The global economy has been one victim of the recent crisis of European sovereign debt, but Europe’s banking sector and the investors who have financed it will be the next. A great deal of pushing and shoving has forced European authorities to accept that there is a problem in their banking sector. Some are working hard to understand the problems and others see themselves as immune, though they probably are not; but all have been tempted to let political factors influence decisions that need to be based on sound economic and regulatory footings.

Thus the first response is to say, “European banks, no thanks!” until banks’ lending problems and capital needs are fully addressed. Sovereign debt in the PIIGS (Portugal, Ireland, Italy, Greece, and Spain) countries has become so large as to generate questions of bank liquidity and solvency. Indeed, in an earlier paper (1) I argued that Greece was bust and the other four PIIGS countries posed daunting problems for the financial system, with perhaps as much as a decade being necessary to prove that cases such as Italy could be managed.

Greece’s sovereign debt problems do not need to be a precedent for the other nations and their banks. Meanwhile, regulators in the other PIIGS countries (a new acronym is needed for PIIGS countries excluding Greece: IPSI perhaps, since some of the others do not sound so nice) and the European authorities in general face a nexus of problems on sovereign debt and banks, and sometimes real estate. Those problems are large enough to threaten European growth if extreme deleveraging becomes the main response to the current problems in the financial system. But it is also important not to go for grand solutions that worsen the problems.

This paper continues my earlier work, with a look at the capital strength of the banks within Europe. My analysis will show that:

1. The amount of capital needed by European banks is substantial, even assuming that economic recovery and sovereign debt resolution proceeded so as not to generate substantial new losses;

2. Greece’s default alone poses no significant direct risk to banks outside Greece, though its default essentially will force a redo of the Greek banking system;

3. Requiring capital buffers to back sovereign bond positions in other PIIGS countries would lead to a further, large incremental need for capital at banks;

4. Allowing banks to write up their capital for gains on German bunds makes little difference except for German banks. And it is in any case an unwise solution if one wants prudential regulation that is truly prudent;

5. Spanish banks have the greatest capital shortfall, a fact to some extent already recognized in the attempts of the Spanish authorities to restructure and recapitalize the cajas (saving banks); and,

6. In spite of all the bad news, it is necessary not to go overboard, for setting excessive capital requirements will surely lead to a deleveraging that damages economies, not just bank investors.

 

Recent Trends in Cross-border Claims

The Bank for International Settlements (BIS) has for decades provided data on the foreign claims of one country’s banks on all entities in a second country. They offer a great place to start examining the cross-border impact of the crisis. (2) As of the end of June this year, Italian borrowers posed the biggest challenge among PIIGS countries, responsible for $939 billion of claims to the other BIS member nations (see Table). The list continues down through Spain, Ireland, and Portugal, eventually reaching Greece, to whom the banks had $131 billion of exposure at the end of June.

Several patterns can be seen. French banks hold the biggest exposure to PIIGS countries, some 27% of the total, trailed by German and U.K. banks at 20% and 14%, respectively; no other nation’s banks exceed 10%. French banks hold 42% of the Greek claims, partly reflecting Greek banks with French parents. French banks are also heavily involved in Italy, in this case close to 44% of banking sector claims. The only comparable degrees of reliance are U.K. banks in Ireland (30%) and Spanish banks in Portugal (43%); German banks are the most prominent outsider in Spanish claims with a 24% share, somewhat ahead of France, and are also noticeably represented in Ireland, with 24% of claims. Historical patterns probably can be cited to explain the role of U.K. banks in Ireland and Spanish banks in Portugal, and may well involve corporate rather than sovereign credit.

By contrast, Japanese banks are essentially uninvolved except for a modest position in Italian claims (less than 5% in any individual PIIGS country). U.S. banks (especially compared to their large size) are essentially uninvolved in any of these markets, with the largest exposure in percentage terms taken by claims on Ireland – even though a U.S. entity, securities company MF Global, (3) was among the first casualties of the European sovereign debt problem.

Claims on PIIGS countries have declined since the end of 2009, the last data before Greece’s problems were made manifest. The European decision to support Greece allowed banks and other creditors to exit sovereign lending to PIIGS nations as existing bonds matured, to be replaced by official creditors such as the European Financial Stability Fund (EFSF) and the European Central Bank (ECB). The few cases of increased exposure all come from low bases, such as Swiss exposure to Spain and U.S. exposure to Portugal, and the Swiss cases may even reflect loans denominated in Swiss francs, given the sharp appreciation of the Swiss franc.

