Sheila Margaret Robinson Cunliffe v Information Commissioner [2024] UKFTT 879 (GRC)

[2024] UKFTT 00879 (GRC)

Case References: EA/2023/0532 EA/2024/0014 FT/EA/2024/0103 FT/EA/2024/0092

First-tier Tribunal

(General Regulatory Chamber)

Hearing Heard on: 26 th September 2024

Date 17 October 2024

Information Rights

Heard by Cloud Video Platform

Before

JUDGE L MOAN

TRIBUNAL MEMBER J MURPHY

TRIBUNAL MEMBER P TAYLOR

Between

SHEILA MARGARET ROBINSON CUNLIFFEAppellant

INFORMATION COMMISSIONERRespondent

In person for Appellant

Did not attend for Respondent

Decision:

1.The Appeal is Allowed.

2.The Respondent shall provide a copy of this Decision to the four respective NHS Trusts (the third parties) who made the original decisions not to provide the requested information.

Substituted Decision Notice: The third parties shall within 35 days of being sent this Decision, respond to the Appellant's Part One requests for information as detailed below without reliance upon section 12 of the Freedom of Information Act 2000.

REASONS

Decision under appeal and background

[1]These appeals are brought under section 57 of the Freedom of Information Act 2000 (“FOIA”). The appeals are against the decision of the Information Commissioner (“the Commissioner”) contained in the following Decision Notices:(i) IC-261370-F4T1 dated 10th November 2023 (case number EA/2023/0532) following a review of the decision made by Kingston Hospital NHS Foundation Trust;

(ii) IC-261362-W9S9 dated 14th December 2023 (case number EA/2024/0014) following a review of the decision made by Croydon Health Services NHS Trust;

(iii) IC-263775-X4Y1 dated 23rd February 2024 (case number FT/EA/2024/103) following a review of the decision made by Epsom and St Helier University Hospitals NHS Trust; and

(iv) IC-261361-V6G6 dated 6th February 2024 (case number FT/EA/2024/0092) following a review of the decision made by St George’s University Hospitals NHS Trust.

[2]The Appellant had made a request of the respective four relevant public authorities to provide information about recruitment to assist with her research and reporting. The request is reproduced as an Annex to this judgment. Each of the public authorities had declined to provide that information due to the cost of extracting and providing that information. The appeals concern the same parties (albeit the public authorities differ in each case), concern the same issue and so have been listed to be heard together with the agreement of the parties.

[3]The Commissioner’s decision was that the respective NHS Trusts were entitled to rely on section 12 of FOIA to refuse to comply with the request for information. The Commissioner did not require the Trusts to take any steps as a result of his decision. Background to the appeals EA/2023/0532

Background to the appeals

[4]On 23rd June 2022 the Appellant wrote to Kingston Hospital NHS Foundation Trust and requested information about job applicants. The Trust responded on 9th September 2022 and refused the requested information at Part One under sections 12 (cost of providing) and 22 (information intended for future publication) of FOIA. The Trust provided information requested under Part Two. The Trust stated that the system could only produce 256 days of data.

[5]The Trust conducted an internal review. In that review, the Trust revised its position regarding its reasons for citing section 12 of FOIA because the system was, in fact, able to produce 400 days of data, as opposed to the previously stated 256 days rendering it unnecessary to “interrogate paper and information from other sources”. However, it still argued that section 12 applied for the following reason - “To produce the requested data for all completed campaign activity would require us to extract the data from TRAC and other sources and undertake a manual analysis to produce the report. This would take over 18 hours and therefore Section 12 is engaged.”

[6]The Appellant complained to the Commissioner on 28th September 2023. The Trust withdrew its reliance on section 22 but maintained that section 12 applied.

[7]The Trust stressed that the Appellant “has repeatedly insisted that the data can be pulled from the Trust’s TRAC [applicant tracking system]”. However, “senior members of staff at the Trust and expert users of the Trust’s system” dispute the accuracy of running reports from this system and consequently being able to provide the requested information.

