The Workload of the Court of Justice of the European Union - European Union Committee Contents


Court of Justice

Evidence was taken from Sir Konrad Schiemann, UK judge at the Court of Justice, Señor Alfredo Calot-Escobar, Registrar of the Court, and Mr Timothy Millett, Deputy Registrar of the Court.

The witnesses were asked what changes could be made to the rules of procedure of the Court without treaty change, and what changes they would advocate if they had complete procedural autonomy. Sir Konrad said that all the Court's procedures took a long time to complete. He could not speak "for the Court" as the Court itself had no policies as such. The Court had no control over its case-load and had to deal with whatever cases came in at whatever time.

Sir Konrad indicated that the Court could reject a case by ordinance. For instance, the Court could say to a submitting court that their preliminary reference was not a matter of EU law, and therefore could not be accepted. However, this would have to be done gently. In the past year, around 40-50 references had been rejected in this way.

Occasionally, the question put to the Court was answered in the meantime in another case. In such circumstances the CJ would write to the referring court asking them whether they wished to pursue their reference. But the Court itself had no power to withdraw a reference unilaterally.

Sir Konrad described the typical progress of a case through the Court of Justice:

A national court makes a request for a preliminary ruling. That court sets out what the problem is, the relevant national provisions and what concerns the court. The Court of Justice has published suggested guidelines for how national courts should submit this request, but there is not a specific template. The guidance asks the national court to find the relevant facts in the case in order to allow the Court of Justice to judge on as narrow an issue as possible. For instance, in the case of English football television rights, the English court asked for a ruling because the case appeared to demonstrate a conflict between previous CJ rulings—the Court was therefore asked to rule on the correct interpretation of the law. Usually, the Court would reformulate the question in such terms as "By its question the national court asks ..."

In regard to whether national judges should give their own opinion on the case, Sir Konrad indicated that this already happened in a number of cases, and was encouraged by the Court's guidance information. This was always encouraged in the English Court of Appeal, and was useful as the national judge was familiar with the practical application of the particular law in that country. However, any ruling by the Court of Justice would bind all 27 Member States, and might therefore have different effects in different States. For this reason, all preliminary references by national courts which were not dismissed by way of ordinance were translated into all official languages and circulated to all Member States and the EU institutions for comments. For the same reason all judgments were translated into all official languages of the Member States.

Señor Calot-Escobar described the procedure for translation both in preliminary reference cases and in the other types of cases—direct actions in which the Commission sued a Member State for failing to fulfil its treaty obligations or in which a challenge was made to the legality of some piece of Union legislation and appeals from the General Court—which come before the Court. In almost all cases, the translations had to be produced within imperative time-limits, for example 20 working days for references for preliminary rulings. In 2009, the Court translated around 800,000 pages of legal documents, and had to identify the correct legal terminology in each language to ensure an accurate translation. The Court needed the resources in order to meet this workload and to respect the translation deadlines. Sir Konrad observed that occasionally the translations were difficult to follow, but that it was difficult to tell whether this was the fault of the translation or of the draft by the submitting national judge. However, mistranslations were normally discovered reasonably quickly, typically by the judge working on the case cross-checking with someone in one of the other cabinets in the Court who spoke the relevant language.

After translation, in all preliminary ruling cases the documents were sent to the Member States, and to the Commission, the Council and the Parliament who then had two months to formulate their observations. In direct actions and appeals the documents were translated into French (for the benefit of the Court). Appendices to the parties' pleadings were only translated if this was requested by the reporting judge.

Sometimes people did not respond. This might happen if a party to a case suspected they would lose but wanted to delay judgment as long as possible.

Cases were sometimes more complex than necessary because the parties were asked by the national judge to draft the submission, with the judge sending it on unamended. Although in theory it was for the national judge to decide whether or not to make a reference, in practice in England a judge would be inclined to make a reference if both parties accepted that the case needed to be referred.

Normally, a response would be received from at least the Member State in which the case had arisen.

