UK Government's Position Paper on CJEU jurisdiction: A Short List of Tricky Issues

Earlier today, the UK Government has published its position paper on the jurisdiction of the CJEU post-Brexit: Enforcement and dispute resolution - a future partnership paper (23 August 2017). The paper has been received as constructive by eg David Allen Green and Prof Armstrong, and Prof Peers has stressed on twitter that there is a clash of redlines despite the effort the paper makes to distinguish issues of enforcement (of individual rights) and dispute resolution (between the UK and the EU). I am sure I have already missed some useful reactions and that the commentary on the position paper will keep piling up in the coming hours.

With this post, I only intend to highlight some of the tricky issues that I have identified on first reading of the paper. They are presented in the same order of the relevant paragraphs of the paper where they first appear, but this does not necessarily reflect their level of trickiness.

  1. The way the position in EU is depicted may be too simplistic, in particular concerning the acceptance of international dispute resolution agreements. For example, paragraph 20 refers to the Association Agreements with Ukraine and Moldova as instances where the EU has accepted submission to binding (international) arbitration mechanisms.

    However, taking the EU-Ukraine Agreement as example, the arbitration mechanism is limited due to the need to ensure CJEU supremacy when it comes to interpretation of EU law. In that regard, Art 322(2) clearly establishes that '[w]here a dispute raises a question of interpretation of a provision of EU law [relating to regulatory approximation contained in Chapter 3 (Technical Barriers to Trade), Chapter 4 (Sanitary and Phytosanitary Measures), Chapter 5 (Customs and Trade Facilitation), Chapter 6 (Establishment, Trade in Services and Electronic Commerce), Chapter 8 (Public Procurement) or Chapter 10 (Competition), or which otherwise imposes upon a Party an obligation defined by reference to a provision of EU law], the arbitration panel shall not decide the question, but request the Court of Justice of the European Union to give a ruling on the question. In such cases, the deadlines applying to the rulings of the arbitration panel shall be suspended until the Court of Justice of the European Union has given its ruling. The ruling of the Court of Justice of the European Union shall be binding on the arbitration panel'. Given that these are matters that would be at the core of an EU-UK agreement, the extent to which agreeing on binding internal arbitration would circumvent (direct) CJEU jurisdiction to interpret EU law and identical provisions can be questioned.

    This is however presented in very soft terms in the position paper. In relation with the EU-Moldova Agreement, and under the heading 'Provision for voluntary references to CJEU for interpretation', the position paper indicates that '[t]his approach can apply in respect of both judicial and political dispute resolution models. For example, Article 403 of the EU Moldova Association Agreement requires that an arbitration panel established to resolve disputes shall, where the dispute concerns interpretation of EU law, refer the question to the CJEU and be bound by its interpretation' (para 56, emphasis added); and that 'In the case of the Moldova Association Agreement, the responsibility to make a reference rests with the arbitration panel ... These examples do not involve one party to the agreement deciding, unilaterally, to seek a binding interpretation of the agreement from the CJEU' (para 58). While this is *technically* correct, it is also presented in a misleading way because should an arbitration panel not seek the CJEU's interpretation, whcih it is required to do so, the final award would clearly not be enforceable. Ultimately, in my view, the restrictions derived from the need to ensure the CJEU's position as sole interpreter of EU law create a much harder and relevant restriction on the design of international arbitration or other dispute resolution mechanisms than the image that evaporates from the position paper.

    In fairness, this is somehow recognised in para 38 of today's position paper: 'there are limitations to the matters on which the EU can subject itself to the binding decisions of a quasi-judicial or judicial authority, like an arbitration panel. The arbitration panel cannot adjudicate on matters of interpretation of EU law so as to bind the EU and its Member States'. However, this is not followed by a view on how to resolve this limitation, should the future EU-UK agreement be subjected to international arbitration--maybe this is just aimed at creating space for negotiations, but a clearer position of the UK Government on the acceptance (or not) of a reference mechanism to the CJEU as part of arbitration-based dispute resolution mechanisms will be needed sooner rather than later and the answer seems constrained to a binary yes/no ...
  2. Whether the EU would accept to the creation of another, parallel court, like the EFTA Court can be highly questioned. The assumption in para 21 that the EFTA court is a 'model' that can be replicated seems to me difficult to accept. In my opinion, the only way of benefiting from that solution would be for the UK to become a member of the European Economic Area (which the UK Government does not want to pursue), or else for the EFTA Court to be reformed to expand its jurisdiction to the EEA + UK (which seems unlikely). In my opinion, the creation of another institution with EFTA Court features but with jurisdiction only for the EU-UK relationship does not seem plausible.

    This has a major effect on the viability of post-Brexit coordination of UK and CJEU case law as discussed in paras 46-51 of the position paper because, as is clear from all the examples in that section, the mechanisms for mutual coordination of jurisprudence have so far only been accepted within the scope of the EEA (+ Switzerland). Outside of the scope of the EEA / EFTA Court jurisdiction, it seems difficult to see the EU accepting this type of mechanisms, which are the historical result of a different time of the European integration process. Moreover, the UK government seems to point at differential approaches to case law coordination when it indicates that 'extent to which this approach may be valuable depends on the extent to which there is agreement that divergence should be avoided in specific areas' (para 51). It seems difficult to accept that the EU can tolerate divergence in any areas that are considered of relevance in the context of the future EU-UK relationship (and those not relevant, are likely to be or end up outside of the framework).
  3. The position that 'in both the UK and the EU, individuals and businesses will be able to enforce rights and obligations within the internal legal orders of the UK and the EU respectively, including through access to the highest courts within those legal orders. This would be the case in respect of both the Withdrawal Agreement, including an agreement on citizens’ rights, and the future partnership' (para 23) seems to simplistic to me. First, because this is precisely one of the redlines of the EU's negotiating position, which has indicated that there has to be a 'possibility of administrative or court proceedings to be initiated post-exit for facts that have occurred before the withdrawal date' (para 16 of EU negotiating guidelines), which implies the need to preserve CJEU intervention for the interpretation of the relevant EU law provisions as they applied at the time of the material facts. Second, because litigation is likely to raise complex issues of conflict of laws that can hardly be addressed unilaterally by either of the legal systems.

    As recognised in yesterday's position paper on cross-border civil and commercial litigation: 'Ending the direct jurisdiction of the CJEU in the UK will not weaken the rights of individuals, nor call into question the UK’s commitment to complying with its obligations under international agreements; where appropriate, the UK and the EU will need to ensure future civil judicial cooperation takes into account regional legal arrangements, including the fact that the CJEU will remain the ultimate arbiter of EU law within the EU' (para 20, emphasis added). The same will, of course, happen in every other dimension of legal relationships and, consequently, the same mechanism to 'take account of the position of the CJEU' will need to be extended universally. In my view, this is far away from the streamlined assumption that litigation will be contained in either of the jurisdictions.

