Free to use research project idea

I am in the process of editing a collection of papers on the Regiopost judgment for a book and, in one of them, my colleague Prof Tonia Novitz raises the point that the ECJ could have taken Directive 2014/24 into consideration even if it was not applicable ratione temporis. I found this a very valid point and it got me thinking about whether the ECJ is consistent (or not) in taking into account new(er) iterations of existing directives when they resolve disputes to which the (now) old directive still applies.

In the specific case of procurement, and based only on the 2016 cases I commented in this blog, I could find that the ECJ has sometimes considered ‘in anticipation’ the 2014 version of the public procurement directive (2014/24/EU) in cases where it was not applicable ratione temporis--and thus decided under the 2004 version (2004/18/EC). This happened, for example, in

  • Judgment of 8 December 2016 in Undis Servizi, C-553/15, EU:C:2016:935
  • Judgment of 2 June 2016 in Falk Pharma, C-410/14, EU:C:2016:399
  • Judgment of 2 June 2016 in Pizzo, C-27/15, EU:C:2016:404
  • Judgment of 7 April 2016 in PARTNER Apelski Dariusz, C-324/14, EU:C:2016:214

The ECJ has also engaged with other procurement directives (on Concessions, Dri 2014/23) before they were applicable, such as

  • Judgment of 14 July 2016 in Promoimpresa, C-458/14, EU:C:2016:558

However, there are also cases where the ECJ rejected to do so, such as

  • Judgment of 10 November 2016 in Ciclat, C-199/15, EU:C:2016:853
  • Judgment of 27 October 2016 in Hörmann Reisen, C-292/15, EU:C:2016:817
  • Judgment of 8 September 2016 in Politanò, C-225/15, EU:C:2016:645

More detailed analysis would be necessary to establish the type of cases in which the ECJ decided (not) to resort to the newer version of the directive, and the reasons it offered (not) to do so. It would also be interesting to expand the study significantly, both to make sure it is exhaustive in the area of public procurement (ie 2014-2017 + checking for additional cases) and to identify some additional area of internal market law to use as a comparator.

Like in (too many) other occasions, I am not sure I will have the opportunity to explore these issues any time soon. So here is the idea for a research project. Anyone that is interested and has the time / mental bandwidth for it, feel free to use it.

[Input sought] Access to procurement remedies and reciprocity in EU/EEA Member States

I have been thinking for a while about a comparative procurement law question on which I would appreciate your help and input (please comment below or send me an email to a.sanchez-graells@bristol.ac.uk if you have information about your jurisdiction that you can share, for which I would be most grateful). The question concerns the extent to which contracting authorities in the EU/EEA Member States apply an access filter for bids coming from non-EU/non-GPA covered states -- that is, whether contracting authorities check that the tenderer/tender are covered by trade-liberalising instruments or not at the initial stages of a procurement process -- and the extent to which that filter or its absence may then carry on to the access of non-EU/non-GPA tenderers to domestic remedies in those jurisdictions -- that is, whether remedies are limited to EU/GPA tenderers or are more broadly available.

My interest in this topic comes from the fact that, one of the issues that keep arising in the context of the Brexit debate (particularly in view of Prof Arrowsmith's proposals, which I criticised here, and Pedro Telles also criticised here) concerns the future access for UK tenderers/candidates to domestic remedies in the EU27/EEA jurisdictions in case of no deal with the EU and the UK resorting to GPA rules. This links to the broader question of which tenderers/candidates have access to procurement processes and to domestic remedies in the EU27/EEA jurisdictions, as mentioned above.

In my view, Member States can either control coverage by EU/GPA rules at the start of the process or not, and this may result from either an obligation to check or discretion to check. Later, in relation to the point on remedies, there are probably only four relevant (legal) options:

(a) all tenderers/candidates have access to all domestic remedies regardless of their nationality (ie totally open remedies system).

(b) all tenderers/candidates have access to some domestic remedies regardless of their nationality, but only EU/EEA tenderers/candidates have access to 'premium' remedies mandated by EU law (ie those with standstill, etc) (ie open EU+ remedies system).

(c) only tenderers/candidates covered by GPA/WTO, EU/EEA or bilateral FTAs have access to all domestic remedies (ie trade-led remedies system).

(d) only tenderers/candidates covered by GPA/WTO, EU/EEA or bilateral FTAs have access to some domestic remedies, but only EU tenderers/candidates have access to 'premium' remedies mandated by EU law (ie those with standstill, etc) (ie trade-led EU+ remedies system).

I am interested in writing a short paper on this issue and would really appreciate your input on: (i) whether there is some other legally-compliant model I may have overlooked and, (ii) more importantly, what is the model in your jurisdiction (specially if you are based in an EU27/EEA country). So far, the information I have been able to gather is as follows [apologies to those of you who have helped me with this if I have misinterpreted it -- corrections welcome; when reading this, please bear in mind that the content evolves as I discuss these issues with national experts and explore the ideas further, particularly in terms of the contours between models (a) and (c)]:

