Matrix Assessment of Pan-European Banks Capital Positions relating to Basel III : Jan. '11

CO

RP

OR

AT

E C

AP

ITA

L

PAN-EUROPEAN BANKS

THE BASEL III GAME CHANGER

• We initiate on the 10 largest European ‘lending’ banks:

BBVA, DNBNOR, Handelsbanken, HSBC, Intesa, Lloyds,

Nordea, Santander, Standard Chartered and Unicredit.

• The Basel Committee proposals (which we tentatively call

‘Basel III’) are likely to be a game changer for the banks.

This is a key part of our report. We see Lloyds and HSBC’s

Core Tier 1 ratios falling substantially, to 4.4% and 6.0%

respectively. We see other banks as adequately

capitalised, with the three Nordic banks having

considerable excess capital.

• We are positive on the group as a whole. Economic growth

is likely to surprise on the upside and we foresee the yield

curve remaining steep for a prolonged time.

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

19TH JANUARY 2010

THE BASEL III GAME CHANGER

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

• We initiate coverage on the large cap European retail and corporate lending banks (which we broadly term ‘lending’ banks) for whom the vast proportion of operating earnings 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 DNBNOR, Handelsbanken and Nordea 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.

• Basel III Capital Analysis. We believe that changes in bank capital regulations are going to be 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 detailed 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. We also see HSBC’s Core Tier 1 ratio falling to 6.0%, significantly below peers. The other banks are reasonably capitalised, with the Nordic banks in particular having substantial excess capital.

• In our Sector and Economic Overview, we conclude that the lending banks are operating in a very attractive economic environment which is likely to last until at least Q4 2010. We show, looking at past recessions, that economic growth is likely to surprise on the upside. We believe the yield curve will remain steep for a prolonged period, with the short end depressed by the magnitude of the output gap and the long end remaining at a high level as governments seek to attract investors to the significant volume of new debt issuance.

• We rate as BUY: DNBNOR (Recovering Norwegian economy, normalizing loan losses, very strong Basel III capitalization, cheap); Handelsbanken (Best quality bank, to be appreciated more when the extent of Basel III excess capital is realized); Nordea (Solid Basel III capitalization, well run bank with good earnings momentum, cheap).

• We rate as HOLD: BBVA (Very high ROE, attractive Latam footprint, near term concerns regarding Spanish loan losses); HSBC (Deposit surplus is a strong competitive advantage, attractive Asian footprint, weak Basel III capitalization); Intesa Sanpaolo (Well run, defensive bank, but with a structurally low ROE); Santander (Well managed for growth and stability, but near term concerns on Spanish loan losses); Standard Chartered (Attractive Asian footprint, large deposit surplus, not expensive but not cheap either).

• We rate as REDUCE: Lloyds (Large Basel III capital shortfall, earnings growth constrained by LTD ratio of 170%, expensive); Unicredit (Structurally low ROE bank, not cheap).

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.

P AN - E U R OP E AN B AN KS

19 J AN U AR Y 20 10 2

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 J AN U AR Y 20 10 3

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 J AN U AR Y 20 10 4

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 J AN U AR Y 20 10 5

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 J AN U AR Y 20 10 6

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

Matrix Mcap Cur Price DCF DCF DCF FV upside/

Risk-free

LT market

Market risk Equity

LT growth COE

Terminal 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 J AN U AR Y 20 10 7

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 CHARTER

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%

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 1.20 1.10 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 CHARTER

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%

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 2.22 9.0% 12.1%

Source: Bloomberg Consensus Estimates


19 J AN U AR Y 20 10 8

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 J AN U AR Y 20 10 9

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 J AN U AR Y 20 10 10

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

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

1948-49

1953-541957-58

1960-611970

1973-75

1980

1981-821990-91

20012008-09 Aug Forecast

2008-09 Dec Forecast

0

2

4

6

8

10

12

14

16

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50

GD

P G

row

th in

nex

t 6 Q

uart

ers

(%)

Peak to Trough GDP Decline (%)


19 J AN U AR Y 20 10 11

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

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

0

2000

4000

6000

8000

10000

12000

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

Billion $

Personal Disposable Income Personal Consumption Expenditure


19 J AN U AR Y 20 10 12

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


Figure 7: Banks Reserves with the Bank of England


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

0

200

400

600

800

1000

1200

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

Billion USD

Total Banks Reserves Required Banks Reserves

Excess Reserves

0

20

40

60

80

100

120

140

160

180

May-06 Oct-06 Mar-07 Aug-07 Jan-08 Jun-08 Nov-08 Apr-09 Sep-09

Billion GBP


19 J AN U AR Y 20 10 13

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


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)


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

-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

0

2

4

6

8

10

12

14

0

50

100

150

200

250

300

350

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

$ Thousand Months Supply


19 J AN U AR Y 20 10 14

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


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

70

80

90

100

110

120

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

Jan 2005 = 100


19 J AN U AR Y 20 10 15

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


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.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

1982-83 1990-92 2001-02 2008-09

Zero indicates recessionary trough


19 J AN U AR Y 20 10 16

Figure 12: Five-Year Breakeven Inflation Rate¹

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

-1.0

0.0

1.0

2.0

3.0

4.0

Jan-08 Apr-08 Jul-08 Oct-08 Jan-09 Apr-09 Jul-09 Oct-09

Percent

US UK


19 J AN U AR Y 20 10 17

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

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.

0.0

1.0

2.0

3.0

4.0

5.0

Curren

t

Ap

r-10

Jul-10

Oct-10

Jan-11

Ap

r-11

Jul-11

Oct-11

Jan-12

Ap

r-12

Jul-12

Oct-12

Jan-13

Ap

r-13

Jul-13

Oct-13

US Fed Funds UK 90-Day Short Sterling

%


19 J AN U AR Y 20 10 18

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 J AN U AR Y 20 10 19

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 J AN U AR Y 20 10 20

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


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


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.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%H

and

elsb

anke

n

No

rdea

DN

B N

OR

Un

icre

dito

Inte

sa

San

tan

der

Sta

n. C

har

tere

d

BB

VA

HS

BC

Llo

yds

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

-2.0%

-1.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

Han

del

sban

ken

No

rdea

DN

B N

OR

Un

icre

dito

Inte

sa

San

tan

der

Sta

n. C

har

tere

d

BB

VA

HS

BC

Llo

yds

Excess core equity capital vs 6.0% minimum, 2012e


19 J AN U AR Y 20 10 21

• 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 J AN U AR Y 20 10 22

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


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.

