2015_03_02_CABK Investor Day Cheat Sheet

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

IMPORTANT DISCLOSURE FOR U.S. INVESTORS: This document is prepared by Mediobanca Securities, the equity research department of Mediobanca S.p.A. (parent company of Mediobanca Securities USA LLC (“MBUSA”)) and it is distributed in the United States by MBUSA which accepts responsibility for its content. The research analyst(s) named on this report are not registered / qualified as research analysts with Finra. Any US person receiving this document and wishing to effect transactions in any securities discussed herein should do so with MBUSA, not Mediobanca S.p.A.. Please refer to the last pages of this document for important disclaimers.

Caixa Bank

02 March 2015 Banks Update

Investor Day Cheat Sheet Andrea Filtri

Equity Analyst

New investment case: from „cash cow to be‟ to ‘Iberian restructuring story‟

CABK has been building up as a story of: domestic banking restructuring, capital

free up from stakes and sound balance sheet from premium capital and coverage

ratios. But the bid on BPI and the upcoming sale of Novo Banco (NB) could

transform CABK’s story. Portugal could be a large swing factor for CABK and the

BPI bid takes full control of it. The potential integration of BPI/NB into CABK

could create ‘Bank of Iberia’. This compelling restructuring story in EU periphery

- just as the economy recovers – would lose CABK the cashback and premium

capital angles. We expect tomorrow’s Investor Day to provide further colour.

Equity stakes: 196bp CET1 cushion vs c.40% of 2015E ROTE

We estimate the capital consumption of each stake at CABK and identify on

aggregate 188bp of CET1 ratio (87bp from REP and TEF, 101bp from FIG stakes).

The full sale of the portfolio would boost CET1 by 196bp, 80% from FIG stakes,

20% from industrial ones. 70% of the CET1 benefit would come from RWA

reduction and 30% from the reversal of goodwill deductions, while capital

deductions offset net losses from MTM. So, stakes represent a thick capital

buffer for CABK to fund growth and/or face regulatory hurdles. 2.7 p.p. ROTE

erosion is the negative, with TEF as the largest contributor (70bp) and Bank of

East Asia (BEA) minimizing RoTE hits while maximizing CET1 boost.

350bp CET1 gearing (M&A and regulation) vs 280bp CET1 cushions (stakes)

CABK reported a high 12.3% FL CET1 ratio, yet excess capital is hard to pinpoint

as we estimate 350bp potential erosion from regulation (140bp) and acquisitions

(210bp). On the former: the lifting of the Danish Compromise (90bp), the

introduction of 10% risk weight on govies (20bp) and their implications for DTAs

(30bp). On the latter, we see the sale of equity stakes (200bp), 25% minorities in

BPI (30bp) and the further re-levering of VidaCaixa (50bp).

CABK/BPI/NB, the ‘Bank of Iberia‟: 6% EPS accretion post €5.5bn capital hike

The BPI bid at 0.9x P/TE looks generous based on the low profitability (post

African exit) and synergy visibility. But we believe this the stepping stone to get

to creating Portugal’s and Iberia’s #1 banks through the CABK/BPI/NB merger.

The Portuguese merger would provide for €291m visible after tax cost synergies,

leading to 6% boost to CABK’s 3-yr EPS. Excluding stake disposals, CABK would

need €5.5bn rights issue to defend 11% CET1 ratio (in line with co. targets).

CABK is a recovery play through balance sheet growth; not a low cost model

Spanish banks overhauled their business model during the crisis with over 1/3

branch closures and 25% loan deleverage. Instead, CABK maintained a large

network (2x the peer avg.) of smaller branches requiring higher fixed costs in

exchange of stronger and cheaper deposit gathering. We estimate realigning

branch productivity to peer avg. would boost EPS by 12% vs 18% by cutting costs

in line with competitors. This means CABK is positioned for superior balance

sheet growth in a recovery while peers embody the operating leverage play.

30% valuation swing to our TP – Staying Neutral while awaiting developments

We estimate 20% upside to our €4.4 TP jointly from the BPI/NB takeover funded

by a cash call (+9%) and from the CABK realignment to peer profitability (+11%

both for cost cutting and branch productivity). Instead, we see 9% downside

from the BPI/NB takeover funded by stake disposals and CET1 gearing. This

indicates that valuation is positively skewed to management action.

+44 203 0369 571

[email protected]

Andres Williams

Equity Analyst

+44 203 0369 577

[email protected]

Source: Mediobanca Securities

Price: € 4.11 Target price: € 4.40 Neutral

2014 2015E 2016E 2017E

EPS Adj (€) 0.11 0.26 0.38 0.48

DPS (€) 0.09 0.15 0.23 0.29

TBVPS (€) 3.56 3.59 3.73 3.99

Avg. RoTE Adj (%) 3.0% 7.2% 10.3% 12.4%

P/E Adj (x) 37.9 16.0 10.9 8.6

Div.Yield(%) 2.3% 3.7% 5.5% 7.0%

P/TBV (x) 1.2 1.1 1.1 1.0

Market Data

Market Cap (€m) 23,489

Shares Out (m) 5,715

Main Shareholder Name (%) 10%

Free Float (%) 60%

52 week range (€) 4.92-3.83

Rel Perf vs STOXX EUROPE 600 BANKS E (%)

-1m -0.7%

-3m -12.9%

-12m -11.6%

21dd Avg. Vol. 22,247,312

Reuters/Bloomberg CABK.MC / CABK SM

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 2


Valuation Matrix



Profit & Loss Acc(€ m) 2014 2015E 2016E 2017E Multiples 2014 2015E 2016E 2017E

Net Interest Income 4,155 4,830 5,225 5,672 P/E 37.9 16.0 10.9 8.6

Growth (%) 5.1% 16.3% 8.2% 8.6% P/E Adj. 37.9 16.0 10.9 8.6

Non-Interest Income 2,785 2,793 3,174 3,431 P/Net Op.Income 7.4 6.3 5.4 4.8

Growth (%) 15.6% 0.3% 13.6% 8.1% P/Revenues 3.4 3.1 2.9 2.6

of which Fee Income 1,825 2,026 2,087 2,107 P/TBV 1.2 1.1 1.1 1.0

of which Financial Income 640 275 275 275 P/Total Deposits (%) 13.0% 12.0% 11.8% 11.5%

Total Income 6,940 7,623 8,398 9,103 Yield (%) 2.3% 3.7% 5.5% 7.0%

Growth (%) 9.0% 9.8% 10.2% 8.4%

Total Costs -3,773 -3,840 -3,991 -4,056

Growth (%) -21.2% 1.8% 3.9% 1.6%

of which Personnel Costs -2,578 -2,625 -2,756 -2,799

Net Operating Income 3,167 3,783 4,408 5,047

Growth (%) 100.6% 19.5% 16.5% 14.5%

Provisions&Write-downs -2,579 -1,703 -1,520 -1,360 Per Share Data (€) 2014 2015E 2016E 2017E

Extraordinary Items na na na na EPS 0.11 0.26 0.38 0.48

Pre-tax profit 202 1,968 2,888 3,687 EPS growth (%) 72.6% 136.1% 46.7% 27.7%

Tax 418 -472 -693 -885 EPS Adj. 0.11 0.26 0.38 0.48

Tax rate(%) -206.9% 24.0% 24.0% 24.0% EPS Adj. growth (%) -55.6% 136.1% 46.7% 27.7%

Minorities and others 0 0 0 0 TBVPS 3.56 3.59 3.73 3.99

Net profit 620 1,496 2,195 2,802 DPS Ord 0.09 0.15 0.23 0.29

Growth (%) 96.2% 141.3% 46.7% 27.7%

Adjusted net profit 620 1,496 2,195 2,802

Growth (%) -53.9% 141.3% 46.7% 27.7%

Balance Sheet (€ m) 2014 2015E 2016E 2017E Key Figures & Ratios 2014 2015E 2016E 2017E

Customer Loans 188,762 205,751 209,866 219,310 Avg. N° of Shares (m) 5,712 5,836 5,836 5,836

Growth(%) -4.7% 9.0% 2.0% 4.5% EoP N° of Shares (m) na na na na

Customer Deposits 180,200 200,022 204,022 208,103 Avg. Market Cap. (m) 25,542 23,987 23,987 23,987

Growth(%) 2.9% 11.0% 2.0% 2.0%

Shareholders' Funds 23,993 24,595 25,473 27,014 NII/Total Income (%) 59.9% 63.4% 62.2% 62.3%

Minorities 9 9 9 9 Fees/Total Income (%) 26.3% 26.6% 24.8% 23.1%

Total Assets 339,252 361,443 368,133 375,612 Trading/Total Income (%) 9.2% 3.6% 3.3% 3.0%

Cost Income ratio 54.4% 50.4% 47.5% 44.6%

Personnel costs/Total costs 68.3% 68.4% 69.1% 69.0%

Impairment/Average Loans 1.3% 0.9% 0.7% 0.6%

NPLs ratio 10.7% 9.0% 8.1% 6.9%

Provisions/Loans 5.9% 5.0% 4.4% 3.8%

Avg. RoTE Adj. (%) 3.0% 7.2% 10.3% 12.4%

ROA (%) 0.18% 0.42% 0.61% 0.75%

Tier 1 ratio 13.1% 13.3% 13.7% 14.1%

Basel III Core Tier 1 ratio 12.1% 12.4% 12.7% 13.2%

3.80

4.00

4.20

4.40

4.60

4.80

5.00

M A M J J A S O N D J F

Caixa Bank STOXX EUROPE 600 BANKS E

2/03/15

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 3


Contents

Valuation Matrix 2

Capital is abundant; not unlimited 4

Stakes: much accomplished; much further to go… 5

Stake sales + mitigation vs regulation threat and M&A 12

Building the second home market 17

CABK is not playing the low cost game 29

30% swing around our €4.4 TP – Staying Neutral

34

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 4


Capital is abundant; not unlimited CABK reported 11.6% CET1 fully loaded ratio in Q414, a high level but down 110bp

from Q314 from the acquisition of Barclays Spain and the change in treatment of

Deposit Guarantee charges. The co. anticipated CET1 could fall to 10.4% if the offer

on BPI completes successfully but it targets 11% post BPI acquisition, implying capital

management actions. This means CABK should still retain a good level of capitalisation

but the premium ratio vs peers will go and there will be little capital excess. CABK has

ample room to act on capital ratios, particularly through the management of equity

stakes. Yet, M&A will likely remain a topic on the table, so we expect capital ratios

and allocation to take the pivotal role at tomorrow’s Investor Day.

CET1 ratio is high but it has been drifting down...

