England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
MR JUSTICE GREEN
[2014] EWHC 3677 (Admin)

		Royal Courts of Justice Strand, London, WC2A 2LL
		01/03/2016

Lord Justice Floyd:

1. The issue in this judicial review appeal concerns the legality of decisions by the respondent, the Secretary of State for Energy and Climate Change ("the SoS"), to bring to a premature close, subject to certain periods of grace, a statutory scheme supporting the generation of electricity from renewable sources. The appellants, who were claimants in the court below, are four companies engaged, amongst other things, in the installation of large-scale solar photovoltaic systems or "solar farms". In his judgment of 7 November 2014 Green J, at a rolled-up hearing, gave permission to seek judicial review but then dismissed the application. The appellants appeal to this court with the permission of Bean LJ, renewing the arguments advanced before the judge.
2. I will first set out the structure of the renewables obligation scheme ("the RO scheme"), the Contracts for Difference ("CfD") scheme which is intended to replace it, and the Levy Control Framework ("LCF"), a policy which plays an important part in the arguments. I will then set out the main historical developments which led to the decisions under attack.
3. On the appeal Mr Michael Fordham QC and Mr Simon Murray represented the appellants. Mr Robert Palmer represented the SoS. The court is extremely grateful to them for the clear and economical way in which they presented their arguments.

The renewables obligation scheme

4. The judge described the RO scheme in the following way:

"(iii) The RO scheme in a nutshell

15. The RO scheme provides financial incentives for the creation of generation capacity from renewable sources. It does this in the form of Renewables Obligation Certificates ("ROC"). These are certificates issued to operators of accredited renewable generating stations in respect of the eligible renewable electricity that they generate. Under the RO scheme licensed electricity suppliers are required to submit a specified number of ROCs to the Gas and Electricity Market Authority ("Ofgem") for each megawatt of electricity that they supply during the period from 1st April to 31st March or, alternatively, pay a penalty. The RO scheme has been the main support mechanism used by the Government for large scale renewable electricity since 2002. …

(v) Eligibility for ROCs

17. To be eligible within the RO scheme electricity must be generated by a station that has been accredited by Ofgem as capable of generating renewable electricity in respect of which ROCs may be issued. The RO scheme is hence a demand led scheme with all generation satisfying the relevant conditions being accredited. For solar PV it is a relevant condition that the generating station must exceed 50 kilowatts in size. Generators are issued with ROCs by Ofgem according to the amount of eligible renewable electricity generated. They may sell their ROCs to suppliers, which allows them to receive a premium in addition to the wholesale electricity price. The price of a ROC is determined by the market, there is no fixed rate. The number of ROCs that an electricity supplier is required to submit for each megawatt of electricity they supply is known as the "Renewables Obligation" or "RO". The level of this is determined by the Secretary of State by 1st October in each year preceding the start of each obligation period.

18. The Secretary of State is required to calculate the level of the obligation in accordance with a formula set out in subordinate legislation. This includes a requirement to estimate the amount of renewable electricity likely to be supplied during the relevant obligation period and having regard to that estimate to calculate the number of ROCs likely to be issued."

5. Thus the RO scheme is funded by a levy on electricity suppliers. It is assumed that the suppliers will pass on these extra costs to electricity customers. Accordingly the levy operates analogously to a tax, just as if the Government has taxed electricity users and paid the sums raised to accredited generators under the RO scheme. It is "demand-led" in the sense that the cost of the levy is heavily dependent on the extent of uptake under the scheme.
6. Since 2009 the RO scheme has provided different levels of support for different technologies according to their "band". The different levels of support reflect different costs associated with the respective technologies. Article 33 of the Renewables Obligation Order 2009 empowers the SoS to conduct a review of banding levels every four years, although he may do so at any time if any of the conditions in Article 33(3) are satisfied. Those conditions include where an event has occurred which is was not foreseen by the SoS when making the banding provisions and had had or is likely to have a material effect on the operation of the scheme.

Contracts for difference

7. In July 2011 the Government, through the Department for Energy and Climate Change ("DECC"), announced new measures to increase the rate of "decarbonisation" of electricity generation. The central component of the proposed reforms was the introduction of a new scheme to support renewable generation based upon CfDs which would in due course replace the existing RO scheme. A CfD is a long term contract between a generator and a Government-owned counter-party. Pursuant to the contract, the counter-party will pay the generator the difference between an estimate of the market price for electricity and a second price being an estimate of the long term price needed to bring forward investment in the relevant renewable technology. Unlike the position under the RO scheme where the generator's return depends on the market price for ROCs, a generator would know in advance the value of the renewable energy which it generated. The purpose of the new CfD scheme was to provide long term price stability for generators of low carbon electricity, but also a subsidy. When this scheme was introduced in the July 2011 Electricity Market Reform White Paper ("the EMR White Paper") it was promoted on the basis that it provided a higher degree of protection from changes in policy to generators and investors.
8. The EMR White Paper made clear that the Government intended to phase out the RO scheme in favour of the CfD scheme. Transitional arrangements were proposed.

