1) lack of substance (an old problem);
2) lack of beneficial ownership (especially with the new OECD Model Treaty guidelines , just published);
3) the permanent establishment attack (spc’s having a “place of management” in the home country of the multinational to which a large part of their profits can be allocated);
4) transfer pricing tax planning via shifting risks and intangibles to entities which do not have the knowledge to manage them.

However, the use of hybrid entities, if structured properly, will not fly in the face of any tax authority, because both of them, in a two country situation, will be confronted with a structure which to them is fairly “normal” ie. they know from their own tax viewpoint what the tax rules for these structures are and they usually show no interest in how they are treated abroad, because to them that is irrelevant anyway.

The topic for today: China
In this article I should like to focus on an excellent new tax planning opportunity for investments into China, offered by the introduction of the Chinese Partnerships Law per 1/7/2007 which deals with Chinese general and limited Partnerships. Such investments, in a Chinese LP, by multinationals from whatever country, via the Netherlands into China, offer these multinationals, even if they are not US-based, a “check the box” type of tax planning tool whereby they can freely choose to set up a Chinese LP structure which is tax wise treated as a subsidiary of its intermediate holding company in the Netherlands whilst in China it is seen as a tax transparent entity so the partners are subject to Chinese tax and not the LP itself.

This mismatch, in combination with the reverse mismatch that Dutch limited entities which invest abroad may also by characterized as a type of “check the box” tax planning, now under Dutch tax law, so they could be seen as tax transparent (“branches”) under the Dutch CIT rules whilst abroad they are treated as foreign legal entities, can lead to a very tax efficient financing of the Chinese operations. This might be achieved either in the way of direct financing of the Chinese operations or via indirect financing: either the operational lease of equipment or the licensing of intangibles. The end result in all three cases being a tax deduction in China without any pick-up of the corresponding income in the Netherlands. This may sometimes even be further combined with a double dip: tax deductible interest in the Netherlands (even after the introduction of a new Bill of Law which was announced in May 2011 to reduce the tax deductibility of interest ) and tax deductibility of that same interest (or an economically corresponding lease fee or licensing fee) in China.

This article has been written in close cooperation with several international tax specialists from Beijing University who have excellent access to the SAT, China’s state administration for taxes. In fact the University of Beijing, today, is the only institution in China which has been able to enter into a number of “advance tax rulings” (ATR’s) with the Chinese government, even though ATR’s do not officially exist in China.

What does “check the box” mean?
US-based tax payers have the option, for most types of foreign subsidiaries, to freely choose whether they will for US tax purposes be treated as tax entities (subsidiaries) or as tax transparent entities (branches). This can be done in the US tax return by checking a box on each foreign operation. When foreign entities are treated (“checked”) as branches, any intercompany agreements with their direct parent companies become invisible for US tax purposes, a feature which has given rise to massive tax planning by US based multinationals since these rules were first introduced in 1994. Tax payers outside the US are not normally offered a choice on how to treat their foreign operations: their home country tax rules will decide whether such foreign operation is to be seen as a shareholding in a foreign entity (subsidiary) or as a foreign branch of the home country enterprise.

This article deals with a dual option right to treat subsidiaries as branches: one in the Netherlands where Dutch subsidiaries of Dutch tax payers may disappear for Dutch tax purposes when entering a tax consolidated group, and the Dutch rules to determine the taxation of a participation by a Dutch tax payer in a Dutch or foreign partnership. The Dutch fiscal unity rules are a real “election” almost like the US check the box procedure; with the partnerships interests the “checking” occurs when determining the details of the partnership agreement. The usual freedom of contract ensures that Dutch tax payers can elect to alter the Dutch tax treatment of a partnership share by merely adjusting a few words in the partnership contract; the insertion or deletion of just one word may be enough to go from tax transparency to full tax liability for a given partnership share. Therefore, as easy as “checking a box.”

Why set up a Chinese LP or LLP if there is no joint venture?
Some readers of my previous articles have pointed out to me that they could not easily relate to my advice to set up hybrid LP’s abroad because their company’s foreign investment plans do not involve any cooperation with a third party: they just want to set up 100% owned foreign operations in case a foreign investment is being planned. To them the only tax question has always been “foreign branch or foreign legal entity?” So apparently it is quite a step already to realize that there is a third way to set up your wholly owned foreign business:

In this manner one combines the “foreign branch” tax rules with the “foreign subsidiary” tax rules. The difference is that usually neither a foreign branch nor a foreign subsidiary is capable of constituting a tax mismatch (other than in “check the box” elections made by US investors that are not the subject of this article). A branch is usually a branch in the tax view of the home country and in the tax view of the investment country. And a foreign subsidiary as seen from the home country is usually also a subsidiary in the tax view of the investment country. But an LP may give a tax payer the option to arrive at a tax mismatch at will. Such a mismatch can either be favourable (income elements not taxable in both countries or cost elements tax deductible in both countries) or unfavourable (income elements taxable in both countries or cost elements non-deductible in both countries) which obviously calls for prudence.

In fact, the tax planning around favourable mismatches (double tax deductions or income not taxable in either country) often has its roots in situations of double taxation: two tax authorities not willing (ie. not able) to take foreign tax aspects into account when deciding on the local tax aspects.

So I recommend that both tax payers and tax advisers, when looking at basic foreign investment scenarios, do not only take the usual branch versus subsidiary distinction into consideration but also spend some time on analyzing the tax effects of a fully owned foreign LP.

The Dutch “check the box” rule for foreign limited liability partnerships
The Netherlands is not officially known to have any “check the box” rule for taxation purposes and generally speaking there is no such rule, except for Dutch and foreign limited liability partnerships. Careful reading of the history of the Dutch CIT Act of 1969 reveals that the Dutch tax legislator has been struggling with the question of what forms of partnership should be treated as transparent for tax purposes and what forms should be subject to corporate income tax. This struggle has not been different from the struggle on exactly the same subject in other countries and each country has in the end taken its own decisions on this point, regardless of what other, even neighbouring, countries have done. They never really cared that this created a very high risk of double taxation, even under tax treaties!

The Dutch aim with defining the tax treatment of local partnerships has been to treat partnership forms which showed substantial resemblance with limited liability companies as subject to CIT and other partnership forms as tax transparent. The Netherlands’ legal system knows a fairly large number of joint venture formats, so in the end a number of compromises found their way into the Dutch CIT Act. It should be kept in mind that in the days that these decisions were taken, Dutch limited liability companies were always joint ventures: a BV had to be incorporated by at least two incorporators, but each incorporator was free to sell his shares to a third party (albeit under a right of first refusal for the other existing shareholders) even seconds after the incorporation. But the JV aspect prevailed, back then.

Therefore the decision was taken to subject so-called “open” limited liability partnerships to Dutch corporate income tax. Whether a Dutch LP is “open” or “closed” has been defined as a situation where upon the entrance of a new partner this decision of the partners meeting would or would not be subject to unanimous approval from all partners. In case such unanimity was part of the document which established a Dutch LP (which used to be and can still be mere contractual arrangements in the Netherlands which can exist without any formal “founding” requirements), the partnership was considered as “closed” and not subject to Dutch CIT.

It follows that a Dutch LP (called a “Commanditaire Vennootschap” in Dutch, usually abbreviated to “CV”) is in fact a check the box entity for Dutch CIT purposes: the founders may freely choose how to word their internal rules for the admission of new partners so they are free to create a CV which is not subject to CIT, if that suits them best, or to create a CV which is taxable for CIT.

It was not until the late nineties of the previous century that the Dutch tax authorities decided to use the same “open” versus “closed” criteria dating back to 1969 for participations of Dutch tax payers in foreign LP’s: this was officially made public via a so-called resolution in which the Dutch Ministry of Finance has laid down the tax criteria for “participations in foreign joint venture formats including LP’s”.

The main criteria for a foreign LP to determine its Dutch tax status as “open” or “closed” and if “closed”, as “comparable to a Dutch CV or not” (the check the box trigger) are:

1) Will the foreign LP own all business assets it uses in its enterprise or can some assets continue to belong to a partner even if used by the LP?
2) Is the capital of the foreign LP divided into shares or does the LP have a similar method to allocate profits to the partners?
3) Are all partners only liable for the debts of the LP for the amounts they have put in or are the partners or some partners liable for the debts of the LP without limitation?
4) Can new partners enter the LP or can partners sell their LP shares to other partners without the unanimous consent of all partners?