    Interestingly and confusingly, para 24 of today's position paper takes a different approach and stresses that 'Ending the direct jurisdiction of the CJEU in the UK will not weaken the rights of individuals, nor call into question the UK’s commitment to complying with its obligations under international agreements. The UK’s commitment to the rule of law has been built over centuries, and reaffirmed time and again by effective, independent courts. That commitment to the rule of law means that anyone seeking redress within the UK’s legal systems will know they will be judged by clear rules applied in accordance with the law by the UK’s expert, independent and internationally respected judiciary.' The extent to which both position papers are in contradiction, or the extent to which the UK government can seriously aim to create CJEU-friendly mechanisms for civil and commercial matters and simultaneously CJEU-avoiding mechanisms for eg public law seems to me to be prone to provoke more than a few headaches for anyone trying to solve the puzzle.

Overall, I think that the conclusion in the position paper that 'there are a number of additional means [not involving the direct jurisdiction of the CJEU] by which the EU has entered into agreements which offer assurance of effective enforcement and dispute resolution and, where appropriate, avoidance of divergence, without necessitating the direct jurisdiction of the CJEU over a third party' (para 67) may be overstated and that the position paper, while more flexible than could have been expected, still seems to head full steam ahead for a clash with the unique position of the CJEU in interpreting EU law and preserving individual (citizens') rights. Time will tell.

Public procurement in the CJEU's 2016 Annual Report - What a Busy Year and How Much New Case Law To Keep Up With

The Court of Justice of the European Union (CJEU) has published its 2016 Annual Report, including a detailed assessment of its judicial activity. Even at first read, it is clear that public procurement features more prominently in this year's edition, where the CJEU offers some comments on key ECJ cases, such as Falk PharmaPartner Apelski Dariusz and Wrocław — Miasto na prawach powiatu, and PFE (see Judicial Activity Report, pp 64-65); as well as some comments on key GC cases on public procurement by the EU Institutions and on Commission's Decisions on Utilities Procurement, such as Österreichische Post v Commission, or European Dynamics Luxembourg and Others v EUIPO on the concept of conflict of interest (currently under appeal) (pp. 182-183).

The 2016 Annual Report also allows for an expansion of previous statistical analysis of the ECJ's and GC's case load in the area of public procurement (see also analyses for 2015, 2014 and 2012).

Interestingly, 2016 data shows a reduction in the accumulation of procurement-related backlog, which is now reduced to its 2010 level if the institution is taken as a whole. As the graphs below show, this reduction in backlog is mainly due to increased decision-making at ECJ level and to a significant reduction of new cases at GC level. It is worth taking a closer look at both issues.

Sharp reduction in number of new cases before the GC

The number of new procurement cases before the GC dropped from 23 in 2015 to only 9 in 2016, which marks the all time lowest level since 2006, when the statistical series started. This sharp reduction in new cases allowed the GC to catch up with some of its backlog from previous periods and to reduce it by one third (from 35 to 24 pending cases in the 2015-2016 comparison). However, the low number of cases at GC level, which mostly concerns challenges to procurement decisions by the EU Institutions, continues to indicate that the creation of more effective remedies mechanisms applicable to EU institutional procurement remains a priority for regulatory reform (as stressed by the Court of Auditors). 

Increased Decision-Making at ECJ Level

It is particularly remarkable that the ECJ managed to significantly increase its decision-making in the area of public procurement, moving from an average of 10 decided cases in the 2010-2015 period (with a highest output of 14 decisions in 2015) to 31 decisions in 2016. Coupled with a reduction in the number of new cases from 26 to 19 in the 2015-2016 comparison, this increased level of decision-making activity has cut the backlog of pending cases by one third, thus bringing it back to its 2014 level and stopping (at least for now) the worrying trend observed throughout this decade.

It is interesting to dig a bit deeper in the analysis of this remarkable surge in decision-making activity by the ECJ. The graph below shows the evolution of the public procurement decisions adopted by the ECJ, breaking down annual totals between those closed by Judgment or Opinion (ie substantive decision-making) and those closed by Order (ie procedural decision-making).

As the graph shows, the increased volume of decision-making has resulted in an increased total volume of both substantive and procedural decisions. The impact of this evolution on the different type of decisions may be easier to grasp through a simple ratio. As the table below shows, the increase in decision-making has in part been the result of the adoption of a larger proportion of procedure-based decisions. This may point towards the need for further case-by-case analysis in order to understand if this reflects any new trends concerning the ECJ's management of procurement cases, or if it is simply the result of the larger overall case load in this area. In any case, what is clear is that 2016 was a year of unprecedented substantive decision-making by the ECJ in the area of public procurement. No wonder it felt like such a busy year and that it was hard to keep up with all developments!

Why an appeal of the High Court Parliamentary approval Brexit judgment will bring the litigation to the cjeu?

The High Court has today issued its Judgment in the dispute about the UK Parliament's necessary approval of a Brexit notification--see R (Miller) -V- Secretary of State for Exiting the European Union [2016] EWHC 2768 (Admin). It has ruled that such Parliamentary approval is indeed required as a matter of UK constitutional and public law. The Government has already announced that it will appeal this decision to the UK Supreme Court (UKSC). The implications of such an appeal are important and need to be carefully considered. One such possible consequence is that the appeal (indirectly) brings the case to the docket of the  Court of Justice of the European Union (CJEU).

In my view, an appeal of the High Court's Judgment before the UKSC will indeed trigger a legal requirement under EU law for the UKSC to send a reference for a preliminary ruling to the CJEU. I have rehearsed most of my arguments on twitter earlier (see here and here) and this posts brings them together.

Basic EU Law Background

Article 267(1)(a) TFEU establishes the monopoly of interpretation of the CJEU and it indicates the Court shall have jurisdiction to give preliminary rulings concerning the interpretation of the Treaties. Article 267(2) then goes on to enable the domestic courts of the Member States to issue request preliminary rulings from the ECJ where questions of interpretation of EU law are raised before them and they consider that a decision on the question is necessary to enable them to give judgment. However, that discretion of domestic courts to request preliminary rulings from the CJEU does not apply to the courts or tribunals of a Member State against whose decisions there is no judicial remedy under national law. In that case, Article 267(3) indicates that where a question on the interpretation of EU law is raised, the highest court  shall bring the matter before the CJEU.

The uncertainties surrounding the interpretation of Art 50 TEU before the High Court

One of the extremely complex issues concerning the UK's potential withdrawal from the EU following the Brexit vote of 23 June 2016 concerns the interpretation of Article 50 TEU (on this, see here). One of the difficult sub-questions concerns the (ir)reversibility of an Art 50 TEU trigger notification. This is an essential element for an assessment of the UK's constitutional requirements for the delivery of such notification, as the High Court's Judgment makes clear.

Indeed, as a preliminary issue, in today's Judgment, the High Court has addressed the problematic interpretation of Art 50 TEU. Unanimously, the High Court has indicated that "Important matters in respect of Article 50 were common ground between the parties: (1) a notice under Article 50(2) cannot be withdrawn once it is given ..." para [10]; and that "Once a notice is given, it will inevitably result in the complete withdrawal of the United Kingdom from membership of the European Union and from the relevant Treaties at the end of the two year period, subject only to agreement on an extension of time ..." para [11].

There are two ways of interpreting the High Court's dealing with the argument on irreversibility of an Art 50 notification. First, that the High Court takes this approach in para [11] because it is common ground between the parties ex para [10]--what I would call the UK procedural approach. Second, that the High Court has of its own interpreted an Art 50 notification to be irreversible ex para [11], which happens to align with the common position of the parties in para [10]--what I would call the EU substantive interpretation approach.

