OPINION OF MR ADVOCATE-GENERAL LAGRANGE

DELIVERED ON 18 OCTOBER 1960 ( 1 )

Index

	I — Jurisdiction and procedure
	II — Substance
	(a) Arguments of the parties
	(b) Terminology
	(c) Legal discussion
	(d) Ratio Legis
	(e) General considerations

Mr President,

Members of the Court,

In a few words I shall first go over the facts of the case which are very simple: Mr Humblet, of Belgian nationality, is an official of the High Authority of the European Coal and Steel Community. Although he is employed in Luxembourg where he resides with his wife, he is regarded as having retained his domicile for tax purposes (domicile fiscale) in Belgium where he also maintains a residence and where his wife receives income: this much is common ground.

The income of Mrs Humblet, which was duly declared, was subjected in Belgium to the personal surtax (impôt complementaire personnel) in the name of her husband as head of family in accordance with the law. Nevertheless in 1959, changing their previous practice, the Belgian fiscal authorities requested Mr Humblet to declare the amount of the remuneration which he received as an official of the High Authority and which was exempt from taxation under the Protocol on the Privileges and Immunities of the Community. The authorities wanted in fact to take the amount thereof into account in arriving at the income of the spouses in order to determine the rate of tax applicable although tax was subsequently to be imposed only on that proportion of the income which was not exempt, in this case the income of the wife. The applicant, Mr Humblet, refused to comply and was issued with an estimated assessment in respect of the years 1957, 1958 and 1959 (for the income in the years 1956, 1957 and 1958) and to this assessment were added what in Belgium are called ‘surcharges’, a term which appears to correspond to what in other countries are less delicately called penalties. The objection which he lodged in accordance with the proper procedure against this assessment was rejected and proceedings in the matter are at present pending before the Cour d'Appel.

Alongside these national proceedings Mr Humblet considered himself entitled also to bring the matter before the Court of Justice in application of Article 16 of the Protocol on the Privileges and Immunities which as you know provides that:

‘Any dispute concerning the interpretation or application of the present Protocol shall be submitted to the Court’.

The High Authority has not intervened and is not involved in the proceedings.

The case raises several important and difficult questions concerning jurisdiction, procedure and the interpretation of the Protocol.

I

I shall first examine the questions of jurisdiction and procedure.

The Belgian State, the defendant, raises first an objection as to the Court's lack of jurisdiction. The Belgian State maintains that in the present instance the case relates solely to the application of national law which is within the exclusive jurisdiction of the national courts. On the one hand indeed the dispute concerns solely the income of Mrs Humblet who is not an official of the Community and to whom therefore the Protocol can in no way be applied. On the other hand the only question at issue concerns the method of calculation in conformity with the Belgian tax law, of the tax on the income which is not exempted, that is to say, income which is exclusively covered by this tax law. The defendant further observes that questions of a similar nature have already been or are in the course of being raised before national courts which have not up to now found it necessary to rule that they have no jurisdiction and to refer the matter to the Court of Justice.

I do not share this opinion.

First of all I should like once and for all to dismiss all arguments raised, one way or the other, to the effect that what is involved is the income of the wife of the official: according to the law, this income is not merely assessed on the husband (where there is no separate assessment) but is also aggregated with that of the head of the family as though both incomes arose from a single source.

It is self-evident that the tax authorities cannot disavow the existence of the husband so as to assert that the income of the wife, who is not an official of the ECSC, may under no circumstances benefit from an advantage accorded by the Treaty — and in the same breath put forward the existence of that same husband (and his exempt salary) as a reason for increasing the tax relating exclusively to the income of the wife! As against this Mr Humblet cannot rely on the fact that the income assessable on him is the income of his wife. In my opinion the various problems raised by the present case arise in exactly the same way whether they concern the income of the taxpayer himself or the income of members of his family assessed on him.

Having said this it appears to me to be obvious that the case now before the Court relates both to the interpretation and the application of one of the provisions of the Protocol, namely Article 11 (b): we are essentially concerned with the question whether the provisions of this subparagraph — in particular the words ‘shall be exempt from any tax on salaries etc’ must be interpreted in the broad sense as prohibiting account being taken of the exempt emoluments either for determining the taxable income or for the calculation of the amount payable in respect of a personal tax on income on a rising scale or whether, on the contrary, the word ‘exempt’ (exonéré) on a strict interpretation must be regarded as merely prohibiting the application of the rate to that part of the income which corresponds to the amount of salary although the rate itself may be determined after taking account of the whole income including the exempt salary. This question is certainly within the jurisdiction of the Court. The Court must not involve itself in the application or interpretation of national tax law; it must take that law as it stands with the wording, the principles on which it is based, and where necessary the national case-law to which it has given rise and, in these circumstances, by interpreting the Protocol the Court determines the scope of the exemption, that is the extent to which the exemption derogates from national law.

