Saturday, December 23, 2006

Christmas (tax) Reading

The year can be closing and some people are thinking in buying last minute presents, writing their Christmas cards and gathering around their family, away, very far away from international taxation.

Then again the “usual suspects” come with large pieces of reading material for the Christmas vacation so that our attention is not focused elsewhere. The “usual suspects” suspects are the OECD and the EU Commission, which released a series of reports that for the ones that attempt to follow the international tax developments, it basically means extra tax reading and no fiction novels. Do not even think about it!

The European Commission (usual suspect n.1) adopted a Communication on a co-ordination of national direct tax systems, a Communication that invites Member States to adopt an EU-coordinated approach on exit taxation and a Communication that invites Member States to adopt an EU coordinated approach to cross-border loss relief.

The aim of the first Communication is to coordinate the national direct tax systems to ensure that they are compatible with EU law and interact coherently with each other. The second Communication provides specific guidance for the EU Member States on designing exit tax mechanisms compatible with EC law principles. Finally, the third Communication follows up the Marks & Spencer case, and invites Member States to develop a coordinated approach to cross-border loss relief that eliminates what the EU Commission says to be one of the “main obstacles to cross-border investments”! Perhaps a bit exagareted, specially if we take what the ECJ said in the M&S case.

Remeber, that the European Commission plans to issue a further Communication in early 2007 to report on progress to date in the CCCTB, which miraculously solves the problems identified on the EU communication. What a coincidence! This is called the inversed stick and carrot approach! The ECJ basically intends to beat the Member States with the soft law stick until they find the light of the CCTB! Now it’s a question to see if it lights the whole Europe or only a few brave bunch!

The OECD (usual suspect n.2) had to respond at the level of its Brussels counterpart and the new impetus to the knotty PE project was the answer. The OECD published the expected new versions of Parts I, II and III of its Report on the Attribution of Profits to Permanent Establishments, along with a cover note containing an update on the status of that project. This means that we may send the previously published discussion drafts of Parts I-III to the trash bin and start reading the new ones in a “fun” game of finding the differences!

The OECD discreetly says that the new versions of Parts I (General Considerations), II (Banks) and III (Global Trading) reflect the broad consensus of OECD member countries around an approach to attributing profits to permanent establishments which is based upon the arm’s length principle as described in the 1995 OECD Transfer Pricing Guidelines. One wonders whether the previous drafts failed that test! As to the announced necessary changes to the language of OECD Model Tax Convention and Commentaries, somewhere in 2007 appears to be the target of the OECD! The Part IV (Insurance) is still a bit behind the rest of the gang but do not stress because the OECD promises us to provide extra reading ASAP!

A special thanks for the ECJ (usual suspect n.3), which has the dignity of closing well ahead for the Christmas season!

Have a nice Christmas (reading)!

Tiago Cassiano Neves

Next post: what to expect from 2007!

Friday, December 15, 2006

OECD New "Service PE" proviso

Did you ever noticed that some source countries are slightly reluctant to adopt the principle of exclusive residence taxation of services performed on their territory, which are not attributable to a Permanent Establishment (PE) in that source country? Believe me, there are enough countries out there supporting service PE clauses...perhaps in Asia....

The OECD did notice and therefore decided to propose (just in case) some clarification into the Commentary on Article 5, reflecting the conclusion that the current provisions of the OECD Model Tax are appropriate to deal with services (i.e. the right result for services is residence-based taxation). Just for the sake of clarity and completeness, the OECD also includes an alternative provision to be added to Art. 5 for those "rogue countries" that for one reason or another which to preserve source taxation rights on profits from certain services.


You shouldn't be. Just remember that the OECD is no longer an exclusive club of rich (ups..high developed) countries supporting residence-based taxation. In recent years, just as the world economy, the OECD has become more flat! (read the book the World is Flat and you will understand why!). The inclusion of a (alternative) provision on the taxation at source of certain services (even if hidden in the deepness of the Commentaries) is a sign that in an increasingly service-based economy, new challenges lie ahead on international taxation. Now countries can decide to use it or not!

