Case Reference: EA/2021/0375
Information Rights
Heard: by CVP
Before
TRIBUNAL JUDGE SOPHIE BUCKLEY
TRIBUNAL MEMBER RAZ EDWARDS
TRIBUNAL MEMBER JO MURPHY
Between
LAW CENTRE NI
Appellant
and
THE INFORMATION COMMISSIONER
Respondent
Decision: The appeal is allowed in part.
Substituted Decision Notice:
Organisation: Department for Communities.
Complainant: Law Centre NI
The Substitute Decision – IC-99362-T3Z0
1. The public authority was entitled to rely on s 12 of the Freedom of Information Act 2000.
2. The public authority did not comply with its s 16 duty to provide advice and assistance.
3. The public authority breached s 1(1)(a) and 10(1) by failing to respond to the request within the time limit.
4. The public authority is not required to take any steps.
REASONS
Introduction
This is an appeal against the Commissioner’s decision notice IC-99362-T3Z0 of 25 November 2021 which held that the Department for Communities (‘DfC’) was entitled to rely on s 12(1) of the Freedom of Information Act 2000 (FOIA). The Commissioner found that DfC complied with its s 16 duty to provide advice and assistance. The Commissioner found that DfC breached s 1(1)(a) and 10(1) by failing to respond to the request within the 20 working day time for compliance.
The Commissioner required no steps to be taken.
The Law Centre NI (‘the Law Centre’) was substituted for Owen McCloskey as appellant by order dated 12 July 2022. Mr. McCloskey represented the Law Centre in the hearing, and it is in that capacity that he is referred to in the reasons below.
Background
The Appeals Service (Social Security Appeals), which is part of DfC, provide administrative support to the independent tribunals set up to hear appeals against the decisions made on Social Security by the decision makers in DfC.
The Law Centre has made an annual request to DfC for the Social Security Appeals statistics for the previous financial year. Prior to the financial year 2019/2020 this was usually provided in the format of a table with the following headings:
No Rep Involved | No Rep Involved total | ||||
Appeal type | Adjourned | Allowed | Disallowed | Withdrawn | |
[By type of benefit] | |||||
Total |
Rep Involved | Rep Involved Total | Total | |||
Adjourned | Allowed | Disallowed | Withdrawn | ||
The table breaks down the outcome of the appeals by each type of benefit and by whether or not a representative was involved.
The Law Centre publishes the information annually and uses it as an evidence base to identify and attempt to tackle any issues and to improve access to justice. The Law Centre has a particular interest in the 2019/2020 statistics so that a comparison can be made between the statistics prior to and following the impact of the Covid pandemic.
At some point prior to the request, the computer system used by the Appeals Service (the Northern Ireland Appeals Processing System – NIAPS) was no longer capable of producing the statistics in the same way they had been produced in previous years. The bespoke defined report that had been created in the past no longer worked. A new system was planned but had not been introduced at the relevant time.
Request and response
On 22 June 2020 Mr. McCloskey made the following the request on behalf of the Law Centre NI:
We would appreciate it if you could provide us with a copy of the finalised Social Security Appeals stats for the financial year 2019/20:
Outcome at Hearing by Appeal Type and Representation
DfC responded on 18 September 2020. DfC provided the total number of appeals received in 2019/2020, the total number of successful appeals and the number of appeals that had representation. DfC stated that in-depth information was not available because the case management system was being upgraded.
Mr. McCloskey asked for an internal review. On 23 December 2020 the internal review found that DfC had not issued a response within 10 working days. DfC withheld the information under s 12. It estimated that the cost of extracting the data from the database would exceed the appropriate limit. DfC stated that it was not able to advise on reforming the request to bring it under the cost limit.
Mr. McCloskey referred the matter to the Information Commissioner on 9 April 2021.
The decision notice
The Commissioner accepted that either to open each case manually to retrieve the information or to ‘buy in’ the necessary skills to extract the information from the raw database would exceed the appropriate cost limit. The Commissioner was satisfied that the estimates were realistic and reasonable. He concluded that DfC was entitled to rely on s 12(1) to refuse the request. The Commissioner was satisfied that DfC had met its obligation under s 16 because there was no feasible way in which the request could meaningfully be reformulated. The Commissioner found that DfC had breached s 1(1)(a) and s 10(1) by failing to respond the request within the 20 working day time for compliance.
The appeal to the tribunal
The ground of appeal is, in summary, that the Commissioner was wrong to conclude that DfC was entitled to rely on s 12 and had complied with s 16 because the majority of the information originally requested was subsequently provided to the BBC as part of a FOI request.
