Version classiqueVersion mobile
OpenEdition Books

La réorientation européenne de la TVA à la suite du renoncement au régime définitif

 | 
Francis Querol

Partie II. Les difficultés liées aux règles actuelles de territorialité du régime transitoire de TVA

EU VAT rate structure: towards unilateral convergence?

Rita de la Feria

Texte intégral

1Within Europe differentiated rates structures date back to the introduction of VAT itself. Although evidence as regards potential negative consequences of applying multiple rates was unavailable at that time, difficulties have been apparent for some decades. In light of this reality, since the late 1980s, there have been several attempts to amend European rates structures under the political guidance of the European Commission. There has however been surprising resistance by Member States to any proposed amendments which might lead to a reduced rate differentiation. Thus giving weight to the intuition that once reduced rates of VAT are introduced it is almost politically impossible to remove then. Indeed the most recent agreed upon amendments to the rates structure have increased the level of differentiation, rather than decrease it, with more goods and services being subject to reduced rates in Europe today than even as recently as ten years ago. Yet this reality seems to be changing in the last few years as a result of the economic and financial crisis which has fallen upon Europe. Since 2009 twenty-five of the thirtythree OECD countries have increased their VAT rates, resulting in a broad convergence of VAT standard rates across the EU around the 20% mark. Furthermore, there has also been a decrease in levels of differentiation with a decrease in the number of VAT rates applicable in many Member States, as well as various base broadening measures. The latest developments raise the possibility that Europe might be finally entering a process of convergence of VAT rate structures, not by EU initiative but by domestic necessity; an unforeseeable, unplanned unilateral convergence, to contrast to the long-sought, but so far unattainable, EU harmonisation.

1. EU VAT Rates: Until 2009

  • 1 First Council Directive 67/227/EEC of 11 April 1967 on the harmonisation of legislation of Member S (...)
  • 2 Article 9 of the Second VAT Directive.
  • 3 See R. DE LA FERIA and R. KREVER, “Ending VAT Exemptions: Towards A Post-Modern VAT” in R. DE LA FE (...)
  • 4 Sixth Council Directive 77/388/EEC of 12 May 1977 on the harmonisation of the laws of the Member St (...)

2The introduction of the European VAT system dates back to 1967, with the approval of the First and Second VAT Directives1. The system put in place under those Directives, however, established only a basic framework, leaving a full autonomy to Member States insofar as rates were concerned: national legislators were free to establish their own rates structure, including number and level of rates2. Primarily for political and practical reasons, Member States used that freedom therefore to largely mimic the rates structures applied under their previous turnover taxes3. With the approval of the Sixth VAT Directive in 1977 there was a significant increase on the level of detail contained as regards the tax base, and a decrease in the level of freedom granted to Member States4. Yet, despite the progress achieved in some areas of the system, as regards other areas such as the rates structure, reportedly the EC Council of Ministers found it impossible to reach agreement and consequently further harmonisation was postponed to a later date. The rules applicable to rates under the original version of the Sixth VAT Directive were therefore similar to those previously applicable under the Second VAT Directive, i.e. there was total lack of specific rules as regards rates structures. For that outcome certainly contributed the opposition adopted by Member States at negotiations-such as France, Germany, and the United Kingdom-keen to maintain the domestic application of reduced VAT rates to specific products.

1.1. Towards an harmonised rates structure

  • 5 Completing the Internal Market – White Paper from the Commission to the European Council, COM (85) (...)
  • 6 White Paper, COM (85) 310, 14 June 1985, n. 5 above, at 52.

3In June 1985, the European Commission presented the so-called White Paper for the completion of the Internal Market, which laid down a series of measures with a view to establishing an internal market by 1992. Under the heading removal of fiscal barriers, the paper contained several measures in the field of VAT5. According to the White Paper a close level of “approximation” within VAT was required in order to establish a true internal market, and in particular progress had to be achieved as regards tax rates. In this context, the European Commission was to present a proposal, which would deal both with the number of rates and level of these rates. In the meantime, according to the White Paper “provisions should be adopted which will exclude the proliferation of VAT rates in Member States, or the widening of the gap between VAT rates, since this would make subsequent adjustment more difficult”6.

  • 7 Proposal for a Council Directive completing the common system of value added tax and amending Direc (...)

4In 1987 the Commission put forward a proposal for a new VAT rates structure, which would be compatible with these objectives7. This new structure was based on three basic principles, as follows.

  • 8 However, although theoretically the United Kingdom had a single VAT rate system, in practice it app (...)
  • 9 See COM (87) 321 final/2, 21 August 1987, n. 7 above, at 1.
  • 10 Completion of the Internal Market: approximation of indirect tax structures and harmonisation of in (...)
  • 11 See Completion of the Internal Market: approximation of indirect tax structures and harmonisation o (...)

5Dual rate system: Although, acknowledging that a single VAT rate system would be ideal, as the simplest and most efficient structure, the Commission considered that, as all Member States at the time (excluding Denmark and the United Kingdom) applied at least two VAT rates8, a multi-rate system would be preferable not to upset the tax structures of the majority of Member States9. However, the question of how many VAT rates this multi-rate system should entail was less clear and the real choice for the Commission lay between a two-rate and a three-rate system. Ultimately, the Commission considered that a three-rate system would create more complications for both taxpayers and national administrations and that it would be simpler and more cost effective to establish a dual rate system10. In light of these conclusions, the rate structure proposed by the Commission entailed a standard rate and one reduced rate. However, instead of fixing these rates, the Commission opted for establishing two rate bands, thereby allowing Member States some degree of flexibility in the choice of the actual rates. The standard rate could therefore vary between 14% and 20%, whilst the reduced rate could vary between 4% and 9%. The choice of bands seemed to have been based more on practical political considerations, than on optimal economic and efficiency calculations. The Commission considered that a 6 point range for the standard rate, and 5 point for the reduced rate, would be sufficient to avoid potential trade distortions, while allowing for a maximum number of Member States to suffer minimum budgetary repercussions11.

  • 12 Mostly if compared with the list which was ultimately approved by Council Directive 92/77/EEC and s (...)
  • 13 COM (87) 320 final, 5 August 1987, n. 11 above, at 10.
  • 14 See COM (87) 320 final, 5 August 1987, n. 11 above, at 2.
  • 15 As discussed infra, the tactic proved ultimately unsuccessful, as the proposal failed to get Counci (...)

6Goods / services would be compulsorily allocated to each rate: Under the proposal the reduced rate would compulsorily apply to the following goods / services: foodstuffs, excluding alcoholic beverages; energy products for heating and lighting; water supplies; pharmaceutical products; books, newspapers and periodicals; and passenger transport. There are two different aspects to this proposed list: firstly, the choice of goods / services subject to the reduced rate; and secondly, the compulsory nature of the list. Both aspects reflect the Commission’s view on the role of reduced rates within the new rates structure. Under the proposed list the amount of goods and services subject to the reduced rate is limited12. The Commission’s selection of items was based on the application of reduced rates which Member States already had in place13. By choosing a limited number of items which were already subject to reduced rates in most Member States, the Commission sought to achieve two interlinked objectives: first to cause minimum budgetary disturbance to a maximum number of Member States, and second to facilitate the acceptance of the compulsory nature of the list. It considered that the allocation of goods and services to each rate (all goods / services not listed would be subject to the standard rate) was necessary in order to ensure that “the same type of products or service is placed under the same category of rate in the various Member States, thus avoiding systematic deflections of trade”14. However, there was awareness that the compulsory nature of the new rates structure was likely to face objections by Member States, as their freedom to establish their own rate structures was significantly diminished. In this context, the Commission argument was that if most Member States already applied reduced rates to the listed items, then the compulsory nature of the list would be almost irrelevant, as it did not imply any change to either their domestic legislations, or budgetary consequences. In this way the Commission sought to facilitate Council’s approval of the proposal15.

  • 16 COM (87) 320 final, 5 August 1987, n. 11 above, at 13. For a more detailed report on the Commission (...)
  • 17 See COM (87) 320 final, 5 August 1987, n. 11 above, at 14.
  • 18 Proposal for a Council Directive instituting a process of convergence of rates of value added tax a (...)
  • 19 See COM (87) 324 final/2, 21 August 1987, n. 18 above, at Article 1.

7Temporary derogations, allowing Member States to apply reduced rates and zero rates, would be repealed: Article 28 (2) of the Sixth VAT Directive allowed Member States to retain the use of reduced rates and zero rates in force on 31 December 1975. The measure was temporary, and destined to be repealed once fiscal frontiers were abolished and taxation within the Community was based on the principle of origin. Attempting to fulfil this objective, the Commission’s proposal envisaged the revocation of Article 28 (2). It considered that the extensive application of reduced rates and particularly zero-rates had the potential to obstruct the Internal Market and distort competition, thus concluding that it “could not recommend that the Community should abandon what has been its considered and settled policy ever since the VAT was first adopted”16. However, as the abolition of this measure was likely to raise difficulties to Member States (and thus, hinder the proposal’s approval by the Council), the Commission clarified that it would not be opposed to the possibility of derogations being granted to Member States facing particular difficulties through the abolition of this measure. In this context, it committed itself “to take a constructive part in the discussion of any derogations for which Member States in real difficulty might feel the need”17. In the interim, until the new VAT rate structure came into force (along with the abolition of fiscal frontiers on 1 January 1993), the Commission considered that it would be beneficial if VAT rates in the different Member States were to be brought to a standstill. To this end it put forward an additional proposal18 according to which Member States should basically refrain from altering their rate structures, unless such alterations would bring their structures closer to that being proposed by the Commission, e.g. reducing or abolishing their increased tax rates or bringing their standard and reduced rates within the bands proposed19.

  • 20 A.J. EASSON expressed a widely felt scepticism when he commented: “to expect to achieve an alignmen (...)
  • 21 The Economic and Social Committee, for instance, recognised that the Commission’s proposal was prag (...)
  • 22 Completion of the Internal Market and Approximation of Indirect Taxes, Communication from the Commi (...)
  • 23 Resolution on completion of the Internal Market: Approximation of indirect taxation in the Communit (...)
  • 24 Additionally, it was also suggested at the time that the proposal would weigh more heavily on some (...)

8The Commission’s 1987 proposals were widely regarded as very ambitious in both their aims and their prospected methods for achieving these aims20. Progress in Council discussions proved slow and, following initiative of the Council and the Commission, several working parties were set up in order to establish the best strategy forward. Set up in the beginning of 1989, one of these concentrated exclusively on the allocation of VAT rates to different product categories. However, major disagreements still existed21. By June 1989 the Commission recognised that, based on Council and Parliament discussions, certain aspects of the 1987 VAT rates proposal were curtailing the possibility of reaching agreement, as follows: the width of the bands, namely in the case of the standard rate, which was often regarded as excessive and likely to give rise to distortions of competition; the range of products to be subject to reduced rate; and the problem of zero-rated products22. The European Parliament’s Committee on Economic and Monetary Affairs and Industrial Policy reported that23, as soon as the debate on the Commission’s proposal was initiated in the European Parliament, it became clear that a great number of national and special interests were anxious to have the list of items, which could be subject to the reduced rate, enlarged24. According to that same report, some fifty amendments to the first version of the European Parliament’s opinion on the proposal were tabled in the Committee.

