Brexit may have negative effects for the control of public expenditure, particularly regarding subsidies to large companies

In the current state of turmoil, it is difficult to speculate on the exact relationship between the EU and the UK that can result from the Brexit vote and the future negotiations to be held under Article 50 TEU, in case it gets triggered. However, in order to contribute to the debate of what that relationship should look like in the interest of taxpayers in the UK, it is important to consider the implications that a post-Brexit deal could have in terms of the potential disappearance of the EU rules applicable to the control of how public funds are spent. A reduction in the control mechanisms applicable to certain types of public expenditure could indeed diminish the effectiveness of policies funded by UK taxpayers and create shortcomings in public governance more generally.

This is particularly clear in the case of the EU State aid rules in Articles 107 to 109 TFEU and accompanying secondary legislation, which ultimately aim to avoid subsidy races, as well as the protectionist financing of national champions by Member States. Ultimately, these rules establish a set of controls over the selective channelling of public funds to companies, be it in the form of direct subsidies, or in more indirect ways such as tax exemptions, special contributions to pension plans, or the transmission of public assets (such as public land) in below-market conditions.

The European Commission has created a framework that allows Member States to use State aid for horizontal purposes (such as the support of environmental, innovation or employment-related activities), but also aims to prevent the use of public funds in order to benefit specific companies, in particular through a subsidisation of their operating costs. The European Commission enforces these rules and can bring Member States that breach them before the Court of Justice of the European Union. Additionally, competitors of the companies that receive State aid can challenge those decisions in their domestic courts.

Even if these rules are admittedly imperfect and their enforcement could be improved,* there is no question that the European Commission has been active and rather effective in combating the use of public funds to benefit specific large companies. Remarkably, Member States need to notify State aid measures to the European Commission and must not provide any aid until the Commission has authorised it. Overall, this means that in cases involving large companies, no State aid contrary to the EU rules is generally put in effect, as demonstrated by the discussions surrounding the Hinkley Point project. Where Member States infringe this standstill obligation, the Commission can force a recovery of the aid. The recent tax avoidance cases involving Starbucks or Fiat are a clear testimony of this important role in controlling the way public funds are spent in support of large companies.

The European Commission is thus heavily involved in the State aid measures aimed at specific large companies and acts as a filter to ensure that the expenditure of public funds pursues a legitimate objective in compliance with EU law. This was particularly the case of the State aid channelled to banks in the aftermath of the 2008 financial crisis.

Overall, then, at least for cases of State aid involving large sums of money and large companies, the Commission acts as an important filter to prevent damaging economic interventions in the economy, which constitutes an important check on how public money is spent. Whether such a tight system could be relaxed in order to enable a more proactive EU-wide industrial policy is a subject of significant debate, but the constraints that EU State aid rules currently impose on the provision of direct and indirect financial support to large companies are certainly not perceived as minor.

The question is thus whether a post-Brexit deal could free the UK Government from such State aid control, at least in the medium to long-run, so that it could engage in largely unchecked public subsidy policies, such as creating particularly beneficial tax conditions in order to try to retain or attract large multinational companies considering relocating elsewhere in the EU, or channelling public funds to chosen companies, either in support of industrial policy goals or otherwise.

These would be policy interventions clearly tackled by the European Commission under existing rules, and they would also be caught by the EFTA Surveillance Authority in case the post-Brexit deal resulted in the UK joining the European Economic Area (the so-called ‘Norwegian option’), which would require compliance with the same rules. However, whether interventions aimed at subsidising large companies would be caught in case of a ‘WTO-based’ trade scenario is less clear because the WTO rules on subsidies are not as tight as the EU’s, and their enforcement ultimately relies on other WTO Members bringing a complaint against the UK to the dispute settlement board, which is a very political decision ultimately reliant on trade calculations. To be sure, the EU itself could bring cases against the UK, but this would be a highly contentious issue in the framework of a relationship already very strained by the UK’s exit from the EU and detachment from the EEA.

Should the UK not be a part of the internal market via membership of the EU or the EEA, and in the absence of effective WTO-based external checks on the use of public funds to provide financial support to large companies, the control of this form of public expenditure would fall solely to Parliament and the domestic UK institutions, such as the National Audit Office.

This can be seen as an advantage by those convinced by arguments of self-control and UK-centric governance, but economic regulatory capture theory, and public policy theory more generally, have repeatedly demonstrated that such a self-policing architecture is unlikely to prevent ‘politicised’ uses of public funds. It seems clear to me that, in that case, the possibilities for any given Government to engage in expenditures of this type would be greater than they currently are, which would not necessarily result in the pursuance of the best interests of taxpayers in the UK.

Therefore, if there is value in having an external control of subsidies to large companies in order to avoid anti-economical protectionist policies or redistributive policies that take money away from other pressing social priorities—and I would certainly argue that there is—it seems clear to me that any post-Brexit deal that does not include the application of EU/EEA State aid rules would imply a net loss in terms of public governance and, in particular, in terms of an effective control of public expenditure, particularly regarding subsidies to large companies. Ultimately, then, from this perspective, it seems to me to be in the interest of taxpayers in the UK to strongly support a post-Brexit arrangement that retains State aid control, either by the European Commission or the EFTA Surveillance Authority.


* A Sanchez-Graells, “Digging itself out of the hole? A critical assessment of the Commission’s attempt to revitalise State aid enforcement after the crisis” (2016) 4(1) Journal of Antitrust Enforcement 157-187.

Another State aid decision by GC follows restrictive approach to standing of interested parties (T-118/13)

Following its previous restrictive case law on the granting of active standing to challenge State aid decisions to competitors of their beneficiaries (see here), the General Court (GC) of the Court of Justice of the European Union (CJEU) reiterated this position in its Judgment of 22 June 2016 in case Whirlpool Europe v Commission, T-118/13, EU:T:2016:365.

The case at hand is a long-lasting saga where producers of large household appliances (Electrolux and by Whirlpool) have been challenging France's restructuring aid to one of their competitors (Fagor France). In this iteration of the approval of the aid and its ensuing challenge, the Commission has adopted the strategy of challenging on of the competitors' standing. Whirlpool has opposed this approach on several basis, including the fact that its legal standing had not been challenged in the previous iteration of approval / challenge, that its market share is affected by keeping Fagor in the market, and due to Whirlpool's very close involvement in the case throughout.

The Commission dismisses all arguments. In the Commission's view,

the fact that an undertaking’s views were heard and that the conduct of the procedure was largely determined by its observations, although a factor which is relevant to the assessment of locus standi, does not relieve that undertaking of having to show that the aid at issue is liable to result in its market position being ‘substantially affected’. As regards that ‘substantial effect’, the Commission states that, in accordance with the case-law, it cannot suffice, in order to prove that the undertaking at issue is individually concerned, to establish that the aid at issue may exercise ‘an influence’ on the competitive relationships and that the undertaking concerned is in a competitive relationship with the addressee of the aid. On the contrary, it should be demonstrated that the applicant was particularly affected by the aid in relation to its competitors (T-118/13, para 28, emphasis added).

In short, the GC has accepted the Commission's arguments and, in particular, stressed that

44 Where an undertaking calls into question the merits of the decision appraising the aid ... the mere fact that it may be regarded as concerned within the meaning of Article 108(2) TFEU cannot suffice to render the action admissible. It must go on to demonstrate that it has a particular status within the meaning of the judgment of 15 July 1963 in Plaumann v Commission (25/62, EU:C:1963:17) ... That applies in particular where its market position is substantially affected by the aid to which the decision at issue relates (see, to that effect, judgment of 13 December 2005 in Commission v Aktionsgemeinschaft Recht und Eigentum, C‑78/03 P, EU:C:2005:761, paragraph 37 and the case-law cited).
45 In that regard, not only the undertaking in receipt of the aid but also the undertakings competing with it which have played an active role in the procedure initiated pursuant to Article 108(2) TFEU in respect of an individual aid have been recognised as individually concerned by the Commission decision closing that procedure, provided that their position on the market is substantially affected by the aid which is the subject of the contested decision. An undertaking cannot therefore rely solely on its status as a competitor of the undertaking in receipt of aid but must additionally show, in the light of its participation in the procedure and the magnitude of the harm to its position on the market, that its factual circumstances distinguish it in a similar way to the undertaking in receipt of the aid (see order of 7 March 2013 in UOP vCommission, T‑198/09, not published, EU:T:2013:105, paragraphs 25 and 26 and the case-law cited; see also, to that effect, judgment of 28 January 1986 in Cofaz and Others v Commission, 169/84, ECR, EU:C:1986:42, paragraph 25, and order of 27 May 2004 in Deutsche Post and DHL v Commission, T‑358/02, EU:T:2004:159, paragraphs 33 and 34).
46 As regards establishing such an effect, the Court of Justice has had occasion to explain that the mere fact that a measure such as the contested decision may have some influence on the competitive relationships existing on the relevant market and that the undertaking concerned was in a competitive relationship with the addressee of that measure cannot in any event suffice for that undertaking to be regarded as individually concerned by that measure (see, to that effect, judgments of 10 December 1969 in Eridania and Others v Commission, 10/68 and 18/68, EU:C:1969:66, paragraph 7, and 22 December 2008 in British Aggregates v Commission, C‑487/06 P, EU:C:2008:757, paragraph 47).
47 According to settled case-law, the applicant must provide evidence to establish the particularity of its competitive situation (order of 27 May 2004 in Deutsche Post and DHL v Commission, T‑358/02, EU:T:2004:159, paragraph 38, and judgment of 10 February 2009 in Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraphs 49 and 51) and demonstrate that its competitive position is substantially affected in comparison with the other undertakings competing in the market at issue (see, to that effect, order of 27 May 2004 in Deutsche Post and DHL v Commission, T‑358/02, EU:T:2004:159, paragraph 41; see also, to that effect, judgments of 10 February 2009 in Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraph 51; 13 September 2010 in TF1 v Commission, T‑193/06, EU:T:2010:389, paragraph 84; 15 January 2013 in Aiscat v Commission, T‑182/10, EU:T:2013:9, paragraph 68; 5 November 2014 in Vtesse Networks v Commission, T‑362/10, EU:T:2014:928, paragraph 55; and 3 December 2014 in Castelnou Energía v Commission, T‑57/11, EU:T:2014:1021, paragraphs 35 to 37) (T-118/13, paras 44 to 47, emphasis added).

Once more, the substantive analysis in which the GC engages in Whirlpool Europe v Commission results in a threshold of 'comparatively more adverse substantive negative competitive impact' that is almost impossible to discharge. This is bound to keep on restricting the number of State aid cases that can be successfully challenged, which will continue to contribute to a reduction in the effectiveness of the State aid control system [as criticised in A Sanchez-Graells, 'Digging itself out of the hole? A critical assessment of the Commission’s attempt to revitalise State aid enforcement after the crisis' (2016) 4(1) Journal of Antitrust Enforcement 157-187]. Thus, this Judgment must also receive criticism for its disproportionately restrictive assessment of the conditions to grant active standing to challenge State aid decisions under Art 263 TFEU.

RegioPost and its implications: personal notes of the full extended discussion at Bristol conference

Pictures by @PetraOden & @asanchezgraells

Pictures by @PetraOden & @asanchezgraells

Spending a whole day with friends and distinguished academics discussing the RegioPost judgment and its implications for the enforcement of labour law standards in public procurement settings under EU law proved to be both intellectually challenging and very rewarding. I am sincerely grateful to all speakers and participants for the excellent exchange of ideas (the event was definitely not short on controversy ...).

It will take us some time to get the formal publication of the proceedings of the conference in place, so I thought it would useful to advance here some of the most salient issues discussed at the event. These are my (lengthy) personal notes and, even if I tried to capture the essence of the speakers' presentations, they may not represent their views and they do not bind them in any way; all mistakes in these notes are my own.

Joanne Conaghan gave us a very warm welcome (literally, for it was a really hot day in Bristol after the gorgeous weekend weather) and stressed the relevance of putting the discussions to be held in the broader context of the constant struggle of harmonising economic and social considerations in any regulatory and policy framework.

With this in mind, the discussions for the day were organised around four panels, covering in turn different perspectives of the RegioPost judgment and, more generally, the difficulties and challenges in enforcing labour standards through public procurement. 

Panel 1: Constitutional and Internal Market Aspects of the Enforcement of Labour Standards in the EU

Constitutional view on RegioPost

Phil Syrpis' paper addressed the constitutional framework considering the dynamic relationship between secondary and primary EU law, and in particular how the posting of workers directive (PWD) impacts on provision of services in the EU. Two constitutional visions: a) all secondary legislation is under the Treaties and, consequently, secondary law cannot have any influence on the way the CJEU interprets primary law; b) more organic or heterarchical model, the CJEU’s monopoly on the interpretation of primary law is necessarily complemented by legislative action whereby the legislative institutions provide input on the context of the rights and, consequently, the CJEU should take this into account when interpreting rights under the Treaties.

Not clear how secondary law in conflict with previous case law of the CJEU should be dealt with, as both models are in tension. The case law of the CJEU is inconsistent and the Court has not clarified what is the right approach in cases of conflict between secondary law and previous case law interpreting primary law. As an aside – these issues raise important points for Brexit, particularly when some directives create individual rights, as well as the possibility for MS to shape social and citizenship rights and influence the case law of the CJEU.

Initial reaction to RegioPost is that it clarifies some issues, but it confuses others too. Essentially, the question was about compatibility with Art 56 TFEU for the regional government to insist on compliance with regional minimum wage. Usefully, the CJEU distinguished between exhaustive and non-exhaustive harmonisation. Where there is exhaustive harmonisation, the framework provided by the secondary legislation displaces that of the Treaty. From a constitutional perspective, that is the easy case. However, Art 26 Dir 2004/18 does not lay down exhaustive rules on contract performance and, consequently, the analysis of legality needs to be carried out under primary law—ie Art 56 TFEU. Moreover, Art 26 explicitly required for contract compliance clauses to be in compliance with Union Law.

However, in para 60 of RegioPost, despite saying they would assess under primary law, the Court actually assessed compliance with the PWD, which is a measure of non-exhaustive secondary legislation as well. In the light of the relationship between primary and secondary law, this approach is strange and complicates the issue of what the hierarchies are. The provision of regional law was determined to be compatible with EU law due to compliance with Art 3 PWD, rather that Art 56 TFEU, which is unusual and confusing.

Distinction between exhaustive and non-exhaustive harmonisation presupposes that most initiatives by the EU legislator will be of exhaustive harmonisation. However, exhaustive harmonisation is never the case, particularly in areas such as social or environmental law. This requires a more sophisticated approach to what to do in these areas in terms of compatibility with EU law.

PWD and Dir 2004/18 seem to be in a strange relationship because the legality under Dir 2004/18 ends up depending on the interpretation of the PWD—does this trigger a privileged position of the PWD over the procurement rules? RegioPost suggests that the CJEU’s controversial interpretation of the PWD is more important to the CJEU than the interpretation of the procurement rules. That is odd. What should the relationship between particular directive and the CJEU’s interpretation of the Treaty? ‘Provided they are compatible with the Treaties’ should imply some independent analysis of compatibility with Art 56 TFEU itself, almost regardless of compatibility with other rules of secondary EU law.

For the future: what is the likely impact for changes in the public procurement rules? Art 26 Dir 2004/18 vis-à-vis Arts 18(2) and 70 Dir 2014/24. Does the subtle wording in Dir 2014/24 alter the position of the procurement rules? Also, in view of the proposals for the revision of the PWD, there are queries as to the permeability of procurement rules to that reform. Finally, it is also unclear to what extent these changes in secondary law are likely to affect the CJEU’s case law and interpretation of primary law (will it reflect the organic model?).

Art 56 TFEU and the principle of proportionality

Piotr Bogdanowicz stressed that RegioPost has been described as the gold standard for the social protection of workers performing public contracts. However, there was no reference to proportionality in the judgment. Generally, there is a three step methodology in this area: restriction, justification and proportionality assessment. It is surprising that RegioPost does not engage in this three-step analysis because it had been used in Ruffert and Bundesdruckerei. The lack of this analysis is a shortcoming of this Judgment and, unfortunately, another case challenging the coherence of the case law of the CJEU.

