APRA's Basel III implementation rationale and impacts

Charles Littrell, Executive General Manager - APRA Finsia Workshop, Sydney

Good morning, ladies and gentlemen. Following Dr. Laker’s introductory remarks, I propose to give a somewhat more detailed view of the Basel III capital and liquidity reforms. Then we will look at some aggregate impacts upon the larger banks, and touch upon cost-benefit considerations.

The capital reforms

Under current prudential arrangements, established by the Basel Committee in 1988, the way we form capital requirements is first to determine risk-weighted assets. Risk-weighting is intended to map capital closer to the true risk of loss, rather than simply to the accounting measure of an asset.

Once the risk assets are established, a total capital requirement is set, and this total capital requirement is then decomposed into a rather complex tiering of common equity, preference equity, and subordinated debt, plus some provisions.

One unsatisfactory outcome from this arrangement is that common equity is insufficiently prominent in the mix of prudential capital requirements.

Under Basel III, we also start with risk-weighted assets, but the capital requirements are reversed and simplified. That is, we start with the common equity requirement, then add a Tier 1 requirement, which is essentially preference shares, then a total prudential capital requirement, which encompasses subordinated debt.

The Basel III approach reflects the global consensus that common equity is the key element in capital. In addition to this fundamental change, Basel III updates the 1988 capital requirements in many other ways. The concept of loss absorbency is a case in point. During the global financial crisis, some governments found themselves bailing out subordinated debt and preference share capital providers, as well as senior creditors. This is not how capital is supposed to work, and the new consensus makes it clear that prudential capital providers are all on the hook for equity exposure, should the relevant bank fail.

In global terms there are tighter definitions on what counts as prudential capital, what assets can be included for capital calculations, and how capital deductions are allocated between common equity and other capital items. Much of this tightening was already in place in Australia, so the local effects are relatively small.

Basel III also introduces the concept of explicit buffers, notably a capital conservation buffer, and in some circumstances a counter-cyclical buffer. Later this morning, my colleague Wayne Byres will discuss APRA’s approach to buffers.

As a separate string to the regulatory bow, we also propose to introduce a leverage ratio requirement.

In addition to many definition and application changes, the Basel III capital reforms call for higher capital requirements. Under Basel III, the common equity requirement more than doubles on a comparison of the minimums, and more than triples when the capital conservation buffer is taken into account.

What effect will these reforms have on Australia’s authorised deposit-taking institutions, or ADIs? The Basel III capital reforms will apply to all ADIs, but the great majority of building societies and credit unions already comfortably exceed the new capital requirements. Any economic impact is likely to fall upon the larger banks.

As is typical in such reforms, APRA has conducted and will continue to conduct a series of quantitative impact studies, or QISs, to estimate the actual impact of the reforms. We collect this information at the individual ADI level, but confidentiality considerations dictate that any public disclosure is at an aggregate level. The numbers discussed here are for Australia’s six largest listed banks, based upon pro forma reports from mid 2011.

What this data shows is that using the minimum global standards, these banks would report an 8 per cent common equity Tier 1 ratio today. As is typically the case, APRA adopts a more conservative approach than the Basel minimum, and the outcome using APRA’s proposed Basel III implementation is a 6.9 per cent common equity Tier 1 ratio.

When we consider Tier 1, there is a complication in that essentially the entire stock of Australian non-common equity will not qualify as Tier 1 under the new Basel III definitions. APRA considers that transition for this stock of instruments is appropriate, and based upon our assumed transition path, Australia’s larger banks would report an 8.1 per cent Tier 1 ratio, if Basel III were in place today.

I emphasise that this is the 2011 pro forma, for a 2013 requirement. Under the Basel rules, the common equity and Tier 1 minima are 4.5 per cent and 6 per cent respectively. The banks in aggregate are already well in excess of these requirements.

There is scope in Basel III for APRA to bring in the 2013 requirements incrementally, over a number of years. We have elected not to do this, for two reasons. First, such a deferral would be pointless, given that Australia’s ADIs are already comfortably in excess of the requirements. Second, we wish to signal Australia’s relative capital strength to the rest of the world. A number of other jurisdictions are considering similar approaches, where their banking sectors can afford a quick transition, and several countries are setting higher minimum capital requirements than the base level in Basel III.

