S. 300E. Summary

1. Starting position

This study has committed itself to an analysis of the German, the UK and the EU GAAR in terms of their effectiveness to counteract tax avoidance when compared with the judicially developed methods against tax avoidance. Additionally, the study thereby also aimed to evaluate whether the GAARs are compatible with the principles of a sophisticated legal culture.

In order to draw the proper conclusions, this study has been structured in three main sections: Main section B paved the way for the further structure of the study as it analysed the judicially developed anti-avoidance doctrines separately. This is because in order to understand the GAARs, one must also understand the judicial approaches behind each GAAR. Main section C then dealt with the GAARs. A comparative investigation into the GAARs has revealed that their functioning and therewith, their requirements of application, can be can be disassembled into various requirements and tests. Eventually, main section D put the judicially developed approaches in relation with the statutory GAARs. In doing so, the reasons for introducing (or suggesting) GAARs became visible. Eventually, this allowed the author to draw a conclusion concerning the value of the GAARs as implemented by Germany and the UK and as suggested by the Commission.

The following sections summarise the main findings of this study.

2. The courts’ approach

Main result of section B: Courts have developed methods to counteract abuse on their own, i.e. by making the finding of abuse a matter of interpretation.

1. Similar starting situations but different developments in the UK and in Germany

The starting positions in the UK and Germany were surprisingly similar: After all, both the House of Lords as well as the Reich Fiscal Court interpreted statutes literally and decided that the form and the legal results of an arrangement are decisive. The House of Lords did so in the Duke of Westminster (1935) and the Reich Fiscal Court did so in Mitropa (1919).

S. 301The German legislator decided to enact a major tax reform in order to overcome the aftermaths of the First World War. A new tax system, following an economic concept of income, was introduced. In addition to the enactment of principle-based taxing statutes, the legislator also introduced in Sec 4 Reich Fiscal Code the concept of the economic perspective (“wirtschaftliche Betrachtungsweise”). This concept required courts and tax authorities to interpret tax legislation by taking into account the legislation’s purpose. Mainly for political reasons, also a GAAR was introduced with this reform in Sec 5 Reich Fiscal Code.

The UK legislator reacted differently to the courts’ approach of interpreting statutes literally. Instead of enacting principle-based statutes, the UK legislator decided to enact a mass of detailed rules, hoping to tie a net around the taxpayer from where it cannot escape. The UK tax system was also not equipped with a concept similar to the economic perspective and courts were therefore not required of having to interpret tax legislation by taking into account its purpose.

What followed were completely different developments, both with regard to the courts’ approach towards tax avoidance as well as with regard to the future framing of the tax system:

Different developments regarding the courts’ approach towards abuse: The German tax reform was successful. German courts did no longer interpret statutes literally nor did they attach too much weight on formalities. This evolution was to a large extent due to the broad and principle-based statutes and the concept of the economic perspective (“wirtschaftliche Betrachtungsweise”). Interpretation turned out to be the key to counteract abusive arrangements. Courts made use of the principle-based statutes, applied various methods of interpretation and were also not afraid from using the methods of analogy or teleological reduction to counteract abusive arrangements. The GAAR was almost not used at all by the Reich Fiscal Court in the first years of its existence. UK courts, in turn, were irresistibly interpreting statutes literally and clearly rejected to consider any underlying statutory purpose, let alone to add, amend or omit words in a statute.

Different developments regarding the future framing of the tax system: Unlike a typical German taxing statute, a UK taxing statute is still intentionally drafted in a rather tight way. The drafters aim to leave no conceivable ambiguity or doubt. Statutes in the UK are much more detailed, by comparison to the open-ended statutory texts on which German judges are nurtured. The foundation stones for S. 302these developments were thereby already laid in the early 20^th century: The German legislator did not have to articulate and operate a too complicated tax system as it could rely on the judiciary that used various methods of interpretation in conjunction with principle-based statutes. The UK legislator in turn, kept enacting ever more detailed provisions which did little more than make it more difficult (and therefore more lucrative) for tax professionals to find more gaps and loopholes.

