense in court cases. These provisions have thus far received only limited application in tax cases, chiefly where the case was in the nature of a criminal proceeding. But the European Court on Human Rights has recently ruled that they were applicable to administrative tax penalties, which were to be, from that point of view, assimilated to criminal penalties.82 Several Western European countries are debating whether to extend all the legal guarantees for the defense in a criminal case to cases of administrative litigation.
III. Interpretation of Tax Laws
A. General Considerations
Like other laws, tax laws are general legal prescriptions. However, a legal rule cannot typically foresee all conditions of its implementation, so that ongoing interpretation (and frequently revision) of tax law is essential to its application. Occasionally, constitutions may provide for interpretation by the legislature itself.83 The legislature may achieve a similar effect by amending an existing law, with or without retroactive effect. Such action by the legislature is common when the legislature wants to reverse the effect of the interpretation of a statute by a court.
Because in most countries implementation of tax laws belongs to the executive branch, the interpretation of tax law falls first to the executive branch, which issues regulations, decrees, circulars, and general rulings (“executive rules”). It also will apply law and interpretation to individual cases through individual rulings and decisions. However, executive rules must be in accord with constitutional and statutory law. Review of these rules is undertaken by independent courts. In addition to reviewing executive rules, courts interpret the tax law and apply it in specific disputes between the taxpayer and the tax administration. This means that the final interpretation of tax laws belongs to the judiciary.
The style in which courts interpret tax law will depend to a large extent on the way in which they interpret statutes in general. Statutory interpretation is a complex topic a full discussion of which is beyond the scope of this book. The style of statutory interpretation differs substantially from jurisdiction to jurisdiction.84 For example, courts differ on whether they even admit that an issue of interpretation exists or that there is more than one possible way to read the statute.85 They also differ on methods for ascertaining the intent of the legislature in enacting the statute, such as in their use of travaux préparatoires (legislative history). A general distinction can be made between common law countries and civil law countries. Courts in common law countries tend to pay close attention to the facts and exercise more freedom in their legal reasoning. Courts in civil law countries tend to take greater interest in the exact wording of the applicable rule and are generally more strict in their legal reasoning. While the style of interpreting tax statutes is influenced by the general approach to statutory interpretation, tax law presents some special considerations.
Everywhere in the world, even in common law countries, tax law has largely become a phenomenon of statutes and regulations. Oddly enough, the most detailed and elaborate statutory provisions are to be found in common law countries, such as Australia, Canada, and the United States. As a consequence, the application of the statutory rule is the basis for interpretation in common law as well as in civil law countries.
In all Western legal systems, the courts apply a specific method of legal reasoning, based on a systematization of facts and legal rules, in order to arrive at the concrete application of the tax law in the individual case. This type of legal reasoning is not peculiar to tax law, but common to all forms of statutory interpretation. Its objective is to answer the specific question whether a tax is due from a specific taxpayer, by applying one or more rules to the facts that are thought to be relevant. The facts are often not raw physical facts but legally constructed facts, such as a company, a sales contract, or an inheritance. Legal reasoning selects and orders these facts, so that they become susceptible to the application of tax rules. The legal rules to be applied are also to be selected from a variety of norms. Again, legal reasoning selects and orders these norms, so as to arrive at a concrete application of the tax law. The objective of this process is to arrive at a clear result (i.e., a tax is due or not due). The objective is not to achieve reconciliation of the taxpayer with the position of the tax administration.
Two competing principles are of overriding importance in the interpretation of tax law. The principle of legality (under which no tax can be imposed except on the basis of law) can be interpreted as providing that a court should not extend the words of a taxing statute to impose a tax in circumstances where the language of the law does not clearly impose it.86 This is the basic argument in favor of a literal interpretation of tax laws.87 However, if tax laws are interpreted rather literally, taxpayers can often arrange their affairs so as to avoid taxation. The countervailing principle therefore is that in enacting a tax law the legislature intends that it be effective, that is, that it not be circumventable through artificial maneuvers. Moreover, the principle of equality would call for interpreting the statute so as to tax equally taxpayers in the same economic circumstances. The tension between these two approaches to interpreting tax laws has been resolved in different ways by courts in different countries; the review of country practice below focuses on this issue. In addition, the variety of tax cases has raised many issues of statutory interpretation that arise with tax laws as with other statutes and that cannot easily be summarized in such a brief discussion.
The basic questions with respect to the interpretation of tax laws considered below are therefore (1) whether tax laws should be interpreted strictly or in a wider sense by the teleological or analogical method, (2) whether the legal form of a transaction should take precedence over the substance of the transaction, and (3) whether tax laws should be subject to a kind of “economic” interpretation, which would not be applicable in other areas of law. These are partially overlapping questions and are answered differently by the case law of various countries.88
As a general rule, in the French tradition, tax laws are interpreted strictly. This is a consequence of the legality principle laid down in article 34 of the Constitution. A clear text cannot be interpreted beyond the literal meaning intended by the legislator.89 Yet, the Cour de cassation and the Conseil d’Etat, the two highest courts to deal with tax cases, do not entirely share the same position on strict interpretation. The Conseil d’Etat, which deals with the majority of the more modern taxes (personal and corporate income tax and VAT), tends to have a more flexible attitude toward the interpretation of tax laws.90 However, even under the traditional rule of strict interpretation of tax laws, the French courts have always recognized the authority of the tax administration to submit evidence about the real nature of the transaction, so that it should be requalified for tax purposes.91 At about the same time, French courts developed the theory of abuse of law in civil law.92 In general terms, this means that a person does not have the right to exercise the person’s rights (e.g., property rights) in an abusive manner so as to injure others. This revolutionary theory would much later play an important role in tax cases in other countries.93
Belgium has a long tradition of strict and literal intepretation of tax laws. This is based on the principle of legality enshrined in the Constitution: no tax is due unless imposed by a law, and the burden of proof for establishing that a tax is due lies with the tax administration. The quintessence of the Belgian jurisprudence on taxation has been laid down in a decision of the Cour de cassation94 in which the court stated that a taxpayer is allowed to choose the “lesser taxed way,”95 and that for the application of the tax laws a legal construction engaged in by a taxpayer will stand, even if the form of the construction is unusual, provided the taxpayer subscribes to all legal consequences of the taxpayer’s construction. The holding of the court was based on the view that the legal system as a whole is consistent and that if the taxpayer took all the legal consequences of the taxpayer’s acts, the tax administration also had to recognize the tax consequences. The court held specifically that in tax law, there was no room for a principle of “economic reality”.96 Generally, it has also been held that there is no room for the application of abuse of law or fraus legis in the area of taxation. This jurisprudence stands for a high degree of legal security for the taxpayer. However, as tax planning became more aggressive, political pressure built up to introduce statutory antiavoidance rules and, in 1993, a general antiavoidance provision was enacted in the Income Tax Code.97
Yet the Belgian courts, like the French courts, applied the doctrine of “simulation” to some more traditional areas of taxation, such as gift and inheritance taxes. There is simulation when the legal act or instrument that is invoked by the parties against the tax administration does not correspond to the underlying legal relationship for which the parties have aimed. For example, a gift subject to substantial consideration to the benefit of the donor or a third party may be requalified as a sale.98 A transfer of immovable property to a newly established company in exchange for shares, immediately followed by the sale of the shares to a third party, has been requalified as a transfer of the real property itself to the third party.99
Germany is an example of a country where the legislator and the courts have over time interfered with each other regarding the interpretation of tax laws. Already in 1919, when the general tax law (Reichsabgabenordnung) was introduced, it provided that the tax laws had to be interpreted in accordance with the economic interpretation;100 the language was broadened in the Steueranpassungsgesetz of 1934.101 The objective of introducing economic interpretation of the tax law as a guiding principle of interpretation was to get rid of the excessively restrictive interpretation of the tax law on the basis of concepts and categories of civil law.102 Particularly between the two world wars, the Reichsfinanzhof was keen on furthering a wide interpretation of tax law. Economic interpretation became an instrument in extending the tax law to fill gaps and loopholes by analogical interpretation.103
The use of economic interpretation as a guiding principle in the interpretation of tax law has gradually been abandoned by the Federal Tax Court of Appeal, and the pre-eminence of the use of civil law concepts in tax law interpretation has been re-established.104 At the same time, the German Constitutional Court has been less clear in its decision on strict or extensive interpretation of tax law. Sometimes, it has spoken out in favor of strict interpretation and against the economic interpretation of tax law;105 at other times, however, the same court has decided in favor of “judicial development of the law.”106 When the new general tax law was adopted in 1977, the general “economic meaning” clause in the Steueranpassungsgesetz was not renewed.107 At the same time, a few specific and one general antiabuse clauses were introduced so as to give the courts more leeway in the interpretation of tax law, particularly in cases of abuse of legal constructions.108
Like France and Germany, the Netherlands at an early stage adopted a general antiavoidance provision.109 However, for quite a long time, this statutory provision on the interpretation of tax law did not influence court decisions because, at about the same time, the Supreme Court introduced the fraus legis doctrine into tax law.110 According to this doctrine, any legal construction resulting in a factual situation that is effectively subject to tax should be similarly taxed if so required by the purpose of the tax law. Originally, the legal construction was set aside under the fraus legis doctrine only when tax minimization was the exclusive reason for the legal construction.111 Gradually, however, the case law developed the doctrine that the legal form of the transaction would be set aside when the tax motive was the dominant or decisive reason for the transaction.112 Whether the tax motive is the dominant reason for the transaction is determined not by the subjective intent of the taxpayer, but by objective facts to be evaluated by the judge. It means that if the taxpayer has objective nontax reasons for the transaction, it will stand the test of fraus legis. In this way, the Dutch courts still maintain the right of the taxpayer to arrange his or her affairs in such a way as to minimize tax liability, provided that the validity of the legal form is well established.113 The fraus legis doctrine has been considered more than adequate to permit the courts to strike down artificial legal constructions, so that in 1987 the Minister of Finance decided to render the statutory antiavoidance provision inoperative, although it is still on the statute books.
F. United Kingdom
The U.K. tax system has no general statutory antiavoidance provision. Interpretation of tax statutes used to be controlled by the case IRC v. Duke of Westminster, where the court stated:
Every man is entitled if he can to order his affairs so as that the tax attaching under the appropriate Acts is less than it otherwise would be. If he succeeds in ordering them so as to secure this result, then, however unappreciative the Commissioners of Inland Revenue or his fellow taxpayers may be of his ingenuity, he cannot be compelled to pay an increased tax.114
This is generally considered to be the leading case for literal and strict interpretation, although the latter principle had already been formulated as follows in an earlier case:
[I]n a taxing Act one has to look merely at what is clearly said. There is no room for any intendment. There is no equity about a tax. There is no presumption as to a tax. Nothing is to be read in, nothing is to be implied. One can only look fairly at the language used.115
However, in 1981, W.T. Ramsay Ltd. v. Inland Revenue Commissioners was decided.116 In this case, the House of Lords struck down a tax-planning device on the basis that it was entitled to look at the overall result of several transactions and need not give tax effect to every single transaction.
[T]he fiscal consequences of a preordained series of transactions, intended to operate as such, are generally to be ascertained by considering the result of the series as a whole, and not by dissecting the scheme and considering each individual transaction separately.117
This doctrine was further developed in Furniss v. Dawson, in which the step-transaction doctrine and the commercial purpose doctrine were formulated as follows:
The formulation, therefore, involves two findings of fact: first whether there was a preordained series of transactions, ie [sic] a single composite transaction; second, whether that transaction contained steps which were inserted without any commercial or business purpose apart from a tax advantage.118
More recently, the House of Lords has limited the scope of the business purpose doctrine and the step-transaction doctrine in a series of cases.119 The court decided that where two courses of action are open to the taxpayer and are actively considered by him, the Government could not deprive him of the tax benefit of one of the alternatives.
