European Crisis – beginning of the end

European Crisis – beginning of the end

Capital deficit: From €106 bn to €76 bn to €53bn to €19 bn
On Wednesday evening the European Banking Authority (EBA) announced a
€106 bn capital deficit for the European bank sector (for deficit by bank see Table
2). This figure includes €30 bn for Greek banks, which is the existing backstop
facility and exceeds the deficit calculated under the EBA scenario. Stripping out
Greek banks, the announced capital deficit falls from €106 bn to €76 bn. We
estimate €53 bn of the €76 bn deficit resides within our coverage universe.
However, when taking account of mandatory convertibles and retained earnings,
the capital deficit falls to what we view as a very manageable €19 billion for our
coverage universe excluding Greece.

EBA targets dividends & bonuses to reduce capital deficits
By the end of 2011 all banks with a capital deficit under the EBA sovereign stress
scenario must submit a plan to their national authorities detailing how they will
eliminate the capital deficit. The EBA explicitly highlights dividends and bonuses
as areas banks should target to reduce deficits. For each bank where the EBA
has identified a capital deficit, we have categorized the dividend risk as high,
medium, and low (Table 5). There are three banks where we forecast a cash
dividend, but highlight the risk as high: SocGen, MPS, and Banco Popolare.

If markets improve, expect €30 to €35 bn capital raised
Moving from the EBA stress test to a Basel III analysis, we have divided our
coverage universe into four categories based on the likelihood of capital raising
(see Table 6). We see eight banks that almost certainly have to raise capital and
another eight banks that we think could raise capital if market conditions improve.
In aggregate, we see the potential for €30-35 billion of capital raising –
representing 5-6% of the market capitalization of the sector.

Investors questioning value of European sovereign CDS
With the second Greek private sector initiative (PSI), investors face a “voluntary”
50% haircut on Greek debt. This programme, with a targeted 95% take-up, would
lower Greek debt by €100 bn and, if a number of other things go positively, could
allow Greece to achieve a debt to GDP ratio of 120% by 2020. Given the
similarities with the first Greek PSI, we can’t imagine the second Greek PSI
programme will trigger CDS. ISDA itself reports “it does not appear to be likely
that the [Greek] restructuring will trigger payments under existing CDS contracts.”

Banks haven’t been big buyers of sovereign CDS protection
We highlight that the total amount of CDS protection written on Greece, Ireland,
and Portugal is less than €10 bn (US$ 13.5 bn). Furthermore, looking at the
difference between gross and net sovereign exposure in the July EBA stress test,
we think the potential amount of CDS exposure held by banks is very limited (see
T able 9).

A €106 bn deficit shrinks to €19 bn
Last week the EBA announced a €106 bn capital deficit for the European
bank sector. Rather than inject government capital into the banking sector,
banks have been given nine months to recapitalize. As a result, banks will
be able to include future retained earnings and convertible notes when
calculating their deficit. Excluding Greece, we think the capital deficit in
our coverage universe as of June 2012 would be a very manageable €19 bn.
In order to meet this capital deficit we expect banks could cut cash
dividends and we highlight SocGen, MPS, and Banco Popolare as the
dividends most at risk. If markets improve we could see banks raise
capital, not only because of the EBA’s sovereign exercise, but also because
of weak Basel III capital ratios. In aggregate, we see €30-35 bn of capital
raising.

EBA announces a €106 bn capital deficit
Table 2 below attempts to show the capital deficit for each of the European banks
included in the EBA’s recapitalization effort. Excluding Greece, the EBA shows a
capital deficit of €76.4 bn. By comparison, on a bank by bank basis we could only
get to €75.6 bn. We think the difference comes from Portugal.

Breaking Down EUR 106 Billion Capital Deficit

€106 bn includes €30 bn for Greece
The EBA’s disclosure on Greece lists a €30 billion capital deficit which is the
existing backstop facility (€10 bn provided in the first €110bn package plus €20
bn that is part of the more recent €109bn package). While the EBA included the
€30 billion figure in estimating the capital deficit, it also discloses that the €30
billion figure exceeds the estimated capital deficit for Greek banks under the EBA
sovereign stress methodology. We only cover four of the six Greek banks
stressed by the EBA. By our calculation, using the EBA stress methodology,
these four banks would have roughly an €11 billion capital deficit. The Greek
deficits look different under the IMF scenario, but we stick with the EBA scenario
in this analysis. In our view, including the actual deficit calculated for Greek
banks would take the €106 billion figure below €100 billion. Similarly, we find it
odd that the EBA chose to include roughly €5 billion of deficit for Dexia and
Volksbank – both of which have undergone significant restructuring. We
conclude that the EBA probably felt pressure to publish a deficit of over €100
billion.