2. Further development and the status quo in Germany

After the end of the Second World War, the Federal Fiscal Court’s jurisprudence in cases involving potential issues of tax avoidance was marked by a struggle between the relationship of civil law and tax law. Up to the middle of the 1960s, the court thereby reinforced the civil law point of view when interpreting terms used in taxing statutes. However, in the middle of the 1960s the pendulum swung back in the other direction and the courts re-adopted the concept of the economic perspective (“wirtschaftliche Betrachtungsweise”). This development was influenced by the jurisprudence of the Federal Constitutional, which attested that civil law and tax law are distinct fields of law and which held that a civil law term must not necessarily have the same meaning as in civil law when it is used in a taxing statute.

Eventually, scholars also started to critically examine the concept of the economic perspective (“wirtschaftliche Betrachtungsweise”). It is due to the work of scholars like Gassner that it has quickly been acknowledged that the concept of the economic perspective is merely a form of teleological interpretation. Accordingly, the economic perspective takes into account that tax law is primarily concerned with transactions happening in an economic environment.

In 1977, some legislative changes were made with respect to the economic perspective (“wirtschaftliche Betrachtungsweise”) and to the GAAR: Since the concept of the economic perspective manifested itself so firmly in the roots of the German tax system, Sec 1 (2) Fiscal Adjustment Act (which was the predecessor provision of Sec 4 Reich Fiscal Code) was considered to be superfluous and hence, deleted in 1977. Additionally, the GAAR found its way into Sec 42 Fiscal Code. Since then, the GAAR has been subject to two further changes made in 2001 and 2008. Nonetheless, these amendments did not change the courts’ (and tax authorities’) approach to deal with matters involving abuse in two alternative ways:

S. 303The Federal Fiscal Court started to use the GAAR more frequently in the 1970s. It is hereby noticeable that the Federal Fiscal Court developed two ways to deal with abuse. These approaches are in an interchangeable manner still applied today and have been described as Innentheorie and Außentheorie. In their core, both theories deal with the legal relevance of the GAAR:

Under the concept of the Innentheorie, the Federal Fiscal Court makes the finding of abuse a question of statutory construction. In doing so, the court interprets the respective provision sought to be exploited or circumvented in a methodologically correct way and assesses whether the facts answer the questions posed by the statute. Supporters of the Innentheorie therefore argue that the finding of abuse is nothing special and merely a matter of interpretation. Hence, it is brought forward that the GAAR (i.e. Sec 42 Fiscal Code) is superfluous. After all, judges already have the tools to counteract abuse on their own. Supporters of the Innentheorie therefore also agree on the idea that it is well within the competence of a judge to apply the mechanisms of analogy and teleological reduction. Broadly speaking, these methods of interpretation decide a conflict between the wording and the telos of a law in favour of the telos of the law. Accordingly, unintended legislative gaps, where the wording of a provision is lagging behind its apparent statutory purpose, must be closed via the mechanism of analogy. Statutes whose wording is unintentionally going beyond their purpose must be dealt with by the mechanism of teleological reduction.

Under the concept of the Außentheorie, the Federal Fiscal Court makes the finding of abuse conditional on a comparison of an inappropriate with an appropriate arrangement. The court’s approach thereby reveals that the Innentheorie and the Außentheorie are more than just two theories. This is because supporters of the Außentheorie initially argued that the GAAR is needed as a legal basis to apply the mechanism of analogy. However, the Federal Fiscal Court already held in the 1980s that there is no prohibition of analogy (also to the detriment of the taxpayer) in tax law. Nonetheless, the discussions dealing with the legal relevance of the GAAR did not cease to exist. This is because the GAAR developed a life of its own. After all, the Federal Fiscal Court uses Sec 42 Fiscal Code to strike down inappropriate arrangements. However, it does not do so by applying the mechanism of analogy but by comparing the actual arrangement with a fictitious arrangement. Accordingly, it is put forward by supporters of the Außentheorie that Sec 42 Fiscal Code is a constitutive anti-avoidance rule, having its own content and that it will apply in the case of inappropriate arrangements.

S. 3043. Further development and the status quo in the UK

When the Duke of Westminster was decided in 1935, the scene was set for a golden age of tax planning in the UK: Judges preferred form over substance and were interpreting statutes literally, irrespective of any underlying purpose. The legislator tried to do the impossible and to keep pace with that development. The inevitable consequence of that cat-and-mouse game was that taxing statutes got even more complex. With that, taxpayers were advised to enter into even more complex schemes. The legislator, in turn, continued with its efforts to plug these holes. The spiral was set in motion.