[8]The Trust acknowledged that it had previously provided figures in response to an earlier request “based on simple running of reports that, on further examination, proved to be inaccurate”. The Trust found issues with the TRAC system when checking it against the ESR [electronic staff record] system. To accurately generate this data “would be an extensively manual process and take considerably longer than eighteen hours”. The report would not answer the questions asked in the request and could not provide the breakdown requested without the manual process. The Trust contended that it would be necessary to download a separate report for Band 2, Band 3 etc and then add them together for quarter one and two. It explained that for quarter three onwards a report would need to be run per question. Additionally, nurses and midwives cannot be reported on separately as they are in the same staff group in TRAC.

[9]The Trust gave the example of a report being run April to September where a job was advertised on 29th September. The job would be included in the ‘number of applicants’ but the outcome would not be included in the ‘shortlisted or ‘offered’ columns. The Trust says that this would look as if it did not employ as many BAME staff because it could not run reports on completed campaigns. An incomplete campaign meant that there was not the data for each stage of recruitment. It stated that it would take three days to download and manipulate into what would be an incomplete view. The Trust argued that running a report was one thing but that the reporting was not accurate for the reasons given. Additionally, TRAC only processed standard campaigns and did not cover all recruitment activity such as recruitment managed separately - international recruitment, recruitment events (appointments made on the same day), and agency managed campaigns. This meant that the TRAC data missed a lot of volume recruitment, and it gave the example of HCAs and nursing. Any campaign managed outside of TRAC would need Kingston to be disaggregated from the other Trusts it supported. To complete what it had detailed would take more than eighteen hours. The Trust ran around 500 campaigns a year with 12,500 applications and it estimated that it would take between fifteen minutes and two hours for each campaign, depending on the volume of candidates and the data source. This could mean anything from 125 hours to 1000 hours and it contended that it would be nearer to the higher figure. To produce accurate data would require the Trust to do the following:(i) Extract the data from TRAC and other sources, undertake a manual review and analysis and produce the requested data set.

(ii) There was no single report available; multiple reports would have to be utilised and then the data manually ‘stitched’ together.

(iii) The data would have to be scrutinised for each individual applicant at each of the six recruitment stages that constitute the life cycle of applicants.

[10]The Trust explained that it was able to “provide an ethnicity profile of their workforce and the ethnicity profile of new starters over a defined period from ESR, which was accurate”. However, it was unable to provide the requested information in its entirety without it exceeding the appropriate limit. It did not accept that the previous information it provided to the Appellant was accurate. Versions of TRAC varied from Trust to Trust as does the information entered into it and “the local system was operated by a specialist team”.

[11]The Appellant referred to a previous request that had been made that was similar but there were fewer profession/grade categories “to which the trust had responded in full”. The Appellant pointed out to the Commissioner that the data was produced from the same system. She also told the Commissioner that the Trust could not provide “recruitment data for some categories as it was not held for a full year”, despite having done so previously.

[12]Additionally, the Appellant said that she had not asked for ‘completed’ campaigns and that it was her view (from familiarity with the system) that “each report should take ten-twenty minutes to produce from start to finish”.

[13]The Commissioner’s decision was that the Trust had cited section 12(1) of FOIA appropriately. He also accepted that the Trust could not offer any meaningful advice and assistance within the context of the request.

[14]The Appellant appealed on 8th December 2023. She understood that the data requested was no longer available but wished to ensure the data for 2023-24 was made available. She said that –(a) Similar data was provided in 2021. Both the 2021 and 2022 requests were for the numbers of applicants, shortlisted applicants and offers (not starters) by each of Asian, Black, Mixed, Other, White and Unknown ethnicities in order to ascertain the likelihood of offer from application by ethnicity in a particular 12-month period. The request was not for data on all recruitment.

(b) Kingston NHS Trust used the same TRAC data to produce their Workforce Race equality Report for NHS England.

(d) It was not the case that TRAC reports need to be run on a band-by-band basis and totalled. Multiple bands can be selected for inclusion in a report.