Sir Konrad explained how the judges to hear the case were chosen:

Firstly, the President of the Court would appoint a reporting judge (juge rapporteur) to the case. The President had to balance considerations of choosing the judge with the most relevant experience against fair distribution of appointments among the judges. The reporting judge made a summary of the case and the arguments of the parties, and would also propose how many judges should take part in the proceedings and whether there should be an opinion by the Advocate General. There was a choice of 3, 5, 13 or all 27 judges, although all 27 had sat in only two recent cases. The decision as to how many judges should hear each case and whether there should be an opinion was always taken by all the members of the Court.

The numbers from 2005-2009 were as follows:

3 judges—35%

5 judges—57%

13 judges—8%

President alone—1%

Most judges were chosen for cases by rota.

With regard to oral hearings, all Member States were entitled as of right to request a hearing and often one or more did. Individuals could submit a reasoned request for a hearing, and such requests were generally granted. Sir Konrad suggested that it would be helpful if the Court could alter its procedure in order to refuse such requests occasionally.

There were around 600 cases per year, of which around 180 featured hearings. To people familiar with the English legal tradition, these hearings appeared unsatisfactory. Some of this was due to the need for simultaneous translation of the proceedings, and some due to the variable quality of the counsel. Sir Konrad noted that British counsel were usually the most effective in oral hearings, and the most expensive.

The hearings were normally restricted to 20 minutes for each party, although this could be extended on request. The members of the court could intervene to ask counsel to elaborate on particular points.

Sir Konrad suggested that, ideally, more of the proceedings could take place in writing, but that this would delay the process even further. Currently in preliminary reference cases the written element amounted to one round of simultaneously submitted material: any further elaboration or rebuttal had to be done orally.

The oral procedure had developed in recent years in part perhaps under the influence of British legal tradition—judges took more part in proceedings than before, rather than adopting the French practice of impassivity.

Sir Konrad suggested that there was little tension between common law and Napoleonic approaches as civil law matters rarely arose. Tension mainly arose in cases of choice of forum or enforcement of judgments, where a common law judge would normally expect to have greater discretion, whereas the Court's case law placed more emphasis on the desirability of clear rules inflexibly applied.

Sir Konrad suspected differences of view among members of the CJEU with regard to allowing preliminary references to be taken by the General Court. He had no firm view himself, but argued that there was a difficulty in determining which cases would be suitable to be dealt with at that level. Some people had suggested that technical VAT measures, for instance, might be taken by the General Court, but it was difficult to predict in advance which cases might raise fundamental questions of EU law. He also suggested that any ruling from the General Court would be subject to appeal to the Court of Justice, which would not alleviate either Court's workload. In any case, the General Court was already overloaded, so it would be difficult to add to its remit without taking other areas away from it.

Señor Calot-Escobar argued that translation was not causing a bottleneck. It was an obligation to translate into the official languages and this was not currently a problem, but could become one if the workload increased. He argued that the Court needed additional resources, and was struggling to receive them. An attempt by the Court to employ one extra English lawyer-linguist (bringing the total from 30 to 31) had met with resistance from the Council. Additionally, he confirmed that recruitment of British lawyer-linguists was difficult because the pay was not competitive for such highly-qualified people. Sir Konrad added that everybody in a particular grade had to be paid the same, regardless of their home Member State. Normally this had been done by adopting the most generous rate, but this had still not resolved the problem for English speaking lawyer-linguists who were relatively few and in world-wide demand.

With regard to unusual languages, when no translator from an unusual language (A) to another unusual language (B) could be found and therefore documents in (A) needed first to be translated into a more common language, the 20 day deadline still applied.

Sir Konrad suggested it would be possible to abandon translation into all 23 languages, but this would be extremely sensitive politically. The essence of EU law was that it was also part of national law; it was therefore understandable that people wanted it translated into their own national languages. He suggested that it might be easier if the working language of the Court were English, but that, politically, the one Member State who could not make this suggestion was the UK.

Regarding the accusation of judicial activism and the expansion of the Court's jurisdiction, Sir Konrad noted that the Treaty calls for an "ever closer Union", and that the Court took seriously its obligation to follow the Treaty. The Court did not see itself as having a mission independent of what the Treaties prescribed.