Why should proportionality apply to RegioPost? Because Art 56 TFEU should apply, mainly due to the fact that procurement rules are a fundamental expression of free provision of services, as well as the fact that the fundamental freedoms are the legal basis for the adoption of the Directives. Furthermore, due to the lack of exhaustive harmonisation.

The analysis is odd because, should Art 26 Dir 2004/18 not have required compatibility with EU law, would this have been an issue? This is particularly relevant because Art 70 Dir 2014/24 does not require it. In Piotr's opinion, there is no need for this requirement to be written in secondary law because it derives simply from supremacy of EU law. And, in order to assess such compatibility, the three-step approach needs to be followed. Restrictions are easy to prove and the CJEU rarely rejects the public interest reasons raised by the Member States, so proportionality ends up being the key issue of analysis in almost all cases.

In procurement, proportionality is particularly relevant because it has been included from the early generations of public procurement directives. Proportionality has always been applied in the procurement setting as a general principle of EU law. It was applied in the case law anticipating RegioPost (Ruffert and Bundesdruckerei) and in both cases the CJEU found that national measures did not pass the proportionality test. The importance of proportionality is also reflected in Art 18(1) Dir 2014/24, where it is explicitly consolidated.

Why did the CJEU not apply proportionality? It is difficult to identify the reasons why, mainly because it derives from the construction of the judgment itself. In para 67, the CJEU stresses that interpretation of Art 26 Dir 2004/18 is supported by Art 56 TFEU, which is a strange argument. That is why the abrogation of the proportionality analysis is an odd analytical strategy. In Ruffert, the CJEU had gone the same way but did apply proportionality to assess compatibility with EU law (after the restriction was demonstrated under the PWD). Even if the CJEU wanted to distinguish RegioPost from Ruffert, it ought to have carried the proportionality analysis. As it is now, it seems that the PWD offers a safe haven for national law, which could have been done 8 years ago (in Ruffert), but not now because Art 26 Dir 2004/18 applied to the case and this should have triggered a different analysis.

This adds complication under Dir 2014/24 because of Art 70 + 18(1), which change wording and potentially the test. The case law of the CJEU shows that the more politically sensitive the case, the less intrusive the scrutiny seems to be. ‘Indulgent’ scrutiny was advocated for by AG Mengozzi because the measure is applicable to workers assigned to public contracts rather than workers in the private sector—this raises an issue of conflict between economic and non-economic goals (competition v social protection). This, however, should trigger a strict proportionality test.

The implications of lack of proportionality in RegioPost are severe. A recent report by Bruegel showed that there is 1.2 million posted workers in the EU, 40% come from Eastern countries (20% come from Poland). The CJEU seems to have adopted a more careful approach to the PWD and this may improve the situation for posted workers. This is something that requires further case law from the CJEU.


focused on complex issues around the way the preliminary reference ‘framed’ the CJEU’s assessment and why the CJEU ‘abrogated’ alternative analyses such as non-discrimination or split of competences. The discussion also covered the difficulty of applying proportionality in fields where we are dealing with incommensurate values. Moreover, this is complicated by the lack of clarity and consistency of the recitals in specifying what specific secondary law interventions aim to achieve—which is particularly muddy regarding Dir 2014/24.

Panel 2: Public Procurement and Labour Standards (I):
EU Procurement Law Perspective

Sustainable procurement and corporate experimentation

Nina Boeger shared some thoughts beyond RegioPost and an enquiry into two developments arising from the financial crisis: a) rise of polarisation of corporate governance models with, on the one hand, market responses to financial crisis that persevere on traditional approach to maximisation of return on investment and, on the other hand and as a counterbalance, new or revived corporate models, such as social enterprise, cooperative enterprise and commons-oriented enterprise; b) rise of sustainable or smart procurement driven by austerity politics and cuts on public budgets, which is also becoming polarised with, on the one extreme, traditional or economically-driven procurement, with focus on immediate value for money obtainable from service outputs and very much focussed on price and concrete delivery and, on the other extreme of the spectrum, smart / sustainable procurement that aims to change perspective of procurement and commissioning practices and moving towards longer-term perspectives on triple bottom line for the communities that are relevant for the procurement or commissioning body, as well as process sensitivity and long term relationships with providers of services.

Smart procurement involves a recognition of the risk that contracting with shareholder-driven corporations entails (large corporations are “too big to fail” and the risk of failure that the public sector faces requires a more diffuse provider base), as well as the passing on of costs to the tax payer. There is also a concern on the risks of insufficient specification (under-specification) and uncertainty, which is an integral part of procurement and commissioning. Corporate governance and the way they are structured will affect the way they deal with these issues of risk and how they will respond. Alignment of corporate governance and risk issues is necessary and contracting authorities are searching for trust and reliability in addition to capacity. New corporate models (such as social enterprises) that are struggling to make their business sustainable are dying to access public demand, both to raise funds for their sustainability and to raise social awareness.

In view of these trends, it looks like there are promising possibilities in procurement in aligning two very important structural developments: emergence of new corporate forms and the development of smart procurement. The question now is whether there is enough flexibility in the legal regime to allow public authorities to procure in this way? [this looks like a watershed moment for EU public procurement law, which is still fulfilling a traditional role of non-discrimination, transparency and workable market openness; but is also fulfilling the role of a change agent by supporting the development of those new models—the tension is not anymore on what the EU / the MS can do, but rather about thinking about new structures of capitalism].

The issue of flexibility then hinges on certain aspects of Dir 2014/24, such as market engagement (Art 40), the possibility of reserving non-profit contracts (Spezzino), grounds for exclusion also help some experimentation (Art 57), reservation of contracts for employee-led organisations (Art 77—which is incredibly narrow) … but the main problem for the development of ‘corporate governance’ clauses in procurement is the requirement of link to the subject matter (LtSM), which has been extended massively compared to Dir 2004/18. This poses the question how do we make sure that we can distinguish corporate governance requirements linked to the provision of the service from the limits on mandating general corporate policies. It is necessary to have these conversations so that the developments can move forward.

Minimum and living wage in public contracts:
Enforceability after RegioPost

Abby Semple started off stressing that the discussion on the enforcement of labour (pay) standards needs to be more specific and we need a taxonomy that distinguishes minimum wage, sectoral minimum wage [in particular, wages specific to public contracts], collectively agreed wage and living wage (which is a voluntary undertaking to pay a wage above the legally-mandated minimum wage). There is also an interesting argument on whether requirements for fair trade that link to wages paid to producers can be included as part of the same analysis. She also stressed that minimum wages show great disparity in the 28 Member States, even when adjusted by parity of purchasing power. This raises significant practical difficulties.

What does RegioPost say? The CJEU distinguished the Ruffert case very strongly because it was then a non-universal collective agreement that set the minimum wage in dispute, whereas in RegioPost the minimum wage is a result of (regional) law, in which case it does not matter that it only applies to the public sector. The CJEU also acknowledged the social protection argument as potential justification (like in Bundesdruckerei), but did not engage in full proportionality assessment (maybe because it did not need to do so, but leaves RegioPost vulnerable to future case law where the CJEU may engage in proportionality review).

Proportionality review should be a mechanism of last resort because Dir 2014/24 sets precise rules and, if they were not shielded from proportionality review, then contracting authorities would have no liberty to rely on the secondary rules.

Interestingly, the CJEU did not distinguish from Bundesdruckerei. The CJEU did not make any assessment / statement on the issue of lack of a cross-border situation in RegioPost. This is important in itself. It is also relevant that the CJEU engaged with an analysis with the PWD, primarily because the referring court had raised it. Also because Art 26 Dir 2004/18 is an empty / reenvoi clause that required engaging with PWD (this is cross-referenced in Recital 34 of Dir 2004/18C and Recital 98 of Dir 2014/24).

It is also relevant to stress that RegioPost makes some conceptualisations of contract compliance / performance clauses problematic. Nord pas de Calais, which reinterpreted Beentjes, also created a circular self-justifying doctrine that indicated that contract performance clauses did not allow for exclusion of those that did not meet the requirements. The RegioPost judgment now allows for exclusion of tenderers that are not committing to comply with contract performance requirements, which conflicts with the Commission’s understanding under the previous case law.

Looking specifically at minimum wage clauses, it is relevant to look at the Fair trade coffee case (C-368/10), where the CJEU stressed that you could have fair trade as an award criterion. Fair trade includes, amongst other things, the payment of wage premia to producers—this creates a problem if a contracting authority can do it in a third jurisdiction, but not in its own jurisdiction. There is evidence that the Commission does not accept that payment of a living wage could be a pass/fail criterion (Scottish government sent repeated letters to the Commission, to which it replied in the negative). How to address living wage as an award criterion? Could you use the wages the tenderer is willing to pay as an award criterion? Semple does not see any reason why it is not possible.

Scenarios for further thought. 1. Could you have a maximum wage clause in the contract for cost-control purposes? 2. Where a contract is affected by TUPE, the obligation to apply same terms and conditions, as the contracting authority, you need to effectively require pre-existing terms and conditions including wages (and you NEED to do that, including where the incumbent is the contracting authority itself in case of outsourcing).

RegioPost allows for legally-embedded sectorial [public sector] minimum wages. It does not deal specifically with living wages but the argument can be made in light of fair trade coffee.

Competition and State aid implications of
minimum wage clauses in EU public procurement

Albert Sanchez-Graells' presentation focused on the trade / social dumping rationale for minimum wage requirements, and I submitted that the PWD is the anti-dumping standard under EU economic law, so that no more demanding standard can be allowed without subjecting them to a very strict proportionality test. I also submitted that the approach of the CJEU to the interpretation of the PWD in the public procurement scenario does not make economic sense because it creates double standards for cross-border and inter-regional situations, which results in an impossible situation for contracting authorities that, at the time of tendering, cannot foresee whether they will manage a purely domestic or cross-border procurement project (as that depends on actual interest from cross-border tenderers).

On that basis, I discussed the problems of reverse discrimination and unforeseen consequences (anti-SME, pro-delocalisation, etc) that can derive from strict enforcement of the contradictory rules created by the Bundesdrukerei-RegioPost tandem. I moved on to stress the competition law implications that can result from this situation and the difficulties in capturing such anticompetitive effects with the competition law prohibitions in Arts 101 and 102 TFEU, included the supporting State action theory.

I finally tackled issues of application of State aid and submitted that the payment of above-market wages by the contracting authorities is an economic advantage that triggers the prohibition of Art 107 TFEU and, in turn, this cannot be justified either under the 2014 general block exemption regulation (it does not fit the regulated categories of disadvantaged/disabled employment, or creation of jobs), or the de minimis regulation (the aid is not transparent). Thus, I submitted that State aid enforcement in this area can raise difficult issues in the near future.


focused mainly on the issues of the social economy model and the extent to which competition law / State aid can apply to these issues at all, as well as the undesirability of a strict proportionality assessment that would kill innovation. It is also discussed to what extent a proportionality analysis is at all possible and whether social requirements are necessarily protectionist or not, and whether this should be dealt with as an issue of direct or indirect discrimination and how EU law would regulate them, which comes back to issues of proportionality of restrictions vis-à-vis their intended goals. The debate then also moved upwards and looked at the different treatment under WTO GPA and the EU rules, as well as the possibility to coordinate them. The issue of the heterogeneity of what is considered a ‘social consideration’ was also explored in some detail.

Panel 3: Public Procurement and Labour Standards (II):
EU Labour Law Perspective

Government as a socially-responsible market actor after RegioPost

ACL Davies took a step back and looked at the discussion from first principles. Use of contracts to pursue contract-unrelated policy is controversial. It is useful to distinguish two dimensions: a) reinforcing existing legal requirements, and b) creating requirements that go beyond existing legal duties. a) may see superfluous but the leverage of contracting can make up for deficiencies in enforcement elsewhere (in the UK case, given the lack of labour inspectorate). b) has attracted significant opposition + international drive to reduce the use of contracting authorities’ discretion is a way of reducing the chance for them to adopt protectionist strategies, willingly or otherwise.

The justification for the use of procurement to enforce social goals is as follows: 1) consistency and trust in government (legislative process not only aspect of democratic mandate given to government), so if the government is committed to a public policy, it should enforce it through contracting (to show consistency and avoid silo mentality, all of these foster trust in government). This is linked to the doctrine of legitimate expectations to get government to follow promises made in the past. This use has a long tradition (government as a model employer, and procurement as an aspect of that). 2) stronger argument in modern times is to prevent the harm that competitive tendering can create. This is linked to ILO standards for government contracting. This is particularly relevant in labour-intensive services, where government can create a level playing field in terms of employment conditions, so that alternative providers compete on other dimensions and labour standards are taken out of competition.

Two issues with RegioPost: lack of understanding of procurement and lack of understanding of wage clauses in procurement. Before the negatives, RegioPost is generally welcome if nothing else because it reverses Ruffert on the basis of Art 26 Dir 2004/18.


Art 56 TFEU and its key test for determining whether something is restrictive requires to assess whether it ‘constitutes an additional economic burden that may prohibit, impede or render less attractive the provision of their services in the host Member State’. This is applied in RegioPost and the minimum wage is characterised as an additional economic burden that makes the activities less attractive. The cost linked to the minimum wage can be a burden, but the cost can be passed on to the government and, from that perspective, there is no impediment or restriction to the provision in another Member State because the additional cost is covered by government. However, the services can be less advantageous by limiting the profit that the provider could achieve. This disadvantage is very artificial because it does not carry a loss for the service provider, but merely a missed opportunity for a larger benefit. This should not be covered by Art 56 TFEU or, at least, given that the disadvantage is very marginal. Maybe it is State aid (the government pays for it and State aid can be discussed), but it cannot also be exclusionary. It is either one or the other.

From the perspective of minimum wages, we need to remind ourselves that EU law does not say a great deal about wages. The difficulty of regulating pay at an EU-wide level is significant. This does not stop the CJEU from scrutinising different issues regarding pay in the internal market perspective, but the CJEU lacks a point of reference in this setting. This is what leads it to rely on the PWD and the importance on the wage being a minimum wage. The interpretation of the PWD in Laval is a clear attempt to elevate the PWD to the standard applicable in terms of social dumping in the EU. What’s wrong with the emphasis on minimum wage? The issue is with the justification on worker protection as the justification for the measure discussed in RegioPost. Minimum wage is of course a protection. The elephant in the room is that the minimum wage is only for public contracts, which diminishes the relevance of the argument as compared with contexts of general economic regulation.

In the public procurement setting, minimum wage is not a concern as in the labour market as a whole, but a protection against the competitive nature of government contracts and the ensuing downward pressure on worker protection. So the analysis should be procurement context-specific.

So, upshot, it is better that Ruffert, but RegioPost misses the difference between procurement and regulation, procurement being the space where the government is ready to put its money where its mouth is. This ought to do with preventing the harm that procurement might otherwise cause.

Labour law in the state of exception

Lisa Rodgers looked at labour law as an exception to economic rules and tried to see how far that helps us when thinking about labour standards in public procurement. It is worth stressing the link between economic and social law, and the theory of the ‘state of exception’ derived from neoliberal thought following the crisis. Labour law is an exception or concession against the basic economic foundation that we have. Following the crisis and austerity measures, this exceptionality has become more clear. The focus on procurement is an expression of this exceptionalism.

The state of exception is a theoretical perspective that uses the importance of the exception itself. The argument is that the ‘exception’ is the most important element in any system of regulation because it is at the edge of the legal system and it shows how politics get involved in the adjudication of legal issues. Used this way, it crowds out political considerations because the law starts covering it all.

Positive claims about inclusion of labour law in economic standards: 1) labour law’s concerns are included within the law; 2) labour law’s concerns become part of the economic paradigm; 3) law is subject to interpretation by favourable CJEU; 4) other Treaty provisions can help; 5) incorporation of international legal standards is facilitated.

Responses to the claims: 1) labour law is an exception or a concession, so it is included (by exclusion) as a discretion rather than a mandatory condition [eg Arts 56(1) and 57(4) Dir 2014/24] and some innovations do not include social/economic concerns [eg Art 68 Dir 2014/24]; 2) as an ‘exception’, labour law is only part of the economic paradigm, which weakens the political power of labour law institutions; 3) the CJEU has given broad discretion to consider employment and social considerations as part of the public procurement process (Beentjes), BUT the link to the subject-matter of the contract (LtSM) is an increasingly constraining economic test, which means that the contracting authority cannot influence the policy it applies; 4) relationship between labour standards, public procurement and other EU law provisions and policies (including soft law) show a shift in those instruments and a prevalence of macroeconomic policies that downgrades labour law standards’ relevance; 5) finally, there is a very limiting approach to the inclusion of international standards [narrow list included in Dir 2014/24 and NO reference to ILO convention 94 on public contracts) + Art 18(2)].

