Cinneadh
Ms P and Local Government Management Agency
Ó Oifig an Choimisinéara Faisnéise
Cásuimhir: OIC-106968-Z3F4Q8
Foilsithe
Teanga: Níl leagan Gaeilge den mhír seo ar fáil.
Ó Oifig an Choimisinéara Faisnéise
Cásuimhir: OIC-106968-Z3F4Q8
Foilsithe
Teanga: Níl leagan Gaeilge den mhír seo ar fáil.
Whether the LGMA was justified in refusing access to records relating to a report on the value for money of the mutual insurance model, on the grounds that the records are exempt under sections 15(1)(d), 35, 36, 37 and 40(1) of the FOI Act
18 June 2024
By way of background, this case relates to a request made on 3 February 2017 for access to a copy of a report commissioned by the LGMA from PricewaterhouseCoopers (PwC) on the value for money of the mutual insurance model used by local authorities and education and training boards; relevant minutes and records of deliberations; and records of a joint steering committee relating to the same issue.
The insurance concerned is provided by Irish Public Bodies Mutual Insurance (IPB), which is an insurance company owned entirely by its policyholders.
The LGMA part granted the request and its internal reviewer affirmed this decision. The applicant applied to my Office for a review of the LGMA’s refusal of the remaining records. This Office issued a decision on 15 September 2017 (OIC Case No. 170136 refers), annulling the LGMA’s decision and directing the release of the records sought. IPB appealed this decision to the High Court.
Following a preliminary hearing, the matter was remitted by consent to my Office for further consideration by way of an Order of the High Court perfected on 18 March 2021. Accordingly, the review was reactivated under reference number OIC-106968-Z3F4Q8 for the purpose of making a fresh determination on the matter.
I have now completed my review in accordance with section 22(2) of the FOI Act. In carrying out my review, I have had regard to the extensive submissions made by the applicant and to those made by the LGMA in support of its decision. I have also had regard to the submissions made by IPB, including its identification of specific parts of records 3 and 6 which it considers to contain commercially sensitive and confidential information. Finally, I have had regard to the contents of the records concerned. I have decided to conclude this review by way of a formal, binding decision.
In the previous case, the LGMA identified eight relevant records, of which it released records 1 and 2 in full and 4, 5 and 8 in part. In submissions to this Office during the course of this review, the LGMA clarified that it was relying on sections 15(1)(d), 35(1)(a), 36(1)(b) and 40(1)(d) to refuse access to the remaining records in full or in part. During the course of this review, IPB also contended that section 37 applied to some of the information contained in record 6.
Accordingly, this review solely concerns whether the LGMA has justified its decision to refuse access to records 3-8 in full or in part on the basis of sections 15(1)(d), 35, 36, 37 and 40 of the FOI Act.
Before considering the exemptions claimed, it is important to note the following matters:
Section 22(12)(b) of the FOI Act provides that a decision to refuse to grant a request under section 12 shall be presumed not to have been justified unless the head of the relevant FOI body shows to the Commissioner's satisfaction that its decision was justified. This means that the onus is on the LGMA to satisfy this Office that its decision to refuse access to the records sought, either in whole or in part, was justified.
While I am required to give reasons for my decision under section 22(10) of the FOI Act, I am also required under section 25(3) to take reasonable precautions to prevent disclosure of information in an exempt record.
This Office has no remit to adjudicate on how FOI bodies perform their functions generally. It is also important to note that the release of records under FOI is, in effect, regarded as release to the world at large given that the FOI Act places no constraints on the uses to which the information contained in those records may be put.
In her submission to this Office dated 11 August 2022, the applicant referred to a 2015 Willis Towers Watson (WTW) report and a 2012 Legalwise report. She sought confirmation of a number of matters, including her understanding that WTW and Legalwise were acting as service providers to FOI bodies. The applicant stated that she was “again seeking” the WTW report, “in addition” to the Legalwise report. This Office’s Investigator informed the applicant of her view that neither report came within the scope of her FOI request in this case, and accordingly, would not form part of this review.
In her response, the applicant stated that she “did not accept” that the WTW and Legalwise reports fell outside the scope of her FOI request. She went on to state that she “never accepted that part of the original decision” and made additional arguments as to the nature of the reports in question. The applicant also appears to be of the view that the WTW report contains information which is more relevant to her request than PwC’s summary reports. While this may be the case, these records were not identified by the LGMA in relation to the applicant’s request in this case.
I note the applicant’s reference to Dublin City Council in her letter dated 11 August 2022, and it appears as though the reports mentioned may have been an issue in another case (OIC Case No. 170565 refers). The reports concerned appear to be those that were found to be outside the scope of that particular case.
It is, of course, open to the applicant to make a new request to the relevant FOI body for these reports. I should also state that even if the two reports were within the scope of her original request to the LGMA, this matter was not raised by the applicant in her internal review request or in her application for a review to this Office of the LGMA’s decision. In any event, I am satisfied that these records do not come within the scope of this review.
The applicant also indicated that, in her view, she had not been accorded “natural justice and fair procedures”, as she was “being asked to comment on submissions from various parties” which she had not seen. She stated that until she was provided with that evidence, she was “severely hampered” in drafting further submissions. On this basis, the applicant submitted an FOI request to this Office for a copy of the “entire file relating to this matter including both internal and external communications as well as all file notes and other papers”. The applicant requested that no further action be taken in this review until her FOI request had been complied with.
I understand that a decision has since issued on the applicant’s FOI request. While Schedule 1, Part 1 of the FOI Act provides that this Office is not subject to FOI other than in relation to records concerning the general administration of its functions, and that the records were not released to the applicant, I also note that she has since made additional submissions in this case in support of her views that the records should be released in full in this case.
It is important to note that reviews undertaken by my Office are inquisitorial, as opposed to adversarial, in nature. Submissions made by one or more parties to a review are likely to contain sensitive information that may not be appropriate for disclosure to others, accordingly, submissions are not exchanged between parties to a review as a general rule. However, we take care to ensure that the parties are notified of material issues arising for consideration.