Chart 1 shows the evolution of the CABK CET1 fully loaded ratio over the last 18 months. The 11.6%

level disclosed on Q414 results is very sound in a European context, but this has slipped by 110bp in

the last two quarters on the back of the change in treatment of the Deposit Guarantee Fund (DGF)

and of the Barclays Spain acquisition and it is likely to further slip by c.110bp if the bid on BPI will

be successful. This would take CET1 ratio to 10.4%, no longer at premium vs peers. The company is

targeting 11% post BPI transaction, implying capital actions are in sight, in our view.

...suggesting that capital will be one of the main pillars of the next Business Plan

We believe that capital will be one of the main pillars of the upcoming Business Plan. The previous

three years have worked towards the repositioning of CABK on domestic retail banking and away

from the origin of holding of stakes. On the capital front, this has been implemented around three

directories:

1) Industrial stake disposals;

2) Acquisitions/integrations of domestic banking competitors;

3) Capital optimisation for Basel 3 implementation.

We believe the next Plan will complete the repositioning process and capital will have to be the

main driver to fund this change.

Chart 1: CABK CET1 ratio fully loaded evolution

Source: Mediobanca Securities, company data

8.3%

11.7%12.1% 12.4% 12.7%

12.3%11.6%

10.4%

0%

2%

4%

6%

8%

10%

12%

14%

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 5


Stakes: much accomplished; much further to go… Since its listing, CABK reduced the weight of equity stakes. Yet, these still generated

37% of 2014 profits while absorbing c.190bp of CET1 ratio (16% of group capital), for

an estimated 18% ROAC. Not all stakes are the same: industrial co. and FIG >10%

absorb c.90bp CET1 ratio each, while FIG <10% require less than 20bp. Yet, of the

c.200bp CET1 boost from an outright sale of the stakes portfolio, the impact from

different stakes is very asymmetric: Repsol is neutral, Telefonica generates 40bp, FIG

>10% 82bp (of which 50bp from the removal of capital deductions) and FIG <10% 77bp

of which 63bp from GF Inbursa. The counter argument for a sale is that the stakes

generate 2.7 p.p. ROTE for CABK out of 7% 2015E at group level. Telefonica is the

largest single contributor (0.7 p.p.), FIG stakes >10% produce only 0.4 p.p. and FIG

<10% are large generate 1.2 p.p.. Our analysis allows us to cherry pick the

management actions which maximise capital benefits from sales while minimising the

negative hit to group ROTE: the 19% stake in Bank of East Asia is top of the list.

Several disposals have already been made...

Since its IPO, CABK started the sale down of the industrial portfolio. We recall:

the 5% stake sale in BME;

the 3.7%, 6.4% and 0.89% stake sales in GF Inbursa;

the issuance of a mandatory convertible bond for a 2.5% stake in Repsol.

Not all operations on the stakes have been disposals. We also remind of the acquisition of 3.5m new

shares in Erste Bank (EBS), taking the overall stake to 9.9% from 9.1% previously. This move signals

CABK has not resolutely embarked on the full stake disposal path, in our view and we expect the

upcoming Business Plan to shed full light on this item. In this chapter we go in detail over the

current capital absorption of stakes and what further

...but stakes still absorb 188bp CET1 FL...

Table 1 shows the summary of the remaining stakes portfolio at CABK. This amounts to €11.4bn

carrying value split in €6.6bn in industrial companies (Repsol and Telefonica) and €4.8bn in financial

companies of which €3bn in banks with over 10% shareholding (Bank of East Asia, BPI, Boursorama)

and €1.7bn in banks with shareholding below 10% (EBS, GF Inbursa).

We calculate the portfolio carries €0.9bn goodwill, €0.8bn latent losses largely concentrated in

Repsol, BEA and BPI and partly offset by the gains in GF Inbursa.

We calculate the stakes portfolio jointly consumes 188bp of CET1 ratio, for 16% of group capital.

Within this, we see three types of stakes whose capital consumption differs from the others:

1) industrial stakes;

2) financial stakes with CABK shareholding >10%;

3) financial stakes with CABK shareholding <10%.

Industrial stakes absorbing 87bp - The first group constitutes of Repsol (REP) and Telefonica (TEF).

The former stake is strategic, the latter is classified as available for sale (AFS). This means that REP

is equity consolidated while TEF enters CABK’s P&L through dividends. These stakes are both risk

weighted at 200% and 150%, respectively, for €10bn total contribution to group capital. While TEF is

AFS and therefore subject to mark to market valuation every quarter, we estimate REP is carrying a

latent loss for €0.7bn.

FIG stakes >10% absorbing 85bp - The second group constitutes of Bank of East Asia (BEA), BPI and

and Boursorama. These stakes are equity consolidated and for capital reasons their BVs (€2.4bn, i.e.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 6


€3bn carrying value minus €0.6bn goodwill) screen against a buffer at 10% of capital (€1.7bn). The

portion below the bar is 250% risk-weighted for €4.2bn RWA; the one above is a straight deduction.

FIG stakes <10% absorbing 16bp - The third group constitutes of GF Inbursa and EBS. These stakes

are equity consolidated and for capital reasons their BVs (€1.7bn minus €0.3bn goodwill) are risk

weighted at 140% for €2bn RWAs.

...so that more core business growth or regulatory pressure could require more sales

€2bn capital locked into the stakes portfolio signals that this is still a very large part of the CABK

group. Hence, we look at stakes as a large reserve for the future and expect management to

indicate how they intend to administer it in the face of any further, material growth in the core

businesses or in case of potential further regulatory pressures which could come from the likely

harmonisation of capital rules.

Complete sale releasing 196bp of CET1 ratio...

Table 2 shows the CET1 impact from the sale of the equity stakes. We calculate the impact from

four factors:

1) the recovery of the goodwill deduction from capital;

2) the mark to market gain/loss of the stake compared to the carrying book value;

3) the RWA reduction;

4) the removal of the capital deduction from the BV of FIG stakes exceeding 10% of capital.

We estimate that the sale tout court of all stakes would generate 196bp higher CET1 ratio, taking

CABK virtually to 13.6%, one of Europe’s highest ratios.

Table 1: summary of CABK stakes and impact to FL CET1 ratio, 2014

Stake Carrying

value Goodwill Mkt Cap Gain/loss

Risk

weight RWA CET1 Goodwill

CET1

today bp

Repsol 11.90% 3,367

22,724 (663) 200% 5,320 619

0.44%

Telefonica 5.25% 3,213

61,196 - 150% 4,819 561

0.40%

GF Inbursa 9.01% 868 299 15,439 523 140% 797 93 299

BEA 18.68% 1,925 568 8,616 (316)

568

EBS 9.92% 870

9,411 64 140% 1,218 142 -

BPI 44.10% 939

1,356 (341)

-

Boursorama 20.46% 178 66 696 (36)

66

Industrial companies

6,580 - 83,920 (663)

10,139 1,180 - 0.87%

FIG stakes >10%

3,042 634 10,668 (692)

4,218 491 634 0.85%

...deductions

721

0.50%

FIG stakes <10%

1,738 299 24,850 587

2,015 234 299 0.16%

Total FIG. Stakes

4,780 933 35,517 (106)

6,232 955 933 1.01%

Tot. Stakes

11,360 933 119,437 (769)

16,371 2,135 933 1.88%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 7


Individually, we identify Repsol as the exposure with the largest negative impact in case of sale

with 3bp erosion. On the other hand, GF Inbursa is the stake with the largest potential positive

impact from a sale with 63bp boost.

...80% from FIG stakes, 20% from industrial stakes...

Chart 2 shows the asymmetry between the distributions of capital absorption and capital release in

case of disposal between industrial stakes, the FIG stakes >10% and FIG stakes <10%. We estimate

that industrial stakes to absorb c.1/2 of capital allocated to stakes through €10bn RWAs. Yet, their

sale would only account for c.1/5 of the capital release from a full stake sale.

Conversely, FIG stakes <10% only absorb c.10% of capital through €2bn RWA but would release c.40%

of the capital release from a full stake sale.

Finally, FIG stakes >10% would generate c.42% of capital release from a full stake sale, currently

consuming 45% of capital.

Table 2: FL CET1 ratio impact from stake sales, 2014

Company Stake

Carrying

value

Good

will

Tot.

mkt

Cap

Gain/

loss

Risk

weight RWA CET1

CET1

today

bp

CET1

from

sale bp

Repsol 11.90% 3,367

22,724 (663) 200% 5,320 619 0.44% -0.03%

Telefonica 5.25% 3,213

61,196 - 150% 4,819 561 0.40% 0.40%

GF Inbursa 9.01% 868 299 15,439 523 140% 797 93

0.63%

BEA 18.68% 1,925 568 8,616 (316)

EBS 9.92% 870

9,411 64 140% 1,218 142

0.14%

BPI 44.10% 939

1,356 (341)

Boursorama 20.46% 178 66 696 (36)

Industrial companies 6,580 - 83,920 (663)

10,139 1,180 0.87% 0.37%

FIG stakes >10%

3,042 634 10,668 (692)

4,218 491 0.85% 0.82%

...deductions

721 0.50% 0.50%

FIG stakes <10%

1,738 299 24,850 587

2,015 234 0.16% 0.77%

Total FIG. Stakes

4,780 933 35,517 (106)

6,232 955 1.01% 1.59%

Tot. Stakes

11,360 933 119,437 (769)

16,371 2,135 1.88% 1.96%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 8


…and mostly driven by RWA free up with deductions offsetting MTM losses

Chart 3 shows the share of the 196bp contribution to CET1 boost from the sale of the stakes

portfolio. This shows that the c.50bp negative impact to capital from the MTM losses on the stakes

is offset by the removal of the deductions from the FIG stakes >10%. Hence RWA contraction and

the removal of the goodwill deductions essentially account for the entire boost to capital from

stake disposals.

...but denting 2.7 p.p. of RoTE...

Table 3 shows the 2015E return on allocated capital (ROAC) and the dividend yield of CABK’s stakes

and estimates the impact to CABK’s 2105E ROTE from a sale. We estimate allocated capital as:

11.6% of RWAs, in line with CABK’s FL CET1 ratio;

Plus goodwill;

Minus capital gains/losses.

This implies €4.3bn capital absorbed by equity stakes and generating €0.8bn profits (and dividends)

for an average ROAC of 18%. In detail:

Chart 2: distributions of CET1 absorption and of capital free-up from stake disposals


Chart 3: share of capital relief from stake disposals


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

CET1 absorption Sale

Industrial

FIG >10%

FIG <10%

31%

-25%

71%

24%

Goodwill

Gains/losses

RWA

Deductions

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 9


Industrial stakes consume €1.8bn capital and generate €347m profits (and dividends), for

19% 2015E ROAC;

FIG stakes >10% consume €2.5bn capital and generate €225m profits, for 9% ROAC;

FIG stakes <10% consume -€52m capital on account of €587m MTM gains more than

compensating for the €299m goodwill and the €234m capital absorption from RWAs. This

makes the €186m profit consolidation essentially an infinite ROAC.

