10.2 Summary

This study contains the results of research with respect to the international aspects of the Dutch substantial interest rules, as laid down in chapter 4 and part 7.3 of the Individual Income Tax Act 2001. The research concerns the rules with respect to emigration, immigration and remigration of the holder of a substantial interest, as well as the tax liability of non-resident taxpayers in connection with the residence fiction and the seat-transfer levy. The emphasis has been put on the two exit levies and the residence fiction relating to non-resident taxpayers. The purpose of this study is in the first place to provide insight into the background and functioning of the rules in national law, as well as to analyze the influence of tax treaties, the Tax Arrangement for the Kingdom and EC law on actual application of the national rules in cross-border situations. Furthermore, it has been reviewed how the national rules can be made more consistent, how the national tax jurisdiction can be preserved to the extent possible under tax treaties and the Tax Arrangement for the Kingdom, and which solutions are conceivable in the EU-context with respect to the topic of exit levies.

In chapter 2 the background of the residence fiction, exit levies and liability of non-resident taxpayers with respect to substantial interest is described. Both the residence fiction for non-resident taxpayers (article 7.5, paragraph 6, Individual Income Tax Act 2001) and the levies in case of emigration of the substantial interest holder (emigration levy) and in case of relocation of the effective management of the company from the Netherlands (seat-transfer levy) were introduced to retain the substantial interest claim in case the substantial interest holder moves his residence abroad and the company moves with him. The residence fiction is only meaningful for non-resident taxpayers and causes a company that was resident in the Netherlands for at least five years, to be deemed to be a resident after the relocation of the effective management from the Netherlands for ten years subsequent to such relocation. As a result of the seat-transfer levy, the non-resident substantial interest holder must pay tax in respect of the accretion in value of the substantial interest arising until the moment of the seat-transfer. The emigration levy results in payment in respect of the value accretion that has occurred until the moment of emigration of the substantial interest holder and relates therefore to the tax liability as a resident taxpayer. Both exit levies are not immediately collected, but result in a conservatory assessment with extension of payment. Finally there is also a residence fiction incorporated in article 4.35 Individual Income Tax Act 2001, which is particularly relevant for the determination of the acquisition price of the substantial interest in case of immigration or remigration of the substantial interest holder. This residence fiction purports to deny to immigrants and returned emigrants a step up in case of previous liability as a non-resident taxpayer based on the residence fiction of article 7.5, paragraph 6, Individual Income Tax Act 2001. The residence fictions have not been coordinated, but as a result of an arrangement in the Regulation Income Tax 2001 the acquisition price in case of immigration and remigration will in principle be determined at the proper amount. The exit levies have been introduced, because the residence fiction for non-resident taxpayers appeared to be inadequate as a result of application of tax treaties. The goal that the value accretion originating in the Netherlands is ultimately taxed by the Netherlands is not always accomplished as a result of the concrete application of the exit levies. This is in particular caused by the various possibilities to cancel the tax liability. Furthermore the relationship between the residence fiction for non-resident taxpayers and both exit charges is problematic. As a result of the levy in case of emigration and/or seat-transfer the Dutch value accretion is already taxed when the residence fiction becomes applicable. Subsequent to emigration and/or seat-transfer the acquisition price is determined at the fair market value, as a result of which the levy based on the residence fiction relates to the subsequent value accretion during the foreign period only. Finally, at the end of chapter 2 questions have been raised with respect to the liability as foreign taxpayer for the substantial interest as such, because both in the substantial interest rules applicable since 1979 and the Dutch treaty policy, the focus is on the levy in respect of the value accrued in the period in which the substantial interest holder was resident of the Netherlands.