The High Court has previously considered the fairness of our procedures in the context of our treatment of submissions. Specifically, Quirke J made the following comments in the National Maternity Hospital case: “I know of no principle of natural or constitutional law or justice which confers upon parties who make submissions to a decision-making body the right to respond to the submissions made by every other party who participates in the process. The review undertaken by the Commissioner was a statutory process which expressly envisaged and permitted the adoption of informal procedures.”
It is also important to note that under section 25(3) of the Act, the Commissioner is required to take all reasonable precautions to prevent the disclosure of information contained in an exempt record or matter that, if it were included in a record, would cause the record to be an exempt record. As submissions made by parties to a review may contain sensitive information that may not be appropriate for disclosure to others, submissions are not exchanged as a general rule.
As I have stated, this Office notifies the relevant parties to a review of new material issues arising for consideration insofar as they affect the interests of the parties concerned. Material issues are issues that are relevant to the outcome of the review. Such issues involve information of significance that is likely to influence the decision the Commissioner will make. Any new matter of which a party to a review is unaware and which is likely to cause the Commissioner to make a decision adverse to the interests of that party is communicated to the party concerned. Such matters generally include applicable exemptions not previously raised, pertinent search details not previously disclosed to the applicant, and any new facts and new legal developments which are likely to have a significant bearing on the outcome of the review.
Under section 45(6) of the FOI Act, the Commissioner has discretion to adopt such procedures as are appropriate in all the circumstances of a case. In all circumstances, this Office aims to ensure that the approach adopted is fair, and seen to be fair, to all the parties concerned.
Furthermore, the Court of Appeal’s judgment in Grange vs Information Commissioner [2022] IECA 153 held that it was a matter for the Commissioner to determine the procedure to be followed in any particular case and stated that “[i]n general, therefore, in my view the practice adopted by the respondent of furnishing a précis or summary of an FOI body’s response to the requestor will amount to a fair procedure, and of course the requestor usually should, as occurred here, be given an opportunity to respond, even though the statute is silent in this regard.”
I am satisfied that the applicant in this case was on notice of all material issues of relevance to the review, including the relevant arguments made by the third party concerned, and was afforded an appropriate opportunity to make submissions on those issues.
The Records at issue
The records at issue in this case comprise the following:
• Record 3 : three reports from PwC covering Phases 1, 2 and 3-4 of its value for money review
• Record 4 : minutes of the steering committee meeting on 25 January 2016
• Record 5 : minutes of the steering committee meeting on 27 June 2016
• Record 6 : PwC’s final compendium value for money report
• Record 7 : PwC’s summary report
• Record 8 : a summary report by the steering committee
Records 3, 6 and 7 were withheld in full by the LGMA and records 4, 5 and 8 in part.
Section 15(1)(d) – administrative refusal
This Office’s decision in the previous case issued on 15 September 2017. In January 2019, records 7 and 8 were published online by the Public Accounts Committee (PAC) following their submission in response to a PAC request. Record 7 and 8 are summary reports following the value for money review. From a careful review of the records concerned, I can confirm that pages 2-13 of record 6 comprise PwC’s summary report – i.e. record 7. The LGMA indicated that it is relying on section 15(1)(d) in relation to its decision to refuse access to records 7 and 8 in full and record 6 in part.
Section 15(1)(d) provides for a discretionary administrative refusal to release records which are already in the public domain.
I note the applicant’s objection to the application of section 15(1)(d) to these records. Among other things, she objected to the Investigator’s view that the LGMA’s reliance on section 15(1)(d) at this point was reasonable in respect of records 7 and 8 in full and record 6 in part, as they were essentially placed in the public domain in 2019 when they were published online by the PAC. The applicant argued that it was a “fundamental flaw” for an assessment by this Office of the reasonableness of the LGMA’s decision made in January 2017 to be based on subsequent events relating to the PAC in 2019. She stated that in line “with the jurisprudence of the courts in judicial reviews” and of other reviewing bodies such as the Financial Services and Pensions Ombudsman, it was the reasonableness of the LGMA’s conduct at the time of its decision which must be assessed.
While I note the applicant’s comments in this regard, it is important to note that a review by this Office of an FOI body’s decision under section 22 of the FOI Act is considered to be de novo as it is based on the circumstances and the law as they apply on the date of the decision. This approach was endorsed by the High Court judgment of Ó Caoimh J in the case of The Minister for Education and Science v Information Commissioner [2001 IEHC 116 ], available on our website at www.oic.ie . Similarly, in The National Maternity Hospital and The Information Commissioner [2007 3 IR 643 ] (the National Maternity Hospital case), the High Court (Quirke J) stated as follows: "The Commissioner was entitled to consider all of the material before her on the date on which she made her decision and to make her decision having regard to the circumstances which existed on [the date of her decision]".
In this case, it is important to note that this Office’s original decision issued in 2017. In this regard, the historical nature of documents has often been a significant factor when considering whether records contain information which is commercially sensitive under the FOI Act. I am satisfied that the context and background concerning the creation of the records at issue, as well as the pertaining circumstances at the time of the review by this Office are very much relevant to this review.
The applicant stated that it took her “some time” to locate the records published by the PAC. She also questioned whether the “placing of correspondence on the website of the [PAC]” meant that the records were actually in the public domain for the purposes of section 15(1)(d), “from the perspective of a layperson seeking transparency on the spending of public funds”.
This Office has found in previous cases that there is a distinction between, say, reports in the media relating to matters contained in records sought under FOI, and the publication of what may be described as primary documents relating to the matters concerned. This Office has also accepted that written replies to Parliamentary Questions published on the Oireachtas website, rather than by the FOI body concerned, were sufficient for the records to be considered to be in the public domain for the purposes of section 15(1)(d) (Case No. 160157 refers).
While I note the applicant’s arguments, I am satisfied that records 7 and 8 in full are publically available at the following link:
https://data.oireachtas.ie/ie/oireachtas/committee/dail/32/committee_of_public_accounts/submissions/2019/2019-01-17_correspondence-paul-dunne-ceo-local-government-management-agency-32r001801-pac_en.pdf Furthermore, I am cognisant of the fact that the FOI Act does not require public bodies to release records that are otherwise publicly available. As records 7 and 8 have been published online by the PAC, I am satisfied that they (and, by extension, pages 2-13 of record 6) are in the public domain. Accordingly, I find that the LGMA has justified its decision to refuse access to records 7 and 8 in full and pages 2-13 of record 6 on the basis of section 15(1)(d) of the FOI Act.
