Upper Tribunal Decision: FOIA Governs Requested Information

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 1 IN THE UPPER TRIBUNAL Appeal No. GIA/224/2016 ADMINISTRATIVE APPEALS CHAMBER

Before: Upper Tribunal Judge K Markus QC

The decision of the Upper Tribunal is to allow the appeal.

The decision of the First-tier Tribunal (General Regulatory Chamber) (Information Rights) made on 4th August 2015 under number EA/2014/0123 was made in error of law. The First-tier Tribunal’s decision is set aside.

The Upper Tribunal re-makes the decision as follows:

“The requested information is not “environmental information” within the meaning of the EIR and the request is governed by the FOIA.

[NAME]’s appeal against the [NAME]’s decision is to be determined by the First-tier Tribunal in accordance with the Directions below.”

The decision is given under section 12(2)(a) and (b) of the Tribunals, Courts and Enforcement Act 2007.

Directions

1. The new First-tier Tribunal should not involve the tribunal judge or either of the two members who were involved in the decision dated 4th August 2015.

2. The new First-tier Tribunal should proceed on the basis that the Freedom of Information Act 2000 governs the request.

3. These Directions may be supplemented by later directions by a Tribunal Judge in the General Regulatory Chamber of the First-tier Tribunal.

REASONS FOR DECISION

Introduction 1. The Driver and Vehicle Standards Agency (”DVSA”) is an executive agency of the Department for Transport (“[NAME]”). DVSA’s role is “to improve road safety in Great Britain”. One of its functions is to perform safety tests of motor vehicles. 2. [NAME], was the owner of a [NAME] R. In November 2011 he submitted a safety defect report to DVSA’s predecessor, the Vehicle and Operation Services Agency (“[NAME]”), requesting that [NAME] investigate a potential throttle defect in his vehicle. [NAME] responded that the problem which he reported was not a safety defect and the case was closed. In April 2012, independently of the above request, DVSA test drove a [NAME] as a

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 2 result of receiving a safety defect report from someone other than [NAME]. It concluded that the vehicle was safe to drive. The information relating to the safety test is the disputed information with which this appeal is concerned.

3. DVSA (and formerly [NAME]) has no role in considering the environmental effects of motor vehicles, nor whether vehicle manufacturers have complied with environmental legislation. DVSA tests motor vehicles for safety. Other bodies, including the Vehicle Certification Agency ([NAME]) are responsible for testing for environmental compliance. 4. [NAME] believes that what he describes as the throttle defect, and which I will refer to more neutrally as the throttle characteristic, was deliberately created by [NAME] in the Cayman R so as to circumvent its legal obligations under both international and European environmental legislation on noise emissions. He says that the effect of that characteristic invalidated the noise emissions tests on that vehicle. [NAME] also believes that, in undertaking the safety test in April 2012, [NAME] tested a type of vehicle which would not have revealed any issue about compliance with environmental regulations in the case of the [NAME] R.

5. On 6 November 2013 [NAME] submitted the following request to [NAME]: “I request under the Freedom of Information Act all information held by [NAME] regarding the [NAME] and in particular the [NAME] safety evaluation of the vehicle throttle malfunction... Please supply all the information you have regarding the DVSA test of the [NAME] throttle malfunction with particular reference to: - the vehicle transmission - the vehicle exhaust system - the vehicle model (Cayman S or Cayman R) - the date and duration of the test - the vehicle VIN number - the location of the test - the gears used in the test - the vehicle speed when the test was performed

and all other test data and correspondence with the member(s) of the public who requested the test, government ministers, [NAME], other government departments and [COMPANY]. You may redact names where appropriate...” 6. On 8 November 2013 [NAME] wrote to [NAME] confirming that it held the information within the scope of the request but declining to provide it under section 44(1)(a) of the Freedom of Information Act 2000 (FOIA). [NAME] complained to the [NAME] who issued a decision notice on 7 May 2014 upholding [NAME]’s decision. Section 44(1)(a) of FOIA is an absolute exemption so that, where it applies, the information need not be provided regardless of any public interest considerations. There is no equivalent provision under the Environmental Information Regulations 2004 (EIR).

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 3 7. [NAME] appealed to the First-tier Tribunal (FTT). He argued that the request should have been considered under the EIR and not FOIA because his request “concerned an activity that directly affected the environment, namely an activity to regulate vehicle noise emissions”.

8. The FTT decided that all of the disputed information was “environmental information” within the meaning of regulation 2(1)(c) EIR. The FTT went on to decide that none of the exceptions in the EIR applied to the disputed information and ordered that it be disclosed.

