18 July 2011

Cross Asset Strategy

Special report
www.sgresearch.com

Less bank stress, sovereign progress?

Credit Research - Banks Hank Calenti, CFA (44) 20 7676 7262
hank.calenti@sgcib.com

Equity Research- Banks Philip Richards (44) 20 7762 5893

philip.richards@sgcib.com

James Invine (44) 20 7762 5558

Michala.marcussen@sgcib.com

Economics Michala Marcussen (44) 20 7676 7813

Michala.marcussen@sgcib.com

Few surprises allow us to say that market reaction should largely be insensitive to those stress tests from a top down perspective. 8 banks are forced to recapitalise quickly, 16 banks by April 2012. However, the test does not reflect current reality, in our view; even if GIIPS sovereign are further stressed within this test, a 22bn shortfall and a relatively healthy average 6.2% core Tier 1 appear. The European banking sector is captive to politics at the moment. A political error can trigger a freeze in money markets, and a liquidity crisis could quickly turn into a solvency crisis. Only improved governance would avoid such a nasty scenario.
The EBA was in a lose-lose situation: too few failures and the test is branded too lenient;

James Nixon (44) 20 7676 7385

James.nixon@sgcib.com

too many failures and some will worry that the system is in worse shape than expected.
Recapitalisation What has been done: Under the adverse scenario, a 2.5bn shortfall is

Rates Strategy

Vincent Chaigneau
(44) 20 7676 7707
Vincent.chaigneau@sgcib.com

recognised after 90 banks from 21 countries had been stress tested. The pure sight of the test forced banks into a beauty contest: some massive 50bn was raised so far this year in core Tier 1 to reach that amount.
Recapitalisation what will be done: Banks with less than 5% core Tier 1 will be forced to recapitalise by end 2011. Those between 5% and 6% are highly recommended to do so by April 2012 (see page 7 for company details). The c. 3,500 datapoints (149 during the 2010 stress tests), should enable analysts to draw their own conclusions on a bank-by-bank basis.

Cross Asset Strategy Ahmed Behdenna (44) 20 7676 7513

ahmed.behdenna@sgcib.com

Contents Summary The Facts Economic standpoint Fixed Income standpoint Bank Equity standpoint Bank Credit standpoint 3 4 8 11 13 17

The stress is stressed enough on a macro level: Total losses per year amount to

c. 200bn, equivalent to the loss rates of 2009 repeated in two consecutive years. GIIPS sovereigns stressed by SG analysts on 40 larger banks: Further stressing GIP sovereign debt with a 50% haircut, and Spain/Italy with 20%, on top of the test conducted, and we see a 22bn shortfall and a still relatively healthy 6.2% core Tier 1 (see page 7 for company details).
Politics and governance now take the lead: Peripheral countries need help, and some form of restructuring should occur. Policy makers should allow for the governance to be at par with what should be qualified as a solid banking system, if contagion is avoided.

Macro

Commodities

Forex

Rates

Equity

Credit

Derivatives

Societe Generale (SG) does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that SG may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. PLEASE SEE APPENDIX AT THE END OF THIS REPORT FOR THE ANALYST(S)

CERTIFICATION(S), IMPORTANT DISCLOSURES AND DISCLAIMERS AND THE STATUS OF NON-US RESEARCH ANALYSTS.

Everything you wanted to know about stress tests

Table Of Contents

3 4 6 7 8 11 13 13 14 14 15 15 15 17 18 19

Summary EBA bank stress tests The Facts Headline results Results for the 90 banks which balance sheets have been stressed Economics: Banking on Sovereign Transparency The Fixed Income Angle The Equity Angle Part 1 EBA Stress Tests GIP Sovereign exposures Absolute amounts GIP Sovereign exposures As % common equity (%) Part 2 What if GIIPS sovereign debt had been stressed? Core Tier 1 Ratios 50% haircuts on GIP sovereign exposures, 20% on Italy/Spain Overall results Core tier 1 ratios under four scenarios (%) The Credit Angle Two-way Causality in the stress tests Adverse Scenario Core Tier 1 Capital vs. 5-year Subordinated CDS on 15 July 2011for select EU banks Stressed Sovereign Core Tier 1 Ratios vs. 5-year Subordinated CDS on 15 July 2011for select EU banks

18 July 2011

Everything you wanted to know about stress tests

Summary
Scepticism around: the EBA was on a lose-lose situation
The EBA was effectively in a lose-lose situation: too few failures and the test is branded too lenient; too many failures and some will worry that the system is in worse shape than they had expected.

Economics: banking on sovereign transparency

The stress tests confirmed two already well know results. First, the bulk of exposure to the weakest sovereigns is held by the domestic banks (in both absolute and relative terms). Taking the example of Greece, of the total exposures to the Greek sovereign reported for the 90 banks taking part in the stress tests almost 60% is held by Greek banks. Second, exposure of non-domestic banks to, respectively, Greek, Irish, Portuguese and Spanish sovereign risk is relatively moderate.

The Fixed Income Angle

The stress tests have brought more details on individual exposure to sovereign risk. What investors still do not know is the extent to which banks, and other investors, will be asked to contribute to fixing the sovereign crisis.

The Equity Angle Part 1: EBA stress test

Equity market reaction to the stress tests is likely to be relatively muted in our view, with few big surprises despite the huge volume of new disclosure. Market sentiment is likely to remain focussed on the fall-out of the Greek crisis and how that spreads further afield.

The Equity Angle Part 2: 'Only' 22bn shortfall if GIIPS sovereign debt is also stressed
The tests will inevitably be criticised for such a small number of failing banks (eight) with an aggregate capital shortfall of just 2.5bn, and the fact that sovereign default is not considered. By including the possibility of sovereign default across multiple jurisdictions (haircut of 50% for GIP sovereigns, 20% for Italy/Spain), our analysis suggests 13 banks of a sub-set of 40 of the larger, quoted banks could fail, with an aggregate capital shortfall of 22bn. The average core Tier 1 capital would remain at a still relatively healthy 6.2%.

The Credit Angle

Like the equity market, credit market reaction to the stress tests is also likely to be relatively muted in our view, with few easily decipherable surprises discovered within a large volume of disclosures. We do not envision a flood of new bank debt issuance in the short term for two reasons. It may be more advantageous to de-lever than raise funds in the international capital markets. This would clearly have attending feedback into the bank-to-sovereigns and economy causality. For more details on the credit angle, please refer to: EU Banks, Stressed but not tested

18 July 2011

Everything you wanted to know about stress tests

EBA bank stress tests The Facts

Introduction
We suspect that the market will be surprised by the fact that only nine banks failed the test. The EBA was effectively in a lose-lose situation: too few failures and the test is branded too lenient; too many failures and some will worry that the system is in worse shape than they had expected.
th

After market-close on Friday 15

July, the European Banking Authority published the long-

awaited results of its 2011 stress tests. The tests covered 90 EU banks, representing 65% of banking sector assets in the EU. To recap, the tests have a 5% Core Tier 1 (EBA definition, similar to Basel 2.5) pass mark for two different macro-based scenarios, a baseline and an adverse scenario. The baseline scenario is based on the Autumn 2010 EC forecast of a gradual recovery in EU growth and the adverse scenario a more severe scenario designed by the ECB.

An impossible task
The outcome of the stress tests were never going to do enough to convince sceptics that European banks have enough capital to withstand a severe downturn, particularly if combined with one or more sovereign defaults. That said, we suspect that the market will be surprised by the fact that only nine banks failed the test. The EBA was effectively in a lose-lose situation: too few failures and the test is branded too lenient; too many failures and some will worry that the system is in worse shape than they had expected.

Whose stress test is it anyway?