The discussion draft entitled “The Tax Treaty Treatment of Services” includes new paragraphs 42.11 to 42.45 to the Commentary on Art. 5 of the OECD Model. The new paragraphs address the appropriateness of the current provisions of the OECD Model to deal with the tax treatment of services, views (and policy reasons) of States that do not agree with the principle of exclusive residence taxation of services and finally a alternative paragraph for Art. 5 of the OECD Model that secures additional source taxation rights, in certain circumstances, with respect to services performed within the territory of the source State.

For the States that wish to secure additional taxing rights on services, the OECD suggest the inclusion of a new paragraph which operates as an extension of the permanent establishment definition in conformity with the following principles:
- taxation of services should not extend to services performed outside the territory of the source State;
- taxation of services should apply only to the profits from these services rather than to the gross amount of the service fees; and
- taxation of services should involve a minimum level of presence in the source State, requiring for example a certain days of presence and a link between the revenues and the services or type of business activities performed.

The alternative OECD service PE clause included in paragraph 42.23 of the discussion draft, reads as follows:

"Notwithstanding the provisions of paragraphs 1, 2 and 3, where an enterprise of a Contracting State performs services in the other Contracting State
a) through an individual who is present in that other State during a period or periods exceeding in the aggregate 183 days in any twelve month period, and more than 50 per cent of the gross revenues attributable to active business activities of the enterprise during this period or periods are derived from the services performed in that other State through that individual, or
b) during a period or periods exceeding in the aggregate 183 days in any twelve month period, and these services are performed for the same project or for connected projects through one or more individuals who are performing such services in that other State or are present in that other State for the purpose of performing such services, the activities carried on in that other State in performing these services shall be deemed to be carried on through a permanent establishment that the enterprise has in that other State, unless these services are limited to those mentioned in paragraph 4 which, if performed through a fixed place of business, would not make this fixed place of business a permanent establishment under the provisions of that paragraph."

If the conditions of such provision are fulfilled, the service activities are therefore deemed to be carried on through a PE and the profits derived from such activities taxable in the source State pursuant to Art. 7 (business profits).

The proposed provision is different than the UN Model service PE provisio. When comparing the existing Art. 5 of the UN Model (2001) with Art. 5 of the OECD Model (2005), it can be said that under the UN Model there are more situations leading to a permanent establishment than under the OECD Model. This isthe case for example under Para. 3 (b) of Art. 5 of the UN Model, “the term "permanent establishment" also encompasses: (…) (b) the furnishing of services, including consultancy services, by an enterprise through employees or other personnel engaged by the enterprise for such purpose, but only if activities of that nature continue (for the same or a connected project) within a Contracting State for a period or periods aggregating more than six months within any twelve-month period.”.

A simple search in IBFD database retrived almost 500 real treaties with this language. This means that the OECD inclusion in the Commentary of a service PE perhaps is intended to brign the Model closer to developing nations and at the same time clarify some basic principles regarding cross-border services.

Finally, the discussion draft takes the opportunity to include additional language to paragraph 10 of the Commentary on Art. 17, allowing taxpayers to be taxed on their net income (instead of gross profits) as if they were residents.

The Tax Treaty Treatment of Services: OECD Public discussion draft on proposed Commentary changes, released on 8 December 2006

Thursday, December 14, 2006


With the exception of the Parent–Subsidiary Directive, there is no specific EC rules concerning dividend taxation. Therefore, it is reasonable to say that dividend taxation falls under the competence of each of the EU Member States, subject to the increasingly important limits under the fundamental freedoms established in the EC Treaty. As we have seen throughout the extensive case-law of the ECJ in the field of direct taxation, EC law impacts on national tax systems as a result of the combined application of the four freedoms and the prohibition of discrimination and discriminatory restrictions.

The decision of the ECJ in the Denkavit II case (C-170/05), which held that the French withholding tax on outbound dividends is incompatible with the freedom of establishment (Art. 43 of the EC Treaty), is a good example of those limits. Under certain tax systems, outbound taxation of dividends was sometimes distinguished from domestic situations, where dividends were paid and receive by resident entities. In the first case, a domestic withholding tax was levied, while in the later case no withholding tax was levied (in order to prevent for example the cascading of tax through chains of companies). This case simply says that in a EU scenario such distinction may prove to be incompatible with the treaty freedoms.