  1. Austria (thanks to Michael Fruhmann): Federal Procurement Law states, that procurement procedures shall be carried out in compliance with the fundamental principles of EU Law, the principles of free and fair competition and the equal treatment of all applicants and tenderers. However, different treatment of applicants and tenderers on grounds of their nationality or of the origin of goods which is permissible under international law remains unaffected by this obligation. The (legal) consequence is, that if no union or international obligations (re latter: this depends on the existence and scope of FTAs, RTAs also) exist to open PP procedures to foreign bidders, contracting authorities are free to admit participation or to deny the participation of such bidders. However, if such bidders are allowed to participate they have the same standing as national/EU bidders (also as regards remedies). In practice this comes down to the question, whether the contracting authority wants such bidders to participate. This is a case by case decision depending i.a. on the subject matter of the contract, the interest to intensify/safeguard competition in a given procedure. This decision (no admittance) can of course be reviewed (and has been reviewed) but the courts confirmed that without any EU/international obligation it's fully within the competence of the contracting authority to decide either way. Generally, this points towards the model being generally (c), but with the possibility of going beyond that and getting closer to (a) depending on the contracting authority's discretion.
  2. Belgium (thanks to Baudoin Heuninckx): a contracting authority may reject the request to participate or tender by undertakings from countries outside of EU/WTO or without an FTA, so there is a potential "filter" at the very beginning of the procedure. In terms of remedies, every candidate or tenderer has access to all remedies regardless of nationality. Potentially, this leads to the remedies model being (c).
  3. Czech Republic (thanks to Jaroslav Mencik): contracting authorities may not restrict participation in public tenders of suppliers from the EU, the EEA, Switzerland, or other states with which the Czech Republic or the EU has concluded international agreements which guarantee that suppliers from such a state will have access to the public contract being awarded. It follows that contracting authorities are required neither to check the nationality of tenderers nor exclude non-EU/non-GPA tenderers (but may choose to do so). Remedies follow model (a), all tenderers participate on equal terms.
  4. Denmark (thanks to Carina Risvig Hamer): it is not foreseen in legislation, but contracting authorities can decide not to allow participation from non-EU/non-GPA tenderers. All candidates and tenderers have full access to remedies. Potentially, this leads to the remedies model being (c). 
  5. Estonia (thanks to Mari Ann Simovart): remedies are available to any "interested party" without any restriction based on the country of origin. In short, model (a) applies. However, a contracting authority can restrict access to a particular procurement procedure for tenderers of EU/EEA/WTO only - in which case, tenderers outside EU/EEA/WTO can be regarded as having no "interest" towards the particular procurement and thus no standing to claim review.
  6. Finland (thanks to Kirsi-Maria Halonen): contracting authorities would not always check whether a tenderer is covered by the agreements, but could do so at the beginning of the tendering procedure. If accepted to participate/tender, the candidate/tenderer would likely have access to all domestic remedies. This leads to the remedies being closer to model (c), but it is possible that de facto, contracting authorities may be granting equal treatment beyond GPA/EU/FTA coverage in sui generis basis (model (a)). It is in the contracting authorities' discretion whether to even look into the matter/exclude. If tenderers are not excluded, they'll have equal rights for remedies. However, it is worth bearing in mind that this is untested in the courts.
  7. Germany (thanks to Gabriella Gyori): not taking into account decentralized matters (due to the differences among the "Bundesländern"), according to the federal public procurement legislation related to above threshold procedures, tenderers from outside of Germany are allowed to participate, treated equally and have equal rights. Remedies follow model (a). 
  8. Greece (thanks to Marios Skiadas): in order to be eligible to participate in a public tender, economic operators must be based in an EU, EEA, GPA or other countries which have signed bilateral or multilateral agreements with the EU in matters related to public procurement procedures. Contracting authorities have a first chance of checking this requirement when they assess the ESPD or equivalent documentation. Additionally, during the final stage of the awarding phase, the winning bidder is required to submit all legal documents regarding company establishment, operation and representation. Therefore, the contracting authority will in practice have a “second chance” to check conformance. Access to remedies applies to all parties with an interest in being awarded a public contract. By combining this to the eligibility criteria stated above,Greece follows model (c).
  9. Hungary (thanks to Gabriella Gyori): economic operators shall be excluded from participating in the procedure as a tenderer, candidate, subcontractor or an organization participating in the certification of suitability, if have their fiscal domicile in a country outside the EU, the EEA or the OECD or in a non WTO/GPA country or outside the overseas countries specified in the TFEU or in a country which has not signed any agreement with Hungary on avoiding double taxation or which has not signed a bilateral agreement with the EU concerning public procurement. Claims can be submitted by a contracting authority, a tenderer(s) or any other interested person whose right or legitimate interest is being harmed or risks being harmed by an activity or default which is in conflict with the procurement legislation. This brings the remedies system close to model (c).
  10. Ireland (thanks to James Farrell): based on long-standing practices, there are not requirements of EU/EEA/GPA membership as a qualifying requirement for tenderers, or references to different treatment of tenderers emanating from non-EU/EEA/GPA countries in any policy or guidance documents issued by the relevant Irish authorities. The general approach, driven by Ireland's open trading policy, is to take value wherever it can be found. There have been no court challenges in Ireland where an unsuccessful tenderer sought to argue that a winning tender should be disqualified because of the country of origin/registration/domicile of the tendering entity. Regarding remedies, apart from reliefs arising under the Remedies Directive there are also domestic reliefs such as Judicial Review, Injunctions etc that would be available to tendering entities regardless of nationality. Therefore, Ireland follows model (a).
  11. Italy (thanks to Roberto Caranta): only tenderers/candidates from MS/parties to GPA/WTO, EU/EEA or bilateral FTAs are eligible to bid. Eligible suppliers then have access to all domestic remedies; so the systems follows model (c).
  12. Lithuania (thanks to Deividas Soloveičik): there is no obligation for contracting authorities to check non-EU/ non-GPA suppliers. Remedies follow model (a), all suppliers participate on equal terms.
  13. Netherlands (thanks to Tim Beukema): Dutch law states that a contracting authority shall not grant any advantage in regard to the tender and the contract that is not granted to parties from countries within the EU. In regard to rejection of participants, contracting authorities may reject the request to participate by undertakings from countries outside of the GPA, EU or FTA. Entities operating in the water, energy, transport and postal services sectors (special sectors) have the possibility to reject a participation or tender if the goods that a party provides consists of more than 50% from countries on which the EU has no obligation to, i.e. countries outside the GPA, EU or FTA. A special sector company has the obligation to decline an offer of such party in the case of an equal bid from a undertaking within the GPA, EU or FTA that has less than 50% of the goods from within these countries. Claims can be submitted by parties who are interested in the tender in the case that his rights are being harmed or could be harmed because of the fact that the tender procedure breaches the procurement rules, which is a remedies system in accordance with model (c).
  14. Norway (thanks to Eirik Rise): follows model (c); only tenderers/candidates covered by GPA/WTO, EU/EEA or bilateral FTAs have access to domestic remedies, and only to the extent that it is covered in the relevant FTA.
  15. Poland (thanks to Paweł Nowicki and Piotr Bogdanowicz):  There is a newly introduced obligation to comply with WTO GPA and other international agreements to which the EU is a party, and there is no explicit obligation to exclude non-EU/non-GPA tenderers. Remedies follow model (a).
  16. Portugal (thanks to Pedro Telles): [not clear yet whether there is an initial filter]. Remedies follow model (a).
  17. Romania (thanks to Dacian Dragos): [not clear yet whether there is an initial filter]. Remedies follow model (a).
  18. Slovenia (thanks to Njives Prelog): suppliers from all over are allowed to participate, treated equally and have equal rights. Remedies follow model (a). 
  19. Spain: at the initial stage, contracting authorities have an obligation to check coverage by EU/GPA rules or to require confirmation of reciprocal access for Spanish tenderers in the country of origin of non-EU/non-GPA tenderers (which are also required to have a branch office in Spain). Remedies follow closely model (a) because remedies are open to all those admitted to tender procedures [ie go beyond (c), but are still somehow trade-led due to reciprocity requirement].
  20. Sweden (thanks to Andrea Sundstrand): there is no check at the start of the procedure and suppliers from all over are welcome to participate on equal terms. Remedies follow model (a) and all suppliers have access to exactly the same remedies regardless of whether they are from countries that Sweden has trade agreements with.
  21. United Kingdom (thanks to Aris Christidis and Pedro Telles for discussions): The UK system replicates the EU Directive in terms of extending equal treatment (which can be seen to include access) to economic operators covered by EU law, the WTO GPA, or other international agreements by which the EU is bound (see reg. 25 PCR2015). The remedies system is limited to those economic operators to which contracting authorities are legally taken to owe a duty to comply with public procurement rules. Effectively, this is limited to economic operators from the EEA, GPA signatories (provided the procurement is covered) and countries with bilateral agreements in force (see regs. 89 and 90 PCR2015).

This initial scoping exercise seems to indicate clustering around models (a) and (c). It would be amazing if we could collectively cover most of the EU27/EEA and complete the exercise, not only in order to gain a better understanding of this issue, but also because this will be relevant for Brexit negotiations around procurement in the immediate future. Your contribution will, of course, be duly acknowledged and gratefully received.

New analysis of joint tendering under EU competition law: a few comments on Ritter (2017)

Cyril Ritter has made a new contribution to the analysis of joint tendering for public contracts under EU competition law in this interesting recent paper. Ritter's paper goes beyond previous discussion of the topic [eg my critical remarks on Thomas (2015), see here] and proposes an alternate analytical approach in many points. I find his analysis of different 'theories of harm' applicable to joint tendering interesting and insightful, and the special criteria he suggests for negotiated procedures and for tenders where one contractor is indispensable to two or more tenderers are thought-provoking. However, there are also aspects of Ritter's proposals which I do not see entirely clear, and where I do not think his paper goes much further than previous discussion of the topic.

