Views from APRA

Ian Laughlin, Executive Member - Insurance Council of Australia Regulatory Update Seminar, Sydney

Thanks for the opportunity to talk with you today.

When you are asked to speak at a meeting like this, and you start thinking through what you might say, inevitably you reach a point where you ask yourself how you can make the speech more interesting.

And when you are talking about regulation, this can seem a bit of a challenge!

So I was scratching my head for a while, and then I realized that this wasn’t any old industry conference.

This is a meeting about regulation. And you have all come here voluntarily.

So you must love regulation for what it is!

So once I understood that I relaxed. I didn’t worry about making it interesting at all, and just got on with the job.

Here’s what I will talk about:

Introductory Comments

I’ve been at APRA about 8 months now – enough to give me a good feel for the job and a much better feel for the GI industry.

And there is plenty going on.

We have the LAGIC capital review, the Basel III developments in banking which will have some flow-on effects to the GI world, a number of developments in insurance regulation internationally, the ongoing development of risk management practices, and the bedding-down of APRA’s remuneration requirements to name a few.

And we have had the extraordinary run of catastrophes, which has stretched the industry on various fronts.

So it’s timely for me to be talking with you.

The Catastrophes

Let me start with the catastrophes.

What an extraordinary period we have had!

As you know, earlier last year we had storms in Perth and in Melbourne, but more recently we have had floods in Queensland and Victoria and parts of NSW, Cyclone Yasi and the Perth bushfires and the devastating earthquake in Christchurch.

These events have had and will have a profound impact on the industry – on operations, on staff, on profitability, on financial strength and of course on reinsurance.

Based on data received directly from a number of general insurers and from the Insurance Council of Australia, the gross insured claims from the recent events for the industry in Australia are likely to be about $4 billion.

These are estimates only at this stage and it will be some time before the true insured claims will be known. Of course, actual losses sustained by the Australian community from these events have been much larger, but not all the losses are covered by insurance.

Only a portion of the gross losses will need to be met from the direct resources of APRA-regulated general insurers, because each direct insurer has itself insured against the risk of these events via its reinsurance arrangements. These programs, which are typically placed with large international reinsurance companies, allow for the claims to be shared beyond Australian shores.

The Christchurch earthquake will result in further claims on the general insurance industry in Australia but, at this early stage, it is difficult to quantify these claims. Gross claims may prove to be larger again than the claims from the recent natural disasters suffered in Australia, but initial feedback from the general insurers affected by the earthquake is that their net claims will not be extremely large. Again, the bulk of the claims will be borne by their reinsurers. At this early stage estimates are around NZ$12 billion in total losses to the community (insured losses being much lower).

We are monitoring developments in Christchurch very closely.

APRA’s Requirements for Large Events; Reinsurance

The resilience of our general insurance industry in these circumstances is a credit to management and boards.

At least in part, it is also reflective of APRA’s prudential regime and, in particular, the strengthened risk-based capital requirements introduced in 2002. These capital requirements are currently being revisited as part of our LAGIC review of capital standards.

One of the aims of this review is to make the capital standards more sensitive to the risk profile of each insurer.

APRA has specific prudential requirements for general insurers in regard to capital and reinsurance that are designed to ensure that the insurers are able to cope with multiple catastrophes.

Many of you will be familiar with these requirements, but let me provide an outline here, as it will help understand our current focus and how recent circumstances might influence our thinking.

Retention and Reinstatement

In simple terms APRA currently requires an insurer to determine the ―probable maximum loss‖ for catastrophic events up to 1 in 250 years likelihood. The insurer is then required to hold capital or reinsurance up to that amount. In practice, insurers generally retain the first slice of risk [red box], and reinsure the balance [green box].

Catastrophe reinsurance is provided on an annual aggregate basis, and thus in the event of a catastrophic event the reinsurance is ―used up‖ by a claim, so it is necessary to reinstate cover after an insurer claims on their reinsurance.

