Bank stress tests - Graham's blog for 8 April

Is there a basic flaw at the heart of the bank stress tests?  Will Portugal's news just highlight it?

• EBA says it will test trading book sovereign exposure --- but we all know that the real problem is the obligations held in the banking book
• If the governments demand banks have contingency plans for failure ---will this be just on the reported test? Therefore not including the assumptions that the market will make by reading across to the disclosed banking book holdings?
• The stated exposures will be measured by the markets against the BIS data for say German Banks exposure to say Greece and if there is a huge gap between the stress test “full disclosure”  and BIS data, then the market will ask why is this, who holds it and why are they not revealing it???
• If there is a real fear that the reason is the numbers are too grim and the governments are turning a blind eye, why should banks continue to lend in the interbank market to banks in states that are hiding from the truth??? i.e. this heightens the risk of an interbank run on many banks in states perceived as weak.
• The GDP-type stress calculations are based on Commission autumn 2010 forecasts… but I cannot see that they have done a re-run based on the cumulative 4% decline in expected GDP (so modest actual falls for eurozone). I wonder what the expected budget deficits would look like and how that would feed through into market expectations about various members states finances???
• Then there is the question of exposure by banks to other banks that fail the stress tests. A real risk that the failures will have credit lines cut pretty quickly by those who feel they themselves could be at risk.

Contagion risk is clear and the real elephant in the room is the banking book sovereign exposures. If these are fudged – yet again – the market is likely to draw conclusions.

Extracts from EBA methodology

The adverse scenario, designed by the ECB, is more severe than the 2010 CEBS’ exercise in terms of deviation from the baseline forecast and probability that it materialises. It includes a marked deterioration in the main macro-economic variables, such as GDP (which falls [actually this is cumulative over the period] four [is this going to be flowed over into the public finances forecast???]percentage points from the baseline compared to three in the 2010 exercise), unemployment, and house prices. The adverse scenario also includes a specific sovereign stress in the EU, leading to further falls in the price of some EU bonds from the already stressed levels seen at end 2010. The sovereign haircuts will apply to positions in the trading book where losses would materialise, and will be accompanied by full disclosure of all relevant [how broad will this actually be????]sovereign debt holdings.

The capital threshold will be focused on a definition of core tier 1 capital which is more restrictive than the tier 1 threshold used last year. The EBA is currently defining common criteria for core tier 1 capital that will be applied consistently across the EU.

For more blogs please click here