Office of the Superintendent of Financial Institutions
In February 2012, OSFI published Draft Guideline E-15 – Appointed Actuary: Legal Requirements, Qualifications and Peer Review. OSFI received 17 submissions from industry associations, companies and individuals following the release of the draft. I would like to thank everyone who provided comments and suggestions.
Guideline E-15 was originally issued in 2003. The concept of external peer review, introduced in this Guideline, has generally been accepted over the last eight years by insurance companies, Appointed Actuaries and the actuarial profession. However, it is OSFI’s opinion that the peer review process can be enhanced to better assist OSFI with its review of insurance companies.
In addition, OSFI has recently changed its Supervisory Framework. One of the changes was a revision in OSFI’s view of how it makes use of the role of the Appointed Actuary. A key component of this change was that the effectiveness of the work of the external peer reviewer needs improvement.
OSFI reviewed the submissions and had follow-up meetings with some groups. Today, OSFI is publishing the final version of Guideline E-15, as well as the Annex to this letter, which provides a summary of public comments received and an explanation of how they were dealt with in the final Guideline.
The revised Guideline is effective for the financial statements covering 2013, and for the DCAT prepared during 2013. The review of 2012 year-end financial statements and DCATs presented during 2012 can follow the previous requirements.
However, where possible, insurers are encouraged to comply with the principles and expectations of this Guideline as of the date of this letter.
There were some questions about the new methodology OSFI has been developing to assess the actuarial function in insurance companies and how this fits with peer review. There was also some question about the role of OSFI’s Actuarial Division as compared to the proposed role of the peer reviewer. Commentators did not always see how this fits with the educational role of peer review
The description of OSFI’s objectives has been more clearly articulated. OSFI’s prime concern is the safety and soundness of insurance companies. From OSFI’s point of view, the prime objective of the revised peer review guideline is to assist OSFI in its assessment of the insurer’s safety and soundness, and the role played by the insurer’s actuarial function within the company. This objective was not clearly indicated in the previous version of the guideline.
A second objective of peer review remains to provide a source of independent consultation advice and education to the appointed actuary (AA).
Several commentators indicated that much of the work OSFI expects the peer reviewer to undertake is already being done by the external auditors, thus leading to duplication of effort and cost. It was noted that the auditors employ actuarial specialists as part of the audit team and that they check to see that the valuation assumptions are within standards.
The CICA standards state that the overall objective of the auditor is to obtain reasonable assurance about whether the financial statements, as a whole, are free from material misstatement.
However, OSFI’s objective and the scope for peer review is to assess the safety and soundness of insurers by reviewing the AA’s work for the financial statements at a more granular level. It is OSFI’s view that each of the assumptions used should be independently reasonable, in accord with accepted actuarial standards, and that the methodology should be appropriate for each valuation model.
It is not OSFI’s intention for peer review to duplicate the work of the auditor. For instance, the peer reviewer is not required to perform any detailed recalculations, since this is part of the audit.
Some commentators questioned the preference the guideline gave to having the peer reviewer be independent of the external audit firm. It was noted that the audit firm is independent and its actuaries are qualified to do the peer review work. Also, using the actuary of the audit firm would be less costly and burdensome for the insurance company.
OSFI agrees that an actuary working for the external audit firm is sufficiently independent to be a peer reviewer. The guideline’s wording has been changed to explicitly state that using an actuary from the external audit firm can accommodate smaller and simpler companies.
Notwithstanding this, OSFI expects large and complex companies to engage a peer reviewer who is not a member of its external audit firm. While OSFI recognizes that external audit firms are independent, it is of the view that separate independent actuarial peer review is desirable to provide additional perspective to large and complex companies.
Several commentators questioned the usefulness of the education component of peer review, particularly in the case of companies having internal resources more extensive than those of typical peer reviewers.
While it is true that the large companies have extensive resources available, OSFI is of the view that it is still useful for the AA from a large company to have a source of independent consultation advice from a third party outside the company. The smaller companies will have more use for advice from the peer reviewer on accepted practice in the industry and on improvements to their work. In any case, peer review is not the only source of professional development for the AA.
Several commentators expressed concerns about the timing of the peer review of the Appointed Actuary’s Report (AAR) in the case of pre-release peer reviews. The AAR is usually not prepared in time for a review in the same timeframe as the Life-1 type of statement.
The section of the guideline dealing with the timing of reporting has been revised to recognize such timing challenges. It now recognizes that there may be more than one peer review report. For instance, a separate report on the contents of the AAR can be submitted subsequent to the peer review report on the financial statement valuations.
Some commentators proposed that the peer review requirements should vary by company. They should be based on the size and risk of the company. There were also suggestions that if an external consulting actuary was the AA, then there need not be an additional peer review.
OSFI believes the basic peer review requirements should be consistent for all companies. The peer reviewer is expected to consider materiality and the circumstances of the company in determining the appropriate level of work to be done. Where an external consultant is the AA, a separate peer review is still required.
Some commentators noted that more guidance should be provided about the level of materiality to be used for the peer review work. This was especially noted for the required annual review of material changes in valuation methods and assumptions.
The guideline has added a new section on materiality. A list of examples of what constitutes a material change in the valuation of policyholder liabilities has been added to provide further clarity.
The guideline does not specify a quantitative materiality level, but relies on the professional judgement of the AA and the peer reviewer in the particular circumstances. The CIA standards of practice state that materiality should be judged from the point of view of the prime user of the work. OSFI’s expectations are at a more granular level than the external audit view of materiality. As such, the guideline states that materiality should be set to be appropriate at the line of business level at which the assumptions are set.
The draft guideline included the reference that peer review should include an examination of the process the company uses for stress testing (Guideline E-18) and Own Risk and Solvency Assessment (ORSA). Some commentators pointed out that these two issues are not in the appointed actuary’s domain since there are no professional standards covering them.
OSFI has accepted this suggestion, and the references to E-18 and ORSA have been removed. The peer reviewer is still expected to assess the DCAT.
Some commentators expressed concern that the draft guideline seemed to require the DCAT to have a full peer review each year. This was considered to be excessive.
This was not the intention of the guideline. The wording has been revised to make it clear that there should only be a full review of the DCAT every three years. The yearly review should only be a brief review with the AA regarding the basis of the scenarios to be used, and not the mechanics of the actual production of the scenario results.
Some commentators expressed concern about the reference to the peer review report having to include a summary of the discussions between the reviewer and the AA, and the resolution of any disagreements. This could inhibit a full discussion if this had to be disclosed to OSFI.
This is a valid point, since the aim of peer review is to provide the AA with a source of independent consultation advice. As such, the requirement to discuss any disagreements has been removed.
Commentators suggested numerous other small changes to specific wording.
Many of the suggestions for specific wording changes were very useful and have been incorporated in the revised guideline.