Gravis Solar 1 Ltd & Anor, R (On the Application Of) v Gas And Electricity Markets Authority

This case concerns the lawfulness of the withdrawal on 2 June 2020 of the accreditation of a small solar photovoltaic, or PV, electricity generating station, Kelly Green, by Ofgem. AMP GM011 Ltd, AMP, the second Claimant, owns and operates Kelly Green. Gravis Solar 1 Ltd, GS1, the first Claimant owns AMP. Ofgem carries out accreditation functions on behalf of the Gas and Electricity Markets Authority.

The primary basis upon which the withdrawal of accreditation is said to have been unlawful is that it breached the rights of the Claimants under Article 1 of Protocol 1 to the ECHR; they were each victims, unlawfully, notably disproportionately, deprived of the “possession” of accreditation of Kelly Green, and its consequent rights and expectations. The principal issues raised were whether GS1 was a victim within s7(1) Human Rights Act 1998, and whether the decision struck a disproportionate balance between its impact on private rights and the protection of the public interest. At its heart, is the way in which Ofgem acted upon its conclusion that accreditation had been obtained through the presentation to it, by or on behalf of AMP, of information known to AMP to be inaccurate, its acceptance that GS1, which did not own AMP until January 2018, was innocent of deception, and that AMP’s former owners were not part of the new corporate ownership structure or now involved at all with Kelly Green.

Ofgem accepted that AMP was a victim for the purposes of s7 HRA; it also accepted, at least for the purposes of this case, that accreditation, and what went along with it, amounted to “possessions” within A1P1 but were only possessions of AMP, and not of GS1. Ofgem did not seek to controvert the submission of Mr Sinclair for the Claimants that its action amounted to a deprivation of possessions rather than being an instance of control.

A statutory scheme known as the Renewables Obligation Scheme, ROS, has been created for subsidising, and so encouraging, the generation of electricity from renewable energy sources. The main legislation for this Scheme is the Renewables Obligation Order SI No. 2015/1947, ROO. The significance of accreditation in the ROS is that it enables Ofgem to issue Renewables Obligation Certificates, ROCs, to renewable generators, for every megawatt hour of electricity they generate from the renewable source. A renewable generating station cannot be accredited until it has been commissioned. These ROCs are valuable because electricity suppliers are obliged to source a proportion of the electricity they supply from renewable sources. Without those ROCs, the generators of electricity might not be able to sell it to the suppliers of electricity at a price which provided a return on the generation investment adequate to justify it. These ROCs can also be sold to intermediaries, but their ultimate value lies in the use which suppliers are required to make of them to prove compliance with their renewable supply obligation. The withdrawal of the accreditation of Kelly Green meant that no further ROCs could be issued, and the owners would be left to what the supplier market would pay, without the incentive of the ROC.

Ofgem withdrew the accreditation because it concluded that it had been misled about the date when Kelly Green was commissioned. By the Renewables Obligation Closure etc (Amendment) Order, ROCA, SI No 2016/457, dated 25 March 2016, amending the Renewables Obligation Closure Order I No 2014/2388, the ROS was closed to the accreditation of small solar PV generating stations, i.e. those with a capacity below 5 megawatts, MW, commissioned after 31 March 2016. Kelly Green’s capacity was 4.9MW. If Kelly Green were commissioned after 31 March 2016, it would have been too late to benefit from the ROC purchase mechanism in the ROS. Thereafter, however, there was a grace period of 1 year, to 31 March 2017. This enabled small PV generating stations, in respect of which, so far as material and putting it broadly, substantial expenditure on construction had already taken place by 31 March 2016, to receive accreditation. This yielded lower rates for the electricity generated.

Accreditation was granted in respect of Kelly Green on 13 January 2017, on the basis that Kelly Green had been commissioned on 15 March 2016. Accreditation, absent various specific statutory reasons, would yield ROCs for 20 years from accreditation, here until 2036. It was granted to AMP GM011 Ltd, the second Claimant, a company set up as a special purpose vehicle for the purpose of owning Kelly Green and operating it. Before GS1’s acquisition of AMP, AMP was a wholly owned subsidiary of Solarplicity Debt Funding Ltd, in turn a wholly owned subsidiary of Solarplicity AS Holdings Ltd. Solarplicity Asset Ltd owned the latter. These also owned other special purpose vehicle companies for owning and operating other solar PV generating stations.

In January 2018, through a share purchase agreement, GS1, set up for the purpose by its fund management owner, Gravis Capital Management Ltd, purchased all the shares in Solarplicity AS Holdings Ltd, becoming the owner, at one remove, of all the shares in AMP. AMP has at all times owned and operated Kelly Green, and was and would be entitled to the ROCs issued in respect of it.

I shall have to go through the way in which Ofgem approached accreditation and later concluded that there had been fraud in the information supplied in respect of commissioning in some detail. But the statutory provisions need to be set out before the Decision Letter.

The Electricity Act 1989, s3A, set the principal objectives for the Gas and Electricity Markets Authority, GEMA as being to “protect the interests of existing and future consumers in relation to electricity….”, which included their interests in the reduction of electricity-supply emissions of greenhouse gases, and in the security of the supply of electricity to them. Mr Sinclair emphasised s3A(5A) which required GEMA to have regard to principles which were “transparent, accountable, proportionate, consistent and targeted only at cases in which action is needed” and any other principles which it thought represented best regulatory practice.

The Electricity Act 1989 is one of the empowering provisions for the Renewables Obligation Order 2015. I do not need to set out the provisions which provide for the mechanisms behind the ROS; I focus on those which deal with accreditation and withdrawal. I do not need separately to deal with ROCs. Article 43 precludes the issue of ROCs in respect of electricity generated by a station which is not accredited. Although at one time, Mr Sinclair raised an issue about the suspension of ROCs, he accepted that there had been no properly formulated challenge to the actions of Ofgem in suspending their issue. There may be such a challenge in other proceedings. The decision on the lawfulness of the withdrawal of accreditation will determine what should happen here in respect of ROCs suspended, revoked or not issued.

Article 89 deals with accreditation: “(2) Where a generating station has been commissioned, the Authority may, upon the application of its operator,…grant the station accreditation.” Article 2 defines “commissioned” as meaning “…the completion of such procedures and tests in relation to that station as constitute, at the time they are undertaken, the usual industry standards and practices for commissioning that type of generating station in order to demonstrate that that generating station is capable of commercial operation.” There is no issue about the meaning or application of that term in these proceedings, though there may be in others. Mr Grodzinski QC for Ofgem pointed out that the definition relates to the completion of tests, and tests which have to demonstrate commercial operability.

This guidance sets out the steps to be undertaken to obtain accreditation and receive ROCs; it repeats what the ROO requires for tests. The conditions of accreditation include a requirement to accept random checks, for example, on the accuracy of information provided at the time of accreditation.

Article 90 covers the withdrawal of accreditation. I note Article 90(2) which permits GEMA to attach to the accreditation “such conditions as appear to it to be appropriate.” Article 90 (3)-(4) provides:

“(3)

Where any of the circumstances mentioned in paragraph (4) apply in relation to...an accreditation which the Authority has granted (whether or not under this Order) and having regard to those circumstances the Authority considers it appropriate do so, the Authority may –

(a)

withdraw the…accreditation in question;

(b)

amend the conditions attached to the…accreditation in question;

(c)

attach conditions to the accreditation.

(4)

The circumstances referred to in paragraph (3) are as follows- …

(c)

the Authority has reason to believe that the information on which the decision to grant the …accreditation was based was incorrect in a material particular, …”

The Ofgem guidance also explains the circumstances in which accreditation may be withdrawn. This includes, [3.93], with a gloss on the wording of the ROO, where “we have reason to believe that the information that the decision to grant the accreditation was based on was incorrect in a way that makes the station ineligible.” It also explains, [3.97 -3.100], that routine checks are carried out on accredited stations to make sure that generators are complying with the scheme rules: “Auditing can help identify and protect against errors and fraud.” The selection of stations for audit may be based on issues which have come to notice but may also be random. These audits review, among other matters, “commissioning evidence” and data behind ROC claims. An audit trail should be kept for at least 6 years. ROCs may be suspended until issues are resolved.

I shall set out the Decision Letter, DL, and then the events leading up to accreditation and its withdrawal, referring to the documents relied on in it. Mr Sozansky, its author, Ofgem’s Head of Compliance, who is responsible for administering the ROS, said that all the information Ofgem had had been taken into account, but identified six items in particular. These were: “the information provided with the accreditation application;

Mr Grodzinski submitted that this was important in understanding what in fact had been considered, and in understanding the DL as the culmination of a series of exchanges in which arguments had been elaborated and responded to.