9In was against this background that the Commission realised that the Council would fail to reach agreement and accepted that a more pragmatic approach would be required. The idea of a transitional phase, which would last beyond 1 January 1993, started to take shape in late 1989. During the period until 1991 a series of key meetings of the ECOFIN Council of Ministers took place, from which emerged the basic shape of the VAT arrangements to be applied to intra-Community trade after 1993. These were to become known as the “transitional VAT system”. The decision to introduce a VAT transitional system had serious implications for the discussions on the harmonisation of VAT rates. Rates approximation was still seen as an absolute necessity if abolition of border controls was to take place, however, a close approximation such as the one put forward by the Commission in its 1987 proposal was no longer required. Moreover, as the rates approximation as per the Commission’s 1987 proposal had given rise to such intensive controversy the decision to introduce a transitional system provided the perfect pretext to bring into force a less extreme approximation. In this context, and with a view to facilitate agreement within the Council, the Commission suggested in its Communication the following alternative rates structure:

  1. Minimum standard rate, without any upper limit being set, instead of the previously proposed standard rate band;

  2. One reduced rate set between 4% and 9%, thus retention of the band proposed in 1987; and

  3. Maintenance of zero-rating for a limited number of products, instead of the previously proposed abolition of zero-rates.

  • 25 Legislative Resolution embodying the opinion of the European Parliament on the Commission proposal (...)

10This alternative rate structure was significantly more moderate and less ambitious than the structure originally proposed by the Commission. However, it was still overambitious for Member States, particularly as regards the reduced rates regime – number of reduced rates allowed (only one), percentage band allowed (4% to 9%), list of items to which they would be applicable (six items), and finally the compulsory nature of this list. In an attempt to facilitate agreement regarding the reduced rates regime, the European Parliament suggested that two lists of goods and services would be set up: a first list, covering six items, for which a reduced rate would be mandatory; and a second list, covering twenty-four items, for which a reduced rate would be optional25. The first list contained most of the goods and services listed in the Commission’s initial proposal, some of which could be traded across frontiers. The second list broadly covered transactions where differences in VAT rates could not distort competition because the products in question were not generally traded between Member States.

  • 26 Council Directive 92/77/EEC of 19 October 1992, OJ L 316, 31/10/1992, 1 (hereafter the “Approximati (...)

11In March and June 1991 the Council finally reached agreement on the essential characteristics of the VAT rate structure, which was to apply within the context of the new transitional system. The agreement, which eventually led to the approval of the Council Directive 92/77/EEC of 19 October 1992, known as the Approximation of VAT Rates Directive26, not only differed significantly from the Commission’s original 1987 proposal, but also differed from the alternative rates structure proposed by the Commission in its 1989 Communication. Interestingly, the European Parliament’s two-list’ suggestion was not followed either.

12The new VAT rate structure, which would apply from 31 December 1992 onwards, was largely a product of political compromises and a good example of a victory of politics over economic efficiency. The price for reaching agreement was an extremely complex system (mostly if compared with the simplicity of the structure initially proposed by the Commission), filled with exceptions and derogations. Overall, the new rate structure comprised two types of rules: general rules; and temporary measures, which in theory would apply only during the transitional system.

1.1.1. General rules

13Under the general rules, Member States must apply a standard rate, which should not be lower than 15%, but no maximum limit was established. This rule followed from the Commission’s 1989 suggestion for a minimum standard rate, instead of the band initially proposed. Member States could also apply either one or two reduced rates, which could not be lower than 5%. These rates could be applied to a range of seventeen goods and services listed in what is now Annex III and additionally, under certain conditions, to the supplies of natural gas and electricity.

  • 27 As discussed above, insofar as reduced rates are concerned, the Commission’s 1989 alternative rate (...)
  • 28 As reported by the Commission in Report from the Commission to the Council in accordance with Artic (...)

14These reduced rates rules were substantially different from the ones originally proposed by the Commission in 198727. The most important difference was the optional nature of the rates: Member States were no longer obliged to apply a reduced rate to a few items, but were instead not only free to choose whether to apply a reduced rate or not, but also free to choose to which items (from the ones listed) that rate would apply. The other obvious differences were the possibility to apply two reduced rates (instead of one) and the increased list of items which could be subject to reduced rates (eighteen instead of six). Despite these differences, however, the Council did adopt the Commission’s approach in relation to the criteria adopted for the choice of items which could be subject to reduced rates28: one list of goods and services which were already taxed by a majority of Member States at reduced rates. This approach was substantially different from the one advocated by the Parliament, whose two-list’ proposal was based on a distortion of competition criteria.

  • 29 A Directive on works of art, antiques and collector’s items was approved in 1994, and in 1998 the C (...)

15Finally, it should be noted that, under the new legislation, certain supplies were not subject to these rates’ rules, namely works of art, antiques and collector’s items, agricultural outputs; and gold. These supplies would be subject to special arrangements and thus, the object of autonomous Directives to be approved on a future date by the Council29.

1.1.2. Temporary measures

16During the transitional period, and until the introduction of a definitive VAT system, Member States were allowed to maintain and/or introduce measures which derogated from the general rates’ rules described above. The introduction of these temporary measures had not been envisaged by the Commission’s 1987 proposal, nor had it been suggested by the Commission in its 1989 alternative rate structure. Thus, it most probably emerged from the Council’s discussions as a method of reiterating the Community’s commitment to a simpler, more efficient and harmonised rate structure, but postponing politically difficult decisions for Member States such as the abolition of zero-rating. The implications of these temporary measures to the overall rate structure were immense.

17Under these temporary measures, Member States were allowed, subject to certain conditions, to: continue to apply reduced rates lower than the 5% minimum; continue to apply zero-rates; continue to apply both reduced rates lower than the 5% minimum and zero-rates to items not listed in Annex III; continue to apply a reduced rate to restaurant services, children’s clothing, children’s footwear and housing; and introduce, dependent on certain requirements, an extra reduced rate, not lower than 12%. Additionally, Greece was also allowed to apply, within part of its territory, rates which were 30% lower than the ones applied in the rest of the country; and Ireland could apply for authorisation to apply a reduced rate to the supplies of energy products for heating and lighting.

18Table 1 provides a comparative overview of the three VAT rate structures discussed above: the structure proposed by the Commission in 1987, the alternative structure suggested by the Commission in 1989 and, finally, the structure which was ultimately approved by the Approximation of VAT Rates Directive.

TABLE 1. VAT RATE STRUCTURES: COMPARATIVE OVERVIEW

COMMISSION’S 1987 PROPOSAL

COMMISSION’S 1989 ALTERNATIVE

PROPOSAL APPROXIMATION OF VAT RATES DIRECTIVE

Two rates system (standard rate and reduced rate)

Two rates system (standard rate and reduced rate)

Five rates system (standard rate, three reduced rates and zero-rate)

Standard rate band (14% to 20%)

Standard rate minimum

Standard rate minimum (15%)

Reduced rate band (4% to 9%)

Reduced rate band (4% to 9%)

Reduced rates minimum (5%) in theory; in practice no minimum applies

6 items which may be subject to reduced rate

6 items which may be subject to reduced rate

22 items which may be subject to reduced rates

Compulsory nature of list of goods / services subject to reduced rate

Compulsory nature of list of goods / services subject to reduced rate

Optional nature of list of goods / services subject to reduced rate

Abolition of zero-rating

Maintenance of zero-rating for a limited range of products

Maintenance of zero-rating

1.2. Post-1992 failed initiatives

  • 30 See A common system of VAT – A programme for the Single Market, COM (96) 328 final, 22 July 1996.
  • 31 This means that in practice even if the proposals presented during this period would have been agre (...)

19The VAT transitional system, including the temporary measures on VAT rates described above, was supposed to be in place for a period of four years after 1 January 1993. A time plan was therefore agreed upon according to which the European Commission would bring proposals forward by the end of 1994, with a view to implementing a definitive VAT system based on the origin principle by 1997. Unfortunately, the Commission was unable to fulfil this time plan and it was not until the summer of 1996 that a work programme was presented for the adoption of the definitive VAT system30. Although formal legislative proposals were never put forward, the programme contained an outline of the envisaged system, as well as a detailed work plan extending through to mid-199931. Amongst the key features of the definitive VAT system, as foreseen under that programme, was the further harmonisation of the main aspects of the VAT system including rates.

  • 32 Council Directive 96/95/EC of 20 December 1996, OJ L 338, 28/12/1996, 89.

20This new attempt too was doomed to fail. The first setback came very soon after the presentation of the 1996 programme, as Member States failed to reach total agreement on the already tabled proposal regarding the establishment of a fixed band for standard rates of VAT. This included a minimum rate of 15% and a maximum rate of 25% and whilst Member States were able to agree on the minimum level, it was impossible to reach unanimity on a maximum level. Ultimately, the proposal was approved but the final text contained no reference to the maximum level of standard rate32. Thus, very little progress was made on the Commission’s proposed 1996 programme and it soon became clear that the degree of harmonisation necessary for the introduction of a definitive VAT system (particularly in terms of VAT rates) would not be achieved.

1.3. Further differentiation of rates structures

  • 33 This was in fact the Parliament’s opinion. According to a research paper published in 1995 by the P (...)
  • 34 See Report from the Commission on reduced VAT rates drawn up in accordance with Article 12 (4) of t (...)

21Since the approval of the Approximation of VAT Rates Directive, VAT rates, far from converging as might have been expected33, can diverge much more than under the legal framework set up in 1992. As reported by the European Commission in 2001, despite its tentative efforts to increase convergence, “when current rates are compared with those applicable in 1997, it is apparent that rates continue to vary considerably”34.

  • 35 Council Directive 1999/85/EC of 22 October 1999, OJ L277, 28/10/1999, 34.
  • 36 Council Directive 2002/92/EC of 2 December 2002, OJ L 331, 07/12/2002, p.27; Council Directive 2006 (...)
  • 37 See Experimental application of a reduced rate of VAT to certain labour-intensive services, Report (...)
  • 38 SEC (2003) 622, 2 June 2003, n. 37 above, at 28.

22The main reason for increased rate differentiation within Europe since 1992 has been the so-called labour-intensive services experiment. Implemented in 1999, the experiment allowed the application of reduced rates to certain labour-intensive services, such as hairdressing and window cleaning, with the aim of testing its impact on job creation and the combat against the “black market”35. Initially intended to last for three years, the experiment was consecutively extended despite disappointing results36. A report from the European Commission, published in 2003, confirmed that the impact of introducing reduced rates on prices of labour-intensive services was minimal. When conducting price surveys, Member States found that reduced rates were only partially reflected in consumer prices or not at all and that at least part of the VAT reduction was used to increase the margins of service providers. Moreover, even where the VAT reduction had been passed on to the consumers, Member States found that this was only a temporary measure and prices would subsequently increase37. Overall, the study concluded that, partially due to the lack of effect on prices, the aims of the experiment, namely to increase employment and to combat the black economy, had not be achieved38.

  • 39 Proposal for a Council Directive amending Directive 2006/112/EC as regards rates of value added tax(...)
  • 40 Council Directive 2009/47/EC of 5 May 2009, OJ L116, 09/05/2009, 18-20.

23Yet, the above results did not prevent Member States from either further extending the experiment, but moreover, from transforming the temporary experiment into permanent measures. In 2008 the European Commission put forward a new legislative proposal, which it designated as “a first action concerning reduced VAT rates” and as a “limited legislative proposal […] relating to urgent issues, which do not require any substantial additional study”39. The proposal had two objectives, both allowing for further differentiation of VAT rates: to make the possibility of applying reduced rates to certain labour-intensive services permanent, and to allow Member States the freedom to apply reduced rates to “locally supplies services”, such as restaurant services. The proposal was approved not long after its presentation, with the final legislative document essentially following its wording – both factors a clear indication that negotiations had been relatively straightforward, and that Member States were broadly in agreement with the new direction taken by the European Commission40.

