Upper Tribunal Upholds Exclusion of Journalism Information from Freedom of Information Act

OUTCOME: Dismissed

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

IN THE UPPER TRIBUNAL Appeal No. UA-2022-000648-GIA (ADMINISTRATIVE APPEALS CHAMBER)

BEFORE UPPER TRIBUNAL JUDGE WEST

Appellant [NAME_1]

and

Respondents (1) [NAME_2] (2) THE [NAME_4]

APPEAL AGAINST A DECISION OF A TRIBUNAL

DECISION OF THE UPPER TRIBUNAL

Decision date: 6 September 2023

Decided after a hearing on 25 May 2023

UPPER TRIBUNAL JUDGE WEST

ON APPEAL FROM

Tribunal: First-tier Tribunal ([NAME_3])

Tribunal Case No: EA/2021/0220

Tribunal Hearing Date: 2/2/2022

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 2

DETERMINATION

The decision of the First-tier Tribunal (General Regulatory Chamber) ([NAME_3]) (which sat on 2 February 2022) dated 7 February 2022 under file reference EA/2021/0220 does not involve an error of law. The appeal against that decision is dismissed.

This decision is made under section 11 of the Tribunals, Courts and Enforcement Act 2007.

Representation: [NAME_7], counsel, for the Appellant (instructed by McCarthy Denning)

[NAME_11], counsel, for the First Respondent (instructed by the [NAME_2])

[NAME_13], counsel, for the Second Respondent (instructed by [NAME_16])

REASONS

Introduction 1. The Appellant is Mr [NAME_1] (!”[NAME_1]”). The First Respondent is [NAME_2] (“the [NAME_2]”). The Second Respondent is the [NAME_4] (“[NAME_16]”).

2. This is an appeal, with my permission, against the decision of the First- tier Tribunal (Judge Sophie Buckley, Tribunal Members [NAME_18] and [NAME_19]) which sat to consider the matter at a video hearing on 2 February 2022 and reached its decision on 7 February 2022. The Tribunal dismissed the appeal against the [NAME_2]’s decision notice IC-76825-W4T0 of 2 August 2021 which held that any [NAME_3] held by [NAME_16] within the scope of paragraphs 1 and 8 of the request would be held for the purposes of

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 3 journalism, art or literature and would therefore fall outside the Freedom of [NAME_3] Act 2000 (“FOIA”). The [NAME_2] did not require [NAME_16] to take any steps. The Factual Background 3. As the Tribunal explained at the outset of its decision:

“3. Since 2010 [NAME_16] has annually commissioned the independent research and polling company, [NAME_20], to conduct a representative survey of the UK public on their perception of [NAME_4] standards (including, but not limited to, impartiality) in [NAME_4] output.

4. The survey in question in this appeal is a 2018 [NAME_20] survey (‘the Survey”). The results relevant to this appeal appear on p 19 of [NAME_16] Group Annual Report and Accounts, which reports that 52% of UK adults think that [NAME_16] is effective at providing news and current affairs that is impartial. It includes a pie chart which shows that 44% of 1,829 UK adults who follow the news, April-May 2019 answered ‘[NAME_16]’ to the question ‘Of all the news sources (TV, radio, newspaper, magazine, website, app or social media) which one source are you most likely to turn to for news you trust the most?”

The Request 4. On 17 June 2020 [NAME_1] made a request to [NAME_16]:

“I refer to page 19 of [NAME_16] Group Annual Report and Accounts 2018/2019 giving the results for several survey questions showing that 52% of people asked think that [NAME_16] provides impartial news and that 44% turn to [NAME_16] if they want impartial news. The source for both is given as [NAME_20]. Please can you provide [NAME_3] and all relevant documents relating to the following for both the 52% and 44% results:

1. A copy of and details of the brief and instructions that were given to [NAME_20] or any relevant meeting notes when they were commissioned to carry out the survey that led to the above two results and any underlying contracts;

2. How the audience sample was chosen and what were the criteria to include or exclude survey participants in each case;

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 4

3. A list of the coding options used and raw data received back from the survey participants in each case;

4. Details as to how the survey answers were coded in each case or otherwise how the raw data was extrapolated to create the percentage results shown in the annual report;

5. Whether reports by the organisation News-watch on the subject of [NAME_16]’s impartiality obligations played any part in the production of the conduct of the [NAME_20] survey and the content of [NAME_16] and, if so, what?

6. A copy of the original report and any interim reports by [NAME_20] to [NAME_16]; and

7. [NAME_16] altered or changed the presentation of the results mentioned above.

8. Please also provide copies of all complaints to [NAME_16] about impartiality from 2015 to date and [NAME_16] responses to the same”.

The Response 5. [NAME_16] replied to the request on 25 June 2020, stating that the [NAME_3] requested, if held, would be held for the purposes of journalism, art or literature and was excluded from FOIA.

6. [NAME_1] referred the matter to the [NAME_2] on 11 December 2020.

7. During the course of the [NAME_2]’s investigation [NAME_16] indicated in a letter dated 19 July 2021 that the [NAME_3] requested in parts 2, 3 and 4 of the request was publicly available and provided a link. It stated that the [NAME_3] requested in parts 5 and 6 was not held.

8. It maintained that the [NAME_3] requested in parts 1, 7 and 8 was held for the purposes of journalism. It stated that the [NAME_3] “related to how [NAME_16] analyses adherence to [NAME_16]’s editorial standards and seeks to understand audience perceptions of its commitment to high editorial

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 5 standards, in particular by reference to impartiality standards. That was clearly a function of the third limb of the [NAME_21] definition of “journalism”; to maintain and enhance standards of output”.

The [NAME_2]’s Decision 9. In the decision notice dated 2 August 2021 the [NAME_2] decided that any [NAME_3] held by [NAME_16] within the scope of parts 1 and 8 of the request would be held for the purposes of journalism, art or literature and would fall outside the scope of FOIA.

10. In relation to the scope of the case, the [NAME_2] considered, for the same reasons, that even if [NAME_16] did hold [NAME_3] within the scope of parts 5 and 6 it would be covered by the derogation. In relation to part 7 it considered that because the raw data from [NAME_20] and [NAME_16]’s annual report were in the public domain [NAME_1] effectively had an answer to that part and the [NAME_2] did not consider it further.

11. The [NAME_2] stated that for the [NAME_3] to fall outside FOIA there should be a sufficiently direct link between the purpose(s) for which the [NAME_3] was held and the production of [NAME_16]’s output and/or [NAME_16]’s journalistic or creative activities involved in producing such an output.

12. In relation to part 8, the [NAME_2] had repeatedly ruled that [NAME_3] relating to complaints about [NAME_16]’s output was [NAME_3] relating to the maintenance of editorial standards. There was a clear and direct link between the complaints which [NAME_16] received, its editorial process or review based on those complaints and its subsequent output. The [NAME_3] fell squarely within the third element of journalism because it related to the maintenance of editorial standards.

13. In relation to part 1 of the request, the [NAME_2] accepted that, whilst some of the [NAME_3] might be held for other purposes, it was also held for journalism. A survey asking people whether they considered output impartial

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 6 was done because [NAME_16] wished to measure the quality of its output. Presumably if the percentages of people who considered [NAME_16] delivered impartial news had been low, [NAME_16] would wish to take action to improve the quality of its output.

14. The [NAME_2] accepted that [NAME_16]’s audience research and the correspondence which preceded it was [NAME_3] which [NAME_16] held for the purpose of monitoring and influencing its output and was covered by the third limb of the journalism exception.

The Decision of the Tribunal 15. [NAME_1] appealed to the Tribunal, but his appeal was dismissed. There were 2 grounds of appeal: (1) the [NAME_2] erred in law in the application of the tests for ambit of the “journalism” derogation and incorrectly applied the test described in BBC v Sugar (No.2) [2012] UKSC 4, [2012] 1 WLR 439 (“[NAME_21]”) (2) the [NAME_2] fettered its discretion by predetermining its response by not giving proper consideration to the additional submissions provided by [NAME_1] dated 28 July 2021.

16. So far as material, the Tribunal held that

“Ground 2

63. The tribunal’s remit is governed by s.58 FOIA. This requires the tribunal to consider whether the decision made by the Commissioner is ‘in accordance with the law, or to the extent that the notice involved an exercise of discretion by the Commissioner, that he ought to have exercised his discretion differently’. As the Upper Tribunal noted in IC v Malnick and ICOBA [2018] AACR 29:

90. ... Although the statutory language is less than helpful, this formulation embraces all errors, and is not limited to the traditional taxonomy of errors of law. As is clear from section 58(2) and [NAME_22] (see paragraph 45 above), the F-tT exercises a full merits appellate jurisdiction and so stands in the shoes of the Commissioner and decides which (if any) exemptions apply. If it disagrees with the

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 7 Commissioner’s decision, the Commissioner’s decision was “not in accordance with the law” even though it was not vitiated by public law error.

...

94. ... the appellate machinery in FOIA is not concerned solely with public law error. As already noted, [NAME_22] makes clear that on a proper reading section 58 is concerned with any error of law or fact or even a difference in view. It follows the F-tT may allow an appeal because it makes a different assessment to that of the Commissioner even though the Commissioner has not made any error of law in the public law sense ... under the FOIA regime it is simply unnecessary to raise any considerations of ultra vires. If the F-tT decides that the Commissioner’s decision was made in error of law but agrees with the decision, then it will dismiss the appeal. If the F-tT decides that the Commissioner’s decision was not made in error of law but disagrees with it, then the appeal will be allowed and a different decision notice will be substituted. The legal validity of the F-tT’s decision (which itself is subject to appeal for error of law to the [NAME_23]) satisfies the rule of law.

...

102. ... The decision in [NAME_22] means that there is no limitation on the issues which the F-tT can address on appeal, and the focus of its task is the duty of the public authority. This means that the tribunal must consider everything necessary to answer the core question whether the authority has complied with the law.

64. The Tribunal may receive evidence that was not before the Commissioner and may make different findings of fact from the Commissioner. The tribunal has a full merits appellate jurisdiction.

65. In those circumstances we do not intend to decide whether or not the Commissioner unlawfully fettered her discretion. As the Upper Tribunal states in [NAME_24], if we decided that the decision was made in error of law but agreed with the decision we would, in any event, dismiss the appeal.

Ground 1

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 8

66. In summary, the Appellant’s argument is that the Commissioner was wrong to conclude that the [NAME_3] fell outside the definition.

67. As a preliminary point, we consider that the appropriate time to consider whether [NAME_3] was held for the purposes of journalism should be consistent with the time for assessing whether [NAME_3] was held, or the time for assessing the public interest. There was no internal review in this case, but given that [NAME_16] effectively issued a new substantive response in July 2021, we find that July 2021 is the appropriate time to determine the purpose for which the [NAME_3] was held.

68. At the start of his request [NAME_1] sets the context. He refers to ‘page 19 of [NAME_16] Group Annual Report and Accounts 2018/2019 giving the results for several survey questions showing that 52% of people asked think that [NAME_16] provides impartial news and that 44% turn to [NAME_16] if they want impartial news. The source for both is given as [NAME_20]’.

69. Having set the context, he asks for ‘[NAME_3] and all relevant documents relating to the following for both the 52% and 44% results’.

Part 1 and part 7 of the request

70. Part 1 of the request specifies the following documents: ‘A copy of and details of the brief and instructions that were given to [NAME_20] or any relevant meeting notes when they were commissioned to carry out the survey that led to the above two results and any underlying contracts.’

71. Specifically he is asking for:

- the brief

- the instructions

- any relevant meeting notes

- any underlying contracts.

72. [NAME_16] argues that these were all held for the purposes of journalism.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 9 73. Part 7 of the request, read with the introduction, is for all relevant documents relating to how [NAME_16] altered or changed the presentation of the 52% and 44% results. The Commissioner took the view that she did not need to address this part of the request because [NAME_1] had access to the original results and [NAME_16]’s presentation of those results. Given that [NAME_16] confirmed that it held [NAME_3] within scope of part 7, and did not suggest that such [NAME_3] was in the public domain, we disagree. In our view, we must determine whether such [NAME_3] as was held by [NAME_16] fell within the definition.

74. We consider parts 1 and 7 together.

75. In July 2021, the [NAME_3] was held by the Audience Research team in [NAME_4]. This team was responsible for commissioning the annual survey of the UK public ‘on their perception of [NAME_4] standards (including, but not limited to, impartiality) in [NAME_4] output’. The team is more broadly responsible for facilitating ways that [NAME_16] can better understand its audience.

76. [NAME_16] explains that the [NAME_20] survey is one of many ways that [NAME_16] seeks to better understand audience perception of [NAME_4] programming with a view to improving the quality and diversity of output. The survey results are analysed by [NAME_16] Board, [NAME_4] Board and Executive Committee. The survey also feeds into annual performance reviews of [NAME_4] output to understand how [NAME_16] is meeting its public service mission under its Royal Charter to ‘inform, entertain and educate audiences’.

77. In addition, [NAME_16] states that the results are used by Editorial Standards and Policy in training sessions to explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape as a way of reinforcing how audiences will regard [NAME_4] coverage. This can directly impact on how content is ‘signposted’ in a programme.

78. [NAME_16] further states that [NAME_3] about the underlying scoping of the survey and the terms provided to [NAME_20] would disclose internal considerations about how [NAME_16] perceives its content and how it seeks to engage audiences. The survey costs derive from editorial budgets determined by editorial teams alongside other budgetary decisions like programming

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 10 costs; more money spent on one area means less available for another.

79. The Appellant argues that the survey is a device to allow [NAME_16] to provide an overview of its activities for public consumption in a reporting period.