In conclusion, labour law is increasingly seen as important in the context of the public procurement directives, but there are some important risks if labour law is seen as an exception or a concession to economic law and economic goals. Labour standards are restricted because they need to comply with economic paradigms, there is a tendency to promote individual over collective rights, the institutions of labour law can find their power to challenge the law reduced.

Scope for collective bargaining in posting and procurement

Tonia Novitz looked at Laval and Ruffert and their impact on collective bargaining in public procurement, and positioned them in historical context (decided in 2007 and 2008), which has changed very significantly as a result of the economic crisis. Elektrobudwa and RegioPost indicate new horizons for trade union protection of posted workers and space for regional protection, but not for collective negotiation. This indicates stark shortcomings in proposals for the revision of the PWD.

Laval was a case of a trade union trying to improve conditions for a group of workers. This was generally seen as problematic. Collective negotiation and action could only be permitted in the context of social dumping. This is a very controversial judgment and has been heavily criticised. Laval had a negative impact on the ability of trade unions to call for action with cross-border effects. There is a strong clash between the EU approach and international standards, which cannot be easily reconciled.

Ruffert adopted the Laval approach whole-heartedly on the basis of another rigorous reading of the PWD and, in particular, about Art 3(8) setting out the ways in which collective agreements can be declared generally applicable. Ruffert failed to meet those. MS could exclude the construction industry from these requirements under Art 3(10) PWD, but that did not save the situation either. The imposition of such large limitations on the collective setting of rights for posted workers’ rights led to an understanding of the PWD as a ceiling rather than a floor of employment rights. This could not be overcome as a result of the Monti II Regulation, which was abandoned for different reasons.

It may be worth revisiting these issues by going back to the basics of the need for the PWD, particularly in view of the vulnerability of posted workers, which supports the case for their access to collective representation. It is also worth stressing that the capacity of labour inspectorates across the EU has been severely eroded as a result of the financial crisis. Additionally, memoranda and other crisis-related recommendations have been in line of flexibility and minimising collective negotiation to enterprise level-agreements. Overall, this makes the requirement of universal application stressed in Laval and Ruffert seems very outdated in the context of 2016.

Currently, the number of posted workers is at about 1.9 million and has increased by 49% compared to prior to the crisis. The treatment of posted workers is also reported to suffer from 51% lower wages, higher instances of work-related accidents and displacement issues. This can be triggering a mild shift of the case law regarding collective bargaining.

In Elektrobudwa, there was support for regional collective action on behalf of posted workers. This is linked to the issue of individual claims because in this specific case, there is direct representation by the trade union. There is nothing to indicate sympathy for collective bargaining for posted workers, but it could be seen as a first step in that direction. In RegioPost you can also identify a first step away from Ruffert, but they stress that there was no collective agreement and, therefore, the distinct treatment between public and private workers did not matter. However, none of these cases make significant inroads into collective bargaining in public procurement. RegioPost was very cautious in the way it treated collective agreements.

Commission’s proposals from February 2016 can potentially make a difference around collective agreements that could have a spillover effect on procurement. Of these proposals, the rules concerning subcontract chains, including those that foresee the posting of workers, can have significant effects. But, for this to be relevant, there has to be a move beyond ‘minimum rates of pay’ (absolute minimum) towards ‘remuneration’ (so as to cover a variety of different elements), which should then be equally applicable to all employees regardless of subcontracting chains. This has a connection with Art 71(6) Dir 2014/24 in terms of control of subcontracting chains. It is also a nod to international standards, but a timid one.

Overall, however, this seems very unlikely to be adopted in the near future, and it may well be only after a monitoring of the implementation of the PWD enforcement measures. However, a case could reach the CJEU on the basis of a case based on strike action including both posted and non-posted workers, which will trigger issues of evidence-based social dumping and may not allow the CJEU to set aside this type of collective action as easily as in the past.


focused on issues of localism and devolution, and the way that this can impact the use of procurement to enforce labour standards, particularly in the UK. It also explored the implications of the post-Lisbon social market economy on the issue of the ‘exceptionalism’ of labour law. The audience also questioned whether Laval is really not still the relevant standard and whether the restrictive approach to collective bargaining is not still the rule. There was also discussion on whether wage-based advantages deserve a different treatment where there are and where there are not posted workers, as well as what actually amounts as a restriction on the freedom of provisions of services in terms of two-tier employment scales and issues of discrimination / disadvantage of workers within one same undertaking.

Panel 4: Procurement and Labour Standards (III):
Domestic UK Perspective

Labour standards, domestic law and public procurement

Michael Ford focused on how non-pay labour standards (such as blacklisting [note: of workers involved in trade union activities, not of undertakings from public procurement procedures] or discrimination rules) can be applied in the public procurement setting. UK history shows that the underlying issues have been discussed repeatedly in the past. However, there have been significant developments, such as the use of procurement to promote non-discrimination law. Section 149 Equality Act 2010 (EA 2010) imposes an explicit duty on public authorities to prevent discrimination and give effect to advancing equality law. It is also important to stress the importance of sectoral collective bargaining and quasi-collective bargaining in the UK, which lead to legally binding pay rates and holiday terms. Currently, though, this has been discontinued and there is no legal mechanism for the determination of sectoral pay. Query is whether this will not change again in the future.

From a practical perspective, looking at procurement litigation, it is worth reminding that this is an area institutionally dominated by large commercial organisations with large budgets to litigate, which skews the litigation field. It is also affected by other pragmatic tensions, which raises issues of the different access to statutory review vs judicial review of procurement decisions. Pragmatically, it is also very difficult to apply standard legal tools (proportionality) to non-pay labour standards and, in particular, in issues of non-discrimination—not least due to the absolute protection of non-discrimination concerns over economic issues. Thus, even if it is fundamentally difficult, pay is relatively straightforward compared to non-pay labour standards.

The counterfactual question is, though, what would have happened in EU procurement law if there had been no PWD? Would the CJEU engaged in an untrammelled proportionality test based on ultimate considerations of discrimination? That would have been highly political and this is why the CJEU has sheltered behind the PWD, which has effectively provided the CJEU with a relatively easy answer compared to the alternative of having to legitimise its decisions. This is exemplified in Bundesdruckerei in that there was not AG Opinion to support the CJEU, which considered the use of the PWD straightforward.

However, let’s assume that the PWD is a useful point of departure. Can the government use 'administrative provisions' [which are not defined, under Art 3(1) PWD] to enforce pay policies, particularly reinvigorating sectoral rate setting mechanism? This raises a set of issues: would it still qualify as minimum wage of pay despite the existence of a lower universally applicable minimum wage? Yes, it would. Would it be possible to give legal effect to the result of the collective bargaining? Yes, it would because the rate, despite the collective bargaining origin, derives from law. Could you use a RegioPost type system where you already have a national minimum wage (ie creating a higher minimum wage only for public contracts)? It looks as if the CJEU considered it central that in RegioPost there was no minimum wage, but it is not clear why it would not be possible to go above that ceiling with a RegioPost-like mechanism.

An alternative to the legal enforcement of rates of pay, which is not in line with UK tradition, would be to rely on Art 70 Dir 2014/24 rather than on Art 3(1) PWD (ie base it on discretion of the contracting authority). However, this is still a confusing environment.

A connected issue is blacklisting, in relation with the ILO conventions, which is now illegal under UK law. However, this could be possible under Art 18(2), Art 57(3) and Art 70 Dir 2014/24 by focusing on the tender rather than the tenderer [as in the enactment of reg.56(2) PCR2015]. This should not be subjected to a proportionality analysis under Art 18(2) Dir 2014/24, which only requires proof of breach of any of the international standards.

Another issue is whether there is a tension between the public sector equality duty under the EA 2010 and public procurement. One difficulty is whether you can use procurement as a preference for undertakings that have equal opportunity policies, equal pay audits, etc. This would fit within Art 70 Dir 2014/24, complemented with the relevant ILO conventions. There is again a tension with the issue of minimum standards in the PWD or not.

The answer is that PWD, even when it applies, it does not give the answer—particularly outside the area of pay, it is very difficult to figure out how to effectively carry out a proportionality assessment when there are procurement measures that go beyond agreed (EU) standards.

Discretion and labour objectives

Richard Craven presented preliminary findings of on-going research funded by the British Academy and Leverhulme, where he assesses practitioners’ reactions to existing rules on the possibility to enforce labour standards within the limits of discretion given to public procurement officers.


focused on the difficulties of extrapolating qualitative insights and the insufficiency of public procurement data, which could allow for complementary quantitative research.

An extended discussion of RegioPost--teaser

On Monday 9 May 2016, and in celebration of Europe Day 2016, at the University of Bristol Law School, we are hosting the event "Public Procurement & Labour Standards–Reopening the Debate after RegioPost", where we will discuss in depth the implications of the Judgment of the Court of Justice of 17 November 2015 in RegioPost, C-115/14, EU:C:2015:760 (for my own views, see here and here).

We will be live tweeting from the event (follow #regiopostandbeyond), and I will post a summary of the discussions next week. For now, as a teaser, these are the slides that I will be using in my own presentation regarding the competition and State aid law implications of RegioPost.

CJEU rules on Greek Support to The Agricultural Sector under the 2008 and 2009 State Aid Frameworks: A Blow to the Commission's Waiver of Discretion? (C-431/14 P)

In its Judgment of 8 March 2016 in Greece v Commission (ELGA), C-431/14 P, EU:C:2016:145, the Court of Justice of the European Union (CJEU) ruled on the compatibility of certain measures of financial support to the Greek agricultural sector in the aftermath of the 2008 financial crisis with the EU rules on State aid--ie mainly, Art 107 TFEU and the Temporary Community Framework for State aid measures adopted by the Commission in 2008 (the 2008 TCF), as amended in 2009 (the 2009 amended TCF).

The Judgment is interesting because it assesses the boundaries of the temporary discretionary measures adopted by the Commission in order to flexibilise the enforcement of EU rules in times of economic and financial distress, on the basis that they aim 'to remedy a serious disturbance in the economy of a Member State', ex Art 107(3)(b) TFEU. In particular, the ELGA Judgment assesses whether Member States can validly raise arguments based on Art 107(3)(b) TFEU directly, regardless of the Commission's delineation of its State aid policy based on that same legal basis. Or, in simple terms, whether a valid Art 107(3)(b) TFEU can exist outside of the (temporary) scope of the 2008 TCF and the 2009 amended TCF. The case may seem very specific because of its link to the economic crisis. However, the CJEU makes some broader points about the Commission's discretion that are worth taking into careful consideration.

This discussion is relevant from a legal perspective, due to the clarification of the so far unknown exemption of the State aid prohibition of Art 107(1) TFEU on the basis of Art 107(3)(b) TFEU regarding aid aimed to remedy a serious disturbance in the economy of a Member State' [see P Nicolaides & IE Rusu, 'The Financial Crisis and State Aid' (2010) 55(4) The Antitrust Bulletin 759-782]. It is also relevant for the policy implications of the CJEU's support for the Commission's intervention [for discussion of a general framework, see H Kassim & B Lyons, 'The New Political Economy of EU State Aid Policy' (2013) 13(1) Journal of Industry, Competition and Trade 1-21; and TJ Doleys, 'Managing the Dilemma of Discretion: The European Commission and the Development of EU State Aid Policy' (2013) 13(1) Journal of Industry, Competition and Trade 23-38].

The case of the Greek support to the agricultural sector through ELGA

The specific case concerns a long-running action of the Greek State for the annulment of a 2011 Commission Decision concerning compensation payments made by the Greek Agricultural Insurance Organisation (ELGA) in 2008 and 2009, which the General Court (GC) upheld on appeal (T‑52/12, EU:T:2014:677). One of the difficulties with this case is the sequence of events. From the regulatory perspective, it is worth stressing that the 2008 TCF, which entered into force in 17 December 2008, did not cover aid to the agricultural sector. This was eventually made clear in the 2009 amended TCF, according to which

The possibility under [the TCF] to grant a compatible limited amount of aid does not apply to undertakings active in the primary production of agricultural products. Farmers, however, encounter increased difficulties to obtain credit as a consequence of the financial crisis ... it is appropriate to introduce a separate compatible limited amount of aid for undertakings active in the primary production of agricultural products.

Specifically, the 2009 amended TCF provided that

The Commission will consider such State aid compatible with the common market on the basis of Article [107(3)(b) TFEU], provided all the following conditions are met: ... (h) … Where the aid is granted to undertakings active in the primary production of agricultural products ..., the cash grant (or gross grant equivalent) does not exceed EUR 15,000 per undertaking ...

This took effect on 28 October 2009, which raises a practical temporary difficulty because, '[f]ollowing protests in January 2009 by a large number of Greek agricultural producers about the losses suffered by them in 2008 as a result of adverse weather conditions..., the Hellenic Republic provided that compensation aid of EUR 425 million would be paid to producers on an exceptional basis by ELGA' (C-431/14 P, para 11). Upon investigation, the Commission found that most of that aid was incompatible with the internal market and, in particular, that '[t]he compensation aid of EUR [387.4 million] granted to producers on dates before 28 October 2009 is incompatible with the internal market' (C-431/14 P, para 14, emphasis added).

The issue is that, in plain terms, the Commission rejected Greece's claims that the exemption foreseen in Art 107(3)(b) TFEU could be directly applied in the case because of the economic difficulties that Greece had been experiencing. The Commission rejected such claim on the basis that Art 107(3)(b) TFEU had to be applied within the boundaries of the policy documents developed to that effect, ie the 2008 TCF and the 2009 amended TCF, which could only apply for the future--that is, only from their respective dates of entry into force--which, as the agricultural sector is concerned, was that of the 2009 amended TCF: 28 October 2009. The GC upheld the Commission's approach in the following terms

185 ... it is clear that, contrary to what the Hellenic Republic claims, the Commission had to base its decision on the [TCF] and not directly apply Article 107(3)(b) TFEU in order to assess the compatibility of the payments made by ELGA in 2009 on account of the economic crisis experienced in Greece.
186 It is clear from the case-law that, in adopting rules of conduct and announcing by publishing them that they will henceforth apply to the cases to which they relate, the Commission imposes a limit on the exercise of its aforementioned discretion and cannot depart from those rules without being found, where appropriate, to be in breach of general principles of law, such as equal treatment or the protection of legitimate expectations (see judgment[s] in Germany and Others v Kronofrance, [C‑75/05 P and C‑80/05 P, EU:C:2008:482], paragraph 60 and the case-law cited, and … Holland Malt v Commission, C‑464/09 P, [EU:C:2010:733], paragraph 46).
187 ... in the specific area of State aid, the Commission is bound by the guidelines and notices that it issues, to the extent that they do not depart from the rules in the Treaty (see judgment in Holland Malt v Commission, [C‑464/09 P, EU:C:2010:733], paragraph 47 and the case-law cited).
188 Therefore, it is necessary to reject the arguments of the Hellenic Republic to the effect that, on account of the serious disturbance in the Greek economy due to the economic crisis experienced in Greece since the end of 2008 and in 2009, the Commission should have declared the payments made by ELGA in 2009 compatible directly on the basis of Article 107(3)(b) TFEU (T-52/12, paras 185-188, emphasis added).