The retention of the first tower is included in APRA’s minimum capital requirement. Often only one ―vertical‖ amount of reinsurance is obtained. Where the reinstatement premium has not been pre-paid, APRA’s minimum capital requirement includes the cost of one reinstatement of the reinsurance.

What I have just outlined is APRA’s current minimum requirements. Of course, each general insurer can make its own arrangements beyond this minimum. This might include, for example, reinsurance to cover more than one event, or to cover the retention for more than one event [the red boxes] should multiple events occur – so-called sideways cover.

What is unusual about the recent past is the number of large adverse events that have occurred in such a relatively short period.

This has had two broad effects:

• there has been a substantial impact on the capital of insurers providing cover against these events (through erosion of a large part of the red boxes), and
• reinstatement and/or back-up reinsurance arrangements [more green] have had to be put in place expeditiously.

So the adequacy of insurers’ reinsurance arrangements has been thoroughly tested, along with their ability to reinstate and or replace part(s) of reinsurance programs.

Overall, APRA does not expect that any general insurer it regulates will have difficulty continuing to operate in the face of the potential losses from these recent disasters, though some smaller insurers have had capital support from parents. We are confident in the capacity of the general insurance industry in Australia to meet all of its claims obligations.

We do expect, however, that a number of general insurers will incur additional costs to reinstate reinsurance cover that has been eroded and, where required, to buy additional reinsurance cover for the remainder of the year. In some cases, the cost of the reinsurance cover being sought could be greater than the net claims incurred from the catastrophes themselves.

Given the unusual circumstances, APRA has established a small team of specialist staff to monitor the insurance claims from these events, and to assess the impact of these claims on individual general insurers and on the industry as a whole. We want to satisfy ourselves that each APRA-regulated general insurer has the financial capacity, now and in the future, to meet our capital requirements and, ultimately, its claims from policyholders.

To help with our assessments, we are seeking updated high level capital plans and financial forecasts from each affected general insurer. As part of our assessment, we are applying a stress test that assumes at least one further major catastrophic event of similar proportion to the largest loss experienced to date before the end of the financial year.

While this work is helping us with the immediate situation, importantly it will also help inform our LAGIC thinking and our future approach to supervision.

In other words, we need to learn from this experience.

Stress Testing

This leads nicely into the general topic of stress testing.

On occasions in the past, APRA has requested insurers to conduct stress tests on their business, based on parameters specified by APRA. The one to which I just referred is an example.

These tests are powerful tools. They help us and management to understand the capacity of the organization to cope with the specified scenario, but they also provide insights into vulnerabilities of each company. This in turn can lead to changes in practice or operations.

More generally, stress tests can be of great help to management and boards in assessing the out-workings of actuarial models and prudential capital requirements, without the need for a detailed understanding of how those models or requirements work.

We want to encourage management and boards to consider stress tests as one of the more important weapons in their armoury, and to use them accordingly.

In a minute, I will give an example of how stress testing could be helpful to boards and management in meeting their LAGIC requirements.

International Developments

Before I talk about LAGIC, it is worth reviewing some international developments in financial regulation to give context.

Recent and current activities include the following:

• Basel 3 in banking, which will see revised capital requirements (amount and quality) and liquidity requirements.
• Solvency II for insurance in Europe, long in the making, which is a fundamental overhaul of capital requirements.
• UK changes in insurance regulation a few years ago.
• International Association of Insurance Supervisors (IAIS) guidance for insurance supervisors (like APRA).
• There are a few points to note here:
• They all use the three pillar concept, which we have adopted for LAGIC.
• They are all moving to greater risk-sensitivity.
• They all have 1 in 200 (or similar) level of sufficiency aims.
• There is increasing alignment of capital requirements across sectors i.e. banking, life and general insurance (one of the aims of LAGIC).
• There is increasing alignment of requirements from regulator to regulator internationally.
• There is also increasing adoption of the Tier1 and Tier 2 concepts of quality of capital (as for LAGIC).
• In most cases, the concept of formally defined capital buffers is being introduced – sometimes called ―hard‖ and ―soft‖ floors.

