No.42 - The New Basel Capital Framework and its implementation in

OCCAS IONAL PAPER SER IESNO. 42 / DECEMBER 2005

THE NEW BASEL CAPITAL FRAMEWORK AND ITS IMPLEMENTATION IN THE EUROPEAN UNION

by Frank Dierick,Fatima Pires,Martin Scheicherand Kai Gereon Spitzer

In 2005 all ECB publications will feature

a motif taken from the

€50 banknote.

OCCAS IONAL PAPER S ER I E SNO. 42 / DECEMBER 2005

This paper can be downloaded without charge from the ECB’s website (http://www.ecb.int) or from the Social Science Research Network

electronic library at http://ssrn.com/abstract_id=807416.

THE NEW BASEL CAPITAL FRAMEWORK AND ITS IMPLEMENTATION IN

THE EUROPEAN UNION *

by Frank Dierick 1,Fatima Pires 1,

Martin Scheicher 2

and Kai Gereon Spitzer 3

* The authors are grateful for research assistance by Sara Testi and comments by Mauro Grande, Panagiotis Strouzas and ananonymous referee. The views expressed in this paper are those of the authors and do not necessarily reflect those of the

ECB, the Eurosystem or the Deutsche Bundesbank.1 Directorate Financial Stability and Supervision of the ECB

2 Directorate General Research of the ECB3 Banking and Financial Supervision Department of the Deutsche Bundesbank

© European Central Bank, 2005

AddressKaiserstrasse 2960311 Frankfurt am MainGermany

Postal addressPostfach 16 03 1960066 Frankfurt am MainGermany

Telephone+49 69 1344 0

Websitehttp://www.ecb.int

Fax+49 69 1344 6000

Telex411 144 ecb d

All rights reserved. Any reproduction,publication and reprint in the form of adifferent publication, whether printedor produced electronically, in whole orin part, is permitted only with theexplicit written authorisation of theECB or the author(s).

The views expressed in this paper donot necessarily reflect those of theEuropean Central Bank.

ISSN 1607-1484 (print)ISSN 1725-6534 (online)

3ECB

Occasional Paper No. 40December 2005

C O N T E N T SABBREVIATIONS 4

ABSTRACT 5

EXECUTIVE SUMMARY 6

INTRODUCTION 7

PART 1 – OVERVIEW OF THE NEW FRAMEWORK 8

1 DEVELOPMENT 8

2 STRUCTURE 9

2.1 General overview 9

2.2 Pillar I – Minimum capitalrequirements 11

2.3 Pillar II – Supervisory reviewprocess 18

2.4 Pillar III – Market discipline 19

3 THE NEW FRAMEWORK IN THE EU 20

3.1 Legal and institutional setting 20

3.2 Specific features of EUimplementation 23

PART II – ISSUES OF SPECIFIC RELEVANCEIN THE EU CONTEXT 29

4 PROCYCLICALITY 29

4.1 Definition 29

4.2 Empirical evidence 29

4.3 Mitigating measures 31

4.4 Monitoring in the EU 31

5 HOME-HOST ISSUES AND THECONSOLIDATING SUPERVISOR 32

5.1 International and Europeancontext 32

5.2 Powers and responsibilitiesof the consolidating supervisor 33

5.3 Assessment of the consolidatingsupervisor 36

6 REAL ESTATE LENDING 37

6.1 Overview 37

6.2 Residential real estate 38

6.3 Commercial real estate 40

7 COVERED BONDS 41

7.1 Concept of covered bond 41

7.2 Capital treatment 42

CONCLUDING REMARKS 43

GLOSSARY 45

REFERENCES 48

LIST OF OCCASIONAL PAPERS 52

4ECBOccasional Paper No. 42December 2005

ABBREVIATIONS

AIG Accord Implementation GroupAMA advanced measurement approachesASRF asymptotic single risk factorBCBS Basel Committee on Banking SupervisionCAD Capital Adequacy Directive (93/6/EEC)CBD Codified Banking Directive (2000/12/EC)CEBS Committee of European Banking SupervisorsCP consultative paperCRD Capital Requirements DirectiveCRE commercial real estateEAD exposure at defaultECAI external credit assessment institutionECB European Central BankEFR European Financial Services Round TableEU European UnionFSAP Financial Services Action PlanIRB internal ratings-basedLGD loss given defaultM effective maturityMiFID Markets in Financial Instruments Directive (2004/39/EC)PD probability of defaultQIS quantitative impact studyRRE residential real estateSMEs small and medium-sized enterprisesS&P Standard & Poor’sUCITS Undertakings for Collective Investment in Transferable Securities

5ECB


ABSTRACT

ABSTRACT

Following the adoption by the BaselCommittee of new capital rules for banks, aprocess is now taking place in the EU totranspose the rules into Community law and,ultimately, into national legislation. This papergives an overview of the main issues that relateto the EU implementation, mainly from theperspectives of financial stability and financialintegration. Although the EU rules are to alarge extent based on the texts of the BaselCommittee, modifications have beenintroduced to account for the specific legal andinstitutional setting, as well as for somefeatures of the European financial system.The paper gives an overview of thesemodifications and deals in greater detail with anumber of selected topics: the monitoring ofprocyclicality, the role of the consolidatingsupervisor and the treatment of real estatelending and covered bonds. The paperconcludes with an outlook for the future.

Key words: banks, Basel II, capitalrequirements, financial regulation, financialstability, financial supervision, risk management.

JEL classification: G21, G28


EXECUTIVE SUMMARY

In June 2004, the Basel Committee on BankingSupervision published its new capital rules forbanks (“Basel II”). Recently, the EuropeanParliament and the Council approvedlegislation transposing these rules intoCommunity law and this legislation will,in turn, be transposed into national law.The European Central Bank (ECB) has aninterest in these developments because of theirpossible implications for financial supervision,financial stability and financial integration. Aswell as providing a non-technical overview ofthe main elements of the new framework, thispaper focuses on issues related to the EuropeanUnion (EU) implementation. The paper isstructured in two parts. Part I gives an overviewof Basel II and the main issues that are relevantfor its implementation in the EU. Against thisgeneral setting, Part II deals in greater detailwith a number of selected topics that areparticularly important for the EU: themonitoring of procyclicality, the role of theconsolidating supervisor and the treatment ofreal estate lending and covered bonds.

Basel II is based on three mutually reinforcingpillars: minimum capital requirements (PillarI), the supervisory review process (Pillar II)and market discipline (Pillar III). A maininnovation is that a set of increasinglysophisticated approaches is now available tobanks to calculate their minimum capitalrequirements. Although the simplest method tocalculate capital for credit risk is based onassessments by rating agencies, under the mostadvanced approaches, banks are allowed to usetheir own estimated risk parameters. Moreover,for the first time, banks are required to holdcapital for operational risk.

While the EU rules are to a very large extentsimilar to those developed by the BaselCommittee, they are not an exact copy. First,the legal and institutional setting is different.The EU rules are legislative in nature andbinding in all Member States. Moreover, therecently established Committee of European

Banking Supervisors (CEBS) will play a majorrole in developing supervisory guidance toimplement the new rules. Second, the specificstructure of the European financial system,such as the role of the Single Market, justifiescertain deviations.

