DECISION BY THE UPPER TRIBUNAL
(ADMINISTRATIVE APPEALS CHAMBER)
The DECISION of the Upper Tribunal is (nominally at least) to allow the appeal.
The decision of the First-tier Tribunal (General Regulatory Chamber) (Information Rights) dated 24 December 2014 involves an error on a point of law. The appeal is therefore allowed and the Tribunal’s decision is set aside. The Upper Tribunal is able to re-make the decision under appeal. The decision that the First-tier Tribunal should have made is to the same effect and is as follows:
“The Tribunal allows the appeal by Mr Henney in relation to the preliminary issue as to whether his request was governed by FOIA or the EIR. The requested information is ‘environmental information’ within regulation 2(1)(c) of the EIR.
The parties are invited to propose further directions for the onward conduct of this appeal, to be sent to the office of the First-tier Tribunal (General Regulatory Chamber) within one month of the date of issue of this decision.”
This decision is given under section 12(2)(a) and 12(2)(b)(ii) of the Tribunals, Courts and Enforcement Act 2007.
REASONS
The underlying question posed on this appeal
1. This appeal concerns the proper demarcation line between the Freedom of Information Act 2000 (“FOIA”) and the Environmental Information Regulations 2004 (SI 2004/3391; “the EIR”). How should decision makers and tribunals decide which of these two access regimes applies to the disputed Information on any given request? It concerns in particular the interpretation and application of regulation 2(1)(c) and (e) of the EIR.
Some more abbreviations
2. In addition to the EIR and FOIA, I also use the following abbreviations in this decision (for convenience some are also spelt out in full on the first occasion they appear):
Aarhus Convention Convention on Access to Information, Public
Participation in Decision-Making and Access to Justice in Environmental Matters (Århus, Denmark, 25 June 1998)
CJEU: Court of Justice of the European Union
DCC: Data and Communications Company
DECC: Department of Energy and Climate Change
the Directive Directive 2003/04/EC
the Electricity Directive: Directive 2009/72/EC
EU: European Union
FTT: First-tier Tribunal
IC: Information Commissioner
Implementation Guide The Aarhus Convention: An Implementation Guide (Second Edition, 2014)
MPA: Major Projects Authority
PAR: Programme Assessment Review
SMP: Smart Meters Programme
The outcome of this appeal in outline
3. In summary I allow the appeal by the Department of Energy and Climate Change (DECC). I conclude that the First-tier Tribunal (FTT) erred in law in the way in which it approached the interpretation and application of regulation 2(1)(c) and (e) of the EIR, which define (in part) the ambit of “environmental information”. I therefore set aside the FTT’s decision. I consider it appropriate to re-make the FTT’s decision. In doing so, I reach the same conclusion as the FTT, albeit by a different route. The requested information is “environmental information” within regulation 2(1)(c) of the EIR. The case must now go back to the FTT for the substantive issues to be addressed. While the “headline” for this appeal can be expressed in that short compass, I apologise in advance for what may be the inordinate length of this decision.
The context to this appeal
4. ‘Smart meters’ are meters that provide near real-time information to consumers (as well as to suppliers and network operators) on the amount of electricity (or gas) being used. The EU Electricity Directive (Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC) requires Member States to carry out a cost-benefit assessment of the large-scale distribution of smart meters and, where the assessment is positive, to provide at least 80 per cent of consumers with smart meters by 2020 (Annex 1, paragraph 2 to the Directive).
5. The UK Government’s assessment of smart metering has been positive, and it has pledged to take all reasonable steps to equip all domestic and smaller non-domestic premises with smart metering by the end of 2020. As time marches on, of course, this aspiration to roll out the Smart Meters Programme (SMP) within such a timescale may seem increasingly ambitious (see further House of Commons Energy and Climate Change Committee, Smart meters: progress or delay? (Ninth Report, Session 2014-15, HC 665)). Indeed, the history of other large-scale Government-backed IT projects has notoriously not always been a happy one. Accordingly a partnership between the Cabinet Office and HM Treasury, known as the Major Projects Authority (MPA), works with central government departments with a view to achieving firmer control over the implementation of such major projects, from policy development through to delivery.
6. The Project Assessment Review (PAR) is one tool that may be used in this process. The PAR is essentially an intensive assurance review designed to provide an objective assessment of the viability of a major project (such as the SMP) or a part of it – in broad terms “to ensure that major projects stay on track, deliver the expected benefits and are delivered on time and within cost” (FTT open bundle, first witness statement of Mr D Blackall at §8). Risks are assessed against a ‘traffic light’ system known as the RAG (Red–Amber–Green) scale. In the present case a PAR was commissioned for one particular aspect of the SMP, namely the communications and data component, i.e. the means by which information is communicated from smart meters to (primarily) suppliers and network operators. Central to the communications and data aspect of the SMP is the Data and Communications Company (DCC), which describes itself on its website as “the company that's delivering Great Britain's data and communications infrastructure to support the mass roll out of smart electricity and gas meters to homes and small businesses” (and see further https://www.smartdcc.co.uk/).
Mr Henney’s request and complaint to the Information Commissioner
7. Mr Henney, who has a longstanding professional interest and expertise in energy usage and policy, made a request for the information in the PAR back in November 2012. He is now three years on and not much further forward. Understandably perhaps, he did not frame his original request in legalistic terms of either FOIA or the EIR:
‘I have been told that there was some form of “independent review” of the roll-out [of smart meters] which endorsed the Department’s opinion of the viability of the roll-out. I would be obliged if you would provide me with the key papers.’
8. DECC provided him with a very heavily redacted copy of the PAR, relying on sections 35(1)(a), 36(2)(b)(ii) and (c), 40(2) and 43(2) of FOIA to justify withholding certain material. These are the FOIA exemptions that relate to the formulation of government policy, prejudice to effective conduct of public affairs, personal information and prejudice to commercial interests respectively. Mr Henney complained to the IC in April 2013, albeit he reasonably took no point on the application of the personal data exemption. However, he did complain that he had been provided by DECC with “three pages of Whitehall waffle and a ‘copy’ of 15 pages of which 14½ were redacted” (letter to IC dated 26 April 2013).
9. A year later, in March 2014, the IC issued his Decision Notice (FS50495646), in which he accepted that the appropriate access regime was FOIA and not the EIR. However, while the IC found that section 35(1)(a) (formulation of government policy) was engaged, he concluded that on balance the public interest favoured disclosure of the requested information. The IC also found that sections 36(2) (prejudice to effective conduct of public affairs) and 43(2) (prejudice to commercial interests) were not engaged at all. Both DECC and Mr Henney appealed to the FTT; DECC against the decision ordering disclosure, Mr Henney as to the applicability of the relevant access regime. Mr Henney explained that his own appeal was prompted by the Government’s decision to veto disclosure of the PAR relating to the HS2 project and his concern that a successful appeal about smart meters under FOIA (as opposed to one under the EIR) might face the same fate. I return to this point (much) later (see paragraph 99).
The First-tier Tribunal’s decision in summary
10. In the event, the FTT dealt only with the preliminary issue as to the appropriate access regime. It held an oral hearing on 4 and 5 November 2014, receiving both legal submissions and live evidence from several witnesses, followed by further submissions from the parties in writing. The FTT’s decision was promulgated on 30 December 2014. Its decision was to allow Mr Henney’s cross-appeal, i.e. the FTT concluded that the information he requested was covered by the EIR and not by FOIA. The FTT subsequently gave DECC permission to appeal to the Upper Tribunal against its ruling on the preliminary issue, noting that the arguments on both sides had merit, that “this was a finely balanced matter”, and reporting that “the FTT debated the matter at considerable length”.
The proceedings before the Upper Tribunal
11. I held an oral hearing of this appeal at the Rolls Building in London on 14 September 2015. DECC was represented by Mr A. Choudhury QC, instructed by the Government Legal Department. The IC was represented by Mr R. Hopkins of Counsel, instructed by the Solicitor to the Commissioner. Mr Henney was ably assisted, if not formally represented “on the record”, by Mr M. Frankel of the Campaign for Freedom of Information. The parties also made further written submissions after the hearing in response to my further directions. I am indebted to all concerned for their written and oral contributions.
12. I apologise for the fact this decision has been rather slower emerging than I had intimated might be the case at the end of the oral hearing. This was partly due to the need to issue further directions but also a consequence of my temporary partial assignment to other judicial duties. I recognise that in one sense Mr Henney is actually no further forward than he was almost a year ago, when the FTT decision on the preliminary issue was promulgated. Such, however, is the course of appellate litigation.
The statutory architecture of the two access regimes
13. There are a number of important differences between the FOIA and EIR access regimes (e.g. as regards the respective exemptions and exceptions that may apply). This is not the place to enumerate them all. I merely accept Mr Choudhury’s characterisation of the EIR as being “the more onerous regime” for public authorities in certain respects. However, I do not regard that as a material consideration when approaching the question of statutory interpretation. More importantly, I recognise the starting point is that the conceptual architecture of each regime is not the same.
