Public consultation on procurement planning by the Spanish Competition Authority now open (until 20/12)

competencia_logo_625.jpg

The Spanish Competition Authority (Comisión Nacional de los Mercados y la Competencia, CNMC) is in the process of revising its 2011 Guide on public procurement and competition to reflect recent developments and the change of regulatory framework derived from the transposition of the 2014 EU Public Procurement Package [for a critical assessment of the original guide, in Spanish, see A Sanchez-Graells, ‘Una Visión Crítica de la 'Guía Sobre Contratación Pública y Competencia' Publicada por la CNC’ (2011) 21 Gaceta Jurídica de la Unión Europea y de la Competencia 15-31].

The CNMC plans to update their guidance in steps, and has started the process by focusing on procurement planning. In order to gather input into the formulation of guidance on procurement planning from a competition perspective, the CNMC has published a short preliminary working paper (in Spanish), is holding a public conference on 3 December (in which I am honoured to participate), and has also opened a public consultation (closes 20 Dec 2019).

Even thought, unfortunately, this is a process mainly conducted in Spanish, I am sure the CNMC would welcome any contributions on best procurement planning practices and on the impact of planning on competition via email: dp.ayudaseinformesnormativos@cnmc.es (subject: “Consulta pública Planificación de la contratación pública", indicating whether your contribution can be published or should remain confidential). In case of interest, below is my own contribution to the public consultation (in Spanish).

Litigation in Spanish railroad electrification cartel highlights further inadequacies of regulation of bid rigger exclusion

train-wreck_137015.jpg

In a new episode of the Spanish sainete of the railroad electrification cartel (see here for an overview), it has now emerged that one of the companies affected by the exclusion ground (prohibición de contratar) declared in the resolution of the Spanish National Commission on Markets and Competition (CNMC) of 14 March 2019 subsequently secured interim measures suspending its effectiveness on 19 July 2019.

The freezing order prevents (Spanish) contracting authorities from relying on the exclusion ground and thus shortens the maximum period of (future) exclusion of the colluding companies, unless the CJEU revises its case law on the time-limit calculation for such grounds established in Vossloh-Laeis (24 October 2018, C-124/17, EU:C:2018:855). The decision also highlights issues concerning the cross-border effects of litigation on exclusion grounds. In this follow-up post, I discuss these two issues.

The interim measures decision

Quick recap: it should be stressed that the Spanish transposition of Article 57(4)(d) has resulted in a system whereby the exclusion of economic operators on the basis of previous infringements of competition law is mandatory under Article 71 of Law 9/2017 on Public Sector Procurement (LCSP). However, the scope and duration of such exclusion generates some difficulties, in particular when they are not established in the original decision declaring the infraction and imposing the measure—which is precisely the case of the railroad electrification cartel. In such cases, a further administrative procedure needs to be completed and the scope and duration of the mandatory exclusion (prohibición de contratar) are to be established by decision of the competent Minister.

The effectiveness of the mandatory exclusion ground in the period running from the initial infringement decision and the further Ministerial decision is contested. Two opposing schools of thought exist. One that gives automatic effect to the exclusion ground despite the future specification of its scope and duration, and the opposing view that considers that the measure is incomplete and cannot generate (negative) effects against the sanctioned undertaking until the Ministerial decision is adopted.

The CNMC expressed the first view in its railroad electrification decision, when it stated that ‘regardless of the time limits within which the duration and scope [of the prohibition] must be set [by the Minister of Finance] ... it is possible to identify an automatism in the prohibition of contracting derived from competition law infringements, which derives ope legis or as a mere consequence of the adoption of a decision that declares said infraction, as established in the mentioned Article 71.1.b) of [Law 9/2017]‘ (page 319, own translation full decision available in Spanish).

The Spanish High Court (Audiencia Nacional), in a Judgment of 19 July 2019 (ES:AN:2019:1673A, hat tip to Alfonso Rincón García-Loygorri for posting it on LinkedIn) adopted the same view and recognised that the measure was bound to immediately restrict the affected undertakings’ ability to participate in public tenders. Considering that it is likely that the final decision on the main appeal of the cartel decision arrives after the expiry of the three year maximum duration foreseen for the exclusion ground and that (should the appellant prevail) the effects of such exclusion would be very difficult, if not impossible to correct at that stage, the High Court decided to suspend the effectiveness of the mandatory exclusion ground.

