What to Look for While We Wait for Europe's Bank Stress Tests
Winston Churchill once quipped about Americans: They always do the right thing—after they have tried everything else. One can now say the same of Europeans and their handling of banks. After steadfastly refusing to do so ever since the beginning of the global financial crisis in 2008, last week the European Council suddenly decided to publish by mid-July the results of stress tests of the 25 largest European banks, followed by a second phase covering additional parts of the EU banking system.1
This really good news came due to persistent financial-market pressure on Spain and the seemingly rapidly diminishing access of Spanish private companies and banks to international credit markets, which finally forced the Spanish government to offer to "show its banks' hands" by publishing the results of stress testing of Spanish banks.
In some respects, the financial markets' focus on the plight of Spanish banks has been surprising. While the condition of regional Spanish savings banks (the Cajas) is well known to be very poor (which is why the Spanish government has already earmarked up to €99 billion in the so-called FROB fund to consolidate and recapitalize the Cajas2), the larger Spanish banks are generally considered to be well-regulated and in good financial health, despite the downturn in the Spanish housing market. In its otherwise very critical May 24, 2010 Article IV report on the Spanish economy, the International Monetary Fund (IMF) noted : "Although impaired assets have increased with the downturn, Spanish banks overall report robust capital and provision buffers, supported by a strong supervisory framework." 3
Ironically, seemingly irrational market concerns may have pressed the Spanish government to do the right thing and publish the stress test results and thereby break ranks with EU governments seeking to "sweep the problems under the carpet," fearful of public hostility toward another round of bank bailouts. The "law of unintended consequences" has therefore worked wonderfully for European economic prospects in this case.
There is no doubt that publication of the stress tests is great news for Europe. If properly executed, it will dispel the lingering doubts that have surrounded the European banking system since the beginning of the global financial crisis. It should also eliminate the risk of the European financial system suffering another complete counterparty risk induced freeze, similar to what occurred right after the collapse of Lehman Brothers in September 2008, which in turn removes the single largest risk that Europe will experience a double-dip recession in the coming years. However, the devil as usual is in the detail, and a host of unanswered questions remain.
First, which banks will be included? EU leaders have signaled that the EU-27's top-25 banks will initially be included, followed by other, larger parts of the EU banking sector. The EU's top banks are listed in table 1:
|Table 1: Europe's top-25 banks by assets (latest available data)|
|Current rank||Bank||Home country||Assets (US$mn)||Balance sheet data on|
|1||BNP Paribas SA||France||2,891,948||31.12.08|
|2||The Royal Bank of Scotland Group plc||UK||2,739,361||31.12.09|
|3||Crédit Agricole SA||France||2,303,497||31.12.08|
|5||Deutsche Bank AG||Germany||2,153,033||31.12.09|
|6||Lloyds Banking Group plc||UK||1,658,736||31.12.09|
|8||Banco Santander SA||Spain||1,462,493||31.12.08|
|10||ING Bank NV||Netherlands||1,441,673||31.12.08|
|11||HSBC Bank plc||UK||1,214,158||31.12.08|
|13||Credit Agricole Corporate and Investment Banka||France||1,194,749||31.12.08|
|14||Bank of Scotland||UK||1,067,890||31.12.09|
|15||Credit Suisse Internationala||UK||975,713||31.12.08|
|16||RBS Holdings NV||Netherlands||929,103||31.12.08|
|17||Intesa Sanpaolo SpA||Italy||886,349||31.12.08|
|19||Fortis Bank SA/NV||Belgium||817,580||31.12.08|
|21||Banco Bilbao Vizcaya Argentaria SA||Spain||756,096||31.12.08|
|24||DZ BANK AG Deutsche Zentral-Genossenschaftsbank||Germany||595,082||31.12.08|
|25||Banque Fédérative du Crédit Mutuel||France||592,480||31.12.08|
|27||Danske Bank A/S||Denmark||572,753||31.12.09|
|a. Credit Agricole CIB and Credit Suisse International included despite being part of other banking conglomerates.|
|Source: Data from balance sheet information available at http://www.bankersalmanac.com/addcon/infobank/bank-rankings.aspx.|
Table 1 shows that banks in Germany, France, UK, Italy, Spain, Netherlands, Belgium, Sweden, and Denmark are likely to be among the top-25 stress tested by mid-July. As such, all the large European economies will have their largest banks included in the initial round of stress tests, which would cover about two-thirds of all EU banking assets.4 Similar to the pressure on other EU countries to follow suit once Spain publishes its stress test results, individual EU countries can be expected to publish stress test results for national banks other than those included in the initial top-25 list to avoid "stigmatizing" any individual nontested bank.