14. FOIA requires a public authority to inform a requester, first, “whether it holds information of the description specified in the request and secondly, if so, to communicate that information (FOIA, section 1(1)) – unless, of course, one of the various exemptions applies. The concept of “information” itself is left at large and, as the definition section reminds us, it means “information recorded in any form” (FOIA, section 84). It follows that under FOIA – subject perhaps only to the limits of the English language – it is the requester who initially sets the limits as to the subject-matter of the information requested. Furthermore, section 39(1) of FOIA provides that environmental information within the EIR is excluded from the ambit of FOIA:
‘(1) Information is exempt information if the public authority holding it—
(a) is obliged by environmental information regulations to make the information available to the public in accordance with the regulations, or
(b) would be so obliged but for any exemption contained in the regulations.’
15. The EIR adopt a very different approach. Putting to one side the important statutory obligation placed on public authorities progressively to disseminate environmental information by electronic means etc (regulation 4), the primary duty is that “a public authority that holds environmental information shall make it available on request” (regulation 5(1), emphasis added). The ambit of the EIR is thus defined by the subject-matter of information in issue, and that category of information is defined in turn by regulation 2(1) (see below). Regulation 12 of the EIR then sets out various exceptions to the duty to disclose environmental information. However, there is no exception under regulation 12 for information governed by FOIA, as there is no need for any such provision, given section 39(1) of FOIA. So in terms of the scope of “environmental information”, the EIR cover what it says on the tin. The question then is what the small print on the tin means when it refers to regulation 2.
16. In her recent inaugural Annual Lecture for the UCL Centre for Environmental Law, Professor Liz Fisher of the University of Oxford observed that notwithstanding the apparent “simple, literal and unadorned language of the EIR”, and the apparent breadth of the definition of “environmental information”, there is “more complexity to the EIR than meets the eye” (‘Environmental Law in the Glasshouse: A decade of the Environmental Information Regulations 2004’, UCL, October 20, 2015). The present appeal bears out the force of that observation.
The statutory definition of “environmental information”
17. The statutory definition of “environmental information” has a well-established pedigree in international law and European law. For present purposes it begins with the Aarhus Convention of 1998 (the Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters). Article 2(3) provides the following definition for the purposes of the Convention:
‘“Environmental information” means any information in written, visual, aural, electronic or any other material form on:
(a) The state of elements of the environment, such as air and atmosphere, water, soil, land, landscape and natural sites, biological diversity and its components, including genetically modified organisms, and the interaction among these elements;
(b) Factors, such as substances, energy, noise and radiation, and activities or measures, including administrative measures, environmental agreements, policies, legislation, plans and programmes, affecting or likely to affect the elements of the environment within the scope of subparagraph (a) above, and cost-benefit and other economic analyses and assumptions used in environmental decision-making;
(c) The state of human health and safety, conditions of human life, cultural sites and built structures, inasmuch as they are or may be affected by the state of the elements of the environment or, through these elements, by the factors, activities or measures referred to in subparagraph (b) above;’.
18. As a matter of EU law, the prime instrument is Directive 2003/4/EC of the European Parliament and of the Council of 28 January 2003 on public access to environmental information (“the Directive”). This repealed and replaced the earlier Council Directive 90/313/EEC. Recital (1) to the Directive notes that “increased public access to environmental information and the dissemination of such information contribute to a greater awareness of environmental matters, a free exchange of views, more effective participation by the public in environmental decision-making and, eventually, to a better environment.” Article 2(1) of Directive 2003/4/EC then defines “environmental information” for the purposes of the Directive in the following terms, echoing the Aarhus Convention:
‘"Environmental information" shall mean any information in written, visual, aural, electronic or any other material form on:
(a) the state of the elements of the environment, such as air and atmosphere, water, soil, land, landscape and natural sites including wetlands, coastal and marine areas, biological diversity and its components, including genetically modified organisms, and the interaction among these elements;
(b) factors, such as substances, energy, noise, radiation or waste, including radioactive waste, emissions, discharges and other releases into the environment, affecting or likely to affect the elements of the environment referred to in (a);
(c) measures (including administrative measures), such as policies, legislation, plans, programmes, environmental agreements, and activities affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements;
(d) reports on the implementation of environmental legislation;
(e) cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c); and
(f) the state of human health and safety, including the contamination of the food chain, where relevant, conditions of human life, cultural sites and built structures inasmuch as they are or may be affected by the state of the elements of the environment referred to in (a) or, through those elements, by any of the matters referred to in (b) and (c).’
19. Regulation 2(1) of the EIR in turn defines “environmental” information for domestic purposes as follows (the parts of the definition which were particularly in issue in the present appeal are underlined):
‘“environmental information” has the same meaning as in Article 2(1) of the Directive, namely any information in written, visual, aural, electronic or any other material form on —
(a) the state of the elements of the environment, such as air and atmosphere, water, soil, land, landscape and natural sites including wetlands, coastal and marine areas, biological diversity and its components, including genetically modified organisms, and the interaction among these elements;
(b) factors, such as substances, energy, noise, radiation or waste, including radioactive waste, emissions, discharges and other releases into the environment, affecting or likely to affect the elements of the environment referred to in (a);
(c) measures (including administrative measures), such as policies, legislation, plans, programmes, environmental agreements, and activities affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements;
(d) reports on the implementation of environmental legislation;
(e) cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c); and
(f) the state of human health and safety, including the contamination of the food chain, where relevant, conditions of human life, cultural sites and built structures inasmuch as they are or may be affected by the state of the elements of the environment referred to in (a) or, through those elements, by any of the matters referred to in (b) and (c);’.
20. We can see from this lineage that whilst the Aarhus Convention set out a three-fold typology of environmental information ((a), (b) and (c)), the Directive and the EIR have adopted a six-fold classification ((a) to (f) inclusive). In effect, the Aarhus Convention’s category (b) has been subdivided by the European and domestic legislation (with some further tweaking) into four separate categories ((b) to (e) inclusive), so that Aarhus category (c) becomes the European and domestic (f). It is also noteworthy that the EIR definition faithfully follows the Directive. To emphasise the point, as the opening words of the domestic definition explain, ‘“environmental information” has the same meaning as in Article 2(1) of the Directive’.
21. In this context I note the helpful analysis of the structure of regulation 2(1)(a)-(c) inclusive that was proffered by the First-tier Tribunal in Uttlesford District Council v Information Commissioner (EA/2011/0269 & 0285):
‘24. The logical structure and priorities embedded within the definition are noteworthy. The primary focus is on the elements covered in (a) the physical and biological elements of the natural environment.
25. The matters covered in (b) are the products and by-products of human ingenuity which are “… discharges and other releases into the environment, affecting or likely to affect the elements of the environment referred to in (a)”. The consequence of this is that, for example, a noise, energy or radiation source within an appropriately insulated building does not fall within (b) since the insulation means that it is not making a discharge to the environment.
26. The policies and other matters in (c) are again logically subordinate to the preceding paragraphs and are within the scope of the regulations if they are “affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements.” The final clause “measures or activities designed to protect those elements” are measures to protect “the state of the elements of the environment”.’
22. I recognise that in the instant case now under appeal the FTT had intimated that it “was not particularly assisted by the Uttlesford case” (reasons at [18]). However, it seems to me that this was because the facts of Uttlesford were a long way removed from those on the present appeal. Indeed, it would have been positively perverse for a tribunal to have reached any other decision on the facts than that reached in the Uttlesford case (namely that information relating to a local government Code of Conduct for elected members was not “environmental information”). The FTT’s comment in the present appeal does not detract from the value of the structural observation made in Uttlesford.
The Information Commissioner’s decision
23. At this juncture it is appropriate to return to the IC’s Decision Notice in the present case. This was a detailed 18-page document. The preliminary issue as to the relevant access regime was dealt with in two pages towards the start of the Decision Notice. This passage read as follows (omitting a footnote):
‘The applicable access-regime – FOIA or the EIR?
7. The requested information in this case relates to a review of a proposed roll-out of advanced smart meters. The smart meters scheme follows a European Directive (2009/72/EC) that requires member states to implement intelligent metering systems that shall assist the active participation of consumers.
8. According to the government’s website, smart meters give a wide range of benefits to the public as energy consumers. These include the following:
· Smart meters give you near real time information on energy use –
expressed in pounds and pence
· You will be able to better manage your energy use, save money
and reduce emissions
· Smart meters will bring an end to estimated billing-you will only
be billed for the energy you actually use, helping you budget
better
· Easier switching-smoother and faster to switch suppliers to get the
best deals
9. It is apparent that an important consideration driving the rollout of smart meters is the expectation that the programme will help the transition to a low-carbon economy and help reduce carbon emissions. These aims reflect the requirements set out in the EU’s Directive, which similarly speaks of the need to promote the ’sustainability’ of energy consumption.
