Financial Pacific - In shape, or not Europe's banks after the workout (third party)

Wealth Management Research 15 September 2011

FinancialsIn shape, or not: Europe's banksafter the workout

• Following the European Banking Authority (EBA) stress test, wehave conducted our own WMR stress test which uses moreconservative assumptions on sovereign exposures for bankingbooks and trading books.

• We simulate a base and a worst-case scenario which showthat banks participating in the EBA stress test need betweenEUR 92bn and EUR 138bn capital, which we believe should bemanageable over the next 18 months, in our view.

• Based on our analysis, each isolated event is not a problem forthe sector, the challenge is if multiple sovereign losses occursimultaneously under a contagion scenario. European leadersrecently agreed to an expanded framework aimed at stemmingcontagion, but the political risks remain high.

Barry McAlinden, CFA, strategist, UBS [email protected], +1 212 713 3261

Claudia Sigl, analyst, UBS AGFabio R.J. Trussardi, analyst, UBS AGJens Anderson, FRM, analyst, UBS AG

This report that was originally published outsideof the US on 26 July 2011. This report has beencustomized for US distribution.

Executive SummaryOne of the major criticisms of the official European Banking Authority(EBA) banking stress test was that only positions held in the tradingbook, which are subject to fair value accounting, were realisticallytested. However, most positions in peripheral European governmentbonds are held in the banking book. In our scenario analysis, weadjusted for this fact by using the data on sovereign debt exposureand calculating the impact of adverse developments with alternativemethodologies , setting a Core Tier 1 (CT1) threshold of 7% andsimulating a base and worst case scenario.

Institutions most affected by capital shortfalls are domiciled in oneof the European peripheral countries. Excluding the non-peripherybanks, French and German banks have the highest sensitivity toperipheral government bonds. Nevertheless, the total capital shortfallof up to EUR 138bn should be manageable over an 18-monthhorizon, mitigating all other regulatory challenges and second-ordereffects. However, we believe that the underlying problems withinthe banking sector will remain in the medium to long term, mostprominently emphasizing the contagion effects from the Eurozonedebt crisis.

Related publications• The debt crisis: Next hurdles for the

Eurozone, 14 September• Greek debt deal placates markets, 22 Jul

2011• Credible stress test results, 19 Jul 2011• The debt crisis: Timing of a Greek default and

contagion effects, 18 Jul 2011• European bank and insurance exposure to

Greece, 21 June 2011

This report has been prepared by UBS Financial Services Inc. (UBS FS). Please see important disclaimers and disclosuresthat begin on page 18.

The crisis is here to stay

The European sovereign debt crisis has entered a phase in which even

major economies are challenged by their high debt levels. The piecemeal

approach employed by European political leaders since early 2010 no

longer suffices to ease the pressure and prevent contagion into core

European economies. We currently see little room for large-scale

initiatives that would rapidly improve the situation considering political,

economic and legal constraints. The EU leader summit failed to ease

market participants' concerns.

Direct sovereign exposure is not the only issue

It is quite difficult to identify all parties holding sovereign government

debt. According to the recent (EBA) data, banks are holding net direct

exposure to stressed sovereigns with a total amount of EUR 678bn

(Fig. 1). We appreciate the additional disclosures from the 2011 bank

stress test which we attach in a detailed list (see Appendix, table 4). The

participating banks account for 65% of the European banking sector.

Using this data, banks domiciled in the periphery (so called 'periphery'

banks) constitute 51 out of 90 banks in the EBA sample and hold almost

EUR 460bn (two-thirds) of the outstanding total peripheral debt. For a

detailed distribution please refer to Fig. 2.

These figures have not changed significantly to those of the previous year,

but have increased slightly. The combined exposure of banks not

domiciled in one of the periphery countries (so called 'non-periphery'

banks) is roughly EUR 220bn and is significantly concentrated in banks'

exposure to Italy (EUR 124bn; Fig. 3).

We think other weak countries would suffer from an increased risk

aversion and see their funding costs rise following a restructuring

announcement of Greece. Besides the other peripheral countries,

Portugal, Ireland, Spain and Italy, we would expect a further widening of

risk premiums for weak borrowers.

We think the most important difference compared to the default of

Lehman Brothers in 2008 is that markets were caught by surprise, other

banks were in a similarly critical situation and the exposures of investors

and business partners to the broker firm were entirely unclear. By

supporting Greece for a prolonged period, the Eurozone grants market

participants time to prepare for the event. Transparency on exposures was

increased with the recently published 2011 banking stress test results.

Theoretically mitigating contagion risk, most European 'non-periphery'

financial institutions should hold a manageable direct exposure to Greek

government bonds. Market disruptions and another temporary freeze of

global credit markets resulting from a default event on the periphery of

this scale should not be underestimated and including all second-order

effects of a sovereign default, we think there would be a sizable

economic impact on the Eurozone as a whole and beyond.

Periphery banks are heavily hit by their challenged sovereign. Historically,

banks domiciled in a stressed or defaulted country experienced multiple

factors of pressure, arising from direct and indirect exposure and further

Fig. 1: Distribution of direct net peripheral sovereign exposures, held by EBA banks

Total EUR 678bn

42%

5%2%

12%

39%

Greece

Ireland

Portugal

Italy

Spain

Source: UBS WMR; EBA, as of 26 July 2011

Fig. 2: Banks' total net direct sovereign exposure and share of domestic banks

in EUR mn

0

50

100

150

200

250

300

Greece Ireland Portugal Italy Spain

Banks' total net direct sovereign exposure

Share of domestic banks' net direct sovereign exposure


Fig. 3: Distribution of direct net peripheral sovereign exposures held by 'non-periphery' banks Total EUR 220bn

19% 13%

3%

8%

57%

Greece

Ireland

Portugal

Italy

Spain


Financials

Wealth Management Research 15 September 2011 2

deterioration of their operating environment. In addition to the above

mentioned contagion risks, those banks had to face bank-runs, dried-up

refinancing channels, extremely high funding costs and last but not least,

a dramatic asset deleveraging if not a wind-down. In a best-case scenario,

such a bank could be recapitalized by a foreign investor after a

deleveraging process.

Stressing European banks – the WMR

approach

The official stress test only stressed positions held in the trading book,

which are subject to fair value accounting. Only negligible fair value

changes were considered in the banking book where, in fact, most

positions in peripheral European government bonds are held. The largest

portion of the banking book is held to maturity, i.e. banks would only

need to realize losses on these positions in case of a haircut or longer-

dated valuation losses. These drawbacks leave room for improvement,

which we intend to capture by using data on sovereign debt exposures

and to calculate the impact of adverse developments with alternative

methodologies simulating a base and worst case scenario.

Our views for the European periphery

As a baseline scenario we assume that Greece will default on its debt and

that such an event may occur any time, driven either by reluctance of its

official creditors to continue extending new loans or unwillingness by the

Greek government to continue with its austerity program. In light of high

political uncertainty, our base case estimate for the timing of a default

event is March 2012. We think there is no obvious reason why a Greek

default would consequently trigger a default by Portugal or Ireland, and

we are convinced that both countries would refuse to participate in a

private sector involvement initiative such as has been recently concluded

for Greece. For Spain, the overall debt situation is sustainable and debt

levels are even much lower compared to peers like Italy and Belgium.

Hence, in our opinion, a default by Spain or Italy is highly unlikely.

