APRA’s discussion paper dated 8 November 2018 has laid out the strong reasoning for requiring domestically systemically important banks (D-SIBs) to hold additional subordinated capital. In short, taxpayers should never be asked to foot the bill for a failing financial institution. Higher capital levels and detailed macro-prudential oversight are the key pillars to avoiding disorderly failures of authorised deposit taking institutions (ADIs).
As per the Murray inquiry, Australia’s banks are partially dependent upon offshore funding and therefore must be “unquestionably strong”. By requiring ADIs to hold more subordinated capital the likelihood of one becoming distressed greatly decreases. By having a larger capital buffer protecting senior creditors, the likelihood of a distressed ADI failing in a disorderly fashion is also greatly reduced. APRA’s proposed method of reducing these risks is pragmatic and entirely appropriate for the Australian banking context.
This submission answers several of the questions posed by APRA in its discussion paper. It argues for (i) faster implementation of higher capital levels, (ii) the necessity of using tier 2 capital, (iii) the implementation of a total capital leverage ratio in conjunction with a total capital risk weighted ratio and (iv) for additional macro-prudential oversight of D-SIBs. This submission does not repeat the key reasons for the necessity of the changes. Narrow Road Capital laid out the key arguments for these changes in 2016 and APRA has made similar arguments in its discussion paper.
Timing for Implementation of Higher Capital Levels
The global economy is now ten years on from the global financial crisis and is exhibiting many late cycle characteristics. With good reasons, many experts are concerned about Australia’s record high levels of household debt. These two factors combined point to the need for APRA to lift the countercyclical capital buffer from its current zero setting to at least the middle of the 0 to 2.5% band. An alternative to this would be for APRA to order D-SIBs to implement at least a 2% (risk weighted) increase in subordinated capital levels by the end of 2019, front loading the implementation of the TLAC reforms.
It is also worth noting that APRA has unnecessarily delayed the release of its discussion paper. By 2016 it was clear that TLAC reforms would require banks to substantially increase their subordinated capital levels, with some regulators having announced their positions. Instead of being on the front foot and potentially adjusting the final capital levels over time, APRA has chosen to delay until almost all other banking regulators have announced their positions. This is not a hindsight criticism, Narrow Road Capital publicly argued the case for action in 2016 including sending a copy of its work to APRA. Asking D-SIBs to take substantial action in 2019 would go some way to correcting the delay.
Necessity of Using Tier 2 Capital
In the discussion paper APRA has argued strongly that the best form of additional subordinated capital, which balances both risk reduction and cost efficiency, is tier 2 (subordinated debt) capital. Immediately upon the release of the discussion paper vested interests began campaigning for Australia to introduce tier 3 (senior subordinated) capital. APRA should stand fast in its position, giving greater weight to (i) financial stability and (ii) the interests of taxpayers rather than the vested interests of banks seeking to lower their cost of funding.
Tier 3 debt is a wolf in sheep’s clothing. It may seem more innocuous than tier 2 debt, but if a bank becomes distressed it will have a very similar outcome. Tier 3 debt can have a shorter maturity period than tier 2 debt, but it is subject to the same regulatory approval for principal and interest payments. If a bank is distressed, its regulator would lock-up its tier 2 and tier 3 debt, delivering the same outcome to holders of both. If a bank becomes insolvent both tiers are likely to be facing haircuts or wipe-out, though tier 3 may receive a better recovery rate. Markets have yet to accept this reality, but once the first test case occurs markets will correct, and the margin gap will narrow.
The main (false) concern raised by proponents of tier 3 debt is that they don’t believe capital markets can digest the increase in tier 2 debt that is implied by APRA’s discussion paper. Estimates are that the four major banks will, as a group, need to issue approximately $20 billion of tier 2 debt per annum. There are several reasons markets are well placed to handle this level of supply.
First, as tier 2 margins are now higher following the release of the discussion paper, the incentive for investors to switch from senior debt to subordinated debt has been sweetened. Second, investors will recognise over time that Australian tier 2 debt represents very good value as it has substantial equity and preference share capital providing protection and doesn’t have tier 3 debt diluting its position. Third, increasing the amount of tier 2 capital is likely to see margins on senior debt reduced, offsetting a material portion of the increased cost of funding. Fourth, capital markets have a long history of finding solutions to business capital requirements. If the price is right, the demand will be there for it. The combination of offshore demand, local institutional demand and local listed note (ASX) demand should comfortably cover the forecast supply at a reasonable margin. APRA should ignore the vested interests and prioritise financial stability, requiring the additional capital to be tier 2 debt.
Requirements for Increased Capital and the Impact on Regional Banks/Smaller ADIs
APRA has left open what it will require from regional banks and smaller ADIs. It has been noted that non-D-SIBs may not be required to hold similarly high levels of capital. This may be due to some non-D-SIBs having a lower risk profile or it may reflect concerns relating to competitive positioning.
For non-D-SIBs, APRA should encourage the use of securitisation to reduce on-balance sheet risk. Where ADIs selldown the entire capital structure in a securitisation transaction, full capital relief should be granted. This would allow these institutions to recycle their capital, instead of looking to create exotic subordinated capital instruments that align with their not for profit ownership structures.
Necessity of Having a Leverage Ratio as well as a Risk Weighted Ratio
The current position of Deutsche Bank is the perfect example of why a risk weighted ratio alone is insufficient. Deutsche Bank’s heavy use of low and zero risk weighted transactions means that it can argue it is well capitalised on a risk weighted basis. However, its miserly market capitalisation to book value ratio shows that the equity markets see its position as mildly distressed. Whilst Australia doesn’t have any banks that are gaming their ratios to the extent that Deutsche Bank is, APRA should specify a leverage ratio to ensure that this doesn’t occur.
The discussion paper points to D-SIBs being required to have a total capital to risk weighted assets ratio of 18.5% to 19.5%. This should be supplemented with a minimum total capital leverage ratio of 8%.
As noted earlier, higher capital levels and detailed macro-prudential oversight are the key pillars to avoiding disorderly failures of ADIs. Regulators should regularly visit banks and review a sample of their loan books to ensure that lending standards are maintained. In crisis after crisis, banks have shown they cannot be trusted to maintain conservative lending standards through a credit cycle without such oversight.
The best example of a regulator reviewing loan quality is the US Shared National Credit program, run by the Office of the Comptroller of the Currency (OCC). This program focusses on corporate and institutional lending, the area that most frequently leads to bank failures. Each year the OCC inspectors review large loans and assign each loan a risk grade. For higher risk and defaulted credits, the regulator specifies the minimum amount of capital the bank must hold against the loan.
By reviewing the same loans and lenders year after year, the OCC gains enormous insight into whether lending standards are improving or deteriorating. When they are deteriorating, the regulator can take action including increasing the amount of capital held against high risk loans or stopping banks from lending to high risk borrowers altogether. The 2014 guidance to banks to limit the total debt to EBITDA ratio at 6.0 times is an example of an appropriate response to deteriorating lending standards.