Rotala Plc (R on the application of) v Greater Manchester Combined Authority

The power to make bus franchising schemes was conferred on certain local authorities, including mayoral combined authorities, by amendments made to Part 2 of the Transport Act 2000 (“the 2000 Act”) by the Bus Services Act 2017. Where such a scheme is introduced for a particular area then, unless excepted from regulation, local bus services can only be provided in that area under a local service contract awarded by the authority, or under a service permit granted by it.

The First Respondent, the Greater Manchester Combined Authority (“the GMCA”) is the mayoral combined authority in Greater Manchester, comprising the Mayor (the Second Respondent) and an elected member from each of the ten constituent Metropolitan Borough Councils. This appeal concerns the decision of the Mayor, on behalf of the GMCA, on 25 March 2021, to introduce the Greater Manchester Franchising Scheme for Buses 2021 (“the Franchising Scheme”). If implemented, the Franchising Scheme will radically alter how bus services in Greater Manchester are delivered.

The GMCA is the first authority outside London to seek to implement a bus franchising scheme, although the Court was informed that other authorities have announced an intention to follow suit.

There were originally two parallel claims for judicial review of the decision, one brought by the largest bus operator in the Greater Manchester area, Greater Manchester Buses South Ltd (“Stagecoach Manchester”) and the other, this claim, brought by the Appellant (“Rotala”), the second largest bus operator in that area. In a reserved judgment handed down on 9 March 2022, [2022] EWHC 506 (Admin), Julian Knowles J. (“the Judge”) dismissed both claims. Only Rotala sought permission to appeal, and on 5 May 2022 Popplewell LJ granted permission on two related grounds.

Although Stagecoach Manchester is not a party to this appeal, its ultimate parent, Stagecoach Group Plc, is an interested party. It has made no submissions on this appeal. The other interested party, (“OneBus”) is a partnership of commercial bus operators in Greater Manchester which promotes the improvement of public transport in that area. It has taken no part in the judicial review proceedings.

Rotala was represented on this appeal by Mr Andrew Singer QC and Mr Piers Riley-Smith; the GMCA and the Mayor were represented by Mr John Howell QC and Ms Amy Rogers. We are grateful to all counsel for their written and oral submissions.

The 2000 Act provides in s.123A(2) that a franchising scheme may not be made unless the franchising authority has complied with the requirements of ss. 123B to 123G of the Act. These requirements are described in more detail in the judgment below at [20]-[30]. Put simply, there is a sequence of mandatory steps which the franchising authority must take, comprising (1) assessment of the proposed scheme under s.123B; (2) obtaining any necessary consent of the Secretary of State under s.123C (it was unnecessary for the GMCA to do so, because it is a mayoral authority); (3) obtaining an independent audit of the assessment under s.123D; (4) public consultation under s.123E on a consultation paper prepared in accordance with s.123F; and (5) publication of a response to the consultation, together with the authority’s decision, under s.123G. In the case of a mayoral combined authority, the decision to make a bus franchising scheme can only be made by the mayor, acting on behalf of the combined authority (s.123G(4)). Importantly, the decision taken under s.123G is whether to make a franchising scheme, not whether to make the originally proposed franchising scheme. The statute expressly envisages that the scheme may be modified by the authority after the consultation: see s.123E(6).

If it is decided to go ahead with a franchising scheme, the scheme must be made and published at the same time as the report on the consultation under s.123G (s.123H(1)).

The initial assessment under s.123B must describe the effects that the scheme is likely to produce, and compare making the proposed scheme with one or more other courses of action (s.123B(2)). It must also comply with the specific requirements of s.123B(3). Those which are relevant to this appeal are consideration of whether the authority would be able to afford to make and operate the scheme, and whether the proposed scheme would represent value for money (s.123B(3)(d) and (e)).

The franchising authority must also have regard to any guidance issued by the Secretary of State for Transport concerning the preparation of the assessment (s.123B (5) and (6)). This, and all other relevant statutory guidance, is to be found in the Bus Services Act 2017 Franchising Scheme Guidance issued by the Secretary of State (“the Franchising Guidance”). That document also contains non-statutory guidance; the statutory guidance, which must be followed, is underlined.