With regard to separate tribunals for intellectual property or trade marks, Sir Konrad suggested it might work if properly financed. Another possibility was the appointment of additional specialist judges as part of the General Court. From the point of view of the overall structure of the CJEU it would need to be determined whether such courts should be truly separate, as was the Civil Service Tribunal, or part of the General Court. There would also be the issue of which court would deal with appeals from the tribunal.

Sir Konrad hoped that, in the light of the new Lisbon provisions, cross-border criminal proceedings could be dealt with more quickly and effectively. He noted that there was a tension between EU and national law in these matters. Overarching EU principle governed the area, in order to ensure that the law did not discriminate and was effective, but individual crimes, and the appropriate procedure to deal with them, were largely left to Member States. There were problems with the adherence of some Member State judgments to EU law, and it was necessary for the Court to provide a consistent interpretation of the provisions of the Treaty. For instance, certain terms, such as "detention in custody", needed to mean the same thing in all Member States.

Sir Konrad explained that many of the problems for the Court arose upstream. The Court had to interpret legislation which had been designed by politicians whose political priority was the achievement of a formula, if necessary at the expense of a clear formula. Where the original legislation was imprecise, the Court was required to intervene. This was often the case with Directives, but could also be seen in the Treaties themselves, particularly in the Treaty of Lisbon's provisions for accession to the ECHR.

Sir Konrad explained that the effect of EU accession to the ECHR would depend on the terms of the accession. Before an action could be brought in Strasbourg, all remedies in a particular nation had to be exhausted. It would need to be determined what equivalent principle would apply if the EU acceded. Sir Konrad argued that it was important that EU bodies had a chance to rule on a matter before it was taken to Strasbourg.

General Court

Evidence was taken from Nicholas Forwood, UK judge at the General Court, Marc Jaeger, President of the General Court, and Arnaud Bohler, Head of the President's Cabinet.

Mr Forwood described how in the early years of the EEC cases brought directly before the ECJ related to competition, trade, and staff matters. The Court of First Instance was established in 1989 to deal with these direct action cases. These cases were fact-intensive. Over time, there was an increase in the areas giving rise to direct action: for instance, trade mark law was originally harmonised by the 1989 Directive, and when national courts faced it, they referred questions to the ECJ. But when the Community trade mark was created in 1994, many cases could be brought directly before the General Court. He speculated that the creation of new agencies, such as the European Chemicals Agency, would give rise to an increase in cases at the General Court.

The General Court was a judicial review court, looking at factual and legal elements of decisions taken at a different level.

Mr Jaeger suggested that, while the Court had no formal role in looking at draft legislation, it might be useful to see impact assessments of the effect of proposed laws on the courts. He pointed to the REACH regulation, where it was originally predicted that 250,000 applications would arise. The most recent estimate was that 2 million applications would be made. A proportion of these would be challenged at the General Court, increasing the workload. Mr Jaeger argued that it should be easier for legislators to assess the potential for litigation provided by their proposals than for the Court.

With regard to the recent decision by the Commission regarding anti-competitive behaviour by European air freight carriers, Mr Jaeger explained that, should the Commission's decision be appealed, this would be dealt with by the General Court. Such competition cases had traditionally been dealt with by the General Court, along with other areas of broad economic interest such as State aid. Because they often had elements which might spill over into wider areas of EU law, it would be difficult to establish a separate competition court.

Mr Forwood explained the procedure of the Court:

Pleadings were exchanged in sequence, rather than simultaneously, as in preliminary ruling cases in the Court of Justice. There were two rounds of pleadings. There was also more scope for interventions from interested parties, which in turn gave rise to procedural issues such as confidentiality.

There were now typically more than 100 distinct procedural documents registered in an average GC case (competition or State aid), i.e. an average of 33 steps per year for a three-year case: this figure had increased in recent years. Case management had become increasingly difficult as there was so much documentation from interested parties. A total of 33,275 procedural documents had been registered in 2010, relating to some 1200 pending cases.

