Upper Tribunal Upholds Ruling on Vexatious Information Request

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

1

IN THE UPPER TRIBUNAL Case No. GIA/1325/2016 ADMINISTRATIVE APPEALS CHAMBER

1. This is an appeal by [NAME_4], the information requester, against a decision of a First-tier Tribunal made on 22 March 2016. For the reasons set out below I dismiss the appeal. Introduction and background 2. [NAME_4] has been concerned since at least 2005 to obtain information from the Second Respondent (the Council) in relation to the 11 plus tests taken by children in the Council’s area. It appears that he has been concerned, in particular, with the extent to which performance in the tests can be improved by coaching specifically for that purpose, thus possibly detracting from the intention that the tests provide a comparison of pupils’ potential academic ability. There is an extensive and complex history of freedom of information requests by [NAME_4] on that subject, in some cases proceeding to complaints to the First Respondent (the [NAME_2]) and further. It will only be necessary for me to refer to a small part of that history in this decision.

3. In March 2010 the Council provided to [NAME_4], pursuant to an information request, 184 pages of spreadsheets, in PDF Form, showing, in respect of each of the years 2007 to 2009, a number of items of information in relation to the 11 plus results of each child taking the test in each school in the Bucks area. For example, the information included the name of the school which the child was attending, and the results of the two tests. As I understand it, at some stage thereafter the same information was provided in relation to 2010. My understanding is that, by way of example, two of those pages, showing the information in relation to about 40 children per page, are at pages 43-4 of [NAME_7] Bundle.

4. Although it is not directly material to this appeal, [NAME_4] contended that, even before the addition (with effect from 1 September 2013) of subsection 1A to s.11 of the Freedom of Information Act 2000, s.11 had entitled him to require the information to be provided in the form of an Excel file, rather than merely a PDF file, and he eventually succeeded in that contention in a decision of the Court of Appeal (Innes v Information Commisioner [2014] EWCA Civ 1086; [2015] 2 All ER 560) given on 31 July 2014. The information was then provided by the Council in that form.

5. As I have noted, with effect from 1 September 2013 subsection (1A) was added to s.11 of FOIA. That subsection provides that where an applicant expresses a preference for information which “is, or forms part of, a dataset” to be provided

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

2 in electronic form, the [NAME_5] authority must, “so far as reasonably practicable, provide the information in an electronic form which is capable of reuse.” 6. Prior to the Court of Appeal’s decision [NAME_4] had then on 19 May 2014 made a further freedom of information request, which is the one at issue in this appeal. It was in the following terms: “It has been indicated to me by the Information Commissioner that the information requested below would indeed fall in the scope of the new data set provisions. Therefore I repeat this request: Could you please provide me with the following information related to 11+ results 1) School 2) VRTS score 3) Attitude to Work 4) Academic Recommendation 5) 1st Test Score 6) 2nd Test Score 7) Both Test Dates 8) Plus, if tested by us other than at a school, the test venue and time for each test 9) Plus, if there has been an application for test modifications, there is a more detail just to record the application process and outcome 10) Plus “Order of Suitability” for the years that this was included.

I require this information for the secondary school entry 2005 to 2012. The information is required for each child at each school in the Bucks area. I require the information in a non-propriety standard such as csv. If your decision is the same as in Oct 2013, then please review that decision. In addition, I would like to make another request. For the years given above, I would like all the additional data (i.e. not part of the data requested above) held in the 2 Microsoft Access data base tables (or anywhere else) concerning any aspect of the 11+ testing process related to each child, again in csv form.” 7. So far as the years 2007 to 2010 are concerned, that request sought the same information as had previously been requested and provided, save that it sought the information in a “non-proprietary standard such as csv”, as distinct from the PDF format in which it had previously been provided. The first part of the

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

3 request also sought the same information in respect of the additional years 2005, 2006, 2011 and 2012. The second part (i.e. the last paragraph) of the request sought substantial additional information in respect of all years.

8. On 6 June 2014 the Council refused to comply with the request on the ground that it was vexatious within s.14(1) of FOIA. On 2 October 2014 the Council maintained that position on review. 9. [NAME_4] applied to the [NAME_2] for a determination as to whether his request had been dealt with in accordance with the requirements of FOIA. By a Decision Notice (the [NAME_6]) dated 29 July 2015 the [NAME_2] found that the Council had been correct to apply s.14(1) to the request. It will be necessary for me to refer later in this decision to some of the [NAME_2]’s reasoning.

10. By the decision now under appeal to me, made on 22 March 2016, [NAME_7] dismissed [NAME_4]’s appeal against the [NAME_2]’s decision. The Council was joined as an additional party to that appeal. 11. [NAME_7] had the benefit of detailed written statements of their case from the three parties, totalling some 37 pages. It held a hearing on 14 January 2016 at which [NAME_4] appeared in person and the Council was represented by counsel. The [NAME_2] was not represented. On 15 January the chairman of [NAME_7] issued directions requiring [NAME_4] and counsel to provide "comprehensive written closing submissions”. Counsel for the Council provided written closing submissions running to 12 pages, and [NAME_4] provided submissions running to some 14 pages. The result was that [NAME_7] had before it written submissions from the parties totalling some 63 pages, in addition to the oral submissions made at the hearing.