Implications in terms of maximum duration of the exclusion

Quick recap: the CJEU has established that ‘where an economic operator has been engaged in conduct falling within the ground for exclusion referred to in Article 57(4)(d) of that directive, which has been penalised by a competent authority, the maximum period of exclusion is calculated from the date of the decision of that authority‘ (Vossloh Laeis, above, para 42).

I criticised the CNMC for creating legal uncertainty by not establishing the scope and duration of the exclusion ground in its initial decision. I argued that the CNMC knew or should have known that, as a matter of directly applicable EU law, de facto the maximum exclusion period can run for three years, up to 14 March 2022. Therefore, by referring the file to the Minister and creating legal uncertainty as to the interim effects of the prohibition to contract with a yet to be specified scope and duration, the CNMC actually bought the competition infringers time and created a situation where any finally imposed prohibition to contract is likely to last for much less than the maximum three years.

The High Court’s Judgment raises the same criticisms. While the High Court explicitly took into account the fact that the undertakings could find themselves in a position of not being easily compensated for the undue exclusion from public tenders in case of prevailing in their appeal of the CNMC decision, the High Court ignored that its freezing order will create the reverse effect in case the appeal is dismissed. By preventing (Spanish) contracting authorities from excluding the competition infringers from tenders for an indefinite period starting on 19 July 2019, the High Court has created the risk that the undertakings are never excluded from public tenders because such exclusion is time barred by the time the CNMC decision becomes final—which does not solely depend on the outcome of the High Court’s proceedings, but is subject to a potential further appeal to the Supreme Court.

This highlights once again the inadequacy—or, at least, partiality—of the CJEU Vossloh criterion that the maximum period of exclusion starts running at the time of adoption of the initial infringement decision. It seems clear that, where that decision is contested and, in particular, where interim measures are obtained to freeze its effects—the maximum period of exclusion needs to be calculated taking that into account. Otherwise, the simple fact of litigating buys competition infringers immunity from the debarment system foreseen in Directive 2014/24/EU and thus excludes its effet utile. That cannot be right.

Territoriality of effects

The new episode of the Spanish sainete also raises questions concerning the cross-border effects of the CNMC decision. While Spanish contracting authorities are effectively enjoined from giving effect to the mandatory exclusion ground, the situation is by no means necessarily the same in other EU/EEA jurisdictions. Non-Spanish contracting authorities could (justifiably) be tempted to apply domestic mandatory or discretionary exclusion grounds based on the fact that the relevant undertakings were sanctioned for bid rigging by the CNMC. This could be the case whether they are aware or not of the High Court Judgment, in particular where they have discretion in this matter.

Should any such decision be challenged, the issue should make its way to the CJEU, which would have a hard time finding ways of squaring this practical difficulty with the differentiated treatment that Art 57 of Directive gives to grounds based on a ‘conviction by final judgment‘ (Art 57(1)) and those based on decisions and judgments not subjected to that finality requirement (notably, Art 57(4)), as well as with the self-imposed constraint of the way the maximum time-limit is calculated as per Vossloh.

Once again, we are yet to see the final act of this sainete…

Bid rigging conspiracy in railroad electrification works: A very Spanish 'sainete'

las-leandras.jpg

A case of bid rigging in works contracts for high-speed and conventional railroad electrification in Spain evidences a number of shortcomings in the domestic transposition of the 2014 rules on discretionary exclusion of competition law offenders from public procurement tenders, as well as some dysfunctionalities of their interpretation by the Court of Justice of the European Union (CJEU) in its Judgment of 24 October 2018 in Vossloh Laeis, C-124/17, EU:C:2018:855. The unilateral price adjustment of live contracts sought by the main victim of the cartel, the Spanish rail network administrator ADIF comes to raise very significant issues on the limits to the ‘self-protection’ (or private justice) for contracting authorities that are victims of bid rigging. In this post, I point to the main issues that puzzle me in this very Spanish sainete. I am sure there will be plenty debate in Spanish legal circles after the holidays…

Legal background: EU level: Art 57(4)(c) and (d) of Directive 2014/24/EU

As is well known, Article 57(4) of Directive 2014/24/EU establishes discretionary grounds for the exclusion of economic operators from public procurement tenders. In relation to economic operators that have breached competition law, there are two relevant grounds.