However, a possible Spanish request for funding support following a

disruptive Greek default would trigger a sharp and persistent rise in bond

risk premiums, also for Italy, triggering sizable valuation losses. We refer

to our recent publication "The debt crisis: Timing of a Greek default and

contagion effects" for a more detailed view on each country.

WMR's stress test approach – tuning our assumptions

Compared to the official stress test, we apply stricter loss assumptions to

reflect our bleak sovereign outlook. In our base scenario, we assume

that Greece defaults on its debt, causing an additional 40% loss on

current book values of Greek government bonds. We assume this event

to trigger a widening of bond risk premiums for Portugal and Ireland by

500 basis points (current 5-year premiums are around 1300 and 1000

basis points, respectively), and by 200 basis points for Spain and Italy

(from currently about 300 basis points).

For a worst case scenario we assume a 60% additional loss on Greek

government bonds, defaults by Portugal and Ireland causing an additional

Financials


loss of 30% and an increase in bond risk premiums for Spain and Italy of

500 basis points.

Table 1: Summary of WMR scenarios 5-year bond durations assumed

Greece Ireland Portugal Spain Italy

40% haircut +500 bp spread widening +500 bp spread widening +200 bp spread widening +200 bp spread widening

(~25% fair value loss) (~25% fair value loss) (~10% fair value loss) (~10% fair value loss)

60% haircut 30% haircut 30% haircut +500 bp spread widening +500 bp spread widening

(~25% fair value loss) (~25% fair value loss)

WMR base scenario

WMR worst case

scenario

Source: UBS WMR.

Based on the disclosures from the EBA stress test, we treated the held-to-

maturity positions (HTM) differently from the trading book and the

available-for-sale category (AfS). Moreover, we summarized the Fair-

Value-Option (FVO) for banking book holdings in the trading book.

Default haircuts were applied across all books, as such an event would

trigger a write-down need for the banking book, too. We calculated

changes in market values for the trading book, the FVO in the banking

book and the AfS book only. Our approach is in line with IFRS accounting

rules.

• The first step we had to prepare was to recalculate the sovereign loss

assumptions already applied by EBA. For instance Greek government

bonds were EBA-stressed across maturities ranging from one month

to 15 years with unrealistically low losses ranging from 0.5% to

26.2%. Similarly, fair value changes for Irish, Portuguese and Italian

government bonds were too modest in the EBA stress test and have

already been materialized in reality.

• Next, we applied our scenarios as outlined in table 1 whereby we

simply assumed the same losses or haircuts across maturities (EBA

has used more differentiated losses based on market rates).

• For the spread impacts we assumed on average a 5-year duration

which translates into 5% loss per 100-basis points widening (100bp

= 1%), i.e. a 200-basis points widening would result in a loss of 10%

on Spanish and Italian governments bonds in our baseline scenario.

Equivalently, a 500bp widening results in a 25% loss. Please see on

the right for major WMR stress test assumptions.

Major WMR stress test assumptions

• a portfolio duration of 5 years on average for

each exposure stressed

• haircuts were applied to EBA reported

exposures as of first quarter end 2011

• heuristically applied duration rule ignoring

convexity effects

• partial analysis: ignores any hedges and

diversification effects

• static holdings until 2012, no further

provisioning ex-interim

• no change in the forecasted earnings to built-

up capital base

• only sovereign exposure was stressed, loan

book was stressed in the EBA stress tests

• linear adjustment of CT1 equity and risk-

weighted assets (RWA)

Haircut

A haircut is frequently defined as a discount to

the nominal value of an asset. While calculating

scenarios for the WMR stress test we faced the

difficulty that book values of assets are

unknown. For simplicity we applied haircuts, i.e.

valuation discounts, on reported book values. In

some cases this may overstate estimated capital

shortfalls. Because the largest amounts of

sovereign debt are held in the banking books (of

which approximately half are held-to-maturity)

and medium-term government bonds of Spain

and Italy were trading close to par at the EBA

reporting date, we believe that our approach

generates reasonable results. Source: UBS WMR

Financials


Table 2: Banks falling below the 7% CT1 ratio threshold in the WMR stress test in EUR mn; sorted by WMR worst case scenario Banks Country

Pre-Stress

2010

EBA

stressed

2012

WMR

base scenario 2012

WMR

worst case scenario

2012

WMR

base scenario

WMR

worst case scenario

Unicredit Italy 7.8% 6.7% 5.9% 4.9% 5'590- 10'973-

Royal Bank of Scotland (RBS) United Kingdom 9.7% 6.3% 6.2% 6.0% 5'788- 6'841-

EFG Eurobank Greece 9.0% 4.9% -2.4% -6.4% 4'627- 5'926-

Agricultural Bank of Greece (ATEbank) Greece 6.3% -0.8% -35.0% -62.7% 5'212- 5'352-

Commerzbank Germany 10.0% 6.4% 5.9% 5.4% 3'560- 5'019-

Societe Generale France 8.1% 6.6% 6.2% 5.9% 3'737- 4'996-

Deutsche Bank Germany 8.8% 6.5% 6.3% 6.0% 3'676- 4'987-

BNP Paribas France 9.2% 7.9% 7.1% 6.4% - 3'969-

Banco Santander Spain 7.1% 8.4% 7.4% 6.6% - 2'358-

NordLB Germany 4.6% 5.6% 5.2% 4.9% 1'898- 2'264-

DZ Bank Germany 8.2% 6.9% 6.1% 5.2% 1'117- 2'244-

Banco de Sabadell Spain 6.2% 5.7% 4.6% 3.0% 1'348- 2'217-

HSH Nordbank Germany 10.7% 5.5% 5.4% 5.2% 1'156- 1'281-

WestLB Germany 8.7% 6.1% 5.8% 5.6% 794- 934-

Bank of Ireland Ireland 8.4% 7.1% 5.9% 5.6% 712- 860-

Banco Pastor Spain 7.6% 3.3% 3.1% 2.7% 720- 785-

LBBW Germany 8.2% 7.1% 6.7% 6.4% 318- 719-

Barclays United Kingdom 10.0% 7.3% 7.1% 6.9% - 698-

Intesa Sanpaolo Italy 7.9% 8.9% 8.1% 6.9% - 346-

Bayerische Landesbank Germany 9.3% 7.1% 7.1% 7.0% - 28-

Core Tier 1 (CT1) ratio Capital Shortfall

Sources: UBS WMR; EBA; MS; RBS

Overall assessment of WMR stress test results

Our sample includes all banks participating in the 2011 EBA stress test.

Table 2 shows non-periphery banks that have the highest exposures to

peripheral government bonds according to the 2011 banking stress test,

considering both the highest absolute positions and highest positions

relative to the bank's Core Tier 1 equity. Table 2 is the reference for the

selection on issuers we comment further below.

Applying a CT1 threshold of 7%, 49 banks need to raise a total capital of

roughly EUR 92bn in our base scenario. Applying our worst case scenario

with a 7% CT1 hurdle, 69 banks need to raise a total capital of up to

EUR 138bn. Most of the affected institutions are domiciled in one of the

periphery countries.

The relevant banks from Greece and Cyprus show a capital shortfall of

EUR 26.6bn (base scenario) and EUR 33.3bn (worst case scenario), the

ones from Portugal roughly EUR 7.2bn and EUR 10bn respectively, while

the Irish banks would only need EUR 712mn and EUR 860mn,

respectively.

Excluding the non-periphery banks, French and German banks have the

highest sensitivity to peripheral government bonds. The three most

affected banks are Unicredit, RBS and Commerzbank (table 2).

How sensitive are our scenarios?