In the interests of completeness, I should state that in its submissions to this Office, the LGMA also seemed to be arguing that records 7 and 8 contained a “distillation” of the information contained in records 3 and 6, and that, therefore, section 15(1)(d) should also apply to records 3 and 6 in full. However, having carefully examined the content of the records concerned, I do not accept that the publication of summary reports in this case is the equivalent of putting the complete versions of PwC’s reports in the public domain, as they exclude a lot of information contained in the full reports. Accordingly, I find that the LGMA has not justified its decision to refuse access to record 3 in full and the remaining information in record 6 on the basis of section 15(1)(d) of the FOI Act.
I also note that the applicant essentially argued that these records were not the full reports sought by her and that the availability or otherwise of these particular records should have no bearing on this Office’s decision in relation to records 3 and 6. I can confirm that each of the records at issue in this case was considered individually by this Office during this review and that records 3 and 6 will be considered further below.
The LGMA refused to grant access to the remaining records concerned (record 3 in full and records 4, 5 and 6 in part) on the basis of sections 35, 36 and 40 of the FOI Act. I consider section 36 to be of the most relevance, so I shall consider its application to the records in the first instance.
The LGMA and IPB have both indicated that they consider the second limb of section 36(1)(b) to apply in this case, i.e. that the release of the records sought could prejudice the competitive position of IPB in the conduct of its business or otherwise in its occupation.
Section 36(1)(b) provides for the mandatory refusal of a request if the record concerned contains financial, commercial, scientific or technical or other information whose disclosure could reasonably be expected to result in a material financial loss or gain to the person to whom the information relates, or could prejudice the competitive position of that person in the conduct of his or her profession or business or otherwise in his or her occupation. It is also subject to a public interest balancing test at section 36(3).
The harm test in the first part of subsection (1)(b) is whether disclosure of the information -could reasonably be expected to result in material financial loss or gain. The Commissioner takes the view that the test to be applied in this regard is not concerned with the question of probabilities or possibilities, but with whether the decision maker’s expectation is reasonable. The nature of the harm envisaged and a basis for a claim that such harm could reasonably be expected to result from disclosure of the particular information in the record(s) at issue should be shown by an FOI body or a third party relying on this provision.
The harm test in the second part of subsection (1)(b) is whether disclosure of the information could prejudice the competitive position of the person in the conduct of his or her profession or business or otherwise in his or her occupation. The only requirement which has to be met in the second part of section 36(1)(b) is that disclosure "could prejudice the competitive position" of the person concerned. The standard of proof necessary to meet this test is considerably lower than the standard required to meet the test of "could reasonably be expected to" in the first part of section 36(1)(b). While the standard of proof to meet this test is reasonably low, this Office nonetheless takes the view that, in invoking the phrase "prejudice", the damage which could occur as a result of disclosure of the information must be specified with a reasonable degree of clarity.
In its submissions to this Office the LGMA did not elaborate on its grounds for relying on section 36 to refuse access to the records sought. It indicated that the ongoing commercial sensitivity of the information in the records was a matter for IPB to address and provided a copy of a letter from IPB dated 2 September 2021 in support of the LGMA’s decision. IPB also provided a copy of this letter with its submissions to this Office. This will be considered further below.
In earlier correspondence with this Office following the reactivation of this review, the LGMA provided a copy of an undated and unsigned document entitled “Reactivation of FOI Request OIC – 106968-Z3F4Q8”, which appeared to reflect IPB’s position rather than that of the LGMA. The document comprised nine pages of dense text and echoed many of IPB’s arguments set out below. I note that that the undated document did not cite section 35 of the FOI Act, although it referred to the records being submitted in confidence and described the information concerned as confidential. The document also referred to section 36(1)(c) of the FOI Act, however, this subsection was not relied upon in the subsequent focused submissions provided to this Office by the LGMA or IPB. In any event, I am satisfied that this document relates solely to IPB’s position, and I will consider it alongside IPB’s own submissions below.
Among other things, this Office asked IPB to identify the specific information (financial, commercial, scientific, or technical or other) contained in each record which it considered should not be released under section 36. IPB provided comprehensive submissions to this Office in response, as well as the following documents:
1. A spreadsheet identifying the particular information in the records which it considered to be “[h]ighly commercially sensitive and confidential and not in the public domain” (set out below).
2. A copy of an expert independent opinion provided by an actuary in support of its position. I understand that this opinion is a 2022 update of a document originally prepared in 2018 for the purposes of the High Court appeal of the Commissioner’s decision.
3. Email correspondence with the actuary who drafted the report, confirming that his views still stood, despite the passage of time since he prepared his first opinion.
4. A copy of a letter from IPB to the LGMA dated 13 November 2018 concerning the request from the PAC for a copy of PwC’s report.
5. A copy of IPB’s letter to the LGMA dated 2 September 2021.
6. A copy of a letter from IPB to PwC on 2 September 2021 following the remittal of this case by the High Court.
IPB stated that it had identified specific information in records 3 and 6 which, if disclosed, could be prejudicial to its competitive position. It provided commentary on each of parts of the records identified, alongside the relevant commentary from the expert opinion mentioned above.
Based on the above, I am proceeding on the basis that IPB has no objection to the release of the information contained in records 3 and 6 other than that it specifically identified to this Office. I also note that IPB made no arguments in relation to the information withheld from release in records 4 and 5.
By way of background, IPB stated that it permitted access to confidential and commercially sensitive information of a proprietary nature by a limited number of LGMA representatives and PwC experts strictly subject to confidentiality and the terms of a non-disclosure agreement (NDA). IPB also stated that it engaged in good faith in recognition of its stakeholder responsibilities, on the understanding that the information provided, particularly its raw actuarial data which it noted was not reproduced in the Summary Reports, would remain confidential. It considered that “significant aspects” of the information in the records had already been provided in the course of addressing the applicant’s FOI request and in engaging with the PAC. Its position was that only “limited” material remained which had not been released following the publication of records 7 and 8.