Ramsay, a case decided in 1981, marked the turning point. It involved another highly complex scheme bought by the taxpayer from the tax advisors’s shelve. Broadly speaking, the case involved self-cancelling and circular transactions. The taxpayer wanted one asset to increas in value at the expense of a decrease in value of the other asset and that one could not occur without the other. The taxpayer hoped that the value of the increasing asset was tax free whereas the value of the decreasing asset was deductible. This time, the House of Lords did not find the legal results of the arrangements to be binding. Instead, the judges limited the impacts of the Duke of Westminster by considering that the series of transactions may be looked at together. In doing so, they regarded the scheme as a whole and held that the taxpayer neither incurred a gain nor a loss.

What followed was a period of uncertainty. This was due to the fact that the nature of Ramsay was initially not clear. Was it a judge-made anti-avoidance rule, superimposed over the whole revenue law? Or was it merely a principle of statutory interpretation?

Initially, it seemed as if the Law Lords developed Ramsay into a judicially developed GAAR (comparable with the German approach of the Außentheorie). This is because the judges, in cases like Furniss v Dawson, created a formula with the help of which they were able to strike down abusive arrangements. This led to great uncertainty within the business community that all of a sudden could no longer rely on the sacred principles established with the Duke of Westminster.

As it turned out, these concerns were unfounded. This is because Ramsay, as some kind of judicial anti-avoidance doctrine, quickly came to an end. Craven v White, a case decided in 1988, showed that taxpayers could avoid getting caught by the application of the formula as developed in earlier cases by better planning their transactions.

S. 305Eventually, the House of Lords realised that Ramsay is nothing but a principle of statutory construction. Although the idea of purposive interpretation was not entirely new, it was only in cases such as McGuckian (1997) that the House of Lords explicitly started to make the finding of abuse a matter of interpretation. However, the Law Lords’ approach was not yet flawless. This only changed when the House of Lords gave decided cases such as MacNiven v Westmoreland and BMBF. Today it is undisputed that Ramsay is a principle of statutory construction and that judges may counteract abuse using methods of statutory construction. Accordingly, the judges’ approach towards tax avoidance is comparable with the German understanding of the Innentheorie.

Viewing Ramsay as a principle of statutory construction seemed to have opened the door to the real issue, which is how statutes should be interpreted: This is because some judges were allegedly no longer able to discern a purpose from highly mechanistic tax legislation. This led to cases such as Mayes (SHIPS 2) were the judges decided in favour of the taxpayer, despite there being highly artificial and abnormal arrangements. However, there have also been cases where the courts found ways to counteract schemes that build on highly mechanistic tax legislation. The judges have thereby shown that they apply methods which a civil law lawyer would refer to as historical or contextual interpretation. Even more, the judges were also willing to add, amend or omit words in a statute (i.e. they were willing to apply the mechanisms of analogy or teleological reduction).

4. The ECJ’s approach

Abuse is a latecomer in the EU. It was only in 1974 that the ECJ, for the first time, referred to abusive practices. Initially, the Court left the impression that abuse only had a limited role to play in the field of the Internal Market. This has been explained by the fact that a development of a doctrine of abuse of the freedoms seemed to serve little purpose in a period of integration-building which sought to facilitate the completion of the Internal Market.

However, as the EU legal order developed, a doctrine of abuse began to evolve. Apparently, defendant governments’ resorted to the doctrine of abuse as a countervailing force towards parties which sought to stretch to the maximum the scope of the freedoms. Additionally, EU legislation started to grow in the field of secondary union law, hence opening up opportunities for the improper use of EU S. 306benefits. It has also been explained that the increase in the volume of free movement within the internal market and the 2004 enlargement contributed to the gaining significance of the concept of abuse of law in the EU.

Viewed retrospectively, the ECJ’s approach in matters involving potential abusive behaviour can be described in several stages: (i) the pre-Emsland Stärke cases (i.e. prior to December 2000), (ii) the development of an abuse test in Emsland Stärke, (iii) the post-Emsland Stärke cases until Cadbury Schweppes and Halifax (i.e. December 2000 – 2006), (iv) the adoption of the abuse test in the field of taxation and (v) the cases decided after Cadbury Schweppes and Halifax.