(e) The standard TRAC Equalities Report can be run and analysed in ten-twenty minutes per category. The TRAC system and the standard reporting available was identical in each NHS trust although each trust may choose to use the system and set the report parameters slightly differently. The ICO appeared to have made no effort to independently establish whether what Kingston were saying about the system and the vast number of hours needed to fulfil the request was true (e.g. by contacting TRAC or speaking to another NHS Trust that uses this system).

(f) The Trust could have provided some professional/grade categories and not others if the data was genuinely not available but this was not offered. The ICO does not appear to have questioned or asked for evidence as to which of the requested categories and to what extent the data is claimed to be inaccurate or made any effort to objectively establish whether the vast number of hours Kingston claim was needed to fulfil the request was correct.

EA/2024/0014

[15]On 23rd June 2022 the Appellant wrote to the Croydon Health Services NHS Trust and requested information about job applicants.

[16]The Trust initially responded on 26th August 2022 refusing to provide the information under section 22 of FOIA. The Trust stated that it intended to publish the information within six months. The Appellant chased the Trust for a response as to when the information was to be published.

[17]On 28th September 2023 the Appellant contacted the Commissioner to complain about the way her request for information had been handled. The Trust provided some information to the Appellant on 19th October 2023 and links to the annual report.

[18]The Commissioner sent a further query to the Trust about some information missing from the request. The Trust sent a further response to the Appellant on 30th November 2023 which included its previous response. It provided some information but cited section 12 of FOIA (cost limit) as its reason for not being able to provide all the information - “To produce the requested data for all completed campaign activity would require us to extract the data from TRAC and other sources and undertake a manual analysis to produce the report. This would take over eighteen hours and therefore section 12 is engaged.”

[19]The Trust is part of “SWL [south-west London] campaigns” which means that “any campaign managed outside of TRAC” would require it to disaggregate from the other Trusts involved. “TRAC reporting functionality does not allow the Trust to scrutinise the data captured for reporting.”

[20]The Trust further went on to explain that TRAC is its applicant tracking system. “The data retention period is 400 days.” It describes the functionality of its reporting system as “very poor” and that it does not provide accurate data. In other words, data produced from TRAC could not be relied on. The Trust told the Appellant that - “TRAC will report on all open campaigns within the period requested – these will be both complete and incomplete campaigns (it does not differentiate). An incomplete campaign means that you will not have data for each stage of the recruitment episode.”

[21]In order to “produce accurate and reliable data” certain actions would be required that would bring the Trust over the fees limit (eighteen hours) as it would require the following -(i) Extract the data from TRAC and other sources, undertake a manual review and analysis and produce the requested data set.

(ii) There was no single report available; multiple reports would have to be utilised and then the data manually ‘stitched’ together.

(iii) The data would have to be scrutinised for each individual applicant at each of the six recruitment stages that constitute the life cycle of applicants.

(iv) This work would take between fifteen minutes and two hours for each campaign dependant on the volume of applicants and the data source.

(v) On average Croydon manages 5,933 campaigns although they are concurrent and so 532 at any one time with 19,632 applicants for 2021/22.

[22]The Appellant referred to a previous request in 2021 where the data had been produced by the Trust and that she considered that the Trust had “cut and pasted” the arguments put forward by another Trust.

[23]The Commissioner accepted that this may be the case but the fact that the Trust uses the same TRAC system means that it was likely to provide a similar response.

[24]Despite the tardy and unsatisfactory responses from the Trust to the Appellant, the Commissioner accepted that responding to the request would exceed the appropriate limit. The Commissioner’s decision was that the Trust had cited section 12(1) of FOIA appropriately. He also accepted that the Trust could not offer any meaningful advice and assistance, given the context of the request.

[25]The Appellant lodged her appeal with the Tribunal dated 11th January 2024. The grounds of appeal replicated those given in the earlier appeal form.

FT/EA/2024/0103

[26]On 23rd June 2022, the Appellant requested information from Epsom and St Helier University Hospitals NHS Trust.

