19TH JANUARY 2010

is likely to surprise on the upside and we foresee the yield is likely to surprise on the upside time. curve remaining steep for a prolonged capitalised, with the three Nordic banks having capitalised, with the three Nordic considerable excess capital. as a whole. Economic growth We are positive on the group The Basel Committee proposals (which we tentatively call The Basel Committee proposals changer for the banks. ‘Basel III’) are likely to be a game We see Lloyds and HSBC’s This is a key part of our report. to 4.4% and 6.0% Core Tier 1 ratios falling substantially, as adequately respectively. We see other banks We initiate on the 10 largest European ‘lending’ banks: the 10 largest European ‘lending’ We initiate on Lloyds, Handelsbanken, HSBC, Intesa, BBVA, DNBNOR, Unicredit. Standard Chartered and Nordea, Santander, • • • THE BASEL III GAME CHANGER THE BASEL III Research Andrew Lim +44 20 3206 7347 [email protected] PAN-EUROPEAN BANKS

CORPORATE CAPITAL THE BASEL III GAME CHANGER

We believe that changes in bank capital regulations are going to be a ‘game changer’ for the sector. In our initiation report on large cap European lending banks, we undertake a detailed analysis of what would happen to banks’ capital ratios. We reach some startling conclusions, most notably that Lloyds’ Core Tier 1 ratio falls to 4.4% and HSBC’s falls to 6.0%. Other banks are seen to be adequately capitalised, with Nordic banks in particular having excess capital.

• We initiate coverage on the large cap European • In our Sector and Economic Overview, we retail and corporate lending banks (which we broadly conclude that the lending banks are operating in a term ‘lending’ banks) for whom the vast proportion of very attractive economic environment which is likely operating earnings is derived from plain-vanilla to last until at least Q4 2010. We show, looking at lending activity (i.e. it is mainly net interest income, past recessions, that economic growth is likely to be it from retail or wholesale lending). This group surprise on the upside. We believe the yield curve comprises the Spanish banks BBVA and Santander, will remain steep for a prolonged period, with the the Italian banks Intesa Sanpaolo and Unicredit, the short end depressed by the magnitude of the output Nordic banks DNBNOR, Handelsbanken and gap and the long end remaining at a high level as Nordea and the UK banks HSBC, Lloyds, and governments seek to attract investors to the Standard Chartered. Our analysis of this group is significant volume of new debt issuance. split into three key components: Basel III Capital • We rate as BUY: DNBNOR (Recovering Norwegian Analysis, Sector and Economic Overview and economy, normalizing loan losses, very strong Basel lastly, Dupont Analysis of Earnings. III capitalization, cheap); Handelsbanken (Best • Basel III Capital Analysis. We believe that changes quality bank, to be appreciated more when the in bank capital regulations are going to be a ‘game extent of Basel III excess capital is realized); Nordea changer’ for the sector. The proposals from the (Solid Basel III capitalization, well run bank with Basel Committee, published in December 2009, will good earnings momentum, cheap). increase the quantum of capital in the system, • We rate as HOLD: BBVA (Very high ROE, improve its quality, force out complexity from attractive Latam footprint, near term concerns balance sheets and ultimately drive down ROE. We regarding Spanish loan losses); HSBC (Deposit undertake a detailed analysis of what happens to the surplus is a strong competitive advantage, attractive banks’ capital in an attempt to replicate as closely as Asian footprint, weak Basel III capitalization); Intesa possible the anticipated findings of the Basel Sanpaolo (Well run, defensive bank, but with a Committee’s own impact study, due H2 2010. The structurally low ROE); Santander (Well managed for results are very interesting. The UK banks Lloyds growth and stability, but near term concerns on and HSBC are significantly impacted by the Spanish loan losses); Standard Chartered proposals. We see Lloyds, in particular, having a (Attractive Asian footprint, large deposit surplus, not new Core Tier ratio of only 4.4% by the end of 2012. expensive but not cheap either). We also see HSBC’s Core Tier 1 ratio falling to 6.0%, significantly below peers. The other banks are • We rate as REDUCE: Lloyds (Large Basel III capital reasonably capitalised, with the Nordic banks in shortfall, earnings growth constrained by LTD ratio particular having substantial excess capital. of 170%, expensive); Unicredit (Structurally low ROE bank, not cheap).

Research Andrew Lim +44 20 3206 7347 [email protected]

Matrix Corporate Capital LLP is authorised and regulated in the United Kingdom by the Financial Services Authority. This document must be treated as a marketing communication as it has not been prepared in accordance with legal requirements designed to promote the independence of investment research.

19 JANUARY 2010

PAN-EUROPEAN BANKS

CONTENTS

Summary 3

Stock Ratings 5

Sector and Economic Overview 8

Basel III Capital Analysis 19

Dupont Methodology For Banks 36

Dupont Analysis of Earnings 39

Valuation 65

Company Summaries 67

BBVA 68

DNBNOR 70

Handelsbanken 72

HSBC 74

Intesa Sanpaolo 76

Lloyds 78

Nordea 80

Santander 82

Standard Chartered 84

Unicredit 86

Appendix 88

19 JANUARY 2010 2

PAN-EUROPEAN BANKS

SUMMARY

We initiate on the large cap European retail and corporate lending banks (which we broadly term ‘lending’ banks) where the vast proportion of the earnings stream is derived from plain-vanilla lending activity (i.e. it is mainly net interest income, be it from retail or wholesale lending). This group comprises the Spanish banks BBVA and Santander, the Italian banks Intesa Sanpaolo and Unicredit, the Nordic banks Danske Bank, DNBNOR, and Handelsbanken, and the UK banks HSBC, Lloyds, and Standard Chartered. Our analysis of this group is split into three key components: Basel III Capital Analysis, Sector and Economic Overview and lastly, Dupont Analysis of Earnings. Overall, we are BUYERS of lending banks, as per our Sector and Economic Overview.

BASEL III CAPITAL ANALYSIS – Page 19 We believe that changes in regulations for bank capital are a ‘game changer’ for the sector. The proposals from the Basel Committee published in December 2009 will increase the quantum of capital in the system, improve its quality, force out complexity from balance sheets and ultimately drive down ROE. We undertake a thorough analysis of what happens to the banks’ capital in an attempt to replicate as closely as possible the anticipated findings of the Basel Committee’s own impact study, due H2 2010. The results are very interesting.

The UK banks Lloyds and HSBC are significantly impacted by the proposals. We see Lloyds, in particular, having a new Core Tier ratio of only 4.4% by the end of 2012. The fall is mainly due to the full deduction from common equity of investments in other financial institutions of £10bn (mainly insurance), which under the present FSA transition rules, is only deducted at the total capital level (not 50:50 from Tier 1 and Tier 2 capital as for most other banks). Basel III is essentially crystallising the problem that Lloyds has been able to get away with for years of double counting the capital in other financial entities on the group balance sheet.

We also see HSBC’s Core Tier 1 ratio falling to 6.0%, significantly below peers and in line with what we would deem to be an appropriate regulatory minimum. The reasons for the substantial fall in the Core Tier 1 ratio are more varied than for Lloyds and arise mainly from the deduction of negative AFS reserves, the deduction of minorities, the increase in market risk weights and the deduction of investments in other financial entities. For the last point, HSBC has, like Lloyds, taken advantage of the FSA transition rules currently in force and opted to deduct investments in financial entities at the total capital level rather than 50:50 from Tier 1 and Tier 2 capital. We believe that management may ultimately desire to raise more common equity to obtain a capital buffer and regain parity with peers.

The other banks are reasonably above the 6.0% level that we would deem a minimum, but some more so than others. The Italian and Spanish banks, together with Standard Chartered, have ratios 1–2% higher than the 6% level. The Nordic banks, however, have substantial excess capital. We see DNBNOR and Handelsbanken in particular as having Core Tier 1 ratios roughly 4% above the minimum, with Nordea about 3% higher.

SECTOR and ECONOMIC OVERVIEW – Page 9 Our view of the economy is closely intertwined with our view of how lending banks will perform. The conclusion is that we should be BUYERS of lending banks, based on the following:

1. There is a strong historical relationship between the severity of a recession and the strength of the rebound. Given the severity of the recession just

19 JANUARY 2010 3

PAN-EUROPEAN BANKS

past, the strength of the economic rebound is very likely to have been under-estimated by consensus.

2. Looking at previous recessions, the economic recovery plays the largest part in restoring health to the banking sector, once the necessary actions to stabilise the financial sector have been undertaken (usually with substantial government intervention and public money). Better than expected economic growth (as per point 1) should therefore translate into a better than expected improvement in credit quality and robust loan growth for the banking system.

3. There is scepticism that there is enough liquidity in the system to fund an increase in demand for credit as the economy recovers. Looking at the enormous rise in bank reserves placed with central banks, it would seem this scepticism is misplaced. The US Federal Reserve now has approximately $1tn of reserves in excess of the required minimum, whilst the BOE has reserves seven times the level that prevailed before the crisis. While liquidity requirements for banks are going to increase under new regulatory proposals, liquidity seems more than ample to both meet these requirements and fund future demand for credit.

4. The steepness of the yield curve is likely to be maintained for the foreseeable future, which is an ideal situation for lending banks. We anticipate the short end being kept low by virtue of the large output gap maintaining significant deflationary pressures on the economy (see point 5), while the long end remains high (or even increases) as governments seek to attract demand for their enormous issuance of new paper to fund public deficits.

5. The US output gap (an economic measure of the difference between the actual output of an economy and the output it could achieve when it is most efficient) has been estimated at a massive 8% of GDP as of mid 2009. The conclusion is that even with an above-trend real GDP growth of 4%, it would take until 2014 for the output gap to be eliminated and for the US unemployment rate to decline back to 5%. This strong deflationary pressure means that it would be a big policy mistake to raise base rates. Indeed, we believe base rates in the US, UK and Eurozone will remain unchanged until at least Q4 2010.

DUPONT ANALYSIS of EARNINGS – Page 39 We present the findings of our Dupont analysis of the lending banks in two parts. We firstly analyse the ROE composition, as estimated at the end of 2009, (note that all the banks under consideration have December year-ends, so there is periodic consistency in the comparison). This allows us to compare the quality of earnings, with higher quality banks deriving the bulk of their revenues from NII and incurring lower costs and LLCs as a % of assets.

We then present a trend analysis of the different Dupont components of the banks’ ROE. This is in the form of historical time series charts, showing the relative performance of each component of the ROE on a quarterly basis since the beginning of 2006. We gain some insightful observations using this analysis, particularly with respect to historical and future expected trends in the NIM in conjunction with the development of the loan to deposit ratio.

The use of the Dupont structure for analysing banks provides us with a more robust framework with which to determine relative earnings development. The Dupont structure is used in conjunction with our earnings models and DCF valuation methodology so that our overall view of the group is internally consistent.

19 JANUARY 2010 4

PAN-EUROPEAN BANKS

STOCK RATINGS

We use the following inputs to arrive at our individual stock ratings for the large cap lending banks:

• Basel III Analysis of capital and leverage ratios • Analysis of relative earnings performance using a Dupont system • Discounted Cash Flow (DCF) valuation methodology The rating for an individual bank is given with respect to how we view the stock price will perform in relation to the peer group, given the above.

BUY DNBNOR – An appealing ROE of 10% should gradually increase as the Norwegian economy continues to improve and loan losses normalise at a lower level (particularly in its corporate business). High loan losses at its DNB NORD subsidiary (which includes its Eastern European operations) are an ongoing issue, but manageable in our view. The bank has substantial excess capital under the Basel III framework and is also very cheap.

Handelsbanken – In our view, this is the best quality bank in the group. It is very low risk, as shown by the marginal deterioration in asset quality over the course of the crisis, and it is the most efficient operator. We see substantial excess capital under the Basel III framework. Our Dupont Analysis shows that the bank has capitalised on its superior franchise by gathering cheap deposits, lowering its LTD ratio and improving its NIM.

Nordea – Nordea completes our trio of Buys on the Nordic banks. It is an efficient bank with attractive ROE and significant excess capital under the Basel III framework. We believe loan losses from its small Eastern European operations are easily manageable and more than outweighed by the expected recovery in the Nordic region.

HOLD BBVA – The high structural ROE is underpinned by attractive Latam footprint. However, poor earnings momentum is expected in 2010, driven by elevated Spanish loan losses, combined with a low coverage ratio and an absence of positive one-offs. Capital is seen to be adequate (but not excessive) under our Basel III Analysis. Our DCF model points to substantial upside, but we have a cautious price target given near term concerns.

HSBC – We see HSBC as a ‘super deposit franchise’, with a LTD ratio now at only 80%. The excess deposits will be a significant relative advantage in funding future loan growth. Asset quality is on the cusp of improving, underpinning strong earnings momentum. However, capitalisation looks weak under our Basel III Analysis, with the Core Tier 1 ratio only just meeting minimum requirements at the end of 2012. Our DCF model points to substantial upside, but accounting also for the results of our Basel III Analysis, we set only a modest price target.

Intesa Sanpaolo – A low ROE bank, which will remain so given that domestic Italian peers (such as Unicredit and smaller domestic banks), remain competitive in the deposit and lending markets. Intesa is a well run bank with good asset quality, loan % deposit ratio just below 100% and adequate CT1 ratio as per our Basel III Analysis. However, we believe the ROE is structurally low and that growth prospects are meagre. Our DCF model points to some attractive valuation upside, but it is not as much as for our favoured banks.

19 JANUARY 2010 5

PAN-EUROPEAN BANKS

Santander – Santander has superior growth opportunities from its Latam operations (mainly Brazil). We worry near term about the continued deterioration in the Spanish economy and the fact that loan losses will remain elevated deep into 2010. Capitalisation under Basel III is adequate. We see some valuation upside at current levels, but the Nordic banks appear more attractive.

Standard Chartered – The attractions of Standard Chartered are clear. The focus on Asia has led to superior growth prospects and low loan losses during the credit crisis. Like HSBC, we also see Standard Chartered as a ‘super deposit franchise’ with a LTD ratio of only 80%. Cheap excess deposits will be a significant advantage in funding future loan growth. The bank has adequate capitalisation under our Basel III Analysis. However, while the bank is not expensive, neither do we see it as cheap. Our DCF model points to the bank being fairly valued.

REDUCE Lloyds – Lloyds is the bank where we have the most concerns. Capitalisation under Basel III is very weak. Transition rules would give it time to fix the situation, but not a reprieve from the need to raise more equity in our opinion. We also see earnings growth and margins as under threat from the structural need to reduce the loan book in relation to customer deposits (note that the LTD ratio is ~170%). Our DCF model points to absolute downside. Conventional P/E metrics show the bank being the most expensive in the group out to 2011.

Unicredit – A very low ROE bank, which will remain so given the continued competitive robustness of Italian banks in deposits and lending (meaning structural margins are low), with no near-term recovery in CEE loan losses expected. Capitalisation is adequate under our Basel III Analysis. Our DCF model points to absolute downside and P/E ratios look unattractive. The bank trades at a low 2010e P/TNAV of 1.1x, but this is expensive given that the low single digit ROE is substantially lower than the COE.

Figure 1: DCF Valuation Table

Cost of Equity Assumptions DCF FV Risk- LT Market LT Terminal Matrix Mcap Cur Price DCF DCF upside/ free market risk Equity growth COE value % Stock rating (€bn) (local) FV P/FV downside rate return premium Beta* rate FV BBVA HOLD 48.3 EUR 12.81 16.00 80% 25% 4.02% 8.50% 4.48% 1.47 5.00% 10.60% 35% DNB NOR ASA BUY 13.9 NOK 69.50 85.74 81% 23% 4.02% 8.50% 4.48% 1.24 5.00% 9.56% 43% HSBC HLDGS PLC HOLD 139.2 GBp 703 915 77% 30% 4.02% 8.50% 4.48% 1.15 5.00% 9.15% 46% INTESA SANPAOLO HOLD 39.2 EUR 3.12 3.57 87% 14% 4.02% 8.50% 4.48% 1.29 5.00% 9.81% 41% LLOYDS BANKING REDUCE 41.9 GBp 56.78 49.06 116% -14% 4.02% 8.50% 4.48% 1.66 5.00% 11.45% 32% NORDEA BANK AB BUY 29.5 SEK 73.80 85.66 86% 16% 4.02% 8.50% 4.48% 1.27 5.00% 9.71% 41% BANCO SANTANDER HOLD 95.3 EUR 11.50 12.84 90% 12% 4.02% 8.50% 4.48% 1.48 5.00% 10.63% 35% SVENSKA HAN-A BUY 12.3 SEK 199.50 245.70 81% 23% 4.02% 8.50% 4.48% 1.14 5.00% 9.12% 47% STANDARD CHARTER HOLD 35.9 GBp 1552 1484 105% -4% 4.01% 8.50% 4.49% 1.55 5.00% 10.99% 32% UNICREDIT SPA REDUCE 43.3 EUR 2.25 2.00 112% -11% 4.02% 8.50% 4.48% 1.68 5.00% 11.57% 30%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 6

PAN-EUROPEAN BANKS

Figure 2: Conventional Valuation Table – Matrix Estimates Matrix Mcap Curr Price Target Upside/ Matrix EPS Matrix P/E Matrix BVPS Matrix P/BV Matrix ROE Stock Rating (€bn) (local) Price d’nside FY09E FY10E FY11E FY09E FY10E FY11E FY10E FY11E FY10E FY11E FY10E FY11E BBVA HOLD 48.3 EUR 12.81 14.25 11% 1.35 1.22 1.46 9.47 10.51 8.80 8.58 9.45 1.49 1.36 14.7% 16.0% DNB NOR ASA BUY 13.9 NOK 69.50 86.00 24% 6.10 5.47 6.92 11.39 12.70 10.04 62.87 67.20 1.11 1.03 9.0% 10.6% HSBC HLDGS PLC HOLD 139.2 GBp 703 800 14% 0.483 0.697 1.028 23.74 16.45 11.16 7.39 8.01 1.55 1.43 9.9% 13.6% INTESA SANPAOLO HOLD 39.2 EUR 3.12 3.55 14% 0.249 0.231 0.286 12.53 13.47 10.89 4.69 4.93 0.66 0.63 5.4% 6.4% LLOYDS BANKING REDUCE 41.9 GBp 56.78 49.00 -14% -10.31 -0.45 4.24 -5.51 -126.21 13.41 63.16 68.00 0.90 0.83 -0.7% 6.5% NORDEA BANK AB BUY 29.5 SEK 73.80 86.60 17% 0.608 0.654 0.783 11.99 11.14 9.31 5.44 5.90 1.34 1.24 12.5% 13.8% BANCO SANTANDER HOLD 95.3 EUR 11.50 13.00 13% 1.06 1.02 1.31 10.82 11.28 8.77 8.93 9.72 1.29 1.18 11.8% 14.1% SVENSKA HAN-A BUY 12.3 SEK 199.50 250.00 25% 16.46 17.19 19.68 12.12 11.60 10.13 139.16 152.46 1.43 1.31 12.9% 13.5% STANDARD HOLD 35.9 GBp 1552 1500 -3% 1.91 2.25 2.71 13.26 11.27 9.36 15.60 18.41 1.62 1.38 17.2% 17.5% CHARTER UNICREDIT SPA REDUCE 43.3 EUR 2.25 2.00 -11% 0.098 0.093 0.181 22.92 24.08 12.37 3.48 3.66 0.64 0.61 2.7% 5.1% AVERAGE 9% 12.3 -0.4 10.4 1.201.10 9.5% 11.7%

Source: Matrix Corporate Capital estimates

Figure 3: Valuation Table – Consensus Estimates Matrix Mcap Curr Price Target Upside/ Cons. EPS Cons. P/E Cons. BVPS Cons. P/BV Cons. ROE Stock Rating (€bn) (local) Price d’nside FY09E FY10E FY11E FY09E FY10E FY11E FY10E FY11E FY10E FY11E FY10E FY11E BBVA HOLD 48.3 EUR 12.81 14.25 11% 1.36 1.30 1.52 9.41 9.84 8.41 8.90 9.87 1.44 1.30 15.8% 16.8% DNB NOR ASA BUY 13.9 NOK 69.50 86.00 24% 5.58 5.22 7.12 12.46 13.31 9.76 65.10 68.74 1.07 1.01 8.1% 10.4% HSBC HLDGS PLC HOLD 139.2 GBp 703 800 14% 0.490 0.674 0.946 23.41 17.02 12.12 7.27 7.83 1.58 1.47 10.3% 13.7% INTESA SANPAOLO HOLD 39.2 EUR 3.12 3.55 14% 0.210 0.236 0.328 14.85 13.21 9.50 4.23 4.42 0.74 0.71 6.2% 8.3% LLOYDS BANKING REDUCE 41.9 GBp 56.78 49.00 -14% -12.60 -2.40 4.60 -4.51 -23.66 12.34 63.60 67.60 0.89 0.84 -3.9% 8.9% NORDEA BANK AB BUY 29.5 SEK 73.80 86.60 17% 0.604 0.457 0.596 12.07 15.95 12.23 5.78 6.19 12.78 11.92 8.4% 10.5% BANCO SANTANDER HOLD 95.3 EUR 11.50 13.00 13% 1.05 1.09 1.34 10.94 10.60 8.61 8.81 9.58 1.31 1.20 14.1% 15.7% SVENSKA HAN-A BUY 12.3 SEK 199.50 250.00 25% 15.87 14.79 17.90 12.57 13.49 11.14 138.88 149.56 1.44 1.33 11.2% 12.8% STANDARD HOLD 35.9 GBp 1552 1500 -3% 1.68 1.88 2.25 15.09 13.48 11.26 13.38 14.75 1.89 1.72 15.3% 16.2% CHARTER UNICREDIT SPA REDUCE 43.3 EUR 2.25 2.00 -11% 0.102 0.132 0.256 22.01 17.01 8.77 3.20 3.34 0.70 0.67 4.2% 7.6% AVERAGE 9% 12.8 10.0 10.4 2.38 2.229.0% 12.1%

Source: Bloomberg Consensus Estimates

19 JANUARY 2010 7

PAN-EUROPEAN BANKS

SECTOR AND ECONOMIC OVERVIEW

The relationship between the banking sector and the economy has become deeper and more complex over recent years. The time when close followers of the sector had to learn a new acronym for the multitude of complex credit assets that the banks had invested in has thankfully come to an end, but careful consideration still needs to be given to other complicated issues that are likely to stay with us for many years, such as quantitative easing

The insightful economic overview provided below has been adapted from Matrix research written by Bechara Madi, Head of Macro, Matrix Money Management. Our view of the economy is closely intertwined with our view of how lending banks will perform. The conclusion is that we should be BUYERS of lending banks, based on the following:

1. There is a strong historical relationship between the severity of a recession and the strength of the rebound. Given the severity of the recession just past, the strength of the economic rebound is very likely to have been under-estimated by consensus.

2. Looking at previous recessions, the economic recovery plays the largest part in restoring health to the banking sector, once the necessary actions to stabilise the financial sector have been undertaken (usually with substantial government intervention and public money). Better than expected economic growth (as per point 1) should therefore translate into a better than expected improvement in credit quality and robust loan growth for the banking system.

3. There is scepticism that there is enough liquidity in the system to fund an increase in demand for credit as the economy recovers. Looking at the enormous rise in bank reserves placed with central banks, it would seem this scepticism is misplaced. The US Federal Reserve now has approximately $1tn of reserves in excess of the required minimum, whilst the BOE has reserves 7 times the level that prevailed before the crisis. While liquidity requirements for banks are going to increase under new regulatory proposals, liquidity seems more than ample to both meet these requirements and fund future demand for credit.

4. The steepness of the yield curve is likely to be maintained for the foreseeable future, which is an ideal situation for lending banks. We anticipate the short end being kept low by virtue of the large output gap maintaining significant deflationary pressures on the economy (see point 5), while the long end remains high (or even increases) as governments seek to attract demand for their enormous issuance of new paper to fund public deficits.

5. The US output gap (an economic measure of the difference between the actual output of an economy and the output it could achieve when it is most efficient) has been estimated at a massive 8% of GDP as of mid 2009. The conclusion is that even with an above-trend real GDP growth of 4%, it would take until 2014 for the output gap to be eliminated and for the US unemployment rate to decline back to 5%. This strong deflationary pressure means that it would be a big policy mistake to raise base rates. Indeed, we believe base rates in the US, UK and Eurozone will remain unchanged until at least Q4 2010.

19 JANUARY 2010 8

PAN-EUROPEAN BANKS

Sector and Economic Overview

Economic Growth Likely to Exceed Expectations For a year now, our forecast envisaged an end to the global recession around mid- year 2009. Whilst this was far from consensus for most of the year, it is now widely accepted that a global recovery is taking hold and spreading. As was the case in previous turnarounds, the early stages of the recovery have been driven by a mix of factors, namely the support from policy stimulus, the turnaround in the inventory cycle and especially important for exporting economies, the revival in world trade. The overwhelming message from the economic data is that the growth recovery in many economies is likely to exceed expectations over the next few quarters.

Despite the evidence, a lot of scepticism and uncertainty remains and we currently stand at a point where the range of opinions on likely economic and market outcomes is very wide. Broadly speaking, this range can be split into three camps.

• The first camp still doubts that this seemingly healthy recovery is sustainable beyond the next couple of quarters. This view is backed up by the belief that the legacy of this deep recession will linger in the form of high indebtedness, rising unemployment, and fiscal austerity which will hold back the main developed economies. It is believed that, after the inventory and stimulus cycles come to pass, growth will falter into a significant rollover (the double- dip or ‘W-shaped’) and potentially a deflationary bust. • The second view is that, following a period of relatively strong growth over the next 2–4 quarters, growth will gravitate towards trend growth or slightly above. This will be sufficient to underscore economic, financial and labour market stability, but not enough to quickly close the unemployment and output gaps. This implies that strong disinflationary forces will remain in place for many years to come. This scenario includes the forecast of ‘low interest rates for longer’. • The third possible outcome is that growth will continue to accelerate to well above trend levels, driven by a new financial asset price bubble fuelled by a generous and irresponsible liquidity glut. This scenario includes an acceleration in inflation, which will lead eventually to a rapid reversal in monetary policy and hence increase the probability of a double-dip. Our long-held base-case forecast for G4 economies (i.e. US, Eurozone, Japan and UK) falls into the second camp, however we acknowledge that there is no absolute certainty. We are simply taking a probabilistic view where we would allocate an 80% probability to our base case scenario while we would view the alternative forecasts as tail scenarios. That said, we cannot think of a precedent when two tail outcomes – an inflationary spike or deflationary bust – could be simultaneously argued for with such great plausibility. As for peripheral developed economies and emerging markets, the forecast is skewed to the upside with the probability of strong growth with higher inflation at 40%.

In support of our base case scenario for G4, and de facto justifying the low probability of the alternative scenarios, we are relying on the following historical facts:

Fact 1: Bubbles and busts are all but unusual. Since 1870, the IMF counts eight major international crises, which means a crisis every 10–15 years. Since 1970, there have been 47 housing-led bubbles/busts and 98 equities-led. Although each bust had idiosyncratic characteristics in terms of causes, depth, and length, there is one shared characteristic: recessions are always followed by recoveries (apologies for stating the obvious).

19 JANUARY 2010 9

PAN-EUROPEAN BANKS

Fact 2: There is a robust historical relationship in the US between the severity of a recession and the strength of a rebound. This relationship is known as the Zarnowitz rule, which essentially says that the deeper the recession, the sharper the rebound. Figure 4 illustrates this point in the US. Comparing the current cycle to previous ones, there are of course idiosyncratic characteristics, namely that the recession was global and synchronised. However, equally unique has been the unprecedented and coordinated policy response. We believe that there is no reason for this cycle to be different. Current expectations, although revised upwards over the past few months, remain way too pessimistic by historic levels.

Figure 4: Comparison of Past US Recessions and Recoveries

16 14 1948-49 12 1953-54 10 1957-58 1960-61 1970 8 1973-75 6 1981-82 2008-09 Dec Forecast 1990-91 4 1980 2 2001 2008-09 Aug Forecast

GDP Growth in next 6 Quarters (%) Quarters 6 next in Growth GDP 0 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50

Peak to Trough GDP Decline (%) Source: Bloomberg, Matrix Money Management. Forecasts are Bloomberg consensus forecasts.

Fact 3: The ‘double-dip’ scenario has a weak historic foundation. A double-dip is defined as two recessions separated by a relatively short period (one or two quarters perhaps) of positive growth. In the US (the country that matters from a global perspective), there has been only one post-war double-dip recession and this had been caused by a sharp tightening of policy. During the cycle of 1980-1983, as the economy in late 1980 was emerging from a short recession, an upsurge in inflation to 15% led the Federal Reserve to respond with a sharp tightening in policy in the autumn of 1981. The Fed Funds rate reached 20%, and this tipped the economy tipped back into a new and longer recession. The other example which some people cite is the Great Depression of 1929 which was followed by the great recovery in 1933 and then again by a sharp recession in 1937, caused by a sharp tightening of money, credit and fiscal policy. This, being so many years later, was arguably an entirely new business cycle rather than a double dip. The point we are making here is that there is a clear recognition – in hindsight - that the policy actions in both 1981 and 1937 were mistakes, which suggests that their repetition in this current cycle is highly unlikely. Given the large output gaps in the US and most other major economies and given that core inflation rates are generally very low and/or declining, it is virtually impossible to foresee a sharp tightening of monetary policies that would induce a double-dip within the next year or two.

Fact 4: Contrary to popular belief, the ‘credit-addicted’ Anglo-Saxon consumer should not be entirely written off in this recovery. Pessimism about US consumer spending (which accounts for 70% of US GDP, and 18% of world GDP) is driven by the assumption that its growth is likely to be restrained not only by the traditional factors (loss of household wealth, high debt, rising unemployment) but also because of the expectation of an extended credit crunch. In response to these concerns, we would highlight that exceptionally easy and cheap credit has been a key phenomenon only in the past decade. Under normal conditions (and we are seeing credit conditions slowly normalising) and when the US consumer savings rate

19 JANUARY 2010 10

PAN-EUROPEAN BANKS

stabilises (whether at 7% or 10%), it would be reasonable to assume that the growth in consumer spending should converge with the natural growth in income. This is indeed a historic relationship that has been proven over decades if not centuries (Figure 5). Any other assumption, namely that the Anglo-Saxon consumer is permanently impaired, is without foundation.

