Home

Court of Justice 28-01-1999 ECLI:EU:C:1999:40

Court of Justice 28-01-1999 ECLI:EU:C:1999:40

Data

Court
Court of Justice
Case date
28 januari 1999

Opinion of Advocate General

Jacobs

delivered on 28 January 1999(*)

In this case, the Court has been called upon once again to rule on the compatibility with Community law of certain aspects of the French road tax system.

The case concerns both the relationship between the coefficients in the tax bands and the method by which fiscal horsepower is determined. Mr Tarantik, plaintiff in the main proceedings, claims that the progression coefficient between on the one hand the top three tax bands, covering vehicles of more than 18 CV, all of which are in fact imported, and the band covering vehicles of 15-16 CV, most of which are manufactured nationally, results in discrimination contrary to Article 95 of the Treaty. The progression coefficient is the factor of progression from one tax band to the next; it is equivalent to the proportion of the coefficient for one tax band to the coefficient for the band below. In addition Mr Tarantik claims in effect that the method which is used to determine the fiscal horsepower of his Jaguar is contrary to Article 95. That method is laid down in an administrative circular dating from 1956, which applies in certain limited cases and differs from the standard method laid down in an administrative circular dating from 1977.

Before turning to the facts of the case I will summarise the relevant French legislation. I would point out — and I will return to this issue — that, since the order for reference contains virtually no information about the content or the working of the national rules on road tax, the account which follows has been almost entirely gleaned from the Court's earlier case-law(*) and the written observations submitted to the Court.

The national legislation

Articles 1599 C to 1599 J of the Code General des Impôts (General Tax Code) lay down the basic provisions governing the differential road tax. The structure of the tax is established at the national level but it is levied by each département. Every type of vehicle approved for the French market is given for tax purposes a ‘power rating’, or fiscal horsepower, calculated according to a complex formula, to which I will return. The power ratings are divided into bands, each containing two or three ratings. A coefficient is fixed at the national level for each tax band. Each year a basic rate of tax is fixed for each département by its own authorities. The basic tax thus fixed is multiplied by the coefficients to determine the amount of the differential tax payable for the corresponding tax bands. Thus, the amount of tax payable on any given vehicle may vary from one département to another but the structure of the tax is established nationwide and is the same throughout the country.

Two successive administrative circulars govern the method of calculating the fiscal horsepower of private cars: the circular of 28 December 1956 and Circular No 77-191 of 23 December 1977. They were given force of law, with retroactive effect, under Article 35 of the Finance (Amendment) Law for 1993.(*)

The formula laid down by the 1956 circular for calculating the fiscal horsepower value is P = KnD Lw. Ρ stands for the power rating, η for the number of cylinders, D for the bore in centimetres, L for the stroke in centimetres, and ω for the rotation speed in revolutions per second, while Κ is a numerical coefficient.(*) It is accepted by Mr Tarantik, the French Government and the Commission that the result of the application of that formula to vehicles is exclusively a function of the engine's cylinder capacity.

A new formula, which originates from the 1977 circular, came into effect on 1 January 1978. Under that formula Ρ = m (0.0458 χ C/K)1.48, where P stands for the power rating, m is 1 for petrol and 0.7 for diesel fuel, C is the cylinder capacity of the engine expressed in cubic centimetres, and K is a parameter representing the transmission of motion obtained by calculating ‘the weighted arithmetic mean of the speeds expressed in kilometres per hour which would in theory be attained by the vehicle at an engine speed of 1 000 revolutions per minute for the various ratios of the forward gears of the gearbox’. The new formula was intended to counteract the perceived effect of the formula in the 1956 circular, which depended directly on cylinder capacity, by encouraging cars with quieter engines and optimum consumption. The formula was thus designed to obtain a power rating for tax purposes which better reflected the intrinsic characteristics of the car's fuel consumption for a given mileage and encouraged efforts to reduce such consumption.(*)

The 1977 circular originally applied to most private cars(*) granted type-approval on or after 1 January 1978. ‘Type-approval’ refers to the approval given to a model on application by the manufacturer; models which are not marketed in France by the manufacturer must in contrast be approved on an individual basis. The field of application of the 1977 circular was extended by Circular No 87-56 of 24 June 1987 to individually approved private cars either matching or considered to be equivalent, as regards calculation of the fiscal horsepower, to a model granted type-approval whose fiscal horsepower value had been calculated in accordance with the 1977 circular. Other vehicles remained subject to the 1956 circular.