9. I gave permission to appeal to the appellants and suspended the effect of the FTT’s decision pending determination of the appeal. [NAME] have acted as one in this appeal, arguing that the FTT erred in law in finding that the disputed information was “environmental information” within the EIR. [NAME] supports [NAME]/DVSA on that ground but also submits that, even if the EIR is the appropriate regime, the FTT made a number of errors of law in its application of that regime.

10. The [NAME] supports the appeals, agreeing with all the appellants that the FTT erred in its approach to environmental information. However the [NAME] is not at one with the appellants as to how I should approach any redetermination of the applicable regime, should I allow the appeal. I address those submissions below.

11. A hearing took place before me over two days. The [NAME] were represented by [NAME], the Third Appellant by [NAME] and the [NAME] by [NAME]. [NAME] acted in person. I am grateful to all for their helpful written and oral submissions. The Environmental Information Regulations 2004 12. Regulation 5(1) EIR sets out the general duty on a public authority that holds environmental information to make it available on request, subject to exceptions contained in Part 3 EIR. Section 39 of FOIA provides a qualified exemption for information under FOIA if the EIR applies. 13. “Environment information” is defined in Regulation 2 EIR as follows: “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);

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 4 (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…” (emphasis added)

14. The Directive referred to in this definition is Council Directive 2003/4/EC on public access to environmental information (“the Directive”). The Directive in turn gives effect to international obligations under the 1998 UN/ECE Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters (the “Aarhus Convention”). The definition of environmental information in regulation 2 EIR is the same as that in the Directive.

15. Recitals to the Aarhus Convention include: “citizens must have access to information, be entitled to participate in decision- making and have access to justice in environmental matters …” and: “improved access to information and public participation in decision-making enhance the quality and the implementation of decisions, contribute to public awareness of environmental issues, give the public the opportunity to express its concerns and enable public authorities to take due account of such concerns.” 16. The recitals to the Directive explain its purpose including, in the first recital: “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”.

17. In Department for Business, Energy and Industrial Strategy v Information Commissioner and Henney [2017] PTSR 1644 (“[NAME]”) at [14]-[17] the Court of Appeal set out two important legal principles to be followed in construing and applying the definition of “environmental information” in article 2(1) of the Directive and regulation 2(1) of the EIR.

18. First, the EIR must be interpreted as far as possible in accordance with the purpose of the Directive and the Aarhus Convention. [ADDRESS] drew attention to the recitals to both instruments to which I have set out above.

19. Second, although the term “environmental information” must be construed broadly, there are limits to the broad approach. Case C-316/01, [NAME] v [NAME] fūr Sicherheit und Generationen (13 June 2003) shows that the broad meaning to be given to the Directive does not mean that it is 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”.

20. The issue in [NAME] was, as the Court of Appeal explained at [6], “when and whether information on a measure which does not itself affect the state of the

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 5 elements of the environment or the factors referred to in regulation 2(1)(a) and (b) of the EIR, can be information “on” another measure which does.” 21. [ADDRESS] of Appeal observed (at [37]) that, although the definition of “information” in section 1(1) FOIA focusses on the information itself, the definition of “environmental information” in regulation 2(1)(c) of the EIR also focusses on the relevant measure. Therefore it is first necessary to identify the relevant measure and “Information is “on” a measure if it is about, relates to or concerns the measure in question.” 22. [ADDRESS] rejected a submission on behalf of the Department that the starting point should be the disputed information. It did not matter where the tribunal started as long as it correctly addressed the crucial question whether the disputed information was “on” a measure within regulation 2(1)(c). 23. [ADDRESS] of Appeal endorsed the statement by Upper Tribunal Judge Wikeley in the decision below that, in identifying the relevant measure, “it is permissible to look beyond the precise issue with which the disputed information is concerned.” [ADDRESS] said: “39. …This does not amount to a finding that it is permissible to look at issues with which the information is not concerned, or at issues with which the information is merely connected. It simply means that the Tribunal is not restricted by what the information is specifically, directly or immediately about. In my judgement, this is consistent with the language used in regulation 2(1)(c). Nothing in that language requires the relevant measure to be that which the information is “primarily” on” 24. There is no requirement that the information must be directly or immediately concerned with a measure which is likely to affect the environment. Moreover, it is possible for information to be “on” more than one measure: [42]. [ADDRESS] said: “43. It follows that identifying the measure that the disputed information is “on” may require consideration of the wider context, and is not strictly limited to the precise issue with which the information is concerned. It may be relevant to consider the purpose for which the information was produced, how important the information is to that purpose, how it is to be used, and whether access to it would enable the public to be informed about, or to participate in, decision-making in a better way. None of these matters may be apparent on the face of the information itself. It was not in dispute that, when identifying the measure, a tribunal should apply the definition in the EIR purposively, bearing in mind the modern approach to the interpretation of legislation, and particularly to international and European measures such as the Aarhus Convention and the Directive. It is then necessary to consider whether the measure so identified has the requisite environmental impact for the purposes of regulation 2(1).” 25. At [45] the Court agreed with the Upper Tribunal Judge that “simply because a project has some environmental impact” it does not follow that “all information concerned with the project must necessarily be environmental”. However, it is not