The CJEU has now taken the same line of argument, but has introduced important nuances in determining that

69 ... as the General Court stated in paragraphs 186 and 187 of the judgment under appeal, the Court has also consistently held that, in adopting rules of conduct and announcing by publishing them that they will henceforth apply to the cases to which they relate, the Commission imposes a limit on the exercise of its aforementioned discretion and, in principle, cannot depart from those rules without being found, where appropriate, to be in breach of general principles of law, such as equal treatment or the protection of legitimate expectations (judgments in Holland Malt v Commission, C‑464/09 P, EU:C:2010:733, paragraph 46, and Banco Privado Português and Massa Insolvente do Banco Privado Português, C‑667/13, EU:C:2015:151, paragraph 69).
70 However, in the specific area of State aid, the Commission is bound by the guidelines that it issues, to the extent that they do not depart from the rules in the TFEU, including, in particular, Article 107(3)(b) TFEU (see, to that effect, judgment in Holland Malt v Commission, C‑464/09 P, EU:C:2010:733, paragraph 47), and to the extent that their application is not in breach of general principles of law, such as equal treatment, in particular where exceptional circumstances, different from those envisaged in those guidelines, distinguish a given sector of the economy of a Member State.
71      Consequently, first, the Commission may not fail to have regard to Article 107(3) TFEU by adopting guidelines vitiated by an error of law or a manifest error of assessment, nor may it waive, by the adoption of guidelines, the exercise of the discretion that that provision confers on it. Further, when, in the exercise of that discretion, it adopts guidelines of that nature, these must be kept under continuous review for the purposes of anticipating any major developments not covered by those measures.
72      Secondly, the adoption of such guidelines does not relieve the Commission of its obligation to examine the specific exceptional circumstances relied on by a Member State, in a particular case, for the purpose of requesting the direct application of Article 107(3)(b) TFEU, and to provide reasons for its refusal to grant such a request, should the case arise.
73      In the present case, it is not in dispute that, precisely because of the effect of the economic crisis experienced by the Member States, and in particular, the Hellenic Republic, on the primary agricultural sector of the European Union, the Commission exercised the discretion conferred on it by Article 107(3)(b) TFEU by adopting the TCF and then the amended TCF, since both the former and the latter expressly mention that sector.
74      However, the fact remains that although the Hellenic Republic claimed before the General Court that Article 107(3)(b) TFEU ought to be applied directly to the facts of the case, notwithstanding the existence of the rules of conduct set out in the TCF and the amended TCF, it did not argue, in support of that claim, that there were, in the present case, specific exceptional circumstances in the primary agricultural sector concerned ...
75      Indeed, it is apparent from the documents in the file that the material that the Hellenic Republic put before the General Court was intended to establish the existence of a very serious disturbance affecting the Greek economy from the end of 2008 and in 2009, but it was not such as to prove to the requisite legal standard that that economy was faced with specific exceptional circumstances that ought, in this case, to have led the Commission to assess the aid at issue directly in the light of Article 107(3)(b) TFEU (C-431/14 P, paras 69-75, emphasis added).

implications of the cjeu elga judgement

In my view, the implications of the case are two-fold, and they concern, first, the relationship between the Commission's disclosed State aid policy and the discretion that Art 107(3) TFEU gives it; and, second, the interpretation of Art 107(3)(b) TFEU in particular.

Regarding the issue of the extent to which the Commission can deviate from adopted and publicised State aid policy, the CJEU has now made it clear that 'in adopting rules of conduct and announcing by publishing them that they will henceforth apply to the cases to which they relate, the Commission imposes a limit on the exercise of its aforementioned discretion and, in principle, cannot depart from those rules without being found, where appropriate, to be in breach of general principles of law, such as equal treatment or the protection of legitimate expectations' (para 69, emphasis added); and that 'the Commission is bound by the guidelines that it issues, to the extent that they do not depart from the rules in the TFEU ... and to the extent that their application is not in breach of general principles of law, such as equal treatment, in particular where exceptional circumstances, different from those envisaged in those guidelines, distinguish a given sector of the economy of a Member State' (para 70, emphasis added). It is thus plain that 'the Commission may not fail to have regard to Article 107(3) TFEU ... nor may it waive, by the adoption of guidelines, the exercise of the discretion that that provision confers on it' (para 71, emphasis added).

Somehow, the CJEU has made it clear that the Commission cannot hide behind its disclosed State aid policy if there are relevant circumstances that require a specific discretionary decision. This can be far reaching because the CJEU ELGA Judgment clearly opens the door to Member States' claims beyond the boundaries set by the Commission in its disclosed State aid policy, and may be the end of an era of increasing push for box-ticking exercises and for the Commission's reliance on its predetermined conditions for State aid exemption under block exemption regulations. This may well lead to an increase in litigation by Member States, which may be more willing to challenge the Commission's 'self-enforcement' approach in its recently adopted State aid 2.0 strategy [for discussion, see A Sanchez-Graells, “Digging itself out of the hole? A critical assessment of the Commission’s attempt to revitalise State aid enforcement after the crisis” (2016) Journal of Antitrust Enforcement, forthcoming].

The bit that puzzles me is that, in the specific circumstances of Art 107(3)(b) TFEU and its use in the aftermath of the economic and financial crisis, the Commission had not disclosed any policy documents prior to the 2008 TCF and the 2009 amended TCF. Thus, the issue whether the Commission could block any claims prior to the entry into force of those instruments could also have triggered an argument of retroactive application of beneficial discretionary measures, which I would have expected to read in a case like this. Somehow, the issue of the inter-temporal validity of policy and legal instruments in EU economic law continues to raise unresolved issues.

Regarding the specific interpretation of Art 107(3)(b) TFEU, the implications of the ELGA Judgment are mixed. On the one hand, it seems clear that the CJEU recognises that Member States can claim the existence of specific circumstances in its economy, and this would tail up with the drafting of Art 107(3)(b) TFEU, which indicates that the exemption is available for aid aimed to remedy a serious disturbance in the economy of a Member State. On the other hand, though, the CJEU seems to require Member States to demonstrate that those circumstances 'distinguish a given sector of the economy of a Member State' (para 70) and, in the specific case, 'specific exceptional circumstances in the primary agricultural sector concerned' (para 74). This seems problematic on two fronts.

First, it clearly goes beyond the wording of Art 107(3)(b) TFEU, which has no reference to specific sectors of the economy and seems to accept the possibility of exceptional rules aimed at a distressed economy as a whole. One is left with the doubt whether this requirement to have demonstrated specific exceptional circumstances in the agricultural sector derives from the CJEU's unwillingness to quash the Commission's decision--reading the case, it seems clear that the controversy about the existence of sufficient evidence in the file could have been a driver for this outcome--or, on the contrary, it is a purposeful interpretation of Art 107(3)(b) TFEU in a way that reduces its scope. If the latter is the real reason, then the CJEU could have been more explicit in determining the parameters of such narrow interpreation, not least because of the absence of a sufficient volume of case law that interprets this provision.

And, second, it seems to create a significant limitation in the Member States' design of their macroeconomic (emergency) policies in a way that some could argue falls foul of the principle of subsidiarity. In that regard, the CJEU could have been more explicit as to the reasons for the imposition of a requirement of economic intervention in the specific sectors affected by the serious economic disturbance--which, in my view, would be relatively easy to support on the basis of the general requirements of suitability and proportionality applicable to State measures that aim to benefit from exemptions of Treaty prohibitions under EU economic law, generally.

State aid in rescue of firms in difficulty, merger control and patent litigation (T-79/14): quite a mix

In its Judgment of 1 March 2016, Secop v Commission, T-79/14, EU:T:2016:118, the General Court (GC) has ruled on the procedural rights of interested parties in a State aid case (for discussion of related case law in this area, see here). The Secop Judgment is interesting because it includes some analysis of the similarities and differences of the rights of interested (third) parties for the purposes of, on the one hand, State aid control (Arts 107-108 TFEU and Reg 2015/1589 and its predecessor Reg 659/1999) and, on the other, merger control (Reg 139/2004) under EU law.

The analysis in the Secop case is complicated by two elements. First, by the fact that the State aid was given under the guidelines on rescue and restructuring aid (in their 2004 version) and, because parts of the restructuring plan implied the acquisition of assets of the financially distressed group (ACC) by a competitor (Secop), this required merger control clearance from the European Commission. Second, the analysis is complicated by the subsequent emergence of a patent litigation between the two industrial conglomerates involved in both State aid and merger procedures (ie between the 'surviving' parts of the distressed ACC group and Secop as the acquirer of some of its assets), which have an open dispute as to whether a valid licence agreement for the use of proprietary patented technology was entered into as part of the rescue plan. This dispute has led to two sets of proceedings concerning those patents, respectively before the German and Italian courts. It is interesting to look at the case and the GC's reasoning.

background of the case

The case concerned two industrial conglomerates: ACC and Secop. ACC was an industrial conglomerate with an Italian holding company and a number of subsidiaries at different levels. For the purposes of the case, it is only necessary to note that HCH was the holding company of the group, ACC Compressors was the operating subsidiary of first level, and ACC Austria was an operating subsidiary of second level. Following financial difficulties within the ACC group, all its subsidiaries and the holding company itself were eventually declared insolvent. As the GC summarises,  'following a call for tenders launched in the context of ACC Austria’s insolvency proceedings, a purchase agreement for the assets of ACC Austria was signed between [Secop] ... and ACC Austria’s insolvency administrators. That contract was made subject to the suspensive condition of a declaration by the European Commission that the transaction was compatible with the internal market' (para 3).

In order to cover the liquidity needs of the ACC group and to allow it to continue its activities pending the preparation of a restructuring or liquidation plan, Italy gave ACC Compressors (the parent company ACC Austria) a State guarantee of 6 months for credit lines in support of liquidity needs of a total amount of EUR 13.6 million. Subsequently, the European Commission decided not to raise objections to the acquisition of ACC Austria’s assets by Secop (see Case No COMP/M.6996 - Secop/ ACC Austria, the ‘merger decision’), thereby validating the contract between Secop and ACC Austria's insolvency administrators. Shortly afterwards, the Commission also decided not to raise objections to the State aid given by Italy to ACC Compressors (see Case No COMP/SA.37640 - Rescue aid for ACC Compressors S.p.A. - Italy, the 'contested State aid decision').

What I find interesting in the case is that the challenger of the State aid (Secop) is the beneficiary of the asset disposal under the merger procedure, which was in turn opposed by ACC Compressors as the parent company of the 'traded subsidiary' under insolvency administration (ACC Austria). Thus, Secop and ACC, as industrial conglomerates, hold opposite interests in the merger and the State aid cases.

It would seem that, by aiming to enforce the exclusive rights deriving from the patents acquired together with ACC Austria's assets against the former parent company (ACC Compressors), as well as challenging the State aid given by the Italian Republic to that same company, Secop is clearly engaging in an all-out strategy to eliminate a competitor at at time when it faces financial difficulties (which would nullify the Italian intervention to rescue it). Conversely, it could also seem that by selling assets linked to specific patents and claiming to have retained a right of use of the patents (through the entering of a valid licence agreement, or otherwise), and at the same time receiving State aid from Italy, ACC could be trying to obtain dual support in times of financial difficulty--ultimately at the expense of a competitor (Secop) that acquired assets at a time of distress. These issues and considerations are not particularly clear in the Secop Judgment, but my intuition is that they influenced the outcome of the case.

In particular, the GC's Secop Judgment refers to the action by Secop seeking the annulment of the State aid received by ACC Compressors after the transfer of ACC Austria's assets took place. For the purposes of our discussion, the two main arguments submitted by Secop are that: 1) the European Commission should have taken into account that, following the transfer of ACC Austria's assets, ACC Compressors would not be legally entitled to keep on using certain patents now held by Secop, which would prevent ACC from carrying on with its industrial activity and, ultimately, infringe the 2004 guidelines for rescue and restructuring aid; and 2) that it is discriminatory for ACC Compressors to have been able to oppose the acquisition of ACC Austria's assets by Secop in the framework of the merger control procedure (where ACC Compressors was recognised as an interested party), whereas Secop has been denied the equivalent possibility in the State aid case because the Commission decided not to open a formal investigation. The discussion focuses on each of these arguments in turn. 

Arguments regarding the use of patents

On the substance of the dispute, primarily, Secop contends that 'following the disposal of ACC Austria’s assets, the patents at issue can no longer be used by ACC Compressors, which must, therefore, be considered to be a firm emerging from the liquidation of an existing firm and, consequently, a newly created firm ... failing the ability to use the disputed patents, ACC Compressors does not have sufficiently developed structures to be eligible for rescue aid' (para 30). This argument concerns point 12 of the 2004 guidelines for rescue and restructuring aid, which indicated that 'a newly created firm is not eligible for rescue or restructuring aid even if its initial financial position is insecure. This is the case, for instance, where a new firm emerges from the liquidation of a previous firm or merely takes over that undertaking’s assets. A firm is in principle considered to be newly created for the first three years following the start of operations in the relevant field of activity. Only after that period will it become eligible for rescue or restructuring aid …’. The GC dismisses this argument on the following grounds:

35 First, ACC Compressors and ACC Austria were initially part of one and the same undertaking in that the two companies produced the same products, on two different sites, but under the same economic management. Upon the transfer of ACC Austria’s earning assets ... it is true that the volume of activity of this firm had been reduced, since the activities corresponding to the production site located in Austria no longer formed part of it. Thus, the undertaking to which the contested aid ... was granted comprised only ACC Compressors’ earning assets. Nevertheless, ACC Compressors managed the undertaking concerned, both before and after the transfer, and ... it carried on ... albeit in a reduced fashion, the production and marketing of compressors, which was the traditional activity of that undertaking. Therefore, contrary to the applicant’s claims, it was the same undertaking as that which had been making compressors since 1960.
36 Second, ... in the situation in which the assets are transferred, it is not the entity formed of the economic activities retained by the transferor company that is relevant, for the purpose of the classification ‘newly created firm’ but the entity made up of the economic activities of the transferee company, within which the transferred assets were integrated. It is also normal and reasonable for a firm in difficulty to dispose of certain assets and focus its activity on its core business, whether from a geographical or sectoral perspective, in order to improve the chances of economic recovery. Point 39 of the Guidelines thus expressly envisages the divestment of assets as a means of preventing undue distortions of competition, in the context of the examination of a restructuring plan for the purpose of granting restructuring aid. It would be contrary to the overall purpose of the Guidelines for such a sale of assets to lead systematically to the exclusion of the transferring company from the benefit of rescue aid.
37 The fact that a legal dispute over the ... patents is under way between ACC Compressors and [Secop] cannot lead to a different assessment.
38 Indeed, at the time the contested [State aid] decision was adopted, the Commission could take into account only the factual and legal situation of ACC Compressors as it was at the date of that adoption; at the most, it had to take into account the foreseeable evolution of that situation, for the period for which rescue aid was granted, namely, six months ... However ... at the date of the adoption of the contested [State aid] decision, ACC Compressors was still using the disputed patents to manufacture compressors ... and there was nothing to indicate that this situation could have changed in the six following months.
39 In addition, the existence of the patent dispute was not relevant for the purposes of assessing the compatibility of the contested aid with the internal market. It is true that, had [Secop] won the case in the patent dispute, it would have been conceivable that ACC Compressors could no longer have used the disputed patents and would, accordingly, have had to cease production of a significant range of compressors ... However, this also depended on the question of whether, after a possible defeat in the courts, ACC Compressors could obtain a user license for those patents. Moreover, it could not be ruled out from the outset that it could offset the possible disposal of its activity producing ... compressors against the development of other lines or activities. In any event, it must be considered that it was not for the Commission to anticipate the outcome of the patent dispute, pending before the national courts at the date of adoption of the contested decision, by substituting its assessment for that of the competent courts, seized of that dispute.
40 Finally, it is appropriate to reject the applicant’s argument ... that the Commission ought to have taken into account that, in the context of the merger procedure, ACC Compressors itself had indicated that, if [Secop] were to purchase the assets of ACC Austria, it could not pursue its production of compressors, since it would not then be able to use the disputed patents any longer.
41  In the merger decision, the Commission considered ACC Compressors’ claims and found that, given, in particular, the patent dispute between the two parties, it was not inconceivable that an agreement on a licence should be concluded between them. The Commission had therefore already found, in the merger proceedings, that ACC Compressors’ claims that it could not pursue the production of compressors when there was no licence for the disputed patents were hypothetical (T-79/14, paras 35-41, emphasis added).

I find the second part of the GC's position difficult to share. In particular, I struggle to understand why the Commission did not require the granting of a sufficient licence as a condition for the clearance of the merger. This would have avoided all issues leading to the existing patent litigation and, in the specific circumstances of the State aid case, it would have also allowed for the rescue and restructuring plan to avoid a major risk of discontinuation of industrial activity by the beneficiary of the aid, which would have seemed desirable.

It is clear that the GC cannot review or alter the merger decision when reviewing the contested State aid decision, but it seems strange that it shows such deference to the Commission's argumentation in the merger decision, which is very weak. Indeed, the Commission's considerations (as presented by the GC in para 40 and 41) are equally hypothetical and rather counterintuitive--why would the companies reach a licence agreement now, when they could have included it in the negotiations leading up to the contract for the purchase of the assets? Were there any impediments for ACC Compressors to obtain that licence via the insolvency administrators of its subsidiary ACC Austria.