One of the key issues that require clarification for the purposes of assessing whether join tendering breaches EU competition law (Art 101 TFEU) as an instance of anticompetitive joint selling concerns whether the members of the joint tender are competitors or not. On that point, Ritter emphasises that "what matters here is whether they are competitors for the purpose of the particular procurement procedure at issue" (p 4). After a review of the relevant ECJ case law, Commission's guidelines and administrative practice in the area of EU competition law enforcement, he proposes that the relevant question is to assess whether a firm has "real concrete possibilities" to bid for the contract being tendered (see p. 6). In his view, the burden of proof rests with the authority, but it can be shifted where the "authority brings substantial evidence that the parties are potential competitors" (ibid). Substantively, his main test requires assessing whether the firms have independent ability to bid for the contract, which is determined by the "ability to meet the tender specifications -- in terms of having sufficient spare capacity, equipment, staff, regulatory permits, quality certifications, etc" (p. 7). Interestingly, Ritter excludes the possibility of carrying out an analysis of the undertakings' intention to bid for the contract (pp. 9-10).

At this point, Ritter reaches the need to assess the extent to which it can be objectively determined that an undertaking had the ability to bid independently for a contract for which it has decided to bid jointly with others. He points out at the disagreement between Thomas an myself (see here) concerning whether the possibility of giving up alternative projects can/should (not) be included in the analysis. Ritter considers that the discussion may be beside the point, and that the issue rather requires an assessment of "what happens when a party to the joint tender would not be able to bid on its own (perhaps because capacity is allocated to other projects), but could have done so by hiring more staff, buying or renting more equipment, or teaming up with someone else? Should it be considered a potential competitor?" (p. 8).

Interestingly, this brings Ritter's proposed test very close to Thomas', where the latter indicates that it is important not to ignore "the possibility that each undertaking might nonetheless be able to submit an independent bid, by bringing in specialist resources from outside. If it were in fact feasible for each undertaking to submit a tender in this way, then surely it cannot be excluded that a joint bid would restrict competition. The real question is rather whether, in the absence of the joint bid, there could in fact have been two or more independent bids". And, more specifically, when Thomas clarifies that "One possible approach to this issue would be to ask whether, in the ordinary course of business, each undertaking would normally bring in such resources from outside. Alternatively, and more precisely, are such resources demonstrably available on reasonable terms and in time to prepare and submit the tender, from an undertaking that is not a competitor in the procurement procedure?".

As I said when I commented on Thomas' paper, I find this line of argument exceedingly restrictive. Conceptually, because it relies on an assessment of whether the parties of the teaming/joint bidding agreement could have cooperated with other undertakings or complemented their capacities in a different way (including the need to source additional capacity from elsewhere), which fundamentally and in itself proves the point that they were unable to submit bids individually or with a total independence from third parties (including suppliers or providers of services, as well as employees, although this raises the tricky issue of the need to contain the analysis within the limits of the concept of undertaking for the purposes of EU competition law enforcement). Once this is clear, I see no good reason for the assessment to rely on whether there were alternative potential partners that joint bidders could have (independently?) teamed up with, not least because this would require an excessive amount of second-guessing by procurement and competition authorities, who may not be the best placed to query business decisions ex post facto.

Indeed, the difficulty with this line of assessment is that it would require second-guessing business strategies and preferences actually revealed by the undertaking -- which decided to participate in the joint bid with its specific partners, rather than engaging in any of the other (theoretically) possible alternative business strategies -- and compare them with an alternative scenario envisaged by the enforcement authority. Even if Ritter advises against extracting hard and fast conclusions from such an analysis (p. 9), he does indicate that "the rule of thumb is that the parties to a joint tender are competitors if it reduces the number of tenders that realistically could have been made otherwise" (ibid).

Overall, this comes to indicate the difficulties in excluding the applicability of Art 101(1) TFEU to cases of joint tendering, which are likely to be considered potentially restrictive of competition in most instances if a strict objective assessment of the joint tenderers' ability to have tendered for the contract (independently, or with others) is carried out, as proposed by Thomas and Ritter. However, this does not necessarily eschew the analysis (although it does effectively reverse the burden of proof) towards the finding of infringements, provided that the possibility of declaring prima facie restrictive joint tendering agreements exempted under Art 101(3) TFEU properly concentrates on the analysis of their efficiency. Ritter addresses this issue towards the end of his paper (pp. 15-16).

In that regard, Ritter considers that the parties to the joint tendering agreement need to be able to show that

  • the joint tender improves the value proposition to the customer, e.g. in terms of price, or, more likely, in terms of quality (first and second conditions of Article 101(3); this assessment may require giving a monetary value to non-price factors);
  • achieving those efficiencies would not have been possible through a less restrictive alternative, such as hiring personnel or equipment, or teaming up with another firm which is not a competitor (third condition of Article 101(3); this assessment may entail an element of counterfactual analysis); and
  • the joint tender does not "afford such undertakings the possibility of eliminating competition" with respect to the procurement procedure at issue, i.e. the joint tender is unlikely to be the only tender (fourth condition of Article 101(3)) (Ritter (2017) 16, emphasis added)

Once more, this test also seems rather stringent and, in particular, its second aspect can be rather problematic. In its literal reading, the equivalent condition of Art 101(3) TFEU requires that the agreement does not "impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives". A strict reading, such as Ritter's, to the effect that this requires that "achieving those efficiencies would not have been possible through a less restrictive alternative, such as hiring personnel or equipment, or teaming up with another firm which is not a competitor (third condition of Article 101(3); this assessment may entail an element of counterfactual analysis)" would create the effect of conflating the test for the application of Art 101(1) TFEU and the exemption of Art 101(3) TFEU with the logically circular and perverse implication that any teaming agreement that is found prima facie restrictive and in breach of Art 101(1) TFEU because the parties could have sought additional personnel or equipment, or teamed up with a third party (itself not a competitor), is also necessarily excluded from exemption under Art 101(3) TFEU precisely because of those reasons.

The need to distinguish the elements for an analysis under Art 101(1) and Art 101(3) TFEU when the assessment includes the need to consider potential competition triggers some difficult issues. In the context of public procurement, this requires settling whether the assessment of the need for the (potential) competitive restriction implicit in the joint tender to generate the claimed efficiencies is, either (a) limited to the agreement under analysis, or (b) should also include the potential alternative business strategy which (theoretical) existence brought the joint tendering agreement under scrutiny in the first place. Existing European Commission Guidelines on  the application of Article 101(3) of the Treaty can provide a framework for this analysis.

The key part of the Art 101(3) TFEU Guidelines is para [76] and, more precisely, the consideration that "It is particularly relevant to examine whether, having due regard to the circumstances of the individual case, the parties could have achieved the efficiencies by means of another less restrictive type of agreement and, if so, when they would likely be able to obtain the efficiencies. It may also be necessary to examine whether the parties could have achieved the efficiencies on their own" (emphasis added). Applied to the specific point, I read this to require an assessment of whether a less restrictive agreement between the same parties would have allowed the joint tender and, potentially, whether they could have generated the same efficiencies (strictly) on their own, quod non because of the previous determination that they would have needed "hiring personnel or equipment or teaming up with a non-competitor" -- which in my view does not fit the counterfactual of an analysis of the ability of the party to bid for the tender all things being equal, which would have determined its classification as an actual competitor. My objection is that proceeding in the way Ritter suggests (ie considering the potential scenario of alterative business strategy both at Art 101(1) and Art 101(3) stages) would create, if not a circular or self-referential logic, at least a double whammy for the joint tenderers because their condition of potential competitors would not only be used to bring their agreement under Article 101(1) TFEU, but also to exclude its exemption under Article 101(3) TFEU -- which does create substantive analytical conflation in my view.