[27]The Trust responded on 27th July 2022 as follows: information was provided for what it describes as “part two” (regarding ethnicity, this was actually in part one of the request) and “part three” (actually part two of the request). Section 12 was cited regarding part one (numbers in roles, labelled 1-14).

[28]Following requests for internal reviews, the Appellant complained to the Commissioner on 12th October 2023.

[29]The Trust responded to the Commissioner on 25th January 2024 citing section 12 FOIA. In its response to the Commissioner, the Trust confirmed that it had been correct in citing section 12 to part one of the request in its original response. The Trust concluded that the answer it had provided was “insufficient, failing to provide the appropriate justification for applying the section 12 exemption”. It explained that “the Trust should have set out how the cost of compliance with the request exceeded the appropriate limit…” It had “concerns over the robustness and accuracy” of the information it had provided.

[30]The Trust said that the concerns raised over TRAC were shared by the Trusts in SWL and the “expert users”:

“1. TRAC reporting will include all open campaigns during a requested period and will not differentiate between complete and incomplete campaigns. An incomplete campaign will not record data for each of the stages of a recruitment campaign. 2. TRAC reporting does not cover all recruitment activity. This included all international recruitment campaigns; recruitment events; and agency managed recruitment. As a result, TRAC data did not include a significant volume of recruitment activity particularly for the recruitment of HCAs and nursing staff. 3. Reporting Equality data from TRAC is significantly different when compared to the new starter data on the NHS Electronic Staff Form (ESR). The SWL Recruitment Hub is currently undertaking work to understand the variance in reporting.”

[31]The Trust said that TRAC did not allow the Trust to scrutinise this data. Producing the “requested data for all completed campaign activity would require extraction of data from TRAC and other sources” before undertaking manual analysis.

[32]The Trust had “approximately 700 recruitment campaigns occurring annually”. It was no longer possible to report on the timeframe specified by the request because data was only retained for 400 days. The Trust estimated that it would take between fifteen minutes and two hours to review each campaign. Based on the estimate, it would take between 175 hours and 1,400 hours to review the 700 campaigns, therefore vastly exceeding the fees limit.

[33]The Appellant provided some information from the equality data step-by-step guide reporting from TRAC that they themselves had provided to a different Trust to demonstrate that what they had requested could be reported on by the Trust in a much shorter timeframe. She said that an individual familiar with TRAC could run a report in twenty minutes. Even allowing extra time – the whole request would not take more than ten hours in total.

[34]The Appellant provided a number of arguments to support why she did not accept that section 12 applied to the request:(i) The request was not for the number of completed campaigns, it was for the number of applicants, shortlisted and offered in a 12-month period. Additionally, “the Trust has given no explanation as to why there should be a difference in likelihood of outcome by ethnicity between these two data sets”.

(ii) The Appellant did not require every single campaign, just requiring a “substantial sample” and did not accept that the amount of manual analysis claimed was necessary. She argued that there was an argument to exclude overseas recruitment campaigns run by an agency in a particular geographical area because they are likely to be disproportionately of one ethnicity.

(iii) “Even where a Trust added successful candidates from the campaigns to TRAC at the offer stage and onboards them, the total number was often able to be identified via the NHS ESR (HR) system where (e.g.) fully qualified nurses will initially be employed at Band 4 pending registration with the NMC and a move to Band 5; only overseas starters are onboarded in this way. In addition, this would only prevent the Trust from providing data for one professional group at one band and not for the whole data request.”

(iv) The Appellant needed information about the ethnicity of applicants and the likelihood of an offer according to ethnicity, not information about the ethnicity of staff.

(v) The Appellant questioned the Trust’s argument that the recruitment data was incorrect and that “it was not possible to draw this conclusion from a simple comment” that the two systems are different.

(vi) The Appellant had “great difficulty in accepting” the Trust’s view on accuracy as it had “used the TRAC system to undertake the required reporting and active planning for the year in question” and this was in the public domain. The Appellant attached Workforce Race Equality Standard data (WRES), WRES dashboard analysis and the Trust’s Workforce Race Equality Standard plan to underpin her argument that the Trust could not have reached conclusions from the same system if the data is unreliable. The Appellant queried how the Trust could comply with its Public Sector Equality Duty.