First, Art 57(4)(c) foresees the possibility of exclusion ‘where the contracting authority can demonstrate by appropriate means that the economic operator is guilty of grave professional misconduct, which renders its integrity questionable‘. This was interpreted by the CJEU as covering entities that had been sanctioned for breaches of competition law in relation to the earlier rules of Directive 2004/18/EC (Art 45(2)(d)) as an instance of their being ‘guilty of grave professional misconduct proven by any means which the contracting authorities can demonstrate’. The CJEU established in unambiguous terms that ‘the commission of an infringement of the competition rules, in particular where that infringement was penalised by a fine, constitutes a cause for exclusion under Article 45(2)(d) of Directive 2004/18’ in its Judgment of 18 December 2014 in Generali-Providencia Biztosító, C-470/13, EU:C:2014:2469 (para 35).

Second, Art 57(4)(d) allows for the exclusion ‘where the contracting authority has sufficiently plausible indications to conclude that the economic operator has entered into agreements with other economic operators aimed at distorting competition‘. The relationship between both exclusion grounds relating to competition law infringements is somewhat debated. I have argued elsewhere that Art 57(4)(c) should still be used as the legal basis for the exclusion of economic operators that have already been sanctioned for previous bid rigging offences, whereas Art 57(4)(d) creates an additional ground for exclusion based on indicia of contemporary collusion. For details, see A Sanchez-Graells, Public Procurement and the EU Competition Rules (2nd ed, Hart, 2015) 296-301.

Of course, discretionary exclusion on grounds of infringements of competition law can be modulated by the rules on self-cleaning in Art 57(6) Directive 2014/24/EU. It is also important to add that these discretionary exclusion grounds can be applied for a period not exceeding three years from the date of the relevant event, as per Art 57(7) Directive 2014/24/EU. The CJEU has interpreted the ‘relevant event’ in this context, and clarified that ‘where an economic operator has been engaged in conduct falling within the ground for exclusion referred to in Article 57(4)(d) of that directive, which has been penalised by a competent authority, the maximum period of exclusion is calculated from the date of the decision of that authority‘ (Vossloh Laeis, above, para 42)

Legal background: domestic level: the transposition by Law 9/2017

The transposition into Spanish law of these provisions has introduced some important modifications.

First, these exclusion grounds have been made mandatory under Article 71 of Law 9/2017 on Public Sector Procurement, as discussed by P Valcarcel, ‘Transposition of Directive 2014/24/EU in Spain: between EU demands and national peculiarities‘ in S Treumer & M Comba (eds), Modernising Public Procurement: The Member States Approach, vol. 8 European Procurement Law Series (Edward Elgar, 2018) 236-237. For a broader description of the Spanish system of mandatory exclusion (ie through ‘prohibiciones de contratar,’ or prohibitions on contracting), see A Sanchez-Graells, 'Qualification, Selection and Exclusion of Economic Operators under Spanish Public Procurement Law' in M Burgi, S Treumer & M Trybus (eds), Qualification, Selection and Exclusion in EU Procurement, vol. 7 European Procurement Law Series (Copenhagen, DJØF, 2016) 159-188.

Second, the grounds in Art 57(4)(c) and (d) of Directive 2014/24/EU have been transposed in a seemingly defective manner. Art 57(4)(d) has been omitted and Art 57(4)(c) is reflected in Art 71(1)(b) of Law 9/2017, according to which there is a prohibition to enter into a contract with an ‘economic operator … guilty of grave professional misconduct, which renders its integrity questionable, in matters such as market discipline, distortion of competition … in accordance with current regulations’ (own translation from Spanish).

Thirdly, Art 72(2) of Law 9/2017 foresees two ways in which the mandatory exclusion ground based on a prior firm sanction for competition infringements can operate. On the one hand, the prohibition to enter into a contract with competition law infringers ‘will be directly appreciated by the contracting bodies when the judgment or administrative resolution [imposing the sanction] had expressly established its scope and duration, and will be in force during the term indicated therein’ (own translation from Spanish). On the other hand—and logically, as a subsidiary rule—it is also foreseen that ‘In the event that the judgment or administrative resolution does not contain a ruling on the scope or duration of the prohibition to contract … the scope and duration of the prohibition shall be determined by means of a procedure instructed for this purpose, in accordance with the provisions of this article’ (own translation from Spanish). Such procedure is rather convoluted and involves a decision of the Minister of Finance on the advice of the State Consultative Board on Public Procurement.