80. We accept that is one of the purposes of the survey. In our view impartiality is a fundamental aspect of [NAME_16]’s journalism output. We find that presenting the figures on impartiality to the public in an annual report is intrinsically and directly linked to its journalistic output.

81. Further, we accept that the survey results are also used by [NAME_16] to ‘better understand audience perception of [NAME_4] programming with a view to improving the quality and diversity of output’. They are also used in training sessions to ‘explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape’ and can directly impact on content.

82. In our view, this falls squarely within ‘journalism’. It concerns the maintenance and enhancement of the standards and quality of journalism (particularly with respect to balance). It can directly impact on the output of [NAME_16].

83. On this basis, we conclude that all the peripheral [NAME_3] which was created in order to produce that survey (the underlying contracts, the brief, instructions, any related meeting notes) was held, at least in part, for the purposes of journalism.

84. As stated above, in our view, the presentation of these figures on impartiality in an annual report is intrinsically and directly linked to [NAME_16]’s journalistic output. Therefore the [NAME_3] which related to how the presentation of the results was changed was held, at least in part, for the purposes of journalism.

85. The Appellant submits that the documents caught by part 1 are ‘of their nature not journalistic in content but more akin to routine financial documents’ so that they fall within the definition relying on [NAME_25] at para 55 of the Court of Appeal judgement in Sugar v BBC [2010] EWCA 715. [NAME_25] states:

In my view, whatever meaning is given to "journalism" I would not be sympathetic to the notion that [NAME_3] about, for instance,

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 11 advertising revenue, property ownership or outgoings, financial debt, and the like would normally be "held for purposes ... of journalism". No doubt there can be said to be a link between such [NAME_3] and journalism: the more that is spent on wages, rent or interest payments, the less there is for programmes. However, on that basis, literally every piece of [NAME_3] held by [NAME_16] could be said to be held for the purposes of journalism. In my view, save on particular facts, such [NAME_3], although it may well affect journalism-related issues and decisions, would not normally be "held for purposes ... of journalism". The question whether [NAME_3] is held for the purposes of journalism should thus be considered in a relatively narrow, rather than a relatively wide, way.

86. In the tribunal’s view the requested [NAME_3], including, for example, the terms upon which the survey was to be produced and the brief is not akin to [NAME_3] about advertising revenue, property ownership or outgoings or financial debt. All of those are only very remotely linked to [NAME_16]’s output.

87. The requested [NAME_3] in this case is much more closely linked to [NAME_16]’s output because it was directly used to commission a survey to be used to influence content and it includes ‘internal considerations about how [NAME_16] perceives of its content and how it seeks to engage audiences’. Looking at the directness of the purpose, we find that the requested [NAME_3] is sufficiently proximate to [NAME_16]’s journalistic purposes and the end product. It was an immediate object of holding the [NAME_3] to use it for one of those purposes.

88. We accept that [NAME_16]’s specific argument about the budget used to fund the survey is not too many steps removed from [NAME_25]’s observations about the link between the [NAME_3] he described and journalism: ‘the more that is spent... the less there is for programmes’.

89. However, [NAME_16]’s argument is more focussed: it is not simply an argument that financial [NAME_3] on spending unrelated to journalism is held for the purposes of journalism because the spending reduces the overall pot available to [NAME_16] as a whole. It is specific to the editorial budget. In our view this supports [NAME_16]’s

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 12 assertion that the [NAME_3] is held for the purposes of journalism. Further this is not simply a request for the cost of the survey. The requested [NAME_3] consists of the contract, the brief etc. which were, as we have stated above, directly used to commission a survey for, at least in part for the purposes of journalism, and in itself contained the internal considerations referred to above.

90. We have considered whether, at the time of the request, the [NAME_3] in parts 1 and 7 was still held for those purposes. We have concluded that the [NAME_3] was still, in July 2021, held for the purposes for which it had originally been held. The reason [NAME_4] holds it remains as it was in 2019. There is no evidence that the requested [NAME_3] has been ‘archived’ in any sense of that word.

Part 8

91. We accept that there is no binding authority to the effect that [NAME_3] held for editorial complaints falls outside the scope of FOIA.

92. The [NAME_3] requested in part 8 of the request is ‘copies of all complaints to [NAME_16] about impartiality from 2015 to date and [NAME_16] responses to the same.’

93. This [NAME_3] is held by [COMPANY_26] who administer [NAME_4] complaints and the Executive Complaints Unit who handle appeals.

94. In support of its submission that these complaints were held for the purposes of journalism [NAME_16] has referred the tribunal to a number of previous decision notices concerning complaints.

95. It is apparent from the [NAME_3] submitted to the Commissioner in this and in other investigations that the consideration of complaints by [NAME_16] is an important tool used by [NAME_16] to monitor, maintain and enhance its journalistic output and to ensure the impartiality of that output.

96. [NAME_16] has explained in the course of previous investigations that [NAME_3] relating to editorial complaints is held for editorial purposes to influence editorial direction and inform future content. It plays a significant role in improving the quality of journalistic output.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 13

97. In our view, it is clear that [NAME_4] uses previous complaints to inform content. In the particular context of bias, this use will undoubtably include the use of previous complaints to monitor, maintain and enhance its journalistic output and to ensure the impartiality of that output.

98. In our view, given the use to which [NAME_16] puts previous complaints, they are clearly held for the purposes of journalism. Their use is directly linked to [NAME_16]’s journalistic output.

99. [NAME_27] has asked us to infer that at least some of the complaints in the requested period will no longer have been held for the requisite purposes at the relevant date. There is no evidence before us that complaints are no longer envisaged as having any current purpose, but stored for historical purposes, after a certain date. There is no evidence before us on which we could base a finding that complaints before a certain date were no longer referred to and therefore could not be seen as work in progress. [NAME_16] stated in July 2021 that the requested [NAME_3] was held by the [COMPANY_26] who administer [NAME_4] complaints and the Executive Complaints Unit who handle appeals. This suggests to us that it was still in current use. On this basis we find that the [NAME_3] was still held, at the relevant time, for the purposes of journalism.

100. On the above grounds, we conclude that the Commissioner reached the correct decision, and we dismiss the appeal.”

Permission To Appeal 17. The Tribunal refused [NAME_1] permission to appeal against its decision on 18 March 2022. He applied to the Upper Tribunal for permission to appeal on 5 April 2022. I directed an oral hearing of his application for permission to appeal on 17 May 2022.

18. On 31 October 2022 I heard [NAME_1]’s application, when he was represented by [NAME_7], and granted him permission to appeal against the decision of the Tribunal.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 14 19. Following my grant of permission to appeal, [NAME_16] applied to be joined as Second Respondent to the proceedings on 24 November 2022 pursuant to rule 9(3) of the Tribunal Procedure (Upper Tribunal) Rules 2009. [NAME_1] nor the [NAME_2] objected to [NAME_16] being joined to the proceedings. Accordingly I granted [NAME_16]’s application to be joined as Second Respondent to the proceedings and made consequential directions on 9 December 2022.

20. In accordance with my directions the [NAME_2] put in a formal response to the grounds of appeal on 12 December 2022 and [NAME_16] on 16 January 2023.

21. It was not possible to hold the hearing on 16 February 2023, as had originally been envisaged, and the matter was relisted for hearing on 25 May 2023. On that occasion [NAME_1] was again represented by [NAME_7]. The [NAME_2] was represented by [NAME_11] and [NAME_16] by [NAME_13]. I am indebted to all of them for their concise and well-argued submissions.

The Legislation 22. So far as material, FOIA provides that

“s.1(1) Any person making a request for [NAME_3] to a public authority is entitled

(a) to be informed in writing by the public authority whether it holds [NAME_3] of the description specified in the request, and

(b) if that is the case, to have that [NAME_3] communicated to him.

…

S.3(1) In this Act “public authority” means—

(a) … any body which, any other person who, or the holder of any office which—

(i) is listed in Schedule 1”.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 15

23. The bodies listed in Part VI of Schedule 1 include

“The [NAME_4], in respect of [NAME_3] held for purposes other than those of journalism, art or literature.”

(Three other public bodies have the same designation: Channel 4 Television Corporation, the Gaelic Media service and Sianel Pedwar Cymru (the Welsh television station known as S4C).

24. S.7(1) provides that

“Where a public authority is listed in Schedule 1 only in relation to [NAME_3] of a specified description, nothing in Parts I to V of this Act applies to any other [NAME_3] held by the authority”.

The effect of s.7(1), read with Part VI of Schedule 1, is that [NAME_3] held by [NAME_16] is not disclosable pursuant to FOIA if it is held for the purposes of journalism, art or literature.

The Relevant Date 25. [NAME_1] originally submitted that the relevant date at which the Tribunal should have considered whether the requested [NAME_3] was held for the purposes of journalism was the date of its decision. It was therefore wrong to decide, as it had done in [67], that the appropriate date at which to determine the purpose for which the [NAME_3] was held was July 2021, the date of [NAME_16]’s replacement substantive decision. It should rather have asked itself whether, at the date of its own decision, namely 7 February 2022, the [NAME_3] was held for the purposes of journalism. That was no academic error because it would have been even less plausible for the Tribunal to consider in February 2022 that the instructions given to [NAME_20] which had led to the survey quoted in the 2018/2019 Annual Report were still being held for the immediate object of journalism (if indeed they were ever so held) than it was for the Tribunal to consider that to have been the case in July 2021. The passage of time was all important, as [NAME_25] MR

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 16 recognised in the observation in the Court of Appeal, quoted in the Supreme Court at [66], that [NAME_3] held for the purposes of journalism might soon stop being held for that purpose.

26. By contrast, it was the [NAME_2]’s position that both the Tribunal and [NAME_1] were wrong and that the relevant date was 25 June 2020 (that being the date of [NAME_16]’s response to the request). [NAME_16] supported that position.

27. In his skeleton argument [NAME_7] ultimately accepted (at least before the Upper Tribunal) that the relevant date was 25 June 2020 (as being the date of [NAME_16]’s response to the request), so that all three parties were agreed on the same relevant date.

28. In my judgment, that agreement was rightly reached and the correct relevant date was 25 June 2020 (as being the date of [NAME_16]’s response to the request).

29. I agree with the [NAME_2]’s submission that the time to assess the application of the derogation should be based on the same principles as apply to the consideration of the applications of the exemptions under FOIA and the application of the public interest test under s.2.2 FOIA. In that context the Upper Tribunal has very recently considered the time at which to apply the public interest test in s.2(2) FOIA in Montague v Information Commissioner [2022] UKUT 104 (AAC). In Montague the Upper Tribunal set out the preceding case law on timing, including R(Evans) v HM Attorney General [2015] UKSC 21, [2015] 1 AC 1787, APPGER v IC and FCO [2015] UKUT 377 (AAC); [2016] AACR 5 and Maurizi v IC and CPS [2019] UKUT 262 (AAC). Having considered that case law the Upper Tribunal held:

“86. ... The public authority is not to be judged on the balance of the competing public interests on how matters stand other than at the time of the decision on the request which it is has been obliged by Part I of FOIA to make.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 17

87. We therefore conclude that the FTT erred in law in its decision ... in not confining itself to assessing the balance of the competing public interests for and against disclosure on the basis of matters as they were at the date of DIT's (initial) refusal decision of 8 February 2018 ...

89. ... The correct approach was for the FTT to ask, in respect of each piece of [NAME_3] separately, whether at the date of the 8 February 2018 refusal decision, the public interest in maintaining a given exemption outweighed that in favour of disclosure, taking account of anything that was already actually in the public domain as at 8 February 2018.”

There is no obvious reason why the timing for considering the application of the derogation should be different from the timing when considering the other provisions under FOIA. Applying the principles in Montague to the derogation, the correct date was 25 June 2020.

The Decision in [NAME_21] 30. [NAME_29] sought disclosure pursuant to s.1 of FOIA of an internal report (the [NAME_29] report) commissioned by [NAME_16] on the quality of its coverage, including its impartiality, of the Middle East. By s.7(1) and Part VI of Schedule 1, [NAME_16] was under a duty to communicate [NAME_3] only if it was “held “for purposes other than those of journalism, art or literature”. [NAME_16] refused to disclose the report and its decision was upheld by the [NAME_2] on the basis that it held the report for the purposes of journalism and that, even if it was also held for non-journalistic purposes, it was not disclosable because the journalistic purposes were manifestly dominant. On appeal, the [NAME_3] Tribunal held that, when originally commissioned, the report was for predominantly journalistic purposes and was not disclosable, but at the time of [NAME_29]’s request it was being used for purposes predominantly other than those of journalism, namely for purposes of strategic policy and resource allocation and so was no longer exempt from production under FOIA. That decision was reversed by Irwin J, who rejected the contention that a document was disclosable unless held exclusively for the purposes of journalism. He

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 18 held that, provided journalism was one of the purposes for which the document was held, that was sufficient to take it out of the ambit of the Act. [ADDRESS] of Appeal upheld that decision.

31. The Supreme Court dismissed [NAME_29]’s appeal. By a majority of 4-1 ([NAME_29] dissenting on the issue of purpose) it held that, having regard to the language and legislative purpose of FOIA, [NAME_3] held by [NAME_16] to any significant degree for the purpose of journalism was not “held for purposes other than those of journalism” within the meaning of Part VI of Schedule 1, even if it was also held for other, possibly more important purposes, and was therefore exempt from production. However, the question whether [NAME_3] was held for the purposes of journalism should be considered in a relatively narrow way and, in determining whether [NAME_3] was currently held by [NAME_16] for those purposes, consideration should be given to whether there remained any sufficiently direct link between the continued holding of the [NAME_3] and the achievement of its journalistic purposes.