The UK procedural approach is saved by the High Court's discretion under Art 267(2) TFEU to consider that the interpretation of Art 50 TEU is actually not necessary for it to adjudicate the matter at hand because this is not part of the controversy between the parties. However, the EU substantive interpretation does trigger some issues because, having recognised the interpretation of Art 50 TEU as an important aspect for the adjudication of the case, the High Court should not have taken it upon itself to interpret it and should rather have requested a preliminary ruling from the CJEU. However, unless under a very expansive interpretation of the principle of sincere or loyal cooperation in Art 4(3) TEU, this does not amount to a breach of EU law.

The uncertainties surrounding the interpretation of Art 50 TEU before the UK Supreme Court

Now, in case of an appeal of the High Court's decision before the UKSC, in my opinion, the referral to the CJEU is legally unavoidable (I will not deal for now with arguments of judicial politics or pragmatic views on the UKSC's likely course of action). Even if the parties do not challenge or even raise to the UKSC's consideration the matter of the (ir)reversibility of and Article 50 notification, it is a logical given that the UKSC needs to take a stance (even if implicit) on this point in order to be able to rule on the case. If it quashes the High Court's decision, it needs to clarify the points of law which the High Court would have gotten wrong--one of which concerns the irrevocability of an Art 50 notification. if it upholds the High Court's decision, it is (implicitly) accepting the assumption that an Art 50 notification is irrevocable. Either way, the UKSC cannot escape a substantial (implicit) consideration of the interpretation of Article 50.

In my view, this engages the UKSC's obligation to request a preliminary ruling from the CJEU under Article 267(3) TFEU and not doing so triggers a risk of infringement of EU law by the UK due to the acts (or omission, in this case) of its highest court.

Semi-Advanced EU Law Background

The UKSC's obligation to request a preliminary reference from the CJEU is controlled by the so-called CILFIT test, which establishes that "a court or tribunal against whose decisions there is no judicial remedy under national law is required, where a question of [EU] law is raised before it, to comply with its obligation to bring the matter before the Court of Justice, unless it has established that the question raised is irrelevant or that the [EU] provision in question has already been interpreted by the Court or that the correct application of [EU] law is so obvious as to leave no scope for any reasonable doubt. The existence of such a possibility must be assessed in the light of the specific characteristics of [EU] law, the particular difficulties to which its interpretation gives rise and the risk of divergences in judicial decisions within the [EU]" (283/81, EU:C:1982:335, para 21).

What does this mean for the UKSC in the Brexit litigation in case of appeal?

In short, my understanding of the CILFIT test is that a highest court of a Member State (the UKSC) must request a preliminary ruling on the interpretation of the Treaties to the CJEU and has no discretion not to do so unless: (a) the question is (objectively) irrelevant for the adjudication of the case, or (b) the provision has already been interpreted by the CJEU, or (c) there is no scope for reasonable doubt in the interpretation of the provision. None of these apply in the specific case of the Article 50 litigation.

First, it is inconceivable to me to argue that the interpretation of Art 50 and the (ir)revocability of a notice under it is irrelevant for the adjudication of this case. A different issue would be whether the UKSC could pragmatically sidestep the need to engage in that interpretation, either by presuming its content (the EU substantive interpretation approach mentioned above), or by insisting on the fact that it is common ground to the parties to the litigation and, therefore, the issue of the (ir)revocability of the notification is not (formally, explicitly) raised before it (the UK procedural approach.

However, in my opinion, neither of these avoidance strategies would meet the basic requirements of good faith in the interpretation of the CILFIT test, coupled with Article 4(3) TEU, which requires the domestic court to assess the need to request a preliminary ruling "in the light of the specific characteristics of [EU] law, the particular difficulties to which its interpretation gives rise and the risk of divergences in judicial decisions within the [EU]". The interpretation of Article 50 TEU is, to put it simply, the most relevant EU constitutional law issue since the OMT litigation and one of the top, if not the top, EU constitutional law issue since the entry into force of the Lisbon Treaty. Engaging in semantics in the analysis of the first prong of the CILFIT test against this background (ie, stretching the narrow interpretation "irrelevant") seems to me logically and legally unacceptable.

Second, it is plain that Art 50 has not been interpret by the CJEU yet. And, thirdly, it is also plain that there is scope (massive scope, a gaping hole) for reasonable doubt in the interpretation of Article 50 TEU. Thus, the so-called acte claire doctrine (ie the counterbalance of the CILFIT test) simply does not apply here.

Overall, in my opinion, the UKSC has an absolute and inexcusable obligation to request a preliminary ruling on the interpretation of Article 50 TEU from the CJEU the moment the appeal against the High Court's Judgment (eventually) reaches its docket. Otherwise, the UKSC risks triggering an infringement of EU law and eventually creating liability in damages under the Kobler / Traghetti del Mediterraneo strand of case law on State liability. Again, I am not dealing with the arguments on the likelihood of an actual infringement case brought forward by the European Commission, or the CJEU's eventual decision. I am, for now, simply stressing the state of EU law, which the UKSC would be well advised to bear in mind and uphold, unless it aims to contribute to the deterioration of the rule of law in the UK and the EU (which is something that keeps me awake at night).

Public procurement in the CJEU's Annual Report 2014 Statistics

The Court of Justice of the European Union (CJEU) has published the full version of its 2014 Annual Report, which allows for an update of the statistics available two years ago (here). The 2014 report offers interesting data about the continued relevance of public procurement in the overall activities of the Court, as well as the evolution of the backlog in the docket, which seems to be needing attention at the highest level. I provide the new data first, both for the CJEU and the GC, and then update the time series I first prepared in 2012.

CJEU 2014 data

In 2014, the CJEU opened 21 new cases on public procurement (3.42% of all new cases), of which 20 were references for a preliminary ruling and the other case was an appeal. It adopted 13 decisions in public procurement cases (11 Judgments and 2 Orders) during the same period (which represent 2.08% of all cases closed). This clearly indicates that the CJEU has accumulated a (further) backlog of around 50% of the 2014 new procurement cases.

A cursory search on CURIA's case finder shows 9 pending procurement cases (below), which would track the 2014 mismatch (although two of them are 2015 cases)--but only assuming there was no backlog of procurement cases at the beginning of that year, which I do not think is correct (see below for some conjectures). In any case, more transparency on the backlog of cases would be desirable.

GC 2014 data

In 2014, the GC opened 17 new cases on public procurement, and it issued 18 decisions (16 Judgments and 2 Orders). The GC is managing to keep the number of pending cases stable at around 35. The fact that the GC publishes explicit statistics on pending cases by subject matter makes things easier.

Time series

The following is an update of the time series I prepared in 2012. Just like then, please note that unfortunately, prior to 2010, the data for the CJEU does not include a separate category for public procurement cases (they were likely to be classified under approximation of laws, or under the relevant fundamental freedom). Therefore, the actual numbers may be higher than the available statistics show but, in my view, the general trends remain clear: backlog is increasing and now reaches about 75 cases. As I mentioned above, more transparency (or a correction of incorrect classification of cases, if there is any) would be much desirable.