If, following the s.123B assessment, the authority wishes to proceed with the proposed scheme, it must obtain a report from an auditor on that assessment (s.123D)(1)). The audit report must state whether, in the opinion of the auditor:

the information relied on by the authority in considering whether it would be able to afford to make and operate the scheme and whether the proposed scheme would represent value for money is of sufficient quality;

the analysis of that information in the assessment is of sufficient quality;

the authority had due regard to the statutory guidance in preparing the assessment (s.123D(2)).

The auditor must have regard to any guidance issued by the Secretary of State on the matters to be taken into account when forming their opinion as to whether the information relied on, and the analysis of that information, is of sufficient quality - s.123D(6). Again, that guidance is to be found in the Franchising Guidance. Paragraph 1.85 of the Franchising Guidance requires the auditor to take into account the quality and timeliness of any information received from bus operators and to consider, in particular:

Whether the information used comes from recognised sources;

Whether the information used is comprehensive or selectively supports the arguments in favour or against any particular option;

Whether the information used is relevant and up to date;

Whether the assumptions recorded as part of the assessment are supported by recognised sources; and

The mathematical and modelling accuracy of the analytical methods used to calculate the impacts of the options.

If, after obtaining and considering an auditor’s report under s.123D, the authority wishes to proceed with the proposed franchising scheme, it must publish a consultation document relating to the proposed scheme (s.123F). At the same time it must also publish the s.123B assessment and the s.123D auditor’s report, and give notice of the scheme in such manner as the authority considers appropriate for bringing it to the attention of persons in the area to which it relates. S.123E(3) prescribes the ingredients of the notice, and s.123E(4) lists the persons that the authority must consult after giving such a notice.

S.123G(1) provides that a franchising authority that conducts a consultation under s.123E must publish a report setting out its response to the consultation and its decision on whether to make a franchising scheme covering the whole or any part of their area.

Transport for Greater Manchester (“TfGM”) is an executive body of the GMCA which is responsible for running Greater Manchester’s transport services. It also has certain statutory powers and functions which are more fully described at [9] of the judgment.

As long ago as 30 June 2017 the GMCA decided that an assessment of a proposed franchising scheme should be prepared pursuant to s.123B of the 2000 Act. The Judge found at [10] that TfGM effectively acted as an officer of the GMCA in carrying out work relating to the proposed franchising scheme. The work was presented to the GMCA, which then voted on whatever recommendations had been made by TfGM in relation to that work.

The assessment pursuant to s.123B was carried out by TfGM and completed in June 2019. On 28 June 2019, the GMCA instructed Grant Thornton to provide an independent audit report on that assessment pursuant to s.123D of the 2000 Act. That report was provided on 26 September 2019. The public consultation under s.123E of the 2000 Act was carried out between 14 October 2019 and 8 January 2020. It therefore concluded before the global Covid-19 pandemic began.

In June 2020, TfGM produced a report on the consultation, but that report did not attempt to consider the potential impact of the pandemic on the proposed franchising scheme. TfGM recommended the adoption of the proposed scheme, subject to some suggested modifications in the light of some of the responses to the consultation. However, they stated that before taking any decision it would be necessary to consider the consequences of Covid-19, once its impacts on the economy and on public transport in Greater Manchester had become clearer.

Having taken stock of the situation, the GMCA produced an “Update Report” on 26 June 2020, which indicated that it did not consider it appropriate at that juncture to publish a formal response to the consultation pursuant to section 123G of the 2000 Act. Instead, paragraph 3.6 of the Update Report stated that:

The GMCA then asked TfGM to prepare a further report to consider how the pandemic had affected the analysis and the key conclusions in the original June 2019 assessment.

On 10 November 2020, TfGM presented a report to the GMCA entitled “COVID-19 Impact on Bus Franchising Report” (“the COVID Impact Report”). TfGM first looked at how the pandemic had already affected the bus market in Greater Manchester. It then postulated four potential future transport scenarios to see how these, and Covid-19, might affect various of the conclusions set out in the original assessment, including the conclusion that the scheme could be successfully procured, and the conclusion that the scheme was affordable. The COVID Impact Report also contained an evaluation of how Covid-19 might affect the alternative options that were considered in the assessment or were put forward by bus operators during the consultation process.