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 6 necessary that “the information itself must be intrinsically environmental”. [ADDRESS] then gave guidance as to how the line is drawn: “47. In my judgment, the way the line will be drawn is by reference to the general principle that the regulations, the Directive, and the Aarhus Convention are to be construed purposively. Determining on which side of the line information falls will be fact and context-specific. But it is possible to provide some general guidance as to the circumstances in which information relating to a project will not be information “on” the project for the purposes of section 2(1)(c) because it is not consistent with or does not advance the purpose of those instruments.

48. My starting point is the recitals to the Aarhus Convention and the Directive, in particular those set out at [15] above. They refer to the requirement that citizens have access to information to enable them to participate in environmental decision-making more effectively, and the contribution of access to a greater awareness of environmental matters, and eventually, to a better environment. They give an indication of how the very broad language of the text of the provisions may have to be assessed and provide a framework for determining the question of whether in a particular case information can properly be described as “on” a given measure.”

26. [ADDRESS] noted that “a purposive approach can be used to interpret a provision more narrowly than its very broad literal meaning” and adopted the statement of the Upper Tribunal Judge that: “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.” 27. Although the focus of [NAME] was the requisite connection between the information and the measure, all the parties to this appeal agreed that the approach of the Court of Appeal was of more general application to the interpretation of regulation 2. The First-tier Tribunal’s decision 28. In its decision the FTT identified two categories of information requested. Despite some confusion in the terminology used by the FTT, the parties agree that the first category was that relating to “all information held by [NAME] regarding the [NAME]”, and the second category was that relating “specifically to the safety evaluation/safety test”. The FTT decided that all the information which was solely within the scope of the first category had been disclosed by the time of the hearing. Thus the FTT was concerned only with the second category. The FTT decided that that information was “environmental information” within regulation 2(1) of the EIR. Its reasons were as follows: “Whether the safety test affects, or is likely to affect, the elements and factors of the environment:

61. As the DVSA put it, the disputed information is about determining whether a component of a [NAME] is likely to cause death or serious injury. Indeed,

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 7 all respondents have asserted that the safety test in this case is not an activity, which can be said to affect the elements of the environment.

62. The Tribunal respectfully disagrees. Whilst being careful not to reveal the contents of closed material, it is clear to anyone that in order to test the issue complained of (i.e. the vehicle throttle response under specific conditions) the vehicle must be driven, or at the very least the engine must be running. Consequently, by conducting the safety test:

- the DVSA caused emissions by driving the vehicle (r.2(1)(b)); - at the very least those emissions affected the air (r.2(1)(a)); - they did so through a measure (a safety test) which was likely to affect the elements (air) (r.2(1)(c));

63. From this analysis it is clear that even if the Tribunal were to reject the argument that the safety test was undertaken due to an environmental concern, the method by which it was undertaken was itself an interaction with the environment and thus caught by the definition of environmental information.

64. Turning to the wider context of the safety test, as argued by the appellant, we have considered the approach adopted in the [NAME] v [NAME] case ("[NAME]"). In response to the proposition (in that case) that the EIRs are overused and often said to apply to anything, the Tribunal noted: "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..."

65. We also refer to the Department for Energy and Climate Change ("DECC") case, specifically para.21 which developed the reasoning in [NAME]: "The Tribunal noted the approach of the FTT in the [NAME] 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."

66. The Appellant asks us to consider the purpose of the defect following this approach. However, we are not persuaded that this is correct. In our view the focus must be on the purpose of the test, which was to check the safety of the vehicle whilst being driven. The purpose was not to test whether the defect / modification had been put in place to circumvent noise emission regulations. The Appellant's reliance on "[NAME] and "DECC" is not redundant however. Applying this approach to the instant case, it is clear to the Tribunal that the DVSA is an executive agency within the Department for Transport. On their website, the DVSA summarise what they do as: "We improve road safety in Great Britain by setting standards for driving and motorcycling, and making sure drivers, vehicle operators and MOT garages follow roadworthiness standards. We also provide a range of licensing, testing, education and enforcement services."

67. The equivalent Department for Transport statement, reads:

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 8 "We work with our agencies and partners to support the transport network that helps the UK’s businesses and gets people and goods travelling around the country. We plan and invest in transport infrastructure to keep the UK on the move."