The EU rules diverge from Basel II in theirscope of application and the range ofapproaches that will be available to institutionsto calculate their capital requirements. Theyalso change the supervisory responsibilitiesby giving a larger role to the consolidatingsupervisor and by requiring supervisorydisclosure. In the Single Market there isa greater need for cooperation betweensupervisors. Some specific features of the EUeconomy and financial system need to beaddressed as well. This includes venturecapital, real estate lending and covered bonds.In addition, the EU rules provide for specificarrangements concerning the issue ofprocyclicality. Empirical evidence indicatesthat capital rules can indeed exacerbate theeconomic cycle. However, in the developmentof the new rules various changes were madeto accommodate these concerns. A studyperformed under the auspices of the EuropeanCommission confirmed that procyclicalityshould no longer be a major problem, althoughit should be kept under review. To that end, theEU rules provide for a monitoring arrangementthat also involves the ECB.

In more detail, the EU rules provide for anenhanced role of the consolidating supervisor,in particular, through the process of approval ofadvanced methods to calculate capitalrequirements on a group-wide basis. Althoughthis extended role does not go as far as somelarge financial groups would like, it willsimplify the interaction between the bankinggroup and its supervisors as well as between thevarious supervisors. It is also likely to bebeneficial for financial stability andintegration, although it raises compleximplementation issues that will need to befollowed up closely.

7ECB


INTRODUCTION

Real estate lending is another important areagiven its significance in bank lending and thatbanking crises have often coincided with crisesin the real estate market. In the treatment of realestate lending, the EU rules seem to achievegreater consistency than Basel II, although thisleads to a somewhat less conservativetreatment of commercial real estate lending.From a financial stability perspective,however, it is important that this more lenienttreatment does not apply to the more risky realestate, the income of which is used for debtservicing. Some of the rules offer further scopefor supervisory convergence, which would bebeneficial for the integration of real estatelending in Europe.

Covered bonds receive a specific treatmentunder the EU rules that is not present in BaselII. These financing instruments, with the“Pfandbrief” as its best-known exponent, areimportant for EU capital markets. Thispreferential treatment, reflecting the additionalsecurity inherent in these bonds, is fullyconsistent with the principles underlying BaselII. It also seems that most of the existingcovered bonds in the EU would qualify for thispreferential treatment.

This paper concludes by listing some keychallenges for the capital rules, first, as regardstheir EU implementation and, second, in along run perspective. In the longer term, thedefinition of regulatory capital and itsinteraction with the new accounting rules willhave to be addressed. Other areas which maywarrant further study relate to the blurring ofthe border between credit risk and market risk,and the possible use of full credit portfoliomodels by banks to calculate their capitalrequirements.

INTRODUCTION

In June 2004, the Basel Committee on BankingSupervision (BCBS) published its newsolvency rules for banks1, the New CapitalFramework (“Basel II”).2 A parallel process

has taken place in the EU to transpose theinternationally agreed capital standards intoCommunity law. Member States will also haveto transpose the EU rules into nationallegislation in due course.

The ECB has an interest in these new rulesbecause of its responsibilities in the areas offinancial supervision and financial stability.3

The New Framework may also affect theEurosystem credit operations through its rulesfor ratings assigned to financial assets used ascollateral for such operations.4 In addition, theproposed EU rules provide for arrangements tomonitor the macro-economic impact of theNew Framework, which includes a specific rolefor the ECB. Finally, the implementation of thenew solvency rules may have implications forfinancial integration, a Community objectivethat is supported by the ECB. The ECB wasactively involved in the discussions on the newrules through its participation in the BCBS andEuropean fora, and via its policy advice.5

This paper aims to give an overview of the mainissues related to the implementation of the NewFramework in the EU, especially from theperspectives of financial stability and financialintegration. Although the proposed EU rulesare to a large extent similar to the onesdeveloped by the Basel Committee, they are notan exact copy of them. First, the legal andinstitutional setting is different. The EUframework is legislative in nature and thereforebinding in all Member States. Moreover,account has to be taken of the new regulatoryand supervisory set-up in the EU as a result ofthe extension of the “Lamfalussy framework”to the banking sector. Second, the specificfeatures of the European financial system, suchas the role of the Single Market, justify somedeviations.

1 In this paper, the term “bank” is used as a synonym for“credit institution”.

2 Basel Committee on Banking Supervision (2004a).3 Art. 105(5) of the Treaty establishing the European

Community.4 See in this respect ECB (2005a).5 See in particular ECB (2005b).


This paper is structured in two main parts. Thefirst one deals with the Basel Framework andits implementation in the EU and the second oneaddresses certain topics that are particularlyimportant in the EU context. In Part I, Section 1briefly recalls the main steps that were taken infinalising the Basel II Framework and Section2 provides an overview of its main elements. InSection 3, the EU implementation of Basel II isdiscussed in general terms. It starts by recallingthe specific legal and institutional setting in theEU and then turns to the main differencesbetween the Basel and the EU rules.

Part II builds on this general analysis andsingles out a number of areas, which are ofparticular relevance for the EU, andinvestigates them in greater detail. Section 4tackles the issue of procyclicality, i.e. theconcern that the new capital rules mayexacerbate economic cycles. It reviews theempirical evidence, the different measures tomitigate such effects and the specificmonitoring arrangement provided for under theEU rules. Implementation issues in a cross-border context are analysed in Section 5. Here,the EU rules diverge from Basel by enhancingthe role of the authority responsible forsupervision on a consolidated basis. Section 6compares the EU and Basel rules in theirtreatment of real estate collateral, both forresidential and commercial real estate lending.The treatment of covered bonds, a financinginstrument that is particularly popular in theEU, is dealt with in Section 7. The paperconcludes by giving a future outlook as Basel IIand its implementation in the EU represent veryimportant steps in a longer term process, butnot its end.

PART 1 – OVERVIEW OF THE NEW FRAMEWORK

1 DEVELOPMENT

With its New Framework for capitalrequirements, the BCBS aims to alleviate someof the drawbacks of the current regime datingback to 1988 (“Basel I”).6 First, the current

Basel I rules offer a simplified and rigidquantification of credit risk, that are not in linewith best practices applied by banks in theirrisk management. Basel II significantly refinesthe framework’s risk sensitivity by requiringhigher levels of capital for high-risk borrowers.By aligning required capital more closely to abank’s own risk estimates, the New Frameworknarrows the gap between regulatory capital andeconomic capital (requirements).7 It thereforeencourages banks to improve their riskassessment methods. Furthermore, theincreasing use of risk mitigation andsecuritisation has created the need to treat themmore extensively. Another drawback of thecurrent framework stems from its unintendedincentives for capital arbitrage throughtechniques such as securitisation. In addition,the current framework lacks rules for propermarket disclosure and therefore does notsupport market discipline. Finally, it offers noguidance for the supervisory review of banks’risk management practices.