From a Dutch CIT viewpoint we will always need a foreign LP interest and not a foreign GP interest
This has to do with fairly old but still prevailing Dutch Supreme Court case law which held that a foreign interest as a GP in a Dutch LP structure must be regarded as directly accruing to the GP even if the Dutch LP is “open”. This case law will then also apply to interests held by a Dutch tax payer as a foreign GP interest and this interest can consequently never be regarded as a subsidiary because the GP interest is then always deemed to stem from a “closed CV” from a Dutch CIT perspective. So in fact a Dutch Open CV is a hybrid all by itself: it is only “open” for the limited partners and not for the general partners; slightly confusing perhaps but something not to miss when structuring the set-up.

Country by country tax research is clearly indicated
In general, the question with regard to an investment by a Dutch tax payer into a foreign LP “is this foreign LP comparable to a Dutch CV and if so, is it “open” or “closed”?” cannot be answered without closely studying the LP rules on a per country basis. Each country has different rules as regards the above four “tax attributes” at stake.

It should be carefully noted that the question “is the foreign LP share subject to foreign CIT as a branch office or as a tax entity?” is not part of the criteria! The Netherlands will treat a foreign LP interest of a Dutch tax payer as a foreign branch if the Dutch rules say so, even if the foreign tax system treats that interest as a tax entity and vice versa.

When studying the Law of the People’s Republic on Partnerships, it became apparent to me that Chinese LP’s seem to have all the characteristics necessary for the Dutch “check the box” election: the partnership can own business assets all by itself, but can also use business assets owned by one of the partners, the managing partners of the LP have unlimited liability and the admission of new partners or the replacement of existing (exiting) partners by other existing (remaining) partners is, according to the Chinese Law, subject to the unanimous voting of the partners meeting “unless arranged otherwise in the LP agreement”. So one can found a Chinese LLP where admission and replacement is subject to unanimous voting of all limited partners so the GP has no vote in this and this will turn the Chinese LP into an “open” LP so the interest in it qualifies under the Dutch rules for participations (ie. subsidiaries and the – in many ways different – Dutch tax rules for foreign branches do not apply.

What this might cause is depicted in three different case study scenarios, which show remarkable economical resemblance but are nonetheless treated differently under tax laws and tax treaties: intra-group financing, intra group operational leasing of equipment and intra-group licensing of an intangible.

“Tax authorities never really cared that their tax treatment proposals for LP’s created a high risk of double taxation even under tax treaties”

http://www.worldcommercereview.com/publications/article_pdf/202

http://www.worldcommercereview.com/publications/article_pdf/243

http://www.worldcommercereview.com/publications/article_pdf/273

http://www.worldcommercereview.com/publications/article_pdf/369

Introduction
Over the last several years many countries have increased the personal income tax burden on carried interest income, earned by private equity investment managers; other countries are contemplating to do the same in the coming years. This is especially true in the European Union. One may wonder if, apart from the investment manager moving abroad and seeking a new job with an investment management group overseas, where the fiscal attitude towards private equity is more favourable (USA, Hong Kong, Singapore) something can be done about this.

Below I will discuss an idea which may help to – sometimes significantly – reduce carried interest taxation. I will address this tax planning opportunity in some detail, because the Netherlands seems to offer an opportunity to do so. I will focus on the EU, where the problems are clearly visible in the form of a drain of well-qualified investment managers to the USA and Asia.

The starting point of my “Dutch carried interest planning” is that certain income elements, if earned in the Netherlands, cannot be taxed again in the home country of the investment manager concerned, due to the presence of a tax treaty between the Netherlands and that other country. Seen in this light, the new carried interest taxation rules which have recently been introduced or will shortly be introduced in many EU countries may prove to be ineffective.

For example the French “Arthuis” rule which went into effect on 01/01/2011. Until then the French tax on carried interest income was a flat 30% income tax, but today carried interest is taxed in France under the progressive income tax scales, which will often imply that a large part is taxable against 50+%. And carried interest will from 1/1/2011 onwards also be subject to the French social security levies of 20+%. One should therefore not be surprised to see future carried interest rewards in France attracting a combined income tax plus social security contributions rate of well over 70%. The situation in other EU countries may be much the same.

But what would happen if this income comes from a Dutch legal entity in the group and qualifies for avoidance of double taxation in France under the Dutch/French tax treaty? Under the proper circumstances, France may not be allowed to tax such Dutch income elements. France may then only apply its progressive individual income tax rates to the French income parts earned by the French resident investment manager, and must leave the Dutch part untaxed. So under certain circumstances and with careful tax planning, France may often not be able to levy its 70% tax plus social security on carried interest income of French resident investment managers, because its tax treaty with the Netherlands got in the way. The same may be true for many other countries which have unilaterally increased carried interest taxation or are planning to do so.

The obvious next question then is: how high is the Dutch income tax on the carried interest rewards which the Netherlands may tax under the treaty? Is the Dutch tax rate better than 70%? The answer is mind boggling: with precise planning, the Dutch tax rate on carried interest rewards earned by non-Dutch residents with a non-Dutch EU passport can be kept at just 10% even for very considerable carried interest sums. This is true for employment income including some forms of deferred employment income like carried interest “bonuses” of up to €80,000 per year, so over an extended period of time for several hundred thousands of euros. Details of how this might work are not disclosed in this article but are based on the work I have done in the past for many HR departments of multinational enterprises and are available on request (jos@merlyn.eu).

Analysis
The above is just a “basic idea”. It requires fine tuning (below) and a word of warning is needed too; with 27 EU countries which all have their own individual income tax and social security tax systems, including different ways to avoid double taxation on foreign income elements, my idea can be made to work in several situations, but not in all. In some EU countries the overall tax savings from routing carried interest rewards or parts thereof via the Netherlands may result in very considerable tax savings, especially those EU countries which apply an “exemption with progression clause” as a result of their tax treaty with the Netherlands, but in other EU countries, especially those where a tax credit system applies to foreign income, the Dutch carried interest planning may not make much difference at all. The question whether the home countries applies a so-called ceiling to the income amount, subject to social security charges is also important: the benefit might even be considerably higher if countries do apply such a ceiling.

For investment managers in non-EU countries, who may also be able to benefit from a tax treaty which their home country has with the Netherlands, the benefit calculation will co-depend on the question whether there is not only a tax treaty with the Netherlands but also whether there is a social security treaty.

So a general outcome cannot be given. Generally speaking the calculations come out best for investment managers from the following EU countries: Belgium/Luxemburg/France/Germany/the Czech and Slovak Republics/Spain/Hungary and Denmark (“exemption with progression” countries). For non-EU countries calculations will have to be made separately.

Investment managers from EU countries such as Italy/the UK/Poland/Ireland/Sweden/Greece/the Baltic States may not be able to realise any benefit from the low Dutch income tax which is also true for certain non-EU countries (“tax credit” countries).

Points of additional attention
I will now guide you through some important implementation details; just a good basic idea is of course never enough: it must be thought out in full, as if it had already been implemented, to the level of the future foreign income tax returns to be filed by the investment managers in their home country in the future which will contain the Dutch income elements. Only then can we be certain what the actual income tax savings for a particular investment manager in a particular country really are.

Challenge 1: the Dutch income must be income from Dutch employment, not from a Dutch shareholding.
Not all income earned from Dutch sources is subject to the avoidance of double taxation rules in tax treaties. Especially carried interest, which often takes the form of shares which the investment manager obtains in a given investment fund. Income from shares is always taxable in the country of residence of the investment manager, even if it would come from shares in a Dutch legal entity. So the first step must be to convert future income from carried interest shareholdings (ie. dividends and capital gains) into Dutch employment income. By itself this does not have to be a difficult exercise: the shares allocated to the investment managers should from the outset, before they have any intrinsic value, be allocated to a Dutch legal entity whereby the investment manager, in lieu for these shares, obtains an entitlement to an incentive payment from the Dutch entity, which qualifies as employment income (a “bonus”). This should be neutral to the Dutch entity: it will only have to pay these bonuses to its new foreign employees if and to the extent it will make a profit on the shares which it obtains from these employees. Investment managers who want to continue owning the shares in their fund(s) themselves cut themselves off from the tax treaty benefits described;

Challenge 2: what kind of Dutch employment contract is required: one for an employee or one for a director?
The big difference between income from regular employment as an employee and income from a directorship for the day–to-day application of tax treaties is the issue of where the work may be performed. In situations covered by the regular employment article of a tax treaty (usually article 15, exceptions noted) the foreign employee must perform his activities IN the Netherlands. This may of course practically limit the effectiveness of my tax panning idea. For a director, there is no need to spend time in the Netherlands; he can continue to work in his home country and/or in the investee company countries like before. So, generally speaking, from a tax treaty perspective, a Dutch directorship is “better” from an execution viewpoint than a regular Dutch employment contract as an employee.

