The duck test is a paradigm of inductive reasoning. If, as Douglas Adams chirpily wrote, “it looks like a duck, and quacks like a duck, we have at least to consider the possibility that we have a small aquatic bird of the family anatidae on our hands.”
The sitting duck test, to coin a phrase, involves inductive reasoning of a different sort, as Citigroup explains in a research note on US banks’ European exposure published Friday evening. “The US seems like a ‘sitting duck’, unable to directly respond,” write economists Steven C Wieting and Shawn Snyder.
So if it looks like dangerous peripheral exposure and walks like dangerous peripheral exposure, it probably is dangerous peripheral exposure, right? And how big is it?
Citi takes as its starting point the IMF’s €300bn ($405bn) estimate for the “macro-level capital hit” to European banks from the deterioration to date in assets issued by Greece, Ireland, Portugal, Spain, Italy and Belgium, and their banks. (The €200bn figure we discussed last week didn’t include losses associated with declining bank asset prices.)
As the IMF noted in box 1.3 of the underlying report (and Citi reiterates), this estimate is neither exhaustive nor the result of the sort of detailed estimate that should be found in stress tests.
But it’s an esimate nonetheless — one that Citi uses to size up US banks’ total exposure to “high-spread” European countries and their banking sectors.
The result: an ugly, sitting duckling.
Interestingly, extrapolating the IMF’s seemingly rough estimate for Europe’s bank losses to the U.S. – using relative exposures reported in first quarter 2011 BIS data – revealed a significantly smaller implied loss than we would have guessed. While a range of different weightings to different countries and banks yields somewhat different results, the U.S. bank exposure appeared to be about 1/9th as large as the European banking system’s exposure to the six nations used to estimate the 300 billion euro capital hit (see Figures 3-4).
A simulated hit of say $45 billion across the thousands of U.S. banks represented in the BIS data would represent 4.5% of the U.S. banks’ now heightened level of tangible common equity. While significant growth concerns away from Europe are at work, U.S. bank equities have already fallen 22% in the past 3 months. A good part of assessing future economic damage to the U.S. should be a judgment that market assessments of risk have already become quite dire.
Here are the results in tabular form, which you can click to expand:
Caveat time!
This is an extrapolation based on an extrapolation; an extrapolation², if you like. The IMF’s calculations have been questioned by eurozone officials and use BIS data, which we originally discussed in these two posts. It’s difficult to get a proper sense of the hedged exposure of European banks, never mind the American ones. Throw in different accounting practices and this estimate needs to be treated with extreme caution.
And let’s not forget, exposure ain’t a rational animal, as Citi points out:
To be fair, the 2008 Lehman event showed that it might take just one systemically important financial institution with worse exposures than the average to do harm in the financial system and ultimately the world economy. In the aftermath of the Lehman event, G-20 leaders vowed to avoid further failures of SIFIs. Followed by actions, this helped in part to restore confidence and functioning to the financial system. We imagine officials hoped to generate such a confidence response this week with quite similar statements. However, political roadblocks – now legislated or otherwise – to additional intervention in markets has raised doubts about both the effectiveness and willingness to act.
Quack. Quite.
Related links:
America’s very own, very different, peripheral exposure [updated] – FT Alphaville
BIS > stress tests – FT Alphaville
The IMF can make your €200bn capital hole disappear in days! – FT Alphaville