The Commissioner’s response
The Commissioner interprets the grounds of appeal to be, in essence:
That the Commissioner failed to take account of relevant evidence, namely the fact that the majority of the information had subsequently been provided to the BBC; and
That the Commissioner provides conflicting information in the decision notice about the future ability to access this data.
The Commissioner submits that the tribunal does not have the remit to review either the way in which the Commissioner carried out his investigation or the way in which the decision notice is drafted.
The Commissioner had reverted to DfC who had responded to confirm that, at the time of the request, DfC was not able to generate a report linking all three elements required to answer the request. The appeals service was able to provide either the outcome of social security appeals and the benefit concerned or the outcome of social security appeals and the representation status.
Mr. McCloskey’s reply
Mr. McCloskey confirms that the substantive findings in relation to s 12 and s 16 are in dispute.
In relation to s 12, Mr. McCloskey notes that there is no explanation in the decision notice of why DfC estimate 32 hours for the work to be done by an ICT organisation, when on 26 October 2021 the case officer had provided Mr. McCloskey with the DfC response which stated that it was estimated to take between 2-4 days (16-32 hours).
In relation to s 16 Mr. McCloskey submits that DfC could have provided him with the information set out in the Commissioner’s response. DfC could have advised Mr. McCloskey to reformulate his request to ask for the outcome of social security appeals and the benefit concerned and, separately, the outcome of social security appeals and the representation status.
Further submissions of the Commissioner
The Commissioner had obtained confirmation from DfC that it was able to provide, with the costs limit, either (but not both) (1) the outcome of social security appeals and the benefit concerned or (2) the outcome of social security appeals and the representation status.
In relation to s 12 the Commissioner submits that the task would require a minimum of 2 visits lasting 2 days each. This is confirmed by DfC at p 121 of the open bundle.
In relation to s 16, in the light of the recent confirmation that DfC could have provided either (1) the outcome of social security appeals and the benefit concerned or (2) the outcome of social security appeals and the representation status, the Commissioner now considers that there is a breach of s 16 to the extent that it did not provide this advice to Mr. McCloskey.
Skeleton argument on behalf of the appellant
The tribunal has taken account of the skeleton argument but understands the principal point to be as follows.
On 19 June 2020, just 3 days before the FOI request, the Minister for Communities provided a parliamentary answer which combined (i) the appeal outcome (ii) the benefit involved and (iii) the representation status for the individual benefit of Personal Independence Payment (PIP). PIP represents by a significant margin the highest volume of all appeals that are heard in the Social Security tribunal (p129 of the open bundle).
It is submitted that the s 12 appeal should be allowed in the absence of detailed explanation of how the minister was able to access information that DfC have stated is inaccessible.
Oral submissions on behalf of the appellant
The tribunal has taken full account of the focussed and ably presented oral submissions made by Mr. McCloskey on behalf of the Law Centre.
Evidence
We read and took account of an open and a closed bundle. The closed bundle contains annex A and annex B to the letter to the Commissioner dated 19 November 2021.
The relevant law
Under s 12(1) a public authority is not obliged to comply with a request for information where:
…the authority estimates that the costs of complying with the request would exceed the appropriate limit.
The relevant appropriate limit, prescribed by the Freedom of Information and Data Protection (Appropriate Limit and Fees) Regulations 2004 (‘the Regulations’) is £600.
In making its estimate, a public authority may only take account the costs it reasonably expects to incur in relation to the request in–
(a) determining whether it holds the information,
(b) locating it, or a document which may contain the information,
(c) retrieving it, or a document which may contain the information, and
(d) extracting it from a document containing it. (See regulation 3).
The Regulations specify that where costs are attributable to the time which persons are expected to spend on the above activities the costs are to be estimated at a rate of £25 per person per hour.
The estimate must be sensible, realistic and supported by cogent evidence (McInnery v IC and Department for Education [2015] UKUT 0047 (AAT) para 39-41).
The test is not a purely objective one of what costs it would be reasonable to incur or reasonable to expect to incur. It is a test that is subjective to the authority but qualified by an objective element. It allows the Commissioner and the tribunal to remove from the estimate any amount that the authority could not reasonably expect to incur either on account of the nature of the activity to which the cost relates or its amount. (Reuben Kirkham v Information Commissioner [2018] UKUT 126 (AAC)).