Our analysis concludes that the full sale of stakes would lose CABK €0.7bn profits (€0.8bn dividends

and profits minus the €4.3bn capital free up invested on Spanish 3 year bonds at 0.45%), or 50% of

2015E profits for a loss of 2.6 p.p. of RoTE. This would take CABK’s 2015E ROTE from our current

7.2% estimate to 4.5%.

Telefonica the largest single ROTE contributor

Looking at the specific components of the portfolio (see Chart 4) we note that out of the 2.7 p.p.

2015E RoTE loss:

Telefonica is the single stake with the highest ROTE contribution at 0.7%;

FIG stakes <10% (EBS, GF Inbursa) jointly contribute 1.2 p.p. of ROTE;

Repsol and FIG stakes >10% are relatively less productive, each generating 40bp ROTE.

Table 3: CABK RoTE erosion

Company Stake

Carrying

value Goodwill

Gain/lo

ss AC

Div

2015

Profits

2015 Yield ROAC

CABK RoTE

2015

Delta CABK RoTE

from sale

Repsol 11.9% 3,367 0 (663) 1,280

174

14% 6.8% 0.4%

Telefonica 5.3% 3,213 0 - 559 173

5.4% 31% 6.6% 0.7%

Industrial companies

6,580 - (663) 1,839

347

19% 6.1% 1.1%

FIG stakes >10%

3,042 634 (692) 2,536

225

9% 7.0% 0.2%

FIG stakes <10%

1,738 299 587 -54

186

n.s. 6.3% 0.9%

Total FIG. Stakes

4,780 933 (106) 2,482

412

17% 6.0% 1.3%

Tot. Stakes

11,360 933 (769) 4,322

759

18% 4.5% 2.7%

CABK

7.2%

CABK ex stakes

4.5%

Source: Mediobanca Securities, BBG consensus, company data

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 10


Bank of East Asia is the first candidate for a disposal...

Chart 5 maps the trade-off from stake sales on the CABK CET1 FL ratio and the 2015E ROTE. FIG

stakes >10% show a minimal negative impact to group ROTE with the highest positive re-rating of

CET1, suggesting this is where management should 1st look to make sales, in our view (Bank of East

Asia above all, given the recent move to take control of BPI).

Repsol takes second place in the disposal priority list, in our view. The negligible impact to core

capital comes with c.40bp erosion to ROTE.

Telefonica and FIG stakes <10%, despite the low strategic rationale, are the ones whose sale would

imply the largest ROTE hits, but would also provide material CET1 boost.

...generating alone 81bps CET1 boost

Chart 4: CABK 2015E ROTE waterfall for stakes disposals


Chart 5: CABK CET1 and RoTE impact from stake disposals


7.2%

4.5%

0.4%

0.7%

0.4%

1.2%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

2015E CABK RoTE Repsol Telefonica FIG stakes >10% FIG stakes <10% 2015E ROTE ex stakes

Repsol

Telefonica

FIG stakes >10%

FIG stakes <10%

-1.00%

-0.90%

-0.80%

-0.70%

-0.60%

-0.50%

-0.40%

-0.30%

-0.20%

-0.10%

0.00%

-0.10% 0.00% 0.10% 0.20% 0.30% 0.40% 0.50% 0.60% 0.70% 0.80% 0.90%

2015E R

OT

E d

elt

a

2014 FL CET1 delta

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 11


Table 4 estimates the CET1 impact from the individual sale of the BEA stake. We calculate this

operation would relieve CABK from capital deductions as the BV of FIG stakes >10% would not

exceed 10% of the bank’s capital. As a result, we estimate the sale would generate 81bp CET1

boost.

Table 4: Bank of East Asia stake disposal to CABK CET1 FL ratio

Status quo BEA sale

10% CET1 threashold 1,687 1,687

TE of FIG stakes > 10% 2,408 1,051

RWA from FIG stakes >10% 4,218 2,628

Deductions from FIG stakes >10% 721 -

Gains/losses

(316)

Goodwill 568 568

CABK RWAs 145,019 143,429

CET1 16,871 17,844

CET1 ratio 11.63% 12.44%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 12


Stake sales + mitigation vs regulation threat and M&A In this chapter we weigh opportunities and threats for CABK’s capital ratio. After

dealing with asset quality in 2014 with the Comprehensive Assessment, we envisage

2015 will be for the ECB the year of regulatory harmonisation. We see little risk for

CABK on credit risk but the potential lifting of the Danish Compromise on insurance

holdings, risk weights on govies and the repercussions on DTA/DTC could erode CET1

by 350bp including acquisitions. Of these, stake sales, the listing of BPI minorities and

mitigation on insurance could offset 280bp, maintaining the adjusted 2014 FL CET1

ratio at 11.5%. This suggests when management is targeting 11% post BPI, they have

already taken into account a combination of capital management actions. We

conclude that CABK has room to manage the capital ratio but also that the potential

gearing factors are material and that capital excess considerations on CABK should be

put in this context.

2015 is the year of regulatory harmonisation for the Banking Union

We are convinced that 2015 is the year of regulatory harmonisation in the Banking Union. This

means that the Common Regulator will have to progressively harmonise capital arbitrage across

jurisdictions and banks. The main topics on the table regard:

Credit risk weights - A-IRB risk weights where internal models assess materially different

risks for the same product or counterpart, implying intrinsically asymmetric risk profiles of

banks which are hard to assess by the market (and by regulators);

Market risk, operational risk, litigation risk – the crisis taught us that banks

underestimated these risks and regulators will likely increase risk weights or demand

buffers to cushion risks coming from such factors;

Government bond risk weights – the crisis showed that government bonds are not risk

free. This is a very delicate topic particularly within the Eurozone where introducing risk

weights on govies could refuel the issue that there is no buyer of last resort;

DTA/DTC treatment into CET1 – peripheral EU countries provided a government guarantee

on a large portion of DTAs to allow their computation into CET1 ratio. If risk weights of

govies are introduced, it is likely that this could trigger a limitation or a discount of such

items into regulatory ratios.

Credit risk weight harmonization is not a threat...

Chart 6 shows the comparison of RWA/Assets and of RWA/Loans for EU banks. CABK stands at 41%

and 74%, respectively, compared with 35% and 84% at aggregate sector level and 39% and 74% for

retail banks. This suggests CABK’s RWA density is in-line with other retail banks and above the

sector, so that a potential harmonisation should not pose a serious threat to capital ratios.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 13


...but Danish Compromise could be waived...

CABK controls 100% of VidaCaixa, one of Spain’s largest insurance companies with c.€36bn technical

reserves. In 2013 the vehicle carried €3.5bn tangible equity, after €1bn goodwill deduction. This

provided for a 10% ratio between tangible equity and technical reserves. Under Basel III, the

treatment of the equity from controlling stakes in insurance companies is part of the 10% and 15%

buffers of FIG stakes and DTA. Any excess is a deduction from capital ratios. The introduction of the

Danish Compromise allowed to account for such exposures through a penalising 370% risk weight.

The ECB’s harmonisation of capital rules could potentially lift the adoption of the Danish

Compromise, for a potential large hit to CABK’s CET1 ratio. In 2014, CABK geared up VidaCaixa by

paying out over €2bn equity to the parent company, hence reducing the potential risk to group

CET1 ratio. On the Q414 call management disclosed the lifting of the Danish Compromise would

account for c.90bp CET1 erosion but could be reduced to less than 50bp from further mitigating

actions. Table 5 shows the estimated evolution of VidaCaixa’s equity and the 2014 adj. equity to

reduce capital deductions. We envisage up to €1bn further gearing at VidaCaixa through the payout

of further relevering of the balance sheet. This would take TE/technical reserves at 2.9%. This

move matches what Credit Agricole (CA) is doing on its huge insurance operations. The current

€11bn equity supporting the €220bn technical reserves could be reduced to €6bn according to the

company, taking the ratio between the two items to 2.7%, in line with CABK’s potential target.

Chart 6: comparison of RWA/assets and RWA/loans, 2014


Table 5: VidaCaixa vs CredAg capital requirements 2013, 2014, adjusted for Basel III FL I l s e

2013 2014 adj. CA 2014 CA adj

Capital 1.3 1.3 1.3

Share premium, reserves and retained earnings 3.2 1.7 0.7

Goodwill 1.0 1.0 1.0

TE 3.5 2.0 1.0 11 6

Technical reserves 35 36 36 220 220

Equity/reserves 10.1% 5.6% 2.9% 5.0% 2.7%


41%35%

39%

74%84%

74%

0%

20%

40%

60%

80%

100%

120%

140%

ALPH

A

PM

I

PIR

BPE

RBI

BBVA

Euro

bank

SAB

NBG

EBS

STAN

PO

P

UBI

UCG

BKT

HSBC

DN

B

ISP

SAN

MPS

BP

CABK

CBK

KBC

RBS

SEB

BKIA

ND

A

BAR

CS

LLO

Y

ING

BN

P

GLE

KN

DAN

SKE

DBK

SW

ED

A

UBS

CASA

SH

B

Banks

Reta

il b

anks

WB b

anks

RWAs/ASSETS RWA/LOANS

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 14


...risk weights on EU govies is likely to be introduced...

Table 6 shows the evolution of CABK’s ALCO portfolio. As of Q414 this was essentially constituted of

€37bn, 80% from Spanish govies, 20% from other securities. Despite the drop in sovereign spreads,

the yield of the portfolio has maintained stable at a staggering 3.4% through the increase in

duration from 2.1 to 3.1 years. Interestingly, the incidence of Spanish govies on total assets has

reverted to 8% in Q414, back to the Dec-13 level and down from a peak of 10% in Jun-14 and Sep-

14. We believe this is the consequence of the compression of sovereign yields. Yet, the likely action

on risk weights on govies from the ongoing ECB work on RWA harmonisation should imply a

continuation in the diversification of the portfolio, in our view. For the sake of our simulation of

sensitivity of capital ratios, we conservatively assume 10% risk weight.

...likely combined with DTA/DTC restrictions

CABK carries €4.6bn government guaranteed DTAs. This represents 27% of CABK’s FL CET1 capital

and 3.2 p.p. of CET1 ratio. We believe that the recognition of state-guaranteed DTAs into CET1 goes

hand in hand with the introduction of risk weights on govies as these types of DTAs represent

implicitly the same risk. For the sake of our simulation of sensitivity of capital ratios, we assume a

10% haircut to DTA recognition into CET1 capital.