It is one thing for the court to treat as a fiscal nullity a purely artificial step which will inexorably be followed by one or more others so as to achieve the desired end result. It is quite another for the court to treat as a fiscal nullity a step which had a commercial purpose in addition to tax avoidance and which in reality at the time it was taken might not have been followed by the other steps.120
This decision was confirmed a few years later, together with associated cases, and Lord Jauncey succinctly stated the position of the House of Lords on tax avoidance:
I conclude my analysis of the three cases by emphasizing that the Ramsay principle is a principle of construction, that it does not entitle the courts to legislate at large against specific acts of tax avoidance where Parliament has not done so and that at the end of the day the question will always be whether the event or combination of events relied on amount to a chargeable transaction or give rise to allowable relief within the meaning of the relevant statutory provisions.121
Now, the question is how long it will take before the Inland Revenue will decide that statutory antiavoidance measures are in order, as has been the case in Canada and Australia.122
In Australia, interpretation of the tax laws was for a long time dominated by literal and restrictive interpretation along the lines of IRC v. Duke of Westminster in the United Kingdom. While the British courts have been gradually taking a more flexible position on interpretation of tax law, the Australian courts persisted in their literal interpretation, thereby extending the doctrine of Duke of Westminster to all kinds of modern and complicated tax planning schemes, and implementing in fact a policy that favored the taxpayer. In Investment and Merchant Finance Corp. Ltd., this literal and strict interpretation was based implicitly on the principle of legality:
It is, of course, true that it is because company dividends are rebatable under s.46 that dividend-stripping is so attractive, and, if it be thought that this is a practice which should be checked, it is to that section that Parliament may choose to direct some of its attention. It is not for the courts, however, to depart from Parliament’s clear statement....123
In 1976, the Privy Council decided under New Zealand tax law the following:
[I]t is not the economic results sought to be obtained by making the expenditure that is determinative of whether the expenditure is deductible or not; it is the legal rights enforceable by the taxpayer that he acquires in return for making it.124
Chief Justice Barwick, who has been held responsible for the extent to which the High Court developed the strict interpretation of tax laws, stated his opinion as follows:
It is for the Parliament to specify, and to do so, in my opinion, as far as language will permit, with unambiguous clarity, the circumstances which will attract an obligation on the part of the citizen to pay tax. The function of the court is to interpret and apply the language in which Parliament has specified those circumstances. The court is to do so by determining the meaning of the words employed by Parliament according to the intention of Parliament which is discoverable from the language used by the Parliament. It is not for the court to mould or to attempt to mould the language of the statute so as to produce some result which it might be thought the Parliament may have intended to achieve, though not expressed in the actual language employed.125
Although the Australian income tax law contained a wide general antiavoidance and antiabuse provision,126 consecutive court cases by strict and literal interpretation of the tax law gradually whittled away the scope of that provision.127 In 1981, the court reversed its stand on literal interpretation and agreed to extend the scope of a statutory provision, although that wider scope was not within the literal meaning of the statute.128 By that time, however, there had been a political reaction and Parliament had inserted a range of general and specific antiavoidance provisions into the Income Tax Assessment Act, culminating in the adoption in 1981 of a new general antiavoidance rule.129
H. United States
Although the Internal Revenue Code contains a limited provision allowing the Commissioner to deny tax benefits from an acquisition, the principal purpose of which is tax avoidance,130 it does not contain a general provision on interpretation of tax law by the courts. Over time, the courts have developed a doctrine allowing them to set aside certain legal constructions that do not have a “business purpose.”131 When a legal construction has as its clear purpose the avoidance of income tax and does not at the same time involve some economic substance, it can be set aside by the courts as having no effect for tax purposes and replaced by another characterization of the underlying factual situation. Starting with the Gregory case, the courts have developed several judicial doctrines, such as constructive income or ownership,132 continuity of business enterprise,133 and the step-transaction doctrine. The step-transaction doctrine allows a court to decompose a transaction into several distinct steps, or to take several separate transactions together, in order to ascertain whether each of the individual steps, or the overall complex transaction, meets the requirements to benefit from certain effects under the tax law.134 The precise methods of applying these doctrines are complex and continually evolving.135
The issues in applying the substance-over-form approach in U.S. tax case law have been summarized well by Bittker & Eustice:
One of the persistent problems of income taxation, as in other branches of law, is the extent to which legal consequences should turn on the substance of a transaction rather than on the transaction’s form. It is easy to say that substance should control, but, in practice, form usually has some substantive consequences. If two transactions differ in form, they probably are not identical as to substance. Even so, they may be sufficiently similar to warrant identical tax treatment....
The foregoing judicial principles and statutory provisions, which often overlap in practice, are useful deterrents to tax-avoidance schemes of varying scope and ingenuity. Forcing transactions heavily freighted with tax motives to withstand judicial analysis in the context of these broad principles and provisions, vague and uncertain in application though they may be, is more salutary than uncompromising literalism in applying the statutory system for taxing corporations and shareholders.136
The often broad way in which U.S. tax courts interpret the tax law should be contrasted with the very close style of legal drafting used in the Internal Revenue Code, which prima facie obliges the courts to make decisions on very narrow rules. In spite of this, U.S. courts stick to their judicial doctrines, probably because of the common law tradition of legal analysis, where interpreting facts and rules with common sense plays an important role.
I. Antiabuse Legislation
Closely connected with the problems of interpretation of tax laws are statutory measures introduced to provide general rules for the application of tax legislation in situations where taxpayers structure transactions in a peculiar legal form so as to obtain a tax benefit unintended by the tax law. Tax laws being general prescriptions, it is inevitable that the legislator cannot foresee all situations in a rapidly changing world, thereby leaving gaps and loopholes in any tax law.137 Also, in many cases, the tax law allows the taxpayer a choice between different legal alternatives to reach factual objectives that are identical or very similar, but with different tax consequences. Depending on the legal choice made by the taxpayer, the same factual objective will result in a lower or higher tax burden. The two basically related questions raised here for the application and interpretation of the tax law are (1) what are the respective roles of the legislator and the courts in filling the gaps and loopholes, and (2) should the tax law attach different tax consequences to different legal situations that result in the same or a very similar factual situation?
The answer to these two questions may be clearer if the so-called antiabuse legislation is considered in the wider context of tax evasion and tax avoidance. In practically all developed tax systems, a distinction is made between tax evasion and tax avoidance. Tax evasion or tax fraud138 is an offense against the tax laws that is punishable by criminal sanctions. It consists of clear violations of the tax laws, such as fabricating false accounts or other documents, keeping parallel accounts, not reporting income, or smuggling or dissimulating goods or assets. The tax consequences of these acts can of course be corrected by the tax administration, but in addition these acts may give rise to criminal sanctions. The statutory measures taken to combat such violations of the tax law are generally not considered to be antiabuse measures.
Tax avoidance, on the other hand, is a behavior by the taxpayer that is aimed at reducing tax liability, but that does not constitute a criminal offense. The distinction between tax avoidance and tax evasion is critical, although sometimes confused, particularly by nonlawyers. Such confusion may be understandable in an economic or moral context, but it is basically wrong in a legal context of administration and implementation of tax law. In principle, most countries recognize the right of the taxpayer to arrange his or her affairs in such a way as to pay less tax.139 The problem is that the lesser tax burden may result from a legal construction or transaction that uses a gap or a loophole in the law to place the taxpayer outside the reach of the tax law or within the reach of a statutory provision providing for a lesser tax burden, or from a legal construction or transaction to which the tax law attaches a lesser tax liability than to another legal construction or transaction with similar factual results. It is clear that on the basis of considerations of economic efficiency (taxing similar economic situations the same way) and of fiscal justice (taxing similar factual situations the same way), there are good reasons to disregard the tax consequences of the legal construction or transaction and to close the gaps and loopholes, subjecting similar situations to the same tax burden. Therefore in some countries some constructions or transactions that constitute tax avoidance, although not a criminal offense, are not recognized for tax purposes either by the courts, or by general or specific antiabuse provisions.
In addition to tax evasion and tax avoidance, there is an activity that can be called tax minimization, which can be defined as behavior that is legally effective in reducing tax liability. It can consist of factual behavior by which taxes are avoided such as not consuming certain products (not smoking tobacco or not drinking alcoholic beverages) subject to tax or not earning certain types of income.140 This factual avoidance of the tax burden is considered perfectly legal and is not subject to statutory antiavoidance measures. According to Rivier, it consists in “using a lacuna intended by the legislator or the freedom allowed by the law to create a factual situation different from that contemplated by the law, whose consequences for the taxpayer are likewise different from those envisaged by the text of the law.”141 By contrast, tax avoidance typically consists not of factual, but of legal behavior, that is, molding factual situations in legal forms that bear less tax than other legal forms. The difficult question is whether a particular instance of such behavior is considered tax avoidance or tax minimization.
The question is whether the refusal to recognize the effectiveness for tax purposes of a legal construction is a task for the legislator or for the courts. The arguments against the courts doing this job are largely based on the principle of legality and the role of the courts vis-à-vis the legislator.142 The doctrine of the separation of powers holds that it is not for the judiciary to legislate. Therefore, when the clear wording of the tax law fails to tax certain situations, thereby leaving gaps and loopholes, even when reasonably and as a matter of tax policy these situations should be taxed, the courts will shy away from imposing a tax when there is no formal legal basis for doing so. Strangely enough, the same courts may fill the gaps and loopholes left by the legislator in other areas of the law. The reason is that for taxes, many countries have an explicit or implicit constitutional provision limiting the authority to tax in a similar way as the authority to impose criminal penalties: no taxation without legal basis. This supposes for an effective implementation of the tax law an all-knowing and infallible legislator who, in reality, does not exist.
With respect to extending the reach of the tax law to legal constructions and transactions having a factual effect similar to situations subject to a heavier tax, many jurisdictions will allow the tax administration to recharacterize a legal construction or transaction, provided it can show that the legal elements for such different characterization exist, but will refuse a recharacterization for tax purposes when only a similarity in fact exists. In more simple terms, this is stated as the problem of the opposition between substance and form. The attitude of the courts again presupposes that the tax consequences attached to each legal construction or transaction are the adequate tax reply to the factual situation covered by the construction or transaction; that is, it presupposes an infallible inner consistency of the law so that each legal form is always the adequate translation of the underlying substance. That unique quality of the legal rule is of course absent in many cases.
The ways in which the courts of various countries have dealt with these problems have been discussed above.143 In some countries, the legislator has judged it necessary to take legislative action in the form of general or specific antiabuse provisions to remedy the courts’ failure to interpret the law in such a way as to cut off abuse. The general antiabuse provisions, on the one hand, call on the courts to apply an extensive or economic interpretation of the tax law and to disregard legal constructions and transactions when they have an artificial flavor. Specific antiabuse provisions, on the other hand, which can be found in nearly all developed tax systems, are aimed at closing particular gaps and loopholes.
It should be noted that there is no clear relationship between the way courts interpret tax law (strictly vs. extensively) and the presence or absence of general antiabuse provisions. Several countries operate their tax system without general antiabuse provisions: Belgium (until 1993), Italy, Sweden (1992–95), Switzerland, United Kingdom, and United States. Except for the United States, in most of these countries, tax law is interpreted in a strict or literal way. The combination of case law and specific antiabuse provisions is apparently held to be adequate in administering the tax system. A second group of countries does have general antiabuse clauses in their tax legislation with rather different results. The most prominent examples are Australia, Austria, France, Germany, the Netherlands, and Spain.144
The original Australian antiavoidance rule provides that contracts are void for tax purposes if they were made in order to alter the incidence of the income tax, or to defeat, evade, or avoid any liability under the Income Tax Assessment Act.145 Although the wording of this section was very broad, in the general climate of literal and strict interpretation that was dominating the interpretation of tax law by the Australian courts,146 the scope of the section was systematically whittled down through the application of the “freedom of choice” doctrine to a narrow rule that became very difficult to apply.147
By 1980, it became clear that the existing Australian setup of general and specific antiavoidance clauses and literal or strict court interpretation was not working.148 In 1981, section 260 was amended to apply only to schemes entered into prior to May 27, 1981, and a whole new set of antiabuse rules applicable to arrangements entered into on or after that date was introduced as Part IVA (“Schemes to Reduce Income Tax”).149 Basically, Part IVA provides that when there is a “scheme” as defined in the statute, the Commissioner has discretionary power to deny a tax benefit or disallow a deduction, which would have been obtained through the scheme, when such scheme satisfies eight conditions set forth in the statute.150 The crucial question in applying the act is what constitutes a scheme. In section 177A(3) and 177D, a scheme is defined as any unilateral scheme, plan, proposal, action, course of action, or course of conduct entered into or carried out for the purpose of enabling the relevant taxpayer or other taxpayers to obtain a tax benefit in connection with that scheme. Contrary to general antiabuse provisions in Europe and even in Canada, the Australian provision follows a very complicated and technically difficult style of drafting.