32. In his judgment, Lord Phillips PSC said that

“64. We are concerned with a provision that provides protection against the disclosure obligations that are the object of the Act. What is the purpose of that protection? It is not, as is the protection against disclosure of documents protected by legal professional privilege, designed to remove inhibition on the free exchange of [NAME_3]. Were that the case the protection would focus on the purpose for which the [NAME_3] was obtained. The protection is designed to prevent interference with the performance of the functions of [NAME_16] in broadcasting journalism, art and literature. That is why it focuses on the purpose for which the [NAME_3] is held …

65. A purposive construction of the definition will prevent disclosure of [NAME_3] when this would risk interference with the broadcasting function of [NAME_16]. This will not depend upon the predominant purpose of holding the [NAME_3]. It will depend upon the likelihood

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 19 that if the [NAME_3] is disclosed the broadcasting function will be affected …

66. [NAME_25] of Abbotsbury MR at para 53 remarked that “today’s journalism is tomorrow’s archive” and at para 58

“In the case of journalism, above all news journalism, [NAME_3] ‘held for purposes ... of journalism’ may soon stop being held for that purpose and be held, instead, for historical or archival purposes”.

… No doubt [NAME_16] has recourse to its archives for journalistic purposes from time to time and, if “held for purposes of journalism” is given a broad meaning it could be said in relation to [NAME_16] that one of the purposes of holding archived material is journalism, albeit a relatively remote purpose.

67. However, [NAME_25] accepted that archived material would not, as such, fall within the protection afforded by the definition. I consider that he was right to do so. Disclosure of material that is held only in the archives will not be likely to interfere with or inhibit [NAME_16]’s broadcasting functions. It ought to be susceptible to disclosure under the Act. If possible “[NAME_3] held for purposes other than those of journalism, art or literature” should be given an interpretation that brings archived material within that phrase. Can this be achieved? I believe that Lord Walker has the answer. He has concluded, as have I, that the protection is aimed at “work in progress” and “[NAME_4]’s broadcasting output”. He suggests that the Tribunal should have regard to the directness of the purpose of holding the [NAME_3] and [NAME_16]’s journalistic activities. I agree. [NAME_3] should only be found to be held for purposes of journalism, art or literature if an immediate object of holding the [NAME_3] is to use it for one of those purposes. If that test is satisfied the [NAME_3] will fall outside the definition, even if there is also some other purpose for holding the [NAME_3] and even if that is the predominant purpose. If it is not, the [NAME_3] will fall within the definition and be subject to disclosure in accordance with the provisions of Parts I to V of the Act.”

33. In his judgment Lord Walker JSC said that

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 20 “75. … In my judgment the correct view is that (as [NAME_25] MR put it at para 44):

“once it is established that the [NAME_3] sought is held by [NAME_16] for the purposes of journalism, it is effectively exempt from production under the Act, even if the [NAME_3] is also held by [NAME_16] for other purposes.”

So in effect there are only two categories: one is [NAME_3] held for purposes that are in no way those of journalism, and the other is [NAME_3] held for the purposes of journalism, even if it is also held for other (possibly more important) purposes.

…

78. In this case, there is a powerful public interest pulling in the opposite direction. It is that public service broadcasters, no less than the commercial media, should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress. They should also be free of inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses. A measure of protection might have been available under some of the qualified exemptions in Part II of FOIA, in particular those in sections 36 (Prejudice to effective conduct of public affairs), 41 ([NAME_3] provided in confidence) and 43 (Commercial interests). But Parliament evidently decided that [NAME_16]’s important right to freedom of expression warranted a more general and unqualified protection for [NAME_3] held for the purposes of [NAME_16]’s journalistic, artistic and literary output. That being the purpose of the immunity, section 7 and Schedule 1 Part VI, as they apply to [NAME_16], would have failed to achieve their purpose if the coexistence of other non- journalistic purposes resulted in the loss of immunity.

79. That is confirmed by the language of these statutory provisions. The disclosable material is defined in terms (“held for purposes other than those of journalism, art or literature”) which are positive in form but negative in substance. The real emphasis is on what is not disclosable – that is material held for the purposes of [NAME_16]’s broadcasting output …

…

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 21 83. In my view the correct approach is for the Tribunal, while eschewing the predominance of purpose as a test, to have some regard to the directness of the purpose. That is not a distinction without a difference. It is not weighing one purpose against another, but considering the proximity between the subject-matter of the request and [NAME_16]’s journalistic activities and end-product. As Irwin J observed in the financial [NAME_3] case, para 87, in the context of a critique of what was “operational”: “The cost of cleaning [NAME_16] is only remotely linked to the product of [NAME_16].”

84. I respectfully agree with the measured comments of [NAME_25] MR (para 55):

“In my view, whatever meaning is given to ‘journalism’ I would not be sympathetic to the notion that [NAME_3] about, for instance, advertising revenue, property ownership or outgoings, financial debt, and the like would normally be ‘held for purposes . . . of journalism’. No doubt there can be said to be a link between such [NAME_3] and journalism: the more that is spent on wages, rent or interest payments, the less there is for programmes. However, on that basis, literally every piece of [NAME_3] held by [NAME_16] could be said to be held for the purposes of journalism. In my view, save on particular facts, such [NAME_3], although it may well affect journalism-related issues and decisions, would not normally be ‘held for purposes . . . of journalism’. The question whether [NAME_3] is held for the purposes of journalism should thus be considered in a relatively narrow rather than a relatively wide way.”

34. For his part, Lord Brown JSC added

“85. All of us agree that on any conventional approach to the construction of the Freedom of [NAME_3] Act 2000 (the Act) and in particular the expression “[NAME_3] held for purposes . . . of journalism” within the meaning of Schedule 1 to the Act, it clearly encompasses the [NAME_29] (the Report) throughout the whole period that [NAME_16] has held it.

…

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 22

103. I turn then briefly to the question whether, in a case where [NAME_3] is held partly for journalistic and partly for non-journalistic purposes, it is necessary to ask which purpose is predominant and to disclose any [NAME_3] held predominantly for non-journalistic purposes. I conclude, in common with Lord Phillips and Lord Walker (and, indeed, with the Court of Appeal), but in respectful disagreement with [NAME_29], that the answer is no. My reasons being essentially the same as those given by both Lord Phillips and Lord Walker (although perhaps more particularly those of Lord Walker), I can explain my concurrence very shortly indeed. …

106. As for the point at which [NAME_3] will cease to be held to any significant degree for the purposes of journalism and become held instead, say, solely for archival purposes, that necessarily will depend on the facts of any particular case and involve a question of judgment. I too agree with Lord Walker that the central question to be asked in such a context will be, not which purpose is predominant, but rather whether there remains any sufficiently direct link between [NAME_16]’s continuing holding of the [NAME_3] and the achievement of its journalistic purposes.”

35. Finally for the majority, Lord Mance JSC stated

“108. The question on this appeal is whether the [NAME_29] commissioned by [NAME_16] in relation to its Middle Eastern coverage and completed in July 2004 constituted “[NAME_3] held for purposes other than those of journalism, art or literature” (within Part VI of Schedule 1 to the Freedom of [NAME_3] Act 2000). The appeal falls to be approached on the basis that the Report was at the material time held predominantly for journalistic but partly also for other purposes. The material time was in 2005, when [NAME_29] first requested disclosure of the Report.

…

110. … I have come to the conclusion that the test applied by Lords Phillips, Walker and Brown is to be preferred …

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 23

111. … I share Lord Walker’s view (para 79) that the real emphasis of the words is on what is not disclosable, so that the exemption applies, without more, if the [NAME_3] is held for any journalistic, artistic or literary purpose ...

112. Lord Phillips discusses the position regarding archived material. We were not given any clear picture when or on what basis archiving might occur. I assume that the reference is to material not envisaged as having any current purpose, but stored for historical purposes or against the possibility of some unforeseen need to revisit, or produce evidence of, past events. A library maintained for current reference would in contrast contain material held for the purposes of journalism, art or literature.”

36. [NAME_29] dissented on the question of predominance of purpose, but in the course of his judgment he also said that

“39. … In what circumstances will [NAME_16] hold [NAME_3] for the purposes of journalism? The Tribunal attempted to answer that abstract question; and the substantial criticism of its decision has been directed not at its analysis but at its application of its analysis to the circumstances in which [NAME_16] held the [NAME_29] report. Within the word “journalism” in the designation (which it described as “functional journalism” – a puzzling qualification in that, without elaboration, it implied the existence of other areas of journalism) the Tribunal identified three types of activity: first, the collecting, writing and verifying of material for publication; second, the editing of the material, including its selection and arrangement, the provision of context for it and the determination of when and how it should be broadcast; and third, the maintenance and enhancement of the standards of the output by reviews of its quality, in terms in particular of accuracy, balance and completeness, and the supervision and training of journalists. In relation to this third type, the Tribunal added, at para 116:

“Self-critical review and analysis of output is a necessary part of safeguarding and enhancing quality. The necessary frankness of such internal analysis would be damaged if it were to be written in an anodyne fashion, as would be likely to be the

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 24 case if it were potentially disclosable to a rival broadcaster.”

40. The Tribunal contrasted the three suggested types of journalistic activity with the direction of policy, strategy and resources which provides the framework within which a public service broadcaster conducts its operations.

41. In the Court of Appeal [NAME_25] said, at para 53, that, at any rate in the present context, he could not improve upon the Tribunal’s general analysis.

42. Apart from pointing out that its tripartite classification does not readily encompass the actual exercise of broadcasting or publishing the material, [NAME_16] does not quarrel with the Tribunal’s analysis of what falls within and without the concept of journalism for the purposes of the Act. In my view, and subject to that point, this court should endorse the Tribunal’s analysis but should decline [NAME_16]’s invitation to clothe it with greater specificity.”

37. [NAME_29] was in the minority on the question of the predominance of purpose, there is nothing to suggest that any of the other member of the Supreme Court disagreed with what he said in those paragraphs of his judgment.

The Grounds Of Appeal 38. In the hands of [NAME_7]’s predecessor, [NAME_27], who has since been ennobled and holds ministerial office, the grounds of appeal as attached to the form UT13 and dated 28 March 2022 were that the Tribunal had erred in law (1) by applying a different test to that set out in [NAME_21] as to the breadth of the journalism exception (2) in assessing when material could properly fall into the category of archival material and could therefore be excluded from the operation of the journalism derogation.

Ground 1 – application of incorrect test in light of [NAME_21] 39. The Tribunal considered the arguments relating to questions (or parts) 1 and 7 of the request at [74-90]. In [76] it accepted [NAME_16]’s contention, asserted in correspondence, that the [NAME_20] survey was to ‘better

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 25 understand audience perception’ with a view to ‘improving the quality and diversity of output’. It was clear from the wording of the paragraph that the Tribunal concluded that the feeding in of [NAME_3] from the survey to the annual performance reviews, as required by [NAME_16] charter, was a secondary purpose. That was an error on the material available to the Tribunal and it was noted that no or no adequate explanation was provided as to how that conclusion was reached.

40. The Tribunal also had no basis upon which reasonably to conclude, as it did in [77], that a principal use of [NAME_3] derived from survey was that it was ‘used by Editorial Standards and Policy in training sessions to explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape’. There was no evidence to support that conclusion beyond one assertion to that effect in [NAME_16]’s correspondence; certainly there was no adequate basis to ‘demote’ the use of the results from the survey in the Annual Review prepared in compliance with [NAME_16]’s charter obligations.

41. From the available evidence it was clear that the [NAME_3] was gathered for the purposes of the preparation of the annual report (being a Charter responsibility) and any other use was merely incidental. The Tribunal erred in ascribing, in the absence of any evidence, primacy to the other purposes.

42. The seat of the Tribunal’s error lay in [82-90]. In particular in [86-89], wrongly rejecting [NAME_1]’s submission, it held that the matters described in [83], namely the underlying contracts, the brief, instructions and any related meeting notes concerning the Survey (‘the underlying material’) were ‘not akin’ to the [NAME_3] referred to by [NAME_25] in [NAME_21] as being too distantly connected to purposes of journalism to benefit from the derogation.

43. That could be seen in [89] where it rejected his contention that underlying material (listed in [83]) was outwith the derogation because the [NAME_3] related to expenditure which was ‘specific to the editorial budget’. That

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 26 appeared to represent the decisive matter, together with the nature of the underlying material, in reaching the conclusion that that material was ‘at least in part for the purposes of journalism’.

44. That was to fall into error in the application of the test provided for in [NAME_21]. [NAME_1] submitted that it must, at least arguably, be that that application of the test was too broad and gave no or no sufficient consideration to [NAME_25]’s dictum that such matters should be considered ‘relatively narrowly’.

Ground 2 – Archive exception error 45. The Tribunal erred in its consideration of the application of the principle enunciated in [NAME_21] that the journalism exception did not have enduring application. The reasoning in [90] was inadequate to sustain the contention that material concerning the survey was still within the ambit of the exception. Moreover, the decision at [67] that the Tribunal should consider matters as from July 2021 was inconsistent with the principle that the Tribunal should consider those issues de novo.

46. [COMPANY_40] v [NAME_2] (EA/2006/0011 and EA/2006/0013) (‘the [NAME_31] case’) the then [NAME_3] Tribunal confirmed relevantly, of the appellate jurisdiction under s.50 FOIA (at paragraph [14]):

(1) that the Tribunal is not bound by the [NAME_2]’s views or findings, but would arrive at its own view (that was underlined by s.58(2) FOIA); it would give such weight to the [NAME_2]’s views and findings as it thought fit in the particular circumstances;

(2) in considering whether the [NAME_2]'s notice was in accordance with the law, the Tribunal must consider whether FOIA had been correctly applied. It was not bound by the [NAME_2]'s views or findings, but would arrive at its own view. In

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 27 doing so it would give such weight to the [NAME_2]'s views and findings as it thought fit in the particular circumstances.