Work on the New Framework started after 1996when the Capital Accord underwent a majoramendment to introduce capital requirements formarket risk.8 At the current juncture, fullimplementation of the Basel II Framework isexpected for the end of 2007. The whole process,therefore, from the initiation of discussions tothe moment when the most advanced calculationmethods will become available, will havestretched out over more than a decade.

Box 1 lists the key development stages andthe next steps in the implementation. Thedevelopment process was characterised by anintensive dialogue with the banking industry,reflected in the various consultation paperspublished by the BCBS. The Basel Committeealso carried out several quantitative impact

6 Basel Committee on Banking Supervision (1988). Basel I isalso called the “Basel Capital Accord”.

7 Regulatory capital is capital that is eligible to meetregulatory capital requirements; economic capital is capitalheld by the bank internally as a result of its own riskassessment. Technical terms are explained in the glossary atthe end of the paper.

8 Basel Committee on Banking Supervision (1996).

9ECB


2 STRUCTURE

Box 1

CHRONOLOGY OF THE BASEL PROCESS

June 1999 First Consultation Paper (CP1)July 2000 Quantitative Impact Study 1(QIS1)January 2001 Second Consultation Paper (CP2)April 2001 Quantitative Impact Study 2 (QIS2)November 2001 Quantitative Impact Study 2.5 (QIS2.5)October 2002 Quantitative Impact Study 3 (QIS3)April 2003 Third Consultation Paper (CP3)2004/2005 Quantitative Impact Studies 4 and 5 (QIS4/5)January 2004 “Madrid compromise”June 2004 Publication of the “New Framework” documentApril 2005 Consultation on the trading book review and double defaultJuly 2005 Publication of the trading book review and double defaultSpring 2006 Scheduled recalibration of the New FrameworkEnd 2006 Scheduled G10 implementation of simpler methodsEnd 2007 Scheduled G10 implementation of advanced methods

studies to gauge the impact of the new rules onbanks’ solvency positions and further refine therules.

Basel II has been designed as an evolutionaryframework, so updates will be made over timeto keep pace with ongoing developments in thefinancial industry. Prior to the implementationof the new rules, the Framework may undergo aquantitative adjustment (“recalibration”) onthe basis of the results of the most recent impactstudies. In addition, some technical changeswere introduced after June 2004 to address the“double default” issue and to bring thetreatment of trading activities more in line withthe New Framework.9 In the longer term, theBCBS intends to address a number of otherareas which are discussed in greater detail inthe last section dealing with the future outlook.

2 STRUCTURE

2.1 GENERAL OVERVIEW

An overview of the New Framework and itsmain components is shown in Chart 1. The darkboxes, which will be discussed in greater detail

below, refer to the components that have beenintroduced or to which there are major changesas a result of the new rules; the bright boxesrefer to the components that have remainedunchanged. Whereas Basel I only coveredminimum capital requirements, the Basel IIFramework now rests on three complementarypillars, namely minimum capital requirements(Pillar I), the supervisory review process(Pillar II) and market discipline (Pillar III). Inorder to effectively support financial stability,the Framework requires a smooth interactionbetween all three pillars.

As regards Pillar I, the minimum solvency ratioof 8% remains unchanged. This ratio expressesthe relationship between the bank’s regulatoryown funds (capital) and its “risk-weightedassets”, a measure of the risks it incurs. Risk-weighted assets are asset values multiplied by afactor (risk weight) that is a proxy of the (credit)risk related to these assets. For operational riskand market risk, the two other risk categories

9 Basel Committee on Banking Supervision (2005b). “Doubledefault” refers to the fact that the risk of both a borrower anda guarantor defaulting on the same obligation may besubstantially lower than the risk of only one of the partiesdefaulting; this feature is not suff iciently recognised in theJune 2004 Framework.


Chart 1 Overview of the New Framework

covered by the Framework, the risk-weightedassets that enter into the capital ratio are derivedfrom the directly calculated capital requirementsby multiplying them by 12.5 (the reciprocal ofthe minimum ratio of 8%). In addition, thedefinition of regulatory capital (the numerator ofthe capital ratio) was basically unaffected.

Pillar I, however, provides a fundamental updateof the Basel I methodology for the calculation ofrisk weighed assets, the denominator of thecapital ratio. First, operational risk is introducedas a new risk category for which the bank has tohold regulatory capital. This risk categorycomprises losses resulting from inadequate orfailed internal processes, people or systems, orfrom external events.

Second, a range of increasingly sophisticatedand risk-sensitive options are now available fordetermining banks’ capital requirements, bothfor credit risk and operational risk. In this way,

the option can be chosen that best suits the bank’sspecific features. Moreover, incentives are inplace for banks to adopt the more sophisticatedapproaches and thus improve their riskmanagement capabilities over time.10 In the areaof credit risk, two methods are available, namelythe standardised approach and the internalratings-based (IRB) approach. The former tiesrisk weights to ratings provided by recognisedrating agencies. The latter uses banks’ ownestimates of certain risk factors; depending onthe risk factors they are allowed to estimate, adistinction is made between a “foundation” andan “advanced” approach. The new rules for creditrisk also cover a detailed treatment ofsecuritisation and credit risk mitigation. Finally,in the area of operational risk, a bank can

10 For example the QIS3 results for the EU showed thatinstitutions adopting the standardised approach would facean increase in capital requirements of 2%, while thoseadopting the foundation IRB and the advanced IRBapproaches would see a decline of 7% and 9% respectively.See European Commission (2003a).

Source: ECB.

Pillar IMinimum

capitalrequirements

Pillar IISupervisory

review process

Pillar IIIMarket

discipline

BASEL II FRAMEWORK

Risk-weighted

assets

Operationalrisk

Standardisedapproach

Own funds

Market risk Core capital Supplementarycapital

Credit risk

Internalratings-based

approach


FoundationIRB

AdvancedIRB

Advancedmeasurement

approach


Internalmodels

approach

Basic indicatorapproach

11ECB


2 STRUCTURE

calculate its capital requirements on the basis ofits gross income (basic indicator approach andstandardised approach) or by using its ownmodel (advanced measurement approach). Asregards market risk, the New Framework leavesthe existing approaches basically unchanged.

2.2 PILLAR I – MINIMUM CAPITALREQUIREMENTS

2.2.1 CAPITAL REQUIREMENTS FOR CREDITRISK

(a) Standardised approachThe standardised approach is closest to thepresent capital rules. Exposures are classifiedinto a set of standardised asset classes (seeChart 2) and a risk weight is applied to eachclass, reflecting the relative degree of creditrisk. As under Basel I, off-balance sheetexposures are for capital purposes transformedinto “assets” through the application of “creditconversion factors”. The main changescompared to Basel I relate to the use of externalcredit ratings as the basis for determining therisk weights and the greater differentiation inthe possible risk weights.

Chart 2 Asset c lasses in the standardised approach

To give an example, risk weights for corporateexposures are now connected with their creditratings as indicated in Table 1 (for illustrativepurposes, ratings by Standard & Poor’s or S&Pare used).