12. I have been provided with the bundle of documents (pages 1 to 202) which was initially prepared for use before [NAME_7] (“[NAME_7] Bundle”). In addition, a number of additional documents which were before [NAME_7] but had not been included in [NAME_7] Bundle have been provided to me in a separate Bundle (“[NAME_7] Additional Bundle”). In the numbering which I propose to use for the purposes of this decision [NAME_7] Additional Bundle is numbered from pages 65 to 205.

13. On 12 February 2016 [NAME_7] members met in order to consider their decision, in the light of all the material then before them. By its reasoned decision issued on 22 March 2016 [NAME_7] dismissed the appeal. 14. [NAME_4]’ grounds of appeal against [NAME_7]’s decision are for the most part contentions that [NAME_7] went wrong in law in failing to provide sufficient reasons for its findings of fact and conclusions. He himself summarised the thrust of his grounds in para. 4:

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

4 “ For the large part, the [FTT’s] Decision contained no reasons for the decisions reached. The Decision’s conclusions, running from paragraph 32 to 40 (9 paragraphs for the 7 grounds of appeal), contained the following phrase ….(or similar)

“The Appellant has failed to persuade us …..” ….on no less than 8 separate occasions but without any substantive reasoning as to why the appellant had failed to persuade the Tribunal, as if the fact that the appellant had failed to persuade [NAME_7] were reason enough. This despite the fact that clear arguments and points of law were raised in the appellant’s submissions both written and oral. This deficiency in the Decision is compounded by the fact that the appealed [NAME_6] from the [[NAME_2]] was itself devoid of any substantive analysis or reasoning – in large part relying wholesale on the submissions of the Council without scrutiny.” 15. In the closing paragraph of his grounds of appeal to the [NAME_8] [NAME_4] sums up as follows: “Finally, I really fail to see, after all the time and expense that is involved in performing a FTT hearing, exactly what has been achieved. I really see no value, no benefit whatsoever in [NAME_7] Decision. There is no analysis, no breakdown of the decision making logic, no reasoning. It simply confirms findings of the [[NAME_2]] without any documented assessment of the appellant’s submissions. Even where the Decision does not agree with the [NAME_6], the [NAME_6] is left as it is. This process really has been a complete waste of time and money.” 16. On reading [NAME_4]’ grounds of appeal and [NAME_7]’s decision I formed the view that, on the face of those documents (and without more) there might well be substance in [NAME_4]’ contentions. I therefore gave permission to appeal, so that the contentions could be fully considered.

17. I have received detailed written submissions in the appeal from all three parties, totalling (including the grounds of appeal) some 45 pages. [NAME_4] requests an oral hearing of the appeal, but the other two parties submit that it can be satisfactorily decided without an oral hearing. Appeal to the Upper Tribunal lies solely on a point of law. In the light of the nature of the issues, and the detailed written submissions before me, I have concluded that I can properly dispose of the appeal without a hearing, and I therefore refuse [NAME_4]’ request. In so doing I have taken into account that [NAME_4]’ grounds of appeal and written submissions have throughout been, if I may respectfully say so, a model of lucidity. He has clearly considered very carefully what points are open to him. He has had the opportunity to reply in writing to the other parties’ submissions. I doubt whether there is much (if anything) more that he could usefully have said at a hearing.

18. Section 14(1) of FOIA provides as follows: “”Section 1(1) does not oblige a [NAME_5] authority to comply with a request for information if the request is vexatious.”

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

5 19. [NAME_7] was referred in paras. 19 to 21 of counsel for the [NAME_2]’s closing submissions to it to the observations concerning s.14 in the decisions of the [NAME_8] and the Court of Appeal in the Dransfield case: [2012] UKUT 440 (AAC); [2015] EWCA Civ 454. I would myself highlight, in the particular context of this appeal, the following passage from the judgment of Arden LJ in the Court of Appeal: “68. In my judgment, the [NAME_8] was right not to attempt to provide any comprehensive or exhaustive definition [of ‘vexatious’]. It would be better to allow the meaning of the phrase to be winnowed out in cases that arise. However, for my own part, in the context of FOIA, I consider that the emphasis should be on an objective standard and that the starting point is that vexatiousness primarily involves making a request which has no reasonable foundation, that is, no reasonable foundation for thinking that the information sought would be of value to the requester, or to the [NAME_5] or any section of the [NAME_5]. Parliament has chosen a strong word which therefore means that the hurdle of satisfying it is a high one, and that is consistent with the constitutional nature of the right. The decision maker should consider all the relevant circumstances in order to reach a balanced conclusion as to whether a request is vexatious. It if happens that a relevant motive can be discerned with a sufficient degree of assurance, it may be evidence from which vexatiousness can be inferred. If a requester pursues his rights against an authority out of vengeance for some other decision of its, it may be said that his actions were improperly motivated but it may also be that his request was without any reasonable foundation. But this could not be said, however vengeful the requester, if the request was aimed at the disclosure of important information which ought to be made publicly available. …..” 20. In CP v IC [2016] UKUT 427 (AAC) at para. 45 Judge Knowles considered that Arden LJ was not intending to say that if there is significant value in the information being sought the request cannot be found to be vexatious. Arden LJ went on to say in para. 68 that “the decision maker should consider all the relevant circumstances in order to reach a balanced conclusion as to whether a request is vexatious”. If, for example, complying with a request would impose a substantial burden, a request may therefore be vexatious even if the information has some value to the [NAME_5].