After the written pleadings closed the reporting judge produced a report for the hearing, which was a synthesis of the principal documents received, and in parallel produced a report for his colleagues on the merits and handling of the case. It was at this point that the Court considered whether to hold a hearing and how best to organise it. However, there had to be a hearing in all substantive cases except trade marks (where the judge and parties could decide to dispense with it) or appeals from the CST.

Mr Forwood explained how the Court's judgments involved not only an assessment of the facts and substantive merits of the case, but also the need to produce a judgment consistent with an ever increasing body of EU case law.

In contrast to the CJ, cases were referred to a particular chamber immediately. In the CJ, where the Court had to give a definitive answer on a point of EU law, it was more important for all 27 judges to follow the case until a relatively late stage. In the General Court, cases involved too many complicated issues of merit and fact for all 27 judges to be expected to follow a case in the same way as the CJ.

Mr Forwood explained why the use of specialised chambers had been abandoned. The system had been first introduced in 1999 to deal with trade mark cases. This was, at the time, a new area of EU law with a large volume of work: in order to achieve consistency of judgment in the evolving case law, all trade mark cases were remitted to two of the then five chambers of the Court. However, by 2003 the law in this area had become more settled, and it had also become apparent that with specialisation the Court risked losing its character as a general court combining the expertise and traditions of then 15 (now 27) Member States—those judges who sat in the specialised chambers would have reduced possibilities to take part in other cases, and those from other chambers would have been unable to contribute to trade mark cases.

Mr Forwood explained that the compositions of the chambers were established at the beginning of each three year mandate of the GC President. The President proposed a composition of the chambers with the aim of spreading expertise among them. The three members of each chamber would normally remain in that chamber for the next three years.

Mr Forwood suggested that the solution to reducing the problems with the GC's workload was the appointment of more judges. This could be accomplished in two ways: (1) an increase in the number of judges in the GC, or (2) the establishment of a first instance court at a tier below the GC, rather like the CST. The Treaty allowed for such courts to be established, and they might be appropriate for dealing with trade mark cases.

With regard to the language regime, Mr Forwood explained that not everything was translated into all 23 languages. Unless a Member State intervened in a case, the case took place in a single language of the applicant's choosing. If this language was not French then the case would be translated into French, the Court's working language. The judgment was then produced in French and translated into the language of the case—but only if the chamber decided it raised issues of general importance would it be subsequently translated into all 23 languages. This happened in around 60-70 % of cases.

Mr Jaeger explained that the pleadings contributing to the report for the hearing were voluminous. In the Cement case, there were some 40 applicants and the judgment was extremely long[199] Mr Jaeger explained that there was sometimes a delay in producing translations quickly, due to the length of the documents involved. However, he argued that this was of less importance in the GC than in the CJ. GC judgments had a more specific effect than those of the CJ, which dealt with fundamental aspects of EU law. It was therefore more important for CJ judgments to be produced in all languages at the same time.

Mr Forwood acknowledged that cases took longer as a result of translation, but stressed that it was a necessary part of the process. He argued that it was less of a delaying factor than many people supposed as translation took up on average only around 3 months (including both translation of the written pleadings and of the judgment) of a case lasting 32 months.

The witnesses stated that while appeals were limited to points of law there was no ability for the CJ to refuse leave to appeal. Mr Jaeger reported that around 25-30% of General Court cases went to appeal, with around 5% overturned by the Court of Justice (so only less than 5% of appealable decisions were totally or partially annulled). In 2009, in 85% of all appeals, the decisions of the General Court were fully upheld by the Court of Justice.[200]

With regard to trade mark cases a much lower proportion (ca. 17%) went to appeal. The proportion of appeals dismissed summarily had also increased as trade mark law had settled down.