10. Taking into account the potentially significant effect that the programme will have on the environment, the Commissioner has firstly considered whether the requested information is ‘environmental’ information and thus should properly have been dealt with under the EIR as opposed to FOIA. When considering this issue, it is necessary to turn to the definition of ‘environmental information’ provided by regulation 2(1) of the EIR.
11. In accordance with the Council Directive 2003/4/EC from which the EIR derives, it is the Commissioner’s view that the definition should be interpreted widely; an approach reinforced by the fact that regulation 2(1) states environmental information is “any information” on the factors and activities described at paragraphs (a) through (f). Importantly, it is not necessary for the information itself to have a direct effect on the environment or to record or to reflect an effect, in order for it to be environmental. On this basis, the Commissioner has invited DECC to consider whether the EIR might apply to any part of the withheld information.
12. In response, DECC has maintained that FOIA is the appropriate access regime. According to DECC, it is correct to say that one of the policy intentions of the programme is to assist with the efforts to reduce the overall usage of energy. However, DECC has advised that the report in question only concerns the area of policy development designed to facilitate and regulate the provision of communication and data services to electricity and gas meters. It therefore talks about the measures that are being taken to secure a system of communication and data services and the outcomes are not, in the view of DECC, outcomes which affect or would be likely to affect the elements of the environment. This contrasts with information relating to some other administrative and legislative activities connected to the smart meters programme – such as work required to choose and regulate different energy suppliers – which DECC considers is more likely to fall under the provisions of the EIR.
13. The Commissioner understands that it will not always be a simple process to decide under which access-regime a request should be handled. This is one of those occasions. However, he is satisfied on the basis of DECC’s explanation that FOIA and not EIR applies. The Commissioner has therefore gone on to consider the exemptions relied on by DECC to withhold information.’
24. In common with many explanations for decisions made by first-tier agencies, this passage is rather stronger on its narrative than it is on its reasoning.
The First-tier Tribunal’s proceedings, decision and reasons
25. At the hearing before the FTT, both DECC and the IC argued that FOIA was the applicable access regime. Mr Henney, on the other hand, contended that the EIR applied. The FTT recorded that, having retired to consider its decision, it “felt that the issue of the applicable access regime had not been very thoroughly considered by either DECC or the Commissioner. Both parties seemed to be operating on the basis that it was a ‘given’ that the appropriate access regime was indisputably FOIA” (FTT at paragraph [17]). It accordingly invited further written submissions on the point to be filed after the hearing by all parties.
26. The core of the FTT’s subsequent decision and reasoning was set out in paragraphs [19]-[25] (emphasis in the original):
‘19. All parties duly responded with further submissions. In their submissions both DECC and the Commissioner contended that the PAR could not be properly categorised as an economic analysis used within the framework of environmental measures and activities. The Commissioner did however concede that the matter was finely balanced. The Commissioner also helpfully referred the Tribunal to relevant First Tier Tribunal decisions whilst conceding that these were not binding. Mr Henney was again hampered by the fact that he did know the contents of the PAR although he asserted that as a matter of common sense the PAR must contain some form of economic analysis. DECC additionally submitted that the Tribunal should be considering only whether the data and communications component of the SMP as opposed to the whole SMP was a measure within Reg 2(1)(a) or (b) EIR but the Tribunal were unanimous in rejecting this as the correct approach.
20. The Tribunal noted in particular the decision in the Southwark v ICO and Lend Lease case (EA/2013/0162). Here the Tribunal ruled that a viability assessment in connection with a housing development fell to be considered under the EIRs. Of particular note are the following paragraphs:
29. We are inclined to agree with Mr. Pitt-Payne QC that there may be a tendency to overuse EIR; almost an assumption that, for example, anything to do with land or anything to do with the planning process in England and Wales is outside the scope of FOIA.
30. The answer to this tendency, it seems to us, is not the development of the vague notion of “remoteness”. Rather it lies in a purposive application to the facts of a case of the definition of “environmental information” in Reg 2(1) EIR. It may be for example that the phrase “the state of the elements of the environment” is not always given sufficient weight.
33. In our judgment the project is so large that it is likely to affect the state of the landscape as an element of the environment. The activity or programme, call it what you will, is therefore a measure which falls within subparagraph (c).
34. In our judgment it also cannot be doubted that the viability assessment including Appendix 22 is an economic analysis used within the framework of that measure and activity. By virtue of subparagraph (e) therefore, the information requested falls within EIR and not within FOIA.
21. The Tribunal noted the approach of the FTT in the Land Lease case - firstly it looked at the programme as a whole and whether that fell within the definition (para.33), even though that was not in itself the focus of the request. Secondly it decided that the viability assessment (which was the focus of the request) is a form of economic analysis used within the framework of that measure and activity and thus falls within part (e) of the definition.
22. The present Tribunal adopted a similar analysis in the DECC case. The Tribunal looked first at the smart meter project as a whole, and noted that it was agreed amongst the parties that this would have environmental benefits through things such as reduced carbon emissions. Therefore it can rightly be said to be a measure which is likely to affect the elements and factors of the environment. Following the Southwark approach, the PAR, being itself a form of viability report against policy objectives, is used within the framework of that measure. Consequently it has the potential to be covered by 2(1)(e) of the EIR.
23. DECC (and to a certain extent the Commissioner) contend that the PAR here is not within 2(1)(e) because PARs in general are not designed to be a form of economic analyses. However, in the Tribunal’s view, whether or not a PAR in general is not a form of economic analysis is irrelevant. In this case the PAR discusses two options, both of which have differing financial consequences. The government is already committed to expenditure on one option (the DCC model - £200m according to Mr. Walker’s open evidence); however through this PAR the government is examining whether an alternative approach (the Decentralised Model, “DM”) will have preferential benefits. The DM would not cost the same as the DCC given that there would be no need to procure either a data or communications provider. So even if there is no explicit mention of costs or benefits within the PAR, the exercise is itself an economic assessment given that the two possible outcomes have differing financial implications.
24. In response to DECC’s particular submissions – the Tribunal considered that the PAR was unquestionably used “within the framework” of the SMP. DECC contended that PARs in general are not designed to be used as part of such projects, (para.14(d) of their recent submissions). But the Tribunal considered that this PAR had a direct bearing on a fundamental element of the SMP and its overall cost. Therefore it must be right to say that the PAR has been used within the SMP policy “framework”.
25. The Tribunal also considered an additional approach or analysis: The Tribunal noted that DECC contends at para.9 of their latest submissions that even if the PAR can be regarded as a measure under EIR 2(1)(c), it does not or is unlikely to affect the state of the elements of the environment. The Tribunal noted that the PAR deals with the communications and data models, an essential and integral part of which is a 2G network. Such a network will produce emissions. Emissions from mobile telecommunications constitute environmental information, as per the findings in Office of Communications v IC & T-Mobile (EA/2006/0078):
27. ...The definition is not intended to set out a scientific test and its words should be given their plain and natural meaning. On that basis we believe that radio wave emissions that pass through the atmosphere from a base station to any solid component of the natural world are likely to affect one or more of the elements listed in subparagraph (a) or the interaction between some of them. Accordingly we conclude that the radiation from a base station falls within the meaning of the expression “environmental information”.
26. Consequently the Tribunal unanimously concluded that the applicable access regime is the EIR and not FOIA.’
27. Contrary to first impressions, there was no mis-numbering or formatting error in the extract in the passage above that the FTT cited from the earlier tribunal decision in Southwark v ICO and Lend Lease. The missing paragraph [31] simply cited the relevant parts of regulation 2(1) of the EIR while paragraph [32] stressed the sheer size of the redevelopment in question in that case, noting the developer was “essentially building an entire town centre at its own risk”. Similarly the IC’s Decision Notice had described the regeneration project as being “the biggest in western Europe” (at paragraph 7).
The parties’ positions (in outline) before the Upper Tribunal on this appeal
28. The parties’ respective positions before the Upper Tribunal can be summarised as follows. DECC argued that the FTT had erred in law in three respects – (1) as to the “measure” under consideration; and so the FTT had misapplied the definition in regulation 2(1) of the EIR; (2) in wrongly concluding that the PAR was an economic analysis within regulation 2(1)(e); and (3) in purporting to apply factual findings from the Ofcom case in the context of regulation 2(1)(c).
29. The IC resisted the appeal on ground (1); whilst the Commissioner himself had concluded in his Decision Notice that the appropriate access regime was FOIA, he did not accept now that the FTT had erred in law in deciding that the EIR applied. The Commissioner agreed in part with ground (2) and in full with ground (3). However, the IC contended that ground (1) was the crux of the case. If the FTT had not erred in law in that regard, then any errors it may have committed in connection with grounds (2) and (3) were simply not material. Assuming for the moment that ground (1) failed, all parties agreed with that analysis of the consequences.
30. Mr Henney supported the FTT’s decision and reasoning on the preliminary issue in all material respects and so opposed DECC’s appeal on each of its three grounds.