But there is of course a general sort of restriction here: one cannot, practically speaking, make all foreign employees who are elected to participate in a Dutch carried interest transformation programme which would replace their original home country carried interest programme, directors of the Dutch legal entity (usually a “BV” company). That just is not workable and also not very logical. The home country may ignore such “directorships” if the person involved is not actually performing a director’s function in the Netherlands based on “treaty abuse”. The Netherlands would do so in the reverse situation, according to a guideline published by the Dutch Ministry of Finance: Dutch residents who want to benefit from exempt director’s income from foreign legal entities but who do not travel to the country of the entity, will be denied tax treaty relief if their directorship is fake (there is no Dutch case law to confirm that this view has ever been the subject of a tax court case in my country, however).

And there is also a tax treaty relevance here: many countries treat foreign income from regular employment in the same manner as foreign directors’ fees, but in some Dutch tax treaties, the method to avoid double taxation differs. Notably the Dutch tax treaties with “good treaty countries” France, Denmark and the Czech and Slovak Republics distinguish between employment income and directors’’ fees, whereby double taxation on directors’ fees is avoided via the tax credit method. As explained above, the tax credit method is not the method which will bring the desired effect. So investment managers from these countries will have to accept a regular employment contract, with the consequence that they must perform their activities IN the Netherlands too. Entering into a directorship agreement would in such cases annihilate the tax benefit.

For social security, there may also be a distinction between an employee and a director: in some EU countries a director is treated as a self-employed person whilst the Netherlands will treat him/her as an employee (my country makes no distinction in these cases). Belgium is a good example here and additional precautions may be needed to avoid double social security levies.

In the French/Dutch situation I have asked my French network partners to make a number of calculations of what French tax and social security contributions savings might be achieved by routing €500,000 of carried interest income via the Netherlands over a 7 year period. As stipulated above, if the income would have been earned in France under the new French rules, these fine people would have paid 70% in France. In the Netherlands they would pay 10% only. Due to the progression clause in the French/Dutch tax treaty and due to the way the social security system in France treats foreign income, there would be some additional 25% French tax plus social security due, which brings the combined Dutch/French levy to 35%. This implies that French investment managers who would obtain their carried interest via a Dutch group entity would save themselves a whopping 50%, or €175,000, of tax NET from the idea as unfolded. How much additional gross income would one have to earn to arrive at a NET additional 175K euros? For some countries (where social security is tied to a “ceiling”) the benefit could be even bigger.

Challenge 3: what Dutch legal presence is required? Do we need a Dutch “fund” or just a Dutch legal entity which owns shares in (a) foreign funds(s)?
It would of course be easiest to implement my carried interest transformation ideas in case the Dutch entity which is required to give the foreign investment managers access to the individual income tax benefits described, if the Dutch entity would be the investment fund itself. This would for instance eliminate much of the discussion above on whether there would actually be Dutch work-days, which may, as explained, be a big practical hurdle to affect my tax planning idea if the Dutch entity is not a fund itself. This question is a tough one to answer, however. From practice I know that the private equity sector is generally not keen on Dutch investment funds (exceptions noted!) for various tax and non-tax (regulatory) reasons. I can’t comment to the regulatory issues for lack of sufficient knowledge, but from a tax viewpoint this question touches upon all the corporate taxation aspects of Dutch investment funds. Just briefly:

1) VAT; with VAT exemptions for banking and insurance activities (where a lot of the private equity money comes from) there is considerable risk of non-recoverable VAT; this risk is augmented by the fact that passive holding companies suffer from VAT exemptions themselves. Many private equity investor groups will therefore avoid any EU country to put their funds in and prefer the Channel Islands or Switzerland to get rid of non-recoverable input VAT.

Working with several non-EU based funds such as Channel Islands companies creates considerable risks in other corporate tax areas, by the way. Such set-ups bring along the need to interpose all kinds of special purpose companies in between, say, Jersey and the investee countries. But the tax authorities of those investee countries are increasingly challenging such SPC’s for lack of substance and lack of beneficial ownership. It is expected that the OECD will soon publish a guideline on beneficial ownership, which will make the situation even worse. Such problems do not exist with investments from most EU jurisdictions, which can be described as “general purpose companies”. But a VAT problem is about “known amounts of money today” and a beneficial ownership or substance issue is about “unknown amounts of money tomorrow” so most private equity investors seem to opt for the short term VAT recovery rather than avoidance of future capital gains tax;

2) The applicability of the Dutch participation exemption to foreign dividends and capital gains on exit; the Dutch tax authorities show an amazing hesitation to apply the Dutch participation exemption to private equity funds, advance ruling are hard to get in my country and the ones I have obtained contain side conditions which make the structure de facto unworkable. Despite the fact that the Dutch tax treaty network all by itself, plus the new participation exemption provisions as from 1/1/2007, would be ideal for private equity investment purposes;

3) Some technical opinions I have seen from other tax advisers that comment to “which country is best for investment funds” stipulate that a Dutch legal entity other than a fund would need to own at least 5% of the shares in a foreign investment fund to ensure the Dutch participation exemption on future dividend income and capital gains; this is not true, however: an entity in the group to which the Dutch entity belongs (in fact: the foreign fund) has to own 5% together with the Dutch entity; the Dutch entity itself may own percentages below 5% (the “carry along rule”).

So despite the fact that on paper the Netherlands is an almost ideal country to host private equity investment funds, this has not materialized. The corporate tax aspects of private equity will be addressed by me in more detail in a future article. But if the doubt which the private equity industry now has toward putting funds into the Netherlands remains, this may have a drawback on the practical feasibility of any carried interest transformation programme, of course.

Conclusions
The Netherlands offers an opportunity for non-Dutch resident employees and directors of Dutch entities to earn a substantial income part of up to several hundreds of thousands of euros over a period which should coincide with the duration of the investment fund, against a 10% effective Dutch tax plus social security rate. This rate is not available to Dutch residents, by the way. If structured well, such Dutch income elements cannot again be taxed in the home country of the investment manager, but must be exempted, under the tax treaty of that home country with the Netherlands. Since carried interest income is subjected to ever increasing tax and social security levels in parts of the world, certainly in the EU, such a “transformation” of carried interest income from locally sourced to Dutch sourced, may bring very considerable tax benefits. Both in the areas of the “progression” clause which is usually part of that same treaty and due to the social security system of the home country, part of the initially very large tax differential between the Netherlands and the home country may wash out, but even then, tax savings of 50% on carried interest income seem possible and plausible.

The precise structuring of the tax solution I have described is an exercise all by itself, due to a number of additional constraints which have to be taken into account as explained. But there certainly is hope!

The use of two hybrid Dutch entities for a chemicals production company that wants to enter Europe in various countries simultaneously

Phase one: avoiding exit tax upon IP transfers to the new jurisdictions

This case involved a company in a high tax jurisdiction that was considering start-of-production locations for their chemical products in various European countries. The company is profitable in its home country and seeks to avoid additional tax there on income related to the transfer of the intangibles it uses at home to the new European production locations; in addition it is interested in keeping foreign profits tax low. Can the Netherlands play a role in the tax planning required?

The first issue to address is clearly to avoid taxation on the transfer of product technology and production technology plus a marketing intangible (the “client list”) from the parent company of the group to the new production locations. An immediate and taxable capital gain on such transfer can be avoided by transferring the intangibles mentioned to newly set up foreign subsidiaries against an “arm’s length” royalty payment but this would still risk additional taxable income, as of day one, in the home country of the parent whilst the new subsidiaries will generate start-up losses so the tax deductibility of their royalty payments do not reduce foreign tax for a number of years.