-5.0%-4.5%-4.0%-3.5%-3.0%-2.5%-2.0%-1.5%-1.0%-0.5%0.0%

DN

B N

OR

Inte

sa

No

rdea

San

tan

der

Han

del

sban

ken

Un

icre

dito

BB

VA

Sta

n. C

har

tere

d

Llo

yds

HS

BC

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


19 J AN U AR Y 20 10 23

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


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).

0.0

10.0

20.0

30.0

40.0

50.0

60.0

Llo

yds

HS

BC

Han

del

sban

ken

No

rdea

Sta

n. C

har

tere

d

Un

icre

dito

San

tan

der

BB

VA

DN

B N

OR

Inte

sa


19 J AN U AR Y 20 10 24

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



19 J AN U AR Y 20 10 25

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


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:


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



19 J AN U AR Y 20 10 26

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 financial institutions

P19 Q3 2009 Report. This is shown as "Other adjustments" in the accounts, but IR explains that this pertains to stakes in financial holdings with a participation >10% (mainly the stakes in Citic Bank and CIFH, and also some insurance companies). 50% (i.e. €1527m) is deducted from T1 and 50% from T2.

Investments in other financial institutions are reported as zero. P33 of Q3 2009 Report.

50% of investments in other financial institutions already deducted from T1 capital. Remaining 50% deducted from T2 capital. However, this only pertains to capital contributions made after 20 July 2006. Under current rules, investments in financial institutions made prior to that are deducted 100% from total capital.

P190, H1 2009 Interim Report. Mainly pertaining to investments in insurance companies. This amount is deducted from T2 capital but not T1 capital.

Figure not disclosed in accounts. IR states it as "a few hundred million".

Excess of expected IRB losses over existing impairment allowances

Generic provisions are not included in Tier 1 capital. We treat this as a buffer which already fully covers any excess of expected losses using IRB over existing impairment allowances.

P33 of Q3 2009 Report. This is 50% of the amount which is deducted from Tier 2 capital.

P34 of Q3 2009 Report. 50% of the excess of expected losses over impairment allowances is deducted from T1 capital and 50% from T2 capital.

P190, H1 2009 Interim Report. 50% of excess of expected losses over impairment provisions (i.e. $3375m) is already deducted from T1 capital, but the other 50% is deducted from T2 capital.

€140m deducted from each of T1 and T2 capital. Disclosed by IR.

Minorities As per the balance sheet. Confirmed by IR. P16 of Q3 2009 Factbook. P190 of H1 2009 Interim Report.

P44 Q3 2009 Report. Confirmed by IR.

Negative AFS and cash flow hedging reserves added back to T1

Spanish banks already deduct all negative AFS reserves from Tier 1 capital. In any case, BBVA did not have any negative reserves.

P33 Q3 2009 Report. Deduction of NOK2284m from T1 for losses made on bond assets recorded at fair value already made.

Negative AFS reserves and cash flow hedges are added back to Tier 1 capital. P34 of Q3 2009 Report.

P190 of H1 2009 Interim Report. Includes the reversal of the deduction arising from the gain on own debt from T1 capital.

Page 44, 9M09 accounts. Negative AFS and cash flow hedging reserves already fully deducted from T1 under Italian regulations.

Deferred tax assets, for losses incurred

P129 of 2008 Financial Accounts. Figure is for DTA for "Tax losses and other". DTA for other items are not eligible deductions from Tier 1 capital according to IR.

DTAs already deducted from Tier 1 capital. P33 of Q3 2009 Report

Deferred tax assets already deducted from Tier 1 capital. P16, Q3 2009 Factbook

P201, H1 2009 Report. Net of deferred tax liabilities.

IR discloses that only EUR600m of gross DTAs relate to losses.

Defined benefit pension deficit

P108 and 109, notes 24 and 25.2 of 2008 Financial Accounts. Net pension liability has been provided for in full.

From Q3 2009 Accounts, p33.

This actually relates to surplus value pension assets rather than a defined benefit pension deficit. The amount is deducted from total capital, not T1 capital.

P212, H1 2009 Report IR states "not material".


Total existing Tier 1 hybrid instruments

As per the balance sheet. T1 hybrids are entirely comprised of perpetual subordinated loan capital.

As per the balance sheet. As per the balance sheet. As per the balance sheet.

Of which total preference shares

As per the balance sheet. DNBNOR has no preference shares.

SEK 5.2bn are non-innovative hybrids (including a staff convertible loan of SEK 2.3bn). The rest of the T1 hybrids are innovative hybrids.

As per the balance sheet. As per the balance sheet.

Add: Eligible Tier 1 hybrid instruments

40% of preference shares considered eligible as Tier 1 capital.

There are no preference shares, so there is no eligible hybrid T1 in our opinion.

Estimate that 40% of the non-innovative hybrids satisfy the new Basel III criteria.

UK FSA excludes nearly all current preference shares as well as innovative 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.

40% of preference shares considered eligible as Tier 1 capital.


Market risks Data provided by IR (not disclosed in reports).

P153 2008 Annual Report. Adjusted for 2009E.

P35 Q3 2009 Interim Report. Adjusted for 2009E.

P191, H1 2009 Report. Adjusted for 2009E.

P45 Q3 2009 Report. Adjusted for 2009E.



19 J AN U AR Y 20 10 27

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 financial institutions

P60 H1 2009 Interim Report. This amount is deducted in T2 capital but not T1 capital. Mainly pertaining to investments in insurance entities (which is £8729m of total).

P61 Q3 2009 Factbook. Relates to deductions for investments in insurance companies. This figure is deducted from T2 capital, not T1 capital.

P15 Q3 2009 Report. IR states that virtually all deductions pertain to investments in other financial institutions. Total amount of €1872m is deducted 50% from T1 and 50% from T2 capital.

P39 H1 2009 Report. "Material holdings and securitisations" are deducted 50% from T1 capital and 50% from T2 capital. The amount deducted here is the amount deducted from T2 capital.

