AG proposes extension of Falk Pharma doctrine to framework agreements, for wrong reasons (C-9/17)


In his Opinion of 13 December 2017 in Tirkkonen, C-9/17, EU:C:2017:962 (not available in English), Advocate General Campos Sanchez-Bordona has proposed the application to a framework agreement for the provision to farmers of advisory services funded by the European Agricultural Fund for Rural Development (FEADER) of the Falk Pharma doctrine (ie that the absence of a choice in concreto of the awardee of a contract by the contracting/funding authority excludes the applicability of the EU public procurement rules; see Judgment of 2 June 2016 in Falk Pharma, C-410/14, EU:C:2016:399, and here).

In his view, the fact that individual farmers—and not the competent authority administering the FEADER funds—could choose the specific rural advisor that would provide them the services carved the framework agreement out of the scope of application of the EU (and domestic) public procurement rules—which were therefore not applicable to the tender of the framework agreement in the first place.

In my view, the Tirkkonen Opinion engages in an unjustifiably expansive interpretation of the Falk Pharma Judgment that both ignores some of the basic elements in the functioning of framework agreements, and takes that Judgment’s functionally-erroneous interpretation of the concept of procurement one step too far. If the Tirkkonen Opinion was followed, in combination with Falk Pharma, it would create a significant risk of ineffectiveness of the EU public procurement rules for aggregate and dynamic contracting mechanisms. Therefore, in this post, I present my reasons for a plea to the Court of Justice of the European Union (ECJ) not to follow AG Campos in this occasion, as I think his approach is problematic, both from a positive and a normative perspective.

Tirkkonen – a bad case raising the wrong issues

Why ignore explicit requirements in secondary EU law?

The way the preliminary reference in Tirkkonen reached the ECJ evidences that this is a bad set of circumstances on which to develop the case law on the scope of application of the EU public procurement rules. In the case at hand, the Finnish Agency for Rural Space (Maaseutuvirasto) tendered a framework contract for the provision of advisory services to farmers. Given the (expected) high volume of demand for advisory services, the framework was intended to include as many qualified rural advisors as possible, subject to their passing of an exam to ensure their knowledge and competence (AGO, C-9/17, para 19). Rural advisors admitted to the framework agreement could then be chosen by individual farmers (who should in principle chose the closest advisor, although some exceptions applied), and their services would be remunerated on the basis of hourly rates paid by Maaseutuvirasto, with the beneficiary farmer covering applicable VAT charges (AGO, C-9/17, para 18). It is not explicitly stated in AG Campos' Opinion, but it is worth stressing that the Maaseutuvirasto had set the hourly rate payable to rural advisors, and that the award (ie admission to the framework contract) was to be decided solely on quality (ie competency to provide the service) (see here for details (in Finnish), and thanks to K-M Halonen for help with the translation). The suppression of price competition will be relevant for the assessment below.

The advisory services organised by Maaseutuvirasto were ultimately funded by FEADER for the period 2014-2020 and, under the relevant rules (Reg 1305/2013/EU, Art 15(3), and Impl Reg 808/2014/EU, Art 7), the selection by the Finnish (and all other national) competent authority of the providers of those advisory services was explicitly subjected to European and domestic public procurement rules, which required for the selection to be made: ‘through calls for tenders. The selection procedure shall be governed by public procurement law and shall be open to both public and private bodies’ (Art 15(3) Reg 1305/2013/EU). It was reiterated that the 'calls for tenders referred to in Article 15(3) of Regulation (EU) No 1305/2013 shall follow the applicable Union and national public procurement rules' (Art 7 Impl Reg 808/2014/EU). The Finnish government had no doubt that EU and domestic procurement rules applied, and thus tendered the contract as described above.

Therefore, against this background, a preliminary reference enquiring about the potential non-applicability of the procurement rules to the tender of the framework agreement despite the explicit requirements in special (in the sense of lex specialis) secondary EU legislation is beyond bizarre (see below). However, AG Campos does not see a problem here, and considers that

… that reference to procurement law must be interpreted in the sense that the procedure for the selection of rural advisors must comply with the principles (of non-discrimination, equal treatment and transparency) that govern that sector of the legal order. It does not portray, in my view, a requirement that implies subjection to each and all of the provisions of the EU Directives on public sector procurement (AGO, C-9/17, para 34, own translation from Spanish).

