Finding success in failure

Wayne Byres, Chairman - Actuaries Institute Banking Conference 'Banking on Capital', Sydney

Thank you for the invitation to be part of today’s event. The Actuaries Institute’s increasing engagement on banking-related matters is a welcome development and, as was the case last year, I’m happy to support it. I am, however, going to risk the ire of the organisers by straying somewhat from the theme of today’s event: rather than ‘banking on capital’, I’m going to talk about why a sole reliance on capital is an unwise strategy.

As you all know, over the past couple of years, there has (quite justifiably) been a great deal of attention given to the implications of the first recommendation of the Financial System Inquiry (FSI): that APRA should set capital standards such that Australian authorised deposit-taking institutions (ADIs) would have capital ratios that are seen to be ‘unquestionably strong’. Since that recommendation was published, much has been written about what it means – and given that it will be a little while before APRA is in a position to make a final determination on how we best meet this recommendation, no doubt much more will be. Today, I’ll simply repeat the point I’ve made previously: that the steady accumulation of capital remains a sensible course of action for most ADIs.

It’s always important to remember, though, that a stable and resilient banking system is not just delivered by more capital. Capital adequacy is a relative concept – is capital adequate relative to risk? When we judge a bank’s capital to be high or low, or something in the middle, we are making a judgement that takes into account a range of issues that impact on bank risk profiles: funding and liquidity, asset quality, governance, risk management and risk culture, to name a few, all come into the equation in some way or another.

We also need to remember there are no guarantees. No level of capital (short of 100 per cent equity funding) can provide creditors with an absolute guarantee against the possibility of bank failure. To pick up today’s theme, we can’t always ‘bank on capital’: something can always go wrong. To attempt to provide the community with an iron clad guarantee that nothing can go awry would require severe limitation on the risk-taking ability of the banking system, and prevent it from fulfilling the vital and productive roles that it plays in intermediating between borrowers and lenders and facilitating the smooth functioning of payments throughout the economy. Put simply, a zero failure regime is not desirable.

Inevitably, we need to find a balance.[1] One implication of that is, even once we define and achieve ‘unquestionably strong’, we won’t be downing tools. In a world of fractional reserve banking, ‘unquestionably strong’ will never equate to ‘invincible’.

That in turn means we need to think about failure.[2] And when I use the word ‘failure’, I’m not just talking about a classic case of insolvency or bankruptcy, but using it more broadly to capture situations in which the on-going viability of one or more financial institutions is in serious jeopardy.

The failure of a bank is not like the failure of a bakery. The potential for contagion to other financial firms, not to mention the widespread adversity it can impose on the broader community, means the failure or near-failure of a bank is no run of the mill matter. So, if we start with the proposition that failures are inevitable, then I’d suggest the regulatory system needs to be assessed against two benchmarks: are failures nevertheless sufficiently rare, and do we have mechanisms to handle them adroitly when they occur? If the answer to both these questions is yes, then community confidence and trust in the financial system is likely to remain high.

With that in mind, we can think of two basic types of failure: orderly and disorderly.

To the extent we have failures, orderly failures are what we aspire to. Orderly failures will typically have been anticipated, produce no loss to protected beneficiaries,[3] and be managed in such a way that the firm’s critical functions are maintained without disruption while the situation is resolved. (Shareholders and other providers of capital of the non-viable firm may very well lose out along the way, but that is their lot in life).

To the extent any failure can be regarded as a success, then an orderly exit from the industry in this fashion is what we should strive for. For example, a firm that identifies it is approaching non-viability or a strategic dead-end, and seeks to merge with a stronger and more viable entity (sometimes with a nudge along the way from APRA) is a far from uncommon event in the Australian financial system. To the extent such exits can be classed as failures, or at least near failures, they are often so orderly that no one notices them.

Disorderly failures, on the other hand, are to be avoided. Disorderly failures will be those which catch everyone by surprise, impose significant losses on those who the regulatory framework seeks to protect, and/or involve significant disruption to the smooth operation of the financial system, and economic activity more generally. HIH Insurance would be the classic of this genre.

So what can we do to maximise the probability that failures will be orderly, rather than disorderly? I’d suggest there’s a number of preconditions that the public sector needs to put in place:[4]

• active supervision is at the heart of identifying the risk of failure early – it will be hard to anticipate problems if there is no one looking for them in the first place;
• powers to intervene must exist – this is critical to being able to address small problems before they become big ones;
• a willingness to intervene is also critical – powers are of no value if there is no capacity or willingness to use them when needed[5];
• planning and preparation – ideally, contingency plans will have been prepared and tested in ‘peace time’ rather than thought up on the run in the midst of a crisis;
• a capacity to maintain critical functions – an orderly resolution of a failure may take time, yet critical functions need to be maintained in the meantime; and
• there will ideally be a safety net or backstop to provide additional confidence at a time of uncertainty – which will help avoid actions that are individually rational but collectively damaging (the typical case being a bank run).