When assessing the sensitivity of our scenarios applying partial analysis,

i.e. assuming all other factors do not change, we gain interesting though

not unexpected insights.

Financials


When increasing the CT1 threshold from 5% to 7%, the capital shortfall

increases from EUR 2.5bn to EUR 41bn, i.e. a further capital shortfall of

EUR 50bn arises when testing our WMR base scenario. The capital

shortfall even increases from EUR 92bn to EUR 138bn under WMR's

worst-case scenario.

Assessing WMR's worst-case scenario, all else equal:

• an increase of a haircut on Greek government bonds by 20%

increases the capital requirements for European banks only by EUR

12bn;

• a 30% haircut on Portuguese government bonds increases the

sector's capital shortfall by less than EUR 2bn;

• a 30% haircut on Portuguese government bonds increases the total

capital shortfall by EUR 7bn;

• each 500bp additional spread widening on Spanish government

bonds increases the capital shortfall by EUR 18.2bn;

• each 500bp additional spread widening on Italian government bonds

increases the capital shortfall for stressed banks by EUR 25.2bn.

In essence, each isolated event is not a problem for the sector. The

challenge is if events occur simultaneously or in chain reactions, which

can be reasonably assumed, since defaults are usually clustered. Further,

the results confirm what markets already knew: Contagion to Spain and

Italy has to be stopped because both an incremental spread widening of

500bp for Spanish and Italian government bonds would, according to our

scenarios (with simplified assumptions), erode capital in the magnitude of

more than EUR 25bn (excluding held-to-maturity holdings!). This is more

than double the amount of a scenario in which haircuts on Greek (60%),

Portuguese (30%) and Irish (30%) sovereign bonds are taken

simultaneously (including held-to-maturity assets).

Banks - country by country

We only focus on the banks treated in table 2 failing the WMR stress test

hurdle.

Peripheral issuers from Greece, Portugal and Ireland

Two Greek banks, Agricultural Bank of Greece and EFG Eurobank, did

not pass the stress test, while Piraeus and Marfin Popular Bank

(Cyprus) passed it with a slight margin. We also highlight that generic

provisions were considered as a mitigation measure (only Spanish banks

and Greek banks did it) and we think that it partially invalidated the

results comparability across the banking sector. Irrespective of the very

weak stress test data, the Greek banks would most likely default along

with the sovereign and public entities.

Under the EBA stress test scenario, all four Portuguese banks passed the

exercise with limited margins even if accounting for the mitigation

measurers. The two major listed banks are Banco Comercial Portugues

(CT1 of 5.4% at 2012 under the EBA stress scenario) and Espirito Santo

Financial Group (CT1 of 5.1% at 2012 under the EBA stress scenario).

The four banks are all loss-making under the stress scenario in 2012, and

hence consuming capital. The four banks will have to implement the

Fig. 4: WMR stress test sensitivities Partial analysis of capital shortfalls

Change in

haircut

Change in

spreads

Geeece +20% +12

Ireland +30% +2

Portugal +30% +7

Spain +500bp +18

Italy +500bp +25

Increase in capital

shortfall

in EUR bn

Source: UBS WMR; EBA

Financials


already declared actions in order to achieve the required 6% CT1 level by

September 2011.In terms of sovereign exposures, the largest holdings are

of Portuguese bonds, while the exposures to the other peripheral

countries appear manageable.

German banks

The 12 participating German banks passed the official EBA stress test.

However, we believe that one of the main factors having contributed to

the good result is the assumption of no credit spread widening in all the

maturities for the German sovereign bonds. Obviously German banks

have a high exposure to their domestic sovereign bonds and an assumed

shock to sovereign credit spreads would have penalized them, resulting in

worse outcomes than the stress test assumes. Referring to the official

test, we do not see any particular implication for German banks.

Applying WMR's scenarios, German banks have the highest sensitivity to

peripheral government bonds, 8 German banks need to raise a total

capital of roughly EUR 12.6bn in our base scenario. Applying our worst

case scenario, 9 banks need to raise a total capital of up to EUR 18bn.

Even if these capital needs should be manageable and the German

banking capitalization is well below European peers, the critical aspect is

within the tighter capital requirements according to Basel 3. Especially

Landesbanks are most vulnerable, as the treatment of existing silent

participations, which form a major part of their current T1 capital, will

mostly not meet the eligibility criteria for T1 capital under Basel 3 (except

for the grandfathering period).

The uncertainty about the German banking sector comes on top of an

already long list of concerns. The critical problems of high fragmentation,

obstacles to consolidation, structurally poor profitability, weak business

models and the future of the Landesbanken remain as yet unresolved.

According to Moody's, revenues in German retail banking have fallen by

more than 20% over the past nine years and are likely to stagnate in the

medium term. This is clearly a credit negative for banks counting on the

domestic market to make up for lost longer-term earnings potential in

international banking or capital market activities.

Deutsche Bank, a globally active player in the banking segment, passed

the official test with a CT1 ratio of 6.5% in the adverse scenario,

accounting for a tough 44% increase in RWA. However in Deutsche

Bank's case we argue that the provisioning impact under the adverse

scenario was not really severe, assuming a debatable 8% decline in 2011

and an almost flat value for 2012 compared to the 2010 value. Referring

to a capital need of EUR 3.6bn to EUR 5bn resulting from WMR's stress

scenarios we judge the position of Germany's market leader comfortable.

We expect a strong organic capital generation due to the benefits from

recent market share gains.

Commerzbank: passed the official test with a ratio of 6.4%; hit by its

commercial real estate exposure and legacy assets. More realistically,

Commerzbank had to face a significant increase of provisions (roughly

+58% in 2012 versus the 2010 level) in the adverse scenario.

Nevertheless, Commerzbank should be able to withstand a loss scenario

without further capital injection needs and funding difficulties, given the

Financials


recent conversion of the majority of silent participations into core capital

and the continued government support. Applying our base and worse

case loss rates would hit core capital significantly. Under the terms of the

state aid approval, Pfandbriefe subsidiary Eurohypo needs to be sold by

2014.

BayernLB is required by the EC to reduce its balance sheet by 2013 to

40-50% compared to 2008 level due to state aid approval. In our view,

capitalization levels are sufficient but not above average. Applying a 7%

CT1 ratio will result in a capital shortfall of EUR 28mn, which is quite

comfortable. The threat of privatization is remote for the time being.

DZ Bank passed our stress test wit a capital shortfall in the worst case

scenario of EUR 561bn. As an integral part of the cooperative Financial

Services Network (FinanzVerbund) DZ Bank serves the local cooperative

banks and benefits from the integration and the comprehensive

protection scheme within the cooperative sector.

HSH Nordbank passed the official test with a CT1 of 5.5%, which is

close to the minimum requirement. This could put the bank under the

closer scrutiny of the regulator and market participants. EBA requested

specific steps to strengthen the bank's capital position. HSHN has a highly

capital-intensive lending franchise and limited access to capital.

Consequently, the bank needs to shrink its balance sheet to release

capital. Its dependence on costly support from owners implies major

uncertainties given the support providers’ own stretched financial

flexibility (the state of Schleswig-Holstein and the city-state of Hamburg).

LBBW already received the EC state aid approval that requires a total

asset reduction of 40% from 2008 to 2013. A mandatory haircut for all

deep subordinated bonds is imposed. Furthermore, the bank should be

converted into a joint-stock company by 2013. The capital shortfall of up

to EUR 719 bn in WMR's stress scenarios should be manageable.