(3) in cases where the correct application of FOIA would depend upon the findings as to disputed issues of fact, the Tribunal might review any finding of fact by the [NAME_2]. It would reach its conclusions on the factual issues upon the whole of the material which was properly before it on the appeal. Having decided the factual issues, it must consider the correct application of FOIA to the facts as found.

47. The guidance in the [NAME_31] case and the principle that the Tribunal could conduct a full merits review were approved by the Upper Tribunal in [NAME_2] v [NAME_24] & anr [2018] UKUT 72 (AAC) (at [45] and [90]).

48. Lord Phillips PSC noted in his judgment in [NAME_21] at [66-67], agreeing with [NAME_25] MR in the Court of Appeal, that archived material would not, as such, fall within the protection afforded by the definition. Disclosure of material which was held only in the archives, he accepted, would not be likely to interfere with or inhibit [NAME_16]’s broadcasting functions. Such material ought therefore to be susceptible to disclosure under FOIA.

49. Lord Walker concluded at [78] that protection was aimed at ‘work in progress’. Lord Phillips summarised the point by stating at [67]: ‘[NAME_3] should only be found to be held for purposes of journalism, art or literature if an immediate object of holding the [NAME_3] is to use it for one of those purposes’.

50. [NAME_1] contended that none of [NAME_3] requested in questions 1, 7 or 8 could be described as having an immediate use for one of the excluded purposes. It did not have such a use when the request was made on 26 March 2020, still less when [NAME_16] refused the requests on 25 June 2020 (and again in July 2021 and when the [NAME_2] refused the appeal on 2 [NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 28 2021 and when the Tribunal was considering those submissions in February 2022, although he ultimately accepted (at least before the Upper Tribunal) that the relevant date was 25 June 2020). Even if it had, it surely could not now be said, at this stage, that the [NAME_3] sought was held with the ‘immediate object’ of one the excluded purposes.

51. The Tribunal erred in accepting without scrutiny the assertions from [NAME_16], ostensibly accepted by the [NAME_2], that the material was ‘directly linked to [NAME_16]’s journalistic output’ (at [98]). Indeed, by commenting ‘There is no evidence before us that [the] complaints are no longer envisaged as having any current purpose, but stored for historical purposes after a certain date’ the Tribunal appeared to have imposed a burden on an applicant not envisaged in the [NAME_31] case. It was unreasonable and wrong in law to place an expectation on an applicant to prove that ‘complaints before a certain date were no longer referred to and therefore could not be seen as work in progress.’

52. [NAME_1] submitted that the data held in relation to the Annual Report and Accounts 2018-2019 in particular could have no immediate object in relation to one of the excluded purposes. The Tribunal failed to address that submission beyond the contention that he had not proven to the contrary. That amounted to the imposition of an impermissible gloss on evaluation of the facts. Once he had raised the reasonable contention that the 2018-2019 report could have no immediate object, the Tribunal, in accordance with the guidance in the [NAME_31] case, should have considered the evidence afresh rather than simply adopting the facts found by the [NAME_2], who in turn had accepted them from [NAME_16] without any, or any adequate, scrutiny. By the logic adopted by the Tribunal, complaints of seemingly any age would still have ‘a significant role in improving the quality of journalistic output’ (at [96]). That overly broad approach – of enduring relevance – adopted by the Tribunal was contrary to the requirement for ‘immediacy’ identified by Lord Phillips in [NAME_21].

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 29 53. In [91-99] the Tribunal failed to address submissions to the effect that, since immediacy of the object was the issue, the effluxion of time was an important consideration – for example complaints from 2015 were less likely to meet the requirement for immediacy than those nearer to the present date. The Tribunal erred in providing insufficient consideration of material over time, finding (at [99]) instead that because the whole of the complaints material from 2015-2020 was ‘held by the [COMPANY_26]’ that was suggestive that it was still in current use.

54. As the Supreme Court made clear in [NAME_21], consideration of the journalism exception required the decision-maker to strike the difficult balance of competing interests, which was the process which Parliament must be taken to have been aiming at in providing for the exception in the Act. Those were difficult decisions for the [NAME_2] and, on appeal, the Tribunal. There could not be (in the words of Davis J, at first instance, in BBC v Sugar [2007] 1 WLR 2583 at [57]) (“[NAME_21] (No.1)”) any ‘unequivocal, bright-line’ test. The broad-brush and unreasoned approach adopted by the Tribunal to the submission in respect of archival material amounted to the impermissible application of such a bright-line test.

55. When [NAME_7] came to address me at the permission hearing, he did not resile from those grounds and continued to rely on them, but put forward a note making certain points which overlapped with what [NAME_27] had originally said. (For the sake of completeness, I should make clear that I intended to grant permission to appeal on the basis of the grounds originally articulated by [NAME_27] and as refined by [NAME_7]. In any event, [NAME_7] had set out his arguments in his skeleton argument which he submitted on 6 February 2023. At the hearing of the appeal, the [NAME_2] and [NAME_16] sought, albeit fairly faintly, to argue that [NAME_7]’s arguments were outwith the scope of the grant of permission to appeal. I made clear, however, that he was not trespassing outside the scope of the grounds of appeal as I understood them to be and the appeal proceeded on that basis. In any event, in the circumstances which happened, because the originally scheduled date

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 30 for the hearing could not proceed, that skeleton argument was produced more than 3 months in advance of the substantive hearing of the appeal and the Respondents had more than adequate time to prepare their response to the appeal on that basis.)

Ground 1 - misapplication of the statutory definition (i) A question of law, not fact 56. For what purposes [NAME_16] held the [NAME_3] was, of course, a question of fact, but whether, in holding the [NAME_3] for those purposes, [NAME_16] was holding it ‘for purposes […] of journalism’ was a question of law because it turned on what Parliament meant by that expression.

57. That difference between, on the one hand, deciding what the facts of a particular case were and, on the other, deciding whether those facts fell within the scope of some statutory concept was a familiar one. To give a recent example, in Centrica Overseas Holdings Ltd v Revenue and Customs Commissioners [2022] EWCA Civ 1520, [2023] 1 WLR 316 the issue was whether certain expenses that the taxpayer had incurred constituted ‘expenses of a capital nature’ within the meaning of the Corporation Tax Act 2009. [ADDRESS] of Appeal, applying earlier authority, held that this was a question of law, adding (at para 78 per Singh LJ):

‘[i]t is therefore common ground that this court can and should arrive at its own conclusion on the Capital Expenditure issue but must do so on the basis of the findings of fact made by the FTT.’

It was submitted that the same was true here. The Tribunal could and should arrive at its own conclusion on the issue as to whether the [NAME_3] the subject of [NAME_1]’s request was (or, rather, was at the relevant time) ‘[NAME_3] held […] for purposes of journalism’ as those words were properly to be interpreted.

58. Of course, not every question as to whether something fell within some statutory definition gave rise to a question of law. So, for example, whether a

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 31 person was guilty of ‘insulting […] behaviour’ under the Public Order Act 1936 was a question of fact: Cozens v Brutus [1973] AC 854. The case of R (Thames Water Utilities Ltd) v Water Services Regulation Authority [2012] EWCA Civ 218, [2012] PTSR 1147 turned in part on a challenge to a regulator’s decision as to whether a development site was or was not ‘served’ with water and sewerage services by an existing undertaker within the meaning of the Water Industry Act 1991. Laws LJ (with whom [NAME_32] LJ and Kitchin LJ (as he was then) agreed) explained at [23-24] what he called ‘the nature of the question […] whether a statutory measure applied to a particular set of facts’, and gave further examples:

‘23. […] [T]his question is ambiguous. It may mean: is the statute to be construed so as to cover the accepted facts? That is a question of law. Or it may mean: are the facts to be judged as falling within the accepted meaning of the statute? That is a question of fact. The first question arises where there is no contest as to the evaluation of the facts, and the only issue is whether the statute is to be interpreted as covering those facts or not. An example far from the present case might be that of an imitation firearm. The statute prohibits the possession of firearms without defining the term. Does the provision on its true construction include the imitation weapon? The second question arises where there is no contest as to the meaning of the statute, and the only issue (an issue for a factual decision-maker) is whether the facts are to be evaluated as falling within the statutory rubric. An example equally far from the present case might be the statutory criminalisation of dangerous driving: the road traffic legislation uses but does not define the adjective “dangerous”. The decision-maker, the criminal court, having found the primary facts, must evaluate them: must decide whether they establish a case of dangerous driving.

24. This second class of case, where the facts must be evaluated to see whether they fall within the statutory rubric, arises where the legislature has used a term whose factual scope is a matter of judgment, even opinion. It may be a matter upon which reasonable people may disagree. In such a case the debate is not about the meaning of the statutory expression, and it will have been the intention of Parliament to consign the issue as to the expression’s application in a particular

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 32 case to the judgment of the appointed decision-maker. In the dangerous driving example there is never an argument in the magistrates’ court or the Crown Court as to what the word “dangerous” means as a matter of law; the argument is all about whether the facts before the court disclose a case of dangerous driving.’

Applying that analysis, the Court of Appeal declined to impeach the judgment of the regulator that the site in question was not ‘served’ by an existing undertaker (at [27]).

59. The present case, submitted [NAME_7], was firmly in Laws LJ’s first class of cases. It did not call for ‘evaluation’ of [NAME_16]’s activities vis-à-vis the requested [NAME_3], as one might evaluate whether a person’s driving ought to attract the adjective ‘dangerous’, or his behaviour the adjective ‘insulting’. The issue was, rather, whether the statute was to be interpreted as covering the facts or not.

60. That was not to say that it was [NAME_1]’s submission that the Upper Tribunal should ignore the decision of the Tribunal below. The Upper Tribunal should treat it with respect and not reverse it unless persuaded that it was wrong (see [NAME_21] in the Court of Appeal [2010] EWCA Civ 715, [2010] 1 WLR 2278 at [78] per Moses LJ). But ultimately, [NAME_7] submitted, it is for the Upper Tribunal to decide the issue, not merely to ask, as [NAME_16] suggested, whether the Tribunal below stated the law correctly, or applied it ‘faithful[ly]’, or reached a conclusion which was ‘open’ to it.

61. It was obvious (and was recognised in [NAME_21]) that, although one of [NAME_16]’s reasons for existence was journalism, [NAME_3] about, say, its financial outgoings on maintaining property was not ‘held for purposes […] of journalism’ as Parliament used the expression, even though having a building in which to work was a necessary element of producing journalism so that in one sense [NAME_16] maintained the property for the purposes of journalism. Likewise, it was obvious that, say, a reporter’s notes of what a source has said about the Prime Minister’s current intentions on some matter was ‘held

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 33 for purposes […] of journalism.’ 62. The issue in the present case was where, on the spectrum between those examples, the [NAME_3] fell.

(ii) The Tribunal’s answer to that question of law was wrong 63. The Tribunal considered, at [80]-[83], that the [NAME_3] relating to the 2018/2019 [NAME_20] survey all fell within the intended meaning of [NAME_3] ‘held for purposes […] of journalism’ because: (a) ‘presenting the figures on impartiality to the public in an annual report is intrinsically and directly linked to [[NAME_16]’s] journalistic output’; and (b) using the survey results to ‘“better understand audience perception of [NAME_4] programming with a view to improving the quality and diversity of output” [as [NAME_16] put it]’, and using them ‘in training sessions to “explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape” [again, the phrasing is [NAME_16]’s]’, such that the results ‘can directly impact on content’, fell ‘squarely within “journalism”’ because it ‘concerns the maintenance and enhancements of the standards and quality of journalism particularly with respect to balance’.

64. Stripped of [NAME_16]’s jargon and expressed more simply, the Tribunal’s conclusion was that [NAME_16] held the [NAME_3] from the 2018/2019 survey - i.e. the [NAME_3] underlying the published claim that 52% of people, when asked, thought that [NAME_16] provided impartial news and that 44% turned to [NAME_16] if they wanted impartial news -‘for purposes […] of journalism’ in the

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 34 sense that it used it when telling the public about public opinion on its degree of impartiality, in reflecting on how it was doing in maintaining impartiality and as a means of reminding [NAME_4] journalists of, and training them about, their obligation to be impartial.

65. [NAME_7] submitted that that was to adopt a much more expansive reading of ‘held for purposes […] of journalism’ than Parliament could have intended. He relied in support of that submission on (inter alia) Lord Phillips’ statements in [NAME_21] at [64] that

‘[t]he protection [i.e. the protection against disclosure afforded to [NAME_16] by the words “in respect of [NAME_3] held for purposes other than those of journalism, art or literature”] is designed to prevent interference with the performance of the functions of [NAME_16] in broadcasting journalism, art and literature’;

and at [65] that

‘[a] purposive construction of the definition will prevent disclosure of [NAME_3] when this would risk interference with the broadcasting function of [NAME_16]. This will not depend upon the predominant purpose of holding the [NAME_3]. It will depend upon the likelihood that if the [NAME_3] is disclosed the broadcasting function will be affected’;

and at [67] that

‘[…] the protection is aimed at “work in progress” and “[NAME_4]’s broadcasting output”. […] [T]he tribunal should have regard to the directness of the purpose of holding the [NAME_3] and [NAME_16]’s journalistic activities’ (all emphases added).

66. He also relied on Lord Walker’s statement at [78] that ‘[…] public service broadcasters, no less than the commercial media, should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress. They should also be free of

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 35 inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses.’

67. [NAME_16] to disclose the [NAME_3] underlying the two survey results published in the 2018/2019 Annual Report would not interfere with the performance of [NAME_16]’s work in progress, or inhibit [NAME_16] from monitoring its output so as to maintain standards.

