OPINION OF ADVOCATE GENERAL
WAHL
delivered on 29 May 2013 (1)
Case C‑101/12
Herbert Schaible
v
Land Baden-Württemberg
(Request for a preliminary ruling from the Verwaltungsgericht Stuttgart (Germany))
(Agriculture – Regulation (EC) No 21/2004 – Identification and registration of ovine and caprine animals – Article 16 of the Charter of Fundamental Rights of the European Union – Article 20 of the Charter – Proportionality – Equality)
1. Combating epizootic diseases is a legitimate aim for the European Union (‘EU’) legislature. However, the new system for the individual electronic identification of sheep and goats, introduced by Council Regulation (EC) No 21/2004 of 17 December 2003 establishing a system for the identification and registration of ovine and caprine animals and amending Regulation (EC) No 1782/2003 and Directives 92/102/EEC and 64/432/EEC, (2) is unsuitable for the attainment of that objective, unnecessarily burdensome and costly, as well as discriminatory. As such, that regulation breaches the freedom of animal keepers to conduct business, enshrined in Article 16 of the Charter of Fundamental Rights of the European Union (‘the Charter’), and the principle of equality under Article 20 of the Charter, and is therefore invalid.
2. That, in essence, is the main argument raised by Mr Herbert Schaible (‘Mr Schaible’), a German keeper of sheep with 450 ewes, in proceedings before the Verwaltungsgericht Stuttgart (Stuttgart Administrative Court) (Germany) which he had brought against the Land Baden-Württemberg in order to be released from a number of obligations laid down in Regulation No 21/2004.
3. In this Opinion, I will try to illustrate why it is my view that the legal arguments raised by Mr Schaible are not well founded. Accordingly, I will propose that the Court state in reply to the national court that consideration of the questions referred has not disclosed any factor of such a kind as to preclude the validity of the provisions at issue. Lastly, to the extent that my analysis shows that the practical application of that regulation can potentially constitute an obstacle to the full achievement of its overarching aim and the overall fairness of the system, I will briefly present some considerations de lege ferenda.
I – Legal framework
4. Recitals 1, 3 and 11 in the preamble to Regulation No 21/2004 state:
‘(1) Pursuant to Article 3(1)(c) of Council Directive 90/425/EEC of 26 June 1990 concerning veterinary and zootechnical checks applicable in intra-Community trade in certain live animals and products with a view to the completion of the internal market, animals for intra-Community trade have to be identified in accordance with the requirements of Community rules and be registered in such a way that the original or transit holding, centre or organisation can be traced. …
…
(3) Rules concerning the identification and the registration of ovine and caprine animals in particular have been laid down in Directive 92/102/EEC. In respect of ovine and caprine animals, experience, and in particular the foot-and-mouth disease crisis, has shown that the implementation of Directive 92/102/EEC has not been satisfactory and is in need of improvement. It is therefore necessary to lay down more stringent and specific rules …
…
(11) In Member States with a relatively small ovine or caprine animal population, the introduction of an electronic identification system may well not be justified; it is therefore advisable to allow such Member States to make the system optional. …’
5. Under Article 1(1) of Regulation No 21/2004:
‘1. Each Member State shall establish a system for the identification and registration of ovine and caprine animals in accordance with the provisions of this Regulation.’
6. Pursuant to Article 3(1) of Regulation No 21/2004 the system for the identification and registration of animals must comprise the following elements: ‘(a) means of identification to identify each animal; (b) up-to-date registers kept on each holding; (c) movement documents; (d) a central register or a computer database.’
7. Article 4 of Regulation No 21/2004, so far as is relevant here, reads:
‘1. All animals on a holding born after 9 July 2005 … shall be identified in accordance with paragraph 2 within a period to be determined by the Member State as from the birth of the animal and in any case before the animal leaves the holding on which it was born. …
2. (a) Animals shall be identified by a first means of identification which complies with the requirements of Section A.1 to A.3 of the Annex, and,
(b) by a second means of identification approved by the competent authority and conforming to the technical characteristics listed in Section A.4 of the Annex.’
8. Article 5(1) of Regulation No 21/2004 provides:
‘1. Each keeper of animals, with the exception of the transporter, shall keep an up-to-date register containing at least the information listed in Section B of the Annex.’
9. Article 9(3) of Regulation No 21/2004 states:
‘3. As from 31 December 2009, electronic identification according to the guidelines referred to in paragraph 1, and in accordance with the relevant provisions of Section A of the Annex, shall be obligatory for all animals.
However, Member States in which the total number of ovine and caprine animals is 600 000 or less, may make such electronic identification optional for animals not involved in intra-Community trade.
Member States in which the total number of caprine animals is 160 000 or less may also make such electronic identification optional for caprine animals not involved in intra-Community trade.’
II – Facts, procedure and the questions referred
10. The main proceedings concern an action brought by Mr Schaible, asking the Verwaltungsgericht Stuttgart to declare that he was not subject to the obligations of individual identification, individual electronic identification and the obligation to keep a holding register as prescribed by Regulation No 21/2004.
11. Entertaining doubts as to the validity of several provisions of Regulation No 21/2004, the Verwaltungsgericht Stuttgart decided to stay the proceedings and refer a number of questions for a preliminary ruling to the Court, asking whether:
‘(a) the claimant’s obligation to identify individual animals under Articles 3(1) and 4(2) of Regulation No 21/2004,
(b) the claimant’s obligation of electronic identification of individual animals under the first subparagraph of Article 9(3) of Regulation No 21/2004 …, and
(c) the claimant’s obligation to keep holding register C in accordance with Article 5(1), read in conjunction with point B(2) of the Annex to Regulation No 21/2004,
are compatible with higher-ranking European Union law and therefore valid’ (‘the questions referred’).
12. Written observations have been submitted in the present proceedings by Mr Schaible, the Land Baden-Württemberg, the French, Netherlands and Polish Governments, the Council and the Commission. Mr Schaible, the French Government, the Council and the Commission also presented oral argument at the hearing on 7 March 2013.
III – Analysis
A – The questions referred for a preliminary ruling
13. By its questions, the referring court seeks guidance on the validity of certain provisions of Regulation No 21/2004, by which three specific obligations are placed upon keepers of ovine and caprine animals, namely: (i) to identify individual animals (Articles 3(1) and 4(2)); (ii) to undertake the electronic identification of individual animals (Article 9(3)(1)); and (iii) to keep an up-to-date register of animals (Article 5(1) and point B(2) of the Annex) (collectively, ‘the obligations at issue’).