The DL then set out the relevant legal provisions, including the power relied on, in particular, Article 90(4)(c) of ROO. It also dealt with the power to revoke ROCs which is not an issue here.

The decision itself followed in these terms:

“For the reasons given in my minded-to letter dated 12 December 2019, and taking into account the information that has come to light since then, I have reason to believe that the decision to grant accreditation to Kelly Green PV was based on information which was incorrect in a material particular, in that the generating station was not commissioned by 15 March 2016 as stated in the documentation provided to Ofgem in support of the accreditation application. Further, I consider that it is appropriate in the circumstances to withdraw the accreditation. In response to our minded-to letter dated 12 December 2019, your solicitors have presented a number of arguments in their letter of 17 February 2020 in respect of Ofgem’s position to withdraw the accreditation. In summary, your solicitors argue that Ofgem’s minded to withdraw position is “based on an erroneous interpretation of the relevant provisions of the ROO 2015” including (i) the definition of “commissioning…; and (2) whether revoking accreditation would be a rational, proportionate and justifiable decision in the circumstances.”

It continued:

“Your solicitors’ criticism of Ofgem’s minded-to position appears to be based on the misunderstanding that the rational for Ofgem’s decision is based on our requiring actual generation at the full capacity of the station.

This is not the case.

As set out above and in my minded-to letter, Ofgem’s position is based on the commissioning definition set out in article 2 of the ROO 2015, which requires the completion of industry standard procedures and tests to demonstrate capability of commercial operation. This takes into account whether the station was fully built at the relevant time so that those procedures and tests were able to be carried out to the station to demonstrate capability of commercial operation.

Your solicitors acknowledge that whether a station has “commissioned” depends on the circumstances of each case. I agree. Further, this is made clear by the commissioning definition in the ROO, the usual industry standard procedures and tests that need to be completed to demonstrate capability of commercial operation, depend on the type of generating station that is commissioned. …

As set out in my minded-to letter, in the case of the Kelly Green station, our auditors found: -

•

The DC test sheets for the station were largely identical to those provided by Solarplicity for other generating stations within the portfolio audited suggesting that the test sheets were not genuine;

•

There was a lack of generation after the stated commissioning date between 15 March and 23 May 2016;

•

That only 70% of the inverters were operational from 23 May 2016;

•

The site representative admitted that certain areas of the station were not complete at the time of initial generation.

Moreover: -

•

Your explanation in response to the audit findings based on information provided by Solarplicity about the apparent relocation of the station was not disclosed to us by Solarplicity in response to our queries during the accreditation process;

•

No documentary evidence has been provided to support this version of events;

•

It does not accord with the planning permission that was granted;

•

No explanation has been provided to satisfactorily explain the lack of operation of the inverters.

Now, the serious matters contained in your letter dated 17 February 2020 have been disclosed.

Some of the material to which I now come was not known to GS1 at the time of its purchase of the shares in the companies in the Solarplicity group, but it would have been known to AMP of course, through its directors and those acting on its behalf on the application. GS1, at purchase, would have known or would have been in a position, by asking for the application documents, to find out the content of the application. It did not know of investigations carried out by Ofgem in 2016, or why it had carried them out. Solarplicity did not tell them. The detail of Gravis Capital Management’s, GCM’s, enquiries before purchase have not been disclosed; Solarplicity’s answers may not have been the whole truth.

The application was made on-line, and is not in the documents before the Court. But, as Mr Grodzinski pointed out, the only person who can make the application for accreditation is the operator of the station. Although AMP was a corporate special purpose vehicle, that is a business description and not a legally distinct form of corporate entity. It was a company with shareholders, albeit corporate, and would

A consultant on behalf of AMP provided evidence to support the asserted commissioning date of 15 March 2016. The documents submitted, signed by the superuser of AMP, Mr Elbourne of Solarplicity, and dated 15 March 2016, appeared to meet the requirements of the Act and guidance: G59 test certificate, Direct Current, DC, string test results, inverter commissioning certificates, photographic evidence of gross meter readings and Half Hour generation Data, HHD, with letter and timeline. It was not an application for accreditation under the grace period provisions.

However, while Ofgem was reviewing the application and its supporting material, a whistle-blower in the Solarplicity group provided Ofgem with information about a number of ROS applications made by Solarplicity group companies at around this same time. Their information concerned the accuracy of the asserted commissioning dates, but it was not supported by evidence. Ofgem’s investigations produced satisfactory answers for some only of the stations. For Kelly Green, Ofgem sought more evidence about the HHD records. The problems were said to arise from metering/data collection issues at the Kelly Green site. Solarplicity gave apparently credible replies to further questions, supported by screenshots of meter connection flows, and confirmation from the operator of the distribution network that 15 March 2016 was the connection date and that the site could generate electricity on to its network at that date.

That satisfied Ofgem, and accreditation was granted to AMP by Ofgem in its letter of 18 January 2017, addressed to the superuser, Mr Elbourne at AMP, with an accreditation date of 15 March 2016, and a rate of ROC issuance of 1.3ROC/MWh. It contained the standard conditions, including one requiring AMP to allow random checks on the accuracy of data submitted for obtaining accreditation. GCM knew of that condition.

Ofgem had also carried out sample site audits of two other stations in respect of which Solarplicity had sought accreditation through other special purpose vehicle companies, putting the accreditation of Kelly Green on hold until satisfied that there was no fraud in relation to those other sites.

After accreditation, and after the purchase by GCM of various Solarplicity companies, Ofgem started carrying out audits of stations with a commissioning date near to the closure date of the RO Scheme to small PV stations. Ofgem used Black & Veatch, B&V, to carry out these audits. In January 2019, B&V carried out an audit of the Kelly Green PV station. It found serious problems of non-compliance concerning the commissioning date. This led it to consider that Kelly Green had not

Ofgem sent the audit report to Mr Parker of AMP on 29 March 2019. GCM responded on 10 May 2019: 70% of the inverters showed no generation because the station had been constructed under overhead power lines and the station had had to be relocated; diagrams showing the original and revised locations were provided. But Ofgem found that the site to which it was said the station had been moved was simply where the planning permission had required it to be in the first place. This explanation was not supported by any contemporaneous evidence, despite Ofgem’s request. On 9 July 2019, Ofgem wrote to AMP saying that it was suspending the issue of further ROCs in the light of the audit findings.

On 12 December 2019, it wrote to Mr Parker at GCM, saying that it was minded to withdraw the accreditation of Kelly Green, setting out its reasons, and giving GCM the chance to reply. The letter set out the categories of data, all dated 15 March 2016, supplied for the grant of accreditation to AMP. It described the investigations carried out by Ofgem before the grant of accreditation, and the replies which it accepted, followed by the audit report and GCM’s replies. Ofgem had then reviewed the material from which it noted that: the asserted relocation had not been mentioned in the accreditation process, nor was it supported by any evidence, or by any explanation for why the Kelly Green station had been constructed where it should not have been constructed, especially as the plan, showing where it should be, also showed the overhead power lines; the accreditation data showed that all inverters were generating electricity on or about 15 March 2016 as they should have been for commissioning, whereas in fact only 30% were generating electricity and the pattern of generation did not support a claim that they had been reduced as the station was moved around, and then resumed at 100%. A partly complete station could not be commissioned for the purposes of the ROO. There was no accurate or reliable data for a commissioning date, which Ofgem then considered was around 23 May 2016.

GCM’s solicitors, CMS, wrote to Ofgem on 17 February 2020 in reply, as did Mr Parker, who had become the superuser after the share purchase. In addition to raising an issue about the meaning of “commissioning”, which was not pursued in these proceedings, CMS contended that it was unfair to “impose” an audit, three years after commissioning. It was disproportionate to “strip” accreditation from a purchaser in good faith which had been entitled to rely on the accreditation after the satisfactory operation of the station for some years, and with ROCs being issued. GCM had bought “in reliance on the actions of Ofgem (in the accreditation of Kelly Green and having taken due care in granting the accreditation.)” Solarplicity could also have made a grace period application if doubts had been raised. In any event, Kelly Green should be accredited under the significant investment grace period, with a commissioning date of 23 May 2016, an accreditation which GCM would accept.