So (perhaps unsurprisingly in the aftermath of the GFC), we are seeing much greater co-operation and alignment of thinking internationally and across sectors, and APRA is part of this.

IAIS

As its name implies, the members of International Association of Insurance Supervisors are regulators like APRA.

APRA is an active member of the IAIS, and is engaged in a number of its activities.

IAIS’s aim is to support the development of high standards of insurance supervision around the world. Accordingly it has developed a set of Insurance Core Principles (ICPs), supported by Standards, and Guidance. These are aimed at the supervisors rather than institutions, and are intended to set minimum standards.

There are three main activities currently being undertaken by the IAIS:

1. A comprehensive review of the ICPs, Standards and Guidance. This work is well advanced.

2. The Financial Stability Board on behalf of the G20, has been working with the Basel Committee on Global Systemically Important Financial Institution (G-SIFIs) for some time, the idea being that such organizations should be subject to more stringent capital and/or supervision. IAIS has started work on an analysis of what might constitute an insurance G-SIFI. This is proving to be quite a difficult issue to grapple with and work will be ongoing.

3. The development of what is called ―ComFrame‖. This is a common framework for the supervision of internationally active insurance groups – that is, those that operate in more than one jurisdiction. The intention here is that there is a consistent and co-operative approach to the regulation of insurers that operate internationally. This will influence our approach to regulation of such companies in due course.

The IAIS is increasingly influential and is recognized by the Financial Stability Board (which supports the G20 leaders) as its primary advisor on insurance. In this sense it is the broad equivalent of the Basel Committee in the banking world.

So international developments are significant, and will be increasingly important in insurance regulation. Note though, that the peculiarities of local markets will always be reflected in local regulation. For example, Solvency 2 needs to take account of the markets and politics of Europe. Similarly, while these developments will influence us, APRA’s requirements will be tailored to our markets and reflect our standards.

LAGIC

Later today, there will be a panel session on LAGIC, and Helen Rowell will be there to give you an update and to answer your questions.

So I don’t want to go into detail about LAGIC, but I do want to cover a few higher level points.

High Level Issues

Process

First, let’s talk about process. As an outsider to APRA, it has been quite interesting to walk in part way through the LAGIC project. I can tell you that the process is a rigorous one, with plenty of intellectual horsepower, challenge and debate. And all of this is done in light of substantial input from industry. The formal consultation process is quite visible, but there is also considerable less formal interaction with industry. So, for example, there have been meetings with the ICA board and an ICA committee, with the Institute of Actuaries, with individual actuaries, and with boards and senior management of a variety of companies. There has been verbal and written feedback. All of this is valuable input.

So I am pretty comfortable that the industry has had ample opportunity to present its case.

But the feedback presents a logistical challenge for us. All of it has to be assembled and digested and assessed – a lot of work.

We are well advanced in that process, and in a few weeks we will issue our response paper, and then conduct another QIS.

Pillar 2 and ICAAP

It was apparent that it was not clear to many how our capital requirements – in particular Pillar 2, ICAAP, and target surplus - would work in practice, so I thought I should spend a few minutes on this.

You will recall that as part of LAGIC we are introducing the three pillar concept. To remind you,

• Pillar 1 – quantitative requirements in relation to required capital, eligible capital and liability valuations;
• Pillar 2 – the supervisory review process which may include a supervisory adjustment to capital; and
• Pillar 3 – disclosure requirements designed to encourage market discipline.

Under Pillar 2, APRA may increase the prescribed capital determined under Pillar 1 if we are of the view that this amount does not adequately account for all of an insurer’s risks. Such an adjustment may increase the total required capital amount and/or strengthen the composition of the insurer’s capital base (i.e. the insurer may have to hold an increased proportion of higher quality capital).

We are also introducing ICAAP – the Individual Capital Adequacy Assessment Process.