The pandemic was a totally unforeseen event which potentially impacted on the reliability of the assessment of what would happen if the scheme were to be introduced, and whether it would represent value for money. However it had no impact on the quality of the historic information that had been used as the basis for the original projections, nor on the methodology that was adopted at that time. TfGM’s exercise was carried out at a time when there were numerous uncertainties, including how much longer the pandemic would continue, how it would affect the future use of public transport, and what the economic repercussions would be for authorities such as the GMCA (or their constituent Boroughs), and it was designed to test whether these uncertainties undermined the conclusions reached in the earlier assessment.

The GMCA asked Grant Thornton to review the COVID Impact Report. Grant Thornton’s review, entitled “assessment of the TfGM Covid-19 Impact on Bus Franchising Report” (“the Assessment Review”), was sent to TfGM on 19 November 2020. In the Assessment Review, Grant Thornton made it plain that they were not carrying out an audit of the COVID Impact Report but rather, providing the GMCA with independent assurance on the approach taken by TfGM in preparing it.

Since the role that was undertaken by Grant Thornton in cross-checking the approach that was taken by TfGM was a completely different role from that which it performed as statutory auditor under s.123D, it might equally well have been carried out by another accountant or management consultant. However, it made obvious sense for the GMCA to engage Grant Thornton given their detailed existing knowledge of the factual background. As the Judge held at [182], Grant Thornton identified a risk that the outturn position may be materially different, but the existence of that risk did not necessarily mean that the key conclusions in the assessment could no longer be relied on, and they did not suggest that that was the case.

Both the Covid Impact Report and the Assessment Review were presented to the GMCA as appendices to a report from its Chief Executive on 27 November 2020. The GMCA concluded that it was the right time to proceed with the modified franchising scheme recommended by TfGM. It approved the publication of the COVID Impact Report and the Assessment Review. It then undertook a second public consultation, in which consultees were afforded an opportunity to comment on the COVID Impact Report and on the question whether to proceed with the proposed franchising scheme now, or whether it should be delayed. The second consultation ran from 2 December 2020 to 29 January 2021, though Stagecoach requested (and was granted) a short extension. There were more than 4,000 responses to that consultation.

On 21 January 2021, Rotala filed its claim for Judicial Review, which was initially aimed at the decision of the GMCA in November 2020 to move ahead with the proposed scheme. Stagecoach filed its claim, which also challenged that decision, on 26 February 2021. After Summary Grounds of Defence to both claims had been filed, the November decision was overtaken. TfGM’s report on the second consultation was presented to the GMCA at a meeting on 23 March 2021, together with the report prepared by the GMCA’s Chief Executive pursuant to s.123G of the 2000 Act, which considered both consultations and the responses to them and made recommendations for the consideration of the GMCA’s members. The impugned decision was taken by the Mayor on 25 March 2021. Both claims for judicial review were duly amended to challenge the lawfulness of that decision.

The hearing of the claims took place in late May 2021. The judgment, which runs to 331 paragraphs, was handed down on 9 March 2022. It addresses the numerous grounds of challenge raised by Stagecoach and Rotala, many of which overlapped but had nuanced differences. The issues on this appeal are a refinement of certain aspects of Rotala’s case as presented to the Judge; indeed, the arguments became even more refined in the course of the hearing, in response to judicial questioning.

In these circumstances, no useful purpose would be served by summarising the Judge’s reasons for rejecting the arguments that were advanced before him. Suffice it to say that I found his analysis to be impeccable, and that for the reasons which will appear below, I agree with him that the decision was neither unlawful nor irrational.

The first ground of appeal, and the one primarily relied on by Rotala, is that as a matter of statutory interpretation, the GMCA were required to obtain and consider another audit report pursuant to s.123D from an independent auditor on the original assessment as affected by the views in the COVID Impact Report, before making the decision. The Judge had held that the statutory requirement to obtain a report under s.123D had been fulfilled in September 2019 and did not need to be revisited; it is contended that this was an error of law.