It is thus apparent that two separate methods of calculating power ratings have existed in parallel in France since 1 January 1978. It appears that the vast majority of private cars granted approval in France are covered by the formula laid down by the 1977 circular. Nevertheless, the formula laid down in the 1956 circular continues to apply to:

  • private cars granted type or individual approval before 1 January 1978;

  • private cars granted individual approval before 24 June 1987;

  • private cars presented for individual approval on or after 24 June 1987 which neither match nor are considered to be equivalent, as regards fiscal horsepower, to a type-approved model whose power rating has been calculated in accordance with the 1977 circular;

  • private cars imported by type on or after 1 January 1978 where a comparable model was approved before 1 January 1978;

  • private cars whose specifications (number of seats, type of engine and type of transmission) are not within those covered by the 1977 circular.(*) This category includes cars with a manual gear box comprising more than five gears and cars with an automatic gear box comprising more than four speeds. Thus, for example, recently developed cars with six gears are subject to the 1956 circular; the Commission stated at the hearing that most such cars are imported and that Article 169 proceedings against France in respect of this aspect of the road tax are envisaged.

The criteria by which the administrative authorities determine whether individually approved cars match or are considered to be equivalent to a model granted type-approval are not set out.

According to the Commission, application of the 1977 circular generally results in a fiscal horsepower which is lower by some 2 CV than that arrived at by application of the 1956 circular. The effect of that difference is particularly marked for vehicles whose fiscal horsepower would be between 21 and 22 CV in accordance with the 1977 circular and above 23 CV in accordance with the 1956 circular. The fiscal horsepower of Mr Tarantik's Jaguar is 24 CV in accordance with the 1956 circular and would be 22 CV in accordance with the 1977 circular. The amount of road tax payable by Mr Tarantik is 50% higher than the amount which would be payable if his car had a fiscal horsepower of 22 CV. Moreover, according to the Commission — which was not contradicted on the point — motor insurance premiums are calculated on the basis of the car's fiscal horsepower.

It may be noted that Circular No 98/58 of 3 July 1998 governs the determination of the fiscal horsepower of new vehicles sold on or after 1 July 1998. The new calculation is based on two variables: the power of the engine, expressed in kilowatts, and carbon dioxide emissions expressed in grams per kilometre. The equation thus takes no account of cylinder capacity, transmission or type of fuel.

The facts and the main proceedings

Mr Tarantik is the owner of a Jaguar car with a fiscal horsepower of 24 CV which was first put on the road on 11 April 1979. In February 1996 he was reported by the police for failing to display a road tax disc for 1995-1996. In May 1996 the Principal Tax Inspector for Bussy Saint-Georges sent Mr Tarantik a notice for payment of FF 19 325, consisting of FF 6 569 in respect of ordinary duties and FF 12 756 in respect of fines and penalties. Having unsuccessfully requested that those duties be set aside on the ground that the tax was contrary to Article 95 of the Treaty, Mr Tarantik brought proceedings before the Tribunal de Grande Instance (Regional Court), Meaux, against the Chief Tax Inspector for the Département of Seine-et-Marne, seeking cancellation of the payment notice and payment of default interest together with FF 7 000 pursuant to Article 700 of the Code of Civil Procedure.(*)

Before the Tribunal de Grande Instance de Meaux, Mr Tarantik contested the lawfulness of the differential road tax, arguing that the Court of Justice had held it to be unlawful on several occasions. He submitted that the tax legislation remained discriminatory, subsequent amendments to it notwithstanding.

The Tribunal de Grande Instance de Meaux accordingly sought a preliminary ruling from the Court on the following questions:

‘Considering the date of approval of the plaintiff's Jaguar car, which has a fiscal horsepower of 24 CV, bears the registration number 197 AT 77, and was first put on the road on 11 April 1979, and having regard to the diagrams and the outline of the progression of the tax submitted by the plaintiff, on the one hand, and to the observations of the French tax authorities, on the other, does the system of taxation applied correspond to objective criteria lacking in any discriminatory effect prohibited by Article 95 of the EEC Treaty ? In particular:

  • Is the progression coefficient existing between the tax band covering an imported vehicle of more than 18 CV and the tax band covering a similar vehicle of 15-16 CV discriminatory or not?

  • Do the circulars of 28 December 1956, 23 December 1977, 24 June 1987, 12 January 1988 and 20 September 1991, as retroactively validated by Article 35 of the Finance (Amendment) Law of 22 June 1993, have the effect of making the tax discriminatory in regard to owners of vehicles for which type-approval has not been granted in France, that is to say, vehicles approved on an individual basis?

  • If the answer is yes, can the owner of a vehicle with a power rating in excess of 100 kW rely on that answer in order to plead, by application of the general principles of Community law such as equality in regard to public charges and the provisions of the European Convention on Human Rights and the Protocols thereto, that the tax is not payable on the ground that it is discriminatory or inequitable?’

The facts set out in the order for reference are extremely sparse; moreover they contain inconsistencies and apparent errors, some of which have now been resolved or corrected by written observations and by answers to written questions put to Mr Tarantik. The following paragraphs, drawn from those sources, supplement the account of the facts given in the order for reference.

