ECB: EU Banking Structures Report -- September 2010

8/8/2019 ECB: EU Banking Structures Report -- September 2010

1/51

E U B A N K I N G S T R U C T U R E S

S E P T E M B E R 2 0 1 0


2/51

EU BANKING STRUCTURESSEPTEMBE R 2010

In 2010 all ECB publications

feature a motif taken from the

500 banknote.


3/51

European Central Bank, 2010

AddressKaiserstrasse 2960311 Frankfurt am MainGermany

Postal addressPostfach 16 03 1960066 Frankfurt am MainGermany

Telephone+49 69 1344 0

Websitehttp://www.ecb.europa.eu

Fax+49 69 1344 6000

All rights reserved. Reproduction for educational and non-commercial purposesis permitted provided that the source isacknowledged.

Unless otherwise stated, this document uses data available as at 31 July 2010.

ISSN 1830-1878 (online)


4/513

ECBEU banking structures

September 2010

EXECUTIVE SUMMARY 5

1 OVERVIEW OF DEVELOPMENTSIN EU BANKING STRUCTURES 7

1.1 Regulatory initiatives 71.2 Bank intermediation 81.3 Consolidation and M&A activity 151.4 Concentration, competition

and capacity indicators 18

1.5 Cross-border intermediation 201.6 Conclusion and outlook 21

2 SPECIAL FEATURE ON THE FUTUREEVOLUTION OF THE EU BANKING SECTOR 23

2.1 Activities, business modelsand strategies 232.1.1 Diversi ed versus

specialised business models 232.1.2 In the medium term:

towards diversi ed, safer and more rational models

and risk practices 262.1.3 Adjustment of business

lines within banks 272.1.4 Other possible

developments 272.2 Funding and capital structures 28

2.2.1 Search for more andhigher-quality capital 28

2.2.2 Search for stable sourcesof funding 29

2.2.3 Banks liability structures:towards more stability

and higher costs 322.3 Conclusion 33

ANNEXES

1 Structural indicators for theEU banking system 34

2 Methodological note on thestructural indicators 48

Appendix (published separately)

Beyond RoE: How to measure bank performance

CONTENTS


5/514

ECBEU banking structuresSeptember 20104

OTHERSEU (EU27) European Union (27 countries after enlargements in 2004 and 2007)EU15 European Union (15 countries before enlargement on 1 May 2004)MU16 Monetary Union (16 countries participating in the euro area as at

31 December 2009) NMS New Member States (12 countries, marked above with *)RoW Rest of the world (non-EU27 countries)

COUNTRIESBE BelgiumBG Bulgaria *CZ Czech Republic *DK Denmark DE GermanyEE Estonia *IE IrelandGR GreeceES Spain

FR FranceIT ItalyCY Cyprus *LV Latvia *

LT Lithuania *LU LuxembourgHU Hungary *MT Malta *

NL NetherlandsAT AustriaPL Poland *PT PortugalRO Romania *SI Slovenia *

SK Slovakia *FI FinlandSE SwedenUK United Kingdom

ABBREVIATIONS


6/515


September 2010

EXECUTIVE SUMMARYThis report analyses the structural developmentsthat took place in the EU banking sector in the

period from 2008 to 2009, as well as pertinentregulatory changes. The analysis is based ona wide range of indicators and has bene tedfrom the exchange and assessment of qualitativeinformation within the Banking SupervisionCommittee (BSC) of the European Systemof Central Banks (ESCB). The BSC comprisesrepresentatives of the central banks and bankingsupervisory authorities of the EU Member Statesand of the European Central Bank (ECB).

The overview chapter starts with a section onregulatory developments focusing on the commonglobal approach taken in response to the nancialcrisis. The planned reforms aim to strengthenthe institutional framework for nancial stabilityand enhance the macro-prudential approach tosupervision, while improving the prudentialregulation of banks. In the EU, the reform of the nancial supervisory architecture will havea signi cant impact on banking supervision andcoordination between the national authorities.The amendments that will be made to the CapitalRequirements Directive (CRD) will also restrictrisk exposures and improve the supervision of cross-border banks.

With regard to the structural trends of 2008-09,longer-term growth in bank intermediationwas disrupted as the nancial crisis intensi ed.Central banks and governments reacted promptly

to support the functioning of their

nancialsystems and these actions helped to revivecon dence in the markets in the course of 2009.The recovery of lending has, however, beenuneven across banks. A number of uncertaintiesstill prevail in the markets, and have evenheightened in the course of 2010 with regard tosovereign credit risk. Although some institutionshave started to reimburse or have already fullyreimbursed the capital support granted by the

public authorities, others are continuing toreceive public support. The strategies guiding

the gradual exit and the related restructuring plans will continue to have an impact on their activities in the future.

Consolidation in the banking sector and amore ef cient use of resources, as measured

by selected capacity indicators, has continued.Market concentration has remained at the levelattained in previous years, with small countriestypically having more concentrated marketsthan large ones. Domestic banks continue todominate the markets in EU Member Statesand have marginally increased their share atthe expense of foreign branches. Signi cantdifferences between countries continue to exist,with subsidiaries with a euro area parent being

prevalent in the New Member States (NMS).

The nancial crisis has had a signi cant effect oncross-border activities, including intermediationand merger and acquisition (M&A) activity.The decline nevertheless came to a halt in 2009,and there is reason to believe that these activitieswill pick up again quickly once economicgrowth resumes.

The second chapter of the report discussesthe future evolution of the EU banking sector in the aftermath of the recent nancial crisis.This special feature is based on interviews heldin workshops with relevant interlocutors inEU banks and other nancial rms, banking and nancial market associations, rating agenciesand consulting companies, and with expertsfrom academia.

The rst section of Chapter 2 examines the

future of banking activities, business models andstrategies. The diversi ed banking model hasshown to have acted as a shock absorber in timesof stress, and market participants expect it toincrease in importance at the cost of specialised

banking models. At the same time, owing to thedrop in pro tability, banks are likely to searchfor additional pro ts and economies of scale inselected areas by focusing on their core markets,activities and clients.

The second section takes a closer look at the

future of bank capital and funding structures.The regulatory reform will inevitably result infunding structures moving from volatile and


7/516


short-term sources towards more stable andlong-term sources, such as capital and deposits.However, the crisis has also increased investorsawareness of banks capital endowments.It is thus likely that market participants willend up requesting additional buffers on topof the minimum regulatory requirements.The limited funding resources together withthe increased demand are likely to result inincreased competition and funding costs in themedium term.

The special feature concludes that theconsequences of the expected developmentsfor nancial stability are expected to be

bene cial, although a number of uncertainties,in particular as to regulatory developments, mayhave an impact on the process. For example,the implementation of the new regulatoryframework is expected to increase the resilienceof the funding structures of European banks,

but rising funding costs may also encouragerisk-taking behaviour by banks seeking torestore pro tability. Finally, diversi ed banksappear to be individually more resilient, yet awide spectrum of business models may makethe system as a whole more stable.


8/517


September 2010 7

1 OVERVIEW OFDEVELOPMENTSIN EU BANKING

STRUCTURES

7

This chapter analyses major structural andregulatory developments in the EU bankingsector in the period from 2008 to 2009. 1 The bankruptcy of Lehman Brothers inautumn 2008 triggered a general loss of con dence in nancial markets and institutionsand was followed by concerted action bygovernments and central banks at theinternational level to support the nancialsystem. The tension began to decrease graduallyin 2009. Banks took recourse to governmentmeasures and, in some Member States, exit fromthem has already started. The recovery has,however, been uneven across banks and Member States. Uncertainties still prevail in the marketsand concerns about sovereign credit risk exacerbated this in the rst half of 2010.

Adapting the banking system structure inline with the new business and regulatoryenvironment is a lengthy process. 2 In terms of

banking structures, the nancial crisis seems tohave had a clear impact on the aggregate levelof assets and on cross-border activities, wherelong-term growth trends have been interrupted.As early as 2009, however, the rst indicationsof a reversal towards the trend seen prior to thecrisis were emerging. The trend towards higher consolidation and operational ef ciency continuedto prevail, while market concentration remainedroughly at levels reached in previous years.

The chapter starts by brie y describing the

regulatory developments both at the EU andinternational level. It then analyses the lateststructural developments in the banking sector interms of intermediation. The next two sectionsmove from describing the size of activitiesto the notion of sectoral ef ciency, which isin uenced by consolidation and M&A activityand expresses itself in the resulting degrees of concentration, competition and operationalef ciency. A section on cross-border activitiesfocuses on nancial integration in the EU.The nal section concludes and provides a brief

outlook for the future.

It should be noted that the EU banking structuresreport 2010 was nalised before the agreement

on the Basel III framework. Consequently, theseregulatory developments are not discussed inthe report.