Dividend taxation under international tax law

The structure of the international income tax, drawn from the so-called international consensus, is based on the assumption that income tax is generally levied (i) on the domestic and foreign income of its residents (residence taxation) and (ii) the domestic income of non-residents (source taxation). Under such system, whilst the residence concept establishes a relationship between a particular jurisdiction and the taxpayer deriving the income, the source concept connects the income itself with a particular jurisdiction.

Under that framework, when considering the taxation of dividends it is then important to distinguish: (i) a withholding tax levied on the dividends paid by the company on behalf of the shareholder at the moment the dividends are paid out; (ii) the taxation at the level of the shareholder receiving the dividends.

Under international income tax law, dividends are usually sourced on the basis of the residence of the company paying them. As such, if a resident of one country earns dividend income from a source in another country, double taxation is likely to arise because one country will tax that income on a source basis (usually through a flat-rate final withholding tax on the gross amount of the dividends) and the other country on a residence basis (usually using a progressive income tax rate scale for individuals or a flat-rate for companies).

In this case, the internationally accepted regime is that the source country has the prior right to tax (although limited by reduced treaty rates), and the residence country is responsible for relieving any double taxation that results. Such relief is generally achieved through the exemption system (whereby the foreign income is exempted from tax in the residence country) or the foreign tax credit system (whereby the tax of the residence country on the foreign income is reduced by the amount of source country tax on the income). Under that model, this case has to be distinguished especially from the so-called economic double taxation, i.e. where two different persons are taxable in respect of the same income or capital.

The domestic withholding tax rate on outbound dividends is typically set between 20%-30%, which is then generally reduced to 5%-15% under the respective tax treaties. A usual feature found in outbound dividend taxation relates to the distinction between direct investment (an investor which has a controlling shareholding) and portfolio investment (where no controlling shareholding exists). This distinction is generally defined through an ownership percentage of the capital (for example the OECD Model uses a 25% ownership test).

It should be noted that in respect of inter-company dividends, many countries have chosen recently in their tax treaties to simply eliminate its dividend withholding tax. This situation is related with the wider trend of lowering corporate income tax rates, inclusion of domestic exemptions on certain outbound payments and the enactment of the Parent-Subsidiary Directive, which was recently extended to Switzerland. Nevertheless, at this stage one can say that withholding tax on outbound dividends is still the rule under treaties. The issue now is to see how the same issue is covered under the EC Law framework.

Outbound dividends in Europe - An (in)complete framework

The area of dividend taxation in Europe, in addition to the bilateral treaties, is fundamentally marked by the existence of a EU directive for the taxation of parent and subsidiary companies. The so-called Parent-subsidiary Directive (Council Directive 90/435/EEC) had an immediate effect on cross-border business transactions in Europe, by providing a comprehensive double tax relief throughout Europe for dividends flowing between companies from different Member States when the companies are in a parent/subsidiary relationship.

The Directive, which deals with issues that were previously the exclusive concern of tax treaties, basically requires that Member States: (i) refrain from imposing withholding taxes on distributions of profits made by subsidiary companies to their parent companies in other Member States; and (ii) to grant parent companies double taxation relief in respect of such income either by exempting it from further tax or by granting relief for the underlying company tax on the profits out of which the distribution is made.

Nevertheless, the scope of the Directive seems narrower (on the first element noted above) than the Dividend article found in tax treaties, since its application is limited to certain types of companies established in accordance with domestic law of the EU member states. Even though there have been recent extensions of its scope, there is a range of situations outside the coverage of the Parent/subsidiary Directive that may need to be assessed applying the fundamental freedoms case law.

Just imagine the treatment of EU inter-corporate dividends paid by a company that does not meet the requirements set out in Art. 2 of the Parent-Subsidiary Directive. For example it may be a dividend paid by a type of company that is not listed in the Annex. In addition, just imagine a payment of inter-company dividends that fails the 20% threshold requirements of Art. 3. On a more extreme scenario, just consider individuals, which are not covered by the Parent-Subsidiary Directive and therefore are required to incorporate their holdings, through a "listed" company, to achieve the same objectives of source taxation minimization.