The letter dated 17 February 2020 from Mr Parker provided Ofgem with significant new information. GCM had been carrying out investigations following the letter of 12 December 2019 saying that Ofgem was minded to withdraw the accreditation. An email of July 2019 from an angry subcontractor to Solarplicity, refused Solarplicity’s request for a statement that the station was 100% complete at 14 March 2016. He had responded that such a statement would be fraudulent; “As you are well aware, on the 14^th March we were nowhere [sic] near the completion”, implying also that commissioning works were still being carried out in June 2016.

Mr Parker was at pains to emphasise that this was not known to him, GCM or GS1 when the share purchase from Solarplicity took place. Indeed, Ofgem only had this particular information because he had sent it to Ofgem, although as the superuser, he was bound to do so. But he recognised the consequences this had for the

“incompleteness and inaccuracy” of much of the important data about the commissioning date supplied by Solarplicity to Ofgem, and to GCM before the share purchase.

Further information known to Ofgem

After the commencement of these proceedings, GS1 sought further information from Ofgem about what information it had received from Solarplicity and third parties concerning the application for accreditation made by Solarplicity, and in particular about the commissioning date. The Claimants were contending that Ofgem ought to have been more astute to follow up indications that the commissioning evidence from Solarplicity was wrong; this was part of its argument that the consequences of the false commissioning date evidence should not be laid at GS1’s door by way of withdrawal of the accreditation, without even a grace period accreditation.

Mr Hargreaves’ witness statement, dated 16 December 2020, provided further detail of Ofgem’s actions following the whistle-blower’s allegations. The additional checks carried out had been met by responses from Solarplicity, via its consultant, which appeared credible. The accreditation had been put on hold while Ofgem audited two other Solarplicity stations about which the whistle-blower had raised concerns. Mr Elbourne of Solarplicity sent an email to Ofgem, of 12 December 2016, explaining that the commissioning tests for those two had not been completed by 31 March 2016, as intended; this had been only an indicative date, a not uncommon practice. But he maintained that commissioning for Kelly Green had in fact been on 15 March 2016. At a meeting between Ofgem and Solarplicity on 20 December 2016 concerning the audit of two other stations, Solarplicity accepted that they were commissioned during the grace period, but that was in contrast to its position over Kelly Green. Mr Hargreaves commented, that in the light of the absence of evidence to substantiate the whistle-blower’s allegations, and the thorough investigation that had been carried out, he was satisfied with the reliability of the information provided by Solarplicity on accreditation. He added “I was also of the view that for a sizeable and seemingly reputable company like Solarplicity to repeatedly lie in writing, verbally and face to face to Ofgem seemed unlikely.”

Mr Hargreaves’ evidence, which had responded to Mr Parker’s first witness statement, led to a further witness statement from Mr Parker criticising the inaction of Ofgem, as he saw it. Much of Mr Parker’s two witness statements comprises argument rather than evidence, and I deal with it when I consider submissions.

This provides:

“Every natural legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.

The preceding provisions shall not however in any way impair the right of a state to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties.”

The standing of GS1 as a victim within s7 HRA is at issue. Although, logically, consideration of whether it is a victim should come before a consideration of the proportionality of the decision, it is the proportionality of the decision which is at the heart of the case, and an undue focus on whether GS1 is a victim may distract from the real issue. This turns very much on the significance to be attributed to the fact that the new owner of AMP is a purchaser for value in good faith, and there is a new superuser. As I have said, it is not disputed for these purposes that “accreditation”, with the consequential right to receive ROCs, is a possession of AMP’s within A1P1, or that Ofgem’s revocation decision has deprived AMP of those possessions. And it has done so without compensation. AMP continues to own the Kelly Green generating station, and can sell the electricity to the market, but has lost the financial value which accreditation creates through ROCs.

Proportionality considers the striking of a fair and proportionate balance between the public interest being promoted and the other interests involved. The approach to be adopted to proportionality was summarised by Lord Sumption in Bank Mellat v HM Treasury (No.2) [2014] AC 700 at [20]:

“Their effect [earlier cases] can be sufficiently summarised for present purposes by saying that the question depends on an exacting analysis of the factual case advanced in defence of the measure, in order to determine (i) whether its objective is sufficiently important to justify the limitation of a fundamental right; (ii) whether it is rationally connected to the objective; (iii) whether a less intrusive measure could have been used; and (iv) whether, having regard to these matters and to the severity of the consequences, a fair balance has been struck between the rights of the individual and the interests of the community. These four requirements are logically separate, but in practice they inevitably overlap because the same facts are likely to be relevant to more than one of them. The question is whether a less intrusive measure could have been used without unacceptably compromising the objective. Lord Reed… takes a different view on the application of the test, but there is nothing in his formulation of the concept of proportionality (see his paras 68-76) which I would disagree with.”

Lord Reed formulated the fourth requirement in this way, at [74], though recognising that there was no difference in substance:

“whether, balancing the severity of the measure's effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter… In essence, the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure.”

The value in this case of what Lord Reed adds is the focus, not so much on the importance of the objective aimed at by the disputed measure, but the extent to which the disputed measure would contribute to its achievement.

It cannot be seriously disputed but that the revocation of accreditation was in accordance with the law. The provisions of Article 90 ROO were satisfied: the authority did have reason to believe that the information on which the decision to grant accreditation was based was incorrect in material particulars. It did consider that it was appropriate to withdraw accreditation. Grounds 2 and 3 of this challenge overlap with the A1P1 challenge: they allege that Ofgem failed to take necessary steps to acquaint itself with material facts, and fettered its discretion or closed its mind to the grant of a grace period accreditation. I deal with and reject those later; for the purposes of A1P1, the decision was in accordance with law.

It is also clear that the deprivation serves legitimate aims, that of protecting the integrity of the ROS, and protecting the interests of consumers. It does so through the prevention or deterrence of fraud in the obtaining of accreditation through the provision of what was known by the applicant to be significantly misleading information, and in removing the benefits of an accreditation which had been obtained in that way. That benefit is access, through ROCs, to a large subsidy from consumers. The context is that the relevant information is in the hands of the applicant; Ofgem cannot verify each component of the information supplied to it by those who seek a large public or consumer subsidy. The previous AMP superuser, Mr Elbourne, signed the standard declarations that “accurate, reliable and materially correct information” had been provided in respect of the accreditation application, and likewise annually was supplied in respect of entitlement to ROCs. That declaration was wrong, and known by Mr Elbourne and AMP at the time to be wrong.

The power to withdraw an accreditation, granted on the basis of incorrect information, is a necessary part of the ROS, sufficiently important to justify the withdrawal of accreditation. Otherwise, schemes which should not have received accreditation, or accreditation on the basis upon which it was applied for, will retain the benefit of inaccuracies and incompleteness, which may extend to fraud.

It is perfectly clear that Ofgem does have the aim of deterring fraud, and furthering its discovery. A suggestion that it did not was simply untenable. What was really meant was that it does not in fact have a policy of “zero tolerance” which was what the DL referred to. Mr Sinclair referred to the other two Solarplicity sites which Ofgem had investigated at the time of its investigation of Kelly Green in 2016. That is wrong on the facts. Ofgem found no fraud or materially misleading information; it accepted that “an indicative commissioning date” permissible practice at the time, had been given. It would be more troubling, if fraud were then involved, if Ofgem had not by 2020 tightened up its response, and sent that firm message to the industry about its intolerance of fraud, which was part of its justification for the withdrawal of accreditation.

It is the third and fourth steps which are at more serious issue, to which I now turn. The considerations overlap, as they do with the first two steps.

I was referred to the ECHR Guidance, as at 31 August 2020, on A1P1. It is not itself authority, but it does contain a summary, and an accurate one, of relevant ECtHR jurisprudence on this Article. I note: (1) there is no fixed list of factors relevant to the proportionality balance; (2) legal certainty and consistency in practice are relevant to the assessment of whether an individual has been required to bear an excessive burden, including whether the public authority acted in good time; (3) automatic and general measures can be disproportionate; (4) the availability of less intrusive measures does not necessarily make the action at issue disproportionate; (5) the conduct of the applicant is relevant, such as taking advantage of weakness by the authority or a loophole in the system, the applicant’s culpability or negligence, qualifications or personal vulnerability, and whether they could reasonably have been expected to be aware of the legal limitations on the property at issue. This includes whether the applicant had been involved in fraud.