21. In the Upper Tribunal in Dransfield Judge Wikeley (at pras. 17 and 27) referred with approval to the statement of the Upper Tribunal in Wise v [NAME_2] (GIA/1871/2011) that: “Inherent in the policy behind section 14(1) is the idea of proportionality. There must be an appropriate relationship between such matters as the information sought, the purpose of the request and the time and other resources that would be needed to provide it.” 22. [NAME_7] in effect substantially upheld the reasoning in the [NAME_2]’s [NAME_6] (although see paras. 64-5 below). [NAME_7] rejected a contention by [NAME_4] that his letter of request dated 19 May 2014 should be analysed as making two requests, which in relation to questions such as the value of the information and the burden imposed by the request should therefore be analysed separately.

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

6 [NAME_7] considered that the letter should be seen as making one request, with the particular consequence that the time and expense involved in complying with the second part of the request should be considered material in determining whether the request as a whole (including the first part) was vexatious.

23. It is in my judgment clear from the conclusions set out in paras. 43 to 50 of the [NAME_2]’s [NAME_6] that in the [NAME_2]’s view (which was upheld by [NAME_7]) the most important factors in determining the issue of vexatiousness in this case were (a) the potential value to the [NAME_5] of the information sought and (b) the extent of the burden which would be involved in complying with (in particular) the second part of the request. That is consistent with the notion of proportionality which is referred to in the authorities. It was further in accordance with [NAME_4]’ own approach in his written closing submissions to [NAME_7], where he said: “3. The focus of these submissions is on the value of the requested information and on the burden on the council in providing this information. I believe that that is where the real issue in this case is.” It is in my judgment therefore appropriate for me to focus particularly on those of [NAME_4]’ grounds which are founded on the adequacy of [NAME_7]’s reasoning in relation to those two issues.

The grounds of appeal: analysis and conclusions Ground 6 24. It is convenient first to consider [NAME_4]’ ground of appeal no. 6, which is that [NAME_7] went wrong in law in treating his letter of 19 May 2014 as making one request, rather than two separate requests. He says (para. 50 of his grounds of appeal): “There is no justification given for treating a set [of] individual requests as a single request. This interpretation is the most restrictive interpretation of the legislation and appears unjustified when considering that the legislation refers to individual requests and makes no provision for combining individual requests into a single request. This could easily lead to the case where a series of requests in a single communication are all deemed vexatious simply because only one of the requests was actually vexatious. This unnecessary and unjustified restrictive approach is not consistent with the constitutional nature of the right.” 25. The [NAME_2]’s [NAME_6] did not, I think, expressly consider the possibility that the letter of 19 May should be analysed for s.14 purposes as making two separate requests. The reasoning referred simply to “the request”, and assumed that it should be treated as one compendious request. [NAME_7]’s only express reasoning on this issue was in para. 32:

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

7 “Although the Appellant’s request is drafted in two parts, for the purpose of this appeal we regard it as one request dated 19 May 2014. In fact the Appellant has also taken this approach except where he seeks to divide the two in the expectation or recognition that one part might be vexatious while the other part not so. For the avoidance of doubt we, taking into account the history and circumstances of exchanges between the Appellant and the Council over a considerable period of time, are of the view that the Commissioner was correct in finding it to be vexatious. In any event, we find both parts of the request are vexatious in all the circumstances. The context and history are relevant as the Commissioner has argued and we look at the request in that regard.” 26. It is right also to point out that in paras. 37 and 38 of [NAME_7]’s decision, which are part of its “Conclusions”, it appears to some extent to deal separately with “the first part of the request” and “the second part of the request”. I do not regard that as detracting from its primary conclusion, expressed in para. 32, that for the purpose of a s.14 analysis it should be regarded as one request. Further, I do not regard para. 37, read in the context of [NAME_7]’s decision as a whole, as intending to state that the first part of the request was vexatious merely because it partly sought information which had been requested previously.