NordLB passed the official test with a CT1 of 5.6%, which is close to the

minimum requirement. This could put the bank under further. EBA

requested specific steps to strengthen the bank's capital position. In our

scenario, NordLB has to face a capital shortfall from EUR 1.9 to EUR

2.3bn. Nevertheless, NordLB continues to benefit from a very high

probability of cross-sector, owner or systemic support.

WestLB passed our worst case scenario with a capital short-fall of

roughly EUR 930mn, mitigating the deleveraging activities currently

passing on. Uncertainties prevail as to whether or not WestLB's business

model will stand the test of time. In our view, its viability is questionable

despite the restructuring efforts. As time is running out for an earnest

search of equity investors or merger candidates, we cannot exclude an

orderly liquidation of WestLB, which we consider as the most sensible

solution.

Helaba decided to step back from the official test, as EBA did not

recognize bindingly agreed measures by its owners for restructuring and

adapting the silent participations in the amount of EUR 1.92 billion to the

Basel 3 criteria. From the point of view of the EBA, the silent participation

Financials


by the State of Hesse would have to be reported together with the "non-

hardened silent participations" and would have brought their CT1 ratio

below the 5% hurdle. In our view, Helaba has relatively stable financial

fundamentals and franchise compared with other Landesbanks. Due to its

conservative business profile, the bank's comparatively low capitalization

level is mitigated.

UK banks

UK banks did well in the EBA stress test. An explanation for this is their

higher starting CT1 ratio, on average near 10%, compared to the pool

average of 8.9%. We highlight however, that under the EBA adverse

scenario, the average UK CT1 ratio fell to 7.6%, close to the average of

7.7% of the 91 measured banks. Obviously the UK banks are more then

averagely leveraged to a potential economic UK downturn. The 2012 core

capital position under the adverse scenario was 7.5% across the four

major UK banks (RBS, Barclays, Lloyds, HSBC). An offsetting factor is that

EBA methodology used static balance sheets that do not reflect the

deleveraging process currently ongoing among the UK banks. For most

UK banks, loan losses have peaked and started to decline. However, a

recovery to "normal" levels may be protracted by a sluggish economic

development. Another challenge is refinancing the large share of short-

term bond market funding. In the WMR worst scenario, the UK banks will

come up with a capital shortfall of up to EUR 7.5bn, which should be

manageable.

RBS passed the EBA test, but with a CT1 ratio of 6.3%, but assumed a

debatable 26% decline in provisions for 2012 under the adverse scenario

compared to the 2010 level. RBS will come up with a capital shortfall of

up to EUR 6.8bn in our worst case scenario which is 91% of the total UK

banks' capital need. The bank announced a five-year restructuring plan,

but we think even after its planned reduction of non-core assets, RBS

would maintain a huge balance sheet and a high sensitivity to stress

events. That said, we think RBS may manage to improve its stand-alone

credit profile towards the start of the gradual introduction of Basel 3 by

2013.

Barclays passed the EBA test with a CT1 ratio of 7.3%, but assumed

declining provisions for 2012 under the EBA adverse scenario compared

to the 2010 level (-1% for 2011 and -5% for 2012). Barclays will come

up with only a capital shortfall of up to EUR 700mn in our worst case

scenario.

French banks

All French banks passed the official EBA test. Nevertheless, the picture

looks different when applying our WMR worst case scenario, under which

most French banks would need EUR 11.8bn of fresh capital. For BNP

Paribas and Societe Generale, the magnitude of decline in CT1 ratios

already under the EBA stress test is remarkable. Apart from the stress test

results, we still have doubts about the likely aggressive RWA calculation

of the French banks, as the ratio of RWA to total assets, for all of them, is

much lower than the sector average.

BNP Paribas showed a comfortable CT1 of 7.9% in the EBA adverse

scenario, despite an almost nil profit calculated for 2012. The CT1 ratio

Financials


shrinks to 7.1% in WMR base case scenario. BNP Paribas does not score

particularly well on the sovereign exposure as it has the highest average

duration of its peripheral exposure (mostly to Italy) among the French

banks with 8 years. Hence, WMR's worst case scenario sheds different

light on BNP's capital position under which the bank would need nearly

EUR 4bn capital. We judge the WMR worst-case capital shortfall as

manageable, even though we had expected higher profitability resilience

(just EUR 61mn profit in 2012 under the EBA adverse scenario). In

addition, BNP Paribas has close to 50% of its debt obligations maturing

over the next two years, which increases refinancing risk. We do not feel

comfortable with its likely aggressive RWA calculation that could lead to

capital shortfall given the current regulatory uncertainty.

Societe Generale passed the EBA test with a capital position of 6.6%,

compared the other French banks. We additionally highlight that the

assumed increase in provisions under the adverse scenario was not

particularly severe (+33% vs. 2010 level for 2011E, but just +6% for

2012E). Under our WMR worst case scenario, Societe Generale would

need EUR 5bn, the largest amount among French banks due to the

mediocre results on its loan books. The exposure to troubled sovereigns

appears comparatively moderate and the duration to these sovereign

bonds is at 3 years below the 6-year sector average.

Credit Agricole passed the EBA adverse scenario with a sound 8.5% CT1

ratio, even increased compared to the 8.2% starting point. In WMR's

worst-case scenario, Credit Agricole's CT1 ratio declines to 7.1% (8.2%

in our base scenario). Peripheral exposure is less than for BNP thanks to

lower holdings in Italian government bonds. However, the 4-year duration

of its peripheral sovereign portfolio is below sector average (6 years).

Regarding the RWA calculation, Credit Agricole has the lowest RWA

compared to the sector average which could be an indication that the

banking group's CT1 is overstated. We do not feel comfortable with

Credit Agricole's likely aggressive risk-weighted assets calculation that

could lead to capital shortfall given the current regulatory uncertainty.

Italian banks

All the five Italian banks participating in the official EBA stress test passed

it with an average 7.3% CT1 ratio under the adverse scenario, accounting

for rights issues decided by April 2011 (roughly EUR 10mn), but excluding

some hybrid instruments usually considered part of regulatory capital. It is

worth mentioning that all the Italian banks were loss-making in the

adverse EBA scenario and hence consuming capital. Moreover, Italian

banks did not include the generic provision as a mitigation measure to

boost the capital adequacy, while Spanish and Greek banks did. As a

reference, generic provisions for the two major Italian banks would have

been roughly EUR 3bn for Unicredit and EUR 2.8bn for Intesa Sanpaolo.

Italian banks stressed the broad definition of defaulted assets (stated non-

performing loans, past-due loans and restructuring loans) while other EU

peers just focused on a part of them (i.e., Spanish banks focused on the

stated non-performing loans only). In our WMR worst-case scenario, we

estimate a total capital shortfall of EUR 23bn for Italian banks.

Intesa Sanpaolo showed the best result with a CT1 ratio of 8.9% in the

EBA stress test. Intesa's CT1 would deteriorate materially to 6.9%

Financials


(marginally below the 7% threshold) under our WMR worst case scenario,

mainly due to its large exposure to Italian government bonds which

accounts for 220% of CT1. Intesa's sovereign exposure to Greece, Ireland

and Portugal is negligible with 3% of CT1.