The additional arguments in relation to section 36 contained in IPB’s submissions, its supporting documentation, including the expert opinion and the undated document provided by the LGMA were broadly as follows:
• The type of information and the level of detail in the records is not in the public domain and is not subject to statutory reporting obligations.
• Within the company itself this information is considered sensitive, and access is primarily restricted to members of IPB’s Executive, Board, auditors and regulator.
• The records contain “acutely” confidential and commercially sensitive financial, commercial, technical and strategic information of “substantial value” and there were “no circumstances” in which a commercial undertaking would share such information, as release could harm its interests.
• The level of information contained in the records meant that it would be possible for potential competitors to use it to their advantage, prejudicing IPB’s competitive position.
• IPB’s chief competition is other insurers offering products to its members and “a number already do so”. The information in the records would allow its competitors to target the most profitable customers, which were, in effect, supporting less profitable segments of its business.
• Release of the records could reasonably be expected to result in a material financial loss to IPB and its members and an unwarranted benefit to competitors. This would jeopardise IPB’s ability to meet its members’ insurance needs and “undermine the mutual insurance model itself”.
• IPB has a competitive advantage in knowing and being able to anticipate its customer needs more precisely than potential competitors, derived from more than 90 years of dealing with public bodies. Release would benefit IPB’s competitors, who do not have the same level of insight into IPB’s customers as it has. This could prejudice that advantage/competitive position.
• The information identified is still relevant and of use to competitors, as it does not just provide a “point in time” view of its business, but would provide insight into ongoing features of its underwriting strategy and could allow competitors to extrapolate IPB’s current pricing structures and prices, to its “grave detriment”.
• In terms of the value of the competitive interests at risk, IPB stated that it competes in Ireland’s general non-life insurance market which recorded gross written premiums of €3.64 billion in 2020. It stated that of that figure, its own proportion was €144.1 million.
• The FOI Act was not designed as a “means by which the operations of private enterprises were to be opened up to scrutiny”.
IPB identified the following information as particular commercially sensitive:
• Record 6 : pages 41-46, 51-61, 64-66, 73-74, 82 (in part), 84 (in part), 93 (in part) and the corresponding pages in Record 3.
• Record 3 : as above, and pages 5-7 of the Phase 2 report and page 5 of the Phase 3 and 4 report (in part).
Essentially, IPB’s position was that, regardless of the age of the information concerned, the particular information it identified in the records comprised a granular level of detail about its business, approach and positions which it was not required to make otherwise publically available. It stated that its release would give its competitors detailed insights into how it prices its insurance products and put them in a position to cherry pick and better target IPB’s most profitable customers, which could reasonably be expected to prejudice its competitive position. It stated that the information identified includes details of its pricing methodology, projections of loss ratios under various scenarios, its approach to business development, including specific underwriting decisions on individual policies, underwriting risk and handling claims, as well as detailed breakdowns of its pricing strategy and approach, and its reserves. It contended that some of its approaches and positions do not necessarily change frequently. It argued that on that basis, the release of the information concerned could reasonably be expected to allow its competitors to extrapolate its current positions and approaches (which could reasonably be expected to prejudice its competitive position).
The applicant made extensive submissions in this and the previous case. She also submitted a number of reports and other documents relating to these matters in support of her views. While I will not address these all below, I can confirm that I have regard to their content.
The applicant was of the view that commercial sensitivity did not arise in this case. She referred to the e-tender website and stated that no local authorities had tendered for liability insurance since 2014, other than Dublin City Council tendering for art property insurance. The applicant contended that the records sought were “now so old” that they could no longer be “asserted to be of current commercial value”. She also argued essentially, that the NDA signed by the parties in 2015 “conceded that this information would no longer be confidential nor commercially sensitive” after a certain timeframe.
In response to the IPB’s arguments as outlined above, the applicant stated that she “rejected totally” what amounted to “mere bare assertions” by IPB. The applicant’s position was that all Irish local authorities are insured with IPB and the release of the records concerned would be “highly unlikely” to change that. She argued that it was not correct for IPB to assert that other insurers could, and in some cases already were, offering products to its members. Essentially, her view was that insurance products offered or provided by other insurers to IPB’s clients were for different areas such as own damage insurance for art works or insurance relating to compulsory Road Traffic Act liabilities. The applicant described IPB as a “monopoly underwriter for liability insurance for local authorities”. In summary, she was of the view that any assertions by IPB that it was operating in a general insurance market such that its competitive position would be prejudiced by the release of details of its business as set out in the records at issue were “disingenuous”.
In the previous case, the Commissioner noted the applicant’s position that information such as insurers' premium income, claims costs, expense ratios and commission payments can be found in Central Bank insurer returns and more detailed information can be found in certain forms which are filed with annual financial statements. In her submissions in this case, the applicant also contended that regulatory reports and independent rating agencies publish substantial information on insurers that provides insight into projections of loss ratios, levels of reserves, underwriting risk and pricing strategies. By way of example she provided a copy of Standard & Poor’s analysis of IPB in 2021, which I note referred to IPB’s “dominant position” and stated that it had a very high retention rate of “nearly 100%”.
In her submissions to this Office, the applicant argued that the summary report (record 7) provided no financial data that could be employed by a researcher to “explore the rationale for moving from self-insurance of Public Liability and Employer Liability … and to determine whether the move to ground up liability insurance delivered value for public money”, which appears to be a reference to Dublin City Council and its move to IPB in 2017. However, as this occurred after her FOI request in this case (made in 2016), any records relating to that decision would be likely to be outside the scope of her request in this case. Furthermore, while the applicant indicated that she is seeking the records at issue for the purposes of research, section 13(4) provides that I cannot have regard to an applicant’s stated or perceived motive for seeking access to records, except in so far as those motives reflect what might be regarded as public interest factors in favour of release of the information.
The applicant also referred to a number of examples which she considered to demonstrate a lack of transparency in relation to public expenditure on insurance generally and Dublin City Council specifically.
Essentially, the applicant’s view is that IPB is a service provider to the LGMA and that the information in the records concerned is controlled by the LGMA, an FOI body. Her position is that the liability cover provided by IPB does not go through a competitive tendering process and, on that basis, IPB is not in a competitive position in relation to other insurers in the context of liability insurance for Local Authorities and ETBs. She is also of the view that the type of information contained in the records is the kind of information which insurance companies are required to publish under various regulatory schemes.
