The recent 336-page High Court judgment in Energy Solutions EU Ltd v The Nuclear Decommissioning Agency (the NDA) highlights the importance of robust evaluation and scoring processes when conducting regulated OJEU tender processes, and specifically the risks associated with:

  • “manipulating” the evaluation process to avoid the disqualification of certain bidders;
  • subject matter experts evaluating and scoring on an inconsistent basis; and
  • not adequately recording scoring and assessment decisions.

This case concerned the competition to become the “Parent Body Organisation” for 12 sites operating “Magnox” nuclear power stations, together with two other sites. US-based Energy Solutions (ES) had spent approximately £10 million preparing its tender and expected to receive approximately £100 million in fees for its role in managing the delivery of the decommissioning work. Ruling on ES’s challenge to the process, Mr Justice Fraser held that the bid submitted by the consortium to which ES was a party would have won if the bids had been evaluated properly.

The question of quantum of damages is still to be decided, but ES is said to be seeking around £100 million in damages. ES’s right to recover damages is being contested separately by the NDA as a preliminary issue on which the NDA has been granted permission to appeal to the Supreme Court.

Mr Justice Fraser’s judgment is attached here.

Background

Following a lengthy competitive dialogue process, a consortium consisting of Bechtel Management Company Limited and Energy Solutions EU Limited was told it had been unsuccessful. It was informed that it had achieved a score of 85.42%, compared to the winning tenderer’s score of 86.48%. Having analysed the responses received from the NDA, ES concluded that the NDA had committed various breaches of the Public Contracts Regulations and raised three claims (later combined as one) seeking damages of £100 million.

Key issues

Judgment rested on consideration of two key features of the evaluation process:

  • The approach to evaluating the bids of each consortium – each separate part of the bid was called a “Node” and contained different matters, which were called “Requirements”. Each Requirement was given a particular score by the team of evaluators consisting of three subject matter experts, who were under the overall management of a Core Competition Team (theCCT); and
  • The clarification process in respect of some of the “Nodes” – if evaluators considered that a part of a bid was unclear, clarification could be sought via bidder clarification requests, which were drafted by the evaluators or the CCT and issued by the CCT to the bidders. The responses would be scrutinised by a member of the CCT and, if the CCT felt the response went further than clarification, that information would be redacted and the clarification then provided back to the evaluators – this fact was disputed by ES. The final scores given to the different Requirements and Nodes by the evaluators were then translated into a percentage result for the overall bid by applying the relevant weighting to the different scores. Different Nodes had different weightings towards the overall percentage result.

Findings on the evaluation process

Mr Justice Fraser’s judgment sets out two different headings of “The evaluation process in theory” and “The evaluation process in practice”, identifying a series of issues including: 

  • informal and unrecorded conversations forming part of the actual evaluation process;
  • inherent risks in the use of “cut and paste” techniques for evaluation;
  • the AWARD electronic system used to process the tender submissions being amended due to “accidental error”;
  • one of the evaluators, who was performing evaluations on more individual Nodes than any other single evaluator, stating during evidence that, at a personal level, he did not feel he needed to be consistent as “that wasn’t in the scoring criteria”;
  • the NDA personnel kept no records of the dialogue which lasted several months (so only memory could be used to ensure consistency) – Mr Justice Fraser adding at para. 188 that logistical difficulties in so doing “would not have been too difficult” to overcome; and
  • notes used for evaluation by evaluators, other than those stored in AWARD, were to be destroyed (there is some discussion in the case about this but according to evidence the evaluators were not allowed to write anything down at all and the policy of keeping notes stored in AWARD was so that they could be “deleted from the system if no longer required/answered during final review stage”). 
    Mr Justice Fraser noted at para. 211 that “[i]n circumstances where there is an express obligation of transparency on the NDA, this approach to note and record keeping, and sensitivity about retaining written material, simply does not seem to me to be justified. That is putting the point at its most favourable for the NDA” and further noted at para. 269 that “[t]he motivation… seems to have been as much to make the NDA immune from challenge, as it was to ensure that there would be no ground for challenge in the first place”.

Judgment

In a wide-ranging judgment covering a larger number of legal issues Mr Justice Fraser’s principal conclusions were that:

  • the evaluators had purported to apply good industry practice during their evaluation process when they manifestly had not;
  • there was “demonstrably different and unequal treatment of the two bidders on the same Requirement being judged against exactly the same scoring criteria” for which there was no rational explanation (para. 466);
  • there were no stated requirements for modelling to have been carried out by any of the bidders, and such lack of modelling did not appear on the scoring criteria, and therefore the evaluators requiring modelling (and marking accordingly) was manifestly erroneous; and
  • various scores awarded were not capable of objective support, even taking into account the margin of appreciation available to the evaluators.

Mr Justice Fraser did, however, accept at para. 747 that the NDA’s approach of having the CCT redact responses going further than clarification prior to providing these to the evaluators was lawful as “[t]he CCT were vigilant in considering the contents of all the clarification requests that came in from bidders, before their contents were uploaded into AWARD, to ensure that only genuine clarification was provided to the evaluators”.

In conclusion, at para. 944, Mr. Justice Fraser found that there was “no escaping the fact that applying the correct legal test to the evaluation exercise leads to the scores having to be reconsidered, with some of them changing”.

The winning bidder should “have been disqualified from the competition, by the application of the very rules contained within the SORR that the NDA itself drew up that governed the competition” and most damningly of all that “the NDA sought to avoid the consequences of disqualification by “fudging” the evaluation of those Requirements… By the word “fudging”, I mean choosing an outcome, and manipulating the evaluation to reach that outcome”, per para. 945 of the judgment.

The consequence of those findings was that, even if the winning tenderer had not been disqualified, ES’s consortium would have bid the most economically advantageous tender.

Comment

The High Court’s decision does not make easy reading for contracting authorities and strongly emphasises the need to ensure:

  • clear and objective marking criteria which are followed consistently;
  • the process is conducted in line with the principles of transparency and equal treatment, with all bidders marked against the same objective criteria;
  • where a threshold condition exists, this must be enforced rather than worked around;
  • records of the evaluation process (including dialogue sessions) are taken and retained, and not deleted or destroyed; and
  • no actions are taken to shield the procurement process from legal challenge.