Although the market is unlikely to give the stress tests credit for it, we suspect that the rash of capital raisings announced earlier this year were triggered, at least in part, by the looming stress test. Without the mitigation measures this year, including capital raising of some 50bn, the EBA stated that 20 banks would have failed the test. We suspect that seeing those banks fail the test and then raise capital/restructure would have been taken positively: a real sign that regulators were getting tough and taking tough decisions. As it is, those capital raisings are now a distant memory, leaving this stress test to be perceived (perhaps unfairly) as a damp squib. But whereas the sector rallied into last years stress tests, share prices have this time been heading down, whether we look in absolute terms, relative to the European market or relative to global banks. So while these tests are not, in our opinion, a powerful trigger for a rally, the sectors valuation and investor sentiment are both low enough that even a mild positive may help share prices.

18 July 2011

Everything you wanted to know about stress tests

European banks performance 120 days before/after stress test (absolute)

130

European banks performance 120 days before/after stress test (rel to euro mkt)
130

European banks performance 120 days before/after stress test (rel global banks)
130

100

100

100

70 -120 -96

-72

-48

-24
2010

+24 +48 +72 +96 +120

2011

70 -120 -96

-72

-48

-24
2010

+24 +48 +72 +96 +120

2011

70 -120 -96

-72

-48

-24
2010

+24 +48 +72 +96 +120

2011

Source: Datastream, SG Cross Asset Research

We have the raw materials to make a more comprehensive test

The main criticism of the tests centres on the fact that it does not incorporate the possibility of a sovereign bond default, despite the fact that this is looking increasingly likely. In the analysis below, we compute our own, tougher, tests on 40 of the 90 banks covered by the EBA typically the larger, quoted European banks.
Stressing Greek debt by 50%, and Italian and Spanish debt by 20%... would trigger 13 of the largest 40 quoted banks failed the test with 22bn capital shortfall.

In addition to the stress applied by the EBA to the banks private sector loan books, we write down Greek, Irish and Portuguese sovereign exposures by 50% and those of Italy and Spain by 20%. On that basis, we see 13 of the 40 banks failing the test, with a total deficit of 22bn, equivalent to 41% of the end-2010 tangible equity of those 13 banks, but just 2% of the end2010 tangible equity for all 90 banks in the test. We suspect that if the EBA stress test had been formulated in that way, with those results, then the market opinion may well have been more positive.

Does it really matter?

Ultimately, of course, even with the stress test results suggesting that the European banks are in a relatively healthy position and able to withstand likely shocks, no regulatory authority can ever prove that is the case. A stress test examines what happens to an asset when macro conditions deteriorate but relies on the bank/regulator estimating how losses will develop as those macro variables change. But while macro conditions have been worse than expected, we believe that the worstaffected banks through the crisis would never have anticipated the losses incurred, even if they had had perfect foresight on the economic outlook.

18 July 2011

Everything you wanted to know about stress tests

Headline results
The total capital shortfall of the 9 banks is a mere 2.5bn. This result is almost identical to the 2010 stress test results, when 7 banks failed the tests with a total capital shortfall of 3.5bn

On the EBAs adverse scenario 82 banks pass and just 8 banks fail versus the 5% Core Tier 1 pass-mark: Five Spanish banks (listed Banco Pastor and four Cajas) Two Greek banks (EFG Eurobank and ATEBank), Austrian Volksbank German Landesbank Helaba (which pulled out of the tests earlier in the week amid protests against the EBA methodology).

The total capital shortfall of the 8 banks is a mere 2.5bn. This result is almost identical to the 2010 stress test results, when 7 banks failed the tests with a total capital shortfall of 3.5bn (Hypo Real Estate, ATEbank and five Spanish Cajas). There were a further 16 near-fail banks with a capital ratio between 5% and 6% in the 2011 tests, which included Banco Sabadell, Banco Popolare, Banco Popular and Bankinter.

Core Tier 1 ratios on EBA Adverse scenario, versus 5% pass mark

25%

20%

Near-fail
15%

Fail

10%

5%

-5%

Source: European Banking Authority, SG Cross Asset Research

BANCA MARCH IRISH LIFE & PERMANENT OT P BANK Sydbank BANQUE CAISSE D'EPARGNE DANSKE Jyske PKO BANK POLSKI OP-Pohjola RABOBANK SEB Landesbank Berlin DEXIA BANK OF VALLETTA CAJA SAN SEBASTIAN KBC AIB Hypo Real Estate Nordea Swedbank MONT E DE PIEDAD Nykredit DekaBank Deutsche ABN AMRO BBVA DnB NOR INT ESA SANPAOLO GRUPO BBK WGZ BANK CAJA VITORIA Y ALAVA ING SHB CREDIT AGRICOLE HSBC SANT ANDER Erste NOVA KREDITNA BNP PARIBAS Raiffeisen Lloyds NBG UBI ALPHA BANK BARCLAYS CAJA INVERSIONES Landesbank Baden-Wrt Bayerische Landesbank BANK OF IRELAND SNS BANK DZ BANK EFFIBANK BPCE CAJA ZARAGOZA UNICREDIT Banco BPI DEUT SCHE CAJA DE BARCELONA COMMERZBANK RBS MPS CAIXA GERAL DEPSITOS COLONYA BANK OF CYPRUS GRUPO BMN WestLB SABADELL BANCO POPOLARE GRUPO BANCA CIVICA CAJA ONTINYENT Norddeutsche Landesbank HSH Nordbank T T HELLENIC BCP BFA-BANKIA CAIXA GALICIA BPE NOVA LJUBLJANSKA MARFIN PIRAEUS BANKINTER BES EFG EUROBANK CAIXA DE CATALUNYA Volksbank CAIXA UNIO DE CAIXES GRUPO CAJA3 BANCO PASTOR CAJA DEL MEDITERRANEO AT Ebank

0%

18 July 2011

Everything you wanted to know about stress tests

Core Tier 1 results

Name 2010 EBA definition 2012 EBA adverse* Name 2010 EBA definition 2012 EBA adverse

BANCA MARCH IRISH LIFE & PERMANENT OTP BANK Sydbank BANQUE CAISSE D'EPARGNE DANSKE Jyske PKO BANK POLSKI OP-Pohjola RABOBANK SEB Landesbank Berlin DEXIA BANK OF VALLETTA CAJA SAN SEBASTIAN KBC AIB Hypo Real Estate Nordea Swedbank MONTE DE PIEDAD Nykredit DekaBank Deutsche ABN AMRO BBVA DnB NOR INTESA SANPAOLO GRUPO BBK WGZ BANK CAJA VITORIA Y ALAVA ING SHB CREDIT AGRICOLE HSBC SANTANDER Erste NOVA KREDITNA BNP PARIBAS Raiffeisen Lloyds NBG UBI ALPHA BANK BARCLAYS CAJA INVERSIONES
*LB=Landesbank - Source:EBA, SG Cross Asset Research

22.2% 10.6% 12.3% 12.4% 12.0% 10.0% 12.1% 11.8% 12.2% 12.6% 11.1% 14.6% 12.1% 10.5% 13.2% 10.5% 3.7% 28.4% 8.9% 8.7% 12.5% 8.8% 13.0% 9.9% 8.0% 8.3% 7.9% 10.2% 10.8% 12.5% 9.6% 7.7% 8.2% 10.5% 7.1% 8.7% 7.4% 9.2% 8.1% 10.2% 11.9% 7.0% 10.8% 10.0% 8.2%

23.5% 20.4% 13.6% 13.6% 13.3% 13.0% 12.8% 12.2% 11.6% 10.8% 10.5% 10.4% 10.4% 10.4% 10.1% 10.0% 10.0% 10.0% 9.5% 9.4% 9.4% 9.4% 9.2% 9.2% 9.2% 9.0% 8.9% 8.8% 8.7% 8.7% 8.7% 8.6% 8.5% 8.5% 8.4% 8.1% 8.0% 7.9% 7.8% 7.7% 7.7% 7.4% 7.4% 7.3% 7.3%

