Is Altmark not applicable to ambulance services? Or is San Lorenzo & Croce Verde not very clear? (C-113/13)

In view of the Judgment in San Lorenzo and Croce Verde Cogema, C-113/13, EU:C:2014:2440, the rules applicable to the provision of emergency ambulance services is definitely clear as mud. In the case at hand, the applicants challenged an Italian law whereby emergency ambulance services must be awarded on a preferential basis and by direct award, without any advertising, to certain voluntary bodies (such as the Red Cross). This rule has, ultimately, constitutional protection in Italy, as 'the Italian Republic has incorporated into its constitution the principle of voluntary action by its citizens. Thus, the last paragraph of Article 118 thereof provides that citizens, acting individually or in an association, may participate in activities of public interest with the support of the public authorities, on the basis of the principle of subsidiarity' (para 9).

The applicants' argument was not necessarily of a constitutional level, but rather that freedom of establishment is unduly restricted by a preferential scheme that excludes the tendering out of those ambulance services. They brought forward arguments based on general free movement provisions, public procurement rules and competition rules. The latter are not examined because the CJEU considered that the public procurement analysis makes it unnecessary (para 64).

In my view, if read paragraph by paragraph, the reasoning of the CJEU is accurate and technically precise, but the overall Judgment is too timid in spelling out the conditions for the application of the 'public service exception' under art 106(2) TFEU (or otherwise) tot he direct award of emergency ambulance services to voluntary action associations. I will try to summarise my criticism and doubts as succinctly as possible. This is an area where more considered research is definitely needed.

On the bright side, I think that some positions of the CJEU can be clearly spelled out. 

(1) When fully applicable, both Dir 2004/18 and Dir 2014/24, preclude legislation such as that at issue in the main proceedings which provides that the local authorities are to entrust the provision of urgent and emergency ambulance services on a preferential basis and by direct award, without any advertising, to the voluntary bodies mentioned in the agreement (para 44). However, Dir 2004/18 does not automatically apply to ambulance services (see 2 below) and art 10(h) Dir 2014/24 clearly excludes these contracts from its scope of application (para 8). Hence, this clear position is not that useful in practice.

(2) Where the Directives are not fully applicable (ie where contracts can be tendered under part B services rules under dir 2004/18, or under the special regime for social services under arts 74-77 dir 2014/24), the general principles of transparency and equal treatment flowing from articles 49 TFEU and 56 TFEU would be applicable (para 45) if the contract is of cross-border interest (paras 46-50). In that case, it is also clear that such a preferential scheme would run contrary to the Directives, which are: 'intended to ensure the free movement of services and the opening-up to competition in the Member States which is undistorted and as wide as possible' (para 51). 

(3) Implicitly, then, where the Directives do not apply at all but the contract is still of cross-border interest (ie the new likely situation under art 10(h) dir 2014/24), the award of the contract is 'merely' subjected to the (residual/general) requirements of articles 49 TFEU and 56 TFEU. In that case (not expressly assessed in the San Lorenzo & Croce Verde Judgment), the contracting authority still would need to go through the applicable assessment under the market access test generally applicable to restrictions of freedom of establishment [for two thought provoking attempts to rationalise this test, see E Christodoulidis, 'The European Court of Justice and Total Market Thinking' (2013) 14 German Law Journal 2005; and MS Jansson & H Kalimo, 'De minimis meets “market access”: Transformations in the substance – and the syntax – of EU free movement law?' (2014) 51(2) Common Market Law Review 523].

Hence, there are always concerns and constraints derived from EU law (either general, or the specific rules of public procurement) if the contract is of cross-border interest. Nonetheless, they are of varying degrees of intensity and it looks as if upon the entry into force of Directive 2014/24, the award of service contracts for emergency ambulance services  (either exclusively, or for most of their value if the contracts include other sorts of ambulance services) will exclusively be governed by the general rules on freedom of establishment.

On the shady side, though, once the potential incompatibility with EU public procurement or general free movement law is established (and, really, there seems to be no escape to 1-3 above except if the contract has no cross-border interest whatsoever--and, on that, see the  Ancona issue here), the CJEU will apply a Sodemare-like test because the provision of ambulance services falls within the (very broad) remit of the organisation of healthcare and social security systems (paras 55-59). In that case, then, it will particularly important that "EU law does not detract from the power of the Member States to organise their public health and social security systems" (para 55), but that "it is for the Member States, which have a discretion in the matter, to decide on the degree of protection which they wish to afford to public health and on the way in which that degree of protection is to be achieved" (para 56). So far, so good.