Initially, U-turn transactions were at the heart of the ECJ’s case law involving abusive practices. These cases concerned situations where persons, services (and capital) or goods moved from one Member State to another Member State although the final destination of the person, service or good was the Member State of origin. The Court’s approach was thereby consequent:

Within the non-harmonised areas of law, the compatibility of domestic (anti-avoidance) rules with primary law was at issue. It is thereby noticeable that the Court ruled out the violation of a freedom for either one of the following two reasons, to be assessed subsequently: In a first step, the Court assessed whether an arrangement falls within the scope of a fundamental freedom. If the (allegedly abusive) arrangement does not fall within the scope of the respective freedom, it did not enjoy any protection. Where there is no ‘institutional abuse’, the Court made the matter one of balancing interests. Hereby the Court examined whether the domestic (anti-avoidance) provision can be justified by overriding reasons of safeguarding the national interest at stake when balanced against the internal market objective. Within the area of harmonised law, the Court made the finding of abuse a matter of interpretation. In doing so, the Court ascertained whether or not a potential abusive arrangement falls within the legislative union law framework, properly interpreted. This approach is then comparable with the Innentheorie and with Ramsay as a principle of statutory construction.

The ECJ set up an abuse test in Emsland Stärke in the year 2000. This abuse test made the finding of abuse conditional on the fulfilment of an objective test and a subjective test. The objective test thereby required to ascertain whether the facts are covered by the rules sought to be applied. The subjective test was aimed at ascertaining whether the person had the intention to obtain an advantage.

S. 307The adoption of the abuse test was unnecessary. The Court could have easily done what it already did in its earlier cases and could have made the finding of abuse a matter of interpretation. After all, also the facts in Emsland Stärke reveal that the persons were involved in a U-turn transaction that was not in line with the provision sought to be exploited.

The abuse test had no influence in the short run. It seemed as if the impacts of Emsland Stärke were only limited to the field of agricultural levies. This is because the cases which were decided by the ECJ after Emsland Stärke, followed the rationale of the cases which were decided before Emsland Stärke. Accordingly, the Court made the finding of abuse either a matter of balancing the interests or one of interpretation.

However, in 2006, the ECJ also adopted the abuse test in the harmonised area of indirect taxation. In Halifax, the Court held that an abusive practice exists where an objective and a subjective test is fulfilled. However, there was also no need to adopt the abuse test in Halifax. The Court itself acknowledged that the arrangements in question are not in line with the principle of fiscal neutrality, as pursued by the Sixth Directive.Halifax has been followed by Part Service and Weald Leasing, two cases where the ECJ also made the finding of abuse conditional on the fulfilment of the two-prong abuse test.

It was also in 2006 that the ECJ extended the concept of abuse to the non-harmonised area of direct taxation. Cadbury Schweppes concerned the question whether UK CFC rules are compatible with the freedom of establishment. The Court thereby also referred to Emsland Stärke. However, the Court did not apply the subjective test and instead concentrated on the objective test. In doing so, the Court for the first time gave a definition of the term “wholly artificial arrangement”. Eventually, the Court’s understanding of a wholly artificial arrangement was the result of an interpretation of the freedom of establishment. This is why it has been put forward that the Court could have neglected the applicability of the freedom of establishment from the outset. However, since the Court made the issue one of balancing interests, it eventually reduced the scope for Member States to justify their measures to matters of institutional abuse.

After Cadbury Schweppes, the Court refrained from further elaborating on the definition of the expression wholly artificial arrangement. Instead, it reconciled what it did in the cases decided before Cadbury Schweppes and immediately S. 308shifted the emphasis towards the proportionality of a restriction. Accordingly, instead of defining a wholly artificial arrangement positively, the Court concentrated on saying when a measure is not only confined at targeting only wholly artificial arrangements. The cases which were to follow reveal that the Court establishes a rather high threshold for a measure to be a proportional restriction of the respective freedom.

After Halifax (and Part Service and Weald Leasing), the ECJ also changed its focus to wholly artificial arrangement in the field of harmonised VAT law. Although the Court still aims to identify abusive practices, it no longer focuses on the two-prong abuse test but on the analysis of the arrangements. Hereby it is noticeable that the Court’s case law can be interpreted in such a way that the Court made the finding of an abusive practice conditional on a (wholly artificial) arrangement defeating the statutory requirements of the respective provision.

5. Conclusion: Courts found ways to deal with matters of abuse on their own

Overall, it can be concluded that each court found ways to counteract abuse on its own, i.e. by making the finding of abuse a question of interpretation: The Federal Fiscal Court does so along the concept of the Innentheorie. The UKSC does so since it has been acknowledged that Ramsay is a principle of statutory construction. More recent cases decided by the ECJ in the field of harmonised VAT suggest that the Court is back on the track of making the finding of abuse a question of interpretation.