CJEU criticisably supports taxi monopoly in State aid case on use of London's bus lanes (C-518/13)

The Court of Justice of the EU has ruled in Eventech, C-518/13, EU:C:2015:9 and has broadly followed AG Wahl's approach to the case (criticised here) to determine that "The practice of permitting, in order to establish a safe and efficient transport system, Black Cabs to use bus lanes on public roads during the hours when the traffic restrictions relating to those lanes are operational, while prohibiting minicabs from using those lanes, except in order to pick up and set down passengers who have pre-booked such vehicles, does not appear, though it is for the referring court to determine, to be such as to involve a commitment of State resources or to confer on Black Cabs a selective economic advantage for the purpose of Article 107(1) TFEU."

The Eventech Judgment is criticisable for the same reasons identified in view of the AG Opinion (see here) and, in my view, constitutes a bad precendent in the treatment of access to (quasi?) essential facilities under public property. The analysis of the economic exploitation of the bus lanes is particularly weak, as it completely avoids the clear issue that black cabs do use that infrastructure in order to develop an economic activity--which, consequently, creates important issues of free access to public goods that the CJEU has simply disregarded. It can just be lamented that the CJEU did not identify the logical traps that affected the AG Opinion and deviated from them. Maybe, at least, the case can be used as yet another clear indication of the need to involve economists in the decision-making process of the CJEU [for some exploratory thoughts, see A Sanchez Graells, The Importance of Assessing the Economic Impact of the Case Law of the Court of Justice of the European Union: Some Exploratory Thoughts (April 18, 2013)]. 

Is the CJEU engaging in 'Judicial Abstinence'? Or, where do we go from here?

The CJEU has released the ebook of the May 2013 conference that celebrated 50 years of its Van Gend en Loos Judgment. The book contains many interesting contributions and, in my view, one of the most thought-provoking is Prof Catherine Barnard's 'Van Gend en Loos to(t) the future'.
Prof Barnard reflects about the future role of the CJEU in a changed and changing EU and identifies a trend of 'Judicial Abstinence' that 'leaves the uncomfortable impression that the Court is reneging on its key function, first articulated in Van Gend en Loos, that EU law must be effective' (p. 122).
Her analysis is relevant and expands well beyond the area of labour law, where she focusses. In my view, recent Judgments in other (even more fundamental?) areas such as the legal position and effectiveness of the Charter of Fundamental Rights of the EU also show the judicial abstinence the CJEU is (selectively) engaging in (for instance, in case C-313/12). And that leaves us with the difficult question of why is the CJEU (suddenly) so averse to (continuying to) act as constitutional court at EU level?
Pessimists could say that we are witnessing a deflation of the EU law supremacy balloon, and that the 'golden age' of the CJEU lasted 50 years and is already over. However, Prof Barnard offers a positive outlook as long as the CJEU departs from 'hardcore' judicial abstinence and opts for a medium-ground strategy (which she terms 'Old-Rules Lite'). As she stresses,
The EU of 2013 is infinitely more complex and any solutions will have to reflect that. There will be no Van Gend en Loos II. However, this does not mean the end of the Court of Justice and its influence. Quite the contrary. It means a repositioning of the Court from standard bearer of EU integration to ensuring that the EU is able to function in its new, more fragmented reality. This paves the way for the Court to develop a new kind of doctrine of effectiveness, one that might mean endorsing the devolution of more decision-making power to the Member States or other actors (p. 122, emphasis added).
The next few years will show if the CJEU is up to the task.

GC on #quality assurance #standards in #publicprocurement: A knee-jerk reaction (T-288/11)

In its Judgment of 6 May 2013 in case T-288/11 Kieffer Omnitec v Commission (only available in French), the General Court of the European Union (GC) was presented with an important issue concerning the proportionality of quality assurance requirements under EU public procurement rules. In a setting that resembled the issue addressed a year ago by the Court of Justice (CJUE) in relation to general corporate social responsibility / fair trade requirements in Commission v Netherlands (Fair trade beverages) (C‑368/10), the GC was asked to consider whether requiring that tenderers be ISO certified for all their maintenance activities is disproportionate and, consequently, breaches the applicable EU rules.

In the case at hand, the European Commission had tendered a contract for the maintenance of HVAC, sprinklers and other equipment in one of its buildings. As a part of the tender requirements, the Commission requested that all tenderers furnished proof of ISO certification valid for the whole of their maintenance activities. A tenderer that failed to provide such proof (but which engaged a third party to ISO-audit its activities in the Commission's building in case it was awarded the contract) and was, hence, not considered for the award of the contract challenged this requirements on various grounds. Amongst the challenges raised by the disappointed tenderer, it is worth noting that it considered that extending the requirement to all maintenance activities instead of limiting it to the activities covered by the contract was excessive and disproportionate.

One of the arguments presented by the disappointed tenderer was that, despite the Commission not being directly subjected to the provisions of Directive 2004/18 on public procurement, the rules established in its article 49 should be taken into consideration. Such provision specifically addresses the issue of quality assurance standards and mandates that:
Should they require the production of certificates drawn up by independent bodies attesting the compliance of the economic operator with certain quality assurance standards, contracting authorities shall refer to quality assurance systems based on the relevant European standards series certified by bodies conforming to the European standards series concerning certification. They shall recognise equivalent certificates from bodies established in other Member States. They shall also accept other evidence of equivalent quality assurance measures from economic operators (emphasis added).
However, before entering the discussion of the proportionality of the requirement, the GC strangely separated itself from the use of Directive 2004/18 as a valid interpretation guide (by analogy). The GC considered that:
22 Before turning to the examination of the matter at hand, it is important at the outset to recall that, regarding the law applicable to procedures for the award of public service contracts undertaken by the institutions of the European Union, these procedures are governed by the provisions of Title V of Part I of the Financial Regulation as well as its Implementing Rules.

23 These provisions are based, of course, on the EU directives in this area (see, to that effect, judgment of 12 July 2007, Evropaïki Dynamiki / Commission, T-250/05, not published in the ECR, paragraph 1, and judgment of 9 September 2010, Evropaïki Dynamiki / EMCDDA, T-63/06, not published in the ECR, paragraph 4). However, Member States are the sole addressees of these directives and, therefore and in principle, these rules only govern public procurement by the institutions of the Member States. Such directives do not apply to public contracts awarded by the institutions of the Union on their own account, save for the question regarding the thresholds that determine the manner of publication, the choice of procedures and the applicable deadlines (judgment of 19 March 2010, Evropaïki Dynamiki / Commission, T-50/05, p. II-1071, paragraph 104).
24 It follows that, in this case, in the examination of the first plea raised by the applicant, only the provisions of the Financial Regulation and the Implementing Rules need to be taken into consideration. Reversely, however, there is no need to take into account Article 49 of Directive 2004/18, cited by the applicant. (T-288/11 at paras 22 to 24, own translation from French).
With these remarks, the GC is departing from its previous practice to consider the rules under Directive 2004/18 as a valid guide for interpretation and is generating a risk of inconsistency in the development of EU public procurement law. Moreover, there is no good reason why the general criteria encapsulated in article 49 dir 2004/18 could not be expressly referred to since, it must be stressed, they are no more than a specification of the general principles of 'technical neutrality' (in broad terms) and proportionality that the GC must take into consideration anyway. Nonetheless, as we shall see, by excluding the use of the general provision as a valid analytical framework and 'chopping off' the last bit of article 49 dir 2004/18, the GC conveniently avoids the issue of having to consider if the Commission failed to accept 'equivalent quality assurance measures'.