APRA, along with our colleagues amongst the Council of Financial Regulators, has spent a great deal of time in recent years looking in a fairly hard-nosed way at how well Australia stacks up against these preconditions. The conclusion was somewhat mixed: there are no glaring deficiencies, but a number of areas for improvement.

• The importance of active supervision, and a willingness to intervene where appropriate, were some of the hard lessons that APRA took to heart following the HIH episode. Justice Owen found APRA under-resourced to identify problems, and slow to respond to them once found. These were fair conclusions, and APRA worked hard under my predecessor to build both its capacity and conviction. Fifteen years on from HIH, efforts to further improve our supervision – to identify risks early and respond promptly – remain at the forefront of our latest strategic plan, and I expect they will always feature prominently in APRA’s priorities.
• When it comes to powers to intervene, the FSI’s Final Report noted there are some gaps and deficiencies in the Australian statutory framework for crisis management and resolution when compared with international standards.[6] This includes the need for such things as broader investigation powers; strengthened directions powers; improved group resolution powers; enhanced powers to deal with branches of foreign banks; and more robust immunities to statutory and judicial managers. In his speech last week, the Treasurer noted the Government’s intention to make improvements in this area, which we see as a very valuable (and low cost) investment in the stability in the financial system.
• Crisis planning is critical. Last year at this event I spoke about our plans for recovery and resolution planning. During the past year, I’m pleased to say larger ADIs have submitted new plans based on updated guidance issued by APRA, and we are now reviewing and benchmarking the plans in order to highlight areas of better practice that will further increase the credibility of plans in subsequent iterations.[7] On resolution planning, more detailed work is also underway with specific firms to consider the planning required to ensure that APRA is able to use our resolution powers when needed. Our focus here is on the assessment of critical functions, intra-group dependencies such as critical shared services, and the identification of potential barriers to resolvability.
• No matter how good the plan, however, stabilising and restructuring a financial firm that is no longer viable in its current form is rarely going to be a quick and easy exercise. So it is important that, while a resolution plan is being implemented, the firm’s critical functions can be maintained so as to reduce potential losses and minimise the disruption to the broader financial system. Key to doing this is that the firm has the financial resources to allow it to continue to operate while its business is reorganised.

Searching for new capital when a business is distressed, and time is of the essence, is ideally to be avoided. And we want to minimise the risk that the taxpayer has to come to the rescue. With that in mind, the post-crisis regulatory framework has built mechanisms that trigger automatic corrective action to help restore a firm’s capital when it has been diminished. There is a lot of discussion and debate on the merits of so-called bail-inable instruments and bail-in powers, including in response to the FSI’s third recommendation that APRA implement a framework for recapitalisation capacity sufficient to facilitate the orderly resolution of an Australian ADI and minimise taxpayer support. But the idea of bail-in is not something completely new: certain recapitalisation mechanisms already exist in the Australian framework. Examples include:

• the use of the capital conservation buffer, which imposes increasing limitations on a ADI’s ability to make discretionary distributions to capital providers and employees as the ADI approaches its minimum regulatory requirements;[8]
• the trigger that exists in Additional Tier 1 instruments (often referred to as ‘hybrids’) that provide for the instrument to be written off, or converted to equity, in the event that an ADI’s capital ratio falls below 5.125 per cent; and
• the point of non-viability trigger in both Additional Tier 1 and Tier 2 instruments, which provides for the instruments to be written off or converted to equity in the event that APRA considers that the ADI would become non-viable without such action (or some other form of support).

These latter two recapitalisation mechanisms, in particular, are designed to provide some ‘breathing space’ to allow for orderly resolution. They are not designed to deliver resurrection, but more modestly to provide scope for an ADI’s services to customers to continue while new owners and managers are being put in place. Strengthening this by increasing loss absorption and recapitalisation capacity further, as recommended by the FSI, remains a work in progress and likely to take some time to complete. We are, as I have said elsewhere, hastening slowly in response to that recommendation given the importance of getting the policy settings right.