27. I have been referred by counsel for the [NAME_2] in this context to the decision of Judge Jacobs in McInerney v IC [2015] AACR 32; [2015] UKUT 47 (AAC), and particularly to para. 55 of that decision, where he said: “….. The form in which a request is presented should not dictate how the section is applied. A series of requests could each be considered vexatious when viewed in the context of the series as a whole. Likewise, when presented with what on its face is a single request, the [NAME_5] authority should not be obliged to dissect it to see whether it could be severed. The [NAME_5] authority, and the First-tier tribunal on appeal, should take an overall view of the circumstances as a whole to decide whether what is before it, whether presented as a series of requests or a single request, is vexatious.” 28. However, one must also have regard to the whole of what Judge Jacobs said in relation to what he described as “the severance issue”. In my judgment what he said is so dependent on the particular facts of that case, that I do not derive great assistance from it in relation to this case.

29. In my judgment in the present case [NAME_7] was entitled to treat this as one request for s.14 purposes. Although the last paragraph of the letter of 19 May 2014 said that “I would like to make another request”, which as a matter of form might be argued to support a contention that there were separate requests, the last paragraph sought information from the same databases in respect of the same years. As counsel for the [NAME_2] submits, it was in essence an expansion of the first part of the request, to capture related data. Further, I doubt whether the Council, knowing that the second part of the request had been made, could (in

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

8 respect of the additional years) properly have complied with the first part of the request without considering the possible data protection implications of complying with the second part of the request: compliance with the second part might have required some details requested in the first part to be redacted. The two parts were as a matter of common sense so interlinked that [NAME_7] was entitled to treat this as one request. Further, although [NAME_7]’s reasoning was very brief, and really did not more than state a conclusion, it was in my judgment adequate in the circumstances. Ground 4 30. I turn next to Ground 4, which counsel for the [NAME_2] describes as “the only ground of broad importance.” 31. S.11(1A) of FOIA provides as follows: “Where (a) an applicant for information makes a request for information to a [NAME_5] authority in respect of information that is, or forms part of, a dataset held by the [NAME_5] authority, and (b) on making the request for information, the applicant expresses a preference for communication by means of the provision to the applicant of a copy of the information in electronic form the [NAME_5] authority must, so far as reasonably practicable, provide the information to the applicant in an electronic form which is capable of reuse.” 32. [NAME_4]’ contention is that where information forming part of a dataset was provided before the date (1 September 2013) when s.11(1A) came into effect, a further request after that date for the same information to be provided in an electronic form that is capable of reuse (i) cannot be vexatious under s.14(1) (or at any rate cannot be so merely because the same information has previously been provided in a different form) and (ii) cannot be regarded as a “subsequent or substantially similar request” for the purposes of s.14(2).

33. I do not consider it necessary to consider section 14(2), because as far as I have been able to discern neither the [NAME_2] nor [NAME_7] in any way founded their decision on that provision. I take that view notwithstanding the references to a “repeated request” in paras. 26 and 33 of [NAME_7]’s decision. Their decisions were that the request was vexatious within s.14(1).

34. In my judgment the [NAME_2] and [NAME_7] were clearly correct to hold that the whole of s.11, which is concerned with the means by which information is to be communicated, is subject to s.14 in the sense that if a request is found to be vexatious under s.14 then there will be no obligation to disclose it under s.1(1) and therefore section 11 will simply not come into play. As counsel for the [NAME_2] submits, s.11(1A) does not take the request outside the scheme of FOIA.

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

9 35. If, therefore, the burden on a [NAME_5] authority in complying with a request for information is excessive and disproportionate, the request could be held vexatious under s.14 notwithstanding that the information sought was part of a dataset and notwithstanding that the information was sought in electronic form, so that (subject to section 14) s.11(1A) would be applicable.

36. In the present case, therefore, if the burden on the Council of complying with the request as a whole is such that, in all the circumstances, the request is properly considered vexatious, that conclusion is not negated merely because, if there were an obligation to provide the information (i.e. if the request as a whole were not vexatious), s.11(1A) would be applicable.

37. I think that the point which [NAME_4] is really seeking to make is that it is not permissible to take into account, in determining whether his request was vexatious, the fact that he was previously provided with the some of the same information in a different form, because s.11(1A) entitles him to have it in a form capable of reuse, as he requested.