It appears from the answer to a written question put to Mr Tarantik by the Court that the fiscal horsepower of his vehicle was calculated on the basis of the 1956 circular, and not, as is twice stated in the order for reference, on the basis of the 1977 circular. The French Government concurs with this.

According to the French Government, Mr Tarantik's Jaguar was a model XJ6, type 4L2, of which production started in 1968 and type approval was granted in France in January 1969. Supplementary type approvals, reflecting subsequent modifications to the model, were granted in April 1974, June 1975, November 1976, and, according to Mr Tarantik, October 1978. Mr Tarantik's car was a 2 Series, production of which started in 1973, which was the subject of the November 1976 supplementary type approval. At that time, the fiscal horsepower of all vehicles was calculated by reference to the 1956 circular.

It thus appears, and has been confirmed by Mr Tarantik in answer to a written question put by the Court, that Mr Tarantik's vehicle was granted type approval, not, as is implied by the national court's second sub-question, individual approval.

Admissibility

In its written observations the French Government contests the admissibility of the reference. It essentially argues that the order for reference does not provide sufficient information. The order mentions neither the Jaguar model at issue nor the date or method of approval. It does not set out the complex method of determining power rating pursuant to either the 1956 or the 1977 circular, nor does it indicate in which cases each method is to be applied; at most, it states incorrectly that the 1977 circular was applied to Mr Tarantik. The French Government accordingly submits that the national court's second sub-question is too general and is irrelevant to the dispute before it and that the order for reference does not permit other Member States or other interested parties to submit observations pursuant to Article 20 of the Court's Statute.

I concur with the French Government's view that the order for reference does not give sufficient information to enable the Court to give a reply to the questions referred on the basis of the order alone.

It is well established that, in order to arrive at an interpretation of Community law which will be helpful to the national court, it is necessary for the national court to define the factual and legislative context of the questions it is asking or, at the very least, to explain the factual circumstances on which they are based.(*) The Court has observed that the need for adequate information is particularly acute in the field of competition law which is characterised by complex factual and legal situations.(*) In my view the requirement is similarly of particular relevance in cases concerning fiscal issues, where the national legislation may well be highly complex and where it is likely to be impossible to assess the full implications of one aspect of the system in the absence of complete information concerning the functioning of the system as a whole.

That requirement is not satisfied in the present case, since as indicated above there are several elements of basic information missing in the order for reference. The order does not state the model of Mr Tarantik's vehicle, the procedure for its approval in France or how its fiscal horsepower was calculated. It does not set out the relevant national rules at all: the two existing formulas for calculating power rating are not described, nor is it explained in precisely which circumstances the formula laid down in the 1956 circular is applied. In addition, as indicated above, the order for reference contains several inaccurate statements of fact with regard to elements essential to an understanding of the method by which the fiscal horsepower at issue was determined. The order does not explain the effect of the national rules on the taxation of motor vehicles imported from other Member States. It follows that the referring court has not explained why the present case raises an issue of the compatibility of the French rules with Article 95 of the Treaty. Lastly, the order does not state in what way the Court's judgment could be relevant for the main proceedings: if the application of the 1956 circular were contrary to Community law, would that mean that the 1977 circular should be applied and, if so, would that result in a different level of taxation ? Those may all to my mind be sound reasons for regarding the present reference as inadmissible. It may be noted that this was the approach taken by the Court in Bresle(*) in similar circumstances.

The Court has however accepted that the requirement for the national court to define the factual and legislative context is less pressing where the questions relate to specific technical points and enable the Court to give a useful reply even where the national court has not given an exhaustive description of the legal and factual situation. (*) In the present case much information can be distilled from the Court's previous case-law, from the observations of Mr Tarantik, the French Government and the Commission, and from the replies of Mr Tarantik to the written questions put by the Court. In so far as the national court's questions can be reformulated so as to permit replies which will enable that court to resolve the dispute before it, it may be appropriate for the Court to answer the questions notwithstanding the manifest inadequacy of the order for reference.

That approach does not, of course, resolve the problem that other Member States or other interested parties have been unable to submit observations based on a full and informative order for reference. It may be, however, that in this case that issue is not fatal since, as will be seen, in my view the national court's questions, in so far as they are relevant to the main proceedings, may be answered on the basis of existing case-law.

I would note that one reason for the inadequacy of the order for reference, and indeed it may be said for similar problems with a number of orders for reference recently lodged at the Court concerning the French road tax,(*) may be that parts of it at least have apparently been drafted in very general terms without regard to the facts and issues characterising the individual case. Any such practice is clearly unsatisfactory for the reasons given above. Moreover, according to the French Government, many of the claims currently before the French courts, some of which have resulted in requests for preliminary rulings from the Court, have been instigated by an association orchestrating resistance to the French road tax system. If the response by national courts to the activities of that association leads to orders for reference and pleadings before the Court which do not explain the specific case in respect of which the order is made, the question arises whether proper use is being made of the system of references for preliminary rulings established by Article 177 of the Treaty.