There was some discussion about how the Court should approach its assessment of proportionality. The Court is not the primary decision-maker. Although it is for the Court to reach a judgment about whether an interference with possessions was proportionate, it has to accord respect to the decision of the relevant body, if that body has reached a decision on that point. This is not a case where the claimant has to show that the decision at issue was “manifestly without reasonable foundation”; the ROS comes from the legislature, but not this decision made by Ofgem under its statutory but discretionary powers. Here, the Court has to consider the decision of a statutory regulator in the application of the statutory scheme, where the decision is seen as having implications beyond the particular case. Ofgem has experience in dealing with the problems of fraud and misleading information for accreditation as present in this particular scheme, and will have a grasp of the issues and of the ramifications of its decisions on the withdrawal of accreditation and grace period accreditation. This matters because the proportionality issue revolves around how Ofgem approached an accreditation which it is justified in concluding was obtained by fraud, where the loss of accreditation is to be suffered, at least indirectly, by an innocent purchaser for value, and where it has to consider the effect on the ROS as a whole of allowing an accreditation obtained by fraud to continue, or to continue in alternative form which could accord to the purchaser all that a non-fraudulent application would have received, here, it is said, a grace period accreditation.

I cannot accept Mr Sinclair’s suggestion that the third and fourth steps in the proportionality assessment were not considered by Ofgem, or were considered in some rather superficial way in view of the brevity of the DL in dealing with them. It is plain that they were considered, and reasons for the decision were given in each respect. The reasoning has also to be seen in the light of the reasoning in the mindedto-revoke letter and the representations made on behalf of GS1 in response.

Mr Parker complained, in his first witness statement, that Ofgem failed to consider (i) the impact upon GS1, (ii) that it had not been fraudulent but rather had assisted Ofgem through the information it provided in response to the minded-to-revoke letter, (iii) the impact or absence of impact on the public purse, (iv) the compliance by GS1 with the ROS, (v) the fact that Kelly Green met the eligibility requirements for accreditation, (vi) the range of alternatives, and (vii) the reliance by GCM upon Ofgem’s accreditation in purchasing the shares in Solarplicity.

I have already set out parts of the evidence of Mr Sozansky. It also responds to the contention that various factors were not considered by Ofgem, a contention which he describes as “unfounded.” He said that he was well aware of the consequences of the withdrawal of accreditation for the operator. But he pointed out the obligation on Ofgem to ensure compliance with the ROS, under the ROO. To him it was clear that accreditation should be withdrawn, because the Kelly Green station ought never to have had the benefit of the ROS, as it was not commissioned before the scheme was closed to such stations. AMP, at all times the owner and operator, had knowingly supplied materially incorrect information. For it to remain accredited would “be contrary to the integrity of the RO Scheme in line with the legislation and its broader duties to consumers.” He acknowledged GS1’s position that it was an innocent victim, but even were that so, he thought the factors in favour of withdrawal outweighed that. I note the rather reserved language which he uses about GS1’s claim to be an innocent victim, but I am satisfied that there is no basis for approaching this claim with some sort of reservation about it. Mr Sozansky emphasised that accreditation could always be withdrawn if further information came to light. The accreditation letter made clear that accreditation was no guarantee that further action would not be taken to check the information provided in its support. GCM was thus on notice that an audit could look at that information, and AMP should have retained it for inspection by GCM. He saw no good reason why the ROS should be compromised by Ofgem:

“permitting accreditation to stations which had commissioned after the RO Scheme had closed (and on the basis of false information) in order to protect the interests of good faith purchasers of such a station. That is because a purchaser can be expected to protect itself by commercial measures such as due diligence and warranties or indemnities.”

He also referred to the absence of a provision in the statutory scheme to the effect that ultimate beneficial ownership of the shares of a company operating or owning a generating station should be material to a decision to withdraw accreditation granted on the basis of materially inaccurate information. The conditions attached to the accredited station, rather than to a particular legal entity. In any event, whatever the change in ownership of AMP, it was AMP which throughout had operated the Kelly Green station and the materially inaccurate information had been provided by or on its behalf.

Mr Sozansky then turned to his consideration of the alternative of granting accreditation based on a commissioning date after the closure of the scheme to small scale stations, under the grace period criteria. No formal application for a grace period commissioning had been made in accordance with the statutory scheme, supported by the necessary documentation. In any event, it would have been inappropriate to grant it in view of “the great lengths to which AMP…had gone to mislead Ofgem.” Transparency was required as to why an operator had been unable to meet the relevant deadline; it was not an appropriate mechanism to cater for the situation where a dishonest attempt had been made to hide the fact that a deadline had been missed, even where there had been a change in ownership of the operating company. Even by June 2020, Ofgem still did not know what the true commissioning date for Kelly Green was: evidence suggested that it had not been completed until June 2016, after what Ofgem initially suggested had been the commissioning date of 23 May 2016.

Ofgem had not concluded that there was a “quantifiable and specific cost to the consumer” if the accreditation were maintained. Its concern was that the financial benefits of the scheme, funded by consumers, should only go to those properly entitled to its support and not to those whose “purported entitlement was obtained by the knowing provision of false information.”

All this material is in my judgment an elaboration of the thinking in the DL, rather than significant new material. Thus, it was obvious that the revocation of accreditation would cause loss to GS1, it was accepted that it was not part of the fraud, and it was plain that Mr Parker had provided information adverse to his case and which Ofgem took into account. But, as superuser, he had been obliged to supply it once he found it. And without it, Ofgem still had a powerful case that incorrect information, which went well beyond mere inaccuracy, underlay the application for accreditation. It did not dispute that GS1 had complied with the ROS, but its decision shows that that was not significant.

The assertion that there would be no impact on the public purse if accreditation were allowed to continue, is not dealt with in those terms, but rather in terms of the integrity of the ROS and its dependence on consumers being required to pay more for electricity through charges to their bills. That issue, properly framed, was considered. If the point made by GS1 were sound, it would be sound for all stations which could have been accredited, even if the date for commissioning were false. It would be available to fraudulent applicants too.

The argument that GCM relied on the fact of accreditation, in its share purchase of Solarplicity companies, was in substance dealt with under the heading “Whether revoking accreditation would be a rational, proportionate and justifiable decision in the circumstances.” After all, the possession of accreditation is a pre-condition to its withdrawal and withdrawal after a period of operation is provided for in the statutory ROS, without time limit after accreditation beyond which withdrawal can no longer take place. GCM knew that. There is no provision preventing withdrawal of accreditation from a station once it has been purchased in good faith by someone other than the applicant for accreditation. There was no evidence of any reliance on the accreditation beyond the simple fact that Kelly Green had been accredited, which can go nowhere under the ROS, as it would be true of every withdrawal of an accreditation after a purchase in good faith. There was no evidence of any especial act of reliance by GCM, or of one conveyed to Ofgem. Ofgem was not said to have any representation made about the reliability of the information upon which accreditation had been based. GS1 was simply making a point available to any purchaser of an accredited station acting in good faith.

Less intrusive alternatives were considered but rejected. The principal contention was that Kelly Green met the requirements for commissioning at some point during the grace period and, by some mechanism, should be granted at least grace period accreditation. This was plainly considered. The various mechanisms suggested by the Claimants do not touch the basis upon which a grace period accreditation was rejected. It was not on the basis of some problem in the mechanisms for achieving such an outcome.

I consider therefore that a fair reading of the DL, with the documents referred to in it, shows that the essence of the points relied on by the Claimants in their case that the decision was disproportionate, were considered. All that entitles Ofgem’s decision to considerable weight.

I also accept Mr Grodzinski’s submission that, as the decision on proportionality falls to be made by the Court, it has to consider the material before it, which can include material which elaborates or explains the reasoning behind a decision, or even makes good omissions in it; R (Friends of Antique Cultural Treasures Ltd) v Secretary of State for the Environment [2020] EWCA Civ 649 [2020] 1WLR 3876 at [98-99]. Its primary focus is on the proportionality of the decision rather than on the quality of the decision, even where, as I consider the position is here, it is dealing with a decision which did address at least the significant proportionality issues, and which is entitled to considerable weight.

The Claimants’ case, at its highest, involves setting to one side the legal personality of AMP which was unchanged from the entity on whose behalf the application was made and the entity now operating the once and formerly accredited station. This presents the case on the basis that GS1 is the victim, and a purchaser for value in good faith without knowledge of the fraud of the applicant for accreditation, or that both together are the victims, and to be treated as a purchaser in good faith. It certainly involved ignoring the corporate structure and the past role of AMP.