There is about a 1.1 per cent difference in common equity Tier 1 ratios between the Basel minima and APRA’s proposals. This difference arises mainly from two sources: deferred tax assets and equity investments in affiliates. In a late change to the Basel rules text, in December 2010 the Basel Committee decided to give national supervisory authorities the discretion to allow banks to avoid capital deductions on a proportion of their capital investments in other (non-consolidated) companies, and their deferred tax assets. The earlier rules text did not include these concessions, and APRA’s current rules require these assets to be deducted.

APRA considers it a relatively straightforward matter that these assets should not be counted. The first would represent a double use of capital, which does not strike us as prudent. The second allows an asset that is unlikely to exist in the case of a failed or severely troubled institution, and again, this does not strike us as prudent.

As part of the Basel III implementation, APRA is proposing to reverse its long-standing treatment of expected dividends. These are currently deducted from capital. We propose to align with the rest of the world, and allow this item to count in capital. Our reasoning here is that the overall thrust of Basel III, particularly the new capital buffers, would probably remove the need for our current dividend treatment. In recent years, Australia’s banks have also demonstrated their willingness to reduce dividends in need, and this gives APRA some comfort in moving away from the dividend deduction.

In addition to direct capital effects, Basel III will have some risk asset effects. If we were following the global rules at their minimum value, the risk--weighted asset increase would actually be greater, as some assets APRA proposes to disallow, would instead receive a risk-weighting. Otherwise, there are some relatively minor adjustments for items associated with markets trading and with securitisation investments.

The Basel III capital reforms focus upon common equity, but also mandate worthwhile improvements to other prudential capital. Most conspicuously, all prudential capital needs to be equity, become equity, or be cancelled in the event of a failure. Because this rule has not applied to the current stock of non-common equity capital, the current Australian stock does not comply with Basel III. APRA proposes to grant limited transition relief, to give the banking sector time to replace its capital other than common equity with Basel III compliant new issues.

As for the leverage ratio, APRA intends to implement the global minimum requirement, in the globally mandated timeframe. APRA has always seen significant prudential benefits in a risk-based capital measure, but the leverage ratio is nonetheless part of the minimum ticket for a nation to achieve Basel III compliance. We note that Australia’s larger banks already comply, and are likely to be even more compliant by the 2018 introduction date. Smaller ADIs should also meet this test without difficulty.

APRA considers that the Basel III capital reforms will materially strengthen the banking capital regimes in some countries. In the Australian case and throughout most of Asia, Basel III will achieve worthwhile but more moderate strengthening, because our previous rules and practices were stronger than the global minimum requirements.

As noted earlier, Australia’s banks already comfortably exceed the 2013 requirements. By the time the capital conservation buffer comes into play in 2016, we expect that the larger banks will add sufficient equity to give themselves an internal operating buffer over the benchmark 7 per cent common equity ratio. On their current earnings and growth profiles, this should be achievable mainly from retained profits and reinvested dividends.

The liquidity reforms

Let’s now move to the liquidity reforms. Unlike capital, where Australia’s banks were and are demonstrably in excess of global minimum expectations, during the global financial crisis our banking system revealed a strategic vulnerability to wholesale foreign funding. The industry, and APRA, and for that matter nearly everybody who took an interest in money markets, assumed that these markets would punish weak borrowers, but would provide funding on at least tolerable terms to sound borrowers. In 2008 this expectation was proven false, as the world’s fixed income investors abandoned structured and unsecured financial sector debt, in favour of sovereign debt.

Recently, Standard & Poors issued new guidelines for bank soundness across countries. The so-called BICRA ratings range from a score of 1 to 10, based upon the joint consideration of a nation’s economic strength, and its banking system strength. Australia scores a 2. Formerly Australia was in the top tier. Only Canada and Switzerland now enjoy a score of 1.

Why has S&P downgraded Australia from the top tier? Let’s look at the subsidiary S&P analysis for system-wide funding. In these subsidiary tables, scoring ranges from a high of 1, to a low of 6. Australia is nearly falling off the wrong end of the chart, with a score of 5. And this is after the banking system has made substantial improvements in the past three years. This liquidity assessment helps to explain why Australia’s banking system is no longer placed by S&P in the very top tier for soundness.

We do not agree with S&P on all of their analysis. Among other things, we think that sourcing long-term funding offshore is sensible, and the fact that Australian bank borrowings are hedged to our sovereign currency is probably understated as a strength. It is clear, however, that Australian banks are an international outlier in their reliance upon foreign and wholesale funding, for domestic and retail assets.

