The Good Law Project (R on the application of) v The Secretary of State for Health and Social Care

This is the judgment of the court to which we have all contributed.

The Minister for the Cabinet Office appeals against the decision of O’Farrell J to allow (in part) a claim for judicial review of his decision to award a contract to Public First Limited without public notice or competition, relying on Regulation 32(2)(c) of the Public Contracts Regulations 2015.

The decision was made in late February/early March 2020 as the Covid-19 pandemic took its grip on the nation. The Director General of the World Health Organisation formally classified the disease as a pandemic on 11 March 2020 and on 23 March 2020 the Prime Minister announced the first “lockdown” with the slogan:

The contractual services were the provision of focus group and communications support services designed to inform government public messaging and communications strategy for the pandemic crisis. Among other things, Public First was responsible for testing the slogan launched at the beginning of the first national lockdown. There is no suggestion that Public First provided its services with anything other than expert skill and care; on the contrary Public First has been described as having been “very insightful”, having done “an excellent job”, “a brilliant job” and having provided “work of a very high standard” that met deadlines. Its services helped to inform government decisions of “vital life-saving importance”. Against a maximum value of £840,000 payable under the contract, a total of £564,393.67 was paid.

The Good Law Project Limited (“Good Law”), a not-for-profit organisation with a stated aim of using the law to protect the interests of the public, commenced proceedings for judicial review on 10 July 2020 on the following grounds:

There was no basis for making a direct award under Regulation 32(2)(c), as the direct award of the contract to Public First was not strictly necessary;

The award of the contract for a period of six months was disproportionate. Even if Regulation 32 was applicable, the contract should have been restricted to the Minister’s immediate, short-term needs, pending a competitive process to procure a longer-term supply of the services;

The decision to award the contract to Public First gave rise to apparent bias contrary to principles of public law. The fair minded and informed observer would conclude that there was a real possibility of bias having regard to the personal connections between the decision-makers and the directors of Public First.

No challenge or complaint was ever raised to the award of the contract by any potential competitor of Public First. The judge held that Good Law had sufficient standing to bring proceedings for the purpose of section 31(3) of the Senior Courts Act 1981 and rely upon the Regulations, as might a commercial entity which considered that it had been deprived unlawfully of the opportunity to bid for the contract. She also concluded that Good Law had standing to mount the public law challenge based on apparent bias despite having no interest in the letting of the contract. The Minister has not appealed that part of the judge’s decision. It was based, so far as concerns the Regulations, on the obiter dicta of this court in R (Chandler) v. Secretary of State for Children, Schools and Families [2009] EWCA Civ 1011 at [77] and [78]. They were summarised in R (The Good Law Project Limited and others) v. Secretary of State for Health and Social Care [2021] EWHC 346 (Admin) by Chamberlain J at [99]. The arguments on standing below did not distinguish between the claim based on the Regulations and the public law challenge based on apparent bias. The question of standing for complete strangers to the procurement process with no commercial interest both under the Regulations and on public law grounds is a question ripe for review when it next arises.

The judge dismissed Good Law’s challenge under Regulation 32 and on the basis that the six-month period of the contract was disproportionate. She allowed the claim that the decision to award the contract to Public First gave rise to apparent bias contrary to common public law principles. This finding was based on a combination of the personal association between Dominic Cummings, the then Chief Adviser to the Prime Minister, and the directors and owners of Public First (which she found did not in itself to give rise to any apparent bias) and the Minister’s failure to consider any other research agency or to keep a record evidencing that objective criteria were used to select Public First over other research agencies.

This is an unprecedented outcome: a party with no potential interest in a contract has not hitherto obtained a declaration of unlawfulness on the basis of apparent bias in respect of a decision by a public body to grant a private law contract.

The Minister appeals against the finding of apparent bias. Good Law cross-appeals against the dismissal of its claim under Regulation 32 and that the six-month duration of the contract was disproportionate.

Good Law submits that the Appellant’s Notice was filed one day late. That is because, although it was filed electronically on the last day allowed by CPR 52.12(2)(b) (30 June 2021) it was not sent by email until 23.47 which was too late. Good Law says that, in accordance with CPR 5.5 and Practice Direction 5B a document sent by email after 16.00 is deemed to have been received the following day: PD5B, paragraph 4.2. The Minister says that the notice was not sent in accordance with Practice Direction 5B but, instead, pursuant to the “Court of Appeal Civil Division urgent business priorities coronavirus update” of 17 May 2021 and in force at the time. This required all documents to be filed with the court electronically and made no mention of any deeming provision. The notice was therefore filed on the last available day, 30 June.

