APRA's Basel III implementation transition, pillar 2 and supervisory issues

Wayne Byres, Executive General Manager - APRA Finsia Workshop, Sydney

Good morning everyone, and let me please add my welcome to that of the previous speakers.



As you can tell from the title of my presentation, my time this morning will be spent on something of a ‘grab bag’ of issues, mainly in relation to capital. But that reflects some of the new features of the regulatory framework being introduced by Basel III, and the need to set out how they will interact with aspects of the existing framework that we think work quite well. In the time I have this morning, I’m going to outline our thinking in relation to:

• the implementation timetable;
• the transitional arrangements for non-complying capital instruments;
• the interaction between Pillar 2 and the capital buffers; and
• Pillar 3.

I’ll offer a few comments on a couple of supervisory issues in relation to implementation at the end of my remarks.

Starting point for implementation

Let me start by repeating some of the material Charles presented earlier. The Australian banking system is already well placed to meet the new Basel III capital requirements: the QIS results for a selection of larger banks show a Common Equity Tier 1 ratio of 6.9% at end June this year. That position is calculated including:

• the fund deduction of all of the new regulatory deductions introduced by Basel III, i.e. no allowance for transition is included;
• the inclusion of the national decisions proposed by APRA, e.g. the continuing deduction of all DTAs and equity investments; and
• continuing with other areas of national discretion from Basel II, e.g. capital requirements for IRRBB and the minimum LGD for residential mortgages.

The results for smaller ADIs will be even more favourable, given their stronger starting capital ratios under Basel II and generally lower level of deductions.

These results should not surprise anyone, since Australian ADIs have traditionally held a higher quality capital base than many of their offshore peers.

The global reform agenda, however, has been driven by the need to substantially overhaul the minimum international requirements, which many countries had adopted. The Basel Committee, faced with a choice of adopting high standards or fast implementation, sensibly and quite rightly chose to aim high but provide a more leisurely timetable for banks to move to compliance. So we see, for example, that although there is a great deal of focus on the 2013 introduction date for Basel III, the only requirements being introduced by the Basel Committee are a CET1 requirement of 3.5% and an increase in the minimum Tier 1 ratio from 4% to 4.5%. And even this is calculated with generous phase-in arrangements for the new regulatory deductions – so generous that none are implemented in the first year! Having learned the lessons of the financial crisis, these are not onerous requirements. The newer parts of the capital regime – the buffers – do not begin to be phased in until 2016. But – and this is important – the Basel Committee said ‘move earlier if you can’.

Reflecting the strong starting point in Australia, APRA has decided to adopt a quicker, and much simpler, implementation timetable. In short, all of the basic components of the minimum Pillar 1 requirements are introduced in full from 2013. Very simple. There is no need to fiddle around with phase-in arrangements – with the exception, as Charles noted earlier, of non-complying capital instruments, and I will say more about those on my next slide. We will have a constant set of minimum capital requirements from January 2013, and the main new feature of the capital framework – the capital conservation buffer – will come into force in full from 2016. That means that pretty much all of the Basel III capital regime will be in place from that point in time.

To repeat the comments made earlier by John and Charles, we are not alone in moving early and see this as a virtue that the Australian banking system can use to its advantage. We also see our approach as simple and easy to understand. In a regime which is otherwise increasingly complex and technical, some moves towards simplicity are no bad thing.

Non-complying capital instruments

One big change that Basel III will make is that non-equity capital instruments will become genuinely loss-absorbing, and a much more permanent feature of ADI’s capital structure, than some of the debt-like instruments that currently receive regulatory recognition. These tighter requirements on capital eligibility mean that virtually all non-equity capital instruments issued by Australian ADIs prior to January 2013 will be non-complying from that date. This is because many include incentives to redeem – such as step-ups combined with call options, which create an expectation that the instrument will be redeemed – and virtually all will lack the conversion trigger associated with non-viability.

These instruments were, however, issued in good faith and with APRA approval under the rules that have applied up until now. So it would be unfair and unnecessarily penal to exclude all of these instruments from regulatory capital immediately from the beginning of 2013. Instead, the Basel Committee proposed, and APRA has adopted, a lengthy phase-out period of these instruments which potentially extends out to 2022. There are, however, some important caveats to be eligible for this transition relief.

• instruments are only eligible for transition until their first available call date – if an ADI can get rid of the non-complying instrument, the loss of eligibility for transition means that it has incentive to do so;
• there is a cap on the eligibility of the portfolio of instruments, which is fixed throughout the transition period, set at 90% of outstanding stock as at 1 January 2013 and reduces 10% each year thereafter; and
• there are separate caps and amortisation schedules for Tier 1 and Tier 2 instruments.

As a result, the actual portion of any instrument that is eligible will be determined by overall maturity profile of the ADI’s stock of non-complying capital instruments. Each ADI will effectively have a forward projection of the run-off of their non-complying capital instruments – with separate projections for Tier 1 and Tier 2 instruments. From the charts shown here1, which show hypothetical run-off schedules, you can see that although by definition each bank will have an initial period where some portion of their capital instruments will not be eligible for inclusion in their capital base, it is possible that, for a large part of the time, ADIs will get full credit for their non-complying instruments.