The Basel III liquidity reforms include worthwhile improvements in risk governance. I don’t propose to discuss these improvements today, other than to note that APRA is already supervising ADIs on the expectation that they meet, or are moving towards meeting, these liquidity risk governance standards.

The Basel III quantitative liquidity standards have received much more attention. These standards are the Liquidity Coverage Ratio, or LCR, and the Net Stable Funding Ratio, or NSFR. In both cases, upon their respective 2015 and 2018 introduction dates, the requirement will be that the ratios exceed 100 per cent.

The LCR and NSFR in Australia will only apply to those ADIs, essentially the banks, which are required by APRA to manage their liquidity using scenario analysis. Other ADIs, largely credit unions and building societies, will not need to meet the LCR and NSFR, and largely will be held to their current liquidity requirements.

A 100 per cent LCR translates into a 30 calendar day survival horizon for a bank experiencing a specified name stress. Broadly, this stress involves the markets refusing to extend new facilities or roll some facilities with the bank, but the stress does not extend to a full scale run. This test is roughly four times more demanding than the current APRA liquidity requirement, which mandates a five business day survival horizon.

The NSFR is not a survival test, but instead assesses the degree to which a bank’s illiquid assets are funded by capital, stable deposits, or long term funding.

The global application of the LCR is based upon several assumptions that do not apply in Australia, or for that matter in New Zealand. The critical assumption is that banks can find a large stock of government paper, well-rated corporate bonds, and other non-bank assets from which to build a pool of liquid assets. In Australia,for the excellent reasons of low government indebtedness, and low corporate leverage, this critical assumption does not hold.

In our QIS work on the LCR, the larger banks are currently running at about a 31 per cent ratio, before considering the RBA’s liquidity facility, vs. the 100 per cent minimum. The ratio is based upon two defined terms: net cash outflows, and high quality liquid assets.

In Australia, based on mid-2011 numbers, net outflows under the LCR stressed scenario would be about $429 billion, and qualifying assets are only $132 billion. For Australia’s banks to meet the LCR, they need to greatly reduce their net outflows, or greatly increase their qualifying assets, or both increase qualifying assets and reduce net outflows.

Reducing net outflows implies that banks reduce the degree to which they provide maturity transformation to the rest of the Australian economy. There is worthwhile scope for improving the net cash outflow, but not quickly. Australia’s LCR issue is not, in the main, that our banks have too great an LCR net cash outflow. It is our lack of a deep pool of non-bank liquid assets that creates LCR difficulties.

Having noted this, we expect to see Australia’s banks chip away at their net outflows, which in aggregate and over time will make a worthwhile contribution to LCR compliance, and improved liquidity in a more general sense. We expect that banks will continue to grow retail term deposits, and continue to term out liabilities where feasible. We also expect some LCR-specific product development, such as 31 day call deposits. ASIC is currently in consultation on an LCR-friendly amendment to the rules regarding early termination of term deposits.

We also expect to see banks improve their access to new sources of term funding, which will also improve their LCR figures. The four major banks, for example, are putting in place arrangements to issue approximately $100 billion in covered bonds. To the extent that these new funds replace shorter-term liabilities, and particularly sub 30-day liabilities, then the LCR will improve.

We also see banks becoming more proactive in sourcing superannuation funding. For many years bankers have complained that superannuation trustees invest insufficiently in bank paper. Banks are now moving to offer attractive term deposit rates on superannuation investment platforms. We expect that in the next few years, banks will become more adept at sourcing term funding from super funds.

If banks cannot and should not greatly reduce their net cash outflows on the LCR calculation, can they find a lot more liquid assets? In short: no. They can find some additional assets, and perhaps encourage more such assets in the Australian markets, but in nothing like sufficient amounts to solve the LCR via direct investment.

Fortunately, the Basel Committee has recognised that alternative arrangements are necessary for countries without much in the way of high-quality liquid assets. Guy Debelle will explain the proposed RBA committed liquidity facility (CLF), so I will defer here, other than to observe that this facility is likely to feature prominently for some of the larger banks, as they meet their LCR requirement.

It is not APRA’s intent that the RBA’s committed liquidity facility should be the first port of call for banks as they become LCR compliant. Rather, banks will need to undertake all the steps they reasonably can in private markets, before filling any shortfall with the LCR. To address these issues, APRA proposes to closely supervise bank access to the RBA facility, including the amount each bank can count towards its LCR calculation. This is a delicate balancing act.