I disagree with this assessment, which is not based on any specific reasons, and which violates the natural reading of Reg 1305/2013/EU and Impl Reg 808/2014/EU. Moreover, it comes to reduce the value of the explicit reference to procurement law in those provisions, and to collapse it into the general principles that are common with general internal market law and, more importantly, the eponymous general principles of EU law—which would be applicable anyway to all activities implementing the relevant instruments of secondary EU law. Therefore, AG Campos’ position not solely deviates from the natural reading of the provisions, but also runs contrary to the functional reasons for the inclusion of the explicit reference to procurement rules (ie to go beyond the general requirements of the always applicable primary EU law). Thus, already on the weakness of the reasons for a deviation from the literal and functional interpretation of those provisions of secondary EU law, I think that the ECJ should largely ignore AG Campos’ Opinion and simply answer the question by confirming the applicability of the EU (and domestic) procurement rules on the basis of the explicit requirements in Reg 1305/2013/EU and Impl Reg 808/2014/EU.

Why not simply state that Finnish procurement law was wrong?

Beyond that first clear-cut solution, which I think highly unlikely the ECJ will adopt, the Court will have to explore the general (as in lex generalis) reasons that still justify the applicability of the EU and (domestic) procurement rules to the case—also contrary to AG Campos’ Opinion. To that end, it is still necessary to understand why the preliminary question was sent to the ECJ—which is explained by a misconstruction of the EU public procurement rules and, in particular, by the harsh consequences of an exceedingly restrictive approach to documentary clarification in the domestic Finnish procurement rules that violates the Manova-Slovensko line of case law (see here, here and here).

In that regard, it is worth noting that the preliminary reference derived from the fact that, in the context of the tender for the framework agreement, Ms Tirkkonen failed to properly complete all required documentation—ie she had failed to indicate whether she accepted or rejected the tender conditions attached to the draft framework agreement (AGO, C-9/17, para 20). She was thus excluded from the framework agreement. Her complaint is fundamentally grounded on the fact that she should have been given the opportunity to clarify whether she accepted the conditions or not prior to her exclusion from the framework agreement.

It is a settled legal fact of the case that, under Finnish law, the omission of that indication of acceptance of the general conditions would only be susceptible if the clarification or correction of the tender was not controlled by public procurement law (which excluded such possibility of clarification), and was rather subjected to general administrative law governing the relationships between citizens and the public administration (AGO, C-9/17, para 3).

Therefore, the harshness of the Finnish procurement rules is behind the interest of the claimant in excluding the tender from the scope of application of domestic procurement rules—which can only be done by seeking a carve-out from the concept of procurement under the EU rules. And, more importantly, the Finnish approach is in contravention of EU law—oddly, as confirmed by AG Campos himself: ‘if Directive 2004/18 was applicable, it would result that the contracting authority would be able to accept, in the context of public procurement, the correction of formal shortcomings that do not imply the submission of a new offer, or substantially altered the terms of the initial offer. On this point, I refer to my Opinion in case MA.T.I. SUD y DUEMMESGR (C-523/16 y C-536/16, EU:C:2017:868)’ (AGO, C-9/17, para 23, fn 7, own translation from Spanish; for discussion of MA.T.I. Sud, see here).

Consequently, the second clear-cut solution for the ECJ is to (i) pick up on the incorrect interpretation of EU public procurement law that underpins the preliminary reference, (ii) reformulate the question and consider that it asked whether the exclusion from the framework agreement due to the formal shortcoming in the documentation and without the possibility to correct it was required or allowed by EU procurement law, (iii) reiterate the Manova-Slovensko case law, and (iv) leave it for the national court to decide on the legality of the exclusion (with a clear hint that exclusion in this case was not justified, due to the logical assumption that would-be rural advisors understood that accepting the general conditions of the draft contract was a requirement for entering into specific contracts, and that confirming such acceptance does not constitute a new offer or substantial modification of the initial offer).

For some reason, however, I am also not optimistic that the ECJ will adopt this second solution and pass on the opportunity to clarify its Falk Pharma case law. Should the ECJ engage with the question and the issues raised by AG Campos, and for the reasons below, I think that the ECJ should provide clarification of Falk Pharma in the opposite direction to that adopted by the Tirkkonen Opinion.

Tirkkonen Opinion ignores how framework agreements work

Once the argument concentrates on the definition of procurement under Article 1(2)(a) of Directive 2004/18/EC, AG Campos summarises the Falk Pharma doctrine as establishing that

… the choice of a tender and, thus, of a successful tenderer, is intrinsically linked to the regulation of public contracts by that directive and, consequently, to the concept of ‘public contract’ within the meaning of Article 1(2) of that directive (AGO, C-9/17, para 37, own translation from Spanish, with reference to Falk Pharma, para 38).

And that

… in the public contracts subjected to Directive 2004/18 a final awardee must exist, which is preferred to the rest of its competitors on the basis of the characteristics of its offer. And this key element is applicable ‘for every contract, framework agreement, and every establishment of a dynamic purchasing system’, for which ‘the contracting authorities are to draw up a written report which is to include the name of the successful tenderer and the reasons why his tender was selected (AGO, C-9/17, para 38, own translation from Spanish, with reference to Falk Pharma, para 39).