  • 41 Proposal for a Council Directive amending Directive 77/388/EEC as regards reduced rates of value ad (...)
  • 42 The Council initiated its formal discussions on the Commission’s proposal at the ECOFIN meeting of (...)

24In the meantime, Commission’s attempts at limiting overall differentiation failed miserably. In 2003 the Commission presented a proposal with a view to “review and rationalise the use of reduced rates”. The proposal left considerably more freedom to Member States to decide on their own VAT rates structure than under previous Commission’s proposals, namely the 1987 and 1989 proposals. Obviously, it considered that by allowing increased freedom, the likelihood of Member States reaching unanimous agreement at the Council would also increase. Although not exceedingly ambitious, however, the proposal did envisage the move to a compulsory natured list of products which may be subject to reduced rates, which seems to have been sufficient to cause concern amongst Member States41. After years of discussions at the Council42, the proposal was finally approved in 2006 but at significant costs: the emphasis was no longer on rationalisation of reduced rates, but rather on the extension of the temporary rates provisions within the VA Directive, as well as on the extension of the list of products to which reduced rates may apply.

  • 43 Council Directive 2006/18/EC of 14 February 2006, OJ L51, 22/02/2006, 12.
  • 44 Study on reduced VAT applied to goods and services in the Member States of the European Union, Taxa (...)
  • 45 Consultation Paper: Review of existing legislation on VAT reduced rates, TAXUD/D1D/24232, 6 March 2 (...)

25The 2006 legislation also included a mandate from the Council to the Commission to present to the European Parliament and to the Council, by the end of June 2007, an overall assessment report on the impact of reduced rates on job creation, economic growth and the internal market43. This mandate has produced quick results. In 2007 the Commission published a study undertaken by Copenhagen Economics on the economic impact of the application of reduced rates44; and in March 2008 it launched a public consultation as part of its aim of “launching a broad debate in the Council, the European Parliament and with other stakeholders to obtain all relevant views before initiating a more far reaching proposal on reduced rates is the most effective approach to develop a sustainable and well balanced proposal in the medium term”45. For those awaiting the presentation of this “far reaching proposal”, early signs were not encouraging. As discussed above the European Commission seemed to be moving in the wrong direction: not only had the most recent proposal on VAT rates been aimed at increasing differentiation of rates, rather than the opposite, but equally the consultation paper expressly stated that the Commission was considering introduction of further reduced rates to, amongst others, environmentally friendly products. As will be seen below, this approach changed radically in the wake of the economic and financial crisis.

1.4. State-of-Play in 2008

  • 46 In addition to the legislation already mentioned, the following has also included some amendments t (...)
  • 47 See COM (96) 328 final, 22 July 1996, n. 30 above, at 5.
  • 48 Council Directive 2006/112/EC of 28 November 2006 on the common system of value added tax, OJ L347, (...)

26As discussed above, it is clear that although the provisions governing the rates structure have been subject to several amendments since the entering into force of the Approximation of VAT Rates Directive46, “the situation has changed little and the level of harmonisation of VAT rates has remained modest”47. At present the rates structure under the VAT Directive is a multiple-rate system, allowing for a standard rate and one or two reduced rates in theory (two more in practice), and subject to a few basic rules, as follows48:

    • 49 The date until which this minimum standard rate level will apply has been consecutively postponed. (...)

    The standard rate cannot be lower than 15% (Article 97 of the VAT Directive)49;

  1. Member States may apply one or two reduced rates to supplies of goods / services specified in Annex III, including labour-intensive services, but not where they are electronically supplied (Article 98 of the VAT Directive);

  2. Subject to certain conditions, reduced rates may also be applied to supplies of natural gas and electricity (Article 102 of the Common VAT System Directive), imports of works of art, collectors’items and antiques and certain supplies of works of art (Article 103 of the VAT Directive);

  3. During the transition period, i.e., until the entry into force of the definitive VAT system, Member States may maintain, under certain conditions, various special measures concerning the application of reduced rates, including: application of reduced rates lower than the authorised 5% minimum; maintenance of reduced rates for goods or services not covered by Annex III; or application of an additional reduced rate, known as the “parking rate”, no lower than 12% (Articles 109 to 122 of the VAT Directive); and,

  4. Finally, some Member States have been allowed to temporarily apply reduced rates to specific transactions (Articles 123 to 130 of the VAT Directive), and special rules also apply to the Austrian communes of Jungholz and Mittelberg and the Portuguese regions of the Azores and Madeira (Articles 104 and 105 of the VAT Directive).

  • 50 For a more detailed analysis of the VAT rates applied in the several Member States, the Commission (...)

27The described rules leave Member States significant freedom to establish their own rates structure. In practice Member States are free to decide on the following: whether to apply one or two reduced rates; whether to apply, subject to special conditions, an extra “parking rate”; the level of standard rate, as long as it is more than 15%; the level of the reduced rate (s), subject to certain conditions, which depend on each Member State’s specific circumstances; and to which goods / services to apply reduced rates too, subject to the conditions set out in the VAT Directive50. Unsurprisingly, this freedom resulted until recently in VAT rates structures within the EU remaining highly discrepant, highly differentiated, and highly complex. The high level of differentiation is particularly worrying, since in itself will almost always result in high level of discrepancy across Member States, and unavoidably in high level of complexity.

  • 51 The data was reported on the Commission’s report on reduced VAT rates, COM (2001) 599 final, 22 Oct (...)
  • 52 Articles 123 to 130 of the VAT Directive.
  • 53 Croatia joined the EU in July 2013.
  • 54 See Council Decision 2006/774/EC of 7 November 2006 authorising Member States to apply a reduced ra (...)

28Unfortunately, and until 2008, the rate differentiation in EU Member States was particularly extensive. As regards the “old” Member States, six (Greece, Spain, France, Ire land, Italy and Luxembourg) applied a reduced rate lower than the minimum laid down in Article 98 of the VAT Directive (a “super-reduced rate”); three (Belgium, Ireland and Luxembourg) applied a reduced rate not lower than 12% (the “parking rate”); five Member States (Belgium, Denmark, Finland, Italy and Sweden) applied a zero rate on a marginal and restricted basis; while Ireland and the United Kingdom continued to make extensive use of this derogation51. The situation was slightly different within the new Member States, but not radically so: six (Cyprus, Czech Republic, Estonia, Latvia, Malt a and Slovakia) applied a zero rate of VAT, and almost all were granted authorisations to introduce / maintain the application of rates which derogated from Articles 98 and 99 of the VAT Directive52. Member States’application of the labour-intensive services experiment was also a good example of the discrepancies that can emerge in the context of the application of reduced rates. In 2009 only eighteen, out of then twenty-seven Member States53, submitted applications to avail of the option to apply reduced rates to labour-intensive services. Of those eighteen Member States, each of them had chosen different services from the ones listed in the old Annex IV (now part of Annex III): twelve had chosen renovation and repairing of private dwellings; eight, small services of repairing; six, domestic care services; seven, hairdressing; three window cleaning and household cleaning services; and one minor services of repairing clothing and household linen54. With the transformation of the labour-intensive services experiment into a permanent feature of the EU rates structure, as discussed above, this level of differentiation looked more likely to increase, rather than to decrease. Instead the financial and economic crisis hit, and Member States approach to rate differentiation changed radically.

2. EU VAT Rates: Post-2008

  • 55 See OECD, Consumption Tax Trends 2012 – VAT/GST and Excise Rates, Trends and Administration Issues (...)

29According to the OECD, after a period of relative stability between 1996 and 2008, the average standard rate of VAT started to rise again after 200855. Indeed between 1 January 2009 and 1 January 2012 many OECD countries increased their standard and / or their reduced VAT rate, particularly EU countries (Czech Republic, Estonia, Finland, Greece, Hungary, Ireland, Italy, Poland, Portugal, Slovakia, Spain, and United Kingdom). Since 1 January 2012 further rate increases have been implemented in Cyprus, Czech Republic, Finland, Netherlands, Poland, Spain, and Slovenia; Italy is also said to increase its standard rate from September 2013, and France both its standard and reduced rates from January 2014. At the same time several EU Member States have also made substantial amendments to their tax base, moving goods and services from reduced to intermediate rate, or from reduced and intermediate to standard rates, including Italy, Portugal, and Spain. Overall twenty-five, out of the thirty-three OECD countries, and over half of EU countries, changed their VAT rate structures during this period.

  • 56 As A. J. EASSON prophetically pointed out in 1993, “the fiscal implications of EMU are obvious and (...)

30These numbers demonstrate the extent to which Member States turned to VAT policy in the wake of the economic and financial crisis. The reason is clear: confronted with high budget deficits and limited (or negative) economic growth, whilst at the same time deprived of the possibility of currency devaluation and bound to a common interest rate, Member States—specifically those which are part of the Eurozone—were confronted with serious limitations on their abilities to respond effectively; it was therefore unsurprising that most turned to tax policy as their preferred means of macro-economic intervention56. Within tax policy the weapon of choice seems to have been VAT.

31This focus at national level on VAT policy in the current economic climate is hardly surprising. Certainly it comes within the context of the general trend for a long-term shift towards indirect taxation, rather than direct taxation. This trend is based largely on the traditional economic view that consumption taxes are relatively more efficient as a revenue source, are less distortive, and have favourable effects on growth and employment. Thus, in many Member States, VAT has become the main source of national revenue: in 2009 it accounted for 21% of the tax revenues of EU Member States, an increase of 12% since 1995. Against the background of the economic crisis, however, these comparative advantages of VAT have become particularly significant: on one hand, national governments need additional revenue, and VAT presents itself as a more reliable and stable source of revenue than profits and income, especially in the current climate; on the other hand, the emphasis is also on economic growth as the only medium to long-term solution, with less distortive taxes becoming particularly appealing.

  • 57 European Commission, Green Paper on the Future of VAT—Towards a simpler, more robust and efficient (...)
  • 58 R. MAAS et al., “VAT Focus—Roundtable discussion: The Green Paper & the Future of VAT” (2011) Tax J (...)
  • 59 See European Commission, Summary Report of the Outcome of the Public Consultation on the Green Pape (...)
  • 60 Future of VAT—Towards a simpler, more robust and efficient VAT system tailored to the single market(...)

32Clearly keen to harness the political momentum, the European Commission presented in December 2010 the Green Paper on the Future of VAT57. The stated of the paper, which was said to be “one of the most important documents issued by the European Commission for some time”58, was to launch a broad based consultation process on the functioning of the current EU VAT system. Indeed, whilst the paper itself was hardly as ambitious as that aim might suggest, it was nevertheless far-reaching, covering many—albeit not all—of the most problematic areas of the system, including harmonisation of rates. A year later, amidst favourable reactions from other European institutions and various stakeholders59, the Commission issued a follow-up Communication, which had two stated purposes: in the long term, to set out the fundamental features of a future EU VAT system—a system which continues to raise revenue but which also increases competitiveness; and in the short to medium term, to list the priority areas for further action in the coming years—with a view to moving towards those objectives60. Amongst these listed priority areas was the review of the rate structure.