14. While the Verwaltungsgericht Stuttgart has specifically identified the obligations at issue, both in terms of their content and their legal bases, it has been rather less specific as to the provisions of EU primary law which might preclude such obligations.
15. Indeed, the referring court merely asks the Court whether the obligations at issue are ‘compatible with higher-ranking European Union law and therefore valid’.
16. The general and imprecise wording of the questions would have confronted this Court with a particularly challenging task had not the remainder of the order for reference, as well as the observations submitted by Mr Schaible and other parties, shed some light on that fundamental aspect. (3)
17. As the Verwaltungsgericht Stuttgart explains, Mr Schaible argues that the obligations at issue are incompatible with his freedom to pursue an occupation, as enshrined in Article 15 of the Charter. The national court points out, however, that the Charter provision in relation to which the issues raised in the main proceedings fall to be examined is rather Article 16, concerning the freedom to conduct a business: a freedom which, in the circumstances, would be ‘more specific’. The national court also expresses doubts as to the legality of the obligations at issue in the light of the principle of equal treatment.
18. Notably, in the observations submitted to the Court by Mr Schaible, legal arguments based on Article 16 of the Charter – rather than Article 15 – have been put forward. In the other observations (apart from those made by the Polish Government), the focus is either exclusively on the freedom to conduct a business (the French Government and the Council), or on both freedoms addressed together, on the view that those freedoms are largely identical in substance (the Netherlands Government and the Commission). All the other parties before the Court (including the Polish Government) have added specific arguments concerning the principle of equal treatment laid down in Article 20 of the Charter.
19. Against this background, I consider it appropriate first to clarify the key legal issues arising and, subsequently, to suggest to the Court how, in my opinion, the three questions referred by the Verwaltungsgericht Stuttgart could be dealt with most efficiently.
20. Essentially, those questions raise two issues. The first concerns the possible infringement of Articles 15 and/or 16 of the Charter, because of an allegedly disproportionate interference with animal keepers’ rights, and the second concerns a possible breach of the principle of equal treatment. While the first issue is raised with respect to all the obligations at issue, because – as Mr Schaible himself repeatedly states – they are inextricably linked, the second concerns only one of the obligations at issue, namely the animal keepers’ obligation under Article 9(3) of Regulation No 21/2004 to undertake the electronic identification of individual animals – and, specifically, the derogations allowed in connection with that obligation under the second and third subparagraphs of that provision.
21. In light of the above, and with due regard for the principle of judicial economy, it seems to me more appropriate to deal with the three questions referred as they relate to those two issues, rather than to address the questions separately, each in turn. For this reason, I will first examine the validity of the obligations at issue from a perspective of proportionality. I will then weigh the merits of the arguments relating to the allegedly discriminatory nature of the obligation referred to in the second question from the Verwaltungsgericht Stuttgart.
B – Breach of the principle of proportionality
1. The freedom to conduct a business
22. The first legal issue raised by the questions referred concerns the proportionality of the obligations at issue.
23. As has already been mentioned above, for the national court the parameter of legality in this case is the freedom to conduct business, as recognised by Article 16 of the Charter, whereas some of the parties who have submitted observations to the Court refer also to Article 15 of the Charter.
24. The wording of those provisions suggests that, while Article 16 is focussed on protecting the business freedoms of entrepreneurs, Article 15 is more concerned with the rights of workers. Such a reading is supported by the ‘Explanations relating to the complete text of the Charter’, which state that Article 15(1) of the Charter also draws upon Article 1(2) of the European Social Charter and point 4 of the Community Charter of the Fundamental Social Rights of Workers. (4)
25. That said, it seems to me that these two freedoms largely overlap in that they both relate to the right of Union citizens freely to engage in and conduct an economic activity.
26. In my opinion, the case-law of the Court confirms that overlap. Even before the Treaty of Lisbon made the Charter legally binding throughout the European Union, the Court had recognised both freedoms as general principles of EU law, the observance of which it is under a duty to ensure. (5) In some of its decisions, the Court has explicitly stated that the freedom to pursue an occupation and the freedom to conduct a business may ‘coincide’. (6) Moreover, in several other cases, the Court has more generally referred to the ‘freedom to pursue a trade or profession’ (7) or to the ‘freedom to pursue an economic activity’, (8) which arguably cover aspects common to both freedoms.
27. It therefore seems unnecessary in the present case to delimit the scope of Articles 15 and 16 of the Charter.
28. Since the national court considers that Article 16 of the Charter is more relevant to the case before it, and given that Mr Schaible is an independent operator, I will base my reasoning on that provision of the Charter. Nonetheless, the conclusions which I reach in this regard will be valid, mutatis mutandis, in relation to Article 15 of the Charter (9) because of the common objectives underlying the two provisions.
29. Under Article 16 of the Charter, ‘[t]he freedom to conduct a business in accordance with Union law and national laws and practices is recognised.’
30. EU primary law thus recognises the fundamental value of that freedom but also makes it clear that exercise of that freedom is not an absolute right. The Charter was in fact framed in terms consistent with the established case-law of the Court, according to which principles such as the freedom to pursue an economic activity or the right to property ‘must be viewed in relation to their social function’. (10)
31. That is reflected in Article 52(1) of the Charter, which lays down the rules relating to the limitations that may be placed on the rights and freedoms recognised by the Charter, such as the freedom to conduct a business. Under that provision, such limitations may be imposed as long as they are provided for by law, they respect the essence of those rights and freedoms and, consistently with the principle of proportionality, they are necessary and genuinely meet objectives of general interest recognised by the European Union or the need to protect the rights and freedoms of others. (11)
2. The principle of proportionality
32. According to well-established case-law, the principle of proportionality is a general principle of EU law and requires that measures implemented through its provisions ‘be appropriate for attaining the legitimate objectives pursued by the legislation at issue and [do] not go beyond what is necessary to achieve them’. (12)
33. For the purposes of judicial review as to whether those conditions have been complied with, the Court has accepted that, in the exercise of the powers conferred on it, the EU legislature is accorded a broad discretion in areas where its action involves political, economic and social choices and where it is called upon to undertake complex assessments and evaluations. (13)
34. This is particularly true of the common agricultural policy, an area where the EU legislature enjoys a broad discretion, consonant with the political responsibilities conferred on it by Articles 40 and 43 TFEU. (14) According to the case-law of the Court, the legality of a measure adopted in that area can be affected only if the measure is manifestly inappropriate in terms of the objective which the competent institution is seeking to pursue. (15) The criterion to be applied in that context is thus ‘not whether the measure adopted by the [EU] legislature was the only one or the best one possible but whether it was manifestly inappropriate’. (16)
35. Admittedly, the term ‘manifestly inappropriate’ as used in that case-law of the Court could well attract criticism on two grounds. In the first place, the expression seems to imply that measures whose legal vices are not absolutely obvious, and ascertainable only on the basis of a more in-depth legal assessment, cannot be annulled by the Court. In the second place, it appears to confine review by the Court to the first step of the traditional proportionality test (suitability of the measure to achieve its stated objective), precluding the full analysis.