Figure 5: US Personal Income and Consumption Expenditure Billion $ 12000

10000

8000

6000

4000

2000

0 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09

Personal Disposable Income Personal Consumption Expenditure

Source: Bloomberg, Matrix Money Management

Fact 5: A robust banking sector is not a pre-condition for a sustainable recovery. It is in fact the economic recovery that plays the largest part in restoring health to the banking sector. Looking at previous recessions, notwithstanding the causality, financial sector stress has almost always interacted with the general weakening of economic activity and exacerbated the downturn. In all previous cases, it was generally necessary to stabilize conditions in the financial sector, usually with substantial government intervention and public money. Once some financial stability is achieved, economic recovery proceeds and, in the reverse of the process that operates during the downturn, it is primarily the economic recovery that restores good health to the financial system. Indeed, the great recovery in the summer of 1933 was achieved with a banking and financial system that had been stabilized but was not robust. More recently, the deep US recessions of 1973–75 and 1980–82 were associated with considerable financial stress. As the recoveries began from these recessions, major financial institutions were in almost as bad if not worse condition than major financial institutions are today.

Fact 6: Contrary to popular belief, banks have more than enough liquidity to fund an increase in demand for credit. Indeed, sceptics question whether the ‘impaired’ banking system will have adequate supply of credit to meet the needs of the recovery. This scepticism is again misplaced. The essence of quantitative easing was to help banks liquefy their balance sheets, restore confidence between banks and ultimately encourage them to resume their function of providing credit to the various economic agents. This ‘liquefaction’ of bank balance sheets is evidenced by the rise in banks reserves with the central banks, mirroring the expansion of central banks balance sheets. In the US for example, as Figure 6 illustrates, banks hold with the Federal Reserve almost $1.06tn in reserves, which includes $995bn in excess reserves (i.e. over and above the reserve requirements), on which banks earn interest of 25bp. As credit markets normalise and confidence rises, it would be natural to assume that banks will slowly begin to withdraw these excess reserves and lend to qualified borrowers at interest rates higher than the current 25bp that is earned on the excessive reserves. Taking into account the money multiplier effects, these excess reserves of $995bn potentially support bank lending of many times that figure (of course subject to capital ratios). Other than a few exceptions (e.g. Iceland

19 JANUARY 2010 11

PAN-EUROPEAN BANKS

where credit is constrained), the reserve situation is more or less the same for the large developed economies of Europe and Japan. For example, the Bank of England balance sheet shows liabilities of roughly £145.9bn (as at 9 December 2009) in the form of reserves deposited by banks (Figure 7). These are seven times the pre-crisis average level of £20bn.

Figure 6: US Banks Reserves Held With The Fed

Billion USD 1200

1000

800

600

400

200 Excess Reserves

0 Jan-06 May-06 Sep-06 Jan-07 May-07 Sep-07 Jan-08 May-08 Sep-08 Jan-09 May-09 Sep-09

Total Banks Reserves Required Banks Reserves

Source: Bloomberg, Matrix Money Management

Figure 7: Banks Reserves with the Bank of England

Billion GBP 180 160 140 120 100 80 60 40 20 0 May-06 Oct-06 Mar-07 Aug-07 Jan-08 Jun-08 Nov-08 Apr-09 Sep-09

Source: Bloomberg, Matrix Money Management

All in all, for the reasons detailed above, we see no reason to doubt that the recovery will follow the historic norms. We believe that growth over the next 4–6 quarters is likely to exceed expectations and thereafter it will naturally gravitate towards potential growth. Of course, there will be legacy problems that will take years to resolve. However these by themselves are not likely to derail the recovery. The economic developments so far, as evidenced by the increase in consumption indicators even in credit-sensitive sectors, suggest that this cycle appears to be more ‘normal’ than many sceptics assume. The latest unemployment data in the US - including strong upward revisions for the previous two months – even suggest that the stabilisation of labour markets is happening much faster (typically they lag the economy recovery by 6–9 months) than many forecasters assume. It may well be

19 JANUARY 2010 12

PAN-EUROPEAN BANKS

that this stabilisation is a function of the over-aggressive US labour shedding that occurred over the last 12 months. To be sure in validating this theory however, we need to wait and see the non-farm payrolls data for the next two or three months.

Figure 8: US Non-farm Payrolls

1000

750

500

250

0

-250

-500

-750

-1000 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09 3Mth MA

Source: Bloomberg, Matrix Money Management

The stabilisation in housing is also occurring in line with previous crises. Home sales have strengthened in recent months and the inventory of unsold new homes now stands at 6.7x the monthly demand (Figure 9), down from a peak above 12x. This is an important indicator because builders historically regained pricing power at the long-term average of six months of supply.

Figure 9: US New One Family Houses - Average Price vs. Inventories (in Months of Sales)

$ Thousand Months Supply 350 14

300 12

250 10

200 8

150 6

100 4

50 2

0 0 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09

New Houses Average Price - LHS New Houses Monthly Supply - RHS

Source: Bloomberg, Matrix Money Management

In our forecast, we also expect that whilst most countries will enjoy reasonably robust and mutually reinforcing recoveries, there are idiosyncratic forces that will differentiate these recoveries. For example, although the UK has a number of characteristics that are, on the surface, similar to the US, the depth of the UK recession has been worse than the US. The UK recession hit the economy hard and across the board, leaving the country with the worst fiscal position amongst the G10. The financial sector, a very important sector in the UK economy, absorbed more than its fair share by virtue of its large international exposure (larger than the US when adjusted by GDP) but also from its domestic exposure to housing. Although

19 JANUARY 2010 13

PAN-EUROPEAN BANKS

financial stabilization has been achieved (albeit slightly behind the US on the ‘normalisation curve’), a recovery in the financial sector and a return to its former glory is a long way off. The loss of income and wealth in the financial sector will probably continue to dampen the recovery in house prices and residential investment, as well as restrain somewhat the growth of consumer spending. The uncertainty surrounding next year’s elections and the expected fiscal tightening (which is certain irrespective of the outcome of the elections) is likely to keep the UK at the bottom of the growth table. It is therefore no surprise that growth expectations, although improved in recent months, remain subdued in both relative and absolute terms. On the positive side, the sharp depreciation of Sterling (Figure 10), most notably against the USD and the EUR, should bring good news to the export sector in the form of bargain hunters in goods, services and (prime) real estate. However, this may take some time to have a significant impact on the overall economy.

Figure 10: Sterling Trade-Weighted Index

Jan 2005 = 100 120

110

100

90

80

70 Jan-07 Jun-07 Nov-07 Apr-08 Sep-08 Feb-09 Jul-09 Dec-09

Source: Bloomberg, Matrix Money Management

In other regions and countries that have typically been less affected by the financial crisis, the cyclical recovery will be more normal and, as we alluded to above, prospects are much stronger. Emerging markets in general, with some rare exceptions, are witnessing classic business cycle recoveries. We believe that most forecasters still underestimate the power of the mutual reinforcement that synchronised recoveries can generate.

As already mentioned above, the key risks to our baseline forecast would be policy errors. In this context, the greatest of these potential errors would be an early withdrawal of stimulus policies and, more specifically, a rapid reversal of QE and/or an aggressive series of rate hikes. Here there are only two countries that really matter from a global perspective, namely the US and China. With Ben Bernanke (a Great Depression expert) at the helm of the Fed, we can safely rule out any premature tightening by the Fed. Remember that the Fed’s mandate (unlike the ECB and the BoE) is defined by financial stability, price stability and full employment, in that order. In China, a monetary tightening by the People’s Bank of China (PBoC) in 2010 has effectively been over-ruled by the Chinese Politburo, which issued a statement confirming the importance of maintaining stability and continuity in macro policy for 2010.

Deflation, Not Inflation, Remains the Key Concern Through 2010 and Even 2011 Despite our expectation of strong growth over the next 12–18 months, we continue to believe that the risks of deflation remain far greater than those of inflation. This view is based on the amount of capacity that exists in the key developed markets, as

19 JANUARY 2010 14

PAN-EUROPEAN BANKS

well as the world at large. Estimates by Michael Mussa – a highly respected US economic forecaster, Senior Fellow at the Peterson Institute for International Economics, former Director of Research at the IMF and former member of the US Council of Economic Advisors – put the US (negative) output gap1 at about 8% of GDP at mid-2009. This means, according to Mr. Mussa, that even with above-trend real GDP growth of 4%, it would take until 2014 for the output gap to be eliminated and for the US unemployment rate to decline back to 5%. Therefore, real output gaps will remain large and effectively keep inflationary pressures at bay over the same time horizon. This forecast is supported by history where the experience of previous business cycles in the US, Euro zone and the UK shows core inflation continuing to decline for up to three years after the end of the recession (Figure 11).

Figure 11: US Core CPI – A Multi-Cycle Comparison

14.0%

12.0% Zero indicates recessionary trough 10.0%

8.0%

6.0%

4.0%

2.0%

0.0% -12-10-8-6-4-2024681012141618202224262830323436 1982-83 1990-92 2001-02 2008-09

Source: Bloomberg, Matrix Money Management

This benign outlook seems to be validated by the inflation markets, even in the two countries where inflation concerns have been stoked by Quantitative Easing. Five- year inflation-linked government bonds (TIPS) suggest US and UK inflation rates of 1.7% and 2.2% respectively, well within the respective inflation targets of 2.0% and 2.5%¹ respectively (Figure 12).

1 The output gap is an economic measure of the difference between the actual output of an economy and the output it could achieve when it is most efficient, or at full capacity. A positive output gap occurs when actual output is more than the full- capacity output. Negative output gap occurs when actual output is less than full- capacity output. Economic theory suggests that positive output gap will lead to inflation as production and labour costs rise, and vice versa.

19 JANUARY 2010 15

PAN-EUROPEAN BANKS

Figure 12: Five-Year Breakeven Inflation Rate¹

Percent 4.0

3.0

2.0

1.0

0.0

-1.0 Jan-08 Apr-08 Jul-08 Oct-08 Jan-09 Apr-09 Jul-09 Oct-09 US UK

Source: Bloomberg, Matrix Money Management. ¹ Defined by the nominal yield less the TIP yield

Monetary Policy: Divergence is a key Theme for 2010 Whilst the market is increasingly obsessed with the timing and mechanism of the exit from quantitative easing and the normalisation of interest rates, we continue to believe that there is very little room for rate normalisation in the G4 before Q4 2010 at the earliest for reasons detailed above. When defining the likely path of monetary policy, we believe that the decision-making over the 12-18 months is likely to remain heavily skewed against a premature withdrawal of stimulus. In simple terms, it is far easier for central banks in highly leveraged economies to deal with inflation than to deal with deflation. Hence, deflation is to be avoided at all costs, even at the risk of keeping rates too low for too long.

Asset inflation will probably emerge as a key topic of discussion in coming months. Critics of the current loose monetary policy point to cheap money as creating a bubble in risky assets. They argue that the sins of the past are being repeated, and hence the need for a tightening of policy. We disagree for two reasons: firstly, the mandate of central banks is goods and services inflation, and not asset price inflation. Secondly, a rally becomes a bubble when excessive leverage is applied and carry becomes negative. This is not the case for now, and is unlikely to be the case for many years. As long as the rally in asset prices is unleveraged, and viewed by policy-makers as virtuous and part of balance sheet healing, rather than driving consumption, it will be unlikely to influence much monetary policy in the G4.

Many Developed and Emerging economies on the periphery have been less impaired by the financial crisis and therefore are more likely to normalise their policy rates over the next 6-12 months. Australia, Norway, Korea, Israel and the Czech Republic are leading this cycle, but are likely to be followed by many others. This divergence between the monetary policies of the ‘financially-impaired’ and the ‘not- impaired’ will set the tone for currency and fixed income markets in 2010.

2010 Market Outlook We believe that the market outlook for 2010 will continue to be dominated by the economic recovery. Broadly speaking, we believe that the rationale for the ‘recovery trade’ remains intact. Although this is now a consensus trade, in the absence of anything derailing the economic recovery and with uncertainty continuing to decline, we see no reason for a halt in the ‘recovery trade’.

The trade is in fact reinforced by the policy of asset reflation. Cash is the lowest yielding asset in the world, and the main reason to hold it, which is uncertainty, is

19 JANUARY 2010 16

PAN-EUROPEAN BANKS

fading. The flow out of cash is lifting all other asset prices in the world. This will end only when G4 central banks start hiking or investors have become underweight cash. We are far from either scenario. For the reasons described above, we are not expecting the major central banks to remove the punch bowl any time soon. Even when policy is tightened, the fact remains that monetary policy has simply been a catalyst to encourage private money (excluding financial institutions) into riskier assets (by making cash expensive to hold) rather than providing the liquidity itself. It will take significant rate hikes to make cash a competitive asset class again.

Bond Markets: The divergence of policy rates between G4 and peripheral markets has key implications for the respective bond markets. In the peripheral markets, the expected tightening of policy suggests a bear market with a flattening of the curve. In the G4, for the reasons detailed above, we strongly believe that rate hikes are unlikely before Q4 2010 and we therefore disagree with the current market pricing (Figure 13). If true, with short-term interest rates anchored by official rates, yield curves should remain steep as the medium-term uncertainty on inflation validates a risk premium. Furthermore, the expected issuance to fund the widening public deficits should lead to additional upward pressures on long-term yields. This is more so in the absence of additional Quantitative Easing to mop up some of the new issuance. In short, with uncertainty regarding the global crisis easing, G4 bond yields have to rise much more to improve their attractiveness relative to higher yielding assets.

Figure 13: US & UK Rate Expectations - Implied from Futures Contracts as at 22 December 2009

% 5.0

4.0

3.0

2.0

1.0

0.0 Current Apr-10 Oct-10 Apr-11 Oct-11 Apr-12 Oct-12 Apr-13 Oct-13 Jan-11 Jan-12 Jan-13 Jul-10 Jul-11 Jul-12 Jul-13

US Fed Funds UK 90-Day Short Sterling

Source: Bloomberg, Matrix Money Management.

Equities: Once again, equity markets correctly led the economic recovery. History, as gauged by the last six US recoveries, shows that equities rally reliably during the last 3 months of a recession and the first 3–4 months of a recovery. The equity rally since 9 March 2009 has fitted this pattern perfectly. Given our economic forecast and given that equities have been the most visible and most common expression of the ‘recovery trade’, we see no reason for a halt in the equity rally. This view is further supported by expectations of continued earnings growth in 2010 (although we are not pinning a lot of hope on a pick-up in dividends any time soon). Admittedly, the risk-reward in equities is less attractive than it was six months ago; however the remaining uncertainties in the market are strong enough to hold back most investors from moving down the liquidity curve into cheaper but less liquid assets. Therefore, we expect equities to remain a favourite asset class to express the ‘recovery trade’, hence widening the valuations relative to the less liquid asset classes.

19 JANUARY 2010 17

PAN-EUROPEAN BANKS

Credit: We prefer Credit over Equities in terms of exposure to the corporate sector. Firstly, Credit offers defensive characteristics against market corrections in risky assets. Secondly, Credit benefits from a positive carry in the form of a coupon (unlike equities where the dearth of dividends is likely to continue), in addition to capital appreciation. Last but not least, Credit should benefit from flows from institutional investors – insurers and banks – which are being forced to move along the risk curve in search for yield and carry. That said, the Dubai crisis was a timely reminder that the legacy of the global credit crunch could remain evident in some highly-leveraged entities. Hence, the need to be more selective within the Credit asset class.

Emerging Markets: We believe EM to be one of the best asset classes to express the ‘recovery trade’. Most Emerging Markets are exiting the crisis relatively unscathed and structurally and financially sounder than the main developed markets. This suggests medium-term outperformance of EM equities, credit and FX, although an underperformance in local sovereign bonds as monetary policy is tightened early in response to the cyclical recovery. The need to be selective and differentiate within EM was also highlighted by the Dubai crisis. In our August update, we detailed a framework to analyse the EM universe and we split them into three broad groups: (1) The larger economies that have sufficiently large populations to generate growth led by domestic demand. China is the leader in this group, which includes India, Indonesia, and to a lesser extent Brazil. (2) The resource exporters. Brazil also falls into this group, as well as most Latam countries, South Africa and the Gulf oil exporters. (3) The processing economies, which effectively represents the resource- short export-oriented economies of Asia and Central Europe. Of the three, we would pick the second group of countries on the basis of the quality and composition of their growth.

Commodities and commodity currencies: We continue to believe that Commodities should be a key beneficiary of the global recovery and as such are also one of our favourite expressions of the ‘recovery trade’. Within the class, Gold has been one of the best performers this year however it is beginning to display characteristics associated with bubbles. Specifically, Gold is supported by arguments for both inflation and deflation. It seems that there is no imaginable scenario which is bad for Gold. That is worrying.

19 JANUARY 2010 18

PAN-EUROPEAN BANKS

BASEL III CAPITAL ANALYSIS

The recent proposals by the Basel Committee (published 17 December 2009), titled Banking Supervision Consultative Document on Strengthening the resilience of the banking sector, are wide-ranging in their scope. We believe that these proposals will be a ‘game changer’ for the sector and will be one of the most important factors to consider for individual banks going forward. The next few years are going to be characterised by the strengthening of capital requirements which, as our analysis shows, is going to negatively affect some banks to a significant degree (such as HSBC and Lloyds) and leave others with considerable excess capital (notably the Nordic banks).

In this section, we summarise the main themes from the consultative document and undertake a quantitative analysis of the effect the proposals have on the capital and leverage ratios of the banks. In close consultation with Matrix’s Fixed Income team, our aim with the quantitative analysis is to try and replicate as closely as possible the results of the Basel Committee’s own impact assessment, due H2 2010. We acknowledge that the proposals are unlikely to be finalised as ‘Standards’ in their current form, given that some of them lack robustness from a conceptual point of view and are likely to be altered in favour of the banks. There are also reasons why our analysis should not be interpreted as harshly as the initial conclusions suggest, (we discuss all of these in Caveats to Implementation of the Basel Proposals, page 33). However, we believe the proposals will largely be adopted as Standards when it finally comes to implementation.

It is clear to us that the regulators have the strong intention of simplifying, harmonising and making more robust the capital structure of banks. It is not inconceivable that they envisage a return to the way banks used to operate 20-30 years ago. Their aim is to improve comparability, drive out complexity from bank balance sheets and improve the capital robustness of the banking system. To this end, we can only conclude that the Basel proposals will ultimately be penal to banks’ return on equity as a whole. Within the sector though, those banks which, by their own conservativeness or by the conservativeness of their national regulator, adhere closest to the proposed regime will suffer the least, whilst those banks at the other end of the spectrum will likely incur significantly reduced profitability as their capital structure is amended. Generally speaking, the Nordic banks fare the best, followed by the Italian banks. The Spanish banks fare averagely so, together with Standard Chartered, whilst the UK banks HSBC and Lloyds are by far the worst affected.

Conclusions from Capital and Leverage Analysis

We conduct an analysis of what the new capital ratios (both core equity tier 1 and tier 1) will look like under the new proposed regime, which we tentatively call ‘Basel III’. We also calculate what the new leverage ratio will look like. We exclude an analysis of tier 2 and total capital ratios as we do not believe this will be of relevance or interest to equity investors. The analysis is an attempt to replicate as closely as possible the likely results of the Basel Committee’s own impact study, due H2 2010.

The chart below shows the estimated 2009 Core Equity Tier 1 ratios if the Basel III proposals are implemented now, and adds the expected contribution from organic capital generation over the period 2010–2012 to obtain the estimated Core Equity Tier 1 ratios as of the end of 2012.

19 JANUARY 2010 19

PAN-EUROPEAN BANKS

Figure 14: Estimated Core Equity Tier 1 Ratios, 2012E

12.0%

10.0%

8.0%

6.0%

4.0%

2.0%

0.0% BBVA HSBC Intesa Lloyds Nordea DNB NOR DNB Unicredito Santander Handelsbanken Stan. Chartered

New Core Tier 1 ratio (YE 2009e) Estimated organic capital generation, 2010-2012

Source: Matrix Corporate Capital Research

Figure 15: Estimated Capital Excess/Deficit vs. Anticipated Minimum Core Equity Tier 1 Ratio of 6.0%, 2012E

5.0%

4.0%

3.0%

2.0%

1.0%

0.0%

-1.0%

-2.0% BBVA Intesa HSBC Lloyds Nordea Unicredito DNB NOR DNB Santander Handelsbanken Stan. Chartered

Excess core equity capital vs 6.0% minimum, 2012e

Source: Matrix Corporate Capital Research

On an individual stock basis, the results of our analysis are quite polarised.

• The Nordic banks are the best capitalised, with Handelsbanken the best capitalised bank in the group. DNBNOR runs it very close in fact if we take into account that its Core Tier 1 ratio would be approximately 2.0% better if full Basel II figures are used as the starting point in our analysis, rather than the transition Basel II figures that it has published. Note that full Basel II figures are used as the starting point in our calculations for all the other banks and that the 2.0% improvement for DNBNOR is only an estimate, as disclosed by management.

19 JANUARY 2010 20

PAN-EUROPEAN BANKS

• The next best capitalised banks in the group are the Italian and Spanish banks, as well as Standard Chartered. They all have similar estimated 2012 Core Tier 1 ratios of 7–8%. • HSBC and Lloyds are markedly worse than the other banks. The saving grace for HSBC is that strong expected organic capital generation will aid its capital ratios up to 2012 (when full implementation of Basel III is expected), but only to the extent that it can meet minimum capital requirements and nothing more. (We use an anticipated minimum Core Tier 1 capital ratio of 6.0%). HSBC may feel the need to raise more equity capital to establish a suitably large buffer above the regulatory minimum capital. Lloyds is in an even worse position in our opinion. By our analysis, it sees its Core Tier 1 ratio fall to the lowest level in the group. The lack of substantial organic capital generation up to 2012 means that there is a real risk of the Core Tier 1 ratio being below 5.0%. Whilst this is below the trigger level at which the £8.5bn of COCOS are converted to equity (at a conversion price of 59.2p) our understanding is that the terms of the COCOS specifically state that the Core Tier 1 ratio is as calculated on a Basel II basis. Dilution via the conversion of COCOS is therefore not a concern from this angle (we do not see the Core Tier 1 ratio on a Basel II basis falling below 5.0%) but it does not detract from the fact that Lloyds is particularly undercapitalised under the new regime and should require new equity capital. We estimate Lloyds’ capital deficit (against a minimum Core Tier 1 ratio of 6.0%) to be approximately £7.8bn, which is 17% of shareholders’ funds expected at the end of 2012. Components of Estimated Change in Core Tier 1 Ratio The chart below shows the various components of the estimated changes in the Core Tier 1 ratios if the Basel proposals are applied as at the end of 2009. Note that the components are not strictly ‘additive’ – although all the deductions are applied to core capital in the numerator, the increase in market risk weights is applied to RWAs in the denominator of the Core Tier 1 ratio. The sum total of the individual effects on the Core Tier 1 ratio, however, is very close to the actual total change.

We believe that the methodology of how the Basel III proposals are applied is extremely important. The proposals are wide ranging and the interpretation of them is tricky, but not impossible. In addition, disclosure of the necessary information by the banks is not quite as transparent as it could be. Most of the information is available in financial reports, but where it is not, we have obtained what we can from investor relations departments. To this end, we have provided a full audit trail of all the changes that we have applied in order to obtain the new Core Equity Tier 1 and Tier 1 capital ratios. Please see the chapter, Adjustments made in obtaining the new Basel III capital ratios, page 23.

19 JANUARY 2010 21

PAN-EUROPEAN BANKS

Figure 16: Components of the Estimated Changes in the Core Tier 1 Ratios, 2009E

0.0% -0.5% -1.0% -1.5% -2.0% -2.5% -3.0% -3.5% -4.0% -4.5% -5.0% BBVA HSBC Intesa Lloyds Nordea DNB NOR DNB Unicredito Santander Handelsbanken Stan. Chartered

100% of holdings in other financial institutions Shortfall in existing stock of provisions Minorities Negative AFS and cash flow hedging reserves added back to T1 Deferred tax assets, for losses incurred Defined benefit pension deficit Market risk increased 3 times

Source: Matrix Corporate Capital Research

On average, the greatest impact on Core Tier 1 ratios comes from the increase in market risk weighted assets. In our methodology, market RWAs are increased threefold, in accordance with the average increase observed from the BIS’ own Quantitative Impact Study, undertaken in October 2009. The methodology is explained in more detail later.

The other components affect the banks to varying degrees. The aggregate impact on core capital for most banks in our group appears reasonably low and manageable.

The exceptions, however, are HSBC and Lloyds, which is mainly by virtue of the less conservative policies permitted by the FSA. Lloyds’ core capital ratio, for example, falls mainly because of the 100% deduction of holdings in insurance entities, which under the FSA’s current Basel II transition guidance is allowed to be deducted 100% from total capital, rather than 50:50 from Tier 1 and Tier 2 capital, (which is the regulatory basis applied to most other banks). As per the transitional provision in the FSA Handbook (http://fsahandbook.info/FSA/html/handbook/GENPRU/TP/7):

‘A firm may elect not to apply GENPRU 2.2.239R (2) to (4) (50:50 split between deductions from tier one capital and tier two capital) to material insurance holdings acquired before 20 July 2006. If a firm elects not to apply GENPRU 2.2.239R (2) to (4), the firm must deduct such material insurance holdings from the total of tier one capital and tier two capital.’

Basel III is essentially crystallising the problem that Lloyds has been able to get away with for years of double counting the capital in other financial entities on the group balance sheet. HSBC also takes advantage of the above provision and, unsurprisingly, has a significant hit to its core tier 1 capital as a result (although not to the same extent as for Lloyds).

Both Lloyds and HSBC are also impacted to a significant degree by the deduction of deferred tax assets. This is more substantial for Lloyds given the large group losses incurred. In Lloyds’ case, we are sceptical that the full value of the DTA can be realised by the time that Basel III is fully implemented by the end of 2012, given the lack of organic earnings generation that we predict in our model. In HSBC’s case, we believe substantial earnings will result in the full realisation of its DTA by the end of 2012.

19 JANUARY 2010 22

PAN-EUROPEAN BANKS

Another issue affecting HSBC in particular is the full deduction of negative AFS reserves. Negative AFS reserves pertaining to bond assets are currently added back in HSBC’s core equity capital. We deduct the full amount in our analysis.

Leverage Ratio In its consultative document, the Basel Committee discusses of the introduction of a leverage ratio as a supplementary measure, with a view to eventually migrating it to a Pillar 1 treatment (i.e. making it subject to a regulatory minimum requirement).

For our estimate of the Basel III leverage ratio, we use the new Tier 1 capital that we have calculated in accordance to the Basel III proposals as the denominator. For the numerator, we use total assets as forecasted at the end of 2009. In the absence of sufficient published information, we exclude off-balance sheet items.

The results show that most of the banks are below a leverage ratio of 30x, which we would envisage as an appropriate regulatory maximum. Other banks are slightly over this limit, but we believe that they would be able to make small adjustments to their balance sheets in order to delever. Lloyds, perhaps unsurprisingly, has a very high leverage ratio of 50x, reinforcing our concerns about the company.

Figure 17: Banks Ranked by Basel III Leverage Ratio

60.0

50.0

40.0

30.0

20.0

10.0

0.0 BBVA HSBC Intesa Lloyds Nordea DNB NOR DNB Unicredito Santander Handelsbanken Stan. Chartered

Source: Matrix Corporate Capital Research

Adjustments made in obtaining the new Basel III capital ratios

The two tables below show the exact numbers used in calculating the new capital ratios.

We also provide an audit trail showing how these figures have been obtained, with reference to earnings reports or to disclosures from investor relations (IR).