The second ground of appeal is that even if the Judge was correct, and there was no statutory obligation to require a further s.123D audit report, it was irrational of the GMCA and the Mayor to proceed without obtaining such a report and therefore the decision was unlawful. It was contended that the Assessment Review provided no actual assurance upon which a rational decision-maker could properly rely.

Mr Singer accepted that by June 2020 all the requisite statutory steps had been taken up to and including the public consultation. There was no criticism of the s.123B assessment, of the auditor’s report on that assessment, nor indeed of the consultation process. He also confirmed that it was not Rotala’s case that, when faced with the uncertainties of the pandemic, the GMCA was legally obliged to go through the whole statutory process from scratch, beginning by carrying out a fresh assessment. That had been the position adopted by Stagecoach, which was addressed and dismissed for cogent reasons by the Judge at [178]–[183] of the judgment.

However, Mr Singer contended that it was implicit in the statutory scheme that if there was a material change of circumstances, the auditor would be obliged to re-assess the audit in the light of that change. He submitted that it was necessarily implicit that if the audit report produced under s.123D(2) was negative, (e.g. if it was critical of the methodology or the robustness or reliability of the information used in the s.123B assessment) the authority could not lawfully proceed with the consultation. The same must apply, he contended, if the change in circumstances were of such a nature as to cast doubt on the continued validity of the initial assessment, or the reliability and robustness of the information underlying it. It could not be right to treat each stage in the statutory scheme as a mere box-ticking exercise, never to be revisited once a particular stage was complete.

It is appropriate to begin by considering the express provisions of the statute. The 2000 Act unambiguously identifies the subject-matter of the audit report under s.123D, namely, an assessment of the scheme by the franchising authority under s.123B. S.123D does not require the auditor to produce a report on some other document. Even if it were possible to carry out something in the nature of a s.123D audit on the Covid Impact Report, it could not be an audit report under s.123D.

The statute makes no provision for a fresh s.123B assessment, or for the revisitation of the original s.123B assessment, in the light of the provision of further information which might have a bearing on the conclusions reached by the authority in the assessment. For example, although the authority is required to consider the proposed scheme against other options when making the assessment, it is not required by the statute to undertake a new assessment, or revise the initial assessment and have it re-audited, if a further option which was not considered at the time of the original assessment is proposed by one of the consultees.

Likewise, the statute contains no express requirement for the revisitation of the s.123B assessment or for a fresh assessment to be carried out if information comes to light in the course of the public consultation which calls into question the robustness or reliability of the information relied on in the original assessment. Indeed, there is nothing on the face of the statute that expressly obliges the authority to revisit its s.123B assessment if the audit report draws its attention to flaws in it, however serious those flaws might be. The absence of any express requirement to revisit the s.123B assessment at any time after it has been audited, is a major obstacle to construing the statute as providing for a fresh s.123D audit.

All the circumstances to which I have adverted – the emergence of a new option for delivery of bus services; the provision of further information in consequence of the consultation; and concerns expressed by the auditors, are entirely foreseeable in the course of the statutory process. If the statute does not require a further assessment and/or audit to be undertaken in circumstances that are foreseeable, it is hard to see how it could be construed as requiring such steps to be taken in circumstances which were entirely unforeseeable, such as a global pandemic.

The 2000 Act also makes plain when the s.123D audit is to be obtained and for what purpose; it takes place after the assessment and before the public consultation on the proposed scheme, and is published with the assessment for the benefit of the consultees. The specific timing for provision of the audit report, the express provision as to what should be done with it, and its specific function, are further factors strongly pointing against a statutory obligation of the kind contended for by Rotala.

A further obstacle to implying the provision for which Mr Singer contended is the absence of anything in the Franchising Guidance to support the concept of a continuing role for the auditor or an obligation on the authority to re-instruct them. Importantly, paragraph 1.87 underlines the point that the function of the auditor is not to pass judgment on the outcomes of the assessment, but merely to consider the process that has been followed and the accuracy and robustness of the information that has been used in the analysis, and to satisfy themselves that the mechanics of the process have been carried out correctly. Once that is done, the auditor’s role is fulfilled.