The Levy Control Framework

9. When the coalition government was elected in May 2010 it conducted a spending review. The 2010 spending review set an overall cap for DECC's tax and spending through policies which entailed levy-funded spending. The LCF policy document was issued by the Treasury in March 2011. Paragraph 1.1 stated that the purpose of the LCF was to ensure that DECC achieved its fuel poverty, energy and climate change goals in a way consistent with economic recovery and minimising impact on consumer bills. Paragraph 1.2 explained that the cap imposed by the 2010 spending review was to be managed through the LCF. Paragraph 1.3 is of importance to both sides' cases:

"1.3 The implementation of the control framework will be consistent with relevant legislation, regulations, the Treasury guidance – Managing Public Money – and Government policy announcements. The Government remains committed to maintaining support levels for those existing investments where it has said it would do so and not to making retrospective changes." (underlining added)

10. The LCF thus sets limits upon the costs of DECC's levy-funded policies. The LCF also sets out what is to happen if the cap is exceeded:

"3.1 Should exogenous factors or updated analysis result in forecasts or actual spend that is greater than the agreed cap, then the Treasury will need to be satisfied that there is a robust, agreed plan in place to bring spend down to within the cap…

3.2 Where forecast or actual spend exceeds the agreed cap, then the presumption will be that the Treasury will deny any changes to policy that do not seek to bring the forecast down, even if such changes are cost-neutral."

11. Paragraph 3.4 of the LCF states:

"3.4 Where spend exceeds or is projected to exceed the range of acceptable headroom, DECC will rapidly agree with the Treasury a plan for bringing spending back down to the agreed profile. This plan will set out the adjustments that DECC proposes to make to its policies to reduce their spend, and the impact by year of taking action. The departments recognise the duty to follow statutory and other requirements before making adjustments and the need to maintain levels of support where it has said it would do so." (underlining added)

12. Paragraph 4.1 warns that DECC will retain the right to make changes to policies within the framework. Paragraph 5.1 warns that it should not be assumed that the cap will expand in line with higher than expected demand-led deployment:

"The cap will not increase in line with changes in forecast tax and spending due to non-policy reasons (including forecast error), for example, higher than expected deployment of renewable energy installations..."

Key events and statements

13. In December 2010 DECC issued a consultation document entitled "Electricity Market Reform" ("the EMR consultation") in preparation for the subsequent EMR White Paper issued in July 2011. This was the first of a number of documents which contain statements on which the appellants rely. The EMR consultation recognises the need for a balance to be struck between the need for industry to have certainty and securing "the best possible deal for consumers". This would be achieved by a new system of long term contracts in the form of Feed-in Tariffs ("FiTs") with CfDs. It stated that the Government would:

"seek to minimise uncertainty and delays to planned investments by establishing appropriate grandfathering arrangements. Feed-in-tariffs could be generally introduced from 2013 once legislation has been passed, but accreditation under RO could also continue until 2017 to minimise the risk of disruption for developers…".(underlining added)

14. The EMR consultation proposed that accreditation under the RO would remain open until 31 March 2017. It also contained a paragraph comparing the Feed-in Tariff (with CfDs) to a low carbon obligation (such as the RO) which clearly implies that levels of support under the RO scheme are subject to policy changes whilst investment in constructing the generator is under way:

"Another benefit of a FIT over a low-carbon obligation is that the investor gets certainty when they sign the contract about the level of support they would receive, rather than the support level being set after construction, once the installation is built and connected to the grid, i.e. with the ROCs generators are exposed to the policy risk that the level of support changes in between an investment decision and the project being accredited under a[n] obligation based scheme."

15. The LCF was issued by the Treasury in March 2011. The EMR White Paper, issued in July 2011, was entitled "Planning our Electric Future – a White Paper for secure and affordable low carbon". The document states that "at the heart of our strategy" was the new system of long-term contracts in the form of Feed-in Tariffs with Contracts for Difference "providing clear, stable and predictable revenue streams for investors in low-carbon electricity generation." Respondents to the EMR consultation had expressed the view that a 2017 cut-off date might not allow a long enough lead-in time for such projects. The Government nevertheless maintained its commitment to the 31 March 2017 date. It made clear, however, that existing accredited generation would continue to be supported under the RO. Paragraph 37 of the EMR White Paper contained the following:

"It is essential that the period of transition between the current and new market arrangements runs smoothly and allows investment to continue. As such, we support the principle of no retrospective change for low carbon investments and have listened to industry views on the best way to transition to a new mechanism. Therefore:

to ensure ongoing Renewables Obligation (RO) stability, existing accredited generation will continue to be supported under the RO;

once the FiT CfD is introduced and until 31 March 2017, to provide flexibility new renewable generation will have a one-off choice between the RO and FiT CfD;

the RO will close to new accreditations on 31 March 2017. No generation will be able to accredit under the RO from that date; and

we will grandfather RO support for all technologies at the rate applicable on 31 March 2017."

16. "Grandfathering" support meant that, although the RO scheme would close to new accreditation on 31 March 2017, generators accredited at that date would continue to be supported at the applicable rate thereafter.
17. Chapter 8 of the EMR White Paper was entitled "Managing the Transition". The summary at the beginning of that chapter reiterates that maintaining stable market conditions and industry confidence throughout the period of change was "a key objective". To support that objective DECC stated that it supported the principle of "no retrospective changes for all low-carbon investments". Paragraph 8.4 stated:

"The Electricity Market Reform consultation document set out proposals for a transitional framework from the current RO system to FiT CfD. The aim of our proposals was to protect existing investments under the RO, and provide investors with confidence that projects under development would not be delayed and that investment would be able to continue in the transition period."