However, the new mechanisms that have already been instituted within existing capital instruments are a very important part of the new regulatory framework. Viewing these capital instruments as simply higher-yielding substitutes for vanilla fixed-interest investments, let alone deposits, is something to be counselled against, since from APRA’s perspective holders of these instruments are providing the important first lines of defence that we can call into action, in some instances even ahead of shareholders, to aid an orderly resolution.

• Finally, despite these new mechanisms, having a safety net or backstop to provide additional confidence before and during a period of resolution is an important component of any resolution framework. Australian depositors have long had the benefit of the depositor preference provisions in the Banking Act 1959. In 2008, this was supplemented by the Financial Claims Scheme (FCS) which seeks to provide certain depositors with prompt access to their funds – up to $250,000 per account holder - in the event an ADI fails.[9]

But to achieve a prompt payout of protected deposits to account holders – something depositors must be confident of for the backstop to work as intended – requires that ADIs be operationally ready, prior to any failure, to meet payment, reporting and communication requirements for the FCS.10 Planning and investing to facilitate their own demise is something that financial firms inevitably struggle to do, so APRA will be reinforcing its expectations in relation to ADI’s FCS testing programs in the near future, with a view to ensuring there is genuine readiness to activate the FCS if it is ever needed.

Concluding remarks

Adequate capital is undoubtedly critical to the stability of any banking system. But to return to today’s theme, we can’t solely ‘bank on capital’ to deliver safety and stability. If we accept that failures, while hopefully still reasonably rare, are nevertheless inevitable, then preparation to minimise their impact is an essential investment. ‘Successful failures’ might seem a contradiction in terms, but they are far better than the alternative.

Footnotes

1. The need for balance is recognised in our enabling legislation, the Australian Prudential Regulation Authority Act 1998, which requires APRA to balance financial safety and financial stability with considerations of competition, efficiency, contestability and competitive neutrality. It is also recognised in the Statement of Expectations that the Government sets for APRA, which acknowledges that it is not possible or efficient for APRA to eliminate all risks, and that trade-offs in risk reductions are necessary. That expectation applies not just to banks, but to all industries APRA regulates.
2. It also means, by the way, that not every failure will necessarily imply a failure of the regulatory regime.
3. Beneficiaries in this context refers to those creditors of a regulated entity that APRA is tasked with protecting: ADI depositors, insurance policyholders and superannuation fund members. They can be thought of as the primary beneficiaries of APRA’s activities.
4. That is not to imply the responsibility rests solely with the public sector. APRA’s supervisory philosophy is built on the premise that the primary responsibility for the prudent management of a regulated firm rests with its board and management.
5. Being willing to use powers can be easier said than done. In an impending failure, information on the state of the firm will likely involve a great deal of uncertainty, and while early action is almost always the best course of action, acting too early can bring legal risks with it. Key to having appropriate powers, therefore, is not just the extent to which the powers exist in statute, but also having as much clarity as possible as to how and when they can be used so that regulators can use the powers as intended with confidence.
6. The relevant benchmark is the Financial Stability Board’s Key Attributes of Effective Resolution Regimes.
7. APRA will also be working with large insurers on recovery planning in 2017, following the industry stress tests this year.
8. The capital conservation buffer is, for most ADIs, a minimum of 2.5 per cent of risk-weighted assets (3.5 per cent in the case of an ADI that APRA has deemed a domestic systemically important bank). The buffer may be extended by up to an additional 2.5 per cent (ie to 5.0 per cent, or 6.0 per cent for DSIBs) through the use of the countercyclical buffer. At present, the countercyclical buffer component is zero in Australia, although ADIs with operations overseas may be required to hold additional buffer if authorities in foreign jurisdictions have applied a non-zero buffer to exposures within their jurisdiction.
   • In cases where an ADI does not hold sufficient CET1 to meet its minimum prudential capital requirement (PCR) plus its required conservation buffer (CB), the ADI is limited in the extent to which it may make discretionary distributions - typically, dividends on ordinary shares, distributions on Additional Tier 1 instruments, and discretionary payments to staff - in accordance with the following table (which shows the relevant ranges in instances when the total buffer is 2.5 per cent, and when the buffer is increased to 3.5 per cent). As noted, these ranges will increase whenever the countercyclical capital buffer is applicable.

1. The general insurance component of the FCS covers most claims up to $5000. Claims above $5000 are also covered for eligible policyholders and certain third parties.
2. For depositors to be confident in the backstop, they also need to be aware of it in the first place. APRA has recently launched a new website – www.fcs.gov.au – to give the FCS greater prominence, and provide the community with improved understanding about how the FCS works.