Two further questions appear to me to be more difficult: the first is whether and in what circumstances an official may apply directly to the Court of Justice; the second is what are the powers of the Court in this respect and what is the scope of its judgment.

As to the first question it may be pointed out that Article 11 of the Protocol establishes by way of the legislative process a number of privileges and immunities for Members of the High Authority and officials of the Community. In this respect the situation appears to differ from that where the privileges and immunities are stipulated in a local agreement (accord de siege) concluded between an international organization and the host state: in this case any difficulties which may arise as to the scope of any provision contained in the agreement are problems between the host state and the organization which are customarily regulated according to a special procedure set up by the agreement itself: such a procedure (for example arbitration) may only be set in motion by the organization or by the State, that is to say, by the parties to the agreement.

Here on the contrary it is a provision of the Protocol, which has the same force as the Treaty and creates a right for the benefit of officials. It is true that according to Article 13 of the Protocol, privileges, immunities and facilities are granted to Members of the High Authority and to officials of the institutions of the Community ‘solely in the interests of the Community’; however, this provision, which explains the reasons for establishing privileges, cannot prevent persons to whom they were granted from acting to defend them. The general nature of the terms used in Article 16 allows such persons, in my opinion, to apply directly to the Court if (and this is the only condition which may be required) there exists a ‘dispute’ concerning the interpretation or application of the Protocol.

In the case of fiscal privileges I think that the existence of a ‘dispute’ is sufficiently established where, as in the present case, the party concerned is in disagreement with the tax authorities and he has brought his appeal before the competent national authorities in the proper manner.

It is true — and the defendant was at pains to emphasize this — that no procedure for reference to the Court of Justice for a preliminary ruling was provided for. This is explained by the fact that disputes liable to arise on the application of the Protocol are of a widely divergent nature and may lead to very different procedures and may not always arise in cases brought before a national court.

Having said this I think — and this is a personal opinion — that the national courts before which a question is raised of interpretation or application of the Protocol which may fall within the jurisdiction of the Court of Justice would be well advised — even if they did not regard themselves as obliged — to order reference to the Court of Justice for a preliminary ruling either at the request of one party or even of their own motion: the jurisdiction assigned to the Court is necessarily exclusive of their jurisdiction and the lack of express provision of procedure for this eventuality should not constitute any obstacle to a reference.

However this may be, if the competent national court did not regard itself bound to order this reference there is no reason why the party concerned should, not himself bring the matter before the Court of Justice. Article 16 in no way prevents this:

‘Any dispute concerning the interpretation or the application of the present Protocol shall be submitted to the Court’.

The dispute can be just as well submitted to the Court by the party concerned himself, if he can prove that there exists a dispute, as by a national court ordering a reference for preliminary ruling.

These few considerations will be of assistance in resolving the second question relating to the powers of the Court of Justice and the scope of its judgment.

The Court clearly has jurisdiction to interpret the provisions of the Protocol with regard to the facts of the case. However, the question arises whether the Court can go further. Has the Court, as the applicant maintains, the power to make an order affecting the national authorities, that is to say in the present case, the power to order the discharge or the reduction of the contested tax and to order that the consequential relief be given? In my opinion it certainly has not; that would be a clear incursion into the jurisdiction which the national courts have retained: the Court may not substitute its judgment for that of the authorities or of the national courts acting within the scope of the national fiscal legislation.

It is true that under Article 16 disputes which must be referred to the Court are not only those which relate to the interpretation of the Protocol but also those relating to its application. However, in my opinion the significance of this provision must not be exaggerated. It is intended to enable the Court — and this is very important — to give a ruling in cases where the difficulty does not relate to a question of the interpretation of a provision of the Protocol which is obscure or regarded as such but to the conditions for its application or for refusal to apply it. However, once again the Court may not substitute itself for the national authorities and still less so for the national courts and may not exercise their powers in their stead.