IR will not disclose figure. They state "it is a substantial part of deductions. Deduction is 50:50 between T1 and T2 capital for Banks' and Financial Institution and 100% from T2 for Insurance". Disclosed deductions from T2 are just above €2bn, so we use €2bn as an estimate.

Excess of expected IRB losses over existing impairment allowances

Lloyds actually books a surplus of £2884bn of provisions against expected losses on underlying IRB portfolios. The positive delta is included in T2 capital (not T1 capital). P60 of H1 2009 Interim Report.

Most recent data is only available from the 2008 Annual Report, Table 43, p50. NOK269mn is the deduction from T2 capital, which is 50% of the total. The other 50% has already been deducted from T1 capital.

Generic provisions are not included in Tier 1 capital. We treat this as a buffer which already fully covers any excess of expected losses using IRB over existing impairment allowances.

P39 of H1 2009 Interim Report. This is 50% of the excess of expected losses over existing impairment allowances which is deducted from T2 capital. NB There is no tax adjustment since this has already been accounted for in its entirety in T1 capital.

Excess of losses over impairment charges is split 50:50 between T1 and T2 capital. We deduct the 50% deducted from T2.

Minorities IR indicates that minorities included in common equity is similar to that disclosed on balance sheet.

As per balance sheet. As per balance sheet. P39 H1 2009 Report Confirmed by IR.

Negative AFS and cash flow hedging reserves added back to T1

Page 60, H1 2009 accounts. Includes cash flow hedging reserve.

P33 Q3 2009 Report Spanish banks already deduct all negative AFS reserves from capital.

P44 H1 2009 Report. Figure needs to be adjusted slightly to obtain the amount added back to Tier 1 capital for bond assets only. Exact figure not disclosed by IR.

Negative reserves already deducted 100% from T1 capital under Italian regulations.

Deferred tax assets, for losses incurred

Page 89, H1 2009 Interim Accounts

P47 Q3 2009 Report IR discloses that it is EUR1.5-2.0bn.

P43 H1 2009 Report DTA stemming from tax loss 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 deficit

P60, H1 2009 Interim Report

P32 Q3 2009. Defined benefit pension deficit, net of pension assets.

IR confirms that it is zero. Exact figure not disclosed by IR. IR states that the deficit will be very small.

The small deduction is actually from an overfunded defined benefit pension scheme (i.e. Surplus assets). P279/280 2008 Annual Accounts.


Total existing Tier 1 hybrid instruments

As per the balance sheet. As per the balance sheet. Hybrid capital is all preference shares. As disclosed by IR.

As per the balance sheet. As per the balance sheet.

Of which total preference shares

As per the balance sheet. Hybrid capital contains no preference shares. Majority is innovative hybrid.

As per the balance sheet. As per the balance sheet. P226 Q3 2009 Report: "Non-innovative capital instruments", which we interpret as preference shares.

Add: Eligible Tier 1 hybrid instruments


There are no preference shares, so there is no eligible hybrid T1 in our opinion.

40% of existing preference shares considered eligible as T1 hybrid under the new proposals.


We assume 40% of total preference shares are eligible as T1 hybrids under the new proposals.


Market risks P62 of H1 2009 Interim Report. Adjusted for 2009E.

P44 Q3 2009 Report. Adjusted for 2009E.

Disclosed by IR. Not available in Q3 2009 Report.

P39 H1 2009 Report. Adjusted for 2009E.

P227 of Q3 2009 Extended Report. Figures adjusted for 2009E.



19 J AN U AR Y 20 10 28

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 J AN U AR Y 20 10 29

• 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 J AN U AR Y 20 10 30

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

COMMON EQUITY

("Common" or "Core")

Exclude hybrid capital, (including preference shares).

Exclude hybrid capital, (including preference shares).

Essentially common shares plus retained earnings, net of regulatory adjustments and deductions (as proposed).

Include minorities. Exclude minorities.

Include defined benefit pension deficits. Exclude defined benefit pension deficits and assets.

No deductions made. Deduct negative AFS reserves (previously added back to Tier 1 capital).

Deduct negative cash flow hedging reserves

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.

TIER 1 CAPITAL

(T1)

Include Tier 1 hybrid capital. Exclude innovative hybrid capital (i.e. with some form of synthetic maturity; this encompasses nearly all hybrids currently in issue). Some preference shares in issue are also excluded; most are permanent capital and have non-cumulative dividends (which satisfies the criteria) but do not have fully discretionary dividends. However, UK FSA seems to ban all existing preference shares from T1.

Remainder of the T1 capital base must be comprised of instruments that are subordinated, have fully discretionary non-cumulative 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. This appears to exclude innovative hybrid capital but allows preference shares in their current form. Note that the UK FSA appears to have additionally moved to exclude preference shares in their current form (because of no loss absorption feature).

Include minorities (part of common equity). Include minorities (but not part of common equity).

Exclude defined benefit pension deficits and surplus assets (part of common equity).

Defined benefit pension deficits and surplus assets now deducted from common.

Add back negative AFS reserves pertaining to bond assets.

Negative AFS reserves now deducted from common.

Add back negative cash flow reserves. Negative cash flow reserves now deducted from common.

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 entities. Other 50% deducted from T2. Some banks deduct 100% from T2.

100% of investment in financial entities now deducted using a "corresponding deduction approach".

Deduct 50% of excess of expected IRB losses over impairment allowances. Other 50% deducted from T2.

100% of excess of expected IRB losses over existing impairment allowances now deducted from common.

TIER 2 CAPITAL

(T2)

Some deductions (as above) for investments in financial entities and excess of expected IRB losses over impairment allowances.

Nearly all deductions now made from common capital. However, investments in the Tier 2 capital of other financial entities needs to be deducted.

T2 capital, available to absorb losses on a "gone" concern basis, reduced to a single level i.e. no lower (LT2) and upper T2 (UT2) only old LT2 i.e. plain vanilla sub debt.

TIER 3 CAPITAL

(T3)

Included in total capital. Eliminated under Basel III proposals. T3 is currently comprised of sub debt used as capital against market risk. Under Basel III proposals, T3 will be eliminated.


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 J AN U AR Y 20 10 31

• 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 J AN U AR Y 20 10 32

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 J AN U AR Y 20 10 33

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 J AN U AR Y 20 10 34

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 J AN U AR Y 20 10 35

• 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 J AN U AR Y 20 10 36

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.