The tenor of Mr Tarantik's written and oral observations suggests that the Court is being asked to deliver a general ruling on the compatibility of the French system with Community law. In my view, that should rather be done in the context of infringement proceedings pursuant to Article 169 of the Treaty. (According to the Commission, the pre-litigation procedure under Article 169 has been opened in relation to two specific categories of cars still subject to the 1956 circular, the vast majority of which are imported, namely vehicles which as a result of new technology are not covered by the 1977 circular (for example, cars with six gears) and vehicles whose actual power rating is over 100 kW and which are approved on an individual basis. The calculation of the fiscal horsepower of those vehicles still includes factor K with a ceiling, and thus to that extent there has not been compliance with the Court's ruling in Feldain,(*) discussed below.) I will accordingly focus in this Opinion on what I have described above as the national court's sub-questions, namely the specific questions which it puts.

The previous rulings of the Court

The Court has already had occasion to give several rulings on the compatibility of the French road tax system with Article 95 of the Treaty. In the first case, Humblot,(*) the Court was asked whether it was compatible with Article 95 of the Treaty to impose both a differential tax on cars rated at 16 CV or less and a special tax, levied at a single and considerably higher rate, on vehicles of more than 16 CV. The Court first of all made the following general observation concerning progressive systems of road tax:

‘... as Community law stands at present the Member States are at liberty to subject products such as cars to a system of road tax which increases progressively in amount depending on an objective criterion, such as the power for tax purposes, which may be determined in various ways’.(*)

The Court added however that such a system of domestic taxation was compatible with Article 95 only in so far as it was free from any discriminatory or protective effect. The Court considered that that requirement was not met by a system such as the French one where the special tax was several times the highest amount of the progressive tax payable on cars of 16 CV or less and where the only cars subject to the special tax were imported, in particular from other Member States.

In order to comply with that judgment, the French legislature abolished the special fixed tax on cars with a power rating of over 16 CV and created four new tax bands with the result that there were nine tax bands in total, namely: up to 4 CV, from 5 to 7 CV, 8 and 9 CV, 10 and 11 CV, from 12 to 16 CV, 17 and 18 CV, 19 and 20 CV, 21 and 22 CV, and 23 CV and over. It also fixed a coefficient for each tax band (1; 1.9; 4.5; 5.3; 9.4; 14.1; 21.1; 31.7; 47.6 respectively).

In the second case, Feldain,(*) the Court was asked whether the system as so modified was compatible with Article 95 of the Treaty. One of the issues was whether it was compatible with Article 95 to establish a system whereby the progression of the coefficients from band to band, which was linear up to 16 CV, became ‘exponential’ beyond the 12 to 16 CV tax band, which was the last one to include vehicles manufactured in France. The Court held that the progression did not display any significant difference of such a kind as to indicate that there was any discriminatory or protective effect. However, the Court did find that the tax system was discriminatory in two respects.

First, the arrangement of the tax bands was such that top-of-the-range cars manufactured in France fell in the 12 to 16 CV tax band and only vehicles from other Member States came within the tax bands above 16 CV. Because the 12 to 16 CV band was a broad one (the only one containing five power ratings whereas all the others comprised two or three power ratings) top-of-the-range cars manufactured in France were protected from the normal progression of the tax.

Second, the method of determining the power rating itself, as laid down in the 1977 circular, was not objective and favoured cars manufactured in France. That method was expressed in the formula Ρ = m (0.0458 x C/K)1.48. As explained above, in that formula, Ρ stands for the power rating for tax purposes, m is 1 for petrol and 0.7 for diesel fuel, C is the cylinder capacity of the engine expressed in cubic centimetres. The cylinder capacity was divided by a factor, known as factor K, which was equal to the average of the speeds which would in theory be attained by the vehicle concerned at an engine speed of 1 000 revolutions per minute for each ratio of the gearbox. In the version of the 1977 circular being considered by the Court, however, the factor Κ was subject to a ceiling of 21 for vehicles whose actual power rating exceeds 100 kW.

The Court established that, by limiting the factor Κ as described above, the rules at issue had the effect of attributing to cars with a manual gearbox and a cylinder capacity exceeding 3 109,7 ce a power rating which exceeded 16 CV, irrespective of any other characteristics which it might display with respect to fuel consumption. Moreover, where the factor Κ was limited the cylinder capacity alone determined the power rating for tax purposes. Since the cylinder capacity was raised to the power of 1.48, the result of any increase in cylinder capacity was to bring those cars within the highest tax bands.