In my opinion, an alternative analysis is preferable, to the effect that 

... undertakings concluding joint bidding and teaming agreements should be able to prove that they can only submit a compliant tender if they participate together, or that the terms of their joint tender are substantially better for the public buyer than those they could offer independently—ie, that there are specific and measurable efficiencies derived from the teaming or joint bidding strategy and that they are passed on to the public buyer. For their part, contracting authorities will need to be on the lookout for potential negative impacts on competition in the market, as well as the inclusion of unnecessary restrictions in the teaming and joint bidding documents (A Sanchez-Graells, Public procurement and the EU competition rules, 2nd edn (Oxford, Hart, 2015) 339, footnote omitted and emphasis added).

Or, in other words, I think that -- for the purposes of the application of Art 101(3) TFEU -- the analysis needs to rest on whether the joint tenderers have limited their collaboration to what was necessary to create the efficiency of their joint bid, or have rather improperly taken that chance to further restrict competition amongst them. But it should not revisit the same theoretical counterfactual analysis that brought the agreement under Art 101(1) TFEU scrutiny to begin with.

Some thoughts on regulatory substitution, public procurement and labour objectives

I am thrilled to participate in the event 'Socially Sustainable Public Procurement' organized by Dr Richard Craven at the University of Leicester tomorrow. Together with Richard and Dr Eleanor Aspey, I will be reflecting on issues around the use of public procurement for the enforcement of labour standards.

My comments (which I expect to reflect the minoritarian view) will be centered on issues of regulatory substitution and the competition implications that the ECJ's case law in Bundesdruckerei and Regiopost have created. This will help me sharpen the arguments for two papers that are in the making. More details soon. The slides for tomorrow's presentation are here:

Separate operational units within a contracting authority and the scope of Directive 2014/24

One of the reforms of EU public procurement rules in 2014 that may well have slipped under the radar concerns the treatment of procurement carried out by separate operational units within a contracting authority. For the purposes of calculating the estimated value of procurement to determine the applicability of the EU rules, Art 5(2) Dir 2014/24 now establishes that "Where a contracting authority is comprised of separate operational units, account shall be taken of the total estimated value for all the individual operational units. Notwithstanding [that] where a separate operational unit is independently responsible for its procurement or certain categories thereof, the values may be estimated at the level of the unit in question."

This seemingly simple rule raises an important number of issues and, most importantly, requires a determination of what is a "separate operational unit" for the purposes of Art 5(2) Dir 2014/24 and the associated anti-circumvention rule. These issues are the focus of the comparative report "Characteristics of Separate Operational Units – A Study on Aggregation Rules under Public Procurement Law", commissioned to Dr Kirsi-Maria Halonen by the Swedish Competition Authority.  

The study includes a comparative overview that is interesting in itself and, of more practical relevance, it also formulates a test for the assessment of whether units within a contracting authority meet the requirements for being considered operationally separate and, thus, able to trigger a differentiated calculation of value thresholds triggering (or most likely, not) the application of EU public procurement rules in Dir 2014/24. The test is presented as follows:

"In order to facilitate the evaluation of a unit’s status, this study identifies six key elements which can be of importance when determining, whether the contract value can be estimated at the level of a separate unit or, whether all purchases of units within the same contracting authority should be aggregated: 

  1. The unit has a separate budget line which is managed by the unit itself and from which the procured items are paid from
  2. The unit runs the tender procedure independently
  3. Competence to make buying decisions and to conclude contracts on behalf of the contracting authority
  4. Is any other part of contracting authority interfering or affecting the contract between the unit and its contractor?
  5. Will other units of the same contracting authority purchase through the contract awarded by the unit?
  6. Obligation to purchase through centralized framework agreements or contracts"

I find the test (which is further detailed in the study) well thought-through and the only addition I would suggest would concern a dimension of supply-side analysis, mainly to assess whether the seemingly separate operational units are supplied by different suppliers / under different terms. That would allow for a final check to be added in order to capture situations where looking only at the demand side (ie at the units within a contracting authority) may mask issues concerning the bigger picture of the procurement/supply relationship between specific suppliers and the contracting authority as a whole.

The report is well worth reading, in particular in countries where the existence of separate operational units has been taken for granted in the past (such as in Spain). This is an area where future empirical research could usefully provide good insights on the way in which the creation of the new rule in Art 5(2) Dir 2014/24 may result in different levels of stringency in the application of EU public procurement rules at domestic level--depending on the extent to which Member States adapt the internal organisation of their contracting authorities to maximise, minimise (or ignore) the new possibilities.

Some initial comments On the Commission's Report on the Effectiveness of the Public Procurement Remedies Directives

The European Commission recently published its long awaited report on the effectiveness of the public procurement remedies directives [COM(2017) 28 final, 24.1.2017], which is accompanied by a much bulkier staff working paper [SWD(2017) 13 final, 24.1.2017]. The report provides an interesting general overview of the situation at Member State level after the last revision of the remedies directives 10 years ago and, by  and large, offers the unsurprising conclusions that (1) the remedies directives have been transposed in all Member States in a relatively homogeneous manner, except in relation to institutional decisions--where there is a split between administrative and judicial first instance review procedures, (2) there is a general consensus that the remedies directives are relevant and that they enhance the effectiveness of the substantive public procurement rules, (3) the costs of compliance with the review system envisaged in the remedies directives is not necessarily higher than that derived from any alternative domestic system the Member States would have in place in its stead, (4) the remedies they create are generally in line with the right to an effective remedy as a general principle of EU law, and (5) that the remedies directives are generally fit for their purpose. 

On the basis of that overall assessment, the report concludes that, generally, "the evaluation identified neither major nor urgent needs to amend the Remedies Directives, [so] it is decided to maintain them in their current form, without any further modification at this stage." This is not to say that the Commission does not recognise shortcomings in the remedies directives and their fitness for purpose going forward, but that it rather strategically decided to avoid a path of legal reform and rather aims to work on soft mechanisms to try and improve the situation (see below). In my view, this is a highly questionable regulatory strategy and the assessment that there is no major or urgent need for an amendment of the remedies directives does not really tail up with the difficulties in ensuring an improvement of the system by alternative means without a review of the remedies directives [for discussion, see A Sanchez-Graells,  "'If It Ain't Broke, Don't Fix It'? EU Requirements of Administrative Oversight and Judicial Protection for Public Contracts", in S Torricelli & F Folliot Lalliot (eds), Administrative oversight and judicial protection for public contracts (Larcier, 2017, forthcoming].

Throughout the report, there are internal evaluative contradictions and clear indications that the Commission has not been ambitious in its assessment of the remedies directives, or even adopted a coherent approach to its overall evaluation, and that it has rather indulged in a complacent strategy that severely limits the ability of the remedies system to continue being fit for purpose in the future. Indeed, the main report itself indicates that 

The evaluation revealed that certain aspects of the Remedies Directives could be made clearer. This is confirmed by the contributions received. This applies, for example, to matters such as the interplay between the Remedies Directives and the new legislative package on public procurement, and the development of criteria to be applied to lift the automatic suspension of the conclusion of the contract following the lodging of a legal action.
The Remedies Directives are in line with the rights and general principles laid down in EU primary law concerning fundamental rights. They lie at the core of public procurement legislation as they allow bidders to enforce their substantive rights. They were found to be generally aligned with the new 2014 legislative package on public procurement, in particular to cover the concessions subject to Directive 2014/23/EU. Nonetheless, as already mentioned, the interplay between these Directives and the new legislative package on public procurement could be further clarified.
... the Commission acknowledges that certain provisions of the Remedies Directives are not entirely clear. In particular, despite the update introduced by the new legislative package on public procurement, some additional needs for clarification have been identified. For instance, references to ‘contract notice’ in the Remedies Directives do not reflect the fact that new Directive 2014/24/EU permits the use of a prior information notice, instead of a contract notice, to call for competition in certain circumstances. It could also be clarified how the Remedies Directives apply to modifications of public procurement contracts and concessions, to the termination of such contracts and to the light procurement regime (footnotes omitted and emphases added).

