Commissioner Justice Owen adopted two key recommendations (16 and 35) based on Prof. Zehnwirth's submissions to the HIH Royal Commission.
The following are extracted from the Commissioner’s report.
Pages 190-191.
Actuarial methodologies
Responsibility for the selection of an appropriate valuation model or methodology lies with the actuary. As I have mentioned, Guidance Note GGN 210.1 provides some guidance.
Do not propose to consider the relative merits of the various methodologies of which I have become aware during the course of the Commission. However, I do wish to make some more general observations. They arise in part at least from the submissions of Prof. Ben Zehnwirth, who was critical of the methodologies adopted by a number of actuaries, including those advising HIH. In essence his criticisms were directed towards the over reliance upon subjective judgment (for example, in determining an appropriate allowance for future claims inflation) at the expense of reliance upon properly constructed probabilistic models and analysis of the trends and variability found in historical experience.
The actuarial science is still developing. It may well be that over time a greater consensus is reached as to the merits of particular methodologies and models for particular types of insurance business. But it seems to me that there will inevitably remain a need for the exercise of subjective judgment on the part of the actuary. To me the concern is not so much with the existence of subjective judgment as with the identification of the judgments that have been made and their impact upon the valuation.
From my consideration of HIH’s actuarial valuations in Chapter 15 it is clear they involved a significant level of subjective judgment—for example, in selecting rates of superimposed inflation which were inconsistent with the historical experience, and in making assumptions about the impact of particular changes in the business being written. HIH’s actuary referred often to the past experience for particular lines of HIH business not being a useful indicator of the future. But often I found it difficult from a reading of the actuarial reports to discern precisely what subjective judgments had been made, let alone the impact of those judgments upon the valuation outcome.
In my view the usefulness of the actuarial advice provided to general insurers would be enhanced by clear disclosure of the existence and impact of such judgments and departures from historical experience. The resulting transparency of the actuarial advice should assist readers (including the board, auditors and APRA) in identifying, testing and assessing the appropriateness of such judgments.
Recommendation 16
I recommend that the Institute of Actuaries of Australia and the Australian Prudential Regulation Authority introduce a requirement for more detailed disclosure of the exercise, incidence and impact of subjective judgment and departure from historical experience.
Important areas for greater disclosure
Two types of information are particularly important in assessing the financial condition of an insurance company. These relate to the accuracy of an insurer’s outstanding claims provisions and its reinsurance arrangements. Inaccurate estimation and reporting of these matters played a major role in the failure of HIH. Measures to improve their accuracy should therefore be encouraged.
Swiss Re discussed both of these issues in its submission:
Swiss Re believes that a lack of transparency destroys shareholder value because the market will discount what it does not understand. For insurance companies, there are two key areas on non-transparency:
• Reserving—the setting of reserves for future claims is subjective and this will always be the case. Changes in assumptions can have a large impact on a company’s reported position and generally it is very hard to tell whether a company is appropriately reserved—either under or over; and,
• reinsurance—the reinsurance programmes of many companies have evolved over a number of years. It is common for insurance companies to have series of reinsurance contracts, each of which is split between many reinsurers of varying credit quality. Understanding the true financial position of an insurance company is difficult because the overall effect is, at best, opaque.[30]
Swiss Re recommended that insurers should be required to publish information on past claims payments that is sufficient to allow outsiders to assess the quality of reserves. Swiss Re also recommended that insurers should be required to publish a reinsurance resume including broad details of key programmes together with limits and terms of coverage and the names of reinsurers.
In regard to outstanding claims provisions, Prof. Zehnwirth suggested that APRA should adopt the US model of collating what is referred to as ‘schedule P’ loss development data (triangulations) and other relevant data on a yearly basis for each line of business for each company.[31]
Recommendation 35
I recommend that information that enables external users to make an informed assessment of an insurer’s outstanding claims provisions and reinsurance arrangements be published by the insurer or the Australian Prudential Regulation Authority. APRA should develop reporting returns for insurers that would enable this to occur if existing returns are insufficient.
In particular, general insurers should publish:
• material equivalent to the ‘schedule P’ loss-development data published in the United States
• a summary of the approved actuary’s valuation of the outstanding claims liabilities, including the methodologies and assumptions underlying that valuation.