The Court also found that no French-manufactured car had a cylinder capacity exceeding 3 109,7 cc. Consequently, the effect of limiting the factor K to 21 was to place in the higher tax bands only imported vehicles, whereas, without such a limitation, those cars would be given lower power ratings for tax purposes. Such a limitation was not justified by considerations relating to fuel consumption. There was no significant difference in that respect between vehicles affected by the limitation of the factor K and comparable cars not affected by it. The Court therefore held that that method of determining the power rating for tax purposes was not objective in character and favoured cars manufactured in France.

The Court came to the same conclusions as those reached in Feldain in the third case, Seguela.(*)

The French legislature adopted new legislation to comply with the judgment in Feldain, splitting the old band of 12 to 16 CV into two and thus creating a total of 10 instead, of 9 bands. Consequently, there is now a band for 12 to 14 CV and another for 15 to 16 CV. Furthermore, Circular No 88-04 of 12 January 1988 modified the method of determining the power rating which was laid down in the 1977 circular by removing the ceiling on the factor K.

In Deville,(*) the Court ruled on the lawfulness of national time-limits applying to claims for refunds of that element of the road tax found by the Court to be unlawful in Humblot and Feldain. Circular No 91-71 was adopted on 20 September 1991 listing the vehicles whose owners could obtain rectification of their fiscal horsepower in the light of those judgments. It permits owners of vehicles whose horsepower was calculated in accordance with the aspects of the 1977 circular held to be unlawful by the Court to obtain rectification of the fiscal horsepower of their vehicles. The 1991 circular is limited to vehicles within the scope of the 1977 circular which were granted type approval on or after 1 January 1978.(*) It is expressed not to apply to vehicles whose fiscal horsepower was calculated in accordance with the 1956 circular.(*)

The Court was thereafter again confronted with the French differential road tax in Jacquier.(*) It was asked in effect whether the fact that the progression coefficient(*) was greater for the three tax bands above 18 CV, which concerned only vehicles imported from other Member States, than for the three bands from 12-14 CV to 17-18 CV, which included both imported and domestic cars, was contrary to Article 95.

The Court noted that whether the increase in the progression coefficient above the 18 CV threshold had a discriminatory or protective effect depended on whether that increase might deter consumers from purchasing vehicles with a fiscal horsepower of over 18 CV, which were all of foreign manufacture, to the benefit of vehicles of domestic manufacture. If the increase in the coefficient for vehicles with a fiscal horsepower of over 18 CV did indeed deter some consumers from buying such vehicles, those consumers would choose a model in the tax band immediately below, the 17-18 CV band, or even a model in the 15-16 CV band. The 15-16 CV and 17-18 CV tax bands, however, included both imported vehicles and vehicles of domestic manufacture. In the 17-18 CV band, the vehicles were nearly all of foreign manufacture, domestic manufacturers having a market share of only about 5% of total car sales in that band. In the 15-16 CV band, the majority of vehicles were of domestic manufacture; consumers nevertheless had a wide choice of imported vehicles in that band and the progression coefficients for the 15-16 CV and the 17-18 CV bands were the same, in round figures, so that consumers who were looking for a vehicle from the top of the range would not thereby be induced to purchase a vehicle in the 15-16 CV band. By this, the Court meant that the progression coefficient from 12-14 to 15-16 (1.22) was almost the same as that from 15-16 to 17-18 (1.23).

The Court accordingly concluded that Article 95 did not preclude such a system provided that the increase in the coefficients did not have the effect of favouring the sale of vehicles of domestic manufacture over the sale of vehicles imported from other Member States.

The Court did not in that case consider the method of determining the power rating of cars.

The first question: the progression coefficients

By its first question, the national court asks whether the progression coefficient between the tax band covering an imported vehicle of more than 18 CV and the tax band covering a similar vehicle of 15-16 CV is compatible with Article 95 of the Treaty. It is therefore concerned with the relationship between the coefficients, that is to say the manner in which there is a progression from one coefficient to the next. That relationship is expressed as a factor which, when multiplied by the coefficient for a particular tax band, produces the coefficient for the next band.