Adjustments to minimum Pillar 1 requirements – Pillar 2 and the buffers

Under the Basel III framework, there are three types of adjustments that will be made to the Pillar 1 capital requirements, and which will therefore influence the capital levels that ADIs will actually target:

• Basel III introduces the capital conservation buffer. It is a permanent feature of the regime, applied to all and (mostly) applied at a constant size to all. Its workings will be set out in the Prudential Standard;
• Basel III also introduces the countercyclical buffer, which works by extending the size of the capital conservation buffer in times of high system credit growth. As such, while applied to all, it will vary in size over time. It will be announced by APRA when switched on and off; and
• finally, each ADI may have a Pillar 2 adjustment. A Pillar 2 adjustment is a supervisory adjustment to the minimum (Pillar 1) capital requirements to take account of institution specific risks. Unlike the other two adjustments, Pillar 2 adjustments are entity-specific, will vary through time, and will not be disclosed.

Unlike the two buffers, Pillar 2 is not a new feature of the capital regime. Pillar 2 has been actively applied by APRA for some years. Importantly, Pillar 2 forms part of the binding capital requirement: it sets the hard floor.

APRA’s general approach to Pillar 2 will not change, but given the much higher minimum Pillar 1 requirements being introduced by Basel III, and the introduction of the buffer regime (about which I’ll say more in a minute), we will need to make some changes to the way we apply Pillar 2:

• current Pillar 2 add-ons will need to be recalibrated (typically they will be lower in size, and my guess is that they will be applied less frequently than they have in the past);
• we will need to redesign them to fit with the buffer regime; and
• we will apply Pillar 2 primarily to CET1 (typically with constant adjustments to Tier 1 and Total capital).

One thing we will not change, however, is the confidentiality of the Pillar 2 regime. Pillar 2 adjustments will remain the subject of discussion between APRA and the ADI concerned only. Confidentiality is critical to avoiding the adverse signalling effects which would otherwise limit the effectiveness of the regime.

This chart2 is relatively straight-forward and shows how the various components stack together. Each ADI will have a stock of capital which should, hopefully, exceed its Pillar 1 requirement by a considerable margin. Coupled with any Pillar 2 adjustment that may be applied, we have the ‘hard floor’ capital requirement. On top of this are the regulatory buffers – the capital conservation buffer and, in times of strong system credit growth, the countercyclical buffer – and then finally the buffer the ADI chooses to hold to ensure it remains in excess of all the regulatory triggers.

Despite this apparent simplicity, the workings of Pillar 2 and the buffers introduced by Basel III has caused APRA to think hard about how they best interact. In this area, there is no international consensus. Despite the fact that many of you think that we love to layer capital with more capital, we are conscious that the minimum (i.e. Pillar 1) Basel III requirements are substantially higher, and simply stacking the buffers on top of Pillar 2 requirements may not always be the right answer. So we have reserved the right, when we apply a Pillar 2 add-on, to reduce the capital conservation buffer by up to a corresponding amount. Importantly, the minimum CET1 requirement plus the capital conservation buffer will always be 7% or more.

While reducing the buffer might at first seem to be a weakening of the buffer requirements, the interaction with Pillar 2 add-ons means the effect is always to raise the level at which the earnings distribution restrictions kick in. The examples shown in this slide3 can be worked through at your leisure, but the main point is that for any row in the table, as the regulatory requirements increase from left to right, the threshold at which the distribution restrictions kick in is higher.

Pillar 3

Let me say a few words quickly on Pillar 3.

Charles has already made the point that we do not accept the argument that international consistency is always an unqualified good. Australian banks have long argued that APRA’s ‘difficult rules’ create difficulties for them because ‘lazy analysts’ just look at the headline number and don’t adjust for APRA’s more conservative calculation methodologies. Our response to that is to say that if there is a transparency problem, it shouldn’t be fixed by giving away important prudential principles. Rather, a transparency problem should be fixed by improved transparency.

With that in mind, the Basel Committee is working on improved disclosure templates which will provide better visibility of standard Basel III vs local ratios. I don’t wish to over-promise here – these templates will not solve all issues, since they will focus on the numerator of the capital calculation and not the denominator – but they should substantially improve the understanding of the impact different local rules have on reported capital ratios. There is also scope for APRA to improve on this when we consult on our local Pillar 3 requirements.

Supervisory matters

I’ll finish now with a final few words on supervisory work leading up to Basel III implementation.

We will be asking banks to develop both capital and liquidity plans to help APRA understand how capital and liquidity positions will evolve over the next few years:

• to help with capital planning, we expect to advise ADIs of their revised Pillar 2 add-ons around the end of Q1 2012;
• during 2012, we will also need to agree with each ADI the treatment of all non-complying capital instruments; and
• in the case of liquidity profiles, we also need to upgrade our review of annual funding plans to understand the extent to which ADIs will need to make use of the RBA committed liquidity facility. AS we have already noted, ADIs are expected to do what they reasonably can to manage their liquidity to minimise their call on the RBA. For example, if an ADI plans to grow its assets with significant reliance on short term offshore funding, and to rely on the RBA facility to provide the necessary liquidity, I would suspect they will find their reliance misplaced. One feature of the new regime is that asset growth will need to be funded with longer term and more stable liabilities than has been the case in the past.

We’ll also be continuing with on-going quarterly data collections from the larger banks to allow us to monitor the impact of the proposed reforms, as well as the trajectory on which ADIs are travelling towards the new requirements.

Finally, we will need to design, and all ADIs will need to implement systems to provide, new quarterly data collections. Our immediate priority will be capital reporting under Basel III, which will come into force in March 2013.

So along with the finalisation of the policy regime, for which we are in the midst of extensive consultation, there is a lot of work to do to operationalise the Basel framework. But while time is short, we are well placed to get the job done.

Thank you.