1.1 REGULATORY INITIATIVES

To address the shortcomings revealed by thecrisis, the G20 Heads of State committed toenhancing regulation and supervision in aglobally consistent way. The need to strengthenthe institutional framework led to the expansionof the memberships of the Financial StabilityBoard (FSB) and the Basel Committee onBanking Supervision (BCBS). In addition, theG20 leaders committed themselves to enhancingthe macro-prudential approach to supervisionthrough the extension of the scope of regulationand oversight to all systemically important nancial institutions, markets and instruments.In the EU, a profound change in the architectureof nancial supervision will take place throughthe establishment of: (i) a European SystemicRisk Board (ESRB) to oversee the stability of the nancial system as a whole; and (ii) a EuropeanSystem of Financial Supervisors (ESFS) toincrease supervisory convergence and cooperationin the supervision of individual institutions. 3

As regards banking regulation, the G20 leadershave agreed to enhance the quantity and qualityof capital, and to mitigate the pro-cyclicality

While the chapter focuses on the period from 2008 to 2009,1in certain cases (i.e. regulatory developments and M&As) theanalysis covers the period up to mid-2010.The statistical data used for the structural analysis of the banking2sector mainly draws from on-balance sheet information (assetsand liabilities) and not income statements. The effect of exchangerate movements during the period of observation on nominaldata equally needs to be disentangled from changes in structuraltrends. In particular, the pound sterling, the Polish zloty and theSwedish krona decreased signi cantly between the data pointsin 2007 and 2008, which decreases the nominal euro value of the balance sheet data items for the UK, PL and SE for 2008.This movement was partially offset in the data for 2009.See the European Commissions proposal for a Regulation on3Community macro prudential oversight of the nancial systemand establishing a European Systemic Risk Board (COM(2009)

499 nal) and the proposals for a Regulation on establishing aEuropean Banking Authority (COM(2009) 501 nal), a EuropeanInsurance and Occupational Pensions Authority (COM(2009)502 nal) and a European Securities and Markets Authority(COM(2009) 503 nal).

1 OVERVIEW OF DEVELOPMENTS IN EUBANKING STRUCTURES


9/518


of capital requirements. Furthermore, theycommitted to discouraging excessive leverageso as to contain unsustainable balance sheetgrowth, and to developing a global framework for liquidity risk requirements. They alsodecided to re ne the incentives for themanagement of securitisation risks and toreform compensation practices to curb excessiverisk-taking. Finally, they agreed thatcross-border resolution of and the moral hazardrisks posed by systemically important nancialinstitutions should be addressed. 4 In the processof putting these commitments into practice,in July 2009 the BCBS published a set of enhancements to the Basel II capital framework,enhancing the coverage of risks such as marketrisk, incremental default risk in the trading

book and risks relating to certain types of securitisation. Supervisory review processes anddisclosure requirements were also addressed. 5 Furthermore, in December 2009 the BCBS

published consultation documents on the qualityof capital, the leverage ratio, and frameworksfor liquidity risk and counter-cyclical capital

buffers. In July 2010 the Governors and Headsof Supervision reached broad agreement on theoverall design of the capital and liquidity reform

package. In particular, this includes the de nitionof capital, the treatment of counterparty creditrisk, the leverage ratio, and the global liquiditystandard. 6 The general framework will be further detailed and nalised by the end of the year.

In the EU, the European Commission isincorporating the BCBS enhancements intoCommunity Law by amending the CapitalRequirements Directive (CRD). The amendmentin May 2009 already addressed large exposures,hybrid capital, management of risks relatedto liquidity and to securitisation exposures,and supervisory cooperation, among other things. The European Commissions proposalof July 2009 for further amendments to theCRD includes higher capital requirementsfor the trading book and re-securitisations,

remuneration policies and the disclosure of securitisation. In addition, the Commissionconducted a public consultation in the rstquarter of 2010 on possible amendments related

to liquidity standards, the de nition of capital,the leverage ratio, counterparty credit risk,countercyclical measures, systemically important nancial institutions, and a single rule book in

banking. Based on the feedback received fromstakeholders and depending on the discussionsat the international level, a legislative proposalis expected in the second half of 2010.

Finally, 2008 and 2009 also saw importantdevelopments in the areas of crisis managementand resolution in the EU. With regard to theformer, a Memorandum of Understandingon cross-border nancial stability among EUsupervisory authorities, nance ministries andcentral banks entered into force in June 2008. 7 As to the latter, the Directive on the depositguarantee schemes was amended in March 2009,following a commitment made by the EU FinanceMinisters in October 2008. Among other things,the amendment increased the minimum level of coverage to 50,000 as a rst step, to be further increased to 100,000 by the end of 2010. 8 The nancial crisis also gave new impetus to thework on cross-border crisis management andresolution. In this regard, in autumn 2009 theEuropean Commission consulted the market ona new EU framework for crisis management inthe banking sector that would comprise tools for early intervention, bank resolution measures andinsolvency procedures in a cross-border context,and published its proposal on an EU network of

bank resolution funds in May 2010.

1.2 BANK INTERMEDIATION

The growth trend of the total assets of creditinstitutions, insurance corporations, investmentand pension funds was interrupted in the second

See, in particular, the leaders statements at the Pittsburgh and4Toronto Summits on 24-25 September 2009 and 26-27 June 2010respectively.Additional adjustments to Basel II were presented in June 2010.5See http://www.bis.org/press/p100726.htm.6

See http://www.ecb.europa.eu/pub/pdf/other/mou- nancialstability72008en.pdf.This further increase was con rmed in the European8Commissions proposal of July 2010 for a thorough revision of the Directive on deposit guarantee schemes.


10/519


September 2010 9


STRUCTURES

9

half of 2008 (see Chart 1). 9 Bank assets stillcontinued to grow in 2008, but the trend haltedin many countries in the course of 2009, withthe exception of assets in the NMS whichcontinued to grow throughout the period.In contrast, the assets of insurance, pension and

especially investment funds began to decreasein 2008 as a consequence of declining asset prices and de-leveraging induced by the nancialcrisis, falling to levels seen in 2006. Followingthe recovery in bond and stock markets in thesecond quarter of 2009, the assets of theseinstitutions began to increase again in almost allcountries. 10 Although the share of banks withinthe EU aggregate declined in 2009, they clearlyremained the dominant suppliers of nancingamong this category of institutions, their assetsrepresenting 75% of the combined assets of

banks, insurance corporations and pension andinvestment funds in the EU (see Chart 2). 11

Despite the nancial crisis and the stagnatingassets in nominal terms, bank intermediation inrelation to GDP continued to increase on averagein the EU, mainly re ecting the declineexperienced in GDP (see Chart 3). Asset growthremained strong in relative terms in the NMS, 12

but also in Denmark, Ireland, Portugal, Finland,Sweden and the United Kingdom. Many banks

These assets cover the assets of credit institutions, insurance9corporations, investment and pension funds established in theEU, as reported in Tables 2, 8 and 9 in the Annex.For a discussion on developments in the bond and stock markets10in 2009, see, for example, Financial Stability Review , ECB,December 2009, p. 69 onwards.Another way to consider the importance of banks as suppliers of 11 nance is to compare the supply of credit with the size of capitalmarkets. In this regard, bank credit comprises half of the sum of credit, stock market capitalisation and outstanding debt securitiesin the EU. The corresponding gure for the United States isaround one quarter (see, for example, Statistics Pocket Book ,ECB, July 2010).An exception was Slovakia, where the ratio of assets to GDP12declined by around a fth between 2007 and 2009.

Chart 1 Total assets of credit institutions,insurance corporations, pension funds andinvestment funds(EUR billions)

0

10,000

20,000

30,000

40,000

50,000

60,000

0

10,000

20,000

30,000

40,000

50,000

60,000

credit institutionsinsuranceinvestment funds

pension funds

1 2 3 1 2 3 1 2 3 1 2 3 1 2 32005 2006 2007 2008 2009

1 EU27

2 MU163 NMS

Source: ECB. Note: Investment funds comprise bond, equity, mixed, real estateand other funds, but do not include money market funds.

Chart 2 Distribution of assets for creditinstitutions, insurance corporations, pensionfunds and investment funds(percentages)

0

20

40

60

80

100

0

20

40

60

80

100

credit institutionsinsuranceinvestment funds

pension funds

1 EU27

2 MU163 NMS

1 2 3 1 2 3 1 2 3 1 2 3 1 2 32005 2006 2007 2008 2009

Source: ECB. Note: Investment funds comprise bond, equity, mixed, real estateand other funds, but do not include money market funds.