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.

ROE = Net Profit = Net Profit x AssetsEquity Assets Equity

= ROA x Equity Multiplier


19 J AN U AR Y 20 10 37

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).

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).

NII = NII x IEAAssets Interest-earning assets Assets

= Net Interest Margin x Earning Asset Ratio

PA

N-E

UR

OP

EA

N B

AN

KS

1

9 J

AN

UA

RY

20

10

3

8

Fig

ure

23:

Du

po

nt

Sys

tem

fo

r B

anks

Sou

rce:

Mat

rix C

orpo

rate

Cap

ital R

esea

rch.

Dup

ont

mo

del a

dapt

ed

for

bank

s fr

om

Suc

cess

ful B

ank

Ass

et/L

iabi

lity

Man

age

men

t: A

gu

ide

to th

e fu

ture

bey

ond

gap

(Joh

n W

iley

& S

ons,

199

2).

RETU

RN O

N

AV

ERA

GE

EQU

ITY

RETU

RN O

N A

VER

AG

E TO

TAL A

SSET

S

TOTA

LASS

ETS

% E

QU

ITY

(i.e.

Equ

ity M

ultip

lier)

NET

INTE

REST

INCO

ME

% A

V. T

OTA

L A

SSET

S

NET

NO

N-I

NTE

REST

OPE

RATI

NG

IN

COM

E%

AV

. TO

TAL A

SSET

S

NET

NO

N-O

PERA

TIN

G IN

COM

E%

AV

. TO

TAL

ASS

ETS

OPE

RATI

NG

CO

STS

% A

V. T

OTA

L A

SSET

S

LOA

N LO

SS C

HA

RGES

%

AV

. TO

TAL

ASS

ETS

GRO

UP

TAX

% A

V. T

OTA

L A

SSET

S

NET

INTE

REST

INCO

ME

%

AV

. IN

TERE

ST-E

ARN

ING

ASS

ETS

(i.e.

Net

Inte

rest

Mar

gin)

INTE

REST

-EA

RNIN

GA

SSET

S %

AV

. TO

TAL

ASS

ETS

(i.e.

Ear

ning

Ass

et R

atio

)


19 J AN U AR Y 20 10 39

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 (Assets

(as % of average

op. income income costs Margin(NII % IEA)

asset ratio

Company % Equity) total assets)

(IEA %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%



19 J AN U AR Y 20 10 40

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


-0.60%-0.40%-0.20%0.00%0.20%0.40%0.60%0.80%1.00%

0.05.0

10.015.020.025.030.035.0


19 J AN U AR Y 20 10 41

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


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


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

100%

NII % operating revenue Non-NII % operating revenue

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

0 bps100 bps200 bps300 bps400 bps500 bps

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


19 J AN U AR Y 20 10 42

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 J AN U AR Y 20 10 43

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 J AN U AR Y 20 10 44

ROE = ROA x Equity Multiplier

Figure 29: ROE, Spanish banks


Figure 30: ROE, Italian banks


Figure 31: ROE, Nordic banks


Figure 32: ROE, UK banks


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).

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

E

BBVA Santander

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

E

Intesa Sanpaolo Unicredito

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

E

DNBNOR Handelsbanken Nordea

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

E

HSBC Lloyds Standard Chartered


19 J AN U AR Y 20 10 45

Figure 33: ROA, Spanish banks


Figure 34: ROA, Italian banks


Figure 35: ROA, Nordic banks


Figure 36: ROA, UK banks


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.

-1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

A

BBVA Santander

-1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

A


-1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

A


-1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

RO

A



19 J AN U AR Y 20 10 46

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


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


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


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


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.

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

AS

SE

TS %

EQ

UIT

Y

BBVA Santander

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

AS

SE

TS %

EQ

UIT

Y


0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

AS

SE

TS %

EQ

UIT

Y


0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

AS

SE

TS %

EQ

UIT

Y



19 J AN U AR Y 20 10 47

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


Figure 43: NII % Assets, Nordic banks


Figure 44: NII % Assets, UK banks


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.

0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

ave

rag

e to

tal a

sset

s

BBVA Santander

0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

ave

rag

e to

tal a

sset

s


0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

ave

rag

e to

tal a

sset

s


0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

ave

rag

e to

tal a

sset

s



19 J AN U AR Y 20 10 48

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


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


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


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


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.

0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-N

II o

per

atin

g in

com

e %

ave

rag

e as

sets

BBVA Santander

0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-N

II o

per

atin

g in

com

e %

ave

rag

e as

sets


0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-N

II o

per

atin

g in

com

e %

ave

rag

e as

sets


0 bps

50 bps

100 bps

150 bps

200 bps

250 bps

300 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-N

II o

per

atin

g in

com

e %

ave

rag

e as

sets



19 J AN U AR Y 20 10 49

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


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


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


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


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.

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

200 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-o

per

atin

g in

com

e %

ave

rag

e as

sets

BBVA Santander

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

200 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-o

per

atin

g in

com

e %

ave

rag

e as

sets


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

200 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-o

per

atin

g in

com

e %

ave

rag

e as

sets


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

200 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

no

n-o

per

atin

g in

com

e %

ave

rag

e as

sets



19 J AN U AR Y 20 10 50

Figure 53: Operating costs % assets, Spanish banks


Figure 54: Operating costs % assets, Italian banks


Figure 55: Operating costs % assets, Nordic banks


Figure 56: Operating costs % assets, UK banks


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.

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Op

erat

ing

co

sts

% a

vera

ge

asse

ts

BBVA Santander

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Op

erat

ing

co

sts

% a

vera

ge

asse

ts


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Op

erat

ing

co

sts

% a

vera

ge

asse

ts


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Op

erat

ing

co

sts

% a

vera

ge

asse

ts



19 J AN U AR Y 20 10 51

Figure 57: Loan loss charges % assets, Spanish banks


Figure 58: Loan loss charges % assets, Italian banks


Figure 59: Loan loss charges % assets, Nordic banks


Figure 60: Loan loss charges % assets, UK banks


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.

