One might think it is relatively easy to identify banks in trouble, as the results of stress tests are known and there are even stress test calculators available that enable the DIY investor to identify the impact on banks of various default scenarios of sovereign bonds in the euro area.
But while instructive, capital (or the lack thereof) is by no means the whole story, so no matter how much fun it might seem to perform one's own stress test to identify banks to short, we have to warn relying on this kind of simplicity. European banks suffer from four interrelated problems:
- Many are underfunded and need more capital to cover exposure
- Banks in the periphery suffer from capital flight
- The money market is not working properly, as banks don't trust one another
- European banks are too dependent on short-term funding
- They're likely to be forced to recapitalize
In order to be able to write-down Greek debt holdings, many banks need to be recapitalized. The situation is so dire that a bank like Dexia (OTC:DXBGF), a Belgium-France-Luxembourg bank, already went over even before any sovereign default has materialized.
If this can happen to Dexia, there are certainly other possible victims out there. One way to find them is through the wonderful little stress test calculator from Thomson Reuters. The test uses the same data as the official test by the European Banking Authority, but applies a haircut to sovereign debt based on market prices for five-year bonds as of Oct. 12.
The total is deducted from the stressed core Tier 1 capital ratio (which you can set yourself in the calculator). The results show how many banks fail, and the amount of capital that they need to get back to a core Tier 1 capital ratio of 7%. You might also want to check what happens if that ratio is set at 9%, however.
This won't be good for equity holders either. According to rumors in the FT, the EU will force higher capital ratios on banks, involving a 9% ratio of core Tier 1 capital to risk-weighted assets for banks across the continent in a six- to nine-month period. If banks have trouble accessing money markets, they're not likely to fare any better in the capital markets. From the Financial Times:
Banks and their advisers said their scope to raise fresh capital from investors was all but non-existent. “I don’t think anyone has access to the markets now,” said one senior European investment banker. Investors are loath to commit to fresh equity injections, in the knowledge that the new money would simply be soaked up by sovereign debt writedowns, bankers said.
The response to this is more likely to be further deleverage by selling assets.
But by shrinking assets – the denominator of capital ratios – many banks believe they can reach the targets without resorting to government recapitalisation. In recent weeks, both BNP and SocGen (OTC:SCGLF) have signalled plans to offload a combined €150bn of risk-weighted assets. Further businesses could now be sold. Italy’s Unicredit (OTC:UNCIF) and Germany’s Commerzbank (OTC:CRZBF) were likely to find themselves under most pressure to deleverage and divest assets, bankers said.
The stress test calculator even shows how many banks fail in each country and the effect of complete bailout on public finances. Before you embark on sorting out which bank might get in trouble next, and short its shares for a profit (not possible in all countries), there are a number of issues to consider:
The euro area is purportedly working on a grand plan to recapitalize banks (and solve the euro problem). It might just work -- you'll never know. So far, markets seem to think so. Additionally, it really is questionable whether lack of bank capital is the only, or even the main, problem.
Remember that Dexia, that bank that just went bust, passed the European stress test with flying colors (it was the 12th safest bank in the EU) and had just $3B of Greek debt on its books. (Greece hasn't even defaulted.) Apparently, there is more going on.
Short-term funding and the money market
Insufficient capital is not the only, or in many cases even the main, problem. Here is why Dexia went belly up (from NPR):
They [Dexia] actually had plenty of capital, which is what these stress tests were designed to test. What it didn't have was access to funding, and a little bit like Bear Stearns in 2008. It needed a lot of short-term capital. And because of the Eurozone debt crisis, banks got increasingly nervous about lending to each other, and so the money just ran out for Dexia.
European banks' short-term financing needs are rather large. Boone and Johnson noted:
European banks also require large short–term funding. According to the recently published European Banking Authority stress test results, the 90 banks covered in that test owe €4.772 trillion within 24 months, equaling 38 percent of European Union GDP and 51% of euro area GDP (European Banking Authority 2011, 17). In France, Italy, and Germany, the largest two banks alone need to roll over 6 percent, 9 percent and 17 percent of national GDP in debt, respectively,within 24 months. This compares to just 1.6 percent of GDP for the largest two banks in the United States.3 Since the ability to raise such large amounts of short–term finance in Europe rests on market confidence that the ECB and banks’ respective sovereigns are standing by as a lender of last resort, any rapid shift toward a regime where bailouts are unlikely could quickly lead to a liquidity crisis for some marginal banks.
Unfortunately there isn't a nifty little calculator anywhere to get more handle on the situation as this problem actually derives from insufficient transparency. If even banks cannot separate the liquid from the illiquid, there is little chance the retail investor is capable of this.
However, isn't it reasonable to assume that there is at least some correlation between a bad stress test outcome and the willingness of other banks to lend the bank in question money through the money markets?
Apparently such correlation, if it exists, isn't a perfect indicator as the Dexia example shows: good stress test performance, but no access to funding. One reason why such correlation breaks down could be the stress test itself, which is why the calculator is useful, as you can perform your own stress test to execute scenarios you deem more relevant.
One problem with the European stress test was that haircuts were applied only to bonds in the trading books of banks, not in their investment books (unfortunately we couldn't find anything about this in the stress test calculator).
At the minimum, one has to inquire what the short-term funding needs of a particular bank are as stress test results are not all determining.
Since talk of default alerted investors to the fact that bonds from countries that issue their own currency are inherently less risky compared to bonds from countries that don't, there was a capital flight out of the latter. But not only that, investors and depositors fled out of banks in those countries as well, as these are usually the biggest holders of those bonds. Here are Boone and Johnson again:
In Ireland the two leading banks lost €65 billion in deposits from the end of 2008 to the end of 2010—equal to 52 percent of Irish gross national product. Similar deposit losses have occurred in Greece. As bond yields rise, the local banks tend to buy, because their futures are inextricably tied to the survival of their sovereign. Foreign institutions tend to sell bonds back to the local banks of the government that issued them. Finally, corporations and households tend to save outside their local banks, and foreign bank branches tend to reduce the size of their balance sheets in troubled nations. All these transactions generate capital outflows.
Much of this capital is parked in banks in the strong countries, which deposit it at the ECB, which then provides the liquidity to the banks in the weak countries to keep them alive. Here you can see the magnitude of it all (from July; things have gotten considerably worse since then).
So location also comes into it, although the determining factor seem to be the amount of toxic sovereign debt is on the books. Since banks are normally holding substantial sovereign bonds in their portfolios, banks in PIIGS are the most vulnerable but this already comes out of the stress test calculator, which correlates much better with capital flight than it does with short-term funding needs.
Let's assume (we're economists, after all) that there will be a substantial (65.5%) Greek default, capital ratios will be set at 9%, but somehow contagion will be avoided, so no haircuts on other sovereigns. What does the stress test calculator say? You can fill it out yourself, but here are some results:
- Overall capital shortfall of 242.6B euro
- 66 banks failing the test
- RBS the biggest casualty with a shortfall of 19.2B euros, followed by Deutsche Bank with a shortfall of 13.5B euros, Unicredit (Italy) 12.7B, SocGen 12.4B and BPCE 12.3B euros.
The stress test calculator gives a first indicator of bank weakness under different sovereign default scenarios, but be aware that not all banks are shortable, and capital ratios aren't the whole story.