68. Therefore, whether one looks at DVSA alone or as part of the wider department, both are concerned with matters, which have a considerable impact on the environment, i.e. transport.

69. It is inescapable that the safety test relates to the safety of the vehicle whilst being driven. Driving is an activity, which has a direct impact on the environment. Government policy on “Reducing greenhouse gases and other emissions from transport" states: “Transport is a major source of greenhouse gases. Around a quarter of domestic carbon dioxide (CO2) and other greenhouse gas emissions in the UK come from transport. Transport is also a source of emissions which make air quality worse.” 70. The safety test, set in its wider context, is therefore a measure (reg.2(1)(c)) which is likely to affect the elements of the environment (i.e. the air - reg.2(1)(b)) through specific factors such as noise and emissions (reg.2(1)(b)). 71. [NAME] argue that the disputed information is about product safety rather than the environment. With respect we say that whilst the central issue is undoubtedly concerned with product safety, the product in question is a motor vehicle. As we have indicated at paras.62-63, motor vehicles have an impact on the environment, even whilst being tested (contrary to their argument), a measure which we find to be within the definition of environmental information.

72. Having considered the above and all evidence before it, and as indicated at the third hearing, the Tribunal rules that the EIRs are the correct access regime in relation to information about the safety test (our categories 2-7).

29. At paragraphs 73-78 the FTT explained why it rejected other arguments made by [NAME] regarding the throttle characteristic, but this does not impact on the FTT’s reasons for finding that the EIR was the correct regime.

30. At paragraph 80 the FTT confirmed that it had decided that the EIR applied to all the information in the second category. Whether the FTT erred in law 31. None of the parties have taken issue with the FTT’s conclusion at paragraph 70 that the safety test is a “measure”. Moreover, the parties are agreed that the disputed information is “on” the safety test. All the information in Category 2 relates specifically to the safety test. The FTT rejected [NAME]’s contention that the purpose of the test was environmental but nonetheless decided that the test was within regulation 2(1)(c).

32. The first line of reasoning by the FTT, at paragraphs 62-63, was that the test involved running an engine and that this caused emissions which affected the air. Thus, said the FTT, the method by which the test was conducted was “an interaction with the environment”.

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 9 33. That reasoning confuses the steps involved in carrying out an activity and the activity itself. Although running a car engine was a necessary element of carrying out the safety test, that did not of itself mean that, on a purposive approach to the EIR, the test affected environmental elements or factors. It fails to reflect the principle established by the Court of Appeal in [NAME] and in [NAME] that information which has only a minimal connection with the environment is not environmental information. That principle must apply not only in deciding whether information is on an environmental matter but whether a measure or activity has the requisite environmental effect. According to the FTT’s argument information about any activity which involves running an engine would be environmental information. Moreover, as [NAME] said, virtually anything done by any person or thing has some level of effect on or interaction with the environment. On the FTT’s reasoning, information on virtually all human activity would fall within the EIR. That plainly extends the regulations along way beyond their intended scope.

34. The second line of reasoning was about the “wider context” of the safety test (at paragraphs 66-70). The FTT did not rely on any environmental characteristic or impact of the test itself. Its reasoning was that the test related to the safety of a vehicle while driven and that driving affects the environment. This reasoning suffers from the same error as set out above. On this reasoning the definition would embrace, for example, information about an employee’s claim for mileage expenses because it would be information about the distance over which a person had driven and produced emissions. Indeed if the line of reasoning is correct then there is no logical reason to limit it to information about vehicles while they are driven as long as it has some connection with the driving of cars. The reasoning would also mean that economic forecasts or policies for future car manufacturing or sales in the UK are environmental information, as those activities impact on the numbers of cars that are made and driven.

35. Although the FTT used some of the language of purpose and context, it did not address whether and in what way knowledge of the disputed information, which concerned the safety of the [NAME], could contribute to the Directive’s purpose. The FTT did not give any consideration to whether access to the information would contribute to greater awareness or, free exchange of views about or more effective participation in environmental decision-making, or to a better environment. The FTT did not address any considerations such as those listed by the Court of Appeal at paragraph 43. Nor did the FTT reflect on whether its literal approach to the definition of environmental information led to an impermissibly broad reading that included information which could not reasonably be said to fall within the regulation. The FTT’s reasoning, set out above, demonstrates a fundamentally flawed approach which is inconsistent with the approach established by European case law and confirmed by the Court of Appeal in [NAME].