I believe that in cases such as the present where the Court of Justice becomes involved while proceedings are taking place in the national courts and where the outcome of the case depends, at least in part, on the judgment of the Court, the procedure should be the same as that for references for a preliminary ruling. In other words it is for whichever party sees fit to produce the judgment of the Court of Justice before the national courts which must, in their turn, deduce the legal consequences for their own decision under national law. The present case is comparable to that which arises where, for example, a question of nationality is raised in the course of proceedings and the party concerned produces in evidence a judgment which that party itself obtained on the point but which was not given following a reference by the court hearing the main action: certainly the latter must determine whether the judgment produced to it was properly given by the competent authority, whether it is a final judgment etc. However, once the necessary verifications have been carried out it must defer to the force of res judicata on the point which was outside its own jurisdiction and draw the appropriate conclusions for its own decision. In my opinion this is the appropriate procedure in the present case.

II

I now come to the substance of the case, that is to say, to the extent of the exemption afforded by Article 11 (b) of the Protocol when it is applied to a personal tax on income levied on a rising scale. We are fundamentally concerned in this respect with determining the meaning of the words:

‘the Members of the High Authority and officials of the Community … shall be exempt from any tax on salaries and emoluments paid by the Community.’

The Court has heard the two views; weighty arguments have been submitted in support of each.

The applicant states that the text is formulated in a general manner and that the system adopted by the Belgian authorities results in the partial taxation of the salary by fixing a rate taking account of the whole of the income including the exempted salary. This, he claims, constitutes an infringement of the Protocol.

Not at all, replies the defendant: the only income which is taxed is that which does not benefit from the exemption. Thus the provisions of the Protocol are complied with. A distinction must, in fact, be made between ‘exempt’(exonéré) income, that is to say income which cannot be subject to taxation, and ‘immune’ (immunise) income, that is to say income which is not taken into account in determining the taxable income. If the authors of the Protocol had intended to adopt the second alternative they would have used different wording. In addition the exemption is personal: it is the official who is exempted not the salary. Provisions relating to fiscal exemption must be strictly interpreted as also must international provisions, in particular those establishing privileges or immunities; in cases of doubt the restrictive interpretation must prevail. This is all the more true in the present case as the ‘immunity’ (immunisation) would strike at the very foundations of the tax in question which is a personal tax on a rising scale taking account of the ‘taxable capacity’ (faculté contributive) of the taxpayer which depends on his total income: to assess the income which is not exempt at a low rate as though the taxpayer had no other source of income would be extremely unjust and, in the absence of a provision expressly requiring such action, it must be rejected. In support of its view the defendant further relies on a number of judgments given both by Belgian courts and by foreign courts, in particular the Federal Court (Tribunal Federal) of Lausanne; he further relies on the practice followed for the application of conventions for the avoidance of double taxation: here the exempt income is nevertheless taken into account to determine the taxable income, in particular to determine the rate of tax. However, this rate is not applied to the amount of income which has been exempted once it is proved that this income was subject to tax in the country of its source.

The applicant replies to all these arguments by contesting the value of a distinction between ‘exempt’ (exonéré) income and ‘immune’ (immunise) income, by pointing out that the Protocol established a privilege which cannot be reduced to a simple arithmetical process by denying that the national law makes any distinction between exempt persons and exempt income, by showing the difference which exists between a privilege creating an exemption and the application of double taxation conventions and by attempting to prove that the system adopted by the Belgian tax authorities does in fact result in taxing the remuneration which he claims to be exempt.

As regards this dispute, I believe, after weighing up all the arguments, that the applicant's view is the correct one.

First, what is to be made of the distinction between ‘immune’ income which is merely ‘exempt’? It is clear that this distinction may, in itself, have some meaning if it genuinely reflects the distinction between the two systems defended by the respective parties. It is conceivable that, in the context of a system of personal taxation on income on a rising scale, an individual source of income may be taken into account in determining the total income subject to tax, in particular for determining the rate of tax, but may subsequently be relieved of the application of this rate which, however, remains applicable to income from other sources. On the other hand it is conceivable that income from a particular source may not be taken into account in determining the taxable income; it is solely for the legislature dealing with taxation to determine these matters. However, the question which concerns us is whether, in ordinary tax terminology, the first system is characterized necessarily, or at least usually, by the use of the word “exempt” (exonéré) while for the latter system the term “immune” (immunise) is exclusively used. It would ev.en be necessary that such a distinction between immunity and mere exemption be so enshrined in international tax terminology that the authors of the Treaty could be regarded as having chosen the former system and rejected the latter solely by reason of their having used the word “exempt” and not the word “immune”.

