In discussing the regulatory outlook today I will cover:
- the new GI regime and what it means for insurers
- some further improvements that we at APRA envisage because you can always improve and from the reauthorisation exercise we have identified some areas where it would be useful to do that
- how APRA will go about its supervision job under the new GI regime
- a new internal risk rating system called PAIRS that we have developed in APRA; and
- what I see as current and upcoming issues at the international level.
A major component of the new GI regime is APRA’s new Liability Valuation Standard which requires a prudential margin to be included so that technical provisions are estimated to be sufficient 75% of the time, and it requires the insurer to get expert advice from its Approved Actuary, who must comply with APRA and professional requirements in doing their calculations and giving their advice. Boards of insurers ultimately decide the level of provisions and they can override the Approved Actuary’s advice, but they will have to report and defend such a decision to APRA and the market. These requirements will lead to prudent and consistent estimation of the largest and most important item in a general insurer’s balance sheet.
Insurer’s also need capital to withstand unexpected and extreme shocks, whether they be natural disasters, economic or financial market downturns, legislative changes or competitive pressures. So APRA’s Capital Adequacy Standard deals with the types of allowable capital and the levels of capital insurers must hold given the risks in their business. Insurers are now allowed to have a mix of Tier 1(i.e. equity type) capital and Tier 2 (i.e. subordinated debt type) capital, so they will have greater capital management flexibility. The minimum level of capital insurers must hold is driven by the types of insurance lines they write, with long tail and inwards reinsurance risks requiring higher capital than short tail lines; the amount and concentration of risk they retain and the quality and concentration of their reinsurance recoveries; and the investment or asset risks they have, with lesser quality and volatile investments, or investment concentrations requiring higher capital.
APRA’s Standard on Risk Management includes numerous corporate governance requirements ranging from the fitness and propriety of directors, management, auditors and actuaries; to board composition; and through to the types of risk we expect insurers to have identified in their operations, how they assess those risks and the limits and controls they have in place to manage those risks.
Reinsurance is crucial for most general insurers as it allows them to serve customers by taking on large individual risks or concentrations of risk in certain business lines or geographic areas, but to lay off some of that risk to reinsurers. These reinsurers have larger risk pools through which they can diversify those risks and they have larger amounts of capital to withstand the extreme claims. APRA also has a Reinsurance Management Standard which requires insurers to have a documented reinsurance management strategy which is approved by the insurer’s board; identifies the insurer’s risk appetite via the type and level of risks it wants to write and retain; specifies how the insurer will manage the risks it wants to lay off; and what type and level of reinsurance arrangements the insurer will put in place.
All general insurers had to be reauthorised, so APRA could assess if they would meet all the new requirements and thus be in adequate shape to continue writing business. We had a form of industry health check whereby insurers provided much material to APRA about their operations, including business plans, capital adequacy calculations, risk management statements, reinsurance management strategies, auditor and actuary details. Some insurers were inadequately capitalised or were otherwise unable to satisfy APRA’s authorisation criteria. Most have since received capital injections or improved their risk management. Ten insurers have gone into run-off and another 15 general insurers have transferred their liabilities and exited the market during the reauthorisation period. In addition, APRA imposed formal conditions on 9 general insurers that have capital transition plans, which allow them to bring their capital positions into line with APRA’s requirements over the next 2 years. The initial governance material provided by companies in their applications for re-authorisation was commonly substandard, and many companies had to substantially re-work their Reinsurance and Risk Strategies to bring them up to scratch.
There are new returns which general insurers will need to submit to APRA under the new regime. An overview and details of these can be found on APRA’s website. The 2 main points I want to convey today are that the new APRA returns have a different accounting basis to that required by AASB 1023 (but there are not significant data capture issues for insurers) and the new returns commence from the Sep 2002 quarter. We expect some teething issues but 2003 should see them resolved.
The new prudential framework that we have put in place significantly strengthens the general insurance regime and brings it more into line with international best practice, although it is setting the pace in some respects and is being adopted by other countries. However, this job is one which is never finished and we always expected that future changes would be needed as we discover implementation issues and the market keeps on developing. The new framework with APRA Prudential Standards subject to Parliamentary disallowance was chosen to allow for such changes to be timely.
Some fine-tuning items already on our agenda include:
- changes to capital factors to improve the risk-sensitivity of the framework;
- a review of the investment concentration capital charge;
- more detailed guidance on the content of Risk Management and Reinsurance Management Strategy documents;
- additional guidance on the calculation of the Maximum Event Retention; and
- a review of the definition of ‘inside Australia’ assets.
Historically insurance supervision has focussed on each licensed insurance company to ensure it is adequately managed and financially sound so it can be ‘ring-fenced’ from problems elsewhere in its wider corporate group. The new GI regime still operates essentially on this basis although we do now have the power to authorise and supervise non-operating holding companies. With more complicated financial group structures, increasing globalisation of many insurers and quicker transmission of information and thus problems around the globe it is more difficult to deal with contagion problems so we want to consider how we can practically supervise insurers at both the stand alone entity level and the wider group level. Group supervision would deal with issues such as double counting of capital, intra-group exposures, group-wide risk management and product badging.