33In the Communication the Commission states that in order to increase the efficiency of the VAT system, it favours restricted use of reduced VAT rates. The use of reduced rates should then be based upon a few guiding principles:

  1. Abolition of those reduced rates which constitute an obstacle to the proper functioning of the internal market;

  2. Abolition of reduced rates on goods and services for which the consumption is discouraged by other EU policies;

  3. Similar goods and services should be subject to the same VAT rate.

  • 61 Ibid, at 11-12.
  • 62 European Commission, Review of existing legislation on VAT reduced rates, Consultation Paper, TAXUD (...)
  • 63 European Commission, Review of Existing Legislation on VAT Reduced Rates, Summary Report of the Out (...)

34The Commission set out the aim of launching in 2012 an assessment of the current VAT rates structure in the light of these guiding principles, and subsequently make proposals along those lines after ample consultation with stakeholders and Member States by the end of 201361. In this context it launched a public consultation in October 2012 on the review of the EU legislation on VAT reduced rates62. As opposed to previous initiatives which were broad in their scope, this was a very targeted public consultation: only nine questions, strictly framed by the guiding principles, eight of which concerned specific sectors of activity, namely the application of reduced rates of VAT to water, energy, waste and e-books. Despite the limited scope of the questions asked, the Commission also asked more generally for any “concrete examples of distortions of competition within the internal market or of specific problems encountered due to the current VAT rules”. Despite this effective broadening of the scope of the consultation, the targeted nature of the questions resulted in a low number of submissions from academics, tax advisors and tax practitioners; and on the contrary, a very high number – more than half of all submissions – from national or European associations, the large majority of them representing sectors currently benefiting from a reduced VAT rate63. Unsurprisingly, the nature of the respondents reflected heavily on the contents of the responses: most were opposed to the abolition of the reduced rates and/or advocating for their extension; and many challenged the general trend of shifting taxation away from labour towards consumption. Some submissions also defended that no further harmonisation should take place, and that the decision on whether or not to apply reduced VAT rates should be left to the Member States.

  • 64 See A Retrospective Evaluation of Elements of the EU VAT System—Final Report, TAXUD/2010/DE/328, FW (...)

35In the context of the outcome of this public consultation, it is pertinent to question whether this latest initiative can be successful. Reviewing the rate structure has been part of every Commission’s attempt to reform the EU VAT system—and with good reason. A recent study commissioned by the EU Commission indicates that a 50% reduction in the dissimilarity in VAT rates structures between Member States could result in a rise of 9.8% in intra-EU trade and an increase in real GDP of 1.1%64. Moreover, this is merely the last of several studies indicating the negative consequences of rate differentiation and its unproven positive effects. Yet, these studies in themselves have traditionally been insufficient to convince Member States to act. On the contrary, what has now made many Member States act at a domestic level has been the pressing need for extra revenue. Whilst no reference is made to this reality in the Communication, it is clear that the Commission is relying on that need in order to push this measure forward—the fact that so many Member States have already taken this political choice at national level might just be enough to create the necessary momentum for agreement at EU level.

  • 65 See R. DE LA FERIA, n. 5 above.
  • 66 This point is developed further in R. DE LA FERIA, “The 2011 Communication on the Future of VAT: Ha (...)

36This signals a significant shift in approach to VAT harmonisation by the European Commission. For the last four decades the Commission’s approach has been primarily to convince Member States that harmonisation is an essential step for the establishment and the functioning of the European Internal Market. Although one can certainly agree with that statement65, the reality is that this approach has failed consistently to create the necessary political enthusiasm for reform. In essence, it has failed to rally the troops. The approach now adopted by the European Commission is very different: there are comparatively few references to the EU perspective, and indeed there is only one reference in the entire Green Paper on the Future of VAT to the “Internal Market”; instead the focus is clearly on “consolidation of public finances” and “sustainable economic growth”. In the midst of the economic and financial crisis, the Commission has clearly re-packaged long-sought reform by offering Member States EU answers to national needs—and that is why this time, it might just succeed66. Furthermore it must also be acknowledged that the limited, or specific, nature of the review now being considered may make it politically easier to attain Member States’ agreement on. This limited nature of the review, however, also raises concerns namely on whether, even if successful, is this proposed review worthwhile? Certainly it would result in an improvement to the current EU VAT rate structure, but not a massive one. Essentially there is a trade of: lower risks, lower returns; such are the costs of political realism.

3. European VAT Rate Structures: Criteria for Reform

37Whether the latest European Commission’s initiative on the review of the EU VAT rate structure gathers the necessary support or not, it is clear that only limited improvements to the structure can be achieved. Therefore, if significant gains are to be achieved, they must come through a different route; and if the political momentum is to be seized, it is necessary to think outside the box. In this context, would it be possible to have significantly improved, converging, European VAT rate structures, through national, uncoordinated, action?

3.1. Ideal VAT rate structure

38As discussed above, the original introduction of reduced VAT rates was based not so much on clearly articulated policy objectives but rather on pragmatic political goals, as designers of the VAT sought to replicate the impact of the predecessor turnover taxes and deflect concerns about the tax on beneficiaries of previous concessions. Over time, however, it was argued that the use of reduced rates achieves social and distributional aims, and namely three ex post facto rationales have been offered, as follows:

  1. Vertical equity: idea that these concessions limit the natural regressivity of VAT, i.e. that the tax weights more heavily on poorer households; so applying reduced rates to key products such as food, energy, healthcare, education, etc, would limit the impact of this tax on those households.

  2. Positive externalities: idea that these concessions increased consumption of socalled merit goods, such as cultural events, books, sport activities, etc.

  3. Increase employment: idea that application of reduced rates will ultimately lead to increase employment in labour-intensive industries (e.g. hairdressing), or areas where price is particularly elastic (e.g. electronics), or both (e.g. restaurants).

39Once it is accepted that the application of reduced rates amounts to tax expenditures, these should be subject to a cost-benefit analysis similarly to direct expenditure programs: what are the benefits of applying reduced VAT rates? I.e. does application of these rates actually achieve social and distributional aims? And even if it does, what are the collateral costs, from both a legal and an economic perspective?

3.1.1. What are the benefits of applying reduced VAT rates?

  • 67 It has been argued that VAT is regressive with respect to income but not necessarily with respect t (...)

40The equity argument derives from the fact that the proportion of income that is saved reduces as income reduces, with the lowest income earners using all their income for consumption and diverting none to savings. As VAT falls only on income used for consumption and exempts income that is applied to savings, the tax is said to fall more heavily on lower income persons than on higher income persons in terms of the proportion of income derived by those persons67. Reduced rates for commodities that form a higher percentage of the spending budget of lower income persons are seen as a way of reducing the tax burden on these persons, and thus increasing their consumption capability.

41The positive externalities rationale for exemptions derives from a belief that the market price for some types of supplies does not fully reflect the overall benefits from consumption of those supplies for society as a whole, and thus government intervention to subsidise consumption of those goods is deemed desirable.

42The job creation argument has been developed relatively recently, when compared with the other two rationales for the use of reduced VAT rates. It derives from the belief that price decreases resulting from the introduction of reduced rates will lead to increase in demand, which in turn will result in increased supply. In labour-intensive services that increased supply will necessarily lead to new job creation. This argument attained political endorsement within the EU in the late nineties, leading to the approval of the so-called labour-intensive services experiment in 1999.

  • 68 See SEC (2003) 622, 2 June 2003, n. 37 above, at 26.

43Crucially, these arguments – i.e. that application of reduced rates contributes to vertical equity, increases the consumption of merit products, and leads to job creation – presuppose that the decrease in the VAT rate is reflected in consumer prices. Theoretically, this should indeed be the case: generally, in a competitive market if costs go down (including taxes), so should prices. However, recent empirical experiments with VAT rates seem to indicate the opposite. The first and most significant has been the labour-intensive services experiment discussed above. The second experiment to assess the impact of reduced rates on prices took place in Ireland. Struggling with high levels of inflation, as a collateral effect of their outstanding economic growth, Ireland decided to reduce the standard VAT rate from 21% to 20% from January 2001. In a speech in December 2000, the Irish Finance Minister stated that: “The government expects to see the VAT reduction passed on to the consumer and not absorbed in higher retail margins. If this does not occur, the wisdom of further VAT cuts will be placed in doubt. We will be monitoring the situation and I hope consumers will be vigilant in seeing that the VAT reduction is passed on to them”. In 2002, Ireland decided to raise back the rate of VAT from 20% to 21%. In a speech in December 2001, the Finance Minister stated that the lower rate of VAT had not been passed on to consumers: “I had reservations about cutting that rate last year. I said that I would be looking to see if it was fully passed on. I am not convinced that this was the case.”68

  • 69 See Copenhagen Economics, Study on reduced VAT applied to goods and services in the Member States o (...)
  • 70 For example, there might be geographical restrains, with only one hairdresser, or repair shop in a (...)

44How to explain this discrepancy between theoretical and empirical results? A convincing explanation has yet to be given. A study published in 2008 has suggested that the empirical results of the labour-intensive services experiment might be due to its temporary nature, i.e. if firms know that a lower VAT rate is temporary, why would they use time and money to expand production capacity and incur costs if they have to revert to their previous production level within a few years69. It is also possible that, labour-intensive services do not operate in fully competitive markets70, and that a decrease of 1% in the rate of VAT is too minimal to be passed on. Finally, it is also worth noting that both experiments took place in a boom economy, where it is possible that demand outweighed supply. Yet, these are merely tentative explanations: in practice, until definite arguments are presented all that can be said with certainty is that evidence so far does not support the argument that reduced VAT rates reduce prices.

  • 71 For early studies on the effects of the VAT temporary standard rate reduction in the UK in 2009, wi (...)
  • 72 See A. CHARLET and J. OWENS, “An International Perspective on VAT” (2010) Tax Notes International 5 (...)

45The recent changes to VAT rate structures, which took place in various Member States in the context of the current economic crisis, will offer new opportunities for assessing the incidence of VAT, and the impact of reduced rates on prices, in particular in the context of a downturn economy71. In fact some initial examples, which took place in the beginning of the economic crisis, are already available, appearing to add support to the results of previous experiments72. In 2009, France dropped the VAT rate from 19.6% to 5.5% for supplies of restaurant and catering services on the assumption that restaurants would reduce prices substantially, raise wages, or create new jobs, and improve compliance. However, according to the French National Institute for Statistics and Economic Studies, the decrease in prices for restaurant services was not minimal – around 1%-but also temporary. According to the French authorities, if only 30% of the VAT cut had been passed on to customer, this would create 6,000 jobs in the long run, but the government stood to lose up to €3 billion in revenue in a full fiscal year from the cut; this would equate to each new job in the sector costing French taxpayers €500,000. In 2010 Germany reduced the VAT rate applicable to the hotel industry as part of a more general tax cut. A recent survey indicated that the cut had not been passed on to consumers, and instead prices had remained the same.

  • 73 See Copenhagen Economics, n. 69 above, at 12.
  • 74 C.L. BALLARD and J.B. SHOVEN, using US data, conclude that differentiation of rates used to partial (...)
  • 75 See OECD, n. 55 above, at 73.
  • 76 Godbout and St. Gerny show that with tax credits targeted at lowest income individuals in place, th (...)