Compared to Basel I, where all corporateexposures are weighted at 100%, there is now aconsiderable differentiation in the risk weights.The weight for investment-grade firms hasdeclined considerably (e.g. to 20 % for AAA),whereas in the non-investment grade segment, arisk weight of 150% applies to firms rated below“BB-”. Furthermore, unrated firms now obtainthe same risk weight as that formerly obtained byall corporates under Basel I.

For claims on banks, the former distinctionbetween institutions from OECD (20% riskweight) and non-OECD countries (100%) is nolonger applied. Instead, two options areavailable to national supervisors. Under thefirst option, the risk weights for banks arederived from the ratings of the country in whichthe bank is incorporated. Under the secondoption, the risk weights are determined on thebasis of the bank’s own rating.

Rating AAA to AA- A+ to A- BBB+ to BB- Below BB- Unrated

Risk weight 20 50 100 150 100

Table 1 Risk weights for corporate exposures under the standardised approach

(percentages)

Corporate exposures

Bank exposures

Sovereignexposures

Retail exposures

Commercialreal estate exposures

Other assetsResidentialproperty

exposures

Asset classes in the standardised approach

Source: ECB.


Retail exposures (75% instead of 100%) andmortgage loans (35% instead of 50%) aretreated more advantageously than underBasel I. Exposures to small businesses mayunder certain conditions also benefit from thepreferred retail treatment.

The rating agencies, or external creditassessment institutions (ECAIs) as they arereferred to in the New Framework, must obtainrecognition from the banking supervisor beforetheir ratings can be used by banks fordetermining risk weights. To get suchrecognition, an ECAI must satisfy each of thefollowing six criteria:

1. Objectivity of the rating or credit riskassessment methodology.

2. Independence. The ECAI must be free frompolitical or economic pressures, whichcould influence the analysis.

3. International access/transparency. TheECAI should offer its services to bothdomestic and foreign firms at similar terms.

4. Disclosure of material information.Includes the rating methodology, thedefinition of default, the time horizon, themeaning of each rating, the actual defaultrates experienced in each assessmentcategory and the transition matrix.11

5. Sufficient resources for offering creditassessments of high quality.

6. Credibility of the credit assessments.

(b) Internal ratings-based approach

Theoretical foundationThe internal ratings-based (IRB) approach tocredit risk is one of the most innovativeelements of the New Framework because itallows banks themselves to determine certainkey elements in the calculation of their capitalrequirements. Hence, the risk weights – andthus the capital charges – are determined

through a combination of quantitative inputsprovided either by banks or supervisoryauthorities, and risk weight functions specifiedby the BCBS. The new methodology isdesigned to be suitable for implementation bybanks of different size, business structure andrisk profile. Due to this characteristic, astandardised approach to modelling credit riskacross all types of banks is used for supervisorypurposes for the first time.

The IRB approach is closely linked to keyresults of academic work on credit riskmodelling.12 Its theoretical basis is theasymptotic single risk factor (ASRF) model ofcredit risk. Here, the likelihood of a borrowerbeing unable to repay his debt is derived fromthe distance between the value of its assets andthe nominal amount of his debt. The value ofthe firm’s assets is modelled as a variablewhich changes over time, in part as a result ofthe impact of random shocks. Default occurswhen a borrower’s assets are too low to coverits debt. The corresponding measure of creditrisk within a certain time frame (commonly setat one year, also in Basel II) is the probabilityof default (PD).

The ASRF characteristic of the model impliesthat it does not take into account borrower-specific idiosyncratic risks, i.e. risks that canbe diversified away in the lending bank’s loanportfolio. Instead, the model measures themarginal risk contribution of an exposure thatit would add to an already well diversifiedportfolio. In this respect the IRB approachdiffers from models that some banks applyinternally which measure a loan’s riskcontribution to a bank’s actual portfolio,inclusive of a potential additionaldiversification effect achieved by adding anexposure to this specific borrower (“credit riskportfolio model”). The IRB approach thereforecontains a deliberate simplification compared

11 The transition matrix provides a distribution of the likelychanges in a borrower’s credit quality (expressed by itsrating) over a certain time period, usually f ixed at one year.

12 Details on the theoretical background are given in BaselCommittee on Banking (2005a) and Gordy (2003).

13ECB


2 STRUCTURE

with the most advanced techniques currentlyapplied. This simplification allows for a modelthat is standardised and can be applieduniformly to banks of different sizes andportfolio compositions. It also avoids theparticular uncertainties connected with theestimation of correlations between the risks ofindividual borrowers.

Under the IRB approach, the requiredminimum capital is based on the probabilitydistribution of losses due to the default risk in aportfolio of loans or other financialinstruments. The horizon of the risk assessmentis set at one year. The IRB model furtherassumes a 99.9% confidence level. This meansthat once in a thousand years, the actual loss isexpected to exceed the model’s estimate.

Designed to address unexpected lossesAs a result of the agreement reached by theBCBS in January 2004 with the “Madridcompromise”, the IRB capital requirementsnow cover only unexpected losses whereaspreviously they were designed to cover bothexpected and unexpected losses. Unexpectedlosses are losses that occur above expectedlevels and this at a certain confidence level(99.9%). Although capital requirements are

designed to cover unexpected losses, bankshave to cover their expected losses on anongoing basis, e.g. through pricing, provisionsand write-offs.

In the IRB framework, banks have to comparethe stock of provisions they have made to coverloan losses with the expected losses based onthe IRB parameters. Any shortfall should bededucted equally from core capital andsupplementary capital13 and any excess will beeligible for inclusion in supplementary capitalsubject to a cap. The cap depends on the risk-weighted assets and is currently set at 0.6%.

Differentiated according to asset classesJust like the standardised approach, the IRBapproach distinguishes between asset classes(see Chart 3) to which different supervisoryrisk weight functions apply. Details on thecapital requirements for standard corporateloans are provided below. The retail portfolio isof considerable importance because it also

Chart 3 Asset c lasses in the internal rat ings-based approach

13 Core capital or Tier 1 capital consists of own fundscomponents of the highest quality, such as fully-paid capitaland disclosed reserves from post-tax retained earnings.Supplementary capital or Tier 2/Tier 3 capital consists ofown funds components of lower quality, such as certain typesof subordinated debt.

Source: ECB.

Asset classes in the internal ratings-based

approach

Retail exposures

Equity exposures

Bank exposures

Sovereignexposures

Specialisedlending

Exposures tosmall andmedium-

sized enterprises

Residentialmortgage

loans

Qualifyingrevolving

retail exposures

Purchasedreceivables

Other retail(incl. loans

to small businesses)

Purchasedreceivables

Othercorporate exposures

Corporate exposures


Chart 4 Basic structure of the internal rat ings-based approach

applies to a sizeable fraction of lending to smallfirms. Subject to certain conditions, the NewFramework permits aggregate exposures to asingle firm of up to €1 million to be treatedunder the IRB approach as retail exposures,which is advantageous compared to thetreatment of other corporate lending. Evenloans to other small and medium-sizedenterprises (SMEs) that do not qualify as“retail” can benefit from a preferentialtreatment based on an adjustment relative to thefirm’s size under the standard corporatetreatment. This reduction applies when thefirm’s total sales are between €5 million and€50 million and its impact declines inproportion to the sales. Economically, the firmsize adjustment can be justified by the fact thatdefault probabilities for smaller firms areobserved to be less correlated with the overallstate of the economy so that they contributerelatively less risk to a well diversified loanportfolio.