The first case involving these provisions to reach the High Court of Australia was Federal Commissioner of Taxation v. Peabody.151 The decision illustrates the complexity of a general antiabuse provision because it had to identify the “tax benefit,” “the scheme,” and “the relevant or other taxpayer.” In this particular case, the taxpayer won on the basis that the Commissioner had allocated the revenue to the wrong taxpayer. The Commissioner also lost the second case brought under this provision on the basis that the dominant purpose of the scheme involved was to make an investment and not to obtain a tax benefit, even though the scheme resulted in earning income that was exempt from tax.152
At the same time that the new general antiabuse provisions were inserted in the Income Tax Assessment Act, Australia amended its Acts Interpretation Act to promote a purposive interpretation of legislation, particularly tax law. The new section reads as follows:
In the interpretation of a provision of an Act, a construction that would promote the purpose or object underlying the Act (whether that purpose or object is expressly stated in the Act or not) shall be preferred to a construction that would not promote that purpose or object.153
The combined effect of the changes to the Acts Interpretation Act, the application of the general antiabuse provision of the income tax law, and changes in the composition of the High Court led to a shift from literal to purposive interpretation of income tax legislation.154
French tax law contains two general instruments to combat tax avoidance: a provision on the “abuse of tax law”155 and the court doctrine of the “abnormal management act,”156 which does not have a direct statutory basis.
The main characteristics of the abuse of tax law provision are that a transaction is subject to sanction only when a specific procedure is followed and, according to the courts, when the transaction has been set up exclusively for tax avoidance purposes. This provision covers transactions where the real legal transaction is hidden by an apparent legal transaction (simulation),157 as well as, according to case law,158 transactions entered into exclusively to obtain a tax benefit (fraude à la loi). Because the burden of proof is on the tax administration and the condition of the exclusive tax avoidance motive is difficult to prove, this weapon is seldom used by the tax administration. The French tax administration is now pushing for an amendment to the statute, so as to apply the abuse of law provision in cases where the tax avoidance motive is the dominant reason and not necessarily the exclusive reason for the transaction.
The abnormal management act doctrine has no specific statutory basis, but has been entirely developed by the courts.159 It is based on the theory that a business taxpayer cannot engage in any activity that is contrary to the taxpayer’s business interest because the purpose of the business is to make a profit. This does not mean that the taxpayer has the obligation to maximize business income under all circumstances, but it allows the tax administration to intervene in situations in which the taxpayer reduces taxable income, by acts against the taxpayer’s business interests, in order to transfer income to another taxpayer who is exempt or who is taxed at a lower rate. Because the burden of proof is less onerous than under the abuse of tax law provision and because there is no specific procedure, the tax administration prefers this court doctrine to combat abuses of taxpayers.160 The application of the abnormal management act doctrine is not subject to any special procedure. In most cases, it presents problems of fact and not of law, so that it is to be distinguished from the abuse of tax law provision of the code of tax procedure. However, the same transaction can reduce a taxpayer’s income by an act against the taxpayer’s business interests, while at the same time having been entered into exclusively for tax avoidance purposes. In such a case, both antiabuse instruments would be applicable.
Germany introduced quite early161 a provision in its general tax laws obliging the courts to follow the economic interpretation of the tax law.162 Gradually, however, the Court of Tax Appeals shifted its interpretation to a more traditional stance, giving predominance to concepts of civil law over tax concepts, so that the taxpayer would be in a position to make a choice between different legal forms of a transaction to minimize the taxpayer’s tax burden.163 Also, in the German tax doctrine, the economic interpretation was not considered specific for tax law, but was a general kind of teleological interpretation.164 When the new General Tax Law was introduced in 1977, the mandatory economic interpretation method of tax laws was abandoned and replaced by several antiabuse provisions.165
The new provisions are contained in DEU AO sections 40 through 42, of which section 42 is the most important for the interpretation of tax law. AO section 40 establishes the rule that transactions will be taxed whether they are legal or not. The effect of this section is to tax profits from illegal activities, like gambling, drug trafficking, and so on, so as to avoid a situation in which illegal activities would benefit from a tax exemption.166 It is important to note that deductions for expenses are also allowed, even when incurring such expenses would constitute an illegal activity.167 AO section 41 subjects to taxation transactions that are legally invalid for nontax purposes under civil or commercial law when the economic substance of the transaction is maintained in spite of its legal nullity. It also disregards sham transactions.168 A sham transaction exists when the parties agree that the transaction should have no legal effect or when one legal transaction is used to hide another legal transaction. Both sections base taxation on the economic or, more generally, the factual substance of a transaction, without regard to its illegality, nullity, or legally fictitious character. In this sense, both sections can be considered a continuance of the economic application of tax law.
The most important general antiabuse clause is contained in AO section 42, providing that tax cannot be avoided by “abuse of legal constructions.”169 When abuse of a legal construction is established, the tax claim will be based on the legal form of the transaction that is appropriate to the legal factual situation. An abuse is considered to exist when the legal form of the transaction or construction used by the taxpayer is not appropriate to the factual economic situation. The key word in this provision is “appropriate.”170 It requires that the factual consequences of a transaction be more or less consistent with its legal form. The abuse consists of the choice of a legal form that is inappropriate for the economic relationship in order to avoid taxes.171 The legal form of a transaction will be considered inappropriate when reasonable persons—in order to achieve a specific economic relationship and, in particular, a specific economic goal—would not choose a particular legal form because they would consider it inadequate.172 The specific characteristic of the German law is that it requires some consistency between the legal form and the economic content of a transaction. In many other tax systems, it suffices to have a business purpose, even if the legal form in which this business purpose is achieved is not entirely appropriate. If a transaction has no business purpose at all, it may be assumed that the legal form is inappropriate and that there is abuse of a legal construction. Generally speaking, for a legal transaction to be effective for tax purposes, it will require (1) a business purpose, and (2) an adequate legal form to achieve the business objectives of the taxpayer. It is clear that when there are several adequate legal forms to achieve these business objectives, the section will not be applicable when the taxpayer chooses the legal form that minimizes the taxpayer’s tax burden.
In the Netherlands, a general antiabuse provision was introduced in the General Tax Law in 1925. Since 1959, it provides that a legal transaction that does not have as its purpose a significant change in the factual circumstances or that would not have occurred but for the fact that it eliminates or reduces the tax liability shall not be taken into account; that is, when the exclusive purpose of a transaction is to minimize the tax burden, it is subject to correction for tax purposes.173 In the Dutch tax literature, this provision is known as “correct taxation.”174 The tax inspector who wants to apply the procedure of correct taxation has to ask for specific advance approval from the Minister of Finance. Given the judicial development of the fraus legis doctrine, the statutory provision has been of limited importance.175
In Spain, the abuse of law doctrine is based on article 6.4 of the Civil Code, which was adopted in 1974.176 This concept of civil law was also used for tax purposes, because although the General Tax Law referred in article 24, paragraph 2 to “abuse of law,”177 there was no clear definition of abuse of law in the tax code.178 In 1979, this provision was implemented by a decree establishing a special procedure for the application of the concept of abuse of law.179 As in France, this procedure is to be followed when a taxpayer is notified that the taxpayer is accused of abuse of law. The burden of proof is with the tax administration. In addition, article 25 of the Spanish tax code provides that taxes should be levied in accordance with the real legal or economic nature of the taxable event.180 When the taxable event consists of a legal transaction, it will be characterized for tax purposes in accordance with its “true legal nature,” regardless of the form of the transaction. When the taxable event is determined by economic concepts, it will be characterized in accordance with “effective economic relationships.” Both provisions seem to indicate a strong bias in favor of economic interpretation of tax law and of substance over legal form.
However, article 24-1 of the General Tax Law contains an explicit prohibition of extensive interpretation of tax law and interpretation by analogy beyond the strict meaning of the words. The resulting legal framework of the antiabuse provisions in Spain is at least confusing, and there is great debate about the exact meaning of the provisions. As a result, these contradictory legal prescriptions have driven the High Court to very divergent applications of tax laws.181 Recently, article 24 on abuse of law has been amended.182 Under the amended language, reference to economic or social interpretation has been eliminated. Taxes will be due on the basis of the “legal nature” of the taxable event. The new Spanish law establishes the “legal reality” of transactions as the sole legal basis for taxation, as opposed to economic or social reality.
J. Specific Antiabuse Provisions
In addition to general antiabuse rules, the tax laws of most countries contain specific antiabuse provisions.183 The approach of the specific provisions is different from the general antiabuse provisions, because in many cases they do not focus on application or interpretation of tax law, but simply mechanically deny certain tax benefits under certain conditions. Their goal is to prevent avoidance or abuse of specific rules in the tax code. It is impossible to make an inventory of all the rules that vary from country to country; some examples are listed below.
Most countries have the following antiavoidance rules in the domestic area: (1) limitation of deductions for entertainment and traveling expenses; (2) rules on taxation of accrued as opposed to effectively paid interest; (3) rules on arm’s-length dealing between related taxpayers, or between taxable and taxexempt taxpayers; (4) rules against dividend stripping; (5) limitations on tax loss carryovers from one taxpayer to another; and (6) limitations on loss deductions by partners and shareholders in companies not subject to corporate income tax.
In the international context, the following rules are common: (1) rules on dealing at arm’s length in international transactions; (2) rules on thin capitalization; (3) rules against the transfer abroad of income-generating assets without payment of tax; (4) rules on controlled foreign corporations; (5) rules limiting the effects of physical emigration of taxpayers; (6) rules limiting tax benefits for income sourced in tax havens; and (7) rules limiting deductions of expenses and losses in corporate headquarters or branches of foreign companies.
This brief survey shows that the problems of tax avoidance and the issues of substance over form are truly universal, although there are variations in each tax system. Basically, there have been two broad alternative legislative and judicial approaches in the countries surveyed. Courts have interpreted tax laws either in a strict and literal way or in a more flexible way that takes into account the economic and social objectives of the tax laws. The way in which courts interpret tax laws will of course depend on the way courts interpret laws in general and on whether over time they have developed special doctrines for the interpretation of tax laws. Because of limits to what courts can or are willing to do to combat tax avoidance by interpreting the tax laws, many legislatures have resorted to the enactment of antiavoidance provisions.
The survey of the general antiabuse and antiavoidance provisions shows that they are a mixed blessing. The best and most consistent results seem to have been achieved in countries that do have a general antiabuse provision on the statute books, but one that is very sparsely used by the tax administration, because the courts have developed a reasonable—and not too strict or literal—approach to the interpretation of tax law.184 Very close is the situation in which there is no general antiabuse provision, but in which the courts have developed a general antiabuse doctrine, like the business purpose test.185 A second-best solution provides for a general antiabuse provision on the statute books, which is sometimes used by the tax administration under strict and narrow conditions imposed by law.186 In Spain, however, there was the problem of the contradiction between the statutory provision on narrow interpretation of tax law and the general antiabuse provision, which has recently been addressed by legislation introducing the concept of the legal nature of the transaction.187 The worst scenario, apparently, is the historic Australian experience in which frequent reliance by the tax administration on a general antiabuse provision is combined with strict and literal interpretation of tax law by the courts.
These experiences suggest that for countries that do not have a long court tradition, a general antiabuse provision should be combined with intense education of judges on how to develop legal reasoning and on how to make a reasonable application of the rule of law in general and the rule of tax law in particular. For countries that do have a long court tradition, the solution is simpler: when court interpretation is flexible, no general antiabuse provisions are needed; however, when court interpretation is strict, it may be preferable to work on the education of judges rather than to introduce a general antiabuse provision.
Finally, an increase in aggressive tax planning and resulting tax avoidance have been caused in part by the increasing complication of tax laws and by the growing burden of taxation. This is an imperative reason for drafting simple tax laws, leaving few options to the taxpayer and reducing to an absolute minimum the possibilities for tax arbitrage between the various options. In the end, the justice of a tax system is better served by simple rules that do not make too many distinctions, but that can be applied effectively, than by rules that try to take into account the very different relative positions of various taxpayers, but that can be avoided by taxpayers rich enough to pay for good tax advice.