As set out above, IPB identified specific information in records 3 and 6 which it considered to be highly sensitive and confidential, and contended that the release of this information could be prejudicial to its competitive position. It stated that even within IPB itself, this information is considered sensitive and access to it is limited. IPB also argued that the information contained in the records is above and beyond what insurers are required to publish under statutory and regulatory reporting obligations. While I am constrained from revealing the content of an exempt record, as noted above, IPB also contended that some of the information contained in the records concerned was “among the most sensitive data held by any insurer”.
IPB’s position is that the passage of time has not negated the sensitivity of this particular information, as it would reveal insights into its methodology, strategy and business which could be of benefit to its competitors. It argued that competitors could use this information to identify its most profitable customers and offer optimal rates to poach these customers from IPB. As I understand it, its position is that such competitors approach its members regardless of whether there are any relevant tender competitions.
The applicant, however, is essentially of the view that whether or not the information was commercially sensitive when it was provided in 2015, that it could not continue to be so at this point. She also refers to a lack of local authority tenders for insurance products. The applicant’s view is that as the public bodies who are members of the mutual insurance model scheme are not inviting tenders for insurance services, there are no opportunities for IPB’s competitors to tender for such business.
As set out above, the harm test in the second part of subsection (1)(b) of section 36 as to whether disclosure of the information concerned “could prejudice the competitive position” of IPB in the conduct of its business or otherwise in its occupation is very low. Although the onus is on the LGMA to justify its decision to refuse access to the records sought, I accept that IPB is in a better position to identify the relevant information and set out the expected harms to its business that may arise from release. In seeking to establish whether the requirements of section 36(1)(b) have been met, it is important to first look at the specific contents of the records themselves and what they do and do not disclose. In this regard, I consider that some information in the records may be very case specific or time-bound, but that other more general information might have broader applicability.
The remaining information identified by IPB as being particularly commercially sensitive includes what it has described as management information, such as ratios calculated on an Underwriting year basis, rather than on a Financial or Trading Year basis; information about its approach to market segmentation, pricing methodology and cross-subsidies; as well as other details of its business which it states it is not required to make publically available.
In the circumstances of this case, I am willing to accept that some of the information contained in the records is usually solely prepared for internal management reporting and is above and beyond what IPB is required to publish under the relevant statutory and regulatory schemes.
In addition, I understand that IPB Insurance primarily provides insurance services to a specific market, i.e. its members. However, I also understand from its publications that it additionally provides insurance services to non-members. I note that while the S&P report refers to a very high retention rate, it has stated that it is not 100%. While this implies that IPB’s business is stable with very little loss of repeat customers, there appears to be some loss. Furthermore, one of the tasks for the value for money review was to consider the option of seeking insurance cover in the open market. I am therefore willing to accept that IPB Insurance has potential competitors.
In the circumstances of this case, I am willing to accept that the release of some of the information IPB identified in records 3 and 6 could prejudice its competitive position. Accordingly, I find that section 36(1)(b) applies to certain information identified by IPB in records 3 and 6.
However, having carefully examined the records in question, I am not satisfied that the LGMA (or IPB) has adequately explained how the release of all of the information identified by IPB in records 3 and 6, or the information redacted from records 4 and 5, could reasonably be expected to cause the same harm to IPB’s competitive position. Furthermore, it seems to me that the information withheld from records 4 and 5 merely echoes similar information contained in records 7 and 8, which is already in the public domain. Having carefully considered the contents of the records and the submissions made by the parties in this case, I am not satisfied that section 36(1)(b) applies to the remainder of records 3, 4, 5 and 6 withheld from release by the LGMA. Accordingly, I find that the LGMA was not justified in refusing to grant access to the following information on the basis of section 36(1)(b) of the FOI Act:
• Records 4 and 5 in full
• Record 6: pages 51, 74, 82 and 84 in part, and page 93 in full (and the corresponding pages in record 3)
• Record 3: as above, and pages 5, 6, 7 and 38 in part (Phase 2 Report) and page 5 (Phase 3 and 4 report)
Having found that section 36(1)(b) of the FOI Act applies to certain information in records 3 and 6, I shall now consider sections 36(2) and (3) of the FOI Act in relation to this information.
Section 36(2) provides for the release of information to which section 36(1) is found to apply in certain circumstances. I am satisfied that none of the circumstances identified at section 36(2) arise in this case.
The applicant’s submissions
Essentially, the applicant was of the view that disclosure of all of the records sought was likely to reveal that there was no public procurement or competitive tendering process involved in the move to using the mutual insurance model and that such a move may have been based on “other reasons that are as yet not open to public scrutiny”. She contended that the “whole issue of liability claims and liability insurance is of immense specific public importance”, and was of “considerable public interest both to ordinary citizens and to the SME sector”. She argued that the Government had invested heavily in investigating these issues, including establishing a Cabinet Sub-Committee on insurance reform. She stated that her request relates to matters which “go well beyond the broad principle of openness”. The applicant stated that the CCPC found that competition in the Irish liability insurance market was “sluggish” overall and she said that IPB “certainly do not face normal competition from other liability insurers”.
The applicant provided a number of insurance industry-related reports and studies to this Office in support of her position. In particular, she referred to a 2020 Competition and Consumer Protection Commission (CCPC) Public Liability Market Study, which described IPB Insurance as “specialis[ing] in providing insurance for the public sector and complementary markets in the semi-state and private sectors”. The study also stated that IPB’s market share had increased from 15% to 22% during the relevant period. The applicant was of the view that this “belies the assertion by IPB that they are operating in the general insurance market overall”. She also queried whether an underwriter could “grow market share by 47% over 5 years if proper competitive procurement were invoked to ensure value for money that comes from the public purse”.
In her submissions to this Office, the applicant referred to questions raised at PAC in 2019 concerning the use of IPB to manage insurance and claims on behalf of the public sector rather than the State Claims Agency. She stated that, in her view, there still remained “many unanswered questions about accountability, transparency and efficiency in the expenditure of public funds in this context which is the focus of my research”. She argued that this was a “‘specific concern’ that ranges beyond the general policy on the broad principle of openness at s11(3) of the FOI Act”.