I will deal with the case on that basis first and then consider the implication of the other factors, notably the unchanged corporate identity of AMP throughout the period from application for accreditation to date. I do so, not just because there is an issue of substance which arguments about corporate personality and victim status may obscure, but also because of the role of the superuser. This gradually emerged during the course of argument. But I am not sure that its significance has been fully examined. The functions of the superuser have to be performed by an individual human being and not by a corporate entity, so that personal liability can be fixed and is at the heart of the application process. The superuser at the time of the application, responsible for all the inaccuracies, was Mr Elbourne of Solarplicity. On

Mr Sinclair made the following points in submission, running AMP and GS1 together as one victim. First, he tackled the public interest points in this way, though there is an overlap with the private interests. Ofgem had identified two areas of the public interest which could be affected: the cost to the public purse, or more appropriately to consumers generally, and the aim of deterring fraud expressed as a “zero tolerance” approach. As to the first, no cost had been identified to the public or consumers generally, although there is a difference between the market price and the value attributed to ROCs, which is what this case is about, and which Mr Parker puts at between £4-6m. The value of electricity generated without accreditation would be approximately half that of accredited generation. He disputed, with Mr Parker’s evidence, what he described as the contention by Ofgem, that the public purse would suffer from either maintaining the accreditation or granting a grace period accreditation: the electricity would be generated as intended and the industry would be supported by the subsidy intended for it. I note that Ofgem does not say that 4-6m is the cost to consumers; no doubt it is very much more complicated than that. But there is a substantial subsidy, and it is met by consumers, not generators.

As to the second, Mr Sinclair submitted that this was “a one strike and out”, “one size fits all” policy, showing Ofgem’s mind to be closed to the specific facts of any individual case. This was not part of the statutory scheme, which gave a discretion to Ofgem as to whether to revoke an accreditation based upon incorrect information. The existence of a discretion did not mean that revocation was necessarily proportionate or fair. It could not necessarily be made fair by the possibility of a purchaser obtaining warranties or indemnities, which might prove to be unenforceable, especially where fraud was involved. Moreover, Ofgem had granted grace period accreditations to Solarplicity on at least two other sites in 2016 which showed that it had sought to persuade Ofgem that commissioning had occurred earlier than in fact it had. So, it was not averse to granting accreditations in such circumstances. Yet it had not investigated the Kelly Green application more closely.

In any event, the station had been commissioned during the grace period, and continued to generate electricity as envisaged at accreditation; the purpose of the ROS would be served by enabling such stations to receive the subsidy which led to their creation. There was no gain to the consumer if its accreditation were modified, with appropriate conditions, so as to treat is as an accreditation granted during the grace period. This would lead to no unexpected cost to the consumer, to no extra subsidy, and no gain to the Claimants from the inaccurate commissioning date information. On the contrary, Ofgem’s decision would undermine support for the generation of renewable electricity; it had a duty to protect the interests of consumers under S3A Electricity Act 1989.

Mr Sinclair then tackled the impact on private interests. First, he pointed to the loss which GCM/AMP would suffer from the loss of the subsidised purchase price for the electricity generated from Kelly Green for which no compensation would be paid. There was no provision for compensation in the statutory scheme, which could be seen as requiring great care over depriving such a purchaser of its possessions.

Second, he submitted that real significance was rightly attributed by GCM to the fact of accreditation by Ofgem, and to the ability of Ofgem, using statutory powers, to obtain information from an applicant for accreditation, which a purchaser could not do. GCM and GS1 had been unaware of the investigations carried out by Ofgem in relation to Kelly Green and the more extensive investigations Ofgem carried out into two other sites, until after the share purchase, and in part, not until it received Mr

Third, GCM/GS1 had carried out an industry standard due diligence process before the share purchase. There was a limit, by duration and value, in obtaining more extensive warranties then they had done from Solarplicity, and especially by comparison with the investigatory powers which Ofgem had on accreditation. These would be of limited value where the vendor engaged in fraud both against Ofgem, and subsequently against the share purchaser. Mr Parker did say that Solarplicity had provided warranties in relation to accreditation, and to the effect that it had no knowledge of any matters that would give rise to its revocation. Mr Grodzinski pointed out that Mr Parker did not set out the terms of any warranties, nor describe any action taken under them against Solarplicity, or on some other basis, in relation to the misrepresentations which must have been made on the share purchase.

Fourth, it was relevant, in considering the position of AMP, and the change of ownership, to recognise the change in superuser, brought about upon the share purchase, from Mr Elbourne to Mr Parker. The role of the superuser, in making applications for accreditation and taking responsibility for the accuracy of the information, was important when considering whether it was right for responsibility for the inaccurate information simply to be laid at the door of AMP and whoever happened to own it. By way of underlining this point, Mr Parker listed many references in the evidence of Mr Sozansky and Mr Hargreaves referring to applications made, or actions undertaken, “by” Solarplicity. This he said ignored the essential role of the superuser, whose identity had changed upon the share purchase.

Fifth, Mr Sinclair contended that Ofgem, first at a meeting in March 2019, had seemingly been willing to consider a grace period accreditation, as he contended had happened in other cases, including two Solarplicity sites where Ofgem had suspected the provision of deliberately inaccurate accreditation information. But Ofgem had changed its mind by the time of the DL. He said that all the material necessary for such an application was available; he had been expecting a request for it after Ofgem’s consideration of the responses to the minded to revoke letter, but no request had come. Such an application could not just be made “a propos of nothing”, as Mr Parker put it, in these circumstances. Ofgem now said that it would not in any event grant one, and that it had no such application with the necessary evidence, supporting a commissioning date in May 2016 or during the grace period. Reliable information as to a commissioning date was no longer possible, said Mr Sozansky in his witness statement.

Therefore, submitted Mr Sinclair, the Claimants should be treated as innocent purchasers for value of the Kelly Green station. Withdrawal of accreditation of the station from such owners, without compensation, was not proportionate. It was unnecessary to uphold the integrity of the scheme by deprivation in such circumstances since those whom the withdrawal would affect adversely were innocent of the fraud or inaccurate information which led to accreditation. GS1, and GCM, had relied on the fact of commissioning and the lapse of time thereafter, without withdrawal of accreditation, in purchasing the station. GCM had done what it was reasonable to do by way of due diligence before purchase, and had obtained, unspecified, warranties and indemnities.

Mr Grodzinski submitted that the DL should be given great weight, and that it, with the evidence of Mr Sozansky and Mr Hargreaves, demonstrated that the withdrawal was a proportionate decision.

I turn to my analysis. I regard it as obvious that, if the accreditation had been obtained by deliberately misleading information, and that the person who committed the fraud still operated the station, there could be no successful challenge to the withdrawal of the accreditation, and no conditioning of it to create a grace period accreditation. Mr Sinclair jibbed at that a little, but not so as to create a doubt in my mind.

There is no UK authority on how, under A1P1, the Court should approach the position of the purchaser in good faith when the possession at issue has been acquired by bad faith on the part of the predecessor in title. Mr Sinclair relied on the ECtHR decision in Beyeler v Italy [2000] I WLUK 51, (2001) 33 EHRR 52. Mr Beyeler bought a valuable painting in 1977, which he ought to have declared to the Italian authorities, which had a statutory right of pre-emption. He did not declare himself to be the purchaser to the authorities until 1983, when he intended to sell it. It was not until 1988 that the relevant Italian ministry stated that it was interested in buying the painting instead of his intended purchaser; Mr Beyeler said that he was willing to sell to the ministry, but it failed to respond in time and he sold the painting, notifying the ministry that he had done so. Six months later, the ministry exercised its right of pre-emption in respect of the 1977 sale, paying only the 1977 sale price, which was well below the current market value. It argued that the loss was caused by Mr Beyeler’s failure to notify the authorities that he was the purchaser in 1977. It won in Italy and lost in Strasbourg.

The case is relevant for what the ECtHR said about the conduct of the parties. Between 1977 and 1983, Mr Beyeler had deliberately avoided a pre-emption order by failing to comply with the notification requirements of Italian law. The fact that the applicant “had not acted openly and honestly carries some weight, especially as there was nothing to prevent him from informing the authorities of the true position before 2 December 1983 in order to comply with the statutory requirements.” However the Italian authorities’ attitude towards the applicant after 1983 had “oscillated between ambivalence and assent and they often treated him de facto as the legitimate title holder under the 1977 sale.” There was no convincing explanation as to why the Italian authorities had not acted in 1984 as they acted in 1988; taking punitive action at that stage on the ground of an incomplete declaration of which they had been aware for nearly five years “hardly seems justified.” The delay had allowed the ministry to acquire the painting in 1988 at well below its market value. It had derived an unjust enrichment from uncertainty during the period after 1983 to which it had largely contributed. that was incompatible with the requirement of a ‘fair balance’”. The applicant had thus borne a disproportionate and excessive burden.