36. However, that would be to misconstrue the case-law.
37. On the first possible point of criticism, it is my opinion that the use of the term ‘manifestly inappropriate’ is intended merely to emphasise that, where the legislature is obliged to take into account a variety of (potentially conflicting) interests, it must of necessity be accorded a certain measure of discretion. Policy choices inevitably involve the weighing and balancing of different public policy goals which cannot all be pursued at the same time and to the same extent. Only the EU legislature has the mandate to undertake this exercise and also to take decisions which may affect social and economic sectors of society unevenly but which are considered at a given moment to be in the best interests of society as a whole. This is why the Court has explicitly stated that the importance of the objectives pursued by the legislature may justify restrictions which have adverse consequences – even substantially adverse consequences – for certain traders. (17)
38. That limited standard of review does not mean, however, that the Court cannot subject contested measures to rigorous scrutiny in order to verify their proportionality. Indeed, the Court may determine, inter alia, whether the EU legislature based its choice on objective criteria, (18) taking into account all the relevant facts and the technical and scientific data available at the time of adoption of the measure in question, (19) and whether these facts and data are capable of supporting the conclusions reached. (20) The Court must also ensure that the relevant EU institution has fully taken into account all the interests involved beyond the main objective pursued (21) and – in this context – that due attention was paid to the interests of individuals negatively affected. (22)
39. Conversely, the Court cannot substitute its own assessment for that of the EU legislature when reviewing the exercise of legislative powers where policy choices were involved. (23) It is not the constitutional role of the EU judicature to carry out complex assessments in matters where political, economic and social considerations of a general nature play a major role – judicial proceedings being, in any event, manifestly ill-suited for that type of evaluation. (24) Nor is it the role of the Court to carry out an ex post evaluation of the effectiveness or appropriateness of measures adopted by the EU legislature. In this regard, the Court has consistently held that the legality of an EU measure cannot depend on a retrospective assessment of its efficacy. Where the EU legislature is obliged to assess the future effects of rules to be adopted and those effects cannot be accurately foreseen, its assessment is open to criticism only if it appears to have been manifestly incorrect in the light of the information available to the legislature at the time of the adoption of the rules in question. (25)
40. On the other possible point of criticism referred to in point 38 above, I would merely refer the Court to the recent Opinion of Advocate General Kokott in Kokopelli, in which she pointed out that the term ‘manifestly inappropriate’ by no means implied that review by the Court was confined to evaluating the appropriateness of a measure in the light of its objective. I agree that the Court’s jurisdiction to review legality extends to all the stages of the proportionality test, so as to verify that the measure in question is both suitable and necessary for achieving the aim pursued, and stricto sensu proportionate, in the sense that it strikes a fair balance between the interests of those affected. (26)
41. In fact, also in cases concerning measures adopted in the field of the common agricultural policy, the Court has time and again reiterated its case-law to the effect that the principle of proportionality requires that ‘where there is a choice between several appropriate measures, recourse must be had to the least onerous; and the disadvantages caused must not be disproportionate to the aims pursued’. (27)
3. Appraisal
42. It is in the light of those principles that the arguments alleging the invalidity of the obligations at issue must be examined.
43. For a better understanding of those arguments, it will be useful first to sketch a brief overview of the key aspects of the old system for the identification and registration of animals, as well as the system subsequently introduced by Regulation No 21/2004.
44. In the beginning, the rules on the identification and registration of livestock, including sheep and goats, were governed by Council Directive 92/102/EEC. (28) At the heart of the old system was the possibility of identifying animals per holding (‘batch identification’). Under Article 4(1)(b) of that directive, any animal keeper had to keep a register indicating the total number of sheep and goats present on the holding each year. The register was also required to record, inter alia, certain data regarding the movements of sheep and goats entering or leaving the holding. Article 5(3) of the directive required keepers to mark their animals with an eartag or a tattoo, so that it would be immediately obvious from which holding they originated.
45. The system introduced by Regulation No 21/2004 focuses, on the contrary, on the individual identification of each animal by means of two identifiers. Save for certain exceptions, these two identifiers are the (traditional) eartag and an electronic device in the form of an electronic eartag, a ruminal bolus, an injectable transponder or an electronic mark on the pastern, which can be read by specific electronic readers. The identity of each animal must also be recorded in a holding register. Moreover, when animals leave the holding, their movements are to be recorded in a document accompanying the animal. In addition, each Member State is required to set up a central register or computer database recording all holdings located on its territory and, at regular intervals, to take an inventory of the animals kept on those holdings.
46. Against that background, let me now try to illustrate why the arguments put forward by Mr Schaible against the obligations at issue are not persuasive.
(a) Appropriateness of the obligations at issue
47. First, I will examine whether the obligations at issue are appropriate for attaining the objectives pursued by Regulation No 21/2004.
48. According to the first recital in the preamble to that regulation, Regulation No 21/2004 was adopted with a view to completing the internal market in the trade in certain live animals and animal products. The need for a coordinated and efficient system for the identification and registration of animals had arisen when the EU legislature decided, by Directive 90/425/EEC, to abolish zootechnical and veterinary checks at national borders in order to ensure the free circulation of animals within the then European Economic Community. (29) Accordingly, Article 3(1)(c) of Directive 90/425 provided that live animals could be traded only if, for instance, they were identified and registered in such a way that their original and transit holdings could be traced.