18. Mr Palmer submits, and I agree, that this passage draws a distinction between existing accredited investments under the RO on the one hand and pipeline investments which had not yet reached the stage of accreditation on the other. It is the former which were going to be "protected".
19. Returning to the LCF, on 8 December 2011 DECC issued a document entitled "Control Framework for DECC levy-funded spending- Questions and Answers" ("the LCF Q&A"). In a general paragraph of introduction the LCF Q&A explains again that the purpose of the LCF is to make sure that DECC achieves its fuel poverty, energy and climate change goals in a way that is consistent with economic recovery and minimising the impact on consumer bills. Two of the questions and answers are relied upon by the SoS:

"Q. Under what circumstances will there be policy changes as a result of this framework and will industry be consulted on policy changes?

Where a policy is forecast to overspend against the envelope, DECC will have to develop plans to bring spend back within the cap, taking into account impact on energy bills and progress towards our targets. How and how quickly these changes are implemented will depend on the various factors pertaining to the policy at the time; however, DECC will follow all required procedures such as statutory consultation and Parliamentary scrutiny".

and:

"Q. What happens if DECC thinks spend might be on course to exceed its envelope?

If we think that spend might be on course to exceeding its envelope (either in the short term or at some point in the future) we will consider making an adjustment to the policy, taking into account the impact on energy bills and progress towards our targets. Decisions on adjustments will be taken by Ministers at the appropriate time and will be judged on a case by case basis. A core principle through all of this is that the Government policy is to maintain support levels for those existing investments where we have said we would and not to make retrospective changes for these investments.

If Ministers take a decision to adjust a policy DECC will follow statutory and other required procedures. " (underlining added)

20. The phrase "to maintain support levels for those existing investments where we have said we would and not to make retrospective changes for these investments", said to be a core principle, echoes the similar phrase in the LCF itself. An important issue between the parties is what investments the first part of this phrase is referring to.
21. It is in the light of the statements in the LCF itself and the LCF Q&A that the judge concluded that anyone seeking RO support either knew or must be taken to have known that the continued currency of the RO was dependent upon expenditure not exceeding its envelope, and any operator who wished to work out what the envelope referred to was could easily have discovered that it meant the cap imposed by the Treasury.
22. In December 2012, the SoS published the results of a banding review for the RO scheme. Levels of support for ground-mounted solar PV in the period April 2013 to March 2017 varied downwardly from 1.6 ROCs per megawatt hour in 2013/14 to 1.2 ROCs per megawatt hour in 2016/17. These could of course be varied further if circumstances justified a further banding review.
23. In July 2013 a fresh consultation was issued. It was entitled "Consultation on the Transition from the Renewables Obligation to Contracts for Difference". In the Executive Summary the Government again stated that the RO scheme would close to new capacity on 31 March 2017. Thus the Executive Summary stated:

"This consultation seeks views on policy and processes for the transition period, from mid 2014 to 31 March 2017, during which the RO and CfD will both be open for applications from new renewable generating capacity.…

On 31 March 2017, the RO will close to new capacity…"

24. In order to implement the policy of transition from the RO scheme to CfDs, the Government needed to introduce new legislation. However, in order to close the RO scheme to new entrants in 2017, the SoS originally intended to use the existing power then contained in section 32C(1) of the Electricity Act 1989. Accordingly, when an Energy Bill was introduced into the House of Commons in 2012 to effect the changes, it did not contain any new power to make a "renewable obligations closure order". Doubts subsequently arose within Government as to whether the existing Electricity Act provision could in practice properly be used to close the RO scheme to all new entrants and whether its effect could extend beyond England and Wales. A Government amendment to the Energy Bill was therefore introduced in the House of Lords to insert into the Electricity Act 1989 what subsequently became sections 32LA and 32LB of that statute. The power so introduced was the power to make a renewable obligations closure order.
25. On 4 November 2013 Baroness Verma explained that the introduction of these new provisions was to address the lack of clarity as to the extent of the existing power to close the RO scheme to new capacity and the ability simultaneously to extend the effect of the closure to Scotland:

"My Lords, Amendment 66 provides the Government with the power to close the Renewables Obligation to new capacity. As noble Lords know, this closure is planned for 31 March 2017 as part of the transition to contracts for difference. We have previously considered that the Renewables Obligation could be closed using existing powers within the Electricity Act 1989. However, we have now concluded that a specific power in this Bill will put the closure arrangements on a more reliable and transparent legislative basis.

To ensure that consumers and industry have confidence that closure will take place consistently across the UK, the amendment provides the power for the Secretary of State to close the RO in England, Scotland and Wales. It enables the Northern Ireland Executive to make similar provision for the Northern Ireland Renewables Obligation. To give industry early certainty on the way in which Government proposes to use this power, we intend to publish this week detailed proposals on RO grace periods for those projects that are delayed due to circumstances beyond their control. These proposals will include a 12-month grace period for projects subject to current investment decisions, giving developers making such decisions this winter substantial reassurance that their investments are not at undue risk from the RO closure date".