The coefficients for the bands from 12 CV upwards are as follows:

Bands

Coefficient

12-14CV

9.4

15-16CV

11.5

17-18CV

14.1

19-20CV

21.1

21-22CV

31.7

23CV & over

47.6

The factors of progression of the coefficients for the bands from 12 CV upwards are as follows:

Bands

Progression factor

from 12-14CV to15-16CV

1.22

from 15-16CVto17-18CV

1.23

from 17-18CVto19-20CV

1.50

from 19-20CVto21-22CV

1.50

from 21-22CVto23CV & over

1.50

Mr Tarantik notes that, since the Court in Jacquier in effect found that the level of tax on vehicles with a fiscal horsepower of 23 or more might induce consumers to buy domestic vehicles from the 17-18 or 15-16 tax band, provided that their minimum requirements regarding performance, size and comfort were met,(*) it was necessary in the present case to refer to the Court the question whether the factor of progressivity between on the one hand vehicles of 19 CV or above (all of which are imported) and on the other hand vehicles of 15-16 (the majority of which are of domestic manufacture) was discriminatory. Mr Tarantik asserts that vehicles of more than 23 CV may in fact have the same technical characteristics as vehicles of 15-16 CV, in particular their actual power rating, price and performance being broadly comparable. The coefficient of an imported vehicle of 23 or 24 CV (such as his own) is however 47.6, whereas that of a similar domestic vehicle of 15-16 CV would be 11.5; the tax payable in respect of the former would be 4.13 times higher than that payable in respect of the latter.

The French Government and the Commission both consider that the national court's first question may be answered by reference to the judgment of the Court in Jacquier,(*) where the Court stated:(*)

‘A system of taxation cannot be regarded as discriminatory solely because only imported products, in particular those from other Member States, come within the most heavily taxed category (Case 140/79 Chemial Farmaceutici ν DAF [1981] ECR 1, paragraph 18, and Case C-132/88 Commission ν Greece [1990] ECR I-1567, paragraph 18).

To determine whether the increase in the progression coefficient of the differential tax above the 18 CV threshold has a discriminatory or protective effect, it must be examined whether that increase may deter consumers from purchasing vehicles with a fiscal horsepower of over 18 CV, which are all of foreign manufacture, to the benefit, of vehicles of domestic manufacture.

If the increase in the coefficient for vehicles with a fiscal horsepower of over 18 CV does indeed deter some consumers from buying such vehicles, those consumers will choose a model in the tax band immediately below, the 17-18 CV band, or even, as the Advocate General notes in point 27 of his Opinion, a model in the 15-16 CV band.

The 15-16 CV and 17-18 CV tax bands, however, include both imported vehicles and vehicles of domestic manufacture. In the 17-18 CV band, it is apparent from the documents in the case that the vehicles are nearly all of foreign manufacture and that domestic manufacturers have a market share of only about 5% of total car sales in that band. In the 15-16 CV band, while the majority of vehicles sold are indeed of domestic manufacture, firstly, consumers do nevertheless have a wide choice of imported vehicles in that band, and secondly, as appears from paragraph 6 of this judgment, the progression coefficients for the 15-16 CV and the 17-18 CV bands are the same, in round figures, so that consumers who are looking for a vehicle from the top of the range will not thereby be induced to purchase a vehicle in the 15-16 CV band.

In a system such as that at issue in the present case, therefore, it does not appear that the increase in the progression coefficient can have the effect of favouring the sale of vehicles of domestic manufacture.’

I concur with the view of the French Government and the Commission, to the effect that the national court's first sub-question can indeed be answered by application of the principles that the Court laid down in Jacquier.(*) The national court will have to ascertain which vehicles are in competition with cars such as Mr Tarantik's Jaguar and analyse the breakdown both of imported and domestic competing vehicles and of the tax bands into which competing vehicles fall. Those findings should enable it to conclude whether the increase in the progression coefficient favours the sale of cars of domestic manufacture, in which case it must conclude that the element of the tax at issue is contrary to Article 95.

The second question: the method of determining fiscal horsepower

By its second question, the national court asks whether the effect of the circulars of 1956, 1977, 1987, 1988 and 1991 is that the tax discriminates against owners of vehicles for which type approval has not been granted in France. That question appears, as pointed out above, to be based on the incorrect assumption that Mr Tarantik's Jaguar was not granted type approval in France. An answer to the question as put cannot therefore resolve the dispute before the national court. It is however possible to reformulate the question in the light of the information provided in the observations to the Court and the replies to written questions put by the Court to Mr Tarantik. On that basis, I shall assume that the national court intended to ask whether the national rules for determining fiscal horsepower have the effect of making the tax discriminatory as against either (i) specifically, owners of vehicles for which type approval was granted in France before 1 January 1978, whose fiscal horsepower is determined in accordance with the 1956 circular or (ii) more generally, owners of vehicles whose fiscal horsepower is calculated in accordance with the 1956 circular. An answer to either one of those questions would enable the national court to resolve the issue before it, namely whether the method by which the fiscal horsepower of Mr Tarantik's car has been determined is discriminatory.

Mr Tarantik, after analysing in great detail the system for determining fiscal horsepower, expounds certain aspects of it which in his view are discriminatory.