68. The case was quite different from, for example, the facts of [NAME_21] itself, where the report whose disclosure was refused was an internal report into the quality of news reporting on a particular subject, such as was therefore likely to be used to change, in specific respects, how [NAME_16] did such reporting on that particular subject of journalistic investigation or coverage in future. Here, by contrast, [NAME_1] was simply concerned to know, in the public interest, how [NAME_16], as a public authority, came to publish the conclusion that 52% of the public in a survey thought that it was impartial and 44% turned to [NAME_16] for impartial news. That part of the [NAME_3], concerned as it was with the basis for those claims made by [NAME_16] in its 2018/2019 Annual Report, was far attenuated from the actual business of producing journalism (and it was irrelevant that routine financial [NAME_3] such as [NAME_3] about property ownership would be even more so).

69. The same, submitted [NAME_7], was true of the Tribunal’s conclusion in respect of complaints and [NAME_16]’s answers to complaints. The fact, if it were such, that [NAME_16] used complaints as a means of trying to ensure that future output is impartial (see [95]-[96] of the Tribunal’s decision) did not mean that the complaints themselves and [NAME_16]’s responses to complaints— which is what [NAME_1] wanted to see—were ‘held for purposes […] of journalism’. The complaints and responses were not, as the Tribunal curiously asserted at [99], ‘work in progress’. [NAME_1] was not asking to see [NAME_16]’s instructions to journalists as to how to carry out their journalism, but only to see what complainants said and what [NAME_16] said in reply. [NAME_16] would not be inhibited in its journalistic activities by being required to disclose that [NAME_3].

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 36

Ground 2 – the passage of time (i) The error as to the date 70. By the time of the hearing [NAME_7] had accepted (at least before the Upper Tribunal) that the relevant date for considering whether [NAME_3] was ‘held for purposes […] of journalism’ was 25 June 2020, being the date of [NAME_16]’s response to [NAME_1]’s request, but he submitted that the Tribunal went wrong in law in its approach to that issue.

71. As regards the 2018/2019 survey materials, the Tribunal’s reasoning consisted of the short assertion in [90] that (i) the [NAME_3] was still held in July 2021 (the date which the Tribunal used at [67]) for the same purposes, and (ii) there was ‘no evidence that [it] has been “archived” in any sense of that word.’

72. The latter was wrong insofar as it placed a burden on [NAME_1] to show that [NAME_16] itself had decided to place the [NAME_3] into a different category. The former was not an adequate analysis. Even if (contrary to [NAME_1]’s case) the [NAME_3] underlying the 2018/2019 survey was ‘held for purposes […] of journalism’ shortly after it was created, could that really have been true on 25 June 2020? Did there not come a point at which such [NAME_3] was so far removed from current journalism as not to be held for such purposes but merely to be of historical interest? The Tribunal’s reasoning did not engage with that at all.

73. As regards the audience complaints from 2015 to 2020 and [NAME_16]’s responses to them, the Tribunal’s conclusion was at [99]. In [NAME_1]’s submission, it was an error of law for the Tribunal thus to lump together all of the complaints and responses. While it was perhaps plausible that a specific complaint and [NAME_16]’s response to it, especially a recent one, might be said to be ‘held for purposes […] of journalism’ in the sense that it could be used to shape some journalistic work currently in progress so that it did not attract the same complaint, the Tribunal had no basis for asserting that that was true of

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 37 all complaints and responses over the whole 5-year period.

74. The fact, if it be such, that [NAME_16] used complaints as a means of trying to ensure that future output is impartial (see [95-96]) did not mean that the complaints themselves and [NAME_16]’s responses to complaints - which was what [NAME_1] wanted to see - were all ‘held for purposes […] of journalism’. The complaints and responses were not, as the Tribunal curiously asserted at [99], ‘work in progress’. [NAME_1] was not asking to see [NAME_16]’s instructions to journalists as to how to carry out their journalism, but only to see what complainants said and what [NAME_16] said in reply. [NAME_16] would not be inhibited in its journalistic activities by being required to disclose that [NAME_3].

75. In any event, to state as the Tribunal did, that the fact that complaints were held by the Audience Complaints Unit and Executive Complaints Unit meant that they were ‘still in current use’ was a plain non sequitur, assuming one meant - as the Tribunal must have meant - still in current use ‘for purposes […] of journalism.’

76. If the Upper Tribunal was with [NAME_1] on that ground, it was asked to remit the appeal to the Tribunal with a direction to redetermine the issue as to whether any item of [NAME_3] created ‘for purposes […] of journalism’ was still held for that purpose at the relevant date.

The [NAME_2]’s Submissions 77. In its response dated 12 December 2022, the [NAME_2] submitted that the Supreme Court in [NAME_21] was the leading authority on the application of the derogation. The Supreme Court concluded that:

(1) only [NAME_3] held for exclusively non-journalistic purposes would be within FOIA ([NAME_29] dissenting and advancing a “predominant purpose” test (at [57]). Lord Walker, representing the view of the majority, set out the purpose of the derogation in [78] and stated that the intention of [NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 38 would not be achieved if the predominant purpose test was adopted. He expanded upon that conclusion in [79] and stated that the “exclusive” or “predominant” test “raise almost insoluble problems in their application” and that Parliament “intended to lay down a workable test”. That was not to say that all of the [NAME_3] held by [NAME_16] could be classed as being held for journalistic purposes, as Lord Walker explained in [83-84]. Lord Mance agreed with Lord Walker at [111].

(2) on the issue of archival material and the point at which [NAME_3] would no longer fall within the derogation:

78. Lord Brown agreed with Lord Walker on the predominant purpose test and added at [106] that the point at which [NAME_3] would cease to be held to any significant degree for the purposes of journalism and became held instead, say, solely for archival purposes, would necessarily depend on the facts of any particular case and involve a question of judgment.

79. Similarly, Lord Phillips agreeing with Lord Walker (at [67]) held of archival material that disclosure of material which was held only in the archives would not be likely to interfere with or inhibit [NAME_16]’s broadcasting functions. It ought to be susceptible to disclosure under FOIA. If possible “[NAME_3] held for purposes other than those of journalism, art or literature” should be given an interpretation which brought archived material within that phrase. That could be achieved by applying Lord Walker’s test, namely that the protection was aimed at “work in progress” and “[NAME_4]’s broadcasting output”. The Tribunal should have regard to the directness of the purpose of holding the [NAME_3] and [NAME_16]’s journalistic activities. [NAME_3] should only be found to be held for purposes of journalism, art or literature if an immediate object of holding the [NAME_3] was to use it for one of those purposes. If that test was satisfied, the [NAME_3] would fall outside the definition, even if there was also some other purpose for holding the [NAME_3] and even if that was the predominant purpose. And see too Lord Mance on the issue of archival material at [112].

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 39 Approach of the Upper Tribunal to Appeals 80. The Upper Tribunal should also not “assume too readily that the Tribunal has misdirected itself just because not every step of its reasoning is set out” (Jones v First Tier Tribunal & CICA [2013] UKSC 19 per Lord Hope at [25].

81. Further, a challenge to factual findings of the First-tier Tribunal could only be advanced on the grounds of perversity, see Bangs v Connex South Eastern Ltd [2005] ICR 763 at [7].

Response to the Grounds of Appeal Ground 1: Application of the incorrect test in light of [NAME_21] 82. [NAME_1]’s Ground 1 concerned part 1 and 7 of the Request. He asserted that the Tribunal erred in applying the “wrong test”. That was simply an attempt to reargue a ground of appeal which failed below. There was no error of law in the Tribunal’s decision, which (i) carefully analysed the judgment in [NAME_21] stating the legal principles which could be derived at [53- 61], (ii) set out the facts at [75-78], and then (iii) at [80-90] carefully applied the test established in [NAME_21] to the factual circumstances, concluding that the requested [NAME_3] was held for journalistic purposes. On any fair reading of the decision as a whole there was no error of law.

83. The reasoning in the Grounds was wrong for the following reasons:

(i) it was not necessary for the Tribunal to conclude, and it did not conclude, that using [NAME_3] from the Survey in the annual reports was a “secondary purpose”.

(ii) if [NAME_1]’s appeal was based on the premise that the Tribunal’s findings of fact were not supported by the evidence available, that argument must be advanced as one of perversity. As explained in the decision there was sufficient evidence before the Tribunal for a reasonable decision maker to conclude that one of the reasons for which the [NAME_3] was held was currently and directly linked to its journalistic output: it certainly did not rise to

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 40 the level of perversity as would be required to conclude there was an error of law (see Bangs at [7]).

(iii) as to his assertion that the decision was inadequately reasoned, in this case the Tribunal provided careful and detailed reasoning, but even if a step in the reasoning was missing the Upper Tribunal should not assume too readily that the Tribunal had misdirected itself ([NAME_33]). That was particularly the case in circumstances where the Tribunal had set out in detail the legal principles it must apply, as derived from [NAME_21].

84. As to the assertion that the Tribunal had no basis on which to conclude that the “principal” use of the [NAME_3] was that it was used by Editorial Standards and Policy, (i) it was not necessary and the Tribunal did not conclude that that was the “principal” use; there was no such conclusion in the decision and it was not necessary for there to be such a conclusion for the derogation to apply; (ii) in circumstances where [NAME_3] had been provided by [NAME_16] in relation to the request including why the [NAME_3] was held and there was nothing to rebut [NAME_16]’s statement, the Tribunal’s findings of fact were reasonable. Again, it certainly did not rise to the level of perversity as would be required to conclude there was an error of law (see Bangs at [7]).

85. [NAME_1] had misread the decision and the legal principles derived from [NAME_21] in focusing on the “primacy” of the purposes for which [NAME_3] was held or produced. [NAME_3] might be produced and held for many different reasons. The Tribunal was not required to decide which one was “primary”; that would be unworkable (as per [NAME_21]). All that the Tribunal was required to do, and all that it did in this case was decide if one of those purposes was journalism.

86. [NAME_1] asserted that the [NAME_3] actually requested in Part 1 of the request, the underlying contracts, the brief, instructions and any related meeting notes concerning the survey (the “underlying material”), should not

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 41 fall within the derogation as they were too distantly connected to the purposes of journalism. That did not identify an error of law in the decision, but re-ran an argument which failed below. The Tribunal set out its reasons at [86-89] for concluding that the underlying materials were not akin to [NAME_3] referred to in the Court of Appeal ([2010] EWCA Civ 715) in the judgment of [NAME_25] MR. It concluded that, unlike [NAME_3] on property ownership or financial debt, the underlying materials which were directly used to commission the Survey were much more closely linked to [NAME_16]’s output. There was no error of law.

87. In summary, contrary to [NAME_1]’s criticisms there was no error of law in the application of the test set out in [NAME_21].

Ground 2: Archive error exception 88. The first issue raised by [NAME_1] was the issue of timing. The Tribunal concluded that the correct time at which to assess the application of the derogation was at the time of [NAME_4]’s new substantive response in July 2021 (see [67]) (in circumstances where there was no internal review). The [NAME_2] understood that to refer to [NAME_16]’s submissions to [NAME_2] dated 19 July 2021. [NAME_1] had originally asserted that the correct time was the date of the Tribunal’s consideration of the matter, on the basis that that was intended to be a full merits review.

89. The [NAME_2] disagreed with both [NAME_1] and the Tribunal. It was the [NAME_2]’s position that the relevant date was be 25 June 2020 (that being the date of [NAME_16]’s Response to the request), as indeed was accepted on all sides by the time of the hearing before me.

90. Turning to [NAME_1]’s main argument under ground 2, that the material requested was archival and not current, he was again advancing as a ground of appeal an argument which simply failed below. There was no error or law by the Tribunal.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 42 91. [NAME_1] complained that the Tribunal accepted [NAME_16]’s evidence, but, as above, that was not an unreasonable approach in circumstances where as the Tribunal held, there was no evidence to the contrary before it and it was only asked by [NAME_1] to “infer” that some of the complaints covered by Part 8 of the Request were not held for current purposes. That was not a perverse decision by the Tribunal, which was the standard which must be met to claim that a finding of fact amounted to an error of law.

92. [NAME_1] complained that the Tribunal did not “consider the evidence afresh”, but it was unclear upon what that assertion was based or indeed what he actually meant by “consider the evidence afresh”.

93. [NAME_1] asserted that because Part 8 of the Request contained a request for [NAME_3] dating back to 2015 it was somehow incumbent on the Tribunal to provide further reasoning as to why the [NAME_3] was still current. However, as explained by the Tribunal the [NAME_3] requested was held by [COMPANY_26]: it had not been archived by [NAME_16] and the Tribunal had no evidence which would support his inference that some of the [NAME_3] was not held for current purposes (see the decision at [99]).

94. The [NAME_2] accepted that there could not be an “unequivocal, bright line test” (as per [NAME_21]), but in the present case no such assertion was made by the Tribunal. Indeed, the decision reflected a careful application of the principles laid down in [NAME_21] to the present request.

[NAME_16]’s Submissions 95. In its formal response to the appeal dated 16 January 2023, [NAME_16] adopted the [NAME_2]’s submissions and submitted that the leading authority on the meaning of the phrase “purposes … of journalism, art or literature” was [NAME_21], from which the following principles could be taken:

(1) the derogation was “designed to prevent interference with the performance of the functions of [NAME_16] in broadcasting journalism, art and literature. That

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 43 is why it focuses on the purpose for which the [NAME_3] is held” (per Lord Phillips at [64]).

(2) the composite phrase “journalism, art or literature” seemed to be intended “to cover the whole of [NAME_16]’s output in its mission (under article 5 of its Royal Charter) to inform, education and entertain the public. On that comprehensive approach the purposes of journalism, art or literature would be, quite simply, the purposes of [NAME_16]’s entire output to the public” (per Lord Walker at [70]).