49. A regulatory system for the identification and registration of animals is thus a corollary of the single market in animals and animal products. Such a system is of particular value in controlling the spread of infectious diseases among animals which are subject to cross-border trade. In fact, Regulation No 21/2004 was passed in the aftermath of the foot and mouth disease crisis of 2001, as the third recital in the preamble to that regulation explicitly states.
50. Mr Schaible does not deny that the objective underlying Regulation No 21/2004 – to combat the proliferation of epizootic diseases – is legitimate. Nor does he dispute the fact that an effective system for the tracing of animals is necessary in this context. He does, however, contend that the obligations at issue are inappropriate for attaining the objective of controlling epizootic diseases. In particular, he argues that the system is ineffective, in that the percentage of transponders which are lost over time or which become defective is not insignificant. In that context, Mr Schaible refers to a study conducted in Germany by the Bayerische Landesanstalt für Landwirtschaft (Bavarian State Research Centre for Agriculture) which concluded that approximately 5% of transponders were lost, or did not function properly, only shortly after they had been applied to the animals.
51. I cannot agree with that contention.
52. First and foremost, the obligations at issue seem to me to be particularly suited to combating the spread of epizootic diseases and, as such, they make an effective contribution to the completion of the internal market within that sector of the economy.
53. The obligation to identify all sheep and goats individually was chosen as a reliable method, for animal keepers as well as veterinary and administrative authorities, of enabling the origin and location of each animal to be traced. It seems to me important to be able to trace systematically all the places through which an animal has passed and where it could have been in contact with other animals. I have no doubt that such data enable the competent authorities to take the measures necessary to prevent or contain contagious diseases.
54. Moreover, it seems reasonable to me to require an electronic means of identification, as this presumably ensures greater reliability and rapidity in the reporting of data, thus making the fight against contagious diseases more effective. This is especially true in the case of animals like sheep and goats, which usually undergo multiple transactions over the course of their lives and are often moved in large groups, sometimes at livestock markets or assembly centres where the composition of herds can easily change. (30) In those circumstances, it would be very difficult to trace the movements of each animal without any means of individual identification and recording.
55. Lastly, as regards the obligation to keep a register for each holding, I concur with the French Government that the system for the individual electronic identification of animals automatically creates the need for such registers. Clearly, the data recorded by the identifiers must be entered in a document which can be rapidly updated and, upon request, easily accessed by the competent authorities. Thus, in order to ensure that the necessary measures can be adopted without undue delay, it is vital that certain data (such as breed, dates of birth and death, and travel movements) relating to the animals held by each animal keeper be provided to the competent authorities in a standard format and promptly. (31) For example, it is only by matching the information on two or more holdings that the authorities are able to find out whether – and, if so, precisely when and where – the paths of specific animals have crossed.
56. I will now address Mr Schaible’s allegations concerning the technical flaws of the identification system. It is hardly surprising if a radical systemic change, such as the one introduced by Regulation No 21/2004, is far from faultless in its initial stages. A certain percentage of errors and defects are, in my view, inevitable when a new project of that kind is launched, especially where it relies considerably on the use of new technologies. Such malfunctions, however, are not in themselves an indication that a system is entirely unsuitable, unless they clearly undermine the objective pursued.
57. That is not the case here.
58. Indeed, the Commission has commissioned several studies on the feasibility and reliability of systems based on the electronic identification of animals. A first study project (the ‘FEOGA’ project), undertaken between 1993 and 1994 by three teams in different Member States and focusing on, inter alia, a total of 5 000 sheep and 2 000 goats, supported the conclusion that the system was sufficiently developed to be further tested on a large scale and under field conditions. (32) A second project (the ‘AIR 2304’ project) – conducted between 1995 and 1998 – involved 10 research teams in 6 different Member States. The aim was to see whether the percentage of total losses of electronic identifiers would increase over a longer period of time (4 years). The final conclusions of that study showed that the electronic identification systems used (electronic ear tags, injectable transponders, ruminal boluses) achieved higher retention rates and readability values than conventional livestock identification systems (ear-tags, tattoos, and so on). (33)
59. A later project, the IDEA project (‘Identification électronique des animaux’) was concerned with the full-scale implementation of the system, including the application of electronic identifiers under a wide range of conditions (such as different animal species, different breeding methods, different transport or slaughtering methods, as well as varying environmental conditions). Altogether, over 500 000 sheep and almost 30 000 goats were tested between March 1998 and December 2001. (34) That study was intended to assess in a particularly meticulous manner the performance (readability, correct functioning, loss rate, recovery rate, and so on) of electronic identifiers in relation to a number of parameters (type of device, type of animal, age of animal, and so on). Among the conclusions presented in the final report of 30 April 2002, the following are noteworthy: the project had ‘reached very positive result[s] on the feasibility to use electronic identification’; ‘the identification of items using transponder technology [was] well developed, commercially available and applied routinely in many industrial areas’; and ‘the industry [was] … in a position to produce within short time periods large numbers of electronic identifiers and corresponding readers for livestock identification’. (35)
60. Lastly, in the course of its progressive implementation of Regulation No 21/2004, the Commission had, in 2007, assessed the information provided to it by Member States which had implemented electronic identification on a voluntary basis or as pilot projects before the system became obligatory for the entire European Union. According to the Commission, the data resulting from this generally confirmed that the ‘electronic identification of sheep and goats functions under a wide range of production conditions’. (36)
61. On that basis, I am convinced that, when it decided to adopt Regulation No 21/2004, the EU legislature was in a position to base its decision on a vast and detailed body of technical and scientific data supporting the idea that a system based on the individual and electronic identification of animals was both feasible and appropriate to the objective of combating epizootic diseases. Moreover, and this is particularly relevant for the present proceedings, that information indicated that, despite some practical issues, such a system would prove more reliable and effective than the system in place at the time under Directive 92/102. In any case, traditional ear tags are still required and thus, in the event of faulty transponders, the identification of the animals can be performed by way of visual inspections, as it was under the previous system.
62. There is nothing in any of the documents referred to by Mr Schaible, including the study prepared by the Bayerische Landesanstalt für Landwirtschaft, to contradict these findings.
63. In light of the above, I believe that the obligations at issue are indeed suitable for attaining the objectives pursued by Regulation No 21/2004. What is more, they appear particularly appropriate to the fight against contagious diseases among ovine and caprine animals.
(b) Necessity of the obligations at issue
64. As a second step, it is necessary to consider whether the obligations at issue go beyond what is necessary to attain the objectives pursued.