26. The Energy Act 2013 received Royal Assent on 18 December 2013. The Explanatory Notes which accompanied it record:

"33. The EMR White Paper proposed that the transition phase would end on 31 March 2017, after which the RO would be closed to new generating capacity. The RO would continue to operate for the generating capacity which accredited under it before it closed to new generating capacity.

282. Subsection (1) inserts new sections 32LA and 32LB to the Electricity Act 1989 to confer a power on the Secretary of State to make a renewables obligation closure order. The closure order will prevent renewables obligation certificates from being issued under any renewables obligation order (whether made by the Secretary of State or by the Scottish ministers) in respect of electricity generated after a specified date (section 32LA(2)). Different closure dates may be specified for different cases or circumstances (section 32LA (3))."

27. Despite all the outward signs that the RO scheme would close to new generating capacity in 2017, on 13 May 2014, in a consultation paper entitled "Consultation on changes to financial support for solar PV" ("the May 2014 consultation paper"), DECC indicated that it had changed its policy in relation to the proposal to close the RO scheme. The closure date for large-scale solar PV was now to be some two years earlier, namely 1 April 2015. The proposals included a grace period to protect those who could demonstrate a significant financial commitment to projects on or before 13 May 2014, that is to say the date on which the consultation was published.
28. DECC justified its change of policy in the May 2014 consultation paper by reference to the higher than expected rate of uptake of solar PV generation since the Government had announced its levels of support under the scheme, and the knock-on effect this would have on the LCF. Thus, in December 2012, PV deployment was anticipated to be around 900 MW by the end of March 2017. However, by the end of March 2014 (after only some 15 months and with 36 months still to run to March 2017), 545 MW had already been accredited and there was evidence indicating that about 3.2 GW of large-scale solar PV could be deployed under the RO scheme by the end of March 2015. Conservative estimates now indicated that there would be a total deployment of 5 GW by the end of March 2017 which exceeded by a considerable margin the upper end of the potential rate of deployment contemplated by the Government in December 2013. In paragraph 20 of the consultation paper, under the heading "The Problem" they said this:

"This is more than we can afford and would have adverse consequences for Government's management and use of the LCF as a whole. The proportion of the LCF which is available for deployment under CfD's would be reduced, as a higher proportion of the LCF would necessarily be allocated to the RO to cover the costs of the additional solar projects. Government's view is that the CfD is a more cost-effective mechanism than the RO. Because the CFD provides for early certainty of support levels than the RO and greater stability of revenue streams by providing a fixed strike price, investors are protected from wholesale price volatility and should therefore be able to reduce the cost of capital, making the development of low carbon generation cheaper for both investors and consumers."

29. Since the publication of the May 2014 consultation paper, work conducted within DECC to calculate the level of the RO for 2015/16 suggested that there would be between 2 and 4GW of solar PV accredited under the RO in 2014/2015. In paragraph 175 of his witness statement in these proceedings Mr Gareth Redmond, head of the Renewables Programme Team within DECC, said that it was estimated that, without any action to constrain further deployment, 6.6 to 10 GW of solar PV could deploy by the end of 2016/17. This would result in £210-£400 million extra spending per year above the EMR Delivery Plan base case. It would result in DECC exceeding the LCF by around £40 million in 2016/17.
30. On 8 September 2014 the SoS, in exercise of the powers conferred on him by sections 32K and 32LA of the Electricity Act 1989, made The Renewables Obligation Closure Order 2014 ("the 2014 order") which came into force on the following day. The 2014 order provided, subject to certain exceptions, that no renewables obligation certificates were to be issued under a renewables obligation order in respect of electricity generated after 31 March 2017: see paragraph 3.

The judicial review proceedings

31. These judicial review proceedings were commenced on 1 August 2014, before DECC had published its consultation response to the May 2014 consultation paper, on the basis that that consultation had put forward proposals which were unlawful. Subsequently the consultation response was published and it is convenient to treat the decisions in that consultation response as being the target of the judicial review.

The decision under challenge

32. The Government's response to the May 2014 consultation was published on 2 October 2014 ("the October 2014 response to consultation"). The decisions were to the effect (a) that the RO scheme will terminate early in 2015 for new solar PV capacity above 5 MW, not in 2017 as originally intended, and (b) that the benefits of the old scheme will be extended but only to those who by a specified, historical, date can show that they meet certain conditions relating to the nature and maturity of the investments they have made in order to bring renewable energy generation projects to fruition.
33. The decision on the closure of the scheme was expressed in the October 2014 response to consultation in the following way:

"10. We have considered very carefully the arguments presented on both sides of this question. We acknowledge that bringing forward closure of the RO for large scale solar PV projects represents a change of policy from that previously announced and given effect in the RO Closure Order 2014, and that the majority of respondents are against us doing so. We have taken into account the fact that large-scale solar PV developers expected that the RO would remain open until 31 March 2017. However, we cannot ignore the very clear evidence that large-scale solar PV is deploying faster than can be afforded and, in addition, that there is significantly more potential deployment of large-scale solar PV than estimated when we published our consultation less than five months ago, and the heightened risk that this poses to the LCF. The position of large-scale solar PV developers who have made significant financial commitments in reliance on their expectation of the previously adopted closure date is addressed by our proposals for a grace period, as discussed on the question four below.

11. The Government has therefore decided to close the RO to new solar PV projects above 5 MW in scale from 1 April 2015, and to additional capacity added to existing accredited stations from that date, where the station is, or would become, above 5 MW."