LB Baden-Wurttemberg Bayerische LB. BANK OF IRELAND SNS BANK DZ BANK EFFIBANK BPCE CAJA ZARAGOZA UNICREDIT Banco BPI DEUTSCHE CAJA DE BARCELONA COMMERZBANK RBS MPS CAIXA GERAL DEPSITOS COLONYA BANK OF CYPRUS GRUPO BMN WestLB SABADELL BANCO POPOLARE GRUPO BANCA CIVICA CAJA ONTINYENT Norddeutsche Landesbank HSH Nordbank TT HELLENIC BCP BFA-BANKIA CAIXA GALICIA BPE NOVA LJUBLJANSKA MARFIN PIRAEUS BANKINTER BES EFG EUROBANK CAIXA DE CATALUNYA Volksbank CAIXA UNIO DE CAIXES GRUPO CAJA3 BANCO PASTOR CAJA DEL MEDITERRANEO ATEbank

8.2% 9.3% 8.4% 8.4% 8.2% 8.3% 7.8% 9.7% 7.8% 8.2% 8.8% 6.8% 10.0% 9.7% 5.8% 8.5% 11.2% 8.1% 8.3% 8.7% 6.2% 5.8% 8.0% 8.9% 4.6% 10.7% 18.5% 5.9% 6.9% 5.2% 7.1% 5.2% 7.3% 8.0% 6.2% 6.4% 9.0% 6.4% 6.4% 6.3% 8.6% 7.6% 3.8% 6.3%

7.1% 7.1% 7.1% 7.0% 6.9% 6.8% 6.8% 6.7% 6.7% 6.7% 6.5% 6.4% 6.4% 6.3% 6.3% 6.2% 6.2% 6.2% 6.1% 6.1% 5.7% 5.7% 5.6% 5.6% 5.6% 5.5% 5.5% 5.4% 5.4% 5.3% 5.3% 5.3% 5.3% 5.3% 5.3% 5.1% 4.9% 4.8% 4.5% 4.5% 4.0% 3.3% 3.0% -0.8%

18 July 2011

Everything you wanted to know about stress tests

Economics: banking on Sovereign Transparency

At the macro-economic level, the most pressing risk in the context of the euro area debt crisis resides in the linkages from sovereigns-to-banks and banks-to-sovereigns. The EBA stress tests offered an unparalleled level of detail on banks sovereign debt exposures, with over 2,800 data points per bank detailing sovereign exposure with regard to issuer, maturity and accounting classification (notably giving the split between trading book and bank book). Combing through the data, we find that - as expected - the bulk of exposure to the weakest sovereigns is held by the domestic banks. Moreover, the direct sovereign exposure to Greece, Ireland and Portugal of other euro area banks are relatively low. The greater concern is potential contagion to other euro area sovereigns and to bank funding conditions stemming from renewed sovereign tension, which are potentially very substantial. When euro area leaders meet on Thursday, July 21 for an emergency summit on Greece, they will do well to keep these channels of contagion in mind in defining the private sector involvement (PSI) in the new Greek financial assistance program. Poorly defined, the risk is to see further contagion domino-ing sovereign bond markets and to European banks.

Sovereigns-to-banks
For Greece, the EBA reports total sovereign exposure at default (EAD) basis to the Greek sovereign at 98.2bn.

The stress tests confirmed two already well know results. First, the bulk of exposure to the weakest sovereigns is held by the domestic banks (in both absolute and relative terms). As seen from the chart below, it is generally true that banks hold a significant share of their sovereign exposure in the home sovereign. For Greece, the EBA reports total sovereign exposure at default (EAD) basis to the Greek sovereign at 98.2bn. Of this total amount for the 90 banks taking part in the stress tests, 67% is held by Greek banks. For Ireland and Portugal, total sovereign EAD and the share held by home banks are reported at, respectively, 52.7bn and 61%, and 43.2bn and 63%. Second, exposure of non-domestic banks to, respectively, Greek, Irish, Portuguese and Spanish sovereign risk is relatively moderate. As seen from the table below (based on gross sovereign exposure) adding Italian sovereign risk to the equation increases the stakes, however. This illustrates the danger of the contagion currently spreading through the euro area bond markets.
Main direct exposure is to the domestic sovereign
Share of total gross sovereign exposure to home sovereign (%, end-2010 EBA stress tests)
100%

80%

60%

40%

20%

0% AT BE CY DE DK ES FI FR GB GR HU IE IT LU MT NL NO PL PT SE SI
Source: EBA, SG Cross Asset Research / Economics

18 July 2011

Everything you wanted to know about stress tests

Greece, Ireland, Portugal and Spain are modest, adding in Italy increases the stakes
Gross sovereign exposure EURbn, end-2010 (selected countries for banks participating in EBA stress tests) Germany Total Greece Ireland Portugal Spain Italy EURbn EURbn % total EURbn % total EURbn % total EURbn % total EURbn % total 556 8 1% 1 0% 4 1% 19 3% 37 7% France Netherlands 504 10 2% 2 0% 5 1% 15 3% 53 11% 169 1 1% 0 0% 1 0% 2 1% 10 6% Italy 223 1 1% 0 0% 0 0% 3 1% 164 74% Spain 350 0 0% 0 0% 5 2% 232 66% 7 2% Greece 64 54 85% 0 0% 0 0% 0 0% 0 0% Ireland 21 0 0% 12 60% 0 1% 0 2% 1 4% Portugal 31 1 5% 1 2% 20 62% 0 1% 1 3% UK 705 3 0% 1 0% 3 0% 12 2% 26 4%

Source: EBA, SG Cross Asset Research / Economics

In considering the linkages and the potential dominos from a contagion scenario, it is also important to take account of other exposures. As seen from the table below, broadening out to include total exposures to all sectors in the economy, this substantially increases the numbers.
As does including the full economy
Total exposure, EAD basis, EURbn, end-2010 (selected countries for banks participating in EBA stress tests) Germany Total Greece Ireland Portugal Spain Italy EURbn EURbn % total EURbn % total EURbn % total EURbn % total EURbn % total 3,755 18 0% 45 1% 18 0% 104 3% 103 3% France Netherlands 4,106 44 1% 21 1% 15 0% 78 2% 261 6% 1,883 1 0% 8 0% 0 0% 4 0% 2 0% Italy 1,895 1 0% 1 0% 2 0% 7 0% 1,262 67% Spain 3,007 0 0% 2 0% 62 2% 2,174 72% 1 0% Greece 339 242 71% 0 0% 0 0% 0 0% 0 0% Ireland 354 0 0% 191 54% 1 0% 5 1% 3 1% Portugal 349 7 2% 2 1% 248 71% 20 6% 2 0% UK 4,467 8 0% 68 2% 14 0% 77 2% 37 1%

Source: EBA, SG Cross Asset Research / Economics

Adding up the total numbers for euro area banks, we find that the pool of sovereign debt that could be involved in PSI over the 3-year horizon, amounts to 34.5bn for Greece (gross exposure, up to and including 3-year maturity), with the bulk held by euro area banks.

A more surprising outcome is found on the maturity of holdings, with a far more mixed picture from institution to institution, breaking somewhat with the general consensus that banks hold the bulk of exposure in short maturity issues. Adding up the total numbers for euro area banks, we find that the pool of sovereign debt that could be involved in PSI over the 3-year horizon, amounts to 34.5bn for Greece (gross exposure, up to and including 3-year maturity), with the bulk held by euro area banks. Note, these numbers include only the European banks taking part in the EBA stress tests, which cover around 65% of total assets in the region. Including all banks would increase the pool for PSI: assuming a simple proportionate increase we find 53bn. At present, several proposals for PSI appear to be on the table, ranging from some form of vehicle to buyback Greek bonds at a discounted price from private investors to various forms of bond exchanges. In all cases, a writedown seems likely to be on the agenda for private investors. The bank stress tests have already made some provisions for such a scenario (please refer to results section), but the risk is to see this exceeded if contagion is not successfully stemmed.