Chapter 3 deals with bilateral treaty situations. In particular the focus has been on the situation in which both the substantial interest holder and the company have moved to one treaty state and on liability as non-resident taxpayer as a result of the residence fiction. This chapter contains in the first place a discussion of article 13, paragraph 5, Dutch Standard Treaty and the application in tax treaty situations of the various alienation fictions. Furthermore a complete overview has been incorporated of the substantial interest carve outs in the Dutch tax treaties and the conditions incorporated therein. In that overview ample attention has been paid to the ‘new style‘ substantial interest carve out, in the form in which the Netherlands tries to realize that in tax treaties negotiated as of 1997. It is determined that the residence fiction is effective in tax treaties with a substantial interest carve out that, in accordance with the Dutch Standard Treaty, requires that the company is a resident of the Netherlands ‘under Netherlands law‘. This is the case for 31 treaties, of which 11 have been concluded in or subsequent to 1997. In 12 treaties with a ‘new style‘ substantial interest carve out the residence fiction is relevant as well, but the levy as a result of liability as a foreign taxpayer is not possible. The ‘new style‘ substantial interest carve outs in fact relate to the application of the emigration levy only. An important goal in this respect is apparently to prevent that the new resident state levies tax in respect of the value accretion contained in the conservatory assessment, and thus to prevent that the Netherlands would have to forego the tax embodied in the conservatory assessment. Under the third important group of treaties, in which the substantial interest carve out requires that the company is a resident of the Netherlands under the residence article (27 treaties and the Tax Arrangement for the Kingdom), the residence fiction is generally not effective. The company can, if it was incorporated under Dutch law, be regarded as resident according to article 4, paragraph 1, OECD-Model Convention, but the tie-break provision of the third paragraph of that article determines that the other treaty state is the state of residence. It is concluded that the Netherlands has safeguarded its substantial interest claim in an ample majority of treaties; either because the tax liability as foreign taxpayer with the residence fiction is effective, or because the unrestricted application of the emigration levy has been provided for. With respect to substantial interest dividends the residence fiction is ineffective. The dividend article allows the Netherlands a limited right to levy only if the company within the meaning of the residence article is a resident of the Netherlands, and as a result of the tie-breaker such is not the case. From the case HR 28 February 2001, BNB 2001/295 (discussed in chapter 4), it must be concluded that it is article 10 that prevents the Dutch levy in such a case. Five recent, but less important treaties contain a special provision that allows the Netherlands, under identical conditions as with respect to the ‘new style‘ substantial interest carve out, the possibility to tax the substantial interest dividend up to ten years following emigration.

In chapter 4 triangular cases have been addressed, in which the substantial interest holder emigrated to State C, and the effective management of the company was moved to State B. As in bilateral treaty situations the residence fiction is effective in relation to substantial interest carve outs that, in accordance with the Dutch Standard Treaty, require that the company is resident of the Netherlands under Netherlands law. That is also true with respect to the ‘new style‘ substantial interest carve out, but only with respect to the application of the emigration levy. More than a third of the treaties, however, contain a substantial interest carve out that requires that the company is a resident of the Netherlands under the rules of the residence article. The chapter contains a thorough discussion of article 4, paragraph 1, OECD-Model Convention and the case BNB 2001/295. Only the residence fiction of article 2, paragraph 4, Corporate Income Tax Act 1969 can result in the tax liability required by article 4, paragraph 1, of the OECD-Model Convention, as a result of which application of tax liability as a non-resident taxpayer with respect to a substantial interest can incur only with respect to a company incorporated under Dutch law. If the company has its residence in State B and the treaty between the Netherlands and State B contains an OECD-Model Convention type residence article, the company cannot be regarded as resident of the Netherlands for the application of OECD-Model Convention type treaties with third states, in this case for application of the treaty Netherlands - State C. The company is on the basis of the operative provisions in the treaty Netherlands – State B no longer ‘fully liable to tax‘ in the Netherlands for the application of the residence article in the treaty Netherlands – State C. This result follows from BNB 2001/295, is also applicable for treaties without the second sentence, and can be regarded as being in accordance with the text of and the commentary to article 4 OECD-Model Convention. As possible exception I see the case in which a company without a real enterprise (not being a fund for common account) and a treaty Netherlands – State B without OECD-Model Convention type other income article. The Netherlands will in particular be able to tax the alienation gains in case the treaty with State B or State C has a residence article that deviates from the OECD-Model Convention. Also with respect to the right to levy in respect of substantial interest dividend, the case BNB 2001/295 plays an important role. In the first place almost all treaties require for the right of the source state to tax in respect of dividend that the company is resident of the source state in the meaning of the residence article of the treaty. As a result of the ‘full tax liability‘-decision of the Hoge Raad, that is generally not the case in triangular cases. In the second place the Supreme Court also decided in this case that in such a situation the Dutch levy is not prevented by the other income article, but by the dividend article itself. This article, in the vision of the Hoge Raad, has other income effect. I find this decision doubtful, but the result is acceptable. Levy in respect of substantial interest dividend in the relation to State C is then in particular possible in case the treaty with State C or State B has a deviating residence article, or in case the treaty with State C has a special provision for substantial interest dividend. Except for the possibility that the treaty Netherlands – State B contains a special residence article, the Dutch levy will in almost all cases be prevented by the prohibition of extraterritorial levy in respect of dividends in that treaty in accordance with article 10, paragraph 5, OECD-Model Convention.