Capital is not an excess resource Table 8 shows the summary of our CABK capital sensitivity where we include impacts from acquisitions, foreseeable regulatory changes, equity stakes sale and mitigating actions. We include:

Acquisitions: Barclays Spain (100%) and BPI (assuming 100% takeout and 25% minority

listing);

Sales: the complete sale of industrial stakes;

Table 6: CABK breakdown of AFS bond portfolio

Govies ES govies Other Yield Duration

govies as %

of assets

Dec-13 27.1 13.5 3.30% 2.1 8%

Mar-14 29.0 12.8 3.40% 2.2 9%

Jun-14 32.8 11.2 3.40% 2.6 10%

Sep-14 31.8 10.5 3.40% 2.5 10%

Dec-14 28.7 7.8 3.40% 3.1 8%


Table 7: DTA/DTC incidence in CABK’s Basel III FL ratio, 2014

2014

DTA/DTC 4,600

CET1 capital 16,871

RWAs 145,019

DTC incidence on RWA 3.2%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 15


Regulatory issues and mitigations: waving of the Danish Compromise on the treatment of

the insurance business and management actions to mitigate this, the introduction of risk

weights on govies and the consequent haircut to DTAs into CET1 capital. We assume no

impact to RWA from credit risk harmonisation.

Overall we see potential for up to 3.5 p.p. gearing of CET1 ratio and 2.7 p.p. of capital free up:

Announced acquisitions potentially gear up CET1 ratio by 2.1 p.p. and could be offset by

30bp assuming 25% minority listing. Equity stakes disposal would offset this with 2 p.p.

accretion;

The change in the treatment of insurance operations could cost 90bp, half offset by

potential management actions;

Regulatory risks could account for 0.5 p.p. gearing.

Overall these factors would take CABK’s Q414 FL CET1 ratio pre-acquisitions to the adjusted level of

11.5%, essentially implying that - excluding the Barclays Spain acquisition - CABK could have enough

in-house means to fully offset balance sheet gearing items.

Yet, we conclude the recently announced acquisitions and the potential future regulatory changes

imply CABK is not an over capitalized bank.

Table 8: CET1 waterfall for acquisitions, disposals and regulatory risks

CET1 Assumptions

2014 FL CET1 (incl. BAR) 12.3%

BAR ES 0.70%

BPI 1.40% 100% takeout

Danish compromise 0.90% full deduction

RWA on govies 0.23% 10% risk weight

DTA 0.32% 10% discount

Stakes disposal 1.96%

BPI minority 0.30% 25% minority listing

VidaCaixa mitigation 0.45%

payout of share

premium


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 16


Chart 7: comparison of RWA/assets and RWA/loans, 2014


12.3%11.5%

0.7%1.40%

0.9%0.2% 0.3%

2.0% 0.3%0.5%

0%

2%

4%

6%

8%

10%

12%

14%

2014 FL CET1

BARES

BPI Danish compromise

RWA on govies

DTA Stake disposals

BPI minority VidaCaixa mitigation

2014 FL CET1 adj

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 17


Building the second home market CABK made a VTO for BPI to gain full control of the Portuguese bank. We attach low

visibility to the €130m synergy target as these imply 25% reduction of BPI’s cost vs a

benchmark of 16% for cross-border deals. Looking at profitability, BPI emerges as a

relative underperformer in Portugal. This is particularly evident if BPI has to sell its

African businesses. Hence, we conclude that on one hand, the valuation offered to BPI

is relatively rich, but reflects the premium required to change BPI’s governance and

could be justified in case of full delivery on synergy targets. The main positive in our

view is that through this deal CABK is regaining full control over its investment case

and particularly to manage directly the potential evolutions on the Portuguese side.

We believe that BPI is only half the story and that the real deal is gaining top spot in

Portugal through BPI’s acquisition of Novo Banco. The latter is up for sale, is twice the

size of BPI, and a merger would deliver Portugal’s #1 bank with c.20-25% market

share. The deal would be transformational for CABK as well. We estimate €291m

visible after-tax synergies coming from the in-market merger and excluding cross-

border synergies announced, leading to 6% 3-yr EPS accretion at CABK following the

sale of Africa and €5.5bn rights issue - and assuming no stake disposal – to defend 11%

CET1 ratio. The deal would transform CABK’s investment case, converting the

potential „cashback‟ story from high capital ratios, cost cutting and stake disposals

into the bank of the Iberian peninsula, i.e. a large cross-border restructuring story to

play the recovery of peripheral Europe. This would position CABK for the forthcoming

cross-border M&A we believe the Banking Union will trigger in coming years.

BPI’s tender offer is only half of the story...

Voluntary tender offer on BPI at 27% premium

On 17 February 2015, CABK launched a voluntary tender offer (VTO) on the 56% of Portugal’s BPI

CABK does not already own. The offer price of €1.329 for 814.5m shares implies a total potential

cash outflow of €1.1bn and a premium of 27% on the previous closing price. The VTO is subject to

the regulatory approvals and to two conditions imposed by CABK:

1) Reaching >50% ownership (i.e. >5.9% additional stake);

2) The removal by BPI’s AGM of the current 20% voting cap.

The second point requires a 75% qualified majority, implying BPI shareholders will have to vote in

mass in favour of the change.

BPI is Portugal’s and Angola’s #4 bank

With €43bn assets, €25bn loans, €24bn deposits and 1.7m clients, BPI is the fourth bank in the

Portugal and in Angola, where BPI controls BFA. BPI has an 8.6% Basel III FL CET1 ratio, 103%

loan/deposit ratio, 61% loan/asset ratio and 88% NPL coverage.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 18


Angola is subsidizing Portugal losses

Table 10 shows the geographical contribution to the BPI group. Portugal accounts for c.80% of costs,

93% of loans, c.80% of assets and 74% of deposits. Yet, Angola returned €126m profits in 2014 vs

€288m losses in Portugal (€28m losses in 2013).

€130m synergies target

CABK is targeting €130m synergies from BPI from ‘sharing of best practices to enable significant

improvements in profitability over time’. This should take BPI’s C/I from the current 85% to 50% by

2017.

Negatives: low visibility on cross border synergies...

CABK’s €130m targeted synergies represent 25% of BPI’s Portuguese costs. This compares with a 10-

year average cost synergies at 16% of the target’s cost base in cross border deals (see Table 11).

Given CABK’s long lasting investment into BPI, we would expect potential cross border synergies

should have been already partly realized. Hence, we see as a the €130m synergies target suffering

from low visibility.

Table 9: BPI 2014 metrics

2014

Assets 41,287

Loans 25,191

Deposits 24,380

RWAs 20,208

NPL 1,219

CET1 ratio FL 8.6%

L/D 103%

Loans/assets 61%

NPL coverage 88%


Table 10: Portugal share of BPI group

BPI Portugal Angola

Portugal % of

group

Costs -672 -530 -142 79%

Profits -162 -288 126 178%

Loans 25,269 23,436 1,833 93%

Assets 42,633 34,851 8,452 82%

Deposits 28,135 20,686 7,449 74%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 19


Table 11: summary revenue and cost synergies, May-04 to Sept-14

Date Revenue Costs

RBS/Charter One May-04 12% 24%

Santander/Abbey Nov-04 5% 17%

Danske/Irish NAB Dec-04 0% 19%

ABN AMRO/Antonveneta Mar-05 5% 15%

UCG/HVB Jul-05 1% 15%

Commerzbank/Eurohypo Oct-05 5% 15%

BNP/BNL Feb-06 7% 14%

Natixis Mar-06 9% 9%

BCP/BPI Apr-06 0% 40%

BIN-SPI Aug-06 3% 19%

BIN-SPI (realised) Aug-06 3% 27%

BPVN-BPI Oct-06 17% 22%

Credit Agricole / Cariparma Oct-06 5% 9%

Danske/Sampo Bank Nov-06 0% 18%

BPU/ BL Nov-06 0% 21%

UCG/CAP May-07 6% 24%

RBS/ABN Amro Jul-07 2% 8%

MPS/Antonveneta Nov-07 6% 25%

ISP/Carifirenze Jun-08 9% 29%

DBK/DPB Sep-08 6% 25%

Lloyds/HBOS Sep-08 0% 15%

BNP/Fortis Oct-08 3% 20%

Bankia May-11 0% 17%

POP/Pastor Oct-11 0% 40%

SAB/CAM Dec-11 4% 32%

BBVA/Unnim Mar-12 3% 40%

CABK/BdV Nov-12 0% 38%

SAN/Banesto Dec-12 4% 51%

BBVA/CX Jul-14 0% 40%

POP/Citi Jun-14 0% 20%

CABK/BARC Sep-14 0% 42%

Average (Mean) 4% 24%

Average in-market (Mean) 4% 31%

Average in Spain (Mean) 1% 36%

Average cross border (Mean) 4% 16%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 20


...and valuation looks rich vs profitability, reflecting premium to change governance

Table 12 shows the comparison between the valuation of BPI and of BCP, according to Bloomberg.

This shows BPI trading on 0.9x TE post CABK’s voluntary offer, a double digit discount vs BCP’s 1.1x.

Table 13 compares the profitability of BPI to the one of BCP and Santander Totta, the main

domestic competitors in 2014-17E. 2014E has been a loss making year for BPI and BCP and is

therefore of little help to assess BPI’s relative profitability. Looking forward, we construct

perspective profitability by:

taking BBG consensus estimates for BPI and BCP and our estimates for Santander Totta;

we grow BPI African businesses’ 2014 reported profits at the 6.6% average growth of the

last six years to estimate the Portuguese profitability.

The indication from this analysis is that BPI suffers from a structurally lower than peer profitability

of c.2-3 p.p. of RoTE. This, despite a relatively stretched CET1 FL ratio of 8.6%.

We believe such relatively high valuation embedded in CABK’s VTO reflects the premium required

to remunerate shareholders for the removal of the 20% voting cap, which requires a vast majority

support at the AGM.

CABK consideration for BPI highly relies on delivery on synergies

Table 14 shows the sensitivity of BPI’s valuation to the sale of the African businesses and to the

synergies targeted by CABK. On the former, we apply a conservative 10x PE multiple to African

earnings. This implies a valuation of €1.3bn for Angola and Mozambique, generating a capital gain

of €0.8bn. This should imply a P/TE valuation of 2.8x, which reflects the 28% RoTE of 2014.