Somehow, it seems that the Commission was cutting corners in its analysis during the merger control procedure, particularly by failing to impose a behavioural remedy that could certainly have dispelled uncertainties in the market prognosis. Then, it seems once again too lenient for the GC to allow the Commission to also cut corners in the State aid case by refusing to open a formal investigation, where it would have had to take Secop's arguments into consideration and dispose of them in a more robust manner. 

Arguments regarding the asymmetrical access by interested parties to merger and State aid procedures

On the procedural side of the dispute, in short, Secop submits that 'it has not had the opportunity to present its views in the State aid procedure, initiated for the benefit of ACC Compressors, in order to oppose the grant of the contested aid to the latter ... On the other hand, ACC Compressors has had the opportunity, as part of the merger procedure, to oppose the takeover of ACC Austria’s assets by [Secop]. In its view, it is a violation of the principle of equal treatment, since the competitive relationship between the ACC group and the Secop group ought to have been assessed in both procedures' (para 61). The GC also dismisses this argument, following this reasoning:

62 ... the principle of equal treatment, as a general principle of EU law, requires comparable situations not to be treated differently and different situations not to be treated in the same way, unless such treatment is objectively justified ...
63 ... both in the context of a State aid procedure and in a merger procedure, the competitors of the firms at issue have no right to be automatically associated with the procedure, and this is particularly so in the context of the initial phase of the procedure, in the course of which the Commission makes a preliminary assessment of either the aid at issue, or the notified merger.
64 Indeed, first, as far as concerns State aid ... It is only in connection with the [the actual investigation stage referred to by Article 108(2)], which is designed to allow the Commission to be fully informed of all the facts of the case, that the FEU Treaty imposes an obligation, for the Commission, to give interested parties notice to submit their comments ... It follows that interested parties, other than the Member State concerned, including competitors of the aid recipient, such as the applicant in the present case, have no right to be associated with the procedure in the preliminary examination stage.
65 Secondly, as regards mergers, ... the Commission may hear — on its own motion — natural or legal persons other than the notifiers and other parties to the proposed merger, but it is obliged to do so only on the two conditions that those persons have a sufficient interest and that they make such a request ...
66 ... ACC Compressors’ position in the merger procedure was not only that of a competitor of [Secop], the undertaking notifying the merger, but also one of an ‘interested party’ ... in that, as ACC Austria’s parent company, all assets of which were to be sold, it had to be assimilated to the vendor of those assets and, therefore, had the status of party to the proposed merger. However, unlike its competitors ... interested parties have the right to express their view at all stages of the procedure, including the preliminary phase ...
67 It must therefore be stated that the situation of the applicant, under the State aid procedure that led to the contested decision, is different from that of ACC Compressors under the merger procedure that led to the decision on the merger, in that ACC Compressors had a right to be heard before the adoption of that latter decision. Consequently, the fact that the Commission did not, before adopting the contested decision, give the applicant the opportunity to state its point of view does not infringe the principle of equal treatment (T-79/14, paras 62-67, emphasis added and references to further case law have been omitted).

I find this analysis too formalistic and, in my view, the GC has ultimately failed to engaged with the argument on discrimination at a substantive level. The recognition of specific rights to interested parties in merger proceedings is not a useful comparator in this case. Rather, the GC could (should) have focused on the different access to the Commission given to competitors in merger cases and in State aid cases, particularly at the initial stage of proceedings, and assessed from a functional perspective whether that difference makes sense (ie is justified and proportionate). In my view, it is not. 

More importantly, the Secop Judgment moves in the same direction as a line of case law where the GC is making it increasingly difficult for competitors to challenge State aid decisions. This is very counter-productive for the consolidation of a State aid 2.0 control system, where the Commission needs to increasingly rely on market intelligence provided by third parties and market complaints raised by competitors. This line of case law will, ultimately, consolidate the ineffectiveness of the EU State aid rules [as discussed in detail in A Sanchez-Graells, “Digging itself out of the hole? A critical assessment of the Commission’s attempt to revitalise State aid enforcement after the crisis” (2016) Journal of Antitrust Enforcement, forthcoming]. This is an undesirable development of EU economic law in this area. 


CJEU consolidates push for overcompliance with EU public procurement rules in the provision of public services (C-446/14)

In its Judgment of 18 February 2016 in Germany v Commission (Zweckverband Tierkörperbeseitigung), C-446/14 P, EU:C:2016:97 (only in German and French), the Court of Justice of the European Union (CJEU) has supported the approach of the General Court (GC) in the assessment of the Altmark (C-280/00, EU:C:2003:415) conditions for the analysis of State aid regarding a system of financial support for a service of general economic interest (SGEI) consisting in the maintenance of reserve animal disposal capacity in the case of epizootic in a public abattoir in the state of Rhineland-Palatinate in Germany.

This appeal was against the GC's Judgment in T-295/12 (EU:T:2014:675, which is discussed by P Nicolaides here), but the analysis of the CJEU was highly relevant for the pending appeal against the GC Judgment in T-309/12 (EU:T:2014:676, discussed here), which has now been abandoned by the appellant (the abattoir, now in liquidation). The intricacies of the case are quite complex, and points of detail are too specific to discuss now. However, there are some general issues to note in view of the CJEU's Germany v Commission (Zweckverband Tierkörperbeseitigung) Judgment. 

From the outset, it must be stressed that the CJEU is following the GC in a trend that may well be modifying the scope of the Altmark test in a way that pushes for overcompliance with the EU public procurement rules as the only effective way in which Member States can achieve legal certainty in the way they organise their SGEIs. This requires to take a long view on some of the arguments in the case.

The CJEU has generally been very clear that 'the four conditions set out in Altmark ... are distinct from one another, each pursuing its own finality' (para 31, own translation from French). In particular, it stressed that the first condition requires 'that the recipient undertaking must actually be required to discharge public service obligations and those obligations must be clearly defined for such compensation to escape the classification as State aid' (para 26, own translation from French), while the fourth condition determines that 'when the choice of the undertaking which is to discharge public service obligations, in a specific case, is not in the framework of a public procurement procedure, the level of compensation needed must be determined based on an analysis of the costs which a typical, well run, undertaking would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging those obligations' (para 29, own translation from French). In that regard, these would seem to require separate, independent assessments of each of the Altmark conditions.

In contrast, in the challenged decision, the GC had determined that
as part of the review of the question whether the fourth Altmark criterion ... is satisfied, there is certainly room to take into consideration the nature of the service in question and the circumstances of the case, and it is therefore possible that this criterion, which requires a comparison of the costs and revenues directly related to the provision of the SGEI, can not be applied strictly to the present case (see, that effect, case BUPA ea / Commission (T-289/03, EU:T:2008:29) paragraph 246). Indeed, the Court has already held that ... although the conditions set out in Altmark ... concern without distinction all sectors of the economy, their implementation must take into account the specificity of a certain sector and, given the particular nature of the SGEI mission in specific sectors, it should be flexible in the application of the Altmark judgment ... in relation to the spirit and purpose that led to the establishment of said conditions, so that they are suitable to the particular facts of the case (see case of 7 November 2012 CBI / Commission, T-137/10 (EU:T:2012:584) paragraphs 85 and 86, and the cases cited) (T-295/12, para 131, own translation from French).
This was criticised by Germany as a conflation of the first and fourth Altmark conditions, particularly because the analysis supported by these general remarks implied dismissing the existence of an SGEI in the specific case in Rhineland-Palatinate, and a general consideration of the costs incurred by undertakings active in the same sector in other German states that, however, may have been subjected to different public service obligations or where no SGEI may have existed at all (T-275/12, para 130). In Germany's submission, this would have led the GC to a tautological conclusion. 

The CJEU dismisses the argument on the following basis:
... the Court cannot be criticized for having reached a tautological conclusion that would have linked the lack of satisfaction of that fourth condition to a finding of lack of qualification of maintaining a reserve capacity as a service of general economic interest [first condition]. Indeed, as is clear from paragraph 130 of the judgment, the Court, first, discussed the situation in which the maintenance of a reserve capacity in case of an outbreak could have validly received such qualification [of SGEI] and on the other hand, felt that, given the obligations of the competent public authorities in all German states to eliminate the largest quantity of substances ... received during an outbreak [regardless of the way they organised the discharge of that public obligation, and regardless of whether they defined an equivalent SGEI], it was necessary to take into account the existing situation in other German states to determine the necessary level of compensation on the basis of an analysis of the costs which a typical undertaking, well run and adequately equipped, would have incurred in meeting the public service requirements (C-446/14 P, para 35, own translation from French).
Thus, the general conclusion of the CJEU is that the GC did not err in law by conflating the different conditions established in Altmark

I disagree with the CJEU because, even if the conditions 'are distinct from one another, each pursuing its own finality', the logic in their application to a same set of factual circumstances requires that, once the scope of the economic activity that the Member State claims is an SGEI is properly established for the purposes of the judicial review (and regardless of whether the first condition is upheld or not in terms of whether those obligations are clearly defined), the analysis of the fourth condition (ie either procurement of that 'alleged' SGEI or analysis of the costs of a notional well-run undertaking providing that 'alleged' SGEI) needs to remain within that context.

Otherwise, the assessment of the notional, well-run undertaking's cost structure outside of the remit of the 'alleged' SGEI under dispute comes to basically neutralise the second alternative test in the fourth condition of Altmark by allowing the Commission and the GC (and ultimately the CJEU) to find any other comparator they deem to be sufficiently close to that economic activity, which nullifies the economic concept of the notional, well-run competitor. Immediately, this pushes Member States to try to avoid in this types of assessment, which can only be done by resorting to (certain types of) public procurement procedures under the first test in the fourth Altmark condition [for discussion, see A Sanchez-Graells, 'The Commission’s Modernization Agenda for Procurement and SGEI', in E Szyszczak & J van de Gronden (eds) Financing Services of General Economic Interest: Reform and Modernization, Legal Issues of Services of General Interest Series (The Hague, TMC Asser Press / Springer, 2012) 161-181]

This may well be cornering Member States in an impossible situation where, regardless of the way they conceive and delineate an SGEI [which they have exclusive competence to do, under Art 14 TFEU and Protocol No (26) therewith, and, currently, reminded in the provisions of Article 1(4) of Directive 2014/24], an assessment of the fourth Altmark condition only allows them to operate with sufficient legal certainty if they contract out the provision of that service in a way that complies with the EU public procurement rules (and not all of them, at that). This is certainly not a desirable outcome and, once more, the developments supported by the CJEU require a fundamental rethinking of the coordination of State aid and public procurement rules, in particular in the area of SGEIs [for discussion, particularly in the setting of procurement challenges, see A Sanchez-Graells, 'Enforcement of State Aid Rules for Services of General Economic Interest before Public Procurement Review Bodies and Courts' (2014) 10(1) Competition Law Review 3-34].

Legitimate expectations claims and EU State aid rules after SAM: some thoughts

One of the points I raised in my paper "Digging Itself Out of the Hole? A Critical Assessment of the European Commission's Attempt to Revitalise State Aid Enforcement after the Crisis" concerned the treatment of legitimate expectations claims in EU State aid enforcement proceedings in the scenario created by the State Aid Modernisation (SAM). In probably not very clear terms, I submitted that
... the Commission will most likely not have the upper hand in withdrawal procedures where Member States (and beneficiaries) will raise important issues related to due process guarantees and good administration duties that limit the Commission’s leeway—unless the old mantra that ‘there is no legitimate expectation to be protected in the field of State aid so as to trump the application of Article 107(1) TFEU’ is extended and applied in an absolute manner—which I do not think possible after the Treaty of Lisbon granted binding force to the EU Charter of Fundamental Rights[68] and, in particular, to the rights to good administration (Art 41)[69] and to an effective remedy and to a fair trial (Art 47).[70]

[68] Charter of Fundamental Rights of the European Union [2010] OJ C 83/389. The specific reasons for this assessment exceed the scope of this paper. For discussion, see R Luja, “Does the Modernisation of State Aid Control Put Legal Certainty and Simplicity at Risk” (2012) EStAL 765-66.
[69] P Craig, “Article 41 – Right to Good Administration”, in S Peers, T Hervey, J Kenner and A Ward (eds), The EU Charter of Fundamental Rights: A Commentary (Oxford, Hart, 2014) 1069-98.
[70] P Aalto et al, “Article 47 – Right to an Effective Remedy and to a Fair Trial”, in S Peers, T Hervey, J Kenner and A Ward (eds), The EU Charter of Fundamental Rights: A Commentary (Oxford, Hart, 2014) 1197-275.
I was rightly challenged on this point by an anonymous reviewer, to whom I am grateful for the opportunity to rethink and expand my arguments on the treatment of legitimate expectations claims for the purposes of State aid enforcement after SAM. I have now addressed the comments and, in the final version of the paper (hopefully soon to be published in the Journal of Antitrust Enforcement), I now explain in more detail what I meant. I hope the argument is now easier to share or, at least, more strongly supported.
... the Commission will most likely not have the upper hand in withdrawal procedures where Member States (and beneficiaries) are likely to raise important issues related to due process guarantees and good administration duties that can limit the Commission’s leeway.[78] It is generally accepted that the principle of legal certainty is one of the general principles recognised in the EU legal order,[79] and that this principle and the corollary protection of legitimate expectations are binding on the Member States and the EU Institutions when they implement EU rules.[80] Nonetheless, the traditional position in this area has been to consider that ‘there is no legitimate expectation to be protected in the field of State aid so as to trump the application of Article 107(1) TFEU’.[81] This has been repeatedly criticised as an inconsistency in the development of the principle of legal certainty and its corollary, the protection of legitimate expectations, in the area of EU State aid law as compared to general EU law.[82] Furthermore, this old mantra may well have been significantly eroded by the entry into effect of the Treaty of Lisbon,[83] which granted binding force to the EU Charter of Fundamental Rights[84] and, in particular, to the rights to good administration (Art 41)[85] and to an effective remedy and to a fair trial (Art 47).[86] Any attempt to transfer the pre-Lisbon ‘no protection of legitimate expectations in State aid law’ paradigm to the post-Lisbon, post GBER paradigm is problematic. The Commission may be tempted to insist that nothing has changed and that, consequently, arguments of legal certainty cannot restrict its ability to dis-apply BER coverage ex post. That would push the old mantra to its extremes and, in my view, would break it. Relatively recent clarifications by the CJEU have tried to establish a balance, whereby recipients of State aid cannot claim legitimate expectations protection if, being diligent, they should have been capable of determining whether or not the EU procedure leading to the award of the aid was complied with or not.[87] Thus, the argument ultimately rests on the observability of the Commission’s ex ante intervention or the absence of such mandatory intervention, where prescribed by EU law (ie Arts 107 and 108 TFEU). In the case of BER protection, this is highly problematic because the restriction of any substantive analysis by the Commission to an ex post phase by necessity requires the recipient to rely on the Member States’ assessment of the BER. As the CJEU has also clarified, ‘a person may not plead breach of the principle of the protection of legitimate expectations unless he has been given precise assurances by the competent authority’.[88] A contrario, such assurances by the Member State as a co-enforcer of EU State aid law in the new post 2014 GBER paradigm may well trigger significant levels of protection of those legitimate expectations.[89] It is submitted, this is likely to increase the weight given to arguments based on legitimate expectations and legal certainty, particularly in the case of attempts to withdraw BER coverage based on a Commission’s ex post assessment that runs contrary to arguments of reasonable reliance (by recipients) on Member State-supported interpretations of the applicable BER,[90] particularly if it derives from a stricter interpretation of the EU State aid rules.[91] This arguments, or at least litigation based on these arguments, can add more layers of ineffectiveness to the post 2014 GBER paradigm based on more withdrawal procedures.
[78] Indeed, this argument is frequently raised in State aid litigation before the EU Courts. For a recent example, see Opinion of AG Whatelet in A2A SpA v Agenzia delle Entrate, C-89/14, U:C:2015:211, paras 44 to 53. However, the AG Whatelet rejected the arguments on the basis of reasons similar to those criticised in the main text.
[79] ISD Polska and Others v Commission, C-369/09 P, EU:C:2011:175, para 122.
[80] Gerekens and Procola, C-459/02, EU:C:2004:454, paras 21 to 24.
[81] However, this is not warranted upon closer examination of the case law, as demonstrated by A Giraud, “A study of the notion of legitimate expectations in State aid recovery proceedings: ‘Abandon all hope, ye who enter here’?” (2008) 45(5) CMLRev 1399-1431.
[82] See T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, OUP, 2006, repr. 2009) 296; W Weiβ and M Haberkamm, “Legitimate expectations in state aid and the CFI” (2010) 9(2) EStAL 537; A Winckler and FC Laprévote, “Reconciling legal certainty, legitimate expectations, equal treatment and the prohibition of state aids” (2011) 10(2) EStAL 321-326.
[83] Similarly, see E Fink, “The Possibility of Protection of Legitimate Expectations in Recovery of Unlawful State Aid” (2013) 1 Juridica International 133-141.
[84] Charter of Fundamental Rights of the European Union [2010] OJ C 83/389. The specific reasons for this assessment exceed the scope of this paper. For discussion, see R Luja, “Does the Modernisation of State Aid Control Put Legal Certainty and Simplicity at Risk” (2012) EStAL 765-66.
[85] P Craig, “Article 41 – Right to Good Administration”, in S Peers, T Hervey, J Kenner and A Ward (eds), The EU Charter of Fundamental Rights: A Commentary (Oxford, Hart, 2014) 1069-98.
[86] P Aalto et al, “Article 47 – Right to an Effective Remedy and to a Fair Trial”, in S Peers, T Hervey, J Kenner and A Ward (eds), The EU Charter of Fundamental Rights: A Commentary (Oxford, Hart, 2014) 1197-275.
[87] Fink (n 83) 136, with reference to France Télécom v Commission, C-81/10 P, EU:C:2011:811, para 59.
[88] AJD Tuna, C-221/09, EU:C:2011:153, para 72; Agrargenossenschaft Neuzelle, C-545/11, EU:C:2013:169, para 25.
[89] At least, where the interpretation by the Member State was reasonable, in line with the original case law in the area of State liability as per The Queen v H.M. Treasury, ex parte British Telecommunications, C-392/93, EU:C:1996:131, para 43 in particular.
[90] The situation is not completely different to that of reliance on legal advisors’ advice, which could erode the argument by reference to Schenker & Co. and Others, C-681/11, EU:C:2013:404. However, this is clearly a controversial area of EU procedural law that requires future developments. In my view, a new wave of protection of legitimate expectations can be expected, particularly where domestic constitutional principles of protection of legitimate expectations as part of the right to good administration are engaged. For discussion, see R Bousta, 'Who Said There is a ‘Right to Good Administration’? A Critical Analysis of Article 41 of the Charter of Fundamental Rights of the European Union' (2013) 19(3) European Public Law 481-488.
[91] Fink (n 83) 139, with reference to Alcoa Trasformazioni v Commission, C-194/09 P, EU:C:2011:497.