34. The decision as to grace periods was expressed in the following way:

"The Government has decided to provide a grace period designed to protect projects with significant financial commitments have been made on or before 13 May 2014, i.e. the day on which we published our consultation. We have also decided to maintain 13 May 2014 as the date by which significant financial commitments must have been made."

35. The requirements for taking advantage of the grace period were expressed in paragraph 1.51 as follows:

"A grid connection offer and acceptance of the offer, both dated no later than 13 May 2014;

A Director's Certificate confirming ownership of the land, the lease of the land or an option to lease or to purchase the land as of 13 May 2014; and

Confirmation that a planning application had been received by the relevant planning authority in respect of the project on or before 13 May 2014."

36. These requirements represented a relaxation of those contained in the May 2014 consultation paper. The consultation had suggested that relevant planning consents should have been obtained by 13 May 2014, not merely a planning application, and the provision of evidence in the form of invoices and payment receipts from the developer or proposed operation demonstrating a minimum of £100,000 of expenditure per megawatt of expected consented capacity, or prove that all material equipment contracts have been entered into by 13 May 2014.
37. Those qualifying for the grace period gained a further year, until 31 March 2016 to submit their evidence that they complied with the requirements for the grace period, and apply for accreditation.
38. The decisions contained in the October 2014 response to consultation were given effect to by the Renewables Obligation Closure (Amendment) Order 2015 ("the 2015 order"). By this order no ROCs were to be issued in respect of electricity generated after the "solar PV closure date" by a "large solar PV station". The solar PV closure date was defined as the later of (a) 31 March 2015 and (b) the last day of the month in which the order comes into force. In practice this meant that the solar PV closure date was 31 March 2015. "Large solar PV station" means one with a total installed capacity of more than 5 MW.

Solar farm investment

39. To put matters in perspective, I should say something about the quantum and timescale of solar farm investment. The evidence of Mr Faulks, the company secretary of the first claimant, was that a typical development schedule for a 5 MW plus solar farm, up to the commencement of construction, spanned some 10 months. The planning application would be made in months 5 to 7. Up to the stage of the planning application some £50,000 would be spent per site. Mr Selwyn on behalf of the second claimant indicated a 41 week schedule up to the commencement of construction, with planning and legal fees spent in weeks 1 to 22 estimated at £200,000 or 2% of overall cost.

The grounds of attack on the decision

40. There are four grounds on which the decision was and is challenged:

i) the decision to implement an early closure of the RO scheme by statutory instrument was ultra vires the powers granted by sections 32 LA and LB of the Electricity Act 1989 because the statutory power was for the purpose of preserving the 2017 closure date and not for extending it, ("issue 1");

ii) the pre-legislative statements that the RO scheme will run until 2017 amounted to the type of assurance which would bind the executive and early closure violated those assurances, ("issue 2");

iii) the statements made by the Government from 2010 onwards that the scheme would not close before 2017 were clear and unequivocal representations giving rise to a legitimate expectation which was not thwarted by any policy consideration, ("issue 3");

iv) the periods of grace are retrospective in effect and therefore unfair in a public law sense, ("issue 4").

The judgment of Green J

41. The judge resolved these challenges in the following way. As to the first challenge (misuse of statutory power), having considered what he considered to be the admissible material, he concluded (paragraph 61) that the purpose of the statutory power was:

"… [to] address the two technical matters identified early on as to the extent of the ability to close RO schemes in their entirety, and, as to the geographical extent of the powers. Those two matters were specifically identified as the reasons why the new statutory powers were introduced. It is not possible, in my view, to discern as a mischief or purpose that the statutory amendments were designed to ensure the continuation of the RO scheme until 2017. Indeed it is hard to see why Parliament would need to enact any new measure to preserve in force a scheme until 2017, for which an existing power had already been exercised."

42. The judge also considered (paragraph 62) that the LCF was relevant in this context, as it acted as "an all pervasive proviso or caveat to any exercise of the statutory power." The fact that the RO scheme was demand-led necessarily meant that its impact upon expenditure and hence the overall policy in the LCF was unpredictable. In that context any statement that the RO scheme would endure until 2017 in order to protect capital expenditure carried within it a systemic risk of change.
43. On the second ground of challenge (express pre-legislative assurances) the judge concluded thus at paragraph 67 of his judgment:

"First, I do not construe the Explanatory Notes as creating any form of assurance…. Second, equally I do not construe the statement of the Minister in Parliament as creating any form of representation which could be elevated into an assurance… And further as to the statements made in White Papers and consultation documents these would… carry materially less weight than Explanatory Notes or a direct statement by a sponsoring Minister. But in any event these admittedly clear statements of intent were not and could not be construed as "assurances". They reflected the policy of the moment which whilst firm was always capable of reversal and the constraints in the LCF represented inescapable context."

44. On the third ground (frustration of legitimate expectations) the judge held at paragraph 78 of his judgment that no legitimate expectation was created by the various statements relied on, because the LCF represented a systemic risk:

"The risk was that if uptake for support led to increases in expenditure beyond the agreed HM Treasury limits that the scheme might (or even would) be curtailed in order to bring expenditure back under control. As such no operator could expect that the system would inevitably or necessarily last until 2017. Put another way, the highest that the legitimate expectation can be put is that the scheme would not be closed absent an increase in expenditure under the scheme which would put the Treasury cap at risk."