IV. Distribution of Tax Law Making Power Between the Legislative and Executive Branches of Government
One of the most perplexing problems that tax officials in developing and transition countries face is in determining the proper role for executive rules to interpret and implement tax laws. It is clear that the legislature is responsible for passing the law, but what is the proper scope for administrative interpretation? Additional questions arise regarding the level of detail to be provided; the type of document to be issued; the name to be given to the document; the organization to issue the document (tax administration, minister of finance, cabinet); the effective date, time, and party to issue the document; and the legal effect to be given to the executive rule.
These questions can be difficult to answer because there are substantial differences in practices from country to country. Moreover, the basic rules governing the legality of permitted practice are often elastic. The legal effect to be assigned to a particular type of executive rule depends on the country’s general constitutional and administrative law, doctrines of legislative interpretation developed by courts or enacted in law, and specific provisions in tax laws that may prescribe the legal effect of particular types of administrative acts.
Not only is there considerable variation on these matters from country to country, but even within the legal tradition of a particular country, it may be difficult to determine the legal effect that courts give to administrative pronouncements. This is because standards for statutory interpretation and the scope of judicial review of administrative action are often quite elusive. Even when courts can agree on general principles, the application of those principles to particular cases can be controversial.
A. Distinction Between Executive and Legislative Functions of Government
Democracies generally subscribe to the doctrine of the separation of powers, according to which there are three independent branches of government: the legislative, the executive, and the judicial.188 Under the general distinction between the legislative and the executive functions of government, lawmaking, in the sense of establishing the general rules that control behavior in the society, is the privilege of the legislative branch (parliament), while the implementation and the administration of the laws pertain to the executive branch. In many countries, the power of the executive branch to implement the laws by government ordinance or decree is based on a general delegation of power in the constitution to implement any law approved by parliament.189 In other countries, the delegation of power must be specifically provided for in the law or is limited in the constitution itself.190 As an exception to this principle, some constitutions assign to the executive branch the power to make law by decree without the consent of parliament, usually strictly limiting this power in scope or in time or permitting it only when a state of emergency or specific authorization by the legislature exists.191
The distinction between lawmaking and administration is not always clear-cut, because administration necessarily involves an element of discretion in interpreting the law. In addition, the administrative branch may be authorized to issue norms with greater or lesser legally binding force in order to carry out the law.
Administrative acts with the greatest legal force are referred to as regulations. (They may also be referred to as orders, decrees, rules, or ordinances.) The relevant minister or the cabinet of ministers may be authorized directly under the constitution to issue regulations to carry out the laws, or tax laws may delegate authority to issue regulations. As long as a regulation is not contrary to the statute, it has the force of law, which means that it is binding on both the taxpayer and the state. Regulations are typically used to fill in gaps and details that are not dealt with in the statute, although they may also fashion rules out of whole cloth when so authorized.
The division of responsibility between laws and regulations varies greatly from country to country, because traditions of administrative law differ among countries. It is therefore important to design tax laws to fit within the country’s scheme of administrative law. In some countries, very short statutes and detailed regulations are routinely written;192 in other countries, the constitution may leave a very narrow scope for regulations, thereby requiring all necessary details to be put into the statute.193
B. Delegation of Power to Make Tax Laws in the Continental European Tradition
In the European continental tradition, the executive branch has the power to establish rules for the implementation or administration of tax laws by way of regulation, provided that the statute approved by parliament contains sufficiently specific rules defining the essential elements of the tax.194 This means that the act of parliament must contain the rules defining the taxpayer, taxable events, tax base, tax rates, and rules for the collection of tax.195 This power of the executive branch of government to execute or implement the tax laws is based on a general or specific delegation of power in the constitution. Tax regulations issued under such delegation of power are limited to the implementation of the law itself and are valid only within the limits of those laws. What can be determined by executive decree are matters of detail, procedure, and administration.196 A regulation that extended the scope of the tax law, changed its conditions, or altered the meaning of the law would have to be declared illegal and inapplicable by the courts.197 The tax administration will be bound by the regulations issued by the executive branch, as long as they have not been declared illegal by a court. In many cases, there will be specific delegation of powers in the tax law, but such specific delegation of power does not add anything to the delegated power of the executive branch of government if a general or specific delegation of such power already exists in the constitution.
In exceptional and very limited circumstances, the legislator may give a full delegation of power to the executive branch to establish tax laws or essential elements of tax laws by decree. Such delegation of power may be specifically provided for in the constitution198 or in the constitutional doctrine.199 In such cases, the law containing the delegation often requires post factum ratification of the decree by an act of parliament.200
C. Delegation of Tax Law Making Powers in Common Law Countries
The power of administrative agencies to make law is viewed somewhat differently in common law countries such as the United States, the United Kingdom, and Canada. Unlike in continental Europe, there is generally no constitutional delegation of tax law making power to the executive branch of government. Rather, such delegation is by statute. For example, in the United States, the Treasury Department issues tax regulations, in conformity with general provisions of administrative law (embodied in part in the Administrative Procedure Act),201 and under the explicit general delegation of authority in the IRC to issue regulations implementing the tax laws.202 In addition, specific provisions of the IRC grant authority to issue regulations. For example, IRC section 7872 grants the Treasury Secretary authority to “prescribe such regulations as may be necessary or appropriate to carry out the purposes of this section.” In allowing regulations to carry out the purposes of the section, this language is broader than the general language in section 7805. However, the specific delegation of regulations authority is confusing, because it is either superfluous or casts doubt on the general delegation of authority in section 7805.
Under the Administrative Procedure Act, regulations are divided into interpretive and legislative regulations, although the distinction between them is not always clear. IRC section 7805 provides sufficient authority for the Secretary of the Treasury to issue regulations interpreting the provisions of the IRC. These will be upheld as valid by a court if they are not inconsistent with the statute. Under broader grants of authority to issue legislative regulations, the regulations may set forth rules that go beyond interpreting the statute. An example is IRC section 385, which authorizes regulations distinguishing between stock and indebtedness, requiring only that the Secretary take certain factors into account. Therefore, as long as statutory authority for a regulation exists, U.S. administrative law does contemplate lawmaking by an administrative agency within the framework of a statute.
The U.S. tax regulations are the most voluminous in the world. Fortunately for those who must consult them, they are numbered according to the sections of the Internal Revenue Code to which they correspond. For example, Treasury Regulation Section 1.117-1 is the first regulations section corresponding to IRC section 117; section 1.117-2 is the second section, and so forth. Most sections are quite lengthy and are subdivided according to a system similar to that used to subdivide sections of the U.S. Code.
Similarly, in the United Kingdom, the executive branch may issue such delegated legislation as it is authorized to do by act of Parliament.203 The power to make laws is vested in Parliament. However, nothing prevents Parliament from delegating this power, in other words, authorizing governmental bodies to make law by administrative order and even to amend acts of Parliament if so authorized.204 Delegated legislation must be within the scope of the delegated power; otherwise, it can be struck down by the courts.205 There is no single name in the United Kingdom for delegated legislation (e.g., regulations, rules, orders), although they are published in a uniform series of statutory instruments.206 In the tax area, there are voluminous regulations, although their text is not as long as that of the laws themselves (about 1½ volumes of statutory instruments to 3½ volumes of laws). This is partly due to the extensive use of schedules to the laws, which often contain what would otherwise be in regulations. In contrast to the tax regulations of the United States, which are arranged according to the arrangement of sections of the statute, the various U.K. regulations stand alone, which obscures their relation to the statute.
D. Administrative Commentaries, Interpretations, and Statements of Practice
In addition to executive decrees and regulations, most tax administrations in continental European countries issue administrative commentaries, instructions (which may relate to specific tax forms or be published separately), guidance to their own staff, and circular letters.207 Such administrative commentaries or instructions are binding only on the administration for which they are intended. They are not binding on the taxpayers or the courts. There are several cases in which the courts have specifically rejected the interpretation of the tax law made by the tax administration in such administrative commentaries or instructions.208
The U.K. Inland Revenue issues “statements of revenue practice.” These are of great importance for the practical administration of the tax system, although they do not have the force of law. Statements of revenue practice generally are interpretations of the tax law by the tax administration. They also include “extra-statutory concessions.” These are written almost in legislative form, although they do not have the same formal status as a statute in the sense that the tax administration does not have a binding obligation to apply them. However, development of administrative case law suggests that the Inland Revenue would not be authorized to deny such a concession to a taxpayer, because such a denial would constitute a breach of duty to act fairly between different taxpayers.209 The ostensible purpose of these concessions is to deal with hardships that are minor or transitory, although in fact they can be more significant.
Revenue Canada issues interpretation bulletins stating its views on how to interpret and apply particular provisions of the tax laws. These administrative bulletins have no legal force, but Revenue Canada, in most cases, follows its own interpretation bulletins. Thus, if a taxpayer also follows them, the tax authorities cannot challenge the taxpayer’s position. If a taxpayer disagrees, the taxpayer can challenge the position of the tax authorities in court.
In the United States, the Internal Revenue Service issues a steady stream of revenue rulings, instructions, and other releases on how it believes the tax laws should be applied. These administrative pronouncements are not binding on the taxpayer, but, until they are withdrawn, they are binding on the lower tax officials.
E. Administrative Rulings
Administrative rulings are an important instrument in the implementation of tax law.210 Some countries, like Australia, Canada, the Netherlands, the United Kingdom, and the United States, have a long tradition of advance rulings. This means that the tax administration will issue a binding application of the tax law to the facts presented by the taxpayer on the condition that the taxpayer give a full and fair representation of all the relevant facts. Such rulings are effective in avoiding conflict and litigation by establishing in advance an authoritative interpretation of the tax law, so that the taxpayer has full security in the way the tax law will work out in a specific situation.
In allowing the tax administration to issue rulings, the following basic questions should be kept in mind: (1) Is the effect of the ruling limited to the taxpayer who requested the ruling, or can other taxpayers also rely on the ruling? (2) Is the ruling regularly published or not? (3) Are there public and private rulings? (4) Which administrative ranks of tax officials have authority to issue a ruling? (5) Are ruling decisions decentralized at the local level or are they centralized at a higher level or issued by a special unit of the tax administration? (6) What is the exact procedure for requesting a ruling and for deciding on and issuing a ruling? (7) What are the conditions under which the tax administration can change its position under a ruling?
In some countries, like the Netherlands, the power of the tax administration goes even further in that the tax inspector can grant a private ruling for a taxpayer by which he or she grants certain concessions. This type of ruling gives enormous flexibility to the application of tax law and permits the establishment of private tax concessions that are not published by way of general rule. Such power can be granted only to tax officials who show great restraint and discipline and are immune from corruption.
In other countries, however, the tax administration cannot issue binding advance rulings. The absence of the power to grant advance rulings has to do with a general view of the role of the tax administration and the role of the tax law and is closely linked to the principle of legality of the tax law and public order, according to which the tax law must be applied strictly and no agreements can be made on its application.211
Sweden has a unique system for advance rulings, whereby these are issued not by the tax administration but by an independent council.212 Decisions of the council can be appealed. “After more than 40 years of Swedish experience with advance rulings, it is quite clear that the cases of advanced rulings being appealed against to the Supreme Administrative Court have delivered an extremely important part of our case law, perhaps the majority of leading cases, and proven of particular value in making court testing, especially of new legislation, possible early enough as to be of real guidance from the outset.”213
France has a specific system of preliminary agreement, which should be distinguished from the ruling system.214 It makes certain tax benefits dependent upon the preliminary fulfillment of certain conditions that are reviewed by an agency other than the tax administration. These preliminary agreements can be found, for instance, with tax benefits granted within the framework of economic development of the regions or the economic restructuring of certain industries. The conditions of economic development or restructuring realized by the taxpayer will be evaluated to see whether the taxpayer meets the requirements for the tax benefit.