What happens in these ""fringe" cases? Is the Member State authorized per se to withhold tax in those situations (or not provide relief) or do we have to read these "fringe" cases in conjunction with ECJ case-law on discrimination? But then comes along a French case on inter-corporate dividends, which fortunately covered taxable years when the Parent-Subsidiary Directive was still not in place.

The Denkavit II Decision

This case involved a dividend distribution from an (almost) fully owned French subsidiary to its Dutch parent company, Denkavit International BV. The problem derived from the fact that domestic dividends were not subject to withholding tax and were 95% exempt at the level of a French parent company, whilst, dividends distributed to a foreign parent were subject to a 25% withholding tax. This domestic withholding tax was nevertheless reduced to 5% under the French–Dutch tax treaty. Since, Denkavit International BV was unable to credit the 5% withholding tax because its dividend income was tax exempt under the Dutch participation exemption regime, it decided to claim a refund of that withholding tax on the basis that a EU parent company could not be treated less favourably than a French parent company.

The ECJ basically followed Advocate General Geelhoed opinion and held that France is precluded under EC Law to impose a withholding tax on dividends paid to a non-resident parent company if it provides an (almost full) exemption of dividend withholding tax to French resident parent companies. In addition, the ECJ held that the dividend withholding tax is prohibited, even if a tax treaty between the Source and Residence State provides for the Residence State taxation to be set off against the Source State withholding tax, if the parent company is unable to set off tax in the residence State, in the manner provided for by the tax treaty.

In taking this decision, the ECJ first pointed out that since the case related to years where the Parent-Subsidiary Directive did not apply, only the relevant provisions of the EC Treaty should be taken into account.

In a rather short decision, the ECJ arrived quickly to the conclusion that the French tax system, irrespective of the effect of the relevant tax treaty, gave rise to a difference in the tax treatment of dividends paid by a resident subsidiary (no withholding tax on domestic dividends and 25% withholding tax on outbound dividends) and that such difference constitutes in principle a prohibited restriction on the freedom of establishment.

The ECJ rejected therefore the French arguments based on the non-comparability between a domestic parent and a EU parent and the justification based on the territoriality principle (which was apparently ignored by the ECJ).

As to the controversial comparability aspect, the ECJ referred that as soon as France, either unilaterally or by way of a treaty, imposes tax on the dividend income, not only of resident shareholders, but also of non-resident shareholders, from dividends which they receive from a resident company, the situation of those non-resident shareholders (which have or not a fixed place of business in France) becomes comparable to that of resident shareholders.

The ECJ on paragraph 37 of the decision followed the preposition of the Advocate General and held that since the domestic exemption on dividends is designed to avoid economic double taxation, France is therefore required to extend such a relief (i.e. exemption) also to non-residents, to the extent that similar domestic double economic taxation results from the exercise of its tax jurisdiction over these non-residents. (1)

(1) According to the Advocate General, this follows from the principle that tax benefits granted by the source State to non-residents should equal those granted to residents in so far as the source State otherwise exercises equal tax jurisdiction over both groups.

The ECJ went on to state that the heavier tax burden on dividends paid to Netherlands parent companies (as compared to dividends paid to French parent companies) constitutes a discriminatory measure incompatible with the EC Treaty. In conclusion, France should not levy a withholding tax on outbound dividends.

The second part of the judgment concerned the equally controversial issue of the effects of tax treaties on the compatibility of domestic law with EC law. The issue here was whether a different answer should be given if a tax treaty exists between the source state and the resident state, whereby a parent company resident in the resident state may offset the withholding tax levied in the Source State, but because of the resident State tax system (i.e. participation exemption) such parent company is simply unable to set off the respective withholding tax. In fact, the Dutch participation regime simply prevented the possibility of offsetting the French withholding tax against Dutch corporate income tax, resulting in an excess tax credit of 5%.

The ECJ was again short in its argumentation and held that in such a case, the combined application of treaty and Residence State participation exemption rules does not serve to overcome the effects of the restriction on freedom of establishment and therefore constitutes incompatible discrimination against foreign parent companies. In reaching its decision, the ECJ rejected the French argumentation that under international tax law it is for the Residence State and not for the Source State, in which the taxed income has its source, to rectify the effects of double taxation.