With regard to increasing the number of judges, Mr Forwood explained that the Statute of the Court would have to be changed by the Ordinary Legislative Procedure as Article 48 set down the number of judges. The Court has not yet made such a request, but in the 1990s had requested more judges from the Council. The Council agreed in principle but could not come to an understanding on how to allocate the six extra judges. In contrast, the mechanism for appointing judges to specialised courts (such as the CST) involved more flexible arrangements and was less likely to cause a repetition of this problem.

Mr Forwood acknowledged that in the longer term, some preliminary ruling jurisdiction might be given to the GC—for instance in the area of trade mark law or competition. However, even in such areas there was a risk of overlap with other areas of law, as would be the case with taxation law impacting on freedom of movement.

It was noted that the last time the GC reduced its workload, by sending cases to the CST, it merely inherited other new cases from the CJ (i.e. those brought by Member States), so was not in a better position as a result.

Mr Jaeger emphasised that the Court needed external help in order to reform its procedures e.g. e-Curia. It would be difficult for the Court to increase its speed without sacrificing quality, which would merely result in an increase in appeals to the CJ.

Civil Service Tribunal

Evidence was taken from Mr Paul Mahoney, President of the Civil Service Tribunal.

Mr Mahoney provided the Committee with a paper based on a talk he had recently given regarding the use of specialised tribunals.

He reported that most people thought that the CST had been a success, and that the advantages of a small tribunal had outweighed the disadvantages. He explained that there was, though, no consensus within the CJEU as to whether a third tier of tribunals was necessary, or as to whether such tribunals allowed proceedings to progress more quickly. Broadly, there appeared to be more enthusiasm for such tribunals within the General Court than within the Court of Justice.

In the CST, judges had one référendaire, as opposed to three in the Court of Justice, but each judge managed to deal with a similar workload.

Mr Mahoney said that a new specialised tribunal could be established along the lines of the CST. The CST had seven judges, which was enough for its purposes. He speculated that, for instance, a specialised competition court would not need to have a full set of 27 judges.

Appeals against the CST, on points of law only, went to the General Court. Were other courts to be established, the General Court might take appeals on points of fact as well. However, there was a risk that this might result in two strands of case-law, with one strand going through the specialised court and the General Court, the other going via preliminary references to the Court of Justice. For this reason consideration would need to be given to appeals going to the Court of Justice rather than the General Court.

A major difference between the CST and other proposed specialised courts was that the cases of the CST were "cuisine interne", i.e. they related only to the functioning of the European institutions, and had no wider ramifications.

Mr Mahoney reported that, when the Tribunal was first established, there were a number of appeals against its judgments, testing the vulnerability of the Tribunal. A similar phenomenon had occurred when the General Court (then, the Court of First Instance) had been set up, but in both cases this had trailed off in time.

He reported that sometimes the GC overturned the verdict of the CST for one reason or another, sometimes it upheld the judgment, but changed the reasoning.

Mr Mahoney described how the practices of the Tribunal had originally been challenged on the grounds that it was being too "activist". The style of cross-examination in the Tribunal was more thorough than people had been used to in the other CJEU courts.

The Tribunal did not look at the merits of disputes (for example, whether somebody should have been promoted) but only the law, except in cases where there had been a manifest error on the merits.

Mr Mahoney explained that only individuals covered by staff regulations could bring a case to the CST. Those staff members could not represent themselves, and had to be represented by a lawyer. Unions had to go to the General Court with cases of more general character; seconded national civil servants could not take cases to the Tribunal, as they were not covered by the staff regulations.

Mr Mahoney described the normal order of proceedings in a case:

1. A staff member makes an application to the Tribunal.

2. The staff member chooses the language in which the case is to be heard; the institution responds to the application in that language.

3. The Tribunal makes a decision on admissibility.

4. The defendant has two months to reply in writing to the application. There is one round of written pleadings as, by that point, the parties will have exhausted all administrative remedies.

5. After a meeting of the relevant chamber, the judge rapporteur gives an opinion.

6. If the case is not thrown out at this point, there is a hearing. Each party has 20 minutes, then there is an hour of questions, followed by 10 minutes of summing up.

7. The chamber of judges then discusses the case immediately after the hearing. This is in contrast to the Court of Justice, where such discussions take place after the Advocate General has presented an opinion, often months later.