While this case makes good Mr Sinclair’s submission, and authority is scarcely required, that the conduct of the parties is relevant to the proportionality assessment, it does not advance his case on the position of a purchaser for value in good faith. Mr Beyeler’s breach of Italian law would have been held against him had the Italian authorities acted in 1984 as they eventually did in 1988. The crucial point was not the lapse of time after the sale in 1977, but the lapse of time after they found out about the sale, which was in 1983. I see no parallel helpful to Mr Sinclair in his contention that Ofgem had the opportunity to investigate Solarplicity’s applications in 2016, but had failed to do so. Indeed, I do not consider that the close examination to which he subjected the detail of Ofgem’s investigation of the other Solarplicity sites served to show that there had been a failing by Ofgem here, to which the Claimants could point, similar to the failings of the Italian authorities in Beyeler.

Such UK authorities as there are on the position of the innocent third party caught up in the failings of others are referred to in R(Global Knafaim Leasing Ltd and CGTSN Ltd) v Civil Aviation Authority [2010] EWHC (Admin) 1348, Collins J. This rather striking case concerned statutory fleet lien powers. An aircraft leasing company, and the owner of a specific aircraft, leased it to what became an insolvent airline. The aircraft was held at a UK airport. It would only be released if the lessors paid BAA and CAA charges, unpaid and due to those bodies, in respect of not just the claimants’ aircraft, but also in respect of all other aircraft operated by the insolvent airline. It was argued among other matters that this was a disproportionate deprivation of property, especially as the claimants had not been aware of the parlous financial position of the aircraft lessee. In effect, they, and not the airport and CAA had to make good the failings of their lessee, extending well beyond those incurred by the lessors’ aircraft. This contention was rejected; and other cases, some relating to control of use rather than deprivation of property were considered, which makes it a useful reference point, and not an encouraging one from Mr Sinclair’s perspective. Collins J said this:

“67.

Once it is accepted that the power is lawfully there, the exercise of it will equally be lawful unless in a given set of circumstances its exercise would be unfair and disproportionate. The defendants submit that the inability of the operator to pay is precisely the situation in which the fleet lien power leading to payment by a lessor is needed. Thus such circumstances as arose in this case cannot be regarded as exceptional or to take this case out of those in which use of the power is permissible. In my view exceptional circumstances are needed. One can well think of possibilities. Examples could include the misleading of the lessor by a possessor of the power or an indication given to a lessor that the power was not to be exercised against his aircraft or there was a deliberate targeting for no good reason of a particular lessor's aircraft. These will depend on the individual facts of a given case. This case, hard though it is on the claimants, has nothing to take it out of the circumstances in which the power can properly be exercised.”

The operator, innocently caught up, had to pay. Collins J considered the competing arguments about whether the claimants or defendants could or should have taken steps to cope with the potential insolvency, and to mitigate the harshness of the outcome for the Claimants. He said this at [56]:

“In the present case, there were steps which the claimants could have taken to provide themselves with more information about Zoom’s financial position. In addition, they could…have forbidden Zoom to fly their aircraft to the U.K. until its financial situation improved or have required (if that was feasible) further security. It is also noteworthy that in the AGOSI case the Government conceded that as a practical matter where a person was free of any fault which could relate in any way to the purpose of the legislation forfeiture could not on any sensible construction of the legislation further its object. One element of the circumstances was the degree of fault or care of the applicant. The court recognised that relevant considerations “included the alleged innocence and diligence of the owner of the forfeited coins” (Paragraph 56). So here, the conduct of the claimants and their failure to make use of their powers to obtain material information or to take steps to avoid the application of the aspects of the fleet lien of which they complain to them is relevant. AGOSI is a strong case in favour of the defendants’ submissions.”

Mr Sinclair also referred to “Bulves” AD v Bulgaria [2009] ECHR 143. The Court said at [70-71] and care needs to be taken to put phrases in context:

“70.

Lastly, as regards efforts to curb fraudulent abuse of the VAT system of taxation, the Court accepts that when Contracting States possess information of such abuse by a specific individual or entity, they may take appropriate measures to prevent, stop or punish it. However, it considers that if the national authorities, in the absence of any indication of direct involvement by an individual or entity in fraudulent abuse of a VAT chain of supply, or knowledge thereof, nevertheless penalise the fully compliant recipient of a VATtaxable supply for the actions or inactions of a supplier over which it has no control and in relation to which it has no means of monitoring or securing compliance, they are going beyond what is reasonable and are upsetting the fair balance that must be maintained between the demands of the general interest of the community and the requirements of the protection of the right of property (see, mutatis mutandis, Intersplav, cited above, § 38).

4.

Conclusion

Considering the timely and full discharge by the applicant company of its VAT reporting obligations, its inability to secure compliance by its supplier with its VAT reporting obligations and the fact that there was no fraud in relation to the VAT system of which the applicant company had knowledge or the means to obtain such knowledge, the Court finds that the latter should not have been required to bear the full consequences of its supplier's failure to discharge its VAT reporting obligations in timely fashion, by being refused the right to deduct the input VAT and, as a result, being ordered to pay the VAT a second time, plus interest.”

I do not consider that to be a comparable situation given the knowledge of the applicant for accreditation, the potential for knowledge to be obtained during the share purchase, and for warranties to be obtained. Nor is it comparable in its outcome to what Mr Sinclair contended for here, whereby the innocent party asserts an entitlement to a continuing benefit created by the fraud of another.

I give considerable weight to the decision of Ofgem and to the considerations which it set out, in elaboration in the evidence of Mr Sozansky. I agree with it, and I consider that the decision was not disproportionate, for the reasons those documents provide. The position of Ofgem as recipient of information from applicants for accreditation, much of which it has to take on trust, and cannot independently check without significant investigation is important, especially when it is the provider of that information which is seeking to access very significant sums of money from the public. The declarations made have to be taken seriously. The integrity of the ROS, and public acceptance of the cost, mean that fraud cannot be tolerated.

I accept that, without revocation, the station would be generating electricity as intended, and would be supported by ROCs as intended by the ROS, but this is true only if the station would have been accredited. It would not have been accredited on the basis of the information supplied. The fraud cannot be treated as a mere, albeit regrettable, incident of accreditation.

That leads on to the contention that Kelly Green should have been treated as commissioned in May or June 2016, and accreditation treated as granted, by some mechanism, within the grace period, as a less intrusive measure. This is particularly where the Claimants rely on the fact that the purchase was made in good faith, as making the crucial difference, and on the change in superuser, as justifying at least a grace period accreditation. Ofgem makes two points, which in my view dispose of that argument. First, although it had thought that commissioning took place in May or June 2016, it has not had the commissioning data necessary to determine the date, and it has never had an accurate and honest version of the accreditation documents and data for such a date. It is entitled to expect that to be supplied by someone seeking a grace period accreditation and especially in these circumstances.

Second, and in any event, the grant of a grace period accreditation on the back of a dishonest application made before the small scheme was closed, and its accreditation properly withdrawn, does not treat the fraud with the gravity it deserves. It merely

I also accept Mr Grodzinski’s submission that the alternative measures, essentially, by one means or another arriving at the equivalent of a grace period accreditation, are not less intrusive measures to the same end. They are less intrusive measures which do not reach the same end in relation to accreditations obtained by fraud. Worse, they run in part at least counter to the aims of upholding the integrity of the ROS and deterring fraud.

I do not consider that other factors prayed in aid by the Claimants have the force necessary to counter Ofgem’s justification for withdrawal of accreditation without any replacement. First, I do not think that the absence of compensation is of any significance in this context. It would undermine the integrity of the ROS as effectively as not withdrawing accreditation at all. Second, the Claimants have no basis for claiming reliance on the fact of accreditation and the passage of time. The statutory scheme is quite clear; there is no time limit on the withdrawal of accreditation, and operators know, and accept in the conditions on the accreditation, that inspection or audit can take place at any time. That should be factored into any pre-purchase enquiries, warranties and valuation. There is no suggestion that GCM made any special enquiries of Ofgem about the accreditation or told Ofgem that reliance was being placed on its accreditation for its purchase, its terms or its valuation. I am sure that GCM would have been disabused of its hopes, and I doubt that it would have then acted differently in any which way. It was an entirely internal source of assurance, if anything.