An exit charge on the IP in the group’s home country could perhaps be prevented by using a hybrid legal entity to transfer the IP to: this entity should form a so-called “permanent establishment” of the parent company abroad, which would avoid the exit charge because no transfer to a separate legal entity will take place under the home country’s tax laws. If this could be coupled to a situation whereby the foreign operations are seen, under the tax rules of the country at the receiving end of the IP transfer, as a “tax entity” one might be able to get a so-called “step up in basis” for the IP which it receives from its foreign parent. Can such a hybrid entity (“permanent establishment” [non-resident tax payer] as seen from the parent in the group, local tax payer as if it was a separate entity [resident tax payer] as seen from the subsidiary itself) be structured in a country which will also allow other multi-country tax benefits? This question was asked by the foreign parent company to several tax advisers in European countries, including to me.

My answer for the Netherlands, but form a multi-country investment purpose, was as follows:

A Dutch “Cooperative Association”’ could well be the hybrid instrument to look for. It has all the hallmarks of a partnership under the tax laws of many foreign jurisdictions, which gives rise to a tax treatment in the home country as a foreign permanent establishment, but nonetheless, from a Dutch corporate income tax viewpoint it is an entity subject to tax itself: the “members” or “partners” will not become taxable in the Netherlands. The Coop will be treated like a Dutch limited liability company (see WCR December 2009[i]).

If a Dutch tax entity acquires IP for book value, this will have to be disregarded for Dutch corporate income tax purposes: assuming that the fair market value of the IP transferred (product technology and production technology plus one or more marketing intangibles) exceeds the book value for which the transfer will unavoidably take place, the Coop can take this difference into account in a tax favourable manner in Holland:

The Coop may capitalize the entire fair market value (to be established on the basis of a transfer pricing report) in its tax balance sheet and depreciate it over the useful economic lifetimes (each IP element may have a different life cycle in this respect). Dutch case law holds that a transfer of a business asset between related parties for a “wrong” price can analytically be seen as 1) a transfer against the right price, in combination with 2) a gift by the parent to the subsidiary of the excess;

An “advance tax ruling” is normally available in Holland to determine, at the outset of the new structure, what amounts the Coop can show as annual tax depreciation in its corporate income tax filings.

Once the above has been completed successfully, the group may, depending on its home country system to avoid double taxation:

1) Have avoided an immediate increase in its home country taxation because of the transfer of the intangibles;

2) Maybe also have avoided a future, annual, increase in its home country taxation because of the IP transfers;

3) Whilst at the same time safeguarding a meaningful tax deduction in the country to which the IP was transferred, if needed via tax loss carry forwards in case the Coop would run into start-up losses as is not unusual when starting up new business sin a new jurisdiction.

My advice, obviously, did not cover the home country tax system of the parent in the group. The obvious tax questions, to be answered by home country tax counsel or the in-house tax department, will have to focus on the ways in which a transfer of IP to a foreign partnership such as a Dutch Coop will affect taxable profit, not only upon the transfer but also in future years. How does the home country tax system deal with foreign “branch” profits? Will they be exempt? Or will there be a credit for the foreign tax incurred? This is beyond the scope of this article, however.

Phase two: expanding the tax effective IP usage to other European countries

The client clearly wanted to set up businesses in several European jurisdictions in a tax effective manner simultaneously, so the tax planning continued. Could we use the IP, transferred to a Dutch hybrid Coop, in other European countries as well and would some further creative tax structuring be possible with the Dutch Coop as the starting point?

In an earlier article (WCR March 2010[ii].) we have discussed the possibility to make a Dutch BV disappear for tax purposes in Holland, whilst it did not disappear in other European countries where it did business. A simple but real hybrid entity therefore. The foreign European operations would then either need to take the form of  genuine branch offices (permanent establishments) which is not a rocket science assignment, or one could work with foreign hybrid LLP’s (the subject of another earlier WCR article I wrote, see WCR June 2010[iii].). Such LLP’s would be tax transparent in the countries of operations but Holland would treat them as foreign subsidiaries under the somewhat peculiar Dutch “foreign entity classification rules” in its Corporate Income Tax Act, which distinguishes between so-called Open Limited Partnership and Closed Limited Partnerships for Dutch purposes. But my country uses these same rules to classify foreign “joint venture” models, to distinguish them between foreign “quasi-subsidiaries” and foreign “quasi-branches”, regardless of how the country of operations classifies these LLP’s itself.

A third and even very intriguing possibility, to arrive at a Dutch tax payer with foreign branch operations, is to create a pan-European legal entity structure in the form of an SE (Societas Europaea), a legal entity proudly introduced by the European Commission in 2004 whereby one only needs one limited liability entity in the EU, which is then governed by the same civil law or common law rules in all EU countries. A big disadvantage of the SE concept has been that if an existing business, already consisting of different legal entities in various European countries (such as an SarL in France, a GmbH in Germany, an A/B in Sweden etc.) would be converted into an SE, this would imply the liquidation, for tax purposes, of all legal entities in all other countries than the country where the SE ends up, so exit taxes would become due so the SE has always been denominated as “the still-born EU child of 2004”. But if one sets up new business, rather than converting an ongoing business, this disadvantage does not apply. In fact the disadvantage turns into an advantage: setting up an SE in one EU country, which subsequently does business in its own name in several other EU countries, will automatically and in very straight forward fashion lead to the desired tax result: a hybrid entity in one EU country with foreign branch operations in all other EU countries.

Note that setting up an SE entity is by no means an easy procedure for a group which does not already possess European legal entities, due to severe restrictions in the EU statute for SE’s, but this aspect, however important, is outside the scope of today’s article.

Turning back to the article I wrote on hybrid BV’s in WCR March 2010²., and emphasizing that “BV” can always be replaced by “SE” because the tax planning consequences will be the same, we will now structure the final piece of the tax planning exercise: the Coop on top of the Dutch structure possesses tax depreciable IP because of the step-up in basis discussed earlier. If it licenses this IP to its subsidiary-SE, which operates the branch offices elsewhere in Europe, the branches, under article 7 of the OECD Model Treaty, which has been incorporated – on this particular subject – in almost every real tax treaty, will become entitled to a tax deduction for their pro rata share of the royalty payments which the Dutch SE has to make to the Dutch Coop. Here the transfer pricing report needed to determine the step-up in Phase I will come in handy again, as it will show the “arm’s length” value of the IP rights.

Let’s say the Dutch SE also sets up business in Italy, Spain, Sweden, Germany, Poland and the UK. These operations will produce chemical products with the help of (perhaps even patented) product technology and production technology. In addition they will obtain the client list for their country and other marketing intangibles, previously in the hands of the parent company of the group before it decided to start operations in Europe.

Provided the royalty payments between SE and Coop are “at arm’s length” and also provided that the division key which the SE uses to spread the royalty payments over its various branches is in order (again: a TP issue), the Italian, German, UK etc. operations will be able to get a tax deduction for these payments, even if the payments do not come from them but from the Dutch head–office (on their behalf).

How big a deal is this, then? We have royalty expense deduction in the foreign branches of the SE (pro rata), in the SE itself (for the Dutch production site) but this leads to royalty income in the Dutch Cooperative. This is a Dutch tax payer, subject to the standard 25% Dutch corporate income tax rate on all business income including the exploitation of intangibles, so why would anyone want to put income into it?

Here’s where the hybrid character of the lower Dutch entity in the structure comes in: Dutch Coops qualify for Dutch tax consolidation, as parent companies. SE’s qualify for Dutch tax consolidation either as parent companies or as subsidiaries (although this is not common knowledge at all in Holland, but we have written confirmation from the Dutch Ministry of Finance on this).

So if I now take you back to my earlier WCR article on hybrid Dutch BV’s (WCR March 2010².), and now knowing that Dutch SE’s are treated the same way, tax wise, as Dutch BV’s and now also knowing that the SE can be part of what is known in Holland as “fiscal unity”, you may draw the proper conclusions together with me:

1) For Dutch corporate income tax purposes the SE does not exist!

