Reg. 46 of the Public Contracts Regulations 2015 (PCR2015) transposes Article 46 of Directive 2014/24 and upgrades the rules on division of contracts into lots. Pedro has already criticised this provision and the "new preference for division into lots" that it encapsulates in his entry of the day, and has raised important points on the effect of lot division on procurement complexity and facilitation of collusion (if improperly carried out). I broadly agree with his views, except on the absolute optionality on the use of lots, and would like to add some comments.
These rules attracted significant attention during the reform process that led to the adoption of these new rules after the European Commission proposed a rule on division of contracts of a value above 0.5mn Euro whereby contracting authorities would have to provide reasons for decisions not to split those contracts into lots--hence setting a presumption or expectation for contracts to be generally divided into lots [see art 44 of the 2011 Proposal and A Sanchez Graells, “Are the Procurement Rules a Barrier for Cross-Border Trade within the European Market? — A View on Proposals to Lower that Barrier and Spur Growth” in C Tvarnø, GS Ølykke & C Risvig Hansen, EU Public Procurement: Modernisation, Growth and Innovation (Copenhagen, DJØF, 2012) 107-133].
The final version of the rule, now in Art 46 Dir 2014/24 is softer, is less prescriptive regarding the need to subdivide contracts into lots, merely indicating that contracting authorities may decide to award a contract in the form of separate lots and may determine the size and subject-matter of such lots [reg.46(1)]. However, the principle of "divide or explain" has remained as a general rule. Indeed, reg. 46(2) PCR2015 requires that contracting authorities provide an indication of the main reasons for their decision not to subdivide contracts into lots.
Consequently, in my view, the rule (or at least the expectation) is that contracting authorities need to make an effort to determine whether they can divide the contract into lots and, only if there is a plausible reason not to do so, tender a single contract for the entirety of the object of the procurement. Moreover, the explanations provided for their decisions not to divide the contract into lots should be open to challenge, among other reasons, on the basis of the general principles of procurement [reg.18 PCR2015] and, notably, the principle of competition.
After setting up this general framework, reg.46 PCR2015 goes on to establish rules whereby contracting authorities can establish restrictions on the possibilities to tender for one, for several or for all of the lots [reg. 46(3)] and, even where tenders may be submitted for several or all lots, limit the number of lots that may be awarded to one tenderer, provided that all this information, including the maximum number of lots per tenderer, is stated in the contract notice or in the invitation to confirm interest [reg.46(4)].
In case any of these restrictions are employed, contracting authorities shall indicate in the procurement documents the objective and non-discriminatory criteria or rules they intend to apply for determining which lots will be awarded where the application of the award criteria would result in one tenderer being awarded more lots than the maximum number [reg.46(5)]. And where more than one lot may be awarded to the same tenderer, contracting authorities may award contracts combining several or all lots where they have specified in the contract notice or in the invitation to confirm interest that they reserve the possibility of doing so and indicate the lots or groups of lots that may be combined [reg.46(6)].
These rules consolidate possibilities that already existed under the previous rules (as none of them were explicitly prohibited) and allow for contracting authorities to take a more sophisticated approach to lot division. The UK legislator has decided not to make use of the possibility foreseen in Art 46(4) Dir 2014/24, whereby Member States may render it obligatory to award contracts in the form of separate lots
under conditions specified in accordance with their national law
and having regard for Union law.
Division of contracts into lots is an issue of particular relevance in terms of SME access to procurement and with an immediate effect on competition for the public contract, and in the market concerned. Hence, applying these rules in an adequate manner is of utmost importance. I discuss these issues at length in Public Procurement and the EU Competition Rules, 2nd edn (Oxford Hart, 2015) 347-352. The following is a simplified version of my comments.