What is (not so) clear from Denkavit II?

The Advocate General Geelhoed intellectual construction, which distinguishes between home and source State obligations, appears to have been accepted by the ECJ. This means that the obligation of the Source State only arises insofar as it exercises its tax jurisdiction over the non-residents. In this case, the source State cannot discriminate between resident and non-resident taxpayers. On Denkavit II, the ECJ appears to oblige the Source State (France) to extend a relief for economic double taxation to the non-resident (Netherlands) equivalent to the relief given in the Source State. But is this case applicable on other types of income?

The particular features of this case appear also to point out that it only applies in circumstances where the Residence State is an exemption country. This link (between the exemption method and the outcome) although not entirely clear may be extracted from paragraph 54, which refers to the combined application of the tax treaty and the relevant domestic legislation (which in that case “does not serve to overcome the effects of the restriction on freedom of establishment”). Is this sufficient to rule out a case under the credit method?

First question: Does the ECJ mean that the case is only applicable to dividend withholding tax situations such as the one found in France-Netherlands? What if the residence state operates through a credit method? What about withholding tax on other types of income?

Another problem, left untouched, is that by doing so (i.e. extending the relief), the Netherlands parent company is slightly better of than a "comparable" French resident company. This is the case, since the exemption in France covers only 95% of the dividend income, whilst the Dutch participation exemption covers the full dividend income (100%). In limit, this should not prevent France to levy a least a withholding tax on the difference (33% of 5%= 1.65).

Second question: Is France then still allowed to tax this residual amount? If yes, at which rate? Treaty rate or CIT rate?

It is needless to say that this decision will have also an impact on Member States that apply a participation exemption regime for domestic parent companies, while applying a withholding tax on outbound dividends paid to EU parent companies, namely when the conditions for exemption under the Parent-Subsidiary Directive are not met. Just imagine, as is the case in various EU member states, that that holding thresholds for the domestic participation exemption and Parent-Subsidiary Directive differ. Nevertheless, governments may easily correct this difference by simply adjusting (upwards) the domestic thresholds or (downwards) the outbound thresholds. The problem here is that experience has shown that the adjustments made as a consequence of ECJ case law have in many instances worsened the position of domestic taxpayers. For example the legislative reaction to the Lankhorst Hohorst case (C-324/00), which found the German rules on thin capitalization in violation of the Treaty, resulted in some cases on the extension of thin-cap rules to domestic creditors.

Third Question: how will governments react to Denkavit II?

Another issue will be, whether such a case would be decided in the same way for portfolio participations that fall under the scope of free movement of capital (Art. 56 EC). The existing ECJ case law seems already to indicate that the same conclusion may be drawn. And what if we are in a third-country scenario? And what if the residence state has a credit system and there is an apparent cash-flow disadvantage on having a withholding tax on the outbound dividend and only later a credit? Here the pending Amurta case (C-379/05) will probably provide an answer in that respect.(2)

(2) The Amurta case, which also focuses on the compatibility of withholding tax on intercompany dividend, deals with the specific case where a company (resident in Portugal) owning 14% of the shares of a Dutch company, receives a dividend from which 25% Dutch withholding tax was withheld. Amurta filed an objection against the levy of withholding tax and argued that such levy violated the free movement of capital as included in Art. 56 of the EC Treaty, arguing that, had the company been resident in the Netherlands, no dividend withholding tax would have been due (based upon the participation exemption rules).

Fourth Question: can we apply the same line of reasoning of Denkavit to cases falling under the free movement of capital? Again, what if the residence state is a credit country and not an exemption?

The ECJ apparently ignored or diplomatically avoided to touch upon the controversial position used by former Advocate General Geelhoed, which distinguishes between “true” and “quasi” restrictions, whereby the latter restrictions fall outside the scope of the Treaty and therefore should only be eliminated by legislative action.(3)

(3) Using the words of the former Advocate General Geelhoed, "Quasi-restrictions result directly and inevitably from the juxtaposition of systems and in particular from: (1) the existence of cumulative administrative compliance burdens for companies active cross-border; (2) the existence of disparities between national tax systems; and (3) the necessity to divide tax jurisdiction, meaning the dislocation of the base."