Finally, specified tax treatment of a transaction may be conditional upon preliminary approval by the tax administration.215 Such a preliminary approval is often used to guarantee that the taxpayer will not abuse the transaction for purposes of tax evasion or tax avoidance. A requirement of preliminary approval may be particularly appropriate for types of transactions that are rare and that, in the absence of an approval requirement, would need complex statutory provisions. In a certain sense, this is a mandatory preliminary ruling. In particular, tax administrations that are not strongly equipped may be tempted to exercise this type of control on taxpayers. However, precisely because such preliminary agreements are often used by weak tax administrations, they can result in corruption and should therefore be implemented only with caution.
V. Division of Tax Powers Between the Central and Local Governments
The allocation of fiscal powers between different levels of government is a complex problem meriting a whole book. Here, a brief overview of the main legal issues is provided.
A. Classification of Tax Powers
Tax law making powers can be divided in different ways. First, a distinction can be made between various types of taxes: income taxes, wealth taxes, turnover taxes, excise and consumption taxes, and so on. The power of taxation with respect to one particular category of tax is often fully reserved for one specific level of government.
A second distinction can be made with respect to the basic elements of any tax. The structure of any tax consists of several elements: the subjects of the tax or the type of taxpayers, the tax base, the tax rate, and the tax procedure. Theoretically, it is possible to reserve the power to legislate with respect to one element of taxation for one level of government (e.g., tax base and rate) and another element for another level of government (e.g., tax procedure).
Finally, a distinction can be made between the levels of implementation of a tax. As explained above, legislative power can be reserved for one branch of government, while administrative implementation is reserved for another. This is a traditional distinction that can be made for any type of lawmaking power. However, the distinction between lawmaking power and administrative implementation may also be made between different levels of government (e.g., central and regional or local government). In such a case, it is necessary to specify which level of government exercises general lawmaking power, and which levels of government exercise various administrative powers of implementation at various hierarchical ranks—for example, executive decrees, regulations, rulings, and instructions.
B. Leading Principles in the Distribution of Tax Powers
1. Federal vs. a Centralized State
The most important factor that determines the distribution of tax law making powers among the various levels of government is whether the state is federal or centralized. In referring to federal and centralized states, it is important to remember that these are simplifications and that the constitutional reality of any particular country may defy easy categorization.
In a centralized state, there are usually only two significant levels of government: the central government and the local government. Intermediate levels of government can exist, but they are usually politically and fiscally unimportant. A federal state is characterized by the fact that in addition to the central and local levels of government, there is a strong intermediate level of government in the form of autonomous or independent regions or states. In several European countries, there is a tendency toward the constitution of a federal state. Formerly centralized states, such as Belgium, France, Italy, Spain, and even the United Kingdom, are all in varying degrees in the process of organizing political and fiscal power at the intermediate level of government. For example, Belgium and Spain have become federal states similar to Austria and Germany. Yet, examples of fairly centralized states continue to exist—Denmark, Ireland, Netherlands, Portugal, and Sweden.
2. Economic and Monetary Union in a Federal State
Within a federal state with a market economy, the preservation of economic and monetary union is a basic element in determining at what level of government certain taxes should be levied.
An economic and monetary union in a free market presupposes a minimum degree of economic cohesion and uniformity. This in turn requires that certain taxes be levied only by the central government. An obvious example is customs duty. If regional governments have the power to levy customs duties, they can obstruct the flow of goods between regions. The first rule of an economic and monetary union is that all taxes related to the import or export of goods are levied by a central authority or at least levied in accordance with rules that are the same for all the component states belonging to the union.216 Other taxes that affect interregional commerce must be levied in a manner that will not unduly impede such commerce.
3. Relation Between Revenue and Expenditure
Another important principle in the distribution of tax law making powers is the balance between revenue and expenditure. A certain overall equivalence between the amount of taxes that can be raised autonomously by local governments217 and the volume of public outlays for which they are responsible is indispensable. This equivalence between taxing power and spending power is an indicator of the true degree of autonomy of local governments.
Of course, the constitutional setup can be organized in such a way that a local government does not raise its own revenue but is subsidized by grants from a higher level of government; that is, the federal government raises the revenue and transfers the funds to local governments. This mechanism is often used in federal states to transfer funds from richer regions to poorer regions.218 The system of financing through grants, in which different governments are responsible for raising revenue and spending it, can lead to problems.
In a system of financing local government through grants, it is difficult to maintain true autonomy of local government. On the one hand, unconditional or unlimited grants can lead to irresponsible behavior by local governments, which will be inclined to spend at the expense of the central government. On the other hand, if the grants are subject to conditions set by the central government, the latter can choke off completely the autonomy of the local governments by imposing strict conditions on these grants or by restricting their amounts.
Besides the democratic substance of the tax system, budgetary principles call for a rough balance between taxing and spending powers: such balance reflects the true allocation of costs of government functions. In a system of financing local governments with unconditional grants, the burden of cost for the operation of a specific level of government is not reflected at the level of government that is spending. Therefore, it will be more difficult to determine the real operating cost of that level of government.
4. Distribution of Tax Law Making Power with Respect to Certain Elements of the Tax
In many cases, full legislative power for all elements of a tax is not vested in one particular level of government, but distributed over several levels of government. This is often the case when the revenue raised from a particular tax is shared by two or more levels of government.
The most frequent model is one in which the central government retains control over the determination of the subjects of taxation, the tax base, and the procedural rules, but the power to fix rates is shared with other levels of government. This model exists in several European countries and in Japan, whereby a surcharge of one or more national taxes is levied to benefit local governments.219
In some cases, besides the power to set the rates, part of the legislative power with respect to the tax base also belongs to regional or local governments. In other cases, simultaneous and full parallel taxing power on the same tax is held by federal and regional levels of government. Examples of this situation are not so common because the coexistence of two levels of legislative power over the same tax is a constant source of conflict. In Belgium, for example, the tax on estimated rental income from real estate is distributed among no fewer than four levels of government: the central state, the regions, the provinces, and the local municipalities. The central state determines the general rules for the tax base and includes this income in the tax base of the progressive income tax. The regions set a separate flat rate on the tax base as determined by the central state, but have the power to introduce certain exemptions from the tax base and to allow certain reductions of the amount of regional tax due. Finally, the provinces and the local municipalities are entitled to a surcharge on the amount of tax levied by the regions without any change of the tax base.
In Germany, the Federal Government theoretically shares its tax law making power with the state governments.220 This parallel power is limited, however, by another constitutional provision stating that the state governments lose their lawmaking power when the Federal Government has legislated in a tax area.221 In Canada, income tax is imposed on individuals and corporations under the Federal Income Tax Act. The provinces have the power to levy income tax on both individuals and corporations; generally, this power is exercised by setting a provincial rate of tax to be applied to the tax base established by the federal act, the tax being collected by the federal administration. Exceptions are Quebec, which has its own income tax law, and Alberta and Ontario, which have their own corporate income taxes.222
In the United States, the states theoretically have full taxing power, except for customs duties.223 This taxing power is subject to some constitutional limitations, the most important of which is the interstate commerce clause,224 which prohibits the states from obstructing interstate commerce by restrictions in the tax laws. States can therefore provide their own definition of taxable income, although in practice the federal definition is the starting point and the deviations from it are relatively limited in scope in most states. Tax rates differ from state to state, and some states do not even have an income tax. As a result of this parallel taxing power, conflicts on tax jurisdiction may arise between the federal and the state level, as well as among the states themselves. In Switzerland, the confederation and the cantons effectively share tax law making power for direct taxes on income and wealth.225 Conflicts between certain types of tax legislation are solved by harmonization of the conflicting tax rules.226
Another example of this setup is the way in which customs duties are administered in the European Union. All the rules with respect to the subjects of taxation, the determination of the base, and the rates are determined by EU law. The tax administration and procedure (i.e., tax returns, control measures, tax protests, and litigation) are administered in accordance with the national law of the member states.
5. Distribution of Tax Law Making Power According to the Level of Implementation of the Tax
Finally, it is possible to distribute tax law making power in accordance with the level of implementation of the tax. In this model, the general rules with respect to the subject of the tax, the tax base, and rates are fixed at the central level of government, while the more concrete details of the implementation of the tax are left to lower levels of government. For example, in Germany, regional tax authorities administer major federal tax laws for the account of the federal treasury.227 Another example of this model can be found in the way the VAT and certain aspects of corporate income tax are implemented in the EU.
The VAT has been introduced in the EU by way of directive.228 A directive is a legislative act issued by the Council of Ministers of all the member states, who decide by unanimous vote (in tax matters) to introduce certain tax rules. In this case, the basic rules determining the subject of taxation, the tax base, and part of the rules of procedure and administration (not the rates) have been determined by directive, leaving certain options to member countries. Each of the member states then implements the VAT through national laws. Disputes may arise when taxpayers argue that the national laws are inconsistent with the directive.229
The same pattern is emerging in the EU with respect to the corporate income tax. Certain requirements with respect to the treatment of corporate reorganizations and intercorporate dividends, which particularly affect corporate groups with members in more than one state, have been imposed by directive.230 Further, in a report on the harmonization of the corporate income tax, a committee of independent experts advised the European Commission to set certain minimum rules with respect to tax rates and tax bases beyond which member states should not go.231 Within the outer Limits established by these minimum rules, member states would retain full taxing power.
6. Deduction or Credit for Regional and Local Taxes
An important question in the distribution of revenue between various levels of government is whether a local tax is deductible from the tax base determined by the central government or whether it can be credited against the amount of tax due to the central government.
If a tax levied by a regional or local government can be credited without limit against a tax levied by the central government, the lower government, by increasing its taxes, can completely wipe out the tax revenue of the central government. An example of this is the tax on estimated rental income from real estate in Belgium. The regional, provincial, and local taxes on estimated rental income can be credited against the progressive personal income tax levied by the Central Government; that is, the amount of tax due to the lower governments is deducted from the amount of tax due to the Central Government and only the balance has to be paid. To prevent the regional and local governments from reducing the Central Government’s revenue by increasing their taxes, the Central Government has set a limit on the amount of tax that can be credited at 12.5 percent of the tax base.
A similar problem arises when local taxes are deductible in determining the base of a tax levied by the central government. Recent examples of this are environmental taxes, such as taxes on litter or the use of water, levied by regional or local authorities, that are deducted from the corporate income tax base. As the local tax burden increases, the tax base for the central government is reduced.
7. Distribution of Tax Law Making Powers in a Centralized State
The distribution of tax law making powers in most centralized states is fairly simple because there are only two significant levels of government: central and local. The local government in most cases is too small to administer any of the important taxes, so the power to impose the most important taxes rests with the central government.
In a typical centralized state, all major modern taxes are levied by the central government. All aspects of legislative power over these taxes rest with the central government, and local governments are not involved in their implementation or administration. Allocation of revenue to local government is typically governed by a law on local finance.232
The problem in such centralized states is that local governments may not have adequate taxing power. While it is not possible to make a complete inventory of taxes levied by local governments, some patterns of taxation do emerge. Local governments in many Western European countries are typically financed by surcharges on personal or corporate income tax, surcharges on national road taxes, taxes on real estate, and taxes on business activity. There is often a ceiling on the amount of local surcharge to be levied. Taxes on the estimated value of real estate, on rented rooms or hotel rooms in tourist sites, or on second residences are categorized as taxes on real estate. Taxes on personnel or equipment used in the exercise of a business, taxes on business offices or the authorization to open a local business, or taxes on turnover or on the exercise of a business are categorized as taxes related to a business or professional activity. Finally, taxes such as those on the collection of refuse and litter, sewer connections, and the delivery of passports and public certificates relate to services provided by the local administration.
As the financial needs of local governments grow, the proliferation of all types of taxes increases the tax burden and can make the local tax “system” incomprehensible and obscure. Therefore, it is preferable to reserve a few major sources of revenue, such as surcharges on personal and corporate income tax, for local governments, so that they are not obliged to raise taxes arbitrarily.
8. Distribution of Tax Law Making Powers in a Federal State
The distribution of tax law making power in a federal state is much more complex than in a centralized state because there is at least one additional level of government (the regional government) large enough to administer a major modern tax system. In a federal system, the question is how to distribute tax law making power with respect to major taxes while maintaining economic and monetary union. In a federal state, both the federal government and the states often have full power to raise important taxes, such as corporate and individual income tax and sales taxes. A single corporation may be liable to corporate income tax in all the states in which it does business. This raises the risk that either (1) the various states in which the corporation operates will each seek to tax more than their appropriate share of the corporation’s income, thereby leading to multiple taxation of the same income, or (2) the corporation will take advantage of the different tax rules operating in each of the states to arrange its affairs so that much of its income escapes taxation.