65. One of Mr Schaible’s main criticisms in relation to the obligations at issue concerns the very raison d’être for Regulation No 21/2004: according to Mr Schaible, the old system for identifying sheep and goats had proved adequate for controlling the spread of contagious diseases. The outbreak of foot and mouth disease in 2001, which prompted the adoption of Regulation No 21/2004, came about – according to Mr Schaible – only because the rules in force at that time had not been properly implemented, not because of a fault in the system as such. Consequently, it was unnecessary for the EU legislature to launch a complete overhaul of the system, since the correct implementation and application of the less stringent rules laid down in Directive 92/102 would have been sufficient to ensure the attainment of the stated objectives.
66. I do not find Mr Schaible’s arguments on the adequacy of the old system convincing.
67. As was mentioned in point 49 above, the new system was established because experience – and, in particular, the foot and mouth disease crisis of 2001 – had demonstrated that the rules on the identification and registration of ovine and caprine animals laid down in Directive 92/102 were not satisfactorily implemented and stood in need of improvement. Thus, while acknowledging that the old rules had not been properly implemented, the EU legislature also found that those rules needed to be revised.
68. Foot and mouth disease is a severe and highly contagious viral disease which, although not readily transmittable to humans, may affect several types of animal, including sheep and goats. (37) The economic and social impact of the 2001 epidemic in the United Kingdom need not be stressed. Suffice it to recall, in this context, that more than 2 000 farms in that Member State were affected by the disease, suffering losses estimated at several billion euros. (38) A number of emergency measures had to be adopted, both at national and European level, leading in particular to the mass slaughter of infected herds (and even of animals which were only suspected of being infected), (39) and a variety of export restrictions within the European Union, as well as a worldwide ban on all exports of livestock, meat and animal products from the United Kingdom. (40) Despite those measures, the virus spread to other Member States (France, Ireland and the Netherlands), although only a few cases were reported there altogether.
69. Under those circumstances, unlike Mr Schaible, I find it reasonable that the EU legislature at the time decided to re-consider the system for the identification and registration of animals in its entirety, rather than simply to try to revise the framework whose limitations had become evident. It seems to me not only legitimate but even prudent, after a disaster of such magnitude, that the European Union undertook a project so extensive in scope and nature.
70. Incidentally, a large-scale project on the identification and registration of animals (the IDEA project, referred to above), coordinated by the EU Joint Research Centre (JRC), was already ongoing in 2001. (41) Thus, when the final report on that project was completed on 30 April 2002, it was only wise for the EU legislature to devote a great deal of attention to the results of that study and the suggestions made therein.
71. One of the findings made in the IDEA Final Report was that the various outbreaks of livestock disease within the European Union had revealed that the livestock identification systems used at the time were ‘not efficient and reliable enough to provide a correct traceability and veterinary monitoring of livestock species’. Indeed, animal identification had been performed manually and by inspecting the existing ear tags by eye, an exercise which was considered ‘subject to errors and fraud’. Closer control of the individual animals and their movements was therefore deemed to constitute ‘a critical issue for … sanitary control and disease monitoring’. To this end, it was judged vital for the ‘efficient and reliable management of EU livestock’ for it to be possible to trace the whereabouts of any individual animal at any given time. Accordingly, in order to make the monitoring of individual animals more efficient, the IDEA Final Report suggested the ‘unique identification of each animal during its entire life, by the application of an electronic identifier’. (42)
72. In the aftermath of the 2001 crisis, the IDEA Final Report was by no means the only expert report to suggest a more in-depth revision of the regulatory framework on the identification of animals. For example, the Final Report of the International Conference on Control and Prevention of Foot and Mouth Disease, held in Brussels on 12 and 13 December 2001, recommended that ‘[a]ll movements [of animals] should be registered and traceable’, in order to prevent the spread of the foot and mouth virus between different holdings. (43) In the same vein, the Anderson Report, commissioned by the UK House of Commons in the wake of the 2001 crisis, recommended the development of ‘a comprehensive livestock tracing system using electronic tags to cover cattle, sheep and pigs’. (44)
73. With Regulation No 21/2004, the EU legislature also responded to requests coming from within the EU institutional framework. For example, a Special Report issued by the Court of Auditors, while noting that the old system for the identification and registration of animals had not been adequately implemented, also emphasised the inherent limits of that system. The report states, inter alia: ‘[t]he fact that Community regulations did not require sheep to be identified individually, and that there were also shortcomings in the system of batch identification, hampered the tracing of animals that were suspected of being infected and so delayed their slaughter. …’ (45)
74. Thus, there are various indicia to suggest that the problem with the old system for the identification and registration of animals lay with the system itself, rather than with the manner of its implementation.
75. Taking into account the conditions in which sheep and goats are often traded, (46) and the way in which epizootic diseases are transmitted, (47) I do not find it unreasonable that the EU legislature subsequently decided that animals should be traceable individually. It further stands to reason that the competent authorities need to be in a position to have expedient access to the data which, thanks to electronic identifiers and holding registers, is available and relevant for the adoption of measures necessary to prevent or limit the spreading of contagious diseases.
76. The alternative identification and registration measures referred to by Mr Schaible – essentially, those provided for in Directive 92/102 –are probably less invasive, as regards the animal keepers’ freedom to conduct a business, than the obligations at issue. Nonetheless, as the studies mentioned above unequivocally show, they have a more limited reach and, as a consequence, are far less effective.
77. In the light of this, I conclude that the obligations at issue do not go beyond what is necessary to attain the objective pursued by Regulation No 21/2004.
(c) Proportionality stricto sensu
78. As a final step, the proportionality stricto sensu of the obligations at issue must be assessed. To this end, I will examine whether the legislature has committed any error in weighing the various interests which can be affected by those obligations.
79. In that regard, Mr Schaible’s criticism is essentially twofold. On the one hand, he alleges that the obligations at issue are incompatible with guiding principles of animal welfare, since the application of transponders produces a high number of wounded animals. On the other hand, he argues that those obligations are excessively burdensome and costly for animal keepers, not only because of the additional material that they are required to purchase (transponders, scanners and so on) but also because of the substantial labour input that the use of that material requires (application of the transponder, reproduction of the codes into the holding register, maintenance of the scanners, more frequent recourse to specialised personnel such as IT specialists and veterinarians, and so on).