Under s 16 a public authority has a duty to provide advice and assistance, so far as it would be reasonable to expect the authority to do so,
The Task of the Tribunal
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, where the Commissioner’s decision involved exercising discretion, whether he should have exercised it differently. The tribunal may receive evidence that was not before the Commissioner and may make different findings of fact from the Commissioner.
Discussion and Conclusions
Having considered the information provided by DfC we make the following findings on the balance of probabilities. At the relevant time the system used by the Appeals Service was no longer capable of producing the statistics in the way that it had previously produced them for the Law Centre on an annual basis. There were problems with the reporting tool and its ability to extract data was limited. The raw data was still held in the system, but extracting it was not as simple as it used to be.
We find, on the balance of probabilities, that it was possible, at the relevant time and without exceeding the appropriate limit, to use the reporting tool to extract the outcome of social security appeals broken down by appeal type/benefit concerned. We find that it was also possible, at the relevant time and without exceeding the appropriate limit, to use the reporting tool to extract the outcome of social security appeals broken down by ‘representation status’ (i.e. whether or not there was representation). However, in order to answer the request, both those figures need to be provided. Further, they need to be combined. For the reasons set out below, we find that this would exceed the appropriate limit.
DfC assert that at the time of the request, the reporting tool could not be used to combine that data as had been done in previous years to produce a table with headings set out under ‘Background’ above. Consequently they assert that extracting the data, whether done internally or by using an external IT expert, would exceed the appropriate limit.
We have taken account of the fact there has been some variation in the explanations provided by DfC as to what data can be provided and why some data cannot easily be provided. In deciding to accept DfC’s assertions and making the findings as set out above, we have taken account of the fact that DfC have willingly provided this data in the past, and that there is no evidence of any motive for refusing to provide similar data for 2019/2020.
Mr. McCloskey relies on information provided by DfC to the BBC (in 2021) and to the Minister for Communities to suggest that DfC could have provided the requested data withing the appropriate limit.
Mr. McCloskey relies on information provided to the BBC in response to a request in 2021. The information provided to the BBC appears at in tables reproduced at pages 28, 29 and 30 of the open bundle. It relates to the financial years 2019/2020 (p28), 2020/2021 (p29) and the first part of 2021/2022 (p30). The tables give outcomes by appeal type/benefit type as follows: confirmed/less advantageous/more advantageous/struck out at hearing/withdrawn at hearing.
The tables do not include the representation status of each appeal and do not combine the data on appeal type/representation/outcome as had been done in the tables provided to the Law Centre in previous years. These tables do not assist us in determining whether or not it was possible to easily extract the data needed to answer the request submitted by the Law Centre.
We note that in their letter to the Commissioner dated 19 November 2021 DfC refer to a request for information from the BBC in August 2019, in response to which they provided statistics for the first quarter of 2019/2020. They state that the report, which included representation by benefit type, was still available at that time. The information that DfC says it provided to the BBC in 2019 is contained in the closed bundle.
It is not clear whether there is a typo in the letter and DfC intended to refer to the 2021 request relied on by Mr. McCloskey or whether DfC are referring to a different request by the BBC made in 2019.
We note that the information on pp 28-30 is different to that contained in the closed bundle. Further we note Mr. McCloskey’s submission that the total number of appeals for 2019/2020 at 28 appear high if the figures are only for the first quarter. Finally we note that DfC refer to the report including ‘representation by benefit type’ which is not included in the information provided to the BBC. For those reasons it seems likely that DfC are referring to a different request in 2019.
Neither the information provided to the BBC in August 2019 (which was provided at a time when it was still possible to easily extract the information) nor the information provided to the BBC in 2021 (which did not combine the data on appeal type/representation/outcome) assist us in determining if it was possible to extract the requested data at the relevant time.
Mr. McCloskey also drew our attention to an answer provided by the Minister for Communities to the Northern Ireland Assembly on 19 June 2020, a few days before he made his request:
Question:
To ask the Minister for communities to detail the number of (i) successful; and (ii) and unsuccessful Personal Independence Payment appeals, broken down by those with and without tribunal representation, over the last three years.
Answer:
During the last three years there have been 14,198 Personal Independence Payment appeals that have had a final decision, of which 8,347 were successful and 5,852 were unsuccessful.
The appellant was represented in 6,547 of the 8,347 successful appeals and the remaining 1,800 were unrepresented.
In relation to the 5,852 unsuccessful appeals 3,607 appellants were represented while the remaining 2,245 were unrepresented.