45. Moreover, the judge held (paragraphs 85-90) that, even if he was wrong about the existence of a legitimate expectation, there were good policy reasons for frustrating such an expectation.
46. Finally as to the challenge based on the allegedly unfair grace period, the judge held (paragraph 101) that there was:

"… a degree of retrospection that applies to [the claimants] in that when they embarked upon their present investments they thought they could plan the pace of expenditure in such a way that they would, in due course, become accredited before the 2017 closure date. I accept therefore… that the change in the rules has exerted a retrospective impact which is more than de minimis."

47. The judge accordingly applied what he described as "a basic fairness or rationality test which takes into account the level of retrospection which applies". He concluded (paragraphs 104-7) that (a) the SoS had addressed himself to all relevant considerations including the plight of pipeline investors; (b) the SoS had not violated any procedural standard because he had consulted on the conditions for the grace period and had taken account of the criticisms made; (c) the SoS had sound policy reasons for wishing to apply the rules in this fashion; (d) the choice of the cut-off date of May 2014 was not unfair as it coincided with the date of the consultation: had a further period of time been given for operators to prove compliance with the stated conditions then this would have risked what had been referred to as the "gold rush" adding between £110-120 million per annum to costs from 2016/17 onwards; (e) it was wrong to suggest that the threshold for protection should have been devised to protect any investment which was more than de minimis: the Minister had to do draw a line somewhere by identifying what in his view was a significant investment. For those reasons the judge considered that the grace periods were fair, and rejected the challenge based on retrospection.

Discussion

48. Because it was put at the forefront of Mr Fordham's submissions, it is convenient to address issue 3 first, before turning to the other grounds of challenge.

Issue 3: legitimate expectations

49. There was no real dispute about the relevant principles which apply to the role of legitimate expectations in a situation in which the state wishes to change a previously announced policy. The principles were elucidated by Laws LJ in Bhatt Murphy v Independent Assessor [2008] EWCA Civ 755 and it is not necessary to repeat them here. Mr Fordham drew our attention to paragraph 34 in that judgment, where Laws LJ said that a promise that a policy would continue "if it points to no particular date or future event to mark the end of the policy" amounted to little more than the ordinary expectation that a policy would continue in force until rational grounds existed for its cessation. Then, at [43] Laws LJ put it in this way:

"43.Authority shows that where a substantive expectation is to run the promise or practice which is its genesis is not merely a reflection of the ordinary fact (as I have put it) that a policy with no terminal date or terminating event will continue in effect until rational grounds for its cessation arise. Rather it must constitute a specific undertaking, directed at a particular individual or group, by which the relevant policy's continuance is assured. Lord Templeman in Preston referred (866 – 867) to "conduct [in that case, of the Commissioners of Inland Revenue] equivalent to a breach of contract or breach of representations"".