19 JANUARY 2010 23

PAN-EUROPEAN BANKS

Figure 18: Numbers Used for Basel III Capital Analysis, Table 1 BANK (local currency) BBVA DNB NOR Handelsbanken HSBC Intesa Core tier 1 capital (YE 2009E) 23,192 84,656 68,619 104,179 28,603 Tier 1 capital (YE 2009E) 30,117 93,381 83,389 119,827 31,603 RWA (YE 2009E) 287,616 1,082,995 618,479 1,170,867 365,535

Core Tier 1 ratio (YE 2009E) 8.06% 7.82% 11.09% 8.90% 7.82% Tier 1 ratio (YE 2009E) 10.47% 8.62% 13.48% 10.23% 8.65%

Deduct from Core tier 1 capital (if not already deducted from T1): 100% of holdings in other financial institutions -1,527 0 -5,777 -10,568 -400 Excess of expected IRB losses over existing impairment allowances 0 -339 -827 -3,375 -140 Minorities -1,279 -3,265 -240 -6,943 -1,126 Negative AFS and cash flow hedging reserves added back to T1 0 0 -1,389 -11,659 0 Deferred tax assets, for losses incurred -1,631 0 0 -6,018 -600 Defined benefit pension deficit 0 -608 -1,647 -4,639 0

New Common capital 18,756 80,444 58,739 60,977 26,337

Adjustments for Tier 1 capital: Total existing Tier 1 hybrid instruments 5,398 8,725 14,770 15,648 3,000 Of which total preference shares 5,398 0 5,200 3,747 3,000 Add: Eligible Tier 1 hybrid instruments 2,159 0 2,080 0 1,200 Add: Minorities 1,279 3,265 240 6,943 1,126

New Tier 1 capital 22,194 83,709 61,059 67,920 28,663

Add to RWA: Increase in risk weighting for assets backing trading Market risks 11,217 27,194 12,658 76,866 17,652 Market risks % RWA 3.9% 3% 2% 7% 5% Market risk increased by 3 times 33,651 81,582 37,973 230,598 52,957

New RWA 310,050 1,137,383 643,794 1,324,599 400,840

New Core Tier 1 ratio (YE 2009E) 6.05% 7.07% 9.12% 4.60% 6.57% New Tier 1 ratio (YE 2009E) 7.16% 7.36% 9.48% 5.13% 7.15% Estimated organic capital generation, 2010–2012 0.95% 0.84% 0.97% 1.38% 1.00%

Capital adequacy versus proposed regulatory minimum: Core Tier 1 ratio, 2012E 7.00% 7.91% 10.09% 5.98% 7.57% Proposed regulatory minimum Core Tier 1 ratio 6.00% 6.00% 6.00% 6.00% 6.00% Excess core equity capital, 2012E (%) 1.00% 1.91% 4.09% -0.02% 1.57% Excess core equity capital, 2012E (amount) 3,422 23,340 30,839 -201 6,554 Excess capital as % of FY 2012E shareholders' funds 9% 20% 29% 0% 11%

Impact on Core tier 1 ratio of individual changes: DEDUCT from Core tier 1 capital: 100% of holdings in other financial institutions -0.53% 0.00% -0.93% -0.90% -0.11% Shortfall in existing stock of provisions 0.00% -0.03% -0.13% -0.29% -0.04% Minorities -0.44% -0.30% -0.04% -0.59% -0.31% Negative AFS and cash flow hedging reserves added back to T1 0.00% 0.00% -0.22% -1.00% 0.00% Deferred tax assets, for losses incurred -0.57% 0.00% 0.00% -0.51% -0.16% Defined benefit pension deficit 0.00% -0.06% -0.27% -0.40% 0.00% Other 0.00% 0.00% 0.00% 0.00% 0.00% ADD to RWA: Market risk increased 3 times -0.58% -0.37% -0.44% -1.03% -0.69%

Leverage ratio: Total assets (excluding off balance sheet assets) 544,968 1,882,750 2,180,415 2,519,227 625,184 New Tier 1 Capital 22,194 83,709 61,059 67,920 28,663 Leverage ratio 24.6 22.5 35.7 37.1 21.8

Source: Matrix Corporate Capital Research

19 JANUARY 2010 24

PAN-EUROPEAN BANKS

Figure 19: Numbers Used for Basel III Capital Analysis, Table 2 BANK (local currency) Lloyds Nordea Santander Stan. Chartered Unicredito Core tier 1 capital (YE 2009E) 37,922 16,628 43,254 16,373 41,128 Tier 1 capital (YE 2009E) 49,221 18,770 51,435 22,301 45,013 RWA (YE 2009E) 470,394 171,938 544,151 209,116 454,694

Core Tier 1 ratio (YE 2009E) 8.06% 9.67% 7.95% 7.83% 9.05% Tier 1 ratio (YE 2009E) 10.46% 10.92% 9.45% 10.66% 9.90%

Deduct from Core tier 1 capital (if not already deducted from T1): 100% of holdings in other financial institutions -10,073 -1,174 -936 -304 -2,000 Excess of expected IRB losses over existing impairment allowances 0 -269 0 -517 -645 Minorities -1,637 -85 -2,680 -238 -3,108 Negative AFS and cash flow hedging reserves added back to T1 -2,069 -1 0 -326 0 Deferred tax assets, for losses incurred -3,913 -18 -2,000 -675 -1,513 Defined benefit pension deficit -515 -209 0 -100 -73

New Common capital 19,715 14,872 37,638 14,213 33,789

Adjustments for Tier 1 capital: Total existing Tier 1 hybrid instruments 11,299 2,142 8,181 6,003 3,885 Of which total preference shares 4,944 0 8,181 2,699 1,498 Add: Eligible Tier 1 hybrid instruments 0 0 3,272 0 599 Add: Minorities 1,637 85 2,680 238 3,108

New Tier 1 capital 21,352 14,957 43,591 14,451 37,496

Add to RWA: Increase in risk weighting for assets backing trading Market risks 18,233 4,414 32,649 18,121 11,759 Market risks % RWA 4% 3% 6% 9% 3% Market risk increased by 3 times 54,698 13,241 97,947 54,364 35,278

New RWA 506,859 180,765 609,449 245,359 478,213

New Core Tier 1 ratio (YE 2009E) 3.89% 8.23% 6.18% 5.79% 7.07% New Tier 1 ratio (YE 2009E) 4.21% 8.27% 7.15% 5.89% 7.84% Estimated organic capital generation, 2010–2012 0.54% 0.99% 1.37% 1.64% 0.70%

Capital adequacy versus proposed regulatory minimum: Core Tier 1 ratio, 2012E 4.43% 9.22% 7.54% 7.43% 7.76% Proposed regulatory minimum Core Tier 1 ratio 6.00% 6.00% 6.00% 6.00% 6.00% Excess core equity capital, 2012E (%) -1.57% 3.22% 1.54% 1.43% 1.76% Excess core equity capital, 2012E (Amount) -7,832 6,486 10,035 3,784 9,306 Excess capital as % of FY 2012e shareholders' funds -17% 25% 11% 9% 12%

Impact on Core tier 1 ratio of individual changes: DEDUCT from Core tier 1 capital: 100% of holdings in other financial institutions -2.14% -0.68% -0.17% -0.15% -0.44% Shortfall in existing stock of provisions 0.00% -0.16% 0.00% -0.25% -0.14% Minorities -0.35% -0.05% -0.49% -0.11% -0.68% Negative AFS and cash flow hedging reserves added back to T1 -0.44% 0.00% 0.00% -0.16% 0.00% Deferred tax assets, for losses incurred -0.83% -0.01% -0.37% -0.32% -0.33% Defined benefit pension deficit -0.11% -0.12% 0.00% -0.05% -0.02% Other 0.00% 0.00% 0.00% 0.00% 0.00% ADD to RWA: Market risk increased 3 times -0.58% -0.47% -0.85% -1.16% -0.44%

Leverage ratio: Total assets (excluding off balance sheet assets) 1,071,486 491,612 1,081,423 430,345 948,866 New Tier 1 Capital 21,352 14,957 43,591 14,451 37,496 Leverage ratio 50.2 32.9 24.8 29.8 25.3

Source: Matrix Corporate Capital Research

19 JANUARY 2010 25

PAN-EUROPEAN BANKS

Figure 20: Audit Trail of Adjustments Made to Obtain New Capital Ratios, Table 1 BBVA DNB NOR Handelsbanken HSBC Intesa Deduct from CT1 capital (if not already deducted from T1): 100% of holdings in other P19 Q3 2009 Report. This Investments in other 50% of investments in P190, H1 2009 Interim Figure not disclosed in financial institutions is shown as "Other financial institutions are other financial institutions Report. Mainly pertaining accounts. IR states it as "a adjustments" in the reported as zero. P33 of already deducted from T1 to investments in insurance few hundred million". accounts, but IR explains Q3 2009 Report. capital. Remaining 50% companies. This amount is that this pertains to stakes deducted from T2 capital. deducted from T2 capital in financial holdings with a However, this only pertains but not T1 capital. participation >10% (mainly to capital contributions the stakes in Citic Bank and made after 20 July 2006. CIFH, and also some Under current rules, insurance companies). 50% investments in financial (i.e. €1527m) is deducted institutions made prior to from T1 and 50% from T2. that are deducted 100% from total capital. Excess of expected IRB Generic provisions are not P33 of Q3 2009 Report. P34 of Q3 2009 Report. P190, H1 2009 Interim €140m deducted from losses over existing included in Tier 1 capital. This is 50% of the amount 50% of the excess of Report. 50% of excess of each of T1 and T2 capital. impairment allowances We treat this as a buffer which is deducted from expected losses over expected losses over Disclosed by IR. which already fully covers Tier 2 capital. impairment allowances is impairment provisions (i.e. any excess of expected deducted from T1 capital $3375m) is already losses using IRB over and 50% from T2 capital. deducted from T1 capital, existing impairment but the other 50% is allowances. deducted from T2 capital. Minorities As per the balance sheet. Confirmed by IR. P16 of Q3 2009 Factbook. P190 of H1 2009 Interim P44 Q3 2009 Report. Report. Confirmed by IR. Negative AFS and cash Spanish banks already P33 Q3 2009 Report. Negative AFS reserves P190 of H1 2009 Interim Page 44, 9M09 accounts. flow hedging reserves deduct all negative AFS Deduction of NOK2284m and cash flow hedges are Report. Includes the Negative AFS and cash added back to T1 reserves from Tier 1 from T1 for losses made added back to Tier 1 reversal of the deduction flow hedging reserves capital. In any case, BBVA on bond assets recorded at capital. P34 of Q3 2009 arising from the gain on already fully deducted from did not have any negative fair value already made. Report. own debt from T1 capital. T1 under Italian reserves. regulations. Deferred tax assets, for P129 of 2008 Financial DTAs already deducted Deferred tax assets P201, H1 2009 Report. Net IR discloses that only losses incurred Accounts. Figure is for from Tier 1 capital. P33 of already deducted from Tier of deferred tax liabilities. EUR600m of gross DTAs DTA for "Tax losses and Q3 2009 Report 1 capital. P16, Q3 2009 relate to losses. other". DTA for other items Factbook are not eligible deductions from Tier 1 capital according to IR. Defined benefit pension P108 and 109, notes 24 From Q3 2009 Accounts, This actually relates to P212, H1 2009 Report IR states "not material". deficit and 25.2 of 2008 Financial p33. surplus value pension Accounts. Net pension assets rather than a liability has been provided defined benefit pension for in full. deficit. The amount is deducted from total capital, not T1 capital. Adjustments for Tier 1 capital: Total existing Tier 1 hybrid As per the balance sheet. T1 hybrids are entirely As per the balance sheet. As per the balance sheet. As per the balance sheet. instruments comprised of perpetual subordinated loan capital. Of which total preference As per the balance sheet. DNBNOR has no SEK 5.2bn are non- As per the balance sheet. As per the balance sheet. shares preference shares. innovative hybrids (including a staff convertible loan of SEK 2.3bn). The rest of the T1 hybrids are innovative hybrids. Add: Eligible Tier 1 hybrid 40% of preference shares There are no preference Estimate that 40% of the UK FSA excludes nearly all 40% of preference shares instruments considered eligible as Tier shares, so there is no non-innovative hybrids current preference shares considered eligible as Tier 1 capital. eligible hybrid T1 in our satisfy the new Basel III as well as innovative 1 capital. opinion. criteria. hybrid capital from T1 capital, stating that hybrid capital must have features "regarding permanence, flexibility of payments and loss absorbency to be eligible as tier one capital." Virtually all preference shares in issue in the UK do not have loss absorbency pari passu with common equity. Increase in risk weighting for assets backing trading Market risks Data provided by IR (not P153 2008 Annual Report. P35 Q3 2009 Interim P191, H1 2009 Report. P45 Q3 2009 Report. disclosed in reports). Adjusted for 2009E. Report. Adjusted for Adjusted for 2009E. Adjusted for 2009E. 2009E.

Source: Matrix Corporate Capital Research

19 JANUARY 2010 26

PAN-EUROPEAN BANKS

Figure 21: Audit Trail of adjustments made to obtain new capital ratios, Table 2

Lloyds Nordea Santander Stan. Chartered Unicredit Deduct from CT1 capital (if not already deducted from T1): 100% of holdings in other P60 H1 2009 Interim P61 Q3 2009 Factbook. P15 Q3 2009 Report. IR P39 H1 2009 Report. IR will not disclose figure. financial institutions Report. This amount is Relates to deductions for states that virtually all "Material holdings and They state "it is a substantial deducted in T2 capital but investments in insurance deductions pertain to securitisations" are part of deductions. Deduction not T1 capital. Mainly companies. This figure is investments in other deducted 50% from T1 is 50:50 between T1 and T2 pertaining to investments deducted from T2 capital, financial institutions. Total capital and 50% from T2 capital for Banks' and in insurance entities not T1 capital. amount of €1872m is capital. The amount Financial Institution and (which is £8729m of total). deducted 50% from T1 deducted here is the 100% from T2 for Insurance". and 50% from T2 capital. amount deducted from T2 Disclosed deductions from T2 capital. are just above €2bn, so we use €2bn as an estimate. Excess of expected IRB Lloyds actually books a Most recent data is only Generic provisions are not P39 of H1 2009 Interim Excess of losses over losses over existing surplus of £2884bn of available from the 2008 included in Tier 1 capital. Report. This is 50% of the impairment charges is split impairment allowances provisions against Annual Report, Table 43, We treat this as a buffer excess of expected losses 50:50 between T1 and T2 expected losses on p50. NOK269mn is the which already fully covers over existing impairment capital. We deduct the 50% underlying IRB portfolios. deduction from T2 capital, any excess of expected allowances which is deducted from T2. The positive delta is which is 50% of the total. losses using IRB over deducted from T2 capital. included in T2 capital (not The other 50% has existing impairment NB There is no tax T1 capital). P60 of H1 already been deducted allowances. adjustment since this has 2009 Interim Report. from T1 capital. already been accounted for in its entirety in T1 capital. Minorities IR indicates that minorities As per balance sheet. As per balance sheet. P39 H1 2009 Report Confirmed by IR. included in common equity is similar to that disclosed on balance sheet. Negative AFS and cash Page 60, H1 2009 P33 Q3 2009 Report Spanish banks already P44 H1 2009 Report. Negative reserves already flow hedging reserves accounts. Includes cash deduct all negative AFS Figure needs to be deducted 100% from T1 added back to T1 flow hedging reserve. reserves from capital. adjusted slightly to obtain capital under Italian the amount added back to regulations. Tier 1 capital for bond assets only. Exact figure not disclosed by IR. Deferred tax assets, for Page 89, H1 2009 Interim P47 Q3 2009 Report IR discloses that it is P43 H1 2009 Report DTA stemming from tax loss losses incurred Accounts EUR1.5-2.0bn. carryforwards are disclosed only in full-year accounts. As of 2008 they were only €1,513m. Table 14.1, P260 2008 Annual Report. Defined benefit pension P60, H1 2009 Interim P32 Q3 2009. Defined IR confirms that it is zero. Exact figure not disclosed The small deduction is deficit Report benefit pension deficit, net by IR. IR states that the actually from an overfunded of pension assets. deficit will be very small. defined benefit pension scheme (i.e. Surplus assets). P279/280 2008 Annual Accounts. Adjustments for Tier 1 capital: Total existing Tier 1 hybrid As per the balance sheet. As per the balance sheet. Hybrid capital is all As per the balance sheet. As per the balance sheet. instruments preference shares. As disclosed by IR. Of which total preference As per the balance sheet. Hybrid capital contains no As per the balance sheet. As per the balance sheet. P226 Q3 2009 Report: "Non- shares preference shares. innovative capital Majority is innovative instruments", which we hybrid. interpret as preference shares. Add: Eligible Tier 1 hybrid UK FSA excludes nearly There are no preference 40% of existing preference UK FSA excludes nearly We assume 40% of total instruments all current preference shares, so there is no shares considered eligible all current preference preference shares are eligible shares as well as eligible hybrid T1 in our as T1 hybrid under the shares as well as as T1 hybrids under the new innovative hybrid capital opinion. new proposals. innovative hybrid capital proposals. from T1 capital, stating from T1 capital, stating that hybrid capital must that hybrid capital must have features "regarding have features "regarding permanence, flexibility of permanence, flexibility of payments and loss payments and loss absorbency to be eligible absorbency to be eligible as tier one capital." as tier one capital." Virtually all preference Virtually all preference shares in issue in the UK shares in issue in the UK do not have loss do not have loss absorbency pari passu absorbency pari passu with common equity. with common equity. Increase in risk weighting for assets backing trading Market risks P62 of H1 2009 Interim P44 Q3 2009 Report. Disclosed by IR. Not P39 H1 2009 Report. P227 of Q3 2009 Extended Report. Adjusted for Adjusted for 2009E. available in Q3 2009 Adjusted for 2009E. Report. Figures adjusted 2009E. Report. for 2009E.

Source: Matrix Corporate Capital Research

19 JANUARY 2010 27

PAN-EUROPEAN BANKS

Timing of Basel III Implementation

In terms of timing:

• Consultation on the proposals will continue until April 2010. • An impact assessment will be carried out in the first half of 2010. • The Basel Committee will then review the regulatory minimum level of capital in the second half of 2010. • The final Standards should be developed by the end of 2010 and phased in, as financial conditions allow, by the end of 2012.

Summary of Main Themes From Consultative Paper

The main themes from the consultative paper are as follows. A more detailed summary of the proposals is provided in Appendix I on page 87.

• Quality consistency and transparency of the capital base will be raised. Common equity will essentially comprise common shares plus retained earnings. All deductions from capital will be made at the common equity level. Additional capital included in Tier 1 capital will include minorities (by our interpretation of the proposals), but exclude non-qualifying hybrids. This accounts for virtually all existing innovative hybrid securities as well as some preference shares currently in issue. We discuss later the criteria which lead to their exclusion. The UK FSA has been even more conservative in its recent (late December 2009) consultative document by appearing to exclude all preference shares in their current form from Tier 1 capital. Note that the proposals do not ban the issuance of new hybrid securities which satisfy the criteria, nor do they prevent banks from altering the features of existing hybrids via a cram down. Also, there will be appropriate grandfathering arrangements for ineligible hybrids instruments and a transition period for implementation of the new capital standards. These will be determined once the impact assessment has been completed, although we note that the UK FSA has already provided key details of its own proposed grandfathering process. • Capital requirements for trading books will be increased. Additionally, counterparty risk exposure arising from derivative, repo and security financing will be raised, whilst collateral and mark-to-market exposures to centralised clearing facilities will qualify for a zero risk weighting. This forces banks to use central clearing for derivative trades and make their balance sheets less complex. It also increases risk weighted assets, putting more pressure on capital ratios. • A gross leverage ratio (i.e. not risk weighted) will be introduced. The leverage ratio is again quite onerous, using as its numerator a simple non risk based calculation of assets and also including off-balance sheet items, standby letters of credit, cancellable commitments etc. The denominator will probably be Tier 1 or common capital, but both smaller under the new Basel III proposals. There will be no netting of derivatives. US banks net off derivatives (therefore reducing the size of assets and liabilities on the balance sheet) but European banks do not (except for Credit Suisse, which uses US GAAP accounting). This will be seen as a negative for banks with large derivative exposures which have previously indicated a much smaller asset base if US-style netting is allowed (e.g. Deutsche Bank). It would also be particularly onerous for US banks of course; however, note that the Basel committee is not a statutory authority in any country and Basel II has not even been implemented in the US.

19 JANUARY 2010 28

PAN-EUROPEAN BANKS

• Pro-cyclical provisioning buffers will be introduced. This builds on the success of the pro-cyclical generic provisioning model in Spain. Of particular interest is the attempt to transfer the risk of being under-provisioned to employees, by restricting the payment of bonuses to them when the bank’s capital is within a buffer range just above the minimum capital requirement. To a certain extent this can be viewed as raising the minimum capital requirement. • A global minimum liquidity standard will be introduced. The necessity for an increased stock of liquid assets ironically increases the demand for the vast issue of government paper to finance national deficits.

Methodology of Analysis

We outline below the scope of our Basel III Analysis, which is conducted on a ‘best efforts’ basis given that the proposals are likely to change in advance of final implementation by the end of 2012, and given the fact that some required information is not disclosed by bank managements (for example, the amount of negative AFS reserves pertaining to bond assets added back to Tier 1 capital is not disclosed by Standard Chartered). Our analysis of the impact on banks’ capital ratios should therefore be seen in the context of direction and estimated magnitude.

The key quantitative parts of our analysis are as follows:

• We analyse the impact on the banks’ Core Tier 1 ratio as of the end of 2009 by applying the deductions to be made to common equity and dividing by risk- weighted assets, which are adjusted upwards for the increase in risk weighting for trading books. • We calculate the new tier 1 capital of each bank (as of the end of 2009) by deducting all ineligible hybrids currently included in Tier 1 capital. This includes nearly all innovative hybrids (i.e. those with a synthetic maturity, usually by virtue of having a step-up feature). It also includes the deduction of preference shares which we think are ineligible on the basis of not having fully discretionary dividends, (note that the vast majority of existing preference shares satisfy the other criteria of being perpetual and non-cumulative). Matrix’s fixed income team estimates that 60% of total preference shares outstanding are ineligible on this basis and we apply this deduction in our model. The exceptions are the UK banks (HSBC, Lloyds and Standard Chartered) where we deduct 100% of preference shares. The FSA, a regulatory body no less, appears to have totally excluded preference shares (in their current form) from Tier 1 capital on account that they do not have loss-absorptive features. Lastly, it should also be remembered that although minorities are excluded from common equity under the new proposals, they are still included in Tier 1 capital as far as we can see. Minorities remain within Tier 1 capital as they did before. • We calculate the new leverage ratio, determined as total assets divided by the new tier 1 capital. In the absence of sufficient information, we do not include off-balance sheet and similar items. A summary of the changes to the key components of the capital structure, in tabular form, is presented below.

19 JANUARY 2010 29

PAN-EUROPEAN BANKS

Figure 22: Summary of Changes to the Key Components of the Banking Capital Structure

Current BASEL II treatment Proposed BASEL III treatment Description of capital under

BASEL III Exclude hybrid capital, (including preference Exclude hybrid capital, (including preference Essentially common shares plus shares). shares). retained earnings, net of regulatory adjustments and deductions (as Include minorities. Exclude minorities. proposed). Include defined benefit pension deficits. Exclude defined benefit pension deficits and assets. No deductions made. Deduct negative AFS reserves (previously COMMON EQUITY added back to Tier 1 capital). ("Common" or Deduct negative cash flow hedging reserves "Core") Exclude DTA pertaining to losses, (net of deferred tax liabilities), if not already excluded. Deduct investments in financial entities using a "corresponding deduction approach". Deduct 100% of excess of expected IRB losses over existing impairment allowances. Include Tier 1 hybrid capital. Exclude innovative hybrid capital (i.e. with Remainder of the T1 capital base some form of synthetic maturity; this must be comprised of instruments encompasses nearly all hybrids currently in that are subordinated, have fully issue). Some preference shares in issue are discretionary non-cumulative also excluded; most are permanent capital dividends or coupons and have and have non-cumulative dividends (which neither a maturity date nor an satisfies the criteria) but do not have fully incentive to redeem. Innovative discretionary dividends. However, UK FSA hybrid capital instruments with an seems to ban all existing preference shares incentive to redeem through from T1. features like step-up clauses, currently limited to 15% of the T1 Include minorities (part of common equity). Include minorities (but not part of common will be phased out. This appears to equity). exclude innovative hybrid capital Exclude defined benefit pension deficits and Defined benefit pension deficits and surplus but allows preference shares in TIER 1 CAPITAL surplus assets (part of common equity). assets now deducted from common. their current form. Note that the UK (T1) FSA appears to have additionally Add back negative AFS reserves pertaining Negative AFS reserves now deducted from moved to exclude preference to bond assets. common. shares in their current form Add back negative cash flow reserves. Negative cash flow reserves now deducted (because of no loss absorption from common. feature). Include DTA pertaining to losses, net of DTL. 100% DTA, net of DTL, pertaining to losses now deducted from common. Deduct 50% of investments in financial 100% of investment in financial entities now entities. Other 50% deducted from T2. Some deducted using a "corresponding deduction banks deduct 100% from T2. approach". Deduct 50% of excess of expected IRB 100% of excess of expected IRB losses over losses over impairment allowances. Other existing impairment allowances now deducted 50% deducted from T2. from common. TIER 2 CAPITAL Some deductions (as above) for investments Nearly all deductions now made from T2 capital, available to absorb (T2) in financial entities and excess of expected common capital. However, investments in the losses on a "gone" concern basis, IRB losses over impairment allowances. Tier 2 capital of other financial entities needs reduced to a single level i.e. no to be deducted. lower (LT2) and upper T2 (UT2) only old LT2 i.e. plain vanilla sub debt. TIER 3 CAPITAL Included in total capital. Eliminated under Basel III proposals. T3 is currently comprised of sub (T3) debt used as capital against market risk. Under Basel III proposals, T3 will be eliminated.

Source: Matrix Corporate Capital Research

Deductions from Common Equity The common equity proportion of Tier 1 capital is where all deductions to capital are now made. We list below the key items.

19 JANUARY 2010 30

PAN-EUROPEAN BANKS

• Investments in financial subsidiaries. Currently, investments in financial entities (including banks and insurance companies) are either deducted 50:50 from Tier 1 and Tier 2 capital or 100% from Tier 2 capital (as is the case for HSBC and Lloyds). The Basel proposals state that there should be a ‘corresponding deduction approach to investments in the capital of other banks, other financial institutions and insurance entities’. Common equity investments should therefore be deducted from common equity, Tier 1 hybrid investments should be deducted from Tier 1 hybrids, and so on. To all intents and purposes, however, a bank solely invests in the common equity of other financial entities, not other parts of their capital structure. We therefore deduct 100% of investments in other financial entities from common equity. The rationale for the proposed deduction is to remove the double counting of capital in the banking sector and limit the degree of double counting in the wider financial system. • Excess of expected IRB losses over existing impairment allowances. A bank permitted to use Internal Ratings Based (IRB) methodology is allowed to calculate expected losses arising from its loan portfolio. The excess of expected losses over existing impairment provisions is currently deducted 50:50 from Tier 1 and Tier 2 capital. If expected IRB losses are less than the existing impairment allowances (i.e. the bank has excess provisions) then the current treatment is to add this to total capital, above the Tier 1 capital line. • Minority interests. The proposed deduction of minority interests from common equity has come as somewhat of a negative surprise to the market. The rationale for the deduction is that, although it absorbs losses within subsidiaries, it is not available to absorb losses at the group core capital level. It should be noted that minorities can still be included in Tier 1 capital. • Negative AFS reserves pertaining to bond assets. This comes under the scope of the proposal that no adjustment should be applied to remove from the common equity component of Tier 1 capital unrealised gains or losses recognised in the balance sheet. In many cases, banks opt to add unrealised losses on AFS bond assets back to Tier 1 capital on the premise that the assets are often held to maturity, whereupon the full fair value will be realised (the assumption being that the discount to fair value is due to illiquid markets rather than actual credit losses). • Negative cash flow hedge reserves. This is a small amount for the banks analysed. The treatment is usually to add this back to Tier 1 capital. • Deferred tax assets, for losses only. The deduction of deferred tax assets will be for previously incurred losses, not for DTAs pertaining to loan loss provisions, for example. DTAs are usually included in Tier 1 capital, although some banks currently deduct it. • Defined benefit pension deficits and assets. Defined benefit pension deficits are currently treated as equity rather than a liability under current Basel rules and are usually not deducted from Tier 1 capital. Pension assets, which we interpret as surplus assets in the scheme over commitments (to the extent that they exist), are sometimes included in total capital. The proposal is that they should be deducted from common equity on the basis that these assets are not an accessible form of capital.

19 JANUARY 2010 31

PAN-EUROPEAN BANKS

In our methodology, our starting point is Core Tier 1 capital (NOT common equity). We start off with Core Tier 1 capital because we want to see what impact the proposals have with respect to the current Core Tier 1 ratio. Core Tier 1 capital is simply Tier 1 capital, less preference shares and other hybrid capital included in T1 capital. Consequently, any deductions already made to T1 have also been deducted from Core Tier 1 capital. We do not therefore need to deduct from Core Tier 1 capital those items that have already been deducted from Tier 1 capital, (for example, the 50% of investments in banking entities that is deducted from Tier 1 capital). Instead, we need to deduct from Core Tier 1 capital the following:

1. Relevant items that lie outside of Tier 1 capital (for example, the 50% of investments in banking entities that is deducted from Tier 2 capital)

2. Items within Tier 1 capital that need to be excluded (for example, minorities and negative AFS reserves).

It is clear that existing deductions from Tier 1 capital are currently made on an inconsistent basis across banks (e.g. some banks fully deduct negative AFS reserves from T1 capital, whereas others do not). This has been taken into account in our analysis so that the banks are presented on a like-for-like basis.

Additional Instruments Included in Tier 1 Capital Aside from common equity, other capital included in Tier 1 must be comprised of instruments that have the following features:

• Are subordinated • Have non-cumulative dividends or coupons • Are fully discretionary • Have neither a maturity date nor an incentive to redeem (i.e. are perpetual) Under these criteria, nearly all innovative hybrids (i.e. those that have a step-up feature or other synthetic maturity) are excluded because they have a degree of ‘non-permanence’. Non-permanent capital is undesirable for a bank in financial distress.

Meanwhile, some preference shares will be ineligible. Although the vast majority of preference shares in issue are permanent capital and are for the most part non- cumulative (i.e. they do not have dividends that can be deferred to a later date), the main limiting criterion in our opinion is the fact that some preference shares in issue are not, strictly speaking, fully discretionary. We do not, for example, consider preference shares which stipulate payment of dividends if there are distributable reserves to be fully discretionary. In our analysis, we estimate the amount of existing preference shares that are ineligible under the new proposals to be 60%.

This treatment of preference shares might seem harsh, but it may not be harsh enough if we take the FSA as a leading indicator of how national regulators will interpret the Basel proposals. The UK FSA, in its own consultative document regarding the implementation of the Basel III proposals (which was released late December 2009) is even more conservative in its acceptance of hybrids within Tier 1 capital. In summary, it states that hybrid capital must have features ‘regarding permanence, flexibility of payments and loss absorbency to be eligible as tier one capital.’ Virtually all preference shares currently in issue by UK banks do not have loss absorptive capability which is pari passu with common equity.

We do not include grandfathering arrangements in our analysis, assuming that the transition to Basel III occurs as of the end of 2009. This may seem harsh, but is done to preserve consistency in our analysis and to highlight that, regardless of the

19 JANUARY 2010 32

PAN-EUROPEAN BANKS

grandfathering timeframe involved, some banks will indeed have their Tier 1 capital ratios negatively impacted more so than others and this will ultimately affect their relative earnings growth.

Aside from common equity and eligible hybrids, we also include minorities in Tier 1 capital. There is nothing in the consultative document which suggests to us why they should not be included in Tier 1 capital, despite their exclusion from common equity.

Increase in Risk-Weighted Assets The increase in risk-weighted assets in our analysis arises from increasing the required capital backing the trading book. We do this by multiplying the market risk component of RWA by 3. This is in accordance with the Quantitative Impact Study (http://www.bis.org/publ/bcbs163.pdf) published by the Basel Committee in October 2009 to assess the effect on banks of revised capital requirements for trading book exposures. The scope of the increased risk weightings encompasses:

• A capital charge for incremental risk (i.e. default risk) • A stressed value-at-risk (VaR) capital charge • Capital charges for securitisation exposures in the trading book • Revised specific risk capital charges for certain equity exposures. The results of the impact study indicate an average increase of 223.7% of market risk capital requirements. We therefore increase the market risk weighting component of risk weighted assets (which is disclosed by all the banks analysed) by 3x to reflect the approximate average increase. Note, however, that the individual impact on banks varies immensely, ranging from a decrease in market risk weighting of 19.5% at the low end to an increase of 1112.8% at the high end.

In the absence of published data, we do not increase RWA for counterparty risk exposure, but acknowledge that banks with substantial derivative, repo and securities financing business will incur a marked increase in capital requirements (therefore mainly applicable to the investment banks).

Calculation of Capital Ratios The Core Tier 1 and tier 1 ratios are calculated, respectively, as the common equity and tier 1 capital divided by the RWAs.

We also calculate the organic capital ratio generation from the beginning of 2010 to the end of 2012. This encompasses growth/contraction in both capital and RWAs. We add this to our new Core Tier 1 and tier 1 ratios to provide an estimate of what capital ratios will look like by the end of 2012, when full implementation of the final Basel III standards is expected.

Leverage Ratio The leverage ratio, as outlined briefly in the consultative paper, is calculated here on the basis of total assets divided by the new tier 1 capital. We exclude off balance sheet items, in the absence of published figures.