(vii) The Appellant considered that the second part of the request has nothing to do with the TRAC system and would “normally be held under a specific corporate or departmental budget heading”.

[35]The Commissioner’s decision was that the Trust had cited section 12(1) of FOIA appropriately.

[36]The Appellant’s appeal to the Tribunal was dated 21st March 2024. The concerns mirrored those in her earlier appeal forms and she added that the NHS England national WRES data used data reported from the TRAC system for the same year as the data she requested and which the Trust are saying they cannot provide due to data integrity concerns. She was concerned that the Trust had breached FOIA in terms of its response times and that it took so long for the Trust to respond to the Commissioner. She was not able within the timeframe allowed, to get information from the company that own TRAC to support her submissions.

FT/EA/2024/0092

[37]On 23rd June 2022 the Appellant requested information from St George’s University NHS Trust. The Trust responded on 19th August 2022 and refused to provide the information citing section 22 of FOIA 2000. The Trust stated that it intended to publish the information within six months. The Appellant repeated the request for information in September 2023 and complained to the Commissioner on 28th September 2023 about the way in which the information request had been handled.

[38]The Trust finally responded in December 2023 withdrawing its reliance on section 22 but citing section 12 of FOIA because it believed that it would exceed the cost limit to provide the requested information.

[39]The Appellant said that the Trust had decided to give the same response as other trusts in the SWL. The Appellant said that the Trust were using the same system to provide information for NHS England. The same data had been provided in 2021 albeit the Trust had later claimed that the data was inaccurate. The Appellant had indicated that groups could be separated off if there was concern about accuracy due to recruitment outside the TRAC system.

[40]The Trust had submitted that they would require between 300 and 2400 hours to provide accurate figures. The Appellant contended that the same data was provided to NHS England from each Trust and so there was no need for it to be disaggregated.

[41]The Trust had indicated that it had changed its response to the Appellant (from reliance on section 22 to section 12) following advice from the SWL Recruitment Hub as well as senior members of staff within the Trust’s People Directorate who had highlighted the inadequacies of the TRAC applicant tracking system for providing the information.

[42]The Trust highlighted that an incomplete campaign will not record data for each of the stages of a recruitment campaign. TRAC did not cover all recruitment activity. Equality data on TRAC was significantly different to the new starter data on the NHS Electronic Staff Form (ESR). There were legitimate concerns about the accuracy of data in reports run from TRAC and staff would need to manually intervene to correct that data. Data would need to be collated form outside TRAC and then amalgamated to the scrutinised data from TRAC. The difficulties would be experienced by all TRAC users and not just the Trust or SWL Trusts.

[43]The Commissioner decided that the Trust had cited section 12 appropriately as an exemption.

[44]The Appellant appealed against the Commissioner’s decision on a notice dated 13th March 2024. She submitted that –(i) The standard TRAC Equalities Report could be run and analysed in ten-twenty minutes per category. The TRAC system and the standard reporting available was identical in each NHS trust although each trust may choose to use the system and set the report parameters slightly differently. The ICO appears to have made no effort to independently establish whether what St George's are saying about the system and the vast number of hours needed to fulfil the request is true (e.g. by contacting TRAC or speaking to another NHS Trust that uses this system).

(ii) Her request was not for data on all recruitment or on completed campaigns as the nature of her research did not require this. She had asked for the number of applications, shortlisted candidates, and offers in a 12-month period by specific professional category. She said that the Trust could have provided some professional/grade categories and not others, but this was not offered. The Trust were using this same data for their own WRES reporting. It is quick and easy to identify the total number of internationally recruited nurses in a particular period and remove them in bulk from totals. The Appellant could advise on how to do this. The Trust was using the same data at a national level without any reference to this data being unreliable.

(iii) The TRAC Equalities Report which was part of the standard reporting suite on all TRAC systems allowed a user to run reports between date parameters set by the user, including a one-year period. The Trust appeared to be referring to a different TRAC report which was not suitable for the purpose of producing this data.