Calculation methodThe calculation of capital requirements for aloan’s default risk under Basel II requires four

input parameters to the supervisory risk weightfunctions (see Chart 4):

1. Probability of default (PD): Estimate of thelikelihood of the borrower defaulting on hisobligations within one year.

2. Loss given default (LGD): Loss on theexposure following the borrower’s default,commonly expressed as a percentage of thedebt’s original nominal value.

3. Exposure at default (EAD): Nominal valueof the borrower’s outstanding debt.

4. Effective maturity of the loan (M).

A “foundation” and an “advanced” version ofthe IRB approach is available, the difference inthe two approaches being in the input variablesfor which the bank can use its own estimates.Both approaches rely on banks’ PD estimates,but banks’ internal estimates of LGD, EAD andM can only be applied in the advanced IRBapproach.

LGD

PD 1)

EAD

M

Risk weightSupervisory

risk weight function

Source: ECB.1) To be estimated by the bank under the foundation variant; for the other risk factors, a value f ixed by the BCBS has to be used. Forthe advanced variant, all four risk factors have to be estimated by the bank. In principle, both the foundation and the advanced IRBapproaches are available for all asset classes, the exception being the retail class, where only the advanced version is available.

15ECB


2 STRUCTURE

The resulting risk weights are plotted in Chart 5with the PDs shown on the horizontal axis. Asdesigned by the BCBS, the capital charge forretail portfolios is significantly below thecharge for corporate loans. This difference is ofconsiderable importance as it indicates that alarge number of small borrowers, includingsmall commercial undertakings, will benefitfrom a more favourable capital treatment thanlarger corporate borrowers.

A particular aim of the BCBS is to preserveoverall capital neutrality compared to thecurrent capital requirements, i.e. the capitalrequired from an average bank should not differmarkedly under the current Accord and BaselII. To this end, the BCBS had introduced intothe IRB formula a multiplier which is currentlyset at 1.06 (not reflected in Chart 5). This valueis based on the studies that the BCBS hasperformed so far on the overall impact of thenew requirements. With future assessments ofthe quantitative impact of the new capital rules,the issue of capital neutrality may have to be re-addressed. This multiplier provides then for asimple way to adjust the overall level of capitalrequired from IRB banks.

Practical implementationThe use of the IRB approach is subject to anexplicit supervisory approval, which dependson meeting certain minimum requirementsfrom the outset and on an ongoing basis. Theserequirements are aimed at the IRB systemproviding an adequate assessment of the bank’sexposures, a meaningful differentiation of riskand a reasonably good estimate of risk. Forexample, a qualifying IRB system is required tohave two separate dimensions, the risk ofborrower default and transaction-specificfactors (e.g. collateral, seniority, producttype). The IRB approach must also beconsistent with the internal use by the bank ofthe estimates it produces (“use test”).

In the practical implementation of the IRBsystem, validation (the assessment of thesoundness of the different system elements)will be important. In this context, the Accord

Implementation Group (AIG) of the BCBShas outlined a number of principles.14 First,validation is interpreted as an assessment of thepredictive ability of a bank’s risk estimates andthe use of ratings in the credit process. Inaddition, the bank has primary responsibilityfor validation. Furthermore, validation of theIRB approaches should encompass bothquantitative and qualitative elements. Finally,validation processes and outcomes should besubject to an independent review.

(c) Asset securitisationSecuritisation is one of the most rapidlygrowing activities of major banks. Banksincreasingly apply it to pools of loans on theirbalance sheets. In parallel, other creditinstruments such as corporate bonds areincreasingly used as underlying assets forsecuritisation transactions. With the explicittreatment of securitisation, Basel II providesfor an internationally harmonised standard forthe supervisory treatment of such transactions.At the same time, by making the capitalrequirements depend on the risk in the

14 Basel Committee on Banking Supervision (2005c).

Chart 5 IRB r isk weights across asset c lasses

(percentages)

Source: ECB.Note: Standard assumptions about the risk factors other thanPD, are made. Risk weights are calibrated to cover onlyunexpected losses, which have to be met via capitalrequirements.

Probability of default

020406080

100120140160180200220

020406080100120140160180200220

0 1 2 3 4 5 6 7 8 9 10

Risk weight

corporate, sovereign and banks exposuresexposure to small and medium size entitiesretail exposure: residential mortgage loansretail exposure: revolving creditsother retail exposures


securitisation positions, Basel II aims atreducing the scope for capital arbitrage, a keydrawback of the current framework andapparently an important securitisation driver.

Pillar I contains detailed rules for thesupervisory treatment of securitisation, whichcover the two roles a bank can play in asecuritisation transaction, namely as originatorand investor. In the first case, the banksecuritises its own assets; in the second case, itbuys and holds tranches of securitisations.Basel II deals with traditional transactions,e.g. the sale of loans through asset backedsecurities, as well as synthetic securitisations,such as the transfer of credit risk through creditderivatives without selling the loans.

Just as for credit risk, a number of approachesof different complexity are introduced to dealwith the wide variety in instruments and degreeof sophistication of the bank. Also here, astandardised and an IRB approach areavailable. The structure of the standardisedapproach for securitisation exposures is similarto the standardised approach for credit risk,although tranches that carry a higher risk or areunrated are dealt with more conservatively(higher risk weight or capital deduction).

Banks which were given supervisory approvalto use the IRB approach for the type ofunderlying exposures securitised must alsoapply the IRB methodology to securitisation.However, instead of a foundation andadvanced approach, there are now threeways of calculating capital requirements:the (external) ratings-based approach, thesupervisory formula approach and the internalassessment approach. The first is similar to thestandardised approach, although a greater riskdifferentiation is provided for. The next twoapproaches apply to unrated exposures, andthe third only to the specific case ofexposures resulting from ABCP (asset backedcommercial paper) programmes.

(d) Credit risk mitigationCompared with the current Capital Accord, theNew Framework contains a wider range ofcredit risk mitigation techniques that mayreceive recognition in the form of lower capitalrequirements, subject to prudent eligibilitystandards. For credit risk mitigation in the formof guarantees and credit derivatives, theborrower’s risk weight is replaced by thatof the protection provider. This “substitutionapproach” was recently modified by the BCBSto take account of the pair-wise correlation ofthe borrower and protection provider’s defaultprobabilities.15

For collateralised exposures, the Frameworkcomprises a range of methodologies that are ofdifferent degrees of sophistication. In general,the more sophisticated the methodology andthe more stringent the application conditions,the wider the range of eligible collateral.Compared to Basel I, the range of eligiblecollateral has also been expanded.