18 July 2011

Everything you wanted to know about stress tests

While a CDS event can likely be avoided, a downgrade to SD on Greece seems all but inevitable. This raises the question on ECB collateral. As we have highlighted on previous occasions, this is primarily an issue for Greek banks. The question remains what the ECB would do, but at present it seems likely that in such an event the ECB would no longer accept Greek bonds as collateral and instead offer Emergency Liquidity Assistance (ELA) to the Greek banks.
A mixed picture on the exposure of maturity to the Greek sovereign
Maturity structure of gross exposure to Greek sovreign (%, end-2010 (EBA stress tests)
100% 80%

60%
40% 20%

0%
ALL AT BE CY DE DK ES 3M
Source: SG Cross Asset Research

FI 2Y

FR GB GR 3Y 5Y 10Y

IE

IT 15Y

LU MT NL

PT SE

SI

1Y

An important point to keep in mind is that PSI for Greece is likely to be seen as a blueprint for Portugal and Ireland, and any other member states that many ultimately need to seek EU/IMF assistance. Poorly defined, the risk is that PSI could trigger a buyers strike in weak sovereign bond markets and also weigh on bank funding. In the context of the bank stress tests, it is important to note that funding and a potential liquidity was not explicitly stressed (please refer to results section).

Banks-to-sovereigns, and the economy

The 2.5bn identified is unlikely to pose any significant problem. The risk is, however, that this number ultimately proves higher.

In discussing the links from sovereigns to banks, it should be recalled that it is a two-way causality. As such, a weak banking sector can feed back to weigh on the sovereign. This is especially true if a bank fails to raise sufficient capital in markets or lose access to funding. In the context of the stress tests, the 2.5bn identified is unlikely to pose any significant problem. The risk is, however, that this number ultimately proves higher (please refer to results section). Linkages to the real economy should not be forgotten either. The euro area debt crisis is no doubt a negative for consumer and business confidence, and fiscal austerity weighs directly on activity. In addition, the real economy is linked directly to the banking system via the credit channel (higher funding costs and lesser supply of capital). Wealth effects also play a role for both financial and real assets as risk-aversion increases. Breaking the channels of contagion is thus today the main challenge for euro area policymakers. Failure to do so would spell a very bleak scenario, far beyond what has been outlined in the stress tests.

10

18 July 2011

Everything you wanted to know about stress tests

The Fixed Income Angle

The stress tests have brought more details on individual exposure to sovereign risk. What investors still do not know is the extent to which banks - and other investors - will be asked to contribute to fixing the sovereign crisis.

From the start, EU officials have treated Greeces and other countries problems as a liquidity crisis. There was the foundation of the muddle-through strategy: help Greece for a while, and give them time to convince investors that they are returning to a sustainable debt path. We have never been fond of that strategy, especially on evidence that time was not being used wisely (indeed Greece has failed to reassure). As a second plan for Greece is in the working, EU policymakers are discovering a new impossible trinity. Just consider the three following pillars: 1) bringing Greece on a sustainable debt path; 2) limiting official lending; 3) limiting the damage for the private sector, including the fragile banking sector.
The best we can hope for, for now, is a respite this summer, on hopes that EFSF will be buying in the secondary market (or lending to countries that can use the funds to buy back their debt).

There is no rabbit in the hat: the only way to satisfy those three objectives is through a credible and effective austerity plan. But the past misachievements and resulting credibility gap have made the above an impossible trinity. The French plan, by transforming credit risk into duration risk, clearly focused on making Private Sector Involvement sweet (pillar 3). But the plan had shortcomings: it failed to bring PSI to the desired 30bn level (so failing pillar 2), and certainly did not address Greeces debt sustainability issue (pillar 1). Given that even this Vienna Initiative does not seem sweet enough to avoid a Selective Default, the plan has been shelved. Policymakers instead now seem to focus on pillars 1 and 2. This has probably not been done the hard way yet (haircuts on principal). Greece is set to receive more loans, and we fear that resources continue to be misallocated. The IMF itself now recognises that only the threat of wider financial contagion allows the IMF to keep lending to a country with such an unstable debt situation. Providing more lending may avoid a traumatic near-term shock (Lehman-like), but not a slow-burning contagion. Policymakers indeed are investigating a tougher treatment of private investors: The IMF now cautions against offering any incentives for creditor participation that would negatively impact on Greeces already difficult debt dynamics. Investors, naturally, are getting worried. Debt swaps will most likely prove bigger and more costly than a Vienna initiative. We have yet to see the details of the debt swaps. Default for now may just mean a temporary Selective Default rating. Later on it will most likely involve haircuts on principal. This is proving extremely stressful for sovereign risk with contagion now going beyond Ireland, Portugal and Spain. Whether there will be an EU summit in the coming week is still uncertain. But even if one occurs, there is no instant cure for the troubles of the eurozone. They may be in abeyance for now. But we continue to see more tensions in September, at which time investors want to be short risky sovereigns again (see Sovereign section).

18 July 2011

11

Everything you wanted to know about stress tests

The best we can hope for, for now, is a respite this summer, on hopes that EFSF will be buying in the secondary market (or lending to countries that can use the funds to buy back their debt)
Hedging against the adverse effects of such a crisis is often costly. We critically discuss a number of protective strategies and recommend a 3M/6M LIBOR basis trade in forward space as the most appropriate hedge (see Focus US, Keep hope alive).

The details of the second aid plan to Greece may not be known before late summer. If for now we can avoid a plan that would be ostentatiously painful for the private sector, we should see some near-term respite. That, along with signs of a V-shape recovery in Japan, fading concerns about China's growth and a pick-up in US economic surprise indices, forms the basis of our small short duration exposure (bias towards a rebound in bond yields in the next few weeks). Our biggest near-term fear, however, continues to be the contagion of the crisis towards the banking sector, given the threat of a heavier private sector involvement. Hence we recommend hedges that would benefit from bank stress. It is tempting to do so in USD, as European banks find it more difficult to raise funding in the dollar markets.

12

18 July 2011

Everything you wanted to know about stress tests

The Equity Angle, Part 1: EBA stress test

Equity market reaction to the stress tests is likely to be relatively muted in our view, with few big surprises despite the huge volume of new disclosure. Market sentiment is likely to remain focussed on the fall-out of the Greek crisis and how that spreads further afield. The tests will inevitably be criticised for such a small number of failing banks (nine) with an aggregate capital shortfall of just 2.5bn, and the fact that sovereign default is not considered. In this section we focus on the outcome for a selection of 40 banks typically the largest, quoted banks in the EU. The chart below shows Core Tier 1 ratios under the adverse scenario. The results vary from over 12% for three Nordic banks to 4.9% for EFG Eurobank, the only bank in our sub-set to fail.

Core Tier 1 Ratios Adverse Scenario

14% 12% 10% 8% 6%

4%
2%
Sy dbank Danske Jy ske SEB Dexia KBC AIB Nordea Swedbank BBVA DnB NOR Intesa SHB Credit Agricole HSBC Santander Erste BNP Paribas Raif f eisen Lloy ds NBG UBI Alpha Bank Barclay s Bank of Ireland Natixis Unicredit BPI Deutsche Bank Commerzbank RBS MPS Sabadell Popolare BCP Popular Piraeus Bankinter BES EFG Eurobank

0%

Source: European Banking Authority, SG Cross Asset Research

Updated sovereign exposures

One clear positive of the stress test exercise is the extra disclosure provided alongside the results. Aggregate sovereign bond exposures to Greece, Ireland, Portugal, Italy and Spain are shown below in absolute amounts and as a percentage of common equity. In absolute terms, the banks with biggest exposures are (in order) Intesa, BBVA, Unicredit, Santander and BNP Paribas (to the home sovereign in the first four cases). However, as a multiple of common equity the biggest exposures are AIB, MPS, Piraeus, Popolare and EFG Eurobank. The Nordic and UK banks stand out as the least exposed banks to sovereign bond default.