50. It is true that this passage is prefaced by reference to a policy with no terminal date or terminating event. Even where an end date is given, however, there can be no immutable presumption on that ground alone that the policy will continue without change until that date, whatever changes in circumstances may arise. That is because Government is entitled to formulate and re-formulate policy when rational grounds exist for doing so, unless to do so would amount to an abuse of power by reason of the manner in which it has previously conducted itself: see per Laws LJ at [41] to [42]. In each case the ultimate question is whether there is a "specific undertaking, directed at a particular individual or group, by which the relevant policy's continuance is assured".
51. Mr Fordham recognised that, for this and the other grounds, the context provided by the LCF was relevant: indeed it was at the heart of the case. If the LCF meant that levy-funded policies such as the RO scheme would have to be promptly re-visited if they threatened to breach the agreed cap, then there are obvious difficulties in finding a sufficiently clear representation that such a scheme would continue until 2017, come what may. However Mr Fordham contended that the respondent and the judge had misinterpreted the LCF, and thus misled themselves as to the impact it would have on the expectations of the industry. He submitted that the central question was whether, as the judge had held, the clear statements as to the original closure date of the RO scheme were qualified or subject to the LCF, or whether, as he submitted, the LCF was itself subject to a clear proviso that the Government would keep promises it had made in relation to maintaining support levels for existing investments (including pipeline investments) until 2017. Mr Fordham also relies on the very clear statements that it was the Government's objective to provide stability for investment in solar PV. It was in the Government's interests to make promises that levels of support which were on offer whilst a project was being developed would not subsequently be withdrawn, because it would thereby stimulate investment in renewable energy.
52. It thus becomes critical to understand what was promised by the Government and, in particular, what was meant by paragraph 1.3 of the LCF (see paragraph 9 above) when the Government spoke of "maintaining support levels for those existing investments where it has said it would do so". Does that phrase include investments which are in the pipeline towards accreditation but not yet accredited?
53. In support of his submission that existing investments included those in the pipeline towards accreditation, Mr Fordham sought to draw on the language used by the SoS' witnesses in their witness statements made for the purpose of these proceedings in describing the history of the RO scheme. For example, in some passages in his evidence Mr Redmond had used the phrase "protecting existing investments" or "protecting existing significant investments" in a context which appeared to cover pre-accreditation investments. I do not consider, however, that this is a legitimate source of context for the purposes of interpreting the LCF. It would not be available even to the best informed of readers.
54. More relevant by way of context, as it seems to me, is what the Government had said up to this point about maintaining support levels. The EMR consultation had made it clear that there was to be grandfathering, which I take to mean continuing the agreed support level in relation to installations which had become accredited by the cut-off date. Beyond that, however, the EMR made clear that the support level was set "after construction and once the installation is connected to the grid", that is to say when the installation is accredited. The EMR consultation is not a statement that pipeline projects would be afforded the support levels on offer to accredited installations.
55. There is also the context of the LCF itself. The LCF makes it quite plain that policies within the framework will be subject to rigorous adjustment where spending threatens to exceed the agreed cap. There is an express warning that higher than expected deployment of renewable energy installations will not cause the agreed cap to be increased, as well as the express reservation of the right to make changes to policies within the framework. This is an unpromising background for a suggestion that the phrase "maintaining support levels for existing investments" was intended to extend to pipeline investments which might be proceeding at higher than expected rates.
56. The White Paper in July 2011 also contains no suggestion that there was any proposal to protect pipeline investments as opposed to accredited installations. Paragraph 8.4 of the White Paper speaks of protecting "existing investments under the RO" (by grandfathering) and draws a distinction with projects which are merely under development (where a choice will exist between the RO and the CfD). Pipeline investments are not investments "under the RO".
57. The LCF Q&A document also makes clear the fact that policy in this area was subject to change if there was higher than expected deployment.
58. Whilst Mr Fordham is correct to say that the SoS had been at pains to reassure the industry that it wanted to provide a stable investment environment, this was not the sole objective of policy in this area. I agree with the judge that, whilst security and stability of the investment environment were important aspects of the policy, there were other aspects to it as well. These included the need to ensure that funds were available for a diverse range of technologies, not just solar PV; the need to ensure funds were available to support CfDs; and the objective of limiting and reducing consumer's bills. Not all these objectives could be achieved without compromise, and solar PV operators could not possibly entertain a secure expectation that their particular scheme would be protected at the expense of others.
59. I therefore reject Mr Fordham's submission that the LCF was subject to a caveat in favour of existing pipeline investments. In my judgment the statement in paragraph 1.3 of the LCF would be understood as applying to accredited installations which would be grandfathered. Moreover, the Government's statements about investment security would not be taken as being immune to change.
60. It follows that I agree with the judge that no legitimate expectation was created that the Government would not change its policy with regard to the March 2017 closure date, and thereby impact those who had made pipeline investments. It must have been apparent to all concerned that, if uptake of solar PV threatened the LCF cap, the Government might, and probably would, bring forward the closure date. On this basis I would reject the challenge inherent in issue 3. It is accordingly not necessary to consider the judge's alternative reason for rejecting it, namely whether any expectation which was created was legitimately frustrated.

Issues 1 and 2: misuse of statutory power and clear assurances

61. Section 32LA of the Electricity Act 1989 provides as follows:

"32LA Renewables obligation closure order

(1) The Secretary of State may make a renewables obligation closure order.

(2) A renewables obligation closure order is an order which provides that no renewables obligation certificates are to be issued under a renewables obligation order in respect of electricity generated after a specified date.

(3)Provision made under subsection (2) may specify different dates in relation to different cases or circumstances.

(4)The cases or circumstances mentioned in subsection (2) may in particular be described by reference to—

(a) accreditation of a generating station, or

(b) the addition of generating capacity to a generating station."

62. It is common ground that the language used in section 32LA is, on its face, wide enough to enable the SoS to curtail the period of operation of the RO Scheme from 2017 to 2015. Mr Fordham submits (in his skeleton argument) that the judge's conclusion that the purpose of the statutory power was to address the two technical matters identified, namely the ability to close RO schemes in their entirety, and for Scotland as well as England and Wales, was an error, because it ignored a third purpose, which was to close it with effect from 1 April 2017. He further submits that if the purpose of the power was to close the RO scheme in its entirety, then the challenge is still a good one because the SoS has purported to exercise the power for the purpose of closing the scheme only in relation to solar PV capacity above 5 MW.
63. I am unable to accept these submissions. As to the first point, I cannot see any reason why Parliament should have intended that the Minister's power should be constrained to closing the scheme as from 1^st April 2017. Parliament gave the Minister a power to close the scheme "to electricity generated after a specified date". Neither the Ministerial statement nor the Explanatory Notes suggest that the purpose of conferring this power was to allow it to be exercised only in relation to a date on or after 1 April 2017. The Minister stated that the closure was planned for 31 March 2017. That was a statement of Government policy: he did not say that that there would be no circumstances in which the RO scheme would be closed any earlier. The Explanatory Notes record, correctly, that the White Paper proposed that the transitional phase to CfDs would end on 31 March 2017, and also correctly described the effect of the power contained in section 32LA. There is nothing in the Notes to indicate that the purpose of the power was to keep the scheme open until 2017. Indeed, as the judge observed, no new power was needed to keep the existing scheme open.
64. None of the material in the case provides a basis for suggesting that the power was only given for the purpose of closing the RO scheme on or after 31 March 2017.
65. As to the second point, it does not follow from the fact that the purpose of the power was to enable the SoS to close the RO scheme in its entirety and for the whole territory that the SoS could not exercise the power for some lesser purpose. Indeed the SoS is expressly empowered to specify different dates in relation to different cases or circumstances. He chose to specify a date in the case of large solar PV, as he was empowered to do.
66. Whilst presented to the judge as an independent point, I do not consider that the appellants' case based on clear assurances (issue 2) can survive the conclusions I have come to on the legitimate expectations case. Reading it in its proper context, I am unable to regard any of the material in this case as providing anything in the nature of a clear and binding assurance, to Parliament or otherwise, that the scheme would not be closed before 2017.