2) As a direct consequence of this fiction, the licensing contract between the Coop and the SE does not exist either! One cannot license intangibles to oneself (in the Dutch tax consolidation concept, the SE has been absorbed by the Coop, it has in fact become a Dutch permanent establishment of the Coop);

3) So from a Dutch corporate income tax viewpoint, there are no royalty payments from the SE to the Coop. And invisible income is non-taxable.

We therefore end up with a situation where, regardless of the fact that in Germany, Sweden, Poland, the UK etc. a tax deduction may be claimed and must be granted by the tax authorities on the basis of the tax treaty with Holland for a pro rata share of the royalty payments between the Dutch SE and the Dutch Coop, there is no income pick-up in Holland.

This “phase 2” set-up has been the subject of Supreme Tax Court litigation between 1987 and 2003 (that is how long it might take for a case to be finally dealt with in my country, but litigation time in Holland is no exception compared to other jurisdictions), with an ultimate total victory for the tax payer concerned. The Dutch Supreme Tax Court not only decided in 2003 that:

(1) Set-ups like the one shown do indeed have the effect of disappearing income but also that:

(2) This tax planning cannot be disregarded as abuse of law either, because in the Supreme Court’s view there is no Dutch tax at stake: if the Coop would not set up an SE to license its IP to, and form “fiscal unity” with it, but would set up its own foreign branch offices, where its IP would be used, Dutch taxable income would equally not show royalty income: all income would be earned by the branches and all branches other than the Dutch one would be exempt from Dutch corporate income tax, under the standard Dutch foreign branch income exemption rules.

Below I have depicted the structure as seen under the tax rules of the countries involved, which clearly show two hybrid entities: the Dutch Coop and the Dutch SE. You will note that such structures are not necessarily restricted to the Dutch operations of multinational enterprise alone but can on the contrary be made to work, not only for many foreign parent company jurisdictions, but also for many operational jurisdictions.

enterprises, WCR Volume 3, Issue 4, December 2009 http://worldcommercereview.com/publications/article_pdf/202

[ii] The use of hybrid legal entities in group structures of multinational

enterprises, Part II, WCR Volume 4, Issue 1, March 2010 http://worldcommercereview.com/publications/article_pdf/243

[iii] The use of hybrid legal entities in group structures of multinational

enterprises, Part III, WCR Volume 4, Issue 2, June 2010 http://worldcommercereview.com/publications/article_pdf/273

After all, freedom of contract is available in most jurisdictions so parties can on paper agree to anything they want. And if money is provided from the top company in a group to its subsidiaries, the parent will often decide on its own what the terms and conditions should look like. Room for toying around, therefore, sometimes with unexpected consequences. Do it wrong and you end up with non-deductible interest expenses in one group entity but taxable interest income in another group entity. But the opposite is also possible: tax-deductible interest on the one hand but tax free interest income on the other hand.

This occurs because tax authorities, since the mid-eighties of the previous century, have found out that tax courts are willing, depending on the circumstances of each case, to reclassify interest payments into “deemed” dividends. The payer of an interest amount which is reclassified as deemed dividend faces non-deductibility (dividends are not tax-deductible in 99% of jurisdictions) but the recipient of a reclassified interest flow may well be able to claim a tax exemption: if real dividends can be received free form corporate income tax, then deemed dividends should be treated the same way. Many Western-European tax jurisdictions offer such a “participation exemption” nowadays: France, Belgium, the Netherlands, Germany, Spain, Sweden etc. The UK is considering participation exemption too.

During the last three decades or so of the 20th century, the Dutch Supreme Tax Court faced a number of cases whereby tax payers defended a different tax treatment of their interest income or expense than the tax authorities. This usually dealt with interest payment between related entities where conditions were imposed which one would hardly find elsewhere.

In a case where the parent of a group provided a loan to a subsidiary company in the group, but it was clear from the internal documentation that the intention had been to provide equity, but parties wanted to circumvent certain legal disadvantages to a capital contribution so they called the provision of the funds a “loan agreement”, the verdict was that this was not a loan but capital for tax purposes despite the wording of the document. The interest income from this loan was subsequently a deemed divided under the Dutch Corporate Income Tax Act (CITA) and remained untaxed (whilst abroad the interest was fully tax deductible).

In a second major verdict, the Supreme Tax Court faced a situation where a subsidiary of a multinational was obliged to redeem a bank loan and asked its parent company to lend her the money to repay the bank. So one loan was replaced by another loan. The verdict here was nonetheless the same. The subsidiary was not doing well at all (which was the reason for the bank to cancel the loan upon expiration date and not refinance the entity via a loan continuation). The Supreme Court held that if a tax payer provides funds to a related party knowing from the outset that the subsidiary will not be able to repay the loan unless circumstances would change dramatically and unexpectedly, the loan is not a loan in a tax sense but the provision of informal capital. If the debtor is Dutch, the interest on such a loan would not be tax-deductible but if the creditor is Dutch, the interest income would be covered by the participation exemption and remain untaxed.

Over the last decade several Supreme Court cases in the Netherlands have dealt with what tax practitioners now describe as “profit participating loans”. Multinational enterprises have increasingly provided funding to their subsidiaries through loan agreements which contain a number of “odd” conditions. Generally speaking such loans have three characteristics which differentiates them from third party loans:

a) A relatively long fixed term (25 years or more);
b) A relatively low fixed interest percentage (usually 1%); in addition there is variable interest which is dependent on the net after tax profit of the subsidiary; with good profitability the variable interest percentage can easily exceed 10%;
c) The loan is subordinated: in case the subsidiary goes bankrupt, all other creditors are paid first before the principal of the loan gets repaid.

There are still many jurisdictions where the tax authorities, to determine whether “interest” is really interest in a tax sense, are not allowed to use an economical analysis but have to treat the loan in the same way it is treated for commercial accounting and/or for legal purposes. A reclassification of interest into a deemed dividend will not easily occur there. But in the Netherlands and a few other countries, the tax authorities, either by law or by jurisprudence, are allowed to follow an economical approach. The most far-reaching tax test would be “could the subsidiary have obtained this loan from a bank?” and if not, the loan is no longer a loan for corporate income tax purposes and the interest thereon is a deemed dividend distribution (non-deductible for corporate income tax and perhaps even subject to a dividend withholding tax). This is the approach in the UK, also for transfer pricing purposes.

The Dutch Supreme Tax Court has followed a middle-road when it judged, in a series of verdicts all covering intra-group loans with variable interest rates:

1) That primarily the legal denomination (ie. the denomination under accounting law) of a loan will determine its tax consequences but:
2) In case the loan is granted under such conditions that the creditor, to a certain extent, participates in the business of the subsidiary, the tax treatment of the loan may require reclassification as the hidden provision of informal capital, so the interest payments no longer qualify as such for tax purposes but become deemed dividend income or expense, which is true if the following three criteria have been met simultaneously:

a) The loan has a maturity date of 50 years or more;
b) The interest is highly dependent on the future profitability of the debtor;
c) The “loan” is subordinated to all other debts of the subsidiary.

This Supreme Court verdict was rendered in the case of a French loan which is commonly known in France as a “Prêt Participatif”. Interestingly, this PPL contained a clause on an early repayment possibility (which at first sight seems to conflict with requirement a)) which did not bring the judges to a different verdict.

The Dutch tax legislator, at first, did not want to accept that many tax advisers started to use the Supreme Court criteria to create profit participating loans (PPL’s) which in the Netherlands would ensure applicability of the participation exemption to the interest income they created whilst knowing that such interest would be tax deductible abroad. For some time the Dutch CITA has therefore contained a special article which made PPL interest income taxable unless the tax payer could prove that the interest was not tax-deductible abroad, but this approach was given up per 1/1/2007 when this anti-abuse article for PPL interest was withdrawn. Consequently, in the Netherlands one may now again apply the old case law criteria to determine whether interest income or expense is also interest income or expense for tax purposes, or a hidden dividend.

With the above explanations in hand one may relatively easily create hybrid intra-group financing structures, via Dutch intermediary holding companies or Dutch group financing companies, from the parent entity in the group to subsidiaries in tax jurisdictions which allow the deductibility of interest on group loans if such loans for commercial purposes are treated as loans, which lead to the tax mismatch described. So one either gets an interest deduction for what in an economical sense is the provision of equity, or a “double dip” (two interest deductions for the same amount) if the group finances the Dutch PPL’s from borrowed money in jurisdictions where interest to finance subsidiary entities is tax deductible.