80. In my opinion, these contentions are also ill founded.
81. Contrary to what is argued by Mr Schaible, the impact of electronic identification on the well-being of sheep and goats was duly considered in the course of the study projects which had been carried out before the enactment of Regulation No 21/2004. The IDEA project, for example, surveyed the main causes of ‘injuries and death after electronic identifier application’ and provided the legislature with useful information on the manner in which the various types of device could affect the health of animals. (48) Moreover, animal welfare considerations were also taken into account at the implementation stage. (49)
82. While it is true that electronic ear tags can sometimes cause wounds on the ears of animals, the same can be said of traditional ear tags, such as those prescribed by Directive 92/102. Admittedly, the data available on electronic ear tags seems to indicate that the problem arises somewhat more frequently in their case because they are heavier. Yet, the difference relied on by Mr Schaible between the two types of device does not seem to be so significant as to call into question the overall evaluation of the system in terms of animal welfare. (50) After all, the benefits of the system in terms of increased security for the whole of society are considerable. In addition, it can be expected that, with future technical developments reducing the size and weight of the transponders, as well as greater experience on the part of farming staff with regard to the application and use of those devices, the situation will inevitably improve.
83. Lastly, the new system – by permitting a more precise identification of those animals which have been in contact during outbreaks – is bound to limit the spread of disease and will accordingly enable the resultant suffering of infected animals to be avoided. Seen from this perspective, the obligations at issue appear to contribute positively to the protection of animal welfare.
84. Moreover, Mr Schaible’s arguments regarding the excessive costs and labour which the obligations at issue are supposed to entail for animal keepers are not convincing to any great extent.
85. As both the Council and the Commission stressed in their observations, those aspects were the subject of detailed studies and were debated at length with Member States and other stakeholders. For example, the JRC published a report in 2009 with the title ‘Economic Analysis of Electronic Identification of Small Ruminants in Member States’. The study analysed all costs relating to the various electronic devices available on the market, both for the purchase of the equipment, and for their daily use (such as tagging and reading operations, or the training of staff). A number of possible ways of implementing the obligation to identify each animal electronically were examined. For each option, cost savings and advantages were compared, with the aim of finding reasonable compromises. (51)
86. The same aspects were also the subject of a survey carried out by the Standing Committee for the Food Chain and Animal Health in 2010, (52) for which Member States provided data on costs.
87. It seems to me, therefore, that the EU legislature had also considered thoroughly all the relevant information on those aspects, including the evaluation of the various options available, and accordingly took a final decision in full knowledge of the various interests at stake. In my view, the EU legislature cannot be criticised for having decided that, in that context, the protection of public health had to take precedence over economic considerations. (53)
88. My conclusion on this point is further supported by three additional considerations.
89. First, beyond pursuing the main aim of combating epizootic diseases, the EU legislature also took into consideration the fact that the obligations at issue could entail a variety of other economic and non-economic benefits to society; for example, with regard to farm management, animal breeding, official controls and food safety. (54) As was stated in the 2009 JRC Report, the new system for animal identification and registration ‘is not only to be seen as [a] cost factor, as its multi-purpose use generates also benefits, not only at farm level but also for official controls and downstream the production chain, e.g. for livestock markets, slaughterhouses and possibly also for the consumers. … [W]hen fully implemented, it could [also] reduce … expenditures for disease eradication measures.’ (55)
90. Second, the EU legislature has, at the same time, adopted a number of measures with a view to alleviating the burden placed on animal keepers by the obligations at issue.
91. To begin with, Regulation No 21/2004 made provision for the new system to be implemented gradually and progressively. Several measures introduced by that regulation were accordingly subject to a transitional period, so as to give animal keepers (as well as Member States’ authorities) sufficient time to adapt to the new requirements and plan in advance the investments needed. Notably, the final date for the obligatory introduction of general electronic identification at EU level, originally set to be 1 January 2008, was later deferred to 31 December 2009. (56) Also, the obligations for animal keepers to keep a holding register (57) and to produce movement documents (58) did not become fully applicable until 9 July 2005.
92. Furthermore, where the impact of the new measures appears disproportionate or the additional risks are deemed insignificant, Regulation No 21/2004 allows derogations. These derogations notably concern all the obligations at issue: the obligation to identify each animal, (59) the obligation to use electronic identifiers, (60) and the obligation to keep a holding register. (61)
93. Third, the EU legislature has further mitigated the additional costs for animal keepers by allowing Member States and regions to grant them, where appropriate, financial aid from EU funds. Accordingly, Council Regulation (EC) No 1257/99 of 17 May 1999 on support for rural development from the European Agricultural Guidance and Guarantee Fund (EAGGF) and amending and repealing certain Regulations (62) was amended shortly before the adoption of the new rules on the identification of animals. (63) In particular, the newly introduced Article 21b(1) of that regulation provides that ‘[t]emporary support intended to contribute partly to costs incurred and income foregone may be granted to farmers who have to apply demanding standards based on Community legislation and newly introduced in national legislation’. That provision applies, inter alia, to the animal keepers affected by the obligations at issue. Importantly, Article 31 of Council Regulation (EC) No 1698/2005 of 20 September 2005 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD), (64) which repealed Regulation No 1257/99 and is currently in force, has preserved the gist of that rule. Furthermore, Article 20(a) of Regulation No 1698/2005 allows a variety of support measures intended to promote the competitiveness of the agricultural sector, including measures for ‘promoting knowledge and improving human potential through … vocational training and information actions, including diffusion of scientific knowledge and innovative practices’.
94. In the light of those considerations, I am of the opinion that the EU legislature duly considered all the interests affected by the obligations at issue, including the interests pertaining to animal welfare and financial interests relevant to individuals, such as animal keepers who are required to bear additional costs in order to comply with the new rules. I do not believe, therefore, that the EU legislature committed any error when weighing those interests against each other and in striking a fair balance between them.
4. Conclusion on the alleged breach of proportionality
95. Concluding on the issue of proportionality, I take the view that the obligations at issue do not breach the principle of proportionality. They are both appropriate to the stated objective and do not go beyond what is necessary to achieve it. Furthermore, there is no indication that the EU legislature committed any error, let alone a manifest error, in considering the advantages and disadvantages of the obligations at issue with regard to all the interests involved.
96. For these reasons, I believe that the obligations at issue are justified and, consequently, that those obligations do not amount to a violation of the freedom of animal keepers to conduct a business.