46If reduced VAT rates cannot reduce prices, then the logical conclusion is that they cannot attain the distributional and social aims that they are set up to achieve. However, even if one assumes that the above empirical results are flawed, and that indeed reduced rates of VAT do affect prices, there are still no certainties that distributional and social aims are, or can be, reached. A recent empirical study seems to indicate that the effectiveness of applying such rates depends on the elasticity of specific products: in the case of basic goods, such as food, consumers react only weakly to lower prices (where consumption is price in-elastic), so production and employment will not increase significantly; in contrast, if consumers react strongly to new prices, as in the case for high value goods, such as package holidays, books, and electronic equipment (where consumption is price elastic), production and employment may increase significantly73. Moreover, other economic studies have consistently shown that since VAT is not an effective method of pursuing distributional goals, and it is far better to tax as broadly as possible74, using the yield to compensate low-income households75. High-income households typically consume more of basic necessities than low-income households. In this context, if items currently subject to reduced rates were fully taxed – personal income tax relief or means-tested social security benefits – the government could more effectively achieve social and distributional aims, and have additional revenue left over to apply to other redistributive programs76. In this sense, lower income persons may be much worse off with a tax system that contains reduced VAT rates designed to assist them, than they would be in a tax system with one single rate and redistribution of the excess revenue raised under a more neutral tax base. In addition, job creation or protection of key sectors of the economy, would also be better achieved through direct subsidies.

3.1.2. What are the costs of applying reduced VAT rates?

  • 77 E.H. DAVIS and J.A. KAY provide amusing examples with reference to the United Kingdom to illustrate (...)

47The benefits of applying reduced VAT rates are therefore questionable. Moreover, the costs of subsidising consumption of target goods and services in this manner are on the contrary likely to be significant. From a legal perspective application of reduced rates gives rise to definitional and interpretative problems, and constitute an incentive to engage in aggressive tax planning. For these reasons reduced rates tend to result in substantial – and increasing – litigation, which in turn results in substantial compliance and administrative costs77.

  • 78 See cases 415/85, Commission of the European Communities v Ireland, [1988] ECR 3097; 416/85, Commis (...)
  • 79 Case C-259/10, Commissioners for Her Majesty’s Revenue and Customs v The Rank Group plc, [2011] ECR (...)
  • 80 See V. SLOANE, “VAT Focus – Rank and Fiscal Neutrality” (2011) Tax Journal 1101, 20, 18 November 20 (...)
  • 81 See European Commission, Taxation: Commission refers France and Luxembourg to the Court of Justice (...)

48Symptomatic of this increase in litigation is the number of cases brought before the Court of Justice of the European Union (CJEU) in relation to the application of reduced rates by Member States to various goods and services78. At stake in many of these cases was the application of reduced rates to specific products, whilst other similar products were subject to standard rates. The Court has consistently emphasised the importance of respecting the principle of fiscal neutrality: the application of reduced rates to certain products must be consistent with this principle that precludes treating similar goods, which are therefore in direct competition with each other, differently for VAT purposes. In a recent case concerning exemptions the Court has gone further in application of the principle of fiscal neutrality by stating that Member States cannot apply different VAT treatments to services that are comparable to each other from the point of view of the customer or meet the same needs of the customer79. Following this decision the debate has been on whether the new criteria will have implications for the Court approach to VAT rates structures. Some have already been defending that it will, stating that it is “highly likely” that the criteria laid down in Rank Group will affect the application of VAT rates, particularly to food80. The big test should come soon with the eagerly expected decisions in the e-books cases, where the Court has been called to decide on whether e-books can be subject to reduced rates of VAT similarly to hardcopy books81.

  • 82 United Biscuits (UK) Ltd (No 2) (LON/91/160) VAT Decision 6344.

49National courts too have been struggling with similar difficulties, and in this regard, two United Kingdom court cases in the last decade are particularly telling: Marks & Spencer and Pringles. Marks & Spencer concerned the classification of a particular type of the retailer’s teacakes: a chocolate covered marshmallow. In the UK cakes are zero rated, whilst biscuits are subject to standard VAT rate. The rationale for this distinction is unclear but it seems to be connected with health considerations, as well as with the idea that some “luxury” food products should not benefit from reduced rates. From 1973 to 1994 Her Majesty Revenue and Customs (HMRC) took the view that Marks & Spencer (M & S) teacakes should be regarded as biscuits and thus subject to the standard rate of VAT, rather than cakes. In 1994 however, HMRC had to rethink this classification in the wake of the landmark UK court decision in Jaffa Cakes82.

50Jaffa Cakes comprise three layers, a sponge cake base, a layer of orange flavored jelly, and a layer of chocolate covering the jam. For several years McVities, the company which produces Jaffa Cakes treated them as zero rates cakes. However, in 1991, this classification was challenged by HMRC, in particular on the basis that Jaffa Cakes are the same size and shape as biscuits. The case was brought before the VAT Tribunal, with a central question: what criteria should be used to class something as a cake, rather than a biscuit? McVities defended its classification of Jaffa Cakes as cakes, by producing a giant Jaffa Cake in order to illustrate that its Jaffa Cakes were nothing more than miniature cakes. It also argued that a distinction between cakes and biscuits is, inter alia, that biscuits would normally be expected to go soft when stale, whereas cakes would normally be expected to go hard. It was demonstrated to the Tribunal that Jaffa Cakes become hard when stale. Other factors taken into consideration by the Tribunal included the name, ingredients, texture, size, packaging, marketing, presentation, appeal to children, and manufacturing process. Ultimately, the Tribunal agreed with the McVities that Jaffa Cakes, where indeed cakes, rather than biscuits, and thus should be zero rated.

  • 83 [1997] VAT & Duties Report 85.
  • 84 [1999] Simon Tax Cases 205, and [2000] Simon Tax Cases 16.
  • 85 Case C-309/06, Marks & Spencer plc v Commissioners of Customs and Excise, [2008] ECR I-2283.
  • 86 This argument is further developed in M. DEVEREUX and R. DE LA FERIA, “VAT – Unjust Enrichment” (20 (...)

51Following the Jaffa Cakes decision, HMRC acknowledged that M & S teacakes should also be classified as zero-rated cakes. M & S therefore asked for a refund of VAT overpaid between 1973 and 1994. HMRC refused to pay the totality of the VAT claimed on the basis that M & S had passed on the VAT to the final consumers, and the case was brought before the UK courts. At the initial hearings the VAT and Duties Tribunal concluded on the evidence of expert witnesses that 90% of the VAT paid by M & S had been passed on to customers and any consequential economic loss would not amount to more than 10% of the tax paid83. Thus the Tribunal concluded that M & S would be unjustly enriched if repaid the full amount of VAT claimed; instead the proper amount repayable under s 80 would be 10% of that amount. On appeal by M & S both the High Court and the Court of Appeal agreed with the Tribunal’s decision on the basis of the economic evidence presented84. The reference to the CJEU arose in the context of M & S’s final appeal to the House of Lords. In its decision, the Court concluded that whether repayment would result in unjust enrichment, and to what extent, could only be established “following an economic analysis in which all relevant circumstances are taken into account”85. The Court concluded that it is for the UK House of Lords to determine whether the appraisal made by HMRC fulfils these conditions. The outcome of the case thus became dependent on the incidence of the tax, which is notoriously difficult to establish86.

  • 87 See G. MORSE, “Proctor & Gamble UK v HMRC (Pringles Two) – a very peculiar UK practice – the charac (...)

52Pringles concerned the classification of their popular snack for VAT purposes. In the UK food is usually zero rated, however potato crisps are specifically excluded from this rule, and are thus subject to standard rate (17.5%). The rationale for this distinction is, as before, unclear but in this case seems to have been health considerations which guided the legislator, although the first instance judge in this case, John Avery-Jones, has recently stated that the had been unable to establish any purpose for this rule. Following a dispute between Procter & Gamble (P & G), Pringles’ manufacturer and HMRC, the case was sent to the VAT and Duties Tribunal, which concluded in 2006 that Pringles must be regarded as “a potato crisp product”, and thus subject to the standard 17.5% rate. P & G appealed, insisting that their product was not similar to potato crisps, because of their “mouth melt” taste, “uniform colour” and “regular shape”, which are not found in nature. In its decision dated July 2008, the High Court upheld the appeal. The Court considered that Pringles should not be regarded as potato crisps. In order to be classified as such, a product “must be wholly, or substantially wholly, made from the potato”, whilst Pringles are made from potato flour, corn flour, wheat starch and rice flour together with fat and emulsifier, salt and seasoning, with a potato content of only 42%87.

53In addition to highlighting definitional and interpretative difficulties, these rulings also demonstrate the pitfalls of attempting to attain distributional and social aims through reduced rates: biscuits and potato crisps were excluded from the scope of application of the reduced rates because the legislator deemed these products as not fulfilling those distributional and social aims; yet, similar products like M & S teacakes, or Pringles are benefiting from reduced rates; the result being that the tax system is de facto subsidising those products, in detriment of competing products. The consequences for fiscal neutrality are obvious: treating competing products differently for VAT purposes is bound to create distortions of competition. In addition there might be unexpected detrimental effects: the ruling in Pringles, for example, creates an incentive for producers of potato crisps to reduce the potato contents in their products in order to benefit from a reduced rate.

  • 88 See OECD, n. 55 above, at 113. For a useful explanation of the methodologies used for establishing (...)

54Whilst difficult to quantify, the costs of these distortions of consumption and investment decisions may be extremely significant. Reduced rates of VAT erode the tax base, and importantly may subsidise inefficient production – since the suppliers of products subject to reduced rates do not have to compete on a level playing field with suppliers of products subject to standard. The result is a significant decrease in efficiency of the tax, as measured by the IMF and the OECD, which shows that European countries’ VAT systems tend to rank below the OECD c-efficiency ratio, or revenue ration, average, which stands at 55 points out of 100 possible – suggesting that about half of the potential VAT revenue is not collected by Member States88.

  • 89 See also S. CNOSSEN, “Value-Added Tax and Excises: Commentary” in S. Adam et al (eds.), n. 74 above
  • 90 See A. CHARLET and J. OWENS, “An International Perspective on VAT” (2010) Tax Notes International 5 (...)
  • 91 See L. EBRILL, M. KEEN, J-P. BODIN and V. SUMMERS, The Modern VAT (Washington DC: IMF, 2001). See a (...)

55The results of the cost-benefit analysis as applied to reduced VAT rates is therefore particularly negative: not only it is unclear whether they accomplish any of the social and distributional objectives that they set out to achieve, but they also carry significant costs beyond the mere loss of potential revenue89. The ideal VAT is therefore a broad-based VAT, with a single rate. This much has been consistently defended by the OECD since the 1980s90, and was recently supported by the European Commission in the 2010 Green Paper on the Future of VAT. This has also been the position of the IMF, which has recommended the introduction of a single-rate VAT system to many countries around the world91. Such a VAT rate structure, however, would be extremely difficult – if not impossible – to implement in Europe; so the question is, what is an achievable VAT rate structure.

3.2. Achievable VAT rate structure

  • 92 As acknowledged by the OECD, see n. 55 above, at 73.

56In light of the above, any reform of national VAT rate structures with a view to having a significantly improved structure, must take in consideration various factors. First, it must be acknowledged that, in the immediate term, moving products from reduced to standard rate is likely to have a significant economic impact, namely in the context of the high standard rates applied in almost all Member States that mean that this move could represent as much as a 20% or 15% tax hike. This economic impact could be reflected in higher prices, which would hit the poorest households the hardest, or in increased unemployment: one can imagine that in price inelastic sectors, such as food or utilities, prices will most likely increase92; in other price elastic sectors, where an increase in price might lead to a considerable contraction in consumption, suppliers may opt to maintain prices, but will need instead to decrease costs, which in a labour-intensive sector, such as restaurant services or tourism, will lead to job loses.

  • 93 See European Commission, n. 63 above.