The Upper Tribunal’s analysis: DECC’s first ground of appeal
Introduction
31. My analysis of DECC’s first and indeed primary ground of appeal proceeds through the following four stages: (i) the general principles of construction of the EIR; (ii) a review of relevant previous FTT decisions on regulation 2; (iii) the parties’ submissions on the decision reached by the FTT in this case as regards the relevant “measure” within regulation 2(1)(c); and (iv) my conclusions, with reasons, on those submissions.
The general principles of construction for the Environmental Information Regulations
32. There was a fair amount of common ground between the parties as to the proper approach to be taken to the interpretation of regulation 2(1) of the EIR. The following general principles were not in dispute.
33. First, regard must be had to the European and international antecedents of the EIR. Thus the definition of “environmental information” in regulation 2(1) of the EIR must be construed in compliance with Directive 2003/04/EC which follows, but also expands upon, the definition of that term in the Aarhus Convention. Recital 5 of the Directive makes it plain that the EU legislature intended to ensure that EU law was compatible with the Aarhus Convention in terms of the right of access to environmental information.
34. Second, while The Aarhus Convention: An Implementation Guide does not have binding force, courts and tribunals are entitled to have regard to its guidance (Solvay and Ors v Région wallone Case C-182/10 at [27]). The Implementation Guide advises that “The clear intention of the drafters … was to craft a definition that would be as broad in scope as possible, a fact that should be taken into account in its interpretation” (p.50). Conversely, just as the definition of “environmental information” must be read broadly, the exceptions in regulation 12 of the EIR must be construed restrictively (see Article 4(4) of the Aarhus Convention and Recital (16) to the Directive).
35. Third, that emphasis on a broad interpretation of the expression “environmental information” has been echoed in the domestic case law; see e.g. Secretary of State for Communities and Local Government v Venn [2014] EWCA Civ 1539 at [10]-[12] per Sullivan LJ, citing the CJEU’s judgment in Lesoochranárske VLK v Slovenskej Republiky (Case C-240/09) [2012] QB 606 ("the Brown Bear case").
36. Fourth, although the expression “environmental information” must be read in a broad and inclusive manner, one must still guard against an impermissibly and overly expansive reading that sweeps in information which on no reasonable construction can be said to fall within the terms of the statutory definition. The CJEU, dealing with the earlier Directive 90/313, held that it was not intended “to give a general and unlimited right of access to all information held by public authorities which has a connection, however minimal, with one of the environmental factors mentioned in Article 2(a). To be covered by the right of access it establishes, such information must fall within one or more of the three categories set out in that provision” (Glawischnig v Bundesminister für Sicherieit und Generationen, Case C-316/01, at [25]).
37. There is also a general acceptance that the principle set out in Glawischnig applies equally to Directive 2003/04/EC as to its predecessor (see Opinion of Advocate-General Kokott in Stichting Natuur en Milieu (Environment and consumers) [2010] EUECJ C-266 at [44] and [58] and also Evans v Information Commissioner (Correspondence with Prince Charles in 2004 and 2005 [2012] UKUT 313 (AAC) at paragraph 235(1) and (2)). As another FTT has neatly put it, one must avoid imperilling “the principle of legal certainty by extending the meaning of words beyond their normal meanings” (Uttlesford at paragraph [28]). In that context it may be significant that the examples given in regulation 2(1)(a)-(c) are illustrative only, rather than exhaustive (referring to the elements of the environment, factors and measures respectively, “such as ….”; see further Implementation Guide p.50). In contrast, the categories of environmental information set out in regulation 2(1)(d) and (e) are expressed in self-contained terms which are essentially parasitic on the preceding provisions.
The previous decisions of the First-tier Tribunal on regulation 2 of the EIR
Introduction
38. I start with the cautionary and self-evident observation that first instance FTT decisions do not carry the status of being legal precedents. They are not strictly a source of case law, however persuasive they may be. Such decisions are fact-specific and it is axiomatic that one FTT decision cannot bind another FTT. Furthermore, there is an ever present danger that engaging in too close an analysis of previous FTT decisions “can elevate issues of fact into issues of law or principle” (see Camden LBC v Information Commissioner [2012] UKUT 190 (AAC) per Judge Jacobs at [20]).
39. That said, in the present appeal both the FTT and the Upper Tribunal received detailed submissions on two particular previous decisions by tribunals which bear on the proper interpretation and application on aspects of regulation 2 of the EIR. The two cases in question are Mersey Tunnel Users Association v Information Commissioner and Halton Borough Council (EA/2009/0001) (or “Mersey Tunnel Users Association”) and London Borough of Southwark v Information Commissioner, Lend Lease and Glasspool (EA/2013/0162) (or “Lend Lease”). These cases are said to illustrate the “bigger picture” approach to the interpretation of regulation 2(1)(c). This approach was explained by Mr Hopkins in his skeleton argument in the following terms (at §19(viii)):
‘When identifying the relevant “measure” for the purposes of regulation 2(1)(c) EIR, it is permissible to look beyond the precise issue with which the disputed information is concerned and to take account of the ‘bigger picture’. The Commissioner accepts, however, that there must be a sufficient connection between the ‘big picture’ and the particular information in dispute, such that the latter satisfies the definition under regulation 2(1) EIR.’
Mersey Tunnel Users Association
40. In Mersey Tunnel Users Association the disputed information was principally commercial and financial in nature, being a tolling analysis for the Mersey Tunnel redevelopment. The IC took the view that the disputed information was covered by the EIR, not FOIA, but the local council disagreed, arguing that the information requested (the tolling data) was too remote to fall within regulation 2(1)(c). The IC’s justification for concluding that EIR applied, as set out in the Decision Notice (FER0138940), is worth citing as it was subsequently relied on by the tribunal:
‘28. The Commissioner has examined all of the withheld information and is of the opinion that it falls within the definition of environmental information set out in regulation 2(1)(c), namely that information relating to the building of a new bridge and to tolling is information on a measure which is likely to affect the elements of the environment referred to in regulation 2(1)(a), in particular the land and the landscape. Building a new bridge inevitably changes the landscape. The proposed bridge would also affect the use of land, as it is intended to divert traffic away from the existing bridge and on to the new one.
29. The Commissioner does not agree that, at the time of the request, the possibility of introducing a congestion charging scheme and building a bridge was too remote to fall within the definition of environmental information. Instead, he has taken into account Council Directive 2003/4/EC (derived from the Aarhus Convention) which is implemented into UK law by way of the EIR. The Directive sets out that one of the purposes of the legislation is to allow the participation of the public in environmental decision making at the earliest stages. This has been interpreted as meaning that information which would help the public contribute to the preparation of a plan which is likely to have an affect on the environment should be dealt with as a environmental information under the EIR. The Commissioner’s view is that the fact a plan may not come to fruition does not prevent the information from being environmental. Further, the Commissioner has interpreted the phrase “information on” widely, to include information which is “about” a particular measure or activity. For these reasons, the Commissioner considers that all of the withheld information falls within the definition of environmental information as set out in Regulation 2(1) of the EIR and should properly have been considered as a request for information under that legislation.’
41. The tribunal in that case agreed with the IC that the EIR was the relevant access regime. Its central reasoning was as follows:
‘67. It appears to us, therefore, that there is no dispute that the Mersey Gateway Project will have a significant impact on the state of elements of the environment, such as, at least, the land and the landscape, and on factors such as emissions, discharges and other releases into the environment, affecting or likely to affect the elements of the environment referred to, such that information relating to it would fall squarely within the definition of environmental information under Regulation 2(1) of the EIR. The question for us is whether information on “tolling” of the Mersey Gateway Project would also fall within that definition.
68. We considered the statement of Mr Nicholson and, in particular, his evidence that by “March 2006, the Government approved the terms under which the Mersey Gateway Project could be funded and these made it clear that the only way in which it could be delivered was by tolling both the proposed new bridge and the Silver Jubilee Bridge.”
69. We are satisfied that tolling is an integral part to the Project and its viability. We agree with the Commissioner’s reasoning and conclude that the information requested falls within the definition of environmental information set out in Regulation 2(1)(c) EIR.’
42. It is therefore plain that Mersey Tunnel Users Association was decided on the basis it was a regulation 2(1)(c) case. There appears to have been no appeal to the Upper Tribunal against the FTT’s decision in that case. Indeed, its decision that the disputed information was “environmental information” seems eminently sustainable on the facts. However, I also observe that in its decision on the present case under appeal the FTT made no direct reference to the Mersey Tunnel Users Association case (to be precise, it is listed in the preamble to the reasons in a list of other “cases considered”, but is not discussed at any point in those reasons).
Lend Lease
43. In contrast the FTT in the instant appeal made extensive reference to the approach taken by the tribunal in Lend Lease, which concerned a major housing redevelopment at the Elephant & Castle in South London. In that case the requester had asked for a copy of the “financial viability assessment” submitted with the relevant planning application. The public authority (the London Borough of Southwark) released much of that viability assessment but also redacted significant elements. The viability assessment itself was accompanied by a number of appendices, which for the most part provided the evidence to support the main text. However, Appendix 22 was different in that it comprised the financial model used by the developer as an analytical tool on such major projects.