CPR 5.5 and Practice Direction 5B govern the sending and filing of documents by email. Although, pre-pandemic, only some documents could be sent or filed by email, the update made clear that those restrictions had been removed because of the pandemic. Instead, all documents, including the Appellant’s Notice, could be filed by email. But nothing in the update abrogated or modified the rule as to deemed receipt after 16.00 set out in PD5B. Furthermore, the general application of that rule is reiterated by CPR 2.8(5) and the note in the 2021 White Book at para 2.8.5. The Appellant’s Notice was thus filed one day late.

That said, we grant relief from sanctions in accordance with the principles set out in Denton v. TH White Ltd [2014] EWCA Civ 906; [2014] 1 WLR 3926. The deemed delay of a day was not serious or significant; the Minister’s late change of mind about appealing provides at least some explanation for the delay; and a consideration of all the circumstances of the case, and in particular the important issues that have been raised by the appeal, makes it just to grant the necessary extension of time.

The factual background is set out in detail at [7] to [61] of the judgment below: [2021] EWHC 1569 (TCC). We shall provide only a summary.

Public First is a consultancy firm specialising in opinion research on complex issues of public policy, as well as policy analysis and communications. It was established by Rachel Wolf and James Frayne, its directors. In 2011 Mr Frayne was appointed to work as Director of Communications for the Department of Education, where he worked alongside Mr Cummings, who was then a special adviser to the Rt Hon. Michael Gove MP. Ms Wolf formerly worked as an adviser to Mr Gove and by 2020 had worked previously for Mr Cummings as well. Through these work connections, Mr Cummings had become a personal friend of Ms Wolf and Mr Frayne, although he had not met Mr Frayne since 2016.

In early February 2020, Public First was engaged by the Minister for the discrete task of providing focus group services to test public opinion and policies in preparation for the Prime Minister’s speech planned for later that month, ahead of the budget. Neither Mr Cummings nor Mr Gove had any involvement in this appointment.

By late February 2020 the growing crisis arising out of the Covid-19 pandemic had become apparent; the public health crisis had developed quickly. The Minister needed an accurate understanding, amongst other things, of what the public knew or understood about Covid-19 and safety information, what was the most vital information and how best to convey it, how the public was behaving, and to check the effect of visual displays. Urgent external help was required to communicate essential health messages to the public.

On 27 February 2020 Mr Alex Aiken, Executive Director for Government Communication, raised with his team the need for urgent focus group testing of Covid-19 issues, the results of which could be provided to No. 10 at 15.00 the following day. Ms Helen Stratton, Head of Insight, Government Communications Service, Prime Minister’s Office and Cabinet Office, knew that Public First was already scheduled to conduct focus groups in Crawley that evening. She recommended that Public First should re-direct those focus groups to conduct the Covid-19 research. That is what happened.

On 28 February 2020, Public First provided the first information gathered from the focus groups. In a meeting held on the same day, a decision was made to continue using focus groups to gather information on public opinion and behaviour. Following a strong recommendation from Mr Cummings, Mr Aiken decided that Public First should conduct those focus groups, alongside Britain Thinks, another public policy research agency. Ms Stratton subsequently met Public First to devise a programme for the focus group work to be carried out over the weekend and the following week. Mr Aiken approved the expenditure to allow initial recruitment to start the following morning. The initial focus group work was carried out at the beginning of March 2020.

Ms Catherine Hunt, Head of Insight and Evaluation, Prime Minister’s Office and Cabinet Office communication team, raised internal requests for a purchase order to cover the initial costs, noting that no contract was yet in place for the earlier work, or the Covid-19 focus group work. Mr Aiken authorised the use of further focus group services and the associated additional expenditure.

On 16 March 2020, at the request of Mr Cummings, Mr Milland of Public First was seconded to the Covid-19 communications hub to provide additional support in respect of the Covid-19 focus group work. By way of email dated 21 March 2020, Mr Aiken requested the Crown Commercial Service (“CCS”) team to put in place a formal agreement for Mr Milland’s secondment and to agree monthly remuneration for Public First.

The working conditions and pressures arising out of the pandemic crisis on those involved during this period were described in the unchallenged evidence served on behalf of the Minister, as demonstrated in the passages set out in Appendix A to this Judgment.

The scope and value of the contract were discussed between March and April 2020. CCS prepared a strategy and award report, dated 26 May 2020, which recommended the direct award of the contract to Public First. The contract was contained in a letter from the Minister dated 5 June 2020 and an email from Public First dated 8 June 2020, with an effective date of 3 March 2020 and an expiry date of 2 September 2020. On 12 June 2020 notice of the contract award was published on the Government “Contract Finder” website.

The services under the contract included:

Recruitment and delivery of focus groups and/or mini groups to an agreed specification, covering the general public and key sub-groups defined by demographic, life stage or other agreed criteria;

Same-day top line reporting and next-day fuller reporting of focus group findings; and

On-site resource to support Number 10 Communications.

Public First provided these services throughout March, April, May and June 2020. In July 2020, the services under the contract were extended to cover qualitative research into EU exit topics and themes, re-building the economy following the Covid-19 crisis and attitudes to the United Kingdom union.