Fourthly, and in an extreme pro-leniency fashion, Art 72(5)II of Law 9/2017 has established that the prohibition to enter into contracts will not apply to economic operators that have self-cleaned and, in particular, to those that have obtained leniency in the context of competition enforcement procedures. That is, there is an exemption from the otherwise applicable exclusion ground based on infringements of competition law for undertakings that demonstrate the ‘adoption of appropriate technical, organisational and personnel measures to avoid the commission of future administrative infractions, which include participating in the clemency program in the field of competition law‘ (own translation from Spanish).

It is also odd that the provision does not require economic operators to have ‘clarified the facts and circumstances in a comprehensive manner by actively collaborating with the investigating authorities‘, which was the main issue at stake in the Vossloh Laeis litigation.

A controversial decision by the Spanish National Commission on Markets and Competition (CNMC)

On 14 March 2019, the CNMC adopted a decision against 15 construction companies finding them responsible for a long-lasting bid rigging scheme to manipulate the tenders for public contracts works relating to different aspects of high-speed and conventional railroad electrification (full decision available in Spanish). One of the novel aspects of the decision is that the CNMC explicitly activated the prohibition to enter into contracts against the competition infringers. However, the CNMC did so in very peculiar manner.

The oddity of the decision mainly lies on the fact that CNMC decided not to establish the scope and duration of the prohibition to contract, but simply to refer the case to the State Consultative Board on Public Procurement (see pages 317-320). This was the object of criticism in a dissenting vote by Councillor María Pilar Canedo, who stressed that the CNMC should have set the scope and duration of the prohibition to contract in its decision (pages 366-370). The position of the CNMC is certainly difficult to understand.

On the one hand, the CNMC stressed that ‘regardless of the time limits within which the duration and scope [of the prohibition] must be set [by the Minister of Finance] ... it is possible to identify an automatism in the prohibition of contracting derived from competition law infringements, which derives ope legis or as a mere consequence of the adoption of a decision that declares said infraction, as established in the mentioned Article 71.1.b) of [Law 9/2017]‘ (page 319). On the other hand, however, the CNMC decided to (potentially) kick the effectiveness of such prohibition into the long grass by not establishing its scope and duration in its decision—and explicitly saying so (unnecessarily…). No wonder, contracting authorities will have some difficulty applying the automaticity of a prohibition which time and scope are yet to be determined.

Moreover, the CNMC was aware of the CJEU decision in Vossloh Laeis (above), to which it referred to in its own decision (in a strange manner, though). In that regard, the CNMC knew or should have known that, as a matter of directly applicable EU law, de facto the maximum exclusion period can run for three years, up to 14 March 2022. Therefore, by referring the file to the Minister of Finance via the State Consultative Board on Public Procurement and creating legal uncertainty as to the interim effects of a seemingly prohibition to contract with a yet to be specified scope and duration, the CNMC actually bought the competition infringers time and created a situation where any fianlly imposed prohibition to contract is likely to last for much less than the maximum three years.

The (for now) final twist: ADIF takes justice in its own hands

As if this was not enough, according to the Spanish press (see the main story in El Pais), the main victim of the cartel—the Spanish rail network administrator, ADIF—has now decided to take justice in its own hands.

According to the report, ADIF has written to the relevant companies announcing claims for damages—which is the ordinary reaction that could be expected. However, it has also taken the decision of demanding an anticipation of the compensation from those companies with which it has ‘live’ contracts, to which it has demanded a 10% price reduction. What is more, ADIF has decided to withhold 10% of the contractual price and to deposit in an escrow account before a notary, as a sort of sui generis self-created interim measure to ensure some compensation for the damages suffered from the cartel. The legal issues that this unilateral act generates are too many to list here. And these will surely be the object of future litigation.

What I find particularly difficult to understand is that, in contrast with this decisively aggressive approach to withholding payment, ADIF has awarded contracts to some of the competition infringers after the publication of the CNMC decision. And not a small number of contracts or for little amounts. In fact, ADIF has awarded over 280 contracts for a total value close to €300 million.