First, he considers the retention of the ceiling on factor K for certain cars whose fiscal horsepower is calculated on the basis of the 1977 circular. However, the question whether the 1977 circular as applied to certain cars in respect of which the ceiling on factor K has not been abolished is discriminatory cannot resolve the issue before the national court in the main proceedings, which concern the determination of the fiscal horsepower of Mr Tarantik's Jaguar which neither is nor allegedly should be determined on the basis of the formula in the 1977 circular including the ceiling on factor K. That argument of Mr Tarantik cannot therefore be regarded as relevant. As indicated above, the Commission proposes to initiate proceedings under Article 169 of the Treaty concerning the retention of the ceiling on factor K in certain cases.

Second, Mr Tarantik considers the allegedly discriminatory effect of the coexistence of several methods of calculating fiscal horsepower of which the least favourable apply to imported cars. He submits that the criteria for distinguishing between the two principal methods are not objective and increase the tax payable by imported vehicles alone.

Mr Tarantik submits that the fiscal horsepower of cars manufactured in France for domestic sale has always been determined on the basis of the most recent circular as at the date of approval, whereas that of imported cars is determined on the basis of several circulars containing different formulae: the result is that certain imported cars are attributed a fiscal horsepower different from that attributed to an equivalent domestic car, with no objective reason for the distinction.

Moreover, the fiscal horsepower of numerous imported cars is determined by reference to the less favourable formula in the 1956 circular, with no objective criterion for the distinction. Where an imported vehicle similar to a domestic one in terms of actual power rating, price and intrinsic qualities suffers significantly higher tax solely because the formula chosen by the Member State for applying to all similar vehicles is unsuitable and the official objective of the latter formula is not attained, the excessive tax payable by the owner of an imported car of more than 16 CV is manifestly abusive and should be prohibited.

Finally, the criterion for determining the amount of tax payable is not objective since the authorities' choice of different formulae applying only to certain imported vehicles is manifestly liable to discourage imports and is not perfectly neutral in effect. The Court has ruled that Article 95 is infringed where the taxation on the imported product and that on the similar domestic product are calculated in a different manner on the basis of different criteria which lead, if only in certain cases, to higher taxation being imposed on the imported product.(*)

The French Government argues that the method of determining the fiscal horsepower of the applicant's vehicle, namely in accordance with the 1956 circular, is compatible with Article 95. It notes that the Court has held that Member States have the right to apply progressive taxation based on objective criteria and that a system of taxation cannot be considered to be discriminatory solely because it applies only to imported vehicles in the absence of domestic production(*) or solely because only imported products, in particular those from other Member States, come within the most heavily taxed category,(*) and that Member States are at liberty to subject products such as cars to a system of road tax which increases progressively in amount depending on an objective criterion, such as cylinder capacity, provided that that system of taxation is free from any discriminatory or protective effect.(*)

The French Government concludes that the method of determining fiscal horsepower laid down by the 1956 circular, which depends on the objective criterion of cylinder capacity, and applies equally to domestic and foreign cars, is compatible with Article 95, noting that the applicant has adduced no evidence of the discriminatory effect of the 1956 circular vis-à-vis foreign vehicles nor demonstrated that the application of the 1956 circular leads consumers to buy French vehicles with an equivalent cylinder capacity.

The Commission refers to my Opinion in Tabouillot,(*) which noted the apparent lack of transparency in the application of the system of road tax and stressed that a system of taxation which was so lacking in transparency that the national courts were unable to decide whether it is discriminatory or not would make it impossible for the taxpayer to rely on Article 95 and would for that reason alone be contrary to Article 95.(*) The Commission considers that the reply to the national court's question may be found in that Opinion, in the sense that the compatibility of the French system with Article 95 is not guaranteed given the persistent and serious lack of transparency of the system.

More specifically, the Commission states that it cannot be sure that the circumstances in which the disadvantageous calculation set out in the 1956 circular is applied to vehicles for which individual approval was granted after 1 January 1978 are neutral and objective as required by Article 95. That depends on whether the authorities consider that the vehicle matches or is equivalent to an approved type; the French Government has not provided details of the criteria applied.

The Commission also notes that the French authorities appear to be extremely demanding with regard to the evidence required of the technical details of vehicles presented individually, needed in order to apply the formula in the 1977 circular; the Commission raises the question whether those requirements are really relevant to the objective pursued and above all proportionate.

The Commission has submitted further arguments going to the legality of the 1977 circular, in particular the retention of the ceiling on factor K for certain categories of vehicles. Those observations appear to be based on the misapprehension — entirely understandable in the light of the order for reference — that the fiscal horsepower of Mr Tarantik's Jaguar was calculated in accordance with the formula in the 1977 circular with factor K subject to a ceiling. They are accordingly not relevant to the present case. As indicated above, the Commission proposes to bring infringement proceedings in respect of that particular aspect of the present application of the 1977 circular, which are clearly a more appropriate forum for an evaluation of the lawfulness of the issue than a preliminary reference from a court in proceedings in which the issue does not arise.

