Practical Law

One might expect there to be little more difficult than decommissioning 12 nuclear facilities, not least given the rather worrying fact that in some cases, “nobody knows how much radioactive waste is present in temporary storage vaults, how radioactive it is, nor how many highly radioactive springs are stored there” and further that “there are few reliable design drawings available of the facilities themselves.”

It would, however, seem from the recent judgment in Energysolutions EU Limited v Nuclear Decommissioning Authority [2016] EWHC 1988 (TCC), that procuring the decommissioner might be just as, if not more, difficult. At 324 pages and 948 paragraphs (not including 5 appendices, some confidential) it is a daunting tome, but it still demands to be read by anyone with an interest in procurement law for its elucidation of a wide variety of procurement and civil litigation issues, from the everyday to the genuinely exceptional.

For those without the time to grapple with Mr Justice Fraser’s comprehensive analysis, I have distilled ten key lessons from his judgment.  

1. Do not “fudge” the scoring process to avoid exclusion of a bidder

This may just be the most important lesson.

As is common in procurements, and particularly unsurprising with nuclear decommissioning, the Nuclear Decommissioning Authority (the “NDA”) established certain mandatory minimum requirements for tenderers. Energysolutions (“ES”) argued that the winning bidder (“CFP”) had failed to meet those requirements and should therefore have been excluded. The NDA argued that was incorrect or, even if it was correct, it had some discretion to waive those requirements.

Fraser J found for ES in respect of at least two mandatory requirements where the scoring of CFP had been “fudged” to avoid automatic exclusion. He explicitly rejected the suggestion that the NDA could somehow waive mandatory requirements after they had been established, not least as the NDA had chosen to establish them as mandatory in the first place: “the rules cannot be changed after the competition has started and the tenders submitted”. Even if this meant that all of the bidders had to be rejected:

“This would have been embarrassing for all concerned, but given this was a £4 billion contract for 14 years, re-running the competition if all the bidders failed would not necessarily have been a commercial disaster.  The alternative would be conducting an unlawful evaluation leading to a potential challenge, claims for damages and even greater embarrassment and financial loss depending on the outcome.”

It would seem that it is ultimately this “alternative” that has come to pass (subject, of course, to appeal).

2. Evaluative judgments in scoring are not immune from review

As to scoring more generally, the NDA had argued that evaluative judgments were immune from review by the Courts. This was roundly rejected by Fraser J: it was precisely the role of the Court to assess whether there had been a “manifest error” in such evaluative judgments, although the public authority would be allowed a “margin of appreciation”.

3. Even if the score is right, you might still be wrong

The NDA also argued that consideration of “manifest error” should be limited to the question of whether the score ultimately awarded was correct. This was again rejected by Fraser J. Instead, he found that in determining whether there had been manifest error it was necessary to consider the application of the tender criteria, the scoring process and the score. If there was a manifest error in any of these elements, only then would the Court consider what the score should actually have been.

4. Minimising written records is not consistent with the duty of transparency (and do not even think about shredding)

With one eye on potential litigation, the NDA understandably sought to train its scorers on the practicalities of the scoring process. Unfortunately, the initial iteration of that training suggested the shredding of all scoring records other than the final consensus reasoning and score (such shredding would have been “wholly unacceptable”). While shredding was abandoned, it was replaced by a prohibition on the making of notes other than the final agreed reasoning and score. Fraser J found that there was little difference in the two approaches and neither was consistent with the duty of transparency. Moreover, the policy only harmed the NDA’s defence of the litigation as there were extremely limited contemporaneous records of why particular scores had been awarded.

It is also worth noting that proper record-keeping has become an explicit requirement under Regulation 84(8) of the Public Contract Regulations 2015.

5. Don’t change agreed scores without recording the reasoning why

To make matters worse, even after a “final” score had been agreed by the scorers, the NDA subsequently changed some of the scores following unrecorded internal discussions or, potentially, advice from lawyers. As to the latter, see below, but Fraser J was as critical of the idea of unrecorded discussions and unrecorded reasons for changing scores as he had been of the record-keeping policy for scorers. Ultimately, the absence of records only made it more likely a manifest error would be found by the Court.