Issue 4: unfair grace periods

67. The main focus of Mr Fordham's submission on this final ground of challenge was the unfairness of specifying a date which has already passed as the date by which an investor in solar PV would have to have satisfied the various criteria for a significant investment, and thereby benefit from the grace period. He submitted that this was conspicuously unfair in a public law sense.
68. Mr Fordham also submits that in applying a basic fairness or rationality test the judge overlooked a prior question of whether Parliament had conferred a power to act retrospectively at all.
69. I will deal with the second point first as it seems to me to be logically prior to the assessment of fairness. Mr Fordham, relying on the judge's conclusion that there was an element of retrospectivity in the changes, submitted that there was no clear power in the legislation to enact retrospective changes at all. He submitted that the judge had been right to find that the legislation had retrospective effect. He took as an illustrative example an investor who had failed to submit a planning application for his solar farm by 13 May 2014. Such an investor would have expected that his investment up to that point would ultimately come to be protected by the RO scheme, because that scheme would continue to be open for accreditation until March 2017. He would not have appreciated that the consequences of not having filed a planning application before 13 May 2014 would be that the benefits of the scheme would not be available to him. The effect of the decisions as to grace periods, however, was to attach new consequences to that omission. That meant that the decision, and the legislation which gave effect to it, had an unlawful retrospective effect.
70. We were referred to a passage from Bennion on Statutory Interpretation (6th Edition):

"It is important to grasp the true nature of objectionable retrospectivity, which is that the legal effect of an act or omission is retroactively altered by a later change in the law. However the mere fact that a change is operative with regard to past events does not mean that it is objectionably retrospective. Changes relating to the past are objectionable only if they alter the legal nature of the past act or omission in itself. A change in the law is not objectionable merely because it takes note that a past event has happened, and bases new legal consequences upon it."

71. In my judgment the changes to the RO scheme which were effected to close the RO Scheme to new generators with effect from 1 April 2015 were not retrospective in this sense at all. The most that could be said was that acts or omissions committed in the past would have different consequences from those which would have been contemplated at the time. This does not alter the legal nature or even the legal effect of those acts or omissions. The legal nature of acts or omissions in respect of installations which have not been accredited has not been altered. In this connection one can contrast the effect of the provisions in play in Secretary of State for Energy & Climate Change v Friends of the Earth and others [2012] EWCA Civ 28 where at [42] Moses LJ pointed out that:

"Modification of the FIT Payment Rate, in respect of installations becoming eligible prior to the modification, would have a retrospective effect. Because the Scheme fixes a rate by reference to the year the installation becomes eligible, reduction of that rate ... would all have a retrospective effect… That entitlement arises on the eligibility date. Any modification of the rate … takes away the owner's entitlement under the Scheme to payment at that fixed and pre-determined rate."

72. No question of modification of an accrued entitlement arises in this case. Accordingly I consider that there is no need to look in the legislation for a power to act retrospectively in this sense.
73. It seems to me that the first point is concerned with the fairness or rationality of the exercise of the power in circumstances where it will frustrate the expectations of those who had committed expenditure which falls short of that required to benefit from the grace period. It is of course relevant, as the judge held, to appreciate that the effect of the changes announced in the decision was to alter the rules of the game after investors had started playing it. But this cannot be more than a factor in the overall assessment of fairness of the exercise of the power which Parliament undoubtedly had granted. As such it is closely related to, if not indistinguishable from the case based on legitimate expectations. Moreover if it was lawful for the SoS to close the RO scheme in its entirety to new entrants with effect from 1 April 2015, I have difficulty in seeing how it could be unlawful to soften that blow by extending the scheme for a further year to those who had reached a particular stage of investment.
74. If the SoS was going to have a grace period at all, then I do not regard it as remotely arguable that he was obliged to extend it to anyone who had made more than de minimis investment as the appellants contend. There is no conceivable reason why the grace periods should have such a long tail. In those circumstances a line had to be drawn. There can be no criticism of the procedural fairness of the process of consultation adopted to determine where the line should be drawn: it in fact resulted in a significant relaxation of the conditions. In setting the conditions the SoS had to balance a number of considerations, as the judge held, including the important one of not accelerating pipeline investment so as to exacerbate the very problem which had caused the date to be brought forward. These were all matters for the decision making of the SoS. There is no basis for interference by way of judicial review.
75. I reject the appellants' suggestions that the grace periods were the subject of retrospective legislation, and that the use of the legislation to enact them was unfair in a public law sense. Accordingly I would reject this ground of challenge as well.

Conclusion

76. For the reasons I have given, none of the judicial review challenges succeed. I would therefore dismiss the appeal.

Lord Justice Treacy:

77. I agree.

Lord Justice Tomlinson

78. I also agree.