Banks using the standardised approach canapply the “simple method” for financialcollateral under which the risk weight of theborrower is replaced by the risk weight thatwould apply to the collateral if it were theexposure. Under both the standardised and thefoundation IRB approaches, there is also a“comprehensive method” for financialcollateral that foresees that the exposureamount under the standardised approach, or thesupervisory LGD under the foundation IRB, isreduced to reflect the adjusted value of thecollateral. In adjusting the collateral value,banks have to take account of the volatility ofits market value. A specific feature of thefoundation IRB is the presence of a supervisorymethod for the recognition of certain physicalcollateral.

For banks using the advanced IRB, the range ofadmissible collateral is unbounded as long asthe bank can demonstrate that it has goodestimates for the collateral value in the

15 Basel Committee on Banking Supervision (2005b).

17ECB


2 STRUCTURE

situation when the borrower has defaulted.While under the standardised approach there isin principle no recognition of physicalcollateral, real estate collateral is an exception.Also under the foundation IRB approach, thereare particular conditions that need to beobserved for this kind of collateral, asdiscussed in greater detail in Section 6.

2.2.2 CAPITAL REQUIREMENTS FOROPERATIONAL RISK

Operational risk has so far not been subject tocapital requirements. However, this does notimply that adding this new component toregulatory capital leads for an average bank tohigher overall requirements. Rather, theadditional requirement for operational riskcorresponds on average to lower capitalrequirements for credit risk compared with thecurrent rules so that the BCBS goal of capitalneutrality can on balance be reached. Foroperational risk, three options of differentlevels of sophistication are introduced.

The two most simple options are based on anindicator, namely gross income, which servesas a rough proxy for the size and the degree ofrisk of the operations. For this purpose, grossincome is defined as the sum of net interestincome and net non-interest income. Fees paidto outsourcing providers are not deducted fromthe income because it was considered thatoutsourcing is not a perfect mitigant foroperational risk and should therefore not leadto lower capital requirements. Gross income isalso adjusted for a number of items that areconsidered irregular and is in additionsmoothed by using a three-year average,excluding negative annual figures. Thisindicator, multiplied with a supervisory factor,delivers the capital requirement.

Under the basic indicator approach, thesupervisory factor (called “alpha”) is appliedto the total gross income. Under thestandardised approach, the gross income issplit out over eight different business lines,namely corporate finance, trading and sales,retail banking, commercial banking, payment

and settlement, agency services, assetmanagement, and retail brokerage. To eachbusiness line a different supervisory factor(called “beta”) is applied, reflecting therelative risk of the business line according tothe BCBS’ expert judgment. The average betaequals the alpha so that clear incentives toadopt the standardised rather than the basicindicator approach are only present for thosebanks that derive most of their gross incomefrom business lines with low betas, such asretail banking. Banks that want to use thestandardised approach have to implement a riskmanagement for operational risk that conformsto a number of qualitative minimumrequirements; such requirements are notmandatory for banks using the basic indicatorapproach.

In contrast to the case of the IRB as the mostadvanced approach for credit risk, the BCBSdoes not resort to a single modelling techniquefor operational risk. Rather, the most advancedoption for determining regulatory capital foroperational risk consists of a class ofapproaches referred to as the advancedmeasurement approaches (AMA). Under theAMA, the regulatory capital requirement iscalculated on the basis of banks’ internaloperational risk measurement systems. Thesehave to take account not only of actual internaland external loss data, but also of scenarioanalyses and factors relating to the banks’business environment and internal controls.Furthermore, the model has to achieve astatistical soundness standard comparable tothat of the IRB approach, where capital chargesare based upon a one-year time horizon and a99.9% confidence level, as described above.Subject to compliance with these modelproperties, banks are free to develop their ownapproach. This freedom is explained by the factthat, until now, no reliable candidate for astandard operational risk model has beenidentified.

In addition to the soundness standards for themodel itself, banks that request approval fromtheir supervisors for their AMA have to comply


with minimum requirements on theiroperational risk management that are moredemanding than for the standardised approach.By taking into account banks’ businessenvironment and internal controls, banks underthe AMA are in principle able to mitigate theiroperational risk capital charge by improvingtheir operational risk management, for instanceby introducing enhanced controls into thebusiness process. However, many openmethodological questions still remain withrespect to how this can be done in a sound andpractical manner. Banks are also allowed torecognise insurance as a risk mitigant for up to20% of their AMA capital requirements; thecondition is that their insurance contracts andproviders meet certain eligibility standards.

In the context of operational risk capitalrequirements, the study by De Fontnouvelle etal. (2004) is of considerable importance. Usingloss data from a number of internationallyactive banks, the authors find that loss data byevent types are quite similar acrossinstitutions. They also show that their resultsare consistent with economic capital numbersdisclosed by some large banks, and with theresults of studies modelling losses usingpublicly available “external” loss data.

2.3 PILLAR II – SUPERVISORY REVIEWPROCESS

Under Pillar II, banks assess their capitaladequacy on the basis of own internal riskmanagement methodology and supervisorsanalyse whether a specific bank’s capitaladequacy assessment is in line with its overallrisk profile and business strategies. Aconsistent application of supervisory practicesacross countries is of great importance to avoidany undue compliance burden and to ensure alevel playing-field. This is particularly true ifbanks belonging to the same group aim to makeuse of group-wide risk management but facedifferent expectations from their respectivenational supervisory authorities. Convergenceand cooperation in supervisory practices mayconstitute a way to alleviate this concern. The

extended role for the authority that exercisesconsolidated supervision, as provided for in theproposed EU rules, deserves particularattention in this regard (see Section 5).

Under the supervisory review process, thequestion will also be addressed whether thebank should hold additional capital againstrisks that are not, or not fully, covered in PillarI, and this may involve supervisory actionwhen this is indeed the case. An active role forsupervisory authorities will give banksincentives to continuously improve their riskmanagement models and systems. Relative tothe present situation, Pillar II requiressupervisors to apply considerably morediscretion in their assessment of capitaladequacy in individual banks.

The supervisory review process relies on fourprinciples:

1. Banks should have a process for assessingtheir overall capital adequacy in relation totheir risk profile and a strategy formaintaining capital levels.

2. Supervisors should review and evaluatebanks’ internal capital adequacyassessments and strategies, as well as theirability to monitor and ensure compliancewith regulatory capital ratios. If they are notsatisfied with the result of this process,supervisors should take appropriate action.

3. Supervisors should expect banks to operateabove the minimum regulatory capital ratiosand should have the ability to require banks tohold capital in excess of the minimum.

4. Supervisors should seek to intervene at anearly stage to prevent capital from fallingbelow the minimum levels required tosupport the risk characteristics of a bank andshould require rapid remedial action ifcapital is not maintained or restored.

The BCBS has outlined some important issuesto which both banks and supervisors should

19ECB


2 STRUCTURE

devote attention in the supervisory reviewprocess. These issues also include riskcategories which are not directly addressedunder Pillar I, such as interest rate risk in thebanking book16 and credit concentration risk.