Respondent’s response to the appeal

[45]The Commissioner’s position has been that the Tribunal should dismiss the appeal for the reasons set out in their decision letters. The Commissioner has investigated each complaint, given extensive reasons in the decision letters and assessed the Appellant’s complaints objectively.

Procedural matters relating to the determination of the appeal

[46]The Tribunal considered the bundles (420 pages, 394 pages, 173 pages and 229 pages) and a 39 page document which contained WRES data. The Appellant sent an email the day before the appeal hearing from the creators of the TRAC system confirming that reports could be run more than once a year and an example of raw data from TRAC as relevant to the information request. The Appellant confirmed that the Tribunal had all the necessary information to hear the appeal.

[47]The hearing was attended by the Appellant. She gave evidence to the Tribunal and made submissions about her appeal. The hearing took place remotely via video (CVP). There were no objections to this as a suitable method of hearing. The Tribunal were satisfied that was a just and expedient way to conduct the appeal.

The Appellant’s oral evidence

[48]The Appellant was an experienced HR professional who had worked extensively for and with various NHS Trusts and had access to the TRAC reporting system in at least three different NHS Trusts. She confirmed that the TRAC system was the same across the Trusts although individual Trusts may set their own parameters. Over the years she had gained significant experience in using TRAC.

[49]She confirmed that she had used similar data received from eighteen Trusts in 2021 for research. She made similar requests in 2022. She received the second tranche of data from thirteen of the eighteen Trusts. In many cases, the Trusts took two-three months to produce the data but it had been produced. Another NHS Trust had not provided the data other than the four Trusts named in this appeal but she had missed the deadline to appeal that refusal. Some Trusts had provided the data as requested, some had produced raw data and she had manually undertaken the necessary calculations, as required.

[50]She said that she was mindful of resources when making the requests but to undertake the reports for the fourteen occupations listed in Part One of her request should have taken no more than thirty minutes which would amount to seven hours in total. She clarified that she had received answers to her Part Two questions from some Trusts and that was not part of her appeal. It was the Part One information that was important.

[51]The Appellant remarked that it was the four NHS Trusts in South West London that had taken the same approach. She had no recollection of the Trust attempting to contact her and she was happy to assist them in producing the reports in the most-effective way.

[52]She told the tribunal that the Trusts had access to the data on the TRAC system and they had a duty to provide the data in their NHS England submissions. She said she had provided evidence that reports could be produced other than quarterly (with reference to the email from Alex Davison dated 24th September 2024) and that the Trust were able to split joint campaign data as they had to do that for their WRES responses.

[53]She was disappointed that neither the Respondent nor the Trusts had attended. She had been caused extra work and frustration by the refusal. Her research was done pro bono and there was public interest in discrimination. She considered that this was an effort to prevent her having access to the data.

[54]She fully recognised that the data requested may no longer be capable of being supplied as it was held only for 400 days but she considered the appeal issues to be of some importance as regards future requests for the information.

The Legal Framework

[55]The Freedom of Information Act 2000 allows any person to make request of public authorities for information. The right is contained in section 1(1) as follows: (1) Any person making a request for information to a public authority is entitled— (a) to be informed in writing by the public authority whether it holds information of the description specified in the request, and (b) if that is the case, to have that information communicated to him.

[56]Subject to the authority requesting further information from the applicant to identify and locate the information, the Act provides for disclosure of the information (not documents) unless one or more exemptions in the Act apply.

[57]An authority may rely on an exemption under Part II of the act before the Tribunal that had not been relied upon previously, subject to the Court’s case management powers.

Compliance costs

[58]The public authority claimed that section 12 of FOIA applied as regards the costs of compliance with the Appellant’s request -(1) Section 1(1) does not oblige a public authority to comply with a request for information if the authority estimates that the cost of complying with the request would exceed the appropriate limit.

(2) Subsection (1) does not exempt the public authority from its obligation to comply with paragraph (a) of section 1(1) unless the estimated cost of complying with that paragraph alone would exceed the appropriate limit.