Caveats to Implementation of the Basel Proposals

The Basel proposals, in our opinion, are undoubtedly very conservative compared to the current Basel II regime. We believe that national regulators will be keen to err on the side of conservativeness in their adoption of the proposals as they compete with each other to attain ‘regulatory kudos’. (The Spanish banking regulator, Banco de Espana has, in our opinion, improved its reputation immensely compared to the somewhat discredited FSA – is this perhaps one driver behind the FSA releasing its own consultative document soon after the Basel proposals were released, which was even more onerous in its interpretation of eligible Tier 1 hybrid capital?).

19 JANUARY 2010 33

PAN-EUROPEAN BANKS

Whilst many of the proposals have a robust conceptual grounding, there are some which we believe the Basel Committee will dilute or eliminate in favour of the banks. There are also some reasons why we think it is not fair to accept the conclusions of our analysis as harshly as has been indicated. We outline these below:

Features of the Proposals We Think May Be Diluted/Eliminated • The key concern about the proposals expressed by some analysts is that it is unfair to deduct minority interests and not make a proportionate deduction for the associated minority RWAs. We think this argument is applicable to the calculation of the Core Tier 1 ratio, which excludes minorities in the numerator but includes 100% of minority RWAs in the denominator. We do not agree with this in the calculation of the Tier 1 ratio, which includes minorities fully in both the numerator and denominator. Eliminating the deduction of minorities would lead to the greatest improvement in Core Tier 1 ratios for HSBC (+0.59% benefit) and Unicredit (+0.68%). Regulators argue that the deduction, as proposed, makes sense because a bank could be fully liable for its subsidiaries’ losses if it goes bankrupt, but minority-owned equity might not support losses elsewhere in the group. The counterargument is that banks could walk away from a bankrupt subsidiary if they wanted to. Although this is a legally viable option, in practice the parent bank nearly always supports the minority owned subsidiary when it is in financial distress. We believe a compromise will be reached where some sort of proportionate deduction from the risk-weighted assets in the calculation of the Core Tier 1 ratio will be allowed. • Regarding the calculation of the leverage ratio, it is perhaps too harsh to disallow netting of derivative assets and liabilities on the balance sheet (therefore reducing total assets). Note that US banks are allowed netting under US GAAP accounting. • Deferred tax assets in Italy arise not just from losses incurred, but from loan losses exceeding 0.3% of total customer loans, which under Italian corporate income tax law, has to be deducted from net income over the following 18 years. The Italian banking regulator has moved to explicitly suggest that a limit be introduced to mitigate this country specific legislation, whereby a deduction of DTAs only above a certain threshold as a percentage of Core Equity Tier 1 capital is necessary.

Reasons Why We Think the Conclusions of the Analysis Could Be Deemed Too Conservative • Some proportion (if not all) of deferred tax assets may be realised by the time that Basel III will be implemented at the end of 2012, meaning that it will be too harsh to deduct them in their entirety. We believe that most of the banks will be able to realise the full value of the DTAs by the end of 2012. The one bank which we believe may not be able to do so is Lloyds, which does not generate enough earnings over the period in our opinion. • Conversion to the new Core Tier 1 ratio will be smoothed by the implementation of transition rules. We do not know what form these transition rules will take at the moment. However, in our opinion, they merely provide time to those banks which are required to make adjustments to their capital base to satisfy minimum capital requirements. They do not exclude those banks from having to raise better quality capital to plug any capital shortfalls, which leads us to believe the market will look through the transition rules.

19 JANUARY 2010 34

PAN-EUROPEAN BANKS

• Grandfathering of ineligible hybrid instruments in Tier 1 capital will permit a smoother transition. The FSA has provided details on its grandfathering proposals, whereby 100% of ineligible hybrid assets in Tier 1 capital can be grandfathered for 10 years from 31 December 2010. For the next 10 years, only 40% of those ineligible hybrids can be grandfathered and for the next 10 years after that, 20% of the ineligible hybrids can be grandfathered. This falls to 0% after 30 years. Note, however, that grandfathering has no impact on the (more important) Core Tier 1 ratio, since core common capital does not include hybrid instruments. • Banks can issue new qualifying hybrid capital or make ‘cram down’ adjustments to existing hybrid capital such that it becomes eligible as Tier 1 capital. Again, however, this is more an issue for transition to the new Tier 1 ratio rather than the more important Core Tier 1 ratio.

19 JANUARY 2010 35

PAN-EUROPEAN BANKS

DUPONT METHODOLOGY FOR BANKS

Dupont analysis is a methodology by which the ROE of a company can be broken down into its constituent parts. We use Dupont analysis in two ways:

1) To undertake a comparison analysis of the individual banks to ascertain the drivers of ROE and the relative quality of earnings.

2) To perform a trend analysis of the ROE drivers such that we can see how the banks have performed relative to each other and why. We undertake this analysis using quarterly data since the beginning of 2006 to encapsulate the impact of the financial crisis on the banks.

ROE is a measure of how successfully the management of a company has deployed the equity to generate a return for its shareholders. However, ROE incorporates leverage in its calculation. For two banks with the same amount of assets and generating the same return on those assets, the bank with the smaller amount of equity (and hence the higher equity leverage) will generate the higher ROE.

ROE can thus be expressed as a follows. Note that average balance sheet numbers are used in all the Dupont equations over the period in question. Equity is as stated by the company and therefore includes intangibles.

ROE = Net Profit = Net Profit x Assets Equity Assets Equity

= ROA x Equity Multiplier

The ROA reflects how effectively the bank’s management is employing the bank’s assets and cannot be skewed by leverage. It is a pure leverage measure by which we can make a consistent comparison between banks. The ratio of Assets % Equity meanwhile, is known as the equity multiplier. When the bank is making profits, then a high equity multiplier will multiply earnings as a percentage of equity and will therefore enhance the ROE. However, this will similarly be the case when the bank is making losses, and exacerbates the risk of bankruptcy. The ratio of assets to equity has obviously become an increasing focal point over the course of the financial crisis, but the key limitation should be highlighted in that it does not take into account the risks inherent in the various underlying assets. We therefore complement the equity multiplier with the Core Tier 1 ratio, which uses risk-weighted assets (RWA) as its denominator.

We disaggregate ROA further into its constituent parts. In doing so, we can isolate the key elements which are driving ROA (and thus also ROE). The key component for a lending bank (i.e. with retail and corporate lending operations) is of course net interest income (NII). Other components are:

• Net non-NII operating income. (This comprises fees and commissions, dividends and profits/losses on investments carried at equity, profits/losses on trading, income from insurance business and other operating income); • Net non-operating income. (This includes goodwill impairments, asset write downs, minority interests, income from discontinued operations and exceptional items); • Operating costs (comprising personnel and administrative costs); • Loan loss charges (on the bank’s loan book); • And finally, taxes.

19 JANUARY 2010 36

PAN-EUROPEAN BANKS

All of these earnings components are expressed as a percentage of assets, so that when added together, they sum to the ROA. We can compare the components of ROA for each bank to judge their relative earnings quality. The market should be prepared to rate a bank more highly if it has better earnings quality (i.e. has a higher proportion of earnings derived from NII rather than more volatile non-operating income and/or has low loan loss charges, pointing towards better asset quality).

The key driver of ROA for a lending bank is the NII. The NII % assets component of ROA can be disaggregated further into a multiplicative equation involving the NII % interest-earning assets (more commonly known by its conventional name of Net Interest Margin, or NIM) and interest-earning assets % total assets (also known as the Earning Asset Ratio).

NII = NII x IEA Assets Interest-earning assets Assets

= Net Interest Margin x Earning Asset Ratio

The bank can improve its ROA if it can improve its NIM relative to other banks. We see in our analysis later that some banks have been able to achieve this over the course of the financial crisis by winning a greater share of the retail deposit market, which provides them with much cheaper funding versus expensive wholesale borrowing. We supplement our analysis of NIM with a trend analysis of loan % deposit ratios and see that those banks that have an improving LTD ratio are also those that have an improving NIM.

The NII % assets can also be improved by increasing the ratio of interest-earning assets to total assets. We see later however that this is not a factor that differs substantially between banks and neither does it change considerably over time. The vast proportion of assets for banks in this report is, unsurprisingly, made up of interest-earning assets (between 85-95%).

The Dupont system has been adapted for banks from the book Successful Bank Asset/Liability Management2 (which we recommend as essential reading for all bank analysts and managers). We show a chart of the system on the next page. There are mathematical symbols between some of the boxes to indicate the operation to perform between the ratios in the relevant equation. We use average balance sheet items in our calculations.

Limitations of Dupont Analysis Dupont analysis is an excellent way to disaggregate the ROE of a company into its constituent parts. However, in the case of banks, the methodology does not encompass analysis of capital adequacy, liquidity or asset quality. We therefore supplement our Dupont analysis with more conventional tools such as Core Tier 1 ratios, loan % deposit ratios and non-performing loan coverage ratios.

Dupont analysis is also a bottom-up approach. To complete the picture for banks we provide a top-down analysis in our Sector and Economic Overview, page 9.

2 Bitner, John W. with Goddard, Robert A, Successful Bank Asset/Liability Management: A guide to the future beyond gap (John Wiley & Sons, 1992).

19 JANUARY 2010 37

PAN-EUROPEAN BANKS

Figure 23: Dupont System for Banks

NET INTEREST INCOME NET INTEREST INCOME % % AV. TOTAL ASSETS AV. INTEREST-EARNING ASSETS (i.e. Net Interest Margin)

NET NON-INTEREST OPERATING INCOME% AV. TOTAL ASSETS INTEREST-EARNING ASSETS % AV. TOTAL ASSETS (i.e. Earning Asset Ratio) RETURN ON AVERAGE NET NON-OPERATING INCOME TOTAL ASSETS % AV. TOTAL ASSETS

RETURN ON AVERAGE EQUITY OPERATING COSTS % AV. TOTAL ASSETS TOTALASSETS % EQUITY (i.e. Equity Multiplier)

LOAN LOSS CHARGES % AV. TOTAL ASSETS

GROUP TAX % AV. TOTAL ASSETS

Source: Matrix Corporate Capital Research. Dupont model adapted for banks from Successful Bank Asset/Liability Management: A guide to the future beyond gap (John Wiley & Sons, 1992).

19 JANUARY 2010 38

PAN-EUROPEAN BANKS

DUPONT ANALYSIS OF EARNINGS

We present the findings of our Dupont analysis of the lending banks in two parts. We firstly analyse the ROE composition, as estimated at the end of 2009, (note that all the banks under consideration have December year-ends, so there is periodic consistency in the comparison). This allows us to compare the quality of earnings, with higher quality banks deriving the bulk of their revenues from NII and incurring lower costs and LLCs as a percentage of assets.

We then present a trend analysis of the different Dupont components of the banks’ ROE. This is in the form of historical time series charts, showing the relative performance of each component of the ROE on a quarterly basis since the beginning of 2006. We gain some insightful observations using this analysis, particularly with respect to historical and future expected trends in the NIM in conjunction with the development of the loan to deposit ratio.

Analysis of Banks’ ROE Composition

We use the Dupont system to analyse the composition of banks’ ROE as of the end of 2009. This enables us to:

• Determine which companies have a high ROE by virtue of having a high ROA, and which rely more on having high balance sheet leverage (as indicated by the equity multiplier, or the assets % equity ratio). • Determine the relative quality of ROA for the companies, by comparing the main constituents of ROA. • Analyse further whether the NII % assets is as a result of a high net interest margin or a high earning asset ratio. • Analyse the development of the NIM with respect to the development of the LTD ratio. We are able to observe, for example, the extent to which an expansion in cheap deposit funding (resulting in a fall in the LTD ratio) results in an improvement in the NIM.

Figure 24: Composition of 2009E ROE Using the Dupont System for Banks ROE equals: ROA equals: NII % Assets equals: ROE ROA x Equity NII + Net non-NII + Net non-op. - Operating - LLC - Taxes Net interest x Earning Multiplier (as % of op. income income costs Margin asset (Assets average (NII % IEA) ratio Company % Equity) total (IEA % assets) total assets) BBVA 18.24% 0.92% 19.7 252 bps 121 bps -3 bps -149 bps -98 bps -31 bps 2.69% 94% DNBNOR 9.90% 0.46% 21.4 123 bps 85 bps 8 bps -103 bps -45 bps -21 bps 1.40% 88% Handelsbanken 13.41% 0.48% 28.2 102 bps 48 bps 0 bps -68 bps -17 bps -17 bps 1.08% 95% HSBC 6.89% 0.29% 23.6 167 bps 123 bps 3 bps -137 bps -116 bps -11 bps 2.19% 76% Intesa Sanpaolo 6.21% 0.50% 12.3 168 bps 111 bps -8 bps -149 bps -55 bps -17 bps 1.94% 87% Lloyds -14.24% -0.48% 29.9 173 bps 157 bps 68 bps -162 bps -303 bps 20 bps 2.51% 69% Nordea 12.89% 0.51% 25.4 109 bps 81 bps 0 bps -90 bps -32 bps -17 bps 1.50% 73% Santander 13.81% 0.81% 17.0 243 bps 119 bps -17 bps -151 bps -91 bps -22 bps 2.75% 88% Standard Chartered 16.33% 0.90% 18.1 176 bps 199 bps -3 bps -189 bps -55 bps -37 bps 1.96% 90% Unicredit 2.74% 0.17% 16.6 173 bps 103 bps -13 bps -150 bps -85 bps -11 bps 1.87% 92%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 39

PAN-EUROPEAN BANKS

Breakdown of ROE into ROA and Equity Multiplier It comes as no surprise that the Spanish banks BBVA and Santander have the highest ROEs, together with Standard Chartered. They are closely followed by the Nordic banks. The Italian banks and HSBC have low ROEs, in the low-to-mid single digit percent range. Lloyds stands out as still making significant losses.

A breakdown of ROE into ROA and the Equity Multiplier provides us with further insight as to how that ROE has been achieved with respect to genuine profitability of the asset base, versus the use of leverage on the balance sheet (i.e. assets % equity).

We see that the high ROEs of BBVA, Santander and Standard Chartered are driven by high ROAs and that the use of balance sheet leverage is moderate.

Then, perhaps surprisingly, there is a marked drop in ROAs for the next group of banks; they comprise the Nordic banks, (Nordea, Handelsbanken and DNBNOR), and Intesa. Within this, the Nordic banks are characterised by having fairly high balance sheet leverage while Intesa has very low leverage.

The third group of banks have the lowest ROAs; it comprises HSBC, Unicredit and Lloyds. It is perhaps surprising to find HSBC in this category, but we see later that this is mainly because of the currently elevated loan loss charges being experienced, which we believe will normalise at a much lower level. HSBC has a slightly greater than average equity multiplier. Unicredit, although also experiencing high loan losses, is a bank which we believe does indeed have inherently low returns, which we examine later. Meanwhile Lloyds, experiencing by far the highest loan loss ratio of all, is making a negative ROA. Although the loan losses will normalise, there are a number of other significant factors which mean that we are very doubtful that the bank will be able to achieve the high returns that it enjoyed in the past. One of these is the very high balance sheet leverage, shown by the equity multiplier of 37x.

We should remember that the simple use of the equity multiplier is not a clear cut indication of inherently higher balance sheet risk. There is no appreciation of risk weightings attributed to the different assets on the balance sheet. Were we to take this into account, then the Nordic banks would have very low risk weighted assets, and indeed very high Core Tier 1 ratios (i.e. core equity tier 1 capital % risk weighted assets). We gain comfort on the Nordic banks from our Basel III Capital Analysis, on page 19. In any case, the leverage ratios of the Nordic banks, while higher than average, are not at a high enough level to raise concerns.

The bank where we do have genuine concerns about balance sheet leverage is Lloyds. The leverage ratio is substantially higher than for the other banks, shown by both the simple equity multiplier and the leverage ratio as per our Basel III Analysis.

Figure 25: Banks Ranked by ROA, FY 2009E Figure 26: Banks Ranked by the Equity Multiplier, FY 2009E

1.00% 35.0 0.80% 30.0 0.60% 25.0 0.40% 20.0 0.20% 0.00% 15.0 -0.20% 10.0 -0.40% 5.0 -0.60% 0.0

Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

19 JANUARY 2010 40

PAN-EUROPEAN BANKS

Breakdown of ROA into Constituent Components As expected for the lending banks, the key component of operating income is NII. On average, NII comprises 60% of operating income for the lending banks analysed. (Operating income is defined as NII plus non-NII operating income). However, within that, some banks obviously make a greater proportion of their income from NII than others, as per the table below. The ‘high quality’ banks which derive most of their operating income from net interest income are Handelsbanken, and the Spanish and Italian banks. At the other end of the scale, banks which derive a greater proportion of their operating earnings from lowly rated sources (mainly trading income, and fees and commissions) are Standard Chartered and Lloyds. It perhaps comes as no surprise that our Basel III Analysis shows that Lloyds and Standard Chartered have some of the largest decreases in their Core Tier 1 ratios as a result of the increase in (trading-related) market risk weighted assets.

Figure 27: Breakdown of Operating Income Into NII and Non-NII Operating Income, FY 2009E

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

NII % operating revenue Non-NII % operating revenue

Source: Matrix Corporate Capital Research

Looking at the breakdown of ROA for each bank (as below), we make the following observations:

Figure 28: Breakdown of ROA into Constituent Components, 2009E. Components Expressed as % of Average Total Assets

500 bps 400 bps 300 bps 200 bps 100 bps 0 bps -100 bps -200 bps -300 bps -400 bps -500 bps

NII Net non-NII op. income Net non-op. Income Operating costs LLC Taxes

Source: Matrix Corporate Capital Research

19 JANUARY 2010 41

PAN-EUROPEAN BANKS

The Spanish banks have exceptionally high NII, more than offsetting fairly high operating costs and loan loss charges. The high NII reflects, in our opinion, the high lending margins coming from BBVA’s and Santander’s Latin American operations (mainly Mexico for BBVA and Brazil for Santander). High LLCs mainly arise from the domestic Spanish operations. In aggregate, the ROAs are amongst the highest in the group, and will get higher still as the LLCs normalise.

Intesa and Unicredit at first appear quite similar in terms of their ROA composition. In aggregate, the overall ROA is average in Intesa’s case, with strong NII and moderate non-NII operating income more than offsetting operating costs and LLCs. However, the higher LLCs at Unicredit, by virtue of its Eastern European exposure, means that it has the second lowest ROA in the peer group after Lloyds.

The Nordic banks are characterised as having low positive contributions to ROA from NII and non-NII operating income. However, they are also the lowest cost operators in the group, with very low loan loss charges as well. Handelsbanken stands out as the most extreme example in the group, having the lowest costs and LLCs, but also the lowest contributions to ROA from NII and non-NII operating income. In our opinion, this reflects the Nordic banks being very efficient, risk-averse lenders, and that the Nordic markets in which they operate are characterised by having very low margins.

HSBC has high positive contributions to its ROA from both NII and non-NII operating income (mainly trading income). However, it also ranks highly in terms of LLCs (mainly from its US consumer finance operations). Our trend analysis later on shows how we believe HSBC has significant potential to increase its ROA as its LLCs normalise. We also believe the bank has potential to improve its NII and NIM as it deploys its excess of deposit funding in new high margin loans.

Standard Chartered, like the Spanish banks, has high positive contributions to its ROA. However, it is skewed more towards lower quality non-NII operating income (mainly trading income) rather than high quality NII. We are struck by how elevated its operating costs are as a percentage of assets, making Standard Chartered the most inefficient bank on this measure. LLCs, however, are extremely low, reflecting the low loss experience currently being enjoyed in Asia.

Lloyds stands out for a number of reasons. It has high positive contributions to ROA, but it is of low quality in our opinion. A not insignificant proportion comes from trading income (which is within the non-NII operating income component). NII, the component that we would deem high quality, is of an average level. Also, Lloyds is the only bank in the group to have notable non-operating income in 2009. This item is from an exceptional gain on the net fair value unwind of loans purchased from HBOS, which we expect to be much smaller going forward.

Regarding the negative components of ROA, LLCs are extremely elevated, largely because of the loan loss provisions taken against the commercial real estate and corporate loan books acquired from HBOS. These should normalise at a lower level of course, particularly since management indicates that it ‘front-loaded’ provisioning of these books in H1 2009.

Aside from the Nordic banks, Lloyds is also one of the lowest cost operators in the peer group. It is already benefitting from efficiency gains as it integrates HBOS. We believe it can yield further cost savings as this process continues.

19 JANUARY 2010 42

PAN-EUROPEAN BANKS

Trend Analysis

Applying a trend analysis to the lending banks using the Dupont system leads us to the following conclusions:

• If a bank operates in a strong competitive environment for deposits and loans, then this would appear to curtail its ability to attract cheap deposits (therefore reducing its loan to deposit ratio) and reduce its funding costs. We see this most markedly in the comparison between Intesa/Unicredit and HSBC/Standard Chartered/Spanish banks. • HSBC and Standard Chartered are, to us, ‘super deposit franchises’. They have been able to benefit the most from the flight to quality of cheap customer deposits. Their LTD ratios are now both about 80%, representing an excess of deposits over loans. HSBC and Standard Chartered now have the unique advantage of being able to use their excess of deposits to fund new lending at very high margins. We believe this will be a driver for the positive development of NIMs going forward for these two banks. • Loan losses are at or very near a turning point for some banks in our opinion, notably for the Nordic banks, Intesa, HSBC and Standard Chartered. In addition, we believe Lloyds, having booked a substantial amount of provisions (indeed, to the extent that there exists an excess of impairment provisions against IRB expected losses in the capital disclosure), will start to show a marked improvement in its loan loss experience. Some banks are still likely to suffer elevated loan losses for the next few quarters in our opinion, principally due to macro concerns. These banks include BBVA and Santander (due to Spanish loan losses) and Unicredit (due to CEE loan losses) • Trading income is predicted to fall going forward. Current high levels of trading income are unlikely to be sustained going forward. The attribution of higher risk weights to trading assets, as proposed by the Basel Committee, is also going to reduce the risk weighted returns for such activity. We reflect this in our models and it can be observed on a Dupont basis in the reduced amount of non-NII operating income going forward. The reduction affects those banks which rely on trading income the most (mainly Standard Chartered, Lloyds and HSBC within our group). We present our Dupont trend analysis in the form of time series charts, starting from the beginning of 2006. For ease of comparison, we show each component of ROE as four charts with the same axes, (one chart each for the Spanish, Italian, Nordic, and UK banks).

The Dupont trend analysis gives rise to some very interesting observations as to how the banks have performed over the course of the crisis. We see, for example, how the NIM has moved in tandem with changes in interest rates and the development of the loan to deposit ratio. This information is also a useful aid in determining how the banks will perform going forward. With respect to the NIM for example, we can judge how this should develop after taking into account changes in the base rate and how the excess (or lack) of deposits will have a bearing on funding costs.

We start off by disaggregating ROE into ROA and the equity multiplier.

19 JANUARY 2010 43

PAN-EUROPEAN BANKS

ROE = ROA x Equity Multiplier Figure 29: ROE, Spanish banks Figure 30: ROE, Italian banks

60.0% 60.0%

50.0% 50.0%

40.0% 40.0%

30.0% 30.0%

20.0% 20.0%

ROE 10.0% ROE 10.0%

0.0% 0.0%

-10.0% -10.0%

-20.0% -20.0%

-30.0% -30.0% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 31: ROE, Nordic banks Figure 32: ROE, UK banks

60.0% 60.0%

50.0% 50.0%

40.0% 40.0%

30.0% 30.0%

20.0% 20.0% ROE 10.0% ROE 10.0%

0.0% 0.0%

-10.0% -10.0%

-20.0% -20.0%

-30.0% -30.0% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

ROEs have obviously come down from the high levels established before the crisis. This has largely been due to higher loan losses. What is notable from the above charts are those banks that haven’t suffered a significant decrease in ROEs, which are Handelsbanken and Standard Chartered. Unsurprisingly, these banks have had the lowest increase in loan loss charges over the course of the crisis. The question going forward is whether the same pre-crisis ROE levels can be achieved by the other banks in coming years. We argue that it will be difficult for most of them to achieve this for several reasons:

1) The majority of banks have already undergone a process of deleveraging by virtue of raising substantial amounts of equity through rights issues, making it difficult to improve ROE on an enlarged equity base.

2) The current buoyant profits from fixed income trading are unlikely to be sustained. Trading activity will also become a lower return activity when Basel proposals to increase market risk weights are implemented.

3) Banks will undergo changes to their capital structure from Basel III which will require them to hold more equity capital. This will negatively impact ROEs. Our Basel III Analysis indicates that this will affect Lloyds the most by far and that HSBC will also be significantly impacted. We have not factored in the impact on ROEs from Basel III into our models, but caveat our analysis with the acknowledgement that we are likely too optimistic in our ROE forecasts.

4) A reduction in the LTD ratio for Lloyds from ~170% currently to a targeted 140% will likely force its ROE lower in coming years (as we discuss later).

19 JANUARY 2010 44

PAN-EUROPEAN BANKS

Figure 33: ROA, Spanish banks Figure 34: ROA, Italian banks

3.00% 3.00%

2.50% 2.50%

2.00% 2.00%

1.50% 1.50%

1.00% 1.00% ROA ROA

0.50% 0.50%

0.00% 0.00%

-0.50% -0.50%

-1.00% -1.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 35: ROA, Nordic banks Figure 36: ROA, UK banks

3.00% 3.00%

2.50% 2.50%

2.00% 2.00%

1.50% 1.50%

1.00% 1.00% ROA ROA

0.50% 0.50%

0.00% 0.00%

-0.50% -0.50%

-1.00% -1.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

THE ROA component of ROE has generally fallen for most banks due to loan loss charges increasing. This has affected HSBC and Lloyds the most over the course of the crisis. We generally see an improvement in ROAs over the next few years to 2012 as loan losses normalise, but believe that the improvement will be more gradual than the upturns in previous crises due to the more shallow nature of the economic improvement expected. This is perhaps not in keeping with our economic analysis, but we prefer to err slightly on the side of caution regarding the development of loan losses.

In our disaggregation of the ROA into its component parts, we examine more closely the drivers of how we expect the ROA to develop over coming years.

19 JANUARY 2010 45

PAN-EUROPEAN BANKS

Figure 37: Assets % Equity (i.e. Equity Multiplier),Spanish banks Figure 38: Assets % Equity (i.e. Equity Multiplier), Italian banks

45.00 45.00

40.00 40.00

35.00 35.00

30.00 30.00

25.00 25.00

20.00 20.00

15.00 ASSETS % EQUITY ASSETS 15.00 ASSETS % ASSETS EQUITY

10.00 10.00

5.00 5.00

0.00 0.00 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 39: Assets % Equity (i.e. Equity Multiplier), Nordic banks Figure 40: Assets % Equity (i.e. Equity Multiplier), UK banks

45.00 45.00

40.00 40.00

35.00 35.00

30.00 30.00

25.00 25.00

20.00 20.00

15.00 15.00 ASSETS % ASSETS EQUITY ASSETS % ASSETS EQUITY

10.00 10.00

5.00 5.00

0.00 0.00 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

The assets % equity ratio (or equity multiplier) has come down for most banks in the group as a result of them raising equity capital, as per the following:

• DNBNOR – NOK14bn raised in December 2009. • HSBC - £12.9bn raised in April 2009. • Lloyds - £13.5bn raised in November 2009 and £4bn open offer in May 2009. • Nordea - €2.5bn raised in April 2009. • Santander - €7.2bn raised in November 2008. • Standard Chartered - £1.8bn in November 2008. • Unicredit - €4bn raised in January 2010. In fact, the only banks in the group that haven’t raised equity are BBVA, Intesa and Handelsbanken.

Additionally, the banks have spoken of the need to reduce leverage. This has obviously occurred to some extent already because of the equity raised, but also as managements have moved to shrink balance sheets. However, yet more shrinkage may occur, particularly in anticipation of the new capital and leverage ratios being proposed by the Basel Committee. We have not modelled a reduction in leverage on this basis going forward, so we caution on our forecasts with the acknowledgement that we have likely estimated the equity multipliers as too high.

The one bank where we do forecast a fall in the equity multiplier is Lloyds, where we model a reduction in assets by virtue of the contraction in the loan book. This arises from management’s ‘soft’ target of bringing down the bank’s LTD ratio from ~170% currently to ~140% over the course of the next three years.

19 JANUARY 2010 46

PAN-EUROPEAN BANKS

ROA = Net Interest Income % Average Total Assets Add Net Non-NII Operating Income % Average Total Assets Add Net Non-Operating Income % Average Total Assets Minus Operating Costs % Average Total Assets Minus Loan Loss Charges % Average Total Assets Minus Group Tax % Average Total Assets Next we can disaggregate the ROA into its (additive) constituent parts. The first component is net interest income.

Figure 41: NII % Assets, Spanish banks Figure 42: NII % Assets, Italian banks

300 bps 300 bps

250 bps 250 bps

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps Net interest income % average total assets Net interest income % average total assets

0 bps 0 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 43: NII % Assets, Nordic banks Figure 44: NII % Assets, UK banks

300 bps 300 bps

250 bps 250 bps

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps Net interest income % average assets total % average income interest Net assets total % average income interest Net

0 bps 0 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

The NII % assets ratio is disaggregated later into the NIM and the Earning Asset Ratio (i.e. interest earning assets % total assets). We make insightful comments about what can be concluded about net interest margins at that point. For the time being, however, it can be observed that:

• The Spanish banks have enjoyed an improving ratio. • The Italian banks have had a modestly contracting ratio. • The ratio for the Nordic banks has been generally stable relative to the other banks. However, within that, DNBNOR and Nordea’s ratios have fallen slightly over the course of the crisis, whilst Handelsbanken’s has improved slightly. • Standard Chartered has suffered a modest contraction. HSBC has had a stable ratio. Lloyds, meanwhile, initially enjoyed an expansion, but then suffered a sharp and significant contraction in NII % assets.

19 JANUARY 2010 47

PAN-EUROPEAN BANKS

Figure 45: Net non-NII operating income % assets, Spanish banks Figure 46: Net non-NII operating income % assets, Italian banks

300 bps 300 bps

250 bps 250 bps

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps Net non-NII Net non-NII operating income % average assets Net non-NII operating income % average assets 0 bps 0 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 47: Net non-NII operating income % assets, Nordic banks Figure 48: Net non-NII operating income % assets, UK banks

300 bps 300 bps

250 bps 250 bps

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps Net non-NII operating income % average assets Net non-NII operating income % average assets 0 bps 0 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Net non-NII operating income comprises fees and commissions, dividends and profits/losses on investments carried at equity, profits/losses on trading, income from insurance business and other operating income. Up until the end of Q1 2009, these items can generally be said to have suffered due to falling volumes and prices in both the equity and credit markets and weakening economies.