The problems involved in limiting the taxing authority of regional governments are beyond the scope of this book.233 In countries where regional governments enjoy fiscal autonomy, there is usually substantial litigation concerning these limitations.
Note: Victor Thuronyi contributed to the writing of this chapter.
A special case is the customs tariffs and the minimum rates of the value-added tax (VAT) in the European Union, which are not determined by the national legislators, but proposed by the European Commission and decided by the European Council of Ministers. Even in this case, a statute would be needed to implement the decision in domestic law.
See Bundes-Verfassungsgesetz [Federal Constitution] art. 18 (AUT); Grondwet [Constitution] art. 170 (BEL); Const. art. 91(3)(CAN); Grundlov [Constitution] § 43 (DNK); Hallitusmuoto [Constitution] § 61 (FIN); Const. art. 34 (FRA); Const. art. 23 (ITA); Const, art 99 (LUX); Const. art. 106(2) (PRT); Const. art. 133 (ESP). Although art. 58 of the Constitution of the People’s Republic of China provides that legislative power is exercised by the National People’s Congress and its Standing Committee, there is no constitutional provision that requires a specific legal basis for imposing taxes. The National People’s Congress can also delegate legislative power to the State Council, which is the highest executive organ of state administration. Xianfa [Constitution] art. 85. As a result, there has been some confusion as to which institution in China has the power to propose and approve tax laws. In 1985, the National People’s Congress authorized the State Council to make provisional laws and regulations with respect to foreign investment and economic reform. See Decision of the Third Session of the Sixth National People’s Congress on Authorizing the State Council to Formulate Interim Provisions or Regulations Concerning the Reform of the Economic Structure and the Open Policy (adopted Apr. 10, 1985), reprinted in Bureau of Legislative Affairs of the State Council of the P.R.C., 1 Laws and Regulations of the People’s Republic of China Governing Foreign-Related Matters 391 (1991).
See, e.g., Const. art. 245, seventh sched., List I, item 82 (IND) (parliament may establish taxes on income other than agricultural income); id. item 92A; id. List II, item 54 (parliament may tax the sale or purchase of goods where the sale or purchase takes place in the course of interstate trade or commerce, but other sales are subject to taxation only by the states); id. List II, item 53 (only the states may tax the consumption or sale of electricity). In 1976, Pakistan, which has a similar setup, amended its Constitution to grant to the Federal Government power to tax “the sales and purchases of goods imported, exported, produced, manufactured, or consumed.” See Const. art. 142, fourth sched., item 49 (PAK).
See Grondwet [Constitution] art. 174 (BEL); Const. art. 47 (FRA) and Ordonnance No. 59–2 of Jan. 2, 1959, Portant loi organique relative aux lois de finances, art. 4, Dalloz, Législation [D.L.] 175 (1959); Guy Gest & Gilbert Tixier, Droit fiscal 33–34 (4th ed. 1986); Const. art. 81 (ITA); Const. art. 134 (ESP). In the United States, the Constitution requites congressional consent for any spending of public money, U.S. Const. art. 1, § 9, but does not require annual consent for taxation. Accordingly, if Congress withheld its consent to public spending, the Government would have to stop spending, but the liability of citizens to pay taxes would remain unaffected.
See Const. art. 42 quater (CHE), translated in XIX Constitutions of the Countries of the World (Albert P. Blaustein & Gisbert H. Flanz eds., 1982) (“The Confederation is entitled to enact regulations, by means of legislation, against arrangements with taxpayers granting unjustified tax advantages”).
As a consequence, the institution of plea bargaining (not contesting a charge of a lesser offense in order to avoid a charge under a major offense), which is well known in the United States, does not exist in these countries in respect of major offenses.
In most countries that have constitutional control by the courts, the constitutional court is competent to check whether a law violates any constitutional provision. This is the case in France, Germany, Italy, and the United States. In some countries, however, the constitutional court has limited control. This is the case in Belgium, where the Cour d’arbitrage can only check violations of the rule of equality and laws violating the constitutional distribution of power and the economic and monetary union of the country.
See Judgment of Dec. 27, 1973, Conseil constitutionnel [Con. const.], 1974 La Semaine juridique (Juris-Classeur Périodique) [J.C.P] II, No. 17691. The decision concerned former article 180 of the General Tax Code, as amended by the 1973 Finance Act, which allowed taxpayers to contest the taxation d’office, under which income tax could be imposed on the basis of the taxpayer’s expenditures, by proving that the expenditures were financed by resources other than taxable income. This opportunity for proof, however, was unavailable to taxpayers whose income exceeded a specified level. It was this denial—to one group of taxpayers only—of an opportunity to prove their case that the court found objectionable.
See Judgment No. 95-369 of Dec. 28, 1995, Con. const., 1996 J.C.P. II, No. 67749. In this case, the court held that a reduction in the inheritance tax on an interest in a business, conditioned only on the heir retaining the property for five years, without being required to participate in the management of the company, discriminated in favor of one type of property without any rational legislative purpose. See also infrach. 12, note 40.
See, e.g., Nordlinger v. Hahn, 505 U.S. 1 (1992); see also Apache Bend Apartments, Ltd. v. United States, 964 F.2d 1556, 1562–69 (5th Cir. 1992). In that case, the court upheld so-called rifle-shot transition rules, which singled out particular taxpayers (usually those with effective lobbying representation) for transitional relief from the application of the Tax Reform Act of 1986. The court refused to find that this type of ad hoc transition relief was so arbitrary as to violate the constitutional requirement of equal protection of the law.
Const. art. 53, cl. 1 (ITA), translated in IX Constitutions of the Countries of the World (Albert P. Blaustein & Gisbert H. Flanz eds., 1987). Art. 53(2) of the Constitution of Romania provides that “[t]he legal taxation system must ensure a fair distribution of the tax burden,” These provisions have probably been inspired by the French Déclaration des droits de l’homme et du citoyen of Aug. 26, 1789, which is an integral part of the present 1958 Constitution of France, and art. 13 of which says: “Pour l’entretien de la force publique, et pour les dépenses d’administration, une contribution commune est indispensable; elle doit être également répartie entre tous les citoyens, en raison de leurs facultés.”
See Judgment of July 6, 1972, Corte costituzionale [Corre cost,], 1972 Giurisprudenza Cositituzionale [Giur. Cost.] 1, No. 120, at 1289; Judgment of Apr. 19, 1972, Corte cost., 1972 Giur. Cost. I, No. 62, at 272; Judgment of Dec. 13,1963, Corte cost., 1963 Giur. Cost. I, No. 155, at 1546.
All shall contribute to the sustenance of public expenditures according to their economic capacity through a just tax system based on the principles of equality and progressiveness, which in no case shall be of a confiscatory scope,” Const. art. 31, § 1 (ESP), translated in XVIII Constitutions of the Countries of the World (Albert P. Blaustein & Gisbert H. Flaz eds., 1991).
The European Commission on Human Rights has stated that tax legislation can be scrutinized under the European Convention on Human Rights on the basis of whether “a reasonable degree of proportionality existed between the means employed and the aim sought to be achieved.” A., B., C., and D. v. The United Kingdom, App. No. 8531/79, 23 Eur. Comm’n H.R. Dec. & Rep. 203, 211 (1981). In this case, discussed in the text at note 47, infra, the Commission found that the retroactive application of the statute was reasonably related to the aim of the legislator (prevention of further use of tax shelters).
E.g., Code civil art. 2 (BEL); Code civil art. 2 (FRA). See Claude Gambier & Jean-Yves Mercier, Les impôts en France §§ 2280–81 (1991) (explaining that, under the civil code, in the absence of an explicit statement in the law, provisions take effect for taxable events occurring after publication in the official gazette; in the case of income tax, this means that if publication occurs before Dec. 31, the current year will be affected, since the taxable event is considered not to occur until the close of the year).
The courts recognize that the legislator may deviate from the ordinary rule of non-retroactivity in light of an overriding interest of public order.” Louis Trotabas & Jean Marie Cotteret, Droit fiscal 138 (1985) (ed. trans.).
See Judgment No. 86-223 of Dec. 29, 1986, Con. const., 1987 J.C.P. II, No. 20903; Judgment No. 95-369 of Dec. 28, 1995, Con. const., 1996 J.C.P. II, No. 67749 (court decisions may be overturned retroactively only for reasons based on the public interest).
Rechtsstaatsprinzip. Similarly, the Polish Constitutional Tribunal struck down income tax amendments that would have come into effect less than one month after the legislation was passed on the basis that taxpayers were given inadequate notice. See Janusz Fiszer, Constitutional Battle over Poland’s 1996 Personal Income Tax Rates, 12 Tax Notes Int’l 246 (1996).
See generally John S. Nolan & Victor Thuronyi, Retroactive Application of Changes in IRS or Treasury Department Position, 61 Taxes 777 (1983). The German Constitutional Court applies its doctrine concerning retroactivity to regulations as well as to statutes. See Judgment of June 8, 1977, BVerfG, 45 BVerfGE, No. 6, at 142.
For example, in the United States, the power to levy taxes is subject to the general limitations on legislative power in the Constitution, such as the due process clause of the Fifth Amendment. In practice, U.S. federal tax legislation is very rarely found to be unconstitutional. An important exception is the Pollock decision. See Pollock v. Farmers’ Loan & Trust Co., 157 U.S. 429 (1895).
See Judgment of Jan. 17, 1957, BVerfG, 6 BVerfGE, No. 9, at 55; see generally 1 Tipke, supra note 53, at 380. Art. 6/1 of the Constitution of the Federal Republic of Germany provides: “Marriage and family shall enjoy the special protection of the state.” VIII Constitutions of the Countries of the World (Albert P. Blaustein & Gisbert H. Flanz eds., 1994).
The Constitutional Court of Guatemala read a provision of the income tax law as taxing an item of income twice and struck it down as violating a constitutional prohibition against double taxation. Cases Nos. 39-88 and 40-88, Corte de Constitucionalidad, in Leyes y Reglamentos de la Reforma Tributaria 83, 91–92 (Luis Emilio Barrios Pérez ed., 1989). See also Pollock v. Farmers’ Loan & Trust Co., 157 U.S. 429 (1895).
See In re James Richardson & Sons, Ltd. et al. and Minister of National Revenue, 9 D.L.R.4th 1 (1984), where the taxpayer sought protection under the law of privacy. Section 231(3) of the Canadian Income Tax Act, as it then was, gave the tax authorities the power to demand from any person any information “for any purposes related to the administration or enforcement” of the act. The tax authorities relied on this provision and required a company, which was a commodities futures market broker, to reveal the names and addresses of its customers for purposes of doing a feasibility study before introducing a new regulation on information reporting. The tax authorities guaranteed confidentiality of the data during the study. Neither the company nor any of its customers were under investigation at the time. The company refused to turn over the information and challenged the power of the tax authorities at court. The Supreme Court of Canada held that “a requirement of information under § 231(3) could only be made where the Minister was conducting a genuine and serious inquiry into the tax liability of specific persons.” Id. at 1 (quoting case summary).
Section 231(3)(b) of the Income Tax Act, as it then was, authorized the Minister to require a lawyer to produce files relating to his client “within such reasonable time as may be stipulated” in a registered letter. In In re Joseph et al. and Minister of National Revenue, 20 D.L.R.4th 577 (1985), the Minister required the lawyer to produce the information “without delay.” The court held that the Minister had no power to demand information to be produced without delay, which means immediately. Id. at 585. Parliament did not mean immediately when using “reasonable time,” Tax authorities must give the lawyer some time to consider whether to produce the information because of the solicitor-client privilege protection.
When a tax official is conducting an inspection or audit in a taxpayer’s residence or business premise, the official must obtain consent from the taxpayer except where a search warrant is issued by a judge. In considering whether to issue a search warrant, the judge must be convinced that there is evidence of violation of the tax law committed by the taxpayer. The search warrant must also describe the premises to be searched. Otherwise, the search is illegal, and the documents seized will be illegal evidence, which cannot be used in a court of law.