11/5110


in the NMS in particular had limited exposuresto foreign structured nancial products, whichoriginally left them relatively isolated from the

problems associated with these products.The ongoing catch-up process followingaccession to the EU, visible in signi cantlylower asset-to-GDP ratios than the EU average,also contributed to the strong and persistentgrowth in bank assets in many of the NMS. 13

The stock of total loans to non-credit institutionsin the EU stagnated in 2008 and decreasedslightly in 2009 (see Chart 4). 14 Loans to

non-

nancial corporations, however, continuedto grow in 2008 in the EU, while lending tohouseholds decreased. In 2009 the roles reversedwith lending to households resuming andgrowing by almost 5%, while corporate lendingdecreased slightly. The contribution of the

NMS to total loan growth was again signi cant(see Chart 5). Loan growth in the NMS wasstrongest in the household sector, standing at22% in 2008 and 8% in 2009. Growth in loans tonon- nancial corporations also began to declinein the NMS in 2009.

According to the ECBs Bank Lending Survey,the reduction in corporate lending in MU16 ismainly attributable to a low demand for loans.

In particular, the negative in uence of weak xed investment, scarce M&A activity and theavailability of alternative sources of nancing 15 on the demand for corporate loans has been

The EU27 and NMS averages for 2009 amounted to around13360% and 120% respectively. Among the EU15, Greece, Finlandand Italy have banking sectors that are less than three times their respective GDP.By contrast, loan-to-GDP ratios resumed growth in 2009,14re ecting the decline in GDP.Internal nancing, loans from non-banks, issuance of debt15securities and issuance of equity.

Chart 3 Total assets of credit institutions

(as a percentage of GDP)

0

100

200

300

400

0

100

200

300

400

2005 2006 2007 2008 2009

EU27EU15MU16

NMS

Source: ECB.

Chart 4 Total loans to non-credit institutionsin the EU

(EUR billions)

0

5,000

10,000

15,000

20,000

0

5,000

10,000

15,000

20,000

2005 2006 2007 2008 2009

total loans to non-credit institutionsloans to non-financial corporationsloans to households

Source: ECB.

Chart 5 Total loans to non-credit institutionsin the NMS

(EUR billions)

0

100

200

300

400

500

600

0

100

200

300

400

500

600

2005 2006 2007 2008 2009

total loans to non-credit institutionsloans to non-financial corporationsloans to households

Source: ECB.


12/5111


September 201011


STRUCTURES

11

stronger than the increasingly positive impactof debt restructuring needs. In addition todemand-related factors, credit standards for non- nancial corporations tightened during2008 and 2009. However, the pace of tighteninghas diminished since the rst quarter of 2009,nearing a reversal to net easing by the end of the year. 16

In the category of lending to households, loansfor housing purchase were severely affectedin 2008, with Belgium and Ireland recordingthe most signi cant declines in 2008. 17 This decrease can be attributed to a tightening of the credit standards on the supply side, but alsoto deteriorated housing market conditions andconsumer con dence. Altogether, lower demandrather than supply seems to have been the mainfactor behind the decline in loans to householdsand speci cally for housing purchase. Housingmarket prospects turned positive in 2009 and thetightening of credit standards started to declinein the second half of 2009, largely owing to

banks improved access to nance outweighingthe increased cost of capital. Loans for housing

purchase seem to have revived during 2009,although the recovery has been uneven acrosscountries. Loans still continued to decline inBelgium, Ireland and in the Baltic countries.By contrast, a signi cant increase took place in

the United Kingdom, despite the total amountstill being below the level of 2007.

On the liabilities side, total deposits of bankscontinued to increase in 2008 and 2009,although at a lower pace (see Chart 6). 18 In theMU16, only Luxembourg and Ireland reporteda decline in 2008. In 2009 there was greater heterogeneity, with deposits also declining inBelgium, Ireland, Greece and the Netherlands.Of these, Greece and the Netherlands hadrecorded strong deposit growth alongsideSpain, Portugal and Finland in 2008.

The increase in deposits was partly aconsequence of an active effort by banks toincrease the share of more stable fundingand reduce their dependence on wholesalemarkets (see Section 2.2 of the special feature).The elevated interest rates in relation to other investment opportunities have indeed attracteddepositors in some Member States. 19 In somecountries, banks also pro ted to a large extentfrom the increased risk awareness of householdsthat shifted funds from non-banking institutionsto banks in the perceived absence of alternative,secure investment opportunities. On the other

The Bank Lending Survey reports the responses of senior loan16of cers in a sample of euro area banks. A net tightening of credit standards or an increase in loan demand means that the

proportion of all respondents that indicated a tightening or anincrease would take place is higher than the proportion of thosewho indicated an easing or a decrease. A diminishing pacerefers to a lower proportion of respondents indicating increasingdynamics. See Euro area bank lending survey , ECB, publishedquarterly and available at http://www.ecb.europa.eu/stats/money/surveys/lend/html/index.en.html.Looking at the nominal values, housing loans in the United17Kingdom seem to have declined the most in 2008. It should,however, be noted that this gure is affected by exchange rateeffects which were signi cant in 2008. For that year, the declinefor the United Kingdom stood at 29% in nominal terms, and 7%in real terms. Conversely, the growth in loans in some of the

NMS in particular may be attributable to the combination of adepreciation and foreign currency nominated loans. This is thecase for Poland in 2008, for example.Some of the declines in the data on individual countries in the18Annex were attributable to exchange rate effects. For example,the United Kingdom and Sweden recorded declines in totaldeposits of credit institutions from non-credit institutions of 12%

and 3% in euro terms in 2008 respectively, but in local currencythey increased by 14% in the United Kingdom and by 12% inSweden.See also Section 1.4 on concentration, competition and capacity19indicators.

Chart 6 Total deposits to credit institutions

from non-credit institutions

(EUR billions)

02,000

4,0006,000

8,00010,000

12,00014,000

16,000

18,000

0

100

200

300

400

500

600

2005 2006 2007 2008 2009

EU27 (left-hand scale)MU16 (left-hand scale) NMS (right-hand scale)

Source: ECB.


13/5112


hand, signs of returning risk appetite are alreadyvisible in many Member States and havecontributed to the easing of non-deposit fundingconditions. Finally, as customer con dence hasalready started to increase in some countries andmay continue to rise in the near future, depositgrowth can be expected to level off again inthe future. 20

Loan-to-deposit ratios that were increasingsteeply until 2008 declined in 2009, owing todecreasing loans, increasing deposits andconscious efforts to limit liquidity risk, a trend

which was partly reinforced by slower balancesheet growth (see Chart 7). 21 Loan-to-depositratios differ to a large extent between Member States, with Denmark and Sweden displayingthe highest ratios (see Chart 8). 22 Loan-to-deposit ratios decreased in almost all of the NMS in the course of 2009, whichcontributed to reducing the dependence onforeign funding substantially, together with theinitiatives of authorities to limit loans in foreigncurrencies. However, signi cant differences asto access to funding may persist for some time,not only across Member States, but also across

banks, particularly as some might need tocontinue to rely on public support measures and

problems will be revealed when these measuresare no longer available. For some of these banks,fundamental restructuring and possibly also a

Shifts from deposits to other investment products owing to20

increased customer con dence have been already observed,for instance in Finland in the last quarter of 2009. See themanager survey by the Federation of Finnish Financial Services,dated 5 October 2009 (Pankkibarometri 3/2009, in Finnish).A possible additional effect arose from banks paying more21attention to internal pricing policies. Misaligned internal pricingof funding costs during the crisis seem to have contributed to theincentives of business units to leverage and maximise volumes.See EU banks funding structures and policies , ECB, May 2009.

Note that the high loan-to-deposit ratio in Denmark is attributable22to the inclusion of mortgage-credit loans which are nanced byissuance of mortgage-credit bonds. In Denmark, mortgage-creditinstitutions do not receive deposits.

Chart 7 Historical development of loan-to-depositratios

(percentages)

80

90

100

110120

130

140

150

80

90

100

110120

130

140

150

2005 2006 2007 2008 2009

MU16 averageEU27 average NMS average

Source: ECB.

Chart 8 Loan-to-deposit ratios in EU Member States, 2009

(percentages)

0

50

100

150

200

250

300

350

0

50

100

150

200

250

300

350

DK SE LV LT IE SL IT FI SK FR AT PT HU MT RO NL BG ES DE PL CY UK GR CZ LU BEEE

Source: ECB.


14/5113


September 201013


STRUCTURES

13

downsizing of balance sheets may be necessary before long-term viability is restored again. 23

In conjunction with the G20 objectives, creditinstitutions are expected to strengthen their capital bases in terms of quality and level.This development will enhance the sustainabilityof bank activities, but will also exert downward

pressure on performance measures such as

return on equity (RoE). Box 1 discusses the problems related to RoE as a measure of sustainable bank performance and alternativeapproaches to it.

For more information on the use of government support23measures, see Box 12 entitled Government measures to support

banking systems in the euro area in Financial Stability Review ,ECB, June 2010.