On the dark side, however, and significantly departing from the more developed approach in Altmark for SGEIs (is the CJEU implicitly recognising--without analysis--that ambulance services are per se SSGIs?), the CJEU has created an economically oriented safeguard that leaves too much room for maneuver by ruling that
Having regard to the general principle of EU law on the prohibition of abuse of rights (see, by analogy, judgment in 3M Italia, C‑417/10, EU:C:2012:184, paragraph 33), the application of that legislation cannot be extended to cover the wrongful practices of voluntary associations or their members. Thus, the activities of voluntary associations may be carried out by the workforce only within the limits necessary for their proper functioning. As regards the reimbursement of costs, it must be ensured that profit making, even indirect, cannot be pursued under the cover of a voluntary activity and that volunteers may be reimbursed only for expenditure actually incurred for the activity performed, within the limits laid down in advance by the associations themselves (C-113/13, para 62, emphasis added).
In my view, this is way too timid. Indeed, the CJEU constructs a rather weak safeguard by not focussing at all in the economic efficiency of the voluntary activities (which, even on a non-profit, reimbursement basis can be extremely inefficient) and imposes a sort of 'anti-fraud' test that, in my view, misses the point. In order to ensure compatibility with State aid provisions (which should not have been set aside so quickly in para 64), an efficiency based test like the one existing in the fourth condition of Altmark should have been imposed [for discussion, see A Sanchez Graells, “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]. 

Indeed, the analysis of the applicability of Art 106(2) TFEU to the case is totally missing and this is strange. It looks like the difference between SGEIs and SSGIs will haunt all of us also under the 2014 Directives and revised guidance from the European Commission is becoming urgently needed, given the implicit vacuum that can exist if Member States maximise the possibilities of direct award under art 10(h) dir 2014/24, but equally under its arts 74-77 (and particularly, the latter).

As briefly mentioned, this is an area where more research is needed. I hope I can convince some colleagues to put together a research project on this soon. Interested contributors, please feel free to contact me at

Risk of 'sweet deals' for public sector #mutuals under the new #EU #publicprocurement rules leading to #monopolisation of social services

Article 76a of the final compromise text for a new Directive on public procurement creates a significant risk of abuse in the award of contracts for social services that may be of particular concern when read side by side with the UK Government's public sector mutualisation strategy.

Article 76a allows contracting authorities to reserve for the participation of given types of organisations (such as ‘public sector mutuals’, for instance) the award of contracts for certain services included in specific categories of the Common Procurement Vocabulary[1] in the areas of health, social and cultural services[2]—which basically comprise all or the most relevant medical services, personal services, educational and training services (including eLearning), sports and cultural services. 

In such cases, the contracting authority will need to make sure that the (type of) organisation chosen to be awarded the contract meets the following requirements: (a) its objective is the pursuit of a public service mission linked to the delivery of the services to be contracted; (b) its profits are reinvested with a view to achieving the organisation’s objective (and where profits are distributed or redistributed, this should be based on participatory considerations); (c) the structures of management or ownership of the organisation performing the contract shall be based on employee ownership or participatory principles, or shall require the active participation of employees, users or stakeholders; and (d) the organisation shall not have been awarded a contract for the services concerned by the contracting authority concerned pursuant to this Article within the past three years. Moreover, the maximum duration of the contract shall not be longer than three years and the call for competition shall make reference to this Article. There is no (maximum) value threshold for this exclusion to be effective.

Given that the Mutuals Taskforce has clearly recommended that the UK Government use the (direct) award of contracts as a tool to support, foster and consolidate the creation of public sector mutuals (Recomm 9), there is a risk that the carve-out (negotiated?) in the new EU public procurement rules leaves the award of such 'start-up contracts' fundamentally unchecked, since there will basically be no EU rules applicable in this case, regardless of the value of the contracts [which is also valid in relation to the EU rules on the control of State aid, which exclude health care and other social services from their scope of application; see art 2 of Decision 2012/21/EU].

Therefore, there is a clear risk that the public sector reform strategy ends up creating (3-year long, local) monopolies for the provision of those services in the hands of the newly spun-off public sector mutuals, which may extend their dominance beyond that point in time as incumbency advantages pile up. That would result in distortions of competition similar to those just identified by the Competition Commission in the market enquiry on private health care services and, in my view, is an undesirable prospect. 