36. Before I conclude this part of the decision, I mention [NAME]’s position on this issue. His submissions are premised on his acknowledgment that it was an error to treat all motor vehicle safety tests as being environmental, and he does not

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 10 seek to defend the FTT’s reasoning in this regard. His submissions in essence seek to support the FTT’s decision on different grounds. They are similar to those which the FTT rejected at paragraphs 73-78 and which the Appellants have not sought to challenge. [NAME]’s submissions do not detract from my conclusion as to the fundamental legal error by the FTT as explained above, they are relevant to my decision as to the applicable regime which I address below. Disposal 37. In the light of the error of law by the FTT, I allow the appeal and set aside the FTT’s decision. [NAME] and [NAME] urge the Upper Tribunal to remake the decision as to the applicable regime. [NAME] submits that I should remake that decision, if possible, but he says whether it is possible to do so depends on the view that I take of his submissions as to the correct approach in law and of the closed material in that context.

38. I am satisfied that I should remake the decision on the applicable regime. As I explain below, I do not consider that I need to obtain further information or evidence in order to do so. I have had the benefit of detailed written and oral submissions about the correct approach, made by reference to the evidence. Remitting the question of the applicable regime to the FTT will mean that the parties will have to go over much of that ground again. Scope of the request 39. A prior issue arises as to the scope of the request. [NAME]’s request was for “information held by [NAME] regarding the [NAME] vehicle, and in particular the [NAME] safety evaluation of the vehicle throttle malfunction”. The FTT added [NAME] as a Respondent to the appeal, on the basis that DVSA was an agency of [NAME] and [NAME] was the relevant public authority. On 14 January 2015 the FTT Judge directed [NAME] “to seek to ascertain what information (if any) it holds across the Department (including its agencies) within the scope of the request.” [NAME] complied under protest, maintaining that such material was outwith the scope of the request. As a result [NAME] provided (in open and closed bundles) other material held within the Department but not held by DVSA. This included material held by [NAME]. I am satisfied that only the materials held by DVSA (formerly [NAME]) are within the scope of the request. [NAME]’s request was clearly limited to that information. He did not ask for information held by other agencies. He understood the different functions of [NAME] and [NAME] but, although he was concerned about [NAME]’s involvement with [NAME], he did not ask for information held by [NAME]. Closed session 41. Part of the hearing took place in closed session in the presence only of the appellants and the [NAME] and their representatives, in order to consider the closed material and for those parties to make submissions which referred to the closed material.

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 11 42. The closed material comprised unredacted documents relating to the safety test, the redacted copies of which were in the open bundle. Counsel for the appellants submitted that none of the redacted text had any environmental content or character. Ultimately whether or not any or all of the information was environmental turned on a small number of redacted passages. [NAME]’s case 43. [NAME]’s case is that the requested information is environmental information within regulation 2(1)(c) EIR. Although in the FTT he also relied on subparagraphs (b), (d) and (f), he does not now do so. The appellants and the [NAME] are unanimous in submitting that there can be no argument that any subparagraph other than (c) applies, and [NAME] has not sought to argue otherwise.

44. The basis on which [NAME] argues that the safety test affects the environment has undergone some changes during the course of this appeal. In his initial response to the appellant’s grounds of appeal, he relied on three factors. The first was that emissions are created during the test and the second was that environmentally compliant vehicles which are cleared by DVSA as safe to drive cause emissions. These are the same or very similar to the arguments adopted by the FTT which I have rejected. In any event, by the time of the hearing [NAME] conceded that these arguments were wrong in that they adopted too broad an approach. 45. [NAME] now focusses his case on the third factor, which relates to his particular environmental concerns about the [NAME]. He says that the throttle characteristic was deliberately fitted in the [NAME] in order to manipulate noise emission testing carried out by [NAME]. He says that the characteristic occurs in the Cayman R when being driven in third and fourth gear when the vehicle is travelling at approximately 50 km/hour. EU regulations require vehicles to be tested for noise emissions when being driven at 50km/hour with the throttle fully open, but he says that the throttle in the Cayman R is not fully open when driven under those conditions and that this enables the vehicle to pass tests which it would otherwise fail. He says that as a result cars are being placed on the market and driven even though their noise emissions breach environmental regulations, and consumers have incorrect information about the true output of the vehicles. 46. [NAME] referred this characteristic to [NAME] because he considered that what he described as a sudden and unpredictable loss of engine power, which he says is a consequence of the throttle characteristic, presented a risk to road safety. A similar complaint by another owner led to the safety test about which [NAME] seeks information. [NAME]’s conclusion was that there was no safety concern and that the hesitation that arose when driving the vehicle at around 50 km/hour was to change to 2nd gear. This has led to a further concern by [NAME], that the vehicle’s compliance with regulations concerning CO2 emissions was not validated on the vehicle being driven at that speed in that gear. He says that the

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 12 conclusion from the safety test as to the vehicle’s gearing ratio contradicts the presumed gearing ratios used in CO2 emissions tests on the vehicle and is, accordingly, information “on” the CO2 test.