Are English Universities likely to stop having to comply with EU public procurement law?

One of the elements implicit in the on-going discussion about higher education reform in England concerns the extent to which changes in the funding and governance structure of HEFCE (to be transformed into the Office for Students, or any other format that results from the consultation run by BIS) can free English universities from their duty to comply with EU public procurement law. 

The issue is recurring in the subsequent waves of higher education reform in England, and the same debate arouse last summer following BIS statements that the most recent reform (lifting the cap on student numbers) would relieve English universities of their duty to comply with EU public procurement law (see discussion here).

Overall, then, there is a clear need to clarify to what extent English universities are actually and currently obliged to comply with EU public procurement rules, both as buyers and as providers of services. That analysis can then inform the extent to which in the future English universities are likely to remain under a duty to comply with EU public procurement rules.

In this study we provide an up-to-date assessment of situations in which universities are bound by public procurement rules, as well as the combined changes that market-based university financing mechanisms can bring about in relation to the regulation of university procurement and to the treatment of the financial support they receive under the EU State aid rules. National differences in funding schemes are likely to trigger different answers in different EU jurisdictions. This study uses the situation of English universities as a case study.
The first part focuses on the role of universities as buyers. The traditional position has been to consider universities bound by EU public procurement rules either as state authorities, or because they receive more than 50% public funding. In the latter case, recent changes in the funding structure can create opportunities for universities to free themselves from compliance with EU public procurement rules.
In the second part, we assess the position of universities as providers. Here the traditional position has been that the State can directly mandate universities to conduct teaching and research activities. However, new EU legislation contains specific provisions about how and when teaching and research need to be procured if they are of an economic nature. Thus, accepting the exclusion of university services from procurement requirements as a rule of thumb is increasingly open to legal challenge.
Finally, the study assesses if and in how far universities can benefit from exemptions for public-public cooperation or in-house arrangements either as sellers or buyers. 
The full paper is available on SSRN:

We have submitted our piece of research to BIS as part of the consultation on the green paper. We hope that our research and the insights it sheds can inform the discussion on the new mechanisms for the allocation of the teaching grant to English universities (and particularly the discussion around Q18 of the consultation).

CJEU takes more refined approach to the 'separability of activities' in the definition of (public) undertakings for the application of EU competition law (C-185/14)

In its Judgment in EasyPay and Finance Engineering, C-185/14, EU:C:2015:716, the Court of Justice of the European Union (CJEU) assessed whether the grant by a Member State of an exclusive right to pay retirement pensions by money order to the national postal operator that formerly held the postal monopoly restricts the rights of alternative postal operators and is detrimental to free competition. The CJEU also assessed whether the granting of such exclusive right to pay retirement pensions by postal money order constitutes State aid or if it can be exempted on the basis that the activity constitutes a service of general economic interest (SGEI).

The case concerned Decree of the Bulgarian Council of Ministers adopted in the year 2000 whereby it ordered that retirement pensions had to be paid through the domestic banks and the post offices of the national postal operator ‘Balgarski poshti’ (BPoshti), which at the time was a single-member commercial company wholly owned by the State. Those postal money orders included the payment of retirement pensions at both the post offices and the address of the beneficiary by a postal worker, which was an activity covered by the universal postal service that only BPoshti was authorised to carry out.

A reform of the postal services act (PSA) to implement postal liberalisation determined that postal money orders are no longer included in the universal postal service and, consequently, under the postal legislation they should be services open to provision by alternative post operators in competition with BPoshti. However, the Council of Ministers maintained the reserve of this activity for BPoshti on the basis that 'the granting and payment of pensions form part of the exercise of State social security functions, which cannot be qualified as an economic activity. ‘Balgarski poshti’ was entrusted under a regulatory act with a public service activity which does not fall within the scope of competition law. The Council of Ministers adds that only that company has a branch network covering all of the territory of Bulgaria, including sparsely populated areas' (C-185/14, para 25). Privately-owned alternative postal operators challenged this reserve of activity under the both EU postal rules of Directive 97/67 (as amended) and Articles 106 and 107 TFEU.

The CJEU first addressed the point of coverage of postal money orders by the postal universal service and confirmed that 'money order services, which consist in making payments through the public postal network to natural or legal persons on behalf of and on the order of others, are not within the scope of Directive 97/67 (see judgment in Asempre and Asociación Nacional de Empresas de Externalización y Gestión de Envíos y Pequeña Paquetería, C-240/02, EU:C:2004:140, paragraph 34)' (C-185/14, para 32). Therefore, given the inapplicability of the sectoral postal regulation, the analysis of the reservation of this activity to BPoshti had to be assessed under the general rules in Articles 106 and 107 TFEU.

In this analysis, the CJEU returns to the definition of undertaking for the purposes of the application of competition law and stresses several aspects regarding the need for an inseparable connection with the national pensions system for an entity to escape the definition of undertaking on the basis that it operates within a system based on the principle of solidarity and oriented exclusively to perform a social function. In terms of the Judgment in EasyPay and Finance Engineering,
37 ... for the purposes of the application of EU competition law, an undertaking is any entity engaged in an economic activity, irrespective of its legal status and the way in which it is financed ...  any activity consisting in offering goods and services on a given market is an economic activity (see judgment in Compass-Datenbank, C-138/11, EU:C:2012:449, paragraph 35).
38 ... the organisations involved in the management of the public social security system fulfil an exclusively social function. That activity is based on the principle of national solidarity and is entirely non-profit-making. The benefits paid are statutory benefits bearing no relation to the amount of the contributions (see, to that effect, judgment in Poucet and Pistre, C-159/91 and C-160/91, EU:C:1993:63, paragraph 18).
39 It is for the referring court to ascertain whether or not the money order operations carried out by ‘Balgarski poshti’, enabling the payment of retirement pensions at issue in the main proceedings, is involved in the functioning of the public social security service and, accordingly, must or must not be regarded as an economic activity falling within the scope of Article 107(1) TFEU.
40 In that context, it must be recalled that, in order to avoid classification as an economic activity, that activity must, by its nature, its aim and the rules to which it is subject, be inseparably connected with the national pensions system (see, by analogy, judgment in Aéroports de Paris v Commission, C-82/01 P, EU:C:2002:617, paragraph 81). Thus, in the main proceedings, any inseparable connection thereto of the activity of money order operations must be taken into consideration.
41 In that regard, it is apparent ... that the old-age benefits granted in the State social security system form part of the task of the [National Social Security Institute] which, in carrying out that task, uses ‘Balgarski poshti’ solely to handle the payments of retirement pensions.
42 In addition ... payment of retirement pensions may also be made through banks. Thus, according to the information provided by the Institut ... approximately 53% of the total number of retirement pensions were paid by bank transfer. Accordingly, the money orders used by ‘Balgarski poshti’ are not actually the sole method of payment of the retirement pensions.
43 Those elements constitute an indication enabling the view to be taken that the activity of money order operations enabling the payment of retirement pensions may be separable from the national pensions system. It is for the national court to assess the relevance of those elements, in particular in the light of the other factual and legal elements before it (C-185/14, paras 37-43, emphasis added).
This is an interesting approach, because the CJEU seems to have deviated from the line of case law that was tending towards an excessively lenient analysis of the 'inseparable connection' between market services and the discharge of public services (see here and, for more details, here). However, the specific circumstances of the case may have prompted this  more nuanced or refined approach, particularly because the postal operator seems to be quite far removed from the core public function its (market) activity supports.

On the point of analysis of the reservation of the activity as an SGEI, which would exclude the existence of State aid under Article 106(2) TFEU, the CJEU focuses the analysis on the level of compensation that BPoshti receives for the discharge of the public service that triggers the reservation of activity. In that regard, it is worth noting that
45 ... a State measure regarded as compensation for the services provided by the recipient undertakings in order to discharge public service obligations, so that those undertakings do not enjoy a real financial advantage and the measure thus does not have the effect of putting them in a more favourable competitive position than the undertakings competing with them is not caught by Article 107(1) TFEU (see judgments in Libert and Others, C-197/11 and C-203/11, EU:C:2013:288, paragraph 84, and Altmark Trans and Regierungspräsidium Magdeburg, C-280/00, EU:C:2003:415, paragraph 87).
46 However, for such compensation to escape classification as State aid in a particular case, a number of conditions must be satisfied (judgment in Altmark, C-280/00, EU:C:2003:415, paragraph 88).
52 Where the undertaking which is to discharge the service of general economic interests ... is not chosen pursuant to a public procurement procedure, it is also for the referring court to make sure, in accordance with the fourth condition laid down in paragraph 93 of the judgment in Altmark (C-280/00, EU:C:2003:415), that the level of compensation is determined on the basis of an analysis of the costs which a typical undertaking, well run and adequately equipped, would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging the obligations (C-185/14, paras 45-46 and 52, emphasis added).
In the case at hand, the remuneration payable to BPoshti was set out in the Decree of the Council of Minister, which foresaw that the National Social Security Institute 'shall pay through its territorial divisions to the territorial divisions of the “Balgarski poshti” 8.5 thousandths of the pensions payable in the month in question for the work performed in connection with the payment of pensions through the postal network' (C-185/14, para 15). In my view, it seems unlikely that such a general figure established 15 years ago (coincidentally) matches the costs of a notional efficient postal competitor.

Thus, in general terms, the conclusion to be extracted from the Judgment in EasyPay and Finance Engineering is that the reservation of the activity of payment of pensions or any other benefits through postal money order (or equivalent) to incumbent postal operators (or any other undertakings) is very likely to fall foul of Articles 106 and 107 TFEU, unless the undertaking is chosen through a procurement procedure (which should be relatively easy to implement) or its remuneration is determined in strict accordance with the standard of a notional efficient postal competitor [for discussion, see A Sanchez-Graells, 'The Commission’s Modernization Agenda for Procurement and SGEI', in E Szyszczak & J van de Gronden (eds) Financing Services of General Economic Interest: Reform and Modernization (The Hague, TMC Asser Press / Springer, 2012) 161-181]. Let's hope that the domestic Bulgarian court follows the very clear cues given by the CJEU in that respect.

An EU Competition Law Primer for Public Procurement Students

My friend and colleague Dr Carina Risvig Hamer asked me to contribute a chapter on EU competition law to her forthcoming handbook on EU public procurement she is about to publish with Djøf Forlag. She is writing it in Danish to support her teaching at the University of Southern Denmark. Thus, the book is unlikely to reach a wider English-speaking audience. This is why I decided to post the chapter on SSRN, in case there are some non-Danish procurement students interested in a first introduction to EU competition law issues.

As the abstract indicates, this chapter aims to identify the key areas where EU competition law is relevant from a public procurement perspective: that is, mainly, the prevention and sanctioning of procurement manipulation by suppliers (bid rigging) and the granting of distortive State aid that advantages some of them over others. It also focuses on potential abuses of market power by undertakings holding a dominant position, but it assesses this potential distortion of competition to a more limited extent. Once these areas are identified, the chapter describes the basic EU competition rules that apply in each of these different cases, as well as their interpretation in the case law of the CJEU. The main goal of this chapter is to provide public procurement students with an overall view and basic understanding of the EU competition rules more directly relevant to procurement practice.