Table 12: BPI valuation multiples vs BCP

BPI BCP

BV 2,129 4,987

Intangibles 25 1,198

TE 2,104 3,789

Market cap 1,847 4,320

P/TE 0.9 1.1

Source: Mediobanca Securities, BBG, company data

Table 13: BPI profitability benchmarking

2014 2015 2016 2017

BPI RoTE -7.7% 7.0% 8.3% 8.5%

BPI ex Africa

1.3% 3.0% 3.2%

BPI ex Africa incl. CABK synergies 1.3% 6.4% 7.3%

BCP RoTE -3.5% 7.6% 12.0% 12.2%

SAN TOTTA RoTE 12.0% 11.3% 10.2% 10.1%

Avg RoTE competitors

9.5% 11.1% 11.1%

Source: Mediobanca Securities, BBG, company data

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 21


Adjusting the consideration paid by CABK for BPI for the potential sale of the African businesses

retrieves a P/TE of 0.6x for the equivalent 2017E RoTE of 3.2%, which would reflect a dear

valuation for a business. Assuming 100% delivery on targeted synergies by 2017E would lift RoTE by

4p.p. to 7.3%, making the 0.6x P/TE valuation an attractive one, in our view.

We conclude that the 0.9x optical valuation for BPI offered by CABK heavily relies on the delivery

on targeted synergies.

Positives: taking control of a large part of CABK’s investment case

We believe the main positive of the CABK VTO on BPI is that with the move CABK is looking at

regaining full control of its own destiny/investment case. In fact, in the context of the upcoming

auction of Novo Banco (former Banco Espirito Santo), CABK was running the risk of having BPI – one

of its main investments – potentially making an offer for the acquisition of Novo Banco with CABK

exposed to a transformation of its investment case without full control on it.

The 20% voting cap of BPI is to blame for this anomaly. In fact, despite CABK’s 44% stake, the voting

cap implies CABK has equal status in the BoD to Santoro (Angolan investors) which own only 18.6%,

leaving Allianz – third shareholder with 8.4% - and BPI management as potentially able to tilt key

decisions.

Hence, CABK’s move on BPI is a way to take the bull by the horns and make a definitive decision on

its Portuguese position, i.e. if the VTO will be successful, BPI will become core business and

Portugal will be CABK’s second home market; if not, we envisage CABK’s and BPI’s destinies will

move apart.

Table 14: BPI valuation excl. African businesses

BCI BFA BPI Africa BPI

Stake 30.0% 50.1%

Profits 11 117 127

P/E valuation 10 10

Price 105 1,169 1,274

Carrying value 55 418 473

RWA

8,124

Gain 50 751 801

BV

456

P/BV implied for BPI Africa

2.8

Consideration paid

1,847

P/TE

0.9

Excl. Africa

0.6

2017E RoTE ex-Africa

3.2%

2017E RoTE ex-Africa incl. synergies

7.3% GGGG

Source: Mediobanca Securities, company data, BBG

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 22


…the real deal is taking the top spot in Portugal Novo Banco’s auction to start at the end of March

Novo Banco is a carve out from Banco Espirito Santo. The only publicly available data reflect the

balance sheet as at Dec-14 (see Table 15). The bank holds €72bn assets, €39bn loans, €27bn

deposits and €50bn RWAs. Main ratios indicate 9.2% CET1 FL ratio, 152% NPL coverage and 141%

loan/deposit. The Bank of Portugal has announced the 15 selected bidders will have to submit their

non-binding offers for Novo Banco by 20 March.

We see two potential types of bidders:

1) domestic competitors extracting cost synergies;

2) Non-European bidders buying the entry ticket to the EU.

In the former group we see larger domestic players (BPI, Santander, BCP) with higher chances given

the large expected cost synergies to be extracted.

In the latter group we see both private equity funds acquiring access to the ECB’s liquidity to fund

their investment portfolios or sovereign wealth funds (Chinese) which are aggressively growing their

peripheral EU exposure to boost portfolio diversification and exposure to Euro denominated assets

with good recovery prospects in the longer term.

BPI-Novo Banco would be a transformational deal...

Table 16 shows the total loans, assets, deposits and tangible equity of BPI and Novo Banco as at

Dec-14 against the Bank of Portugal aggregate figures for banks. The combination of the two banks

would create a bank with 20-25% market share in Portugal as the figures in the table include the

international businesses of BPI which should be stripped out.

Table 15: Novo Banco 2014

Novo Banco 2014

RWAs 49,906

CET1 FL 4,610

CET1 ratio FL 9.2%

NPL 3,462

Coverage 5,248

Coverage ratio 152%

Net assets 72,465

Loans 38,569

Deposits 27,281

Loans/assets 53%

L/D 141%

BV 5,577

Intangible 336

TE 5,241


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 23


Geographically, the combination would provide 23% market share in branches at national level, with

a fairly homogeneous distribution across the different districts (see Chart 8).

Hence, the potential combination would be a transformational deal for BPI, almost trebling the size

of the Portuguese bank in the country and constituting the undisputed leader in Portuguese

banking.

...but no different from other national champions

Table 17 compares the national market share of loans and deposits for the main national champions

in Europe. The combination of BPI and Novo Banco would position between the levels of LLOY in UK

and Credit Agricole Group in France so that we see little risk of material antitrust issues from a

potential merger, in our view.

Table 17: comparison of national champions market shares, Q314

BPI

Novo

Banco

BPI/Novo

Banco ISP LLOY CA

Loans 10% 15% 24% 15% 19% 29%

Deposits 10% 11% 21% 16% 22% 31%


Table 16: BPI/Novo Banco market share, 2014E

BPI Novo Banco

BPI+Novo

Banco Portugal Mkt share

Loans 25,191 38,569 63,760 263,893 24%

Assets 41,287 72,465 113,752 449,356 25%

Deposits 24,380 27,281 51,661 250,571 21%

Capital 2,104 5,241 7,345 29,305 25%

Source: Mediobanca Securities, company data, APB

Chart 8: branch market share

Source: Mediobanca Securities, company data, APB

23% 24% 24%20%

25% 25% 24%22% 22%

9%

22%

0%

5%

10%

15%

20%

25%

30%

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 24


In-market synergies would be much more visible for the market

There is very little the market knows about Novo Banco. The opening balance as at Dec-14 is the

main information available and there is no detail on P&L as yet. Therefore, to assess the magnitude

of potential cost synergies, we estimate the cost base by applying a 1.5% of total assets in line with

the Portuguese business of BPI. This provides €1.1bn total costs at Novo Banco. Synergy targets in

line with the aggressive cost cutting plans of the Spanish integrations of the recent years (see Table

11) would provide for €0.4bn cost synergies or €0.3bn after tax.

Base case: 6% EPS adj. accretion with €5.5bn rights issue keeping CET1 at 11%...

Table 19 simulates the merger between CABK, BPI and Novo Banco to assess the impact to CABK

adj. EPS, ROI and impact to Basel III fully loaded CET1 ratio.

By definition, a three-way merger is a complex operation with higher execution risks. We believe

modelling the CABK/BPI/NB combination is particularly challenging given the following

uncertainties:

The little knowledge available on Novo Banco, requiring a number of benchmarking

exercises to assess costs - and subsequent restructuring costs and cost synergies - and a

fair value to calculate the financials of the deal on, including the potential badwill

generation;

The potential requirement by regulators for BPI to dispose of African businesses, with little

reference as to the market value of those assets;

The redefinition of BPI ex-Africa and its perspective growth in earnings and RWAs;

The mechanics of the potential deal, which require making an assumption on which is the

integrating entity and if the Portuguese business would indeed remain listed or merged

into CABK;

The approach of CABK to capital ratios and the will to use equity stake disposals to fund

the acquisition;

The dividend policy post deal.

Our analysis follows from the following assumptions:

We assume the integration to follow a two-step process with BPI buying Novo Banco in cash

(as Portugal would likely prefer this to a paper deal) through a capital raise and CABK

supporting BPI’s capital raising and in turn raising capital to defend an 11% FL CET1 ratio

(hence no sale of equity stakes and no loss of earnings);

We assume CABK to completely absorb the two Portuguese banks, leaving no listed

minorities, and to proceed with the sale of BPI’s African businesses as per Table 14 and

therefore deduct the corresponding profits and the current pro-rata BPI contribution to

Table 18: BPI/Novo Banco synergy estimate

2014 2015 2016 2017

Novo Banco costs @ 1.55% of assets -1,123

Synergies @ avg Spain in market deals

399

Phasing

50% 75% 100%

Synergies

200 299 399

After tax synergies 146 219 291


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 25


CABK earnings embedded in 2015-17E. The sale also results in €0.8bn boost to CET1 ratio

(or a smaller capital raising);

We assume the BPI/Novo Banco merger to provide cost synergies (no revenue synergies) as

per Table 18 and assume restructuring costs at 40% of Novo Banco’s costs to be booked

over first two years at a 75-25 split;

We discard the €130m synergy target of CABK on BPI given the lower visibility and the large

risk of double counting with the BPI/Novo Banco synergies;

We consolidate €62bn RWAs from BPI/Novo Banco (€20bn from BPI, €50bn from Novo

Banco), excluding €8bn RWA from BPI’s African businesses;

We use BPI consensus estimates to project future profits and deduct African profits

assuming a 6% CAGR as per historical average. For Novo Banco, we apply the average

expected ROTE from peers (BCP consensus and SAN Totta from MBe). We grow BPI/Novo

Banco Portuguese RWA in line with our RWA growth estimate at CABK.

The combination results in 340bp CET1 stretch, taking 2015E FL CET1 ratio for CABK to 8.5%. We

assume CABK considering 11% the minimum level in line with the indication provided on the BPI VTO

call. This would require €5.5bn capital increase, which would imply 1.4bn new CABK shares at the

current €4.0 share price.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 26


In summary, the deal would result in:

3% 2015ED EPS dilution and 6% accretion in 2016-17E for 30% increase in CABK net profit;

17% ROI over three years;

11% fully loaded CET1 ratio with no equity stake disposal.