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

loss

cha

rges

% a

vera

ge

asse

ts

BBVA Santander

-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

loss

cha

rges

% a

vera

ge

asse

ts


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

loss

cha

rges

% a

vera

ge

asse

ts


-250 bps

-200 bps

-150 bps

-100 bps

-50 bps

0 bps

50 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

loss

cha

rges

% a

vera

ge

asse

ts



19 J AN U AR Y 20 10 52

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


Figure 63: Group tax % average assets, Nordic banks


Figure 64: Group tax % average assets, UK banks


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.

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Gro

up ta

x %

ave

rag

e as

sets

BBVA Santander

-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Gro

up ta

x %

ave

rag

e as

sets


-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Gro

up ta

x %

ave

rag

e as

sets


-100 bps

-50 bps

0 bps

50 bps

100 bps

150 bps

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Gro

up ta

x %

ave

rag

e as

sets



19 J AN U AR Y 20 10 53

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


Figure 67: Net Interest Margin, Nordic banks


Figure 68: Net Interest Margin, UK banks


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.

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

inte

rest

-ear

ning

ass

ets

BBVA Santander

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

inte

rest

-ear

ning

ass

ets


0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

inte

rest

-ear

ning

ass

ets


0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Net

inte

rest

inc

om

e %

inte

rest

-ear

ning

ass

ets



19 J AN U AR Y 20 10 54

Figure 69: Loan % Deposit ratio, Spanish banks


Figure 70: Loan % Deposit ratio, Italian banks


Figure 71: Loan % Deposit ratio, Nordic banks

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

Figure 72: Loan % Deposit ratio, UK banks


• 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.

50%

100%

150%

200%

250%

300%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

% d

epo

sit r

atio

BBVA Santander

50%

100%

150%

200%

250%

300%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

% d

epo

sit r

atio


50%

100%

150%

200%

250%

300%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

% d

epo

sit r

atio


50%

100%

150%

200%

250%

300%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Loan

% d

epo

sit r

atio



19 J AN U AR Y 20 10 55

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 J AN U AR Y 20 10 56

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


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


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


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.

60.00%

65.00%

70.00%

75.00%

80.00%

85.00%

90.00%

95.00%

100.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Inte

rest

-ear

ning

ass

ets

% t

ota

l ass

ets

BBVA Santander

60.00%

65.00%

70.00%

75.00%

80.00%

85.00%

90.00%

95.00%

100.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Inte

rest

-ear

ning

ass

ets

% t

ota

l ass

ets


60.00%

65.00%

70.00%

75.00%

80.00%

85.00%

90.00%

95.00%

100.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Inte

rest

-ear

ning

ass

ets

% t

ota

l ass

ets


60.00%

65.00%

70.00%

75.00%

80.00%

85.00%

90.00%

95.00%

100.00%

1Q06

2Q06

3Q06

4Q06

1Q07

2Q07

3Q07

4Q07

1Q08

2Q08

3Q08

4Q08

1Q09

2Q09

3Q09

4Q09

e

1Q10

e

2Q10

e

3Q10

e

4Q10

e

1Q11

e

2Q11

e

3Q11

e

4Q11

e

1Q12

e

2Q12

e

3Q12

e

4Q12

e

Inte

rest

-ear

ning

ass

ets

% t

ota

l ass

ets



19 J AN U AR Y 20 10 57

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 J AN U AR Y 20 10 58

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 J AN U AR Y 20 10 59

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 J AN U AR Y 20 10 60

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 J AN U AR Y 20 10 61

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 J AN U AR Y 20 10 62

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 J AN U AR Y 20 10 63

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 J AN U AR Y 20 10 64

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 J AN U AR Y 20 10 65

VALUATION

We adopt a discounted cash flow (DCF) valuation approach for the lending banks. We like this approach because it can be linked to our Dupont analysis (which we have modelled up until the end of 2012) and to our long-term earnings and balance sheet models for each and every bank. This provides a sanity check that the 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)


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 conventional valuation table displays our targets prices as well. We use the DCF valuation of the banks as a guide in determining our target prices in most cases. The 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

DCF FVStock 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 % FVBBVA 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%

Cost of Equity Assumptions

Our DCF valuation is linked to our Dupont analysis and long-term financial models to provide a sanity check that the valuation methodology is grounded in realistic earnings assumptions.

The DCF valuation is used as the basis for our target prices in most cases, except BBVA and HSBC.


19 J AN U AR Y 20 10 66

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/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%

Matrix ROTNAVMatrix TNAVPS Matrix P/TNAVMatrix P/EMatrix EPS


19 J AN U AR Y 20 10 67

COMPANY SUMMARIES

This page has been left blank intentionally.


19 J AN U AR Y 20 10 68

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 FV 80% EPS 1.35 1.22 1.46 2.06

Price 12.81 Upside/Downside 25% P/E 9.47 10.51 8.80 6.22

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%


Investment Summary

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 set to post the worst GDP growth in developed Europe in 2010 according to consensus estimates, and where the unemployment rate should exceed 20%.

Our Basel III Analysis points to adequate capital adequacy, but note that the Core Tier 1 ratio is the third lowest in the peer group. We estimate that on a Basel III basis, it will be 7.0% by the end of 2012.

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

Company Description

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

Analyst: Andrew Lim +44 20 3206 7347 [email protected]

Divisional breakdown of operating profit, 9M09

Spain & Portugal

61%Wholesale Banking &

Asset Management

7%

Mexico (Bancomer)

17%

USA4%

South America

11%

Divisional breakdown of group loans, 9M09

Spain & Portugal

62%

Wholesale Banking &

Asset Management

12%

Mexico (Bancomer)

8%

USA11%

South America

7%


19 J AN U AR Y 20 10 69

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%



19 J AN U AR Y 20 10 70

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 FV 81% EPS 6.10 5.47 6.92 8.80









Investment Summary

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 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 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.

Our DCF valuation points to considerable upside. DNBNOR also has one of the lowest P/TNAV in the group. This is not warranted in our opinion given the attractive ROE and low cost of capital, which we see as underpinned by the relatively small capital, liquidity and earnings risks.

Company Description

DNBNOR is Norway’s largest financial services group. The bank is broadly split into Retail Banking, Large Corporate and International Operations (mainly wholesale lending to the group’s largest Norwegian and international customers), life insurance and asset management (Vital being the key life business), DNBNOR Markets (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).

NORD L/B owns the other 49% of DNB NORD. DNBNOR is currently undertaking an 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.