The paper's full reference is: A Sanchez-Graells, 'An EU Competition Law Primer for Public Procurement Students' (October 18, 2015). Available at SSRN:

CJEU implicitly rejects GC's views on subjective assessment of two-part State aid measures under Art 107(1) TFEU (C-15/14)

In its Judgment in Commission v MOL, C-15/14, EU:C:2015:362, the CJEU upheld the previous Judgment of the GC where the selectivity of two-part State aid measures was assessed with very generous deference towards the State's exercise of regulatory powers (which I criticised here). 
The CJEU assessed the criticism by the Commission of the GC's position (T-499/10, paras 64 and 65) that the presence of a selective advantage cannot be deduced from the mere fact that the operator is left better off than other operators when the Member State concerned justifiably confined itself to exercising its regulatory power following a change on the market. 
Remarkably, the Commission took issue with the fact that the General Court linked "the assessment of the selective nature of the ... agreement, and therefore the measure at issue, to whether or not the Member State concerned had the intention, at the time of concluding that agreement, of protecting one or more operators from the application of a new fee regime" (C-15/14, para 85, emphasis added). As the CJEU stresses
According to the Commission, the General Court thus disregarded the settled case-law of the Court of Justice to the effect that Article 107(1) TFEU defines State interventions on the basis of their effects, and independently of the techniques used by the Member States to implement their interventions (see, inter alia, judgments in Belgium v Commission, C‑56/93, EU:C:1996:64, paragraph 79; Belgium v Commission, C‑75/97, EU:C:1999:311, paragraph 25; British Aggregates v Commission, C‑487/06 P, EU:C:2008:757, paragraph 89; and Commission v Government of Gibraltar and United Kingdom, C‑106/09 P and C‑107/09 P, EU:C:2011:732, paragraphs 91, 92 and 98) (C-15/14, para 86).
I had also criticised the GC for the inclusion of the element of "intention" in its previous Judgment. However, I also expressed doubts as to the CJEU's willingness to side by the GC. In my view back then,
If Article 107(1) TFEU is meant to avoid distortions of competition in the internal market, when confronted with sequential, two-part or complex aid measures, the fact that they all formed part of a 'master plan' from the outset or are the 'random or supervening' result of discrete interventions should be irrelevant. Otherwise, the burden of proving 'distortive intent' from the outset may simply make it impossible to pursue these cases. However, it may well be that the remarks made by the GC in para 67 of MOL v Commission will remain a 'mere' obiter dictum and that the assessment of two-part or complex measures will remain much more objective in the future.
Consequently, I was hoping that the CJEU would quash this part of the Judgment in T-499/10. However, the CJEU rejected the argument of the Commission and determined that the GC's argumentation in paras 64 to 67 and 82 of the Judgment in T-499/10 was not vitiated by any error of law. I disagree with the CJEU's arguments to support the GC's position, which deserve close scrutiny (below). However, given that the CJEU has managed to uphold the GC's reasoning and at the same time stress that two-part or complex State aid measures must be assessed without any reference to the "intention" of the Member State, I agree with the outcome of the case.
According to the CJEU,
92 ... the General Court stated, in paragraph 67 of the judgment under appeal, that [under] the case-law of the Court of Justice, ... for the purposes of Article 107(1) TFEU, a single aid measure may consist of combined elements on condition that, having regard to their chronology, their purpose and the circumstances of the undertaking at the time of their intervention, they are so closely linked to each other that they are inseparable from one another (judgment in Bouygues and Bouygues Télécom v Commission and Others and Commission v France and Others, C‑399/10 P and C‑401/10 P, EU:C:2013:175, paragraphs 103 and 104 and the case-law cited).
93 In that context, the General Court emphasised, in paragraph 67 of the judgment under appeal, that a combination of elements such as that relied upon by the Commission in the decision at issue may be categorised as State aid when the State acts in such a way as to protect one or more operators already present on the market, by concluding with them an agreement granting them fee rates guaranteed for the entire duration of that agreement, while having the intention at that time of subsequently exercising its regulatory power, by increasing the fee rate so that other market operators are placed at a disadvantage, be they operators already present on the market on the date on which that agreement was concluded or new operators.
94 It was in the light of those considerations that the General Court, in paragraph 68 of the judgment under appeal, decided that it was necessary to examine whether, in those proceedings, the Commission was entitled to consider that the contested measure was selective.
95 It follows from the foregoing that, as MOL contends, paragraphs 64 to 67 of the judgment under appeal do not, as such, concern the examination of the selectivity of the 2005 agreement, but are preliminary explanations aimed at introducing the relevant framework in relation to which the General Court examined whether the Commission was correct in finding that the measure at issue was selective (sic).
96 As the Advocate General stated in points 107 and 114 of his Opinion, by those preliminary explanations, the General Court in fact sought to deal with the issue of the links existing between the 2005 agreement and the 2008 amendment, which the Commission had not specifically addressed in the decision at issue, and more particularly, to underline the fact that, given that there is no chronological and/or functional link between those two elements, they cannot be interpreted as constituting a single aid measure.
97 By those preliminary explanations, the General Court merely applied the case-law laid down by the Court of Justice in the judgment in Bouygues and Bouygues Télécom v Commission and Others and Commission v France and Others (C‑399/10 P and C‑401/10 P, EU:C:2013:175), to which the General Court also expressly referred in paragraph 67 of the judgment under appeal, and according to which, since State interventions take various forms and have to be assessed in relation to their effects, it cannot be excluded that several consecutive measures of State intervention must, for the purposes of Article 107(1) TFEU, be regarded as a single intervention. That could be the case, in particular when consecutive interventions, having regard to their chronology, their purpose and the circumstances of the undertaking at the time of those interventions, are so closely related to each other that they are inseparable from one another (C-15/14, paras 92 to 97, emphasis added).
I find the reasoning of the CJEU very poor. By artificially breaking up paragraph 67 of the GC's Judgment in paras 92 and 93 of its own Judgment, the CJEU attempts to limit the requirement of the element of "intention" to some mysterious "preliminary explanations" excluded from the selectivity assessment, and this is very unsatisfactory and unconvincing.

In my view, the CJEU should have plain and simply said that the GC would have been wrong to include an element of "intention" in the test applicable to two-part or complex State aid measures, which assessment needs to be carried out in view of objective factors such as 'their chronology, their purpose and the circumstances of the undertaking at the time of their intervention, [or whether] they are so closely linked to each other that they are inseparable from one another' as per Bouygues and Bouygues Télécom v Commission and Others and Commission v France and Others.

Allowing the GC to save face by limiting its erroneous interpretation of that case law in para 67 of T-499/10, or failing to stress the fact that it was an unfortunate expression made obiter dictum (if they wanted to remain deferential) pays lip service to legal certainty. In my view, the CJEU could have decided otherwise because the element of "intention" is actually not assessed at any point of the GC's Judgment and the CJEU was ready to accept the selectivity analysis carried out by the GC. Consequently, there was no need for the strange and convoluted analysis in paras 92 to 97 of the Judgment in C-15/14. 

Be it as it may, the silver lining is in the fact that the CJEU has clearly rejected that the test it progressively laid down for the analysis of two-part or complex State aid measures encompasses any subjective element of "intention" on the part of the granting Member State. Consequently, the analysis of the selectivity of measures closely connected will continue to have to be carried out on the basis of purely objective factors, such as 'their chronology, their purpose and the circumstances of the undertaking at the time of their intervention, [or whether] they are so closely linked to each other that they are inseparable from one another. All is well that ends well.

Social housing, State aid and procurement show up again in EU Courts' case law (T-397/12)

In its Judgment in Diputación Foral de Bizkaia v Commission, T-397/12, EU:T:2015:291 (only available in ES and FR), the General Court (GC) of the Court of Justice of the European Union (CJEU) has decided a case involving State aid and public procurement for social housing. The issues raised in this case are technically different from those discussed in previous cases and, particularly, in Libert and Others (joined cases C-197/11 & C-203/11, EU:C:2013:288). However, there are some elements worth discussing. 

As a preliminary point, it is worth stressing that the case only concerned violations of State aid rules under Arts 107 and 108 TFEU. However, there was a very significant violation of EU public procurement rules as well--as a result of the direct award of over €150mn worth of works or supplies (depending on how pre-fabricated modular homes are categorised). Nonetheless, given that this was not discussed in the case, the comments are limited to the State aid dimension of the Spanish infringement of EU law.

The claimant, Diputación Foral de Bizkaia (DFB) is a regional authority in Spain and it owns 100% of public corporation Bizkailur SA (Bizkailur), which is dedicated to urban development, particularly for industrial purposes, with the ultimate aim of attracting business to the region. Through Bizkailur, DFB entered into two contracts with Habidite Technologies País Vasco SA (Habidite) for the set-up of a construction module factory in Alonsotegi and the delivery of 1,500 modular homes. 

According to the first contract, DFB and Bizkailur would purchase a land plot and adapt it for industrial use to prepare the setting up of a Habidite factory in Alonsotegi. Under the second contract, the public authorities committed to purchase from Habidite a total of 1,500 homes constructed with modules produced in Alonsotegi in order to sell them as social housing [see the Commission's press release].

The Commission found that the contracts contained illegal state aid because no private player would have accepted to contract on such terms. In the Commission's view, the maximum support that could be granted to the project was of €10.5 mn, but the actual advantage given to Habidite was much larger [see Commission's Decision for details]. Moreover, given that the measures had not been notified before they were granted to Habidite, the aid was unlawful.  DFB challenged the Commission's decision, but the GC has dismissed the appeal. A further appeal before the CJEU seems unlikely, given that the Habidite project was eventually abandoned by DFB due to lack of financial resources.

Some of the arguments submitted by DFB to the GC are so shocking that they deserve some discussion before laughing them off. In particular, I find it ridiculous for DFB to have submitted that no aid existed on the basis of a circular argument ultimately based on its own failure to comply with EU law, which would suffice to trigger a constructive estoppel under Spanish law (as, indeed, later declared by the competent domestic courts; see press release). 

The argument goes as follows: DFB signed a contract with Hadibite through Bizkailur for interests in land and property, which subjected it to Spanish civil rules (ie private law). A fundamental element in the Spanish contract formation rules is the requirement of Art 1258 Civil Code, according to which "Contracts are perfected by mere consent, and since then bind the parties, not just to the performance of the matters expressly agreed therein, but also to all consequences which, according to their nature, are in accordance with good faith, custom and the law" (emphasis as per GC's extracts). The aid given to Hadibite through the contracts was illegal under Arts 107 and 108 TFEU. Hence, the contract was null and void and, ultimately, there was no aid because DFB (through Bizkailur) was not boud to comply with an illegal contract. 

This is a circular and preposterous argument, as it would mean that there would never be illegal and unlawful State aid measures because they would be legally non-existent as a result of the supremacy and direct effect of Arts 107 and 108 TFEU, regardless of the specific (private or public) contract rules of the specific Member State. And, in any case, domestic theories of estoppel could well de-articulate the argument on strictly contractual terms--which would immediately trigger issues of good faith in negotiations and pre-contractual liability for the authorities granting aid, which is also prevented by the effet utile of Arts 107 and 108 TFEU.

Along these lines, and more elegantly, the GC considered that "as to the applicant's argument that there can be no breach of EU law because national law requires a prior notification of the planned aid to the Commission [without which there is no legally binding and unconditional commitment to actually grant the aid], it suffices to state that it cannot be declared that no infringement of EU law existed for the mere fact that [the contracts] also violated national law" (T-397/12, para 36, own translation).

Finally, it is worth mentioning that the Judgment is also interesting for the discussion of the procedural rights of sub-central public authorities in State aid infringement procedures opened against the Member State of which they are part. The GC reiterated the standard position that (given that they are interested parties, but not parties of the procedure) such authorities cannot rely on the rights of defense or try to maintain an open debate with the Commission (paras 53-63).

New SSRN paper on State aid enforcement after the crisis

I have uploaded a new paper on the University of Leicester School of Law Research Paper SSRN Series. It is entitled "Digging Itself Out of the Hole? A Critical Assessment of the European Commission's Attempt to Revitalise State Aid Enforcement after the Crisis" and has the following abstract:

This paper aims to assess the likelihood that State aid enforcement can be revitalised in the post-crisis period as a result of the 2012-2014 State aid modernisation process (SAM). The paper takes the view that State aid enforcement was left in a difficult impasse as a result of the extraordinary measures the Commission implemented during and immediately after the 2008 economic breakdown, which left the Commission in a difficult position due to the unavoidable concessions and lowering of standards that dealing with the soaring volume of State aid required. The paper builds on this premise to critically assess whether a scenario of stronger enforcement can be foreseen under the modernised, post-2014 procedural framework of SAM. It pays particular attention to the need for the European Commission to (re)engage in a more substantive assessment of aid measures and to promote judicial (or private) enforcement of State aid rules in an effective manner. It concludes that revitalisation of State aid enforcement under SAM is highly unlikely.

I have attempted some statistical analysis to support my view that State aid enforcement is not being efficient. As a taster (full details in the paper), I argue that 'it seems conservative to estimate at around 100 billion Euros the amount of (non-investigated) illegally-granted State aid in the EU28 between 2008 and 2013' and that the Commission is accumulating a significant backlog of State aid cases (of around 500 in the same period), despite having expanded its State aid workforce by 53% between 2007 and 2011.

I also argue that the Commission's push for more transparency of the awards of State aid will not result in an actual involvement of private parties and society at large as stewards of EU State aid rules, in particular due to the restriction of the locus standi to submit (admissible) complaints to the Commission (following Sarc v Commission and the rules under the revised art 11a of reg 794/2004) and the compounded effect of the mandatory use of a standard form that requires significant information.

I will present a reworked version of this paper at the Antitrust Enforcement Symposium held by the Centre for Competition Law and Policy of the University of Oxford in June, where I am honoured to share a session on Competition and the State with such distinguished scholars and practitioners as Conor Quigley QC, Damien Geradin, James Cooper, David Szafram, Isabel Taylor, Angus Johnston and Ioannis Lianos. As you see, not the easiest audience. So all comments that can help me improve the paper are most welcome! I already thank my colleague Dr Paolo Vargiu for his first reactions.
The full citation for the paper is: A Sanchez Graells, "Digging Itself Out of the Hole? A Critical Assessment of the European Commission's Attempt to Revitalise State Aid Enforcement after the Crisis" (May 5, 2015) University of Leicester School of Law Research Paper No. 15-15. Available at SSRN:

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)]. 

Exclusive rights, State aid and lottery: a winning ticket worth an extended monopoly? (T-58/13)

In its Judgment in Club Hotel Loutraki and Others v Commission, T-58/13, EU:T:2015:1, the General Court (GC) has confirmed the previous Decision of the European Commission and considered that Greece had not granted illegal State aid to Organismos Prognostikon Agonon Podosfairou AE (OPAP) through the simultaneous extension of its existing exclusive right to operate certain games of chance and the granting of a new exclusive right to exploit 35,000 Video Lottery Terminals (‘VLTs’) for a period of 10 years in Greece. 

The key to the analysis conducted by the Commission and now upheld by the GC is that by overpaying for the extension of the existing exclusive right, OPAP has been able to secure a much larger exclusive right to operate VLTs in Greece. As the GC summarises:
10 As regards, first, the Addendum [which extended the existing exclusive rights for the period 2020-2030], the Commission observed that the study provided by the Greek authorities was based on sales projections elaborated by an independent company specialised in the gambling sector. The net present value of the Addendum was calculated on the basis of those projections, which were considered by the Commission to be reliable.
11 Following that calculation, the Commission found that the amount paid by OPAP in exchange for the Addendum, including the levy imposed by the Greek State corresponding to 5% of the gross gaming revenues generated by the games concerned for the period from 13 October 2020 to 12 October 2030 (see paragraph 4 above), was higher than the net present value of the Addendum.
12 As regards, secondly, the VLT Agreement, the Commission also calculated its net present value on the basis of the study commissioned by the Greek authorities.
13 On the basis of that calculation, the Commission stated that the net present value of the VLT Agreement was significantly higher than the amount of EUR 560 million provided for in the VLT Agreement, which would economically advantage OPAP.
14 However, the Commission stated that it was logical for the conformity of the VLT Agreement and the Addendum with Article 107(1) TFEU to be assessed jointly. In that way, the overpayment by OPAP for the Addendum was taken into account in order to assess the conformity of the VLT Agreement with that article. The Commission stated that the overpayment reduced the gap between the net present value of the VLT Agreement and the amount of EUR 560 million owed by OPAP
[...] (T-58/13, paras 10-14, emphasis added).
Even if it is true that the Commission managed to impose an additional payment on VLT revenues to further close the economic gap as an amendment to the State aid scheme, the crucial point remains:
17 [...] Referring to the amendment introduced by the Greek authorities, and taking account of the overpayment for the Addendum, the Commission found, on average, OPAP would pay more than the value of the VLT Agreement.
18 In other words, the Commission took the view that, following the amendments to the initial notification, OPAP would pay the Greek State a higher amount than the cumulated values of the exclusive rights granted by the VLT Agreement and the Addendum (including a reasonable return for OPAP)
(T-58/13, paras 17-18, emphasis added).
Hence, as mentioned, the crucial point for the legality of the (conflated) scheme is still the fact that the overpayment for the extension of an existing exclusive right is used to secure the approval of the underpayment in the granting of a new exclusive right. Moreover, the final finding of the European Commission simply makes no sense, as no market agent would pay a higher price for those exclusive rights than their accumulated value, as this would not be a rational investment decision. Consequently, there are many issues that would require some deeper scrutiny.

More importantly, in my view, the general acceptance of the 'cross-overpayment' amounts to allowing dominant undertakings with exclusive rights to buy their way into an extended monopoly (in a rather evident economic leverage) and, consequently, the case should be criticised--and quashed by the Court of Justice upon appeal (if it gets further appealed). Not least because it follows an emerging trend of improper assessment of two-part State aid measures (in favour of former State companies) that I find worrying and potentially dangerous for a credible and effective State aid control regime (see a previous instance here). The reasoning followed by the Commission and the GC, then, deserves some analysis.

