An IMF tick for Australian banks

It is curious how some ‘'research in progress'' by a couple of International Monetary Fund researchers has morphed, in local reports, into a full-blown warning from the IMF that the major Australian banks need to raise more capital if they are to withstand the impact of a combined residential property market collapse and corporate loan losses.

While the paper is an IMF working paper, it doesn't represent the IMF's views and is more of a position paper designed to encourage discussion than anything definitive.

And, while the paper does conclude that ‘'more robust capital levels for systemically important banks'' would be beneficial, particularly in times of market uncertainty, the paper in its entirety is actually a glowing endorsement of the major Australian banks and their regulator's conservatism.

The paper correctly identifies the main potential vulnerabilities of the big banks: their exposure to highly-indebted households through their mortgage lending and their sizeable offshore borrowing requirements.

It also, however, notes that the banks are strongly positioned, with conservative lending practices, robust supervision, a low non-performing loan ratio and small exposures to high-risk mortgages because of relatively conservative loan-to-valuation and debt-servicing ratios. Most of the mortgage debt is held by higher income households and the full recourse nature of the lending helped limit loan defaults, the researchers said.

The Australian banks have more and higher quality capital than most of their international peers – the Australian Prudential Regulation Authority's approach to calculating capital adequacy is significantly more conservative than even the Canadian system, the other banking system that came through the 2008-09 crisis relatively unscathed. APRA requires the banks to assume higher notional loss rates when risk-weighting their assets than the Basel II rules require and has a tougher definition of eligible capital than most other jurisdictions.

If the Australian banks were able to use the Canadian approach to risk-weighting residential mortgages, the paper says, their total capital ratio would rise by more than 120 basis points and their tier one ratio by about 100 basis points. If other differences, notably the eligible capital were factored in, Westpac has estimated, the paper says, that its common equity ratio of 8 per cent in March last year would have been 13 per cent under Canada's rules.

At the heart of the paper is an interesting exercise. The researchers factored the experience of the Irish banks' residential mortgage experiences during the financial crisis into the Australian banks' balance sheets. Ireland, of course, saw 13.6 per cent unemployment in 2012 and a 46 per cent fall in house prices from their 2007 peak. The Irish banks also had, compared with the Australian banks, high loan-to-valuation ratios.

For the purposes of the paper, the researchers assumed the share of the three riskiest categories for residential mortgages for the four majors would rise to those of the Irish banks, which would increase the banks' probability of default from four per cent to 11 per cent, generate losses larger than their existing loan loss provisions and eat into their capital bases.

Despite that reasonably extreme scenario, the banks' tier one capital ratio would fall by only 1.5 percentage points and would remain well above the regulatory minimum. Even if the assumed loan defaults and risk-weights were increased 1.5 times (which the researchers said was unlikely) one bank would have a total capital ratio of six per cent, but the other three would remain above eight per cent.

So, a housing crisis by itself doesn't blow up the system. What about a housing crisis and a spate of corporate collapses? Well, if the estimated peak losses from the corporate sector of about six per cent of their loans to corporates during the crisis were to be repeated, it would, in association with the hypothetical residential mortgage lending losses, cost them about two percentage points of their total capital ratio, which would fall to about seven per cent. They would – as they did during the crisis – need to raise some more capital.

As it happens, APRA – with some pointless resistance from the banks – plans to implement its (conservative) version of the new Basel III regime as it applies to capital adequacy on a far more aggressive timetable than international regulators have recommended. It sees the virtue in the Australian system maintaining its reputation for prudential soundness and having some insurance against another bout of crisis, even if there is a cost to the banks shareholders and/or customers.

While a new crisis could, given the state of the eurozone, emerge abruptly, the strong condition of the banks, which are holding excess capital and liquidity, and the potential for the government funding guarantees to be revived, means the system should be able to cope. A residential property meltdown would take much longer to unfold, by which time the banks should have even more capital and liquidity.

Far from being alarmist, the paper is actually an endorsement of the strength of the Australian system. While it might use the Irish experience as a mechanism for stress-testing the Australian system, it is apparent from the commentary in the paper that its authors recognise that Australia is no Ireland.