38. However, even if his request had been limited to the same information as he had already been provided with for the years 2007 to 2010, and was merely for that information in a format which s.11(1A) now requires, it would still as a matter of law in my view have been possible for the request to be held vexatious, on the ground that as a matter of the structure of the legislation one does not get to s.11 unless there is an obligation to provide the information under s.1. It would as a matter of law be permissible to find that on the particular facts it was vexatious to request information which had already been provided, albeit not in the format now required by s.11(1A). In the present case [NAME_4] accepts that he could himself convert to csv format the PDF or Excel data which he has been provided with (although it may not necessarily follow that the same assumption should be made in relation to members of the [NAME_5] generally). Ground 1 39. I turn then to Ground 1, which is that [NAME_7] did not come close to providing sufficient reasons for its finding in relation to the purpose and value of the information requested. The directly relevant paragraphs in [NAME_7]’s decision are paras. 8, 14 to 16, and 38. Para. 38, in [NAME_7]’s conclusions, is as follows: “The approach to the Commissioner’s assessment of the second part of the request is essentially one of the serious purpose of the request as evidenced by the weight to be given according to the value of the information. The Commissioner did acknowledge that the requested information had some value but that this value was weakened by the changes to the system, which have rendered direct comparisons meaningless. The Council confirmed that the changes to the system would render the requested information of little or no[w] value. The Appellant has failed to persuade us that the Commissioner erred in placing little by way of significant value

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

10 to the requested information. We do not accept the Appellant’s contention that the requested information has a high absolute value per se.” 40. In considering the adequacy of [NAME_7]’s reasons in relation to the value of the information it is of course necessary to have regard to (i) the [NAME_2]’s findings on this point, to which [NAME_7] referred and which it considered itself to be upholding and (ii) the contentions which were made to [NAME_7].

41. The most relevant paragraphs of the [NAME_2]’s [NAME_6] are paras. 36 to 38, and 44 to 50.

42. In para. 36 the [NAME_2] summarised the Council’s contention as being that the request had no serious purpose or value in that (i) the nature of the 11+ test had changed and “this would render any direct comparisons meaningless” and (ii) the grammar schools which utilise the present 11+ tests are now academies free from local authority control. “[The Council] therefore considers that any conclusions the complainants might reach from any data disclosed would be of no great value now, and the council could take no specific action, in relation to the 11+ test anyway.” 43. However, in paras. 37 and 38 the [NAME_2] said: “37. The Commissioner considers that there is a value in older data being disclosed to interested parties. Although it could have no direct effect in the current system an analysis of previous results could feasibly allow interested parties to highlight flaws or areas of concern which the new system might not have addressed. It might also allow an analysis of how the newer system compares with the older one if similar information could be obtained from the academies on their results.

38. Nevertheless it is fair to say that the changes which have occurred will weaken the argument for the data to be disclosed. The tests administrators have changed, in some cases the tests themselves have changed in order to try to reduce the ability to coach children, and the responsibility for admissions has moved out of local authority control to the schools themselves. The value of the information being disclosed has therefore weakened substantially.” 44. In setting out its conclusions in paras. 44 to 50 of the [NAME_6] the [NAME_2] did not really add to its stated conclusion, in para. 38, that the value of the information had “weakened substantially”, for the reasons there given.

45. I have taken [NAME_4]’ contentions to [NAME_7] on this point to be set out in paras. 9 to 33 of his closing written submissions (pages 183 to 187 of [NAME_7] Additional Bundle). He contended that “the grammar school selective system and the tests which support the selection process have and continue to be a topic of enormous and relevant [NAME_5] interest.” I think that the [NAME_2] and [NAME_7] accepted that. (Although not material to my decision, I observe that recent policy proposals in this field will, if anything, have heightened this general [NAME_5] interest).

46. In Annex 1 to his grounds of appeal to [NAME_7] he had set out some of the conclusions which he had drawn from information already available, and in

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

11 particular from the differences in performance between children at well performing and poorly performing schools. 47. [NAME_4] repeatedly asserted that the data which he had asked for would continue to be useful in assessing how the new (CEM) version of the selection tests is working: “The new method of 11+ testing (now called the Transfer Test) developed by CEM (Centre for Evaluation and Monitoring) of Durham University which purports to be coach proof has been introduced in Bucks and other selective areas. The CEM have declared their product is coach proof but have not been obliged to provide any evidence that this is the case. The children now taking the 11+ (Transfer Test) are guinea pigs of a new system where there is much to be learned from comparing childrens’ performance with the old (11+), and the new (CEM) version of the selection tests. …………………………………………. It is very misleading of the Council to suggest that results from the old 11+ test cannot meaningfully be compared with the new by suggesting that too much has changed about the system for grammar entry …… …………………………………………….. The Council view of ‘historic 11+ data’ seems to typify their wish to keep their head in the sand and ignore any evidence which points to some failure in the 11+ system. In this case whether the CEM transfer tests really have made things better. The issues with CEM testing is just one area where the historic data can be used to investigate matters.” 48. In my view the nature of the evidence and of the contentions before [NAME_7] on this issue did not permit any detailed and closely reasoned assessment of the value which might be derived from the requested information, having regard in particular to the changes in the nature of the testing, and to the lapse of time. (I note that the 2005 and 2006 data requested were some 8 to 9 years old by the time of the 2014 request). In particular, [NAME_7] did not have details of the precise changes in the nature of the test, and was substantially in the dark about the nature of much of the information which would be revealed by the second part of the request. As to this latter point, [NAME_4] himself asserted, as far as I can see with justification, that he did not know what heads of information were or might be contained in many of the (as alleged by the Council) 240 fields comprised in the databases. [NAME_7] had no more evidence as to that than he had. In addition, it was surely not possible to predict to precisely what uses researchers and others might want to put the information. It was in my view not possible for [NAME_7] to make more than a very broad and substantially uninformed assessment of the extent to which detailed information about the children and their performance in tests taken in past years would be useful in the future, having regard to the lapse of time and the change in the nature of the test and possibly changes in many