18 July 2011

13

Everything you wanted to know about stress tests

GIP Sovereign exposures Absolute amounts (bn)

In absolute terms, the banks with the biggest exposures are (in order) Intesa, BBVA, Unicredit, Santander and BNP Paribas (to the home sovereign in the first four cases).
70,000 60,000 50,000 40,000 30,000 20,000 Spain Italy Portugal Ireland Greece

10,000
Intesa BBVA Unicredit Santander BNP Paribas MPS Dexia Barclay s Commerzbank NBG Credit Agricole HSBC Deutsche Bank Popolare UBI RBS Popular EFG Eurobank Natixis Piraeus KBC BCP Sabadell AIB Ireland BPI Alpha Bank Bankinter BES Erste Danske SEB Raif f eisen Nordea Jy ske Lloy ds SHB Swedbank DnB NOR Sy dbank
-

Source: European Banking Authority, SG Cross Asset Research

GIP Sovereign exposures As % common equity (%)

However, as a multiple of common equity those with the biggest exposures are AIB, MPS, Piraeus, Popolare and EFG Eurobank. The Nordic and UK banks stand out as the least exposed banks to sovereign bond default.
800% 700%

600%
500% 400% 300%

Spain Italy Portugal Ireland

200% 100% Greece

Source: European Banking Authority, SG Cross Asset Research

Incorporating sovereign exposures into the tests

As noted above, the EBA stress tests do not account for the possibility of a sovereign default, despite the fact this has become an increasingly likely scenario. In the following section we therefore incorporate into the EBA scenarios: 1. The potential impact of a sovereign default in Greece, Ireland and Portugal, by applying a 50% haircut to sovereign exposures to each country. 2. Apply an additional 20% haircut to sovereign bond exposures to Italy and Spain. We caveat that this is far removed from our base-case expectation.
14 18 July 2011

AIB MPS Piraeus Popolare EFG Euro BPI BBVA NBG Intesa Sabadell BCP Commerz Bankinter UBI Popular Unicredit Dexia Alpha Bank Santander Ireland BNP Paribas KBC BES Barclay s Deutsche Cred Agricole Natixis RBS HSBC Erste Danske Raif f eisen Jy ske SEB Nordea Lloy ds SHB Swedbank DnB NOR Sy dbank

0%

Everything you wanted to know about stress tests

The Equity Angle, Part 2: What if GIIPS sovereign debt had been stressed
By including the possibility of sovereign default across multiple jurisdictions, our analysis suggests 13 banks of a sub-set of 40 of the larger, quoted banks could fail, with an aggregate capital shortfall of 22bn.

Applying 50% haircuts on GIP exposures

The chart below shows estimates Core Tier 1 ratios based on the EBAs adverse scenario, incorporating a sovereign default in each of Greece, Ireland and Portugal. On this basis, 8 of the 40 banks fail the 5% CT1 pass mark, versus just one bank as reported by the EBA excluding sovereign stresses. Perhaps unsurprisingly, the newly failing banks are from those countries included within the sovereign default all four Greek banks fail, one Irish and all three Portuguese banks. The aggregate capital shortfall of the failing banks is EUR16bn.
Core Tier 1 Ratios 50% haircuts on GIP sovereign exposures
14%

On this basis, 8 of the 40 banks fail the 5% CT1 pass mark, versus just one bank as reported by the EBA excluding sovereign stresses

12% 10% 8% 6% 4% 2% 0% -2% -4%

Source: European Banking Authority, SG Cross Asset Research

Applying 50% haircuts on GIP AND 20% haircut on Italy/Spain exposures

If we additionally include a 20% haircut on all Italian and Spanish sovereign bond exposures, 13 of the 40 banks fail with an aggregate capital shortfall of 22bn. As well as those failing on the GIP haircuts, two Italian banks would fail (MPS and Popolare) and the three small Spanish banks (Sabadell, Bankinter and Popular).

Sy dbank Danske Jy ske SEB KBC Nordea Swedbank BBVA Dexia DnB NOR Intesa SHB HSBC Credit Agricole Santander Erste Raif f eisen AIB Lloy ds BNP Paribas UBI Barclay s Deutsche Bank Unicredit BPCE MPS RBS Commerzbank Popolare Sabadell Bankinter Popular Bank of Ireland BES Alpha Bank BCP BPI EFG Eurobank NBG Piraeus

18 July 2011

15

Everything you wanted to know about stress tests

Core Tier 1 Ratios 50% haircuts on GIP sovereign exposures, 20% on Italy/Spain
14% 12% 10% 8% 6% 4% 2% 0% -2% -4%

Source: European Banking Authority, SG Cross Asset Research

Finally, the chart below shows the overall results for each bank on 3 different scenarios, ranked by the most severe scenario (scenario 3 below): 1. EBA Adverse scenario 2. EBA Adverse scenario with additional 50% haircut to GIP sovereign exposures 3. EBA Adverse scenario with additional 50% haircut to GIP sovereign exposures and 20% haircut to Italy/Spain sovereign exposures. The banks which standout as best performing under the scenarios are all Nordic banks, KBC, the Austrian banks and large Spanish banks. The middle ground typically includes the French, UK and large Italian banks, with the most impacted unsurprisingly being the smaller banks from the GIIPS countries. Of the larger banks, Commerzbank, RBS and Unicredit are below average. For RBS this largely reflects the stringent rules applied to securitisation positions, whereas for Unicredit and Commerzbank it reflects the sovereign haircuts we apply. Looked at from the opposite perspective, we are surprised by just how resilient some of the banks are. Despite an economic recession (adverse scenario), a triple sovereign default and 20% haircuts on Italian and Spanish sovereign debt, the average core Tier 1 ratio for our subgroup is a still relatively healthy 6.2%. Furthermore, most of the larger banks have ratios in excess of that average.
Overall results Core tier 1 ratios under four scenarios (%)
Despite an economic recession (adverse scenario), a triple sovereign default and 20% haircuts on Italian and Spanish sovereign debt, the average core tier 1 ratio for our sub-group is a still relatively healthy 6.2%. Furthermore, most of the larger banks have ratios in excess of that average.
15.0%

10.0% Adv erse 5.0% GIP GIIPS

0.0%

Source: European Banking Authority, SG Cross Asset Research

16

18 July 2011

Sy dbank Danske Jy ske SEB Nordea Swedbank KBC DnB NOR SHB HSBC Credit Agricole Erste Raif f eisen Lloy ds AIB Dexia Santander Barclay s BNP Paribas Intesa Deutsche Bank BBVA Natixis RBS UBI Unicredit Commerzbank Popolare Ireland Sabadell Bankinter Popular BES Alpha Bank MPS BCP BPI EFG Eurobank NBG Piraeus

-5.0%

Sy dbank Danske Jy ske SEB Nordea Swedbank KBC DnB NOR SHB HSBC Credit Agricole Erste Raif f eisen Lloy ds AIB Dexia Santander Barclay s BNP Paribas Intesa Deutsche Bank BBVA BPCE RBS UBI Unicredit Commerzbank Popolare Bank of Ireland Sabadell Bankinter Popular BES Alpha Bank MPS BCP BPI EFG Eurobank NBG Piraeus

Everything you wanted to know about stress tests

The Credit Angle: Two-way Causality in the Stress Tests

Please see the following report for details on trade recommendations: EU Banks: Stressed, but not tested.