Retail Banking

24%

Large Corporates

and International

43%

DnB NOR Markets

27%

Life and Asset

Management6%


Retail Banking

62%

Large Corporates

and International

31%

DnB NORD7%


19 J AN U AR Y 20 10 71

Figure 82: DNBNOR Financial Summary

DNB NOR ASA FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E

Income Statement (NOK m)




Other revenue 2,202 2,977 3,830 3,959 4,092


Operating costs -18,719 -19,215 -19,795 -20,556 -21,676





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


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 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)


Customer deposits 1,203,464 1,126,571 1,172,314 1,268,951 1,373,553




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%


LLC % gross loans -0.33% -0.72% -0.67% -0.42% -0.22%



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%


RWA % total assets 65.55% 57.52% 57.12% 55.37% 53.63%



19 J AN U AR Y 20 10 72

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 FV 81% EPS 16.46 17.19 19.68 23.27









Investment Summary

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

Svenska Handelsbanken is keen to highlight its decentralised branch network. Each branch is responsible for all customers within its geographic area, including the largest companies. The main division in the group is simply called ‘Bank Operations’, 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. Sweden is its home market. The bank has 461 branches in Sweden, 54 in Denmark, 45 in Finland, 48 in Norway and 66 in Great Britain.



Bank operations

total83%

Handelsbanken Capital markets

16%

Handelsbanken Asset

Management1%


19 J AN U AR Y 20 10 73

Figure 84: Handelsbanken Financial Summary

SVENSKA HAN-A FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E

Income Statement (SEK m)




Other revenue 975 487 500 511 522


Operating costs -13,137 -14,820 -15,067 -16,107 -17,342




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 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 assets 2,158,784 2,180,415 2,357,009 2,566,602 2,795,591


Liabilities (SEK m)






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%


LLC % gross loans -0.12% -0.25% -0.26% -0.19% -0.09%



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%


RWA % total assets 33.44% 28.37% 27.30% 27.14% 26.97%



19 J AN U AR Y 20 10 74

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 FV 77% EPS 0.48 0.70 1.03 1.41









Investment Summary

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.

• HSBC’s loan loss experience should improve earlier than peers. Recent results point to the first reduction in impairments in the US Household Personal Finance division since the beginning of 2006. The US loan book is, in our opinion, relatively mature given its heavy weighting towards early cycle subprime mortgages.

• HSBC has been able to improve its LTD ratio from just below 100% at the beginning of 2007 to 79% as of H1 2009, as deposits have undergone a flight to quality. Our Dupont analysis shows that this has enabled HSBC to maintain a steady NIM throughout the crisis, whilst many expected the fall in base rates to have led to a contraction.

The excess of deposits over loans means that HSBC has considerable capacity to write high margin new loans to increase its NII and NIM (via a positive mix effect).

Our key concern with HSBC, however, (and the reason we initiate with a Hold rating and not a Buy) is that under our Basel III Analysis, the bank’s Core Tier 1 ratio falls 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.

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.

Company Description

HSBC is a truly global bank with operations in Europe, Asia, the Americas, Middle East and Africa. Asia, accounting for 25% of operating profit (including Hong Kong) is a key part of the franchise given its growth potential. HSBC manages its business through two customer groups, Personal Financial Services and Commercial Banking (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 businesses. HSBC plans to list in Shanghai in the first half of 2010.


Divisional breakdown of operating profit, H1 2009

Personal Financial Services

43%

Commercial Banking

17%

Global Banking and

Markets37%

Private Banking

3%

Divisional breakdown of group loans, H1 2009

Personal Financial Services

43%

Commercial Banking

22%

Global Banking and

Markets31%

Private Banking

4%

Other0%

Geographical breakdown of operating profit, H1 2009

Europe32%

Hong Kong15%

Rest of Asia Pacific10%

Middle East5%

North America

27%

Latin America11%


19 J AN U AR Y 20 10 75

Figure 86: HSBC Financial Summary

HSBC HLDGS PLC FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E

Income Statement (US$m)




Other revenue 13,999 3,642 6,520 6,994 7,294


Operating costs -38,535 -34,600 -37,426 -40,201 -42,912



Investment income 197 823 1,400 1,400 1,400


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


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 assets 2,527,465 2,519,227 2,698,947 2,896,843 3,114,796


Liabilites (US$m)






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%


LLC % gross loans -2.55% -3.06% -2.60% -1.90% -1.20%



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%


RWA % total assets 45.42% 46.48% 45.14% 43.75% 42.33%



19 J AN U AR Y 20 10 76

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










Investment Summary

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.

Our Basel III Analysis points to an adequate Core Tier 1 ratio of 7.6% by the end of 2012. The fall in the Core Tier 1 ratio is one of the lowest in the group, reflecting the conservatism of its banking regulator and the fact that it does not have a corporate structure involving significant minority interests (unlike Unicredit).

Our DCF valuation points to attractive upside for the company. However, we see extra upside, greater growth potential and even more robust balance sheets in the Nordic banks. We therefore initiate with a HOLD.

Company Description

Intesa Sanpaolo is Italy’s second largest bank by market cap, just behind Unicredit, although it has the largest presence by share of loans and deposits in the domestic Italian market. The bank was created in September 2007 by the merger of Intesa and San Paolo IMI. The main division of the group is Banca Dei Territori, which encompasses its numerous retail banking franchises in Italy and accounts for 48% of operating profit. Corporate and Investment banking accounts for 32% of operating profit. The group’s international retail banking exposure is relatively modest and is comprised mainly of operations in the CEE. Other smaller divisions include Public Finance, Eurizon (an asset manager with €135bn under management) and Banca Fideuram, a high quality financial advisor network in Italy.