3. The GAARs’ approach

Main result of section C: GAARs are self-defeating, not more effective than the judicially developed methods against tax avoidance and detrimental to the legal culture

1. Common requirements for the application of a GAAR

The Federal Fiscal Court’s jurisprudence along the Außentheorie and the GAARs as implemented in the UK and as suggested by the EU, have much in common. The finding of abuse is made conditional on the fulfilment of several requirements:

S. 309A first requirement for the application of the GAAR is that the taxpayer’s conduct must manifest itself in an arrangement (like a transaction or a scheme). The finding of abuse is additionally subject to the requirement that the arrangement deals with a tax that falls within the scope of the GAAR (like income tax). These two requirements set a rather low threshold.

At the core of each GAAR stands the issue of having to identify an arrangement that is“abusive”. Each GAAR thereby acknowledges that only certain arrangements will be “abusive”. This implies that the application of the GAAR requires establishing a borderline between “abusive” and “non-abusive” arrangements.

In order to draw such a line, GAARs (respectively the courts applying the GAAR) recognise that the finding of an “abusive” arrangement first of all requires the finding of an appropriate benchmark against which the arrangement can be evaluated. An analysis revealed that GAARs can develop a life of their own and that this line can therefore be drawn in two ways:

Theoretically, the borderline between abusive and non-abusive arrangements should only be drawn by the objective test. This test has the avowed purpose of distinguishing “abusive” from “non-abusive” arrangements. Properly interpreted, each GAAR therefore demands that the benchmark against which the (potentially abusive) arrangement needs to be evaluated can only be the respective taxing statute sought to be circumvented or exploited. An arrangement will be “abusive” where it defeats the statutory requirements of the respective taxing statute.

The cases decided by the Federal Fiscal Court along the Außentheorie reveal that there is the danger that GAARs might in practice develop a life of their own. When the Federal Fiscal Court invokes Sec 42 Fiscal Code, it typically does not make the finding of abuse conditional on an objective test. Rather, the court makes the finding of abuse conditional on a subjective arrangement test. The court thereby draws the borderline between “abusive” and “non-abusive” arrangements by pursuing the idea that “abusive” (i.e. inappropriate) arrangements show some specific characteristics which can be used to distinguish them from “non-abusive” (i.e. appropriate) arrangements. Accordingly, the court does not ascertain whether an arrangement falls within the scope of a provision, properly interpreted.

S. 310Eventually, the Federal Fiscal Court’s approach along the concept of the Außentheorie reveals a significant difference when compared with the subjective tests (consisting of an arrangement test, a purpose test and a double-reasonableness test and the requirement of a tax advantage) as required by the UK and EU GAAR:

The EU GAAR also requires an arrangement test. This arrangement test focuses on the form of the transaction and will typically be fulfilled where the arrangement is “artificial”. An “artificial” arrangement for purposes of the EU GAAR will thereby often also be “inappropriate” for purposes of the jurisprudence of the Federal Fiscal Court along the Außentheorie. After all, both tests focus on the form, effects or nature of the arrangement(s). This is where the difference as described above emerges: While the Federal Fiscal Court employs the arrangement test instead of the objective test, the EU GAAR applies the arrangement test in addition to the objective test. The effects of an arrangement are even less important for purposes of the UK GAAR, which considers them to be only indicators of abuse.

The other subjective tests of the UK and EU GAAR also need to be applied inaddition to the objective test: A purpose test thereby asks whether the arrangement had the intention to obtain a tax benefit. A double-reasonableness test asks whether there can be a reasonable held view that also a reasonable person would have carried out the arrangement.

Last but not least, each GAAR also connects the finding of abuse with the presence of some kind of tax advantage. Closely connected with the identification of such an advantage are the legal consequences of applying the GAAR. After all, GAARs want to restore the situation that would have prevailed, had the arrangement not been considered to be abusive.

2. GAARs are self-defeating and not more effective than the judicially developed methods against tax avoidance

The idea of GAARs is self-defeating: Each GAAR requires in the course of the objective test that the finding of abuse is made conditional on an arrangement defeating the statutory requirements of the provision sought to be circumvented or exploited. However, aGAAR is not needed to conclude that an arrangement is ‘abusive’ (or whatever one wants to label the result of interpretation) where it fails to be in line with the statutory requirements of the relevant provision sought to be applied by the taxpayer.