In chapter 5 attention has been paid to bilateral and trilateral situations in which the Tax Arrangement for the Kingdom plays a role. The Tax Arrangement for the Kingdom contains a deviating capital gains article and also a number of interesting antiabuse provisions. The substantial interest carve out requires, inter alia, that the company is resident of the Netherlands within the meaning of the Tax Arrangement for the Kingdom, which implies that levy in respect of alienation gains can only occur if the company is incorporated under Dutch law. Dutch residence is in bilateral situations in which both the substantial interest holder and the companies are in the Netherlands Antilles not possible, because the tie-breaker points to the place of effective management. The capital gains article is then entirely not applicable; it is the other income article that prevents the Dutch levy. In the triangular case with the company in State B, the company is already not resident of the Netherlands as a re-sult of the lack of ‘full tax liability‘ in the Netherlands, also required by the residence article in the Tax Arrangement for the Kingdom. With respect to substantial interest dividend, the Netherlands cannot levy dividend tax in the mentioned bilateral and trilateral situation, because the company is not resident in the Netherlands. The residence fiction is thus neither effective under the substantial interest carve out nor under the dividend article. However, the Tax Arrangement for the Kingdom does contain a special article with complicated anti-abuse provisions that should give the Netherlands the possibility to tax substantial interest income, in particular in cases in which the substantial interest holder or the company is subject to a special regime. Remarkable is that the Dutch levy based on the liability as a foreign taxpayer, including the residence fiction, can be applied without restriction. Application of these provisions, however, seems illusory, while they also largely lost their value as a result of the emigration levy. With respect to the levy of dividend tax the Tax Arrangement for the Kingdom contains a special anti-abuse provision that withdraws the tie-breaker, if the predominant reason for the move of the effective management was to prevent the levy of dividend tax. The result of application of this provision is that the Netherlands can effectuate the right to levy pursuant to the dividend article in a bilateral situation or a trilateral situation within the Kingdom.

In chapter 6 the emigration levy and the seat-transfer levy have been tested against tax treaties. The emigration levy can be applied without a problem under treaties with a ‘new style‘ substantial interest carve out if the company is (deemed) resident in the Netherlands, as well as pursuant the other tax treaties entered into as of 1997. From HR 24 October 2003, BNB 2004/257, it follows that the emigration levy is not problematic either with respect to the Tax Arrangement for the Kingdom. With respect to the other tax treaties, the question is whether the emigration levy is contrary to good faith. As is known, the Hoge Raad has prevented the application of various Dutch fictions in the fictitious wage– and the pension emigration cases, because there was a unilateral posterior change of national law, resulting in a change of the right to levy based on the nature of the income in the applicable treaty. With respect to the emigration levy, however, in the first place it can be mentioned that article 13, paragraph 5, OECD-Model Convention does not prevent the application of an exit charge at the border in respect of the value accrued until that time of the substantial interest. The OECD-Model Convention allows a value accretion tax, and the substantial interest levy can be regarded as a combination of a capital accretion tax and the liquidity principle, in which a charge in case of migration is necessary. In the second place the emigration levy is different from the provisions tested in the fictitious wage- and pension emigration cases. For example, the emigration levy relates to a real advantage that by its nature is covered by the capital gains article, as a result of which there is no reallocation, and the Netherlands does not effectuate its charge to the extent the new residence state taxes the income included in the conservatory assessment. In my view also the seat-transfer levy is possible under treaties that are based on the substantial interest regime applicable as of 1997. Treaties with a ‘new style‘ substantial interest carve out do prevent this levy, because it occurs during the tax liability as a non-resident taxpayer of the substantial interest holder. Whether the seat-transfer levy is allowed under treaties from prior to 1997 is uncertain. Reliance on article 13, paragraph 5, OECD-Model Convention is not possible with respect to the seat-transfer levy. With a view to the conditions applied by the Hoge Raad in his fictitious wage- and pension emigration cases, arguments in favour of a seat-transfer tax are that there is a real advantage, no reallocation from one to another treaty article and that the reverse tax credit prevents international double taxation.