Banca Dei Territori

48%

Corporate and

Investment Banking

33%

Public Finance

4%

International Subsidiary

Banks11%

Eurizon Capital

1%

Banca Fideuram

3%


Banca Dei Territori

52%

Corporate and

Investment Banking

29%

Public Finance

11%

International Subsidiary

Banks8%


19 J AN U AR Y 20 10 77

Figure 88: Intesa Sanpaolo Financial Summary

INTESA SANPAOLO FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E




Trading profits/losses -53 1,170 1,439 1,511 1,586

Other revenue 708 584 684 763 843


Operating costs -9,936 -9,379 -9,771 -10,371 -10,973




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)


Interbank loans 56,371 55,000 57,750 63,525 69,878



Total assets 636,133 625,184 636,623 665,285 697,923


Liabilities (€m)






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%


LLC % gross loans -0.69% -0.90% -0.82% -0.67% -0.47%



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%


RWA % total assets 60.22% 58.47% 58.28% 59.19% 59.88%



19 J AN U AR Y 20 10 78

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 FV 116% EPS -10.31 -0.45 4.24 7.68

Price 56.78 Upside/Downside -14% P/E -5.51 -126.21 13.41 7.39



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%



Investment Summary

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 suffer pressure on its net interest earnings growth as it tries to reduce its loan to deposit ratio from ~170% to ~140%. The reliance on wholesale funding is a key reason why HBOS got into trouble in the first place. It is a structural problem that Lloyds has inherited and we fear the market has ignored the impact on earnings that rectifying it will have.

Our DCF valuation points to absolute downside. Lloyds also looks expensive on conventional valuation metrics over the next few years given that earnings will be 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 the fact that the ROE does not exceed the cost of capital until 2012 – and that is without taking into consideration the need to solve the Basel III capital dilemma.

Company Description

Lloyds acquired HBOS in January 2009. The combined group is a largely UK-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 to common equity when the Core Tier 1 ratio falls below 5.0%).



Retail39%

Wholesale46%

Insurance9%

Wealth and International

6%


Retail63%

Wholesale36%

Wealth and International

1%


19 J AN U AR Y 20 10 79

Figure 90: Lloyds Banking Group Financial Summary

LLOYDS BANKING FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E

Income Statement (£m)



Trading profits/losses -9,186 874 1,552 1,475 1,401

Other revenue 8,799 9,555 8,208 8,208 8,208


Operating costs -6,000 -12,201 -13,955 -12,354 -11,754




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


Net profit 772 -3,591 -286 2,697 4,890

Assets (£m)


Interbank loans 38,733 37,398 38,146 39,687 41,290



Total assets 436,033 1,071,486 1,061,729 1,057,280 1,055,177


Liabilities (£m)






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%


LLC % gross loans -1.25% -3.34% -2.39% -1.70% -1.05%



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%


RWA % total assets 39.10% 43.90% 44.30% 45.38% 47.31%



19 J AN U AR Y 20 10 80

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 FV 86% EPS 6.16 6.63 7.94 9.07









Investment Summary

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 the bank to have excess capital once the principles of Basel III are formalised.

Our DCF valuation points to Nordea having attractive upside, albeit not as much as for DNBNOR and Handelsbanken, our more favoured Nordic plays. Conventional P/E ratios place the bank as one of the cheaper stocks in the group on our 2010 and 2011 estimated earnings.

Company Description

Nordea is the largest financial institution in the Nordic area, which is considered its home market. It has the broadest presence in the region amongst its Nordic peers, with approximately 30% of the Nordic loan book in Denmark, 28% in Sweden, 22% in Norway and 19% in Norway.

The present form of the group, under the name of ‘Nordea’, has only been in existence since December 2001 and has been achieved through a number of cross-border Nordic mergers and acquisitions. Nordic Banking accounts for 80% of the group’s loan book. It is also present in Poland, Russia and the Baltic countries, which account for 5% of the group’s loan book. Other notable operations are Shipping Oil Services, Private Banking and Life Insurance.

Nordea’s two largest shareholders are Sampo and the Swedish state, with shares of 18.3% and 19.8% respectively. Sampo, the Nordic insurance group, has made clear its intention to participate in consolidation of the Nordic banking market via its shareholding.



Nordic Banking

65%

New European Market

7%

Financial Institutions

5%

Shipping Oil Services &

International7%

Other Customer Operations

(incl Pr. Banking &

Life)16%


Nordic Banking

80%

New European Market

5%

Financial Institutions

1%

Shipping Oil Services &

International5%

Other Customer Operations

(incl Pr. Banking &

Life)9%


19 J AN U AR Y 20 10 81

Figure 92: Nordea Financial Summary

NORDEA BANK AB FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E




Trading profits/losses 833 1,747 1,963 1,994 2,061

Other revenue 391 477 287 298 311


Operating costs -4,338 -4,350 -4,449 -4,648 -4,883


Total provisions -466 -1,536 -1,327 -844 -492



Other income 0 0 0 0 0

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


Net profit 2,671 2,454 2,641 3,163 3,616

Assets (€m)


Interbank loans 23,903 16,587 17,465 18,562 19,752



Total assets 474,074 491,612 521,565 557,531 593,054


Liabilities (€m)






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%


LLC % gross loans -0.17% -0.52% -0.43% -0.26% -0.14%



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%


RWA % total assets 44.99% 39.78% 40.58% 41.10% 38.63%



19 J AN U AR Y 20 10 82

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










Investment Summary

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.

We worry, however, about the prospect of elevated loan losses in Spain, a country which is forecast to have one of the worst GDP growth rates in developed Europe in 2010 (-0.40%) and where the unemployment rate is predicted to worsen to over 20%. Coverage ratios for the group have been run down to a lowly 73%. While lower 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 other banks.

Our Basle III Analysis shows Santander’s Core Tier 1 to be 7.5% at the end of 2012, 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 compared to the upside available at more favoured stocks in the group. Conventional valuation metrics also do not point to Santander being particularly cheap on 2010 P/E or P/TNAV. We note that the growth in earnings makes the stock 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.

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 acquisition strategy beginning with the purchase of Abbey National, whilst BBVA 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.



Continental Europe42%

United Kingdom

13%

Latin America43%

Sovereign2%


Continental Europe49%

United Kingdom

32%

Latin America14%

Sovereign5%


19 J AN U AR Y 20 10 83

Figure 94: Santander Financial Summary

BANCO SANTANDER FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E





Other revenue 927 686 745 774 804


Operating costs -14,949 -16,092 -16,528 -17,240 -17,997





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


Net profit 8,877 8,664 8,316 10,690 13,354

Assets (€m)


Interbank loans 64,731 46,658 48,543 52,545 56,876



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)






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%


LLC % gross loans -1.09% -1.49% -1.43% -1.05% -0.67%



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%


RWA % total assets 48.97% 50.32% 48.33% 48.31% 48.26%



19 J AN U AR Y 20 10 84

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 FV 105% EPS 1.91 2.25 2.71 3.20

Price 1551.50 Upside/Downside -4% P/E 13.27 11.27 9.36 7.92








Investment Summary

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, provides Standard Chartered with a considerable growth advantage versus peers.