Box 1

PERFORMANCE MEASURES FOR BANKS

Recent events have proven that the most common measure of a banks performance, i.e. return onequity (RoE), is only one part of the story since a good level of RoE may re ect either high pro tabilityor more limited equity capital. In addition, the traditional decomposition of RoE (i.e. looking at

banks operational performance, risk pro le and leverage) may have been useful to assess banks performance during benign times. This approach has clearly not proven adequate in an environmentwith much higher volatility, such as during the global nancial crisis, where operational performanceis at the root of all of the uctuation in RoE and it does not help to provide an understanding of the

potential trade-off between risk and return in performance. This may actually explain why some rmswith high RoE performed particularly poorly during the crisis, held back by rapid leverage adjustment.

The BSC has recently examined: (i) why the commonly used RoE measure may not be suf cientto characterise banks performance; (ii) what may be missing in this type of approach; and(iii) potentially complementary approaches to RoE. For its analysis, the BSC used the capacity togenerate sustainable pro tability as a de nition for describing banks performance. A bank must

be able to generate such pro ts in order for it to continue operating and for investors to obtainfair returns, but it is also important for supervisors as it guarantees more resilient solvency ratios,even in a context of a riskier business environment. Indeed, retained earnings appear to belong to

the most important drivers of Tier 1 ratios.

The main drivers of banks pro tability remain earnings, ef ciency, risk-taking and leverage.In this respect, RoE has a number of limitations. First, RoE is not risk-sensitive. A decompositionof RoE shows that a risk component represented by leverage can boost RoE substantially.By contrast, other risk elements are missing in RoE (such as the proportion of risky assets and thelevel of solvency). RoE is thus not a stand-alone performance measure and decomposition or further information is necessary in order to identify the origin of developments and possible distortionsacross time. RoE has proven to be a point-in-time measurement without signalling power or aforward-looking perspective. Indeed, the crisis demonstrated that RoE fails to distinguish the best

performing banks from others in terms of the sustainability of their results. RoE is a short-termindicator and must be interpreted as a snapshot of the current health of institutions.

Finally, RoE measures can be misleading or manipulated and can provide wrong incentivessince data are not always reliable, given that they are in uenced by quite strong seasonal factors.RoE measures can also expose banks to higher unexpected risk levels.


15/5114


RoE must thus be re ned in order to provide a more informed assessment of banks performance. In particular, measures of banks performance should ideally be forward-lookingand comparable, and should also measure stability across time.

Alternative approaches to measuring banks performance may require looking in more depthat how banks run their business, making use of their stress test results, and further enhancinghigh-level discussions with supervisors on the consistency between performance and businessstrategy. This may require greater transparency from banks with regard to their pro tabilitystructure, and some adjustment in the governance process as suggested in the proposals for

enhancing Basel II. Among others, these measures comprise a reassessment of the independenceof the risk management function with respect to the bank in question, the available tools and anadequate level of risk awareness at the top-tier management level. As a result, they may presentan opportunity for regulators to address these issues with bank managers.

As a conclusion, the main messages stemming from this analytical work may be summed up asfollows.

RoE may be less of a performance benchmark than a communication tool in the relationship1. between banks and markets.

A comprehensive performance analysis framework needs to go beyond this kind of 2.

indicator though not excluding it - and provide for a more informed assessment using banks business-based data and qualitative information.

The consistency of risk appetite with the business structure and strategy appears to be one of the3.most crucial elements in assessing an institutions capacity to perform well in the future. Againstthis backdrop, sustainable indicators constructed on the basis of economic capital models and nancial planning frameworks inside the banks may become even more relevant. For instance,risk-adjusted types of indicator for returns, such as the risk-adjusted return on capital indicator,may bene t from higher disclosure and explanation to the markets, or at least to the supervisors.

Measures of banks performance should ideally encompass more aspects of the performance than4. just pro tability as is the case for a purely market-oriented indicator such as RoE. In particular,

it may be useful to take into account the quality of assets, funding capacity and the risk associatedwith the production of value. In that context, a good performance measurement framework shouldincorporate more forward-looking indicators and be less prone to manipulation from the markets.

In the context of achieving a comprehensive analysis for all business areas, data availability5.and comparability are key factors. This may call for enhanced disclosure (both to thesupervisors and, where possible, to the public) and improved market discipline.

As regards governance, the adoption of a more comprehensive and more forward-looking6.assessment of performance may represent a rst step towards intensifying the dialoguewith banks top-tier management, related to the coherence between economic performance,the respective business model and supervisory and nancial stability issues.

The full analysis is presented in the Appendix: Beyond RoE How to measure bank performance,September 2010, available at http://www.ecb.europa.eu.


16/5115


September 201015


STRUCTURES

15

1.3 CONSOLIDATION AND M&A ACTIVITY

Consolidation of the EU banking sector continued in 2008 and 2009. Excluding theeffect of a reclassi cation in Ireland in 2009, thenumber of credit institutions has declined at asteady pace (see Chart 9). 24 The decline was

particularly marked in Cyprus, as a consequenceof the consolidation of its credit cooperativessector, but pronounced declines also took placein Denmark, Germany, France, the Netherlandsand Sweden. Notable exceptions to this trendwere the Baltic countries, which saw an increasein both domestic and foreign banks, indicatingthat there was still room for new service

providers in these markets. 25

The number of M&As in the EU dropped bya quarter in 2008, bringing the total number tothe lowest point throughout the period under observation (see Chart 10). In terms of the totalvalue of transactions, and leaving aside thespecial effect of the acquisition of ABN Amro

by the consortium of Royal Bank of Scotland,

Fortis and Santander in 2008, the M&Adata revealed a signi cant decline, with EUcross-border and outward transactions beingmost affected (see Chart 11). 26 M&A activitystarted to pick up in 2009, with the clearestincrease taking place in the sub-category of domestic deals. 27 The values of the deals haveremained modest, however, indicating a clear

This reclassi cation of 419 credit unions as credit institutions in24Ireland resulted in a slight increase in the overall number of creditinstitutions in the EU in 2009. More details on the reclassi cationcan be found on the ECBs website at http://www.ecb.europa.eu/

press/pr/date/2009/html/pr090113.en.html.In Estonia, this included the establishment of the domestic25AS LHV Bank and a branch of Bank Snoras, a rst foreign branchof the Lithuanian bank. New banking licenses were granted tothe Latvia Post Bank and a branch of Balti Investeerigute GrupiPank in Latvia, and, in Lithuania, to the domestic bank Finasta aswell as to SEB and to Handelsbanken branches.The ABN Amro deal represented 90% of the total value of 26cross-border transactions in 2008. Other large cross-border transactions in 2008 include the acquisition of Citibank Privatkunden by Banque Fdrative du Crdit Mutuel, as wellas the acquisition of Alliance and Leicester and Bradford andBingley by Banco Santander Central Hispano.Domestic deals denote deals that take place within national27

borders. In this report, EU-wide deals are referred to ascross-border M&A (see also the notes for Charts 10 and 11).

Chart 9 Number of credit institutions

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

2005 2006 2007 2008 2009

EU27MU16 NMS

Source: ECB.

Chart 10 Bank M&As number of transactions

0

20

40

60

80

100

120

140

160

0

20

40

60

80

100

120

140

160

2000 2001 2003 2005 2007 20092002 2004 2006 2008 2010H1

domesticcross-border outwardinward

Source: Zephyr, Bureau Van Dijk. Notes: M&As include both controlling and minority stakes.Cross-border M&As refers to intra-EU27 transactions involving

a non-domestic acquirer. Inward refers to M&As by non-EU27 banks in the EU27 and outward indicates M&As by EU27 banksoutside the EU27.


17/5116


tendency towards smaller deals. 28 Important dealsin 2009 and early 2010 include the acquisitionsof Dresdner Bank by Commerzbank and HBOS

by Lloyds TSB as domestic deals, but alsoFortis by BNP Paribas as an example of a cross-

border deal 29 and Mellon United National Bank by Banco Sabadell as an example of an outwarddeal. Most of these deals were accelerated or

induced by the

nancial crisis.