Bundling and Aggregation of Contracts. Similarly to what was discussed in relation to centralisation of purchases, it is clear that these two trends ‘should be carefully monitored in order to avoid excessive concentration of purchasing power and collusion, and to preserve transparency and competition, as well as market access opportunities for SMEs’(rec (59) dir 2014/24). Hence, the rules applicable to the bundling and aggregation of contracts should have a prominent position under the applicable procurement rules. This was not the case under the 2004 EU public procurement directives, which regulated neither the division of contracts into lots, nor the bundling of those lots or the aggregation of contracts by the public buyer. The only rules regarding division of contracts into lots aimed at establishing specific criteria for the calculation of the value of the tendered contracts for the purpose of determining the applicability of the EU public procurement regime (art 9(5) dir 2004/18, which is now contained in art 5(8) and 5(9) dir 2014/24)—and, more specifically, with the purpose of preventing the circumvention of EU rules by the artificial division of contracts into lots whose value remains below the thresholds that trigger their application. Other than that, reference to the division of contracts into lots, their bundling or aggregation was only made in relation to contract notices—which, where the contracts are subdivided into lots, must indicate ‘the possibility of tendering for one, for several or for all the lots’ (Annex VII A dir 2004/18).
Therefore, Member States seemed to hold substantial discretion to set domestic public procurement rules on the division of contracts into lots, the bundling or aggregation of lots and contracts to be tendered together, the establishment of rules allowing or not for multiple and/or conditional tendering for different lots in a given tender procedure, etc. The situation has now been slightly altered by the inclusion of more specific rules concerning the division of contracts into lots in article 46 of Directive 2014/24, which fundamentally rest on a general expectation that contracting authorities will consider the possibility of dividing contracts into lots and, where they decide against it, provide a justification (ie, ‘divide-or-explain’ requirement). Indeed, according to article 46(1), contracting authorities ‘may decide to award a contract in the form of separate lots and may determine the size and subject-matter of such lots’ and, when they decide otherwise, they ‘shall … provide an indication of the main reasons for their decision not to subdivide into lots’, which may be included in the procurement documents. This duty to ‘divide-or-explain’ is conceived as a soft requirement not amenable to review, as indicated in recital (78) of Directive 2014/24, where it is explained that: ‘The contracting authority should have a duty to consider the appropriateness of dividing contracts into lots while remaining free to decide autonomously on the basis of any reason it deems relevant, without being subject to administrative or judicial supervision. Where the contracting authority decides that it would not be appropriate to divide the contract into lots, the individual report or the procurement documents should contain an indication of the main reasons for the contracting authority’s choice’.
Moreover, it should be taken into consideration that under article 46(4) of Directive 2014/24, ‘Member States may [render] it obligatory to award contracts in the form of separate lots under conditions to be specified in accordance with their national law and having regard for Union law’—which, a contrario, implies that Member States are free to reduce the requirements concerned with the division of contracts into lots to that soft requirement of ‘divide-or-explain’. Consequently, given the full discretion that Member States retain when deciding whether to implement the more specific rules in article 46 of Directive 2014/24 or not, and whether to make them mandatory or voluntary, it is appropriate to take a general approach to the competition analysis of the decisions (and rules) concerning the aggregation and bundling of contracts (or, reversely, of their division into lots).
In this connection, it should be stressed that the bundling of requirements into a single contract or the division of that same contractual object into several lots, as well as the rules imposing the minimum or maximum number of lots a single tenderer can bid for, allowing or excluding conditional or ‘package’ bidding and so on, can generate significant effects on competition for those contracts and in the market concerned. The bundling of independent requirements into a single contract (or the aggregation of otherwise independent contracts) by one or several public buyers may restrict the number of potential bidders and, therefore, generate anticompetitive effects, and alter the structure of the markets. Put otherwise, dividing contracts into several lots may in most instances increase competition, not only for the specific public contract but also for future contracts, and in more general terms, in the market from which the public buyer is procuring goods and services. The (sub)division of contracts into lots can particularly promote participation by SMEs —thereby broadening competition to the benefit of contracting authorities, as well as reducing the need to resort to more restrictive ‘secondary policies’ aimed at encouraging SME participation (such as set-asides or mandatory subcontracting). Therefore, in general terms, dividing contracts into lots or avoiding the aggregation of otherwise independent requirements into a single contract can have significant pro-competitive effects both on the tender and the market. Directive 2014/24 includes most of these economic insights in recital (78) and, consequently, article 46(1) aims at requiring contracting authorities to at least consider the possibility of dividing contracts into lots in order to achieve those benefits.