44. At the outset Southwark treated the request as falling under the EIR, and the IC’s Decision Notice proceeded on that unchallenged assumption (FER0461281). Subsequently the public authority changed tack, arguing (supported by the developer) that FOIA was the appropriate legal regime. The IC and the requester maintained that the EIR applied. The FTT in Lend Lease then addressed the issue as to the relevant regime in paragraphs [29] to [34] of its decision (the material passages being cited by the FTT in the present appeal at paragraph [20] of its decision; see paragraph 26 above). The kernel of that tribunal’s somewhat compressed reasoning occurs at paragraphs [33] and [34] of Lend Lease, where it said:
‘33. In our judgment the project is so large that it is likely to affect the state of the landscape as an element of the environment. The activity or programme, call it what you will, is therefore a measure which falls within subparagraph (c).
34. In our judgment it also cannot be doubted that the viability assessment including Appendix 22 is an economic analysis used within the framework of that measure and activity. By virtue of subparagraph (e) therefore, the information requested falls within EIR and not within FOIA.’
45. I pressed Mr Hopkins at the oral hearing as to how he read the FTT’s decision in Lend Lease. Acknowledging that the tribunal’s reasoning was “concise” (undoubtedly a more polite description than mine as to it being “compressed”), his answer was that the tribunal there had concluded that some of the disputed information was environmental information on regulation 2(1)(c) grounds (at [33]) whilst other information (most notably Appendix 22) was environmental information on the basis of regulation 2(1)(e) (at [34]). It was, accordingly, a two-stage or “belt and braces” decision applying both regulation 2(1)(c) and 2(1)(e). He maintained that position in his written submission after the hearing.
46. I must confess that at the hearing I found that answer rather persuasive. In the cold light of day I am much less convinced. My reasons are as follows. In Lend Lease the FTT’s analysis of the planning background (the discussion at paragraphs [15]-[20] of its decision) indicates that the viability assessment as a whole was a form of cost-benefit analysis. Appendix 22 to the viability assessment was a detailed economic model (see its decision at paragraph [23]). The FTT’s reasoning must be read against that fact-finding context. In particular, on further reflection I am satisfied that paragraphs [33] and [34] of the FTT’s reasons in Lend Lease must be read together. In paragraph [33] that tribunal was simply acknowledging that the sheer scale of the proposed redevelopment project meant that those building plans were “a measure which falls within paragraph (c)”, as they were bound to impact on the local landscape (and so affecting or likely to affect the factors in regulation 2(1)(a)). However, that was not an independent finding that the financial viability assessment (or any part of it) was itself any one of the types of measures listed by way of example in regulation 2(1)(c).
47. Rather, the tribunal in Lend Lease was using that finding as a building block on the way to its conclusion (at paragraph [34]) that the viability assessment as a whole was an economic analysis “used within the framework of the measures and activities referred to in (c)” (and so fell within regulation 2(1)(e)). The specific reference to Appendix 22 in paragraph [34] was simply for clarification and completeness, given the tribunal’s earlier recognition that it was of a rather different nature (see paragraph 43 above). Obviously the FTT’s decision should not be read as a statute. However, the formulation of its conclusion is telling, namely that “by virtue of subparagraph (e) therefore, the information requested falls within EIR and not within FOIA” – and not “by virtue of subparagraphs (c) and (e) therefore, the information requested falls within EIR and not within FOIA”.
48. I am therefore satisfied that Lend Lease was, on its facts and applying the law correctly, exclusively a regulation 2(1)(e) case; the references to regulation 2(1)(c) were necessary simply because of the cross-reference in regulation 2(1)(e) to subparagraph (c). It is not a pure “bigger picture” case in the same way that the Mersey Tunnel Users Association case was.
Conclusion
49. I reiterate that neither the Mersey Tunnel Users Association case nor Lend Lease is a formal precedent. They are both helpful “worked examples” of how regulation 2 of the EIR may be applied in practice. However, the two cases are not on all fours. The disputed information in Mersey Tunnel Users Association was found to be environmental information by virtue of regulation 2(1)(c), whereas the tribunal’s decision in Lend Lease was predicated on regulation 2(1)(e). This distinction is highly material in the context of the present appeal, as will be seen.
The parties’ submissions on the FTT’s decision on the relevant “measure”
50. As has already been noted, the only heads of the statutory definition of “environmental information” that were potentially in issue on the facts of this case were regulation 2(1)(c) and (e) of the EIR. For convenience they are repeated here:
‘(c) measures (including administrative measures), such as policies, legislation, plans, programmes, environmental agreements, and activities affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements;
…
(e) cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c);.’
51. At the FTT hearing, and in the submissions that preceded it, the dispute between the parties was framed in terms of the potential applicability of regulation 2(1)(c). Both DECC and the IC had argued that the disputed information (the PAR) was about the communication and data components of the SMP, and so at one remove from information on a policy likely to affect the environment within regulation 2(1)(c). Accordingly, both argued that FOIA, and not the EIR, was the appropriate access regime. Mr Henney, on the other hand, although at an obvious disadvantage in that he had not seen the redacted material, contended that the choice of the Data and Communications Company (DCC) model would affect the speed of the roll-out of the SMP and as such could delay the environmental benefits of smart metering, and so on his argument regulation 2(1)(c) was met (FTT decision at paragraphs [15] and [16]).
52. In the Upper Tribunal, Mr Choudhury, for DECC, argued that the FTT had impermissibly confused and conflated the requirements of regulation 2(1)(c) and (e). The high point of his submission was that the FTT had not actually reached a properly reasoned conclusion on the application of regulation 2(1)(c) at all. In that context Mr Choudhury drew attention in particular to the final sentence of the FTT’s decision at paragraph [19], recording (but dismissing) DECC’s submission that the FTT “should be considering only whether the data and communications component of the SMP as opposed to the whole SMP was a measure within Reg 2(1)(a) or (b) EIR” (see further paragraph 26 above). This, he argued, simply made no sense at all, not least as the concept of a “measure” is singularly absent from both sub-categories (a) and (b). Further, and more generally, Mr Choudhury submitted, the FTT’s decision was inadequately reasoned as regards the purported application of regulation 2(1)(c).
53. Mr Hopkins, for the IC, acknowledged that the FTT’s reasons in this case were “not especially elaborate”. His central argument was that the FTT had reached positive and separate conclusions that both the regulation 2(1)(c) and (e) limbs applied so as to bring the disputed information within the statutory definition of “environmental information”. As regards regulation 2(1)(c), he submitted that the FTT had adopted the “bigger picture” approach and had been entitled to find that the disputed information could properly be described as information on the SMP as a whole (which was a relevant “measure” within regulation 2(1)(c)). Although the IC had originally been with DECC on this point before the FTT, his position had now altered – the “bigger picture” analysis was permissible and the tribunal’s evaluation of what the disputed information is “on” was fact-specific, such that the Upper Tribunal should not readily interfere. Furthermore, in Mr Hopkins’s submission, if DECC’s Ground 1 failed, neither of the remaining two grounds needed to be considered, as on that basis alone the FTT’s decision was sustainable on regulation 2(1)(c), and any error made by the tribunal in applying regulation 2(1) as regards DECC’s other two grounds of appeal was immaterial.
54. Mr Frankel, for Mr Henney, supported Mr Hopkins’s submissions, at least as far as this first ground of appeal was concerned. Mr Frankel’s position was that the SMP as a whole, and not the communication and data services component, was correctly regarded as the relevant “measure” for the purpose of regulation 2(1)(c), as the latter was fundamental and integral to the former. He further argued that the FTT had correctly addressed itself to the applicability of regulation 2(1)(c). Furthermore, the simple insertion of a reference to regulation 2(1)(c) at the end of the third sentence of paragraph [22] (“Therefore it can rightly be said to be a measure which is likely to affect the elements and factors of the environment”) demonstrated that the FTT had indeed reached a definitive and sustainable conclusion on the application of regulation 2(1)(c) to the facts of this appeal.
The Upper Tribunal’s conclusions on the FTT’s treatment of regulation 2(1)(c)
55. I start by emphasising that I have read the whole of the FTT’s decision many times, with a view to trying to make sense of the passages which cause me difficulty, just as Judge Turnbull did in another case coincidentally involving the same Government department (Amin v IC and DECC [2015] UKUT 527 (AAC) at paragraph 37). It is trite law that an appellate court or tribunal must read the FTT’s decision and its reasons as a whole, and must guard against the temptation to seize on a single sentence out of context that might otherwise suggest the tribunal has misdirected itself in law.