Like the equity market, credit market reaction to the stress tests is also likely to be relatively muted in our view, with few easily decipherable surprises discovered within large volume of disclosures. The focus on the relatively low level of capital required as a result of these tests may cause market participants to dismiss them. Indeed, during the analysts call on Saturday, 16 July 2011, the Chairman of the EBA himself noted that he is uncomfortable with the 2.5bn of required capital raising in light of how the sovereign situation has deteriorated and stated that he believed more strengthening of capital is needed. If there is any good news concerning this exercise for the credit markets, it is that it is over. However, we do not envision a flood of new bank debt issuance in the short term for two reasons. One, the European bank earning season is upon us with attending constraints on the ability of some to raise funding during the quiet period. As institutions publish their results, we are likely to see meaningful attempts by the primary funding markets to re-open. Two, credit market sentiment is likely to remain focussed on the fall-out of the Eurozone and possible US budget crisis. These will continue to influence the overall cost of capital and drive bank managements decisions with respect to how to fund their operations. It may be more advantageous to de-lever than raise funds in the international capital markets. This would clearly have attending feedback into the bank-to-sovereigns and economy causality.

We do not envision a flood of new bank debt issuance in the short term.

SovX Index and iBoxx subordinated financials index

500
450

1.2 1
0.8

400
350

0.6 0.4
0.2

300 250
200

0 -0.2
-0.4

150 Ja n-'11

-0.6 Feb-'11 Ma r-'11 Apr-'11 SovX Ma y-'11 Jun-'11 Jul -'11

i Boxx Fi nancial Subordinated

Source: SG Cross Asset Research, Markit

30 da y l a g correlation: SovX vs Sub Fins

Liquidity and funding matter to credit markets

The European Banking Authority 2011 stress test does not include a stress of liquidity while funding stress is not really tested, in our view. The framework of this test is largely concentrated on the asset side of the balance sheet to derive capital requirements, while the complexity of adding liability stresses to the process are largely absent. However, during the conference call on Saturday, 16 July 2011, the EBA noted that it had recently undertaken a non-publicly disclosed liquidity stress scenario test and discussed these results with national

18 July 2011

17

Everything you wanted to know about stress tests

regulators. When pressed on this issue during Q&A, the EBA noted that it believed it was not appropriate to publish the results of this test.
It is also important to note that these tests are further constrained by the lack of harmonized definitions across the EU with respect to certain important measures.

Funding costs are assumed to be driven by credit spreads which are linked to the sovereign in the stress test, again highlighting the sovereign-to-bank causality. However, these are modelled within a constrained methodology, as the recent jump to default in the credit spreads of some countries is clearly not captured in the funding costs exercise. In addition, the tests permit the re-pricing of loan portfolios to capture 50% of the increase in funding costs while new loans are assumed to replace maturing loans, in the constant balance sheet assumption. This could lead to the overstatement of interest revenues and an understatement of interest expenses in the real world, but clearly as noted (please refer to Does it really matter) there are limits to what can be modelled. It is also important to note that these tests are further constrained by the lack of harmonized definitions across the EU with respect to certain important measures. For example negative Available for Sale Reserves were subject to prudential filters at national discretion and this obviously impacts the starting line with respect to available Core Tier 1 capital. In addition, it is well publicized that the non-inclusion of pro-cyclical loan loss reserves for Spanish banks and the exclusion of silent participations in the German banks had a detrimental impact on their respective starting Core Tier 1 levels, despite inclusion at the national level. Theses impacts can be adjusted for, but focusing on the headline number as a true cross-European comparison may lead to type two errors, that is, false conclusions.

The adverse scenario & bank credit default swaps

Amongst the largest EU banks, the results of the adverse stress case scenario relative to 5year subordinated credit default swaps are shown below. There are clear outliers, but for the most part, the market tends to favour those banks with higher core Tier 1 ratios in an adverse scenario.
Adverse Scenario Core Tier 1 Capital vs. 5-year Subordinated CDS on 15 July 2011 for select EU banks
1000

Popular 900 800 700 600

500 Popolare Bankinter Dexia

Sabadell
MPS

R = 0.2385 Unicredit Commerzbank

UBI

Lloy ds

400
300

BBVA Santander Intesa

KBC

RBS
Raif f eisen Natixis Credit Agricole

Barclay s Erste BNP Paribas Deutsche Bank HSBC Swedbank Nordea DnB NOR SEB Danske

200 100 0 4.0%

SHB

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

11.0%

12.0%

13.0%

14.0%

Source: SG Cross Asset Research, European Banking Authority, Bloomberg

18

18 July 2011

Everything you wanted to know about stress tests

Stressing sovereign exposure & bank credit default swaps

As per the test in (please refer to Incorporating Sovereign Exposures), applying the 50% haircut on sovereign exposures to Greece, Ireland and Portugal, as well as a 20% haircut to sovereign exposures to Spain and Italy, shifts Spanish and Italian banks downwards in the core Tier 1 equity tables. Relative to 5-year subordinated credit default swaps, we have a better fit, in terms of correlation, than is seen in the EBA adverse scenario core Tier 1 test results. However, we caution that this is a static, one-step, assessment with other dynamic factors that would likely occur in this scenario, such as increased funding costs and other losses, not accounted for. Nevertheless, from this example, it appears as though the credit markets are applying a much more stringent test than the EBA to some EU banks.

Stressed sovereign core Tier 1 ratios vs. 5 year subordinated CDS on 15 July 2011 for select EU banks
1000

Popular 900 R = 0.556 800 700 600 Unicredit

500 Popolare Dexia

Bankinter Sabadell
MPS

UBI

BBVA Commerzbank Intesa RBS Credit Agricole Raif f eisen Erste Deutsche Bank HSBC Swedbank SEB Nordea Danske Santander Lloy ds KBC

400
300

Barclay s BNP Paribas

200 100 0 2.0%

SHB

DnB NOR

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

11.0%

12.0%

13.0%

14.0%

Source: SG Cross Asset Research, Bloomberg

18 July 2011

19

Everything you wanted to know about stress tests

APPENDIX
ANALYST CERTIFICATION The following named research analyst(s) hereby certifies or certify that (i) the views expressed in the research report accurately reflect his or her personal views about any and all of the subject securities or issuers and (ii) no part of his or her compensation was, is, or will be related, directly or indirectly, to the specific recommendations or views expressed in this report: Hank Calenti, CFA, Philip Richards, James Invine

SG RATINGS BUY: expected total return of 10% or more over a 12 month period. HOLD: expected total return between -10% and +10% over a 12 month period. SELL: expected total return of -10% or worse over a 12 month period. Sector Weighting Definition: The sector weightings are assigned by the SG Equity Research Strategist and are distinct and separate from SG research analyst ratings. They are based on the relevant MSCI. OVERWEIGHT: sector expected to outperform the relevant broad market benchmark over the next 12 months. NEUTRAL: sector expected to perform in-line with the relevant broad market benchmark over the next 12 months. UNDERWEIGHT: sector expected to underperform the relevant broad market benchmark over the next 12 months.