This however is by no means so.

It is not so even within the national terminology. In his reply the applicant showed by various examples that the Belgian legal language whether it comes from the pen of the legislature, of the tax authorities or of the courts, uses in the same sense the term “exempt” (exonéré, exempté) or “immune” (immunise) income and that the same applies in the Dutch text. The defendant recognizes this fact in its rejoinder and does not pursue its arguments in this connexion.

The same applies in France with the difference that the legislature rarely uses the word ‘immune’ (immunise). Its most apt method, when it decides that income of a certain kind is not to be taken into account in determining the taxable income, is to state in full: ‘The following shall not be taken into account in determining the total net income …’(‘N'entrent pas en compte pour la determination du revenue net global …’) followed by a list: this is the form of words used in Article 157 of the General Tax Code (Code General des Impôts). However, as it is difficult to repeat this rather long sentence each time, the legislature often uses the single word ‘exemption’ or ‘exempt’ (exoneration or exonéré) to signify the same thing (cf. Article 158(3); Article 159 (1) and (2); Article 159 bis); sometimes the two terms are used in the same sense in a single article (Article 157 (11))! The meaning is always identical: the income declared ‘exempt’ is income which is not taken into account in order to determine the total net taxable income. Thus, looking at only one of the languages of the Community, which is both one of the national languages of Belgium and the national language of France, we find that there exists no uniformity of terminology either between the two countries or even within each one of them. We even find that the word ‘exempt’ applied to income is frequently used in the two countries to designate income which is not taken into account to determine the taxable income. These few remarks, restricted to two countries of the Community which, however, do have undeniable similarities in the field of taxation, are sufficient to show that the word ‘exempt’ cannot by itself be regarded as showing clearly the intention of the authors of the Treaty to permit, despite the exemption granted, the inclusion of the exempt salary as part of the taxable income for the purpose of calculating tax in so far as such tax falls on other sources of income; the contrary is true. Finally, it must be borne in mind not only that the authors of the Protocol had to legislate for six countries and therefore were unable to rely on the technical terminology of tax law, uncertain as we have seen it to be, but that the exemption provided by them should apply to all taxes on salaries and not merely the general surtax on income. They found the term which was at once the most simple and the most general.

As we have said the defendant did not pursue this matter in its reply but relied mainly on the argument that the text of the Protocol exempts the person and not the property. It is the Members of the High Authority and the officials of the Community who are exempted and not the salaries.

Although I have given the matter considerable thought, I am unable to grasp this distinction or its extent. I am familiar with a distinction in tax matters between a personal tax and a proprietary tax. We are concerned here with personal taxes. However, as for all personal taxes, there is a subject and an object: the subject is in this case the official carrying out certain duties; the object (the ‘property taxable’ (‘matière impossible’) is the salary he receives. What is the difference between wording such as ‘the salaries and emoluments paid by the Community to its officials shall be exempt from all taxes’ and that of the text ‘the officials of the Community shall be exempt from all tax on salaries and emoluments paid by the Community’? If one wished at all costs to discover a distinction between the two forms of wording I believe that the text which places the accent on the personal nature of the exemption tends more than the other to remove the person entitled to the exemption from any ‘contact’ with the tax law in respect of emoluments which he receives from the Community.

I shall now leave the question of terminlogy in order to attempt to consider whether the system advocated by the Belgian authorities can be regarded as being within the normal scheme of an aggregate personal tax on income. We have seen that in practice there exist numerous cases in Belgium and in France and most certainly in other countries of the Community where certain income is not taken into account for determining the total income. Wehavealsoseen thatthe word ‘exemption’ (exoneration) isoften used to describe this benefit; this may be wrong but it is done in practice. On the other hand it appears extremely rare — and apart from the case of double taxation to which I shall return I am not even sure whether it exists — to find an example where the legislature decided to include certain income fictitiously in the total taxable income for personal tax and at the same time provided that the rate was not applicable to it. However, one does fi nd, at least in those countries where the personal tax is in the nature of an additional tax as compared to tax on various sources of income, the example where the tax exemption is restricted to the special tax(‘scheduled’ tax (impôt ‘cedulaire’) according to the old French terminology) and is not applied to the surtax. This is the situation in Italy in particular where there even exists a general provision under which, save where the contrary is expressly provided, income exempted from the special tax to which it is normally subject is nevertheless taken into account in determining the taxable income for the surtax. This rule which is by no means abnormal, is of no interest to us as the exemption established by the Protocol applies to all taxation, including therefore personal taxes on income whether or not they are additional taxes.