Unicredit passed the EBA stress test with a CT1 ratio of 6.7% in the

adverse scenario in 2012. However, it failed both the WMR base and

worst case scenarios, mainly due to its weaker lending portfolio. Unicredit

did not include the EUR 3bn convertible in its capital calculation, but just

indicated it as a mitigation measure (unlike Santander for example). Thus,

excluding the convertible, Unicredit would need about EUR 11bn to meet

the 7% hurdle in our WMR worst case scenario and still EUR 6bn under

the WMR base case scenario. As for all Italian banks, Unicredit's Achilles

heel is clearly the material sovereign bond exposure.

Spanish banks

Five Spanish banks failed the EBA stress test with a threshold of 5% CT1,

and in total 12 banks fall below the 6% level. The EBA's 2010 starting

point definition of core capital disadvantaged Spanish banks as it did not

allow the use of mandatory convertible notes or existing anti-cyclical

provisions. However, we note that generic provisions were considered as

an allowed mitigation measure (besides Spanish banks, only Greek banks

used generic provisions as mitigation measure) and the treatment of the

convertibles differs bank by bank, i.e., Santander included it for the 2012

capital calculation whereas BBVA did not. We also point out that in the

stress test, the Spanish banks just considered the stated non-performing

loans, without including sub-standard loans, past-due loans, restructuring

loans and real estate repossessed assets. As other EU banks have used a

more stringent approach, we think that all these discrepancies invalidate

the comparability of results across the banks and possibly draw a better

portrait of the Spanish banking system than exists in reality. As the

outcome from the stress test, Banco Pastor and four Cajas (Catalunacaica,

CAM, Unnim and Caya3) failed the test. The remaining Spanish domestic

banks were between 5% and 6% capital under the EBA adverse scenario.

Applying WMR's worst-case scenario, Spanish banks would need EUR

31.5bn to fulfil the 7% CT1 criteria.

Santander, while having passed the EBA adverse scenario under which

the bank stays profitable, did not pass the WMR worst case scenario.

Under the EBA stress test the calculated 2010 CT1 ratio was 7.1% versus

the 8.8% CT1 calculated by the company itself. The difference mainly

comes from the roughly EUR 4bn of intangibles and EUR 7bn mandatory

convertible bond with EUR 14 strike price and maturity in 2012, which is

fully accounted as core capital in the EBA stress test. As the convertible

was sold to Santander's retail clients, we doubt that Santander will finally

ask its clients to convert at current adverse conditions and bear the full

loss. That is why our stress test considered just EUR 4bn instead of EUR

7bn (55% of the EUR 14 conversion price, market price of EUR 7.98 per

share). On the positive side, Santander's international presence mitigates

the performance pressure coming from the Spanish market. The banking

group has a market share around 20% in Spain, including some 5%

through its subsidiary Banco Espanol de Credito (Banesto).

Based on our WMR worst case scenario, we estimate that Santander

Financials


would need at least EUR 2.3bn fresh core capital outright plus an

additional EUR 3bn we think it would need in order to refinance part of

the above described convertible note out of the money. Should Santander

take an active role in supporting the Spanish banking sector, i.e., through

bailouts or takeovers of Cajas (probable through its subsidiary Banesto),

capital amounts would be higher.

BBVA remained profitable under the EBA adverse scenario and passed

the WMR worst case scenario. We highlight that BBVA, unlike Santander,

did not include a mandatory convertible of EUR 2bn issued in July this

year in its calculation. The convertible was just indicated as a mitigation

measure (same treatment of the convertible as Unicredit). Similar to

Santander, BBVA's credit quality is also supported by its international

diversification but suffers from elevated NPLs in Spain. Despite the fact

that our WMR stress test did not reveal any immediate capital needs for

BBVA, we are cautious on the name due to sovereign contagion concerns

and potential headline risks from the Caja sector. Moreover, loan-loss

coverage ratios particularly on commercial real estate are below sector

average as for all Spanish banks.

Banco Pastor didn't pass the WMR stress scenario test, ending it with a

CT1 of just 2.7% and a consequent capital need of roughly 785mn.

Similarly the bank had not passed the EBA stress test, achieving a CT1 of

just 3.3% under the adverse scenario. We believe that Banco Pastor’s

profitability is likely to remain under pressure, as the bank is exposed to

Spain’s property bubble and overleveraged economy, leading to major

asset quality issues and subdued or even negative loan growth. We

expect loan loss provisions to remain high for some time, even though we

acknowledge that the anti-cyclical generic provisions might ease some of

the pressure, and are concerned that a rising interest rate environment

could hamper the fragile stabilization in asset quality.

Banco de Sabadell passed the EBA stress test with a stretched CT1 ratio

of 5.7% even accounting for the recognized mitigating measures under

the adverse scenario. However in the WMR stress scenario, the bank

failed, showing a capital shortfall of EUR 2.2bn. We believe that

Sabadell’s profitability is likely to remain under pressure as the bank is

exposed to Spain’s property bubble and overleveraged economy, leading

to major asset quality issues and subdued or even negative loan growth.

We expect loan loss provisions to stay high for some time, even though

we acknowledge that the anti-cyclical generic provisions might ease some

of the pressure, and are concerned that a rising interest rate environment

could hamper the fragile stabilization in asset quality. We expect high

funding costs to abide, owing to the very competitive deposit market.

Dexia Group is heavily exposed to peripheral sovereign debt, to Greece

in particular. The group passes our worst case scenario which builds on a

EBA CT1 of 15.2bn. Excluding the phasing-in of deductions according to

future Basel 3 regulations will significantly weigh on Dexia's capitalization

needs: given our worst case scenario, Dexia would face a capital shortfall

of another EUR 5bn. Even if capital injections and liquidity facilities

provided by the majority owners – the French and Belgian governments –

should help Dexia Group to withstand our worst-case scenario, we remain

negative on Dexia's fundamentals.

Financials


Conclusion

We think the low EBA stress test headline capital need of EUR 2.5bn does

not help make market participants more comfortable with banks' risks in

light of a sovereign debt crisis. As stated, the test results provided a

sugar-coated picture of the banking industry's stress resistance. However,

a 7% CT1 ratio requirement instead of the used 5%, and tougher

assumptions on government bond losses should enhance the value of

such a stress test.

Validity of WMR stress test results

Each stress test is based on assumptions. Therefore, we stress that

absolute loss amounts and amounts of capital needed shall be interpreted

rather as an indication and not used as the exact number. Despite the

usage of simplified assumptions, we do think that our results give a good

indication regarding the vulnerability and sensitivity of banks to the

various peripheral European sovereign exposures.

Concerns within the banking sector remain

Our largest concern regarding the debt crisis and a Greek default is

contagion to other weak peripheral countries. Further, we are concerned

by second-order effects (renewed disruptions in the interbank market in

Europe; general rise in risk premiums).

We believe that the underlying problems within the banking sector will

remain in the medium term: a lack of growth, a large dependence on

wholesale funding, regulatory pressure, significant exposure to sovereign

risk, high risk of political intervention, protracted subdued profitability

and a lack of accounting comparability.

Uncertainty about the banking sector comes on top of an already long list

of concerns. Weaker banks will likely face lower net interest income

owing to higher funding costs as a result of possible rating downgrades,

and owing to shrinking loan books as result of resized balance sheets.