The LGMA’s submissions
The LGMA stated that PwC’s review found that IPB provided good value for money and that the mutual insurance model was fit for purpose. It also said that this information had been made available to the public by way of the publication of records 7 and 8 by the PAC. Its position, essentially, is that “the public interest has been met” by the publication of these records. However, I note that it has not specified what it considers that public interest to be.
IPB submissions
IPB referred to the findings of the Supreme Court in the Enet case as cited above. It also stated that the “overarching public interest” in relation to the records concerned was whether IPB is offering its public body customers a valuable service for the cost. Its position was that PwC’s assessment found that IPB was providing value for money and that the public interest in transparency and accountability in the context of public expenditure by the relevant public bodies has been met by the “comprehensive information” that is already in the public domain. It stated that the public would not “enjoy any additional benefit” from having access to specific pricing and other commercially sensitive information that PwC relied on to make its assessment. Its position was that this information was only of benefit to IPB’s competitors. IPB further stated that, essentially, the purpose of the FOI Act was to ensure the openness and accountability of public bodies in relation to public expenditure and government affairs, rather than to open up the affairs of private enterprise.
IPB noted that PwC’s Summary Report (record 7) provided a high level overview of Phases 1-4 of the review process. However, it was of the view that the Phase 4 Report, in particular, contained “largely unqualified commentary and speculative recommendations… arising from arbitrary opinions” on IPB’s strategic direction. In this regard, it stated that it was “unclear” what public interest would be served by the disclosure of this information, as it would not “assist the public in better understanding how public funds are spent and managed”. IPB also argued that it was difficult to see any public interest in the disclosure of this information when there was “a clear and obvious material harm to IPB” if such information was available to its competitors.
IPB also argued that if information outlining its methodology, approach to pricing and other supporting information was released in this case, its competitors could extrapolate IPB’s current and likely future pricing from the records, which could be deemed to be price signalling. Its position is that this would be a breach of EU competition law. IPB was also of the view that disclosure of the records sought would undermine its members’ ability to obtain information and conduct future such reviews, which it said would be contrary to the public interest.
Analysis
In considering where the balance of the public interest lies in this case, I have had regard to section 11(3) of the FOI Act which provides that in performing any functions under the Act, an FOI body must have regard to, among other things, the need to achieve greater openness in the activities of FOI bodies and to promote adherence by them to the principles of transparency in government and public affairs and the need to inform scrutiny, discussion, comment and review by the public of the activities of FOI bodies. However, in doing so, I have also had regard to the judgment of the Supreme Court in The Minister for Communications, Energy and Natural Resources and the Information Commissioner & Ors [2020] IESC 57 (“the Enet case”). In that case, the Supreme Court found that a general principle of openness does not suffice to direct release of records in the public interest and “there must be a sufficiently specific, cogent and fact-based reason to tip the balance in favour of disclosure”.
As a general principle, I accept that the FOI Act was not intended as a means by which the operations of private enterprises were to be opened up to scrutiny. For those records to which I have found section 36(1)(b) to apply, I have accepted that their release could prejudice IPB’s competitive position. I believe that the degree of potential harm is something to be considered when weighing up this public interest factor, and this will be dependent on the specific information contained in each individual record and the extent to which it could be used by competitors or others to the detriment of IPB.
Essentially, IPB’s position is that the public interest in ensuring the openness and accountability of public bodies in relation to public expenditure and government affairs is satisfied by the “extensive” disclosures made to date. It also argued that release of the records sought would give its competitors an unwarranted and inequitable insight into its activities by way of disclosure beyond regulatory requirements. Its view was that this would be “fatal” to competitive markets and contrary to the legitimate public interest in private commercial entities being able to engage with FOI bodies without suffering commercially.
I believe that for anyone doing business with a public body that is subject to the FOI Act, and where there are established rules around procurement and value for money, there should be no expectation that such business would be conducted in complete secrecy. In this regard, I note that IPB has not objected to the release of all of the information in the records concerned in this case. I also note that it has not objected to the release of some of the material identified in its own expert’s original report as being commercially sensitive. For instance, I note that the actuary’s report contended that page 78 of record 6 contained information relating to IPB’s solvency capital requirement for 2012 and 2013, which was not required to be published at the time (I understand that that the relevant Solvency II requirements did not come into force until January 2016).
I fully accept that there is a significant public interest in enabling democratic accountability and allowing for public scrutiny of public expenditure. Indeed, it seems to me that the disclosure of the price at which services are supplied to a public body is a significant aid to ensuring effective oversight of public expenditure, to ensuring the public obtains value for money, to preventing fraud and corruption and to preventing the waste or misuse of public funds. However, this does not mean that all such information should be released. The public interest in favour of disclosure has to be balanced against the harm which might be caused by disclosure.
I consider that there is a significant public interest in knowing how the mutual insurance model is working and whether it provides value for money. However, where the information of a third party and FOI body overlap, my Office aims to strike a balance between ensuring openness on the part of an FOI body and limiting the impact of disclosure on the affairs of a third party. In this case, it seems to me that by releasing the majority of the information in the records at issue whilst redacting certain identifying information, the LGMA has sought to strike a similar balance between allowing the applicant and the world at large an insight into how the mutual insurance model is working, whilst simultaneously seeking to ensure that the IPB’s competitive position was not unduly adversely affected.
As I have indicated above, the LGMA released records 1 and 2 in full and records 4 and 5 in part to the applicant, and records 7 and 8 are in the public domain. While I note that the applicant does not agree, I am satisfied that the availability of the records which have been released or otherwise made available to date could be said to address the public interest in allowing for public scrutiny of public expenditure, at least to some extent.
I accept that the release of additional information concerning the value for money review would address the public interest in release further. However, I am satisfied that the specific remaining information at issue in this case is far more detailed than the information that is publicly available. In my view, the release of this more detailed information could cause harm to IPB. In my view, it is not in the public interest to enhance transparency in public spending regardless of potential impact.
Having carefully weighed the competing public interest factors in favour of and against release of the relevant parts of the records at issue, I find that the public interest would, on balance, be better served by withholding the information concerned. I find, therefore, that the LGMA was justified in its decision to refuse to grant access to the specific information contained in records 3 and 6 as identified by IPB under section 36(1)(b) of the FOI Act, other than the information set out above, to which I have found section 36 not to apply.