Thus there normally exist only two situations: either the income is taken into account for determining the total income or it is not taken into account for this purpose and in the latter case it is usually said that the income is ‘exempt’ (‘exempte’ or ‘exonéré’). Occasionally it is taken into account in part (cf. Article 158 (5) of the French General Tax Code (Code General Francais des Impôts).

This is quite understandable. Indeed in all personal taxation on income on a rising scale there exists a close link between the determination of the total tax able income and the fixing of the rate. This link is of the very essence of such a tax which seeks to take account, by means of a tax, the rate of which increases proportionately to the income, of the taxpayer's taxable capacity (facultés contributives). The tax is levied not on the sum of a certain number of incomes from different sources (that is the function of scheduled taxes where they exist)but the total income. Doubtless the sovereign legislature may, for reasons of fairness or the general interest, decide not to subject (I am avoiding the word ‘exempt’) a particular category of income to tax. In this case the income in question is not included in the amount of the total taxable income because if it were so included, even if it were subsequently not itself charged to tax, the amount of the tax calculated at a higher rate would, in view of this income's being taken into account, necessarily be levied thereon at least in part. If the legislature seeks to steer a middle course it may provide that the income in question should only be taken into account as to a certain part thereof for the purpose of determining taxable income (I have given one example of this procedure) but in every case all the taxable elements must be added together in order to obtain the total: at this moment both the origin and the nature of each source of income are lost sight of; what is done is to apply purely and simply the rates laid down by the law with any appropriate reductions or reliefs, taking account of the personal position of the taxpayer, such as those in respect of family responsibilities.

Another factor which confirms the abnormal nature of a system such as the one applied in the present case is the calculations to which it leads. In Belgium and also in France, there exists a system of bands (tranches) each of which is subject to a specified rate and these rates rise progressively. Of course these bands are ‘not identifiable’ (‘anonyme’) if I may express it in that way; they do not correspond to particular types of income; this is merely an arithmetical device to facilitate a progressive increase in tax. What course is to be followed in these circumstances? Is the exempt income to be placed in the lower bands or in the higher bands or in the middle bands? If I have properly understood the system it appears that one first of all calculates the rate of the tax as though the exempt remuneration were taxable, that is by applying the successive rates corresponding to the different bands; subsequently from the figure so ascertained a deduction is made proportionate to the amount of the exempt remuneration. Such procedure is perhaps fair but it is certainly arbitrary.

The question arises whether this is compatible with Belgian law. This court does not have jurisdiction to decide that question. I may merely point out that in France the intervention of the legislature was deemed necessary, or at least preferable, to establish a system of the same kind. This is contained in Article 99 of the Finance Law (Loi de Finances) of 26 December 1959 to which reference was made in the course of submissions.

However, all that need be borne in mind as a result of this examination is that the procedure in question is in itself contrary to the very nature of a personal tax on total income as it exists in each of the six countries of the Community whether this tax is a single tax (as in France since a short time ago, in Germany, in the Netherlands and in Luxembourg) or whether it is of the nature of a tax additional to other taxes on individual sources of income (Belgium and Italy). Normally exemption from tax granted in respect of an individual source of income where the exemption is applicable to the total personal tax, means that the income in question is not taken into account in determining the aggregate income for the purpose of this tax. There is therefore no reason for thinking that the authors of the Treaty intended to decide otherwise when they established the exemption in respect of salaries received by officials of the Community.