Interest rate risk in the banking book has beengiven specific attention because it is seen byregulators as a material risk in the bankingsystem. Although the market risk amendmentof 1996 introduced interest rate risk as aseparate category, it was aimed at theexposures in the trading book. Hence, the 1996modification focused on the risk of lossesarising from, for example, bond portfolios orother categories of fixed-income instrumentsheld for trading. The BCBS acknowledged thisgap in the treatment of losses coming fromsharp interest rate changes. While it did not adda separate capital requirement, wording hasbeen included under Pillar II that requiressupervisors to identify those banks as outliersthat would experience a loss amounting to morethan 20% of their capital from a simulatedstandardised interest rate shock. Supervisoryauthorities must take measures to address thesituation of such outlier banks.

As regards the treatment of creditconcentration risk, there are twocomplementary motivations. First, severalbanking crises have been clearly linked tomaterial risk concentrations in bank portfolios.Second, the ASRF model, which is the basis forthe IRB approach, relies on two assumptionsabout a significant diversification in bankportfolios. First, the ASRF assumes that noborrower accounts for more than a very smallshare of total portfolio exposure. Second, itassumes that banks are well diversified acrossgeographical areas and industrial sectorswithin a large economy. Both assumptions maybe valid for the majority of exposures of someof the larger banks, but in the EU, the IRBapproach is likely to also be applied to smallerbanks, where concentration of exposure toindividual borrowers or certain sectors may bemore substantial. In this context, it should berecalled that the rules on large exposures in the

Codified Banking Directive (2000/12/EC)already serve to limit single-nameconcentration risk. Theoretical analysis hasshown that a ratings-based approach to settingcapital requirements needs the twoaforementioned assumptions, at leastimplicitly. Therefore, the concentrationproblem cannot be addressed merely bymodifications to the IRB risk-weights. Hence,under Pillar II, supervisors will analysepotential risk concentrations and may alsopotentially develop appropriate Pillar II capitalbuffers against such risk concentrations.

2.4 PILLAR III – MARKET DISCIPLINE

Under Pillar III, banks will be required topublish information focused on the keyparameters of their business profile, riskexposure and risk management. Suchdisclosures are seen as a precondition for theeffective working of market discipline onbanks. For banking groups, the requirementsapply to the top consolidated level of thebanking group.

Both qualitative and quantitative informationmust be disclosed. Hence, disclosure isrequired on the structure and adequacy ofcapital, and should therefore include details onthe core capital. It is envisaged that credit,market and operational risk are addressedseparately. For the disclosure of credit risk, it isalso planned to publish data on the portfoliostructure, the major types of credit exposure,the geographical and sectoral distribution andimpaired loans. In addition, information oncredit risk mitigation techniques and assetsecuritisation has to be provided. Banks will berequired to outline some details on their use ofIRB approaches, which represent a majorcomponent of the New Framework. Regardingmarket risk, banks have to summarise the key

16 The banking book is the bank portfolio which consists off inancial instruments that are not held for trading. Thetrading book is the bank portfolio that consists of f inancialinstruments that are held for short-term trading purposes, i.e.they are held intentionally for short-term resale and/or withthe intent of benef iting from short-term price movements orto lock in arbitrage prof its.


details of their internal models whereapplicable and to describe their use of stresstesting and back testing. Disclosurerequirements further cover the management ofand the compliance with requirements onoperational risk. Finally, the New Frameworkrequires that information on equity holdingsand interest rate risk in the banking book to bepublished.

3 THE NEW FRAMEWORK IN THE EU

3.1 LEGAL AND INSTITUTIONAL SETTING

One month after the BCBS published itsdocument on the New Framework, theEuropean Commission released its ownproposals on new capital requirements forbanks and investment firms in the EU.17 Theproposals reflected to a large extent Basel II,but were at the same time tailored to thespecific features of the EU market. TheCommission’s initiative is part of the widerFinancial Services Action Plan (FSAP)launched in 1999. The FSAP outlined a numberof policy objectives and measures to improvethe Single Market for financial services. One ofits objectives was to ensure financial stabilityby keeping pace with state-of-the-artprudential rules and supervision, in particularin the area of banks’ solvency requirements.The Single Market is an important factor thatexplains why in some cases the proposed EUrules go further than the Basel rules.

Thereafter, the Commission’s proposals havebeen subject to scrutiny by the EuropeanParliament and the Council and to tripartitenegotiations with the Commission. Thesenegotiations resulted in a package of almost600 amendments, subject to which both theParliament and Council approved the proposalsin September and October 2005. In this way,they cleared the way for the new capitalrequirements to enter into force in the EU asscheduled by the BCBS.

Although there are a large number ofamendments, they leave the content and thrustof the Commission’s proposals intact. Theamendments introduce the trading book reviewand double default treatment as released by theBCBS in July 2005, reduce the number ofnational discretions, make some minortechnical corrections and take account ofseveral specific concerns relating to MemberStates’ national markets. Finally, theamendments also provide a preliminarysolution to the long debated issue ofinstitutional balance in the “comitology”procedure as part of the Lamfalussy approach,which is discussed in more detail below.

An important distinguishing feature of theimplementation of the Basel rules in theEuropean Union is the legal nature of thecapital adequacy framework. While the BaselII Framework takes the form of an accord oragreement amongst national bankingsupervisors represented in the BCBS – thusimplementation remains in principle voluntary– in the EU, the framework is legislative andbinding in all EU Member States. The Basel IIFramework will be transposed into EUlegislation by means of the CapitalRequirements Directive (CRD). Technically,this has been done via a recasting of theexisting Codified Banking Directive or CBD(2000/12/EC) and the Capital AdequacyDirective or CAD (93/6/EEC). The re-castingtechnique, established by the interinstitutionalagreement 2002/C77/01 of 28 November 2002,allows the incorporation in current legislativetexts of both amended and unchangedprovisions.

Another significant element is the regulatoryand supervisory setting that is now in placein the EU as a result of the “Lamfalussy”approach (see Box 2). This approach wasinitially applied to the securities sector18 but

17 European Commission Services (2004).18 The Lamfalussy approach was set out by the Committee of

Wise Men on the Regulation of European Securities Markets,chaired by Baron Alexandre Lamfalussy, in its “FinalReport” dated 15 February 2001.

21ECB


3 THE NEWFRAMEWORK

IN THE EUwas later extended to all financial sectors,including banking. It was introduced with theaim of improving the speed and flexibility of

the rulemaking process, so that regulation canbetter keep up with developments in financialmarkets.

Box 2

THE LAMFALUSSY APPROACH

Level 1

Proposal for Community legislation advanced by the European Commission and adopted underthe co-decision procedure by the Council and the European Parliament. The legislation takesthe form of directives or regulations. It should be limited to framework principles and definethe powers for the Commission to implement the necessary technical rules.

Level 2

The European Commission enacts legislation containing the technical details for theframework principles approved at Level 1. This requires the intervention of a regulatorycommittee under the “comitology procedure”1. These regulatory committees are chaired by theCommission and composed of high-level representatives from Member States. The ECB hasobserver status in the banking, securities and financial conglomerates committees.