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

12 other variables. [NAME_7] had to do its best, looking at the matter broadly, on the (in my view necessarily) limited evidence available to it. The weight of the [NAME_5] interest in favour of information being disclosed is of course something which FTTs are routinely required to consider, more usually in the context of the [NAME_5] interest balancing exercise. 49. [NAME_7]’s reasoning in para. 38 was certainly very thin. The only two sentences which actually stated a conclusion or reasons were the last two sentences. However, when para. 38 is read as a whole, against the background of paras. 14 to 16, it is in my judgment sufficiently clear that [NAME_7]’s intention was to agree with the [NAME_2]’s finding on this point, and for the same reasons. I cannot agree with [NAME_4]’ general contention (para. 4 of his grounds of appeal) that the [NAME_2]’s [NAME_6] was itself “devoid of any substantive analysis or reasoning – in large part relying on the submissions of the Council without scrutiny.” 50. I have had some concern about the statement in the penultimate sentence of para. 38 that the [NAME_2] “plac[ed] little by way of significant value to the requested information”. It is well arguable that that statement, viewed alone, understates the value as found by the [NAME_2], which was that there was “a value”, albeit a much weakened one. But in para. 15 [NAME_7] expressly acknowledged that the [NAME_2] had found that the requested information had “some value”. It was of course open to [NAME_7] to disagree with the [NAME_2], but in that case it would have had to make clear that it did so, and do its best to explain why. In my judgment [NAME_7]’s intention was to agree with the [NAME_2]’s assessment, and for the same reasons.

51. In my judgment the [NAME_2]’s reasons on this point, in effect adopted by [NAME_7], were adequate, having regard to what was realistic, given the constraints to which I have referred.

Ground 7 52. I turn to Ground 7. The heading to this ground adopted by [NAME_4] is: “Request 2 not as burdensome as the Council would claim.” Summarised in that way this ground appears to be simply an attempt to challenge [NAME_7]’s finding of fact on the issue of the burden which would be involved in complying with the request. Indeed, counsel for the [NAME_2] submits that “the notice of appeal’s treatment of these issues is an analysis of the competing submissions before [NAME_7], with an assertion that the Appellant is correct.” However, in my judgment Ground 7, as developed by [NAME_4], plainly goes beyond that and includes contentions that (i) [NAME_7] did not give sufficient reasons for its decision and (ii) it did not have sufficient evidence before it to come to a properly reasoned finding. In setting out his argument on those points, [NAME_4] necessarily has had to explain what his contentions to the ICC were and why he contends that they were not adequately dealt with in [NAME_7]’s decision.

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

13 53. The directly material paragraphs of [NAME_7]’s decision are paras. 28 to 30, and 39, which I find it necessary to set out:

28. The Council asserted that the breadth of information fields that have been requested would create unique profiles allowing the identification of children whether by a ‘motivated intruder’ or by mosaic effect. The Appellant confirmed that any fields and notes, which could potentially identify a child, could be removed from all records without the need to analyse each record. The Appellant based his determination of burden on the “notes” in the data, but the Council and Commissioner considered the matter in terms of “information.” 29. However, the Appellant’s argument was that the Commissioner has misunderstood how information is stored and extracted in the databases, as any annotations of a field would be added to the specific field for the specific child, not stored in the same field. In this way he distinguished Department for [NAME_9] v [NAME_10] and McInerny EA/2013/0270 as the large amounts of personal data in that instance were contained in unstructured and unpredictable ways on pages, requiring each page to be read in order to redact the information. Contrast this with a database, where in only one or two records need be analysed in order to isolate the potential fields in order to exclude them when a report on the whole data set is generated. The Appellant conceptualised the question of burden as assessing which fields should not be released on the grounds of potential identification, rendering the number of children involved irrelevant.

30. The Commissioner informed the Tribunal that he has had sight of screenshots from the database leading him to conclude that identifying information would not be as simple as the Appellant suggests. Whilst the Commissioner accepted the distinctions the Appellant has drawn, he also accepted that the Council would have to examine a large number of entries to determine whether individual children could be identified. The Council also maintained its position that it would need to review information on a child-by-child basis, which is an obviously oppressive task.