47. I emphasise that the above is only a very broad summary of the much more detailed technical issues raised by [NAME], but it is sufficient for the purposes of this appeal. In essence [NAME]’s case is that the information from the safety test might help to show that the environmental tests for noise or CO2 emissions were invalid. He submits that disclosure of information about the safety test may shed light on why [NAME] appears to have hidden an environmental concern and will give a better awareness of environmental matters and, ultimately, contribute to a better environment.

48. This case theory is important context to this appeal. It is important to note that the case theory is hotly contested by [NAME]. For reasons which I explain, I do not need to decide whether or not it is correct. Analysis: whether the information falls within regulation 2(1)(c) EIR 49. As the Court of Appeal noted in [NAME] at paragraph 37, unlike the definition of “information” in FOIA, the definition of “environmental information” in the EIR focusses on the relevant measure (or activity) and not solely the nature of the information itself. It is first necessary to identify the measure or activity that the disputed information is “on”. The measure or activity must affect or be likely to affect the elements or factors in subparagraphs (a) or (b).

50. The parties’ submissions, and in particular those of counsel, have been structured around four possible candidates for the measure or activity in the present case. It being common ground that information can be “on” more than one measure or activity, counsel have suggested that the information in this case might be said to be on any of the following: the safety test, the throttle characteristic, the decisions as to the design of the throttle characteristic, and the CO2 test. In my view the attempt to distinguish between the first three of these is somewhat artificial but it does not affect my conclusions, for reasons which I now explain.

51. The first candidate is the safety test itself. As I have said, there is no doubt that the information was on the safety test. There is also no dispute that the safety test was concerned with the safety of the vehicle and not environmental matters. For the reasons which I have given for allowing the appeal, the mere fact that the safety test involved driving vehicles or permitted approved vehicles to be sold and driven does not bring the activity within regulation 2(1)(c). 52. [NAME] says that the position where, as he submits, the [NAME] R was not compliant with environmental standards and that, in clearing it as safe to drive, DVSA permitted non-environmentally compliant vehicles to be driven on the roads in the UK which thereby affected the elements in (a) (air) and/or the factors in (b) (noise).

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 13 53. [NAME] strenuously denies the claim that the vehicle was not environmentally compliant. However, even assuming for present purposes that the vehicle was not environmentally compliant, I reject [NAME]’s argument that this would mean that the safety test affected or was likely to affect any of the elements or factors in regulation 2(1). The state of the environmental compliance of the vehicle does not alter the nature or purpose of the safety test. The asserted connection between the test and the environment is too remote to bring it within regulation 2. [NAME]’s submission fails by reference to every one of the questions posed by the Court of Appeal in [NAME] at paragraph 43. The test was conducted by a body that had no environmental remit. The test was not intended to be used for an environmental purpose. Access to information about the safety of the vehicle, its performance in a safety test, or the methodology of the test would not enable the public to be better informed about or participate in decision-making regarding the environment. Information about the safety test would not reveal the vehicle’s compliance or otherwise with environmental regulations.

54. I am reinforced in this conclusion by the decision of the CJEU in [NAME], referred to by the Court of Appeal in [NAME] (see paragraph 19 above). That case concerned EU requirements for labelling foodstuffs produced from genetically modified organisms (GMOs). [NAME] sought information about the administrative measures which had been taken to check and enforce compliance with the labelling requirements. The question for the CJEU was whether this information was environmental information within the definition in the directive which was the predecessor to Directive 2003/4/EC and which included “information on activities or measures affecting or likely to affect” various environmental factors. [NAME]’s argument had strong similarities to that made by [NAME]. She said that placing GMO products on the market affected the environment as did the labelling requirements, and information about administrative measures for the control of the labelling requirements was also environmental information. [ADDRESS] rejected her argument. It said that information on measures of control is not environmental information “even if those controls concern activities or measures which for their part affect or are likely to affect one or more of the environmental factors.” Similarly, in the present case, there is a difference between putting non-environmentally compliant vehicles on the market and safety tests which form part of the controls over whether the vehicles are placed on the market. 55. [NAME] seeks to address this by arguing that the safety test was itself, at least in part, concerned with environmental matters. I note that a similar argument did not assist [NAME], even though in her case the marketing controls related to a labelling regime which (on her case at least) had an environmental dimension. But in any event I reject the factual premise asserted by [NAME], which is an inference based on documents showing that [NAME] or DVSA, or both, were in communication with [NAME] about the test. [NAME] says that these documents show that there was an underlying connection between the safety test and environmental matters, but I do not consider that this follows from the fact