6. Beware involving lawyers in a scoring process

Again, concerns arose in respect of involvement of lawyers in the procurement process. The NDA had invited external lawyers (who subsequently became their lawyers in the litigation) to review the scoring process. As a result of that review, some scores had been changed.  However, the NDA claimed legal professional privilege (“LPP”) over all communications with their lawyers. Fraser J was not critical of the NDA for claiming LPP, which was their right, but did observe that it created yet further gaps in the scoring record, which only made it more difficult for the NDA to explain particular scores and therefore made a manifest error more likely.

If authorities want to involve lawyers in a scoring process, it seems likely they are going to have to choose whether to waive LPP or take the risk that defending a claim will be rendered more difficult by the gaps in the record.

7. Get your reasoning right from the beginning

Perhaps unsurprisingly given the above, the NDA sought to argue that, when considering lawfulness, the Court should review any and all reasoning available to it, including reasoning developed / presented after the litigation had commenced. Fraser J rejected this argument, at least for the purposes of determining whether the process had been lawful, when the Court should limit itself to considering reasoning at the time of scoring and presented to the losing bidders before the litigation commenced.

Fraser J did, however, accept that such ex post facto reasoning would be relevant to the subsequent question of causation, and that a finding of unlawfulness would not therefore necessarily entitle a challenger to damages. Still, it is clearly preferable for public authorities to ensure that all relevant reasoning has been provided in advance of litigation commencing. As observed by the Court, this “should not be seen as requiring anything out of the ordinary for a sensibly organised procurement exercise that is conducted transparently.”

8. Witness training is not always a good thing

The Court had much to say about the style and credibility of the witnesses in the litigation, much of which would prove useful reading for witnesses in procurement cases. One issue common to most of ES’s witnesses was a particular style of answering questions, that the Court concluded was likely to have arisen from training. This essentially involved the witness avoiding answering the question by explaining its “context” in a discursive manner more akin to a “corporate presentation” than an answer. The Court considered this “increasingly unhelpful” and an approach “to be discouraged”.

9. Do not agree to pay witnesses win bonuses (but if you do, ensure you disclose those agreements)

Extraordinarily, immediately before the handing down of the draft judgment, it was disclosed that certain of ES’s witnesses had agreements with the claimant that they would be paid a “bonus” if the claim was successful (in one case potentially in excess of £100,000). Those witnesses were re-called and cross examined alongside the internal and external lawyers for ES.

Fraser J reiterated in strong terms that such agreements were contrary to public policy and also in contravention of the SRA Code of Conduct (he also criticised ES for sexism in its approach to the amount of win bonuses offered). However, he refused the Defendant’s application to strike out the case on the basis that the existence of the agreements went only to the weight that should be given to the evidence. In this respect, he effectively exonerated the witnesses themselves for any wrongdoing, since they could not have been expected to understand that such agreements were inappropriate, and that they had been sought for genuine, understandable reasons resulting from the time spent on the litigation in circumstances where a number of them had been made redundant due to the loss of the tender. He further found that their credibility was generally unaffected and, in any event, he considered that their evidence did not affect the outcome of the proceedings.

Fraser J also exonerated the external lawyers who had been unaware of the agreements until immediately before they were disclosed.  Instead, he saved his criticism for the Claimant’s internal lawyer who drafted the agreements in knowledge they were contrary to the SRA rules and who subsequently decided they did not need to be disclosed under the ongoing duty of disclosure.

10.  Defensiveness is not necessarily the best form of defence in litigation

Perhaps the overall lesson from this litigation is that adopting a defensive position in response to procurement litigation is not always the best policy. Clearly, careful design and appropriate training in preparing for a procurement is crucial, but the NDA’s attempts to minimise records to make future litigation more difficult only ended up harming their defence. Moreover, the judge was sharply critical of many of the NDA’s witnesses, who, in his view, adopted a stance of “obstinate refusal to accept that any mistakes or errors had been made at all by the NDA” irrespective of the correctness of the proposition put to them.  This resulted in “logic becom[ing] an early casualty during the NDA evidence” and the “tortured misuse of the English language.”

Ultimately, at all stages it is simply a question of keeping it simple, transparent, consistent and fair. That shouldn’t be so hard, should it?