APRA already expects insurers to have in place a process to assess their capital needs and manage their capital levels. ICAAP formalises those requirements. In particular, we want insurers to assess their own risk profile and the capital needed to support the risks they undertake, and to carry out appropriate capital projections and stress testing.

The insurer’s ICAAP is expected to go beyond the need to meet regulatory capital requirements. It must include development of appropriate internal capital targets, which reflect the insurer’s risk appetite. There should also be an adequate process for monitoring actual capital levels relative to the insurer’s own capital targets and regulatory capital levels.

Some of the feedback that we received indicated concern that we were being overly conservative and potentially requiring buffers on buffers above PCR. That is not our intent. Let me explain using this slide.

Control Levels

Recall my reference earlier to the concept of a hard floor and a soft floor for capital requirements. The Prudential Capital Requirement (Pillar1, 2) is a hard floor which must not be breached.

Breaching PCR is like an insurer ―putting its foot in boiling water‖. It needs to get it out pretty quickly to avoid APRA taking serious measures and potentially taking control out of the hands of the board and management if the breach is not rectified very quickly.

Above PCR, APRA expects an insurer to set its target position based on its ICAAP. We are not prescriptive about an insurer’s approach to setting target surplus – it can be a range or a single target level.

The insurer is expected to manage its capital according to its ICAAP and target surplus policy. At times actual surplus may drop below target position. This is quite acceptable as long as the situation is addressed by its ICAAP and managed accordingly.

Note that stress testing is likely to be a very useful tool in developing these plans.

As you can see from the diagram, the supervisory attention paid to an insurer will increase in intensity as the insurer’s actual capital approaches the PCR.

The two main points that I want to make here are that:

1. Pillar 2 provides a powerful tool for APRA to reflect a company’s quality of risk management in its capital requirements (PCR on slide). Or to turn it around, subject to the lower bounds of Pillar 1, the company can influence its capital requirements through the quality of its risk management
2. The board and management have responsibility for setting target surplus and managing the capital position around that target in accordance with their ICAAP. It is critically important that this is done well.

Pro-cyclicality

I briefly want to mention pro-cyclicality – the magnifying effect that can sometimes occur in adverse conditions, which then makes the conditions worse. When capital is eaten away by adverse conditions, prudential capital does not reduce in lock step. So it is inevitable that prudential capital requirements have some element of procyclicality.

In reviewing our LAGIC requirements, we have been consciously seeking to reduce this effect where we thought it was prudent to do so. As a result a number of changes will be made to our LAGIC proposals.

Risk Appetite

As you know, APRA makes it clear that the board is primarily responsible for the Risk Management Framework (RMF), and it defines the Risk Management Framework quite broadly. Just to remind you, it refers to ―the totality of systems, structures, processes and people within the insurer‖ in the context of risk management.

That’s pretty comprehensive!

The Risk Management Framework includes the Risk Management Strategy which must, amongst other things, set out the company’s Risk Appetite.

So the board is responsible for setting the company’s Risk Appetite – in simple terms, a clear statement of the degree of risk the company is willing (and able) to take in pursuit of its goals.

A well considered, clearly articulated Risk Appetite is the very foundation of sound risk management. Without this, risk management throughout the business will be carried out with unclear boundaries and expectations.

So setting Risk Appetite should be top of the Risk Management list for the board.

But setting Risk Appetite is much easier said than done – in fact, it is quite difficult to do well, and this is reflected in the practices we are seeing.

In recent times, we have reviewed the Risk Appetite statements from a number of GI and life insurers, and spoken to a number of CEOs and boards about the engagement of the board in the Risk Appetite process.