Prudential supervision systems around the world have for sometime included specific requirements on capital and risk management and active supervision and visits by the supervisor. But now they are also beginning to include specific requirements to disclose company specific prudential information in order to make market disciplines act for the prudential supervisor. Quantitative data such as capital levels and coverage of the MCR, exposures by line of business, breakdowns of reinsurance counterparties, the historical development of claims reserves, and qualitative information on governance and risk management processes are being considered and some already exist in other countries. We will be looking at the practicality of such disclosures over the coming year.
We are concerned that in the current environment of increasing premiums small business, professional associations and community groups are seeking lower cost alternatives and may turn to unregulated providers without fully understanding the consequences. Unauthorised foreign insurers can distribute insurance in Australia through insurance brokers and registered foreign agents, but there must be clear disclosure to the consumer that they are not regulated in Australia. We have been warning consumers that they should check APRA’s website for insurers authorised in Australia and check if the foreign insurer is actually authorised and regulated in its home country, as recently there have been 2 instances where this was not the case.
Discretionary mutuals are mutual aid schemes in which members pool resources to provide coverage to their membership on a ‘discretionary basis’. They are not authorised under the Insurance Act because there are no contractual terms and conditions governing the payment of a claim, so claims are only paid if there is sufficient funds at the time. While such schemes appear to work effectively they are generally unstable and usually unsustainable in the long run because all the risks, including the extreme ones are often not properly allowed and funded for. They have strong incentives to keep subscriptions low rather than adequately fund for all the risks.
Ongoing APRA supervision obviously includes monitoring companies operations and financial health via analysis of returns and on-site visits. Recently we have increased supervision resources within Diversified Institutions Division (my division) by about 15 people to create a 4th supervision branch and spread the workload to allow closer scrutiny by more senior staff. We have also added a specialist Insurance Risk Team to our internal Consulting Services area so we can conduct more detailed on-site visits looking at insurer’s underwriting, pricing, claims management, claims reserving and reinsurance processes to see that insurers are in fact doing what they said they would do in their Risk and Reinsurance Management Statements.
APRA’s supervision will remain risk-based in focussing more resources on the more risky institutions, but in will involve more drilling down, or ‘tyre kicking’, on specific issues as opposed to showroom shopping.
In order to best allocate our scarce supervisory resources, we internally rate all the institutions we supervise and we have spent about 12 months upgrading this rating system into a new one called the Probability and Impact Rating System, or PAIRS.
This PAIRS system is more sophisticated and objective than our past one. Importantly, it takes account of the impact of a regulated entity’s failure, as well as its probability of failure. We combine the probability and impact measures to create a single numerical Threat Index.
The probability component of the rating system is a systematic and structured measure of each entity’s inherent risk, minus the mitigating effect of its risk controls and capital support to arrive at the net risk. We consider an entity’s risk profile and level of risk by considering risks such as asset quality risk; market value movement risk; insurance risk; operational risk, liquidity risk; legal & regulatory risk; strategic risk and contagion risk. We then consider carefully the quality of management and internal control systems which can work to reduce these inherent risks. Finally we look at the capital surplus, profitability and parental or outside capital support that is potentially available to deal with problems if they occur.
The impact component of the system is currently based on size, measured by assets - but we will be incorporating more sophisticated measures of impact in the next generation of PAIRS.
The composite Threat Index is a non-linear measure of the "threat" an entity poses to its stakeholders and the financial sector as a whole. As probability of failure and impact of failure rise, the composite Index goes up exponentially.
We won’t be disclosing these ratings publicly, but regulated entities will be in no doubt a how we view them as we progressively rate all entities using PAIRS. While it will take us about two years to re-rate all 3,000 or so entities we supervise, the vast majority of the larger institutions including the insurer will be re-rated within months.
Once we have determined the PAIRS risk assessments and rating for an entity it will drive our supervisory stance and strategies towards that entity. We will be using 4 supervisory stances. In ‘normal’ or ‘routine’ mode we will be collecting and analysing regular returns, and making routine visits. For an entity in ‘oversight’ or ‘heightened’ supervisory mode we will be collecting greater amounts of information, asking more detailed questions, conducting detailed and more frequent visits, speaking to auditors an actuaries, and possibly commissioning special work by external experts to understand in more depth how the company is dealing with what we regard as higher than desirable risks and to ensure the company’s management and board are aware of our heightened interest. Where relevant we will lift minimum capital requirements to compensate for higher than desirable risk. Where we decide an entity is in ‘mandated improvement’ or ‘intense supervision’ mode we regard them as operating in an unsustainable way and we will be requiring them to quickly develop and implement a business plan which rectifies the concerns we have and restores their operations to what we regard as sustainable. In essence we will be saying you can not continue as is, but we will give you the chance to effect your solution otherwise we will impose one on you. Where we regard entities at risk of near-term failure our supervisory stance will shift to ‘restructure’ or ‘protective’ mode where we will be using our formal enforcement powers to restructure a companies operations and protect beneficiaries (such as policyholders) interests. Such action could involve us replacing people, freezing assets, directing certain business lines to be stopped or exited and ultimately applying for wind-up.