In that regard, with little more detail, the Staff working paper indicates that

several stakeholders underlined that some provisions of the Remedies Directives could be more precise. In particular, more clarity would be welcome in a number of areas related to institutional aspects (for example, professional standards for members of an administrative review body), procedural aspects (for example, criteria for lifting the automatic suspension, for granting interim measures and to award damages) and the interplay between the Remedies Directives and the new Public Procurement Directives (for example, how the Remedies Directives apply to the modification and termination of public contracts and concessions and the so-called ‘light regime’) (p 43, emphasis added). 
It is reported, however, that certain areas could be clarified by the Commission, for example, in the form of guidelines (e.g. fees, requirements for first instance administrative bodies and their work organisation). Following the adoption of the new legislative package on public procurement, more clarity would be welcome with regard to the interplay between the Remedies Directives and the new substantive rules (e.g. references to "contract notice" in the classic Remedies Directive does not reflect the fact that the new classic procurement Directive enables a prior information notice to be used, instead of a contract notice, to call for competition in certain circumstances; it could be clarified how the Remedies Directives apply to modifications of public contracts and concessions, termination of such contracts and the light regime for such contracts) (p. 68, emphasis added). 

The Commission plans to provide such coordination between the remedies directives and the 2014 public procurement package without legal reform, and through "guidance on some outstanding aspects of the Remedies Directives in order to increase the understanding of some provisions and to guarantee their effectiveness. Aspects that could be covered include the interplay between the Remedies Directives and the new legislative package on public procurement and the development of criteria to be applied to lift the automatic suspension of the conclusion of the contract following the lodging of a legal action. Based on evidence gathered so far, the Commission will engage in a dialogue with Member States and stakeholders to identify other specific areas that require clarification."

I find this problematic for two main reasons: First, in my mind, it is not clear at all that the discoordination between the remedies directives and the 2014 public procurement package derives from lack of clarity, rather than from the limited scope of the remedies directives, which do not make provision for important issues such as the regulation of remedies for dynamic procurement mechanisms, or the rules applicable to excessive contractual modification or the improper avoidance of mandatory termination of contracts. They also completely fail to regulate situations that arise from the increased justiciability of interim decisions by the contracting authority (notably, on exclusion and the new possibility of self-cleaning). An adaptation of the remedies directives to these situations is not a matter of guidance, but rather of legislation--as there is no obvious way forward, there is no case law of the Court of Justice of the EU on which the Commission can rely to establish baseline proposals, and the competential limits of the EU can easily be exceeed, in particular regarding the interaction between the remedies directives and the new rules on contract modification and contract termination, where the development of EU rules keeps building on the flimsy residual competence of Article 114 TFEU.

Second, and more importantly, the issuance of guidance in this field strikes as a strange policy avenue and one that the Commission is avoiding in other areas (such as the provision of guidance on the interpretation of the 2014 public procurement package itself). It also strikes as an avenue of dubious effectiveness, in particular if/where the solutions proposed in that guidance are difficult to coordinate with domestic procedural and administrative requirements. In the report itself, the Commission indicates that "The evaluation also made it possible to identify problems that persist at national level. In particular, various stakeholders confirmed in the context of the public consultation that problems identified are rooted either in national legislation beyond the Remedies Directives or in national practices, and not in the Remedies Directives themselves"; and concluded (rather conveniently that) "Even if specific concerns are reported in some Member States, they usually stem from national measures and not from the Remedies Directives themselves".

This assessment should be carried to its ultimate consequences, and there are only two possible scenarios: (a) either the Commission plans to open infringement proceedings against the Member State where the domestic legislation and practices are incompatible with the Remedies Directive (which does not seem either the case, or a priority for the Commission), or (b) the only way to erode those national rules and practices is to create additional rules that make them incompatible with EU law (thus promoting legislative reform at Member State level, or creating the possibility for more clearly grounded infringement proceedings in the future). But this is not what the Commission suggests is the best way forward. Its proposal is, on the contrary, to suggest that there should be simple (soft) guidance and to hope that it will be voluntarily adopted in a way that demolishes contrary domestic rules and practices--if not voluntarily, then maybe as a result of the increased interaction between first instance review bodies it is also hoping to create as part of its future proposals.

I think any practitioner will agree that this strategy is very unlikely to create any meaningful change at Member State level. There are many reasons for that. It is not possible to rely directly on Commission interpretation before domestic courts, and it clearly does not trigger obligations of consistent interpretation as a revised remedies directive would. Commission guidance can, at best, be used as persuasive authority, but always under the risk of the Court of Justice eventually determining that the European Commission had gotten it wrong in its guidance and that domestic review bodies and courts should have asked for the proper interpretation of the Remedies Directives (but this is not an issue of lack clarity, as above), or else establishing that domestic rules breach some general principles of EU law. On that point, the general statement by the Commission that "The Remedies Directives are in line with the rights and general principles laid down in EU primary law concerning fundamental rights" can hardly provide any reassurance, because the difficulty is not in accepting that the remedies directives do not breach that general principle, but rather in establishing whether they do enough to uphold it in an effective manner going forward.

Overall, I find the Commission's decision not to review the remedies directives disappointing, in particular because it has some echoes that remind us of the difficulty of getting technical rules approved or reviewed at EU level, as well as the Commission's conformity with that situation. I can only hope that evidence on the unsuitability of the remedies directive in relation with the novelties of the 2014 public procurement package will emerge, to the point of prompting the European Commission to reconsider this strategy in the medium (or nearest possible) future.

A disappointing Brexit White Paper makes for disappointing comments

Theresa May's Government has published the White Paper on The United Kingdom’s exit from and new partnership with the European Union earlier today. It is an extremely disappointing document that, unfortunately, only allows for disappointing comments. The only remarkable aspect of the Brexit White Paper is the number of dimensions in which it is disappointing.

Its timing is probably one of its most disappointing aspects. Given that the House of Commons debated the European Union (Notification of Withdrawal) Bill and voted its passing to Committee stage only yesterday, today's publication of the Brexit White Paper mimics an absurd claim of power by the Government over the Parliamentary scrutiny of Brexit (the power to unduly limit and restrict it).

On the contrary, publication of the White Paper before the Parliamentary debate would have created the double effect of avoiding the impression that Government is only very reluctantly complying with the UK Supreme Court's requirement for Parliamentary approval of the giving of notice under Article 50 TEU (which would have been positive for the Government), but would also have allowed MPs to criticise the Government for the appallingly poor quality of the document (as per below, which would have been clearly negative for the Government) before even moving to the discussion of amendments.

Clearly, then, the timing of the publication of the Brexit White Paper demonstrates the Government's insecurity about its content and its overconfidence about the ability to push for (hard) Brexit no matter what. In my view, this is a dangerous combination of attitudes.

The content of the Brexit White Paper is also extremely disappointing. It is not more than a minimally expanded version of the speech given by Theresa May PM some weeks ago, coloured with some superficially analysis macroeconomic data, and most of its sections are simply a very superficial account of the current state of EU law coupled with the (unfounded) hope and half promise that Theresa May PM's Government will obtain a great deal from the EU.