Some banks have been more adept than others at positioning themselves for improving asset prices as the equity and bond rally began in March 2009, benefitting either via increased trading flows or taking long positions themselves on proprietary trading books. For other banks, the lack of improvement in non-NII operating income can be put down to an unwillingness or lack of foresight to play the asset reflation trade, or to the fact that their non-NII operating income is more oriented towards the health of the economy (which continues to be subdued) as opposed to the buoyant capital markets. We note, for example, that dividends and insurance income at most of the banks in the group have remained at low levels for most of 2009.

Banks which have enjoyed a rebound in non-NII operating income include Standard Chartered, Lloyds and the Italian banks. Going forward, we do not expect trading orientated profits (particularly fixed income derived) to be sustained as the rally in capital markets runs its course. Trading income is also likely to be negatively impacted by the increase in risk weights expected for such activity under the Basel proposals. The expected fall in trading income is factored into our models going forward. In the Dupont charts above, this is reflected in the muted development of non-NII operating income, with some banks being affected more than others depending on their reliance on trading, offset by the expected recovery in other income lines such as insurance and dividends.

19 JANUARY 2010 48

PAN-EUROPEAN BANKS

Figure 49: Net non-operating income % assets, Spanish banks Figure 50: Net non-operating income % assets, Italian banks

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps

-200 bps -200 bps Net non-operatingNet income % average assets non-operatingNet income % average assets

-250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 51: Net non-operating income % assets, Nordic banks Figure 52: Net non-operating income % assets, UK banks

200 bps 200 bps

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps

-200 bps -200 bps Net non-operatingNet income % average assets Net non-operatingNet income % average assets -250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Net non-operating income includes goodwill impairments, asset write downs, minority interests, income from discontinued operations and exceptional items.

Banks have generally been affected by sporadic negative write-offs, pertaining for example to the Madoff fraud case (affecting a number of banks such as BBVA, Santander and HSBC) and asset and goodwill write downs (affecting HSBC and Intesa in particular).

Lloyds is the one notable bank which has had positive non-operating income over the course of the crisis. The acquisition of HBOS has led to a positive fair value unwind on some of the HBOS loans acquired, reflecting the application of market based credit spreads to HBOS’s lending portfolios and own debt. We expect this to be still substantial in H2 2009, but to reduce to a small positive effect after that.

19 JANUARY 2010 49

PAN-EUROPEAN BANKS

Figure 53: Operating costs % assets, Spanish banks Figure 54: Operating costs % assets, Italian banks

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps

-200 bps -200 bps Operating costs % average assets costs average % Operating assets costs average % Operating

-250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 55: Operating costs % assets, Nordic banks Figure 56: Operating costs % assets, UK banks

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps Operating costs % average assets average % costs Operating -200 bps -200 bps Operating costs% average assets

-250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Operating costs as a % of assets has remained quite stable for most banks throughout the course of the crisis. There have been some modest and gradual efficiency gains for the likes of BBVA, Nordea and DNBNOR, with a notable improvement for HSBC. We forecast this ratio to be stable for these banks going forward.

Lloyds, meanwhile, has benefitted substantially from improved efficiency as it integrates HBOS. We expect the operating costs % assets to be maintained going forward, which is in keeping with an improvement in the cost % income ratio of 200bps for the next two years as per our earnings model.

19 JANUARY 2010 50

PAN-EUROPEAN BANKS

Figure 57: Loan loss charges % assets, Spanish banks Figure 58: Loan loss charges % assets, Italian banks

50 bps 50 bps

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps

Loan loss charges % average -200 assets bps Loan loss charges % average -200 assets bps

-250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 59: Loan loss charges % assets, Nordic banks Figure 60: Loan loss charges % assets, UK banks

50 bps 50 bps

0 bps 0 bps

-50 bps -50 bps

-100 bps -100 bps

-150 bps -150 bps Loanloss charges % average assets

Loan loss charges % average -200 assets bps -200 bps

-250 bps -250 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

For lending banks, loan loss charges have obviously been the key driver causing the negative delta in ROA. (Unlike capital markets banks, they have obviously not had to contend as much with toxic asset write downs). We generally expect a gradual improvement in loan losses to normalised levels over the course of the next three years as the economy recovers, but with some banks experiencing a more delayed improvement versus others given the nature of their exposure to different geographies and risks.

We have concerns about exposure to a Spanish economy which is still weak, and may get weaker still if excessive public spending is reigned in. We see BBVA as possibly having a more pronounced and deeper loan loss experience than Santander in the near term. The most recent Q3 2009 results saw an exceptional €830m gain booked to generic reserves, which was nearly wholly used up by specific loan losses. This is an indication to us that the underlying loan loss experience is likely worse than has been interpreted by the market. BBVA also has a greater proportion of its loan book exposed to Iberia. We discuss this in more detail in the specific Dupont profile of BBVA, on page 57.

Intesa is, in our opinion, already on the path to improving LLCs. Unicredit, however, is seen to experience a deeper and more prolonged loan loss experience given its exposure to the CEE region. We expect Russia and Ukraine, the main problem countries, to take a longer time to show recovery in the credit cycle.

The Nordic banks have had a very benign loan loss experience. We believe the decentralised branch system of Handelsbanken had led to it being the best performer on this metric, but with Nordea and DNBNOR not far behind. We see all of the Nordic banks as on the cusp of improving loan losses.

19 JANUARY 2010 51

PAN-EUROPEAN BANKS

The loan loss experience of Standard Chartered has also been benign given the economic strength of the Asian region. We believe that the bank is already on the path to improvement going forward.

HSBC has suffered high LLCs from its US consumer finance business, but this is set to improve in the near term in our opinion following promising H1 2009 results and Q3 2009 trading statement which indicated the first fall in impairments in two years for the US Finance Corporation division. We discuss this more in HSBC’s Dupont profile.

Regarding Lloyds, having ‘front-loaded’ loan loss provisions in H1 2009 for the CRE book acquired from HBOS, management has indicated much improved loan losses in H2 2009 and beyond. We discuss this in more detail in the specific Dupont profile for Lloyds, on page 60.

Figure 61: Group tax % average assets, Spanish banks Figure 62: Group tax % average assets, Italian banks

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps

0 bps 0 bps Group tax% average assets Group tax% average assets -50 bps -50 bps

-100 bps -100 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 63: Group tax % average assets, Nordic banks Figure 64: Group tax % average assets, UK banks

150 bps 150 bps

100 bps 100 bps

50 bps 50 bps

0 bps 0 bps Group tax % average assets Group tax % average assets -50 bps -50 bps

-100 bps -100 bps 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered

Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Given that most banks in the group have continued to post profits over the course of the credit crisis, they have continued paying normal rates of corporation tax.

The Italian banks notably had positive non-recurring tax items in Q4 2008, relating to the recognition of future tax benefits from the deduction of goodwill from corporate income, as well as the use of deferred tax assets.

The group loss reported by Lloyds in H1 2009 resulted in the use of a tax credit, as well as a policyholder interests related tax credit offsetting in full the charge for policyholder interests included in the Group’s profit before tax.

19 JANUARY 2010 52

PAN-EUROPEAN BANKS

NII % Assets = NII % Interest-earning assets (i.e. NIM) x Interest-earning assets % total assets (i.e. Earning asset ratio) Figure 65: Net Interest Margin, Spanish banks Figure 66: Net Interest Margin, Italian banks

3.00% 3.00%

2.50% 2.50%

2.00% 2.00%

1.50% 1.50%

1.00% 1.00%

0.50% 0.50% Net interest income % interest-earning assets Net interest income % interest-earning assets 0.00% 0.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 67: Net Interest Margin, Nordic banks Figure 68: Net Interest Margin, UK banks

3.00% 3.00%

2.50% 2.50%

2.00% 2.00%

1.50% 1.50%

1.00% 1.00%

0.50% 0.50% Net interest income % interest-earning assets % interest-earning income interest Net

0.00% assets % interest-earning income interest Net 0.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

The analysis of the NIM is best undertaken in conjunction with analysis of the loan to deposit ratio, shown in the following charts. We observe the following.

• The Spanish banks have simultaneously experienced a decrease in LTD ratios and an improvement in NIMs. We believe the improvement in NIMs for both banks has been due to: 1) The fall in the LTD ratio, driven by both banks being able to grow their deposit bases quicker than their loan books. This makes the overall cost of funding cheaper, therefore improving their NIMs. As the crisis unfolded, deposits have sought safer havens at the financially more robust, larger cap banks. We believe this has particularly been the case in Spain with respect to the retrenchment of the caja and domestic orientated banks versus BBVA and Santander.

2) The improvement in relative competitive dynamics. The retrenchment of competitors from the lending market has favoured BBVA and Santander in terms of allowing them to have stronger pricing power for loans.

3) The fall in Euro base rates from 4.25% in September 2008 to 1.0% by May 2009. This resulted in an expansion in asset spreads, given that asset repricing generally lags the immediate fall in base rates. Note that there has been a simultaneous contraction in liability spreads as the base rate has fallen, but this has been mitigated by the increase in the deposit base, which has reduced the cost of funding.

19 JANUARY 2010 53

PAN-EUROPEAN BANKS

Figure 69: Loan % Deposit ratio, Spanish banks Figure 70: Loan % Deposit ratio, Italian banks

300% 300%

250% 250%

200% 200%

150% 150% Loan % depositratio Loan % depositratio

100% 100%

50% 50% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 71: Loan % Deposit ratio, Nordic banks Figure 72: Loan % Deposit ratio, UK banks

300% 300%

250% 250%

200% 200%

150% 150% Loan % deposit ratio % deposit Loan Loan % deposit ratio % deposit Loan 100% 100%

50% 50% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research. Note for Handelsbanken: The Source: Matrix Corporate Capital Research sale of SPP in 3Q07 resulted in a disproportionate reduction in the deposit base versus loans.

• The NIMs for the Italian banks have been squeezed, notably for Intesa, as the base rate has fallen. The fact that Intesa and Unicredit both have LTD ratios just below 100% means that funding costs are already very low and cannot reduce further. The contraction in deposit spreads is therefore significant and this has more than offset the expansion in asset spreads. Note also that the LTD ratios for both Italian banks have remained unchanged, reflecting the continued robustness of domestic competitors in the market for deposits. Relative to the Spanish banks, and indeed HSBC and Standard Chartered as we shall see later, the Italian banks have not benefitted from competitors retrenching from the deposit and lending markets. • The Nordic banks have had relatively stable NIMs throughout the course of the crisis. Within that, Nordea and DNBNOR’s NIMs have deteriorated slightly, whilst Handelsbanken’s has improved slightly. This is consistent with our observations for the respective development of the LTD ratio. We see that the LTD ratio has not changed that much for DNBNOR while for Nordea it has deteriorated slightly (i.e. increased). There has been a marked improvement for Handelsbanken. It would appear from the charts that Handelsbanken has had the best NIM development out of the Nordic banks over the course of the crisis, which we would put down to the relative improvement in the LTD ratio (i.e. reflecting an expansion in its customer deposit base, which reduces its funding costs). One might have expected that, as for the Spanish banks, the fall in Nordic base rates would lead to an expansion in the NIM via widening asset spreads. However, the differentiating factor is that competition for deposits in the Nordic region intensified strongly over the course of the crisis, to the extent that the reduction in deposit spreads has offset the increase in asset spreads.

19 JANUARY 2010 54

PAN-EUROPEAN BANKS

Given the high LTD ratios, one might have been concerned by the impact that the withdrawal of wholesale funding might have had on the Nordic banks. However, the vast proportion of funding, (other than customer deposits), is not short-term wholesale funding, but long term covered bonds (usually 5–10 years maturity), senior unsecured bonds (with maturities ranging from 2–5 years), and commercial paper and certificates of deposit (with maturities up to one year). • HSBC has had a stable NIM over the crisis, against some analysts’ expectations that it would fall given that it has a LTD ratio just under 100% (like Intesa). However, we can see also that the LTD ratio has improved substantially by some 20 percentage points to ~80% as deposits have undertaken a flight to quality. This very cheap source of funding has been used by HSBC to support incremental lending with very high margins, which in our opinion has been one of the key reasons why HSBC’s NIM has not fallen (unlike Intesa). • Standard Chartered, like HSBC, has enjoyed a significant improvement in its LTD ratio to ~80%. Unlike HSBC, the NIM fell noticeably in H2 2008, but we believe this was due to factors specific to Standard Chartered, such as increased funding costs in Korea and Taiwan due to intensified competition. • Lloyds, again, is somewhat of a special case. The NIM increased markedly throughout the course of 2008 due to improvements in product margins. However, pro-forma for the acquisition of HBOS for H1 2009 onwards, the group suffered significantly as liquidity in the wholesale market was withdrawn. Lloyds was forced to borrow £165bn from the UK government at high (albeit undisclosed rates), which resulted in the NIM contracting sharply. The high LTD ratio of the combined group of ~175% is indicative of the reliance on wholesale funding. A significant proportion of this was short term (less than 1 year), although this has now been termed out by virtue of the use of government funding, which is of longer duration than the short-term wholesale funding that it replaced. NIM forecasts Going forward, we generally expect movements in the NIM in accordance with the expectation that base rates will increase very slightly and taking into account the lagged effect of assets being repriced downwards.

This is reflected in the modest contraction in NIM for BBVA and Santander, where in the near term, the effect of assets being repriced downwards is expected to compress asset spreads. Looking further ahead, an expected modest increase in base rates should initially result in a contraction in asset spreads, higher wholesale funding costs, and a contraction in total spreads.

We believe in the near term that margin compression for the Italian banks has reached an end. Looking beyond that, a modest increase in base rates should lead to expanding NIMs. This is because they are largely deposit funded. Their funding costs will remain close to zero, meaning liability spreads will expand and total spreads will improve.

The NIM for Nordic banks is expected to remain insensitive to a modest increase in base rates. We believe that changes in the relative NIM for Nordea and Handelsbanken have been more due to their respective ability to gather cheap deposits, rather than what has happened to base rates, and so we assume the NIM for both to be stable going forward in an environment where the base rate increases only modestly. DNBNOR, in our opinion, is more likely to reverse the fall in its NIM if base rates rise.

We see HSBC and Standard Chartered as special ‘super deposit franchises’, given the improvement in their LTD ratios to ~80%. These banks now have a surplus of

19 JANUARY 2010 55

PAN-EUROPEAN BANKS

cheap deposits with which they can fund high margin incremental loan growth. It appears from analysis of their balance sheets that these deposit surpluses are currently being invested in low yielding cash and highly rated government securities. Redeployment of the deposits in new, higher margin loans would improve their NIMs given the positive mix effect.

For Lloyds, management guides to an improvement in NIM for H2 2009, on the basis of reduced wholesale funding costs, albeit not to the same level achieved in 2008. (The CDS spread is an indicator for the movement in wholesale funding costs and has indeed come down post the successful issue of new equity and COCOS in November 2009). This has been modelled in our earnings forecasts. However, we are sceptical that the NIM can be maintained at this level in light of the fact that the LTD ratio needs to come down to ~140% from ~170%. We therefore model a fall in the NIM after the initial improvement.

Figure 73: Interest-earning assets % total assets, Spanish banks Figure 74: Interest-earning assets % total assets, Italian banks

100.00% 100.00%

95.00% 95.00%

90.00% 90.00%

85.00% 85.00%

80.00% 80.00%

75.00% 75.00%

70.00% 70.00% Interest-earning assets % total assets total % assets Interest-earning Interest-earning assets % total assets total % assets Interest-earning 65.00% 65.00%

60.00% 60.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

BBVA Santander Intesa Sanpaolo Unicredito Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

Figure 75: Interest-earning assets % total assets, Nordic banks Figure 76: Interest-earning assets % total assets, UK banks

100.00% 100.00%

95.00% 95.00%

90.00% 90.00%

85.00% 85.00%

80.00% 80.00%

75.00% 75.00%

70.00% 70.00% Interest-earningassetstotal% assets

Interest-earning assets % total assets 65.00% 65.00%

60.00% 60.00% 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 1Q06 2Q06 3Q06 4Q06 1Q07 2Q07 3Q07 4Q07 1Q08 2Q08 3Q08 4Q08 1Q09 2Q09 3Q09 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e 4Q09e 1Q10e 2Q10e 3Q10e 4Q10e 1Q11e 2Q11e 3Q11e 4Q11e 1Q12e 2Q12e 3Q12e 4Q12e

DNBNOR Handelsbanken Nordea HSBC Lloyds Standard Chartered Source: Matrix Corporate Capital Research Source: Matrix Corporate Capital Research

A bank with high interest earning assets as a proportion of total assets will tend to have more NII as a percentage of total assets, all other things being equal. A low ratio is synonymous with the bank relying on other (lower quality) assets in order to generate its return on assets. This ratio has generally remained high and unchanged for the lending banks. We do not foresee any major changes in the ratio for the banks analysed and in any case, the sensitivity of NII % total assets to changes in the earning asset ratio is not actually significant.

It is worth noting that for Handelsbanken, there was a change in Q1 2008 in the classification of interest bearing securities, resulting in new assets being included in interest-earning assets (and thus resulting in an increase in the earning asset ratio). Nordea had a significant decrease in the earning asset ratio due to the increase in derivative assets and liabilities in Q4 2008 and Q1 2009, which enlarged the asset base.

19 JANUARY 2010 56

PAN-EUROPEAN BANKS

Dupont Profile of Individual Banks

We summarise below the key features of the Dupont analysis for each bank, supplementing it with important conclusions regarding liquidity and asset quality. We do not incorporate the impact on capital (and indeed returns arising from capital deficits/surpluses) from the proposals by the Basel Committee in our earnings model, which is addressed in the Basel III Capital Analysis chapter.

BBVA BBVA is notable in having a high ROE and ROA versus peers, underpinned by its very high NIM. We believe BBVA is a high quality bank which can maintain these structural advantages over the long term, mainly because of its exposure to Latin America. However, in the very near term, we predict that the positive ROA gap is set to narrow due to higher Spanish loan losses and the absence of the large positive one-off €830m capital gain that aided it in Q3 2009.

We note that BBVA has significantly run down its coverage ratio to only 68% as of Q3 2009 (including generic provisions). This was the same level of coverage as at H1 2009, despite BBVA allocating the entire €830m capital gain in the third quarter from the sale and leaseback of properties to generic provisions. Indeed, generic provisions only increased by €140m in Q3 2009. Management indicates to us that the increase in generic provisions was offset by:

• The compensation of specific provision needs • The release of generic provisions due to an increase in quarterly loan volumes (mainly coming from corporate loans, with higher generic consumption requirements • A negative FX effect.

This implies to us that nearly all of the capital gains were absorbed by actual loan losses. We judge, on this basis, that the underlying loan loss experience in Q3 2009 was more severe than the market believes and this is the basis of our near term concerns. Elevated loan losses should be offset by a further capital gain from the sale and leaseback of properties amounting to a few hundred million euros.

BBVA has enjoyed an improving NIM over the course of the credit crisis. A falling base rate has led to expanding asset spreads, whilst BBVA’s improved competitive position, arising from the retrenchment of domestic Spanish banks (mainly the caja savings banks), has allowed it to a) charge more for new loans and b) obtain a greater share of the deposit market and therefore reduce its funding costs (note the slight 10pp improvement in its loan % deposit ratio since the beginning of 2008).

We predict that the NIM will come under pressure in the near term as the lagged effect of assets being repriced downwards now catches up with the fall in the base rate, causing asset spreads to compress. We also believe, looking further ahead, that a modest increase in base rates will have the initial effect of compressing asset spreads further, with the expansion in liability spreads not being able to compensate as much due to the increase in wholesale funding costs (note that BBVA is more dependent on wholesale funding than some of its more deposit funded peers, as can be seen by its high LTD ratio).

19 JANUARY 2010 57

PAN-EUROPEAN BANKS

DNBNOR We see DNBNOR as a very well run bank which has the benefit of operating in a strong, oil-based, Norwegian economy, which has been relatively insulated from the financial crisis. Indeed, GDP growth is predicted by consensus to be 2.3% in 2010. Set against this background, we expect loan losses to be slightly better in 2010 than in 2009, incorporating a recovery in Norway but offset to some extent by continued high loan losses at DNB NORD, the 51% subsidiary which accounts for the group’s Eastern European operations. DNB NORD accounts for only 7% of the group’s loans but approximately half of loan losses as of 9M09.

We expect the ROE of DNBNOR to only gradually recover in 2010 as the normalisation of loan losses is offset by trading income being lower than the high quarterly run rate achieved in 2009. Regarding loan losses, large charges were incurred in Q4 2008 in DNBNOR’s Large Corporates and International division (mainly from international shipping and Norwegian real estate) and its DNB NORD subsidiary, which comprises its Eastern European operations. Loan losses in both entities have remained elevated, although the most recent Q3 2009 results showed an improvement versus Q2 2009. In our model we have assumed a continued improvement for the Large Corporates and International division as corporates benefit from a better Norwegian economic environment. A more modest improvement is modelled for DNB NORD where the economic situation in Eastern Europe, particularly the Baltic countries, is expected to take longer to recover.

We see that the NIM has weakened slightly over the course of the crisis, during which time the Norwegian base rate has fallen from 5.75% in September 2008 to 1.25% in June 2009. The base rate has increased to 1.75% over H2 2009 and is set to increase further to 3.00% in 2010 according to consensus forecasts. However, while we expect deposit spreads to expand, we also expect assets spreads to initially contract due to the lag of asset repricing, meaning that the NIM should again be insensitive to the change in base rate.

DNBNOR does not have a heavy reliance on short-term wholesale financing, despite the LTD ratio being apparently quite high at near to 200%. Covered bonds and other long term securities comprise a large proportion of non-customer deposit funding, which has been a key reason why the NIM stayed stable even though the wholesale funding market dried up. We note that about NOK120bn of covered bonds (about 23% of securities issued on the balance sheet) were swapped through the central bank swap facility in early 2009, which was at a very favourable rate compared to the prevailing market price at the time. The covered bond market has since recovered such that market rates are now more favourable than the government swap facility, so we are quite sanguine on DNBNOR’s funding options.

HSBC We view HSBC as one of the banks which suffered earliest in the credit crisis from rising loan loss charges, by virtue of its exposure to early cycle US consumer finance business, and specifically subprime mortgages. We now see it as one of the earliest to exit the cycle as its loan loss experience normalises in the US, and Asia continues to be economically robust. Synonymous with that is an expectation that HSBC is one of the first lending banks to see an improvement in its ROE and ROA.

19 JANUARY 2010 58

PAN-EUROPEAN BANKS

We see HSBC’s loan loss experience as being at the bottom (or very near the bottom) of the cycle. US impairment trends are a key component of the group’s loss experience. In the most recent Q3 2009 results, trends were better than the market expected, with the HSBC Finance Corporation division in the US reporting the first QoQ fall in impairments since the beginning of 2006 (from $3.4bn in Q2 2009 to $3.0bn). Asset quality in other regions is stabilising, with the overall loan loss charge for the Group being lower in H1 2009 than H2 2008 and the LLC % assets falling versus H1 2009 (although there is some element of seasonality in this). We model an earlier improvement in HSBC’s asset quality versus later cycle peers such as BBVA.

HSBC is an immensely strong deposit franchise, benefitting as deposits undertook a flight to quality. Our Dupont analysis shows how this has reduced its funding costs over time, helping to support its NIM despite the fall in base rates. We see that at the beginning of 2008, its LTD ratio was just below 100% (the same as Intesa). The LTD ratio improved to 80% by H1 2009 as deposits took a flight to quality (whilst Intesa’s stayed roughly the same). Over the same period, as base rates fell globally, HSBC’s NIM still stayed roughly the same, whilst those at Intesa fell by 20bps as its deposit margins were squeezed.

Going forward, we see HSBC’s NIM improving as it seeks to deploy its excess deposits in new high yielding loans. (Excess deposits are currently being invested in low yielding financial instruments such as cash and government debt securities). If we make the simple assumption that the 20% surplus of deposits over loans (derived from the LTD ratio of 79%) can be deployed immediately to write loans yielding 4%, then this would increase NII by a considerable 46% and the NIM would rise from 2.16% to 3.17%. Our expectation for interest rates is that they will stay low for an extended period of time (see our Sector and Economic Overview), but should they rise then HSBC would be one of the main beneficiaries, with funding costs staying close to zero and assets yields being repriced upwards.

HSBC, more so than most of its peer lending banks, has a substantial proportion of operating income coming from non-interest sources. Fees and commissions have been relatively resilient throughout the crisis, but trading activity in particular has been a volatile component of non-interest income. Trading activity in H1 2009 was strong, based on very high customer appetite for corporate bonds and wide spreads. We forecast that trading income will be fairly robust but not as elevated as in the past.

Intesa Sanpaolo Intesa has a low ROE which we believe is structural in nature. We see only modest potential for improvement. The main reason for the low ROE is the low NIM, which in our opinion is characteristic of the robust competition and low margins in Italy.

Intesa’s NIM has contracted over the course of the credit crisis due to deposit spread compression, caused by falling base rates and exacerbated by the fact that, as a largely customer deposit funded bank, its funding costs cannot fall much. Intesa has a loan-to-deposit ratio of ~93%.

Unlike HSBC, which had a similar LTD ratio to Intesa just before the crisis started, Intesa’s LTD ratio has stayed roughly the same throughout the crisis whilst HSBC’s has improved to 79%. We conclude that the more competitive landscape for deposits in Italy has limited Intesa’s ability to grow its deposit base, reduce funding costs and aid NIM expansion. It might also be concluded that the same competitive landscape curtails Intesa’s ability to charge higher margins on new business versus peers such as BBVA and HSBC. We believe this also limits improvement in the NIM.

19 JANUARY 2010 59

PAN-EUROPEAN BANKS

Unlike the Nordic banks, which also have low NIMs, Intesa does not have as efficient a cost base, resulting in its ROE being in the mid single digits whilst those for the Nordic banks are in the low to mid teens.

Intesa’s loan loss experience has been relatively benign (although by the same token, there will be more limited positive delta going forward as the loan loss experience normalises). Although Italy is suffering one of the lowest GDP growth rates in the EU, the country is far less leveraged than ‘Anglo-Saxon’ countries and those in Eastern Europe, with loan-to-value ratios for residential properties being relatively low. This has limited the deterioration in asset quality.

We believe that the potential for Intesa’s ROE to improve is limited given that the delta from loan loss normalisation is low and that we expect base rates to remain depressed for a prolonged period. Additionally, Intesa has benefitted from substantial trading profits in 2009, which we expect to normalise at a lower level.

Lloyds We see Lloyds as having a low (albeit improving) ROE for the next few years as it contends with elevated loan loss charges and structural changes that will result in lower returns. Our main concern is that Lloyds will be unable to achieve the high (~25%) ROEs that it used to as it brings down the LTD ratio from ~170% to ~140% and as its lack of capital adequacy is highlighted by the Basel III changes.

Loan losses increased sharply in H2 2008 and particularly in H1 2009, largely arising from the wholesale division. The weak UK economy, which forced down prices of commercial property valuations, combined with the decision by Lloyds’ management to apply conservative assumptions in its review of the acquired HBOS loan portfolio, resulted in a prudent and very large impairment charge in the CRE book.

We expect the wholesale impairment charge to fall significantly in H2 2009 and fall further still in 2010, given that there should less incremental impairments following the prudency applied in H1 2009. This should be offset to some extent by modest increases in impairments for residential mortgages and commercial loans as economic conditions continue to be challenging. We believe the overall loan loss rate should be much lower going forward, (albeit elevated compared to peers).

Lloyds suffered a sharply contracting NIM from H2 2008 to H1 2009 (pro-forma for the merger with HBOS) as liquidity in the wholesale funding market dried up. Post its acquisition of HBOS, Lloyds’ reliance on wholesale funding increased substantially (the LTD ratio increased from ~140% to ~175%), which put it in a critical liquidity situation when the wholesale markets effectively closed in 3Q08. Lloyds was forced to obtain £165bn of funding from the UK government at very high (albeit undisclosed) rates. The £165bn represents 23% of the bank’s total funding, 50% of wholesale funding and 93% of short-term wholesale funding (i.e. < 1 year). The funding is substantially weighted towards the more expensive, longer term Credit Guarantee Scheme (CGS) as opposed to the cheaper, short term Special Liquidity Scheme (SLS), although the exact amounts are undisclosed. We know that the CGS provides staggered maturities out to 2014, whilst the SLS has maturities falling mainly in 2011. We are informed by the company that wholesale funding rates, for the respective maturities, have come down to a level that is markedly cheaper than the rate applicable for the CGS funding and in line with the rate for the SLS funding – however, this is aside from the fact that £165bn is an enormous amount that will require a liquidity premium.

19 JANUARY 2010 60

PAN-EUROPEAN BANKS

In terms of the future development of the NIM, management has disclosed that there should be a substantial improvement for 2010, but that the level achieved should not be as high as that in 2008. This is due to the marked improvement in wholesale funding rates available in the open market (which can be judged by the fall in Lloyds’ CDS). We have factored in a 50bps improvement versus 2009 in our model, but this is still 65bps shy of the NIM achieved in 2008. Thereafter we predict a gradual contraction in the NIM to 2008 levels. This is where we believe we deviate most with consensus in terms of earnings development. By the same token that we expect HSBC to improve its NIM by deploying its surplus deposits, we believe the move by Lloyds to reduce its LTD ratio from ~170% to ~140% will have a significant negative impact on margins and ROE. Management intends to achieve this by running down a large proportion of its loan book (we estimate ~20% is necessary). We simply cannot see how letting such a huge amount of loans expire without being replaced will not be detrimental to NII and NIM.

Regarding other notable aspects of the Dupont profile, we see that Lloyds has already established significantly better efficiency versus its peers as it benefits from the integration of HBOS. We believe that Lloyds will continue to have a cost efficiency advantage going forward as the integration process continues.

Nordea We see Nordea as quite typical of the Nordic banks in general, in terms of having a relatively low NIM, but offset by being very cost efficient. The ROA is very similar to Handelsbanken’s but has come down from higher normalised levels.

In most respects, however, Nordea is inferior to Handelsbanken when seen through the lens of Dupont. The NIM has not performed as well as Handelsbanken’s over the course of the crisis. We believe this is mainly because Handelsbanken has been better at accumulating cheap customer deposits, which has lowered its funding costs. This is reflected in the better development of its LTD ratio.

Nordea is not as cost efficient as Handelsbanken, but remains one of the most efficient in the group nonetheless. Also, notably, Nordea has enjoyed good momentum in reducing costs as a percentage of its asset base. We believe that this can be maintained for a few quarters more.