Subject to certain exceptions. Sections 2(a) and 3 of Article VIII of the IMF’s Articles of Agreement prohibit IMF members from imposing restrictions on payments and transfers for current international transactions, or from engaging in multiple currency practices or discriminatory currency arrangements. These provisions prohibit the authorities of member countries from imposing some types of tax measures through their exchange systems. For example, the imposition by a member country of a tax on the purchase or sale of foreign exchange will give rise to a multiple currency practice (Article VIII, Section 3) if the tax exceeds 2 percent of the amount purchased or sold. Moreover, a restriction on payments and transfers for current international transactions (Article VIII, Section 2(a)) will arise if the authorities of a member country, before permitting a nonresident to transfer abroad the proceeds of current international transactions (e.g., profits and dividends), require the nonresident to pay outstanding taxes that are not related to the amount to be transferred. See generally International Monetary Fund, Selected Decisions and Selected Documents of the International Monetary Fund 354, 366–68 (20th issue 1995).
In Case 270/83, Commission v. France, 1986 E.C.R. 285, the Court of Justice of the European Communities held that France discriminated against French branches of nonresident EU companies because it denied a tax credit on French-source dividends paid to such branch offices. The argument of the French Government, that it was justified in making an internationally accepted distinction between resident and nonresident taxpayers, was dismissed by the court.
In Case C-175/88, Biehl v. Administration des contributions du grande-duché de Luxembourg, 1990 E.C.R. 177, the Court of Justice held that a Luxembourg tax law violated the nondiscrimination rule because taxpayers who during the tax year moved abroad were denied the right to claim a refund on the excess withholding tax on wages when their annual tax liability on Luxembourg-source income, because of the move, fell below the amount of taxes on salary that had been withheld during their stay in Luxembourg. The court was of the opinion that this disadvantage would hit nonresidents much more often than residents and, therefore, constituted a violation of art. 48 of the treaty.
In Case C-279/93, Finanzamt Köln-Alstadt v. Schumacker, 1995 E.C.R. 225, the Court of Justice held that tax law may make a distinction between resident and nonresident taxpayers. However, for example, if both categories are basically under the same circumstances, when a nonresident earns the major part of his income in another member state, then resident and nonresident taxpayers should be treated identically. In particular, a nonresident taxpayer should benefit from the same refunds on progressive income taxes as a resident taxpayer.
In Case 433/85, Feldain v. Directeur des services fiscaux du département du Haut-Rhin, 1987 E.C.R. 3521, a French law imposing a progressive motor vehicle tax, depending on the horsepower of the car, was held to violate the nondiscrimination principle, because the progressivity of the rate scale, although couched in general terms, was structured in such a way that only foreign cars were subject to the highest tax brackets of the rate scale, resulting in a considerable tax advantage for French domestic luxury cars.
In Case 171/78, Commission v. Denmark, 1980 E.C.R. 447, the Court of Justice held that a lower excise tax on aquavit (the Danish national drink) than on whiskey and gin constituted a violation of the nondiscrimination principle, when in fact the largest part of aquavit was manufactured domestically, while whiskey and gin were mainly imported. The fact that the tax rule did not make a specific distinction between imported goods and domestically manufactured goods was considered to be irrelevant.
Case 3/1993/398/476, Benjenoun v. France of Feb. 24, 1994, série A, No. 284. See Guy Gest et al., Convention europeénne des droits de l’homme et fiscalité—Bilan et perspectives, Les petites affiches, No. 80 (1994).
E.g., Const. art. 205(1) (HND); Decreto No. 115 of Nov. 4, 1966, Gaceta No. 19,011 (HND); HND IR art. 24; Const. art. 58(3) (KGZ). In Belgium, Parliament historically had the power to make interpretive laws. See Law of Aug. 4, 1832 on the Organization of the Supreme Court (Cour de cassation), arts. 23-24, 1832 Pasinomie 469 (abolished by the law of July 7, 1865) (BEL).
Under art. 67 of the Constitution of the People’s Republic of China, the Standing Committee of the National People’s Congress has the power to interpret the Constitution and other national statutes. This means that authoritative interpretation of laws, including tax laws, is in the first place the work of the legislator. However, the Chinese tax legislator has not made frequent use of this power. Article 89(18) of the Constitution allows a delegation of this power to lower agencies. In this way, the constitutional provision is used to grant regulatory power to the Ministry of Finance and to the State Administration of Taxation to issue interpretive regulations of the tax laws, as is the practice in many countries of the Organization for Economic Cooperation and Development (OECD).
See Judgment of July 8, 1992, Conseil d’Etat, 1992 Recueil des décisions [arrêts] du Conseil d’Etat [Lebon], No, 88734. at 284; See also older cases cited in Jean-Jacques Bienvenu, Droit fiscal Nos. 52–54 (1987).
This is the theory of “simulation,” or sham. See Judgment of Feb. 15, 1854, Cour de cassation (civile), 1854 Recueil Dalloz périodique et critique [DP.] I 51: Judgment of Dec. 11, 1860, Cour de cassation (civile), 1861 D.P.I 25; Judgment of Aug. 20, 1867, Cour de cassation (civile), 1867 D.P. I 337.
See discussion under Abus de droit in Encyclopédie juridique, 1 Répertoire de droit civil 28 (Dalloz 1951); See also infrasec. III(E) for the discussion of interpretation of tax law in the Netherlands.
See BEL CIR art. 344 (permitting the tax administration to set aside any legal qualification of an act or a transaction by a taxpayer, when the purpose of such act or transaction was tax avoidance, unless the taxpayer can show a legitimate business purpose).
See Judgment of Dec. 6, 1883, Leuven, Recueil général de l’enregistrement et du notariat [Rec. Gén. Enr. Not.] 10.272; Judgment of Jan. 4, 1900, Brussels, Rec. Gén. Enr. Not. 13.221; Judgment of March 3, 1912, Brussels, Rec. Gén. Enr. Not. 15.129.
DEU Steueranpassungsgesetz § 1/11 (according to which the interpretation of the tax law had to consider “the social viewpoint, the purpose, and the economic significance of the tax laws and the development of the (economic) relationships” (ed. trans.)).
In Germany, this narrow and literal interpretation was called Begriffsjurisprudenz (conceptual jurisprudence) and subject to attack by the end of the nineteenth century. See Karl Larenz, Methodenlehre der Rechtswissenschaft (1983).
… das Steuerrecht wird von der Idee der ‘primären Entscheidung des Gesetzgebers über die Steuerwürdigkeit bestimmter generell bezeichneter Sachverhäte’ getragen und lebt dementsprechend ‘aus dem Diktum des Gesetzgebers,’” Judgment of Jan. 24, 1962, BVerfG, 13 BVerfGE, No. 32, at 318, 328 (“tax law is based on the idea of the ‘primary decision of the legislator concerning the tax treatment of certain generally defined circumstances’ and therefore draws breath ‘from the statement of the legislator’” (ed. trans.)).
Der finanzgerichtlichen Rechtsprechung ist es insbesondere nicht von vornherein verwehrt, im Wege der Rechtsfortbildung veränderten wirtschaftlichen Situationen Rechnung zu tragen....” Judgment of Mar. 12, 1985, BVerG, 69 BVerfGE, No. 12, at 188, 203 (“Judicial decisions in fiscal law are not prohibited from giving significance to changed economic circumstances by way of development of the law …” (ed. trans.)).
According to Tipke, this was because it was considered unnecessary, the approach of Begriffs-jurisprudenz (see note 102 supra) having been abandoned. See 3 Tipke, supra note 53, at 1239. DEU AO §§ 40–42 does contain a few specific antiavoidance provisions, some of which may be interpreted as the continuance of economic interpretation. These provisions, however, have a clear legal meaning.
See Judgment of May 26, 1926, Hoge Raad [HR], 1926 Nederlandse Jurisprudentie [N.J.] 723. The Swiss courts have applied an interpretation of tax law that is very similar to the Dutch theory of fraus legis. There is an abuse of law when the legal form of a transaction is unusual, it was entered into with the intent of obtaining a tax benefit, and the benefit must effectively have been realized. See Jean-Marc Rivier, Droit fiscal Suisse: L’imposition du revenu et de la fortune 61 (1980); Ernst Höhn, Steuerrecht 17 (1972).
See Judgment of Dec. 19, 1990, HR, 1990 B.N.B. 121. A more recent case is discussed in Dick Hofland & Kees van Raad, Dutch Consolidated income That Erodes Interest Payment to Foreign Parent Company Is Not an Abuse of Law, 11 Tax Notes Int’l 1143 (1995).
Commissioners of Inland Revenue v. Duke of Westminster, 1936 App. Cas. 1, 19, 19 T.C. 490. For a comparative study of the interpretation of tax laws in France and the United Kingdom, see Stefan Frommel, United Kingdom Tax Law and Abuse of Rights, Intertax 54 (1991/92); L’abus de droit en droit fiscal britannique, Revenue internationale de droit comparé 585 (1991) (same paper in French).
Investment and Merchant Finance Corp. Ltd. v. Federal Commissioner of Taxation, 125 C.L.R. 249, 265 (1971); See also Curran v. Federal Commissioner of Taxation, 131 C.L.R. 409 (1974); South Australian Battery Makers Proprietary Ltd. v. Federal Commissioner of Taxation, 140 C.L.R. 645 (1978).
See W.P. Keighery Proprietary Ltd. v. Federal Commissioner of Taxation, 100 C.L.R. 66, 92 et seq. (1956–57); Cecil Bros. Proprietary Ltd. v. Federal Commissioner of Taxation, 111 C.L.R. 430, 441 (1962–64); Mullens v. Federal Commissioner of Taxation, 135 C.L.R. 290, 302 (197–576).
See West Coast Marketing Corp. v. Commissioner, 46 T.C. 32 (1966); American Potash & Chemical Co. v. United States, 399 F.2d 194 (U.S. Ct. Cl.), motion denied, 402 F.2d 1000 (Ct. Cl. 1968); King Enterprises, Inc. v. United States, 418 F.2d 511 (U.S. Ct. Cl. 1969), later proceeding 190 Ct. Cl. 947 (1970).
Loopholes can also result from a disorderly legislative process. Sometimes chaotic amendments are made at the last minute without an opportunity to consider all their ramifications and make the necessary adjustments.
To avoid any confusion in terminology, it should be noted that “tax evasion” is translated in French as fraude fiscale and in German as Steuerhinterziehung, whereas “tax avoidance” is respectively translated as évasion fiscale and Steuerumgehung.
For the United Kingdom, Commissioners of Inland Revenue v. Duke of Westminster, 1936 App. Cas. 1, 19 (Lord Tomlin comments, “[e]very man is entitled if he can to order his affairs so that the tax attaching under the appropriate Acts is less than it otherwise would be”); for the United States, Gregory v. Helvering, 69 F.2d 809, 810 (1934)(Judge Learned Hand stating, “[a]ny one may so arrange his affairs that his taxes shall be as low as possible; he is not bound to choose that pattern which will best pay the Treasury; there is not even a patriotic duty to increase one’s taxes,”), aff’d, 293 U.S. 465 (1935); for Australia, Jaques v. Federal Commissioner of Taxation, 34 C.L.R. 328, 362 (1924) (Judge Starke wrote, “[t]here is nothing wrong in companies and shareholders entering, if they can, into transactions for the purpose of avoiding, or relieving them of taxation …”); for Belgium, Judgment of June 6, 1961, Cour de cassation, 1961 Pas. Bel. I 1082, 1089 (“considering that there is neither a prohibited fabrication with respect to the fisc, nor one which constitutes fraud, when the parties, in order to benefit from a more favorable tax regime, taking advantage of the freedom of contract and without violating any legal obligation, establish legal acts all of the consequences of which they accept, even if the form that they give them is not the most usual one” (ed. trans.)).
Belgium (BEL CIR art. 344, as amended in 1993) and Canada (CAN ITA § 245, introduced in 1988) also have general antiabuse provisions, but they are too recent to be able to evaluate their impact on interpretation of tax laws by the courts. Sweden abolished the general antiavoidance provision in 1992 and reintroduced it in 1995.