It is now simple to dismiss the line of argument which the defendant derived from the application of the conventions on double taxation. As was quite correctly observed in the course of the submissions, exemption in such cases is by no means intended to set upa privilege in favour of an income of a particular nature or from a particular source by relieving that income of taxation or even, as in the present case, of all taxation, but merely to avoid taxing the same income twice. The taxpayer benefiting from this measure must not profit from it in order to escape from the ordinary application of the general tax legislation to which he issubject. The judgment of the Federal Court (Tribunal Federal) of Lausanne which is contained in the file of the case is quite significant in this respect: the taxpayer concerned whose domicile for tax purposes was in Switzerland had received certain income in Germany which had been assessed to tax in that country but which, apart from the convention, would also have been assessable to tax in Switzerland. The court held that in spite of the exemption granted by the convention, the income in question had to be taken into account in determining the rate applicable to other income which in Switzerland remained subject to a tax determined on the basis of the total income (in that case the national defence tax). It is clear that the revcerse procedure would have resulted in giving more favourable treatment to a taxpayer who receives income abroad than if he had received the same income in his own country; this would be contrary to the object sought by the conventions on double taxation: avoidance of double taxation must not have the effect of creating a privilege. In addition the most recent conventions of this kind expressly settle the question in this way if one judges for example on the basis of two which have come before us: Article 19 of the Franco-Luxembourg Convention of 1 April 1958(Journal Officiel de la République Française of 11 April Bulletin Législatif Dalloz 1960, p. 300); Article 19 of the convention between France and Finland of 25 August 1958(Journal Officiel de la Republique Française of 27 August 1959, Bulletin Legislatif Dalloz 1959, p. 1107). One can also compare the compensatory measures which have to be used in countries such as Germany where there exists a single tax on income and where, however, certain categories of income such as wages have already been taxed by deduction at source.

In the present case we are not concerned with the avoidance of a double imposition of tax but with creating what in international language is called a ‘privilege’ but which is in reality nothing more than a straightforward exempt ion: there appears to be no reason why the interpretation should differ according as theexemption is provided for by national law or by an international treaty which moreover has been duly incorporated into the internal national legislation as a result of its ratification.

Following this purely legal discussion it may also be of some value to carry the examination on to a broader plane and to examine why the exemption was established.

In my opinion it was not established to ensure the independence of officials vis-à-vis the State of which they are nationals (although this idea was perhaps not entirely foreign to the establishment of the exemption) but chiefly to guarantee real equality of remuneration to officials of the Community whooc-cupy the same posts in the same circumstances. A general principle to this effect governs the whole administrative and financial sphere of the Community with regard to the status of its officials: the absolute equality of remuneration without account being taken of the sometimes substantial differences which, as the Court is aware, exist between Member States as regards the level of salaries and wages. It is possible that the national of such one particular State may find it more profitable to come to Luxembourgor Brussels than the national of some other State who has the same qualifications: this is of no importance. Apart from the extreme difficulties entailed by its application and in view of the difficulties of establishing the bases for comparison in such a sphere any differentiation based on such criteria would have looked like a form of discrimination fundamentally incompatible with the very concept of the Community. It is clear that the subjection of the salaries of officials of the Community to necessarily differing national taxes would in fact have breached this equality. This would come about not so much because of or mainly because of, the differences in rates which exist between the taxes of the six countries but rather by reason of the divergences between the tax systems themselves and the innumerable inequalities which their simultaneous application would necessarily have introduced between officials in the same situation within the Community. In this sphere we are still far from unification or even from mere ‘harmonization’, the more modest goal which is one day to be achieved by the European Economic Community.

First as I have already pointed out, two countries out of the six (Italy and Belgium) have the two-tier system whilst the four other countries have only a single tax on the income of physical persons (although it is true that the Netherlands also has a tax on salaries and wages which, however, is of a territorial nature). The exemption necessarily applies to the tax on salaries and wages and it follows that, on the assumption that the total tax charge should be equal in the different Member States, the effects of the system adopted by the Belgian authorities would in principle be felt more strongly in those of the States where scheduled tax does not exist.

Having said this I must say that various broad principles in mattersof taxation are common to the six countries, in particular in the sphere which concerns us here: that of personal tax on income.

In a general sense there is differentiation according as the taxpayer has or has not what is best called a ‘domicile for tax purposes’ (‘domicile fiscal’) in the country concerned. In theory persons having this domicile for tax purposes, whether they are nationals or aliens, are subjected to what in Germany is called ‘unrestricted taxation’ (unbe-schrankte Steuerpflicht), that is to say taxation in respect of the aggregate of their income including that from a source outside the country. On the contrary those persons who do not have a domicile for tax purposes are only subject to tax on that part of their income which has its source in the country in question.

However, this principle is far from being uniformly accepted. Thus in Italy, Italian citizens and aliens residing in Italy are only assessable to surtax on income arising abroad in respect of that part of such income which they enjoy in Italy.