Once the assessment is carried out precisely in terms of the proportionality of the ISO requirement, the GC finds that:
38 [...] the requirement [of full ISO certification] does not appear disproportionate to the extent that, on the one hand, Article 137, paragraph 3a of the Implementing Rules provides that "[w]hen the contracting authorities require the production of certificates drawn up by independent certification bodies attesting that the economic operator complies with certain standards of quality assurance, they shall refer to quality assurance systems based on the relevant European standards certified by bodies conforming to the European standards series concerning certification."

39 Moreover, as the Commission has rightly pointed out in its defense, in the selection of tenderers, when it comes to ensuring their technical capacity, the ISO certification must necessarily target the agent itself and not the contract to be awarded. In fact, ISO 9001 specifies requirements for the quality management system when an organization needs to demonstrate its ability to provide a product that meets customer and regulatory and legal requirements. This standard of "quality" is applicable to the process that a company uses to make its products or services and so can attest to the effectiveness and quality of its organization and its ability to provide the deliverables covered by the contract.

40 It is true that, except for ISO certifications attesting to the quality of the organization of the company, there are ISO certifications to attest to the quality of products or specific projects. However, as pointed out by the Commission, only the former may be required under the selection criteria of a given tender. The latter can only be used, as appropriate, as a contract performance condition, since they can only be obtained once the contract is in place in order to certify that the project or the product has been made in accordance with ISO standards.

41 Contrary to what the applicant claims, the Court considers that the requirement of a certificate attesting that the bidders to comply with ISO [for all their maintenance activities] is proportionate to the subject of the contract. (T-288/11 at paras 38 to 41, own translation from French, emphasis added).
In my view, there are several objections to be raised to the finding of the GC. Firstly, as mentioned in passing, this 'maximalistic' approach to quality control that links it to a selection criteria may run contrary to  the approach taken by the CJEU in Commission v Netherlands (Fair trade beverages). Indeed, the CJEU took a very restrictive approach to the use of general technical requirements that go beyond those specified in art 48 dir 2004/18. As the CJEU clearly put it 'Article 48 exhaustively lists the factors on the basis of which the contracting authority may evaluate and assess the technical and professional abilities of the tenderers' (C-368/10 at para 105). 

In that regard, it is worth stressing that art 48 dir 2004/18 only mentions quality assurance in the following respects: i) an indication of the technicians or technical bodies involved in quality control [art 48(2)(b)]; ii) a  description of the technical facilities and measures used by the supplier or service provider for ensuring quality [art 48(2)(c)]; and, only in relation to specific products to be supplied, certificates drawn up by official quality control institutes or agencies of recognised competence attesting the conformity of products clearly identified by references to specifications or standards [art 48(2)(j)(ii)]. Therefore, when art 49 dir 2004/18 refers to the 'production of certificates drawn up by independent bodies attesting the compliance of the economic operator with certain quality assurance standards', it can be said that it is only referring to the requirement of art 48(2)(j)(ii)--and clearly sets a strong link (and limitation) with the specific products (not services) to be supplied in a given contract. Failing that, and in any case, art 49 requires contracting authorities to accept 'equivalent quality assurance measures'.

In my view, then, the finding of the GC in Kieffer Omnitec is irreconcilable with the case law of the CJEU on selection criteria and with the foreseeable interpretation of art 49 dir 2004/18. Moreover, even from a broader perspective--and similarly to what I have argued elsewhere [Sanchez Graells, Public Procurement and the EU Competition Rules (Oxford, Hart, 2011) pp. 315]:
Even if rules on qualitative selection and non-discrimination requirements are formally complied with in a given tender, the adoption of certain award criteria could generate the same results as an infringement of those rules. That could be the case if the award criteria or their weighting favoured tenders submitted by certain operators on the basis of conditions that could not have been used for the purposes of the qualitative selection of candidates or that automatically exclude de facto a significant number of tenders (or even restrict the number of compliant tenders to one). For instance, they could do so by requiring the implementation of quality management systems for the purposes of the specific contract that would have proven excessive or irrelevant for the purposes of assessing the general suitability of the tenderer; or that exclude certain operators because they focus on requirements whose implementation would be impossible for tenderers that did not comply with these or other requirements beforehand, or whose partial implementation would not be economically viable with regard exclusively to the specific contract.[1] In these instances, the adoption of such award criteria could generate significant distortions or restrictions of competition—without, it must be admitted, generating a substantial potential for discrimination. Therefore, such a strategy should be banned and contracting authorities should guarantee that the award criteria and their weighting ensure equality of opportunity of all tenderers and, consequently, should not focus on or advantage compliance with criteria not restricted to the tender itself—ie criteria that undertakings would be in a position to comply with or not depending on previous or general conditions unrelated (or not specifically related) to the subject-matter of the contract.[2]

[1] In similar terms, rejecting the possibility of establishing general requirements that go further than required by the object of the contract, see P Trepte, Regulating Procurement. Understanding the Ends and Means of Public Procurement Regulation (Oxford, Oxford University Press, 2004) 197–8.

[2] For instance, if certifying compliance with a given quality standard for the product required the previous certification of the general operations of the undertaking as being compliant with a more general quality control system, and the tender documents did not require tenderers to be certified under that standard—then, giving better evaluations to certified than to non-certified products would generate a distortion of competition by de facto excluding or reducing the chances of award to non-certified undertakings (which would not be in a position to get the products certified only for the purposes of the tender). Therefore, by indirectly advantaging or requiring compliance with a condition not imposed at the qualitative selection stage, which refers to more general conditions unrelated to the specific contract, the contracting authority would be distorting competition in a way that should be declared to run contrary to the directives.
As the discussion above shows (despite it being referred to award criteria), the GC has opened the door to the requirement of general certification for undertakings to  participate in tenders. In my view, this is incorrect, in that contracting authorities can only be concerned with the quality assurance of the products they are supplied or the services they receive, but cannot use procurement as a regulatory tool to mandate quality assurance compliance that goes beyond the remit of the contractual object--in the same manner that the CJEU clearly said in Commission v Netherlands (Fair trade beverages) that public procurement cannot be used to mandated corporate social responsibility.

In my opinion, the Judgment of the GC in Kieffer Omnitec is a knee-jerk reaction to an action brought by a disappointed bidder that clearly did not meet the technical requirements (properly) set by the European Commission. It will be desirable to hear the CJEU interpret art 49 dir 2004/18 and to rule on its (analogous) application to the procurement conducted by the EU Institutions--particularly because the trends to potential inconsistent development of EU public procurement law and the regulatory use of procurement for quality control purposes are not desirable at all.

CJEU puts a noose around its neck: Again on hypertrophy of Art 267 TFEU (C-416/10)

In its Judgment (Grand Chamber) of 15 January 2013 in case C-416/10 Križanand Others, the Court of Justice of the EU (CJEU) has reiterated its constant case law whereby internal constitutional rules cannot trump or diminish its role as the only authentic interpreter of EU Law (art 267 TFEU). 