Nordea’s loan loss experience has been worse than Handelsbanken’s due to its greater exposure to Eastern Europe (albeit still small at only 6% of the loan book). Whilst we believe that Nordic loan losses will improve, loan losses in Eastern Europe are expected to take longer to reach normalised levels, mainly because of the acute economic imbalances in the Baltic countries.

Santander Santander, like BBVA, is a high ROE bank. Historically, its ROE has been much more stable than the ROE of its Spanish peer, by virtue of a lower propensity to book one-off gains/losses. We do not model greater volatility in BBVA’s ROE (since, of course, it is impossible to model one-offs far into the future) but point out that this is likely to be the case going forward.

Like BBVA, Santander has enjoyed an improving NIM over the course of the crisis, and indeed since the beginning of 2006. Falling base rates and asset repricing have helped asset spreads. Over the course of the financial crisis, we can also conclude that the fall in the LTD ratio (as deposits have undertaken a flight to quality) has helped reduce Santander’s funding costs.

19 JANUARY 2010 61

PAN-EUROPEAN BANKS

Going forward, we are concerned that the NIM will come under pressure as the lagged fall in asset repricing now comes to bear and the improvement in the LTD ratio comes to a halt. Looking further ahead, a modest increase in base rates will lead to an initial compression in NIM as asset spreads are compressed and the expansion in liability spreads fails to compensate given that Santander still has a relatively high reliance on wholesale funding rather than deposit funding (its LTD ratio is 146%).

We are concerned that loan losses in Spain will continue to be elevated for a prolonged time, which we take account of in our earnings model. Loan losses (net of releases from generic provisions) have increased as the Spanish economy has weakened. The coverage ratio has been run down to a very low 73%. We believe the market will be wary of coverage ratios being run down even more as loan losses remain at high levels. Our concern with the Spanish economy rests with the fact that consensus forecasts a 200bps increase in unemployment to over 20% in 2020, whilst the budget deficit balloons to over 10% as the government tries to stimulate an economic recovery. The main mitigating factor for Santander versus BBVA is that Iberia comprises a smaller proportion of the group’s total loan book (49% versus 62% for BBVA).

Standard Chartered Standard Chartered’s focus on Asia has put it in good stead throughout the course of the crisis. ROE and ROA have virtually remained at pre-crisis levels. However, within that, the quality of the earnings has changed for the worse.

Analysing the components of ROA using Dupont, we see that the NIM has fallen slightly. Looking more closely at the earnings for Standard Chartered, weakness in the NIM and lending volumes in Consumer Banking has been mainly due to: compressed liability spreads as base rates have fallen; a devaluation of the Korean Won, a shift in focus to secured lending from unsecured lending. The weakness in consumer Banking has been more than offset by the strength in NIM and lending growth in the Wholesale Banking division, where dislocation in the markets and the retrenchment of competitors has led to market share gains and expanding asset spreads.

There has also been a marked increase in low quality trading profits (as can be seen by the increase in non-NII operating income % assets). We do not believe these trading profits can be sustained at such high levels and we model a decrease to more normalised levels.

On the plus side, Standard Chartered, like HSBC, has been able to accumulate a vast amount of cheap deposit funding, such that its LTD ratio now stands at 80%. We believe that the benefits of this are twofold: the bank will be able to improve its NIM going forward as it uses this excess of deposits to fund high margin new lending (i.e. a positive mix effect); and secondly, loan growth is unencumbered by funding constraints, which is a significant relative advantage in an Asian market hungry for credit. We therefore model for Standard Chartered continued superior loan growth, as well as an improving NIM.

If base rates increase, then this will be an additional relative positive for Standard Chartered. Given its excess of deposits, funding costs will remain very low and liability spreads will expand. Asset spreads will gradually expand as assets are repriced, resulting in a net increase in interest margins.

19 JANUARY 2010 62

PAN-EUROPEAN BANKS

Standard Chartered stands out as the least efficient bank in terms of costs% assets. We are not sure why this is the case, but hazard a guess that the disparate geographical nature of its operations means that it is more difficult to use centralised IT systems that would make it more efficient. We nevertheless model an improvement in its efficiency going forward, on the basis that its asset base should grow substantially more than its cost base (i.e. as a high growth company, it should benefit incrementally from economies of scale).

Svenska Handelsbanken Handelsbanken is an exceptional bank in quite a few respects. It is, in our opinion, the most risk averse in the group, as denoted by its exceptionally low LLCs. We believe it is able to achieve this due to its decentralised system, whereby loans are approved by a local branch manager who has astute knowledge of the client and of the local market. Going forward, we see a very modest improvement in loan losses, given that deterioration has been so modest in the first place.

On the flipside, it appears to us that the adherence to low risk lending principles also means that the margins on lending are low (i.e. it lends at low competitive rates only to high quality clients). This is reflected in the fact that Handelsbanken has the lowest NIM in the group.

We see, additionally, that the NIM has improved very slightly over the course of the crisis, and is discernibly better than its Nordic peers. We attribute this to Handelsbanken being the better deposit taking franchise. It has, over the course of the crisis, been able to attract more cheap customer deposits than its peers, as reflected by the relative improvement in it LTD ratio. We believe this is the key reason for the relative improvement in Handelsbanken’s NIM versus DNBNOR and particularly, Nordea.

Whilst the LTD ratio appears very high, it is not reflective of a reliance on short term wholesale funding. Common to other Nordic banks, a substantial proportion of funding is comprised of long term covered bonds and medium term CP and CD securities. Handelsbanken has actually been a net lender to the Swedish government and the central bank throughout the crisis.

Handelsbanken more than makes up for its low NIM by being the lowest cost operator. The Nordic financials have long been noted for their cost efficiency, but Handelsbanken excels even within this lauded company.

The net result is that, in spite of the low NIM, Handelsbanken has quite an attractive, and robust, ROE. It should come as no surprise that Handelsbanken’s ROE has not deteriorated that much over the course of the credit crisis. The potential improvement is limited in our opinion, driven by the slight normalisation in loan losses and its funding costs becoming relatively cheaper as it capitalises on its superior deposit taking ability. However, the modest positive delta is a credit to the fact that management already runs the bank exceptionally well.

Unicredit In our opinion, Unicredit shares with its Italian peer, Intesa, a few similarities that mark it out as a structurally low return bank. The NIM is structurally low compared to other banks in the group, and has suffered from the fall in the base rate (which has led to a compression of the deposit margin). Given that we see little potential for increases in the base rate, the NIM is likely to stay at low levels.

19 JANUARY 2010 63

PAN-EUROPEAN BANKS

We believe that Unicredit is subject to relatively robust competition for deposits and loans in Italy, which we deduce from comparing the development of NIM and LTD ratios with those of HSBC. This is the same conclusion as we have reached for Intesa. As such, neither Intesa nor Unicredit have been able to improve their LTD ratios during the course of the crisis, which will be a relative negative in terms of funding loan growth.

Unlike the Nordic banks, which also have low NIMs, (and again similar to Intesa), Unicredit does not have as efficient a cost base, resulting in its ROE being in the low single digits whilst those for the Nordic banks are in the low to mid teens.

Unicredit is also hampered by elevated loan losses from its CEE operations. We do not see these improving quickly. The loan losses in the region mainly arise from Russia and Ukraine, where we foresee a more prolonged road to recovery in asset quality.

The basis for concluding that there is only modest improvement in the ROE is therefore: low NIM for a prolonged period; elevated loan losses in the CEE region for an extended period of time; a fall in trading profits from the significant levels achieved in 2009.

19 JANUARY 2010 64

PAN-EUROPEAN BANKS

VALUATION

Our DCF valuation is linked to our We adopt a discounted cash flow (DCF) valuation approach for the lending banks. Dupont analysis and long-term financial We like this approach because it can be linked to our Dupont analysis (which we models to provide a sanity check that have modelled up until the end of 2012) and to our long-term earnings and balance the valuation methodology is grounded sheet models for each and every bank. This provides a sanity check that the in realistic earnings assumptions. earnings growth and balance sheet metrics of the banks make sense on an absolute and relative basis.

Assumptions Used for Consistency of Valuation Methodology We have worked to ensure consistency in our valuation methodology across the banks in the group. The first three years of the models, over 2010–2012, are characterised by our stock specific assumptions regarding the various drivers of earnings growth. Thereafter, we assume a fading of annual pre-tax profit growth of 15% and 10% in 2013 and 2014 respectively. We assume long-term annual pre-tax profit growth of 5%. The long-term dividend payout ratio is set at 40%.

Cost of Equity Derived from CAPM We use the Capital Asset Pricing Model to derive the cost of equity for each bank.

The risk free rate is taken as the yield on the 30-year European generic bond.

Equity beta is derived direct from Bloomberg. It is calculated by comparing the price movements of the security and the representative market index for the past two years. The market indices used are the main national indices of the market where the stock is listed; for BBVA for example, the index is the IBEX 35, whilst for HSBC it is the FTSE 100.

The long-term annual equity market return is assumed to be 8.5%.

Figure 77: DCF Valuation Table (Stocks ordered alphabetically by ticker) DCF FV Cost of Equity Assumptions Stock name Matrix Mcap Curr Price DCF DCF Upside/ Risk-Free LT Market Market risk Equity LT Growth COE Terminal Rating (€bn) (local) FV P/FV Downside Rate Return premium Beta* Rate Value % FV BBVA HOLD 48.3 EUR 12.81 16.00 80% 25% 4.02% 8.50% 4.48% 1.47 5.00% 10.60% 35% DNB NOR ASA BUY 13.9 NOK 69.50 85.74 81% 23% 4.02% 8.50% 4.48% 1.24 5.00% 9.56% 43% HSBC HLDGS PLC HOLD 139.2 GBp 703 915 77% 30% 4.02% 8.50% 4.48% 1.15 5.00% 9.15% 46% INTESA SANPAOLO HOLD 39.2 EUR 3.12 3.57 87% 14% 4.02% 8.50% 4.48% 1.29 5.00% 9.81% 41% LLOYDS BANKING REDUCE 41.9 GBp 56.78 49.06 116% -14% 4.02% 8.50% 4.48% 1.66 5.00% 11.45% 32% NORDEA BANK AB BUY 29.5 SEK 73.80 85.66 86% 16% 4.02% 8.50% 4.48% 1.27 5.00% 9.71% 41% BANCO SANTANDER HOLD 95.3 EUR 11.50 12.84 90% 12% 4.02% 8.50% 4.48% 1.48 5.00% 10.63% 35% SVENSKA HAN-A BUY 12.3 SEK 199.50 245.70 81% 23% 4.02% 8.50% 4.48% 1.14 5.00% 9.12% 47% STANDARD CHARTER HOLD 35.9 GBp 1552 1485 105% -4% 4.01% 8.50% 4.49% 1.55 5.00% 10.99% 32% UNICREDIT SPA REDUCE 43.3 EUR 2.25 2.00 112% -11% 4.02% 8.50% 4.48% 1.68 5.00% 11.57% 30%

Source: Matrix Corporate Capital Research

We also include below a conventional valuation table (using Matrix estimated earnings and TNAV) to act as a sanity check against our DCF valuation.

The DCF valuation is used as the basis The conventional valuation table displays our targets prices as well. We use the DCF for our target prices in most cases, valuation of the banks as a guide in determining our target prices in most cases. The except BBVA and HSBC. exceptions are BBVA and HSBC.

For BBVA, we see substantial long-term value from our DCF model, which is in keeping with our view that it is a well run bank with a high structural ROE and adequate capital. However, near term, we are quite concerned about the development of loan losses relative to the peer group, which is largely down to fears that Spain faces significant economic headwinds. We would look to become more

19 JANUARY 2010 65

PAN-EUROPEAN BANKS

positive on a target price for BBVA once we can gain clarity that the bottom in credit quality deterioration has been reached.

For HSBC, there appears to be substantial valuation upside. However, we do not incorporate the impact of the Basel III proposals on the bank’s capital ratios in our earnings model. Were we to do so, then we would see HSBC only just meeting minimum capital requirements in 2012. The absence of excess capital versus the other banks in the group, and the potential desire of the bank to raise more in order to establish a buffer above the minimum requirement, means that we envisage only modest upside to the current price.

Figure 78: Conventional Valuation Table Stock name Matrix Mcap Curr Price Target Upside/ Matrix EPS Matrix P/E Matrix TNAVPS Matrix P/TNAV Matrix ROTNAV Rating (€bn) (local) Price Downside FY10E FY11E FY10E FY11E FY10E FY11E FY10E FY11E FY10E FY11E BBVA HOLD 48.3 EUR 12.81 14.25 11% 1.22 1.46 10.51 8.80 6.19 6.86 2.07 1.87 20.4% 22.1% DNB NOR ASA BUY 13.9 NOK 69.50 86.00 24% 5.47 6.92 12.70 10.04 57.71 62.04 1.20 1.12 9.8% 11.6% HSBC HLDGS PLC HOLD 139.2 GBp 703 800 14% 0.697 1.028 16.45 11.16 5.62 6.17 2.04 1.86 13.1% 17.7% INTESA SANPAOLO HOLD 39.2 EUR 3.12 3.55 14% 0.231 0.286 13.47 10.89 2.50 2.73 1.25 1.14 10.4% 11.8% LLOYDS BANKING REDUCE 41.9 GBp 56.78 49.00 -14% -0.45 4.24 -126.21 13.41 52.63 57.05 1.08 1.00 -0.8% 7.7% NORDEA BANK AB BUY 29.5 SEK 73.80 86.60 17% 0.654 0.783 11.14 9.31 4.77 5.22 1.53 1.40 14.3% 15.7% BANCO SANTANDER HOLD 95.3 EUR 11.50 13.00 13% 1.02 1.31 11.28 8.77 6.05 6.84 1.90 1.68 16.8% 19.2% SVENSKA HAN-A BUY 12.3 SEK 199.50 250.00 25% 17.19 19.68 11.60 10.13 127.30 139.62 1.57 1.43 14.2% 14.7% STANDARD CHARTER HOLD 35.9 GBp 1552 1500 -3% 2.25 2.71 11.27 9.36 12.09 14.76 2.10 1.72 22.5% 22.1% UNICREDIT SPA REDUCE 43.3 EUR 2.25 2.00 -11% 0.093 0.181 24.08 12.37 2.19 2.38 1.03 0.94 4.4% 8.0% AVERAGE 9% -0.4 10.4 27.7 30.4 1.58 1.42 12.5% 15.1%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 66

PAN-EUROPEAN BANKS

COMPANY SUMMARIES

This page has been left blank intentionally.

19 JANUARY 2010 67

PAN-EUROPEAN BANKS

BBVA

Figure 79: BBVA Valuation Table BBVA 17/01/2010 Target Price: 14.25 Rating: HOLD Stock ticker BBVA SM DCF Fair Value 16.00 FY 2009E FY 2010E FY 2011E FY 2012E Currency EUR Price/DCF FV80% EPS 1.35 1.22 1.462.06 Price 12.81 Upside/Downside 25% P/E 9.47 10.51 8.806.22 Market Cap (€bn) 48.3 DCF Assumptions TNAV per share 5.65 6.19 6.86 8.08 Dividend Yield FY 2010E 3.77% Risk-free rate 4.02% P/TNAV 2.27 2.07 1.87 1.59 PEG Ratio 0.35 LT Market Return 8.5% ROE 18.24% 14.69% 16.00% 20.29% Loans % Deposits FY 2010E 128% Equity beta 1.47 ROTNAV 26.19% 20.41% 22.12% 27.33% Assets % Equity FY 2010E 18.4 COE 10.6% ROA 0.92% 0.80% 0.88% 1.14% Tier 1 Ratio FY 2010E 10.5% Long-term growth rate 5.0% Core Tier 1 Ratio 8.06% 8.26% 8.56% 9.40%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on BBVA with a HOLD rating.

BBVA has a high ROE versus peers by virtue of its LATAM (mainly Mexico) exposure. We believe its LATAM businesses will remain robust. However, we forecast that BBVA will struggle to post positive earnings growth versus the peer group of lending banks in 2010. More specifically:

• BBVA benefitted from a €830m capital gain from the sale and leaseback of properties, which it booked wholly to generic provisions, resulting also in a sharp spike in the provisioning rate. The market has largely interpreted this as a one-off provisioning, but look closer and one will see that the absolute level of provisions did not actually increase much and that the coverage ratio stayed the same QoQ at 68%. Management tells us that the generic provisions were used up by specific losses. We are therefore concerned that the underlying loss rate is actually worse than the market expects. • We are generally concerned about the economic prospects of Spain, which is Divisional breakdown of group loans, set to post the worst GDP growth in developed Europe in 2010 according to 9M09 consensus estimates, and where the unemployment rate should exceed 20%. South America USA 7% Our Basel III Analysis points to adequate capital adequacy, but note that the Core 11% Tier 1 ratio is the third lowest in the peer group. We estimate that on a Basel III Mexico basis, it will be 7.0% by the end of 2012. (Bancomer) 8%

Wholesale Spain & Our DCF valuation points to significant upside, but our near-term concerns lead us to Banking & Portugal Asset 62% have only a modest target price for now. We would be looking to upgrade to BUY Management 12% once we have transparency that the asset quality and economic situation in Spain is close to reaching a turning point.

Company Description Divisional breakdown of operating profit, BBVA is comprised of 5 business units. Chief amongst them is the Spain and 9M09 Portugal division, comprising retail and corporate banking activities. It accounts for South America 62% of group loans, but only 37% of operating profit due to its lower ROE versus the 11% USA LATAM divisions. BBVA’s LATAM exposure is mainly via the Mexico division (where 4% the Bancomer subsidiary is Mexico’s biggest bank). Mexico accounts for only 8% of Mexico (Bancomer) the group loan book but 27% of operating profit. Other LATAM operations are 17% grouped together in the South America division. The global Wholesale Business Spain & Portugal division includes investment banking, asset management and private banking Wholesale 61% Banking & operations. The USA division, which is comprised of Compass Bank and three Asset Management smaller Texan banks, accounts for 7% of operating profit. BBVA also has a 15% 7% stake in China Citic Bank.

19 JANUARY 2010 68

PAN-EUROPEAN BANKS

Figure 80: BBVA Financial Summary BBVA FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (€m) Net interest income 11,686 13,718 14,233 15,057 15,892 Net fees & commissions 4,527 4,400 4,510 4,735 4,972 Trading profits/losses 1,558 1,277 1,354 1,381 1,408 Other revenue 1,206 928 1,075 1,101 1,128 Total operating revenues 18,977 20,323 21,171 22,274 23,400 Operating costs -8,455 -8,127 -8,765 -9,354 -9,843 Operating profit 10,522 12,196 12,406 12,920 13,557

Total provisions -4,371 -5,626 -6,531 -5,811 -3,307 Investment income 775 566 461 461 461 Impairments on other assets 0 0 0 0 0 Goodwill impairment 0 0 0 0 0 Pre-tax profit 6,926 7,136 6,336 7,570 10,710

Taxes -1,541 -1,703 -1,521 -1,817-2,571 Minorities -365 -407 -289 -345-488 Other non-operating items 0 0 0 0 0 Net profit 5,020 5,026 4,526 5,408 7,652

Assets (€m) Loans to customers 335,260 325,355 350,185 379,365 411,781 Interbank loans 34,234 22,777 24,655 26,687 28,887 Total securities 128,115 144,010 155,881 168,731 182,640 Intangible assets 8,439 8,292 8,976 9,715 9,715 Total assets 542,650 544,968 590,179 645,250 700,112 Net interest-earning assets 510,514 510,621 550,722 597,399 649,997

Liabilities (€m) Interbank borrowings 66,804 75,253 81,456 88,170 95,439 Customer deposits 383,800 378,187 409,362 448,935 485,942 Total shareholders' equity 25,656 29,454 32,169 35,414 40,005 Tangible net asset value 17,217 21,162 23,194 25,699 30,290

Important Financial Ratios ROA 0.96% 0.92% 0.80% 0.88%1.14% ROE 19.04% 18.24% 14.69% 16.00%20.29% ROTNAV 27.86% 26.19% 20.41% 22.12%27.33% Cost/income -44.55% -39.99% -41.40% -42.00% -42.06% Tax rate -22.25% -23.86% -24.00% -24.00% -24.00% Payout 37.41% 30.07% 40.00% 40.00%40.00% Net interest margin 2.37% 2.69% 2.68% 2.62% 2.55% LLC % gross loans -0.89% -1.57% -1.79% -1.46% -0.73% Non-performing loans % gross loans 2.46% 4.00% 4.25% 3.50% 2.40% NPL coverage ratio 88.68% 67.00% 76.00% 90.00% 120.00% Loans % deposits 131.35% 129.18% 128.45% 126.07% 126.42% Tier 1 ratio 7.90% 10.47% 10.53% 10.70% 11.42% Core tier 1 ratio 6.23% 8.06% 8.26% 8.56% 9.40% Equity % total assets 4.73% 5.40% 5.45% 5.49% 5.71% RWA % total assets 52.96% 52.78% 51.72% 50.20% 49.10%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 69

PAN-EUROPEAN BANKS

DNBNOR

Figure 81: DNBNOR Valuation Table DNB NOR ASA 17/01/2010 Target Price: 86.00 Rating: BUY Stock ticker DNBNOR NO DCF Fair Value 85.74 FY 2009E FY 2010E FY 2011E FY 2012E Currency NOK Price/DCF FV81% EPS 6.10 5.47 6.928.80 Price 69.50 Upside/Downside 23% P/E 11.39 12.70 10.047.90 Market Cap (€bn) 13.9 DCF Assumptions TNAV per share 53.98 57.71 62.04 67.54 Dividend Yield FY 2010E 2.51% Risk-free rate 4.02% P/TNAV 1.29 1.20 1.12 1.03 PEG Ratio 0.47 LT Market Return 8.5% ROE 9.90% 8.97% 10.64% 12.57% Loans % Deposits FY 2010E 189% Equity beta 1.24 ROTNAV 10.97% 9.80% 11.56% 13.57% Assets % Equity FY 2010E 19.4 COE 9.6% ROA 0.46% 0.46% 0.55% 0.65% Tier 1 Ratio FY 2010E 8.8% Long-term growth rate 5.0% Core Tier 1 Ratio 7.82% 8.05% 8.34% 8.75%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on DNBNOR with a BUY rating.

Operating predominately in a strong Norwegian economy, DNBNOR is set to benefit from better loan growth and improving asset quality than most of its peers. Norway is expected to have GDP growth in 2010 of 2.3% according to consensus forecasts, the highest in developed Europe apart from Sweden.

The positive momentum in Norway should more than outweigh the expected fall in trading profits and continued high loan losses from DNB NORD, the subsidiary comprising its Eastern European operations. We see nearly 10% better pre-tax income growth in 2010 versus 2009, as shown in the Financial Summary table.

Our Basel III Analysis indicates that the Core Tier 1 ratio of the bank will be one of the highest in the group, at approximately 10% by the end of 2012. (Note that in using the Divisional breakdown of group loans, 9M09 full Basel II Core Tier 1 ratio as a starting point in our analysis, we assume that there will be a 2.0% uplift to the published transitional Core Tier 1 ratio, as has been DnB NORD 7% indicated in past earnings reports as the approximate benefit). If we use an anticipated minimum Core Tier 1 ratio of 6.0%, then this points to substantial excess capital. Large Corporates and International Our DCF valuation points to considerable upside. DNBNOR also has one of the 31% Retail Banking lowest P/TNAV in the group. This is not warranted in our opinion given the attractive 62% ROE and low cost of capital, which we see as underpinned by the relatively small capital, liquidity and earnings risks.

Company Description Divisional breakdown of operating profit, 9M09 DNBNOR is Norway’s largest financial services group. The bank is broadly split into Life and Asset Retail Banking, Large Corporate and International Operations (mainly wholesale Management Retail Banking 6% 24% lending to the group’s largest Norwegian and international customers), life insurance DnB NOR and asset management (Vital being the key life business), DNBNOR Markets Markets 27% (Norway’s largest provider of securities and investment banking services) and DNB NORD (the 51% owned subsidiary which accounts for DNBNOR’s Eastern European operations; the core markets being Estonia, Latvia, Lithuania and Poland). Large Corporates and International NORD L/B owns the other 49% of DNB NORD. DNBNOR is currently undertaking an 43% evaluation of DNB NORD, ending 31 July 2010, after which it will have the right to acquire NORD L/B’s interest, but in that event, NORD L/B would have the right to take over DNB NORD’s Polish operations. If DNBNOR chooses not to acquire NORD L/B’s interest in DNB NORD, NORD/LB will be entitled to transfer its ownership interest in DNB NORD to DNBNOR or to take over the ownership interest of DNBNOR.

19 JANUARY 2010 70

PAN-EUROPEAN BANKS

Figure 82: DNBNOR Financial Summary DNB NOR ASA FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (NOK m) Net interest income 21,910 22,854 24,048 25,365 27,218 Net fees & commissions 6,895 6,593 6,792 7,029 7,310 Trading profits/losses 3,340 6,278 5,154 5,257 5,362 Other revenue 2,202 2,977 3,830 3,959 4,092 Total operating revenues 34,347 38,702 39,823 41,609 43,982 Operating costs -18,719 -19,215 -19,795 -20,556 -21,676 Operating profit 15,628 19,486 20,028 21,053 22,307

Total provisions -3,509 -8,414 -7,874 -5,162 -2,915 Impairments on other assets 0 0 0 0 0 Goodwill impairment 0 0 0 0 0 Net gains on fixed and intangible assets 52 7 0 0 0 Pre-tax profit 12,171 11,079 12,153 15,892 19,392

Taxes -3,252 -3,965 -3,646 -4,450-4,848 Minorities 293 1,468 406 -172-218 Other non-operating items 0 0 0 0 0 Net profit 9,212 8,582 8,914 11,270 14,326

Assets (NOK m) Loans to customers 1,191,635 1,133,544 1,181,477 1,246,629 1,349,391 Interbank loans 59,717 71,049 73,934 76,936 80,060 Total securities 334,020 480,203 519,787 562,634 609,013 Intangible assets 8,480 8,409 8,409 8,409 8,409 Total assets 1,831,699 1,882,750 1,973,111 2,118,016 2,275,673 Net interest-earning assets 1,599,680 1,674,301 1,762,212 1,870,451 2,019,657

Liabilities (NOK m) Interbank borrowings 178,822 300,078 312,262 324,942 338,135 Customer deposits 1,203,464 1,126,571 1,172,314 1,268,951 1,373,553 Total shareholders' equity 77,065 96,334 102,403 109,455 118,426 Tangible net asset value 68,585 87,925 93,994 101,046 110,017

Important Financial Ratios ROA 0.56% 0.46% 0.46% 0.55%0.65% ROE 12.25% 9.90% 8.97% 10.64%12.57% ROTNAV 13.73% 10.97% 9.80% 11.56%13.57% Cost/income -54.50% -49.65% -49.71% -49.40% -49.28% Tax rate -26.72% -35.79% -30.00% -28.00% -25.00% Payout 0.00% 32.15% 31.91% 37.42%37.38% Net interest margin 1.51% 1.40% 1.40% 1.40% 1.40% LLC % gross loans -0.33% -0.72% -0.67% -0.42% -0.22% Non-performing loans % gross loans 1.11% 2.00% 1.60% 0.80% 0.80% NPL coverage ratio 45.50% 48.00% 50.00% 50.00% 50.00% Loans % deposits 199.52% 188.77% 189.08% 184.31% 184.31% Tier 1 ratio 6.66% 8.62% 8.82% 9.08% 9.46% Core tier 1 ratio 5.84% 7.82% 8.05% 8.34% 8.75% Equity % total assets 4.21% 5.12% 5.19% 5.17% 5.20% RWA % total assets 65.55% 57.52% 57.12% 55.37% 53.63%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 71

PAN-EUROPEAN BANKS

HANDELSBANKEN

Figure 83: Svenska Handelsbanken Valuation Table SVENSKA HAN-A 17/01/2010 Target Price: 250.00 Rating: BUY Stock ticker SHBA SS DCF Fair Value 245.70 FY 2009E FY 2010E FY 2011E FY 2012E Currency SEK Price/DCF FV81% EPS 16.46 17.19 19.68 23.27 Price 199.50 Upside/Downside 23% P/E 12.12 11.60 10.13 8.57 Market Cap (€bn) 12.3 DCF Assumptions TNAV per share 115.40 127.30 139.62 155.06 Dividend Yield FY 2010E 2.48% Risk-free rate 4.02% P/TNAV 1.73 1.57 1.43 1.29 PEG Ratio 0.71 LT Market Return 8.5% ROE 13.41% 12.92% 13.50% 14.53% Loans % Deposits FY 2010E 232% Equity beta 1.14 ROTNAV 14.78% 14.17% 14.75% 15.79% Assets % Equity FY 2010E 27.4 COE 9.1% ROA 0.48% 0.47% 0.50% 0.54% Tier 1 Ratio FY 2010E 14.1% Long-term growth rate 5.0% Core Tier 1 Ratio 11.09% 11.85% 12.14% 12.49%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Svenska Handelsbanken with a BUY rating.

Handelsbanken strikes us as an exceptional bank in a number of respects. In our opinion it is the lowest risk lender, shown by having the most benign loan loss experience in the group over the course of the crisis. We believe its focus on risk aversion and a decentralised branch system results in a very low net interest margin (i.e. it lends at competitive rates to the highest quality clients) but that this is more than made up by the fact that it is also the lowest cost operator, resulting in an attractive ROE in the mid teens.

Sweden is expected to have the highest GDP growth in Europe in 2010 according to consensus forecasts, underpinning our anticipation of good loan growth and a modest improvement in loan losses. Additionally, our Dupont Analysis shows that Handelsbanken has been able to improve its deposit funding slightly versus its Nordic peers (shown by the fall in its loan to deposit ratio), which has been one of the drivers of the relative improvement in its net interest margin. We do not model a continuation of this relative improvement in NIM in the near term but acknowledge that Handelsbanken may further capitalise on its superior deposit taking franchise.

Our Basel III Analysis shows that Handelsbanken is one of the best capitalised in the group. The Core Tier 1 ratio is approximately 10% at the end of 2012, indicating to us that it has substantial excess capital.

A DCF valuation of the bank indicates considerable upside. The bank is in line with the group average on conventional P/E and P/TNAV metrics, but a premium is warranted given a very low cost of capital, underpinned by its capital strength, risk aversion and attractive Swedish footprint.

Company Description Divisional breakdown of operating profit, Svenska Handelsbanken is keen to highlight its decentralised branch network. Each 9M09 branch is responsible for all customers within its geographic area, including the Handelsbank Handelsbank en Capital en Asset largest companies. The main division in the group is simply called ‘Bank Operations’, markets Management 16% 1% accounting for 83% of group operating profit. Other divisions include Asset Management and Capital Markets.