The fact that certain of the Commission's arguments are based on a misconception of the facts underlying the main proceedings perfectly illustrates the difficulties which arise from the order for reference in this case. In particular, two fundamental matters were still not clear by the end of the written procedure, namely whether the fiscal horsepower of Mr Tarantik's car was determined in accordance with the 1956 or the 1977 circular and whether that car was granted individual or type approval. On both issues, it appears from the order for reference that the national court referred the questions for preliminary ruling on an incorrect basis. There has therefore been no adequate exchange of argument. I do not consider it appropriate in those circumstances to deal in detail with the interaction of the two systems. Instead, I shall limit myself to the following observations on the opacity of the national legislation, which in large measure echo points I made in my Opinion in Tabouillot.

It is clear from the Court's case-law that progressive taxation is permissible, provided it is neutral in its effects on domestic products and imports. That requirement of neutrality also applies, in the case of motor vehicles, to the method of determining power rating. Where two different methods are used which result in different levels of taxation, the system must again be neutral: the rules determining the choice of method must be based on objective criteria which do not discriminate against similar imported products and are not such as to afford protection to domestic production.

It is by no means clear that the French system is in accordance with those requirements; there appears moreover to be a serious lack of transparency in its application. The scope of the two circulars is far from easy to understand, the relationship between the two systems is complex and obscure and there appear to be no objective criteria for the exercise of the discretion which apparently exists to determine whether in certain cases cars are covered by one or the other circular. The Court has stated that, where there is a lack of transparency, the onus is on the Member State concerned to prove that the system in question does not engender any possibility of discriminatory effects.(*) Although that statement was made in infringement proceedings, brought under Article 169 of the Treaty, it must clearly be of general application. Article 95 of the Treaty has direct effect and must be enforceable in the national courts. A system of taxation which is so lacking in transparency that the national courts are unable to decide whether it is discriminatory or not would make it impossible for the taxpayer to rely on Article 95. Such a system must for that reason alone be contrary to Article 95. The effectiveness of the prohibition of fiscal discrimination could be singularly reduced if a Member State were allowed to introduce or maintain a non-transparent tax system, in particular where the tax is differential and progressive and applies to a product which is highly tax sensitive and traded across borders in large quantities, as is the case with cars.

The third question: is the tax payable?

By its third question, the national court asks in effect whether, if the tax is contrary to Article 95, it is open to the owner of a vehicle with a power rating in excess of 100 kW not to pay the tax.

The reference to vehicles with a power rating in excess of 100 kW appears yet again to be irrelevant to the case before the national court: the 100 kW threshold is significant only for certain vehicles whose actual power rating exceeds it, whose fiscal horsepower is calculated in accordance with the 1977 circular and which are not covered by the 1991 circular. Since the fiscal horsepower of Mr Tarantik's vehicle was not calculated in accordance with the 1977 circular, an answer to the question as put cannot assist the national court to resolve the dispute before it.

I would none the less briefly note that it is clear from the case-law of the Court that fiscal charges held to be contrary to Article 95 cannot lawfully be demanded by the authorities and, once paid, may be recovered. Claims for recovery are governed by the applicable national rules, provided that those rules are no less favourable than the conditions applicable to similar domestic claims and do not make it impossible in practice or excessively difficult to exercise rights which the national courts have a duty to protect.(*)

Conclusion

For the reasons given above, it is doubtful whether all the questions, as put by the Tribunal de Grande Instance de Meaux, arise. It is however possible to provide answers to the questions, where necessary reformulated, as follows:

  1. It is for the national court to determine, in the light of the judgment of the Court in Case C-113/94 Jacquier ν Directeur General des Impôts [1995] ECR I-4203, whether the progression coefficient existing between the tax band covering an imported vehicle of more than 18 CV and the tax band covering a similar vehicle of 15-16 CV is discriminatory and thereby infringes Article 95 of the Treaty.

  2. Article 95 of the Treaty does not preclude the application by a Member State of two sets of rules for establishing the fiscal horsepower of vehicles provided that the choice between the two sets of rules is based on objective criteria which do not discriminate against similar imported products and are not such as to afford protection to domestic production. Where however there is a lack of transparency in the system used by a Member State to establish the fiscal horsepower of vehicles, the onus lies on the Member State to satisfy the national court that the system does not engender any possibility of discriminatory effects.

  3. Fiscal charges held to be contrary to Article 95 of the Treaty cannot lawfully be demanded by the authorities and, once paid, may be recovered. Claims for recovery are governed by the applicable national rules, provided that those rules are no less favourable than the conditions applicable to similar domestic claims and do not make it impossible in practice or excessively difficult to exercise rights which the national courts have a duty to protect.