Although the 2000 Act makes provision for further regulations (s.123U) or guidance (s.123W) to be issued by the Secretary of State, there is nothing of that nature pertaining to a further independent audit following the initial public consultation. There is nothing in the Franchising Guidance, for example, to require the authority to revise its assessment in the light of the answers to the s.126E consultation and, to the extent that the consultees have produced further information which was not available at the time of the original assessment of the scheme, to obtain another audit of that information (or of a revised assessment of the scheme which takes it into account).

Paragraphs 1.84 to 1.86 of the Franchising Guidance are directed at the evaluation of the quality of information relied upon by the authority in the s.123B assessment. The guidance then goes on, in paragraph 1.88, to describe what happens after the audit report has been prepared, namely, the public consultation. It is not envisaged that there would be any further interaction between the authority and the auditor thereafter.

Paragraph 1.86 of the Franchising Guidance obliges the auditor to advise the authority if they consider that one of the specified criteria (set out in paragraph 1.85 and quoted in paragraph 12 above) have not been satisfied. I accept that that obligation is not limited to reasons which existed at the time of the assessment. If something came to light between the assessment and the audit which threw doubt on the accuracy of the information that had been used by the authority in its assessment, the auditor would be obliged to draw it to the attention of the authority. However, the guidance is silent about what should happen after it has done so. It does not envisage the auditor continuing to be involved once it has given that advice.

If Parliament had intended to impose an obligation on an authority to commission a further audit in certain circumstances, one would not expect both the statute and the statutory guidance to be silent on that topic. Yet Rotala’s case on this issue depends on reading words into the statute which do not appear on its face. It would also involve making corresponding changes to the guidance.

The parties agreed that the leading authority concerning when it is permissible to read words into a statute in order to give effect to Parliament’s intention is Inco Europe and others v First Choice Distribution and others [2000] 1WLR 586. After emphasising that this power is reserved for cases of clear drafting error, at p. 592F-G Lord Nicholls set out the three criteria of which the court must be “abundantly sure” before interpreting a statute in this way, namely:

Mr Singer also relied upon a passage in the speech of Lord Hodge DPSC in R (on the application of O (A Child)) v Secretary of State for the Home Department [2022] UKSC 3, [2022] 2 WLR 343 at 29-31, which describes the process of statutory interpretation. Lord Hodge (with whom Lord Briggs, Lord Stephens and Lady Rose JJSC agreed) said, at 31, that:

None of those principles is controversial but, given that on any view we were being invited to read words into s.123D(2) which do not appear there, rather than to interpret the express provisions of the statute (which are clear enough) it seems to me that Inco Europe is the more relevant of these authorities. It is plain that Lord Nicholls’ three-stage test is not met in the present case.

The primary purpose of the statutory requirement of an audit under s.123D is to provide consultees with some independent quality assurance of the underlying information and the methodology used by the authority to assess it. That purpose is fulfilled when the audit is completed and sent out with the assessment under s.123B and the consultation document. A secondary purpose is to draw to the attention of the authority any problems identified with the information or the methodology used in the assessment (something which the Franchising Guidance expressly obliges the auditor to do) so that if need be, the authority can address them. That purpose is fulfilled when the authority receives the audit report.

There is no need to read any further words into s.123D to ensure that either of the intended statutory purposes is fulfilled. Parliament did not intend the audit to be revisited in the light of the consultation responses, and therefore it cannot have been taken to have intended it to be revisited in the light of any other subsequent events. There was no obvious inadvertent drafting error here.

In any event, when put to the test, the third of Lord Nicholls’ requirements turned out to be impossible to fulfil. The Court asked Mr Singer to formulate the substance of the revised version of s.123D(2) for which he contended. Initially he submitted that the phrase “the information relied on by the authority … is of sufficient quality” in s.123D(2) (a) and (b) should be read as meaning “is and remains of sufficient quality until the decision under s.123G is taken”.