Financials


Table 3: Overview of WMR stress test results

in EUR mn; by countryBanks Country

Pre-Stress 2010 EBA stressed 2012

WMRbase scenario

2012

WMRworst case scenario

2012

WMRbase scenario


Oesterreichische Volksbank AG Austria 6.4% 4.5% 4.3% 4.2% 909- 958-

Erste Bank Austria 8.7% 8.1% 8.0% 7.8% - -

RBI (Raiffeisen Bank International) Austria 8.1% 7.8% 7.8% 7.8% - -

Dexia Belgium 12.1% 10.4% 8.6% 7.0% - -

KBC Bank Belgium 10.5% 10.0% 9.4% 8.6% - -

Marfin Popular Bank Cyprus 7.3% 5.3% 0.5% -2.0% 1,851- 2,399-

Bank of Cyprus Cyprus 8.1% 6.2% 2.6% 0.3% 1,165- 1,666-

Danske Bank Denmark 10.0% 13.0% 12.9% 12.8% - -

Jyske Bank Denmark 12.1% 12.8% 12.6% 12.4% - -

Nykredit Denmark 8.8% 9.4% 9.3% 9.3% - -

Sydbank Denmark 12.4% 13.6% 13.6% 13.6% - -

OP-Pohjola Group Finland 12.2% 11.6% 11.6% 11.6% - -

Societe Generale France 8.1% 6.6% 6.2% 5.9% 3,737- 4,996-

BNP Paribas France 9.2% 7.9% 7.1% 6.4% - 3,969-

BPCE France 7.8% 6.8% 6.6% 6.4% 2,171- 2,889-

Credit Agricole France 8.2% 8.5% 8.2% 7.9% - -

Commerzbank Germany 10.0% 6.4% 5.9% 5.4% 3,560- 5,019-

Deutsche Bank Germany 8.8% 6.5% 6.3% 6.0% 3,676- 4,987-

NordLB Germany 4.6% 5.6% 5.2% 4.9% 1,898- 2,264-

DZ Bank Germany 8.2% 6.9% 6.1% 5.2% 1,117- 2,244-

HSH Nordbank Germany 10.7% 5.5% 5.4% 5.2% 1,156- 1,281-

WestLB Germany 8.7% 6.1% 5.8% 5.6% 794- 934-

LBBW Germany 8.2% 7.1% 6.7% 6.4% 318- 719-

WGZ Bank Germany 10.8% 8.7% 6.6% 4.4% 91- 561-

Bayerische Landesbank Germany 9.3% 7.1% 7.1% 7.0% - 28-

Dekabank Germany 13.0% 9.2% 9.0% 8.8% - -

Hypo Real Estate Germany 28.4% 10.0% 9.3% 8.2% - -

Landesbank Berlin Germany 14.6% 10.4% 10.0% 9.7% - -

Landesbank Hessen-Thüringen (Helaba) Germany 0.0% - - -

National Bank of Greece Greece 11.9% 7.7% 0.4% -3.6% 4,738- 6,831-

EFG Eurobank Greece 9.0% 4.9% -2.4% -6.4% 4,627- 5,926-

Agricultural Bank of Greece (ATEbank) Greece 6.3% -0.8% -35.0% -62.7% 5,212- 5,352-

Piraeus Bank Greece 8.0% 5.3% -3.3% -8.2% 4,028- 5,206-

TT Hellenic Postbank Greece 18.5% 5.5% -37.4% -77.6% 3,023- 3,068-

Alpha Bank Greece 10.8% 7.4% 3.2% 0.9% 1,916- 2,852-

OTP Bank Hungary 12.3% 13.6% 13.6% 13.6% - -

Bank of Ireland Ireland 8.4% 7.1% 5.9% 5.6% 712- 860-

Allied Irish Banks (AIB) Ireland 3.7% 10.0% 10.0% 8.2% - -

Irish life & Permanent Ireland 10.6% 20.4% 17.8% 17.3% - -

Unicredit Italy 7.8% 6.7% 5.9% 4.9% 5,590- 10,973-

Monte dei Paschi die Siena Italy 5.8% 6.3% 4.1% 0.8% 3,239- 6,656-

Banco Popolare SC Italy 5.8% 5.7% 4.7% 3.3% 2,230- 3,545-

UBI Banca (Unione die Banche Italiane) Italy 7.0% 7.4% 6.5% 5.1% 480- 1,800-

Intesa Sanpaolo Italy 7.9% 8.9% 8.1% 6.9% - 346-

Banque et Caisse d'Epargne de L'Etat (BCEE) Luxembourg 12.0% 13.3% 11.3% 8.8% - -

Bank of Valletta Malta 10.5% 10.4% 10.2% 10.1% - -

SNS Bank Netherlands 8.4% 7.0% 6.3% 5.7% 134- 263-

ABN Amro Bank NV Netherlands 9.9% 9.2% 9.1% 8.9% - -

ING Bank Netherlands 9.6% 8.7% 8.4% 8.1% - -

Rabobank Nederland Netherlands 12.6% 10.8% 10.7% 10.7% - -

DnB NOR Bank Norway 8.3% 9.0% 9.0% 9.0% - -

PKO Bank Polski Poland 11.8% 12.2% 12.2% 12.2% - -

Banco Commercial Portuguese (BCP) Portugal 5.9% 5.4% 3.5% 1.9% 2,391- 3,373-

Caixa Geral De Depósitos, SA Portugal 8.5% 6.2% 5.0% 3.7% 1,639- 2,548-


Financials


Table 3: Overview of WMR stress test results

in EUR mn; by countryBanks Country

Pre-Stress 2010 EBA stressed 2012

WMRbase scenario

2012


2012

WMRbase scenario



Espirito Santo Financial Group (ESFG) Portugal 6.4% 5.1% 4.3% 3.8% 2,106- 2,443-

Banco BPI Portugal 8.2% 6.7% 3.2% 0.4% 1,017- 1,667-

Nova Ljubljanska Slovenia 5.2% 5.3% 5.2% 5.1% 280- 301-

Nova Kreditna Slovenia 7.4% 8.0% 8.0% 8.0% - -

BFA-Bankia Spain 6.9% 5.4% 4.5% 3.1% 5,450- 8,242-

Caja de Ahorros y Pensiones de Barcelona (La Caixa)

Spain 6.8% 6.4% 5.9% 5.1% 1,834- 3,103-

Banco Popular Espanol Spain 7.1% 5.3% 4.7% 3.9% 2,172- 2,926-

Caja Ahorros Del Mediterraneo (CAM) Spain 3.8% 3.0% 2.6% 2.0% 2,124- 2,412-

Banco Santander Spain 7.1% 8.4% 7.4% 6.6% - 2,358-

Banco de Sabadell Spain 6.2% 5.7% 4.6% 3.0% 1,348- 2,217-

Caixa D'Estalvis De Catalunya, Tarragona I Manresa

Spain 6.4% 4.8% 4.3% 3.5% 1,343- 1,734-

Caixa De Aforros De Galicia, Vigo, Ourense E Pontevedra

Spain 5.2% 5.3% 5.0% 4.6% 1,097- 1,326-

Grupo Banca Civica Spain 8.0% 5.6% 5.2% 4.4% 859- 1,183-

Caixa D'Estalvis Unio De Caixes De Manlleu, Sabadell I Terrassa (Unnim)

Spain 6.3% 4.5% 3.1% 0.8% 678- 1,022-

Grupo BMN (Banco Mare Nostrum) Spain 8.3% 6.1% 5.5% 4.6% 613- 962-

Bankinter Spain 6.2% 5.3% 4.9% 4.4% 673- 843-

Banco Pastor Spain 7.6% 3.3% 3.1% 2.7% 720- 785-

Caja España De Inversiones, Salamanca Y Soria, Caja De Ahorros Y Monte De Piedad

Spain 8.2% 7.3% 5.9% 3.8% 269- 776-

Grupo Caja3 Spain 8.6% 4.0% 3.2% 2.0% 528- 682-

Caja De Ahorros Y M.P. De Zaragoza, Aragon Y Rioja (Ibercaja)