Table 19: simulation of merger model between CABK, BPI and Novo Banco

Impact on Earnings, € Mn 2014 2015E 2016E 2017E

CABK standalone adj. net profit 620 1,496 2,195 2,802

CABK diluted shares, €m 5,712 5,841 5,841 5,841

CABK diluted EPS before rights issue, €sh 0.11 0.26 0.38 0.48

BPI/NB net profits (ex int.)** 552 665 687

Restructuring cost, net -337 -112

Excluding current BPI domestic contribution 65 54 55

Excluding BPI int. contribution** -134 -143 -153

Cost synergy, net 146 219 291

Revenue synergy (lower funding cost), net 0 0 0

BPI/NB Net profit contribution 291 683 881

CABK+BPI/NB adj. net profit 620 1,787 2,877 3,682

CABK diluted shares, €m 5,712 5,841 5,841 5,841

Additional CABK shares from rights issue

CABK new EPS, €/sh 0.11 0.31 0.49 0.63

EPS impact 0% 19% 31% 31%

ROI and Implicit PE 2014 2015E 2016E 2017E

ROI 4.8% 11.3% 14.5%

CABK P/E pre-merger 37.0 15.7 10.7 8.4

Implied PE post-synergies 13.1 8.1 6.4

Impact in capital ratios, €Mn 2014 2015E 2016E 2017E

CABK old RWAs* 145,019 154,875 157,973 164,949

CABK old Core tier 1 capital 16,871 18,473 19,351 20,892

Fully loaded CET1 ratio pre deal 11.6% 11.9% 12.2% 12.7%

BPI/NB (ex. Int) RWAs** 61,990 66,203 67,527 70,510

CABK new RWAs 207,009 221,078 225,500 235,459

CABK FL CET1 capital, pre deal 18,473 19,351 20,892

+Capital increae from BPI/NB 5,549 5,549 5,549

+Capital gain on sale of Africa 801 801 801

-Total goodwill / badwill 0 0 0

+BPI/NB net profit contribution*** 291 633 1,073

CABK FL CET1 capital, post deal 24,313 25,533 27,514

Fully loaded CET1 ratio post deal 11.0% 11.3% 11.7%


*Basel III FL, **BPI int. 6% profit CAGR, ***50% cash payout from 2016 onwards

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 27


...but equity stake sales and minorities could minimise CABK’s capital raising

The Base Case simulation above is the one which attempts to minimise the moving parts from the

potential three-way merger. Clearly there is a multitude of permutations available involving:

Minorities listing of BPI from CABK’s VTO and from the subsequent BPI/Novo Banco merger

which would clearly reduce profit contribution and capital stretch for CABK;

The potential further stretching of CABK’s CET1 FL ratio which would reduce the assumed

cash call;

The disposal of part or all of CABK’s equity stakes to minimise the capital raising at CABK

but with the inevitable side effect of a short term hit to CABK’s profitability.

We have little to guesstimate how successful CABK’s VTO will be on BPI, whether BPI will indeed bid

and win the auction for Novo Banco and if CABK will fund the acquisition and use equity stakes to

minimise capital. Hence we model the ‘full Monty’ version of this potential integration and refer to

the previous chapter of this note to assess how the use of equity stakes could impact EPS accretion

and capital requirements.

Creating Iberia’s bank CABK potentially representing 14-18% of the Iberian market...

Chart 9 shows the market shares of CABK/BPI/Novo Banco in Spain, Portugal and the Iberian

peninsula. The three banks would hold a joint market share of 16%, 14%, 16% and 18% in Iberian

loans, assets, deposits and branches. The merger would effectively create the incumbent player of

the peninsula, in our view.

...positioning for European cross border M&A

In potentially delivering the BPI/Novo Banco acquisition, CABK would position well for European

cross border M&A, in our view. We believe the European Banking Union is starting to be

implemented faster than the market has realised. So far this year, shortly after the Comprehensive

Assessment publication and the beginning of the SSM regime we have seen:

SAN capitulating on capital raising after four years with €7.5bn ABB;

Chart 9: branch market share


24%25%

21%23%

15%

12%

15%17%16%

14%16%

18%

0%

5%

10%

15%

20%

25%

30%

Loans Assets Deposits Branches

BPI/Novo Banco (Portugal) CABK (Spain) CABK+BPI+NB (Iberia)

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 28


Italy approving a lightning reform of Popolari banks after 20 years of chit-chat;

SocGen appointing a chairman very close to the ECB, separating the executive and non-

executive powers previously merged in the chairman/CEO role;

Spain changing the approach to accounting of the Deposit Guarantee Fund costs of banks,

with non-negligible impacts to banks’ CET1 ratios;

Banks geographically diversifying their sovereign bond holdings.

In an interview early this year President Draghi said that for a Banking Union effective in cushioning

asymmetric shocks the creation of larger cross border banks is a necessary factor. We believe that

national markets are quickly proceeding with further consolidation to prepare the forthcoming cross

border combinations.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 29


CABK is not playing the low cost game Over the crisis of the last five years Spanish banks transformed through M&A and the

rethinking of their distribution networks. Spanish banks cut over 1/3 of branches, 25%

of loans and 10% of deposits while CABK instead maintained branches and loans

stable, and increased staff and deposits by 10% and 27%. This makes CABK the

mammoth of Spanish banking with a network twice the size of the average

competitor. While the latter concentrated staff in fewer, larger branches to cut costs,

CABK continues with a larger number of smaller branches. We estimate that closing

the branch productivity gap with peers would entail 12% EPS boost from €24bn

additional loans. Conversely, adopting the competitors’ setup would allow for 10%

cost cuts, boosting EPS by 18%. The different distribution model – including the full

range of wholly owned product factories - implies higher fixed costs at CABK, but also

higher and cheaper deposit raising capacity, which should play well for a recovering

environment.

Spanish banks structurally transformed over the last five years

Chart 10 shows the 2014/2010 growth in staff, branches, deposits and loans for the major Spanish

banks. We note the very asymmetric trends experienced across banks with those which underwent

sizeable acquisitions exhibiting spectacular growth (SAB above all) and others implementing a strict

diet. In particular:

BKIA is the bank which shrunk the most, as expected: -40% and -53% in staff and branches,

-23% and -32% in deposits and loans;

SAN – the only domestic bank which did not undertake any M&A – follows with -26% and -

28% in staff and branches and boosting deposit gathering with only 5% contraction vs 30% in

customer loans;

SAB is the bank which underwent the largest transformation with a leap in size summarized

by: +63% in staff, +62% in branches, +78% in deposits, +50% in loans.

Hence, there are no clear trends among Spanish banks. CABK grew staff by 10% while cutting

branches by 3% and boosting deposits by 27% with loans up 5%.

Chart 10: 2010-14 growth comparison

Source: Mediobanca Securities, company daya

10%

-3%

27%

2%8%

-4%

42%

5%

63% 62%

78%

50%

-40%

-53%

-23%

-32%

-26% -28%

-5%

-30%

1% 3%

21%

-17%

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

100%

Staff Branches customer deposits customer loans

CABK POP SAB BKIA SAN BBVA

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 30


Mammoth network (smaller branches) requires higher fixed costs…

Chart 11 shows the comparison at 2014YE across Spanish banks (Spanish businesses for SAN and

BBVA) of branches (lhs), cost/deposits, cost/loans. CABK holds a network of over 5,200 branches

before the Barclays Spain acquisition. This is already twice the average network of competitors and

1.6-1.7x larger than SAN and BBVA, making CABK the dominant franchise on the Spanish streets.

Comparing cost/deposits and cost/loans should provide us with indications on:

1) Economies of scale

2) The exploitation of the network and the room for future growth

In relative terms, CABK stands at 10% premium vs the peer average. BKIA emerges as the low cost

player with the smallest network after the halving of it over the last four years and costs 16-18%

below peers. SAN is the only player with more deposits than loans, reflected in the cost/loans ratio

well above the cost/deposits one.

...with cheaper and smaller branches, but dearer staff...

Chart 12 and Chart 13 map the comparison of cost/employee and cost/branch across Spanish banks

and the evolution of the average size of Spanish banks’ branches, respectively. The former shows

that Spain currently has a multitude of models:

BKIA shows higher staff remuneration and branch costs in line with peers, likely reflecting

the sharpest staff reduction of the panel (-40% in 4 years);

SAN and SAB are at opposite ends but not far from the average, with the former showing

higher staff remuneration and branch cost – also likely reflecting the skew towards city

locations – and the latter with lower staff remuneration and lower branch costs, probably

reflecting the lower levels of the large acquisitions made in the cajas space;

POP and BBVA go hand in hand with branch costs in line with the average and with staff

remuneration 15% below peers;

CABK is far from all peers, with relatively high remuneration (14% premium) but branches

coming at half price.

Chart 12: staff cost/staff vs G&A/branch comparison for Spanish banks, 2014

Chart 11: Spanish banks comparison for branches, cost/deposits, cost/branches, 2014


0.012

0.014

0.016

0.018

0.020

0.022

0

1,000

2,000

3,000

4,000

5,000

6,000

CABK POP SAB BKIA SAN BBVA avg peer

€m

n

Bra

nch

es

Branches (lhs) Costs/deposits (rhs) Costs/loans (rhs)

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 31



Chart 13 shows that CABK runs much smaller branches with less than six staff per branch or c.25%

below the peer average. This explains the much lower cost of CABK’s branches and the intrinsically

different business model operated by the huge Catalan bank. In essence, while other players have

redesigned their networks closing branches and focusing on fewer, larger ones to cut costs and

cushion the revenue attrition from the crisis, CABK has maintained a more dense network of smaller

outlets, with the higher fixed and running costs this implies.

...but stronger and cheaper deposit gathering should support stronger growth in a

recovery

Chart 14 shows the comparison of Q414 time deposit remuneration on front book and back book.

With 48bp, CABK is the bank with the second lowest cost of new deposits, after BBVA, but with the

same back book cost of SAB or POP. We believe this shows CABK’s superior pricing power than other

CABK

POP

SAB

BKIA

SAN

BBVA avg peer

150,000

200,000

250,000

300,000

350,000

400,000

450,000

60,000 65,000 70,000 75,000 80,000 85,000 90,000 95,000

G&

A /

bra

nch

Staff cost / employee

Chart 13: comparison of staff/branch evolution, 2010-14


5.2 5.1

5.9

7.3 7.4

7.7

4

5

6

7

8

9

10

2010 2012 2014

CABK POP SAB BKIA SAN BBVA avg peer

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 32


domestic peers, possibly deriving from the more dense branch network which captures deposits in

more remote markets characterized by lower pricing competition.

Chart 15 shows the mapping of loans and deposits per branch for Spanish banks in 2014. The slope

of 0.82 suggests banks are not running huge funding gaps. We note that CABK and SAN stand below

the regression line either suggesting they are the banks with the largest potential loan growth (from

the realignment to the regression line) or those with a lower productivity.