56. To that extent I do not pay particular heed to the passage at the end of paragraph [19] of the FTT’s decision, which formed part of Mr Choudhury’s critique. Taken purely by itself, I view the reference in that sentence to regulation 2(1)(a) and (b) as evidence of a typographical error or elision in drafting rather than any conceptual confusion. I bear in mind that the various heads of the definition in regulation 2(1) internally reference each other – thus category (c) refers back to both categories (a) and (b) while category (e) in turn refers back to category (c). It is perfectly legitimate to read in words that have been accidentally omitted from the reasons a busy tribunal gives for its decision. What the FTT was doing at paragraph [19] was rejecting DECC’s submission that it “should be considering only whether the data and communications component of the SMP as opposed to the whole SMP was a measure [affecting or likely to affect matters] within Reg 2(1)(a) or (b) EIR” (with the underlined words added to make sense of the statement). The sentence then makes perfect sense, irrespective of whether or not Mr Choudhury agrees with the conclusion.
57. Accordingly, I did not find that the wording of paragraph [19] of the FTT’s decision materially advances Mr Choudhury’s argument to any significant degree. However, this was simply one part of Mr Choudhury’s wider submission, made on a number of fronts, that the FTT had failed to reach a properly reasoned determination that the disputed information fell within regulation 2(1)(c). Thus the critical issue is whether or not the FTT reached independently sustainable conclusions that both regulation 2(1)(c) and (e) applied. On this wider point, as regards the adequacy of the FTT’s reasoning as a whole, I am with Mr Choudhury for the following three reasons.
58. First, and again reading the FTT’s decision as a whole and in context, it is clear that the parties’ submissions at the original FTT hearing were focussed exclusively on the potential application of regulation 2(1)(c) alone (see paragraphs [15] and [16] of the FTT’s decision). The FTT further (and, in my view, accurately) considered that “the issue of the applicable access regime had not been very thoroughly considered by either DECC or the Commissioner” (at paragraph [17]). The FTT then, of its own initiative – and entirely consonant with its inquisitorial function – identified regulation 2(1)(e) as a possible alternative route, and sought further written submissions on that issue (see the detailed post-hearing directions set out at paragraph [18]). The FTT’s reasons (at paragraphs [19]-[24]) then addressed those further arguments about regulation 2(1)(e) in some detail before advancing an alternative and “additional approach or analysis” at paragraph [25], which raised an entirely separate issue. This referred to emissions from the smart meter network in the context of regulation 2(1)(c) (the subject of DECC’s third ground of appeal). The FTT then reached the unanimous conclusion that the applicable access regime was the EIR and not FOIA (at paragraph [26], where the conclusion is put in general terms, rather than by reference to either regulation 2(1)(c) or (e) or both). Reading that passage as whole – i.e. paragraphs [15] through to [25] – the frame of reference for the discussion shifts from regulation 2(1)(c) to regulation 2(1)(e), but with the afterthought of an “additional approach” applying regulation 2(1)(c) to the emissions. It follows that the conclusion at paragraph [26] flows naturally from the FTT’s principal reasoning that regulation 2(1)(e) applied, backed up by the “additional approach” in paragraph [25] applying regulation 2(1)(c). Here I should also anticipate my conclusion below (see paragraphs 70-73 below) that DECC’s Ground 3 succeeds.
59. Second, and attractively as though the arguments were put by both Mr Hopkins and Mr Frankel, I was ultimately not persuaded that the FTT did actually make a clear finding that regulation 2(1)(c) applied independently (beyond the “additional approach” in paragraph [25]), and irrespective of regulation 2(1)(e). Mr Hopkins observed in the course of oral argument that the FTT had “expended most of its explanatory energy” on regulation 2(1)(e), but there was, he submitted, sufficient to show that it had addressed its mind to the proper test under regulation 2(1)(c) and reached a positive conclusion on that point. Mr Frankel put the same point in a slightly different way, noting that the FTT had expressly referred to regulation 2(1)(c) at paragraphs [15], [18] and [25] of its decision. However, the discussion at paragraph [15] was premised on the hypothesis that the disputed information was about the roll-out of the SMP as a whole and occurred before the FTT switched the focus to regulation 2(1)(e). The references to regulation 2(1)(c) in paragraph [18] are all made in the context of the drafting of regulation 2(1)(e), which draws in “cost benefit and other economic analyses and assumptions” but only where they are “used within the framework of the measures and activities referred to in (c).” Mr Frankel’s proposed insertion of a reference to regulation 2(1)(c) at the end of the third sentence of paragraph [22] does not save the FTT’s decision, as that is simply a stepping stone to the FTT’s conclusion that regulation 2(1)(e) applied. Finally, the treatment of regulation 2(1)(c) in paragraph [25] was part of a stand alone discussion on the narrow point about emissions (on which see DECC’s successful Ground 3).
60. Third, and crucially, the FTT drew a direct analogy with Lend Lease case. As has been explained above, I accept Mr Choudhury’s submission that on any proper reading that was actually a regulation 2(1)(e) case. The two-stage test involved identifying the measure or programme as a whole, even though that was not the subject matter of the request (the housing redevelopment project in Lend Lease and the SMP in the DECC case), and then considering whether the disputed information (the viability assessment in Lend Lease and the PAR in the DECC case) was a form of economic analysis used within the framework of that overarching measure or programme. This involves using “the measures and activities referred to in (c)” as a building block and reference point for the economic analysis which is the primary focus of regulation 2(1)(e). It did not involve a clear and unequivocal finding that the PAR itself was a measure “affecting or likely to affect the elements and factors referred to in (a) and (b)” such that regulation 2(1)(c) applied. It is furthermore noteworthy the FTT in this case made no specific reference to the Mersey Tunnel Users Association case (other than in its list of “cases considered”), which involved the direct application of regulation 2(1)(c). There may well be a “bigger picture” argument, by analogy with Mersey Tunnel Users Association, but that is not the way that the FTT reasoned its decision.
61. I therefore uphold DECC’s first ground of appeal to the extent that I find the FTT erred in law by failing to explain adequately its reasoning as to the application of regulation 2(1)(c) of the EIR.
The Upper Tribunal’s analysis: DECC’s second ground of appeal
62. Had DECC’s first ground of appeal been unsuccessful, there would have been no need to address its second and third grounds of appeal. This would be because on that hypothesis the FTT’s decision applying regulation 2(1)(c) would have stood on its own terms, so any error of law it may have committed with regard to the other two freestanding challenges would have been immaterial to the outcome. However, DECC’s first ground has been upheld. It follows that the other two grounds of appeal need to be considered.
63. DECC’s second ground of appeal is that the FTT erred in law by concluding that the PAR was an “economic analysis” within the meaning of regulation 2(1)(e), even if there is no explicit mention of costs or benefits within it, simply because the PAR considers “two possible outcomes [having] differing financial implications” (at [23]). Regulation 2(1)(e), it will be recalled, extends the definition of “environmental information” to include “(e) cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c).” Mr Choudhury developed three arguments in support of this ground of appeal, each of which alone he submitted made good this challenge. Mr Hopkins, for the Commissioner, disagreed with the first and third arguments but, after some initial resistance, accepted the force of the second argument. Mr Frankel resisted this second ground of appeal on all fronts.
64. Mr Choudhury’s first argument was that as the FTT had erred in identifying the correct “measure” for the purposes of regulation 2(1)(c) – i.e. in focussing on the SMP rather than on the PAR – it necessarily followed that its conclusions under regulation 2(1)(e) were also wrong. I accept this argument to the extent that I have allowed the appeal on the first ground; if the reasoning in relation to regulation 2(1)(c) was flawed, it follows that the FTT’s decision under regulation 2(1)(e) is likewise infected by the same error.
65. Mr Choudhury’s second argument was that the FTT had fallen into error by concluding that a report with any financial implications would be sufficient to render it an “economic analysis” for the purposes of regulation 2(1)(e). The FTT dealt with this issue at paragraph [23] of its decision. In my view the FTT was right to start from the proposition that what mattered was whether this particular PAR constituted an economic analysis, and not what the position is with PARs in general. However, the FTT then continued as follows (emphasis added):
‘23. … In this case the PAR discusses two options, both of which have differing financial consequences. The government is already committed to expenditure on one option (the DCC model - £200m according to Mr. Walker’s open evidence); however through this PAR the government is examining whether an alternative approach (the Decentralised Model, “DM”) will have preferential benefits. The DM would not cost the same as the DCC given that there would be no need to procure either a data or communications provider. So even if there is no explicit mention of costs or benefits within the PAR, the exercise is itself an economic assessment given that the two possible outcomes have differing financial implications.’
66. My initial response to Mr Choudhury’s argument was that this was all a question of fact and degree, and that the conclusion reached by the FTT was one that was within its province as the primary fact-finder. However, like Mr Hopkins, as the arguments have unfolded I have been brought round to agreeing with Mr Choudhury’s second argument (at least in part). Mr Choudhury argues that in failing to consider whether the primary or substantial purpose of the PAR was a cost benefit or other economic analysis, the FTT effectively ignored the requirements of regulation 2(1)(e). Given the general principles of construction that apply in the application of the EIR, it may be that Mr Choudhury seeks to put the bar too high. I did not hear detailed argument on this specific point and would be reluctant to go as far as Mr Choudhury, not least as it is arguable his “primary or substantial purpose” test may impose too demanding a hurdle in the light of the philosophy of the Aarhus Convention.