Handelsbanken has branches across the Nordic region and also in Great Britain. Bank Sweden is its home market. The bank has 461 branches in Sweden, 54 in Denmark, operations total 45 in Finland, 48 in Norway and 66 in Great Britain. 83%

19 JANUARY 2010 72

PAN-EUROPEAN BANKS

Figure 84: Handelsbanken Financial Summary SVENSKA HAN-A FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (SEK m) Net interest income 19,223 22,232 22,788 24,633 26,900 Net fees & commissions 6,795 7,092 7,447 7,819 8,210 Trading profits/losses 3,169 2,852 2,675 2,755 2,838 Other revenue 975 487 500 511 522 Total operating revenues 30,162 32,662 33,410 35,718 38,469 Operating costs -13,137 -14,820 -15,067 -16,107 -17,342 Operating profit 17,025 17,842 18,342 19,610 21,128

Total provisions -1,605 -3,732 -3,859 -3,027 -1,524 Impairments on other assets 0 0 0 0 0 Goodwill impairment -92 -84 0 0 0 Other income / (expense) -2 1 0 0 0 Pre-tax profit 15,326 14,027 14,483 16,584 19,604

Taxes -3,382 -3,712 -3,766 -4,312-5,097 Minorities 0 0 0 00 Other non-operating items 0 19 0 0 0 Net profit 11,944 10,333 10,718 12,272 14,507

Assets (SEK m) Loans to customers 1,481,475 1,458,797 1,537,119 1,670,900 1,814,105 Interbank loans 164,981 153,809 163,223 176,678 191,242 Total securities 239,480 183,002 198,088 214,416 232,091 Intangible assets 7,057 7,179 7,395 8,004 8,004 Total assets 2,158,784 2,180,415 2,357,009 2,566,602 2,795,591 Net interest-earning assets 2,087,556 2,026,696 2,144,546 2,337,058 2,554,959

Liabilities (SEK m) Interbank borrowings 319,113 201,203 213,539 231,142 250,195 Customer deposits 1,439,469 1,594,187 1,725,599 1,882,001 2,052,765 Total shareholders' equity 74,963 79,127 86,756 95,050 104,676 Tangible net asset value 67,906 71,947 79,361 87,046 96,672

Important Financial Ratios ROA 0.59% 0.48% 0.47% 0.50%0.54% ROE 15.98% 13.41% 12.92% 13.50%14.53% ROTNAV 17.55% 14.78% 14.17% 14.75%15.79% Cost/income -43.55% -45.37% -45.10% -45.10% -45.08% Tax rate -22.07% -26.47% -26.00% -26.00% -26.00% Payout 36.54% 31.52% 28.82% 32.41%33.64% Net interest margin 1.03% 1.08% 1.09% 1.10% 1.10% LLC % gross loans -0.12% -0.25% -0.26% -0.19% -0.09% Non-performing loans % gross loans 0.36% 0.90% 2.00% 1.30% 0.80% NPL coverage ratio 51.07% 55.00% 60.00% 60.00% 60.00% Loans % deposits 272.45% 238.47% 232.14% 228.61% 224.86% Tier 1 ratio 10.51% 13.48% 14.14% 14.26% 14.45% Core tier 1 ratio 8.90% 11.09% 11.85% 12.14% 12.49% Equity % total assets 3.47% 3.63% 3.68% 3.70% 3.74% RWA % total assets 33.44% 28.37% 27.30% 27.14% 26.97%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 73

PAN-EUROPEAN BANKS

HSBC

Figure 85: HSBC Valuation Table HSBC HLDGS PLC 17/01/2010 Target Price: 800 Rating: HOLD Stock ticker HSBA LN DCF Fair Value 914.72 FY 2009E FY 2010E FY 2011E FY 2012E Currency GBp Price/DCF FV77% EPS 0.48 0.70 1.03 1.41 Price 702.50 Upside/Downside 30% P/E 23.75 16.46 11.17 8.17 Market Cap (€bn) 139.4 DCF Assumptions TNAV per share 5.27 5.62 6.17 6.94 Dividend Yield FY 2010E 3.97% Risk-free rate 4.02% P/TNAV 2.18 2.04 1.86 1.65 PEG Ratio 0.39 LT Market Return 8.5% ROE 6.89% 9.93% 13.56% 16.86% Loans % Deposits FY 2010E 76% Equity beta 1.15 ROTNAV 9.37% 13.09% 17.70% 21.69% Assets % Equity FY 2010E 21.0 COE 9.2% ROA 0.29% 0.47% 0.65% 0.82% Tier 1 Ratio FY 2010E 10.5% Long-term growth rate 5.0% Core Tier 1 Ratio 8.90% 9.17% 9.68% 10.43%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on HSBC with a HOLD rating. We see HSBC as a very strong deposit franchise in the group of lending banks. The ROE is currently depressed by elevated loan losses, but we see this situation alleviating quickly. From a Dupont perspective, the improvement in the ROE comes from an early normalisation in the loan loss experience, continued growth in Asia, and having the advantage of a large excess of deposits, which can be used write high margin new loans.

Geographical breakdown of operating • HSBC’s loan loss experience should improve earlier than peers. Recent profit, H1 2009 results point to the first reduction in impairments in the US Household Latin America 11% Personal Finance division since the beginning of 2006. The US loan book is, Europe 32% in our opinion, relatively mature given its heavy weighting towards early cycle subprime mortgages. North America • 27% HSBC has been able to improve its LTD ratio from just below 100% at the Middle East beginning of 2007 to 79% as of H1 2009, as deposits have undergone a flight 5% Hong Kong to quality. Our Dupont analysis shows that this has enabled HSBC to maintain 15% Rest of Asia Pacific a steady NIM throughout the crisis, whilst many expected the fall in base rates 10% to have led to a contraction.

Divisional breakdown of group loans, H1 The excess of deposits over loans means that HSBC has considerable 2009 capacity to write high margin new loans to increase its NII and NIM (via a Private positive mix effect). Banking Other 0% 4% Our key concern with HSBC, however, (and the reason we initiate with a Hold rating Global Banking and Personal Markets Financial and not a Buy) is that under our Basel III Analysis, the bank’s Core Tier 1 ratio falls 31% Services 43% significantly to only 6.0% by the end of 2012. This is considerably less than peers and may lead management to ultimately consider raising equity to regain parity in capital strength.

Commercial Banking 22% Our DCF valuation points to attractive upside. However, given the results of our Basel III Analysis, we believe there is significant concern to warrant a target price only modestly higher than the current share price. Divisional breakdown of operating profit, H1 2009 Company Description Private Banking HSBC is a truly global bank with operations in Europe, Asia, the Americas, Middle 3% East and Africa. Asia, accounting for 25% of operating profit (including Hong Kong) Global Personal is a key part of the franchise given its growth potential. HSBC manages its business Banking and Financial Markets Services through two customer groups, Personal Financial Services and Commercial Banking 37% 43% (accounting for 37% and 15% of operating profit respectively); and two global businesses, Global Banking & Markets and Private Banking (accounting for 32% and 3% of operating profits). PFS incorporates the Group’s consumer finance Commercial Banking businesses. HSBC plans to list in Shanghai in the first half of 2010. 17%

19 JANUARY 2010 74

PAN-EUROPEAN BANKS

Figure 86: HSBC Financial Summary HSBC HLDGS PLC FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (US$m) Net interest income 42,563 42,072 46,455 51,286 56,085 Net fees & commissions 20,024 17,913 18,808 19,749 20,736 Trading profits/losses 6,560 10,344 8,672 8,672 8,672 Other revenue 13,999 3,642 6,520 6,994 7,294 Total operating revenues 83,146 73,971 80,455 86,701 92,787 Operating costs -38,535 -34,600 -37,426 -40,201 -42,912 Operating profit 44,611 39,371 43,030 46,500 49,875

Total provisions -24,937 -29,169 -26,040 -20,987 -14,624 Investment income 197 823 1,400 1,400 1,400 Impairments on other assets 0 0 0 0 0 Goodwill impairment -10,564 0 0 0 0 Pre-tax profit 9,307 11,025 18,390 26,913 36,651

Taxes -2,809 -2,787 -4,597 -6,728-9,163 Minorities -770 -859 -1,448 -2,119-2,886 Other non-operating items 0 0 0 0 0 Net profit 5,728 7,378 12,344 18,065 24,602

Assets (US$m) Loans to customers 932,868 919,241 1,020,866 1,135,055 1,257,620 Interbank loans 153,766 184,089 191,526 199,263 207,314 Total securities 756,097 860,561 889,100 925,019 962,390 Intangible assets 27,357 29,396 30,584 31,819 33,105 Total assets 2,527,465 2,519,227 2,698,947 2,896,843 3,114,796 Net interest-earning assets 1,866,594 1,984,126 2,122,543 2,281,240 2,450,111

Liabilites (US$m) Interbank borrowings 130,084 131,734 137,056 142,593 148,354 Customer deposits 1,295,020 1,380,833 1,512,475 1,657,210 1,816,364 Total shareholders' equity 93,591 120,692 127,936 138,612 153,211 Tangible net asset value 66,234 91,296 97,352 106,793 120,106

Important Financial Ratios ROA 0.23% 0.29% 0.47% 0.65%0.82% ROE 5.17% 6.89% 9.93% 13.56%16.86% ROTNAV 7.41% 9.37% 13.09% 17.70%21.69% Cost/income -46.24% -46.26% -45.72% -45.63% -45.56% Tax rate -30.18% -25.28% -25.00% -25.00% -25.00% Payout 134.58% 39.87% 39.12% 39.40%39.56% Net interest margin 2.22% 2.19% 2.26% 2.33% 2.37% LLC % gross loans -2.55% -3.06% -2.60% -1.90% -1.20% Non-performing loans % gross loans 2.65% 4.00% 3.00% 2.00% 1.50% NPL coverage ratio 94.31% 87.00% 92.50% 97.50% 97.50% Loans % deposits 83.64% 75.25% 75.80% 76.45% 76.83% Tier 1 ratio 8.30% 10.23% 10.45% 10.91% 11.62% Core tier 1 ratio 6.99% 8.90% 9.17% 9.68% 10.43% Equity % total assets 3.70% 4.79% 4.74% 4.78% 4.92% RWA % total assets 45.42% 46.48% 45.14% 43.75% 42.33%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 75

PAN-EUROPEAN BANKS

INTESA SANPAOLO

Figure 87: Intesa Sanpaolo Valuation Table INTESA SANPAOLO 17/01/2010 Target Price: 3.55 Rating: HOLD Stock ticker ISP IM DCF Fair Value 3.57 FY 2009E FY 2010E FY 2011E FY 2012E Currency EUR Price/DCF FV87% EPS 0.25 0.23 0.29 0.35 Price 3.12 Upside/Downside 14% P/E 12.53 13.47 10.89 8.86 Market Cap (€bn) 39.2 DCF Assumptions TNAV per share 2.32 2.50 2.73 3.06 Dividend Yield FY 2010E 2.40% Risk-free rate 4.02% P/TNAV 1.34 1.25 1.14 1.02 PEG Ratio 0.58 LT Market Return 8.5% ROE 6.21% 5.42% 6.42% 7.46% Loans % Deposits FY 2010E 93% Equity beta 1.29 ROTNAV 12.90% 10.36% 11.82% 13.12% Assets % Equity FY 2010E 11.6 COE 9.8% ROA 0.50% 0.47% 0.56% 0.66% Tier 1 Ratio FY 2010E 9.1% Long-term growth rate 5.0% Core Tier 1 Ratio 7.82% 8.27% 8.50% 8.93%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Intesa Sanpaolo with a HOLD rating.

Intesa is, in our view, a high quality bank in many respects, but with a structurally low ROE versus the peer group. We believe there is only modest potential for an improvement in ROE versus peers for the following reasons:

• We believe strong competition in Italy for deposits will limit improvement in the NIM (all other things being equal). Our Dupont analysis shows that Intesa has shown an inability to improve its loan-to-deposit ratio over the course of the credit crisis, unlike some peers who have enjoyed an improvement as competitors retrench (e.g. HSBC and BBVA). These latter banks have enjoyed falling funding costs and a stable/improving NIM, whilst Intesa has suffered a contracting NIM. • Intesa has incurred only modest deterioration in its loan portfolio, with increases in loan loss charges and decreases in coverage being relatively small. However, on the flipside, this means that there will be less growth in earnings going forward as the cycle normalises. Divisional breakdown of group loans, Our Basel III Analysis points to an adequate Core Tier 1 ratio of 7.6% by the end of 9M09 2012. The fall in the Core Tier 1 ratio is one of the lowest in the group, reflecting the International Subsidiary conservatism of its banking regulator and the fact that it does not have a corporate Banks 8% structure involving significant minority interests (unlike Unicredit).

Public Finance 11% Our DCF valuation points to attractive upside for the company. However, we see Banca Dei Territori extra upside, greater growth potential and even more robust balance sheets in the 52% Corporate Nordic banks. We therefore initiate with a HOLD. and Investment Banking 29% Company Description Intesa Sanpaolo is Italy’s second largest bank by market cap, just behind Unicredit, Divisional breakdown of operating profit, although it has the largest presence by share of loans and deposits in the domestic 9M09 Italian market. The bank was created in September 2007 by the merger of Intesa

Eurizon Banca and San Paolo IMI. The main division of the group is Banca Dei Territori, which International Capital Fideuram Subsidiary 1% 3% encompasses its numerous retail banking franchises in Italy and accounts for 48% of Banks 11% operating profit. Corporate and Investment banking accounts for 32% of operating Public profit. The group’s international retail banking exposure is relatively modest and is Finance Banca Dei 4% Territori comprised mainly of operations in the CEE. Other smaller divisions include Public 48% Finance, Eurizon (an asset manager with €135bn under management) and Banca Corporate Fideuram, a high quality financial advisor network in Italy. and Investment Banking 33%

19 JANUARY 2010 76

PAN-EUROPEAN BANKS

Figure 88: Intesa Sanpaolo Financial Summary INTESA SANPAOLO FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (€m) Net interest income 11,630 10,618 10,375 11,160 11,931 Net fees & commissions 5,872 5,257 5,779 6,068 6,372 Trading profits/losses -53 1,170 1,439 1,511 1,586 Other revenue 708 584 684 763 843 Total operating revenues 18,157 17,628 18,277 19,502 20,732 Operating costs -9,936 -9,379 -9,771 -10,371 -10,973 Operating profit 8,221 8,249 8,506 9,131 9,758

Total provisions -2,884 -3,767 -3,391 -2,899 -2,203 Investment income 266 78 150 150 150 Impairments on other assets -949 -175 -100 -75 -50 Goodwill impairment -1,065 0 0 0 0 Pre-tax profit 3,589 4,385 5,166 6,307 7,655

Taxes -180 -1,101 -1,653 -2,018-2,450 Minorities -147 -93 -105 -129-156 Other non-operating items -709 -12 -450 -500 -550 Net profit 2,553 3,180 2,957 3,660 4,500

Assets (€m) Loans to customers 395,189 364,597 368,743 384,502 403,040 Interbank loans 56,371 55,000 57,750 63,525 69,878 Total securities 115,462 140,809 144,166 148,325 152,691 Intangible assets 27,151 26,000 26,000 26,000 26,000 Total assets 636,133 625,184 636,623 665,285 697,923 Net interest-earning assets 555,130 539,406 549,660 575,352 604,608

Liabilities (€m) Interbank borrowings 51,745 42,952 43,597 45,367 47,209 Customer deposits 405,778 387,411 394,499 412,261 432,485 Total shareholders' equity 48,954 53,508 55,577 58,368 62,209 Tangible net asset value 21,803 27,508 29,577 32,368 36,209

Important Financial Ratios ROA 0.42% 0.50% 0.47% 0.56%0.66% ROE 5.08% 6.21% 5.42% 6.42%7.46% ROTNAV 10.73% 12.90% 10.36% 11.82%13.12% Cost/income -54.72% -53.20% -53.46% -53.18% -52.93% Tax rate -5.02% -25.10% -32.00% -32.00% -32.00% Payout 0.94% 0.00% 30.00% 40.00%40.00% Net interest margin 2.21% 1.94% 1.91% 1.98% 2.02% LLC % gross loans -0.69% -0.90% -0.82% -0.67% -0.47% Non-performing loans % gross loans 2.87% 5.00% 5.00% 4.00% 3.00% NPL coverage ratio 49.00% 41.00% 48.00% 55.00% 55.00% Loans % deposits 97.39% 94.11% 93.47% 93.27% 93.19% Tier 1 ratio 7.07% 8.65% 9.08% 9.26% 9.64% Core tier 1 ratio 6.28% 7.82% 8.27% 8.50% 8.93% Equity % total assets 7.70% 8.56% 8.73% 8.77% 8.91% RWA % total assets 60.22% 58.47% 58.28% 59.19% 59.88%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 77

PAN-EUROPEAN BANKS

LLOYDS

Figure 89: Lloyds Banking Group Valuation Table LLOYDS BANKING 17/01/2010 Target Price: 49.00 Rating: REDUCE Stock ticker LLOY LN DCF Fair Value 49.06 FY 2009E FY 2010E FY 2011E FY 2012E Currency GBp Price/DCF FV116% EPS -10.31 -0.45 4.247.68 Price 56.78 Upside/Downside -14% P/E -5.51 -126.21 13.417.39 Market Cap (€bn) 41.9 DCF Assumptions TNAV per share 54.12 52.63 57.05 63.10 Dividend Yield FY 2010E 0.85% Risk-free rate 4.02% P/TNAV 1.05 1.08 1.00 0.90 PEG Ratio N/A LT Market Return 8.5% ROE -14.24% -0.71% 6.46% 10.78% Loans % Deposits FY 2010E 158% Equity beta 1.66 ROTNAV -17.34% -0.84% 7.72% 12.78% Assets % Equity FY 2010E 26.3 COE 11.5% ROA -0.48% -0.03% 0.25% 0.46% Tier 1 Ratio FY 2010E 10.3% Long-term growth rate 5.0% Core Tier 1 Ratio 8.06% 7.86% 8.29% 8.74%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Lloyds with a REDUCE rating.

Lloyds has enjoyed a positive response from the market in being able to extricate itself from the APS and with it, government control. Bullishness has also been based on management guiding to a significant reduction in loan losses (having somewhat ‘kitchen-sinked’ the amount of provisioning for HBOS’ CRE portfolio in H1 2009) and a marked improvement in the net interest margin as wholesale funding costs have come down. However, we believe this is now appreciated by the market and that looking forward, there are significant structural concerns for the bank.

Our key concern is that the Core Tier 1 ratio looks to us very weak under a Basel III framework, at only 4.4% by the end of 2012. The fall is primarily due to the full deduction of interests in financial companies from common equity, where currently this is deducted 100% at the total capital level (as allowed under the FSA transition provision). Other significant impacts come from the deduction of deferred tax assets (which we do not see Lloyds realising before 2012 given low earnings), minorities and negative AFS reserves.

We are also concerned that after the initial NIM improvement, Lloyds is likely to Divisional breakdown of group loans, H1 suffer pressure on its net interest earnings growth as it tries to reduce its loan to 2009 deposit ratio from ~170% to ~140%. The reliance on wholesale funding is a key Wealth and International reason why HBOS got into trouble in the first place. It is a structural problem that 1% Lloyds has inherited and we fear the market has ignored the impact on earnings that

Wholesale rectifying it will have. 36% Our DCF valuation points to absolute downside. Lloyds also looks expensive on

Retail conventional valuation metrics over the next few years given that earnings will be 63% depressed by high (albeit lower) loan losses. The P/E is the most expensive in the group until at least 2011. The P/TNAV looks cheap at 1x, but does not square with Divisional breakdown of operating profit, the fact that the ROE does not exceed the cost of capital until 2012 – and that is H1 2009 without taking into consideration the need to solve the Basel III capital dilemma.

Wealth and International 6% Company Description

Insurance 9% Lloyds acquired HBOS in January 2009. The combined group is a largely UK- Retail 39% focussed retail and corporate bank. Operating profit is split 46% wholesale banking, 39% retail banking, 9% insurance and 6% Wealth Management and International. Towards the end of 2009, Lloyds successfully raised £13.5bn via an equity rights issue and £8.5bn from issuing COCOS (contingent convertible notes which convert Wholesale 46% to common equity when the Core Tier 1 ratio falls below 5.0%).

19 JANUARY 2010 78

PAN-EUROPEAN BANKS

Figure 90: Lloyds Banking Group Financial Summary LLOYDS BANKING FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (£m) Net interest income 7,718 13,030 16,652 14,343 12,678 Net fees & commissions 2,537 1,371 2,686 2,686 2,686 Trading profits/losses -9,186 874 1,552 1,475 1,401 Other revenue 8,799 9,555 8,208 8,208 8,208 Total operating revenues 9,868 24,830 29,098 26,712 24,973 Operating costs -6,000 -12,201 -13,955 -12,354 -11,754 Operating profit 3,868 12,629 15,143 14,358 13,219

Total provisions -2,876 -22,822 -15,480 -10,510 -6,295 Impairments on other assets -136 -78 -465 -465 -465 Goodwill impairment -100 0 0 0 0 Other income/loss 4 5,221 400 400 400 Pre-tax profit 760 -5,049 -402 3,783 6,860

Taxes 38 1,509 113 -1,059-1,921 Minorities -26 -50 3 -27-49 Other non-operating items 0 0 0 0 0 Net profit 772 -3,591 -286 2,697 4,890

Assets (£m) Loans to customers 240,344 627,449 594,857 569,695 558,159 Interbank loans 38,733 37,398 38,146 39,687 41,290 Total securities 105,187 272,576 261,782 251,416 241,460 Intangible assets 2,453 6,574 6,705 6,976 7,258 Total assets 436,033 1,071,486 1,061,729 1,057,280 1,055,177 Net interest-earning assets 284,085 755,423 727,238 707,424 701,453

Liabilities (£m) Interbank borrowings 66,514 107,264 96,805 87,367 78,849 Customer deposits 263,904 711,282 715,605 719,972 724,382 Total shareholders' equity 9,393 41,037 40,216 43,302 47,437 Tangible net asset value 6,940 34,463 33,510 36,325 40,179

Important Financial Ratios ROA 0.20% -0.48% -0.03% 0.25%0.46% ROE 7.17% -14.24% -0.71% 6.46%10.78% ROTNAV 9.32% -17.34% -0.84% 7.72%12.78% Cost/income -60.80% -49.14% -47.96% -46.25% -47.07% Tax rate 5.00% -29.88% -28.00% -28.00% -28.00% Payout 83.94% 0.00% 0.00% 14.26%40.00% Net interest margin 2.90% 2.51% 2.25% 2.00% 1.80% LLC % gross loans -1.25% -3.34% -2.39% -1.70% -1.05% Non-performing loans % gross loans 3.53% 8.50% 8.00% 7.00% 6.00% NPL coverage ratio 33.06% 45.00% 50.00% 60.00% 70.00% Loans % deposits 140.60% 168.78% 158.43% 150.22% 145.72% Tier 1 ratio 8.04% 10.46% 10.26% 10.65% 11.01% Core tier 1 ratio 5.60% 8.06% 7.86% 8.29% 8.74% Equity % total assets 2.15% 3.83% 3.79% 4.10% 4.50% RWA % total assets 39.10% 43.90% 44.30% 45.38% 47.31%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 79

PAN-EUROPEAN BANKS

NORDEA

Figure 91: Nordea Valuation Table NORDEA BANK AB 17/01/2010 Target Price: 86.60 Rating: BUY Stock ticker NDA SS DCF Fair Value 85.66 FY 2009E FY 2010E FY 2011E FY 2012E Currency SEK Price/DCF FV86% EPS 6.16 6.63 7.94 9.07 Price 73.80 Upside/Downside 16% P/E 11.98 11.14 9.308.14 Market Cap (€bn) 29.5 DCF Assumptions TNAV per share 44.30 48.31 52.92 58.20 Dividend Yield FY 2010E 3.55% Risk-free rate 4.02% P/TNAV 1.67 1.53 1.39 1.27 PEG Ratio 0.66 LT Market Return 8.5% ROE 12.89% 12.48% 13.82% 14.54% Loans % Deposits FY 2010E 185% Equity beta 1.27 ROTNAV 14.94% 14.31% 15.68% 16.33% Assets % Equity FY 2010E 23.9 COE 9.7% ROA 0.51% 0.52% 0.59% 0.63% Tier 1 Ratio FY 2010E 10.9% Long-term growth rate 5.0% Core Tier 1 Ratio 8.50% 8.57% 8.67% 9.54%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Nordea with a BUY rating.

Typical of the Nordic banks in general, Nordea is a well run, very efficient bank. We see its Nordic footprint as very attractive. Norway and Sweden are forecast by consensus to have the highest GDP growth rates in developed Europe in 2010, (at 2.3% and 2.5% respectively), which should underpin a near term recovery of loan losses in the region.

On the negative side, loan losses from Nordea’s New European Market division (primarily as a result of the Baltic countries) should remain elevated, despite recent improvements. Extraordinarily large gains in 2009 from investing in bonds, in anticipation of interest rates falling, are expected to normalise at a lower level, bringing down trading income.

Our Basel III Analysis points to Nordea having one of the highest Core Tier 1 ratios in the group, at 9.2% by the end of 2012. We believe that the market will consider Divisional breakdown of operating profit, the bank to have excess capital once the principles of Basel III are formalised. 9M09 Our DCF valuation points to Nordea having attractive upside, albeit not as much as Shipping Oil Other Financial Services & Customer for DNBNOR and Handelsbanken, our more favoured Nordic plays. Conventional Institutions International Operations 1% 5% (incl Pr. P/E ratios place the bank as one of the cheaper stocks in the group on our 2010 and Banking & New Life) 2011 estimated earnings. European 9% Market 5% Company Description Nordea is the largest financial institution in the Nordic area, which is considered its Nordic Banking home market. It has the broadest presence in the region amongst its Nordic peers, 80% with approximately 30% of the Nordic loan book in Denmark, 28% in Sweden, 22% in Norway and 19% in Norway.

Divisional breakdown of operating profit, The present form of the group, under the name of ‘Nordea’, has only been in 9M09 existence since December 2001 and has been achieved through a number of cross-

Other border Nordic mergers and acquisitions. Nordic Banking accounts for 80% of the Customer Operations group’s loan book. It is also present in Poland, Russia and the Baltic countries, (incl Pr. Banking & which account for 5% of the group’s loan book. Other notable operations are Shipping Oil Life) Services & 16% Shipping Oil Services, Private Banking and Life Insurance. International 7% Financial Institutions Nordea’s two largest shareholders are Sampo and the Swedish state, with shares of 5% New European 18.3% and 19.8% respectively. Sampo, the Nordic insurance group, has made clear Market Nordic 7% Banking 65% its intention to participate in consolidation of the Nordic banking market via its shareholding.

19 JANUARY 2010 80

PAN-EUROPEAN BANKS

Figure 92: Nordea Financial Summary NORDEA BANK AB FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (€m) Net interest income 5,093 5,265 5,272 5,553 5,867 Net fees & commissions 1,883 1,666 1,786 1,875 1,970 Trading profits/losses 833 1,747 1,963 1,994 2,061 Other revenue 391 477 287 298 311 Total operating revenues 8,200 9,155 9,307 9,720 10,208 Operating costs -4,338 -4,350 -4,449 -4,648 -4,883 Operating profit 3,862 4,805 4,858 5,072 5,325

Total provisions -466 -1,536 -1,327 -844 -492 Impairments on other assets 0 0 0 0 0 Goodwill impairment 0 0 0 0 0 Other income 000 00 Pre-tax profit 3,396 3,269 3,531 4,228 4,834

Taxes -724 -809 -883 -1,057-1,208 Minorities -1 -7 -7 -8-10 Other non-operating items 0 0 0 0 0 Net profit 2,671 2,454 2,641 3,163 3,616

Assets (€m) Loans to customers 265,100 285,548 300,954 320,227 340,524 Interbank loans 23,903 16,587 17,465 18,562 19,752 Total securities 63,436 72,844 78,849 85,349 92,384 Intangible assets 2,535 2,714 2,714 2,714 2,714 Total assets 474,074 491,612 521,565 557,531 593,054 Net interest-earning assets 340,047 360,814 381,936 408,108 436,603

Liabilities (€m) Interbank borrowings 51,932 55,045 59,583 64,494 69,811 Customer deposits 257,580 264,577 286,387 313,469 339,309 Total shareholders' equity 17,725 20,368 21,966 23,807 25,911 Tangible net asset value 15,190 17,654 19,252 21,093 23,197

Important Financial Ratios ROA 0.62% 0.51% 0.52% 0.59%0.63% ROE 15.35% 12.89% 12.48% 13.82%14.54% ROTNAV 18.08% 14.94% 14.31% 15.68%16.33% Cost/income -52.90% -47.52% -47.80% -47.82% -47.83% Tax rate -21.32% -24.73% -25.00% -25.00% -25.00% Payout 19.43% 40.75% 39.49% 41.81%41.80% Net interest margin 1.56% 1.50% 1.42% 1.41% 1.39% LLC % gross loans -0.17% -0.52% -0.43% -0.26% -0.14% Non-performing loans % gross loans 0.77% 1.30% 1.10% 0.80% 0.80% NPL coverage ratio 52.61% 51.00% 52.00% 58.00% 66.00% Loans % deposits 178.41% 190.19% 185.19% 178.52% 175.37% Tier 1 ratio 9.35% 10.92% 10.89% 10.92% 11.91% Core tier 1 ratio 6.71% 8.50% 8.57% 8.67% 9.54% Equity % total assets 3.74% 4.14% 4.21% 4.27% 4.37% RWA % total assets 44.99% 39.78% 40.58% 41.10% 38.63%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 81

PAN-EUROPEAN BANKS

SANTANDER

Figure 93: Banco Santander Valuation Table BANCO SANTANDER 17/01/2010 Target Price: 13.00 Rating: HOLD Stock ticker SAN SM DCF Fair Value 12.84 FY 2009E FY 2010E FY 2011E FY 2012E Currency EUR Price/DCF FV90% EPS 1.06 1.02 1.31 1.64 Price 11.50 Upside/Downside 12% P/E 10.82 11.28 8.777.02 Market Cap (€bn) 95.3 DCF Assumptions TNAV per share 5.44 6.05 6.84 7.82 Dividend Yield FY 2010E 3.55% Risk-free rate 4.02% P/TNAV 2.11 1.90 1.68 1.47 PEG Ratio 0.42 LT Market Return 8.5% ROE 13.81% 11.82% 14.06% 16.04% Loans % Deposits FY 2010E 138% Equity beta 1.48 ROTNAV 19.53% 16.85% 19.17% 20.93% Assets % Equity FY 2010E 15.9 COE 10.6% ROA 0.81% 0.75% 0.89% 1.03% Tier 1 Ratio FY 2010E 10.1% Long-term growth rate 5.0% Core Tier 1 Ratio 7.95% 8.67% 9.05% 9.56%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Santander with a HOLD rating.