S. 311The objective tests are only codifications of something that the courts already didwithout the ‘benefit’ of a GAAR: they made the finding of abuse a matter of statutory construction.

The remaining tests and requirementsare unnecessary: An arrangement does not more or less defeat the statutory requirements of a provision where it is ”artificial” or where it has been implemented with the “purpose” of generating a “tax advantage” or where “reasonable” taxpayers would not have carried out the arrangements.

‘Subjectivity’ may only play a role at the level of the facts: Taxing statutes (typically) do not distinguish between taxpayers acting in good or bad faith. Crucially, the subjective non-tax reasons need to be taken into account at the level of the facts. Overall, it boils down to a basic question of legal methodology: The court needs to find the applicable provision and needs to identify the relevant facts. It needs to have clarity concerning the facts of the case to which the legal provision is to be applied. Furthemore, the respective legal provision, properly interpreted, will give the answer whether the non-tax reasons are relevant or whether they have nothing to do with the statutory requirements of the respective provision sought to be circumvented or exploited.

3. GAARs are dangerous and detrimental to the legal culture

In addition to the objective test, GAARs would also require the presence of some subjective tests and the presence of a tax advantage. These requirements are unnecessary, meaningless, and dangerous and, as a consequence, detrimental to the legal culture:

Requiring subjective tests is unnecessary: An arrangement does not more or less defeat the statutory requirements of a provision where it is ”artificial” or where it has been implemented with the “purpose” of generating a “tax advantage” or where “reasonable” taxpayers would not have carried out the arrangements.

Requiring subjective tests is meaningless: The subjective tests are so vague so that they could always be fulfilled. After all, tax authorities or courts may always find ways to argue that an arrangement was “artificial”. Similarly, a taxpayer will have no chance of fending off the accusation of having entered an arrangement S. 312for the (main) “purpose” of achieving a “tax advantage”. Last but not least, also the application of the double-reasonableness test is only made conditional on having to ascertain whether a “reasonable” taxpayer would have carried out the same arrangements.

Requiring subjective tests is detrimental to the legal culture: It must be borne in mind that requirements and tests, which come into play in addition to the objective test (and hence, in a scenario where an arrangement already defeated the statutory requirements), give tax authorities, courts and taxpayers alike great power:

Subjective tests entail the risk that taxpayers may refute the existence of abuse, despite their arrangements defeating the statutory requirements: Taxpayers may bring forward that their circular or self-cancelling arrangements were not “artificial” on the grounds that also sale-and-lease back transactions or repo-transactions are often considered to be ‘genuine’ transactions. Additionally, they might refer to some established practice and might argue that also “reasonable” third parties would have entered into the arrangements. Taxpayers might also argue that they did not enter into an arrangement for the (main or essential) “purpose” of obtaining a tax advantage. They will put forward that the tax advantage is merely a side effect of the arrangement and they will underpin this claim by creatively presenting non-tax reasons or economic reasons, hoping to convince the court of their decent conduct. They might also claim that the arrangement does not result in a “tax advantage” in the sense of the GAAR or that the advantages are outweighed by potential disadvantages. Eventually, motives will never be completely transparent to others and those taxpayers who receive good advice will be better off because they can more easily present their case in an acceptable way.

Subjective tests also entail the risk that tax authorities or courts will themselves take advantage of the margin of discretion they enjoy under such tests and requirements: Tax authorities or courts might be tempted to focus too much on the “artificiality” of an arrangement or on the “purpose” for carrying out an arrangement. Especially in the light of tax provisions which are difficult to interpret, they might be tempted to take an easier route and not to pay attention to the statutory requirements. Tax authorities or courts might then simply conclude that an arrangement defeats the statutory requirements just because the taxpayer was said to have had the intention to obtain a tax benefit or because reasonable third parties would have acted differently.

S. 313Subjective tests also entail the risk that tax authorities or courts could ‘pardon’ a taxpayer: This could happen in situations where the authorities or courts do not consider that the taxpayer entered into an arrangement (which defeats the statutory requirements) with the purpose of obtaining a tax advantage. Alternatively, they might look at some established practice and might conclude that the arrangement is not abusive because also reasonable third parties would have acted in such a way. They might also consider that the tax advantage has been or might be outweighed by a disadvantage.