(3) there was a “powerful public interest” in relation to public service broadcasters, including [NAME_16], pulling against the strong public interest in requiring public authorities to provide [NAME_3] about their activities. That flowed from the fact that “public broadcasters, no less than the commercial media, should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress. They should also be free of inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses (per Lord Walker at [78]: see also per Lord Phillips at [64]).

(4) as to what constituted “journalism”, the Supreme Court approved the three types of activity which the First-tier Tribunal had identified:

(i) the collecting, writing and verifying of material for publication;

(ii) the editing of the material, including its selection and arrangement, the provision of context for it and the determination of when and how it should be broadcast; and

(iii) the “self-critical review and analysis of output is a necessary part of safeguarding and enhancing quality. The necessary frankness of such internal analysis would be damaged if it were to be written in an anodyne fashion, as would be likely to be the case if it were potentially disclosable to a rival

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 44 broadcaster” (per [NAME_29] at [39], affirming the analysis at [107-109] of the [NAME_3] Tribunal’s decision)

(5) the proper approach to construction of the ‘derogation’ was that “once it is established that the [NAME_3] sought is held by [NAME_16] for the purposes of journalism, it is effectively exempt from production under the Act, even if the [NAME_3] is also held by [NAME_16] for other purposes” (per Lord Walker at [75], approving the statement from [NAME_25] MR in the Court of Appeal) (emphasis added). In other words, the only category of [NAME_3] that [NAME_16] had to disclose was “[NAME_3] held exclusively for non-journalistic purposes” (per Lord Walker at [73]) (emphasis added). There were only two categories: “one is [NAME_3] held for purposes that are in no way those of journalism, and the other is [NAME_3] held for the purposes of journalism, even if it is also held for other (possibly more important) purposes” (per Lord Walker at [75]) (emphasis added).

(6) that approach did not mean that all of [NAME_16]’s [NAME_3] fell outside FOIA. It was necessary for the decision-maker to have “some regard to the directness of the purpose” (per Lord Walker at [83], original emphasis). That required the decision-maker to consider “the proximity between the subject- matter of the request and [NAME_16]’s journalistic activities and end-product” (id). Hence, the cost of cleaning [NAME_16] boardroom, or [NAME_3] about advertising revenue, property ownership or outgoings and financial debt, although remotely linked to the output of [NAME_16], would not be held “for purposes … of journalism”.

(7) the specific question for the Supreme Court had been whether archival material (the [NAME_29] commissioned by [NAME_16] in relation to its Middle Eastern coverage) fell within scope of the derogation. In that context, the relevant question was said to be whether there was a “sufficiently direct link” between the holding of that [NAME_3] and “the achievement of its journalistic purposes” (per Lord Brown at [106]). Material which did not “interfere with or inhibit [NAME_16]’s broadcasting functions” (per Lord Phillips at [67]), or which

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 45 was “not envisaged for any current purpose” (per Lord Mance at [112]), was not held for the purposes of journalism.

96. A similar set of principles to those set out above were summarised by the First-tier Tribunal in [NAME_32] v [NAME_2] & [NAME_4] (EA/2014/0298) at [17-18].

97. The principles in [NAME_21] had been considered and applied in subsequent cases.

98. In University and Colleges Admissions Service v Information Commissioner and Lord Lucas [2014] UKUT 0557 (AAT) at [53-55], the Upper Tribunal affirmed that the question of whether [NAME_3] was “held for purposes other than” a designated function could not be resolved by a consideration of the “predominant purpose”. The correct test, as set out in [NAME_21], was whether there was a “sufficiently direct link” between the holding of that [NAME_3] and the exercise of the designated function (id).

99. In [NAME_34] v [NAME_2] & [NAME_4] (EA/2018/0264), the Tribunal considered that the Supreme Court in [NAME_21] “was not primarily concerned…with the distinction between current and archival material but with the directness of the connection between the requested [NAME_3] and its journalistic purpose” (at [50]). The relevant question was simply whether there was a “sufficiently proximate relationship” between the [NAME_3] requested and [NAME_16]’s journalistic purpose (id).

100. In [NAME_35] v [NAME_2] (EA/2017/0017), the Tribunal considered whether an internal memorandum drafted by [NAME_16]’s former Head of the Political Research Unit (which included his analysis of the EU referendum poll) was held for the purposes of journalism. The appellant argued that the memorandum was not held for the purposes of journalism because it was “not a collecting or gathering of materials for publication” but rather “disclose[s] policy assumptions behind [NAME_16] editorial process and

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 46 assertive assumptions conditioning [NAME_4] writing and outputs” (at [14]). He explained that the basis of his appeal was that “the defence of “editorial process” did not apply because the memo revealed a collective mindset behind all editorial process, deeper than the process itself and conditioning it, going against balanced editing and journalism” (at [17]).

101. The First-tier Tribunal rejected the appellant’s reliance on “strong public interest considerations” (of the kind described above) as “misconceived” (at [18]). It explained that the nature of the derogation, which was such that the right of access did not apply at all, meant that there was “no scope for applying a public interest balance, in deciding whether the [NAME_3] in question should be disclosed” (at [19]). The Tribunal further held that:

(1) the memorandum was “in essence, advising those involved in the editorial process relating to [NAME_16]’s news output, of conclusions that might be drawn, following the EU referendum result” (at [20]).

(2) it was “impossible to conclude that the memorandum had no direct link with journalism. To do so would be diametrically opposed to the majority judgments of the Supreme Court in [NAME_21]” (at [21]).

(3) as regards the “three elements of journalism” (as set out above), the memorandum was “editorial in nature as well as being directly relevant to the third element, and in particular to ‘the maintenance of the standards and quality of journalism (particularly with respect to accuracy, balance and completeness)’ and to the training and development of journalists” (at [21]).

102. In [NAME_1] v [NAME_2] (EA/2021/0290), the Tribunal considered whether (i) material relating to Editorial Guidelines, training and additional guidance given to those reporting on news and current affairs; and (ii) editorial complaints was held for the purposes of journalism. The Tribunal concluded that they were (see also [NAME_36] v [NAME_2] (EA/2022/0175)).

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 47

Ground 1: the [NAME_2] did not apply a different test from that set out in [NAME_21] as to the breadth of the journalism exception 103. Ground 1 related to the [NAME_3] falling within Parts 1 and 7 of [NAME_1]’s request. The relevant reasoning was at [74-90]. His complaint was that

(1) it was not open to the Tribunal to conclude that that the “principal use” of the [NAME_3] was for the purposes of journalism

(2) in any event, the primary purpose for which the [NAME_3] was held was the preparation of [NAME_16]’s Annual Report, and any other use “was merely incidental” and

(3) the material was, in any event, “too distantly connected to purposes of journalism to benefit from the derogation”.

All three complaints were without merit.

104. There had been no misdirection or misinterpretation of law by the Tribunal regarding the principles set out in [NAME_21]. It accurately set out the relevant principles from [NAME_21] at [53-61]. That summary of the law was unimpeachable and was materially consistent with that which had been applied in subsequent case-law. There was no basis on which [NAME_1] could impugn the Tribunal’s comprehensive and careful consideration of those principles.

105. As regards (1), that was a direct challenge to the Tribunal’s finding of fact. Insofar as it was contended that its findings were not supported by the evidence available to it, [NAME_1] must satisfy the perversity threshold and did not come close to doing so. The material relied upon by the Tribunal was set out in detail at [75-81]. It was based on material contained in the [NAME_2]’s

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 48 Decision Notice at [29-42], and the communications from [NAME_16] dated 19 July 2021.

106. [NAME_16]’s letter stated (citations omitted):

“22. [NAME_3] about how [NAME_16] instructs [NAME_20] under terms and reference, including underlying internal correspondence and notes about the process; the cost of commissioning the survey; and how it presents that [NAME_3] in the Annual Report & Accounts and how it is interpreted and digested internally, are held for journalistic purposes and so not subject to the FOI Act.

23. The [NAME_20] survey is one of the many ways that [NAME_16] seeks to better understand audience perception of [NAME_4] programming with a view to improving the quality and diversity of output.

24. [NAME_3] about the underlying scoping of the survey, and the terms provided to [NAME_20] by [NAME_16] – beyond that which is already publicly available – would disclose internal considerations about how [NAME_16] perceives of its content and how it seeks to engage audiences. The survey costs and associated [NAME_3] like [NAME_16]’s terms of engagement with [NAME_20] are also considered by [NAME_16] to be journalistic in nature, deriving from editorial budgets that are determined by editorial teams alongside other budgetary decisions like programming costs. Any decision taken on costs has a direct impact on the creative scope for the programme and for other programmes because more money spent on one area or one programme means less available for another. Decisions of [NAME_2] have consistently determined that programme costs derived from content or news budget are held for the purposes of journalism.

25. The survey results are analysed by [NAME_16] Board, [NAME_4] Board and Executive Committee which in 2018 included the Head of News and Editorial Policy & Standards. The survey also feeds into annual performance reviews of [NAME_4] output to understand how [NAME_16] is meeting its public service mission under its Royal Charter to ‘inform, entertain and educat[e]’ audiences.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 49 26. In addition, the results are used by Editorial Standards and Policy in training sessions with [NAME_4] staff to help explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape – for instance how elections and referendums impact – as a way of reinforcing how audiences will regard [NAME_4] coverage. This can directly impact on how content is ‘signposted’ in a programme. For instance, explaining that certain content like a tweet from a politician on a specific issue of interest that is included in a news segment, is only one part of a debate on that issue.”

107. On the basis of the material that was before it (which was not challenged, or contradicted by any evidence submitted by [NAME_1]), it was plainly not perverse for the Tribunal to conclude that one of the reasons for which the [NAME_3] was held was directly linked to [NAME_16]’s journalistic output.

108. As regards (2), his complaint was based on a misunderstanding of the test in [NAME_21]. His complaint, in essence, was that the Tribunal erred in demoting the preparation of [NAME_16]’s Annual Report to a “secondary purpose”. In fact, there was no such finding. The Tribunal did not identify any “dominant” or “secondary” purpose. That was entirely consistent with the approach mandated by the Supreme Court’s decision in [NAME_21]. In that case, the Supreme Court rejected the dominant purpose test. As the majority made clear, “once it is established that the [NAME_3] sought is held by [NAME_16] for the purposes of journalism, it is effectively exempt from production under the Act, even if the [NAME_3] is also held by [NAME_16] for other purposes” (at [75]). The only category of [NAME_3] that [NAME_16] had to disclose was “[NAME_3] held exclusively for non-journalistic purposes” (at [73]). That was not this case.

109. As regards (3), [NAME_1] asserted that the [NAME_3] falling within Part 1 of his request was “too distantly connected to purposes of journalism to benefit from the derogation”. He relied on [NAME_25] MR’s examples of [NAME_3] which would not “normally” be held for journalistic purposes each fell within a narrow category of operational [NAME_3] (i.e. “advertising revenue, property ownership or outgoings, financial debt”) (at [55] of the [NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 50 of Appeal’s judgment ([2010] EWCA 715]). The Tribunal considered that argument in detail and rejected it for the reasons given at [86-89]. That reasoning was unimpeachable. [NAME_16] emphasised the following points:

(1) a correct statement of the law on the journalism derogation was contained in the full set of principles set out in [NAME_21]

(2) Lord Walker’s adoption (at [84]) of [NAME_25] MR’s statement must be read in the context of the preceding paragraph in which he addressed the need for “directness of the purpose” (at [83]). The settled approach to determining directness was to ask whether there was any “sufficiently direct link” between the holding of the [NAME_3] and the achievement of the journalistic purposes (per Lord Brown at [106]). Lord Walker’s statement did not represent any departure from that approach (and was not treated as such by the other majority judgments in [NAME_21])

(3) [NAME_25] MR’s examples of [NAME_3] which would not “normally” be held for journalistic purposes each fell within a narrow category of operational [NAME_3] (i.e. “advertising revenue, property ownership or outgoings, financial debt”). The [NAME_3] requested by [NAME_1] was altogether different and it was plainly open to the Tribunal to find that the requested [NAME_3] “is much more closely linked to [NAME_16]’s output because it was directly used to commission a survey to be used to influence content and it includes ‘internal considerations about how [NAME_16] perceives of its content and how it seeks to engage audiences’” (at [87]). That was a faithful application of the test in [NAME_21] and [NAME_1] had identified no error of law.

Ground 2: the [NAME_2] did not err in its consideration of the application of archive exception 110. Ground 2 related to the [NAME_3] falling within Parts 1, 7 and 8 of the request. [NAME_1] contended that archived material would not fall within the journalism exemption and that the Tribunal erred in rejecting his submission at

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 51 least some of this material falls within the archive exception. That complaint was also without merit.

111. [NAME_16] endorsed the submissions of the [NAME_2] in relation to timing and agreed that the relevant date was 25 June 2020, that being the date of [NAME_16]’s response to [NAME_1]’s request.

112. [NAME_1] had identified no error of law on the part of the Tribunal. In its consideration of Parts 1 and 7 (at [90]) and Part 8 (at [98-99]), the Tribunal considered the issue of timing and concluded that the [NAME_3] was still held for the purposes for which it had originally been held. That was unsurprising given the nature of the requests and the material which was before the Tribunal.

113. As regards Parts 1 and 7 of [NAME_1]’s request:

(1) he submitted that that the data held in relation to the Annual Report “could have no immediate object in relation to one of the excluded purposes”. That submission was inconsistent with the material which was before the Tribunal (see, in particular, paragraphs 25-26 of [NAME_16]’s 19 July 2021 letter and the Decision Notice at [31]).