In that regard, then, the OFT (and, soon enough, the CMA) seems to have a mounting amount of pressure to uphold its commitment in the 2013-14 Annual Plan to "focus on IT and local government issues in particular and work with government partners on a range of issues relating to the public sector reform agenda to ensure that government interventions maintain competitive markets. In addition to advocacy and influencing, [the OFT] will consider using the full range of tools at our disposal to tackle any breaches of competition law identified in public service markets" (emphasis added).

Otherwise, the structural changes that non-competitive mutualisation under the umbrella of Article 76a of the new EU public procurement rules can create may be difficult (if not impossible) to revert in the future. At any rate, however, the difficulty derived from the blurred application of competition rules in the health care sector after the National Health Service (Procurement, Patient Choice and Competition) (No. 2) Regulations 2013 were approved may require some creative enforcement efforts on the part of the OFT (and the European Commission).

[1] See the consolidated version available at, last accessed 2 September 2013.
[2] The specific references are 75121000-0 (Administrative educational services), 75122000-7 (Administrative healthcare services), 75123000-4 (Administrative housing services), 79622000-0 (Supply services of domestic help personnel), 79624000-4 (Supply services of nursing personnel), 79625000-1 (Supply services of medical personnel), 80110000-8 (Pre-school education services), 80300000-7 (Higher education services), 80420000-4 (E-learning services), 80430000-7 (Adult-education services at university level), 80511000-9 (Staff training services), 80520000-5 (Training facilities), 80590000-6 (Tutorial services), from 85000000-9 to 85323000-9 (fundamentally, all types of medical services), 92500000-6 (Library, archives, museums and other cultural services), 92600000-7 (Sporting services), 98133000-4 (Services furnished by social membership organisations), and 98133110-8 (Services provided by youth associations). 

#CJEU shows excessive deference towards #socialpolicy in #publicprocurement: #socialhousing schemes may not be public contracts (C-197/11)

In its Judgment of 8 May 2013 in Joined Cases C-197/11 & C-203/11 Libert and Others, the Court of Justice of the European Union has quashed the Belgian 'Living in Your Own Region' scheme, whereby the acquisition of land and property in certain parts of the country was restricted on grounds of social policy. Moreover, when a building or land subdivision authorisation was granted, Belgian law imposed 'social obligations' on economic operators, such as subdividers and developers, so that they had to either pay a contribution to the commune (in cash or in kind), or discharge such obligation through certain sale and lease schemes, or any combination thereof. Interestingly, the CJEU has found that this latter part of the 'authorisation + social obligation' scheme, which potentially implies mandatory sales of property to public entities, may not fall under the scope of Directive 2004/18 on public procurement.

In the first part of its Libert Judgment, the CJEU has followed its prior approach against restrictive development schemes in Konle (C-302/97), and has quashed a decree of the Flemish Region of 27 March 2009 on land and real estate policy that linked the transfer of immovable property in certain Flemish communes to the condition that there existed a 'sufficient connection' between the prospective buyer or tenant and the relevant commune--ie that provided for a prior authorisation procedure to determine whether there was such a ‘sufficient connection’ (based on previous residence or on professional, family, social or economic connections with the commune in question) and, in reality, amounted to prohibiting certain persons from purchasing or leasing for more than nine years land or the buildings thereon (see CJEU press release).

It is interesting to see that the Belgian Government had tried to justify the scheme on social policy grounds and, particularly, by the objective of responding to the housing needs of the less affluent local population in the target communes. However, the CJEU has decided to quash the scheme for its lack of proportionality. The Court has rightly considered that none of the conditions used to assess the existence of a 'sufficient connection' directly reflected the socio-economic aspects relating to the objective of protecting exclusively the less affluent local population on the property market--since they could be met not only by the less affluent local population but also by other people with sufficient resources who, consequently, had no specific need for social protection on the property market. Moreover, such social goal could be achieved through other, less interventionist measures, such as housing subsidies specifically designed to assist the less affluent. So far, the case does not depart from what could be expected from the CJEU.

In its second part of the Libert Judgment, the CJEU considered that the mandatory discharge of social contributions by developers and subdividers can be a justified restriction of the fundamental freedoms involved (mainly, free circulation of capital). The Court considered that such a restriction, in so far as its purpose is to guarantee sufficient housing for the low-income or otherwise disadvantaged sections of the local population, may be justified by requirements relating to social housing policy in a Member State as an overriding reason in the public interest (and left it for the referring court to assess whether such an obligation satisfies the principle of proportionality, that is to say, whether it is necessary and appropriate to attain the objective pursued). Again, the CJEU has followed an approach that was to be expected.