Equity rating and dispersion relationship

250 48% 40%

200

150 50% 100 39% 13% 50 48%

0 Buy
Companies Covered

Hold
Cos. w/ Banking Relationship

Sell

Source: SG Cross Asset Research

MSCI DISCLAIMER: The MSCI sourced information is the exclusive property of Morgan Stanley Capital International Inc. (MSCI). Without prior written permission of MSCI, this information and any other MSCI intellectual property may not be reproduced, redisseminated or used to create any financial products, including any indices. This information is provided on an as is basis. The user assumes the entire risk of any use made of this information. MSCI, its affiliates and any third party involved in, or related to, computing or compiling the information hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability or fitness for a particular purpose with respect to any of this information. Without limiting any of the foregoing, in no event shall MSCI, any of its affiliates or any third party involved in, or related to, computing or compiling the information have any liability for any damages of any kind. MSCI, Morgan Stanley Capital International and the MSCI indexes are service marks of MSCI and its affiliates or such similar language as may be provided by or approved in advance by MSCI. IMPORTANT DISCLOSURES Banco De Sabadell SG acted as joint bookrunner in Banco Sabadell's bond issue (4.5% 11/02/13 EUR). Banco Popolare SG acted as co-lead manager in Banco Popolare's capital raising. Bankia SG is acting as as co-lead manager in Bankia's IPO Barclays SG acted as bookrunner in Barclays's covered bond issue. BBVA SG acted as Joint Bookrunner in BBVA's rights issue. BBVA SG acted as joint bookrunner of BBVA's bond issue. BNP Paribas SG is acting as financial advisor to SFPI, 100% owned by the Belgian State, which holds a 25% stake in Fortis Bank SA/NV Crdit Agricole SG acted as joint bookrunner in Crdit agricole's covered bond issue. Crdit Agricole SG acted as joint book runner in Credit Agricole's covered bond issue. EADS SG is mandated lead arranger of the loan granted to Republic of Brazil to finance the acquisition of helicopters from EADS Group. HSBC SG acted as joint manager in HSBC's High Grade Covered bond. HSBC SG acted as co-manager in HSBC's bond issue. HSH Nordbank SG acted as joint book runner in HSH's bond issue (2.5% 28/07/14 EUR). Intesa Sanpaolo SG acted as co-lead manager in Intesa Sanpaolo's rights issue. Intesa Sanpaolo SG acted as Joint lead manager in Intesa San Paolo's bond issue (3.25% 01/02/13 EUR). Intesa Sanpaolo SG acted as joint lead Manager in Intesa San Paolo's senior bond issue (4% 08/11/18 EUR). Intesa Sanpaolo SG acted as joint bookrunner in Intesa San Paolo's subordinated bond issue (9.5% 01/01/49 EUR).

20

18 July 2011

Everything you wanted to know about stress tests

Intesa Sanpaolo La Caixa (Caja de Ahorros y Pensiones de Barcelona) La Caixa (Caja de Ahorros y Pensiones de Barcelona) Nordea PKO Bank Polski Royal Bank of Scotland Santander Santander SNS REAAL Swedbank UniCredit UniCredit UniCredit

SG makes a market in Intesa Sanpaolo warrants SG acted as joint lead manager in La Caixa's bond.

SG is acting as joint bookrunner in La Caixa's senior bond issue.

SG acted as joint bookrunner in Nordea's covered bond issue. SG acted as joint bookrunner in PKO's senior bond issue. SG acted as joint lead manager in RBS's covered bond issue (4% 15/03/16 EUR). SG acted as joint bookrunner in Santander's covered bond issue (4.625% 21/06/16 EUR). SG acted as joint lead manager in Santander's senior bond issue (4.125% 04/10/17 EUR). SG acted as joint bookrunner in SNS Reaal's covered bond issue. SG acted as joint bookrunner in Swedbank's covered bond issue. SG acted as joint bookrunner in Unicredit's bond issue (18/07/12 EUR). SG acted as joint lead manager in Unicredito Italiano's senior bond issue (14/09/12 EUR). SG makes a market in Unicredito warrants

Director: A senior employee, executive officer or director of SG and/ or its affiliates is a director and/or officer of UniCredit SpA. SG and its affiliates beneficially own 1% or more of any class of common equity of BBVA, Intesa Sanpaolo. SG or its affiliates act as market maker or liquidity provider in the equities securities of BBVA, BNP Paribas, Banco De Sabadell, Banco Popolare, Banco Popular, Bankinter SA, Commerzbank, Crdit Agricole SA, EADS, ING Group, Intesa Sanpaolo, Santander, UniCredit SpA. SG or its affiliates expect to receive or intend to seek compensation for investment banking services in the next 3 months from BNP Paribas, Bank of Ireland, Commerzbank, Crdit Agricole SA, Dexia, EADS, ING Group, Intesa Sanpaolo, KBC, Royal Bank of Scotland, Santander, UniCredit SpA. SG or its affiliates had an investment client relationship during the past 12 months with BBVA, Banco De Sabadell, Banco Popolare, Bankia, Barclays, Crdit Agricole SA, HSBC, HSH Nordbank, Intesa Sanpaolo, La Caixa (Caja de Ahorros y Pensiones de Barcelona), PKO BANK POLSKI, Royal Bank of Scotland, SNS REAAL NV, Santander, Swedbank, UniCredit SpA. SG or its affiliates have received compensation for investment banking services in the past 12 months from BBVA, Banco De Sabadell, Banco Popolare, Bankia, Barclays, Crdit Agricole SA, HSBC, HSH Nordbank, Intesa Sanpaolo, La Caixa (Caja de Ahorros y Pensiones de Barcelona), Nordea, PKO BANK POLSKI, Royal Bank of Scotland, SNS REAAL NV, Santander, Swedbank, UniCredit SpA. SG or its affiliates managed or co-managed in the past 12 months a public offering of securities of BBVA, Banco De Sabadell, Banco Popolare, Bankia, Barclays, Crdit Agricole SA, HSBC, HSH Nordbank, Intesa Sanpaolo, La Caixa (Caja de Ahorros y Pensiones de Barcelona), Nordea, PKO BANK POLSKI, Royal Bank of Scotland, SNS REAAL NV, Santander, Swedbank, UniCredit SpA. SGAS had a non-investment banking non-securities services client relationship during the past 12 months with BBVA, BNP Paribas, Banco Popular, Bank of Ireland, Barclays, Commerzbank, Crdit Agricole SA, Dexia, HSBC, ING Group, Intesa Sanpaolo, KBC, PKO BANK POLSKI, Royal Bank of Scotland, Santander, UniCredit SpA. SGAS had a non-investment banking securities-related services client relationship during the past 12 months with BNP Paribas, Banco Popolare, Barclays, Commerzbank, Crdit Agricole SA, Dexia, HSBC, ING Group, Intesa Sanpaolo, KBC, PKO BANK POLSKI, Royal Bank of Scotland, Santander, UniCredit SpA. SGAS received compensation for products and services other than investment banking services in the past 12 months from BBVA, BNP Paribas, Banco Popolare, Banco Popular, Bank of Ireland, Barclays, Commerzbank, Crdit Agricole SA, Dexia, HSBC, ING Group, Intesa Sanpaolo, KBC, PKO BANK POLSKI, Royal Bank of Scotland, Santander, UniCredit SpA. SGCIB received compensation for products and services other than investment banking services in the past 12 months from BNP Paribas, Banco Popolare, Banco Popular, Bank of Ireland, Barclays, Commerzbank, Crdit Agricole SA, Dexia, EADS, HSBC, ING Group, Intesa Sanpaolo, KBC, PKO BANK POLSKI, Royal Bank of Scotland, Santander, UniCredit SpA.

FOR DISCLOSURES PERTAINING TO COMPENDIUM REPORTS OR RECOMMENDATIONS OR ESTIMATES MADE ON SECURITIES OTHER THAN THE PRIMARY SUBJECT OF THIS RESEARCH REPORT, PLEASE VISIT OUR GLOBAL RESEARCH DISCLOSURE WEBSITE AT http://www.sgresearch.com/compliance.rha or call +1 (212).278.6000 in the U.S.