Two factors have contributed to the decline of the non-domestic component of M&A in

particular. First, strategic expansion was notconsidered a high priority during the crisis by

banks that were facing signi cant losses andwrite-downs and focused on repairing their

balance sheets. Efforts to boost their capital positions and cash reserves led to the spin-off of certain non-core activities (see also Section 2.1.3),either in terms of business or geographical

location.30

The uncertainties related to economic prospects and forthcoming regulatory changesfurther have continued to act as deterrents of non-domestic M&A activity. Second, although

the data assessed in this section do not includethe participation of governments in creditinstitutions, a signi cant shift in the ownershipstructure in this direction has occurred in somemajor EU banks. The European Commissionsconditions for state aid have furthermore led todivestment of activities and markets, partly inorder to avoid distortions in competition. 31

There is, however, reason to believe that theobserved decline in cross-border and outwardM&A is only temporary. First, the number of cross-border deals has already picked up sinceearly 2009. In this regard, the acquisitions of Fortis by BNP Paribas and of UK banks bySantander are examples of strategic cross-border acquisitions by institutions that were in a positionto pro t from the opportunities that arose duringthe crisis. Second, the limited duration of government recapitalisation measures may offer M&A opportunities in Europe in the near future.Indeed, exit from recapitalisation measureshas already begun. 32 Third, an ESCB surveyconducted in May 2009 revealed that, rather

It should also be noted that the value of the deals is not reported28in the data for all M&As. However, the average M&A values for the available data have also decreased since 2008.BNP Paribas acquired 75% of Fortis Bank from the Belgian State.29In this context, reference can be made to the acquisition of the30German retail banking activity of Citigroup by Crdit Mutuland to the acquisition of HSBC Merchant Services by GlobalPayments Inc.The former has been the case, for example, for ING which is31divesting its insurance operations, for Dexia which will reduceits balance sheet by 35% by 2014, largely via divestment of itsSlovakian operations, and for KBC, which will run down itsnon-core activities in particular in Central and Eastern Europe.Examples of the latter include the newly created Lloyds BankingGroup as a result of the acquisition of HBOS by Lloyds TSB,which is to divest its core UK retail banking business. Also INGand KBC will divest some core units in their home marketsto spur competition. See also the European CommissionsCommunication on the return to viability and the assessment of restructuring measures in the nancial sector in the current crisisunder the State Aid rules (OJ L 195 of 19.8.2009, pp. 9-20).It should also be noted that most capital injections were made32through the acquisition of preference shares without votingrights, with the aim of improving the capital position of banksand ensuring the priority of public sector claims, but leavingtheir ownership in the private sector. See also Measures taken

by euro area governments in support of the nancial sector,Monthly Bulletin , ECB, April 2010, pp. 75-90. For a discussion of

the development of M&As in the nancial sector, see also Back to the domestic future , PriceWaterhouseCoopers, March 2009.For cross-border M&As see Special Feature A, Bankingintegration and supervision in the EU, in the ECB report on

Financial Integration in Europe , April 2010, pp. 31-44.

Chart 11 Bank M&As value of transactions

(EUR billions)

0

20

40

60

80

100

120

140

0

20

40

60

80

100

120

140

2000 2001 2003 2005 2007 20092002 2004 2006 2008 2010H1

domesticcross-border outwardinward

Source: Zephyr, Bureau Van Dijk. Notes: M&As include both controlling and minority stakes. For some of the deals, the value is not reported. Cross-border M&Asrefers to intra-EU27 transactions involving a non-domesticacquirer. Inward refers to M&As by non-EU27 banks in theEU27 and outward indicates M&As by EU27 banks outside theEU27.


18/5117


September 201017


STRUCTURES

17

than revising their internationalisation strategies, banks have temporarily stopped or delayedtheir plans (see Box 2). High growth potentialstill remains by far the most important driver for expansion abroad. Given the opportunitiesarising from the exit from government measures

and based on the survey results, cross-border M&A activity is expected to pick up quicklyonce the economic cycle turns.

Section 1.5 contains more information on thecross-border activities of EU banks.

Box 2

THE EFFECT OF THE FINANCIAL CRISIS ON THE INTERNATIONALISATION STRATEGIESOF LARGE EUROPEAN BANKING GROUPS

In the rst half of 2009, the ESCB conducted a survey on the internationalisation plans of major cross-border banking groups in the EU. 1 The survey was sent to 43 major banking groups withsigni cant cross-border banking activities and head of ces in EU Member States. The resultswere compared with those of a similar exercise in 2006. As the survey was conducted usinga sample, its results were intended to be purely indicative of the trends in internationalisationand, in particular, to offer qualitative background information on the motives behind thedecisions taken.

The survey was conducted in the middle of the nancial crisis, and therefore the entire effect of the crisis could not be deduced from the answers at that point in time. Some indication could,however, be obtained on future developments in the eld of internationalisation plans. It wasremarkable that high growth potential remained the main driver for expansion abroad, with over 90% of the banks in the sample stating it as a key factor. The next two most common reasons,namely following clients and limited growth potential in the home country, were cited by aroundtwo-thirds of the respondents. When asked about the impact of the nancial crisis, more thanhalf of the respondents stated that they had amended their internationalisation strategies as aconsequence. The crisis induced banks to refocus on core business activities and to slow down,temporarily stop or delay cross-border activities. Rather than radical changes of strategy, thesereplies thus point towards a temporary prudent attitude taken by banks.

Finally, the crisis changed the respondents risk perceptions for country, exchange rate andreputational risks. Country risk was perceived to have increased in emerging and developingcountries and exchange rate volatility was expected to increase. The crisis also heightened theawareness of reputational risk and the need to manage it. However, the ranking of these risks interms of their importance did not change.

In conclusion, the effect of the crisis seems to be of a temporary nature, and banks do not seemto consider the crisis a reason to change the fundamentals of their internationalisation strategies.It is thus likely that cross-border M&A activity will pick up quickly once economic conditionsimprove, both within the EU and in emerging markets in particular.1 More precisely, the qualitative questions of the survey were answered in May 2009.


19/5118


1.4 CONCENTRATION, COMPETITIONAND CAPACITY INDICATORS

Market concentration, as measured with theHer ndahl index and with the share of totalassets held by the ve largest institutions,increased over the aggregate period 2008-09, inline with the increased consolidation in the EU

banking sector. 33 Both indices peaked in 2008and decreased slightly in 2009, althoughremaining well above the 2007 levels. Withregard to individual Member States, the pictureremains largely unchanged, with larger countries, such as Germany, Italy and the UnitedKingdom, but also Austria, having morefragmented markets, and smaller countries,especially some of the NMS, being characterised

by more concentrated banking sectors.

Banks increased need for stable funding, asdiscussed in Section 1.2, also intensi ed thecompetition on deposits in the EU during the nancial crisis. In addition, competition fromoutside the EU increased temporarily, owingto the entrance of a group of Icelandic banksin some Member States in 2007 and 2008, withinterest rate offers for deposits that exceeded themarket average sometimes by several percentage

points. 34 Fierce competition on deposits andhigh interest rates continued to prevail in someMember States in 2008 and 2009. 35 In somecountries, this led to an increased competitiveadvantage for the established brands, since they

are often able to offer higher rates and sincedepositors perceptions seem to have changedin the direction of valuing large, well-knownand highly diversi ed banks. Competition ondeposits was also high in some NMS such asBulgaria, Hungary, Lithuania and Slovenia,

partly owing to the ongoing catch-up processfollowing accession to the EU and the relatedmismatch of available domestic savings andinvestment opportunities in the country. Depositrates started to decline in many countries inthe course of 2009, possibly also signalling a

slight easing of competitive conditions in someof them, but also related to other factors suchas decreasing central bank rates. Competition

on the credit side has, by contrast, been low, partly owing to the dif culties in accessingwholesale funding, the increased cost of fundingand the impaired macroeconomic conditions.Signs of competition picking up in the creditmarkets have, however, recently emerged in theMU16. 36

Measured in terms of bank assets per employee,the EU banking system continued to becomemore ef cient over the period from 2007 to 2009(see Chart 12). 37 Selected capacity indicators for EU Member States are presented in Table 1.Overall, these indicators continue to vary acrossthe EU, on account of the different businessstrategies followed by banks, but also owing tofactors like population density. 38 The sharpincrease in the indicators related to the number

As a general rule, a Her ndahl index below 1,000 signals33low concentration, while an index above 1,800 signals highconcentration. For values between 1,000 and 1,800, an industry

is considered to be moderately concentrated. Note that theseindicators are calculated on a non-consolidated basis, meaningthat banking subsidiaries and foreign branches are consideredto be separate credit institutions. For more information, see themethodological note in Annex II.Following a major disturbance to the Icelandic nancial system34in Autumn 2008, these banks subsequently disappeared from theEU market in the course of the second half of 2008 and the rsthalf of 2009. See Financial Stability Report , Central Bank of Iceland, 2009, for more information.This is also evident in the negative deposit margins for euro area35monetary and nancial institutions. See Chart S98 in Financial Stability Review , ECB, June 2010.The ECBs Euro area bank lending survey of April 2010 reports36that competition has had a dampening effect on credit standardsin both the household and enterprise sectors, and in the former sector, this impact has gradually increased. Besides banks,

competition from non-banks and especially market nancingseems to have dampened the rise of credit standards for corporateloans for particularly large enterprises. However, this effectdisappeared in the second quarter of 2010. Section 2.1.4. of thespecial feature further discusses the potential effect of regulationon competition arising from non-banks.The term ef ciency in this context refers to the operational37ef ciency of the use of resources (e.g. personnel) with regard tosome measure of output that can be associated with the amountof services provided (e.g. assets). Note that the MU16 aggregatedecreased somewhat from 2008 to 2009; this effect may however also be related to the employee data that is not available for alarge number of countries.For example, Belgium, which is densely populated, has a below38average gure for population per ATM, whereas the densely

populated Netherlands has high gures for both population per

ATM and per branch. At the other end of the scale, scarcely populated Sweden, for example, also has high gures for both population per branch and per ATM, whereas e.g. Cyprus has below average gures for population per branch.