The protection that CJEU has built around its ultimate jurisdiction concerning EU Law interpretation has been strengthened in several decisions adopted since 2010 and, in my view, the wording of the Križanand Judgment is definitive:
68 A rule of national law, pursuant to which legal rulings of a higher court bind another national court, cannot take away from the latter court the discretion to refer to the Court of Justice questions of interpretation of the points of European Union law concerned by such legal rulings. That court must be free, if it considers that a higher court’s legal ruling could lead it to deliver a judgment contrary to European Union law, to refer to the Court of Justice questions which concern it (Case C378/08 ERG and Others [2010] ECR I1919, paragraph 32; and [Case C-173/09 Elchinov [2010] ECR I8889], paragraph 27).
69 At this stage, it must be noted that the national court, having exercised the discretion conferred on it by Article 267 TFEU, is bound, for the purposes of the decision to be given in the main proceedings, by the interpretation of the provisions at issue given by the Court of Justice and must, if necessary, disregard the rulings of the higher court if it considers, in the light of that interpretation, that they are not consistent with European Union law (Elchinov, paragraph 30).
70 The principles set out in the previous paragraphs apply in the same way to the referring court with regard to the legal position expressed, in the present case in the main proceedings, by the constitutional court of the Member State concerned in so far as it follows from well-established case-law that rules of national law, even of a constitutional order, cannot be allowed to undermine the unity and effectiveness of European Union law (Case 11/70 Internationale Handelsgesellschaft [1970] ECR 1125, paragraph 3, and Case C-409/06 Winner Wetten [2010] ECR I-8015, paragraph 61). Moreover, the Court of Justice has already established that those principles apply to relations between a constitutional court and all other national courts (Joined Cases C-188/10 and C-189/10 Melki and Abdeli [2010] ECR I5667, paragraphs 41 to 45). [...]
72 Finally, as a supreme court, the Najvyšší súd Slovenskej republiky [Slovak Supreme Court] is even required to submit a request for a preliminary ruling to the Court of Justice when it finds that the substance of the dispute concerns a question to be resolved which comes within the scope of the first paragraph of Article 267 TFEU. The possibility of bringing, before the constitutional court of the Member State concerned, an action against the decisions of a national court, limited to an examination of a potential infringement of the rights and freedoms guaranteed by the national constitution or by an international agreement, cannot allow the view to be taken that that national court cannot be classified as a court against whose decisions there is no judicial remedy under national law within the meaning of the third paragraph of Article 267 TFEU. (C-416/10 paras 68 to 72, emphasis added).
In my view, paragraph 72 of the Križanand Judgment broadens the scope of the obligations of national Supreme Courts and imposes a counterintuitive (and somehow extensive) interpretation of the concept of domestic court "against whose decisions there is no judicial remedy under national law" (interpreting "remedy" as "full review", I would say). A straightforward reading of the Križanand Judgment seems to imply that all Supreme Courts and all Constitutional Courts of the Member States (where both of them exist) are bound to submit preliminary references under Article 267 TFEU, regardless of the system of checks and balances between both institutions established under the relevant national constitution.

This seems a logic consequence of the supremacy of EU Law and the central position of the CJEU in its interpretation. However, it may generate an even larger number of preliminary references to CJEU--which volume is already generating significant management difficulties. As I already indicated, the preliminary reference system is under significant pressure and risks hypertrophy. 

 In my view, this implies that we need to allow all domestic courts, including the highest courts of the Member States against whose decisions there is no judicial remedy under national law to "ask responsibly". Otherwise, we face a significant risk of hypertrophy of the preliminary ruling instrument. Therefore, in my opinion, the  boundless wording of the Križanand Judgment scenifies the CJEU putting a noose around its own neck.

In-house providing and (minimum) "effective" public control: Sunset or breaking dawn for purely public (commercial) service providers? (C‑182 and 183/11)

In its Judgment of 29 November 2012 in Joined Cases C‑182/11 and C‑183/11, Econord SpA v Comune di Cagno and Comune di Varese (C-182/11) and Comune di Solbiate and Comune di Varese (C-183/11), the Court of Justice of the EU has offered a succinct reminder of its case law on in-house providing as an exception to the applicability of the EU public procurement Directives.  

According to this line of case law, contracting entities can award contracts directly (ie without a competitive tender) where they exercise over the contractor a control similar to that which they have over their own departments, and the contractor carries out the essential part of its activities with the contracting authorities to which it belongs. In those cases, it is assumed that there is no potential for competition and that the market is not affected by the decision of the contracting authority to retain the activity "in-house".

However, in Econord, the CJEU has taken an additional step in the fine tuning of the concept of "similar control" required under the in-house providing exception. In its Judgment, the CJEU has stated that:
27 According to settled case-law, there is ‘similar control’ where the entity in question is subject to control enabling the contracting authority to influence that entity’s decisions. The power exercised must be a power of decisive influence over both the strategic objectives and the significant decisions of that entity (Parking Brixen, paragraph 65; Coditel Brabant, paragraph 28; and Sea, paragraph 65). In other words, the contracting authority must be able to exercise a structural and functional control over that entity (Commission v Italy, paragraph 26). The Court also requires that this control should be effective (Coditel Brabant, paragraph 46).
28 According to the case-law, where use is made of an entity jointly owned by a number of public authorities, the ‘similar control’ may be exercised jointly by those authorities, without it being essential for such control to be exercised individually by each of them (see, to that effect, Coditel Brabant, paragraphs 47 and 50, and Sea, para. 59). 
29 It follows that, if a public authority becomes a minority shareholder in a company limited by shares with wholly public capital for the purpose of awarding the management of a public service to that company, the control that the public authorities which are members of that company exercise over it may be categorised as similar to the control they exercise over their own departments when it is exercised by those authorities jointly (Sea, para. 63). 
30 In those circumstances, although, where a number of public authorities make use of a common entity for the purposes of carrying out a common public service task, it is certainly not essential that each of those authorities should in itself have an individual power of control over that entity, nevertheless, if the very concept of joint control is not to be rendered meaningless, the control exercised over that entity cannot be based solely on the controlling power of the public authority with a majority holding in the capital of the entity concerned
31 Where the position of a contracting authority within a jointly owned successful tenderer does not provide it with the slightest possibility of participating in the control of that tenderer, that would, in effect, open the way to circumvention of the application of the rules of EU law regarding public contracts or service concessions, since a purely formal affiliation to such an entity or to a joint body managing it would exempt the contracting authority from the obligation to initiate a tendering procedure in accordance with the EU rules, even though it would take no part in exercising the ‘similar control’ over that entity (see, to that effect, Case C-231/03 Coname [2005] ECR I-7287, paragraph 24).
32 Consequently, in the cases before the referring court, it is for that court to verify whether the signing, by the Comune di Cagno and the Comune di Solbiate, of a shareholders’ agreement conferring on them the right to be consulted, to appoint a member of the supervisory council and to nominate a member of the management board, in agreement with the other authorities concerned by that shareholders’ agreement, can enable those municipal councils to contribute effectively to the control of Aspem.
33 In the light of the foregoing, the answer to the question referred is that where, in their capacity as contracting authority, a number of public authorities jointly establish an entity with responsibility for carrying out their public service mission, or where a public authority subscribes to such an entity, the condition established by the case-law of the Court to the effect that, in order to be exempted from their obligation to initiate a public tendering procedure in accordance with the rules of EU law, those authorities must jointly exercise over that entity control similar to the control they exercise over their own departments, is fulfilled where each of those authorities not only holds capital in that entity, but also plays a role in its managing bodies. (Joined Cases C‑182/11 and C‑183/11, paras. 27 to 32, emphasis added).
In my view, the Judgment of the CJEU must be interpreted in a functional manner and has refined the requirement for similar control and transformed it into a requirement for "similar, active and effective control". The requirement for contracting authorities to "play a role" in the management bodies of the entities that are considered to remain "in-house" must be active and effective, and it will not suffice that they (jointly) "take a seat" in the relevant boards (as that would fall short for ensuring that they have (more than) "
the slightest possibility of participating in the control of that tenderer" and that they "
take [...] part in exercising the ‘similar control’ over that entity"