Last but not least, it needs to be considered that GAARs also address the legal consequences. As a result of finding an abusive arrangement, GAARs will thereby either ignore the results of the abusive arrangement and will base taxation on the idea that the taxpayer would not have carried out the arrangement at all. GAARs additionally also allow for the taxation of a fictional arrangement. In this context, the legal consequence of applying the GAAR lies in the taxation of a hypothetical arrangement that the taxpayer would have carried out instead of the actual abusive arrangement. This approach will inevitably cause problems as one can be never certain about what the taxpayer would have done if it had known that the tax authority or the court was not going to grant the tax advantage. Would it have done nothing? Would it have arranged its affairs in such a way that it would receive a treatment closer to the one it tried to achieve? If there are several different options, which one would it have chosen?

4. Valuation of the GAARs

Main result of section D: GAARs are introduced to compensate (i) legislative cowardice and (ii) judicial and administrative shortcomings in statutory interpretation and should at best be seen as reminders of having to interpret the law

1. The reasons for and consequences of introducing GAARs:

Courts already developed methods to counteract abuse on their own, i.e. by making the issue one of statutory construction. This is why the idea of GAARs is self-defeating: After all, GAARs can only be applied where an arrangement defeats the statutory requirements of the provision sought to be exploited or circumvented. This is where the ‘true’ reasons for introducing GAARs emerge:

GAARs are introduced for political reasons: Both the UK and the EU GAAR were introduced or suggested in times of a severe financial crisis and in the enviS. 314ronment of a BEPS-world. Accordingly, the GAARs simply formed part of the politicians’ reaction to the rising public anger at a steady flow of press revelations about aggressive tax planning schemes used by certain businesses and individuals. Also the German GAAR was implemented into the major tax reform mainly for political reasons.

GAARs are introduced to compensate legislative cowardice and to compensate administrative and judicial shortcomings in statutory interpretation: German judges never had to deal with such ‘bizarre’ tax schemes as their colleagues from the UK. This is largely due to the fact that the German tax system never provided taxpayers with the possibilities to tax planning/avoidance that are opened up for taxpayers in the UK. Accordingly, it can safely be said that the breeding ground for tax avoidance lies within the depths of a country’s tax system. An increasing complexity of the tax system typically also leads to an increasing number of tax planning schemes.

Lawyers from both Germany and the UK seem to agree on the idea that overly complex tax systems are not desirable. It is thereby interesting to note that many scholars from the UK are therefore pushing for the enactment of broader and more principle-based statutes. German courts additionally even admit that statutes cannot and do need not be ‘perfect’.

Accordingly, the complexity of a tax system also correlates with the legislator’s confidence in the judiciary to handle the respective tax system they have to operate in. Historically, it can therefore be explained why the UK legislator tended to draft its taxing statutes in a very detailed manner: After all, the judiciary interpreted statutes only literally and did not pay attention to any underlying purpose.

However, in the meantime the judiciary developed Ramsay into a principle of statutory construction and acknowledged that the finding of abuse is a matter of statutory construction. The approach of the judiciary would therefore enable the legislator to steer away from highly detailed provisions. This is crucial as tax avoidance can only be effectively dealt with if both the judiciary and the legislator find the rightbalance: Where legal texts are poorly drafted, even the ‘best’ methods of interpretation might not suffice to counteract abuse by way of interpretation. Hence, it would actually be for the legislator take the next step and to simplify the tax system.

S. 315However, instead of implementing reforms and instead of enacting less mechanistic tax laws, legislators might introduce GAARs. However, the introduction of a GAAR misses the point and is on top of that a futile attempt if one considers that GAARs will only work if the underlying law is based on a clearly stated principle, because without such a principle or objective it is impossible to decide whether there has been abuse of the legislation.

The enactment of a GAAR in a tax system with statutes than can hardly be interpreted might in the worst case even be considered as an invitation addressed by the legislator to the courts and tax authorities to develop the GAAR into something comparable to the Federal Fiscal Court’s approach along the Außentheorie. Accordingly, this unfortunate result could then be traced back to the legislator not willing to address the real problem of an overly complex tax system and to the authorities and judiciary not willing (and needing to) further develop the methods of statutory construction.