Spain 9.7% 6.7% 6.0% 4.8% 252- 510-

Effibank (Grupo Banco Área) Spain 8.3% 6.8% 6.5% 6.1% 160- 276-

Monte De Piedad Y Caja De Ahorros De Ronda, Cadiz, Almeria, Malaga, Antequera Y Jaen (Unicaja)

Spain 12.5% 9.4% 8.1% 6.1% - 169-

Caja De Ahorros Y M.P. De Ontinyent Spain 8.9% 5.6% 5.5% 5.4% 10- 11-

Colonya - Caixa D'Estalvis De Pollensa Spain 11.2% 6.2% 6.2% 6.2% 1- 1-

Banca March Spain 22.2% 23.5% 23.5% 23.4% - -

BBVA Spain 8.0% 9.2% 8.6% 7.6% - -

Caja De Ahorros De Vitoria Y Alava Spain 12.5% 8.7% 8.0% 7.0% - -

Caja De Ahorros Y M.P. De Gipuzkoa Y San Sebastian (Kutxa)

Spain 13.2% 10.1% 9.9% 9.5% - -

Grupo BBK Bank Spain 10.2% 8.8% 8.1% 7.1% - -

Nordea Bank Sweden 8.9% 9.5% 9.5% 9.5% - -

SEB Sweden 11.1% 10.5% 10.4% 10.3% - -

Svenska Handelsbanken Sweden 7.7% 8.6% 8.6% 8.6% - -

Swedbank Sweden 8.7% 9.4% 9.4% 9.4% - -

Royal Bank of Scotland (RBS) United Kingdom 9.7% 6.3% 6.2% 6.0% 5,788- 6,841-

Barclays United Kingdom 10.0% 7.3% 7.1% 6.9% - 698-

HSBC Holdings United Kingdom 10.5% 8.5% 8.4% 8.3% - -

Lloyds Banking Group United Kingdom 10.2% 7.7% 7.7% 7.7% - -

Total 91,724 138,001

Sources: UBS WMR; EBA Stress Test 2011; RBS; MS

Financials


Table 4: Net direct exposure to stressed periphral sovereigns per country

in EUR mnIssuer Greece Ireland Portugal Italy Spain

Erste Bank Austria 346 39 106 604 139 1,234 11.7% bi

RBI (Raiffeisen Bank International) Austria 1 0 2 449 3 455 6.0% bi

Oesterreichische Volksbank AG Austria 113 13 29 149 64 368 20.9%

Dexia Belgium 3,462 0 1,927 15,010 1,444 21,843 128.5% bi

KBC Bank Belgium 443 269 159 5,569 1,419 7,859 67.1% bi

Marfin Popular Bank Cyprus 3,407 39 0 0 0 3,446 171.0%

Bank of Cyprus Cyprus 2,406 322 0 36 58 2,822 132.2%

Danske Bank Denmark 1 368 41 350 111 871 6.0% bi

Jyske Bank Denmark 64 22 19 0 15 120 7.1%

Sydbank Denmark 0 0 0 0 0 0 0.0%

Nykredit Denmark 7 0 0 88 0 95 1.4%

OP-Pohjola Group Finland 3 40 0 0 0 43 0.8%

BNP Paribas France 4,985 407 2,033 24,115 3,901 35,441 64.0% bi

Credit Agricole France 655 136 1,109 10,122 2,772 14,794 32.0% bi

BPCE France 1,262 312 318 3,495 380 5,767 18.1%

Societe Generale France 2,651 442 631 3,042 2,219 8,985 32.3% bi

Deutsche Bank Germany 1,510 476 86 3,436 2,083 7,591 25.0% bi

Commerzbank Germany 3,044 26 976 10,133 3,166 17,345 64.9% bi

LBBW Germany 768 0 95 1,412 512 2,787 28.3%

DZ Bank Germany 731 51 992 2,710 4,126 8,610 118.0%

Bayerische Landesbank Germany 145 20 0 507 662 1,334 11.6% bi

NordLB Germany 150 41 256 1,873 498 2,818 70.9%

Hypo Real Estate Germany 0 44 494 7,139 3,394 11,071 199.9%

WestLB Germany 343 35 0 1,103 746 2,227 52.8%

HSH Nordbank Germany 101 0 62 658 178 999 22.5%

Landesbank Berlin Germany 449 0 0 327 371 1,147 22.2%

Dekabank Germany 86 30 31 262 165 574 17.1%

WGZ Bank Germany 315 221 359 1,402 1,169 3,466 182.0%

EFG Eurobank Greece 8,763 0 0 100 0 8,863 206.3% bi

National Bank of Greece Greece 12,882 18 0 0 0 12,900 158.2% bi

Alpha Bank Greece 5,476 0 0 0 0 5,476 103.8%

Piraeus Bank Greece 8,114 0 0 0 0 8,114 267.0%

Agricultural Bank of Greece (ATEbank) Greece 7,850 0 0 0 0 7,850 991.2%

TT Hellenic Postbank Greece 5,313 0 0 0 0 5,313 434.3%

OTP Bank Hungary 0 0 0 0 0 0 0.0%

Allied Irish Banks (AIB) Ireland 0 0 0 0 0 0 0.0% bi

Bank of Ireland Ireland 0 3,269 0 30 0 3,299 46.9% bi

Irish life & Permanent Ireland 0 1,852 0 0 0 1,852 110.2%

Intesa Sanpaolo Italy 639 114 70 57,622 763 59,208 226.3% bi

Unicredit Italy 636 58 84 47,446 1,909 50,133 140.4% bi

Monte dei Paschi die Siena Italy 8 0 202 32,018 284 32,512 516.0%

Banco Popolare SC Italy 87 0 0 11,758 199 12,044 220.0%

UBI Banca (Unione die Banche Italiane) Italy 25 0 0 9,944 0 9,969 152.0%

Banque et Caisse d'Epargne de L'Etat (BCEE) Luxembourg 84 0 179 2,387 171 2,821 181.9%

Bank of Valletta Malta 10 7 3 4 0 24 6.8%

ING Bank Netherlands 745 90 692 5,700 1,762 8,989 29.1% bi

Rabobank Nederland Netherlands 376 59 37 431 154 1,057 3.8% bi

ABN Amro Bank NV Netherlands 0 128 0 1,310 99 1,537 13.3%

SNS Bank Netherlands 47 157 0 763 57 1,024 57.5%

DnB NOR Bank Norway 0 0 0 0 0 0 0.0%

PKO Bank Polski Poland 0 0 0 0 0 0 0.0%

Caixa Geral De Depósitos, SA Portugal 51 23 6,531 0 197 6,802 104.5%

Banco Commercial Portuguese (BCP) Portugal 727 213 5,828 50 0 6,818 193.7% bi

Espirito Santo Financial Group (ESFG) Portugal 309 0 2,685 0 54 3,048 67.4% bi

Banco BPI Portugal 325 283 3,895 972 0 5,475 256.7%

Nova Ljubljanska Slovenia 20 15 15 96 26 172 21.2%

Nova Kreditna Slovenia 0 0 0 0 0 0 0.0%

Banco Santander Spain 177 0 3,632 260 42,258 46,327 110.3% bi

BBVA Spain 128 0 647 3,898 53,452 58,125 233.1% bi

BFA-Bankia Spain 55 0 0 0 25,387 25,442 183.5%

Caja de Ahorros y Pensiones de Barcelona (La Caixa) Spain 0 0 26 1,269 34,332 35,627 320.7% bi