67. However, it does seem clear to me that the FTT set the bar too low in the italicised passage (above) in paragraph [23] of its decision. The FTT appears to be saying in terms that simply because the two outcomes have different financial implications, then that is enough to make the PAR an economic analysis. That seems to be going some way further than the type of document envisaged by the Implementation Guide, which justifies the inclusion of this sub-category of information on the basis that “the quantification of environmental values and the ‘internalization’ of environmental costs are among the most difficult questions for economists. It is therefore also important to be able to analyse the assumptions behind economic modelling used in environmental decision-making” (Second Edition, 2014, p.54). True, the EIR must be read broadly, but at the same time one must avoid imperilling “the principle of legal certainty by extending the meaning of words beyond their normal meanings” (Uttlesford at paragraph [28]). The FTT in the present case did just that. I also bear in mind that whereas regulation 2(1)(a)-(c) are expressed in open terms, regulation 2(1)(e) uses closed language. The mere inclusion of some financial data in the PAR, or reference to different financial outcomes, is not sufficient of itself to bring the document within the category of “cost-benefit and other economic analyses and assumptions” for the purposes of regulation 2(1)(c).
68. Mr Choudhury’s third argument was that the FTT had erred by finding that the PAR was “unquestionably used ‘within the framework’ of the SMP” (at paragraph [24]). This challenge is not persuasive. I agree with Mr Hopkins and Mr Frankel that the expression “within the framework” is phrased in broad terms and so should not be construed in a restrictive manner, given the general principles of construction that apply to the EIR. It matters not, therefore, that the PAR is a separate mechanism to the SMP, managed by an independent authority. What matters is whether or not the PAR contained “information … on” (to quote the opening words of regulation 2(1)) the relevant type of analysis, in this instance “cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c)” (within the terms of regulation 2(1)(e)). The FTT (subject to the second argument above) was perfectly entitled to conclude that the PAR had been used within the SMP policy “framework”.
69. It follows that Mr Choudhury has made good his second ground of appeal, at least with regard to two of the three arguments that he prayed in aid, even if I would not put the point quite in the way that he has. To that extent the second ground of appeal also succeeds.
The Upper Tribunal’s analysis: DECC’s third ground of appeal
70. I can deal with this third ground, which concerns regulation 2(1)(c), relatively shortly. Towards the end of its decision, the FTT considered what it described as “an additional approach or analysis”. In doing so, the FTT found “that the PAR deals with the communications and data models, an essential and integral part of which is a 2G network. Such a network will produce emissions. Emissions from mobile telecommunications constitute environmental information, as per the findings in Office of Communications v IC & T-Mobile (EA/2006/0078)” (at paragraph [25]). The only way to read this paragraph is that the FTT concluded, independently of any other arguments, that the PAR fell within regulation 2(1)(c) as the network adopted would produce radio emissions and those in turn would be likely to affect one or more of the elements listed in sub-paragraph (a).
71. Mr Choudhury attacked this finding on two fronts. First, he submitted the finding was procedurally unfair, as the matter had not been raised in argument before the FTT at any stage. Second, he argued the FTT simply had no evidence before it on which to reach such a conclusion. Mr Hopkins agreed with the second limb of Mr Choudhury’s argument and accepted the FTT had erred by borrowing a fact-specific conclusion from a different FTT case. Mr Frankel argued the FTT was entitled to reach the conclusion it had, not least as the FTT’s decision in what is now generally referred to as the Ofcom case had gone unchallenged to the Supreme Court (and indeed beyond to the CJEU).
72. I do not need to address Mr Choudhury’s procedural challenge. This is because I accept his other argument on this third ground of appeal – the FTT cannot simply borrow a factual conclusion from one case (relating to radio wave emissions from mobile phone masts) and apply it in another case (concerning radio emissions in smart meter networks) without further ado. It has to have some evidential basis on which to act. The issue is not whether smart meter networks produce emissions. The issue is not whether such emissions are safe. The issue is whether there was evidence the PAR amounted to information on “measures … affecting or likely to affect the elements and factors referred to in (a) and (b)” within regulation 2(1)(c) with regard to radio emissions. It is not obvious there was any evidence before the FTT to that effect.
73. However, I also agree with Mr Hopkins and Mr Frankel that had DECC’s first ground of appeal been unsuccessful, it would have been entirely possible in effect to sever paragraph [25] from the rest of the FTT’s decision. Accordingly the FTT’s error of law on this point was entirely immaterial in any event.
A summary of the Upper Tribunal decision on appeal from the FTT’s decision
74. I therefore allow DECC’s appeal on all three grounds (albeit not for all the reasons that have been advanced by Mr Choudhury). I accordingly set aside the decision of the FTT as erroneous in law.
Where we go from here
75. In these circumstances I have three options. I could leave the FTT’s decision intact; however, that would not be appropriate given the successful grounds of challenge. Alternatively I can remit the appeal for re-hearing or re-make the decision myself. I invited the parties’ submissions on these two options both at the oral hearing and subsequently in writing in response to my further directions.
76. In the event that DECC’s appeal on the first ground was successful, Mr Hopkins, for the Commissioner, expressed a “marginal preference” for the matter remaining with the Upper Tribunal, insofar as the preliminary issue of “FOIA or EIR” was concerned. He raised the possibility that I might invite further evidence on the question of whether or not the disputed information is “environmental”.
77. On that same scenario Mr Choudhury, for DECC, stated a clear preference for the preliminary issue to remain with the Upper Tribunal. This was because I had had the benefit of detailed submissions on the matter, I have access to the arguments made before the FTT, the issue involves complex points of law and such an approach would save costs and time.
78. Mr Frankel agreed with Mr Hopkins’s arguments. I surmised from the Upper Tribunal oral hearing that Mr Henney himself – entirely understandably – just wanted to get on with the substance of the matter, given the lengthy delays he has already experienced to date.
79. I am satisfied that it is right to re-make the FTT’s decision on the preliminary issue. This is the fair and just approach, not least as it is the most proportionate option. It reduces both costs and delays. I have considered Mr Hopkins’s guarded invitation to invite further evidence. Mr Choudhury did not take up that suggestion and I do not consider it is necessary, given the detailed evidence and submissions available to me.
The re-making of the First-tier Tribunal’s decision
80. I therefore proceed to re-decide the preliminary issue afresh as to whether the appropriate access regime is FOIA or the EIR.
81. I consider first whether the disputed information falls within regulation 2(1)(c) of the EIR. I remind myself that this category covers information on:
‘(c) measures (including administrative measures), such as policies, legislation, plans, programmes, environmental agreements, and activities affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements;’.
82. In applying this definition, I have reminded myself of the general principles of construction for the EIR, derived from the Aarhus Convention and the Directive, as set out above (at paragraphs 32-37).
83. As a matter of law, when identifying the relevant “measure” for the purposes of regulation 2(1)(c), I also find it is permissible to look beyond the precise issue with which the disputed information is concerned and to have regard to the “bigger picture”. This approach is consistent with a broad interpretation of the EIR as mandated by the Aarhus Convention and the Directive. In this context I also bear in mind that on a proper reading only the Mersey Tunnel Users Association case is a worked example of the pure “bigger picture” approach in the context of regulation 2(1)(c), given that Lend Lease was actually a regulation 2(1)(e) case.
84. Mr Hopkins correctly did not seek to argue that Mersey Tunnel Users Association is in any way binding on me or that the “bigger picture” approach is itself determinative. Put another way, the IC accepted that even though a large complex project may have obvious environmental effects, it does not necessarily follow that information on every component of the project will be environmental information within the regulation 2(1)(c) definition. In his skeleton argument, Mr Hopkins framed the “bigger picture” test in these terms:
‘There must be a sufficient connection between the ‘big picture’ and the particular information in dispute, such that the latter satisfies the definition under regulation 2(1) EIR. The component to which the disputed information relates must play a sufficiently important role in the large project and in the environmental aspects of that project’ (Commissioner’s skeleton argument at §20).
85. Mr Frankel put the point in similar terms, arguing that the question was “whether there is sufficient connection between the requested information, which may not expressly refer to environmental matters, and the project with clear environmental implications to which it relates” (Mr Henney’s skeleton argument at §32).
86. Mr Choudhury, on the other hand, submitted that the “bigger picture” approach, even if permissible in some cases, was entirely inappropriate in the circumstances of the present case – in effect, the decisions and reasoning in Mersey Tunnel Users Association (and in Lend Lease), although in no sense binding, could in any event be distinguished on their material facts. This submission was premised on four contentions, none of which I found to be persuasive.