The LGMA relied section 35(1)(a) of the FOI Act to refuse access to the records sought. I shall now consider this exemption in relation to the remaining information contained in records 3, 4, 5 and 6, which I have not already found to be exempt above. Section 35(1)(a) provides for the protection of information given to an FOI body in confidence.
However, before I consider the applicability of section 35(1)(a), I must consider whether section 35(2) applies. Section 35(2) provides that section 35(1) shall not apply to a record which is prepared by a member of the staff of an FOI body, or of a service provider, in the course of the performance of his or her functions, unless disclosure of the information concerned would constitute a breach of a duty of confidence that is provided for by an agreement or statute or otherwise by law and is owed to a person other than an FOI body or a member of the staff of an FOI body or service provider.
The LGMA stated that records 3 and 6 were created by PwC on behalf of the LGMA, and that PwC was acting as a service provider in this regard. I am satisfied that this was the case. I am also satisfied that record 4 and 5 were created by the steering committee, which consisted of two representatives of Local Government, one representative of the Education and Training Boards and one representative of the Office of Government Procurement. Therefore, I am satisfied that the records at issue were prepared by a member of the staff of an FOI body, or of a service provider, in the course of the performance of his or her functions.
The LGMA and IPB made arguments that IPB was not a service provider in the context of its provision of the information concerned, however, it seems to me that this is not necessarily the determining factor as to whether section 35(2) serves to disapply section 35(1) in the particular circumstances of this case.
Where an FOI body or a service provider prepares the relevant record(s), this Office must be satisfied that a duty of confidence is owed - by an agreement or statute or otherwise by law - in order for an exemption under section 35(1)(a) to apply to the records. The parties have argued that the duty of confidence in this case was owed to IPB by an agreement, that is, a non-disclosure agreement (NDA) signed by the parties concerned. The applicant’s position was that while she acknowledged that an NDA was in place, that such agreement “cannot be of interminable duration”, as this would comprise a “potential abuse” of FOI.
Regardless of any arguments by the LGMA or IPB, the NDA clearly states that the parties agree to be bound by its terms for six years from 8 September 2015. Therefore, I am satisfied that the terms of the NDA lapsed in September 2021. None of the parties have argued that the agreement was signed under circumstances which would invalidate its terms. Furthermore, neither the LGMA nor IPB has argued or provided any evidence that anything unexpected or unusual has occurred in the interim that was unforeseen when the NDA was signed. My understanding is that the parties freely entered into a mutual agreement on the understanding that the confidentiality agreement would last for a specified time span of six years, which has since passed. On that basis, I do not accept that a duty of confidence still exists on foot of the agreement. Therefore, I find that section 35(2) applies and that the LGMA has not justified its decision to refuse to grant access to the remaining information in the records concerned on the basis of section 35(1) of the FOI Act.
As I have found section 35(2) to disapply section 35(1), I do not need to consider the parties’ remaining arguments in relation to section 35(1).
Section 40(1)(d) is a discretionary exemption, which provides for the refusal of an FOI request if, in the opinion of the head, access to the record(s) concerned could reasonably be expected to result in an unwarranted benefit or loss to a person or class of persons.
Section 40(1) is a harm-based provision. Where an FOI body relies on section 40(1) it should identify the harm and show how release of the record could reasonably be expected to cause the harm envisaged, i.e. it should show the link between granting access to the record concerned and the harm identified. It should do this by reference to the specific record being considered for release: what is it about the particular record or the particular information in the record which, if released, could reasonably be expected to cause the harm envisaged? An FOI body’s submissions to this Office should be sufficiently detailed to demonstrate that link.
Section 40(2) sets out specific examples of the type of record that may be covered by section 40(1). The categories of records listed at 40(2) may qualify for exemption under section 40(1), provided that the relevant test in subsection (1) is met. Thus, an FOI body may invoke section 40(2) only in conjunction with one of the paragraphs of section 40(1) and it should show that the requirements of subsection (1) are met.
The LGMA has relied on section 40(2)(h) in this regard, i.e. it considered the records to relate to “proposals in relation to expenditure by or on behalf of the State or a public body including the control, restriction or prohibition of any such expenditure”. I am not sure that a value for money review of the mutual insurance model could strictly be said to relate to proposals in relation to expenditure by or on behalf of the State. However, while section 40(1) applies to the various categories of records set out in subsection 2, it is important to note that it does not apply to these categories exclusively.
The LGMA stated that while all insurers, including IPB, have specific statutory and regulatory disclosure requirements, certain types of information such as projections of loss ratio, are not subject to public disclosure. Essentially the LGMA argued that if the specific information contained in records 3 and 6 is released, IPB’s competitors would be in a position to strategically alter their own pricing in order to most effectively target IPB’s profitable customers. The LGMA was also of the view that this could then result in IPB no longer being in a position to offer services to its members on the same basis as it does currently. It noted that the current offering was found to comprise value for money by PwC. IPB made similar arguments that release of the records concerned would result in an unwarranted benefit to competitors and a loss to IPB and its members.
The LGMA further argued that the use of the information in the records concerned in this manner could jeopardise public bodies’ reliance on IPB for its services, which “would… constitute an unwarranted loss to the State” as it would be “difficult, costly and/or high risk” for these State bodies to source alternative insurance solutions on the same basis as those currently provided by IPB.
I note that the LGMA seems to be arguing on the one hand that IPB’s competitors will undercut its prices to attract IPB’s current customers and on the other hand that public bodies will find it harder to source alternatives to IPB. Taking its first argument a step further, it seems to me that if IPB’s competitors used the information in the records to poach its customers, it would be likely to do so by offering a better or cheaper alternative, at least initially. In those circumstances, while such an outcome could of course, cause harm to IPB’s business, it seems to me that it could arguably also result in a benefit to the State.
In any event, as set out above, I have already found the majority of the specific information contained in records 3 and 6 identified by IPB as particularly commercially sensitive (and presumably of most interest to its competitors) to be exempt from release under section 36(1)(b). In the circumstances, and having carefully considered the matter, I am not satisfied that the release of the remaining information in the records concerned would be reasonably likely to cause the harms envisaged by the LGMA and IPB in relation to section 40. Therefore, I find that the LGMA has not justified its decision to refuse to grant access to the remaining records in part on the basis of section 40(1)(d) of the FOI Act.