Total exposure Total exposure

as % of FY10 CT1 Domicile of the issuer

Financials


Table 4: Net direct exposure to stressed periphral sovereigns per country

in EUR mnIssuer Greece Ireland Portugal Italy Spain Total exposure Total exposure

as % of FY10 CT1 Domicile of the issuer

Effibank (Grupo Banco Área) Spain 37 0 16 0 2,925 2,978 112.1%

Banco Popular Espanol Spain 0 0 643 210 8,874 9,727 145.2%

Banco de Sabadell Spain 0 38 91 0 7,295 7,424 211.7%

Caixa D'Estalvis De Catalunya, Tarragona I Manresa Spain 0 0 0 0 2,839 2,839 91.5%

Caixa De Aforros De Galicia, Vigo, Ourense E Pontevedra

Spain 0 0 134 151 3,813 4,098 143.9%

Grupo BMN (Banco Mare Nostrum) Spain 0 0 88 0 3,619 3,707 112.2%

Bankinter Spain 0 0 0 1 2,533 2,534 132.0%

Caja España De Inversiones, Salamanca Y Soria, Caja De Ahorros Y Monte De Piedad

Spain 0 0 27 0 7,557 7,584 365.3%

Grupo Banca Civica Spain 5 0 0 0 4,747 4,752 128.9%

Caja De Ahorros Y M.P. De Zaragoza, Aragon Y Rioja (Ibercaja)

Spain 0 0 0 384 2,909 3,293 143.2%

Monte De Piedad Y Caja De Ahorros De Ronda, Cadiz, Almeria, Malaga, Antequera Y Jaen (Unicaja)

Spain 6 0 0 309 2,950 3,265 130.5%

Banco Pastor Spain 41 0 116 103 2,182 2,442 175.0%

Grupo BBK Bank Spain 0 4 3 0 3,112 3,119 104.6%

Caixa D'Estalvis Unio De Caixes De Manlleu, Sabadell I Terrassa (Unnim)

Spain 0 13 0 11 2,574 2,598 243.9%

Caja De Ahorros Y M.P. De Gipuzkoa Y San Sebastian (Kutxa)

Spain 0 0 0 0 1,512 1,512 78.2%

Grupo Caja3 Spain 0 8 0 0 1,514 1,522 130.8%

Banca March Spain 0 0 0 0 150 150 7.1%

Caja De Ahorros De Vitoria Y Alava Spain 0 0 0 0 598 598 77.5%

Caja De Ahorros Y M.P. De Ontinyent Spain 0 0 0 0 6 6 10.5%

Colonya - Caixa D'Estalvis De Pollensa Spain 0 0 0 0 26 26 127.4%

Caja Ahorros Del Mediterraneo (CAM) Spain 0 15 5 20 5,585 5,625 305.2%

Nordea Bank Sweden 0 1 0 97 64 162 0.8% bi

SEB Sweden 122 0 132 288 86 628 6.5% bi

Svenska Handelsbanken Sweden 0 0 0 0 0 0 0.0%

Swedbank Sweden 0 0 0 0 0 0 0.0%

Royal Bank of Scotland (RBS) United Kingdom 1,154 403 208 4,654 379 6,798 11.5% bi

HSBC Holdings United Kingdom 919 134 320 3,856 636 5,866 6.7% bi

Barclays United Kingdom 93 407 1,174 2,915 5,495 10,085 21.8% bi

Lloyds Banking Group United Kingdom 0 0 0 32 63 95 0.2% bi

Total 82,672 10,662 37,209 283,081 264,742 678,365

Sources: UBS WMR; EBA Stress Test 2011; RBS; MSbi = bi-annual risk-assessment by EBA

Financials


Appendix

Terms and AbbreviationsTerm / Abbreviation Description / Definition Term / Abbreviation Description / DefinitionA actual i.e. 2010A NV Neutral View: The stock is expected to neither

outperform nor underperform the relevantbenchmark nor significantly appreciate ordepreciate in absolute terms.

p.a. Per annum (per year) Shares o/s Shares outstandingUP Underperform: The stock is expected to

underperform the sector benchmarkWMR UBS Wealth Management Research

Financials


Appendix

Global Disclaimer

Wealth Management Research is published by Wealth Management & Swiss Bank and Wealth Management Americas, Business Divisions ofUBS AG (UBS) or an affiliate thereof. In certain countries UBS AG is referred to as UBS SA. This publication is for your information only and is notintended as an offer, or a solicitation of an offer, to buy or sell any investment or other specific product. The analysis contained herein is basedon numerous assumptions. Different assumptions could result in materially different results. Certain services and products are subject to legalrestrictions and cannot be offered worldwide on an unrestricted basis and/or may not be eligible for sale to all investors. All information andopinions expressed in this document were obtained from sources believed to be reliable and in good faith, but no representation or warranty,express or implied, is made as to its accuracy or completeness (other than disclosures relating to UBS and its affiliates). All information andopinions as well as any prices indicated are currently only as of the date of this report, and are subject to change without notice. Opinionsexpressed herein may differ or be contrary to those expressed by other business areas or divisions of UBS as a result of using different assumptionsand/or criteria. At any time UBS AG and other companies in the UBS group (or employees thereof) may have a long or short position, or deal asprincipal or agent, in relevant securities or provide advisory or other services to the issuer of relevant securities or to a company connected withan issuer. Some investments may not be readily realizable since the market in the securities is illiquid and therefore valuing the investment andidentifying the risk to which you are exposed may be difficult to quantify. UBS relies on information barriers to control the flow of informationcontained in one or more areas within UBS, into other areas, units, divisions or affiliates of UBS. Futures and options trading is consideredrisky. Past performance of an investment is no guarantee for its future performance. Some investments may be subject to sudden and largefalls in value and on realization you may receive back less than you invested or may be required to pay more. Changes in FX rates may havean adverse effect on the price, value or income of an investment. We are of necessity unable to take into account the particular investmentobjectives, financial situation and needs of our individual clients and we would recommend that you take financial and/or tax advice as to theimplications (including tax) of investing in any of the products mentioned herein. This document may not be reproduced or copies circulatedwithout prior authority of UBS or a subsidiary of UBS. UBS expressly prohibits the distribution and transfer of this document to third parties forany reason. UBS will not be liable for any claims or lawsuits from any third parties arising from the use or distribution of this document. Thisreport is for distribution only under such circumstances as may be permitted by applicable law.Distributed to US persons by UBS Financial Services Inc., a subsidiary of UBS AG. UBS Securities LLC is a subsidiary of UBS AG and an affiliateof UBS Financial Services Inc. UBS Financial Services Inc. accepts responsibility for the content of a report prepared by a non-US affiliate whenit distributes reports to US persons. All transactions by a US person in the securities mentioned in this report should be effected through aUS-registered broker dealer affiliated with UBS, and not through a non-US affiliate. The contents of this report have not been and will not beapproved by any securities or investment authority in the United States or elsewhere.Version as per June 2011.© 2011. The key symbol and UBS are among the registered and unregistered trademarks of UBS. All rights reserved

Financials


Financial Pacific - In shape, or not Europe's banks after the workout (third party)

Business

banks portugal

sovereign debt exposure

european sovereign debt

german banks

banks account

europes banks

periphery250 banks

domestic banks stress