The Minister’s appeal is advanced on three grounds. The first challenges the judge’s conclusion that, in circumstances where the relationship between Mr Cummings and Public First did not give rise to apparent bias, the failure to keep a clear record of the objective criteria used to select Public First over other research agencies ([147]), and the failure to undertake some form of comparative procurement exercise ([158] to [164]) did amount to apparent bias.

The second (not advanced before the judge) contends that the concept of apparent bias at common law was irrelevant because the provisions of Regulation 24 concerning conflicts of interest provides a complete code under the Regulations for these purposes. However, on behalf of the Minister, Sir James Eadie QC concedes that, at least for the purposes of this appeal, it does not matter whether apparent bias is considered at common law or under the Regulations. The second ground of appeal has therefore largely fallen away.

The third ground of appeal is that if, as the judge found, this was a situation of extreme urgency under Regulation 32 which dispensed with the need for an open procurement exercise, the rules relating to bias (whether under Regulation 24 or the common law) had no further application to this case.

Good Law’s cross-appeal goes to the judge’s conclusion that this was a situation of extreme urgency, and therefore covered by Regulation 32(2)(c). For three reasons, concerned with alternative suppliers, duration and scope, it was Good Law’s submission on appeal that, whether or not this was a situation of extreme urgency it was not ‘strictly necessary’ to engage Public First under this contract.

We shall consider the issues that arise under Regulation 32 first.

The Regulations are enormously detailed encompassing 122 regulations over 116 pages (together with schedules). They introduced into domestic law the requirements of European Union law. They are designed to provide a complete code governing the procurement of public contracts by public bodies, referred to as “contracting authorities”. Those bidding to carry out these public contracts are referred to as “economic operators”.

Part 1 of the Regulations is general. The opening regulations of Part 2 concern scope. The General Rules relating to procurement begin at Regulation 18 and continue to Regulation 84. Most of those, from Regulation 25 onwards, are concerned with what are called “Procedures”.

Part 3 of the Regulations is concerned with Remedies. Regulation 89 imposes a duty on a contracting authority to comply with the provisions of Parts 2 and 3 of the Regulations. A breach of that duty is actionable “by any economic operator which, in consequence, suffers, or risks suffering, loss or damage”: see Regulation 91. The Regulations do not expressly envisage any third party (i.e. someone other than an economic operator) making any claim for breach of the Regulations.

Regulation 26 is headed ‘Choice of Procedures’. Regulation 26(1) mandates contracting authorities to apply the procedures set out in Part 2 of the Regulations, and Regulation 26(2) provides that “such contracts may be awarded only if a call for competition had been published in accordance with this Part except where Regulation 32 permits contracting authorities to apply and negotiate a procedure without prior publication” (emphasis added). Thus, the only way in which a contracting authority can avoid the call for competition for a contract of sufficient value to fall within scope (and all that goes with it) is by showing that, on the facts, Regulation 32 applies.

The relevant part of Regulation 32 provides as follows:

In summary therefore, a negotiated procedure is permitted by Regulation 32(2)(c) where, for reasons of extreme urgency, brought about by events which are unforeseeable by and not attributable to the contracting authority, the ordinary time limits set out in the Regulations cannot be complied with. The use of the negotiated procedure must, however, be ‘strictly necessary’ in any given case.

Only one domestic case dealing with this Regulation was drawn to our attention. That was Salt International Ltd v. Scottish Minsters [2015] CS1H 85; [2016] S.L.T. 82, which concerned a predecessor provision of Regulation 32 with almost identical wording. The Court of Session (Inner House) confirmed the finding of the commercial judge that, in a contract relating to “Winter 1”, the public authority had been justified in negotiating and securing a supply of de-icing salt from a company called INEOS. The situation was one of extreme urgency because of unforeseeable events and was strictly necessary. However, in relation to a contract relating to “Winter 2”, the authority could not rely on the predecessor provisions of Regulation 32 because they had failed to heed the warnings of the previous winter. On that basis, the need for salt in Winter 2 was not unforeseeable and was in any event attributable to the contracting authority.

We note two aspects of that decision. First, the negotiated procedure which was found to be strictly necessary in relation to Winter 1 was a negotiation with just one company, INEOS. There was no competition involving other economic operators. Secondly, the pursuers argued that the judge had been wrong to find that the salt would not have arrived on time unless the negotiated procedure was used. They said that the shorter time limits permitted by an “accelerated procedure” (a defined process now set out in Regulations 27 to 29) could have been met. The court rejected that argument, pointing out at [46] that any form of competition involved much more than just a consideration of the periods themselves:

The judge considered, first, whether Regulation 32 applied at all (i.e. was this a situation of “extreme urgency”?). She concluded that it was. That conclusion is not the subject of the cross-appeal.