Sometimes nationality appears alongside the concept of domicile for tax purposes. Thus in France whilst French nationals domiciled for tax purposes in France are subject to tax on all their income even if this is received abroad and they can only escape this with the help of a convention for the avoidance of double taxation, an alien domiciled for tax purposes in France canavoid the same tax merely by establishing that the income which originated abroad has in fact been taxed in its country of origin (Article 164 of the General Tax Code (Code General des Impôts)). The criterion of nationality was also taken into consideration in drafting Article 99 of the French Law of 26 December 1959 to which I have referred above and which contains the same rule as that advocated by the Belgian authorities without the benefit of express provision to that effect: this law is only applicable to international officials of French nationality.

Finally, even where there does exist a common rule it may be differently applied. This is the case of Belgium as compared to France and here the difference affectsofficials of the Community directly. The French authorities consider that international officials, even if they are of French nationality, normally have their domicile for tax purposes at the place where they hold their post and where they are obliged to reside: that is ‘the centre of their interests and of their business’ (centre de leurs interets et de leurs affaires) according to the established formula defining the domicile. When they hold a post abroad these officials are therefore not taxed in France save in respect of their income arising from a French source. On the other hand — and confirmation of this is provided by the present case — the Belgian tax authorities consider that an official of the Community employed in Luxembourg has retained his domicile for tax purposes in Belgium provided that he has a residence there.

These anomalies may be further aggravated since the entry into force of the Treaties of Rome by reason of Article 13 of each of the Protocols on the Privileges and Immunities whereby, as the Court is aware, officials of the Community are deemed to have retained their domicile for tax purposes in their country of origin provided that it isoneofthe Member States. If this provision were to be regarded as applicable to the common institutions, the result would be that a different tax system would be applied for example to two French officials working in Luxembourg, one at the Court of Justice or at the European Parliamentary Assembly and the other at the High Authority!

It would be easy to give many examples of anomalies (the last of which is, it is true, caused in part by the lack of uniformity between the Treaties of Paris and of Rome). 1 have given a few to show that the solution of total exemption is the only one which enables the principle of equality of remuneration to which I referred above to be adhered to: until the tax rules have been unified or at least closely coordinated any national intervention exclusively in matters of tax, however laudable the aim, can only affect the rule of equality for the worse. It would also indirectly but none the less certainly affect the powers of the Community authorities who are the only bodies with the power to fix the remuneration of their servants. Finally, although this is clearly no more than an argument of convenience, it creates the risk of hindering, more than is generally appreciated in the Member Countries, the successful recruitment of officials of the Community.

In conclusion may I say only that I can well understand the preoccupations and inclinations which influence national authorities be they administrative — and not only tax authorities — or judicial and which, I must admit, often reflect the preoccupations and inclinations of part of public opinion. In the same way as public opinion these authorities are quite rightly struck by the difference which exists with regard to tax between the position of national officials and that of international officials and this difference appears all the more disturbing to them where the international official is a national of the State where the institution for which he works has its seat (or is situated at the given time).

However, seen from the point of view of the international organization, the question takes on a totally different aspect: from this angle adherence to the principle of absolute equality between officials of the institution whatever their nationality and their origin is of supreme importance.

Thus there exists a conflict between national interests and the interests of the Community. Doubtless that is the reason why the authors of the Treaty gave jurisdiction to settle disputes which might arise in this respect to the Court of Justice, one of the essential roles of which is specifically to arbitrate in conflicts between the interests of the Community and interests of the Member States. It is a task of this nature which the Court is called upon to carry out in this case. I believe that even if the legal arguments which I have set out are not entirely convincing or if a doubt remains in your minds, the over riding necessity for adhering to the full to the principle of equality between officials of the Community should take precedence over national preoccupations, however legitimate they may in some ways be: this idea certainly lies behind Article 11 (b) of the Protocol.

In fact the only means of giving recognition — although I admit that it is only partial — to these national preoccupations is to introduce as soon as possible and on just principles the Community tax provided for in Article 12 of both the EEC and Euratom Protocols.

My opinion is therefore as follows:

—

The provisions of Article 11 (b) of the Protocol on the Privileges and Immunities of the Community should be interpreted as meaning that they prohibit account being taken of the salary and emoluments paid to officials of the European Coal and Steel Community for determining taxable income and for the calculation of any personal tax on income assessed on the official concerned;

—	The further conclusions contained in the application should be rejected;

—	The costs should be borne by the Belgian State either in whole or in part; this matter I leave to the discretion of the Court.

---

( 1 ) Translated from the French.