Realignment to the mean would generate boost EPS by 12%

Realigning CABK to the regression line in Chart 15 would imply 13% increase in loans per branch

(from €36m to €40.5m loans per branch, i.e. €34m deposits x 0.82 + 12.5) or €24bn additional loans

at group level. Applying the 2014 NIM of 2.15%, 54% C/I and 24% tax rate would retrieve €0.2bn

additional profits, for 12% 2015E EPS boost. This shows the potential for CABK in an expansionary

Chart 14: comparison of back book and front book remuneration of time deposits, Q414


Chart 15: loans/branch vs deposits/branch comparison for Spanish banks, 2014, € m


0

20

40

60

80

100

120

140

160

180

BKT POP SAN SAB CABK BBVA

FRONT BOOK

BACK BOOK

CABK

POP SAB

BKIA

SAN

BBVA

avg peer

30

35

40

45

50

55

60

65

30 35 40 45 50 55 60

loan

s/b

ran

ch

deposits/branch

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 33


environment, i.e. management is accepting higher structural costs from a larger network in

exchange of higher potential balance sheet expansion in the recovery.

10% cost cuts from a change of model for c.18% EPS boost

If instead of realigning productivity of branches to the average, CABK management decided to

adopt their peer’s distribution model, i.e. cutting branches and staff to reduce cost, we estimate

the boost to EPS would be of c.18% according to this methodologies:

Realigning cost/deposits to peers would imply €344m lower costs at CABK given €180bn

2014 deposits;

Realigning deposits/branch to that of peers would imply closing 1/3 of branches which at

an average peer’s cost/branch of €0.93 would imply €372m cost savings

Both methodologies point towards the same magnitude, i.e. c.10% cost cuts which, at 24% tax rate

would imply 17-19% boost to 2015E EPS.

CABK is positioned for economic recovery

CABK’s CEO explained they are happy with their different distribution model and that they have no

intention of changing it. We conclude that CABK is positioned to capture the recovery through

higher than peer balance sheet growth on the assumption that the larger deposit gathering provided

by the larger network will outperform the benefits of the higher operating leverage they would gain

from reducing costs at perhaps the cost of lower loan growth.

Table 20: simulation of CABK costs

CABK CABK adj. Cost savings

As % of

CABK costs

As % of

2015E

Costs/deposits, € m 0.021 0.019

Costs, € m -3,770 -3,426 344 -9% 17%

Deposits, € m 180,200 180,200

Deposits/branch, € m 34 49

Branches 5,522 3,664 -1,858

Costs/branch, € m (0.72) (0.93)

Costs, € m -3,770 -3,398 372 -10% 19%


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 34


30% swing around our €4.4 TP – Staying Neutral We run the sensitivity of our €4.4 2015E Target Price (TP) to the four scenarios

depicted in the note, namely: 1) the takeover of BPI/Novo Banco funded by a €5.5bn

cash call, 2) the same deal but funded by stake disposals, 3) the realignment of CABK

profitability through higher branch productivity, 4) cost cutting through branch

closures in line with peers. We conclude that scenario 1 plus 3 or 4 would deliver

+20% boost to our TP, while 3 would provide for -9% swing. This shows CABK valuation

is positively skewed to management action and we stay Neutral on CABK while

awaiting inputs on these items and on management’s plans to boost organic

profitability from tomorrow’s Investor Day.

+20% from cash call-funded M&A and organic profitability improvement; -9% from

stake disposal

We hold a Neutral rating on CABK and €4.4 2015E Target Price (see Table 21) based on the following

methodology:

1) We allocate a 10% minimum capital requirement to the €149bn RWAs to obtain allocated

capital;

2) We apply a 12.1% over the cycle ROAC which implies a higher profitability from CABK in a

normalized macroeconomic environment compared to the 10% level we retrieve on 2015E;

3) We estimate a cost of capital of 8.3% by multiplying the risk free rate to the stock’s beta

and to the 5.5% market premium we adopt across all European banks;

4) This retrieves a 1.45x valuation multiple providing for €21.7bn value;

5) To this we add the excess capital deriving from our estimated 2015E FL CET1 ratio minus

the allocated capital and the MTM of equity stakes.

We then analyze the sensitivity of our Target Price to the M&A and restructuring scenarios depicted

in the note, namely:

1. BPI/NB takeover and €5.5bn cash call: we assume a 12.8% sustainable RoAC at CABK

based on the discounting of the 2017E earnings boost from the acquisitions of BPI and NB

and an increase of allocated capital in line with the additional RWAs. On the dilutive side,

the latter would limit excess capital to €1.8bn, coupled with the new shares deriving from

the €5.5bn rights issue. This would lead to a fair value of €4.8, 9% above MB's TP and 18%

above the current price.

2. BPI/NB takeover and equity stakes disposal: this scenario assumes the funding of the

acquisition of BPI and NB through the sale of CABK’s stakes portfolio and a partial balance

sheet gearing. This would lead CABK to a CET1 ratio below the 10% capital allocation

constraint despite the RWA deconsolidation from stake disposals. This would result in €1bn

capital deficit and a fair value of €4.0, 9% below MB's TP and in line with the current price.

Table 21: CABK 2015E valuation

PTP Taxes Adj. Net

Profit RWAs

MB

capital

/RWA

Weight

range

Allocated

capital ROAC

Over the

cycle ROAC

Cost of

capital

P/TB

V Value

Capital

deficit/

excess

Per

share

CaixaBank 1,968 -472 1,496 149,001 10.0% 6-9% 14,900 10.0% 12.1% 8.3% 1.45 21,722 4,017 4.4


Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Caixa Bank

02 March 2015 ◆ 35


3. Branch productivity realignment: we analyze the impact to valuation of CABK increasing

its branch network productivity in-line with peers and increase loans by €24bn (€18bn

RWAs) and profits by €0.2bn. On the positives, we value extra earnings at 10x PE for €2bn

additional valuation. On the negatives, the higher RWAs imply higher allocated capital and

lower excess capital (€2.2bn). These lead to a CABK fair value of €4.9, 11% above MB's TP

and 20% above the current price.

4. Cost cutting realignment: we analyze the impact of CABK adopting competitors’

distribution model and cutting costs. On the positives, we value extra earnings at 10x PE

for €2.7bn additional valuation. There is no other change to valuation, leading to a CABK

fair value of €4.9, 11% above MB's TP and 20% above the current price.

Interestingly, scenario 3 and 4 result in the same fair value as the higher allocated capital of the

former triggered by the additional RWAs drives a valuation re-rating given the 1.5x valuation

multiple which compensates for the lower excess capital and the lower valuation boost from

additional earnings compared to scenario 4.

Chart 16: sensitivity of CABK Target Price to M&A and restructuring scenarios, 2015E


Our analysis highlights that CABK could be before a binary call on valuation; on one hand, stake

sales without a rights issue would leave 9% downside from our current TP, while a positive evolution

of cost controls and the acquisitions of BPI/NB would lead to a 20% upside, implying a net positive

skew of CABK valuation to management actions.

Staying Neutral while awaiting news on M&A, stake disposals and organic growth from

the Investor Day

Before the BPI deal, CABK could have been considered a safe play on Spain with upside potential

from the capital redeployment angle. But after the balance sheet gearing, we see the disposal plan

as a necessity and CABK increasing its beta gearing on Iberian macro. We stay Neutral while

awaiting to receive inputs from management at tomorrow’s Investor Day on the key swing factors

analyzed in this note and on management’s intentions to drive profitability up.

4.4

4.8

4.0

4.9 4.9

3.5

4

4.5

5

Inital TP BPI+NB takeover and capital raise

BPI+NB takeover and stake sales

Realign profitability

Cut costs

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Disclaimer

02 March 2015 ◆ 36

GENERAL DISCLOSURES

This research report is prepared by Mediobanca - Banca di credito finanziario S.p.A. ("Mediobanca S.p.A."), authorized and supervised by Bank of Italy and Consob to provide financial services, and is compliant with the relevant European Directive provisions on investment and ancillary services (MiFID Directive) and with the implementing law.

Unless specified to the contrary, within EU Member States, the report is made available by Mediobanca S.p.A. The distribution of this document by Mediobanca S.p.A. in other jurisdictions may be restricted by law and persons into whose possession this document comes should inform themselves about, and observe, any such restrictions. All reports are disseminated and available to all clients simultaneously through electronic distribution and publication to our internal client websites. The recipient acknowledges that, to the extent permitted by applicable securities laws and regulations, Mediobanca S.p.A. disclaims all liability for providing this research, and accepts no liability whatsoever for any direct, indirect or consequential loss arising from the use of this document or its contents. This research report is provided for information purposes only and does not constitute or should not be construed as a provision of investment advice, an offer to buy or sell, or a solicitation of an offer to buy or sell, any financial instruments. It is not intended to represent the conclusive terms and conditions of any security or transaction, nor to notify you of any possible risks, direct or indirect, in undertaking such a transaction. Not all investment strategies are appropriate at all times, and past performance is not necessarily a guide to future performance. Mediobanca S.p.A. recommends that independent advice should be sought, and that investors should make their own independent decisions as to whether an investment or instrument is proper or appropriate based on their own individual judgment, their risk-tolerance, and after consulting their own investment advisers. Unless you notify Mediobanca S.p.A. otherwise, Mediobanca S.p.A. assumes that you have sufficient knowledge, experience and/or professional advice to undertake your own assessment. This research is intended for use only by those professional clients to whom it is made available by Mediobanca S.p.A. The information contained herein, including any expression of opinion, has been obtained from or is based upon sources believed to be reliable but is not guaranteed as to accuracy or completeness although Mediobanca S.p.A. considers it to be fair and not misleading. Any opinions or estimates expressed herein reflect the judgment of the author(s) as of the date the research was prepared and are subject to change at any time without notice. Unless otherwise stated, the information or opinions presented, or the research or analysis upon which they are based, are updated as necessary and at least annually. Mediobanca S.p.A. may provide hyperlinks to websites of entities mentioned in this document, however the inclusion of a link does not imply that Mediobanca S.p.A. endorses, recommends or approves any material on the linked page or accessible from it. Mediobanca S.p.A. does not accept responsibility whatsoever for any such material, nor for any consequences of its use. Neither Mediobanca S.p.A. nor any of its directors, officers, employees or agents shall have any liability, howsoever arising, for any error, inaccuracy or incompleteness of fact or opinion in this report or lack of care in its preparation or publication.

Our salespeople, traders, and other professionals may provide oral or written market commentary or trading strategies to our clients and our proprietary trading desks that reflect opinions that are contrary to the opinions expressed in this research. Our proprietary trading desks and investing businesses may make investment decisions that are inconsistent with the recommendations or views expressed in this research. The analysts named in this report may have from time to time discussed with our clients, including Mediobanca S.p.A. salespersons and traders, or may discuss in this report, trading strategies that reference catalysts or events that may have a near-term impact on the market price of the equity securities discussed in this report, which impact may be directionally counter to the analysts' published price target expectations for such stocks. Any such trading strategies are distinct from and do not affect the analysts' fundamental equity rating for such stocks, which rating reflects a stock's return potential relative to its coverage group as described herein.