Fifth Question: does this mean that the distinction between “true” and “quasi” restrictions is still hanging in the air?

In the end, litigation in this area will result in the necessity of symmetry between thresholds of participation exemption. In that regard, another area now left open by this case is its impact on "fringe" cases that fall outside the directive, such as dividends paid by type of companies that are not listed in the Annex of the parent-subsidiary Directive. This discussion of course raises the issue of the relationship between primary (EC Treaty) and secondary Community law (Directives), which is far from clear. One can say, that even if a certain area has been harmonized through a Directive, this does not entail the creation of an invulnerable domain, immune from the influence of the fundamental principles set out by the EC Treaty.

Sixth Question: can we apply the same line of reasoning of Denkavit II to cases outside the subjective scope of the Parent-Subsidiary Directive?

And what about individuals? Is it another ball game altogether as some say or can we also apply the same principles? As the EU Commission rightly puts it in the communication "Dividend taxation of individuals in the Internal Market (COM (2003) 810), "a Member State cannot levy tax a withholding tax on outbound dividends and exempt domestic dividends, as it would tax outbound dividends higher than domestic dividends." Nevertheless, the Commission also alerted to the fact that in assessing the higher burden, a simple comparison of the withholding tax rates is not sufficient. In fact, the basis of comparison should be for the domestic dividends the combined effect of any domestic withholding tax rate plus the domestic income taxation and for the outbound dividend the withholding tax rate on the outbound dividend. Looking at the existing case-law of the ECJ, which has mainly focused on inbound dividends, also here some doubts start popping!

Seventh Question: what about individuals? Is it another ball game altogether?

Any other questions?


Tuesday, December 05, 2006

Letter from Saint Nicholas to the Tax Inspector

Keeping up with the Dutch tradition of Sinter Klaas, which arrives in Mid-November by steamboat with his helper Zwarte Piet (Black Peter), I share a letter from Saint Nicholas to the Tax Inspector.

All credit goes of course to my colleague Dan Geddes also known as “The Satirist”. You can also read last year e-mail sent to Sinter Klaas by an egger US tax advisor.

Irving R. Sheen, Tax Inspector
From: Saint Nicholas of Myra, alias "Sinter Klaas", "Santa Claus"
Taxpayer ID: 000-00-0002
Date of Birth: 5 December, 343 A.D. (12-05-0343)
Occupation: Bishop Emeritus
Subject: Amended Tax Return

Dear Mr. Sheen:

I was greatly upset as I read your last FAX,
that I owed many thousands of dollars in tax,
plus penalties and interest over time that have waxed
into fines more suited to hard criminals' acts.

I have worked very hard to meet your request,
filing a second return was like taking a test.
The time spent was great. I wished to protest.
To gather my documents alone was a Quest.

My faithful assistants, a brigade of Black Petes,
searched high and low for the countless receipts,
stored long in my basement, many millions of sheets!
In order to prove I'm not someone who Cheats.

So I had to obtain rather costly advice
from a tax planning firm (they were all very nice).
They have studied my filing (and, yes, checked it twice),
and now expect from my REFUND a bountiful slice.

They found real problems in your unjust assessment,
forgetting my charity, the great sums that were spent.
And though "substantial" enough, and quite "permanent"
My steamship is really no corporate establishment.

'Twas not wise to cause us this unearned vexation.
For we too collect mountains of information,
books filled with the deeds of the men from all nations,
old scrolls dating back from the time of Creation.

Your own tax forms do not withstand close review,
your errors and omissions number more than a few,
we found that your filing contains claims quite untrue!
It is very disturbing, what we've learned about you.

Your own get-rich schemes are vast and assorted.
You're a far richer man than you've ever reported.
Your investment portfolio looks awfully contorted.
You've hid much of your wealth, or simply "off-shored it."

And your dog does not count as a fourth tax dependent.
Your wages could never buy a house so resplendent.
And where did your wife get that HUGE golden pendant?
You should really reflect on your life's path, and then mend it.

Your finances are more complex than a maze!
Our review of your deeds never ceased to amaze.
But there's one chance you have to change your foul ways.
File a new return in the next ninety days.


Saint Nicholas of Myra
Bishop Emeritus

© 2006 Dan Geddes