87. The first was that Mersey Tunnel Users Association and Lend Lease both involved major infrastructure construction projects with obvious, immediate and significant environmental impacts. In contrast, the SMP did not involve large scale earthworks or other environmentally disruptive steps. I simply do not accept this distinction is material. It is rather like saying that Donoghue v Stevenson [1932] AC 562 is a case solely about the liability of Scots drinks manufacturers for decomposing snails found in opaque bottles containing ginger beer, rather than about the duty of care and the neighbour principle in the tort of negligence more generally (see e.g. W. Twining and D. Miers, How to Do Things with Rules, 5th edn (2010), pp.307-09 and 386-87). In effect, Mr Choudhury is inviting me to adopt an exclusively “bricks and mortar” or “cranes on the skyline” understanding of projects that have a potential environmental impact. In my judgment such a narrow approach is wholly inconsistent with the objectives and philosophy of both the Aarhus Convention and the Directive.
88. Mr Choudhury’s second contention was that the very fact that the SMP itself is, as he put it, “not inherently about the environment” is sufficient to preclude the “bigger picture” approach. I readily accept that the objectives of the Electricity Directive are principally concerned with improving the EU internal market by both enhancing competition amongst suppliers as well as improving information flows to consumers. This much is plain from an inspection of the Recitals to the Electricity Directive. As Mr Choudhury explained, the SMP is “not primarily about saving the planet”. Rather, reducing carbon emissions is an aim rather than the aim of both the Electricity Directive and the SMP. However, the test under regulation 2(1) is not framed in terms of the primary intention or motivation behind the measure in question. Rather, the legislation focuses on the more pragmatic question as to whether the disputed information is “on” any of the various matters enumerated in categories (a) to (f) inclusive.
89. Third, Mr Choudhury argued that the PAR is only concerned with the communications and data component of the SMP. Accordingly, it does not contain information about the state of the elements within category (a) of the definition in regulation 2(1) or measures affecting or likely to affect those elements within category (c). However, this argument only works if the “bigger picture” approach is not legitimately available. If the “bigger picture” approach is, as a matter of law, properly open to a tribunal, then the evaluation as to what the disputed information was “on” becomes factually and contextually sensitive.
90. Fourth, and finally, Mr Choudhury noted that the PAR was not a viability assessment of the SMP as a whole. Instead, he reiterated, the PAR was devoted to one aspect of the SMP, the communications and data component. This is really another way of putting the previous point and faces the same difficulty, namely that the evidence may justify a finding, as part of the bigger picture, that the PAR was information “on” the SMP, which itself was clearly a measure within regulation 2(1)(c).
91. Mr Choudhury argued that both respondents’ arguments involved the misconceived notion that simply because a project has some environmental impact, then all information concerned with that project must necessarily be environmental information. I did not understand either Mr Hopkins or Mr Frankel (or indeed Mr Henney) to adopt that extreme position. Indeed, they all expressly disavowed any such suggestion. They all accepted there are clearly going to be some types of information that are relevant to a project (which itself has some environmental impact) but which do not amount to environmental information within regulation 2(1). An example – not one canvassed in argument, but one which I simply suggest here by way of illustration – is that a report exclusively focussed on the public relations and advertising strategy to be adopted for the SMP is unlikely to come within the scope of the regulation 2(1) definition (which would mean, of course, that any such request fell for consideration under FOIA).
92. So, given that the “bigger picture” approach is permissible, does it actually lead to the conclusion that the PAR is environmental information within regulation 2(1)(c) of the EIR? My conclusion is that it does. The argument proceeds in a series of stages.
93. First, regulation 2(1)(c) must be liberally construed whilst not losing sight of the statutory language. The SMP itself is on any reckoning a “measure”, in that it is a plainly a policy, plan or programme in the ordinary meaning of those terms.
94. Second, the next question is whether the SMP is a measure “affecting or likely to affect the elements and factors referred to in (a) and (b)”, those elements and factors respectively including “air and atmosphere” and “energy” (given as examples of an “element” and a “factor” in regulation 2(1)(a) and (b)). Certainly, the SMP is intended to implement the Electricity Directive by providing consumers with objective and transparent consumption data (see Recitals (50) and (55)). I acknowledge that the Electricity Directive is primarily driven by competition policy. However, it clearly has an impact on energy policy and on the environment. According to Recital (5), “a secure supply of electricity is of vital importance for the development of European society, the implementation of a sustainable climate change policy, and the fostering of competitiveness within the internal market” (emphasis added). DECC’s own Impact Assessment of the “Smart meter rollout for the small and medium non-domestic sector” (30 March 2011) also made several references to the role of the SMP in reducing CO2 emissions and hence helping to deliver environmental policy objectives. It was common ground between Mr Hopkins and Mr Choudhury that “likely” on the context of regulation 2(1)(c) does not impose the relatively high threshold of the balance of probabilities. Rather, it denotes something more substantial than a remote possibility. For the reasons above, on that basis the SMP is at the very least “likely to affect the elements and factors referred to in (a) and (b)”.
95. Third, I acknowledge that the argument so far has been focussed on the SMP, and not the PAR. The primary focus of the PAR is the communications and data components of the SMP. I shall assume for the present – although I recognise too that Mr Henney disputes this – that the PAR itself is not a measure “affecting or likely to affect the elements and factors referred to in (a) and (b) above”. However, the definition of “environmental information” in regulation 2(1) must be read in its entirety. It includes “any information … on” any of the matters enumerated in sub-paragraphs (a) to (e) inclusive. Taking a broad view of the regulation, and bearing in mind the “bigger picture”, it is accurate to say that the PAR is information on the SMP as a whole, which (as noted above), is plainly a relevant measure for the purpose of regulation 2(1)(c). In reaching this conclusion I find that the PAR does not contain information on some incidental aspect of the SMP that could be easily hived off. The communications and data system underpinning the SMP is integral to its success. As Mr Hopkins pointed out, the official consultation paper described the DCC, the government’s chosen vehicle for delivery of the data and communications component, as a “key element” in the rollout of the SMP. The establishment of the DCC’s services were likewise said to be “critical to the success” of the programme as a whole (DECC, Smart Metering Implementation Programme: A consultation on the detailed policy design of the regulatory and commercial framework for DCC (2011), FTT open bundle, p.232). As Mr Frankel put it rather more bluntly, but equally accurately, without the communications and data system there is no SMP.
96. I therefore conclude that the contents of this PAR, with its focus on the communications and data component, is sufficiently closely connected to the success of the SMP overall. Furthermore, the SMP’s objectives include relevant environmental impacts. The disputed information accordingly falls within regulation 2(1)(c).
97. Having decided that the PAR falls within the definition of “environmental information” in regulation 2(1)(c) of the EIR, I do not need to consider whether it also qualifies for that description by virtue of regulation 2(1)(e).
Two other matters
98. There are two other matters I should touch on simply by way of clarification.
99. First, Mr Henney has made it clear throughout these proceedings that one of his prime motivations for arguing that the disputed information is covered by the EIR, and not FOIA, is his concern that if the relevant regime is FOIA, and the FTT confirms the IC’s Decision Notice and orders disclosure of the unredacted PAR, then the Government will resort to use of the veto under section 53 of FOIA. There is no equivalent to the section 53 veto power in the EIR. Moreover, of course, it is one of the peculiarities of information rights law that whereas the veto may be invoked in appropriate circumstances under FOIA, it has no place under the EIR regime, being inconsistent with the Directive (R (on the application of Evans) v Attorney-General [2015] UKSC 21 at paragraphs [108] and [113]). However, I have not had regard to Mr Henney’s concern. The preliminary question of whether the disputed information is “environmental information” within regulation 2 of the EIR is a logically and legally prior question to whatever may be the possible response by one of the parties to such a ruling, once the appellate process has played out.
100. Second, I make the obvious point that the substantive issue in these proceedings – was the IC correct in ordering disclosure of the bulk of the disputed information? – remains at large. The FTT did not get to the point where it had to resolve that issue, and I heard no evidence or submissions on that question. Therefore I am in no position to comment on the merits of the underlying appeal (which includes the contentious ‘safe space’ issue under section 35 of FOIA), other than to observe that while in practical terms Mr Henney may have won this particular battle, the war apparently goes on.
Conclusion
101. I therefore conclude that this appeal by DECC on the preliminary issue must be allowed. However, I remake the FTT’s decision in terms which arrive at the same practical outcome. The disputed information is governed by the EIR (by virtue of regulation 2(1)(c)) and not by FOIA. The underlying appeal is remitted to the FTT for consideration of the substantive issues arising under the EIR.
102. It is not appropriate for me to seek to micro-manage the remitted proceedings. I therefore invite the parties to propose further directions for the onward conduct of this appeal, which should be sent to the office of the First-tier Tribunal (General Regulatory Chamber) within one month of the date of issue of this decision. I make no direction as to whether the substantive FTT appeal should be heard by the same panel as before or by an entirely fresh panel. That is a decision best left to the good judgment of the Chamber President of the General Regulatory Chamber of the First-tier Tribunal, who will be able to take into account the availability of the original panel and all other relevant considerations. The parties may wish to make observations on panel composition in their proposals for further directions.
Signed on the original Nicholas Wikeley
on 07 December 2015 Judge of the Upper Tribunal