Some of the arguments presented by the applicants have (willfully?) not been properly understood, nor analysed by the GC. Amongst other important arguments, the applicants clearly referred to the problem of the extension of the existing exclusive rights by cross-subsidisation in the following terms (in the words of the GC):
79 The applicants claim first of all that the Commission recognised, in paragraph 37 of the contested decision, that the Addendum and the VLT Agreement refer to distinct markets. Nevertheless, the Commission assessed them jointly. The applicants submit that the existence of an advantage for the purpose of Article 107(1) TFEU must be assessed for each market and not on the basis of joint consideration of similar measures concerning different markets, even though the measures examined concern the same recipient. If it were otherwise, the protection of competition would be incomplete because measures constituting an anti-competitive advantage for the purpose of Article 107(1) TFEU in a given market might escape the prohibition laid down in that provision on the basis of a joint assessment. Conversely, measures which grant no economic advantage in a given market might nevertheless be covered by that provision on the basis of a joint assessment with a measure affecting another market. [...]
81 The applicants claim that the VLT and slot machine market cannot be assessed jointly with the 13 games of chance covered by the Addendum since they have no relation to the market of the 13 games of chance on which OPAP has an absolute legal monopoly. By virtue of that monopoly, OPAP could carry out cross-subsidisation practices allowing OPAP to undercut the applicants’ prices on the VLT and slot machine market, by financing that operation by a price increase on the market for the 13 games of chance. However, the joint assessment of the notified measures does not take into account the possibility of such practices (T-58/13, paras 79 and 81, emphasis added).
To be fair, if the arguments were presented in this way (but this seems open to debate), it takes some digging to see that there are two layers of potential cross-subsidy. The first one, which is the one criticised above, is that the overpayment in one leg of the measure (extension of monopoly) secures State aid compatibility of the other leg of the measure (creation of an additional monopoly over VLTs). The second one concerns the operation of the rights in case they had been assigned to different operators, as it would concern a situation in which both OPAP and third parties had been granted licences for the operation of VLTs. The second argument is, in my view, moot or improperly addressed, as it refers to a hypothetical, counterfactual scenario. However, the first argument should have been enough to quash the Commission's Decision. Nonetheless, the GC decided differently.

In its analysis of the fourth plea submitted by the appellants of the Commission's Decision (the other three are basically procedural, so I am skipping them for now), the GC found that:
94 As regards [...] the applicants’ argument relating to subsidisation practices made possible by OPAP’s monopoly over the 13 games of chance covered by the Addendum, it should be noted, first, that it is based on the assumption that OPAP is free to increase prices at will on those 13 games in order to compensate for lower prices on the VLT market. The applicants accordingly submit that OPAP will not sustain competitive pressures in its pricing policy. That argument is not, however, substantiated. In fact, the applicants do not support or demonstrate that the 13 games in question are not subject to competition from other games of chance.
95 Next, the applicants do not explain why the alleged practices of cross-subsidies between the lower prices on the VLT market and the higher prices on the market of the 13 games covered by the Addendum preclude the two notified measures being jointly assessed. Indeed, if such practices were to exist, they would create a link between the VLTs and the 13 games of chance, which instead supports the two measures being jointly assessed.
96 It follows from all the foregoing that the applicants have not demonstrated the existence of an error of law when the Commission carried out a joint assessment of the VLT Agreement and of the Addendum
(T-58/13, paras 94-96, emphasis added).
This is troubling because the GC inverts the order of the arguments on cross-subsidisation and dismisses them in the wrong way. Firstly, it is hard to see how the GC can rely on a theoretical competitive pressure on OPAP when the situation is that it holds basically exclusive rights on all relevant games of chance in Greece. Secondly, it is unacceptable that the GC buys a justification for the joint analysis of the measures precisely because OPAP engages in cross-subsidisation. If this is not a clear deductive fallacy, there is none. Overall, then, the arguments of the GC are disappointingly thin, or simply incongruous.  Consequently, for all the above, I hope the CJEU will receive better economic advice and will reverse the Hotel Loutraki Judgment. Otherwise, the game will be over for the analysis of two-part or leveraged instances of clear State aid.

Some thoughts on a paper on the Concessions Directive and competition law [Farley-Pourbaix, (2015) JECLAP 6(1): 15-25]

Martin Farley and Nicolas Pourbaix have recently published a paper on the interaction between competition and public procurement law in light of the rules of new Directive 2014/23 on concession contracts. The paper is 'The EU Concessions Directive: Building (Toll) Bridges between Competition Law and Public Procurement?' (2015) 6(1) Journal of European Competition Law & Practice  15-25. 

The paper is extremely thinly researched in an area that is generating a significant amount of scholarly commentary and, as such, it is rather disappointing because the authors seem to be (re)discovering powder by emphasising the interaction between procurement and competition law rules. However, some of the main points the authors make in relation to the pre-existing case law of the CJEU are worth considering.

Firstly, they stress the practical complications that the open-ended definition of concession creates, particularly in terms of the difficulty of assessing when the transfer of risks to the concessionaire suffices to be covered by Directive 2014/23 instead of Directive 2014/24 or Directive 2014/25 [for discussion, see C Risvig Hansen, Contracts not covered or not fully covered by the Public Sector Directive (Copenhagen, DJOF, 2012)76-102; A Sanchez-Graells, 'What Need and Logic for a New Directive on Concessions, Particularly Regarding the Issue of their Economic Balance?' (2012) 2 European Public Private Partnership Law Review 94-104; and R Craven, 'The EU's 2014 Concessions Directive’ (2014) 23 Public Procurement Law Review 188-200].

Secondly, they explore the applicability of Art 101 TFEU to bidders that opt to team up or bid jointly for concession contracts. Their remarks are interesting and topical, as the recent publication of the 'Consortium Bidding' guidelines by the Irish Competition and Consumer Protection Commission evidences. I found their warning on the need to limit the exchanges of information between consortium partners particularly relevant (pp. 19-20), as joint participation in selected procurement projects could be the conduit for cartelising behaviour and this is an issue that requires careful consideration.

Thirdly, they revisit the never-ending discussion on the exclusion of contracting authorities from the concept of undertaking for the purposes of the application of (EU) competition law on the basis of the FENIN-SELEX line of case law [FENIN v Commission, C-205/03, EU:C:2006:453; and Selex v Commission, C-113/07, EU:C:2009:191] [for discussion, see A Sanchez-Graells, 'Distortions of Competition Generated by the Public (Power) Buyer: A Perceived Gap in EC Competition Law and Proposals to Bridge It' (2009) University of Oxford, Center for Competition Law and Policy, CCLP (L). 23]. 

On this point, it is interesting to see how Farley and Pourbaix stress that utilities concessions may trigger the application of competition law because, almost by definition, the contracting entity will be engaged in 'downstream' economic activities. Their discussion of the Luton Airport case is certainly informative [Arriva the Shires Ltd v London Luton Airport Operations Ltd [2014] EWHC 64 (Ch)].

This may be a point to take into consideration in the future to (possibly) limit the FENIN-SELEX exemption in case contracting authorities outside the utilities sector engage in (partial) downstream economic activity, which is likely to be the case of some in-house or public-public cooperation arrangements, which can now offer up to 20% of their supplies or services in the 'private market' under the rules of Directive 2014/24. This would be particularly easy on the basis of the 'severability' of activities for the purposes of competition law [Aéroports de Paris v Commission, C82/01, EU:C:2002:617], which in my view would be a most welcome development of this area of the law.

Finally, Farley and Pourbaix focus on specific competition law aspects of the new EU Concessions Directive. Of the issues they mention (other than the duration of the concession contract), the most interesting are the possibility to exclude infringers of competition law (on which see the recent case law of the CJEU here), and the interaction between State aid rules and the modification of concession contracts [for discussion, see A Sanchez-Graells, 'Public Procurement and State Aid: Reopening the Debate?' (2012) 21(6) Public Procurement Law Review 205-212]. 

On the issue of exclusion, the paper stresses burden of proof difficulties and advocates for a careful enforcement of the power to exclude undertakings suspected of competition violations, and points (without mentioning) at corporate human rights such as the presumption of innocence, which would have deserved more detailed consideration [for general discussion, see A Sanchez-Graells and F Marcos, '"Human Rights" Protection for Corporate Antitrust Defendants: Are We Not Going Overboard?' (2014) University of Leicester School of Law Research Paper No. 14-04]. 

On the issue of State aid being (implicitly) granted as a result of a modification of a contract during its term, the paper emphasises that the increased flexibility in the choice of procedures and the possibility to modify the contract (potentially without value limit, despite the stress on 50% that Farley and Pourbaix wrongly put in p. 24-25) in a relatively generous array of cases restricts the 'Altmark' presumption and requires a substantive assessment of the conditions of the contract [something already advocated for in A Sanchez-Graells, Public procurement and the EU competition rules (Oxford, Hart, 2011) 118-121 and, in more detail, in ibid, 'The Commission’s Modernization Agenda for Procurement and SGEI', in E Szyszczak & J van de Gronden (eds) Financing SGEIs: State Aid. Reform and Modernisation, Legal Issues of Services of General Interest Series (The Hague, TMC Asser Press / Springer, 2012) 161-181].

A point of contention, though, refers to the treatment of concession contracts as conduits for State aid. Farley and Pourbaix consider that:
Contracting Authorities may be able to take a certain amount of comfort from the fact that many concessions may not qualify as State aid in any event, on the basis that the remuneration was not granted through State resources. This will at least be the case in those situations where the concessionaire is remunerated entirely by third parties. Following the CJEU’s ruling in PreussenElektra [PreussenElektra, C-379/98, ECLI:EU:C:2001:160] this will still be the case even if the State sets the price that third parties need to pay for the relevant goods or services. (P. 24).
Even if they indicate that mixed arrangements which include some sort of subsidy could erode this possibility to duck State aid rules, I think that they present the situation in a way that excessively narrows down their application. Indeed, on that point, it may worth stressing that the CJEU has relatively recently adopted a less formalistic approach and considered that certain aspects of public control over third party revenue (which are common to concession contracts) may trigger the dis-application of the PreussenElektra exception (see comment here). 

In particular, in Vent De Colère and Others, C-262/12, EU:C:2013:851, the CJEU found that:
Article 107(1) TFEU must be interpreted as meaning that a mechanism for offsetting in full the additional costs imposed on undertakings [...] that is financed by all final consumers [...] constitutes an intervention through State resources (C-262/12, para 37).
Hence, even decisions concerning authorizations to raise user fees (without offering any additional public support or implying any extension of the length of the concession) may trigger State aid application, which is a case most concession contracts usually contemplate. Hence, the interaction between the prohibition of State aid in Art 107(1) TFEU and the rules on modification of concession contracts in Directive 2014/23 is more intense than Farley and Pourbaix's paper presents.

Overall, then, the paper is not groundbreaking and, if the existing literature had been researched, it would probably have been of a higher academic interest (as it is published, though, it certainly is oriented to practitioners) and could possibly have reached a deeper level of analysis. In any case, given the novelty of Directive 2014/23, Farley and Pourbaix's paper can certainly raise awareness of the important issues they mention.

Paper on centralisation of procurement and competition law

Ignacio Herrera Anchustegui, from BECCLE - University of Bergen, and I have just completed a working paper on the new rules on centralisation and occasional procurement under articles 37-39 of Directive 2014/24. The paper assesses the risks, rationale and justification for the rules on centralisation and aggregation of public procurement in Directive 2014/24. The paper is entitled "Impact of public procurement aggregation on competition. Risks, rationale and justification for the rules in Directive 2014/24" and is now part of the University of Leicester School of Law Research Paper Series.
The paper explores the justifications advanced for the aggregation of purchasing and the countervailing risks it generates. In both cases, it focusses in economic and administrative aspects. It then proceeds to a summary overview of the new rules for the aggregation of public procurement in Directive 2014/24, and emphasised how the Directive is expressly recognising possibilities that clearly exceed the more modest approach in Directive 2004/18. Moving on, it then focusses on the potential justification for certain activities now permitted by the 2014 rules, and engages in a critical assessment of their competitive impact. The paper briefly highlights the far-reaching and not necessarily positive implications that a maximisation of the centralisation and aggregation possibilities under Directive 2014/24 could have, and proposes that strict competition law enforcement will be necessary to avoid undesired consequences. Some suggestions for further research are provided by way of conclusions.
The full paper is available for download on SSRN. Its full citation is:

Sánchez Graells, Albert and Herrera Anchustegui, Ignacio, Impact of Public Procurement Aggregation on Competition. Risks, Rationale and Justification for the Rules in Directive 2014/24 (December 5, 2014). University of Leicester School of Law Research Paper No. 14-35. Available at SSRN:

Competition lawyers, please, please, please be aware of public procurement rules: A comment on Bornico & Walden (2011)

I have just read L Bornico & I Walden, 'Ensuring Competition in the Clouds: The Role of Competition Law?' (2011) 12(2) ERA Forum 265-85 (part of the largest QMUL Cloud Legal Project) and have been, once more, surprised at the complete oversight of the public procurement rules that would have been relevant to the competition law analysis.

The paper engages in an exploratory analysis of the role of EU competition law could have in keeping the cloud computing industry competitive and, if possible at all, free from (potential) abuses of dominance by its main players. The paper has the good intuition to suggest that public procurement decisions by governments may play a key role in either the promotion of undistorted competition (if they opt for transparent standards based on interoperability) or, on the contrary, the creation of a very concentrated and potentially monopolistic market structure (if they unduly impose specific technological solutions). This is a very important point, and one that public procurement economists and commentators have been stressing for a long time.

However, when the paper moves on to suggest how to legally prevent and control those issues, it is completely oblivious to the existence of EU public procurement rules. Indeed, Bornico & Walden indicate that, where the contracting authority imposes a given (propietary) technological solution
... harmed competitors may challenge the choice of the public administration alleging that their specifications fit best the requirements ... or may challenge the behaviour of the firm whose specifications were chosen, but only if the firm can be considered dominant. More importantly competitors may challenge the choice of the public administration under Article 107 TFEU if the outcome of public procurement distorts competition.  The choice of formal specifications may soon be a source of disputes in the EU market, along the lines of the Google dispute in the US [by reference to  Google Inc. and Onix Networking Corporation v. The United States and Softchoice Corporation (United States Court of Federal Claims 2011)]; although it is too early to tell how technological choices made by public administrations will be dealt with by competition authorities in the EU. (p. 27, emphasis added).
There are three important points to stress here. Firstly, this is nothing new, but State aid litigation based on public procurement decisions is very limited, generally unsuccessful, and likely to be 'phagocytised' by 'pure' procurement litigation [for an extended discussion, see A Sanchez Graells, 'Enforcement of State Aid Rules for Services of General Economic Interest before Public Procurement Review Bodies and Courts' (2014) 10(1) Competition Law Review 3-34]. 

Secondly, aggrieved competitors would have a much better shot under the applicable rules on the design of technical specifications. Indeed, it has long been the position of the ECJ, now consolidated in the applicable Directives 2004/18 (and/or 2014/24, where transposed), that '[u]nless justified by the subject-matter of the contract, technical specifications shall not refer to a specific make or source, or a particular process, or to trade marks, patents, types or a specific origin or production with the effect of favouring or eliminating certain undertakings or certain products. Such reference shall be permitted on an exceptional basis, where a sufficiently precise and intelligible description of the subject-matter of the contract ... is not possible; such reference shall be accompanied by the words "or equivalent".' (emphasis added) [art 23(8) dir 2004/18, and now art 42(4) dir 2014/24; for discussion, see S Arrowsmith, The Law of Public and Utilities Procurement. Regulation in the EU and UK, 3rd edn, vol. 1 (London, Sweet & Maxwell, 2014) 254-55 and 1068 ; and A Sanchez Graells, Public procurement and the EU competition rules (Oxford, Hart Publishing, 2011) 271-72]. Consequently, specific technological choices that excluded equivalent solutions would immediately be in breach of EU public procurement rules.

Thirdly, a breach of those rules gives aggrieved bidders and other interested economic operators a ground to challenge the procurement procedure before domestic courts or procurement complaints boards, under the provisions of Directives 89/665 and 92/13 (as amended by dir 2007/66). This is a much clearer litigation path and one that would yield much better results to disappointed bidders and competing (technological) firms.

Consequently, in this specific area, competition law is not the best tool to achieve pro-competitive results in the public procurement setting. Public procurement law is. So, competition lawyers, please familiarise yourselves with public procurement rules. In the end, they are two sides of the same coin [C Munro, ‘Competition Law and Public Procurement: Two Sides of the Same Coin?’ (2006) 15 Public Procurement Law Review 352; and A Sanchez Graells, 'Competition Law Against Public Restraints in the Public Procurement Field: Importing Competition Considerations into the EU Public Procurement Directives' (2010)].