Twitter is already full of criticism based on obvious mistakes in the Brexit White Paper, and I am sure that the analysis that will emerge in blogs and newspapers in the next few hours and days will not be more positive. However, I also find it unlikely that much of the criticism will be able to go beyond general issues or points already raised against Theresa May PM's speech, as the Brexit White Paper has not brought any meaningful additional detail that we can analyse. In any case, it may be worth highlighting that I found the sections on the creation of an alternative dispute resolution mechanism to substitute the ECJ's jurisdiction (section 2 plus Annex A) and on trade with the EU (section 8) particularly disappointing.

Section 2 on 'Taking control of our own laws' is very confusing and seems to me to miss several important points. The fundamental issue it does not address is the extent to which future case law of the ECJ will still need to be applied and followed by UK courts and in the UK more broadly, simply as a result of the incorporation of EU law into domestic law, or as an non-renounceable element of the EU's regulatory architecture of the single market.

This has implications in a number of dimensions, such as the difficulty in coordinating the effects of the Great Repeal Bill (which is meant to consolidate EU law into UK law as of the time of withdrawal) with the effects of the ECJ's interpretation of EU law, which is most frequently not time bound and thus has retroactive effect. A simple example would imply a situation where UK withdrawal from the EU happens in 2019 and later, say in 2021, the ECJ interprets a provision of EU law that was incorporated into UK law by the Great Repeal Bill. In that case, lawyers will feast with the litigation that will ensue from the difficult issue of determining the interpretation to be given to the 'nationalised' provision of EU law and the extent to which it would be consistent with the (Parliamentary) intention of using the Great Repeal Bill to 'download' EU law into UK law without any amendment at that point (or barring ulterior Parliamentary intervention or explicit reform through secondary legislation).

But even without going that far in creating a severance between UK and EU law, there is an unknown number of trade-related areas that will require continued compliance with ECJ case law as a technical matter and as far as the UK intends to have any access to the single market. For example, if the UK wants to engage meaningfully in trade with the EU, one of the main issues will be the need for continued compliance with technical standards (to which the Brexit White Paper also refers, but in a confusing or slightly misleading way in section 8), and these are bound to be increasingly subject to ECJ interpretation, particularly after the assertion that they are part of EU law and thus subject to its jurisdiction in the recent case of James Elliott Construction Limited v Irish Asphalt Limited (C-613/14, EU:C:2016:821).

Disposing of these very relevant difficulties in a simple paragraph that boasts that "We will bring an end to the jurisdiction of the CJEU in the UK. We will of course continue to honour our international commitments and follow international law." is a grossly misleading oversimplification.

Section 8 on 'Ensuring free trade with European markets' is not better. From a legal perspective, the point I find most internally contradictory in this section is the fact that the Brexit White Paper indicates continuously how several areas of regulation of the internal market for services hinge on the existence of a single regulatory framework at EU level, on legal certainty (which is logically and legally based on the interpretation of that regulatory framework by the ECJ) and on an effective system of civil judicial cooperation as well as cooperation between regulators and independent enforcement agencies.

Not get bogged down on detail, I fail to understand how this is a model that can be replicated without the need for the UK to comply with EU law (as interpreted by the ECJ, see above) and, even if that is possible, how could that be in line with a reduction of red tape and administrative burden for companies that would, by implication, need to comply with more than one regulatory framework--unless they were only active in the UK. Generally, the logic of wanting to create mutual recognition and at the same time pushing for regulatory disparity seems starkly at odds with the logic of regulatory architecture of the single market. Similar problems arise with the Brexit White Paper veiled insinuation that the UK can still be member of pan-EU agencies despite not being a Member of the EU/EEA. Overall, this section simply does not offer a logic that could pass critical muster.

From that perspective, the Brexit White Paper correctly identifies that "Unlike other trade negotiations, this is not about bringing two divergent systems together. It is about finding the best way for the benefit of the common systems and frameworks, that currently enable UK and EU businesses to trade with and operate in each others’ markets, to continue when we leave the EU through a new comprehensive, bold and ambitious free trade agreement." But it is plainly wrong in the implications it tries to derive from this. The blatantly obvious impossibility of this logic is that, whereas in other types of trade negotiations the harmonisation of systems will result in a reduced administrative burden for both the public and private sectors, in this case the need to dissociate a truly integrated system into two coordinated systems will necessarily create those burdens for both the public and the private sector. And this is what makes this negotiation so riddled with impossibilities and so suicidal: it is a negotiation to move from a win-win to a lose-lose scenario, and the only thing the negotiating parties can hope is to minimise the loss. In my mind, this is an irrational process to engage with, and the only justification for it is that Theresa May PM's government hopes for benefits that no one else identifies. If nothing else, the Brexit White Paper has done nothing to provide evidence of the existence of those potential gains or of the feasibility of the (under-worked) plans to unleash them.

Overall, thus, I find the Brexit White Paper extremely disappointing. And I can only blame myself for having had any hopes that it would not be so.

 

 

Interesting Preliminary Reference on Interaction between Competition and Public Procurement Law (C-531/16) [guest post* by Dr Deividas Soloveičik]

This guest post by Dr Deividas Soloveičik provides interesting background and critical remarks on a Lithuanian reference to the European Court of Justice (ECJ) for a preliminary ruling on issues concerning the interaction of public procurement and competition law--see case Specializuotas transportas, C-531/16. It will be interesting to keep an eye on the case, as it brings an opportunity for the ECJ to expand its limited case law in this area.

Unravelling conflicts of interest

The Supreme Court of Lithuania has been extremely active during recent years in developing the case-law related to both the concept and a genuine practice of the conflicts of interest in public procurement law. It’s interesting to note that the ideas the national judges are bringing “on the table” are much more than the inside legal practice of a Member State. The questions regarding the subject have been referred to the Court of Justice of the European Union (CJEU) and, therefore, became of cross-border interest.

It started from eVigilo case [1] where the Supreme Court sought and received the answers from the CJEU based on which it farther ruled inter alia on how the members of the evaluation committee of the contracting authority must behave during the procedures. First, the national court explained that the concept of the conflict of interest is to be understood broadly and in case of even a minor doubt that any member, employee or whomever might be biased in favor of a particular supplier during the public procurement procedures, the public buyers must take all the measures and actions to eliminate this doubt. Thus, the failure to prove that the relevant actions have been taken might suffice to repeal the subsequent decisions made by the contracting authority.

Secondly, the court explained that the burden of proof always rests on the shoulders of the contracting authority. In other words, namely the latter must prove that it has taken all required measures to inspect and eliminate any alleged or real conflict of interest (by suspending the concrete member of the evaluation committee, etc.) and not the other way around.

Finally, the Supreme Court held that the declarations of honesty or other documents alike do not have a priori legal power and thus do not automatically mean that the contracting authority has passed the conflicts of interest test. As mentioned before, the burden of proof that the transparency prevails belongs to the buyer, not the seller.

It is interesting to note that in the text of the decision in eVigilo the Supreme Court referred to the above-mentioned situations as the “outside conflicts of interest”. However, in later practice, as it will be shown, the Supreme Court named them as the “vertical conflicts of interest”.

The development of the concept did not stop there. Very recently, the national Court has confirmed the second part of the doctrine by referring to it as the “horizontal conflicts of interest”. This was done by two main cases both heard in 2016. Before providing their overview, it has to be mentioned that by “horizontal conflicts of interest” the Lithuanian case-law means situations of bid rigging and any kind of agreements which in whatsoever way distort the competition in a particular public procurement procedure, including the ones that are being caught by the competition law rules on the forbidden arrangements and the ones, that were entered into or agreed upon by the suppliers, the tenderers. The contracting authority is not involved in these cases. Hence, from a very rough point of view there is no conflict of interest, as in such cases the tenderers have not the conflict but, on the contrary, the one same goal – to deceive the contracting authority. However, in the eyes of the law, at least how the Supreme Court views it, this is the horizontal conflict of interest.