There are a number of live and somewhat difficult issues being worked on at the international level.
The International Accounting Standards Board has a project underway to develop an international accounting standard for the accounting of insurance contracts. This is a large and extremely complex project which is reconsidering the fundamentals of insurance accounting and is currently aimed at general purpose financial reporting so stock markets are well informed, rather aimed at reporting for prudential regulatory purposes. The IASB is proposing a ‘fair or market value’ style of accounting rather than a book value style, but is also focussing on asset/liability driven approach where profit is determined from the change in the balance sheet, rather than a deferral and matching of income and expenditure approach where the balance sheet is determined by changes in profit or loss. In some senses this is revolutionary for many countries and while it is less so for Australia there would still need to be major adjustments locally.
Reinsurance is a global business and especially post Sep 11 questions are being asked worldwide about the financial strength of some reinsurers and the potential for problems in the reinsurance market to flow into problems in the whole financial system, e.g. because some reinsurers are currently via credit derivatives taking on some of the credit risk of banks. The Financial Stability Form (ie mostly heads of Treasury Depts and central banks from major countries) only last week said that while the reinsurance industry had coped with recent shocks it is difficult to assess the impact of problems in that sector on the whole insurance industry and on financial stability generally. So the FSF called for improved reporting on, and disclosures by, reinsurers and the FSF strongly supported the efforts of the International Association of Insurance Supervisors to develop an efficient global framework for the supervision of reinsurers, which are currently not supervised in some countries.
The IAIS, which sets non-mandatory standards for insurance supervision, is actively working on developing a solvency and capital adequacy framework or standards which could be applied around the world. Varying insurance products, legal systems, cultures, accounting approaches and even tax systems mean this is complex, but none the less the IAIS sees it as crucial. High level principles have already been developed and the IAIS work program over the next 2 years includes the production of 9 separate guidance papers or standards on solvency and capital adequacy.
Electronic commerce is also an area of high interest to insurance supervisors within the IAIS and work on the risks posed to insurers by E-commerce and principles for managing those risks is already underway.
I earlier mentioned APRA is looking at the practicality of requiring insurers to publicly disclose prudential information to inform the market and consumers and thereby impose an extra discipline on the insurers. The IAIS is also keenly interested in this issue and has a taskforce working to develop standards on disclosure of each of technical (i.e. underwriting or insurance) performance and investment performance. I expect that these will cover more than simple loss and operating ratios, but will also cover risk and investment exposures by type and size, so a proper assessment of the performance against the risks run can be made.
So where are we 12 months on from September 11?
I think it is obvious that the world we live in today is more uncertain than that of the past. Changes happen more quickly and the extent and impact of these changes or events, whether natural or man made, is larger than previously. This is either because the events or changes are in themselves more dramatic (such as Sep 11) or their impact ripples more than before. With instant communications we live in a ‘smaller’ world as the knowledge of events transmits quickly and widely, with many more people considering the event and how it affects them. This is clearly true of investment markets where international events impact traders views and market values of investments virtually instantaneously. With Sep 11 we saw share market drops around the world, and for a short while in Australia we saw life insurers and funds managers have to suspend investment redemptions because they could not determine accurately enough exit prices for many of their overseas investments. This is one simple example of the increasing interconnectedness of the world and financial markets of today.
It is in this type of world that prudential regulators are looking to set benchmarks for financial institutions risk management practices and systems and to get institutions to improve their actual practices. This will mean they are more prepared to deal with the increased risks they may have to face and they think about the secondary and lower order knock on effects that certain events may have on them rather than only the obvious primary risks they face.
Regulators will always do some form of financial analysis of the health of financial institutions, where it currently stands and how it is trending. But that is increasingly being supplemented with assessing the adequacy of institution’s risk management systems and if they are in fact working in practice. It is a move away from supervision out of the rear view window where we looked at whether institutions met regulatory solvency requirements at a point in the past to supervision out the front windscreen where we assess where institutions are headed and how well placed they are to cope with a variety of unexpected or extreme shocks which may come.
Prudential supervision is, and always has been, more an art involving fine judgements than only a mechanical application of specified rules. Such rules set a framework within which financial institutions should operate and if they are designed in a risk-sensitive way they can provide a good set on incentives for institutions to follow prudent behaviour. But the rules can never deal with each and every situation or problem that may emerge, so they must be complemented by supervisor’s judgement.
We believe the new GI regime and the changes to APRA’s supervisory processes are significant improvements and should work to reduce failures of insurers, provide policyholders with greater protection and improve confidence in the insurance industry.