The analyst(s) responsible for preparing this report receive compensation that is based on various factors including SGs total revenues, a portion of which are generated by investment banking activities. Non-U.S. Analyst Disclosure: The name(s) of any non-U.S. analysts who contributed to this report and their SG legal entity are listed below. U.S. analysts are employed by SG Americas Securities LLC. The non-U.S. analysts are not registered/qualified with FINRA, may not be associated persons of SGAS and may not be subject to the FINRA restrictions on communications with a subject company, public appearances and trading securities held in the research analyst(s) account(s): Hank Calenti, CFA Socit Gnrale London, Philip Richards Socit Gnrale London, James Invine Socit Gnrale London IMPORTANT DISCLAIMER: The information herein is not intended to be an offer to buy or sell, or a solicitation of an offer to buy or sell, any securities and has been obtained from, or is based upon, sources believed to be reliable but is not guaranteed as to accuracy or completeness. SG does, from time to time, deal, trade in, profit from, hold, act as market-makers or advisers, brokers or bankers in relation to the securities, or derivatives thereof, of persons, firms or entities mentioned in this document and may be represented on the board of such persons, firms or entities. SG does,, from time to time, act as a principal trader in debt securities that may be referred to in this report and may hold debt securities positions. Employees of SG, or individuals connected to them, may from time to time have a position in or hold any of the investments or related investments mentioned in this document. SG is under no obligation to disclose or take account of this document when advising or dealing with or on behalf of customers. The views of SG reflected in this document may change without notice. In addition, SG may issue other reports that are inconsistent with, and reach different

18 July 2011

21

Everything you wanted to know about stress tests

conclusions from, the information presented in this report and is under no obligation to ensure that such other reports are brought to the attention of any recipient of this report. To the maximum extent possible at law, SG does not accept any liability whatsoever arising from the use of the material or information contained herein. This research document is not intended for use by or targeted to retail customers. Should a retail customer obtain a copy of this report he/she should not base his/her investment decisions solely on the basis of this document and must seek independent financial advice. The financial instrument discussed in this report may not be suitable for all investors and investors must make their own informed decisions and seek their own advice regarding the appropriateness of investing in financial instruments or implementing strategies discussed herein. The value of securities and financial instruments is subject to currency exchange rate fluctuation that may have a positive or negative effect on the price of such securities or financial instruments, and investors in securities such as ADRs effectively assume this risk. SG does not provide any tax advice. Past performance is not necessarily a guide to future performance. Estimates of future performance are based on assumptions that may not be realized. Investments in general, and derivatives in particular, involve numerous risks, including, among others, market, counterparty default and liquidity risk. Trading in options involves additional risks and is not suitable for all investors. An option may become worthless by its expiration date, as it is a depreciating asset. Option ownership could result in significant loss or gain, especially for options of unhedged positions. Prior to buying or selling an option, investors must review the "Characteristics and Risks of Standardized Options" at http://www.optionsclearing.com/publications/risks/riskchap.1.jsp. Important European MIFID Notice: The circumstances in which material provided by SG European Fixed Income (Credit) & Forex Research, SG Commodity Research, SG Convertible Research and SG Equity Derivatives Research have been produced are such (for example, because of reporting or remuneration structures or the physical location of the author of the material) that it is not appropriate to characterize it as independent investment research as referred to in the European Markets in Financial Instruments Directive and that it should be treated as marketing material even if it contains a research recommendation (recommandation dinvestissement caractre promotionnel). However, it must be made clear that all publications issued by SG will be clear, fair and not misleading. For more details please refer to SGs Policies for Managing Conflicts of Interest in Connection with Investment Research posted on SGs disclosure website referenced herein. Notice to French Investors: This publication is issued in France by or through Socit Gnrale ("SG") which is authorised and supervised by the Autorit de Contrle Prudentiel and regulated by the Autorite des Marches Financiers. Notice to U.K. Investors: This publication is issued in the United Kingdom by or through Socit Gnrale ("SG"), London Branch . Socit Gnrale is a French credit institution (bank) authorised and supervised by the Autorit de Contrle Prudentiel (the French Prudential Control Authority). Socit Gnrale is subject to limited regulation by the Financial Services Authority (FSA) in the U.K. Details of the extent of SG's regulation by the FSA are available from SG on request. The information and any advice contained herein is directed only at, and made available only to, professional clients and eligible counterparties (as defined in the FSA rules) and should not be relied upon by any other person or party. Notice to Polish Investors: this document has been issued in Poland by Societe Generale S.A. Oddzial w Polsce (the Branch) with its registered office in Warsaw (Poland) at 111 Marszakowska St. The Branch is supervised by the Polish Financial Supervision Authority and the French Autorit de Contrle Prudentiel. This report is addressed to financial institutions only, as defined in the Act on trading in financial instruments. The Branch certifies that this document has been elaborated with due dilligence and care. Notice to U.S. Investors: For purposes of SEC Rule 15a-6, SG Americas Securities LLC (SGAS) takes responsibility for this research report. This report is intended for institutional investors only. Any U.S. person wishing to discuss this report or effect transactions in any security discussed herein should do so with or through SGAS, a broker-dealer registered with the SEC and a member of FINRA, 1221 Avenue of the Americas, New York, NY 10020. (212)-278-6000. Notice to Singapore Investors: This document is provided in Singapore by or through Socit Gnrale ("SG"), Singapore Branch and is provided only to accredited investors, expert investors and institutional investors, as defined in Section 4A of the Securities and Futures Act, Cap. 289. Recipients of this document are to contact Socit Gnrale, Singapore Branch in respect of any matters arising from, or in connection with, the document. If you are an accredited investor or expert investor, please be informed that in SG's dealings with you, SG is relying on the following exemptions to the Financial Advisers Act, Cap. 110 (FAA): (1) the exemption in Regulation 33 of the Financial Advisers Regulations (FAR), which exempts SG from complying with Section 25 of the FAA on disclosure of product information to clients; (2) the exemption set out in Regulation 34 of the FAR, which exempts SG from complying with Section 27 of the FAA on recommendations; and (3) the exemption set out in Regulation 35 of the FAR, which exempts SG from complying with Section 36 of the FAA on disclosure of certain interests in securities. Notice to Hong Kong Investors: This report is distributed in Hong Kong by Socit Gnrale, Hong Kong Branch which is licensed by the Securities and Futures Commission of Hong Kong under the Securities and Futures Ordinance (Chapter 571 of the Laws of Hong Kong) ("SFO"). This document does not constitute a solicitation or an offer of securities or an invitation to the public within the meaning of the SFO. This report is to be circulated only to "professional investors" as defined in the SFO. Notice to Japanese Investors: This publication is distributed in Japan by Societe Generale Securities (North Pacific) Ltd., Tokyo Branch, which is regulated by the Financial Services Agency of Japan. This document is intended only for the Specified Investors, as defined by the Financial Instruments and Exchange Law in Japan and only for those people to whom it is sent directly by Societe Generale Securities (North Pacific) Ltd., Tokyo Branch, and under no circumstances should it be forwarded to any third party. The products mentioned in this report may not be eligible for sale in Japan and they may not be suitable for all types of investors. Notice to Australian Investors: This document is issued in Australia by Socit Gnrale (ABN 71 092 516 286) ("SG"). SG is regulated by APRA and ASIC and holds an AFSL no. 236651 issued under the Corporations Act 2001 (Cth) ("Act"). The information contained in this document is only directed to recipients who are wholesale clients as defined under the Act. Notice to Canadian Investors: This document is for information purposes only and is intended for use by Permitted Clients, as defined under National Instrument 31-103, Accredited Investors, as defined under National Instrument 45-106, Accredited Counterparties as defined under the Derivatives Act (Qubec) and "Qualified Parties" as defined under the ASC, BCSC, SFSC and NBSC Orders http://www.sgcib.com. Copyright: The Socit Gnrale Group 2011. All rights reserved. This publication may not be reproduced or redistributed in whole in part without the prior consent of SG or its affiliates.

22

18 July 2011