Mr. McCloskey’s submission, in essence, is that in order to provide the Minister with this information, DfC must have extracted the data that he requested for 2019/2020 for PIP appeals. When it came to respond to the Law Centre’s request, DfC should therefore have been able easily to access the data for PIP appeals. PIP appeals account for about 80% of all social security appeals. The time that it would take to find the remaining 20% of the requested information would be much reduced – probably by 80%.
Mr. McCloskey also questions DfC’s assertion that it was not possible (at least not without exceeding the appropriate limit) to combine the data by appeal type/representation/outcome, when this is precisely what was done in order to provide the Minister with the information needed to answer the question on 19 June 2020.
Although the issue of information provided to the Minister has been addressed in general by the DfC in their correspondence with the Commissioner, DfC have not been asked to explain (i) what information or data was provided to the Minister to enable the answer to be given (ii) how that data was extracted from the system, given that a report combining the data could no longer be provided at the relevant time and (iii) to what extent, if at all, that extracted data or information avoided the need for data to be extracted for the purposes of the request in issue in this appeal.
We considered whether to require further information from the DfC before we reached our decision. We took account of the fact that DfC had been asked and answered further questions from the Commissioner on a number of occasions and had been invited to join the proceedings. We also took account of the expense and delay that would be caused and considered proportionality. We decided that it was in the interests of justice and in accordance with the overriding objective to proceed on the basis of the information before us.
The answer given by the Minister shows that it was possible for data to be extracted for ‘the last three years’ which showed the number of successful and unsuccessful PIP appeals which had a final decision, broken down by those with and without tribunal representation.
There is no evidence before us on how long it took to extract that data or how it was extracted. We do not know if that data was easily available, or if it took longer than the appropriate limit to extract.
The answer concerns the number of PIP appeals over the last three years. We do not know the date when ‘the last three years’ runs from. We do not know whether, when the underlying data was extracted, it was recorded in a way which would enable the numbers for 2019/2020 to be ascertained. We do not know if the data was recorded at all, or for how long the written record was retained by DfC, if at all.
Although it is clear from information in the bundle that PIP appeals form the majority of social security appeals by number of appeals, there are approximately 20 other appeal types. The information provided to the Law Centre for 2018/2019 (p 64 of the open bundle) shows that there were 9,713 PIP appeals and 3,353 other appeals. Even if the information on PIP appeals had been readily available, DfC would still have had to extract the information on approximately 3,500 other appeals. Because of the uncertainties set out above we do not accept that it is appropriate to reduce the estimate in the light of the Minister’s answer, but we have considered what the effect of a reduction would have been in any event in our consideration of the estimates below.
In relation to the estimated cost of extracting the data, we accept, on the balance of probabilities, that the two alternatives available to DfC at the time were (i) to extract the raw data manually and (ii) to pay for external expertise to resolve the reporting problems.
In relation to (i) we accept, on the basis of the sampling exercise carried out, and the information provided by the DfC on the method that would need to be adopted, that the cost of extracting the data would have far exceeded the appropriate limit. This is the case even if 3,500 rather than 12,718 records had to be reviewed.
In relation to (ii) we accept, on the basis of the evidence in the closed bundle, that the estimate given by the external expert for how long the work would take was 32 hours. The work was intended to assess, advise on, and if time allowed, to fix the reporting issues. The time estimate would therefore not be affected even if the PIP data provided to the Minister was readily available.
For all those reasons we accept that the estimates provided were sensible, realistic and supported by cogent evidence and that it would exceed the appropriate limit to answer the request.
We note that DfC have indicated that they had decided not to extract the data in any event because of the risk of permanent damage to the system/data. Because of our conclusions above, we do not need to make findings on this.
S 16
The Commissioner had indicated that he now considers that there was a breach of s 16 to the extent that DfC did not advise the Law Centre that it could have provided either the outcome of the social security appeals and the benefit concerned or the outcome of the social security appeals and the representation status within the appropriate limit. We agree that this is a breach of s 16.
We also find that it was a breach of s 16 for the DfC to advise the Law Centre that it could produce the information provided to the BBC and/or the information provided to the Minister for the parliamentary answer dated 19 June 2020 and to invite him to submit a reformulated request if he wished to have that information. As Mr. McCloskey now, in effect, has that advice we do not require DfC to take any steps.
Note on whether this information will be available in the future
Mr. McCloskey highlighted his concerns about inconsistencies as to whether the requested information would be available in the future. Whilst we acknowledge the importance of this information to the Law Centre, it is outside our remit to make any findings on this issue.
Signed Sophie Buckley Date: 17 October 2022
Judge of the First-tier Tribunal
Promulgated Date: 17 October 2022