See W.P. Keighery Proprietary Ltd. v. Federal Commissioner of Taxation, 100 C.L.R. 66, 92 (1957)(“Whatever difficulties there may be in interpreting s. 260, one thing at least is clear: the section intends only to protect the general provisions of the Act from frustration, and not to deny taxpayers any right of choice between alternatives which the Act itself lays open to them.”); Cecil Bros. Proprietary Ltd. v. Federal Commissioner of Taxation, 111 C.L.R. 430, 441 (1964)(“Indeed, s, 260 does not authorize the Commissioner to do anything; it avoids as against the Commissioner arrangements, etc. as specified and so leaves him to assess taxable income and tax on the facts as they appear when the avoided arrangements, etc. are disregarded.”); Mullens v. Federal Commissioner of Taxation, 135 C.L.R. 290 (1976).
“This Part applies to any scheme … where … (a) a taxpayer (in this section referred to as the “relevant taxpayer”) has obtained … a tax benefit in connection with the scheme; and
(b) having regard to—
(i) the manner in which the scheme was entered into or carried out;
(ii) the form and substance of the scheme;
(iii) the time at which the scheme was entered into and the length of the period during which the scheme was carried out;
(iv) the result in relation to the operation of this Act that, but for this Part, would be achieved by the scheme;
(v) any change in the financial position of the relevant taxpayer that has resulted, will result, or may reasonably be expected to result, from the scheme;
(vi) any change in financial position of any person who has, or has had, any connection (whether of a business, family or other nature) with the relevant taxpayer, being a change that has resulted, will result or may reasonably be expected to result, from the scheme;
(vii) any other consequence for the relevant taxpayer, or for any person referred to in subparagraph (vi), of the scheme having been entered into or carried out; and
(viii) the nature of any connection (whether of a business, family or other nature) between the relevant taxpayer and any person referred to in subparagraph (vi),
it would be concluded that the person, or one of the persons, who entered into or carried out the scheme or any part of the scheme did so for the purpose of enabling the relevant taxpayer to obtain a tax benefit in connection with the scheme or of enabling the relevant taxpayer and another taxpayer or other taxpayers each to obtain a tax benefit in connection with the scheme (whether or not that person who entered into or carried out the scheme or any part of the scheme is the relevant taxpayer or is the other taxpayer or one of the other taxpayers).”
Acts Interpretation Act, 1901, as amended, 1901 Austl. Acts 2, § 15AA(1). Sec. 15AB of the Act also contains rules with respect to the extrinsic materials that should be taken into consideration for the interpretation of an act. See infrach. 3, sec. III(C) for discussion of Interpretation Acts.
See FRA LPF art. L. 64 to L. 64 B (prohibiting abus de droit). This provision was introduced for indirect taxes by an act of July 13, 1925, and for income taxes by an act of Jan. 13, 1941. Act No. 87–502 of July 8, 1987, introduced an optional ruling procedure (known as rescrit) for its application.
Judgment of June 10, 1981, No. 19,079, Conseil d’Etat, Lebon 248; Judgment of Apr. 19, 1988, No. 86.19079, Cour de cassation, Chambre commerciale, Revue de jurisprudence fiscale 1989, No. 2, at 47. See also Cyrille David et al., Les grands arrêts de la jurisprudence fiscale, Thème 9, 106 et seq. (2d ed. 1991).
Código Civil art. 6, ¶ 4 (ESP)(stating “acts concluded within the scope of the text of a rule which pursue a result prohibited by the legal regulation or contrary to it, shall be considered as executed as a fraud on the law and shall not thwart the proper application of the norm that was sought to be avoided” (ed. trans.)).
ESP LGT art. 24, ¶ 2 (providing, in part, “to avoid fraud on the law it will be understood, for purposes of the previous paragraph, that there is not an extension of the taxable event in the case of taxation of actions realized for the proven purpose of evading the tax, as long as they produce a result equivalent to that derived from the taxable event” (ed. trans.)).
For a full discussion of the abuse of law provisions in Spain, see Escuela de Inspección Financiera y Tributaria, Ministerio de Economía y Hacienda, Compendio de Derecho Tributario Español 79-88 (4th ed. 1984).
Real Decreto [Royal Decree] 1.919/1979 of June 29, 1979, por el que se regula el procedimiento especial de declaración de fraude de Ley en materia tributaria, Boletín Oficial del Estado de 6 de agosto.
A full discussion of these rules can be found in the relevant chapters of the material tax law throughout the book. Provisions of an intermediate nature are also possible, for example, a denial of deductions incurred in a contract lacking a real economic purpose. See Daniel Deak, New Anti-Avoidance Legislation Enacted in Hungary, 12 Tax Notes Int’l 446 (1996).
See, e.g., Const. arts. 38, 92 (FRA). The French Constitution also provides in art. 37 for a general power to make regulations on matters that are not within the scope of lawmaking under art. 34. This means that regulations can be made by the executive under art. 37 without the explicit delegation of authority by a law. See also Grundlov [constitution] § 23 (DNK); Const. art. 86 (ESP).
The Constitution of Guatemala prohibits tax regulations from modifying the statutory liability to pay tax and confines them to procedural issues. Art. 239 provides:
The provisions, hierarchically inferior to the law, which contradict or twist the sense of the legal provisions regulating the bases of tax collection, are ‘ipso jure’ void. Regulatory provisions cannot modify said bases and will provide specific rules for the administrative collection of taxes and establish the procedures facilitating their collection.
For example, under art. 34 of the French Constitution, “the basis, the rate and the methods of collecting taxes of all types” must be determined by an act of Parliament. See VII Constitutions of the Countries of the World (Albert P. Blaustein & Gisbert H. Flanz eds., 1988). Art. 37 provides that “[m]atters other than those that fall within the domain of law shall be of a regulatory character.” Id. Accordingly, matters such as administration and procedure may be dealt with by regulation. See Loïc Philip, Droit fiscal constitutionnel 29 (1990). Moreover, while a strict reading of art. 34 would require all rules concerning the basis, rate, and methods of collecting taxes to be enacted by Parliament, leaving no room for regulations, given the impracticality of such an approach, the French courts have recognized that while the basic rules of taxation must be contained in the law, regulations may provide for the application of these rules. See id. at 30; See also Judgment No. 86–223 of Dec. 29, 1986, Con. const., 1987 J.C.P. II, No. 20903. On the limitations of tax law making powers, see Judgment of Oct. 12, 1983, Con. const., 1985 Recueil Dalloz-Sirey, Jurisprudence, Informations rapides 351; Judgment of May 23, 1984, Conseil d’Etat, 1984 Lebon 188. Thus, while in principle there is a constitutional limitation on what may be provided in regulations, as opposed to laws, as a matter of practice, many rules of taxation are provided by regulation. The Code général des impôts (CGI) does not specifically authorize regulations, since this is unnecessary under the constitutional system. The French regulations are published in a companion volume to the CGI and are about equal in length to the Code.
See GG art. 80 (DEU). On the basis of this constitutional provision, a regulation (Durchführungsverordnung) has been issued for practically all the major taxes. See also Judgment of Mar. 5. 1958. BVerfG, 7 BVerfGE, No. 36, at 282, 301 (DEU) (“Art. 80 of the Constitution is intended to force the legislator itself to set the rules that are decisive for the regulation of an area, and to the extent that details are left to the executive, to determine their direction and extent, in such a way that the possible contents of the regulations can be foreseen” (ed. trans.)).
For instance, the models of tax forms to be filed and the annexes to be joined, the tax rules that are specific to a certain industry in applying a tax, schedules for depreciation and stock valuation, rules specifying evidence for certain business expenses, specific accounting requirements for tax purposes, implementing rules for tax registration, and rules containing filing requirements.
Section 7805(a) of the U.S. Internal Revenue Code provides that “the Secretary shall prescribe all needful rules and regulations for the enforcement of this title....” The authority for Canadian Income Tax Regulations follows basically the same pattern as that of the United States. Section 221(1)(j) of the Canadian Income Tax Act (CAN ITA) gives broad powers to make regulations “generally to carry out the purposes and the provisions of the Act.” However, regulations that are inconsistent with the provisions of the Act will not be applied by the courts. See Charos v. Minister of National Revenue, 62 Dominion Tax Cases [D.T.C.] 273 (1962). Regulations are published in the Canada Gazette. CAN ITA § 221(2).
E.g., Verwaltungsanordnungen in Germany. In France, the tax administration publishes instructions and circular letters; these are binding on it. It also publishes annually an explanatory treatise, the Précis de fiscalité; unlike administrative commentaries and interpretations in circular letters or instructions, the Précis de fiscalité is not binding on the French tax administration.
For Belgium, Judgment of Nov. 22, 1949, Cour de cassation, 1950 Pas. Bel. I 182, 1949–50 Algemeen Fiscaal Tijdschrift 258; for Canada, Stickel v. Minister of National Revenue, 72 D.T.C. 6178 (1972) and Canadian Pacific Ltd. v. the Queen, 76 D.T.C. 6120 (1976); for Germany, Judgment of May 31, 1988, BVerfG, 78 BVerfGE, No. 20, at 214, 227; for Spain, Escuela de Inspección Financiera y Tributaria, supra note 178, at 57.
See Gambier & Mercier, supra note 45, at §§ 2260–70 (les agréments fiscaux). A similar procedure is applicable in the United States with respect to certain tax benefits, for example, the historic rehabilitation tax credit, which is based on approval of a project by the Interior Department. See USA IRC §47(c)(2)(C).
See, e.g., USA IRC 5 367(a)(1981) (transfer to foreign corporation is taxable event unless the Secretary issues a ruling pursuant to a ruling request filed within 183 days after the beginning of the transfer). In 1984, the requirement to obtain a ruling was repealed.
See, e.g., for Germany, the Bundesfinanzausgleich [Federal equalization of finances] in art. 107 of the Constitution. The United States had so-called revenue-sharing provisions for a time, but they have been dropped. State and Local Fiscal Assistance Act, Pub. L. No. 92–512, Title I, 86 Stat. 919 (1972).
See GG art. 108 (DEU). On the basis of this article, regional tax authorities administer the personal and corporate income tax, the business tax (Gewerbesteuer), the VAT, and inheritance and gift taxes, as well as the toad tax. See 3 K. Tipke, supra note 53, at 1130.
Sixth Council Directive 77/388 of May 17, 1977, on the Harmonisation of Laws of Member States Relating to Turnover Tax—Common System of Value Added Tax Uniform Basis of Assessment, 1977 O.J. (L 145) 1; Second Council Directive 67/228 of Apr. 11, 1967, on the Harmonisation of Legislation of Member States Concerning Turnover Taxes—Structure and Procedures for Application of the Common System of Value Added Tax, 1967 O.J. (L 71) 1303; First Council Directive 67/227 of Apr, 11, 1967, on the Harmonisation of Legislation of Member States Concerning Turnover Taxes, 1967 O.J. (L 71) 1301.
See Case C-35/90, Commission v. Spain, 1991 E.C.R. 5073; Case C-31/89, Commission v. Spain, 1990 E.C.R. 2139; Case C-120/88, Commission v. Italy, 1991 E.C.R. 621; Case 50/87, Commission v. France, 1988 E.C.R. 4797; Case 122/87, Commission v. Italy, 1988 E.C.R. 2685; Case 249/84, Ministère Public and Ministry of Finance v. Profant, 1985 E.C.R. 3237.
See Council Directive 90/434 of July 23, 1990 on the Common System of Taxation Applicable to Mergers, Divisions, Transfers of Assets and Exchanges of Shares Concerning Companies of Different Member States, 1990 O.J. (L 225) 1; Council Directive 90/435 of July 23, 1990 on the Common System of Taxation Applicable in the Case of Parent Companies and Subsidiaries of Different Member States, 1990 O.J. (L 225) 6.
See generally, for Belgium, A. Alen, Treatise on Belgian Constitutional Law 256–62 (1992); for Switzerland, World Tax Series: Taxation in Switzerland, supra note 225, at 103–106; for the United States, 1 Jerome R. Hellerstein & Walter Hellerstein, State Taxation, chs. 4, 5 (2d ed. 1993).