The fact that Spain, which has a very clear national interest in publishing its national stress test results in as comprehensive a manner as possible in order to assuage international investors' fears in its financial system, is expected to publish results first is a further advantage for transparency in Europe. While a certain degree of coordination among Europe's banking regulators seems likely following the endorsement of stress tests by the EU Council on June 17, as first mover, Spain is likely to set a high bar for stress test transparency, which other countries will be under acute market pressure to follow.
Second, what will be included in the EU stress tests? To reassure financial markets, published stress test results must include details of all methodological considerations and assumptions applied at the individual bank level. Moreover, to be credible, stress test results must include results for "highly adverse economic scenarios" and in particular ensure that highly unlikely events—tail-end risks—are adequately addressed. Without such details, no additional value will be derived from publishing stress tests and a mockery will have been made of EU leaders' declaration that the "transparency of the [EU] banking sector must be ensured."5
In the 2009 US Supervisory Capital Assessment Program (SCAP),6 such scenarios included specific assumptions for somewhat severe (too timid in the eyes of many observers at the time) but plausible developments in US GDP growth, unemployment, and not least house prices. However, including "plausible tail-end risk events" presents European bank regulators with an acute problem of a more delicate political nature than merely making assumptions about the likely percentage decline in national housing prices. European banking regulators need to decide whether to include among "tail-end risks" that at least Greek government bonds (and possibly other sovereigns) will be restructured and in the process see their net present value decline very substantially through principal reductions and maturity extensions.
On the one hand, without these details, the EU bank stress tests will ignore the "elephant in the room"— tail-end sovereign risk— in Europe's banking system today and will be a largely meaningless rhetorical exercise. However, on the other hand, by including these details EU regulators will implicitly acknowledge that a Greek sovereign debt restructuring is at least possible if not likely, something EU leaders and the European Central Bank (ECB) have so far strenuously denied is even a remote option.
Moreover, including provisions for a Greek sovereign restructuring in EU stress tests will with certainty show that all Greece's major domestic private banks—or at least the four large ones, National Bank of Greece, Alpha Bank, EFG Eurobank, and Piraeus Bank, that are not foreign-owned—will need major recapitalizations to avoid insolvency. Such recapitalizations after a possible Greek sovereign default seem certain to exceed the €10 billion currently available in the Financial Stability Fund (FSF) under the IMF/EU program to bolster the Greek banking sector's capital base.7 Currently, FSF funds from the IMF/EU are envisioned to be available to private Greek banks under stress tests that do not include the possibility of a Greek sovereign default.
It can be hoped that Spanish banking regulators—whose banks have very little exposure to Greek sovereign debt according to the latest BIS data8—will be courageous enough to simply include these details in their published results, thereby essentially compelling other EU countries' banking regulators to follow suit. Considering that German banks are expected to have much direct exposure to Greek sovereign debt, this move would offer Spanish authorities a compelling chance to retaliate against the numerous media leaks leading up to the June 17 EU Council from German official sources regarding the need for Spain to seek access to the European Financial Stability Facility (EFSF).
However, if Spain (and other EU regulators) refrains from the "best practice" of simply including provisions for a Greek sovereign debt restructuring in their official stress tests, other policy options are plausible.
Greek sovereign debt restructuring could be included in the EU-wide stress test scenarios but under the assumption that Greek banks would not be stress tested, since the country (and its domestic banks through the FSF) is already under an IMF/EU program.
Under another option, EU banking regulators could publish sufficient details about individual banks' holdings of/exposure to Greek sovereign debt and associated risks but not any "official own loss estimates" of the precise impact of an "unmentionable" restructuring. Such transparency would then enable financial markets to "do the math themselves" and insert their own assumptions concerning the most likely shape of a probable Greek "tail-end event."
Simply publishing individual banks' exposure to Greece (or other specified tail-end events) without specifying precise possible losses, however, could raise problems for governments called upon to recapitalize some of their individual national banks after the stress test results have been published. One way or the other, coming up with an appropriate amount of public money to "bail out a stricken bank" that both credibly turns the bank into a sustainable "going concern" as well as lets taxpayers off as cheaply as possible demands that governments pick an expected losses figure, irrespective of the political implications at the EU level.
