A calculated risk for banking
International banking regulators yesterday decided to press ahead as planned with the new Basel III prudential reforms despite intense lobbying from the international banking sector for delays and changes to the new rules. The eurozone crisis has illustrated some of the complexities and challenges those reforms will generate, for banks and their governments.
One of the key elements of the reforms is the imposition on the banks of a requirement to hold more and higher quality liquidity, sufficient to tide them through at least a month of acute stress. The focus is on highly-rated sovereign bonds and bills, although highly-rated corporate debt and covered bonds also qualify.
There are two potential problems with the range of securities deemed eligible liquidity for banks and the basis of their eligibility. One is that ratings, as we've seen consistently, are a lagging indicator – they change after it is well and truly obvious that credit has deteriorated – and the other is that the eurozone experience, and perhaps even the experience in the US, signals that developed world sovereign debt isn't risk-free, or even low-risk.
Banks have always been required to hold reasonable levels of liquidity as insurance against the mismatch between their assets and liabilities created by their core activity of maturity transformation, or borrowing short and lending long.
Before the global financial crisis, sovereign debt issued by the major developed economies was regarded generally as risk-free or, at worst, very low risk. So banks held considerable amounts of sovereign debt, particularly debt issued within their own core jurisdictions. The flight to liquidity and security triggered by the crisis saw bank holding of sovereign issues rise further.
What the crisis also revealed – and the eurozone crisis has illustrated dramatically, and intensified – is the relationship between sovereigns and their banks.
In Europe and the US, the already toppy levels of public sector debt built up during the balmy decade that preceded the crisis were significantly increased by the massive bailouts of financial institutions and the big fiscal stimulus packages that were an attempt to blunt the impact of the financial crisis on real economies.
The legacies of the pre-crisis profligacy and the response to the crisis itself are unsustainable levels of public debt in Europe and the US – and European and US banks stuffed full with massive holdings of government paper.
In the eurozone, that has created two deeply inter-related crises – a sovereign debt crisis entwined with a banking system crisis. If the eurozone were to splinter, deep holes would be gouged in bank balance sheets across the region, requiring more massive taxpayer bailouts from economies that can't afford them.
For the moment, the European strategy appears to be to use the European Central Bank to pump near-costless liquidity into the banks in the hope they will then invest the funds in sovereign debt, putting a floor under demand for the debt, a cap on its price and stabilising their own balance sheets in the process. The strategy doesn't seem to be working.
There is another dimension to the inter-relationship between sovereign debt and banks. The ratings agencies are now explicitly factoring into their assessments the extent to which governments have the capacity and will to support their banks. There are, therefore, a set of circular relationships between the sovereign debt rating, the rating of the banks within that economy and both the public and private balance sheets and costs of funds.
In Europe and the US, unlike Australia and Canada, the bank regulators have been part of that problem.
The regulators in Europe explicitly designate sovereign debt issued by eurozone members as having zero risk weights, as does the US in relation to its own issues. The Europeans also exempt sovereigns from the normal limits on banks' large exposures to individual risks.
In a sense that approach is understandable. Why would regulators concede there was even the remotest possibility that there might be a default on their own sovereign debt? Until the crisis destroyed the illusion of risk-free sovereign debt, the explicit acknowledgement by a regulator of even the slightest of risks might itself have had adverse financial and economic impacts.
In Australia and Canada banks aren't allowed to treat sovereigns differently to other risks – they have to assess the individual risk latent in all the securities they own and hold capital against the perceived risk.
The major Australian banks were early movers to the Basel II internal ratings approach, where it is their responsibility to calculate all their specific risks and provide for them rather than adopt the broad-brushed standardised approach that used to be the norm.
The problem for the eurozone banks is that a proper approach to their sovereign holdings would be to provide against both the ultimate risk of default – which is real in a number of eurozone states – and the losses in the value of their bond holdings that have developed in line with the crisis as sovereign credit spreads have blown out.
A significant proportion of the sovereign debt holdings of eurozone banks aren't marked to market – only the securities within their trading books impact their P&L statements – unlike the treatment used for most of their banking activities.
As the system moves towards Basel III over the course of this decade – and the liquidity requirements are among the earlier changes – it is obvious that the Eurozone and US banks will first have to deal with the legacy issues created by their existing holdings, which will require more capital to support their continuing sovereign exposures and cover their losses on government paper as well as to meet the significant increases in general capital requirements under Basel III.
There will also be an issue – which the Australian banks, the Australian Prudential Regulation Authority and the Reserve Bank have been grappling with – of finding sufficient high-quality liquidity while minimising the amount of capital required to be held against that liquidity now that it has been demonstrated that the concept of sovereign issues by developed economies being risk-free is a myth.
The probability of a default by, say, AAA-rated Australia, might be so remote today as to be almost inconceivable, but it isn't zero.
The relative dearth of available Australian government securities for the banks to hold as 'level one' liquidity caused APRA and the Reserve Bank to create arrangements of last recourse, under which the banks could get liquidity – at a significant cost and after lodging eligible collateral – in an emergency.
Like banks elsewhere, they would no doubt argue that introduction of the Basel III liquidity regime ought to be delayed and the range of securities that qualify as ‘high quality' should be expanded in the light of what has happened in Europe, the redefinition of risk-free securities that has occurred (that there is no such thing) and the adjustment to bank balance sheets and liquidity strategies that will have to occur as a consequence.
In this market, however, those kinds of pleas have been falling on deaf ears. APRA has proven itself ahead of the regulation game and is determined to stay there and the international regulators, deeply conscious of the extent of the eurozone crisis, have ignored the pleas for delays and still recommitted themselves to their process.
It may be easier for the Australian banks to meet the new liquidity and capital requirements than it will be for most of their global peers, but given what has occurred and is still occurring around them it won't necessarily be easy and it certainly won't be costless.