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 14 that information was shared with [NAME]. There could be many reasons for [NAME]’s involvement. A number of issues had been raised about the Cayman with a number of agencies, including [NAME], and not only by [NAME]. It cannot be assumed that any interest by [NAME] in the safety test was because the test was itself concerned with environmental issues. [NAME] has a safety role as well as an environmental role. In any event [NAME]’s environmental remit does not mean that all information which comes into its hands is environmental information. 56. [NAME] also relies on a letter from the chief executive of [NAME] dated 8 January 2013 which stated that the [NAME] remained on sale because a serious risk to road safety or the environment had not been demonstrated. The letter referred to the United Kingdom’s obligations to comply with EU legislation by reference to both road safety and environmental risks, but there is nothing which indicates that information relating to the former was itself environmental.

57. The second candidate identified by counsel is the throttle characteristic. As I have said, it seems to me that it is artificial to draw a line between information on the safety test and information on the throttle characteristic. The open material makes it clear that the focus of the safety test was on what is described as the engine’s “hesitation” under certain accelerating conditions. That is a description of the throttle characteristic and its effects. The throttle characteristic was not something that arose incidentally within the safety test. It was the reason for the safety test being carried out, and it was what was tested. In a sense it characterises the safety test.

58. However, staying with [NAME]’s approach for the time being, having uncoupled the throttle characteristic from the safety test, [NAME] (with whom [NAME] agrees) argues that the throttle characteristic is not itself a measure or activity. He says that the manner in which the throttle operates is simply a feature of the Cayman and is no more a measure or activity than is, for instance, the weight of a smart phone. I agree. It strains both the language and the concept too far to say that a feature of a product is a measure or activity.

59. Alternatively, if I take my preferred approach which is to address this issue on the basis that the safety test is about the throttle characteristic and that is a relevant consideration in deciding whether the requested information is environmental, it does not change my conclusion that the information is not environmental. [NAME]’s case about the operation of the throttle characteristic is inextricably bound up with his claim that the vehicle if not environmentally compliant, and I have addressed that above.

60. The third candidate in counsels’ analysis is the decision about the design of the throttle characteristic. [NAME]’s case is that the throttle characteristic was designed in order to manipulate the environmental test results. [NAME] says that it could be argued that the request was for information on that decision. He says that, although the throttle characteristic is a feature of the vehicle and not an activity or measure, it may be a consequence of a decision about its design and a design decision can be a measure or activity. That does not mean that

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 15 information about a feature will always be information about a design decision. Every feature of a car is the product of the design of the car, and the manner in which a feature functions is in some way a consequence of design decisions. If all such decisions were measures or activities within subparagraph (c), the effect would be to extend the scope of that paragraph very considerably. As [NAME] makes clear, in order to fall within (c) on the basis that information is on a design decision, there must be a nexus which is more than incidental or remote between the information requested and the design decision itself. [NAME] says that there is no such connection here.

61. Again in a sense this too is a somewhat artificial position. The request was for information relating to the safety test. The task of the tribunal is to consider the information and decide whether it is on a measure or activity which affects or is likely to affect the environmental elements or factors. If the requested information included information which related to [NAME]’s case about the alleged purpose and effect of the throttle characteristic, that will be relevant to that decision. However, whichever of the two approaches is adopted, the result is the same: the information is not environmental.

62. As [NAME] and [NAME] submit, the requested information in this case is all about the safety test and does not relate to the design of the throttle characteristic. [NAME] says that the case should be viewed differently. Although the safety test was not concerned with decisions about the design of the throttle characteristic, it does not follow that the information relating to the safety test contains nothing which might shed light on the design decision or the environmental matters with which [NAME] is concerned. He points out that some of the redacted parts of information are written in technical language which is opaque to a non-expert. He says that, properly understood, it may be that some of this technical information is integrally linked to [NAME]’s concerns about the design and purpose of the throttle characteristic. He submits that, if this could determine whether the information is environmental, then the Upper Tribunal cannot decide the question without obtaining an explanation of the technical content of the report. 63. [NAME] further submits that the requested information may be on a measure or activity which affects or is likely to affect the environment, if [NAME]’s theory about the throttle characteristic is correct. He says that [NAME] was interested in understanding how DVSA had scrutinised the throttle characteristic when conducting the safety test because that information might cast light on his theory. [NAME] acknowledges that the evidence does not establish [NAME]’s theory on balance of probabilities. However, he says that the Upper Tribunal need only find that the theory is plausible. He points to the material before me which shows that the fruits of the test were shared with [NAME], and to the materials which [NAME] claims suggest that the throttle characteristic may have been introduced with environmental testing in mind. [NAME] also submits that, if the outcome of the appeal turns on the merits of [NAME]’s theory, further evidence may be needed in that regard.