This leads AG Campos to argue that, in the framework tendered in Tirkkonen, ‘it is not possible to identify the existence of award criteria of the advisory services contracts, but solely of criteria for the selection of economic operators with capability to offer those services (sic)’ (AGO, C-9/17, para 39, own translation from Spanish and emphasis added). AG Campos continues with a discussion of the distinction between selection and award criteria as per Ambisig (C-601/13, EU:C:2015:204, paras 40 and ff, see here), which I consider irrelevant—for the crucial point is that, in multi-supplier framework agreements (as well as in dynamic purchasing systems, as discussed here), the inclusion in the framework does not (ever) imply the choice of the ‘winner’ of the (call-off) contracts but, conversely, exclusion from the framework does prevent the excluded economic operators from providing the service.

In my view, this is the relevant aspect, for the inclusion in the framework is not simply an identification of the capable or qualified economic operators, but the limitation to those included in the framework of the possibility of entering into specific contracts in the terms set in the framework. AG Campos’ maximalistic position would lead to the inescapable logical conclusion that framework agreements are not public contracts for the purposes of EU public procurement law, despite being explicitly regulated, quod non.

The flawed logic of the premise established by AG Campos in para 39 of his Opinion makes the rest of his reasoning crumble. In my view, this defect affects his reasoning that

… what is determinative, in relation to the contracts subject to Directive 2004/18, is not the checking of the economic operators’ capability to provide the advisory service (qualitative selection criterion), but the comparison of the offers of the competing tenderers, once considered capable, with a view to finally chose that or those which will be entrusted with such provision (award criterion) (AGO, C-9/17, para 44, own translation from Spanish).

And that

… the selection that matters, for the purposes of the concept of public contract in Directive 2004/18, is that which results from the comparison between the capabilities and merits of the offers of the different candidates. That is, what is decisive is the final award, comparatively or by contrast, to the best offer, not the initial selection by reference to a threshold meeting which does not imply competition between the candidates (AGO, C-9/17, para 45, own translation from Spanish).

Ultimately, following this same reasoning, AG Campos takes issue with the fact that there was no competition between the candidates that expressed interest in being included in the framework agreement because the contracting authority ‘did not restrict ab initio the number of potential providers of the services, nor did it carry out a comparison of the offers between them, or chose in a definitive manner one or several of them, on the basis of a comparative evaluation of their respective contents, to the exclusion of the rest’ (AGO, C-9/17, para 48, own translation from Spanish).

However, this triggers two issues. First, under Dir 2004/18/EC, there was no obligation to establish a maximum number of economic operators to be admitted to a framework agreement. Art 32(4) Dir 2004/18/EC solely established a minimum of three for multi-supplier framework agreements, but did not require a maximum number. Second, it is in the nature of framework agreements—particularly those involving mini-competitions, as per Art 32(4)II Dir 2004/18/EC—that the contracting authority, at the point of deciding which economic operators are included in the framework, does not ‘chose in a definitive manner one or several [offers], on the basis of a comparative evaluation of their respective contents, to the exclusion of the rest’ for the purposes of the award of the relevant call-off contracts—which is the situation comparable to Tirkkonen. In particular, it is possible that an economic operator included in a framework agreement is never awarded a call-off, especially if there are mini-competitions, which in my view deactivates the functional reasoning of AG Campos.

In my view, AG Campos also misinterprets the implications of the fact that the framework agreement in Tirkkonen was closed to the economic operators not initially admitted to it, in relation to the ECJ’s Judgment in Falk Pharma. In that regard, it is relevant that the argument was made that the closed nature of the framework agreement distinguishes it from the open-ended mechanism discussed in Falk Pharma, which AG Campos rejects in the following terms:

It is true that, strictly, by limiting the contracting system, during its term, to the economic operators initially admitted by the Agency [Maaseutuvirasto] (which prevents access by new advisors) a certain quantitative restriction is being imposed. However, this is but a consequence of the pure and rigorous temporary limitation of the system of funding for advisory services, which is parallel to the program of rural development for continental Finland 2014-2020 (sic).

For the rest, the reference by the Court of Justice in Falk Pharma to the permanent openness of the contracting system to new tenderers was not, in my view, the ratio decidendi of that case, but rather a statement made ad abundantia. What was determinative in that occasion was that the contracting authority had not awarded, in exclusive, the contract to one of the tenderers [Falk Pharma, para 38].