Thus, the first case was Maniga.[2] This company participated in a public tender arranged by the Klaipėda city municipality. It lost to another company – Sankryža. There was one more participant in a public tender – Biseris. Maniga, a claimant, argued that Sankryža‘s bid must have been rejected because it was directly connected and interdepended with Biseris’. The claimant contended that Sankryža and Biseris breached the principle of transparency, non-discrimination and fair competition, since the former company was 100% owned by the latter, they shared the same managing employees, their bids were resembling in a way of the same writing style, etc., and, therefore this meant that these companies exchanged information while participating in the same tender. Thus, Maniga argued that both of their bids had to be rejected and the claimant had to be awarded a public contract.

The Supreme Court started by the endorsement of the fair and transparent competition in every single public procurement procedure that takes place. The Court mainly referred to the national and EU competition law rules (including Art 101 TFEU) and ruled that the principle of fair competition is an inseparable part of a public procurement process. Hence, each participant must obey and follow that principle. The Court maintained the idea that in order to establish the legal connection between the tenderers, the competition law rules on related economic operators must also be applied.

Farther, the Supreme Court went on with the analysis of the concerted practices in public procurement procedures. The Court ruled that it was unfair to require from the interested party to prove the illegal arrangements between the other suppliers as the information the former tenderer has is always very limited. Thus, by making a reference to the eVigilo findings, the Court concluded that in order to prove the concerted practices in public procurement, it is enough to prove the irrational actions or bizarre business behavior of the relevant suppliers. To enforce this line of reasoning the Supreme Court backed it up with the case-law of the CJEU, namely Ahlström Osakeyhtiö v. Commission [3]. Hence, the Court found in favor of the claimant.

The Maniga case entrenched the foundations for the prohibition of the horizontal conflicts of interest. For the sake of clearness, it is important to mention that neither the case-law nor the law itself hinder the participation of the related companies in the same public procurement process. However, this case highlighted the importance of the principle of fair competition and made it as a ground for the rejection of the bids in public procurement in case the principle is breached. Besides, the ratio of the Maniga decision enabled the contracting authorities to reject the bids of all suppliers in case the signs of the concerted practices are detected. However, needless to say, from the point of the wording of the EU public procurement directives, especially the package of 2004, including the Lithuanian legislation, this ground for rejection is not directly regulated. Therefore, even if the idea of the Court is to promote and extremely safeguard the genuine competition among suppliers is very rational, however, legal technique to implement it remained undeveloped.  

And exactly the latter point shall be resolved not only by the Supreme Court of Lithuania but also by the CJEU, which has been referred to by the former Court for a preliminary ruling in a VSA Vilnius case [4].

This case is similar with its facts to the Maniga case. There was a public tender regarding waste management and utility. There were four tenderers. The first two were interrelated and shared the same shareholder and management board directors. Hence, the third-place bidder started legal proceedings against the contracting authority as well as the two related bidders and claimed that their bids had to be rejected on the legal grounds that were provided in the Maniga case, and that the public contract had to be awarded to the claimant.

The Supreme Court started by defining the scope of the case. The Court outlined that the relevant legal questions were the following: (i) the ex ante obligation of the suppliers and the tenderers to inform the contracting authority of their interdependence; (ii) the obligation of the contracting authority to be pro-active in search of the legally related tenderers and the formal outcome of a failure to act so; (iii) the rules on burden of proof in cases of the concerted practices or unfair competition; (iv) the applicable legal norms in cases of concerted practices or unfair competition which is present during the public procurement procedure. These are not only the questions the Supreme Court raised but also the dilemmas both national Court and the CJEU have to solve. As it might be understood most of the mentioned questions were referred to the Court of Luxembourg for the clarification from the standpoint of the EU public procurement law.

Dilemma #1. Hence, the first important question the Supreme Court faced in VSA Vilnius case is the following: do mutually legally related tenderers (e.g. two subsidiaries of one holding company, etc.) have the obligation to inform the contracting authority of their interdependence and possibly common economic ties? Contrary to the Maniga case, where the Court started with the promotion of the principle of fair competition as a guiding light in these kind of situations, the reasoning in VSA Vilnius started from the analysis of the EU and national legal regulation and coming to the conclusion that neither set of laws provides expressis verbis such requirement for the tenderers. However, the Court reasoned that even in the absence of such regulation this kind of legal obligation should arise from the general public procurement principles – equality, non-discrimination and transparency. On the other hand, namely the principles of equality and transparency require the contracting authority to strictly stick to its own procurement documents in accordance to which the purchase procedure is being organised. Thus, how could the public buyer require that tenderers ex ante inform it on their own initiative about the common economic ties if such obligation has not been explicitly envisaged in the tender documentation?

Dilemma #2. On the same note, the Supreme Court addressed the second question, related to the contracting authorities’ obligation to be pro-active and engage in the investigation on whether the particular tenderers are economically connected. On the one hand, the Court referred to eVigilo case and the conclusions made by itself and the CJEU, namely regarding the very active role of the contracting authority in cases of even alleged (not yet proved!) conflicts of interest. However, on the other hand, it remained unclear to the Court if the passive role of the public buyer should be qualified as a breach of the principle of transparency, keeping in mind that overall the related parties are not forbidden from participation in the same tender. Hence, the question is the following: does the fact that the related tenderers de jure are competing per se mean the presence of the reliable information which, as it has been decided in the eVigilo by the CJEU, first, would be sufficient to hold the procurement in breach of the principle of transparency and the whole procedure illicit and, second, would it require the contracting authority to start the investigation on the merits of this kind of information?

Dilemma #3. The third enquiry must have been the least sophisticated to the Court, since it addressed the purely procedural issue of how to prove the horizontal conflicts of interest in the form of concerted practices, bid rigging, etc. The Supreme Court relied on the standards and jurisprudence in the field of competition law, which allows indirect evidence as admissible in these kinds of situations. In its reasoning, the Court cited the CJEU Eturas case [5] and the conclusions made by the European Court regarding the investigation and the evidence in such cases. Thus, the clearest rule that was made in the VSA Vilnius by the Supreme Court is the one that the indirect evidence while detecting and proving the horizontal conflicts of interest is very welcome.

Dilemma #4. The latter one is the most serious issue at hand. The Maniga and VSA Vilnius cases proved that the crux of the concept of the horizontal conflicts of interest is based on the distorted competition made by the suppliers that are mutually related. In other words, horizontal conflicts of interest appear when two or more tenderers pretend to engage in artificial competition while having a disguised intent to deceive the contracting authority and to gain a non-market price for the supply of goods, execution of works or provision of services. The economic relationship among the tenderers in not a legal problem. However, if such relationship leads to the concerted practices, this is a breach of principle of transparency. It is obvious that the law on public procurement heavily relies on the rules of competition law, especially regarding the notion and forms of bid rigging, concerted practices, etc.

However, the paradox is that from the perspective of the competition law it is not illicit for the two or more inter-related parties to participate in a public procurement procedures, exchange of information, co-ordinate practices, etc. since these economic operators are held as one unit under the doctrine of single economic unit. Hence, by making references to the CJEU Corinne Bodson [6], Viho [7], Akzo Nobel [8] and other cases the Supreme Court expressed the following doubts. First, the Court held that on the one hand, horizontal conflicts of interest imply the breach of both the principle of transparency provided by the public procurement law and the competition law rules on concerted practices. However, on the other hand, the doctrine of single economic unit allows the latter practices (in cases of participation of related parties). This leads to a situation where public procurement law relies on competition law to deem the behavior of the tenderers as illegal, meanwhile, in the light of the competition law such actions are fully available (as mentioned, in case of related economic operators).