The insertion of a Dutch group entity into such PPL structuring is highly recommended because the interest paid by the subsidiaries should qualify as interest for local corporate income tax, in which case it will likely also qualify for local interest withholding tax. Using the Netherlands in such a set-up implies that one will have access to the Dutch tax treaties which almost invariably imply a zero foreign withholding tax rate. After all, the Netherlands has no interest withholding tax itself so its tax treaties reflect this and the treaty partner is usually willing to give up its own interest withholding tax if the recipient is a resident of the Netherlands. The usual tax treaty requirement that the Dutch entity must be the beneficial owner of the interest income might have to be achieved by taking a final tax planning step (the Beneficial Ownership Booster, BOB) which involves a relatively simple extra step in the set-up which we will gladly explain to anyone who wishes to set up tax effective intra-group financing via PPL contracts with our assistance. In this respect it may be interesting to know that advance tax rulings from the Dutch ruling authority are available for PPL’s.

We conclude with an example how a hybrid PPL might bring benefits to a Swedish tax payer with a French subsidiary by inserting a Dutch holding company in the group structure.

Dutch LLP’s: Dutch corporation tax act allows tax payers to freely choose between a “tax entity” or a “tax transparent limited partnership”
Article 2 of the Dutch Corporation Tax Act contains a definition of which Dutch legal or contractual entities are subject to Dutch corporate income tax. As far as Dutch LLP’s are concerned, this article subjects to tax the so-called “open limited partnerships”. In doing so the law apparently distinguishes between open limited partnerships and closed limited partnerships. The latter category is not subject to Dutch corporate income tax.

A Dutch LLP is known in the Netherlands by its abbreviation in the Dutch language: CV (Commanditaire Vennootschap). This is the notion we will use below for such Dutch entities.

A CV has at least one general partner and one limited partner. Limited partners are liable only for the debts of the CV up to the amount of their partnership contribution. General partners are fully liable for the debts of the CV and are therefore usually limited liability companies, also in other jurisdictions.

The question whether a CV is “open” or “closed” has been dealt with in legislative history and case law. An “open CV” is any CV which is not considered “closed”. A “closed CV” is a Dutch limited partnership which has quite severe limitations to the access of new partners and the voluntary transfer of a partnership share from one partner to another partner. Only in case such accessions and transfers are subject to the express approval of ALL partners, is the CV a Closed CV and transparent for Dutch corporate income tax purposes. In such a case the partners are subject to corporate income tax in the Netherlands themselves if they are Dutch residents and if they are foreign residents, in case they operate a Dutch permanent establishment or own Dutch real estate.

By playing with the CV’s “articles of establishment”, tax payers are therefore entirely free to choose whether to set up a CV which is taxable for Dutch CIT by itself or to create a CV which is transparent for Dutch CIT purposes. All will depend on the founding documents, especially those concerning admission of new partners and transfer of partnership interests between existing partners.

The foreign tax denomination of a Dutch CV will invariably be dependent on foreign tax rules (foreign entity tax classification rules), so a tax mismatch (good or bad) can rather easily occur: a Closed CV may well count as a Dutch tax entity abroad even if it is not subject to Dutch corporate income tax and an Open CV may well be seen as a tax transparent partnership abroad even if it is subject to Dutch CIT itself. CV’s are therefore “tricky” entities to work with, from an international tax perspective. One may run into double taxation before one knows it, but the opposite (no taxation at all) is also possible. This implies, as always, that one cannot really ignore the tax rules because if one does, things may go terribly wrong, with double taxation as a result.

Foreign LLP’s under Dutch corporate income tax principles
The Netherlands, like any other country, uses its own criteria to determine if foreign LLP’s must be considered taxable entities in which case they are seen as “participations” which qualify for the participation exemption, or as transparent entities in which case they should be seen as a foreign branch office of the Dutch participant/limited partner subject to the Dutch foreign branch income exemption. It should be noted that the two Dutch tax exemptions, one for income from and capital gains realised with participations and the other one for branch income, differ markedly from each other so changing the one exemption for the other may make considerable financial difference. The foreign tax criteria (like the question “is the LLP subject to tax itself under foreign tax law?”) play no role in the Dutch entity tax classification process either.

The Dutch Ministry of Finance offers tax payers guidance in the “Dutch tax classification of foreign entities” process via a so-called Resolution, dated 18/12/2004, which contains the following criteria:

1) Can the foreign joint venture, under its own legal system, own the assets with which the joint venture is conducted?
2) Is there at least one participant in the joint venture who is liable for the debts of the joint venture without limitation?
3) Does the joint venture have a capital dividend into shares?
4) Can new participants access the joint venture or can participant transfer their share in the joint venture to other participants without the unanimous acceptance by all participants?

The answers to the above four questions will basically determine whether from a Dutch corporate income tax viewpoint the foreign joint venture qualifies as a tax entity or as a tax partnership. In case the foreign joint venture is legally comparable to a Dutch CV (which is the case if questions 1) and 2) above have been answered affirmatively, the foreign LLP is a “CV lookalike” in which case criterion 3) loses its significance and criterion 4) needs to be looked at in detail, in which case a set of additional Dutch rules apply, as follows:

a) the foreign joint venture conducts an enterprise in its own name;
b) there is at least one general partner and one limited partner;
c) the general partners are liable for the debts of the joint venture without limitation (although they might be LLC’s themselves);
d) the limited partner is only liable up to the amount of his capital contribution;
e) the limited partner does not act towards third parties as representing the joint venture.

These Dutch criteria are not part of any foreign entity tax classification rules, so a mismatch between the Dutch and the foreign tax take of a foreign LLP can also very easily occur. A good example of this would be the German KG: This entity very much resembles a Dutch CV (the words even mean the same in the two languages), so from a Dutch corporate income tax viewpoint, German KG structures where a German limited liability company acts as the general partner (“GmbH & Co KG” structures) are “CV lookalikes”. It will then depend on the internal rules in the KG as concerns the access of new partners and the transfer of partnership shares between partners, whether the German KG is seen as a tax entity (“Open KG” from a Dutch tax viewpoint) or as a tax transparent entity (“Closed KG”). Regardless of the fact that under German law a KG is always tax transparent!

The Dutch distinction between Open CV’s and Closed CV’s, in use to distinguish foreign LLP interests in “foreign participations” and “foreign branch offices”, based on the internal LLP rules concerning their accessibility to new partners and to the transferability of partnership interests between partners, is a rather unpractical one. Obtaining consent from all other participants for each and every change in the partnership composition is in fact unworkable in partnerships with more than just a few partners. However, this implies that foreign LLP’s which resemble their Dutch CV counterparts will usually be regarded as “Open LLP’s” in which case they are treated as “participations” of the Dutch limited partner who participates in such a joint venture even though abroad they are treated as tax transparent. A mismatch between the Dutch and the foreign tax treatment of LLP’s is therefore often unavoidable.