(3) In subsections (1) and (2) “the appropriate limit” means such amount as may be prescribed, and different amounts may be prescribed in relation to different cases.

[59]In summary, section 12 provides that a public authority is not obliged to comply with a request for information if the authority estimates that the cost of complying with the request would exceed the “appropriate limit” as prescribed in the Freedom of Information and Data Protection (Appropriate Limit and Fees) Regulations 2004. The public authority can still provide the information if they choose to, they are simply not required to. The exemption does not include informing the applicant whether it holds the information requested unless the estimated cost of complying with that would also exceed the appropriate limit (section 12(2).

[60]The 2004 Regulations give guidance as to the appropriate limit and what activities can be included in estimating the cost of compliance –(i) The appropriate limit is £ 600 for central government and the armed forces (Reg 3(2));

(ii) The appropriate limit is £ 450 in the case of other public authorities (Reg 3(3));

(iii) Costs are estimated at a rate of £ 25 per person per hour (Reg 4(4));

(iv) The limit of £ 600 means that central government can be expected to undertake up to and including 24 hours of work. For non-central government, this limit is exceeded after eighteen hours;

(v) The costs which a public authority can take into account are set out in Regulation 4(3) as follows: (a) determining whether it holds the information; (b) locating the information, or a document which may contain the information; (c) retrieving the information, or a document which may contain the information; and (d) extracting the information from a document containing it. However, the costs of determining whether an exemption applied or the public interest test could not be included in the cost estimate.

[61]The limit applies where there are several related requests as the total cost of compliance of those requests (Reg 5).

[62]The Upper Tribunal has issued the following guidance on estimates as to the costs of compliance –(i) The public authority must provide an estimate of what the compliance costs would be – the costs are those that the authority expects to incur;

(ii) The estimate is based on the way the local authority hold the information;

[63]A public authority does not have to provide an invoice style breakdown of their costs but there must be some basis of their calculations as to the estimated cost of compliance. Estimates must be reasonable and reasoned. The appropriate limit is assessed on the basis of the information storage and retrieval systems that a public authority actually has - not the ideal systems, or the systems that an Appellant thinks a public authority ought to have- Commissioner of Police for the Metropolis v Information Commissioner and Mackenzie [2014]UKUT 0479 (AAC).

[64]The Powers of the Tribunal on appeal are provided by section 58(1) of the 2000 Act: If on an appeal under section 57 the Tribunal considers—(a) that the notice against which the appeal is brought is not in accordance with the law, or

(b) to the extent that the notice involved an exercise of discretion by the Commissioner, that he ought to have exercised his discretion differently, the Tribunal shall allow the appeal or substitute such other notice as could have been served by the Commissioner; and in any other case the Tribunal shall dismiss the appeal.

[65]The powers of the Tribunal were considered by the Upper Tribunal in Information Commissioner v Malnick and the Advisory Committee on Business Appointments

[2018] UKUT 72 (AAC)

who confirmed that the Tribunal conducts a full merits review of the Commissioner’s decision albeit the starting point was the Commissioner’s decision. The Tribunal will give such weight as it considers fit to the Commissioner’s views and findings; and will determine whether the Commissioner’s decision was in accordance with the law. The appeal process is not adversarial, it is inquisitorial by nature.

Analysis of the evidence and findings on appeal

[66]The issue for the Tribunal is whether the costs of complying with the request of the Appellant would exceed the Regulations (i.e. over eighteen hours work). This is the exemption claimed by the Trusts which was supported by the Information Commissioner’s decision. The Trusts had claimed that each recruitment campaign (and the Trust ran between 500 and 1200 campaigns annually each) would require between fifteen minutes and two hours’ work to extract the relevant data manually including from TRAC. The position of the Appellant is that the information is already on the TRAC system and a report can be run for each Band in Part One of the request (there were fourteen Bands in the request).