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 16 64. I am satisfied that I do not need any further explanation of the technical material which is within the closed bundle in order to decide this question. The safety test was not intended to address the design of the throttle characteristic, nor whether it was motivated by considerations relating to environmental testing. This was no concern of [NAME]’s. The safety test tested whether the throttle characteristic affected the safety of the vehicle and, to that end, would have been interested in the way in which the throttle characteristic operated. Even if some parts of the information shed light on the underlying design decision and even if they showed, or helped to show, that the design of the characteristic was in some way relevant to the vehicle’s performance in environmental tests, that information would be no more than incidental to the report. On assumptions most favourable to [NAME], the connection between the information sought and the design or purpose of the throttle characteristic would be tenuous at best. Even if it is permissible to ask whether certain passages taken in isolation were environmental, that question has to be considered in the relevant context which, in this case, is the nature and purpose of the safety test. If any information within the report sheds light on [NAME]’s case theory, it is not integral to the safety examination or report. In my judgment, the argument advanced by [NAME] and [NAME] impermissibly looks at issues with which the information is not concerned but with which it is merely connected ([NAME] at paragraph 39) and, to adopt a passage from the decision of Judge Wikeley in those proceedings, involves an “impermissible and overly expansive reading that sweeps in information which on no reasonable construction can be said to fall within the statutory definition”.

65. I have serious doubts that the materials raise a plausible case to support [NAME]’s theory but, even if they did, it would not affect my conclusion and so I do not need any more evidence in that regard. The fact that a requester may wish to use information in order to advance an environmental case does not mean that the information itself is environmental. The submission introduces a degree of uncertainty into a legislative definition which cannot have been intended by parliament nor by the institutions responsible for the Directive and the Aarhus Convention. The consequence of the submission is that, in order to decide whether the EIR applies, the public authority (and the [NAME], if the case goes further) must first assess the merits of the case theory. And it requires decision-makers to engage in a substantial inquiry going way beyond the intended scope of the EIR or the Directive. The Directive is concerned with enabling citizens to have access to information in order to assert their right to live in an adequate environment, including to participate in decision-making and have access to justice to that end. On [NAME]’s case, however, a public authority holding information, and the [NAME] and the FTT, would be called on to determine the underlying environmental grievance held by the requester in order to decide whether the information requested is environmental information. That puts the cart before the horse.

66. I bear in mind the need for a broad and purposive approach to the interpretation and application of the EIR, and that “the tribunal is not restricted by what the

[NAME], [COMPANY] v [NAME] and [NAME] [2018] UKUT 127 (AAC)

GIA/224/2016 17 information is specifically, directly or immediately about” ([NAME] in the Court of Appeal at paragraph 39). Nonetheless, for the reasons which I have explained above, the link between the requested information in this case and [NAME]’s case theory is too tenuous to support a conclusion that the information is on that theory ([NAME]’s approach); and for the same reasons any link between the activity (the safety test) and any environmental issues is too tenuous to mean that the safety test affected or was likely to affect the environmental elements or factors (my approach).

67. The final candidate upon which [NAME] relies is the CO2 emissions test on the Cayman. [NAME] says that, although the request for information did not include anything directly concerning a CO2 test, the information requested was on the CO2 test because the information within or resulting from the safety test contradicted the validity of the CO2 test. The factual premise of this submission is also denied by [NAME] but, even if [NAME]’s allegation is correct, it does not mean that the information is on the CO2 test. The fact that information from one source might prove to be relevant to the evaluation of separate environmental information does not mean that the former is a measure which affects or is likely to affect the environment. Even if information relating to the safety test sheds light on the validity of the CO2 test, that is only incidental to the safety test. Information on the safety test is not also on the CO2 test. Conclusion 68. The appeal is allowed because the FTT erred in deciding that the EIR applied. I conclude that, on the correct approach, none of the requested information is environmental information within regulation 2 of the EIR and so the request for information must be determined in accordance with FOIA.

69. It is not appropriate for this Tribunal to determine whether the information is exempt under FOIA. I have not been addressed on those issues, I do not have the benefit of any findings of fact by the FTT in that regard and there is no good reason for this Tribunal to take on the role of primary fact finder. I remit the appeal to the FTT for consideration of those issues in accordance with the Directions set out above.

Signed on the original Kate Markus QC on 12th April 2018

Judge of the Upper Tribunal