Secondly, she determined whether it was “strictly necessary” for the Minister to negotiate with and award the contract to Public First. The judge rejected Good Law’s arguments that: (i) there was time to run an accelerated procurement; (ii) there was no need to contract with Public First because there were existing suppliers; (iii) the contract was unnecessarily long in duration; and (iv) the scope of the contract was unnecessarily wide (see [101] to [123]). The cross-appeal seeks to challenge the judge’s conclusions on (ii) to (iv) but not the finding that an accelerated procedure could not have been run.

The judge’s distilled thoughts on Regulation 32 come together at [124]:

The judge’s answers to the questions arising under Regulation 32(2)(c) had a significant effect on the remainder of Good Law’s challenge.

They meant that there was no need for the Minister to engage in the call for competition, which lies at the heart of the Regulations. It is unnecessary to set out all the Regulations which depend, directly or indirectly, on the procedures and processes triggered by the call for competition. In a situation of extreme urgency, however, none of those Regulations would apply, because the Minister was entitled to award a contract by a negotiated procedure without prior publication instead. Furthermore, they also meant that, if a negotiation with just one supplier could be shown to be strictly necessary, there was no requirement for any sort of comparative tender exercise at all. That is what had happened in Salt International.

As an example of the effect of the judge’s conclusions about Regulation 32, it is instructive to consider Regulation 67, to which Mr Coppel QC, for Good Law, drew our attention. This sets out complex provisions relating to what the tender documents need to contain by way of contract criteria and how the competing tenders are to be ‘marked’ by the contracting authority. Many procurement challenges under the Regulations relate either to the contract criteria themselves or to the contracting authority’s failure fairly to adjudge the competing tenders by reference to their own contract criteria. Mr Coppel suggested that this was a Regulation which would still apply in full even if it was a situation of extreme urgency.

We disagree. Its complex provisions might be thought to be the antithesis of urgency. Regulation 67 presupposes that there was a competitive tender process which the judge concluded did not need to happen. The extent (if at all) to which Regulation 67, or any of the other Regulations in Part 2, will apply in a situation of extreme urgency will always depend on the facts. But if a negotiated procedure without prior publication with a single economic operator was justified under Regulation 32, because in all the circumstances that was what was strictly necessary, many of them will not apply at all.

In this way, a conclusion such as that reached by the Judge on Regulation 32 leads to a significant departure from the ordinary position under the Regulations.

Mr Coppel also developed arguments by reference to Regulations 18 and 24. Regulation 18 provides:

Good Law has no pleaded claim, even indirectly, by reference to Regulation 18 because, although it sought to make a late amendment raising Regulation 18, that application was refused by the judge. However, Mr Coppel maintained that, even in a situation governed by Regulation 32(2)(c), Regulation 18 would still apply and that this would be part of the background against which the challenge on the grounds of apparent bias fell to be considered. This argument appears to be artificial, at least in the context of circumstances where only one service provider has been considered for a job.

Regulation 24 provides:

Regulation 24 was not pleaded by Good Law, and only arose tangentially before the judge, although that was because both parties were content to argue apparent bias by reference to the common law instead. Its precise scope in the context of procurement exercises is not before us. It is concerned with conflicts of interest which are not exhaustively defined. It requires “appropriate measures” to “effectively prevent, identify and remedy” such conflicts with a view to ensuring that competition is not distorted and that there is equal treatment of economic operators. We can readily accept that if the strict criteria of Regulation 32 are met it is not likely that Regulation 24 (whatever its scope) will have been breached independently. But we do not accept that there are no circumstances in which it might have some application. Sir James did not press his third ground, which in any event was not argued below and so we say no more about it. In common with the approach of the parties we will deal with these issues through consideration of the common law apparent bias challenge.

Once extreme urgency has been established for the purposes of Regulation 32, what dictates whether a particular negotiated procedure without prior publication was permitted is whether it can be shown to have been “strictly necessary”. Mr Coppel advanced three reasons why, contrary to the judge’s conclusions, the contract with Public First was not strictly necessary in accordance with Regulation 32(2)(c). We deal with each in turn.

Mr Coppel argues that it was not strictly necessary to make a direct award to Public First because the Minister already had existing contracts with suppliers which it could have used to commission this work. The judge rejected that submission on the facts at [108] to [113] explaining convincingly her reasoning. Mr Coppel submits that the judge erred in principle because, as a matter of law, it could not have been strictly necessary for the Minister to negotiate a contract with Public First if other suppliers, who were already the subject of lawfully procured contracts, were able to do the work instead. We reject that submission, just as the judge did. The question whether the negotiated procedure is strictly necessary for the purposes of Regulation 32 is one of evaluation based upon the complete range of factual circumstances accepted by the judge to have been in play. It is not subject to artificial constraints of the sort contended for by Good Law. It would be wrong in principle to find that a contracting authority in a situation of extreme urgency could only contract with existing suppliers irrespective of their judgement about who was the most appropriate supplier of the services urgently needed.