39. The Commissioner has accepted that the Appellant did not intend to cause annoyance and disruption but has stated that on the whole the disruption would be disproportionate in all the circumstances. The Tribunal have looked at the spread sheets and screen shots provided by the Council and considered the need to search the records of 58,000 children with up to 240 data fields for each child and the need for redaction throughout. We accept the Council’s concern about the burden that would be imposed on them to disclose the requested information. The Appellant has, again, failed to persuade us that the Commissioner erred in finding the burden of disclosure of the requested information in the second part of the Appellant’s request exceeded any value that might be provided therefor.” 54. In very broad summary, the issue was as to the extent of the work which it would be necessary for the Council to do (if necessary by redaction) in order to ensure that disclosure of the information did not enable identification of the particular child to which part or parts of the data related. As was common ground, the cost of this work was not a qualifying cost for the purpose of the costs limit in s.12 of FOIA, because it was a cost involved in determining whether an exemption

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

14 (namely that in s.40(2) of FOIA – personal information) applied, but it was further common ground that it could be taken into account for the purpose of determining whether the request was vexatious. The Council did not claim that any great cost would be involved in relation to the new information sought in the first part of the request (i.e. for the years 2005, 2006, 2011 and 2012), but contended that the potential problem was with the information sought in the second part of the request (in conjunction with that already disclosed, and sought, in the first part).

55. As [NAME_7] in my view considered itself to be upholding the [NAME_2]’s findings in the [NAME_6] on this issue, it is again important, in considering the adequacy of [NAME_7]’s reasoning, to have regard to (i) the [NAME_2]’s findings and (ii) the contentions which were before [NAME_7].

56. The [NAME_2] dealt with the parties’ contentions in more detail than [NAME_7] did. The most relevant paragraphs of the [NAME_6] are in my view paras. 23 to 31.

57. The essence of the matter is that [NAME_4] contended that the risk of identification of particular children could be minimised with substantially less work than the Council contended. The problem concerned the (as the Council asserted) up to 240 ‘fields’ of information sought, in relation to some 58,000 children, by the second part of the request.

58. The [NAME_2] and [NAME_7] had before them the ‘screenshots’ from the two databases of which copies were at pages 95 to 112 of [NAME_7] Bundle, and of which [NAME_7] had the benefit of the more legible copies now at pages 119 to 137 of [NAME_7] Additional Bundle. As regards the submissions to [NAME_7], I note that, in addition to the contentions in the grounds of appeal and written submissions, [NAME_4] and the Council had made a number of contentions on this issue in the email exchanges in January 2016 which are now at pages 104 to 115 of [NAME_7] Additional Bundle.

59. In essence, the issues turned on whether [NAME_4] was likely to be correct in asserting that it would not be necessary, as contended by the Council, to examine the individual fields in relation to each child, because (i) fields likely to contain identifying information could simply be excluded in relation to each child and (ii) a process of what he described as ‘automation’ could be adopted in order, in the case for example of a school with a small number of relevant children, to prevent identifying information being disclosed; a formula or process could be devised and utilised in each such case.

60. The most detailed rebuttal of [NAME_4]’ contentions seems to have been in paras. 37 to 40 of counsel for the Council’s closing submissions to [NAME_7] (p.179 of [NAME_7] Additional Bundle).

61. My impression is that it was difficult for [NAME_7], on the information before it, to form a view as to even approximately how much work was likely to be involved, and in particular as to the extent to which it would be possible to avoid examining

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

15 the data for individual children, a task which would be likely to be excessive. [NAME_7] did not know the nature of the information in (if there were indeed 240 of them) many of the fields, or how it was structured. But it was told by the Council (para. 38 of the Council’s closing submissions) that “it is possible to add notes or add information [to the fields] and each would need to be checked to see if information is actually held which relates to a child and could be used to identify the child including through other information in the [NAME_5] domain.” 62. It seems to me, as [NAME_4] has in effect suggested, that it might have been possible for the Council to evidence its case more convincingly by taking an example of a particular pupil, possibly one who was at a school with a small number of relevant pupils, and showing exactly, by reference to at least some of the fields, what it would be necessary to do in order successfully to anonymise the data relating to that pupil, or even to check whether anything needed to be done by way of redaction. It is possible, of course, that even an attempt to provide such an example might itself have involved a substantial amount of work, which was what the Council was concerned to avoid by invoking s.14. But nevertheless, my impression is that an example might have been helpful and much more convincing. But, in order to avoid disclosing exempt information to [NAME_4] himself, it might have been necessary either to present the information to [NAME_7] in ‘closed’ form, or to do so in already redacted form, in either of which cases [NAME_4] would have been hampered in responding to it.