Coverage universe accounts for €53 bn deficit
Ignoring Greece, the EBA shows a €76 billion capital deficit. Of this €76 billion,
€53 billion comes from our coverage universe. Focusing on our coverage
universe excludes the following significant capital deficits: €5 billion from
Portuguese banks, €4 bn at Dexia, €4 bn from Cypriot banks, €3bn at Austrian
banks (covered by our CEE counterparts), €3 bn at BPCE (parent company to
Natixis), and €1 bn from non-listed German Landesbanks. For those banks
outside our coverage universe we think equity raising might be very difficult –
either because the deficit exists at a non-listed bank or we don’t think there is
much of an equity story (which explains why we don’t cover the stock).

€53 bn deficit falls to €19 bn deficit by June
The EBA exercise was based on June 2011 balance sheets. Banks have until
June 2012 to come into compliance. Therefore, the capital deficit announced by
the EBA fails to account for 12 months of earnings, mandatory convertibles in
Spain, disposals, and RWA reduction. Table 3 below shows the capital deficit on
June 2012 after we account for convertibles and earnings.

Estimating June 2012 Capital Deficit For BofAML Coverage Universe

Deficit manageable for some, not all, banks
A €19 billion capital deficit represents roughly 3% of the market capitalization of
the European bank sector. While this is a very manageable number in aggregate,
capital isn’t raised in aggregate, it is raised by individual banks. Table 4 on the
following page shows the estimated June 2012 capital deficit as a percent of the
current market capitalization.

June 2012 Deficit As A Percent of Market Capitalization

By the end of 2011 all banks with a capital deficit under the EBA sovereign stress
scenario must submit a plan to their national authorities detailing how they will
eliminate the capital deficit. According to the EBA release, “targets will have to be
achieved avoiding excessive deleveraging, so as to contain the potential impact
on the real economy. To reach the targets, banks will be expected to withhold
dividends and bonuses.”

We highlight the following items that could help banks reduce their capital deficit:

* Unicredit – The €5.8 billion can be reduced by another €3 bn if cash were
included. However, as we have written in previous research, we see a
roughly €7 billion capital deficit at UC.
* Spanish Banks – Have all stated that they won’t raise capital, but rather will
reduce the deficit by disposing of non-core businesses and de-leveraging.
For more information, see our note from Thursday.

* MPS – In calculating MPS’ capital deficit, we believe the EBA has included
Tremonti bonds (€1.9 bn), the recent capital raising (€2.15 bn) and the
Chianti deal (~€400 mn). We don’t think the EBA has accounted for €1.3 bn
of FRESH capital (FRESH 2 €318m; and FRESH 3 €950m) – so that could
help reduce the deficit. While MPS has claimed that the capital shortfall is
manageable – we think it is significant. At a minimum, we can’t see MPS
repaying the Tremonti bonds, which was the rationale for its recent capital
increase.

* UBI – We think UBI could benefit further from the full impact of moving to
Basel II advanced. UBI management claim the bank will not raise capital, but
we can’t exclude UBI executing its soft convertible (€639m). At current share
prices, the conversion would be quite dilutive for existing shareholders on our
estimates.

It is also worth noting that the Greek banks need to be at a 10% core tier 1 ratio
as per the IMF agreement (10%). Clearly the Greek banks will need capital, and
in our view it is very unlikely that this capital will come from the market.

Dividends could provide further capital buffer
Table 5 on the following page shows the amount of dividends we expect banks in
the stress test to pay between now and June 2012 as well as the risk to the cash
component. We also differentiate between cash dividends and scrip dividends.

Risk to Dividends – Stress Test Banks

Capital deficits need to account for Basel III
In Table 6 on the following page we divide our coverage universe into four
categories based on the probability of each bank raising capital. Table 6 takes
into account the EBA stress test, but is really driven by our analysis of Basel III
capital positions. We assume that by year end 2013 banks will want to be
between 9% and 10% on a fully phased in Basel III basis. We recognize that
banks have until YE 2018 before fully phased in Basel III rules apply, and even
then the required capital will be lower than the 9-10% we target (at least for banks
that aren’t globally systemically important financial institutions). Nevertheless, if
market conditions continue to improve, we think some of those banks that have
passed the EBA stress would be tempted to raise capital.

In addition to categorizing banks by likelihood of capital raising, Table 6 also
attempts to quantify the amount of capital the bank might raise. While we
have used our year-end 2013 Basel III forecasts as a guide, Table 6 below is
as much art as science, as we have taken into account each bank’s earning
generation, volatility of earnings, and management’s willingness to raise
capital.

If, for the sake of estimating how much capital could be raised, we put a 100%
probability on the first category, a 50% probability on the second category, and
10% probability on the third category, we end up with an estimated €30 bn to €35
bn of capital raising from our coverage universe.