20/5119


September 201019


STRUCTURES

19

of employees during 2009 may partly be relatedto ad hoc measures taken during the crisis andshould thus be interpreted with caution. 39 By contrast, the number of automated teller machines (ATMs) continued to increase in thelarge majority of Member States, mainlyre ecting the substitution of employees withATMs. Branch networks in Poland, Hungaryand Portugal continued to grow. By contrast,the high growth in branch networks inRomania witnessed in previous years levelledoff in 2009.

Note that the increase in Bulgarian employee data in Table 2 in39the Annex is related to the fact that average numbers are used;a year-end gure would show a clear decline in the number of employees in Bulgaria.

Chart 12 Assets per employee

(in EUR millions)

0

2

4

6

8

10

12

14

16

0

2

4

6

8

10

12

14

16

2005 2006 2007 2008 2009

EU27MU16

NMS

Source: ECB. Note: Data not available for BE, DE or FR.

Table 1 EU banking sector capacity indicators in 2009

CountryPopulation

per CIPopulationper branch

Populationper ATM 1)

Populationper employee

Assets peremployee

Populationdensity

BE 103,750 n.a. 692 n.a. n.a. 326BG 252,133 1,253 1,489 221 1,107 68CZ 187,336 5,251 3,059 273 4,173 133DK 33,671 2,767 1,783 110 22,046 128DE 42,030 2,077 1,033 n.a. n.a. 229EE 74,467 6,293 1,442 235 3,748 30IE 8,962 3,634 1,307 117 34,669 64GR 170,612 2,761 1,446 171 7,463 85ES 130,482 1,034 739 173 12,840 91FR 90,581 1,676 1,203 n.a. n.a. 117IT 75,235 1,771 1,093 187 11,445 200CY 5,148 858 1,300 64 11,138 87LV 60,943 3,614 1,784 182 2,420 35

LT 39,755 3,436 2,284 306 2,401 51LU 3,385 n.a. 1,062 19 30,206 192HU 52,748 2,822 2,173 235 2,961 108MT 17,207 3,562 2,425 108 10,757 1,291

NL 56,023 5,268 1,901 150 20,155 405AT 10,586 2,007 1,090 108 13,419 100PL 53,675 2,867 2,809 208 1,498 118PT 64,051 1,654 629 171 8,360 116RO 510,984 3,340 2,325 316 1,272 90SI 81,661 2,892 1,169 168 4,382 101SK 208,375 4,405 2,403 289 2,905 111FI 15,298 3,471 1,655 215 15,581 16SE 51,894 4,351 3,294 190 19,045 21UK 158,784 4,997 960 131 19,998 252

MU16 50,959 1,766 1,033 165 13,616 127EU27 59,860 2,131 1,168 171 13,156 115

Sources: Calculations based on gures in the Annex, the ECB Blue Book and United Nations data. Notes: CI stands for credit institution. Assets per employee are measured in EUR thousands. Population density is expressed as inhabitants per square kilometre. MU16 and EU27 averages exclude Member States with incomplete data.1) 2008 data.


21/5120


1.5 CROSS-BORDER INTERMEDIATION

Domestic credit institutions increased their sharein the EU banking sector in 2008, but their sharedecreased again in 2009. By the end of 2009,73% of total assets were owned by domesticinstitutions, marginally up from 71% in 2007(see Chart 13). The overall decline in the shareof foreign banks was almost entirely attributableto a decline in the share of branches from 15%to 12%. Foreign subsidiaries increased their

presence slightly in 2009 to 15%. The decline inthe market share of foreign branches was almostentirely attributable to institutions domiciled inthe EU. In contrast, the United Kingdom andBelgium in particular substantially increasedtheir foreign share in 2009, mainly because of the acquisitions by Santander and the acquisitionof Fortis Bank by BNP Paribas.

Chart 14 shows that foreign banks continued to be more prevalent in the NMS, although their

Chart 13 Market share of foreign bankbranches and subsidiaries in terms of totalassets, EU average(percentages)

0

10

20

30

40

50

60

70

80

90

100

0

10

20

30

40

50

60

70

80

90

100

2005 2006 2007 2008 2009

branches EU branches RoWsubsidiaries EUsubsidiaries RoWdomestic CIs

Source: ECB.

Chart 14 Market share of foreign bank branches in EU Member States in 2009

(percentages)

0

10

20

30

40

50

60

70

80

90

100

0

10

20

30

40

50

60

70

80

90

100

branches EU branches RoWsubsidiaries EUsubsidiaries RoWdomestic CIs

NL SE ES DE FR IT AT DK PT SL GR CY IE UK HU BE FI PL LV RO MT LT BG CZ LU EE SK

Source: ECB.


22/5121


September 201021


STRUCTURES

21

aggregate share dropped from 72% to 69%in 2009. Unlike for the EU aggregate in whichsubsidiaries seemed to gain in importance,the decline was relatively even across branchesand subsidiaries in the NMS. The majorityof banking assets in the NMS were held byforeign subsidiaries (61%), with those with anEU parent accounting for 57% of the market.By comparison, only 26% of banking assetsin the EU15 were held by foreign entities,split fairly evenly between branches andsubsidiaries.

The marked decrease in cross-border assetsobserved in wholesale and securities-relatedactivities 40 in 2008 levelled off during 2009(see Chart 15). Integration in the retail bankingsegment, having generally lagged behind that inthe other banking segments, seems to have beenaffected by the nancial crisis to a lesser extent.These developments are also re ected incross-border liabilities (see Chart 16). In all,

banks have clearly relied on their domesticcounterparties rather than on their international

peers during the nancial crisis. However, these

developments are expected to be only temporaryand the increasing trend of nancial integrationin the banking markets is expected to resumesoon. Integration of the up to now fragmented

retail markets is also expected to get newimpetus from the advancing integration of therelated market infrastructures, and in particular from the wide-ranging application of theSingle Euro Payments Area (SEPA) in thecoming years. 41

1.6 CONCLUSION AND OUTLOOK

After the exceptional events of 2008, duringwhich the global banking system suffered froma general loss of con dence and erosion of value

as a result of the

nancial crisis, the year 2009marked the start of a gradual recovery in the EU nancial sector.

The main structural developments in the EUcontinued in line with trends observed inthe years leading up to the nancial crisis.Consolidation of the banking sector and amore ef cient use of resources, as measured bycapacity indicators, continued over the medium

The category shares and other equity includes holdings of 40

securities which represent property rights in corporations or quasi-corporations and mutual fund shares.See also the ECB report entitled Financial Integration in Europe41 ,

published in April 2010 and available at http://www.ecb.europa.eu/pub/pdf/other/ nancialintegrationineurope201004en.pdf.

Chart 15 Cross-border provision of financialservices in the euro area assets

(percentages)

0

10

20

3040

50

60

0

10

20

3040

50

60

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

cross-border non-bank securities other than sharescross-border interbank loanscross-border loans to non-bankscross-border non-bank shares and other equity

Source: ECB. Note: Cross-border activity is expressed as a percentage of thetotal euro area provision of nancial services of each nancialservice or instrument in question.

Chart 16 Cross-border provision of financialservices in the euro area liabilities

(percentages)

0

10

20

30

40

0

10

20

30

40

cross-border interbank depositscross-border non-bank deposits

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Source: ECB. Note: Cross-border activity is expressed as a percentage of thetotal euro area provision of nancial services of each nancialservice or instrument in question.


23/5122


term, and concentration remained roughly atthe level attained in previous years. Growth in

bank assets stagnated, however, and loans for housing purchase declined for the rst timeduring the period under examination owing toa number of factors affecting both the demandand the supply side. Nevertheless, rst signs of a return to growth were already visible at theend of 2009.

A decline followed by a gradual return to pre-crisis trends was also witnessed in thecross-border activities of EU banks. Foreign

branches lost market share to domesticinstitutions, and there was a pronounced declinein cross-border M&As as banks shifted their focus from pursuing growth opportunities torepairing their balance sheets.

Although the next chapter will take a closer look at the future of the EU banking system,in particular as regards the evolution of businessmodels and funding structures, some generalexpectations can already be stated. Owing tothe ongoing regulatory initiatives, EU banks areexpected to hold a higher level and quality of capital and more liquid assets, and to improvetheir credit and liquidity risk management.Although banks have improved the share of stable funding in the course of 2009, the cost of funding is expected to increase in the mediumterm, as banks will need to close the gap

between their current stable funding and that

required in the future. As the economy recoversand market functioning is restored, cross-border and outward M&A activity is likely toresume quickly, as those banks dealing moreeffectively with the challenges of the crisis willseek to take advantage of pro t and growthopportunities abroad. Given that the governmentrecapitalisation measures are transitional, theymay also offer opportunities for acceleratedM&A activity in the future; indeed, exit fromthese measures has already begun.