IPB stated that section 37 applied in relation to particular information relating to a specific claim contained on page 68 of record 6. In submissions to this Office, it stated that this part of the record identified an individual and referred to a fatal accident. While no individuals are named, I note that page 68 refers to a specific incident in 2007. I also note that page 69 refers to a number of other incidents in 2015.
In her submissions to this Office, the applicant stated that although it was difficult for her to comment without sight of the record concerned, she was not seeking the names of any identifiable individuals contained in the records. She stated that she was solely concerned with statistics and hard data.
Section 37(1), subject to other provisions of section 37, is a mandatory exemption, which provides for the refusal of access to a record containing personal information. I am satisfied that the information concerned relates to the employment or employment history and/or the medical history of identifiable individuals. In the circumstances of this case I am satisfied that it would be a reasonably simple matter for the individuals concerned to be identified if the record was put in the public domain. Therefore, I find that section 37(1) of the FOI Act applies to this information.
Section 37(2) provides that section 37(1) does not apply in certain circumstances. I am satisfied that no such circumstances arise in this case and that section 37(2) does not, therefore, apply.
Section 37(5) provides that a request that would fall to be refused under section 37(1) may still be granted where, on balance (a) the public interest that the request should be granted outweighs the right to privacy of the individual to whom the information relates, or (b) the grant of the request would benefit the person to whom the information relates. In considering section 37(5), I consider that only section 37(5)(a) is relevant in this case. This section provides that a request that would fall to be refused under section 37(1) may still be granted where, on balance, the public interest that the request should be granted outweighs the public interest that the right to privacy of the individuals to whom the information relates should be upheld.
Both the language of section 37 and the Long Title to the FOI Act recognise a very strong public interest in protecting the right to privacy (which has a Constitutional dimension, as one of the un-enumerated personal rights under the Constitution). Unlike other public interest tests provided for in the FOI Act, there is also a discretionary element to section 37(5)(a), which is a further indication of the very strong public interest in the right to privacy. Privacy rights will therefore be set aside only where the public interest served by granting the request (and breaching those rights) is sufficiently strong to outweigh the public interest in protecting privacy.
In considering where the balance of the public interest lies in this case, it is important to note that the release of records under FOI is regarded, in effect, as release to the world at large, given that the Act places no constraints on the uses to which a record released under the Act may be put. I am satisfied that the disclosure of the identifying information in question would provide minimal additional insight into the LGMA’s management of the mutual insurance model or the value for money derived from IPB’s offering.
I am satisfied that there is no sufficiently specific, cogent and fact-based reason for finding that the public interest that access should be granted to this particular information outweighs the right to privacy of the individuals to whom the information relates. I find, therefore, that section 37(5)(a) does not apply.
In conclusion, I find that the information concerned is exempt from release under section 37(1) of the FOI Act.
I direct the LGMA to release the records at issue to the applicant as follows:
Records 4 and 5 in full and records 3 and 6 in part, as follows:
Record 6:
• Redact pages 2-13, 41-46, 52-61 and 64-66,
• Page 51 – release first 2 paragraphs only
• Redact the content of pages 73-74 of record 6, except the table on page 74
• Redact the bullet pointed examples and the reference to a specific building in the first paragraph relating to Continuity of Cover Year on Year on page 82
• Redact the final paragraph “We note... sector” on page 84
Record 3 Phase 1 report
• Redact Pages 35-40
Record 3 Phase 2 Report
• Page 5 , release except the second and third paragraphs under Pricing: Target Combined Operating Ratio; both graphs; and the sentence beginning “at 31 December 2014” under Reserve adequacy and Prior Year run-off.
• Page 6 , release except paragraphs 2-5 of Target Combined Operating Ratios; all of the content headed “The pricing process”; and the second sentence and paragraphs 3-4 under the heading “Cross subsidisation”.
• Page 7 , release except: all of the content under the heading “Cross subsidisation between classes”; paragraphs 2, 4 and 6 under “Historical premium rate changes”; all of the material under the heading “Reserve adequacy: Prior year reserves run-off” except the first sentence; and the second paragraph under the heading “Explicit additional reserves”.
• Redact the content of pages 37-38 , except the table on page 38
• Redact pages 16-25 and 28-30
Record 3 Phases 3 and 4 report
• Redact the bullet pointed examples and the reference to a specific building in the first paragraph relating to Continuity of Cover Year on Year on page 9
• Redact the final paragraph “We note... sector” on page 11
Although each case is considered on its own merits, it may be useful to distinguish the circumstances and findings on section 36 in this case from those in Case No 170136. In considering the public interest, it should be noted that the Senior Investigator’s decision issued before the Enet judgment and did not have the benefit of its guidance as summarised above.
Since this Office’s decision in that case, some of the records concerned have been published online. Furthermore, as noted above, the IPB made substantive submissions concerning particular information contained in the records which it considered to be commercially sensitive. This information was not before this Office in the previous case.
Having carried out a review under section 22(2) of the FOI Act, I hereby vary the LGMA’s decision. I affirm its decision to refuse access to records 6 (in part) and 7 and 8 (in full) on the basis of section 15(1)(d), as they are already in the public domain. I affirm the LGMA’s decision to refuse access to certain information contained in records 3 and 6 on the basis of section 36(1)(b). I find that that public interest, on balance, does not favour its release. I affirm its decision to refuse to grant access to additional information contained in record 6 on the basis of section 37 of the FOI Act. I annul the LGMA’s decision to refuse to grant access to the remaining information in the records on the basis of sections 35, 36 and 40. I direct the release of records 4 and 5 in full and records 3 and 6 in part as set out above to the applicant.
Section 24 of the FOI Act sets out detailed provisions for an appeal to the High Court by a party to a review, or any other person affected by the decision. In summary, such an appeal, normally on a point of law, must be initiated by the applicant not later than eight weeks after notice of the decision was given, and by any other party not later than four weeks after notice of the decision was given.
Ger Deering
Information Commissioner