ADDITIONAL DISCLAIMERS TO U.K. INVESTORS: Mediobanca S.p.A. provides investment services in the UK through a branch established in the UK (as well as directly from its establishment(s) in Italy) pursuant to its passporting rights under applicable EEA Banking and Financial Services Directives and in accordance with applicable Financial Services Authority requirements.

ADDITIONAL DISCLAIMERS TO U.S. INVESTORS: This research report is prepared by Mediobanca S.p.A. and distributed in the United States by Mediobanca Securities USA LLC, which is a wholly owned subsidiary of Mediobanca S.p.A., is a member of Finra and is registered with the US Securities and Exchange Commission. 565 Fifth Avenue - New York NY 10017. Mediobanca Securities USA LLC accepts responsibility for the content of this report. Any US person receiving this report and wishing to effect any transaction in any security discussed in this report should contact Mediobanca Securities USA LLC at 001(212) 991-4745. Please refer to the contact page for additional contact information. All transactions by a US person in the securities mentioned in this report must be effected through Mediobanca Securities USA LLC and not through a non-US affiliate. The research analyst(s) named on this report are not registered / qualified as research analysts with Finra. The research analyst(s) are not associated persons of Mediobanca Securities USA LLC and therefore are not subject to NASD rule 2711 and incorporated NYSE rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst.

ADDITIONAL DISCLAIMERS TO U.A.E. INVESTORS: This research report has not been approved or licensed by the UAE Central Bank, the UAE Securities and Commodities Authority (SCA), the Dubai Financial Services Authority (DFSA) or any other relevant licensing authorities in the UAE, and does not constitute a public offer of securities in the UAE in accordance with the commercial companies law, Federal Law No. 8 of 1984 (as amended), SCA Resolution No.(37) of 2012 or otherwise. This research report is strictly private and confidential and is being issued to sophisticated investors.

REGULATORY DISCLOSURES

Mediobanca S.p.A. does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Mediobanca S.p.A. or its affiliates or its employees may effect transactions in the securities described herein for their own account or for the account of others, may have long or short positions with the issuer thereof, or any of its affiliates, or may perform or seek to perform securities, investment banking or other services for such issuer or its affiliates. The organisational and administrative arrangements established by Mediobanca S.p.A. for the management of conflicts of interest with respect to investment research are consistent with rules, regulations or codes applicable to the securities industry. The compensation of the analyst who prepared this report is determined exclusively by research management and senior

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Disclaimer

02 March 2015 ◆ 37

management (not including investment banking). Analyst compensation is not based on investment banking revenues, however, compensation may relate to the revenues of Mediobanca S.p.A. as a whole, of which investment banking, sales and trading are a part.

For a detailed explanation of the policies and principles implemented by Mediobanca S.p.A. to guarantee the integrity and independence of researches prepared by Mediobanca's analysts, please refer to the research policy which can be found at the following link: http://www.mediobanca.it/static/upload/b5d/b5d01c423f1f84fffea37bd41ccf7d74.pdf

Unless otherwise stated in the text of the research report, target prices are based on either a discounted cash flow valuation and/or comparison of valuation ratios with companies seen by the analyst as comparable or a combination of the two methods. The result of this fundamental valuation is adjusted to reflect the analyst's views on the likely course of investor sentiment. Whichever valuation method is used there is a significant risk that the target price will not be achieved within the expected timeframe. Risk factors include unforeseen changes in competitive pressures or in the level of demand for the company's products. Such demand variations may result from changes in technology, in the overall level of economic activity or, in some cases, from changes in social values. Valuations may also be affected by changes in taxation, in exchange rates and, in certain industries, in regulations. All prices are market close prices unless differently specified.

Since 1 July 2013, Mediobanca uses a relative rating system, based on the following judgements: Outperform, Neutral, Underperform and Not Rated.

Outperform (O). The stock's total return is expected to exceed the average total return of the analyst's industry (or industry team's) coverage universe, on a risk-adjusted basis, over the next 6-12 months.

Neutral (N). The stock's total return is expected to be in line with the average total return of the analyst's industry (or industry team's) coverage universe, on a risk-adjusted basis, over the next 6-12 months.

Underperform (U). The stock's total return is expected to be below the average total return of the analyst's industry (or industry (team's) coverage universe, on a risk-adjusted basis, over the next 6-12 months.

Not Rated (NR). Currently the analyst does not have adequate confidence about the stock's total return relative to the average total return of the analyst's industry (or industry team's) coverage, on a risk-adjusted basis, over the next 6-12 months. Alternatively, it is applicable pursuant to Mediobanca policy in circumstances when Mediobanca is acting in any advisory capacity in a strategic transaction involving this company or when the company is the target of a tender offer.

Our recommendation relies upon the expected relative performance of the stock considered versus its benchmark. Such an expected relative performance relies upon a valuation process that is based on the analysis of the company's business model / competitive positioning / financial forecasts. The company's valuation could change in the future as a consequence of a modification of the mentioned items.

Please consider that the above rating system also drives the portfolio selections of the Mediobanca's analysts as follows: long positions can only apply to stocks rated Outperform and Neutral; short positions can only apply to stocks rated Underperform and Neutral; portfolios selection cannot refer to Not Rated stocks; Mediobanca portfolios might follow different time horizons.

Proportion of all recommendations relating to the last quarter:

Outperform Neutral Underperform Not Rated

51.72% 43.94% 3.43% 0.92%

Proportion of issuers to which Mediobanca S.p.A. has supplied material investment banking services relating to the last quarter:

Outperform Neutral Underperform Not Rated

12.50% 12.86% 14.29% 33.33%

The current stock ratings system has been used since 1 July 2013. Before then, Mediobanca S.p.A. used a different system, based on the following ratings: outperform, neutral, underperform, under review, not rated. For additional details about the old ratings system, please access research reports dated before 1 July 2013 from the restricted part of the "MB Securities" section of the Mediobanca S.p.A. website at www.mediobanca.com.

COMPANY SPECIFIC REGULATORY DISCLOSURES

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

Disclaimer

02 March 2015 ◆ 38

MARKET MAKER Mediobanca S.p.A. is currently acting as market maker on equity instruments, or derivatives whose underlying financial instruments are materially represented by equity instruments, issued by CAIXABANK.

RATING In the past 12 months, the rating on CAIXABANK has been changed. The previous rating, issued on 10/10/2012, was UNDERPERFORM. The present rating in regard to CAIXABANK has not been changed since 23/10/2014.

INITIAL COVERAGE CAIXABANK initial coverage as of 10/10/2012.

COPYRIGHT NOTICE

No part of the content of any research material may be copied, forwarded or duplicated in any form or by any means without the prior consent of Mediobanca S.p.A., and Mediobanca S.p.A. accepts no liability whatsoever for the actions of third parties in this respect.

END NOTES

The disclosures contained in research reports produced by Mediobanca S.p.A. shall be governed by and construed in accordance with Italian law.

Additional information is available upon request.

Un

auth

ori

zed

red

istr

ibu

tio

n o

f th

is r

epo

rt is

pro

hib

ited

.T

his

rep

ort

is in

ten

ded

fo

r A

nd

rea.

Filt

ri@

med

iob

anca

.co

m f

rom

MB

.An

dre

a.F

iltri

@m

edio

ban

ca.c

om

MEDIOBANCA – Banca di Credito Finanziario S.p.A. Piazzetta Enrico Cuccia, 1 - 20121 Milano - T. +39 02 8829.1 33 Grosvenor Place – London SW1X 7HY – T. +44 (0) 203 0369 530

Mediobanca S.p.A. Antonio Guglielmi

Head of European Equity Research +44 203 0369 570

[email protected] ANALYSTS

European Banks

Alain Tchibozo France/IBK +44 203 0369 573 [email protected]

Adam Terelak France/IBK +44 203 0369 574 [email protected]

Andrea Filtri Spain/Italy +44 203 0369 571 [email protected]

Andreas Williams Spain +44 203 0369 577 [email protected]

Riccardo Rovere Italy/Scandinavia/CEE/Germany +39 02 8829 604 [email protected]

European Insurance Gianluca Ferrari Italy and Reinsurance +39 02 8829 482 [email protected]

Simonetta Chiriotti Nordics +39 02 8829 933 [email protected]

Italian Research Alessandro Tortora Building Materials/Industrials/Capital Goods +39 02 8829 673 [email protected]

Andrea Scauri Oil & Oil Related/Capital Goods +39 02 8829 496 [email protected]

Chiara Rotelli Branded Goods/Consumers Goods +39 02 8829 931 [email protected]

Fabio Pavan Media/Telecommunications/Consumer Goods +39 02 8829 633 [email protected]

Javier Suárez Utilities +39 028829 036 [email protected]

Massimo Vecchio Auto & Auto Components/Industrials/Holdings +39 02 8829 541 [email protected]

Niccolò Storer Auto & Auto Components/Industrials/Holdings +39 02 8829 444 [email protected]

Nicolò Pessina Consumer Goods/Infrastructure +39 02 8829 796 [email protected]

Simonetta Chiriotti Real Estate/ Industrials +39 02 8829 933 [email protected]

FOR NON US PERSON receiving this document and wishing to effect transactions in any securities discussed herein, please contact:

Mediobanca S.p.A. Charlotte Roden

Head of Equity Sales +44 203 0369 537

[email protected] SALES

Angelo Vietri

+39 02 8829 989 [email protected]

Christopher Seidenfaden


Lorenzo Angeloni


Timothy Pedroni


Stephane Langlois


European Spec Sales

Gaelle Jarrousse Banks +44 203 0369 530 [email protected]

Carlo Pirri Banks +44 203 0369 531 [email protected]

Gert-Jaap Kraan Insurance +44 203 0369 510 [email protected]

Mediobanca S.p.A. Dominic Bidwell

Head of Equity Trading and Sales Trading +44 203 0369 627

[email protected] SALES/TRADERS

Alessandro Gobbi


Matteo Agrati


Michael Sherry


Roberto Riboldi +39 02 8829 639 [email protected]

FOR US PERSON receiving this document and wishing to effect transactions in any securities discussed herein, please contact:

Mediobanca Securities USA LLC Pierluigi Gastone

Head of Mediobanca Securities USA LLC +1 212 991 4745

[email protected]

Massimiliano Pula


Robert Perez


mailto:[email protected]






























Documents

2015_03_02_CABK Investor Day Cheat Sheet