Mr Coppel submits that it was not strictly necessary to award Public First a six-month contract from March 2020, and to continue commissioning services from them in April, May, June and July 2020. The Judge rejected that submission at [114] to [117] of her judgment.

The first difficulty with Mr Coppel’s submission is that it seems to us to be inescapably dependent on the benefit of hindsight, as the Judge found at [116]. The underlying premise is that six months was obviously too long a duration for the contract. But at the height of the original Covid-19 crisis, it was impossible for anyone to say that such a period was too long or more than was strictly necessary.

Furthermore, it could not be said that the duration of the Public First contract was a matter of indifference to the Minister. The original proposal had been for nine months, and that had been reduced to six. The question of duration had therefore been actively considered. Mr Coppel argues that, when applied to contract length, the expression “strictly necessary” meant “until such time as a properly procured contract could be put in place”. But there was nothing to say that six months was not a reasonable estimate of how long it might be before such a contract was placed. The letting of a contract like this, even under the accelerated procedure, is to be measured in months, as the evidence suggested. And it is not just the prescribed time periods in the Regulations that matter: the passage in the judgment in Salt International, set out above, makes clear that much else has to be reflected in an estimate of how long such a process may take.

Furthermore, we note that the contract was a pay-as-you-go contract, so that Public First were only paid for the services that they were actually asked to provide, on a case-by-case basis. There were also various termination provisions which in certain circumstances allowed the contract to be terminated sooner than six months. As a result of these features, it seems to us that the overall duration of the contract was of secondary importance.

In the round, therefore, we are unpersuaded that the judge was wrong to reject that, in some way, the six-month contract was longer than was “strictly necessary” in all the circumstances.

Mr Coppel submits that the scope of the contract was not “strictly necessary” because work that was unrelated to the pandemic was carried out in the later stages of the Public First contract. The Judge rejected that submission at [118] to [123]. She said that the scope of work was defined in general terms and that if, in June/July 2020, work was done under the contract that did not relate to the pandemic, that was a question of contract performance, rather than the terms of the contract and the procurement decision.

We agree with the judge. There was nothing objectionable about the original scope of work, which was widely defined in the contract itself (presumably to give the contracting authority the maximum flexibility). Furthermore, if there had been a complaint that the services being carried out by Public First pursuant to those instructions went beyond that which was legitimate, and amounted to an improper modification of the contract, then Regulation 72 provides the mechanism by which that extension could have been challenged by an economic operator. There has never been a challenge to the Public First contract under Regulation 72 nor did Good Law rely upon it. For those reasons, therefore, we reject the submission that the scope of the work in the Public First contract was not strictly necessary.

We reject the arguments that the negotiated procedure without prior publication that led to the contract with Public First was not strictly necessary. Accordingly, the cross-appeal fails. The arguments in respect of apparent bias therefore fall to be considered in the context that:

This was a situation of extreme urgency;

The negotiated procedure without prior publication with Public First, and the contract that resulted, were strictly necessary in accordance with Regulation 32(2)(c).

A claim to bias, actual or apparent, is a serious allegation which should be determined by reference to the pleaded case, not least because that is the case to which evidence will be directed. Paragraph 32 of the Amended Statement of Facts and Grounds contended:

In summary the judge reasoned as follows:

The fact that individuals at Public First were known to and had worked with those involved in the decision making, including the Minister and Mr Cummings, was insufficient to establish apparent bias. However, the existence of personal connections between the Minister, Mr Cummings and the directors of Public First was a relevant circumstance that might be perceived to compromise their impartiality and independence in the context of a public procurement. As such, it was “incumbent” on those involved in the appointment of Public First to ensure that there was a “clear record of the objective criteria used to select Public First over other research agencies” so that they could “allay any suspicion of favourable treatment based on personal or professional friendships”;

The permitted departure from the usual procedural requirements of the Regulations did not constitute a circumstance giving rise to apparent bias either. However, in the absence of a tender competition, it was incumbent on the Minister to ensure that he could demonstrate that the procurement was nonetheless fair and impartial, namely, “by producing evidence that objective criteria were used to select Public First over other agencies”;

On the question of the failure to consider other providers, the judge determined that the difficulty with the Minister’s justification was that it was “not part of the decision-making process at the time that the decision was taken to appoint Public First.” Further, she stated that it did not “stand up to scrutiny” for reasons which she identified. She went on to say that she recognised that everyone was acting under immense pressure and that the urgency of the pandemic crisis did not allow time for reflection. The time constraints justified derogation from the usual procedures in the Regulations, but “they did not exonerate the [Minister] from conducting the procurement so as to demonstrate a fair and impartial process of selection”;

Concern expressed by various Cabinet Office officials regarding the appointment of Public First and alleged unhappiness among anonymous market researchers carried very little, if any, weight.