A major problem with that formulation is that the statute expressly requires the auditor to provide one report on the assessment carried out by the franchising authority under s.123B and to do so prior to any consultation. Once that task is complete, as I have said, his role is fulfilled; the decision under s.123G is some way off at that stage. The time at which the auditor must be satisfied of the quality of the information is obviously the time of the audit report. If the authority still wishes to go ahead with the scheme, both the assessment and the audit report are disseminated to the consultees with the consultation document. There is nothing in the statute which requires the auditor to remain involved in the process after delivery of the audit report, let alone during or after the public consultation, and I can see no justification for implying those words into s.123D. It is impossible to infer that they were omitted by oversight, not least because they are contrary to the natural and obvious interpretation to be placed on the express language of that section, objectively ascertained.

Mr Singer’s second attempt at formulating the substance of the insertion was:

Moreover, this formulation falls short of imposing a positive obligation on the authority to carry out another s.123B assessment, (or amend the original in the light of the material change in circumstances) without which there can be no audit under s.123D. As I have already pointed out, the only statutory audit that can be undertaken is an audit of an assessment under s.123B. At the time when the statutory obligation is said to arise, there has already been one audit of the only assessment under that section, and that is the only audit for which the statute provides.

The Covid Impact Report was not an assessment under s.123B, and although at one point Mr Singer sought to characterise it as a continuation of the original assessment, it is obvious that it was of an entirely different nature. As the Judge found at [177], the primary purpose of the Covid Impact Report was to test the analysis in the assessment in the light of the pandemic, to see if the recommendations made on the back of that assessment, in the light of the initial consultation, should stand. Mr Howell correctly characterised it as a form of sensitivity analysis which “stress-tested” the key conclusions in the assessment on affordability and value for money. However one construes the statute, it cannot give rise to an obligation to carry out a s.123D audit of a document which falls entirely outside the statutory scheme, either alone or in conjunction with the original assessment.

I do not even accept Mr Singer’s initial premise that as a matter of necessary implication the authority would be prohibited by the statute from going ahead with the consultation (or seeking to implement the scheme) in the event of what Mr Singer termed a “negative” audit report. As I have said, the audit report is designed to act as an independent assurance for the consultees that the franchise authority has gone about the assessment process in an appropriate way. If the audit identified serious flaws in the assessment, in practice there would be little purpose served by an authority taking the next steps towards implementing the scheme without first addressing them. The authority would be inviting a claim for judicial review if it went ahead on the basis of an assessment which it was obliged to tell the consultees its auditors considered to be seriously flawed. However, all that the statute requires is that the audit of the assessment be obtained and that it be provided, with the assessment, to the consultees if the authority wishes to go ahead. The fact that it would probably be a recipe for litigation if the authority were to do so in the teeth of serious criticisms in the audit report, is no justification for reading into the statute words that are not there.

There are good reasons for not doing so. The omission of a prohibition on proceeding to the next stage in those circumstances is not a case of obvious drafting error. The draftsman could not possibly provide for the range of views that the auditor might take, nor for the impact those views might have on perceptions of the reliability of the s.123B assessment. The concerns raised by the auditors might be capable of being addressed relatively simply and swiftly, or it might even be possible for the authority to demonstrate that the concerns were misplaced, in which case there is no justification for treating the “negative” audit as an absolute bar to progress in the manner suggested by Mr Singer. It is a question of fact and degree in each case whether a reasonable authority would take the next steps towards introducing the scheme, armed with the information in the assessment and the statutory audit, all of which it would have to provide to the consultees.

That does not mean that once it is signed off, the audit report presents an insuperable obstacle to challenging a decision to adopt the scheme proposed in the assessment (with or without modification) if it subsequently transpires that the information used by the authority in making the assessment was materially out of date, incomplete or otherwise flawed. In the course of argument, my Lord, Lord Justice Snowden, postulated the scenario in which it comes to light after the public consultation that key documents locked in a data room were inadvertently overlooked by the authority at the stage of carrying out its s.123B assessment. It may transpire that the further information which was not taken into account is of such importance as to make it irrational, in the Wednesbury sense, for the authority to seek to go ahead with the proposed scheme without first considering the fresh information and, if need be, revisiting the assessment. But then it would be for the authority to decide how to address the situation. Parliament has not prescribed by statute what should happen in those circumstances.