Therefore, the answer in view of the specific circumstances of the cases joined in Econord, where the contracting authorities merely entered into "a shareholders’ agreement conferring on them the right to be consulted, to appoint a member of the supervisory council and to nominate a member of the management board, in agreement with the other authorities concerned", should be that they do not exercise a similarly effective control over the contractor as they do with their own administrative units.

If that is the correct interpretation of the Econord Judgment, it would generate difficulty for the creation of purely public (commercial) service providers, whereby a public authority would create and retain majority control of an entity entrusted with the provision of SGEIs, SSGIs or other local services and then offer its services to other contracting entities that would acquire a minority stake and not get involved in its day to day operations. In my view, such development would be welcome and a consistent complement to the competition rules in articles 106 and 107 TFEU. If contracting authorities want to cooperate directly (thorugh public-public partnerships) or indirectly (through instrumental entities), they need to remain actively engaged in the provision of the services contracted out (in-house). 

Otherwise, if the contracting authorities want to disengage from the direct management of those services and take the back seat (eg in a board of directors), there is no reason to see why public contractors should be shielded from the competition of private contractors, since both would be offering a commercial relationship to the outsourcing contracting authority and there would be an effective risk of generating relevant distortions of competition [see Sanchez Graells, Public Procurement and the EU Competition Rules (Oxford, Hart Publishing, 2011) 240-242]. Therefore, in the lack of a sufficiently active involvement, in the absence of an actual organic link between the contracting authority and the "in-house" entity, there is no good reason to exclude the application of the EU public procurement rules, as the CJEU has quite clearly stressed.

Therefore, it will be interesting to see what is the final decision of the Italian courts in the domestic cases leading to Econord, but a decision that upheld the applicability of the in-house exception would be, in my opinion, an inappropriate reading of the CJEU's Judgment.

Distinguishing 'lawyers' and 'legal advisors': The CJEU has its say: It all rests on independence

In its Judgment of 6 September 2012 in Joined Cases C‑422/11 P and C‑423/11 P - Prezes Urzędu Komunikacji Elektronicznej and Republic of Poland v European Commission, the Court of Justice of the EU (CJEU) has faced the tricky question of whether 'legal advisors' are qualified to represent clients before the EU Courts on equal footing with 'lawyers'.

In some Member States this discussion can be moot, since there is no distinction between 'legal advisors' and 'lawyers' (and all legal professionals must belong to the same bar, and are usually considered lawyers regardless of their actual role or of whether they ever appear in court). However, in countries where there is a distinction between both professions (such as in Poland or Denmark), or where there is a similar distinction under other names (such as the distinction between barristers and solicitors in the UK, although it is growing increasingly diffuminated), this issue may be highly relevant.

In the case at hand, a Polish entity was represented before the European General Court by its in-house legal advisors (not lawyers), which had an employment relationship with the entity. The GC considered that this fell short from the requirement in Article 19 of the Statute of the Court of Justice of the European Union of being represented by a 'lawyer' and, consequently, dismissed the action as inadmissible.

In view of the GC, it was irrelevant that the professionals where formally qualified as 'legal advisors' rather than 'lawyers', but the disqualifying circumstance for them to represent the entity lied in 'the existence of a subordinate relationship within the [entity] – even if only to its Director General – when their sole function is to assist [it], [which] implies a degree of independence less than that of a legal adviser or a lawyer practising in a firm that is external to their client.' (GC Order in case T-226/10 Prezes Urzędu Komunikacji Elektronicznej v Commission, at para. 21, emphasis added).

The CJEU has concurred with the GC and has insisted on the basic elements of the profession of a 'lawyer' under EU Law. In my opinion, it is worth highligting that
23. [...] the conception of the lawyer’s role in the legal order of the European Union, which is derived from the legal traditions common to the Member States, and on which Article 19 of the Statute of the Court of Justice is based, is that of collaborating in the administration of justice and of being required to provide, in full independence and in the overriding interests of that cause, such legal assistance as the client needs (see, to that effect, Case 155/79 AM & S Europe v Commission [1982] ECR 1575, paragraph 24; [Case C-550/07 P Akzo Nobel Chemicals and Akcros Chemicals v Commission and Others], paragraph 42; and [Joined Cases C-74/10 P and C-75/10 P EREF v Commission] , paragraph 52).
24. The requirement of independence of a lawyer implies that there must be no employment relationship between the lawyer and his client (see EREF v Commission, paragraph 53 and the case-law cited). As the General Court correctly held at paragraph 18 of the contested order, the concept of the independence of lawyers is determined not only positively, that is by reference to professional ethical obligations, but also negatively, that is to say, by the absence of an employment relationship (Akzo Nobel Chemicals and Akcros Chemicals v Commission, paragraph 45).
The CJEU also clarified that adopting this interpretation is not in breach of Member States' competences for the regulation of the legal professions--even if, indeed, in the absence of specific EU rules in the field, each Member State is free to regulate the exercise of the profession of lawyer in its territory and, therefore, the rules applicable to that profession may differ substantially from one Member State to another. The CJEU saved this obstacle by considering that
34. Although, as noted in paragraph 23 [...] the conception of the lawyer’s role in the legal order of the European Union derives from the legal traditions common to the Member States, in the context of disputes brought before the Courts of the European Union, that conception is implemented objectively and is necessarily independent from the national legal orders.
This generates a logic puzzle, since it is for Member States to determine who qualifies as a 'lawyer' under their national laws in the first instance, and generates a worrying pressure for potential reform in those Member States where some legal professionals face the difficulty of being qualified to represent clients under national law, but unable to appear before the EU Courts due to this additional independence requirement.

A different reading could be that, in view of the concept of 'lawyer' under EU Law, all legal professionals that do not meet the stated independence requirements should not be called 'lawyers' and, consequently, that all Member States may now face pressure to distinguish between 'lawyers' and 'legal advisors' in their domestic rules governing the legal professions.

Either way, in my view, the Judgment of the CJEU in Prezes Urzędu Komunikacji Elektronicznej can be seen as an uncomfortable reminder that the harmonisation of the rules governing the legal professions in the EU are still a pending item in the creation of a truly unified European Legal / Judicial Space, and that more efforts will be needed in this area in the coming years.