Our Dupont Analysis has shown that the quality of earnings has deteriorated in recent quarters, with the proportion from net interest income becoming less, but made up by substantially higher (but lower quality) trading profits. We expect the mix of operating earnings to reverse in coming quarters towards better quality net interest income (as the excess of deposits are used to fund new lending) and away from trading profits, (which we do not see as sustainable at current levels and where higher risk weightings under Basel III will reduce returns).

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 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.

Despite all these positive points, however, our main concern is that the price, whilst not particularly expensive, is also not cheap. Our DCF valuation already shows the 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.

Company Description

Standard Chartered has a clear focus on emerging market territories, particularly Asia. The exposure within Asia is quite broad, but perhaps notable for the relative emphasis on Hong Kong, Singapore and Korea.

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



Consumer Banking

23%

Wholesale Banking

71%

Central Items6%


Consumer Banking

45%Wholesale Banking

55%

Geographical breakdown of group loans, H1 2009

Hong Kong16%

Singapore14%

Korea17%

Other Asia Pacific20%

India5%

Middle East & other S

Asia10%

Africa2%

Americas UK & Europe

16%


19 J AN U AR Y 20 10 85

Figure 96: Standard Chartered Financial Summary

STANDARD CHARTERED PLC FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E

Income Statement (US$m)




Other revenue 1,235 1,790 2,427 2,775 2,864


Operating costs -7,611 -8,184 -8,889 -9,779 -10,725





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


Net profit 3,408 3,901 4,787 5,742 6,775

Assets (US$m)


Interbank loans 46,583 45,366 45,366 47,199 49,106



Total assets 435,068 430,345 494,835 569,359 655,448


Liabilities (US$m)






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%


LLC % gross loans -0.64% -1.01% -0.73% -0.47% -0.30%



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%


RWA % total assets 43.40% 48.59% 45.71% 42.97% 40.37%



19 J AN U AR Y 20 10 86

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


Price 2.25 Upside/Downside -11% P/E 22.92 24.08 12.37 7.77








Investment Summary

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 main reason for Unicredit’s low ROE, according to our Dupont Analysis, is the low NIM which, unlike the Nordic banks, is not compensated for by an efficient cost base. The NIM has stayed roughly the same over the course of the credit crisis. The lack of improvement versus other franchises like HSBC, BBVA and Standard Chartered is a reflection in our opinion of the continued robustness of its competitors in most of its key markets, (particularly Italy). The competition makes it relatively difficult for Unicredit to increase its share of deposits and reduce its funding costs (note that its LTD ratio has stayed roughly the same) and to charge higher margins for new loans through better pricing power.

Additionally, Unicredit is suffering elevated loan losses from its CEE portfolio (with Russia and Ukraine having the most acute credit problems). We expect loan losses to be deeper for longer in these countries given economic imbalances. Exceptional trading profits made in 2009 are also expected to normalise at a lower level.

At 7.8% by the end of 2012, we consider Unicredit’s Core Tier 1 ratio as adequate under our Basel III Analysis.

Our DCF valuation points to absolute downside in the share price. We see Unicredit also being one of the most expensive banks on 2010 and 2011 estimated P/E 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).

Company Description

Unicredit Group is Italy’s largest bank by market cap, just ahead of Intesa Sanpaolo. It is more geographically diverse than Intesa. Corporate and Investment banking is the largest single division, accounting for 50% of operating profit and involved mainly in corporate lending. Unicredit has core retail lending operations in Italy, Germany and Austria, which account for 18%, 1% and 2% of operating profit respectively. Unicredit also has notable CEE operations across a variety of countries, the most significant being Poland, Croatia, Turkey and Russia. CEE in aggregate contributes 25% of operating profit despite accounting for only 14% of the loan book. Other smaller divisions include private banking, asset management. Having considered 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.


Breakdown of loans in the CEE region, 9M09

Croatia16%

Turkey14%

Russia13%

Czech Rep.12%

Hungary7%

Kazakhstan7%

Bulgaria7%

Ukraine6%

Romania5%

Slovakia4%

Slovenia4%

Bosnia2%

Baltics2%

Serbia1%


Retail banking -

Italy18%

Retail banking -Germany

1%

Retail banking -Austria

2%

Corporate and

investment banking

50%

Private banking

2%

Asset mgmt2%

Poland5%

CEE20%


Retail banking -

Italy21%

Retail banking -Germany

6%

Retail banking -Austria

3%

Corporate and

investment banking

55%

Private banking

1%

Poland3%

CEE11%


19 J AN U AR Y 20 10 87

Figure 98: Unicredit Financial Summary

UNICREDIT SPA FY 2008 FY 2009E FY 2010E FY 2011E FY 2012E




Trading profits/losses -1,968 1,899 2,006 2,106 2,212

Other revenue 1,379 841 940 1,014 1,090


Operating costs -16,692 -14,988 -14,611 -15,522 -16,754




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)


Interbank loans 80,827 100,000 100,000 105,000 110,250



Total assets 1,045,612 948,866 939,245 996,760 1,066,099


Liabilities (€m)






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%


LLC % gross loans -0.60% -1.41% -1.54% -1.14% -0.68%



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%


RWA % total assets 49.02% 47.92% 48.41% 48.89% 49.48%



19 J AN U AR Y 20 10 88

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 J AN U AR Y 20 10 89

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 J AN U AR Y 20 10 90

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 J AN U AR Y 20 10 91

This page has been left blank intentionally


19 J AN U AR Y 20 10 92



19 J AN U AR Y 20 10 93



19 J AN U AR Y 20 10 94

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 J AN U AR Y 20 10 95

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.

Matrix Corporate Capital LLP

One Vine Street

London, W1J 0AH

+44 (0) 20 3206 7000

Matrix Assessment of Pan-European Banks Capital Positions relating to Basel III : Jan. '11

Documents

Transcript of Matrix Assessment of Pan-European Banks Capital Positions relating to Basel III : Jan. '11