Europe’s latest bank stress test fails to impress
Not testing hold-to-maturity books or liquidity risk diminishes credibility; Sterner tests conducted in private might be a better use of time
On being confirmed in March as the first executive director of the European Banking Authority, Adam Farkas, former head of the Hungarian financial supervisory authority, declared: “The success of the EBA will be key to enhancing the financial stability and soundness of financial institutions and ultimately to rebuilding the trust of citizens in the European financial system.” Farkas must know that the authority, founded in January this year, has not got off to an auspicious start. The EBA, the collective of European national bank regulatory bodies, has just begun its latest stress test of banks accounting for 65% of total assets in the EU banking system. Results will be published in June and will gauge banks’ ability to withstand hypothetical stress events including, at worst, an economic contraction of 0.4% of GDP in 2010 and zero growth in 2011 across the EU accompanied by rising unemployment, higher interest rates on government bonds, a sovereign debt shock, a 15% stock market sell-off, rising overnight and three-month money market funding rates and falling house prices.
The last European banking stress test was generally perceived to be a damp squib. The market initially reacted favourably to the benign outcome when results were published in July 2010, with just seven out of 91 tested banks deemed to have failed the test with a combined capital shortfall of €3.5 billion. European bank credit default swaps rallied by 20 basis points in the week following announcement of the test results. “There was a dearth of news at the time, and the initial panic around sovereign debt had died down,” recalls one FIG DCM banker. “So while investors had initially been sceptical of the tests, they reacted positively.” But this didn’t last long. Within months the Irish banks that had passed the test required bailouts and the credibility of the 2010 exercise was for ever tarnished, especially in comparison to the US bank stress test of 2009 that had required 10 of the country’s biggest banks to raise an additional $75 billion in capital.
The 2011 European tests began against the worrying background of a disappointing outcome of the Ecofin summit at the end of March, with delays to the capitalization of the European Stability Mechanism by member states, and amid renewed worries over Portugal’s ability to roll over its finances without EU and IMF assistance.
The initial assessment of the latest stress test is that it looks likely to repeat the mistakes of the past by setting out to produce a calming result, suggesting that the banking system can largely withstand the adverse conditions described, and so ensuring that the test lacks the very credibility it needs to have any impact on markets.
Most analysts are damning, few more so than Hanz Lorensen, Matt King and Michael Hampden-Turner, credit analysts at Citi, in a note to investors. “Sure the economic scenario seems mostly fairly severe [but] the translation into market impact is hardly severe. Equity prices are assumed to fall by 15% – the DAX has just fallen 12.5% in three weeks. Valuation haircuts on sovereign bonds are very small, especially at the front end where banks hold much of their exposure.”
It’s perhaps unfair to say that the EBA has already constructed a test that most banks will pass. It has not declared what capital ratio banks must show they can maintain through these stress conditions in order to achieve pass marks. One banker tells Euromoney: “This test is at least one turn of the screw tighter than the last because banks will be measured against the capital they have now, not against capital ratios they intend to maintain in future to meet new regulatory requirements. Also remember that the last test did at least flag the importance of recapitalizing the Spanish savings banks, a process now well under way. Investors may need to think of a series of annual tests with each chipping away at the weakest links in the system.”
But scepticism rules the day. The Citi analysts conclude: “Quite simply you can fail individual banks, but you can’t fail the banking system – at least not without a much more effective, and less political, recapitalization mechanism. So the stress test must reflect what the system can take and, judging by these parameters, that’s still not very much.”
There are two key omissions from the test. It applies sovereign shocks only to government bonds held in banks’ trading books and not to their much larger exposures in hold-to-maturity books. The tensions around Portugal now are focusing attention on this shortcoming, as market pricing indicates steeper discounts already being assumed than those envisaged in the most adverse conditions in the test.
Meanwhile continuing tensions around haircuts to senior unsecured creditors of Irish banks focus attention on the second key omission. The EBA clearly states: “Liquidity risk is not specifically assessed as part of the stress-testing exercise.” Most analysts agree that this is a big disappointment.
It seems that the EBA is stuck in an impossible position. If it constructed the test that investors and analysts really want, this could be seen as an explicit admission by the official sector that Europe faces a solvency crisis at the sovereign level rather than a liquidity crisis and that this might engulf the banks once more. The fear might be that such a test would itself trigger a collapse of confidence in the financial system.
Perhaps a better procedure would be to conduct more severe tests in private, so that regulators and government might at least gauge the scale of the challenge should the worst begin to unfold.
Simon Samuels, banks analyst at Barclays Capital, sees little utility in watered-down tests. “I’m not sure that these will be a turning point for the sector. Rather, they are something of a sideshow. It may sound strange to hear an analyst calling for less transparency rather than more, but it really might be much better if the European bank regulators conducted these stress tests privately, rather than in the public glare. This is what happens regularly between regulators and banks such as in the US.
“Private tests would allow for much sterner stress scenarios to be examined, perhaps involving multiple sovereign government defaults. That would at least allow banks and regulators to assess potential damage to their hold-to-maturity books and overall balance sheets in such a calamity. It wouldn’t necessarily matter if most banks failed such private tests, because regulators might in turn attach an extremely low likelihood of the scenario happening.”