C – Breach of the principle of equality
97. The other main legal issue raised in the present proceedings is whether the obligations at issue are consistent with the principle of equality.
98. At the outset, it seems useful to recall that the principle of equality – a general principle of EU law now expressly laid down in Article 20 of the Charter – requires that comparable situations must not be treated differently, and different situations must not be treated in the same way, unless such treatment is objectively justified. (65) In this regard, the Court has stated that a difference in treatment is justified when it is based on an objective and reasonable criterion. (66)
99. To the extent that Mr Schaible is requesting this Court to review the discretion exercised by the EU legislature in an area where political and economic choices are involved, the Court’s jurisdiction to review legality is necessarily limited, as illustrated above. The Court has made it clear that in such a case it is not for the Court to substitute its own assessment for that of the EU legislature and, as a consequence, it must confine itself to examining whether the assessment made by the EU legislature contains a manifest error or constitutes a misuse of powers, or indeed whether the EU legislature clearly exceeded the limits of its legislative discretion. (67)
100. It is in the light of those principles that I will examine Mr Schaible’s arguments on the principle of equal treatment. Essentially, two different arguments are put forward. First, Mr Schaible emphasises that an obligation to identify each animal electronically is not imposed upon the keepers of bovine and porcine animals, despite the fact that these animals are also exposed to a large number of contagious diseases. Second, Mr Schaible takes issue with the derogation provided for in Article 9(3) of Regulation No 21/2004 (‘the derogation at issue’) which allows Member States with a reduced ovine or caprine animal population to make the system of electronic identification optional. He argues that this potentially gives animal keepers located in as many as 14 Member States a competitive advantage vis-à-vis those located in other Member States with a larger ovine or caprine animal population.
101. For other reasons, the Polish Government also maintains that Article 9(3) of Regulation No 21/2004 is invalid. According to that government, there is no justification for restricting availability of the derogation at issue solely to animals ‘not involved in intra-Community trade’.
102. I cannot agree with any of those arguments.
103. As concerns the first argument put forward by Mr Schaible, I do not believe that any meaningful inference can be drawn from a comparison with the rules applying to bovine and porcine animals.
104. In my opinion, it can hardly be contested that the breeding, transport and commercialisation of those different types of animal occur under conditions which are not assimilable in every respect: a number of particularities exist with regard to each of those types of animal. (68) Moreover, those animals are usually prone to different kinds of disease and the risk profiles established for the different types of animal may differ. (69)
105. Those differences justify a different regulatory framework for each type of animal. (70) The fact that the EU legislature based its choice on an objective and reasonable criterion cannot therefore be called into question.
106. In any event, as the Council has correctly pointed out, the Court has consistently held that, when the EU legislature is called upon to restructure and establish a complex system, it is entitled to have recourse to a step-by-step approach (71) and to proceed in the light of the experience gained. (72)
107. That case-law seems to me to be of great relevance to the case under consideration. That does not mean, however, that a certain degree of consistency and uniformity is not required, on the part of the EU legislature, when, over a certain period of time, it adopts measures intended to regulate situations which have certain features in common, as could be the case with the identification of different farm animals.
108. Lastly, I am also not convinced that the animal keepers located in Member States with a larger ovine and caprine animal population would suffer any discrimination. In the light of Mr Schaible’s arguments, it must be determined whether the derogation at issue creates a disadvantage for some farm operators in relation to others, without that difference in treatment being justified by the existence of substantial objective differences. (73)
109. It is clear that the derogation at issue entails a different legal treatment for animal keepers, depending on the Member State where they are located. Yet I do not think that this amounts to an impermissible discrimination. That contention is based on two grounds: the criterion triggering the derogation at issue cannot be criticised as lacking objectivity or reasonableness; nor does the derogation place some animal keepers at a significant competitive disadvantage vis-à-vis some of their competitors.
110. On the first point, it cannot be denied that the criterion which makes the derogation at issue available – the overall animal population of a country – is entirely objective. Member States may or may not fulfil that criterion, irrespective of their absolute size. Moreover, the situation of some Member States may well change over time since their animal population is necessarily subject to variations. Thus, Member States which in principle fall within the scope of the derogation at issue at any given moment may still be required to introduce electronic identification in the future if the relevant circumstances change.
111. Nor does the criterion appear to be unreasonable. Even if one may regret that Regulation No 21/2004 does not explain why the EU legislature considered that the introduction of an electronic identification system ‘may well not be justified’ in Member States with a relatively small population of ovine and caprine animals, (74) the clarifications provided in the course of these proceedings by the French and Polish Governments, and the Council seem to me persuasive.
112. As Mr Schaible observed, the creation of a fully operational electronic identification system requires considerable investment on the part of both animal keepers and national administrations. In Member States where the animal population is low and individual animal keepers mostly keep a small number of animals, such an investment may be excessive. This is particularly true when considered in the light of a cost-benefit analysis. In the case of an outbreak in such a Member State, the cost of eradicating the disease (through, for example, mass slaughter or vaccination) and, where applicable, the cost of compensation to be paid to operators who have been adversely affected would, in all likelihood, be lower than the investment required to set up the system. In such a case, it might be more reasonable to maintain the status quo, or at least to endorse a more balanced and progressive introduction of the new rules.
113. It is also clear that, where holdings are for the most part small, with a limited number of animals, it is possible, despite the lack of an electronic identification system, for data to be collected relatively swiftly and reliably in response to urgent requests from the competent authorities. The fact that the code of each animal has to be observed visually and then manually noted on the relevant documents is not likely to result in any undue delays. That would not be the position, however, if those actions had to be performed by a considerable number of holdings with a large animal population.
114. In these circumstances, I do not find it unreasonable that the eligible Member States were left free to choose whether it was more appropriate to introduce the new system for the electronic identification of sheep and goats or to avail themselves of the derogation at issue.
115. Obviously, that line of reasoning applies only in so far as the exoneration of some Member States does not lead to heightened risk within the internal market. However, in my view, that is not the case here. In fact, the derogation at issue allows Member States to make electronic identification optional only with respect to animals which are ‘not involved in intra-Community trade’. Accordingly, any animal that crosses national borders and may thus come into contact with animals located in other Member States necessarily requires electronic identification, irrespective of whether or not its Member State of origin decided to make use of the derogation at issue.
116. In that way, any increased risk of contagion is normally confined within the borders of the Member States which decided to make use of the derogation at issue. Those Member States accept that they (and the animal keepers located on their territory) may have to bear more significant economic consequences in the event of an outbreak.