57Secondly, it must be accepted that in the current financial environment that most European countries find themselves in – and not just the ones which benefited from a bail-out agreement – in the context of problematic budget deficits, and significant financial restrains, the likelihood of introduction of measures at personal income tax or social security level, to compensate the VAT hike, is small at best. This concern was indeed expressed by several respondents to the European Commission’s latest public consultation on review of reduced rates: in the current economic climate respondents expressed fear that there might be no national compensating measures, or that they would be insufficient93.

  • 94 There is a growing literature on the intersection between tax and state aid law, in particular foll (...)

58Finally, any reform of nationals VAT rate structures must take into consideration EU law limitations. As opposed to exemptions, application of reduced rates under the VAT Directive is non-compulsory, i.e. the Directive establishes maximum standards of differentiation – number of rates, number of products to which reduced rates can be applied – but does not establish a minimum level of differentiation; Member States are free to apply reduced rates to as limited number of products as they wish, and ad extremis are even free to apply only one rate. Therefore extension of the VAT base through elimination of reduced rates is not subject to any EU law limitations. However, the implementation of compensatory measures may be; in particular, the freedom to introduce measures to compensate labour-intensive or key economic sectors for the increase in VAT rates in might be severely reduced. Within the EU, national subsidies to specific industries, either in the form of tax relief / incentives or direct subsidies, are limited by state aid law94.

59In light of the above limitations, what would be suitable criteria for better, more efficient, more neutral, European VAT rate structures? Four criteria are proposed, as follows.

  1. Criterion 1: Elimination of application of reduced rates of VAT, where the rationale for its application is the creation of positive externalities and/or correction of externalities. There are various arguments to support this criterion. First, it is notoriously difficult (and subjective) to attach positive externalities to specific products; for example, few may argue against the positive externalities of reading, yet do all books or magazines hold positive externalities? Do celebrities’ biographies, or astrology books? And even if so, are these potential positive externalities sufficient to justify a government subsidy? Different people will hold different views. Second, goods or services which are usually perceived as holding positive externalities, such as books or cultural events, are statistically much more likely to be consumed by high-income households. So that applying reduced rates to these products constitutes a de facto subsidy from poor-income to high-income households, thus holding negative distributional effects. It has been argued that the maintenance of reduce rates for these products has an aspirational value. Even if that is the case, is it legitimate to ask low-income households to subsidise attendance to theatre plays, or the opera, by high-income households? Third, these products are by nature price elastic, so it is unclear to what extent prices will be affected by a VAT rate increase. It is possible that they will be an effect on employment, in the context of a possible need to decrease costs, but it is worth keeping in mind that these are not usually labour-intensive industries for unqualified workers, but quite the opposite: they tend to employ small number of qualified workers. Finally, in the context of the current financial and economic crisis, encouraging the consumption of products which hold positive externalities is hardly a priority!

  2. Criterion 2: Maintaining the application of reduced rates of VAT where the rationale for is application is vertical equity. The basis for this criterion is the low price elasticity of these products; prices will most likely increase, hitting low-income households hardest. However, given that high-income households consume considerably more of these products, it makes sense to limit the application of reduce rates to those categories of goods and services which are truly essential, such as food.

  3. Criterion 3: Maintaining the application of reduced rates of VAT where its elimination would have a serious impact on industries which are either labourintensive or key for economic recovery. The arguments in favour of maintaining reduced rates for these sectors are based on keeping competitiveness of national products in the international market, and employment concerns. In principle the tax hike could be absorbed by suppliers by decrease in their margins, but considering the size of the hike it is likely that at least part of the increase will have to be passed on to consumers in higher prices, or to employees in lower salaries / job losses. Both options carry economic risks for key sectors of the economy and those which are labour-intensive: if passed on in higher prices there is a risk of decrease competitiveness for exporting sectors of the economy, which in labour-intensive sectors can have the added effect of increased unemployment; even for non-exporting sectors, if price elastic, it is more likely that the VAT hike would be passed on employees, as increase in prices would lead to contraction in consumption, and then again there would be a significant risk of job losses.

  4. Criterion 4: Rationalisation of categories of goods and services to which reduced rates of VAT apply, by eliminating distinctions within categories, and limiting the use of different rates to different products within the same category. Distinctions within categories are the main sources of interpretative and definitional difficulties; elimination of these distinctions would therefore lead to higher legal certainty, be a disincentive to planning, abuse and fraud, and decrease significantly the potential for litigation – all of which would in turn result in lower compliance and administrative costs. Elimination of distinctions would also avoid other economic distortions, such as product manipulation so as to avail of the reduce VAT rate.

60The proposed criteria were used as a basis for the Portuguese VAT rate structure reform in 2012. Under the bail-out agreement signed with the EU and the IMF in 2011, Portugal was required to reform its VAT, which was deemed to be highly inefficient and if reformed offered potential to help the Portuguese Government reduce its budget deficit. The required refo rm, based on the above criteria, was implemented in the 2012 State Budget, and it resulted in the following key changes to the existing rate structure:

  1. Cultural events, sports activities and environmentally friendly products were moved from the reduced and intermediate rates to the standard rate (on the basis of criterion 1).

  2. Non-essential food and beverages, take-away and restaurant services were moved from the reduced and intermediate rates to the standard rate (on the basis of criterion 2).

  3. Hotel accommodation and tourism-related services, as well as agricultural inputs have been kept at reduced and intermediate rates (on the basis of criterion 3).

  4. Distinctions within categories of foodstuff have been eliminated, so that specific categories are either subject to reduced or to standard rates (on the basis of criterion 4).

  • 95 Although it has been recently suggested that an imperfect extension of the VAT base to cover more b (...)

61The reform resulted in a 30 points reduction in tax expenditure, as well as a significant increase in the C-efficiency level, which before the reform stood at 44 points95. Consumption has contracted significantly; however, until reliable price data is available, it is difficult to dissociate the extent to which the contraction resulted from the VAT base broadening, from the contraction that it would have happen as a result of the economic and financial crisis regardless of any tax hikes. Consequently VAT revenue has increased, but at lower levels than expected.

  • 96 See “Taxing Matters”, Society Now, Spring 2013, 25.

62Despite the somewhat disappointing short-term results in terms of revenue collected, the Portuguese reform of the VAT rate structure was broadly complimented by the EU and the IMF. The IMF Country Report on Portugal at the time of the reforms stated that as a result of these the VAT tax base levied at the standard rates was enlarged from 60 to 80% of the total base, which would generate savings of about 1.2% of GDP96. The report from the European Commission referred to additional revenues of 1.4% of GDP, stating:

  • 97 See European Commission, The Economic Adjustment Programme for Portugal – Third review-Winter 2011/ (...)

63“Following past increases in the VAT rates, the 2012 budget focused mostly on broadening the tax base […] In order to protect vulnerable groups, many essential goods remain subject to the 6% reduced rate and this rate also continues to be applied to goods considered crucial for domestic production, such as wine. Overall, the measures will help to significantly increase VAT efficiency.”97

64The success of these measures – even if more limited than expected – allowed the Portuguese Government to focus on introducing amendments to the VAT legislation to promote growth, in particular by helping small and medium-sized businesses, in the 2013 State Budget. Measures introduced included the simplification of the bad debts regime, and a cash-flow tax accounting scheme for companies with turnover below €500,000.

4. Conclusion: Unilateral Convergence of European VAT Rate Structures

65EU agreement on reduced VAT rates is difficult to achieve; and even if achievable, it will result in only minor improvements to the current EU VAT rate structure. In this context it is necessary to consider whether it would be possible to have significantly improved, converging, European VAT rate structures, through national, uncoordinated, action.

66Implementation of an ideal VAT by Member States – i.e. a single-rate system with compensatory measures low-income households, and key sectors of the economy – is conditioned by political constrains present in most European countries, as well as significant budgetary limitations. In that context the criteria proposed for reform of national VAT rate structures will not result in the best VAT possible, but rather in the best VAT Member States can possible have given the circumstances. A broader-based VAT, resulting in increased revenue, decreased administrative and compliance costs, less susceptible to fraud, avoidance and planning; overall a more efficient, more neutral VAT. In the process the holy grail of EU VAT might be finally attained: decreased divergence and even approximation of VAT rates structures across the EU. Not through a process of EU harmonisation, but instead through a process of natural convergence of national VAT policies – a rare case of significant gain, with limited pain.

Notes

1 First Council Directive 67/227/EEC of 11 April 1967 on the harmonisation of legislation of Member States concerning turnover taxes, OJ 71, 14/04/1967, 1301 (hereafter “First VAT Directive”); and Second Council Directive 67/228/EEC of 11 April 1967 on the harmonisation of legislation of Member States concerning turnover taxes - Structure and procedures for application of the common system of value added tax, OJ P 71, 14/04/1967, 1303 (hereafter “Second VAT Directive”).

2 Article 9 of the Second VAT Directive.

3 See R. DE LA FERIA and R. KREVER, “Ending VAT Exemptions: Towards A Post-Modern VAT” in R. DE LA FERIA (ed.), VAT Exemptions: Consequences and Design Alternatives (The Hague: Kluwer Law International, 2013), 3-36. See also S. CNOSSEN, “What Rate Structure for a Value-Added Tax?” (1982) National Tax Journal 35 (2), 205-214, at 209; V. LENOIR, “April 1954–April 2004–VAT Exemptions: The Original Misunderstanding” (2004) European Taxation 10, 456−459, at 456−457.

4 Sixth Council Directive 77/388/EEC of 12 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes – Common system of value added tax: uniform basis of assessment, OJ L 145, 13/06/1977, 1 (hereafter “Sixth VAT Directive”).

5 Completing the Internal Market – White Paper from the Commission to the European Council, COM (85) 310, 14 June 1985. For a detailed analysis of the VAT measures in the White Paper see R. DE LA FERIA, The EU VAT System and the Internal Market (Amsterdam: IBFD, 2009), at 57 et seq.

6 White Paper, COM (85) 310, 14 June 1985, n. 5 above, at 52.

7 Proposal for a Council Directive completing the common system of value added tax and amending Directive 77/388/EEC – Approximation of VAT rates, COM (87) 321 final/2, 21 August 1987.

8 However, although theoretically the United Kingdom had a single VAT rate system, in practice it applied zero rating in a large scale.

9 See COM (87) 321 final/2, 21 August 1987, n. 7 above, at 1.

10 Completion of the Internal Market: approximation of indirect tax structures and harmonisation of indirect tax structure, Global Communication from the Commission, COM (87) 320 final, 5 August 1987, n. 6 above, at 10. However, it should be noted that, at the time, the threerate system under consideration by the Commission entailed standard, reduced and increased rates, and not standard and two reduced rates, as currently applies.

11 See Completion of the Internal Market: approximation of indirect tax structures and harmonisation of indirect tax structure, Global Communication from the Commission, COM (87) 320 final, 5 August 1987, at 11-12; and COM (87) 321 final/2, 21 August 1987, n. 7 above, at 2.

12 Mostly if compared with the list which was ultimately approved by Council Directive 92/77/EEC and still currently in force, see Annex H to the Sixth VAT Directive. However, as the Commission significantly pointed out, these items represented approximately one third of the common Community tax base, see COM (87) 320 final, 5 August 1987, n. 11 above, at 11.

13 COM (87) 320 final, 5 August 1987, n. 11 above, at 10.

14 See COM (87) 320 final, 5 August 1987, n. 11 above, at 2.

15 As discussed infra, the tactic proved ultimately unsuccessful, as the proposal failed to get Council approval due to a great extent to the compulsory nature of the list.