8. A text of the final judgment is produced and translated.

Because the judgments of the Tribunal had no wider effect in Member States, they did not have to be translated into all 23 languages. However, this made access to the case-law problematic. There was always a version in French, as the working language of the Court, but translations into other languages depended on the language in which the case was brought.[201] Therefore the entire case-law of the Tribunal was available only in French.

The CST could get away with not translating its judgments because they had no wider implications, but this would not be the case for another area of law, for instance intellectual property.

Mr Mahoney explained that the CST was based on classic French administrative law and could award unlimited compensation. It could annul decisions taken by the institutions, but could not direct them to act in a particular way.

Member States could occasionally intervene in cases. Notably, Italy and Spain had intervened in cases relating to the language regime of the institutions. Because applicants for positions in the institutions needed to speak one of English, French and German, it was perceived that citizens of other Member States were at a disadvantage. Italy and Spain had challenged this in the General Court, and intervened in certain cases where their citizens were involved.

Mr Mahoney noted a certain fragility of the Tribunal in relation to its small roster of judges. If one of the seven were ill, or unavailable for another reason, that had a greater impact on the CST than it would on a larger court. Mr Mahoney suggested that this could be overcome by drafting in retired ECJ judges on an ad hoc basis, but this would require a change in Statute of the Court which would need to be enacted by the Ordinary Legislative Procedure. The Court of Justice was responsible for the overall structure of the CJEU, but the Tribunal was able to initiate changes by sending proposals to the CJ.

On timings, Mr Mahoney explained that CST judgments had to be executed immediately by the relevant institution. However, were the institution to appeal, it would normally take around two years for the General Court to rule on the appeal. It was not possible to refuse leave to appeal, as it was regarded as almost a right, due to the influence of French administrative law.

There might be a temptation for specialised courts to go their own way, and for this reason it was important that parties were able to appeal to a more general court.

Mr Mahoney explained that the Tribunal had 200 cases pending. It had inherited 122 from the General Court when it had been established, and had been receiving all the new cases in the area since then. He reported that the Tribunal was slowly reducing the backlog.

There were fewer cases coming through for two reasons:

(1)  There had been a reduction in the number of cases brought, as the rules on costs had been changed. Previously, staff would not have to pay the costs of the institution, even if they lost. The regime had been changed so that the loser paid all the costs, and this had reduced the number of cases. Mr Mahoney pointed out, though, that the costs of bringing a case were not high. The Commission, for instance, used its Legal Service in such cases, so did not have to pay for outside lawyers—typically, a losing litigant would pay costs amounting to about €250-400. Other institutions, for instance the European Investment Bank and European Central Bank, used outside lawyers, and in these cases the costs would be €6,000-10,000.

(2)  Lord Kinnock, in his time as a Commissioner, changed the staff regulations, resulting in a substantial amount of litigation. The effect of this had now worked through the system, and the levels had reduced.

In relation to changing the rules of procedure of the CJEU as a whole, Mr Mahoney described the process as cumbersome, as it had to be decided by the Ordinary Legislative Procedure. While he could accept Member State oversight of the CJEU, he pointed to a conflict of interests in that the Member States are often parties to cases. Mr Mahoney argued that the rules should have general enabling clauses to allow the CJEU to adapt its procedures where necessary.

Mr Mahoney pointed to differences between UK and EU employment tribunals. The CST undertook a lot of the work in writing in advance of any hearing. Parties would submit papers, which the CST would then question. Though it was possible for the CST to question witnesses, this was never used. Because of this, the CST had less grip on the facts of a case. Occasionally, a case would be annulled on procedural grounds if the facts were unclear. This at least had the advantage of speed.

199   Mr Jaeger showed the Committee a copy of the judgment, which ran to over 1000 pages.  Back

200   Annual Report 2009, Table 17, p 181. Back

201   Mr Mahoney explained that in the first year of the Tribunal's operation, 85% of cases were brought in French; in 2010, this figure was 60%. Back

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2011