(2) he submitted that the Tribunal failed to address his submission “beyond the contention that the Appellant had not proven to the contrary”. That was wrong. The point was addressed at [90] and the Tribunal made clear that it concluded, on the basis of the [NAME_3] before it (addressed at [75-89]), that “the [NAME_3] in parts 1 and 7 was still held for those purposes”. That was the Tribunal’s assessment of the evidence that was before it. That assessment was not in any sense perverse. There was no “impermissible gloss on evaluation of the facts”, and no failure to consider the “evidence afresh”.

114. As regards Part 8 of the request:

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 52

(1) the Tribunal and the [NAME_2] had consistently accepted that one of the purposes for which [NAME_16] held [NAME_3] about complaints was for the purpose of journalism (see, e.g., the recent decision in [NAME_37] v [NAME_2] (EA/2021/0065P) at [18] and [NAME_1] v [NAME_2] (EA/2021/0290) at [59]). That did not appear to be disputed by [NAME_1].

(2) his sole complaint appeared to be that at least some of the complaints mighty now be archived and therefore no longer held for the purpose of journalism. That argument was considered and rejected at [99]. The Tribunal’s finding of fact was based on the fact that [NAME_16] had indicated in its 19 July 2021 letter that the [NAME_3] “is held by [COMPANY_26] who administer [NAME_4] and the Executive Complaints Unit who are the appellate unit in [NAME_16] responsible for handling escalated editorial complaints under [NAME_16]’s Complaints Framework” (emphasis added). It was plainly not perverse for the Tribunal to reach the conclusion which it did in light of the material which was before it.

(3) for completeness, [NAME_16] also referred to the evidence considered by the First-tier Tribunal in [NAME_1] v [NAME_2] (EA/2021/0290), which expressly addressed the issue of retention of complaints (see [12] of the witness statement extracted at [40] of that judgment) and was consistent with the material before the Tribunal in the present case and with the factual conclusion reached by it.

Discussion Ground 1 115. The first part of the first ground of appeal in the form originally articulated by [NAME_27], and as set out in paragraphs 39 to 41 above, was fundamentally misconceived. The Tribunal did not ascribe dominance or primacy and secondary or incidental classification to [NAME_16]’s purposes. Indeed to have done so would have been inconsistent with the decision in [NAME_21], which had eschewed any test of predominance of purpose (as to

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 53 which see also Upper Tribunal Judge Wikeley in University and Colleges Admissions Service v Information Commissioner and Lord Lucas [2014] UKUT 0557 (AAT) at [53-55]). One can therefore see why it was that [NAME_7] chose to plant his standard on a hill adjacent to that of his predecessor in his reformulation of the ground of appeal.

116. For his part [NAME_7] accepted, at least for the purposes of the hearing before me, that the Tribunal was entitled to accept [NAME_16]’s explanation on the factual question of how it in practice used the [NAME_3] referred to in parts 1 and 7 of the request (in my judgment he could hardly have done otherwise). The appeal had therefore to proceed on the basis that [NAME_16]’s purposes in holding the [NAME_3] were as found by the Tribunal and the question was whether the holding of the [NAME_3] for those purposes was [NAME_3] held for the purposes of journalism, which was a question of law.

117. I shall proceed therefore to consider the first ground of appeal in the way formulated by [NAME_7] on [NAME_1]’s behalf before turning to consider the original formulation of the ground of appeal.

118. In the present case, the Tribunal considered at [80-83] that the [NAME_3] relating to the 2018/19 [NAME_20] survey all fell within the intended meaning of ‘[NAME_3] held for purposes […] of journalism’ because ‘presenting the figures on impartiality to the public in an annual report is intrinsically and directly linked to [[NAME_16]’s] journalistic output’ and using the survey results to ‘“better understand audience perception of [NAME_4] programming with a view to improving the quality and diversity of output” [as [NAME_16] put it]’ and using them ‘in training sessions to “explain how perceptions of impartiality relate to [NAME_16]’s editorial landscape”, fell ‘squarely within “journalism”’ because it ‘concerns the maintenance and enhancements of the standards and quality of journalism particularly with respect to balance’.

119. I agree with [NAME_7] that on the facts of [NAME_21] it is unsurprising that

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 54 [NAME_16] won that case. As Lord Brown said at [85]:

‘All of us agree that on any conventional approach to the construction of […] the expression “[NAME_3] held for purposes … of journalism” within the meaning of Schedule 1 to the Act, it clearly encompasses the [NAME_29] report […] throughout the whole period that [NAME_16] has held it.’

The [NAME_29], the subject of that case, though not as closely linked to the activity of journalism as, say, a reporter’s notes of what a source had just told her, was nevertheless closely linked to it. It was a report about [NAME_16]’s coverage of the Israel/Palestine conflict. As the [NAME_3] Tribunal put it (see Lord Brown at [102]), ‘Self-critical review and analysis of output is a necessary part of safeguarding and enhancing quality.’ The point was that the report consisted of a review of [NAME_16]’s existing output in a specific area, with a view to improving its output in that field in the future.

120. By contrast, argued [NAME_7], in this case the Tribunal had adopted a more expansive reading of ‘[NAME_3] held for purposes […] of journalism’ than Parliament could have intended. In support of his submission he relied on what Lord Phillips said at [64-65] and [67], what Lord Walker had said at [78] and [83-84] and Lord Brown at [106].

121. [NAME_16] to disclose the [NAME_3] underlying the survey results published in the 2018/2019 Annual Report would not, argued [NAME_7], interfere with the performance of [NAME_16]’s work in progress or inhibit [NAME_16] from monitoring its output so as to maintain standards. The case was quite different from the facts of [NAME_21] itself. [NAME_1] was simply concerned to know, in the public interest, how [NAME_16], as a public authority, came to publish the conclusion that 52% of the public in a survey thought that it was impartial and 44% turned to [NAME_16] for impartial news. That was far attenuated from the actual business of producing journalism (and it was irrelevant that routine financial [NAME_3] such as [NAME_3] about property ownership would be even more attenuated). There was little proximity, and

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 55 certainly no direct link, between the [NAME_3] and [NAME_16]’s journalistic activities and end-product. [NAME_7] accordingly contended that the Tribunal reached the wrong answer on that question. (And if, contrary to [NAME_1]’s case as now advanced, he needed to show that the Tribunal reached a conclusion which was not open to it, he argued that he could.)

122. [NAME_7] frankly accepted that the first and second sentences of paragraph 80 of the Tribunal’s decision were questions of fact and that the Tribunal was right (or at least entitled) to reach the conclusions which it did (see paragraph [116] above), but he said that the third sentence was a question of law and that the Tribunal was wrong to conclude as it did. Paragraph 81 was a matter of fact and the Tribunal was right in its conclusion, but paragraph 82 was a matter of law and the conclusion was wrong.

123. Thus, to put the question in context, in paragraphs 80 to 83 [NAME_7] accepted that he could only challenge the words which I have italicised below:

“80. We accept that is one of the purposes of the survey. In our view impartiality is a fundamental aspect of [NAME_16]’s journalism output. We find that presenting the figures on impartiality to the public in an annual report is intrinsically and directly linked to its journalistic output.

82. In our view, this falls squarely within ‘journalism’. It concerns the maintenance and enhancement of the standards and quality of journalism (particularly with respect to balance). It can directly impact on the output of [NAME_16].”

124. In that last paragraph the Tribunal was mirroring very closely the language of [NAME_29] in [NAME_21] at [39] where he referred to the third category of journalistic material, namely that concerning “the maintenance and

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 56 enhancement of the standards of the output by reviews of its quality, in terms in particular of accuracy, balance and completeness, and the supervision and training of journalists”. I do not accept the argument of [NAME_7] that what [NAME_29] said about the ambit of journalistic activity at [39] should not be regarded as binding. Although the other member of the Court did not explicitly endorse [NAME_29]’s tripartite definition, they did not dissent from it (in contrast to his minority view as to the predominance of purpose test) and I see no reason to doubt that his tripartite definition accurately describes the ambit of journalistic activity. Moreover, [NAME_29] noted (with apparent approval) that in the Court of Appeal [NAME_25] MR said at [2010] 1 WLR 2278 at [53] that, at any rate in the present context, he could not improve on the tribunal’s general analysis. In that respect I can see no inconsistency between the views of [NAME_29] and the conclusions of the majority.

125. I accept that the material encompassed in parts 1 and 7 of the request was not as closely linked to [NAME_16]’s journalistic purposes as the journalist’s notes of an interview with a politician and that it was not as closely linked to those purposes as the [NAME_29] report. Equally, I accept that [NAME_3] about advertising revenue, property ownership or outgoings, financial debt and the like would not normally be held for the purposes of journalism. I equally accept that the question whether [NAME_3] is held for the purposes of journalism should be considered in a relatively narrow, rather than a relatively wide way, lest it be said that literally every piece of [NAME_3] held by [NAME_16] could be said to be held for the purposes of journalism (see [NAME_25] at [55] in the Court of Appeal and Lord Walker at [84] in the Supreme Court).

126. However, wherever along the spectrum between the journalist’s interview notes and the [NAME_29] report on the one hand and the cost of cleaning the board room on the other the line is to be drawn, in my judgment the material encompassed in parts 1 and 7 of the request was closer to the journalist’s interview notes and the [NAME_29] report than to the cost of cleaning the board room or routine financial [NAME_3] and was therefore entitled to the

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 57 protection of the derogation from disclosure for material held for the purposes of journalism.

127. [NAME_7] argued that the [NAME_3] in parts 1 and 7 of the request was not directly about something, unlike the case of the [NAME_29] report, and that the [NAME_3] sought was at a much higher level of generality than the [NAME_29] report. I accept that, but that does not detract from the conclusion that the material encompassed in parts 1 and 7 of the request was closer to the journalist’s interview notes and the [NAME_29] report than to the cost of cleaning the board room or routine financial [NAME_3] and was therefore entitled to the protection from disclosure.

128. The requested [NAME_3] in parts 1 and 7 of the request was much more closely linked to [NAME_16]’s output because it was directly used to commission a survey to be used to influence content and included internal considerations about how [NAME_16] perceived its content and how it sought to engage audiences. So far as the directness of the purpose is concerned, the requested [NAME_3] was (and is) sufficiently proximate to [NAME_16]’s journalistic purposes and its end product. It was indeed an immediate object of holding the [NAME_3] to use it for one of those purposes. It cannot be said of this material that it was only “remotely” linked to the product of [NAME_16] in the way that the cost of cleaning [NAME_16] boardroom is (see Irwin J in BBC v Information Commissioner [2009] EWHC 2349 (Admin), [2010] 1 WLR 2278, [2010] EMLR 121 at [87], in the context of a critique of what was “operational”, cited with approval by Lord Walker in [NAME_21] at [83]).

129. To make good his contention, [NAME_7] would have to show that the requested [NAME_3] was [NAME_3] held for purposes which were in no way those of journalism; he would not succeed if the [NAME_3] was held for the purposes of journalism, even if it was also held for other (possibly more important) purposes.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 58 130. I do not therefore accept [NAME_7]’s submission that the material now sought in the request was far attenuated from the actual business of producing journalism or that there was little proximity, and certainly no direct link, between the [NAME_3] and [NAME_16]’s journalistic activities and end- product. There is nothing in the judgments in [NAME_21] to compel any other conclusion, whether of Lord Phillips [64-65] and [67], Lord Walker at [78] and [83-84] or Lord Brown at [106] or otherwise. On the contrary, they are all supportive of the position of the [NAME_2] and [NAME_16], particularly in what Lord Phillips said to the effect that:

65. A purposive construction of the definition will prevent disclosure of [NAME_3] when this would risk interference with the broadcasting function of [NAME_16]. This will not depend upon the predominant purpose of holding the [NAME_3]. It will depend upon the likelihood that if the [NAME_3] is disclosed the broadcasting function will be affected.”

131. I am therefore satisfied that the Tribunal was correct in its conclusion as a matter of law when it held that

“86. In the tribunal’s view the requested [NAME_3], including, for example, the terms upon which the survey was to be produced and the brief is not akin to [NAME_3] about advertising revenue, property ownership or outgoings or financial debt. All of those are only very remotely linked to [NAME_16]’s output.

87. The requested [NAME_3] in this case is much more

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 59 closely linked to [NAME_16]’s output because it was directly used to commission a survey to be used to influence content and it includes ‘internal considerations about how [NAME_16] perceives of its content and how it seeks to engage audiences’. Looking at the directness of the purpose, we find that the requested [NAME_3] is sufficiently proximate to [NAME_16]’s journalistic purposes and the end product. It was an immediate object of holding the [NAME_3] to use it for one of those purposes.”

132. It must follow from that that the Tribunal was also correct to hold that

“83. On this basis, we conclude that all the peripheral [NAME_3] which was created in order to produce that survey (the underlying contracts, the brief, instructions, any related meeting notes) was held, at least in part, for the purposes of journalism.

84. As stated above, in our view, the presentation of these figures on impartiality in an annual report is intrinsically and directly linked to [NAME_16]’s journalistic output. Therefore the [NAME_3] which related to how the presentation of the results was changed was held, at least in part, for the purposes of journalism.”

133. As Lord Walker said in [NAME_21] at [75]

“In my judgment the correct view is that (as [NAME_25] MR put it at para 44):

So in effect there are only two categories: one is [NAME_3] held for purposes that are in no way those of journalism, and the other is [NAME_3] held for the purposes of journalism, even if it is also held for other (possibly more important) purposes.”