CJEU stresses that consistent interpretation of domestic laws with EU provisions cannot give rise to, or aggravate liability in, criminal law

The Court of Justice of the European Union's Judgment of 28 June 2012 in Case C-7/11 Criminal proceedings against Fabio Caronna has stressed once again the general principle of EU law that the duty for national courts to interpret and apply domestic laws in a manner that is consistent with EU law (ie consistent interpretation, on the basis of art 288 TFEU) has the clear limit that it must not give rise to, or aggravate liability in, criminal law (see press release:

The case concerned the wholesale distribution of medicinal products in Italy. Generally, wholesale distribution of such products is regulated under EU law by Directive 2001/83/EC on the Community code relating to medicinal products for human use, as amended by Commission Directive 2009/120/EC. The Directive requires wholesale distributors to obtain a special authorisation, for which certain harmonised requirements are set. However, relevant Italian domestic legislation sets a different requirement, since it allows pharmacists and companies of pharmacists in possession of a licence to operate a pharmacy (ie to sell medicinal products at retail level) to also operate as wholesale distributors (and that 'standard' authorisation does not impose the same requirements that the Directive sets for the 'special' wholesale authorisation).

The common understanding, even by public prosecutors in the specific case, of the Italian rules  on retail and wholesale authorisations was that a pharmacist already authorised to retail medicinal products was exempt from the obligation to obtain the special authorisation required under the material national and European Union rules applicable. On its face, the argument seemed deffective (as the referring Court of Palermo implicitly indicated by requesting the CJEU to issue a preliminary ruling), and it seems rather clear that Italian law has not properly transposed Directive 2001/83/EC, since it requires no special authorisation for wholesale distribution on top of the 'standard' retail authorisation given to pharmacists.

The point of interpretation of domestic law in light of the requirements of EU law was, then, to determine whether pharmacists engaging in wholesale trade on the basis of their 'standard' authorisation to distribute medicinal products at retail level were doing so unlawfully under EU rules. The issue was not minor since this could generate criminal liability, as Italian legislation also determined that any person that engages in unauthorised wholesale distribution of medicinal products was liable to punishment in the form of imprisonment for a term of six months to one year and a fine of between €10,000 and €100,000.

In the case at hand, the CJEU has been clear in distinguishing the consequences of consistent interpretation of Italian laws with Directive 2001/83/EC in general and for the more specific purposes of determining criminal liability.

In general terms, it has clearly indicated that the Directive imposes on Member States a general obligation to make the wholesale distribution of medicinal products subject to possession of a special authorisation, and that the requirement under the Directive to obtain authorisation for the wholesale distribution of medicinal products is applicable to a pharmacist who, as a natural person, is also authorised under domestic law to operate as a wholesaler in medicinal products; since it cannot be presumed from the simple fact that pharmacists satisfy the conditions laid down by the Member States for retail supply that they also comply with the conditions laid down by harmonised rules at EU level for wholesale distribution.

However, as such finding could determine criminal liability for the pharmacists that behaved in accordance with national law but breached EU law, the CJEU reminded that
51 It should be noted, as correctly pointed out by the Commission, that even though national courts are required to interpret domestic law, so far as possible, in light of the wording and purpose of a Directive in order to achieve the result sought by the Directive and, accordingly, to comply with the third paragraph of Article 288 TFEU, that obligation to interpret national law in conformity with European Union law is subject to certain limits in criminal matters.
52 As the Court has held, the principle of interpreting national law in conformity with European Union law is limited by general principles of law which form part of the European Union legal system, in particular, the principles of legal certainty and non-retroactivity. Thus, a directive cannot, of itself and independently of a national law adopted by a Member State for its implementation, have the effect of determining or aggravating the liability in criminal law of persons who act in contravention of the provisions of that directive (see, inter alia, Case C‑60/02 X [2004] ECR I‑651, paragraph 61 and the case‑law cited).
55 [... ultimately] the principle that criminal penalties must have a proper legal basis, enshrined in Article 49(1) of the Charter of Fundamental Rights of the European Union, would prohibit the imposition of criminal penalties for such conduct, even if the national rule were contrary to European Union law (see, by analogy, X, paragraph 63).

In this other group of cases (the vast majority, predictably), the case law of the CJEU imposes a much more demanding exercise of consistent interpretation (which could almost go so far as to impose a 'disguised' contra legem interpretation). As explained elsewhere,

given that the limits of the principle of consistent interpretation remain somewhat blurry and that the CJEU has adopted an expansive approach to the issue of the obligation of Member States to guarantee the effectiveness of directives, it is submitted that the limits of legal construction of Member States’ law with conformity to EU directives should be interpreted restrictively in order to favour to the maximum extent the (indirect) effectiveness of EU law and the goals pursued by EU directives. This is all the more necessary in view of recent developments of the rules of construction developed by the CJEU that are superseding the traditional boundaries of the theory of direct effect and point towards a more general doctrine of ‘legality review’ of the legislative actions of Member States [S Prechal, ‘Does Direct Effect Still Matter?’ (2000) 37 Common Market Law Review 1047, 1051. cf C Hilson, ‘Legality Review of Member State Discretion under Directives’ in T Tridimas and P Nebbia (eds), European Union Law for the Twenty-First Century: Rethinking the New Legal Order (Oxford, Hart Publishing, 2004) 223] and towards the expansion of the boundaries of legal interpretation that conform to the TFEU and secondary rules (in what has been termed as leveraged development) [S Treumer and E Werlauff, ‘The Leverage Principle: Secondary Community Law as a Lever for the Development of Primary Community Law’ (2003) 28 European Law Review 124]—which seem to have overcome the notorious difficulties that the early developments of the direct effect doctrine generated (although they may pose some other interpretative problems of their own) [D Curtin, ‘The Province of Government: Delimiting the Direct Effect of Directives in the Common Law Context’ (1990) 15 European Law Review 195, 220–23]. The limit seems to lie where consistent interpretation requires national courts and authorities to overcome ‘merely’ interpretative functions (broadly defined) and to assume legislative functions [R Alonso García, ‘La interpretación del Derecho de los estados conforme al Derecho comunitario: Las exigencias y los límites de un nuevo criterio hermenéutico’ (2008) 28 Revista española de Derecho europeo 385, 401]. Nonetheless, drawing the dividing line will usually be a difficult task and, as already mentioned, the clear prevalence of a pro communitate interpretative principle must be identified in the relevant case law.
To sum up, it is submitted that (except in those cases where there is potential for criminal liability), Member States are under an almost absolute obligation to guarantee that domestic legislation is interpreted and applied in a manner that is consistent with EU law and, in particular in the case of directives, to ensure that their goals and intended effects are attained through national legislation—regardless of whether that legislation was adopted for the sake of transposing those directives, and regardless of the proper or improper transposition of those directives.
In my opinion, this seems like an area of EU law where case law will continue to develop in the recent future, particularly if national courts continue to refer cases for interpretation, for instance, in relation with the ever increasing number of soft law and hybrid instruments issued by EU institutions.
Even if the Caronna Judgment does not advance EU law in this field, it is a very clear reminder of the limits of consistent interpretation in cases where criminal liability is concerned, where the exercise required from national courts is much more restricted than in cases where no criminal liability is involved.