"Cover-up" was clearly the preferred policy option for European politicians until the Spanish government decided to move toward publishing its stress test results. Today, however, the politics of EU stress tests has been completely changed. As is the case with all bank bailouts and required infusions of taxpayers' money, existing common shareholders must be completely wiped out to avoid the risk of moral hazard and transfers from taxpayers to existing irresponsible/incompetent owners and management. Considering the current public hostility toward banks in Europe and the associated bailout fatigue, the political advantages to the sitting government of wiping out/severely punishing existing common shareholders in a "second round of EU bank bailout" are straightforward.
Moreover, the already existing government influence in the European banking sector—particularly in the two countries generally expected to see the largest potential exposure to a Greek sovereign debt restructuring, Germany and France—is so pervasive that it would be a big mistake to doubt the willingness of EU sovereigns to "step up" and again recapitalize their banks, if called upon to do so by the stress test results.
In fact, quite a lot of Europe's nominally private banking exposure is already within the public sector. This is particularly true of Germany's Landesbanken, which with the two largest included in table 1 are certain to all be subjected to stress tests. All Germany's Landesbanken are owned by regional German state governments (together with regional savings banks) and as such will almost invariably require any additional capital from the German government. This would have been clear from the moment the German government agreed to submit them to stress tests.
Moreover, earlier during this crisis, bankrupt Hypo RealEstate was quickly taken over by the German government, which has also provided capital to WestLB and Commerzbank. Indeed, the German government today already has a 25 percent stake in Commerzbank as a result. Similarly in France, the government already owns 17 percent of BNP Paribas and historically (as with, for instance, Credit Lyonnais) the French treasury has had few qualms about supporting its large domestic banks—usually picking fights with EU competition authorities over state aid issues in the process.
In Germany, the key political fight will be on the issue of states' rights between the federal government (long intent on forcing consolidation of the Landesbanken) and the regional state governments (long intent on preserving their local fiefdoms). However, in a scenario where Landesbanken—generally considered among the most reckless and incompetent lenders in Europe—require additional capital, the main taxing entity in Germany, i.e., the federal government, will surely win this fight. As such, Berlin's suddenly agreeing to publish bank stress tests should be viewed also as a final crushing blow in a long war to rein in German state governments' fiscal independence. The recent unexpected resignation of Roland Koch, the premier of the state of Hesse and an ardent opponent of Landesbank consolidation, is likely a sign of the times and hopefully of imminent consolidation of Germany's Landesbank sector (similar to what is happening among Spain's Cajas).
In short, there is simply no reason to doubt that European governments will not be willing to step up to the plate and recapitalize their domestic banks, if (read when—in case of credible stress tests) this will be required.
The only way one could doubt this resolve is if the real question concerns European governments' ability to recapitalize their banking system, or in other words if one believes that in reality European countries are insolvent, when including the liabilities of their private banks. That, however, is a very difficult scenario to imagine, given how most losses in the European banking system from the imminent stress testing are expected to reside with German and French banks, both of which have—relatively—solid sovereign credit standings. Probably, Belgium—with at least two large and troubled banks in Fortis and Dexia (not included in table 1) and an already high total sovereign debt at nearly 100 percent of GDP—seems the country most immediately exposed to the results of the European stress tests.
But, in the event that Belgium gets into debt trouble, the country could likely access the new EFSF for financial support— like what happened to Troubled Asset Relief Program (TARP) funds. That would be a very small price to pay for getting transparency of Europe's banking system .
1. See "EU to test banks for slower growth; markets rally," Reuters, June 18, 2010.
2. See a detailed description of the function of the Fund for the Orderly Restructuring of the Banking Sector (FROB) in Jose Maria Roldan, 2010, The Spanish Banking Sector: Outlook and Perspectives.
3. International Monetary Fund, Spain—2010 Article IV Consultation, Concluding Statement of the Mission, Madrid, May 24, 2010.
6. For details, see Board of Governors of the Federal Reserve System, The Supervisory Capital Assessment Program: Design and Implementation [pdf], April 24, 2009.
7. For details, see International Monetary Fund, Greece: Staff Report on Request for Stand-By Arrangement [pdf], May 2010, page 15ff.