Examples
In case a foreign LLP is considered a “participation” (ie. a “subsidiary”) from a Dutch corporate income tax viewpoint whilst the foreign jurisdiction, considers it tax transparent, like in a KG situation, the following might happen:
a) The Netherlands will normally exempt any and all income from such a foreign LLP from Dutch corporation tax under its “participation exemption” (eg. interest payments);
b) The foreign tax authorities may equally exempt the Dutch share in what they see as a tax transparent LLP from local profits tax; this would especially be true in case the LLP share does not constitute a permanent establishment of the Dutch limited partner in the foreign country under foreign tax law;
c) Even if the foreign tax authorities would consider the Dutch limited partner taxable in their country, eg. for operating a permanent establishment there or for owning real estate, they may allow for tax deductions for a variety of expenses (especially under a tax treaty with the Netherlands which resembles the OECD Model Tax Treaty’s article 7-3, which is true for 99% of the Dutch tax treaties). However, such foreign tax deductible items may not picked up in the Netherlands as income (by reducing exempt foreign branch income), as a result of which a “double dip” in expense deduction may easily occur: certain expenses incurred by the Dutch “partner” for the KG will be tax deductible in both countries;
d) It may even become possible for the tax payer to create a tax deduction in the KG without income pick-up elsewhere in the group for internal expenses in the Dutch group which owns the KG interest. For instance, for mortgage interest which the Dutch participant in the KG must pay to its Dutch parent company that finances the mortgage loan from equity. The above KG example also works for many other countries;

In case a Dutch LLP is considered a taxable entity in the Netherlands (“Open CV”) whilst abroad it is seen as a tax transparent partnership, and the foreign jurisdiction is where the “parent” of the Dutch CV is, the following might happen:
a) Upon a transfer of intangibles against book value from the foreign parent to the Dutch CV, depending on the “foreign entity tax classification rules” to which the parent company is subject, which will likely deviate from the Dutch rules, there might not be a gain recognition abroad; after all, the parent transfers intangibles to itself (ie. its foreign branch office). However, the Netherlands may consider the CV as a taxable entity (Open CV) and will, upon the tax payer’s request and based on a transfer pricing report, recognize a transfer against fair market value and the intangibles may then be shown for their fair market value in the tax balance sheet of the CV and be depreciated over their useful lifetime (with a five year minimum depreciation period). This will create sometimes very substantial tax deductions, within a multinational group without pick-up elsewhere in the group.

Conclusions
Like Dutch NV’s, BV’s, SE’s and Cooperative Associations, Dutch LLP’s are open to so-called hybridization, in which case they are treated differently under Dutch corporate income tax law than under foreign corporate income tax law of their parent company. The same is true for LLP’s in which a Dutch tax payer participates. This may easily lead to double taxation (for which a tax treaty may not offer any solution). However, the opposite is possible as well: LLP’s, both Dutch and foreign, may give rise to double non-taxation, or to “double dipping” or to the tax deductibility of expenses in one country without income pick-up elsewhere in the group in another country.

Tax authorities show little or no interest in aligning their tax rules (“entity tax classification rules”) with those of other countries. They will therefore – albeit unintentionally – continue to subject tax payers to double taxation and in such cases, tax treaties offer no help at all. Is it then strange if tax advisers do the opposite and advise their clients to deliberately enter into certain hybrid entity structures whereby double taxation does not only disappear, but even turns into double non-taxation because part or even all of their corporate income disappears from the tax radar screen? The one comes with the other, in my view.

Working with foreign joint venture formats is never easy from a tax perspective, in any country. Still, this is the way in which business deals unavoidably develop. Tax payers doing business abroad usually have no choice than to work with a foreign joint venture format. Tax payers are therefore well advised to take a thorough look at their home country tax definitions of any foreign joint venture and to take nothing for granted, to avoid nasty surprises. But it is good to know that often, with some more tax planning, a possibility might exist to turn these tax risks into tax benefits. This can be achieved by deliberately opting for a foreign joint venture format which causes a mismatch between the home country and the investment country’s tax systems, so part of corporate income may fall “between the ship and the shore” or certain business expenses might become tax deductible in both tax jurisdictions.

Next time we will take a closer look at “hybrid financing” where one country sees a loan for tax proposes whilst the other country defines the financing arrangement as the provision of equity.

Introduction
Hybrid entities are legal entities which in the country of the parent of the multinational group are seen as legal entities, so they are considered to be subject to tax of their own accord, whilst the foreign investment jurisdiction considers them as “partnerships”. Such entities are not subject to tax themselves but their owners/shareholders may be, if they are deemed to run a permanent establishment in the investment country. Or the other way around (“reverse hybrids”). Many legal entities can be hybridised or are even hybrid by nature, due to the differences in the classification rules for foreign entities between the tax laws of countries;
This may well lead to double taxation (one country taxes the entity, the other country taxes the partners on the same income). However, the opposite is also possible: the entity is not taxed by either tax authority of the countries it does business in; this is especially true for Europe where many countries exempt foreign dividends from local tax;
Countries do not normally take foreign legal or tax aspects into account when establishing their tax classification rules for foreign entities: they invariably use their own criteria. These criteria differ manifestly per country, however. There are no signs that countries are planning to, or are even willing to, align their entity tax classification rules with one another, any time soon.
In many cases one can make deliberate use of “automatic” hybrids: no tax authority, neither at home nor abroad, will consider them odd or artificial, because they are widely used by others in a non-hybrid way.

The automatic hybridisation of a standard Dutch BV
The Netherlands, like many other tax jurisdictions, recognizes the notion of tax consolidation, widely known amongst tax practitioners in the Netherlands as “fiscal unity”. Any Dutch parent company which was incorporated in the format of an NV a BV, an SE or a Cooperative Association (“Coop”) can form a tax consolidated group with one or more of its subsidiary NV’s, BV’s or SE’s (Coops can only act as parents in such a group, not as subsidiaries). NV’s and SE’s are not different from BV entities as regards hybridization and will therefore not be discussed separately in this article. What I describe for a BV will also work for an NV or an SE in my country.

When taking a closer look at this feature, a few key points come to light which make the Dutch tax consolidation rules differ markedly from those of other countries.

Dutch “fiscal unity” is optional and a request must be filed on a subsidiary by subsidiary basis to achieve such a position. A Dutch parent can elect to form a fiscal unit with any combination of its BV, NV and SE subsidiaries which it deems appropriate. If it is better to leave one or more subsidiaries out of the fiscal unit, one is fully free to do so: “junction” is always possible at a later moment. Tax payers are also free to choose the moment when the fiscal unity starts and stops: this can be any workday throughout the year, also on a subsidiary by subsidiary basis.

A crucial element of a subsidiary entering a Dutch fiscal unit is, that from that moment onwards, the subsidiary becomes invisible for Dutch corporate income (CIT) tax purposes. The subsidiary, for CIT, ceases to exist. Its assets and liabilities are deemed to have been absorbed by its parent and the entity, from the junction moment onwards, is treated as a Dutch branch office of the parent. As a consequence, the subsidiary can no longer file a Dutch CIT return of its own.

Many tax practitioners worldwide are familiar with this Dutch “fiscal unity” procedure and it is often applied almost automatically: if a Dutch entity incorporates or acquires a Dutch subsidiary of which it owns 95% or more, a request for fiscal unity is filed with the tax authorities, often without further input from tax counsel: “the bookkeeper can do it” (and he/she usually does so).

This automatic pilot approach is by no means obvious, however. Fiscal unity in the Netherlands is subject to strict anti-abuse rules, both for a subsidiary entering a fiscal unit and for a subsidiary leaving the tax group, to avoid that tax claims get lost in the process. Such provisions could well kick in, also in cases of non-abuse, because of the way they were written. Or they may lead to unwanted side-effects. This should really be looked at in some depth before a fiscal unit is requested or enlarged and I have seen many cases whereby avoiding fiscal unity gave much better tax results long term than filing for it. After all, saving yourself the expenses of filing an extra CIT return should not be the driving factor, but this is often not well understood.

In this article I will not go into these side-effects of the fiscal unity concept, however. I just want to address an – equally often overlooked – main effect: if a Dutch company, which is a subsidiary of another Dutch company, files for fiscal unity, it ceases to exist only for Dutch CIT purposes and ONLY in the Netherlands. If such a company has business abroad, it will continue to be subject to foreign corporate income tax abroad in its own name and for its own account, regardless of the fact that it no longer exists for Dutch CIT purposes! After all, the company continues to exist for Dutch legal purposes and for all Dutch taxes except CIT.

Filing for Dutch fiscal unity treatment, for Dutch BV entities which operate foreign permanent establishments or possess foreign assets which give rise to foreign tax liabilities (eg. foreign real estate), therefore implies an automatic hybridisation of such entities: they cease to exist for CIT purposes in the Netherlands but continue their lives for CIT purposes abroad.