However, the return to normality may be prolonged by concerns regarding the sovereignsector and the potential impact of scalconsolidation. The funding requirements that

governments will face in the near future areexpected, in particular, to add additional pressureon funding costs for banks. In addition, therecovery may be highly uneven across individualinstitutions. The government measures and therelated restructuring plans may have a particular impact on asset growth for those banks that arestill receiving support.


24/51


25/5124


crisis in 2008, their credit default swap (CDS) premia were generally higher than those of diversi ed banks in Europe (see Chart 18),indicating that markets had greater fears aboutthe former.

The severity of the impact of the crisis hasvaried among institutions, depending on their

business models and exposure to certain risk-taking practices:

A selective extinction occurred, generallyhitting specialised players. These includespecialised lenders (consumer nance,automotive nance, real estate banks, etc.)and pure investment banks (mainly in theUnited States). The investment bankingindustry has been strongly affected, owingto its high reliance on wholesale andcapital markets for funding. This may beexplained by the focus of investment bankson originate-to-distribute activities andengagement in arbitrage or other complex nancial transactions.

Irrespective of the activities of the institutionand the risks taken, all banks, includingdiversi ed banks, may face severe dif cultiesas long as risk management is inadequate.

Chart 17 Net income: European investmentversus diversified banks

(in EUR billions)

0

5

10

15

20

-15

-10

-5

-15

-10

-5

0

5

10

15

20

31 December 200731 December 200831 December 2009

Investment banks

Diversified banks

1 UBS2 Crdit Suisse3 Deutsche Bank 4 Commerzbank

5 BNP Paribas6 Socit Gnrale7 CASA8 Santander

9 BBVA10 Lloyds11 HSBC12 Barclays

1 2 3 4 5 6 7 8 9 10 11 12

Source: Bloomberg. Notes: For this chart, investment banks comprise UBS, CrditSuisse and Deutsche Bank. Diversi ed banks comprise BNPP,Socit Gnrale, Crdit Agricole, Santander, BBVA, Lloyds,HSBC, Barclays and Commerzbank. See footnote 43 for thede nition of investment banks.

Chart 18 CDS premia: European investment versus diversified banks

(in basis points)

0

50

100

150

200

250

300

350

400

0

50

100

150

200

250

300

350

400

minimum investment banksmaximum investment banksaverage

Sep. Jan. May Sep. Jan. May

2008 2009 2010

0

50

100

150

200

250

300

350

0

50

100

150

200

250

300

350

400 400

Sep. Jan. May Sep. Jan. May

2008 2009 2010

maximum diversified banksminimum diversified banks

average

Source: Bloomberg. Note: For these charts, investment banks comprise UBS, Crdit Suisse and Deutsche Bank. Diversi ed banks comprise BNPP,Socit Gnrale, Crdit Agricole, Santander, BBVA, Lloyds, HSBC, Barclays and Commerzbank. See footnote 43 for the de nition of investment banks.


26/5125


September 201025

2 SPECIAL FEATUREON THE FUTURE

EVOLUTION OF THEEU BANKING SECTOR

25

The greater resilience of diversi ed banksis based on clear synergies between private

banking, retail and corporate banking andinvestment banking. 42 In this model, oneactivity supports another, so that the decisivefundamentals of funding, revenue and customersare secured:

Diversi ed banks are built on strongcustomer relationships and more stablefunding sources, such as deposits, whichmake them less reliant on wholesalefunding and on average more resilient toliquidity shocks.

They can count on balanced sources of

revenue that result from the cross-sellingof products, which helps to maintain pro tability (see Chart 19).

Diversi ed banks take advantage of thescale economies that result from cost-sharingacross geographical areas and businesses,which is essential in order to capture agreater share of customers and processingas well as of high-volume activities.

THE DIVERSIFIED MODEL HAS BEEN LESS

VOLATILE THAN THE SPECIALISED MODELThe performance of diversi ed institutions isalso less volatile than that of specialist banks innormal times. On a worldwide scale, the

investment banks in the sample 43 report anaverage long-term pro tability (1997-2006) of 14.8%, surpassing the 13.1% recorded bydiversi ed banks. 44 But their performance ismore volatile and more vulnerable to nancialdistress. The trend is quite similar at theEuropean level (see Chart 19): the pro tabilityof the investment-oriented European banksreached 25% in 2006, then fell to -31% in 2008,

before rallying sharply in 2009 to 8.5%. Thisnonetheless remains substantially below thelong-term average pro tability of 13% (from2001 to 2006). By comparison, Europeandiversi ed banks had an average pro tability of 19% (from 2001 to 2006), which has declinedless steeply since 2008 to reach about 10%.

More speci cally, retail and corporate activities share branches42and back-of ce infrastructures, and retail savings provide animportant source of funding for corporate lending. Likewise,investment banking activities can share some of their productsand risk management skills with corporate banking activities.Private banking can expand upon relationships with customersfrom other banking channels and share some of the basic bankinginfrastructure. Conversely, the combined bank can bene t fromfunding obtained from private banking customers savings. See,for example, De Nederlandsche Bank NV, The Dutch nancial

system: an investigation of current and future trends , 2009.Banks considered as investment banks in this special feature43are those for which the nancial and investment business haveaccounted for over 50% of their average income over the last threeyears (Morgan Stanley, Goldman Sachs, Credit Suisse, UBS andDeutsche Bank). These banks are assumed to be investment banks

for the entire period of observation for the sake of simplicity.A sample of 12 banks are considered as diversi ed banks for the44

purposes of this special feature: HSBC, CASA, Socit Gnrale,BNP Paribas, Santander, UniCredit, Barclays, RBS, JP Morgan,Citigroup, Bank of America and Wells Fargo.

Chart 19 Return on equity of investment and diversified banks from 31 December 2001to 31 December 2009(percentages)

-60

-40

-20

0

20

40

-60

-40

-20

0

20

40

averagerange

2002 2003 2004 2005 2006 2007 2008 2009

-20

0

20

40

-20

0

20

40

-40 -402002 2003 2004 2005 2006 2007 2008 2009

Sources: Bloomberg and Banque de France calculations. Note: Investment banks: Deutsche Bank, UBS, Crdit Suisse. Diversi ed banks: BBVA, Banco Santander, Barclays, BNPP, HSBC,Intesa San Paolo, Socit Gnrale, UniCredit.


27/5126


2.1.2 IN THE MEDIUM TERM: TOWARDSDIVERSIFIED, SAFER AND MORE RATIONALMODELS AND RISK PRACTICES

Many of the shortcomings revealed by the crisis(such as too cheap and abundant liquidity, toolax risk management practices, misalignedcompensation frameworks) are being addressedthrough regulatory reforms. However, theindustry has a role to play in re-examining its

business models.

DIVERSIFIED BANKS: COMBINATIONOF DIVERSIFICATION AND FOCUS ON CORECLIENTS AND MARKETSBanks return on equity 45 is likely to decreaseunder the in uence of several drivers, especiallygiven that banks are expected to continue todeleverage. This means that they will haveto discover new ways to generate revenue.One possibility may be for banks to maintaina diversi ed pro le, while at the same timeconcentrating on their core markets, activitiesand clients, thus ensuring guaranteed revenues.Banks may attempt to increase their sharesin market segments where high volumes areessential for pro tability (consumer lending,global custody, asset management, paymentsand systems, brokerage). Several factors suggestthat such developments are likely:

Most banks cannot afford to continuefunding unpro table businesses, which may

curb cross-subsidisation among activities,at least in the coming months.

A broad spectrum of businesses can bedif cult to control. Furthermore, the recentcrisis showed that, whatever activities

banks conduct, they often deal with similar products.

There are far too many interconnectionsin the global economy for business andgeographic diversi cation to provide a

reliable defence against a breakdown of business fundamentals.

SPECIALISED BANKS: SAFER AND MORE RATIONALRISK PRACTICESOver the short and medium term, the maindif culty facing specialised banks will be accessto funding sources. The cost of funding is likelyto rise with the phasing-out of quantitativeeasing measures and the potential increase ininterest rates, and the securitisation market stillsuffers from uncertainty. 46 These factors maytempt specialised banks to transfer the additionalcost to their clients. This would compress their net interest margins, since specialised bankshave smaller liquidity or funding cushions thandiversi ed banks engaged in deposit-takingactivities.

Some specialised players may choose todiversify. In the United States and Europe, thiswas already observed to a certain extent in theearly stages of the crisis, when a number of investment banking operators either teamed upwith more diversi ed operators 47 or were forcedto adjust. 48

However, there is still room for niche activities,since these market players genuinely have the

potential to create value added in their segments.In the coming years, the specialised bankingindustry is likely to be divided into two clustersof businesses:

Large players with large market shares: insome speci c segments, only large players

with high skills and market power willoperate. Their leading positions will helpdrive scale bene ts and will allow them tokeep their cost-to-income ratios low.