First, we intend to ensure that each bank reasonably optimises its use of Commonwealth Government Securities and semi-government securities, which are the most liquid assets in our market. But at the same time, holdings of this stock cannot allow the liquidity in these markets to be soaked up.

Then, as a general principle, we will be looking to ensure that the assets held as collateral for the RBA facility are diversified in terms of type of instrument and type of issuer. Banks will need to optimise their holdings of non-bank securities. These securities include AUD-denominated bonds from domestic and foreign issuers. Over time there may be some well-rated corporate bond issuance in Australia, and other non-bank securities. Here again, however, we need to ensure that the larger banks don’t simply buy and hold all these securities, destroying the liquid markets that Australia needs to create. We also need to ensure that we don’t inadvertently encourage banks to hold low quality credit risks in order to satisfy their liquidity requirements.

Third, banks will inevitably hold considerable assets issued by other banks, either directly as bank bills, bonds, and certificates of deposit, or on a secured basis through securitisations or covered bonds. The issue for APRA is not so much one of drying up liquidity in a traded market, but avoiding unduly large credit concentrations between Australia’s larger banks. We do not want difficulties in one large bank automatically becoming a major credit concern for all large banks.

On our calculations, even with reasonable diversification by the larger banks among the above asset classes, in many cases they will remain well short of the collateral necessary to back the RBA’s facility. As a pragmatic solution, this short-fall will be made up of self-securitised paper. Although this arrangement is Basel III compliant, we are in essence allowing banks to treat some of their home loans as self-liquidating for LCR purposes. This is a more generous interpretation than APRA would normally take in its prudential standards, and is a matter of necessity rather than preference. We would not be looking to see the reliance on self-securitisation in the collateral pool growing over time.

In addition to finding liquid assets, APRA will expect Australia’s banks to continue terming out their liabilities. As noted earlier, this can be pushed too far, and we have no intention of discouraging banks from offering short-term deposits and Australian money market facilities. However, using foreign wholesale short-term funding to cover domestic retail long-term assets has much less appeal.

The Net Stable Funding Ratio

The NSFR is, happily, a somewhat easier story. In common with many banking systems around the world, Australia’s banks are currently somewhat short of a 100 per cent NSFR. The current pro formas indicated that the larger banks have about $1.3 trillion in stable funding vs. $1.7 trillion in required stable funding, an 80 per cent ratio.

How will the larger banks move beyond 100 per cent on the NSFR measure? First, we should note that some of the arithmetic associated with meeting the LCR requirement will flow through to meeting the NSFR requirement. As a rough estimate, moving to 100 per cent LCRs would solve about a quarter of the current NSFR shortfall.

From that point, the answer is that Australia’s banks need to grow their deposits and other stable funding sources faster than they grow their illiquid assets. As a matter of arithmetic, not of policy intent, I note that seven years of 8 per cent vs. 5 per cent growth in the relevant numbers would achieve the required outcome. This is similar to the pattern of the past three years, when deposits have grown at 11 per cent, vs. credit at 5 per cent.

Basel III cost/benefit analysis

It is easy to think about Basel III as a complicated technical challenge, and it is that. More importantly, Basel III is meant to make the world a safer place for all those who rely upon or who are exposed to large, internationally active banks. This includes essentially everybody in Australia. The global financial crisis demonstrated, to those who still needed the lesson, that when large financial institutions fail, it is not only their shareholders and employees who get hurt. Sufficiently large failures can damage an economy and a society as well. Dr. Laker earlier noted some of the macro costs associated with financial crises; these are a large share of a nation’s GDP, and can blight the economic prospects of a generation.

As we survey the economic landscape in Europe and the United States, and as we remember similar reverses in other economies, one cannot help but be struck by the thought that a sound banking system is a remarkably valuable national asset. Conversely, a weak banking system is a remarkably pernicious national liability. One of APRA’s purposes is to ensure, to the extent reasonably possible, that Australia’s banks are part of the solution, rather than part of the problem, in any period of Australian economic disruption.

APRA considers that there are three major benefits flowing from Basel III. First, many banking systems in other countries will become a lot stronger, in both capital and liquidity terms. This reduces Australia’s exposure to foreign economic disasters.

Second, Australia’s banks will become moderately stronger in capital terms, and a lot stronger in liquidity terms.