For all the above reasons, I have no hesitation in rejecting the first ground of appeal.

I can deal more shortly with the second ground of appeal. Mr Singer contended that it was irrational for the GMCA to proceed on the basis of the Assessment Review instead of commissioning a re-audit of its assessment (in accordance with s.123D and the Franchising Guidance) in the light of the pandemic. He submitted that the Assessment Review provided no objective judgment on the reliability of the financial models that were used by TfGM in the original s.123B assessment. It merely certified that the approach taken in the COVID Impact Report in considering affordability and value for money of the proposed franchising scheme was “appropriate”. It also expressed Grant Thornton’s opinion that the information and analysis of the information in the COVID Impact Report was “of sufficient quality for the purposes of the report” but that was not the same as an application of the quality assessment criteria in the Franchising Guidance to the information used in the assessment itself.

Mr Singer also complained that the reference to Grant Thornton as “auditor” created a false sense of security. However, as TfGM said in response to the Second Consultation, Grant Thornton’s covering letter was clear about the nature of their work, and there was no scope for any consultee to be misled about that.

In my judgment, once it is established that there is no statutory obligation to carry out a fresh audit, such an obligation cannot be imposed on the GMCA by the back door, on the basis that this would be the only rational step for it to take in these circumstances. A reasonable authority which has gone through all the steps it is legally obliged to carry out before making a decision, and has sensibly put the decision on hold, has a choice as to how it goes about evaluating whether (and if so, when) it should still go ahead with the proposed scheme in the light of the uncertainties created by an unforeseen global pandemic.

Mr Singer did not submit that it was irrational for the GMCA not to carry out a fresh assessment of the scheme, even though further information may have come to light in the course of the consultation process which would have a bearing on it, such as the emergence of a new potentially viable alternative. Once it is accepted that the GMCA does not have to start the process all over again, there is no justification for imposing a mandatory requirement upon it to carry out a further audit of the original assessment or to reappraise the financial models that were used in that assessment. A further audit was one option, certainly, as the Judge recognised, but it was not the only one that was rationally open to the GMCA.

The Assessment Review addressed the question whether the key conclusions in the original assessment remained valid given the uncertainties – see the Judgment at [92] and [162]. The assessment itself was not updated, but Grant Thornton used the conclusions in it as a starting point, see the Judgment at [214] to [216]. The scenarios which were used in the exercise by TfGM were intended to cover the range of hypothetical possibilities, though the extreme scenarios were unlikely to arise in practice.

As the Judge stated at [217], the exercise in which TfGM and the GMCA were engaged was shot through with the need to exercise judgment on complex economic and policy questions, a process which had been made all the more difficult by a public health emergency which was unprecedented in modern times. He rightly characterised the question whether the COVID Impact Report had suitably addressed the sensitivity of the conclusions in the assessment in the light of the pandemic as “one of those matters of judgment”. Grant Thornton gave their professional view on that question. It was plainly rational to ask Grant Thornton to check that the exercise undertaken by TfGM had been done properly, and to draw comfort from their opinion that it had.

The consultees had the benefit of seeing the COVID Impact Report and Grant Thornton’s Assessment Review during the second consultation process and it would appear from their responses that they were under no illusions as to the nature of the exercise undertaken by Grant Thornton. TfGM, in its Report on the Second Consultation, answered the criticisms made of the Assessment Review, including that it was not a further audit report; and the Judge comprehensively rejected those criticisms in his judgment for good and sufficient reasons.

It was not irrational for the Respondents to rely on the Grant Thornton Assessment Review. It fulfilled the purpose for which it was commissioned, which was to provide an independent assurance of the appropriateness of the means by which TfGM chose to check whether the conclusions reached in the original assessment could still be relied on in the light of the uncertainties introduced by the pandemic. The original audit was not concerned with the reliability or otherwise of the conclusions, but only with the robustness and reliability of the means by which they were reached. The two exercises were completely different. The course which the GMCA took cannot be castigated as one which no reasonable authority in their position would be entitled to take.

The second ground of appeal also fails.

Accordingly, for the reasons stated above, I would dismiss this appeal.

I agree.

I also agree.