The stunning overall tax effects of an automatic hybridisation of a Dutch BV
It is easy to overlook the ultimate consequence of the notion, in Dutch corporate income tax law, that upon tax consolidation, the Dutch subsidiary company of a Dutch parent company “ceases to exist”, because it is not primarily a tax effect which arises, but a legal effect. But this legal effect may have huge tax consequences:

If a company which forms part of a Dutch tax consolidated group “ceases to exist”, as article 15 of the Dutch corporate income tax act describes it, this may have very significant consequences for any agreements which this entity may have entered into with the other companies in the Dutch tax consolidated group, including its parent company:
1) Intra-tax-group contracts will have to be ignored for Dutch CIT purposes as well. One cannot borrow money from oneself, one cannot enter into a licensing agreement with oneself, one cannot put employees to the disposition of oneself, one cannot rent equipment out from oneself, etc. etc.
2) Any income resulting from these intra-tax-group contracts therefore becomes invisible for corporate income tax purposes in the Netherlands. And invisible income cannot be taxed…

So the bottom line, for the Netherlands taxation of such a hybrid standard Dutch BV entity, is:

An entity which disappears for Dutch CIT purposes upon filing a Dutch fiscal unity request, will see all of its intra-tax-group contracts disappear in the Netherlands, which in its turn leads to disappearing income in the Netherlands from such contracts, which might otherwise, if earned outside a fiscal unity context, have formed regular taxable income in the Netherlands.

As observed, the Dutch entity – if it has foreign operations – does not disappear abroad. The entity remains to be a foreign tax payer with the obligation to file an annual CIT return. Its agreements with other Dutch entities in the Dutch tax consolidated group do not become invisible abroad and the income from such agreements does not disappear from the foreign tax radar either.

Income for one group company, under an intra-group agreement of whatever form, means expenses for the other group company. Hybridising a Dutch BV with foreign operations may therefore cause very significant tax reductions for the multinational involved, since one country must allow tax deductions for expenses incurred whilst the country where these expenses form income, does not recognize this income for CIT purposes.

An example in the area of “intra-group leasing of trucks”:
Say a US fruit transportation company “Frutrans Inc.” seeks to invest in similar activities in Europe, for instance the transportation of fruit from South-Spanish harbours (where it comes in from fruit producers in Northern-Africa) to France, Holland, Germany, Sweden and other areas north of Spain. Frutrans Inc. sets up a Dutch (Parent) Company X BV which sets up a 100% Dutch Subsidiary Company Y BV, and both BV’s form a Dutch tax consolidated group by filing a standard request form. The Dutch tax authorities will normally always accept such a request as the law contains no grounds for a refusal.

X BV subsequently purchases a number of specialized trucks and trailers, suitable for fruit transportation. This is done from a capital infusion which it will obtain from its US parent Frutrans Inc. Let’s say BV X invests €100 million. The trucks and trailers are subsequently leased out to Y BV which sets up a substantial branch office in Spain (virtually all business activities of BV Y will take place in Spain where the customers are which seek transportation of their fruit to Northern-Europe). The lease contracts between BV X and (the Spanish branch of) BV Y are drafted in such fashion that they meet the “arm’s length” standard for transfer pricing: BV X charges BV Y no more than a third party would charge BV Y for similar lease contracts under similar circumstances.

Let’s assume that a transfer pricing study shows that the arm’s length lease fee which Y BV has to pay to X BV for these trucks and trailers amounts to €17.5 million per annum. The Spanish branch office of X BV will be able to deduct these rental fees from its taxable profits in Spain, both under regular Spanish corporate income tax law and if needed under article 7-3 of the Dutch/Spanish tax treaty, which would override any Spanish limitations. The Spanish CIT reduction attributable to the lease contracts are around €5.5 million per annum.

But because Y BV does not exist in the Netherlands, its rental agreement with X BV also does not exist for Dutch corporate income tax purposes, so the Dutch tax consolidated group “X BV” (Y BV has disappeared) cannot recognize the rental income which Y BV paid to X BV for Dutch corporate income tax purposes and this income remains untaxed in the Netherlands. The net/net tax benefit of this setup is then equal to the Spanish tax savings from being able to deduct lease fees rather than depreciation of, some €10 million per annum, which equals €2.5 million each year; if the lease continues after the 10 year depreciation period, the tax benefit increases to 35% of €17.5 million or almost €6 million per year.

Intercompany contracts might of course cover all kinds of relationships, not just the rental of vehicles. Also in these other areas, the Dutch income disappears and the multinational group will be able to save itself millions of tax in the countries of their operations. You might think of:

a) A production environment, where the foreign branches of Y BV produce(s) goods under a licensing agreement for the US patents in use (X BV obtains the exploitation rights for these US intangibles, for use outside the USA, from its parent company in the form of a contribution to capital and licenses them to (the foreign branch of) Y BV;
b) A sales environment: X BV enters into a commissionaire agreement with sales and service subsidiary Y BV, which operates via permanent establishments (branch offices) abroad. The commissionaire agreement substantially reduces Y BV’s taxable income abroad but the income which flows to X BV as a result of this remains untaxed in the Netherlands because the commissionaire agreement does not exist from a Dutch viewpoint, so X BV files for a much higher amount of foreign profits (exempt from CIT in the Netherlands) than actually reported abroad.

A case like this has been litigated in the Netherlands with astonishing results
The Dutch Supreme Tax Court decided in a long and well documented decision which it handed in 2003, after more than 12 years of litigation concerning a case of “disappearing income” as described above, that the phenomenon of disappearing entities in Dutch tax consolidated groups do indeed lead to disappearing income from intra-tax-consolidated-group transactions and that the tax payer concerned was right in not reporting taxable income from such transactions as profit for Dutch corporate income tax purposes. The Supreme Court saw no reason whatsoever, despite repeated pleas by the Dutch Ministry of Finance during the appeals process, to take any tax effects in the country of the foreign operations of the Dutch subsidiary company into account. The verdict was firm and unambiguous.

The Supreme Tax Court also made it clear that this was not a decision in the case at hand only (factual decision), but an affirmation of a cornerstone of the Dutch CIT system, also in other cases where income might disappear as a result of the fiscal unity rules. The court even emphasized that the tax planning behind these structures can in no way be regarded as ‘’abusive’’ under any Dutch “abuse of law’’ rule.

Will the law now change in the Netherlands to prevent this outcome?
Tax legislators have the opportunity, if court cases get lost by the revenue service, to adapt the law if they can convince Parliament that the outcome is unwanted. This was also the base attitude in the Dutch Ministry of Finance when the case was first litigated in the late eighties of the previous century. In fact, an anti-abuse article was introduced in the Dutch CIT Act to make sure that the Dutch foreign income exemption for a tax payer of running foreign operations in its own name (via branches or the possession of foreign real estate) would not be higher than the income declared abroad.

But after the Dutch Supreme Court decided in 2003 that the tax payer was not only right, under the old law, to not report certain income under the Dutch fiscal unit rules, but that this approach also did not constitute any form of abuse under Dutch tax law and tax principles, the Ministry of Finance decided to remove the anti-abuse rule. It did so per 1/1/2005 when several fiscal unity rules were up for revision and clarification.

Per today there is only one anti-abuse element still in force: if a group entity in a Dutch fiscal unit grants a loan to another tax group member, which loan is used in a foreign permanent establishment of the subsidiary entity or for the financing of foreign real estate owned by that subsidiary entity, the interest on that loan will be added to taxable income in the Netherlands even if it does not exist under fiscal unity principles, unless the tax payer proves that the interest has not been tax deductible abroad. All other types of intercompany agreements between members of a Dutch fiscal unit have been taken out of the anti-abuse rule and tax payers are free to use the Dutch fiscal unity rule, which allow for disappearing income, to their financial advantage.

Multinationals without Dutch business can also benefit from the Supreme Court verdict
Setting up a Dutch intermediate holding company, which incorporates a Dutch subsidiary which operates foreign branches, can not only be financially beneficial for Dutch multinationals. These tax planning structures are equally open to multinationals from other countries. Nothing should stop a Swedish multinationals with plans to start production in Singapore to put the above described Dutch “sandwich” in between: Singaporean CIT will substantially be reduced, the Netherlands will levy no tax and Dutch dividends will be tax exempt in Sweden. In fact, almost any multinational in a country which exempts foreign dividend income from taxation can benefit.

Even multinationals in “tax credit” countries such as the USA, Australia, the UK, Canada, China and Japan may benefit substantially from this “Holland routing” of part of their corporate income. This might be the case if they are looking for low-taxed foreign source income to avoid excess credit situations at home, or it might, in the absence of CFC type tax legislation at home, lead to at least very substantial tax deferrals for the group, with a corresponding rise in earnings per share!