Third, although Ofgem has statutory powers of enquiry and investigation which a purchaser lacks, a purchaser can make whatever enquiries it wants and seek whatever it wants from the seller. If the seller baulks at its provision, or if the purchaser is not satisfied with the answers, it can walk away, lower the price, or seek assurances, warranties, or reserve parts of the purchase price against years of ROC earnings and so on. It simply has to decide whether the purchase is worth the risk. It cannot make Ofgem responsible for the risks it decides to undertake. The Claimants decided to undertake what they call industry standard due diligence. No doubt it did not suspect fraud, and though there is no evidence as to the state of GCM’s knowledge one way or the other, I would be surprised if the likes of GCM were unaware of a scramble for accreditation as the small station scheme came to a close, with the risk that some were misleading on dates of commissioning. But I do not see that as a basis upon which the Claimants should be safeguarded from fraud by a relaxed attitude on the part of Ofgem, supposedly protective of the interests of consumers, but in fact protecting the generator.

Mr Grodzinski emphasised that GS1 had not provided the terms of the warranties it obtained. Indeed, the nature of the investigations it carried out are only disclosed in the most general way. (I did not find Mr Sinclair’s references to the preliminary issue judgment in Breyer Group plc v Department of Energy and Climate Change [2014] EWHC (QB) 2257, of any relevance to what I have to decide.) Warranties may not always be of practical value; those who would be liable may be made of straw; but that risk would have been the same for any warranty given to GCM. GS1 has given no information about any action it as undertaken or proposes to undertake against Solarplicity group companies or its directors. If GS1 is waiting to see how this claim fares, that rather highlights the justification for the thinking behind the choice which Ofgem resolved against it.

Fourth, I do not think that any suggestion that Ofgem were somehow on notice of Solarplicity’s fraud, and that that adds to the balance in favour of GS1, is wellfounded. Ofgem was not on notice of fraud by Solarplicity in relation to Kelly Green; it examined the application more closely following the whistle-blower’s allegations. It was satisfied by Solarplicity’s explanations. Mr Sinclair’s submissions require a critique of that investigation and a view to be formed about whether a site visit should have been undertaken, all to show that GS1 should retain an accreditation obtained by the fraud of another, or have it changed to what the fraudster might have been entitled to had it been honest, whilst absolving itself of responsibility for its own acts and omissions.

Ofgem examined more closely applications by Solarplicity in relation to two other generating stations, but was satisfied by Solarplicity’s explanation about the indicative nature of the commissioning dates. The fact that fraud may be difficult to uncover does not mean that Ofgem owes a purchaser the responsibility for uncovering it, and has to accept the consequences of fraud, in effect bearing the purchaser’s responsibility for investigating the purchase and protecting itself, or protecting it from the consequences of a fraud which neither had uncovered.

Mr Parker, responding in detail in his second witness statement, dated 11 January 2021, 15 days before the hearing, to information in the exhibits to Mr Hargreaves’ witness statement, said that Ofgem were on notice about a further 3,315 cases of fraud by Solarplicity, allegations contested with factual material in Mr Grodzinski’s Skeleton Argument. This cannot sensibly be taken into account at this stage and would require a significant and costly diversion on a point of rather limited value.

Accordingly, taking the Claimants’ case at its highest, I find against it.

That means that, when the position is considered without GS1, and on the basis that there is only the one victim, AMP, the claim is hopeless. That asserted Claimant victim is the same corporate entity as the fraudulent applicant. It cannot say that the fraudster was a director, or the superuser, of which it was the corporate victim. It is

For completeness, I will deal with whether GS1 is a “victim”. First, the position of AMP as a special purpose vehicle does not entitle it to any special regard in this respect. Mr Grodzinski cited a passage from the judgment of Chadwick LJ, with whom Moore-Bick LJ and Sir Christopher Staughton agreed, in Wood v Holden (Inspector of Taxes) [2006] EWCA Civ 26, [2006] 1WLR 1393, at [25] approving Park J at first instance:

“" . . . It is also necessary to keep in mind that, while the cases which I have referred to so far all involved the residence of companies with active continuing businesses, it is possible (and is common in modern international finance and commerce) for a company to be established which may have limited functions to perform, sometimes being functions which do not require the company to remain in existence for long. Such companies are sometimes referred to as vehicle companies or SPVs (special purpose vehicles). 'Vehicle' has a belittling sound to it, but such companies exist. They can and do fulfil important functions within international groups, and they are principals, not mere nominees or agents, in whatever roles they are established to undertake. They usually have board meetings in the jurisdictions in which they are believed to be resident, but the meetings may not be frequent or lengthy. The reason why not is that in many cases the things which such companies do, though important, tend not to involve much positive outward activity. So the companies do not need frequent and lengthy board meetings."

The question of whether a victim a company can be, as the share-holder of another company, has been considered by the ECtHR in a number of A1P1 cases. The general principle is that the distinction between a corporate entity and its individual shareholders is to be maintained. The most recent Strasbourg case is Albert v Hungary ECtHR [2020] ECHR 5294. At [119-145], it set out “General principles regarding victim status of shareholders of a company”, drawing upon its earlier judgments. It identified two groups of cases:

“123.

In the former group, shareholders themselves may be considered victims within the meaning of Article 34 of the Convention. In such cases the difference between the rights of the company and the rights of the shareholders is maintained and the company's legal personality remains intact, as the complaints and the Court's substantive analysis concern the rights and the situation of the company's shareholders and not those of the company…

124.

In the latter group the general principle is that shareholders of companies cannot be seen as victims, within the meaning of Article 34 of the Convention, of acts and measures affecting their companies. The Court has recognised that this principle may be justifiably qualified in two kinds of situations, firstly, where the company and its shareholders are so closely identified with each other that it is artificial to distinguish between the two...” [The second is immaterial].

This case does not concern those cases where a distinction is properly to be drawn between acts affecting the company as such and acts affecting the rights of shareholders as such. It is the second group of cases which matters. The ECtHR turned to them, under the heading “Examination of cases where the company and its shareholders are so closely identified with each other that it is artificial to distinguish between them”. That is the basis of Mr Sinclair’s contention that GS1 is a victim, as well as AMP, because it, as shareholder, and AMP itself are “so closely identified with each other that it is artificial to distinguish between them”. The ECtHR said at [135-137]:

“135.

Although companies with a separate legal personality are not normally to be identified with their shareholders, in some of its previous cases the Court has accepted that there are situations where it would “serve no purpose to distinguish between the two” … and has allowed the shareholders to proceed with their complaints about the proceedings or events affecting their companies. For instance, in the Pine Valley Developments Ltd and Others judgment (29 November 1991, … where the third applicant (Mr Healy) was the sole shareholder of the second applicant (Healy Holdings) which wholly owned the first applicant (Pine Valley), the Court noted that “... Pine Valley and Healy Holdings were no more than vehicles through which Mr. Healy proposed to implement the development ...”.

136.

The Court underlines that the reason for accepting victim status in such cases is that there is “no risk of differences of opinion among shareholders or between shareholders and a board of directors as to the reality of infringement of Convention rights or to the most appropriate way of reacting to such an infringement” (see Ankarcrona v. Sweden …, ECHR 2000-VI).

137.

This group has included cases brought by shareholders of small or family-owned or family-run companies or cooperatives, notably where a sole owner of a company has complained about the measures taken in respect of his or her company (see, as early examples, Yarrow and Others v. the United Kingdom, no. 9266/81, Commission decision of

18 January 1983, D.R. 30, p. 155, where the victim status of the

first applicant, who was the sole shareholder, was accepted without discussion of the issue, and Dyrwold v. Sweden, … Commission decision of 7 September 1990; ….”

The Court then considered cases involving “exceptional circumstances precluding the affected companies from bringing the cases to the Court in their own name.” That does not arise here.

This ECtHR decision is entirely in line with the Court of Appeal judgment in Bank Mellat v HM Treasury (No.5) [2017] QB 67, where Lord Thomas LCJ said:

“28.

In my view the general principle applied by the Strasbourg court is clear. Save in exceptional circumstances, it is the company and not its shareholders who have the status and standing as a victim to bring the claim for the loss sustained by the company ….”