Nonetheless, it must be taken into account that the division of contracts into lots also presents some difficulties or undesirable effects and can generate some additional costs. Firstly, division of a given contract into lots may not be feasible in the light of the respective works, supplies and services. Therefore, rules regulating the division of contracts into lots should allow for sufficient flexibility so as not to impose artificially the fractioning of the contractual object where it is technically or economically unfeasible, or where it would substantially impair the effectiveness of the procurement process or raise the procurement costs disproportionately. This is encapsulated in Directive 2014/24, which clearly indicates that the reasons that could justify avoiding the division of a given contract into lots include the ‘risk [of] rendering the execution of the contract excessively technically difficult or expensive, or that the need to coordinate the different contractors for the lots could seriously risk undermining the proper execution of the contract’ (rec (78)). On the other hand, public procurement rules should restrict the ability of contracting authorities to bundle or aggregate artificially otherwise independent needs or requirements if doing so generates a competitive distortion—ie, if it excludes potential tenderers with a more limited capacity of supply, not integrated vertically, or otherwise not able to meet the bundled requirements, while they would be able to meet some of the requirements if unbundled or not aggregated. Therefore, it is to be praised that the 2014 public procurement rules encourage lot division unless it proves to be inadequate or disproportionate to the nature and amount of works, supplies and services concerned.
Secondly, economic theory has stressed that the division of the contract into lots might yield pro- or anti-competitive results depending on the relationship between the number of lots and the number of interested bidders. This is also reflected in Directive 2014/24, which allows deciding against allotting contracts where ‘the contracting authority finds that such division could risk restricting competition’ (rec (78)). Indeed, one of the potentially negative effects of the division of the contract into lots is the facilitation of collusion. Therefore, setting a number of lots that generates difficulties for coordination and allocation of lots amongst potentially colluding tenderers is desirable. In this regard, economic theory seems to provide two general criteria: the number of lots should be smaller than the expected number of participants (so that the impossibility of allocating lots to all interested tenderers diminishes the stability of collusion and forces it to spread over several tenders, thereby increasing the likelihood of detection), and the number of lots should exceed the number of incumbent contracts by at least one (implicitly reserving at least the additional lot for a new entrant or new contractor). Therefore, it also seems undesirable to adopt rigid rules setting a specific number of lots into which the contract should be automatically divided, since it could easily fall outside the desirable range for specific contracts and tendering procedures. In that regard, Directive 2014/24 rightly leaves it to the contracting authority to decide the number of lots to be created and whether to do so according to quantitative or qualitative criteria.
Finally, another important issue in the design of rules regarding lot division is to determine whether bidders can engage in multiple or ‘package’ bidding—and, if so, what are the minimum and maximum number of lots for which they can bid—and if conditional bidding is allowed, thus permitting bidders to offer varying conditions dependent upon the number and mix of lots awarded to them. In this regard, economic theory again stresses the importance of setting flexible rules that allow for a trade-off between fostering competition by smaller bidders and allowing larger bidders to exploit economies of scale, as well as for independent decisions to be made by tenderers—since multiple or package bidding will encourage bidders to submit more competitive offers for given packages than they would for independent lots or for all the lots. In this regard, it has been stressed that contracting authorities should not limit the number of lots tenderers can bid for in a way which would impair the conditions for fair competition, with maybe the only restriction of setting a relatively low maximum number of lots that a single bidder can be awarded at one time (which constitutes a specific case of awarding constraint). Therefore, it seems desirable to adopt rules so that the public buyer can reduce the minimum size of contracts/lots, and thereby maximise the number of smaller suppliers otherwise excluded, without hindering the ability of larger suppliers to bid for large sets of contracts in the event of their being characterised by positive complementarities.
To sum up, economic theory seems to support the finding that public procurement rules should be designed so as to encourage the division of contracts into lots whenever this is technically and economically feasible, and to allow the contracting authority to set the specific number of lots according to the circumstances of the tender. Similarly, contracting authorities should be able to restrict the maximum number of lots that a single tenderer can be awarded—if awarding the entire contract to a single contractor can generate a negative impact on competition; and particularly when ensuring that one or more lots are available for non-incumbent contractors is relevant to preventing distortions of competition in future contracts and/or in the market concerned. Finally, conditional and ‘package’ bidding should be allowed, in order to minimise the potential inefficiencies that lot division could generate. These insights of economic theory are now reflected to a large extent in article 46 of Directive 2014/24, and further considered in recital (79).