She concluded that the Minister’s failure “to consider any other research agency, by reference to experience, expertise, availability or capacity, would lead a fair minded and informed observer to conclude that there was a real possibility, or a real danger, that the decision-maker was biased.”

The common law rules against bias date back to the 1860s with the decision of Blackburn J in R v. Rand [1866] LQR 1 B 230 which established judges cannot determine an issue in which they have any pecuniary interest. The rules are rooted in the context of judicial and quasi-judicial decision-making. Procedural fairness requires that the decision-maker should not be biased or prejudiced in a way that precludes fair and genuine consideration being given to the evidence and arguments being advanced by the parties. It aims at preventing a hearing or decision-making process from being a sham or a ritual because the decision-maker is not open to persuasion.

Actual bias has been described as rare and difficult to prove (see Broadview Energy Developments Ltd v. Secretary of State for Local Communities and Local Government [2015] EWHC 1743 (Admin) at [47]); the courts are therefore more commonly asked to look at the circumstances of a case to see if there is an appearance of bias, an allegation which should only be made on a proper basis. The rules against bias are an aspect of the principles of natural justice. The relevant test is now well-established: the court must first ascertain all the circumstances which have a bearing on the suggestions that the decision maker was possibly biased. It must then ask whether those circumstances would lead a fair-minded and informed observer to conclude that there was a real possibility that the decision maker was biased (see Porter v. Magill [2001] UKHL 67; [2002] AC 357 at [102] to [103]).

The fair-minded and informed observer is someone who reserves judgment until both sides of any argument are apparent, is not unduly sensitive or suspicious, and is not to be confused with the person raising the complaint. This observer considers the evidence carefully, having particular regard to the specific factual circumstances, taking a balanced approach and appreciating that context forms an important part of the material to be considered (see Helow v. Secretary of State for the Home Department [2008] UKHL 62; [2009] 2 All ER 1031; [2008] 1 WLR 2416 at [1] to [3]; Almazeedi v. Penner and others (Cayman Island) [2018] UKPC 3; [2018] 2 WLUK 600 at [20]; Gillies v. Secretary of State for Work and Pensions [2006] UKHL 2; [2006] 1 All ER 731; [2006] 1 WLR 781 at [17]).

These principles have been extended to apply to wider extra-judicial decision making, but always and only in an adjudicative context, such as local authority and planning committee decision-making (see for example R (Lewis) v. Redcar & Cleveland Borough Council [2008] EWCA Civ 746; [2009] 1 WLR 830); or a process to determine which of a number of hospitals should conduct specific treatments (R (Royal Brompton and Harefield NHS Foundation Trust) v Joint Committee of Primary Care Trusts [2012] EWCA Civ 472).

In support of its submission that the common law principles of bias applied to the instant facts, Good Law referred to R (ex p Kirkstall Valley Campaign Ltd) v Secretary of State for the Environment) [1996] 3 All ER 304. The relevant question in that case was whether a decision to grant outline planning permission was tainted by bias on the basis that the chairman of the local planning authority committee making the planning decision had a vested interest in the land under consideration. At page 324g Sedley J (as he then was) stated:

That common law principles of fairness applied on the facts of Kirkstall is unsurprising. The planning committee was considering a formal planning application in the context of an adjudicative process, determining whether or not to grant the planning permission sought.

Turning to the present case, however, there was very specifically (and, as the judge held, justifiably) no competitive procurement process or, for example, an application by Public First as part of an adjudicative procedure of any sort. Rather, the Minister was entering directly into a private law services contract with Public First. It is difficult to see how any analogy can be drawn between the award of such a contract and the adjudicative context in which the rules against bias have hitherto been engaged. Unlike in a competitive procurement process, even one conducted outside the Regulations, the Minister was not assessing one or more applications and then making a determination. The Minister was thus not carrying out any adjudicative (and obviously not a quasi-judicial or judicial) function.

The lack of appropriate analogy can be demonstrated by a consideration of waiver: it has long been held that a party may waive his objections to a decision-maker who would otherwise be disqualified on the ground of bias: see, for example R v. Bow Street Metropolitan Stipendiary Magistrate , ex p Pinochet Ugarte No.2 [2000] 1 AC 199, 1441A; R v. Secretary of State for the Home Department ex parte Fayed [2001] Imm AR 134 at [85]; Locabail (UK) Ltd v. Bayfield Properties Ltd [2000] QB 351 at [15] and [26]. Here there was no relevant third party for the purpose of considering whether or not to waive any alleged bias. Instead, Good Law assert that the process was biased against other potential providers of the service and, in particular, the two identified in the pleading (see [60] above). Yet the evidence was that neither was a suitable candidate: see [79] and following below.