16 COM (87) 320 final, 5 August 1987, n. 11 above, at 13. For a more detailed report on the Commission’s position regarding zero-rating see at 12-14.

17 See COM (87) 320 final, 5 August 1987, n. 11 above, at 14.

18 Proposal for a Council Directive instituting a process of convergence of rates of value added tax and excise duties, COM (87) 324 final/2, 21 August 1987. In fact, as the name indicates, the proposal did not regard exclusively VAT, but also excise duties.

19 See COM (87) 324 final/2, 21 August 1987, n. 18 above, at Article 1.

20 A.J. EASSON expressed a widely felt scepticism when he commented: “to expect to achieve an alignment of tax rates by 1992, by the end of the century or even by the middle of the next one, is to be completely unrealistic. The approach which has been exposed by the Commission for the past twenty-five years or so is doomed to fail”, in “The Elimination of Fiscal Frontiers”, in R. BIEBER et al (eds.), 1992: One European Market? A Critical Analysis of the Commission’s Internal Market Strategy (Baden-Baden: Nomos Verlagsgesellshaft, 1988), 241-260, at 260.

21 The Economic and Social Committee, for instance, recognised that the Commission’s proposal was pragmatic and represented an intermediate position, it considered however that it could still create severe problems for some Member States, and thus, urged the Commission to examine alternatives to its present proposals, see Opinion of the Completion of the Internal Market: Approximation of indirect tax rates and harmonisation of indirect tax structure (Global Communication from the Commission), 88/C237/05, OJ C237, 12/09/1988, 14.

22 Completion of the Internal Market and Approximation of Indirect Taxes, Communication from the Commission to the Council and to the European Parliament, COM (89) 260 final, 14 June 1989, at 1.

23 Resolution on completion of the Internal Market: Approximation of indirect taxation in the Community up to 1993 and thereafter– Report of the Committee on Economic and Monetary Affairs and Industrial Policy on the approximation of VAT rates, Rapporteur: Alan Metten, A3 427/92, OJ C183, 15/07/1991, 293.

24 Additionally, it was also suggested at the time that the proposal would weigh more heavily on some Member States than others, namely on Denmark, United Kingdom, Ireland and Luxembourg, see M. BOS and H. NELSON, “Indirect Taxation and the Completion of the Internal Market of the EC” (1988) Journal of Common Market Studies XXVII (1), 27–44, at 38-39.

25 Legislative Resolution embodying the opinion of the European Parliament on the Commission proposal for a Council Directive supplementing the common system of value added tax and amending Directive 77/388/EEC–Approximation of VAT rates, 12 June 1991, OJ C183, 15/07/1991, 122.

26 Council Directive 92/77/EEC of 19 October 1992, OJ L 316, 31/10/1992, 1 (hereafter the “Approximation of VAT Rates Directive”).

27 As discussed above, insofar as reduced rates are concerned, the Commission’s 1989 alternative rate structure did not differ from its original 1987 proposal.

28 As reported by the Commission in Report from the Commission to the Council in accordance with Articles 12 (4) and 28 (2)(g) of the Sixth Council Directive of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes – Common system of valueadded tax: uniform basis of assessment, COM (94) 584 final, 13 December 1994.

29 A Directive on works of art, antiques and collector’s items was approved in 1994, and in 1998 the Council approved the Directive on investment gold. These are Council Directive 94/5/EC of 14 February 1994, OJ L60, 03/03/1994, 16; and Council Directive 98/80/EC of 12 October 1998, OJ L281, 17/01/1998, 31, respectively. In 1994 the Commission put forward a proposal in respect of agricultural outputs – Proposal for a Council Directive amending Directive 77/388/EEC on the common system of Value Added Tax (taxation of agricultural outputs), COM (94) 584 final, 13 December 1994 – however, the Council was unable to reach an agreement and the Commission had no other option except to withdraw this proposal.

30 See A common system of VAT – A programme for the Single Market, COM (96) 328 final, 22 July 1996.

31 This means that in practice even if the proposals presented during this period would have been agreed upon by Member States, the definitive VAT system would not be in place until 2001 at the earliest, as the Commission envisaged at least a two year period for implementation of the adopted measures, in order to give businesses and tax administrations time to adapt, see COM (96) 328 final, 22 July 1996, n. 30 above, at 34.

32 Council Directive 96/95/EC of 20 December 1996, OJ L 338, 28/12/1996, 89.

33 This was in fact the Parliament’s opinion. According to a research paper published in 1995 by the Parliament, Member States were “likely to face the fiscal consequences of VAT-rate approximation whether this is a result of Community legislation or of market forces. Indeed a general convergence of tax systems is likely as the Single Market develops”, in Options for a definitive VAT system, Working Paper, Economic Affairs Series, E 5, October 1995, at 87. This was also the view expressed by several authors and Member States’ officials following the White Paper strategy, as reported by European Commission staff members, see P. GUIEU and C. BONNET, “Completion of the Internal Market and Indirect Taxation” (1987) Journal of Common Market Studies XXV (3), 209-222, at 215.

34 See Report from the Commission on reduced VAT rates drawn up in accordance with Article 12 (4) of the Sixth Council Directive of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes – Common system of value added tax: uniform basis of assessment, COM (2001) 599 final, 22 October 2001, at paragraph 19.

35 Council Directive 1999/85/EC of 22 October 1999, OJ L277, 28/10/1999, 34.

36 Council Directive 2002/92/EC of 2 December 2002, OJ L 331, 07/12/2002, p.27; Council Directive 2006/18/EC of 14 February 2007, OJ L345, 28/12/2005, 19-20; Council Directive 2004/15/EC of 10 February 2004, OJ L52, 21/02/2004, 61.

37 See Experimental application of a reduced rate of VAT to certain labour-intensive services, Report from the Commission to the Council and to the European Parliament, COM (2003) 309 final, 2 June 2003; and Evaluation report on the experimental application of a reduced rate of VAT to certain labour-intensive services, Commission Staff Working Paper, SEC (2003) 622, 2 June 2003.

38 SEC (2003) 622, 2 June 2003, n. 37 above, at 28.

39 Proposal for a Council Directive amending Directive 2006/112/EC as regards rates of value added tax, COM (2008) 428 final, 7 July 2008, at 2. This is the sixth formal proposal by the Commission exclusively on VAT rates (excluding informal suggestions).

40 Council Directive 2009/47/EC of 5 May 2009, OJ L116, 09/05/2009, 18-20.

41 Proposal for a Council Directive amending Directive 77/388/EEC as regards reduced rates of value added tax, COM (2003) 397 final, 23 July 2003.

42 The Council initiated its formal discussions on the Commission’s proposal at the ECOFIN meeting of 7 October 2003 in Luxembourg (see Preparation of Eurogroup and Council of Economics and Finance Ministers, Luxembourg, 6-7 October 2003, MEMO/03/191, 06/10/2003). Although the Commissioner, Frits Bolkestein, was eager to have the proposal approved before the end of the year, the Council’s initial discussions seemed to indicate that reaching Member States’ agreement regarding this proposal might be difficult and lengthy (see 2530th Council Meeting – Economic and Financial Affairs – Luxembourg, 7 October 2003, C/03/274, Pres/03/2 74, 07/10/2003 and Results of the Council of Economics and Finance Ministers, 25th November 2003 – financial services and taxation, MEMO/03/241, 26/11/2003). During 2005 a substantial push was given to this proposal by both Luxembourg and the United Kingdom – which held the Presidency of the Council during the first and second half of 2005, respectively – leading to its final approval in 2006, see Results of Council of Economic and Finance Ministers, Brussels, 6-7 December 2004, MEMO/04/289, 08/12/2004, and Results of the 2688th ECOFIN Meeting, Press Release 13678/05, Brussels, 8 November 2005, 21.

43 Council Directive 2006/18/EC of 14 February 2006, OJ L51, 22/02/2006, 12.

44 Study on reduced VAT applied to goods and services in the Member States of the European Union, Taxation Papers, Working Paper No. 13, 2008. A summary of the results of the study, as well as an outline of the Commission’s forward strategy was then published in Communication from the Commission to the Council and the European Parliament on VAT rates other than standard VAT rates, COM (2007) 380 final, 5 July 2007.

45 Consultation Paper: Review of existing legislation on VAT reduced rates, TAXUD/D1D/24232, 6 March 2008, at 2.

46 In addition to the legislation already mentioned, the following has also included some amendments to VAT rates provisions: Council Directive 92/111/EEC of 14 December 1992, OJ L 384, 30/12/1992, 47; Council Directive 94/5/EC of 14 February 1994, OJ L 60, 03/03/1994, 16; Council Directive 95/7/EC of 10 April 1995, OJ L 102, 05/05/1995, 18; Council Directive 96/42/EC of 25 June 1996, OJ L170, 09/07/1996, 34; Council Directive 98/80/EC of 12 October 1998, OJ L 281, 17/01/1998, 31; Council Directive 1999/49/EC of 25 May 1999, OJ L 139, 02/06/1999, 27; Council Directive 2000/17/EC of 30 March 2000, OJ L84, 05/04/2000, 24; Council Directive 2001/4/EC of 19 January 2001, OJ L 22, 24/01/2001, 17; Council Directive 2002/38/EC of 7 May 2002, OJ L 128, 15/05/2002, 41; Council Directive 2005/92/EC of 12 December 2005, OJ L51, 22/02/2006, 12-13; Council Directive 2007/75/EC of 20 December 2007, OJ L346, 29/12/2007, 13-14; and finally, with effect from 1 January 2010, Council Directive 2008/8/EC of 12 February 2008, OJ L44, 20/02/2008, 11-22.

47 See COM (96) 328 final, 22 July 1996, n. 30 above, at 5.

48 Council Directive 2006/112/EC of 28 November 2006 on the common system of value added tax, OJ L347, 11/12/2006, 1–118 (hereafter “VAT Directive”).

49 The date until which this minimum standard rate level will apply has been consecutively postponed. This date currently stands at 31 December 2015, see Council Directive 2010/88/EU of 7 December 2010 amending Directive 2006/112/EC on the common system of value added tax, with regard to the duration of the obligation to respect a minimum standard rate, OJ L326, 10/12/2010, 1-2.

50 For a more detailed analysis of the VAT rates applied in the several Member States, the Commission has been publishing, on an annual basis, a document listing the VAT rates applied to a range of products across the EU, VAT Rates Applied in the Member States of the European Community. For an overview of the rates applicable before the economic and financial crisis see the document dates January 2008. See also A. Mathis, VAT indicators, Taxation Papers, Working Paper No. 2, April 2004.

51 The data was reported on the Commission’s report on reduced VAT rates, COM (2001) 599 final, 22 October 2001, n. 34 above. An analysis of the rates in force on 1 January 2009 shows that the situation had not improved and, if anything, it had worsened by then.

52 Articles 123 to 130 of the VAT Directive.

53 Croatia joined the EU in July 2013.

54 See Council Decision 2006/774/EC of 7 November 2006 authorising Member States to apply a reduced rate of VAT to certain labour-intensive services (OJ L314, 15/11/2006, 28-32), and Council Decision 2007/50/EC of 30 January 2007 authorising Romania to apply a reduced rate of VAT to certain labour-intensive services (OJ L222, 31/1/2007, 14-15).

55 See OECD, Consumption Tax Trends 2012 – VAT/GST and Excise Rates, Trends and Administration Issues (Paris: OECD Publishing, 2012), at 15.