The CJEU then engages in an assessment of a part of the social housing scheme that implies tax incentives and subsidy mechanisms for developers from a State aid perspective. However, the CJEU has very limited information and limits itself to refresh the conditions for the exemption of aid, with an express reference to the old 2005 SGEI Decision--now substituted by the 2011/12 'Almunia Package' [for commentary, see Sánchez Graells, 'The Commission’s Modernization Agenda for Procurement and SGEI' in Szyszczak & van de Gronden (eds.), Financing SGEIs: State Aid Reform and Modernisation (TMC Asser Press/Springer Series Legal Issues of Services of General Interest, 2012)]. 

The key issue in this area clearly seems to be how to determine the remit of the 'public service'/ 'social service' obligation, and to find a valid benchmark to ensure that there is no excessive remuneration. In this regard, it is worth stressing that, following Altmark (C-280/00), the Commission is increasingly relying on the conduct of public procurement procedures to exclude the existence of an undue economic advantage for the provide of the 'public service'/ 'social service'. Consequently, the following part of the case becomes all the more relevant.

In that regard, it is key to stress that, in the final part of Libert, the CJEU has addressed whether this authorisation scheme fell within the remit of the EU public procurement rules. The CJEU was asked the following question:
'Should the concept of “public works contracts” in Article 1(2)(b) of Directive 2004/18… be interpreted to mean that it is applicable to a scheme whereby, when a building or land subdivision authorisation is granted in respect of a project of a certain minimum size, it is linked by operation of law to a “social obligation” entailing the development of social housing units, amounting to a certain percentage of the project, which are subsequently to be sold at capped prices to a public institution, or with substitution by it?'
In its (partial) reply to this question, the CJEU heavily relies on the (in)existence of an express written agreement, and considers that
109 [...] it should be borne in mind that, in accordance with Article 1(2)(b) of Directive 2004/18, read in conjunction with Article 1(2)(a) thereof, public works contracts result where four criteria are fulfilled, that is to say, they are contracts for pecuniary interest, concluded in writing, between an economic operator and a contracting authority, which must have as their object either the execution, or both the design and execution, of works related to one of the activities within the meaning of Annex I to that directive or a work, or the realisation, by whatever means, of a work corresponding to the requirements specified by the contracting authority. [...]
111 So far as concerns, in particular, the existence of a contract concluded in writing, it follows from the order for reference that the Constitutional Court is uncertain as to whether that criterion has been met in the present case, inasmuch as the social obligation entailing the development of social housing units is imposed in the absence of an agreement concluded between the housing authorities and the economic operator concerned. According to the order for reference, the social obligation is imposed directly on subdividers and developers by the Flemish Decree and is applicable to them merely because they own the land in relation to which they have applied for the grant of a building or land subdivision authorisation. 
112 In that regard, it should be borne in mind that, in order to establish that some kind of contractual relationship existed between an entity which could be regarded as a contracting authority and a subdivider or developer, the case-law of the Court requires [...] a development agreement to be concluded between the housing authorities and the economic operator in question for the purpose of determining the work to be undertaken by the economic operator and the terms and conditions relating thereto
113 Where such an agreement has been concluded, the fact that the development of social housing units is a requirement imposed directly by national legislation and that the party contracting with the authorities is necessarily the owner of the building land in question does not preclude the existence of a contractual relationship between the authorities and the developer in question (see, to that effect, Case C‑399/98 Ordine degli Architetti and Others [2001] ECR I‑5409, paragraphs 69 and 71). 
114 However, although it is true that Article 4.1.22, first subparagraph, of the Flemish Decree expressly requires an administration agreement to be concluded between the subdivider or developer and the social housing organisation, it is apparent from the order for reference that that agreement does not, in principle, regulate the relationship between the contracting authority and the economic operator concerned. In addition, such an agreement does not appear to concern the development of social housing units, but only the next stage which entails placing them on the market
115 It is therefore for the referring court to determine, in the light of all the applicable legislation and the relevant circumstances of the case in the main proceedings, whether the development of social housing units at issue in the main proceedings is within the framework of a contractual relationship between a contracting authority and an economic operator and whether the criteria referred to in paragraph 109 above have been met. (C-197/11 at paras 109 to 115, emphasis added).
In my opinion, the CJEU leaves the door excessively open to a finding that the scheme is not covered by Directive 2004/18 due to the lack of a 'proper' or 'sufficient' contract. In my view, for the purposes of controlling the award of the rights to divide land, develop property, and then sell it to a public institution in the social housing market, the CJEU should have adopted a more functional approach and indicated that the submission of an authorization followed by its approval (by means of an administration agreement) suffices for the establishment of a contractual relationship between the authority and the developer or subdivider, which would then automatically include the 'social obligation'. 