The judge and the parties proceeded on the premise that the common law principles of apparent bias were applicable to the facts of the instant case. We are in some doubt that the common assumption was correct. But that issue is not before us as part of the appeal. What follows assumes in Good Law’s favour, however, as the judge did, that the common law principles relating to apparent bias are properly engaged.

The central context for an assessment of the fair minded and informed observer’s belief is the emergency conditions arising out of the pandemic which, in turn, led to the engagement of Regulation 32. There is a tension between the judge’s finding on the one hand (for the reasons set out in [124] of the judgment) that the Minister was entitled to rely on Regulation 32 in awarding the contract, and on the other hand the conclusion (at [164]) that the Minister was nevertheless required (i) to consider other research agencies by reference to experience, expertise, availability and capacity and (ii) to keep a clear record of the objective criteria used to select Public First over other research agencies as part of the process in order to avoid an appearance of bias.

Regulation 32 allowed the Minister to proceed without a competition. That conclusion effectively disposed of the allegation at paragraph 32(ii) of the Amended Grounds: there was no obligation on the Minister to carry out any form of competition. The allegation pleaded in paragraph 32(iii) (ability of others to perform the contract) was not established on the facts; and there were no adverse findings in relation to the amount (in fact the maximum value) payable under the contract. At [167] of the Judgment, the judge said that the amount was irrelevant to the question of apparent bias.

That left only the first allegation (pleaded at paragraph 32(i)), namely the relationship between the directors and owners of Public First and Mr Cummings and the Conservative Party. The judge found in terms that the relationship did not create any apparent bias. Having regard to the specialised nature of the public policy and communications research industry, it was unsurprising that those involved at Public First might have developed professional and/or personal friendships over the years working within government departments. This finding disposed of the claim, at least as pleaded, or as Sir James put it, on the basis of that conclusion there was no “springboard” for a consideration of apparent bias.

The effect of the judge’s conclusions was to find breach on the part of the Minister of an unspecified obligation to carry out a process that involved a formally documented consideration of other research agencies (by reference to experience, expertise, availability and capacity) which gave rise to apparent bias.

This conclusion is, we suggest, at odds with the finding that the Minister was at the same time justified in using a negotiated procedure without prior publication, something which did not require consideration of any other agencies. The question of identifying and evaluating the capacity and suitability of other tenderers in these circumstances did not arise at all. We are unable to accept that in these circumstances the impartial and informed observer would, in effect, require the creation of a common law “procurement regime-light" in the absence of which he would think there was a real possibility of bias. This is sufficient to determine the appeal.

With respect to the judge, we also consider that there are real problems in the approach to the evidence. Under the heading “Failure to consider other providers” starting at [154] the judge set out some of the evidence served on behalf of the Minister explaining the reasons for choosing Public First.

Mr Aiken stated that there was no alternative but to appoint Public First for this work:

Ms Hunt stated that Public First had particular experience and expertise in the services that were required and there was no suitable alternative at that time:

The fair minded and informed observer should be taken to be aware of this evidence (and indeed all evidence from the Minister’s officials who dealt with this matter). The judge concluded that the (first) difficulty with the Minister’s justification was that it was “not part of the decision-making process at the time that the decision was taken to appoint Public First”. (This could only have been a reference to the reasoning not being part of a formal, documented process: it clearly was part of the actual thinking of those involved at the time).

The judge had earlier noted (at [141]) that a decision-maker cannot meet a claim of apparent bias “simply by giving evidence, either detailing what was in his or her mind at the time he or she took the impugned decision, or asserting that he or she was not biased”. She relied on Porter v Magill at [104] and R (Georgiou) v London Borough of Enfield [2004] LGR 497 at [36]. Those authorities indeed confirm that assertion of a lack of bias will not advance matters, but they do not go so far as to say that evidence detailing what was in the decision-maker’s mind, or he knew or did not know, at the material time will not be relevant. Nor are contemporaneous records of thought processes necessary.

Decision-makers cannot refute an allegation of bias simply by asserting that they had an open mind and were not prejudiced. But where, as here, the judge was imputing to the informed observer a view that there should have been consideration of other providers, it obviously would be important to know why the decision to contract with Public First and not others was taken. In so far as the fair-minded and informed observer could not find any publicly documented reasons for the decision at the time, then they would ask for an explanation before reaching any firm conclusions. As Good Law itself points out, the hypothetical observer is “informed” and aware of all the circumstances, including facts ascertained on investigation by the court (see In Re Medicaments and Related Classes of Goods (No 2) [2001] 1 WLR 799 at [83(4)]). Those facts are not limited to the facts available to the hypothetical observer at the time of the decision, or to “publicly available information” (see R (Condron) v National Assembly for Wales [2006] EWCA Civ 1573; [2007] 2 P&CR 4 at [50]; Virdi v Law Society [2010] EWCA Civ 100; 2010 1 WLR 2840 at [37] to [44]).