In chapter 7 the case law of the ECJ in the area of application of treaty freedoms in tax matters has been analyzed, and the residence fictions and the exit levies have been tested against the treaty freedoms. The emigration levy in its current form has been found in accordance with EC law by the ECJ in the N-Case (Case-470/04). From this decision it follows that an emigrating substantial interest holder with a 100% interest can always invoke the freedom of establishment, also in case the company in fact has its residence in the Netherlands. In other cases the free movement of EC nationals can be relied on. Notwithstanding the unconditional extension of payment and the cancellation for decreases in value arising after emigration, the emigration levy does result in a restriction according to the ECJ. This restriction, however, is justified based on the allocation of the right to levy based on the territoriality principle. The court does find the emigration levy in accordance with the fiscal territoriality principle that forms the basis of article 13, paragraph 5, OECD-Model Convention and which embodies the principle that capital gains can be taxed in the state of residence of the alienator, and capital gains attached to the ‘temporal component‘ of the stay of the taxpayer in the Netherlands during the period in which the gain has arisen, is allowed. I have questioned this justification, because the various cancellation possibilities erode the territoriality of the rules, just like the liability for non-resident taxpayers in respect of the substantial interest. The chance that the seat-transfer levy will be deemed in accordance with EC law, in my view, is remote. This levy shall, just as the emigration levy, result in a restriction. A justification is in my view not present, since the rules are not coherent and many tax treaties do not give to the Netherlands a right to levy if the company has its effective management outside the Netherlands. The seat-transfer levy, other than the emigration levy, cannot be deemed in accordance with the principle of territoriality on which article 13, paragraph 5, OECD-Model Convention is based and which assigns the right to levy to the residence state of the substantial interest holder. The deemed residence of article 4.35 Individual Income Tax Act 2001 seems problematic only to the extent it would be applied with respect to the transit rules. If a shareholder in a company that was incorporated under Dutch law, but which has its effective management in a foreign country, would be excluded from the facility, such would be a restriction that cannot be justified; already because the arrangement does not meet the requirements of proportionality as a result of the application of the deemed residence. The deemed residence for the tax liability for non-residents renders it less attractive to have the company resident in the Netherlands for 5 years or longer to then move it from the Netherlands. It seems that a justification cannot be found. Invoking fiscal coherence or fiscal territoriality is in my view not possible, because application of the residence fiction, results, contrary to its goal, only in a levy in respect of value accretion originating outside the Netherlands. Finally I have researched whether in the European context a solution is possible for exit levies, as a result of which the emigration levy could be dropped. That solution cannot restrict the freedoms, must preserve the substantial interest claim originated in the state in which the value accrued and must be practical. A system in which the new residence state of the substantial interest holder acquires the claim of the state of departure against immediate payment (clearing), meets these requirements. Less practical, but probably politically more realistic, is the idea to give to the departure state a right to levy in respect of the substantial interest profit actually realized later, combined with mutual assistance.

Chapter 8 deals with the German substantial interest rules. The substantial interest rules are comparable with the Dutch rules, but deviate from these in essential ways. For example, a step-up is granted only in the computation of the emigration levy (Wegzugssteuer), and the emigration levy is only applied if the substantial interest is in a German company. The SEStEG introduces a new structure of the rules: a general step-up arrangement in the case of immigration (but only to the extent there was a charge abroad), the emigration levy is in the future independent from the residence of the company, and in case of emigration to a member state of the EU or the EEA extension of payment will be granted without security. It is interesting that the Ger-man rules only contain a cancellation possibility for decreases in value of the substantial interest after emigration and then only if these decreases will not be taken into account by the new residence state. The German rules are possibly not entirely EC proof as a result of the restricted step-up in case of immigration and the somewhat extensive annual filing requirements. The substantial interest carve outs in German tax treaties often relate to the application of the emigration levy and often result in the grant of a step-up by the new residence state. They offer advantages as compared to the Dutch ‘new style‘ substantial interest carve out because of their simplicity and the more liberal grant of taxation rights to the departure state.

Chapter 9 contains conclusions and recommendations. It is indicated in which concrete treaty situation the tax liability as non-resident resulting from the residence fiction is effective and how the effectiveness of the residence fiction could be strengthened. With a view to the findings with respect to the goals and the consistency of the substantial interest rules, as well as the effectiveness and the relationship between the residence fiction and exit charges under tax treaties, the Tax Arrangement for the Kingdom and EC law, it is, however, recommended to choose for the residence principle as exclusive basis of the substantial interest rules. That results in the termination of the rules for the tax liability of non-residents in respect of substantial interest and the residence fictions, and a full focus on the emigration levy. The emigration levy must to the extent possible lead to a real charge in respect of the value accretion that occurred during the residence of the substantial interest holder. To accomplish this result, the cancellation after 10 years must be withdrawn and the cancellation for decreases in value post emigration must be restricted to those cases in which the new residence state does not take the substantial interest loss into account. The substantial interest carve out should in my view seamlessly fit the emigration levy, which means in particular that the residence requirement must be dropped and it would ideally prescribe that the new residence state allows a step-up to the amount taken into account by the departure state in the computation of the emigration levy.