Moreover, even if the latter situations were allowed, the Supreme Court expressed doubts on whether it did not mean that the tenderers would be allowed to submit the alternative bids when such bids would be forbidden. In other words, if it is held under the rules of competition law that two related economic operators comprise a single economic unit, doesn’t that mean that if such unit participates in a public tender it submits alternative bids (in the meaning of Art. 45 of Directive 2014/24) and not the two separate bids? Again, the latter situation is hindered by the public procurement regulation and usually by the tender documents.

Based on the doubts raised, the Supreme Court referred the following questions to the CJEU for a preliminary ruling:

1. Do the principles of equality and transparency, envisaged in Art. 2 of the Directive 2004/18 due to the freedom of movement, provided in TFEU Art. 45 and 56, including the principles of free and fair competition (construed together or separately) must be understood and interpreted as:
In case mutually related suppliers, whose economic, management, financial and other relations may objectively cause doubts regarding the independence and the protection of the confidential information and (or) may cause (potentially) the advantage against the other suppliers, if they decide to submit separate (their own) bids in the same public tender, do they have in any case an obligation, notwithstanding whether such obligation of theirs is provided in the national legislation, to disclose to the contracting authority this kind of relationship, even if the contracting authority does not require so on its behalf?
2. If the answer to the first question is:
a) Positive (i.e. the suppliers in any case must disclose to the contracting authority their relationship), does it mean that the fact of the failure to execute this obligation fully or partially is sufficient for the contracting authority and for the judicial review body to deem that the related tenderers, who have submitted the separate bids in the same public tender are participating without the competition (fraud)?
b) Negative (i.e. there is no obligation for the suppliers – neither provided by the legislation nor by the purchase documents – to disclose their relationship), does the contracting authority bear the risk of the participation of such related suppliers, including the outcomes that arise thereof, in case the contracting authority does not include this kind of obligation for the tenderers in a purchase documents?
3. Despite the answer to the first question and taking into consideration the decision of the CJEU in eVigilo case, do the legal norms indicated in the first question as well as the third section of the first paragraph of Art. 1(1) of the Directive 89/665 and the Art. 2 (1) (b) of the same Directive (separately or together, without a limitation to these legal norms), must be understood and interpreted that:
a) In case the important relationships (connections) among some of the tenderers become clear in any form during the public procurement procedures to the contracting authority, does the latter, despite its own evaluation of this or other facts (e.g. dissimilarity of the bids of the suppliers, including their content, the public oath of the supplier to compete with the other tenderers in a fair manner, etc.), have an obligation to address separately the related tenderers of a public tender and to seek from them the explanation on if and how their personal condition is in line with the free and fair competition of the suppliers?
b) The contracting authority, having this kind of obligation, however, failing to execute it, is it a sufficient ground for the court to consider its actions, which have not guaranteed the transparency and the objectivity of the procedures, as illegal by not requiring from the claimant or without deciding on its own initiative regarding the influence of the state of the related parties to the outcome of a public tender?
4. Do the legal norms, indicated in the third question as well as the provisions of Art. 101 (1) of the TFEU (together or separately, but not limiting thereof) and taking into consideration the decisions of the CJEU in eVigilo, Eturas and VM Remonts, must be understood and interpreted that:
a) In case the claimant having found about the rejection of the bid of one of the related tenderers, meanwhile, awarding the contract to the other one as well as taking into consideration the other circumstances related to their (related companies) participation in a public tender, such as the same management of the companies, the sole holding company, which has not taken part at the public tender, the fact that the related suppliers have not disclosed their inter-relationship and have not provided any explanation regarding this, inter alia, because there was no request, etc., and when in the light of the afore-mentioned the contracting authority has not taken any action, is this information at its least is sufficient to claim the actions of the contracting authority as illegal and the ones that have not ensured the transparency and objectivity of the public tender, without the separate requirement from the interested party to prove the unfair activities of the related tenderers?
b) The related tenderers do not prove the actual and fair competition in a public tender only because one of the tenderers voluntarily provided the declaration of honesty as well as it has quality management systems implemented and the content of the bids is not similar?
5. Do the actions of the tenderers, who participate separately, taking into consideration the submission of the declaration of honesty voluntarily by one of the bidders, maybe be in principle evaluated from the perspective of the Art. 101 TFEU and the related case-law of the CJEU?

A couple of closing remarks regarding the aforementioned.

Firstly, the questions of the Supreme Court of Lithuania have cross-border interest. The answers are important for the notion, understanding and implementation of the concept of conflicts of interest in a public procurement practice. Moreover, this will have the impact on the understanding the content of Art. 24 of the Directive 2014/24.

Secondly, the CJEU most likely will have to provide the wide interpretation of the concept at issue, because the public procurement regulation must tackle the practices that distort competition.

Finally, in situations which are covered by the doctrine of the single economic unit (in cases of the same business group companies, etc.) it is very possible that the public procurement law would prevail over the general rules of the competition law, as in any case the public procurement procedure may not allow the situations where the companies of the same economic group would agree on the commercial conditions that would be profitable to them but expensive in terms of the financial funds, quality, even the prestige to the contracting authority and the public in general.

Dr. Deividas Soloveičik, LL.M

Dr Deividas Soloveičik is a Partner and Head of Public Procurement practice at COBALT Lithuania. He represents clients before national courts at all instances and arbitral institutions in civil and administrative cases, provides legal advice to Lithuanian and foreign private clients and contracting authorities, including the European Commission , on the legal aspects of public procurement and pre-commercial procurement.

Dr Soloveičik is an Associate Professor and researcher in commercial law at Vilnius University and a contributor to legal publications. He also closely cooperates with globally recognized academic members of the legal profession. Since 2011, MCIArb. Dr Soloveičik is a member of the Chartered Institute of Arbitrators; since 2016, he is a member of the European Assistance for Innovation Procurement – EAFIP initiative promoted by the European Commission and a recommended arbitrator at Vilnius Court of Commercial Arbitration.

Guest blogging at HTCAN: If you would like to contribute a blog post for How to Crack a Nut, please feel free to get in touch at a.sanchez-graells@bristol.ac.uk. Your proposals and contributions will be most warmly welcomed!

___________________

[1] Judgment of the Court of Justice of the European Union of 12 March 2015, eVigilo Ltd v Priešgaisrinės apsaugos ir gelbėjimo departamentas prie Vidaus reikalų ministerijos, Case C-538/13, ECLI:EU:C:2015:166.
[2] Judgement of the Supreme Court of the Republic pf Lithuania of 4th March,2016 in a civil case No. e3K-3-155-415/2016.
[3] Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 ir C-125/85 until C-129/85, ECLI:EU:C:1988:447.
[4] Ruling of the Supreme Court of the Republic pf Lithuania of 11th October,2016 in a civil case No. e3K-3-439-469/2016.
[5] Case C-74/14, Eturas and Others v Lietuvos Respublikos konkurencijos taryba, ECLI:EU:C:2016:42.
[6] Case C-30/87,Corinne Bodson v SA Pompes funèbres des régions libérées, ECLI:EU:C:1988:225.
[7] Case C-73/95, Viho Europe BV v Commission of the European Communities,  ECLI:EU:C:1996:405.
[8] Case, C-550/07, Akzo Nobel Chemicals Ltd and Akcros Chemicals Ltd v European Commission, ECLI:EU:C:2010:512.