In this case, just like in the Falk Pharma case, there has not been any element of true competition between the candidates, to evaluate which of their offers is the best and displaces, simultaneously, the remaining other (AGO, C-9/17, paras 51-53, own translation from Spanish)

The reasoning in these paragraphs is strongly skewed towards a very narrow understanding of procurement as implying the award of contracts solely to a winning tenderer, which is not the way framework agreements (and dynamic purchasing systems) operate. I cannot share the analysis in any of these steps of the reasoning.

Firstly, I think that a temporary justification for the irrelevance of the selective nature of a framework agreement is a logical non sequitur. The fact that the funding is limited to the period 2014-2020 can be used to justify the creation of a framework of six years’ duration, but it can have no bearing on the fact that a restriction of the potential suppliers derives from the framework agreement itself. The Maaseutuvirasto could have chosen a fully open licensing system, which would then have avoided the situation of excluding would-be rural advisors as a result of the one-off chance of being accepted into the system (which is a structural result of the framework agreement).

Secondly, in Falk Pharma, the ECJ did not simply consider the lack of choice of a specific supplier and consider the open-ended nature of the ‘authorisation procedure’ ad abundantia, but rather made this a crucial aspect of the analysis, by establishing it as a defining characteristic of the mechanism (see C-410/14, para 14). This is particularly clear on the explicit distinction the ECJ made with framework agreements when it stressed that

it should be noted that the special feature of a contractual scheme, such as that at issue in [Falk Pharma], namely its permanent availability for the duration of its validity to interested operators and, therefore, its not being limited to a preliminary period in the course of which undertakings are invited to express their interest to the public entity concerned, suffices to distinguish that scheme from a framework agreement (C-410/14, para 41, emphasis added).

Finally, the third point on absence of competition is also problematic. Taken to its logical extreme, this would mean that contracting authorities could avoid compliance with procurement rules where they set ‘take it or leave it’ conditions for the provision of services or supplies. This makes no sense because, particularly where there is scarcity in the number of awards (in this case, a limit of total available funding, as well as the restriction in the number of potential awardees that results from the closing of the framework agreement at the initial stage of the 2014-2020 period), there is always implicitly an element of competition (ie to tender or not, and tendering results in a constraint on the overall number/value of awards available to the other competitors) and the fact that the contracting authority limits the dimensions in which the tenderers compete (in Tirkkonen, and implicitly, their geographical coverage) should not exclude this from compliance with procurement rules.

For all the reasons above, I think the Tirkkonen Opinion misconstrues the relevance of the openness of the system in Falk Pharma, and the explicit distinction made by the ECJ between that system and framework agreements. Moreover, the Opinion gives excessive weight to the need to compare tenders or offers (and the choice of one, and almost only one, to the exclusion of all others) for (covered) procurement to take place. In particular, it misrepresents some of the particular features of framework agreements and opens the door to their de-regulation where contracting authorities set ‘take it or leave it’ conditions (eg, in this case, provision of services at rates established by the contracting/funding authority) and then delegate or decentralise decisions on call-offs, even if they provide general guidelines on the way they should take place. For the reasons set out below, I think the Opinion is not only inaccurate from a positive legal analysis perspective (as discussed so far), but also from a normative perspective.

The undesirable combined effect of Falk Pharma and Tirkkonen

Should the ECJ follow the Tirkkonen Opinion, and as a result of the cumulative effect of the resulting expanded Falk Pharma doctrine, Member States willing to avoid compliance with EU public procurement rules could easily do so by creating systems of ‘user/beneficiary choice’. This could be quite problematic particularly in the context of services and supply contracts, where the existence of end users detached from the contracting authority can always enable this type of mechanisms.

In the extreme, if central purchasing bodies created this type of mechanisms for use by individual decision-makers (eg civil servants or public employees), the atomisation of procurement that would ensue could well result in a de-regulation of the procurement function. Procurement rules would not apply to the CPB because it would not ‘choose definitely’ the specific supplier or provider, and they may not apply to the decision to call-off that does exercise that choice if the value of the call-offs is small enough—which would then trigger litigation around the legality or less of the atomisation of the procurement decision on the last stage, for which analysis the concept of ‘separate operational units’ in Art 5(2) of Directive 2014/24/EU (see also recital (20)) would become highly relevant; see K-M Halonen, 'Characteristics of Separate Operational Units – A Study on Aggregation Rules under Public Procurement Law' (2017) report for the Competition Authority; see here. There is thus a functional need to keep proper checks and balances at the level of creation of the mechanism.

On the whole, I was already concerned that Falk Pharma was eroding the scope and effectiveness of the EU public procurement rules, but Tirkkonen could magnify such undesirable effect. Moreover, this would simply displace the problem towards general EU free movement law, which is not a sensible approach in view of the more developed criteria and rules in the EU public procurement framework. Thus, also from a normative perspective, I would plea to the ECJ not to adopt the same approach of AG Campos on this occasion.