Thus, ADIF has largely carried out its business as usual in the award of public works contracts, both ignoring the rather straightforward argument of automaticity of the prohibition to contract hinted at by the CNMC— though based on a convoluted and rather strained interpretation of domestic law (Art 72(2) Law 9/2017)—and, more importantly, the discretionary ground for exclusion in Art 57(4)(d) of Directive 2014/24/EU.

There will certainly be some more scenes in this sainete…


Spanish competition watchdog CNMC issues report on health care outsourcing procurement

The Spanish Competition Authority has recently published a report on the application of its Guide on Public Procurement and Competition to public health care provision-related procurement in Spain (only available in Spanish: Aplicación de la Guía de Contratación y Competencia a los procesos de licitación para la provisión de la sanidad pública en España). 

The report is interesting to read and it identifies some common trends in competition-reductive procurement practices (if not fully suppressive of meaningful competition) and areas for massive improvement in Spanish health care-related procurement. 

Some of them may offer valuable insights for other countries that also organise their health care provision around a national health system. These are some of the aspects of the report that I find more interesting:

1. The report is mainly concerned with outsourcing processes, whereby the competent (regional) public authorities tender contracts for the construction and management, or only the management, of health care facilities (mainly hospitals). This is an area that will remain lightly regulated in the future EU Directive on concessions (Art 17) and in the new version of the Directive on public sector procurement (Arts 74 to 76a). Consequently, the recommendations and best practices identified in the CNMC report may help in the construction of a fuller set of (binding and non-binding) guidelines for health care management outsourcing.

2. The report offers a radiography of the hospital sector in Spain, which shows that it is rather large and that there is a very relevant presence of private investment in the sector. Overall, there are 789 hospitals in Spain (162,070 beds), which means that each hospital serves an average of roughly 59,300 inhabitants (290 inh/bed). 

However, there are significant regional differences in availability of total hospital services, ranging from Andalusia at 378 inh/bed to Catalonia at 218 inh/bed. Furthermore, it is also interesting that only 325 of the 789 hospitals are public (41%), but they accumulate almost 67% of available beds--which means that the availability of public hospital services actually ranges between Catalonia at 523 inh/bed and Aragon at 308 inh/bed. All regions have schemes of arrangement with private hospitals, so that they extend 'public' coverage through private hospitals (49% of private hospitals are included in such schemes, again with large variations ranging from 100% of private hospitals being included in the 'extended public network' in La Rioja to only 22% of private hospitals in Catalonia). 

The big discrepancies between the availability of total and public hospital services shows large regional differences in private investment and alternative (ie non-public) health care management strategies. This also seems to show that private hospitals tend to be smaller than public hospitals (116 v 334 beds on average)--and, probably, easier (but more expensive) to manage, at least in terms of general costs if economies of scale are properly exploited in the public system (a big if, I think, although the report offers no data to test this). It may also be worth stressing that 21% of private hospital capacity (by number of beds) is controlled and run by the Catholic church and religious organisations. The next larger private (or non-public) player only reaches 4%.

The distribution by areas of activity is also relevant, and it is worth noting that generalist, geriatric and psichiatric hospitals accumulate almost 90% of the available beds--which seems to indicate that there is room for further specialisation in the sector.


The report also offers more detailed analysis of the regions where there has already been some outsourcing of public health care management: Catalonia, Madrid, Valencia, La Rioja and Navarra.

3. The main body of the report focusses on the 5 aspects of health care management outsourcing that are more susceptible to create distortions of competition: (i) the design of the tender procedure and the setting up of the technical specifications, (ii) the setting up of selection criteria, (iii) the choice and weighting of award criteria, particularly those related to (non-measurable) qualitative elements, and (iv) issues related to contract modification.

