This probably would have been more useful before July 23, but oh well.
Much like the US stress tests conducted in the spring of 2009, the European version possessed a couple of pretty significant loopholes for participating banks. One has to do with accounting (think FAS 157 in the US), the other with balance sheet management.
Accounting - As noted previously on FT Alphaville, the CEBS only stress-tested assets on banks’ trading, or Available for Sale (AFS), books. Those classified as Held to Maturity/Loans & Receivables were exempted. A quick reclassification of sovereign exposures from AFS to HTM would have helped banks significantly in the stress tests.
This is something that probably would have been happening within banks anyway, as their sovereign exposures deteriorated at the start of this year. Some examples we know of include National Bank of Greece, which seems to have switched a portion of its Greek government bonds to HTM early in 2010, while Commerzbank reclassified some of its Public Finance portfolio to loans and receivables in Q1.
Balance sheet: What’s more, European banks knew they were being assessed on the basis of their Tier 1 capital ratios — which meant they knew which regulatory number to window-dress.
This was once a criticism of the (now-exalted) US bank tests.
From Richard Bove — a man whose banking notes come with a warning:
There can be three results from these tests none of which are positive:
- * The regulators can claim that the test has successfully demonstrated that the banking system is not in jeopardy. This is, of course, what happened in the European test when it was discovered that 7 small banks failed out of the 91 tested. This result satisfies no one and it leads to a belief that the test is merely a public relations stunt or a cover-up.
- * A second result might be to suggest that there is a major problem – e.g., some 25 to 30 banks have failed and 3 to 4 countries are in danger of default. To claim this would create a global financial crisis which could easily end up in a decade long Depression. It is not an option no matter what is discovered.
- * A third option is to create some convoluted ratio and call it Tier 1 Capital or some such silly name. Tell every bank how this is calculated and then tell the bank it will be evaluated based upon this metric. The banks will then adjust their balance sheets so that they pass the test. In the particular case of Tier 1 Capital the way to do this is to dump loans and add cash. This changes the risk weighted assets or the denominator in the calculation of Tier 1 Capital and allows the banks to pass the stress test. (A review of American bank balance sheets shows how dramatically they have adjusted to meet this country’s stress testing.)
The third option is generally put in place. The result can be seen in the chart on the next page concerning American banks that have been stress tested. Loans plummeted. Not shown is the fact that cash is soaring in the American banking system. Thus, the result of the stress test is to lower bank lending and through this the money supply and the through that economic activity.
Oof. Let’s hope not.
Related links:
Accounting stress – FT Alphaville, 2009
Accounting for European stress – FT Alphaville
Stress test tension and the question of converting capital – FT Alphaville, 2009
Stress tests’ sovereign support = senseless – FT Alphaville