[67]None of the Trusts had joined in the appeal. Noting the issues concerned the technical ability of their recruitment data and the nature of the submissions made by the Appellant, that was disappointing. Between them it was not unreasonable for them to nominate one person to assist in the appeal. They were as TRAC subscribers best placed to provide evidence about TRAC reporting but chose not to do so. No expert evidence about TRAC, no witness evidence from users or the Recruitment Hub or anyone who had concerns about the system had been provided to the tribunal and had not been provided by the Commissioner as evidence provided to him in the first instance. This was both concerning and surprising noting the submissions made. The only objective and independent evidence about TRAC had been supplied by the Appellant.

[68]The Appellant was a knowledgeable and reliable witness who proved that she had expertise in the TRAC system’s capabilities. Her assertions about the TRAC reports were credible and supported in that thirteen other Trusts were able to produce the information. She provided an example of the raw data that she had been provided by one other Trust which was sufficient for her research. She accepted that she may have to undertake some calculations based on the raw data.

[69]The Trusts who are subject to this appeal had not provided any of the Part One data. They had not provided what information they had or in a format that they could provide. It was not just a case that they could not break down the data to the extent requested; they simply had not provided the data at all. That was a less credible stance for the Trusts to take. The four Trusts that were subject to these appeals were all part of the same area. When she had countered the arguments of the Trust with the Commissioner, there had been no rebuttal reply and indeed the Trust have not sought to challenge her submissions at this appeal hearing.

[70]There was also further support in the form of a witness statement from Roger Kline dated 30th March 2024. In his covering email, he said that the NHS Trusts were required to maintain the data as part of the NHS Standard Contract and he was unaware of any shortcomings in the TRAC data noting that he had seen hundreds of WRES submission reports based on TRAC data. Mr Kline was heavily involved in creating the WRES standards. In his witness statement, he said he was surprised to hear that some NHS Trusts were suggesting that the data they held was not sufficiently accurate and the TRAC system was so poor that it would take hundreds or thousands of hours for them to manually create the data requested. His witness statement supported the Appellant and equally his evidence stood uncontroverted at the appeal hearing.

[71]Three of the Trusts had originally cited section 22 as an exemption and when chased for a response, had changed the basis of their refusal to section 12. Considering that the data was held for only 400 days, this was an egregious position for those Trusts to take noting that by the time the data was chased, it was no longer available. This impacted on the credibility of the position of the Trusts. In accordance with best practice, the data (or what data they had) should have been retained.

[72]Each of the Trusts claimed that it would take between fifteen minutes and two hours to provide the data for a campaign. It was not clear why there was such disparity between fifteen minutes and one hundred and twenty minutes. The estimate was not reasonably explained or why the estimate per campaign varied so wildly.

[73]It is recognised that TRAC data may be incomplete as some recruitment took place outside of TRAC. This should be an explanation to the data held and not a reason not to provide it. FOIA is a request for information held. There was no objective support for the Trust’s contention that the TRAC data was unreliable or rebuttal of the Appellant’s submission (and that of Mr Kline) that the WRES submissions were based on the same TRAC data. The Tribunal considered it very probative that the information had been provided before, although it is noted with less breakdown of the Bands, and crucially, that it had been provided that year by thirteen other Trusts, without difficulty.

[74]The Tribunal did not accept that the TRAC system varied from area to area. It would be the same system. The Tribunal accepted that a Trust may use the software in a different way or with different parameters and this was accepted by the Appellant. It may have been sensible and in the spirit of the Act for the Trust to work with the Appellant collaboratively to identify what data could be retrieved from TRAC in the most-efficient way.

[75]The Tribunal did not agree that the requested information was exempt from disclosure under section 12. Neither the Trust not the Commissioner could rely on that section and to that extent the Commissioner’s decision was wrong.

[76]The Tribunal’s powers are limited to replacing the decision of the Commissioner which we so do. However, it is recognised by the Tribunal and the Appellant that the information for the period requested may no longer be available. Nonetheless, the Trust are required to respond to Part One of her respective requests within the time frame set out at the top of this document.

[77]The appeal is allowed. District Judge Moan sitting as a Judge of the First Tier Tribunal 2nd October 2024 Promulgated on: 17/10/2024