In summary, here’s what we have found so far:

• In some cases there is no clear statement of Risk Appetite, or no obvious understanding of what it actually is in concept.
• There is a wide range of approaches to articulating Risk Appetite - from short high-level statements to a few pages of detailed thoughts. This diversity is not necessarily a bad thing, but we need to better understand the practical implications of it.
• The quality of the statement of Risk Appetite ranges from poor to quite good.
• There is a lack of analysis of Risk Appetite through the use of scenario analyses, stress testing etc.
• It is not always clear that the board has been heavily engaged in setting the Risk Appetite.
• In some cases there is a dis-connect between the Risk Appetite statement and its translation into operational management.
• And last, with some subsidiaries or branches of foreign-owned insurers, we see an adoption of group risk management practices without necessarily full and proper engagement of local management and/or board.

Risk Appetite – What supervisors should look for

As a first response, we have been developing a checklist for our supervisors – what they should look for in a company’s approach to risk appetite.

This is very much a work in progress and will be refined based on experience in the field.

In due course, we will issue guidance for insurers, but in the meantime, it will help you to understand the intended focus of our supervisors.

Four broad areas have been identified for assessment – Governance, Risk Management, Implementation and Communication.

This is what our supervisors will be looking for:

Risk Appetite Statement

First, under Governance

1. Evidence that the Board has been engaged in the development of the risk appetite and has demonstrated ownership by its approval of relevant documents.
2. The risk appetite considers the interests and expectations of relevant stakeholders - depositors and policyholders as well as shareholders.
3. The risk appetite is periodically reviewed in the context of the prevailing economic conditions and the competitive environment.
4. There is a feedback loop that enables the Board to
   
   i) understand how management interprets and applies risk tolerances and limits;
   
   ii) be satisfied that management’s interpretation and application of the risk appetite is appropriate; and
   
   iii) review the risk appetite statement in light of these factors.
5. The risk appetite seems reasonable, makes sense and is able to be understood. 
6. Within group structures
   
   a) there is a group RAS; and
   
   b) there is a RAS for each entity which is consistent with the group RAS.
   
   
   
   Now let’s look at Risk Management
   
    
7. The risk appetite considers the material types of risks of the entity.
8. The risk appetite statement is core to the risk management framework and makes clear boundaries and expectations. 
9. The risk appetite is supported by tolerances and limits, which are embedded in the business. 
10. The risk appetite is considered when assessing the adequacy of controls in mitigating risks.
    
    
    
    And Implementation
    
     
11. The entity’s strategic objectives and business plans must be consistent with the risk appetite. 
12. The entity’s capital management plan is consistent with the risk appetite; and capital levels and targets are monitored and reported against the risk appetite. 
13. The risks necessary to produce required returns on capital are within the risk appetite.
    
    
    
    Lastly, Communication
    
     
14. There is relevant communication of the entity’s risk appetite and/or its derived tolerances and limits to each part of the entity’s operations.
15. The board regularly receives reporting on the entity’s level of risk relative to its risk appetite e.g. via metrics for measuring position against tolerances and limits which are derived from the risk appetite; and is engaged in assessing relevant implications of this information. 
16. Scenario analysis and stress testing is performed against the risk appetite and reported to the board in order to confirm that the risk appetite does not overstate its risk capacity.

I hope this helps understand the process we plan to follow.

In any event, you should expect that APRA will increase its focus on Risk Appetite - and its supervisory skills and expertise in this area will continue to develop.

Remuneration

Before I finish, some words on remuneration.

As you know, APRA imposed new requirements for remuneration as from 1 April last year.

Since then, our frontline teams have been doing compliance checks focused on process – ensuring governance arrangements are in place, a remuneration policy has been prepared etc.

In the second half of last year, we conducted peer reviews. Basically we compared how policies were being implemented across a group of similar institutions, and we then provided feedback.

The next step in the process is for us to meet with the Remuneration Committees (NEDs) of our larger institutions to talk through how they are managing the new requirements in practice, issues they may be struggling with.

So we see this as very much an evolving field, and we are keen to work with the industry to ensure a high quality and robust outcome.

Conclusion

Let me finish now.

I’ve covered quite a lot of ground, but I hope my comments have given you some insights into our current thinking and priorities.

You may even have found some of it interesting!

Thank you for your time.