Third, and this is critical in a country so reliant upon foreign funding, APRA’s proposed Basel III implementation keeps Australia in the global "A" league of well-regulated banks. Our somewhat more conservative and more rapid implementation of the capital rules continues APRA’s strategy of presenting our banks to the world as being regulated and supervised to a level materially higher than the global minimum expectation.

There are other worthwhile benefits. As the banking system comes to grips with the LCR and NSFR requirements, the swing item for compliance will be wholesale funding, particularly short-term wholesale funding, and to some extent foreign funding. Gradually reducing the banking system’s reliance upon what we now know are unreliable funding sources, will reduce Australia’s strategic vulnerability to offshore money markets.

As another benefit, bank depositors will continue to get the better deals they started receiving from around 2008. From the mid 1990s to the mid 2000s, banks typically did not see depositors as funding and liquidity generators. A more balanced and useful view of depositors should apply going forward.

All the above benefits add up to a reduced chance of Australia experiencing a financial crisis. If we should experience such a crisis, it will be less damaging than would have been the case absent the Basel III reforms.

There is no such thing as a free lunch, so substantial benefits must generate substantial impacts, and inevitably involve costs.

In Australia’s case the costs and benefits will be concentrated on the larger and foreign banks. Credit unions and building societies will be held to higher levels of risk governance, particularly for liquidity, but in the main will not need to increase their capital or their liquidity.

For the larger banks, we expect to see capital effects and liquidity effects.

The cost of preference equity

Let’s move to a simple case: the cost of preferred equity. Equity is also funding; raising equity is said by bankers to "cost" money but in fact more equity means more profits, because some funding cost is saved.

In the case of preferred equity, let’s take the simple assumptions that the pretax equivalent spread is three per cent, and the spread for senior debt of equivalent tenor is one per cent, leaving a two per cent pretax cost of preference equity vs. unsecured term debt.

This means that increasing the ratio of preference equity funding a loan portfolio by one per cent, costs one per cent times two per cent across the whole portfolio. This is two basis points; hardly an earthshaking figure.

The cost drops when risk-weightings are considered. The typical large bank’s average risk-weighting is about 50 per cent, and the relevant arithmetic indicates that each one per cent of Tier 1 ratio increase costs…wait for it…one basis point per annum.

The basic lesson here is that no matter how it is calculated, the cost of additional Tier 1 sourced from preference equity is low.

The cost of common equity

The main Basel III capital effect will be that the larger banks hold more common equity on their balance sheets, than would otherwise be the case. The reforms from the 1988 capital rules to the Basel III rules focus upon more common equity, and upon better quality capital, but do not encourage material increases in preference equity and subordinated debt.

There is no single right answer to the cost of equity, but there are three competing models of how to figure this cost, or if not the cost at least the effect on the lending spread.

The first technique is to ask: "How much additional margin would it take to maintain the lender’s current return on equity?"

The second technique is to ask: "How much additional margin would it take to maintain the shareholder’s expected return?" Which, as it happens, is close to asking what margin is necessary to maintain the current share price.

The third technique is to apply the Modigliani-Miller critique of capital structure, which suggests that no change in margin is necessary, because increased capital will be balanced by lower required returns on capital.

Let’s use the $1 trillion home loan portfolio of the major banks to try out these techniques. I am simplifying these numbers for the sake of the example.

As core assumptions, let’s say that the risk-weighting on the portfolio is 20 per cent, and the common equity target ratio applied by a bank is 8 per cent. At this capital level and after expected loan losses and other expenses, I will assume the banks generate a one per cent return on the portfolio. So we have $10 billion in pretax earnings on $16 billion in common equity, a reasonably acceptable 62.5 per cent return on equity. Here we also see the joys of a 62.5 to one leverage ratio. In real life, averaged across a bank’s entire balance sheet, we do not allow such high leverage ratios.

Now using the first technique for cost of equity, we apply 62.5 per cent. As it happens, the pretax cost of equity used by many institutions for their management accounting is around 16 per cent, so I borrow that figure here for the second technique. Modigliani-Miller helpfully comes in at zero.

So what happens if APRA moves the required common equity to risk assets ratio up by 2 per cent? I am assuming here that the funding benefit of the new equity is 6 per cent.

In the first example, the banks want to maintain a 62 per cent return, so they must increase home loan spreads by 23 basis points.

In the second example, the increase is 4 basis points.