The main attraction of Santander is that it is a fast-growing bank with a high ROE, underpinned by its LATAM operations. (The ROE, incidentally, is historically more stable too, versus BBVA). The bank has also proved itself a competent acquirer of UK banking assets in the past few years, purchasing cheaply and integrating seamlessly for cost efficiencies.

Divisional breakdown of group loans, We worry, however, about the prospect of elevated loan losses in Spain, a country 9M09 which is forecast to have one of the worst GDP growth rates in developed Europe in

Sovereign 2010 (-0.40%) and where the unemployment rate is predicted to worsen to over 5% Latin America 20%. Coverage ratios for the group have been run down to a lowly 73%. While lower 14% coverage ratios may be within the tolerance of the group’s risk management, the momentum is negative versus the positive improvement already expected at some Continental Europe 49% other banks.

United Our Basle III Analysis shows Santander’s Core Tier 1 to be 7.5% at the end of 2012, Kingdom 32% a level which we consider to be adequate but not indicative of excess capital.

Our DCF fair value indicates upside from the current share price, but it is modest Divisional breakdown of operating profit, compared to the upside available at more favoured stocks in the group. 9M09 Conventional valuation metrics also do not point to Santander being particularly Sovereign cheap on 2010 P/E or P/TNAV. We note that the growth in earnings makes the stock 2% cheap again on 2011 P/E. We may be looking to upgrade the stock at that point once clarity that asset quality deterioration in Spain has reached its nadir. Continental Europe Latin America 42% 43% Company Description Santander is Spain’s largest bank by market cap. Santander matched BBVA in market cap until 2004; thereafter embarking on an aggressive UK-orientated United Kingdom acquisition strategy beginning with the purchase of Abbey National, whilst BBVA 13% focussed more on organic growth. Santander is now about twice the size of BBVA. The Continental Europe division comprises Santander’s European retail operations. This includes the Santander Network and the 88% owned Banesto in Spain; Santander Totta in Portugal; and Santander Consumer Finance, which operates in Spain, Germany, Italy and the Nordic countries. Following its acquisition of Abbey National, Alliance & Leicester and Bradford & Bingley, Santander has become the third largest bank by deposits in the UK. Latin America accounts for 14% of the group loan book but 43% of operating profit. The main operation there is in Brazil. Santander also now wholly owns the troubled US retail bank, Sovereign, which accounts for 5% of the group’s loan book.

19 JANUARY 2010 82

PAN-EUROPEAN BANKS

Figure 94: Santander Financial Summary BANCO SANTANDER FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (€m) Net interest income 20,945 25,933 25,410 26,518 27,920 Net fees & commissions 9,020 8,963 9,411 9,882 10,376 Trading profits/losses 2,597 3,061 3,122 3,185 3,248 Other revenue 927 686 745 774 804 Total operating revenues 33,489 38,643 38,689 40,359 42,349 Operating costs -14,949 -16,092 -16,528 -17,240 -17,997 Operating profit 18,540 22,551 22,161 23,118 24,352

Total provisions -6,601 -9,735 -9,740 -7,550 -5,252 Impairments on other assets -91 -358 -358 -358 -358 Goodwill impairment 0 0 0 0 0 Other income / expense -426 -1,041 -1,041 -1,041 -1,041 Pre-tax profit 11,421 11,418 11,022 14,170 17,701

Taxes -2,391 -2,352 -2,315 -2,976-3,717 Minorities -473 -402 -392 -504-629 Other non-operating items 319 0 0 0 0 Net profit 8,877 8,664 8,316 10,690 13,354

Assets (€m) Loans to customers 634,884 646,199 670,453 729,225 794,854 Interbank loans 64,731 46,658 48,543 52,545 56,876 Total securities 227,878 259,986 281,417 304,615 329,725 Intangible assets 18,836 23,474 23,474 23,474 23,474 Total assets 1,049,632 1,081,423 1,148,455 1,243,534 1,347,521 Net interest-earning assets 947,457 937,376 982,661 1,067,168 1,160,655

Liabilities (€m) Interbank borrowings 79,795 76,091 80,748 87,404 94,609 Customer deposits 672,531 700,358 738,195 799,046 864,913 Total shareholders' equity 57,587 67,848 72,837 79,251 87,264 Tangible net asset value 38,751 44,374 49,363 55,777 63,790

Important Financial Ratios ROA 0.90% 0.81% 0.75% 0.89%1.03% ROE 15.67% 13.81% 11.82% 14.06%16.04% ROTNAV 22.91% 19.53% 16.85% 19.17%20.93% Cost/income -44.64% -41.64% -42.72% -42.72% -42.50% Tax rate -20.94% -20.60% -21.00% -21.00% -21.00% Payout 54.22% 50.92% 40.00% 40.00%40.00% Net interest margin 2.36% 2.75% 2.65% 2.59% 2.51% LLC % gross loans -1.09% -1.49% -1.43% -1.05% -0.67% Non-performing loans % gross loans 2.22% 4.00% 4.25% 3.20% 1.70% NPL coverage ratio 90.64% 73.00% 75.00% 85.00% 120.00% Loans % deposits 156.37% 140.90% 137.75% 138.42% 139.39% Tier 1 ratio 9.12% 9.45% 10.14% 10.41% 10.82% Core tier 1 ratio 7.58% 7.95% 8.67% 9.05% 9.56% Equity % total assets 5.49% 6.27% 6.34% 6.37% 6.48% RWA % total assets 48.97% 50.32% 48.33% 48.31% 48.26%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 83

PAN-EUROPEAN BANKS

STANDARD CHARTERED

Figure 95: Standard Chartered Valuation Table STANDARD CHARTER 17/01/2010 Target Price: 1500 Rating: HOLD Stock ticker STAN LN DCF Fair Value 1481.86 FY 2009E FY 2010E FY 2011E FY 2012E Currency GBp Price/DCF FV105% EPS 1.91 2.25 2.71 3.20 Price 1551.50 Upside/Downside -4% P/E 13.27 11.27 9.367.92 Market Cap (€bn) 35.9 DCF Assumptions TNAV per share 9.88 12.09 14.76 17.94 Dividend Yield FY 2010E 6.50% Risk-free rate 4.01% P/TNAV 2.57 2.10 1.72 1.41 PEG Ratio 0.58 LT Market Return 8.5% ROE 16.33% 17.15% 17.46% 17.45% Loans % Deposits FY 2010E 84% Equity beta 1.55 ROTNAV 22.37% 22.53% 22.12% 21.43% Assets % Equity FY 2010E 16.6 COE 11.0% ROA 0.90% 1.03% 1.08% 1.11% Tier 1 Ratio FY 2010E 11.1% Long-term growth rate 5.0% Core Tier 1 Ratio 7.83% 8.49% 9.24% 10.06%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Standard Chartered with a HOLD rating.

Standard Chartered has a very attractive Asian growth footprint. Perhaps relatively underappreciated is the fact that the bank is also what we would call a ‘super deposit franchise’, by virtue of having improved its loan to deposit ratio to ~80% from ~100% over the course of the crisis as customer deposits undertook a flight to quality. This excess of deposit funding, together with still significant demand for credit in Asia, Geographical breakdown of group provides Standard Chartered with a considerable growth advantage versus peers. loans, H1 2009 Hong Kong Our Dupont Analysis has shown that the quality of earnings has deteriorated in Americas UK 16% & Europe 16% recent quarters, with the proportion from net interest income becoming less, but Africa 2% made up by substantially higher (but lower quality) trading profits. We expect the mix Singapore 14% of operating earnings to reverse in coming quarters towards better quality net Middle East interest income (as the excess of deposits are used to fund new lending) and away & other S Asia India 10% 5% from trading profits, (which we do not see as sustainable at current levels and where

Korea higher risk weightings under Basel III will reduce returns). Other Asia 17% Pacific 20% Our Basel III Analysis shows that Standard Chartered should have a Core Tier 1 ratio of 7.4% by the end a 2012, a level which we consider adequate. It is notable Divisional breakdown of group loans, H1 2009 that the bank’s capital ratios are not as badly affected by the proposals as the other UK banks. In our opinion, this is down to the management being more conservative and focussed on core lending principles.

Consumer Banking Despite all these positive points, however, our main concern is that the price, whilst Wholesale 45% Banking not particularly expensive, is also not cheap. Our DCF valuation already shows the 55% stock to be fully valued. Conventional P/E and P/TNAV metrics point to the same conclusion for 2010, although on 2011 estimated earnings, the stock appears attractive again. We are not yet prepared to add to the consensual Buys on the stock. Divisional breakdown of operating profit, Company Description H1 2009 Central Items 6% Standard Chartered has a clear focus on emerging market territories, particularly Consumer Asia. The exposure within Asia is quite broad, but perhaps notable for the relative Banking 23% emphasis on Hong Kong, Singapore and Korea.

The bank’s divisions comprise simply Consumer Banking and Wholesale Banking.

Wholesale Banking 71%

19 JANUARY 2010 84

PAN-EUROPEAN BANKS

Figure 96: Standard Chartered Financial Summary STANDARD CHARTERED PLC FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (US$m) Net interest income 7,387 7,605 8,724 10,199 11,956 Net fees & commissions 2,941 3,386 3,521 3,662 3,809 Trading profits/losses 2,405 3,433 2,920 2,699 2,740 Other revenue 1,235 1,790 2,427 2,775 2,864 Total operating revenues 13,968 16,215 17,592 19,335 21,368 Operating costs -7,611 -8,184 -8,889 -9,779 -10,725 Operating profit 6,357 8,031 8,703 9,557 10,643

Total provisions -1,321 -2,381 -1,924 -1,427 -1,056 Impairments on other assets -469 -51 -39 -39 -39 Goodwill impairment 0 0 0 0 0 Other income 234 19 19 19 19 Pre-tax profit 4,801 5,618 6,760 8,110 9,568

Taxes -1,290 -1,597 -1,825 -2,190-2,583 Minorities -103 -119 -148 -178-210 Other non-operating items 0 0 0 0 0 Net profit 3,408 3,901 4,787 5,742 6,775

Assets (US$m) Loans to customers 174,178 198,269 234,191 273,574 318,148 Interbank loans 46,583 45,366 45,366 47,199 49,106 Total securities 145,477 131,905 142,194 153,974 167,461 Intangible assets 6,361 6,532 6,796 7,071 7,356 Total assets 435,068 430,345 494,835 569,359 655,448 Net interest-earning assets 390,399 386,751 433,462 486,979 547,487

Liabilities (US$m) Interbank borrowings 31,909 35,652 40,818 46,733 53,504 Customer deposits 257,455 260,630 303,088 352,499 410,005 Total shareholders' equity 22,140 25,638 30,176 35,608 42,045 Tangible net asset value 15,779 19,106 23,380 28,538 34,688

Important Financial Ratios ROA 0.89% 0.90% 1.03% 1.08%1.11% ROE 15.85% 16.33% 17.15% 17.46%17.45% ROTNAV 22.53% 22.37% 22.53% 22.12%21.43% Cost/income -54.49% -50.47% -50.53% -50.57% -50.19% Tax rate -26.87% -28.44% -27.00% -27.00% -27.00% Payout 34.18% 40.75% 40.77% 40.83%40.71% Net interest margin 2.14% 1.96% 2.13% 2.22% 2.31% LLC % gross loans -0.64% -1.01% -0.73% -0.47% -0.30% Non-performing loans % gross loans 1.35% 1.80% 1.50% 1.20% 1.20% NPL coverage ratio 65.88% 73.00% 73.00% 73.00% 73.00% Loans % deposits 74.43% 82.84% 83.88% 84.01% 83.76% Tier 1 ratio 9.92% 10.66% 11.11% 11.66% 12.30% Core tier 1 ratio 7.51% 7.83% 8.49% 9.24% 10.06% Equity % total assets 5.09% 5.96% 6.10% 6.25% 6.41% RWA % total assets 43.40% 48.59% 45.71% 42.97% 40.37%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 85

PAN-EUROPEAN BANKS

UNICREDIT

Figure 97: Unicredit Valuation Table UNICREDIT SPA 17/01/2010 Target Price: 2.00 Rating: REDUCE Stock ticker UCG IM DCF Fair Value 2.00 FY 2009E FY 2010E FY 2011E FY 2012E Currency EUR Price/DCF FV112% EPS 0.10 0.09 0.180.29 Price 2.25 Upside/Downside -11% P/E 22.92 24.08 12.377.77 Market Cap (€bn) 43.3 DCF Assumptions TNAV per share 2.07 2.19 2.38 2.62 Dividend Yield FY 2010E 1.25% Risk-free rate 4.02% P/TNAV 1.08 1.03 0.94 0.86 PEG Ratio 0.32 LT Market Return 8.5% ROE 2.74% 2.72% 5.09% 7.67% Loans % Deposits FY 2010E 93% Equity beta 1.68 ROTNAV 4.81% 4.38% 7.96% 11.59% Assets % Equity FY 2010E 14.3 COE 11.6% ROA 0.17% 0.19% 0.36% 0.54% Tier 1 Ratio FY 2010E 10.4% Long-term growth rate 5.0% Core Tier 1 Ratio 9.05% 9.56% 9.68% 9.84%

Source: Matrix Corporate Capital Research

Analyst: Andrew Lim Investment Summary +44 20 3206 7347 [email protected] We initiate on Unicredit Group with a REDUCE rating.

Unicredit has one of the lowest ROEs in the peer group. We believe this is structural in nature and do not anticipate a substantial relative improvement near term. The Divisional breakdown of group loans, 9M09 main reason for Unicredit’s low ROE, according to our Dupont Analysis, is the low CEE Retail NIM which, unlike the Nordic banks, is not compensated for by an efficient cost 11% banking - Poland Italy base. The NIM has stayed roughly the same over the course of the credit crisis. The 3% Private 21% banking lack of improvement versus other franchises like HSBC, BBVA and Standard 1% Retail Chartered is a reflection in our opinion of the continued robustness of its competitors banking - Germany in most of its key markets, (particularly Italy). The competition makes it relatively 6% difficult for Unicredit to increase its share of deposits and reduce its funding costs Retail Corporate banking - (note that its LTD ratio has stayed roughly the same) and to charge higher margins and Austria investment 3% for new loans through better pricing power. banking 55%

Divisional breakdown of operating profit, Additionally, Unicredit is suffering elevated loan losses from its CEE portfolio (with 9M09 Russia and Ukraine having the most acute credit problems). We expect loan losses Retail to be deeper for longer in these countries given economic imbalances. Exceptional CEE banking - Retail 20% Italy banking - trading profits made in 2009 are also expected to normalise at a lower level. 18% Germany Poland 1% 5%

Retail At 7.8% by the end of 2012, we consider Unicredit’s Core Tier 1 ratio as adequate banking - Asset mgmt Austria under our Basel III Analysis. 2% 2% Private banking 2% Corporate Our DCF valuation points to absolute downside in the share price. We see Unicredit and investment also being one of the most expensive banks on 2010 and 2011 estimated P/E banking 50% earnings multiples. The P/TNAV may appear cheap at only 1.0x, but is not justified by an ROE in the low-to-mid single digits (i.e. below the cost of capital). Breakdown of loans in the CEE region, 9M09 Company Description Unicredit Group is Italy’s largest bank by market cap, just ahead of Intesa Sanpaolo. Slovenia Bosnia Baltics Serbia Slovakia 4% 2% 2% 1% 4% It is more geographically diverse than Intesa. Corporate and Investment banking is Croatia Romania 16% the largest single division, accounting for 50% of operating profit and involved mainly 5% Ukraine in corporate lending. Unicredit has core retail lending operations in Italy, Germany 6% Turkey 14% and Austria, which account for 18%, 1% and 2% of operating profit respectively. Bulgaria 7% Unicredit also has notable CEE operations across a variety of countries, the most

Kazakhstan significant being Poland, Croatia, Turkey and Russia. CEE in aggregate contributes 7% Russia 25% of operating profit despite accounting for only 14% of the loan book. Other Hungary 13% 7% Czech Rep. smaller divisions include private banking, asset management. Having considered 12% raising capital via the issue of Tremonti bonds (which, incidentally, would not have counted towards Core Tier 1 capital), shareholders instead went on to approve a €4bn rights issue to be undertaken in January 2010.

19 JANUARY 2010 86

PAN-EUROPEAN BANKS

Figure 98: Unicredit Financial Summary UNICREDIT SPA FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E Income Statement (€m) Net interest income 18,373 17,238 15,924 16,989 18,684 Net fees & commissions 9,093 7,546 8,295 8,710 9,145 Trading profits/losses -1,968 1,899 2,006 2,106 2,212 Other revenue 1,379 841 940 1,014 1,090 Total operating revenues 26,877 27,524 27,165 28,820 31,131 Operating costs -16,692 -14,988 -14,611 -15,522 -16,754 Operating profit 10,185 12,537 12,554 13,298 14,377

Total provisions -4,043 -8,950 -9,181 -7,148 -4,852 Investment income 207 65 250 400 500 Goodwill impairment -750 0 0 0 0 Other income -140 -351 -200 -200 -200 Pre-tax profit 5,458 3,301 3,423 6,350 9,825

Taxes -627 -1,084 -1,095 -2,071-3,189 Minorities -518 -312 -279 -528-813 Other non-operating items -301 -259 -250 -250 -250 Net profit 4,012 1,646 1,798 3,501 5,572

Assets (€m) Loans to customers 612,480 553,836 537,154 577,984 629,248 Interbank loans 80,827 100,000 100,000 105,000 110,250 Total securities 270,112 213,429 220,851 228,643 236,826 Intangible assets 5,593 5,100 4,900 4,700 4,500 Total assets 1,045,612 948,866 939,245 996,760 1,066,099 Net interest-earning assets 971,071 873,766 864,754 918,627 983,574

Liabilities (€m) Interbank borrowings 177,677 117,906 114,393 122,608 132,715 Customer deposits 591,290 584,654 578,726 614,165 656,889 Total shareholders' equity 54,999 65,034 67,096 70,485 74,900 Tangible net asset value 28,517 39,934 42,196 45,785 50,400

Important Financial Ratios ROA 0.39% 0.17% 0.19% 0.36%0.54% ROE 7.12% 2.74% 2.72% 5.09%7.67% ROTNAV 13.07% 4.81% 4.38% 7.96%11.59% Cost/income -62.11% -54.45% -53.79% -53.86% -53.82% Tax rate -11.50% -32.85% -32.00% -32.61% -32.46% Payout 0.00% 0.00% 30.00% 40.00%50.00% Net interest margin 1.91% 1.87% 1.83% 1.91% 1.99% LLC % gross loans -0.60% -1.41% -1.54% -1.14% -0.68% Non-performing loans % gross loans 4.56% 5.75% 5.45% 4.60% 3.80% NPL coverage ratio 63.63% 62.00% 66.00% 70.00% 70.00% Loans % deposits 103.58% 94.73% 92.82% 94.11% 95.79% Tier 1 ratio 6.80% 9.90% 10.41% 10.50% 10.60% Core tier 1 ratio 6.00% 9.05% 9.56% 9.68% 9.84% Equity % total assets 5.26% 6.85% 7.14% 7.07% 7.03% RWA % total assets 49.02% 47.92% 48.41% 48.89% 49.48%

Source: Matrix Corporate Capital Research

19 JANUARY 2010 87

PAN-EUROPEAN BANKS

APPENDIX

Key features of the Basel Committee on Banking Supervision Consultative Document on Strengthening the Resilience of the Banking Sector – 17 December 2009

The proposals are a ‘key’ element of the reform package addressing the lessons learned from the credit crises.

TIMING • Consultation on the proposals will be until April 2010. • An impact assessment will be carried out in the first half of 2010. • The Basel Committee will then review the regulatory minimum level of capital in the second half of 2010. • The final Standards should be developed by the end of 2010 and phased in, as financial conditions allow, by the end of 2012. We tentatively term these proposals ‘Basel III’.

MAIN THEMES • Quality consistency and transparency of the capital base will be raised. Common equity will essentially comprise common shares plus retained earnings. All deductions from capital will be made at the common equity level. Additional capital included in Tier 1 capital will include minorities (by our interpretation of the proposals), but exclude non-qualifying hybrids. This accounts for virtually all existing innovative hybrid securities as well as some preference shares currently in issue. We discuss later the criteria which lead to their exclusion. The UK FSA has been even more conservative in their recent (late December 2009) consultative document by appearing to exclude all preference shares from Tier 1 capital. Note that the proposals do not ban the issuance of new hybrid securities which satisfy the criteria, nor does it prevent banks from altering the features of existing hybrids via a cram down. Also, there will be appropriate grandfathering arrangements for ineligible hybrids instruments and a transition period for implementation of the new capital standards. These will be determined once the impact assessment has been completed, although we note that the UK FSA has already provided key details of its own proposed grandfathering process. • Capital requirements for trading books will be increased. Additionally, counterparty risk exposure arising from derivative, repo and security financing will be raised, whilst collateral and mark-to-market exposures to centralised clearing facilities will qualify for a zero risk weighting. This forces banks to use central clearing for derivative trades and make their balance sheets less complex. It also increases risk weighted assets, putting more pressure on capital ratios. • A gross leverage ratio (i.e. not risk weighted) will be introduced. The leverage ratio is again quite onerous, using as its numerator a simple non risk based calculation of assets and also including off balance sheet items, standby letters of credit, cancellable commitments etc. The denominator will probably be Tier 1 or common capital, but both smaller under the new Basel III proposals.

19 JANUARY 2010 88

PAN-EUROPEAN BANKS

There will be no netting of derivatives. US banks net off derivatives (therefore reducing the size of assets and liabilities on the balance sheet) but European banks do not (except for Credit Suisse, which uses US GAAP accounting). This will be seen as a negative for banks with large derivative exposures which have previously a much smaller asset base if US-style netting is allowed (specifically Deutsche Bank). It would also be particularly onerous for US banks of course, but note that the Basel committee is not a statutory authority in any country and that Basel II has not even been implemented in the US. • Pro-cyclical provisioning buffers will be introduced. This builds on the success of the pro-cyclical generic provisioning model in Spain. Of particular interest is the attempt to transfer the risk of being under-provisioned to employees, by restricting the payment of bonuses to them when the bank’s capital is within a buffer range just above the minimum capital requirement. • A global minimum liquidity standard will be introduced. The necessity for an increased stock of liquid assets ironically increases the demand for the vast issue of government paper to finance national deficits.

KEY CHANGES FOR BANKS 1. Changes to the Capital Base: • Tier 1 (T1) capital must be fully available to absorb losses on a going concern basis. The predominant form of T1 must be common and retained earnings. • The remainder of the T1 capital base must be comprised of instruments that are subordinated, have fully discretionary noncumulative dividends or coupons and have neither a maturity date nor an incentive to redeem. Innovative hybrid capital instruments with an incentive to redeem through features like step-up clauses, currently limited to 15% of the T1, will be phased out. • Distribution of T1 capital will be deducted from distributable earnings / profit. • Tier 2 (T2) capital, available to absorb losses on a gone concern basis, will be reduced to a single level i.e. no lower (LT2) and upper T2 (UT2) capital; only old LT2, as in plain vanilla subordinated debt. The current limitation on T2 not exceeding T1 will be eliminated. • Tier 3 (T3) capital, previously subordinated debt used as capital against market risk, is eliminated. • Deductions from capital will take place at the common level rather than previously from T1 or from total capital. There are a range of deductions proposed, including minorities, deferred tax assets, defined contribution pension deficits etc. • Grandfathering will be allowed but only securities issued prior to the publication of the consultative document will be eligible (i.e. do not expect a rash of innovative T1 to be issued).

2. Capital requirement for trading and counterparty risks • Increased capital requirements for trading books. • Increased capital requirements for derivative, repo and securities financing. • Distinct capital charge for mark-to-market credit valuation adjustment on counterparty risks. • Collateral and mark-to-market exposures to central clearing will qualify for a zero risk weighting. • Risk weights on counterparty exposures to other financials will be higher than exposures to non-financial counterparties.

19 JANUARY 2010 89

PAN-EUROPEAN BANKS

3. Leverage ratio • The leverage ratio will be based on gross exposures. There will be no netting for collateral CDS or other credit mitigation strategies. • The leverage ratio will be comparable across jurisdictions and adjusted for accounting differences. • The numerator is a simple non risk based calculation of assets and the denominator is probably T1 or common T1. • The numerator includes shadow commitments; liquidity facilities, stand by LOCs, cancellable commitments, etc. • Off balance sheet exposures are included.

4. Pro-cyclical provisioning buffers • The aim of the buffers is to dampen cyclicality of capital requirements, promote more forward looking provisions, conserve capital to build buffers and protect banking sector from excessive credit growth. • There will be the possible use of Probability of Default based on the worst of historical averages. • There may be a potential capital surcharge for systemically important banks. • The bank will be required to provision earlier in a more forward looking basis and if provisions are too low, the shortfall will be deducted from common equity. • If the bank’s capital is in the buffer range (which is some capital range above the minimum capital requirement), the bank may be required to conserve capital by limiting the payment of dividends, other distributions to capital holders and in paying employee bonuses. • There may be a move to adjust the capital buffer range upwards in times of excessive credit creation.

5. Liquidity standard • There will be a potential 30 day liquidity coverage ratio. • There will be a potential longer term ratio to address liquidity (asset / liability) mismatches. • A stock of liquid assets to be held.

19 JANUARY 2010 90

PAN-EUROPEAN BANKS

This page has been left blank intentionally

19 JANUARY 2010 91

PAN-EUROPEAN BANKS

This page has been left blank intentionally

19 JANUARY 2010 92

PAN-EUROPEAN BANKS

This page has been left blank intentionally

19 JANUARY 2010 93

PAN-EUROPEAN BANKS

Important Disclosures

General Disclosures

MCC is committed to publishing accurate but non-independent research reports. MCC may provide financial advice to, and undertake corporate finance and other transactions on behalf of, companies on which it publishes research. Therefore, research reports published by MCC have not been prepared in accordance with legal requirements designed to promote the independence of investment research; nor is MCC subject to any prohibition on dealing ahead of the dissemination of investment research. Accordingly, this research report should be treated as a marketing communication.

This report was first released on 19 January 2010.

Unless otherwise noted, the securities prices noted in this report were market prices on 15 January 2010.

We seek to update our research as appropriate, but various regulations may prevent us from doing so. Our reports are published at irregular intervals as appropriate in the analyst's judgement.

MCC's price targets are based on several methodologies, which may include, but are not restricted to, analyses of market risk, growth rate, revenue stream, discounted cash flow, EBITDA, EPS, cash flow (CF), free cash flow (FCF), EV/EBITDA, P/E, PE/growth, P/CF, premium (discount)/average group EV/EBITDA, premium (discount)/average group P/E, sum of the parts, net asset value, dividend returns, and return on equity over the next 12 months.

This report has been prepared by the individual(s) whose name(s) appear(s) on the front cover. Unless it has been otherwise specified, the preparers are employed by MCC as research analysts. Research analysts are paid in part based on the profitability of MCC, which includes remuneration received from investment banking transactions. Specific Disclosures

Company Disclosure BBVA (BVA) None Banco Santander (SAN) None DNBNOR (DNBNOR) None HSBC Holdings (HSBA) None Handelsbanken (HVB) None Intesa Sanpaolo (ISP) None Lloyds Banking Group (LLOY) None Nordea Bank (NDA) None Standard Chartered (STAN) None Unicredit (UCG) None

1. Matrix Corporate Capital acts as broker 2. Matrix Corporate Capital acts as NOMAD 3. Matrix Corporate Capital acts as market maker

MCC, its partners, employees and any affiliated undertaking may have a position or holding in any of the securities mentioned in this report or in a related instrument. Disclaimer

This report is issued by Matrix Corporate Capital LLP (‘MCC’) which is authorised and regulated in the United Kingdom by the Financial Services Authority (FSA) and is a member of the London Stock Exchange.

19 JANUARY 2010 94

PAN-EUROPEAN BANKS

This report is not directed to, or intended for distribution to, or use by, any person or entity who is a citizen or resident of, or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject MCC to any registration or licensing requirement within such jurisdiction.

This report is for the use of the addressees only and is intended for use only by a person or entity that qualifies as an authorised person or exempt person within the meaning of section 19 of the Financial Services and Markets Act 2000 (the ‘Act’) or qualifies as a person to whom the financial promotion restrictions imposed by the Act do not apply by virtue of the Financial Services and Markets Act 2000 (Financial Promotion) Order 2005, or is a person classified as an ‘professional client’ for the purposes of the Conduct of Business Sourcebook of the FSA. Consequently, this report is intended for use only by high net worth entities or by persons having professional experience in matters relating to investments. This report is not intended for use by other persons, in particular, ‘retail clients’ as defined by the rules of the FSA. Any such person who receives this report should not act upon the contents of this report.

The information and materials presented in this report have been obtained or derived from sources believed by MCC to be reliable. Such information and materials have not been independently verified and MCC makes no representation as to their accuracy or completeness. Neither MCC, nor any of its directors or employees accepts liability for any loss arising from the use of material presented in this report.

Any forecasts or target prices shown for companies or securities noted in this report may not be achieved due to multiple risk factors including, without limitation, market volatility, sector volatility, corporate actions, the unavailability of complete and accurate information or assumptions made by MCC or other sources relied on in the report subsequently transpiring to be inapposite.

The information and materials presented in this report are provided to you for information purposes only and are not to be used or considered as an offer, or the solicitation of an offer, to sell or buy or subscribe for securities or other financial instruments. This report was prepared for general circulation and does not give investment advice or personal recommendations specific to individual investors. As such, the companies and securities discussed in this report may not be suitable for all investors who must make their own investment decisions based on their specific investment objectives and financial situation utilising their own financial advisors as they deem necessary. MCC will not treat recipients of this report as clients by virtue of their receiving it.

Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect the preparer's best judgement at the date of publication and are subject to change without notice. The price, value of, and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. You may not get back as much as you invest. Fluctuations in exchange rates could have adverse effects on the value or price of, or income derived from, certain investments.

None of the material presented in this report, nor its content, nor any copy of it, may be altered in any way, transmitted, copied or distributed to, any other party, without the express prior written permission of MCC.

19 JANUARY 2010 95

Matrix Corporate Capital LLP One Vine Street London, W1J 0AH +44 (0) 20 3206 7000