Niche and high-growth players requiringlow capital: these players will need

See also Box 1 in the overview chapter for a discussion of return45on equity as a performance measure.See also Section 2.2 on funding patterns.46Examples include the deals between Merrill Lynch and Bank 47of America and between Bear Stearns and JP Morgan, and theacquisition of 30% of Deutsche Postbank by Deutsche Bank.Examples include the conversion of Goldman Sachs and Morgan48Stanley into bank holding companies in September 2008.


28/5127


September 201027



27

either scale economies coupled withhigh technology (e.g. prime brokerage)or expertise and strong relationships(e.g. advisory businesses).

2.1.3 ADJUSTMENT OF BUSINESS LINES WITHINBANKS

The adverse environment has already led banksto rethink their business mix. According tomarket participants, different business lines such as investment banking, asset managementand retail activities have experienced differenttrends since 2007. In particular, the contributionof investment banking to total revenue has

been highly volatile and subject to a strongoverall reduction. Revenue from retail activitieshas remained steady, whereas that from assetmanagement business temporarily decreasedat the height of the crisis. Finally, the overallcontribution of specialised nancial servicesseems to have increased slightly.

These developments lend support to the viewthat banks are likely to review the content andstructure of their business lines in the near future.They can adjust their operations through twomain channels: by developing a pro table cross-selling mix of products (probably more basic than

before the crisis) and by streamlining the costs ineach business (reaping more economies of scale,dedicating more capital to low-risk activities,ending the cross-subsidisation of unpro table

activities, and improving distribution channelsand the productivity of the sales force). Whileevery business line is likely to experience suchdeep changes, these adjustments would probablynot fundamentally call into question somespeci c sectors such as the asset managementand the bank-insurance models. For example,asset management activities are likely toincrease, with the size criterion being the mostimportant driving factor. 49

2.1.4 OTHER POSSIBLE DEVELOPMENTS

The development of diversi ed banking modelsappears to be the most plausible scenario,considering the different factors of in uence.

However, this may not be the case if somefeatures that are assumed to be minor in this

baseline scenario gain in importance. While theshort memory of investors may result in high

performance expectations in the wake of the next boom in the long term, certain proposals in thenew regulatory framework have the potential toset an alternative direction for business modelsalready in the medium term.

SOME REGULATORY PROPOSALS: POSSIBLE CURBON THE DEVELOPMENT OF THE DIVERSIFIEDMODELCurrent proposals of the Basel Committee onBanking Supervision (BCBS) on the deductionof holdings in nancial institutions could havean impact on the diversi ed business model.The Basel III framework aims to strengthen thequantity and quality of own funds. In particular,it will require banking institutions to deduct their

banking and insurance equity holdings fromtheir core Tier 1 capital. The impact could besigni cant on large nancial groups offering acombination of banking and insurance activities,and in particular on banking groups implementinga bank-insurance development strategy. If the twoactivities are fully integrated, it would be dif cultto separate them in practice (e.g. same distributionnetworks or information technology structures).This impact could weaken these nancial groupswhich bene t from more diversi ed revenues(e.g. cross-selling) and funding sources and couldcall into question the diversi ed business model.

In a similar vein, regulatory initiativesconcerning systemically important institutionsmay put a brake on diversi cation, in particular if the proposals related to limiting the scope of activities or narrow banking are integrated intolegislation. 50 Given that these initiatives havestill not been fully detailed, it is not yet possibleto estimate their full effect on the industry.

Examples of recent acquisitions in this eld include the49acquisition of Barclays Global Investors by Black Rock and

the merger between SG Asset Management and CA AssetManagement in 2009.The same argument would apply to the possible implementation50of a surcharge for institutions that are considered systemicallyimportant.


29/51


30/5129


September 201029



29

other, the crisis has increased investorsawareness of banks capital endowments.Greater awareness is to be found not onlyamong equity investors, but also amongdebt holders, as higher capital buffers alsoreduce the risk of a bank defaulting on itsdebts. Whatever the effect of the regulatoryreforms on capital requirements, market

participants may demand buffers on theregulatory capital buffers.

Both regulatory developments and thecurrent economic and nancial environmentwill affect the capital structures of banks.The expected increase in the cost of risk willcontinue to consume bank capital in the near future, and the supervisory requirementsrelating to risk weights on a wider range of asset classes will probably be permanentlyhigher for the foreseeable future.

EU banks have already raised their Tier 1 andcapital adequacy ratios by roughly 2 percentage

points (see Chart 20). However, futuredevelopments are likely to be affected by two

factors: rst, the ability to tap markets will differ between banks, and, second, governments arenow important shareholders in the bankingsector of some EU countries. 54

2.2.2 SEARCH FOR STABLE SOURCESOF FUNDING

In the period preceding the crisis, the funding of banks was characterised by low interest rates,low risk premia and thus an inadequate pricingof the cost of risk. Wholesale and interbank sources of funding had continuously grown inimportance, whereas funding through depositswas considered unattractive.

THE RACE FOR DEPOSITSSince 2009, a growing interest in deposits has

been observed, re ected in a rise in depositsupply by the non- nancial corporation sector

A mitigating factor may emerge if banks succeed in increasing54 their pro tability, which would enable them to generatecapital internally and attract new investors. However, asargued in Section 1.2.1, this is not a likely development in theshort term.

Chart 20 Capital ratios of EU banks

(as a percentage of risk-weighted assets)

Tier 1 ratio weighted average Capital adequacy ratio weighted average

4

6

8

10

12

14

4

6

8

10

12

14

2004(347)

2005(578)

2006(643)

2007(687)

2008(716)

2009(104)

4

6

8

10

12

15

4

6

8

10

12

15

14 14

2004(420)

2005(664)

2006(736)

2007(884)

2008(1,023)

2009(113)

Source: Bankscope. Note: Figures in brackets indicate number of observations. Fixing the sample with available 2009 observations does not alter the resultssubstantially.


31/5130


(see Chart 21). Since the crisis, the interest ratesoffered by banks on customer deposits havesteeply increased in comparison with interbank rates, and the deposit margins for banks havedeclined.

Deposit funding is part of the current back to basics policy implemented by numerous banks.It will continue to be a heavily demanded sourceof funding in the medium term. As for other sources of funding, the crisis has resulted inan increased awareness of differences between

banks, with banks with established brandsgaining a competitive advantage vis--vis their weaker competitors.

This increase in deposits has put the fundingsituation of banks on a more stable footing sincethe crisis: loan-to-deposit ratios decreased in thecourse of the second half of 2009 (see Chart 22),not only owing to increased deposit supply butalso as a result of a reduction in lending activity.However, the competition for deposits between

banks can be expected to continue for tworeasons:

First, the saving capacity of depositors andtherefore the deposit supply are nite, andinvestment funds, stocks and bonds will

probably increase again in attractiveness assubstitutes for deposits for investors in thenear future.

Second, the ongoing initiative to harmonise deposit guarantee schemes in the EuropeanUnion is likely to increase the attractivenessof cross-border deposits for customers.Thus, competition for deposits will increase.Moreover, the cross-border provision of deposits may gain some additional impetusfrom the spread of internet banking.Integration in retail banking is alsoexpected to gain pace with the wide-rangingapplication of the Single Euro PaymentsArea in the medium term.

This increased competition will probably in uencethe usually low sensitivity of deposits, makingdeposits a less sticky source of funding.

Chart 21 Total deposits by sector

(annual growth rates as percentages)

-2

02

46

8

10

1214

16

-2

02

46

8

1012

14

16

non-financial corporations (total)households (total)

1999 2001 2003 2005 2007 2009

Source: ECB.

Chart 22 Loan-to-deposit ratios of 17 largeEuropean banks, 10 for H1 2009

(annual growth rates as percentages; maximum, minimum and

inter-quartile distribution)

0

50

100

150

200

250

300

350

0

50

100

150

200

250

300

350

2001 2003 2005 2007 2008 2009H1H2 H1H2

2010 2012

Historical dataPossiblescenario

medianscenario in which loan-to-deposit ratios returnto levels recorded in early 2000s 1)

scenario in which ratio growth returns to levelsrecorded in 2005-2007 2)

low dispersion of ratio among largeEuropean banking groups 3)

high dispersion of ratio among largeEuropean banking groups 4)

Source: Bloomberg. Note: Future dispersions are calculated from historical data by duplicating strong and weak past dispersions and addinga dynamic.1) this may be due to weak economic growth and bankswillingness to adjust their loans to their deposits2) this may be due to aggressive strategies by banks with easier access to wholesale funding3) all banks demon

ECB: EU Banking Structures Report -- September 2010

Documents

Transcript of ECB: EU Banking Structures Report -- September 2010