Otherwise, there seems to be excessive room for strategic behaviour on the part of contracting authorities to avoid compliance with public procurement rules through 'unstructured' documentation of their planning decisions and their economic relationships with developers. Given that only those developers that obtain authorisation will be able to develop and then sell property to public institutions under the 'social housing' scheme, it seems clear that the award of the authorisation implies an economic exchange (or that they are concluded for a pecuniary interest, even if it is deferred) and that the authority will (indirectly, at least) benefit from the development in the execution of its social housing policy. Therefore, the CJEU seems to have adopted a (mild) formalist approach that shows deference towards the implementation of social policies via (quasi)procurement schemes.

The only (implied) safeguard to this approach would be that the CJEU assumed that there would be a procurement procedure when the developments are complete and the property is 'sold at capped prices to a public institution, or with substitution by it', ie in the 'next stage which entails placing them on the market' (para 114). However, that seems highly unlikely, given that the administrative agreement entered into with the developer at the time of granting the authorisation seems to cover that--and, in any case, the contracting authority would probably try to rely on aspects of 'exclusivity' due to the location of the property, or excuse the purchase as a legal obligation, in order to avoid procurement procedures at that stage.

Moreover, in my opinion, the CJEU goes out of its way to provide the referring court with several additional reasons why, even in the presence of a written contract, the scheme may not be covered by Directive 2004/18. As the Court stresses, the scheme may well be below the relevant thresholds (which is a fair remark), or be exempted as part of an in-house scheme or a public-public cooperation scheme (see the European Commission's guidance for further details). These latter considerations are unwarranted by the question referred to the CJEU, which shows no element of collaboration between contracting authorities.  Indeed, the CJEU notes that:
116 [...] on the one hand, the application of Directive 2004/18 to public works contracts is nevertheless subject to the condition that the estimated value of the contract reaches the threshold set out in Article 7(c) of that directive and that, on the other, there are, as is apparent from the settled case-law of the Court, two types of contracts entered into by a public entity that do not fall within the scope of EU public procurement law
117 The first type of contracts are those concluded by a public entity with a person who is legally distinct from that entity [under the in-house exemption] where, at the same time, that entity exercises over the person concerned a control which is similar to that which it exercises over its own departments and where that person carries out the essential part of its activities with the entity or entities which control it (see Case C‑159/11 Ordine degli Ingegneri della Provincia di Lecce and Others [2012] ECR I‑0000, paragraph 32 and the case-law cited). 
118 The second type of contracts are those which establish cooperation between public entities with the aim of ensuring that a public task that they all have to perform is carried out. In those circumstances, the EU rules on public procurement are not applicable in so far as, in addition, such contracts are concluded exclusively by public entities, without the participation of a private party, no private provider of services is placed in a position of advantage vis-à-vis competitors and implementation of that cooperation is governed solely by considerations and requirements relating to the pursuit of objectives in the public interest (see Ordine degli Ingegneri della Provincia di Lecce and Others, paragraphs 34 and 35). (C-197/11 at paras 105 to 119).
 The Court concludes with a tautology:
In the light of all the foregoing considerations, the answer to the eleventh question in Case C‑203/11 is that the development of social housing units which are subsequently to be sold at capped prices to a public social housing institution, or with substitution of that institution for the service provider which developed those units, is covered by the concept of ‘public works contract’ contained in Article 1(2)(b) of Directive 2004/18 where the criteria set out in that provision have been met, a matter which falls to be determined by the referring court (C-197/11 at para 119, emphasis added).
In my opinion, the answer provided by the CJEU to this last question in the Libert case is at the same time too vague and too lenient with the social housing scheme under consideration, and opens a door to a dangerous path of limited or non-application of public procurement rules in relevant areas of social policy and social services, such as social housing. In that regard, it will be important to see how this (emerging?) trend of case law gets squared with the foreseeable future rules applicable to the procurement of social services of general interested, such as those included in the 2011 Proposal for a new Public Procurement Directive (art 74 to 76). 

Definitely, both avenues of development of public procurement law point towards a light-touch regulation of social services / social policy-related procurement, but I am not sure that they are moving in the same direction, since the CJEU seems to be keen to exclude (certain) 'social' contracts from procurement rules, whereas the Commission would like to keep some (restricted) control over them. It will be highly relevant to see how this area develops in the near future, since legal uncertainty is bound to be coupled with intense executive action.