Contagion has become the watchword recently. This can be seen in the wide spreads for the euro-denominated interbank market, which hit 100 basis points on December 1, whereas trading more typically would occur in the range of slightly more than 20 basis points. Some of the possibilities of contagion can be addressed with the BIS data. A more noticeable involvement of U.S., U.K., and Japanese banks can be seen in the French and German markets. U.K. and U.S. banks each have around 20% of the non-French exposure to French entities, followed in third place by German banks. Meanwhile, French banks have led with somewhat more than 15% of the cross-border claims on German entities, followed closely by the Italian, U.S., and Dutch entities. Italy’s rather high position in the German claims may reflect the fact that one large Italian bank acquired a fairly substantial German bank a few years ago. While banking entities outside the eurozone are not heavily exposed to the problems of PIIGS sovereign debt, a contagion that spreads to the sovereign debt of France or Germany would draw more attention to the U.S. and U.K. banks; the Japanese bank exposures appear to be light despite the attention recently given to the claims on Italian sovereigns held by Nomura Securities. (4) BIS Rules and the Banks’ Capital Requirements

Before evaluating bank capital needs, it is worthwhile to do a quick review of a few key points of the BIS regulatory system. The discussions on the BIS rules began in the 1980s, and reflected U.S. fears that the lack of uniform capital requirements diminished the competitiveness of U.S. banks in the face of lax capital requirements in Japan. The rules focus on risk assets (not necessarily the same as risky assets) and the Tier I and Tier II capital required to back those assets. Different assets have different risk weights, ranging from 0% to 100%. Exposure to sovereign debt issued by developed nations was not counted as a risk asset. In early days this was identified as a philosophical defect, but until recently it was ignored as a likely source of serious problems. Some private sector exposures, especially securitizations and mortgages, draw less than a full weight; weights assigned to the property sector vary widely across countries to reflect differences in the quality of the collateral and historical experience. Not surprisingly, the 0% risk weighting on sovereign debt of developed nations encouraged banks to take on government bonds after the collapse of Lehman Brothers in 2008, a repeat of the yield curve play that has been a feature of many bank recapitalizations over the decades.

Early versions were very generous in their definitions of Tier I capital and in the ability to count Tier II capital (subordinated structures); the current rules sometimes appear too strict, or at least arbitrary, with some mandatory convertibles excluded from the EBA calculations of Tier I capital because the conversion date is later than the middle of 2012. The rules are also flexible in their determination of whether specific assets were impaired, particularly those deemed substandard when payments are still being made. A fear that regulators have been dilatory in assessing banks’ risks is one of the factors exacerbating the current crisis.

Another weakness of early versions of the BIS standards was the inclusion of deferred tax assets (DTAs) in capital calculations when it could take years, even decades, to make enough money to transform DTAs into tangible capital. The pretense that DTAs always represent core capital was a key feature in Japan’s prolonged restructuring, and it was not addressed finally until 2003.

The second round of BIS requirements, BIS II, was implemented in a relatively favorable economic environment, so had modest effects in most developed markets except Japan. The third round, BIS III, will be implemented over the rest of this decade, but is being done at a more trying time for banking systems throughout the developed world. The BIS III rules extend the requirements to other areas such as liquidity in addition to capital. But for now the main problem seems to be capital; if there were more, the market would not be worrying about liquidity and deposit flight at European banks.

 

 

1 Richard P. Mattione, “Et tu, Berlusconi? The daunting (but not always insuperable) arithmetic of sovereign debt,” October 27, 2011. Available at www.gmo.com

2 By definition, the data aggregate all claims, rather than separating central government, other levels of government, etc. Thus these data measure (for example) the overall exposure of French banks to Italy, not the exposure to the Italian sovereign alone. These data reflect BIS membership, not quite identical to eurozone membership, nor identical to Europe’s geography. And they do not measure Italian banks’ exposure to Italian entities at all, by definition of being cross-border claims.

3 MF Global would not have been captured in this data because it was not a bank.

4 Nomura is a broker, not a bank, so its positions would not have been captured in the BIS data.

5 European Banking Authority, “2011 EU-Wide Stress Test Aggregate Report,” July 15, 2011.

6 We excluded one bank, Dexia, from our analysis. Dexia was given a clean bill of health, but has since entered workout proceedings.

 

 

(c) GMO

www.gmo.com