Firstly, article 46(2) ab initio allows for multiple bidding, indicating that the contract notice or in the invitation to confirm interest should indicate whether tenders may be submitted for one, for several or for all of the lots in which a given contract is divided. Secondly, article 46(2)II of Directive 2014/24 allows for restrictions on the number of lots that can be awarded to the same tenderer, establishing that even where tenders may be submitted for several or all lots, contracting authorities may limit the number of lots that may be awarded to one tenderer. In that case, the Directive sets up a double requirement of transparency to avoid distortions in the award decision-making, establishing that such awarding constraint will be applicable provided that the maximum number of lots per tenderer is stated in the contract notice or in the invitation to confirm interest, and as long as the procurement documents disclose the objective and non-discriminatory criteria or rules the contracting authority will apply for determining which lots will be awarded where the application of the award criteria would result in one tenderer being awarded more lots than the maximum number. All these rules remain, however, discretionary. Hence, it is still necessary to rely on the principle of competition and its requirements in order to inform their implementation.
The general criterion, in our view, should then be that in the exercise of this discretion as regards the division or aggregation of requirements, the fixing of the number of lots tendered, and the rules for conditional and ‘package’ bidding, contracting authorities must ensure that competition in the market is not distorted and, where possible and feasible, promote competition for the contract—particularly by avoiding the configuration of contracts which result in potentially interested competitors being excluded. As a default rule, division into a large number of lots will be preferable to a division into an insufficient number of exceedingly big lots, since tenderers could compensate for such an ‘excessive fragmentation’ of the object of the contract by submitting bundled offers—while an insufficient division of the object of the contract cannot be compensated by tenderers submitting partial offers or offers for amounts smaller than the object of the tender (as those bids would be considered non-compliant and, hence, rejected).
Arguably, in order to be effective, the rules and decisions on lot division will need to be complemented with clear award criteria as regards the comparability of offers for a different number of lots, as well as with rules applicable in case the offers submitted do not cover all the lots tendered. In this case, asking bidders to submit offers for the entire contract, for each individual lot and for the packages of lots that they would like to be awarded (with different prices and conditions) would arguably eliminate all the benefits of lot division, since tenderers that could not bid for the entire contract (even under less favourable conditions than they could offer for a given lot or group of lots) would be excluded anyway. Therefore, a preferable rule seems to be to allow the submission of bids for independent or grouped lots, without mandatory requirements regarding the entire contractual object. In case one or various lots could not be covered in the initial tendering, the contracting authority could then engage in re-tendering the pending lots by following a subsequent negotiated procedure with all the participating tenderers, or a new procedure, depending on the circumstances (on the rules and criteria regarding procedure selection). Under exceptional circumstances, the option should also be available to the contracting authority not to award any of the lots for which it has received offers if it is clear that this would jeopardise the effectiveness of the follow-up tenders for the remaining lots—which should then be re-tendered in a single contract. However, if the design of the lots was properly conducted in the first place—ie, if lots had been designed according to sensible functional and economic criteria, and an effort had been made to ensure their balance—this situation should be largely marginal. Along these lines, but covering the separate option of whether contracting authorities can ‘cherry-pick’ or ‘mix-and-match’ offers for different lots, article 46(3) of Directive 2014/24 clearly indicates that this is possible, provided certain conditions are met. Indeed, where more than one lot may be awarded to the same tenderer, contracting authorities may award contracts combining several or all lots where they have specified in the contract notice or in the invitation to confirm interest that they reserve the possibility of doing so and indicate the lots or groups of lots that may be combined (on criteria applicable to such conditional award rules).
As a preliminary conclusion, it is submitted that despite the discretionary terms of article 46 of Directive 2014/24 and on the basis of the final goal of maximising competition, contracting authorities should resort to division of contracts into lots whenever it is not unfeasible technically or economically, and should set rules that ensure that, while still giving tenderers the largest possible flexibility to submit package and conditional bids, competition is not distorted by undue contract division or aggregation. Rules on contract division should be complemented and reinforced by consistent award criteria and rules on the re-tendering of unawarded lots.