56 As A. J. EASSON prophetically pointed out in 1993, “the fiscal implications of EMU are obvious and considerable. Of the three main instruments of economic policy, Member States will soon relinquish all control over exchange rate policy and will gradually, over the next five years or so, lose much of their control over monetary policy. They will thus be left with fiscal policy alone”, see Taxation in the European Community, European Community Law Series (London & Atlantic Highlands, NJ: The Athlone Press, 1993), at 18-19.

57 European Commission, Green Paper on the Future of VAT—Towards a simpler, more robust and efficient VAT system, COM (2010) 695 final, December 1, 2010. See also, Accompanying document to the Green Paper on the Future of VAT—Towards a simpler, more robust and efficient VAT system, Commission Staff Working Document, SEC (2010) 1455 final, December 1, 2010.

58 R. MAAS et al., “VAT Focus—Roundtable discussion: The Green Paper & the Future of VAT” (2011) Tax Journal 1094, 12.

59 See European Commission, Summary Report of the Outcome of the Public Consultation on the Green Paper on the Future of VAT—Towards a simpler, more robust and efficient VAT system (December 1, 2010-May 31, 2011), taxud.c.1 (2011) 1417007, December 2, 2011; European Parliament, Report on the Future of VAT (2011/2082 (INI)), Committee on Economic and Monetary Affairs, A7-0318/2011, September 30, 2011; European Economic and Social Committee, Opinion on “Green Paper on the Future of VAT—Towards a simpler, more robust and efficient VAT system”, OJ C318, 14/07/2011, 87; and European Commission, Future VAT System: pro-business, pro-growth, Press Release IP/11/1508, December 6, 2011.

60 Future of VAT—Towards a simpler, more robust and efficient VAT system tailored to the single market, Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee, COM (2011) 851 final, December 6, 2011.

61 Ibid, at 11-12.

62 European Commission, Review of existing legislation on VAT reduced rates, Consultation Paper, TAXUD/C1, October 2012.

63 European Commission, Review of Existing Legislation on VAT Reduced Rates, Summary Report of the Outcome of the Public Consultation (8 October 2012-4 January 2013), taxud.c.1. (2013) 708070, 12/04/2013.

64 See A Retrospective Evaluation of Elements of the EU VAT System—Final Report, TAXUD/2010/DE/328, FWC No. TAXUD/2010/CC/104, December 1, 2011.

65 See R. DE LA FERIA, n. 5 above.

66 This point is developed further in R. DE LA FERIA, “The 2011 Communication on the Future of VAT: Harnessing the economic crisis for EU VAT reform” (2012) British Tax Review 2, 119-133.

67 It has been argued that VAT is regressive with respect to income but not necessarily with respect to consumption, see S. CNOSSEN, “The Value-added Tax: Key to a Better Tax Mix” (1989) Australian Tax Forum 6 (3), 265-281. For a general discussion on the regressive nature of VAT, or otherwise, see also G.N. CARLSON and M.K. PATRICK, “Addressing the Regressivity of a Value-Added Tax” (1989) National Tax Journal 42 (3), 339-351; and N. WARREN, ‘‘A Review of Studies on the Distributional Impact of Consumption Taxes in OECD Countries’ ’(2008) OECD Social, Employment, and Migration Working Paper 64, June 2008, at 24.

68 See SEC (2003) 622, 2 June 2003, n. 37 above, at 26.

69 See Copenhagen Economics, Study on reduced VAT applied to goods and services in the Member States of the European Union, Taxation Papers, Working Paper No. 13, 2008, at 13.

70 For example, there might be geographical restrains, with only one hairdresser, or repair shop in a specific area.

71 For early studies on the effects of the VAT temporary standard rate reduction in the UK in 2009, with the stated aim of boosting consumption see T. CROSSLEY, H. LOW and M. WAKEFIELDS, “The Economics of a Temporary VAT Cut” (2009) Fiscal Studies 30 (1), 3-16; R. BARREL and M. WEALE, “The Economics of a Reduction in VAT” (2009) Fiscal Studies 30 (1), 17-30; and R. BLUNDELL, “Assessing the Temporary VAT Cut Policy in the UK” (2009) Fiscal Studies 30 (1), 31-38.

72 See A. CHARLET and J. OWENS, “An International Perspective on VAT” (2010) Tax Notes International 59 (12), 943, at 950.

73 See Copenhagen Economics, n. 69 above, at 12.

74 C.L. BALLARD and J.B. SHOVEN, using US data, conclude that differentiation of rates used to partially replace the income tax would do little to mitigate the adverse distributional impact of the change, see “The Value-Added-Tax: The efficiency cost of achieving progressivity by using exemptions” in M.J. BOSKIN (ed.), Modern Development in Public Finance: Essays in Honor of Arnold Harberger (Oxford, B. BACKWELL, 1987), 109-129. Recently I. CRAWFORD, M. KEEN and S. SMITH argue convincingly that the case for differential consumption taxation rests primarily on efficiency considerations, rather than distributional grounds, see “Value-Added Tax and Excises” in S. ADAM, T. BESLEY, R. BLUNDELL, S. BOND, R. CHOTE, M. GAMMIE, P. JOHNSON, G. MYLES and J. POTERBA (eds.), Dimensions of Tax Design – the Mirrlees Review (Oxford: Oxford University Press, 2010), 275-422.

75 See OECD, n. 55 above, at 73.

76 Godbout and St. Gerny show that with tax credits targeted at lowest income individuals in place, the effective tax rate of consumption taxes increases with income, see C. GODBOUT and S. St. GERNY, “Are Consumption Taxes Regressive in Quebec?” (2011) Canadian Tax Journal 59 (3), 463-493.

77 E.H. DAVIS and J.A. KAY provide amusing examples with reference to the United Kingdom to illustrate the shortcomings of using the VAT structure as a mean to diminish its regressivity, see “Extending the VAT base: problems and possibilities” (1985) Fiscal Studies, 6 (1), 1-16, at 11-12.

78 See cases 415/85, Commission of the European Communities v Ireland, [1988] ECR 3097; 416/85, Commission of the European Communities v United Kingdom of Great Britain and Northern Ireland, [1988] ECR 3127; C-276/98, Commission of the European Communities v Republic of Portugal, [2001] ECR I-1699; C-481/98, Commission of the European Communities v French Republic, [2001] ECR I-3369; C-83/99, Commission of the European Communities v Kingdom of Spain, [2001] ECR I-445; C-384/01, Commission of the European Communities v French Republic, [2003] ECR I-4395; C-109/02, Commission of the European Communities v Federal Republic of Germany, [2004] ECR I-12691; C-309/06, Marks & Spencer plc v Commissioners of Customs and Excise, [2008] ECR I-2283; C-462/05, Commission of the European Communities v Portuguese Republic, [2008]; C-492/08, Commission of the European Communities v French Republic, [2010] ECR I-5471; C-3/09, Erotic Center BVBA v Belgische Staat, [2010] ECR I-2361; C-41/09, European Commission v Kingdom of the Netherlands, [2011] ECR I-831; C-49/09, European Commission v Republic of Poland, [2010] ECR I-10619; C-94/09, European Commission v French Republic [2010] ECR I-4261; C-441/09, European Commission v Republic of Austria, [2011] ECR I-72; C-453/09, European Commission v Federal Republic of Germany, [2011] ECR I-74; and C-108/11, European Commission v Republic of Ireland, [2013] ECR I-10947.

79 Case C-259/10, Commissioners for Her Majesty’s Revenue and Customs v The Rank Group plc, [2011] ECR I-000. For an analysis of this case see R. DE LA FERIA “Rank Group. VAT exemption on gambling. Principle of fiscal neutrality. Court of Justice” (2012) Highlights & Insights on European Taxation 1; and R. DE LA FERIA, “Novo rumo para a aplicação do princípio da neutralidade fiscal às isenções em IVA?” (2011) Revista de Finanças Públicas e Direito Fiscal 4 (4), 295-309.

80 See V. SLOANE, “VAT Focus – Rank and Fiscal Neutrality” (2011) Tax Journal 1101, 20, 18 November 2011. See also G. MORSE, “Sub One Ltd v HMRC: applying EU law principles to the zero-rating provisions of UK VAT law; the limits of discretion” (2013) British Tax Review 1, 28-34.

81 See European Commission, Taxation: Commission refers France and Luxembourg to the Court of Justice over reduced VAT rates on ebooks, Press Release IP/13/137, 21/02/2013.

82 United Biscuits (UK) Ltd (No 2) (LON/91/160) VAT Decision 6344.

83 [1997] VAT & Duties Report 85.

84 [1999] Simon Tax Cases 205, and [2000] Simon Tax Cases 16.

85 Case C-309/06, Marks & Spencer plc v Commissioners of Customs and Excise, [2008] ECR I-2283.

86 This argument is further developed in M. DEVEREUX and R. DE LA FERIA, “VAT – Unjust Enrichment” (2008) Tax Journal, 12 May, 13-15.

87 See G. MORSE, “Proctor & Gamble UK v HMRC (Pringles Two) – a very peculiar UK practice – the characterisation of food products for zero-rating” (2009) British Tax Review 1, 59-67; and G. MORSE, “HMRC v Proctor & Gamble UK (Pringles Two) in the Court of Appeal: determining classification cases for VAT – a short practical answer and an end to ‘almost mind-numbing legal analysis’?” (2009) British Tax Review 4, 401-405. See also I. ROXAN “Interpreting exceptional VAT legislation: or, are there principles in Pringles?” (2010) British Tax Review 6, 699-716.

88 See OECD, n. 55 above, at 113. For a useful explanation of the methodologies used for establishing the efficiency of VAT systems see F. Borselli, S. CHIRI and E. ROMAGNANO, “Patterns of Reduced VAT Rates in the European Union” (2012) International VAT Monitor 1, 13-21.

89 See also S. CNOSSEN, “Value-Added Tax and Excises: Commentary” in S. Adam et al (eds.), n. 74 above.

90 See A. CHARLET and J. OWENS, “An International Perspective on VAT” (2010) Tax Notes International 59 (12), 943, at 949.

91 See L. EBRILL, M. KEEN, J-P. BODIN and V. SUMMERS, The Modern VAT (Washington DC: IMF, 2001). See also J. NORREGAARD and T.S. KAHN, “Tax Policy: Recent Trends and Coming Challenges” (2007) IMF Working Papers WP/07/274, at 38.

92 As acknowledged by the OECD, see n. 55 above, at 73.

93 See European Commission, n. 63 above.

94 There is a growing literature on the intersection between tax and state aid law, in particular following the CJEU decision in Gibraltar, case C-106/09, European Commission v Gibraltar, [2011] ECR I-11113. See amongst many others, T. LYONS, “Commission and Spain v Gibraltar and the United Kingdom: a landmark case on engineering tax regimes” (2012) British Tax Review 1, 55-63; and A. Rust and C. MICHEAU, State Aid and Tax Law (The Hague: Kluwer Law International, 2012).

95 Although it has been recently suggested that an imperfect extension of the VAT base to cover more but not all services, might be welfare inferior to the baseline, see B. BYE, B. STROM and T. AVITSLAND, “Welfare effects of VAT reforms: a general equilibrium analysis” (2012) International Tax and Public Finance 19, 368-392.

96 See “Taxing Matters”, Society Now, Spring 2013, 25.

97 See European Commission, The Economic Adjustment Programme for Portugal – Third review-Winter 2011/2012, April 2012.

Auteur

Professeur à l’Université de Durham, UK

© Presses de l’Université Toulouse 1 Capitole, 2014

Conditions d’utilisation : http://www.openedition.org/6540