The Judge also stated (as a second reason for dismissing the Minister’s justification) that the reasons advanced in favour of Public First’s appointment did not “stand up to scrutiny”. She summarised the Minister’s position as follows:

Only two companies in the market had the scale, expertise and experience to provide the requisite services in March 2020, Public First and Britain Thinks;

Public First was trusted and known to be capable of undertaking the required services speedily and effectively, and of debriefing under pressure;

Public First was already in place conducting the research; therefore, using them was the most efficient and effective way of obtaining urgently needed research;

Other companies, such as Jigsaw, YouGov and Kantar, did not have the relevant policy experience or had not carried out similar focus group work.

In dismissing these reasons, the judge was concerned that no one identified the objective criteria for appointment; that no one undertook any assessment of Public First and that neither Mr Cummings nor Mr Aiken considered the possibility of an alternative agency; and that the Minister had failed to use the Research Marketplace Dynamic Purchasing System (“the RM DPS”) to identify potential suppliers.

There is force in the criticism advanced by Sir James that these comments ignore key aspects of the evidence, including Mr Aiken’s (unchallenged) evidence as to the rationale for Public First’s appointment and that use of the RM DPS took six to eight weeks, which was impractical in the circumstances. The judge also effectively rejected Ms Hunt’s unchallenged evidence on what were central issues, including whether there was any other agency that could meet the Minister’s urgent needs.

The general rule is that the evidence of a witness is accepted unless given the opportunity to rebut the allegation made against them, or there is undisputed objective evidence inconsistent with that of the witness that cannot sensibly be explained away so that the witness’s testimony is manifestly wrong. A court hearing a judicial review will generally accept the evidence of the public authority: and will not normally decide contested issues of fact: see, for example, R v. Board of Visitors of Hull Prison ex p St. Germain (No. 2) [1979] 1 WLR 1401 at page 1410H and R (Watkins-Smith) v. Aberdare Girls High School [2008] EWHC 1865 (Admin) , [2008] FCR 203 at [135]; R (Safeer) v Secretary of State for the Home Department [2018] EWCA Civ 2518 at [18]).

In the circumstances of this case, there was no basis for rejecting the evidence either of Mr Aiken or Ms Hunt. The Minister was under a duty of candour. The evidence of both Mr Aiken and Ms Hunt was not contradicted by objective evidence elsewhere. Ms Hunt set out in her second witness statement to expand on why each other agency was not suitable, agency by agency and including Jigsaw. Quite apart from Mr Cummings, experienced civil servants, including Mr Aiken who was responsible for letting the contract, knew what services were urgently needed in a public health emergency and were intimately familiar with the market for their provision.

The question of apparent bias ought to have been assessed on the basis that the reasons behind the decision to award the contract at the time that it was taken included, as set out in the evidence for the Minister, that:

Only two companies in the market had the scale, expertise and experience to provide the requisite services in March 2020, Public First and Britain Thinks;

Public First was trusted and known to be capable of undertaking the required services speedily and effectively, and of debriefing under pressure;

Public First was already in place conducting the research; therefore, using them was the most efficient and effective way of obtaining urgently needed research;

Other companies, such as Jigsaw, YouGov and Kantar, did not have the relevant policy experience or had not carried out similar focus group work.

The fair-minded and informed observer would have been aware of these reasons and the following material context:

There was nothing unlawful in the involvement of Mr Cummings in the decision-making process;

The award of the contract was approved by Mr Aiken;

The extent of the emergency arising out of the pandemic and the position of those involved in the process at the time, as set out in the unchallenged witness evidence of those involved;

The extreme urgency brought about by these unforeseeable events was such as to engage Regulation 32. The use of a negotiated procedure without prior publication was strictly necessary;

There was thus no requirement on the Minister to carry out any procurement process, and no need for the Minister to consider other agencies. He was entitled to award the contract directly;

Specifically, given the urgency of the need for the research in question, it would have been “utterly impractical to instruct someone else” on Thursday/Friday 27/28 February 2020 (as Mr Aiken said). There was an urgent need for focus group testing on the Thursday evening, the results of which were to be provided to No 10 the following day. After that work had been presented, the decision was taken to continue with Public First;

It was vital that the services could be provided immediately and reliably, and that their output could be trusted;

The Minister was not carrying out any adjudicative procedure, but rather making, and entitled to make, his own evaluative assessment in a small close-knit market as to which agency was best suited to his needs.

Against this background, the fair-minded and reasonably informed observer would not have concluded that a failure to carry out a comparative exercise of the type identified by the Judge created a real possibility that the decision-maker was biased. Equally, the fair-minded and informed observer, realising, amongst other things, that the use of a negotiated procedure without prior publication (with Public First) was strictly necessary because of the pandemic emergency, would not have found the absence of any formal record of the decision-making process indicative of apparent bias.

For these reasons, the appeal will be allowed and the cross-appeal dismissed.