It is remarkable that, in all of these areas, the CNMC has identified specific examples of very clear distortions of competition. It is worth noting, for instance, that:

a) There has been an excessive degree of bundling of specialist and general services in hospital outsourcing (sometimes forcing the hospital concessionaire to enter into existing public services contracts with third party providers of specialist services, such as image diagnostics or laboratory analysis).

b) Regional authorities have not availed themselves of proper strategic division of tenders into lots and the dominant strategy (one lot, one hospital) may have facilitated collusion.

c) Initial contract duration may have been excessive, with a median of 30+ years for works concessions (building + managing hospitals) and 10 years for service concessions/public service contracts (management only of an existing hospital). Some of them also include relatively generous extension/renewal provisions.

d) Of the 19 contracts that included health management (others were limited to the management of the premises, but included no sanitary provision), 15 were awarded to the only tenderer submitting an offer. In the other 4 instances, only 2 offers were received. This seems to indicate that participation requirements were exceedingly restrictive (or, in an alternative and very personal view, that there was no expectation of effective competition, either due to the existence of a market sharing agreement or widespread corruption, particularly in the case of Valencia and Madrid, where criminal investigations are underway).

e) The setting of very demanding selection criteria (particularly in terms of financial standing and previous experience) have limited dramatically the number of potential offerors and been particularly alienating for temporary unions of undertakings, as a relevant part of the tender documents required that each of the undertakings individually considered met all of the requirements. This is a stark breach of procurement law and, as such, should have been the object of legal challenges.

f) There was an insufficient publicity and advertisement of the tendering for public service concessions worth Eur 4,000 mn in the Madrid region (advertised only in the region itself). This indicates that, in reality, there may be some need for the extension of publicity requirements to concession contracts as the future Directive aims to do. However, this may also have been a breach of EU law requirements, given that the contracts seem to have (at least potential) cross border interest.

g) There was an insufficient disclosure of information with relevant financial implications, such as the personnel costs to be assumed by concessionaires of existing hospitals, or the system of mutual invoicing between public hospitals (which made it difficult to calculate the cost and revenue structure of the concession, particularly for relatively unexperienced tenderers). The information asymmetries were even higher when it came to disclosure of health planning and other requirements.

h) There was widespread misuse of the price criterion as one of the key elements to award the contract. Price assessment formulae based on average prices, or that gave a very low weight to prices (of 30% in construction concessions), or that included irrelevant criteria (such as giving 30% of weight to the establishment of a stock-option scheme by the concessionaire) might have limited the ability of regional authorities to obtain value for money in the outsourcing of hospital management.

i) There were several instances of double-count of elements as both selection and award criteria, particularly as previous experience is concerned. This is another blatant breach of procurement law and, as such, should have been the object of legal challenges.

j) Insufficient or too basic quality control mechanisms and penalties for breaches thereof were included in a significant number of concession schemes. Also, remuneration was always calculated on a per capita basis, so that concessionaires and public service providers would always be remunerated almost regardless of the level of quality or actual provision of services (80% of the per capita support working as a common floor or minimum remuneration).

k) Most tender documentation either imposed or facilitated subcontracting of up to 50-60% of the contract and no proper oversight mechanisms were in place, so that concessionaires were basically free to subcontract very significant parts of their contracts as they saw fit.

l) Excessive resort to contractual modifications: "Of the 38 contracts for which information is available, there have been changes in 24 of them (64%). In 7 of the 24 contracts modified there have been two changes to the contract."

4. In its conclusions (a bit too mild in my opinion, particularly in view of the major irregularities documented in the report), the CNMC recommends, among others, the following measures (see press release in English):

  • When designing tender processes, the open procedure must be used whenever possible, as that procedure is the most conducive to competition and precludes contracts that cannot be justified on account of the pay-back times for investments.
  • As regards access to tenders for participants, publication should be more widespread in order to open up access to the highest number of potential bidders possible.
  • With respect to the weighting of criteria and the procedure for the award of contracts, a suitable weighting should be attached to the variables to avoid leaving excessive discretion to the award body. In the case of healthcare services, the overarching goal is to ensure quality in the provision of services to patients, so that a balance must be struck between competition in the price variable and the quality of the service.
  • Lastly, as regards the implementation of contracts, it is proposed, among other recommendations, that the specifications should describe the elements that define the quality of contract performance and should contain credible and robust mechanisms for monitoring and penalising failures to meet the requirements of those elements. The specifications should also lay down remuneration and transparency mechanisms that encourage the awardee to provide high quality services (emphasis added).
In my view, this Report brings to light a very serious problem and a massive challenge in the modernisation and reform of health care management in Spain. I started wondering if a sectoral regulator would not be necessary, as the ones existing in England (Monitor) or The Netherlands (NZa), as this sector seems to really be crying for some close scrutiny...