134. The conclusion which I have reached is in no sense surprising. It is consistent with the special consideration to which Parliament gave effect

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 60 when enacting the derogation enshrined in the journalism exception. To quote Lord Mance in [NAME_21]

“111. In the present case, the special consideration to which the legislator gave effect was the freedom of [NAME_16] as a public service broadcaster in relation to its journalistic, artistic and literary output. [NAME_3] held for any such purposes of journalism, art or literature was absolutely exempt from disclosure. The legislator was not content with the more qualified protection from disclosure, often depending on a balancing exercise or evaluation, which would anyway have been available under section 2, read with sections 28, 29, 36, 41 and 43. To read into the words “[NAME_3] held for purposes other than those of journalism, art or literature” a need to evaluate whether such purposes were dominant seems to me unjustified. I share Lord Walker’s view (para 79) that the real emphasis of the words is on what is not disclosable, so that the exemption applies, without more, if the [NAME_3] is held for any journalistic, artistic or literary purpose. That conclusion is to my mind also fortified by consideration of the exemption relating to certain functions of the Bank of England.”

135. That special consideration also underlay the judgment of Lord Walker when he said that

“78. In this case, there is a powerful public interest pulling in the opposite direction. It is that public service broadcasters, no less than the commercial media, should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress. They should also be free of inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses. A measure of protection might have been available under some of the qualified exemptions in Part II of FOIA, in particular those in sections 36 (Prejudice to effective conduct of public affairs), 41 ([NAME_3] provided in confidence) and 43 (Commercial interests). But Parliament evidently decided that [NAME_16]’s important right to freedom of expression warranted a more general and unqualified protection for [NAME_3] held for the purposes of [NAME_16]’s journalistic, artistic and literary output. That being the purpose of the immunity, section 7 and Schedule 1

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 61 Part VI, as they apply to [NAME_16], would have failed to achieve their purpose if the coexistence of other non- journalistic purposes resulted in the loss of immunity.”

136. As Lord Walker concluded in [NAME_21] at [84], there is a difficult balance to be struck in weighing the competing interests for which Parliament must be taken to have been aiming. That will leave difficult decisions for the [NAME_2] and, on appeal, the Tribunal, but there cannot be, in the words of Davis J ([2007] 1 WLR 2568 at [57]) any “unequivocal, bright-line” test.

137. I therefore turn to the remainder of the original formulation of the first ground of appeal, as stated by [NAME_27]. I can dispose of it briefly. I am satisfied that the Tribunal set out the law correctly in its exegesis of [NAME_21] in paragraphs 54 to 61. It is apparent that the Tribunal there set out the effect of the decision in [NAME_21] in some detail. There is nothing to suggest that it misquoted or misunderstood the law which it set out in those paragraphs. Indeed, given the extent to which the decision was set out, it would be difficult to argue that it had misquoted or misapplied the law. I am also satisfied that it then applied the law correctly, or at least reached a decision which it was entitled to reach, in paragraphs 80 to 90 and that the original formulation of the first ground of appeal does not advance [NAME_1]’s cause.

138. The upshot of this is that, whether this case falls within Laws LJ’s first class of cases and calls only for the resolution of the issue as to whether the statute was to be interpreted as covering the facts or not, or whether it falls within his second class of cases, so that the Upper Tribunal only has to decide whether the Tribunal below stated the law correctly, or applied it correctly, or reached a conclusion which was open to it, the result is the same and the ground of appeal fails.

139. Thus, whether the ground of appeal is (as now) put on the basis of the Tribunal put an incorrect construction on the words “held for purposes of journalism” or whether (as originally formulated) it is put on the basis that the

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 62 Tribunal applied an incorrect test in the light of [NAME_21], that makes no difference to the outcome of this appeal.

140. I therefore dismiss the appeal on Ground 1.

Ground 2 141. Part 8 of the request was

“8. Please also provide copies of all complaints to [NAME_16] about impartiality from 2015 to date and [NAME_16] responses to the same”.

142. [NAME_16]’s response dated 19 July 2021 stated that

“17. This [NAME_3] is held by [COMPANY_26] who administer [NAME_4] and the Executive Complaints Unit (“the ECU”) who are the appellate unit in [NAME_16] responsible for handling escalated editorial complaints under [NAME_16]’s Complaints Framework.

…

27. In a recent decision by [NAME_2] regarding a request to [NAME_16] for editorial complains, the Commissioner explained how [NAME_3] about editorial complaints was used by [NAME_16] for editorial purposes:

“[NAME_3] about complaints that [NAME_16] receives, including the number of complaints, is derogated [NAME_3]. This type of [NAME_3] is associated with [NAME_16]’s output because [NAME_16] will use [NAME_3] generated by the number and type of complaints it receives to make editorial decisions about its output; either its broadcast news content or content it publishes on its [NAME_4] website” (decision notice IC-90030-D3L5 of 31 March 2021 at [20]).

143. The Tribunal found that

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 63 “95. It is apparent from the [NAME_3] submitted to the Commissioner in this and in other investigations that the consideration of complaints by [NAME_16] is an important tool used by [NAME_16] to monitor, maintain and enhance its journalistic output and to ensure the impartiality of that output.

99. [NAME_27] has asked us to infer that at least some of the complaints in the requested period will no longer have been held for the requisite purposes at the relevant date. There is no evidence before us that complaints are no longer envisaged as having any current purpose, but stored for historical purposes, after a certain date. There is no evidence before us on which we could base a finding that complaints before a certain date were no longer referred to and therefore could not be seen as work in progress. [NAME_16] stated in July 2021 that the requested [NAME_3] was held by the [COMPANY_26] who administer [NAME_4] complaints and the Executive Complaints Unit who handle appeals. This suggests to us that it was still in current use. On this basis we find that the [NAME_3] was still held, at the relevant time, for the purposes of journalism.”

144. In the light of the evidence submitted by [NAME_16] in this case, I am satisfied that the Tribunal was entitled to reach the conclusions which it reached in paragraphs 95 to 98 of its decision. The real dispute was as to the inference which [NAME_1] invited the Tribunal to make, namely that at least

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 64 some of the complaints in the requested period were no longer held for the requisite purposes at the relevant date of 25 June 2020.

145. The practical problem with that submission on the facts of this case is that there is no way of dividing the complaints by date as to those which are still regarded as having some current purpose and those which are stored for historical or archival purposes after a certain date. There was nothing in the material before the Tribunal which would have entitled it to draw a line at say, 1 January 2016, or 31 December 2017, or any other date.

146. That is why the Tribunal found that there was no evidence before it on which it could base a finding that complaints before a certain date were no longer referred to and therefore could not be seen as work in progress. I can see no error of law in that conclusion.

147. [NAME_16] stated in July 2021 that the requested [NAME_3] was held by the [COMPANY_26] who administered [NAME_4] complaints and the Executive Complaints Unit who handled appeals. That suggested that it was still in current use. On that basis the Tribunal found that the [NAME_3] was still held, at the relevant time, for the purposes of journalism. (The relevant time is now agreed to be 25 June 2020 rather than 19 July 2021, but that does not affect the conclusion which the Tribunal reached.)

148. It is incontrovertible that [NAME_16] stated in July 2021 that the requested [NAME_3] was held by the [COMPANY_26] who administered [NAME_4] complaints and the Executive Complaints Unit who handled appeals. From that the Tribunal concluded that that suggested that it was still in current use. Again, on the evidence before the Tribunal I can see no error of law in that conclusion. Like the Supreme Court in [NAME_21], the Tribunal was not given any clear picture at to when or on what basis archiving might occur (see Lord Mance at [112]), but without it there was nothing in the material before the Tribunal which would have entitled it to draw a line at a particular date or with reference to a particular subject matter.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 65

149. That is not to place any sort of burden on [NAME_1], as [NAME_7] sought to argue, nor does it import any sort of presumption. It is simply a conclusion on the state of the evidence before the Tribunal.

150. [NAME_7] criticised the use of the term “work in progress” to describe the complaint and responses. I agree that it is curious to describe the complaints and responses as “work in progress”, but that does not avail him. What Lord Walker said in [NAME_21] at [78] was that

“In this case, there is a powerful public interest pulling in the opposite direction. It is that public service broadcasters, no less than the commercial media, should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress. They should also be free of inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses.”

151. So he was not confining himself to saying that public service broadcasters should be free to gather, edit and publish news and comment on current affairs without the inhibition of an obligation to make public disclosure of or about their work in progress alone. They should, in addition, be free of inhibition in monitoring and reviewing their output in order to maintain standards and rectify lapses. It is in that context that the complaints and responses are relevant.

152. As Lord Brown said in [NAME_21] at [106] there is a point at which [NAME_3] will cease to be held to any significant degree for the purposes of journalism and become held instead solely for archival purposes. That necessarily will depend on the facts of any particular case and involve a question of judgment. The central question to be asked in such a context will be, not which purpose is predominant, but rather whether there remains any sufficiently direct link between [NAME_16]’s continuing holding of the [NAME_3] and the achievement of its journalistic purposes. Clearly, material not

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 66 envisaged as having any current purpose, but stored for historical purposes or against the possibility of some unforeseen need to revisit, or produce evidence of, past events would not contain material held for the purposes of journalism, in contrast to a library maintained for current reference (see Lord Mance at [112]).

153. Archived material would not, as such, fall within the protection afforded by FOIA (see Lord Phillips at [67] and Lord Walker at [83]). That is because disclosure of material which is held only in the archives will not be likely to interfere with or inhibit [NAME_16]’s broadcasting functions and should be disclosable. When deciding whether such material is disclosable, the Tribunal should have regard to the directness of the purpose of holding the [NAME_3] and [NAME_16]’s journalistic activities. [NAME_3] should only be found to be held for purposes of journalism, art or literature if an immediate object of holding the [NAME_3] is to use it for one of those purposes. If that test is satisfied the [NAME_3] will fall outside the definition, even if there is also some other purpose for holding the [NAME_3] and even if that is the predominant purpose. If it is not, the [NAME_3] will fall within the definition and be subject to disclosure in accordance with the provisions of FOIA.

154. What the Tribunal must essentially consider in this context is the proximity between the subject-matter of the request and [NAME_16]’s journalistic activities and end-product. Something which is only remotely, or relatively remotely, linked to the product of [NAME_16] will not attract the protection of the derogation.

155. What, then, of [NAME_25]’s comments in the Court of Appeal at [53] and [58] that “today’s journalism is tomorrow’s archive” and “In the case … news journalism, [NAME_3] ‘held for purposes ... of journalism’ may soon stop being held for that purpose and be held, instead, for historical or archival purposes”?

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 67 156. [NAME_7] asked rhetorically: does there not come a point at which such [NAME_3] is so far removed from current journalism as not be held for such purposes but merely of historical interest? Clearly there does, but the short answer is that it is not demonstrated on the facts of this case. A request made on 17 June 2020 and answered 8 days later seeking copies of all complaints and [NAME_16]’s responses from, say, as recently as March 2020 would in all likelihood be held for the purposes of current journalism. There is no reason to believe that the answer would be different for complaints generated in the previous autumn, but there is no material in the present case to show where such a line might be drawn. [NAME_7] suggested that material from the beginning of 2015 was too old to be other than archival, but there is no reason in principle why that should not be so. I am bound to say that a period of 5½ years is not such as necessarily to lead to a conclusion that the material is not held, and cannot be held, for the purposes of current journalism.

157. Nevertheless, there will come a point where the protection from disclosure ceases to apply and where it can be said that the material is now being held solely for historical or archival purposes. Without deciding the point, it might well be the case that material from more than a decade ago could no longer be said to have sufficient proximity between the subject- matter of the request and [NAME_16]’s current broadcasting output and journalistic activities. The passage of more than 10 years may lead to the conclusion that the material in question is now sufficiently relatively remotely linked to the current product of [NAME_16] that it will not (or no longer) attract the protection of the derogation, even if it has not been formally archived or separately deposited in a historical archive. On the other hand, there may be cases such as Newbery, where even the passage of 14 years between the preparation of the [NAME_29] and the FOIA request is not such as to lead to the conclusion that there was not a sufficiently proximate relationship between the [NAME_16]’s journalistic purpose that the derogation from FOIA applied to the requested material, see [48-51] of that decision.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 68 158. [NAME_13] for [NAME_16] accepted that the archive exception had a relatively limited period of time. He had no instructions to proffer a particular period of time within which the exception would operate, although he submitted that a period of 50 years might be difficult to justify. Without deciding the point, it seems to me that any period of that or similar length would be almost impossible to justify. After that length of time, and given that the tribunal must have regard to the directness of the purpose of holding the [NAME_3] and [NAME_16]’s journalistic activities, it would be fanciful to suggest, after that lapse of time, that an immediate object of holding the [NAME_3] was to use it for the purposes of journalism, art or literature.

159. In my judgment, the problem with [NAME_1]’s request in part 8 was that it was far too widely drawn to be effective for his purpose. Making a request for copies of all complaints to [NAME_16] about impartiality from 2015 to the relevant date in June 2020 and [NAME_16] responses to them without any differentiation as to date or any refinement as to subject matter was to use a blunderbuss rather than a rapier. A request, for example, in the summer of 2023 for copies of complaints in the first few months of 2015 about a particular programme might, depending on the facts, fall to be treated differently.

160. I therefore dismiss the appeal on Ground 2.

Conclusion 161. The Tribunal made its findings of fact and gave adequate reasons for reaching the conclusion which it did. I can see no error of law in the way in which it went about its task or in the decision which it reached or in the adequacy of the reasons which it gave for that decision. The function of the First-tier Tribunal is to assess whether the [NAME_2]’s decision notice “against which the appeal is brought is not in accordance with the law” (s.58 of FOIA). That the First-tier Tribunal has done. I can detect no error of law in its decision.

[NAME_1] v (1) [NAME_2] (2) [NAME_4] [2023] UKUT 228 (AAC)

[NAME_1] v (1) [NAME_2] (2) [NAME_4] UA-2022-000648-GIA 69 162. For the reasons as set out above I am satisfied that the Tribunal was correct in the conclusions which it reached and that the appeal should be dismissed.

Mark West Judge of the Upper Tribunal

Authorised for issue 6 September 2023