In the third example, zero.

Which of these is the right number? Well, the first example is chosen to make the point that you need a ridiculously high cost of equity to generate anything like a noticeable change in the lending rate. APRA thinks the second cost is closer to the mark. We think that equity has a cost expressed in lending spreads, but it isn’t large in the grand scheme of things. But on the other hand, more equity in the balance sheet clearly reduces risks, so perhaps there is some pull downwards towards zero, per Modigliani-Miller.

The cost of liquidity

For liquidity, there are two costs. The obvious cost is any commitment fee paid to the Reserve Bank. The less obvious cost is the need to hold more liquid assets than would otherwise be the case, and the returns on these assets will not necessarily exceed the cost of the liabilities funding them.

We know that the RBA commitment fee is 15 basis points per annum. We also know that the larger banks need to hold about 20 per cent of their assets in liquidity, as opposed to an illiquid loan portfolio. This generates the overly simple answer that liquidity costs 20 per cent of 15 basis points, or 3 basis points across a loan portfolio.

In fact, APRA does not intend that the larger banks can simply access the RBA facility using self-securitised home loans, and avoid holding any third-party assets for liquidity. Such holdings generate interest income, but government securities typically pay less than the bank liabilities used to fund them. When APRA started its work on liquidity reform, we informally estimated that the funding cost of improved liquidity would be less than 10 basis points across the lending portfolio. I note that one of the major bank CEOs concurrently estimated a 7 basis point figure.

As we have resolved our work, and considered the effect of the RBA facility, APRA is increasingly of the view that our initial estimates were pretty good. As a rough rule of thumb, and subject to the current round of consultation on LCR and NSFR implementation, I suggest that a five basis point cost across the loan portfolio is a reasonable estimate.

This brings me to a critical issue when considering the cost of liquidity: the dominant impact of the cost of money. Up to 2006, Australia’s banks and many other good credits could borrow term funding at small spreads over equivalent government rates. During the global financial crisis, credit spreads blew out to remarkable levels, on the order of 300 basis points in some cases. Now spreads have retreated, to say 100 basis points depending upon the instrument and issuer in question. We have seen what is apparently a long-term shift in the terms of trade between fixed-income borrowers and investors. This shift in the wholesale terms of trade has also resulted in better retail deposit rates.

I note that my kerbside estimate of a five basis point cost for liquidity reform could be challenged a few basis points each way, but the cost is nothing like recent changes in the overall terms of trade for money. Five basis points is a rounding error in the context of 100-plus basis point moves in credit spreads.

Net cost and benefit

Where would this leave us overall? After Basel III, Australia’s banks will clearly be stronger. Their liquidity will be about four times better, expressed in regulatory minimum terms, than was the case in 2006. Their common equity, which has already increased since the global financial crisis, is also likely to expand gradually from now through 2016.

From my above admittedly simple calculations, let’s say that the increased cost of liquidity and the cost of equity are both 5 basis points. In APRA’s judgement, an Australia with a sounder banking banking system, which is internationally recognised for this soundness, but faces a 10 basis points increase in its overall capital and funding costs, would be pretty clearly superior to an Australia in which the banks are less sound, through lack of reform.

Any increase in funding cost does not necessarily translate into an equivalent increase in lending rates. Depending upon competitive dynamics at the time, funding cost increases will be distributed between borrowers, creditors, and shareholders.

I emphasise that the above numbers are illustrations. APRA will conduct a more detailed cost/benefit and impact analysis as we progress the Basel III reforms. We do observe from time to time in the press, the suggestion that the Basel III reforms might add something like two or three per cent to lending rates. This is simply fanciful.

Basel III produces material safety benefits, at modest impact on the cost of borrowing. Some of these impacts, moreover, are a redistribution rather than a true cost to society. Although APRA is not in the shareholder protection business, Australian banks have done pretty well by their shareholders, so a conservative prudential regime is not contrary to shareholder interests.

In summary, APRA expects that the Basel III reforms will lead to stronger bank capital positions, and much stronger and clearly necessary liquidity improvements. The Australian rules will be stronger than the global minimum rules, which troubles APRA not at all. Some of the rest of the world, particularly in the United States and Europe, are more affected by Basel III, which means eventually they will close the conservatism gap on Australia. Which will doubtless cause APRA, at some point in the years to come, to reconsider how best it might ensure that Australian banks remain among the strongest in the world.

Thank you for your attention.