I see no value in a prolonged analysis of earlier Strasbourg cases. Mr Sinclair’s strongest case was Pine Valley Developments Ltd and Others v Ireland 1987. Mr Healy controlled the two relevant companies, which were in administration. They were seen as merely vehicles for his activities, and their financial position and the position of the receivers, did not require consideration of the detailed submissions which the Court would otherwise have had to engage with. The reason given in Albert v Hungary for the Strasbourg Court’s approach is clear: there is no risk of differences of opinion between shareholders or between shareholders and a board of directors as to whether rights have been infringed or in the response to the infringement alleged.

The cases in which that has arisen have involved a personal shareholder and not a single corporate shareholder. I very much doubt that Strasbourg would see the ownership of AMP by GS1 as having that close relationship in which distinction was artificial. After all, its starting point is that the distinction between the corporate entities has to be respected, and have legal consequences. I am not certain however that a personal and corporate relationship, rather than corporate to corporate relationship, is the exclusive basis upon which its decisions have been arrived at, although the point is far more readily seen in such cases.

The real problem with its application here is in the relationship between the two companies. The contention has to be, not just that they have the same interest, but that they are “so closely identified with each other that the distinction between them is artificial.” I accept that they both want to maintain accreditation or obtain a grace period accreditation and that there is no conflict between their shareholders or shareholder and boards. But if the two companies are so closely identified with each other that distinction is artificial, the role of the corporate entity, AMP, has to be identified with GS1. That enmeshes GS1 with the fraudulent AMP. AMP have not been left out of the claim. GS1 then has to disavow the corporate actions of AMP, since it is a new owner. But that is to assert a distinction of crucial significance between the two corporate entities. GS1 has been brought in to enable the sins of AMP to be overlooked in favour of the interests of its shareholder. The shareholder claims for its reflective loss as shareholder, and does so for a claim for the same injury which would be denied to the company it owns, AMP, and with which it asserts distinction to be artificial. The need for GS1 to participate disproves the contention it has to make in order to do so. It cannot be joined successfully to evade the consequences of the corporate failings of AMP, which mean that, victim though AMP is, it is entirely the author of its own, and hence of GS1’s misfortunes.

Ground 1 is dismissed.

Mr Sinclair raised eight points here, presented as six: (i) Ofgem had failed to find out the extent to which Gravis or Solarplicity would suffer the detrimental effect from its decision; (ii) it failed to analyse whether sending a message to the market about fraud would work or harm the market through creating uncertainty about the durability of accreditations; (iii) it had thought that withdrawing accreditation would save the consumer money, but later in its case had accepted that it would not do so but would simply mean that the subsidy was not paid; (iv) Ofgem did not find out what GCM/GS1 did or say or what else they should have done to ascertain the true position about the accreditation at the time of the share purchase, especially in the light of the failure of its own investigations in 2016; (v) it did not take reasonable steps to investigate the suitability of a grace period accreditation; (vi) it failed to take into account GS1’s full co-operation with the investigation in 2019-2020; (vii) it failed to take into account the effect of its accrediting Kelly Green on the purchaser, which made the share purchase, it is asserted, in reliance on the accreditation; (viii) there would be an adverse effect on the emission of greenhouse gases.

As Mr Sinclair accepted, there is a considerable overlap with points already considered under ground 1. I accept Mr Grodzinski’s submission that not all considerations which are material may be, in the sense of not being irrelevant, have to be considered, and that not all that are considered need to appear in a decision letter. None of the points now raised are statutory or mandatory considerations. All those raised by GS1 in its response to the minded-to-revoke letter were considered. The duty of enquiry only covers those factors which any reasonable authority in the position of Ofgem would ascertain before reaching a decision.

I deal with Mr Sinclair’s points briefly. (i) Ofgem took it that the loss would fall on one or other of the Claimants in the first instance, but it did and does not know whether they would be able to recover anything from Solarplicity. It did and does not know and does not need to know the extent to which any loss could be recovered. GS1 has not said either. (ii) Ofgem was in a position to reach this judgment and did not need to make or reveal further enquiries; it is obvious as well that the market would understand that fraud in accreditation would lead to withdrawal, and uncertainty would be felt in consequence by those whose accreditations were founded on fraud, or did not know that they were not. (iii) Ofgem’s position on the effect of fraudulent accreditations on the consumer did not change and cannot be regarded as ignoring a material consideration. (iv) What GS1 did or did not do over the share purchase is not a matter for Ofgem to investigate; it considered what the Claimants told it, and what they did not say. (v) This was not for Ofgem to investigate; if GS1 wanted a grace period accreditation directly or by means of conditions, it was for GS1 to supply the material which accreditation required. In any event, Ofgem decided that that was not a course of action it was prepared to pursue. (vi) Ofgem obviously took account of the good faith of GS1; part of that involved the disclosure by Mr Parker, but also that he was bound to make them, as superuser. (vii) Ofgem would have known that GS1 would rely on the fact of the accreditation in its purchase, in the sense that accreditation was a necessary part of the transaction, as it would be of all purchases of accredited stations; but that would have gone along with GS1’s knowledge of Ofgem’s discretionary power to withdraw accreditation when it discovered circumstances such as those which it found out had occurred. There was no special reliance on the fact of accreditation, at least notified to Ofgem. It is difficult to see how the mere fact of accreditation can be a factor material to the decision to withdraw it, beyond that it exists to be withdrawn. (viii) Ofgem was well aware that the purpose of the ROS was to encourage the generation and use of renewable electricity. The effect on greenhouse gases would only arise if the station ceased to generate or the electricity were not used. This was not an issue raised in the Claimants’ representation to Ofgem. It is not a mandatory or rationally necessary consideration. And it is a consequence inherent in the power to withdraw accreditation, assumed in the existence of the power, available to be exercised in the circumstances which had arisen to protect the integrity of the ROS.

There is nothing in ground 2.

Mr Sinclair criticised Ofgem for closing its mind to any alternative to the total withdrawal of accreditation. (i) It was absurd to criticise the absence of an application for grace period accreditation as the time for making such an application had expired. Ofgem was wrong to criticise the absence of data to support a grace period accreditation as Ofgem had not asked for it. (ii) Ofgem suggested that commissioning might have occurred in May 2016, but later said that it had not taken a view as to when Kelly Green was commissioned. It could not therefore have examined a grace period accreditation with an open mind. (iii) A zero tolerance approach to fraud also meant that no alternative to total withdrawal of accreditation existed once Solarplicity’s actions were described as fraud.

These points have been covered essentially under ground 1. Ofgem did not approach the question of a grace period with a closed mind; it is clear that it gave it active consideration. It concluded that it should exercise its discretionary power to withdraw accreditation, and not to accede to requests for conditions, or the like, to create the effect of a grace period, because it considered the application by AMP to have been fraudulent, and that such an application should not be allowed to turn into an effective albeit less valuable grace period accreditation, as if that had been applied for in the first place. The zero tolerance approach was then considered appropriate not just for the original fraudster but also to a purchaser in good faith, for the reasons which it gave. That was not a closed mind, but a mind exercising its discretionary power, where it had found fraud, and a purchaser in good faith. I do not accept Mr Grodzinski’s submission that “zero tolerance” means only that great weight will be given to fraud. But it is difficult to see that the approach is unlawful anyway. If unlawful, however, I cannot see that the factors in favour of the Claimants could outweigh the great weight which on any view Ofgem would rightly give to fraud, for the reasons which it deployed. So, if error, the error, on one aspect, could make no difference to the outcome. Nor could it affect the assessment of a fair balance.

I do not think that the Claimants are right to treat as absurd the comments by Ofgem about the absence of an application for a grace period application, when the time for such an application had passed. It was factually correct, although an honest application could have been made at an earlier stage, albeit not after GS1’s purchase. So GS1 was seeking an indulgence in the form of mechanisms to achieve the same effect, as it could not comply with the requirements for what it actually sought.

I disagree with Mr Sinclair’s contention that it showed a closed mind for Ofgem not to ask for the data showing a commissioning date as if for a grace period accreditation. GS1 was not entitled to anything, but at the very least, to make its plea good, it had to supply the data as if for a grace period application. I can see no ground for any complaint about Ofgem’s reaction. The evidence for a grace period accreditation has not been supplied. If it cannot now be supplied, it is hard to see why that becomes a basis upon which an accreditation should be granted on some weaker material. I see no reason for Ofgem to have to stick to indications that commissioning may have occurred in May or indeed in June 2016. It was entitled to require proof from GS1 of what Solarplicity had not provided, the date of commissioning. Mr Sozansky says that Ofgem does not know that date, which I accept.

There is nothing in ground 3.

This claim is dismissed.