2. GAARs as reminders of having to interpret the law

GAARs may only be helpful where they are seen as reminders of having to interpret the law. The decisive question will then be to ascertainhow statutes can be properly interpreted:

In Germany, for example, it is settled that judges may apply various methods of interpretation and the mechanisms of analogy or teleological reduction, even to the detriment of the taxpayer. Accordingly, supporters of the Innentheorie rightly argue that Sec 42 Fiscal Code is not needed, not even as a reminder of having to interpret the law.

However, the relevance of a GAAR might be slightly different in other jurisdictions. In the UK, for example, judges only officially refrained from interpreting taxing statutes literally in the 1990s. Cases like Mayes (SHIPS 2) additionally revealed that judges might not (yet) be fully aware of the various methods of interpretation available to them: Although the results were said to have been unintended by the legislator, the judges still claimed to not have found any underlying purpose and they did not consider to follow the rationale of different cases, where different judges already omitted or amended words in a statute (i.e. applied, in the terminology of a civil law lawyer, the mechanisms of analogy or teleological reduction). Accordingly, the GAAR might serve as a reminder for judges to take into account various methods of interpretation.

S. 316If the importance of GAARs is reduced to the requirement to remind judges of having to interpret the law properly, then this is also where the circle closes: Again, a GAAR will not work unless the underlying law is based on a clearly stated principle. Accordingly, in order for the judges to be able to properly interpret taxing statutes and consequently, in order for the GAAR to be useful as a reminder of having to interpret the law, the legislator has to have the confidence to enact statutes that the judges can actually work with.

3. Outlook

This study asked in the very beginning that comparative research needs to be carried out so that one country might learn from the experiences made in another country with respect to the same or similar problems. This is all the more important under consideration that the EU GAAR is merely a recommendation and needs to be implemented by national legislators.

The finding of abuse should only be a matter of interpretation: It seems to the author that German courts (as well as other courts which have to deal with a GAAR patterned after the EU GAAR) may learn from the concept of the Innentheorie and the approach pursued by the courts in the UK, that the finding of abuse should only be a matter of interpretation. The House of Lords already acknowledged that Ramsay, as something that goes beyond statutory construction, cannot and should not work and is detrimental to the legal culture. Accordingly, also other courts may learn from these experiences and should acknowledge that the finding of abuse should not be based along a Ramsay doctrine or the Außentheorie (i.e. by assuming that ‘abusive’ arrangements show some specific characteristics which can be used to distinguish them from ‘genuine’ transactions).

There are various methods of interpretation: Judges from the UK (and judges from other jurisdictions which have to deal with a GAAR patterned after the EU GAAR) may learn a lot from the experiences made by the ECJ and German courts with respect to the methods of statutory construction. Judges sitting at the ECJ or at German courts tend to appreciate that the methods of interpretation are not limited to the purposive construction of single words used in a statute. Rather, they would typically apply a more nuanced approach and would distinguish literal from historical, contextual and teleological means of interpretation. Especially German judges also do not tend to be afraid from applying the mechanisms of analogy and teleological reduction (even to the detriment of the taxpayer). It is thereby also crucial to note that also judges from the UK may in the meantime fall back on these methods of interpretation. However, at least in the case of the UK, these methods are not yet wide spread among the judges. Accordingly, the UK (as well as other countries) may learn from the positive experiences made in Germany with respect to a broad range of methods of interpretation.

S. 317Statutes do not have to be “perfect”: Legislators may also learn from the experiences made with respect to the German and UK drafting style of the tax legislation. Viewed in isolation, it is not surprising that both German and UK scholars have not become tired from criticising the complexity of the respective tax system they have to operate in. However, the comparison of the taxing statutes of these two tax systems revealed that the complexity of a tax system is all a matter of perspective. Compared with the complexity of the UK’s tax system, the German tax system might still be regarded as an exemplary model. This is because the German tax system never opened up to taxpayers such possibilities to tax planning and tax avoidance as they have been (and still are) available in the UK. Additionally, the principle-based nature of a typical German taxing statute also contributed to the tax system not getting overly complex. Accordingly, other countries may heed the fact that tax systems do not have to be (nor can they be) ‘perfect’. It can be learned from the UK’s experience that complex rules are not the solution but actually the source of tax avoidance schemes. Hence, legislators may enact principle-based statutes which the judiciary can actually work with. Apart from their potential function of reminding courts of having to properly interpret the legislation, there is no room for GAARs in this setting.