63. At the end of the day I have come to the conclusion that [NAME_7] did not go wrong in law in not requiring the Council to attempt to evidence its case more convincingly. It was entitled to decide the case on the basis of the evidence which it had. On that evidence, it was entitled to accept that the work would be very substantial. Para. 39 of [NAME_7]’s decision is, again, very thin in terms of its reasoning, as [NAME_4] rightly emphasises. But it seems to me to evince an intention to accept the findings of the [NAME_2], and for the same reasons. [NAME_7] had referred to the parties’ contentions in paras. 28 to 30. It was not really possible for [NAME_7], on the evidence before it, to do more than reach a broad conclusion as to whether [NAME_4]’ suggested methods of avoiding the need to look at multiple individual records and fields were likely to be effective. Against that background, I think that its reasoning, brief as it was, was just about adequate.

Ground 2 64. I turn to Ground 2. [NAME_4] contends that, in stating in para. 39 that “the Commissioner has accepted that the Appellant did not intend to cause annoyance and disruption …”, [NAME_7] overlooked or misunderstood what the [NAME_2] said in paras. 40 to 42 of the [NAME_6]. I think that the third sentence of para. 17 of [NAME_7]’s decision, coupled with para. 39, indicates that [NAME_7] did in this respect misunderstand what the [NAME_2] said in the [NAME_6]. The reference in para. 17 to the [NAME_2]

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

16 having “accepted that there is weight to the Appellant’s denial that he had no intention to cause annoyance and disruption” seems to be a misunderstanding of the [NAME_2]’s statement in para. 42 of the [NAME_6] that the [NAME_2] gave “some weight to this argument”, which was the Council’s argument that the request must have been partially designed to cause disruption or annoyance.

65. However, I agree with counsel for the [NAME_2]’s submission that that does not demonstrate an error of law in [NAME_7]’s decision. First, [NAME_7] made clear that it did not itself find that there was an intention to cause disruption or annoyance. The finding against [NAME_4] by the [NAME_2] therefore did not “live on” in the [NAME_2]’s [NAME_6]. Secondly, and in any event, it is in my judgment sufficiently clear that this factor was of relatively minor significance in the [NAME_2]’s and [NAME_7]’s reasoning.

Ground 3 66. I turn to Ground 3. [NAME_4] contends that [NAME_7] failed to give sufficient reasons for its decision in that in paras. 32, 33 and 37 it said that the context and history were relevant, but without stating why and what it made of the competing contentions of the parties. In essence, [NAME_4] contended to the [NAME_2] and before [NAME_7] that the Council had asserted, and the [NAME_2] had apparently accepted, that [NAME_4] had made many more requests, and with substantially less justification, than was in fact the case. He contended that much of the correspondence and of his applications to the [NAME_2] and appeals had been caused by the Council’s inadequate responses to his requests.

67. A proper examination of that contention would have required a detailed review of the somewhat complex history. [NAME_7]’s reasoning does not indicate that it undertook such an exercise.

68. However, I have again come to the conclusion that this did not render its decision wrong in law. First, there were certain aspects of the context and history which could not be disputed. These included that [NAME_4] had previously sought and been given the information in the first part of the request for the years 2007- 10, albeit not in csv form. The first part of the request therefore included a request for information for years (2005 and 2006) which were more distant in time than those for which the equivalent information had already been requested and provided; see, in particular, the reasoning in para. 33 of [NAME_7]’s decision. Secondly, as I have already said, in my judgment [NAME_7] in substance agreed with the [NAME_2]’s approach in paras. 43 to 50 (“Conclusions”) of the [NAME_6] that the most important factors for consideration in this case were the value of the information and the burden which would be involved in providing it. Despite the references in [NAME_7]’s decision to the context and history, that factor was in my judgment considered by both the [NAME_2] and [NAME_7] to be of relatively little importance. An error (even of law) made in the course of reaching a decision does not necessarily mean that the decision itself is wrong in law.

[NAME_1] v [NAME_3] County Council [2016] UKUT 0520 (AAC)

17

Burden of proof 69. [NAME_4] contends that, in saying, on a number of occasions, that “the Appellant has failed to persuade us that the Commissioner erred …..”) [NAME_7] went wrong by in effect placing on him the burden of establishing that the request was not vexatious. I have some sympathy with this objection to the way in which [NAME_7]’s decision is worded. It can make it appear that [NAME_7] has started from the position that the [NAME_2]’s findings are correct, and then proceeded to examine whether [NAME_4] was able to rebut them. But I do not agree with him that that is what [NAME_7] has in substance done. In my judgment [NAME_7] reconsidered afresh, as it was required to do, whether the [NAME_2] was right to decide that the request was vexatious. It did not start from the position that it was for [NAME_4] to establish that the request was not vexatious, or from the position that the [NAME_2]’s findings were correct unless [NAME_4] could demonstrate to the contrary.

Conclusion 70. For the reasons set out above, and contrary to the impression which I formed at the time of giving permission to appeal, none of the grounds of appeal in my judgment demonstrates an error of law in [NAME_7]’s decision.

Charles Turnbull Judge of the Upper Tribunal 7 November 2016