117. The criterion triggering the derogation at issue also ensures – and here I come to the second aspect referred to in point 112 above – that competition between animal keepers located in different Member States is not distorted to any appreciable extent. In fact, all animal keepers wishing to sell their sheep and goats beyond the territory of their home country will have to comply with the same rules. This means that there is no distortion of competition among keepers active in intra-Community trade. Moreover, a keeper located in a Member State which has taken up the derogation at issue cannot gain leverage through his potential cost advantage in any other Member State.
118. The only scenario where I could actually see a competitive disadvantage is in the case of animal keepers located in Member States without a derogation, who export livestock to Member States which apply the derogation at issue. It is true that, in such a case, local animal keepers could have a cost advantage over their competitors in the exporting State. Questioned on that point at the hearing, neither the Council nor the Commission contested this fact. That said, given the smaller herds which are usually to be found in holdings located in Member States applying the derogation at issue, and the fact that exporters tend to be of a larger size and therefore benefit from economies of scale, any distortions within the local markets are likely to be negligible. In any event, this limited difference of treatment seems to me justified for the objective reasons set out above.
119. Lastly, I turn to the argument, put forward by the Polish Government, concerning the scope of Article 9(3) of Regulation No 21/2004. Suffice it to say that the derogation at issue is crucial to ensuring both that the overall risk for public health in the European Union is not unduly increased, and that competition within the internal market is not distorted to any significant extent. Consequently, the criticism by the Polish Government on this point is manifestly unfounded.
120. In conclusion, I am of the opinion that the derogation at issue applies in situations which are objectively different from those falling under the general rule. Consequently, the arguments alleging breach of the principle of equal treatment should be rejected.
D – Final remarks de lege ferenda
121. Although my legal analysis has not revealed any grounds to support a finding that the obligations at issue are unlawful, it has nevertheless demonstrated that the practical application of the derogation at issue can constitute an obstacle to the full achievement of the overarching aim pursued by the regulation, as well as to ensuring the overall fairness of the system.
122. As regards the achievement of the overarching aim pursued by the regulation, I must again stress that Regulation No 21/2004 was adopted in order to combat epizootic diseases in a more effective manner, so as to facilitate the completion of the internal market in that sector of the economy. Yet, as a consequence of the derogation at issue, the free circulation of some animals within the European Union is prevented and also the effectiveness of the new system for the identification and registration of animals is limited within the Member States which apply the derogation at issue. Moreover, certain spill-over effects for the rest of the European Union cannot be excluded.
123. As Mr Schaible correctly points out, in Member States with a low animal population, the risk that an epizootic disease may spread uncontrollably is not necessarily lower than in Member States with a large animal population. Since diseases are mainly transmitted by animal contact, it would seem to me that the chances of serious and rapid contagion depend on a variety of factors, such as the number of transactions and movements; whether the animals are sold in small or large lots; and whether the animal lots are mixed or remain essentially unaltered. More importantly, given the varying geographic configuration of the EU Member States, if anything, it is the density of animals within one country which matters, rather than the absolute number of animals there.
124. Even though the reason for making the derogation at issue conditional upon involvement in intra-Community trade is to ensure that the increased risk stemming from the circulation of animals which are not identified electronically remains confined to their home markets, it cannot be excluded that, in exceptional circumstances, this may not be enough to establish a watertight system. Indeed, the fact that different kinds of data are available for animals identified electronically as compared with animals identified by traditional ear tags might, in the case of an outbreak, delay or hamper the tracking and matching efforts undertaken by the authorities in order to detect all risky cross-border movements by animals.
125. A certain margin of risk, moreover, will always remain, where – owing to fraud or merely to error – some animals might be traded across a border even though they are not electronically identified. After all, sheep and goats are often sold in large flocks and it may not always be easy to guarantee that all the animals in a flock are fully compliant in this regard.
126. In relation to the overall fairness of the identification system, I actually have to admit that the electronic identification system provided for in Regulation No 21/2004 does not represent a perfect solution. As I have already observed, through the derogation at issue, some animal keepers may benefit, within their home markets, from a cost advantage as compared with competitors from other Member States. Furthermore, it is conceivable that the derogation at issue could indirectly bring about additional economic repercussions for animal keepers who have to use electronic identification. Indeed, even if these animal keepers do not export livestock to Member States applying the derogation at issue, it is possible that further down the production chain the products deriving from their animals (such as meat, wool or milk) may, at some point, enter into competition with those produced in the States where the derogation at issue applies. It is safe to assume that the higher cost to be borne by the animal keepers of larger Member States may, to a certain extent, affect the price of the end products placed on the market and thus make those products somewhat less competitive.
127. The distinction arising as a result of the derogation at issue may well be justified in the current situation, but that may be less true in the future. In fact, as many of the parties to the present proceedings have observed, the technology required by Regulation No 21/2004 is becoming more and more economical, owing to rapid technical development and economies of scale. The ratio underlying the derogation at issue may therefore become less and less obvious.
128. Here it is important to note that Regulation No 21/2004 in fact contains a provision which allows the Commission to modify the demographic thresholds for the derogation at issue. (75) However, in the light of the foregoing, I am inclined to believe that the EU legislature will eventually have to reflect on the possibility of doing away altogether with the derogation at issue. This would, in my opinion, further advance the fight against epizootic diseases, improve the free circulation of sheep and goats within the European Union and ensure a level playing field for all animal keepers in the European Union, whatever the circumstances.
IV – Conclusion
129. In conclusion, I propose that the Court answer the question referred for a preliminary ruling by the Verwaltungsgericht Stuttgart (Germany) as follows:
Consideration of the question referred for a preliminary ruling has disclosed nothing to affect the validity of
– the animal keepers’ obligation under Articles 3(1) and 4(2) of Council Regulation (EC) No 21/2004 of 17 December 2003 establishing a system for the identification and registration of ovine and caprine animals and amending Regulation (EC) No 1782/2003 and Directives 92/102/EEC and 64/432/EEC to identify individual animals,
– the animal keepers’ obligation under Article 9(3)(1) of Regulation No 21/2004 to identify individual animals electronically, or
– the animal keepers’ obligation under Article 5(1) of Regulation No 21/2004, read in conjunction with point B(2) of the Annex thereto, to keep holding register C.