BofAML Estimated 2013 Fully Phased In Basel III Core Tier 1 Ratios

Banks not big users of Sovereign CDS
Following a second Greek PSI, this time with a 50% haircut, investors have
started to question the value of CDS protection on European sovereign
debt. We highlight that the total amount of CDS protection on Greece,
Ireland, and Portugal is less than €10 bn (US$ 13.5 bn). Furthermore,
excluding domestic banks, the total difference between gross and net
exposure on Greek, Irish, and Portuguese sovereigns is only €4.5 bn.

Greek solution calls for voluntary 50% haircut
As part of the EU Summit European policy makers have called for private sector
holders of Greek debt “to develop a voluntary bond exchange with a nominal
discount of 50% on notional Greek debt.” According to policy makers, this
voluntary exchange, should it achieve the targeted 95% take-up, would lower
Greek debt by €100 bn. It is hoped that this debt reduction, together with an
ambition reform programme and a global economic recovery, will allow Greece to
achieve a debt to GDP ratio of 120% by 2020.

Greek PSI won’t trigger CDS in our view
While we don’t think the 50% haircut prescribed by the Greek Private Sector
Initiative (PSI) will be voluntary for the banks in our coverage universe, we don’t
expect it to trigger CDS. The determination of a credit event is made the
International Swaps and Derivatives Association (ISDA). According to ISDA “it
appears from preliminary news reports that the bond restructuring is voluntary
and not binding on all bondholders. As such, it does not appear to be likely that
the restructuring will trigger payments under existing CDS contracts.”
Given that ISDA has already examined the structure when the first PSI was
announced, we would be very surprised if the outcome of the second PSI was
any different. The following comes from ISDA’s Web site:

“A CDS is triggered when a Credit Event occurs. There are three Credit
Events that are typically used for Western European Sovereigns, they
are: 1) Failure to Pay; 2) Repudiation/Moratorium and 3)Restructuring.
We will focus on Restructuring for these purposes. The Restructuring
Credit Event is triggered if one of a defined list of events occurs, with
respect to a debt obligation such as a bond or a loan, as a result of a
decline in creditworthiness or financial condition of the reference entity.
The listed events are: reduction in the rate of interest or amount of
principal payable (which would include a “haircut”); deferral of payment
of interest or principal (which would include an extension of maturity
Regarding the trigger, an outstanding obligation); subordination of the
obligation; and change in the currency of payment to a currency that is
not legal tender in a G7 country or a AAA-rated OECD country. The
decline in creditworthiness or financial condition requirement is intended
to filter out restructurings that occur as a result of improved financial
condition”

Given that a triggering event has to occur in a form that binds all holders of the
restructured debt, we simply don’t see how it could apply in this case.

Greek PSI allows ECB not to take a haircut
European policy makers appear to be going to great lengths to avoid triggering a
Greek CDS event. In our view this would be to avoid having to recapitalize the
European System of Central Banks (ESCB). While we don’t know how much
Greek debt the ECB has purchased, we believe it would be in the order of
magnitude of €50 billion. While the IMF has preferred creditor status, the ECB
doesn’t. Therefore any restructuring would result in losses for the European
System of Central Banks and would most likely need to result in a
recapitalization.

Yet Greek CDS continues to imply a default
As can be seen in Chart 1 below, the 5-year Greek CDS continues to imply a high
probability of default.

Greek 5-YR Senior CDS

Our credit / rates / FX strategists believe that PSI 2 will not be the end of the
Greek crisis. As highlighted above, the PSI only targets a debt to GDP ratio of
120%, it assumes that Greece is able to achieve all of its targeted reforms, it
assumes an economic recovery, and it assumes that Greece will be under an IMF
programme for another decade. While all of that is certainly possible, it is also
still possible that Greece could default.

Investors concerned banks own CDS protection
With the value of all European sovereign CDS being called into question on the
back of the Greek PSI, investors are increasingly worried about which European
banks may own CDS protection on peripheral sovereigns.

Table 8 below shows the net notional amount of CDS outstanding. The net
notional outstanding CDS represents the total amount of outstanding notional to
settle between the protection buyer and seller in a credit event. As can be seen
in Table 8, the total amount of CDS outstanding is a relatively small number (less
than €3 billion in the case of Greece).

Net Notional Amount of CDS Outstanding

Very little sovereign CDS protection at banks
We have taken the data disclosed by the EBA with the July stress test and
compared banks’ gross and net amounts of sovereign exposure to Greece,
Ireland, Portugal, Spain and Italy. Of the 90 banks in the exercise, just over half
had no difference between gross and net positions. Table 9 below shows the
difference between gross and net Sovereign exposure for those banks that
reported a difference.

It would be incorrect to assume that the entire difference between gross and net
positions is CDS. The gross position is total lending, while the net position is net
government bonds. We suspect the biggest differences between gross and net
lending is direct lending. If we assume fears are limited to Greece, Ireland, and
Portugal, than the figures below are very manageable (excluding domestic
banks).

Difference Between Gross & Net Positions of Sovereign Debt Holding

Stocks mentioned