The “Level 2” regulatory committeesBanking European Banking Committee (EBC)Securities and investment funds European Securities Committee (ESC)Insurance and pension funds European Insurance and Occupational Pensions

Committee (EIOPC)Financial conglomerates European Financial Conglomerates Committee (EFCC)

Level 3

Level 3 committees are entrusted with the task of facilitating the day-to-day implementation ofCommunity law with the goal of converging both supervisory practices and the application ofCommunity legislation, and enhancing supervisory cooperation. Guidelines, interpretativerecommendations, common standards or best practices may be issued, but these are not legallybinding and implementation remains voluntary. Level 3 committees also assist theCommission in drafting the more technical provisions of the legislation enacted at Level 2. Thesupervisory committees are composed of high-level representatives from the competentnational supervisory authorities.

The “Level 3” supervisory committeesBanking Committee of European Banking Supervisors (CEBS)Securities and investment funds Committee of European Securities Regulators (CESR)Insurance and pension funds Committee of European Insurance and Occupational

Pensions Supervisors (CEIOPS)Financial conglomerates At present, there is no Level 3 committee

1 The procedure whereby the Commission is assisted by a Committee comprising representatives from Member States in theadoption of implementing measures for Community legislation. Council Decision 1999/468/EC of 28 June 1999 laying down theprocedures for the exercise of implementing powers conferred on the Commission specif ies the types of comitology proceduresgoverning the adoption of implementing measures.


Level 4

The European Commission is responsible for ensuring that Member States’ national lawcomplies with Community law and, if needed, to take enforcement action. Legal action againstMember States can be taken before the European Court of Justice. Strengthening enforcementis underpinned by enhanced cooperation between Member States, the regulatory bodies and theprivate sector.

The increased flexibility of the legislativeprocess envisaged by the Lamfalussy approachis reflected in the CRD transposing thecapital requirements into EU law. Whereasamendments to the main principles still have tobe made via co-decision, amendments to themore technical provisions can be introducedvia the more flexible and faster “comitology”procedure. However, the CRD is not apure “Lamfalussy directive” as it combinesboth framework principles and technicalimplementation rules in the same legalinstrument.

The issue of “comitology” and the institutionalbalance between Parliament, Commission andCouncil was long debated given that the currentprocedure does not grant the Parliament theright to “call back” the “comitology” powers ofthe Commission, although such a right wasintended to be introduced as part of theConstitutional Treaty. A last minutecompromise between Commission, Counciland Parliament as part of the above-mentionedamendments means that the current comitologysystem – which provides no formal “call-back”right for Parliament – can be used to implementand update the Directives for a maximum oftwo years or until 1 April 2008 at the latest.After this period, aforementioned powerscan be renewed only with the agreement of thethree institutions. This timeframe should allowfor reflection on a possible recalibration asscheduled by the BCBS for Spring 2006.

The implementation of the New Frameworkrepresented in principle a “window ofopportunity” to structure the new rules alongthe lines of what had been introduced for thesecurities sector. This would have entailed a

clear distinction between framework principles(Level 1 acts) and technical implementationrules (Level 2 acts), with the adoption ofregulations for the latter, wheneverappropriate. In general, recourse to regulationswould reinforce convergent implementationacross the EU, given that they are directlyapplicable in all Member States, as opposed todirectives, which need to be transposed intonational law. This in turn would facilitatecompliance by cross-border groups andcontribute to promoting a level playing-fieldand further integration. This opportunity wasnot exploited and hence, the current legalframework should be viewed as a first step in alonger term process where the ultimate goalwould be to arrive at a uniform and directlyapplicable set of European rules for financialinstitutions.

Regulatory convergence and a consistentimplementation of rules across Member Statesare important to ensure a level playing-field inthe Single Market. However, the CRD stillincludes several national discretions, meaningthat the scope for potential divergent nationalimplementation is considerable. In addition, theuse of terms that are not clearly defined providesnational authorities with considerable leeway,which may also result in significantly differentinterpretations. In this respect, the work by theLevel 3 banking committee – the CEBS – iscrucial. The CEBS has given priority to theidentification and reduction of national options.The progress achieved thus far by the CEBS isreflected in the final CRD text and must beacknowledged, while the pursuit of more workin this field should be strongly encouraged. Asthe EU financial systems become increasinglymore integrated, some of the remaining national

23ECB


3 THE NEWFRAMEWORK

IN THE EUdiscretions may become obsolete over time,necessitating further revisions.

Furthermore, in its work on supervisoryconvergence regarding the new capital rules,the CEBS is focusing on a number of importantareas.19 Foremost is the development of acommon reporting framework for the newsolvency ratio, which is especially importantfor cross-border banking groups that arepresently confronted with a variety of nationalreporting schemes. The CEBS is also workingon guidance for the supervisory reviewprocess, which comprises the bank’s internalcapital adequacy process and the supervisoryreview and evaluation process. As regards thestandardised approach, a common approach isunderway for the recognition of external creditassessment institutions (ECAIs), as well as fordeveloping the criteria to transform (“map”)ratings into risk weights. For the advancedcalculation methods, a common approach tovalidation principles is being defined. Anotherimportant task is to foster cooperation betweenhome and host authorities. Finally, the CRDnot only provides for disclosure by banks butalso by supervisors, for which the CEBS isestablishing a common framework.

3.2 SPECIFIC FEATURES OF EUIMPLEMENTATION

The implementation of the new capital rules inEurope will not fully mirror the Basel IIFramework; rather the EU rules have beenadapted in order to reflect the specific SingleMarket context, which encompasses suchfeatures as the single banking licence, homecountry control and minimum harmonisation ofprudential requirements. The main specificitiesof the European setting that have been taken intoaccount when developing the EU rules arepresented below.

(a) Scope of applicationOne of the main diverging aspects concernsthe scope of application of the rules. Just likeBasel I, the Basel II Framework is envisagedto apply only to internationally active banks. The

European implementation of the capitalrequirements, by contrast, will in principle applyto all banks and investment firms,20 independentof their size or the geographical scope of theiractivity. Financial institutions dealing in thesame activity or providing similar services willtherefore be subject to the same capitalrequirements, thus ensuring a level playing-fieldwithin the EU. In this respect, it should berecalled that following the CAD, banks andinvestment firms in the EU are already subject tothe same capital rules. This similar treatment isrooted in the fact that the CBD allows theuniversal banking model, i.e. the combination ofbanking and securities activities in the samelegal entity or within the same financial group.

Another difference in the scope of applicationis the level at which the new rules apply. TheBasel II rules apply to internationally activebanks at every layer of the banking group ona (sub)consolidated basis. No general explicitcapital requirements are formulated forindividual banks, but when an entity of thegroup itself qualifies as an internationallyactive bank, the Basel II capital requirementsmust also be met by this entity. The EU rules,by contrast, apply in principle both on aconsolidated and an individual (solo) basis.However, subject to certain conditions, theCRD contains the possibility to waive the solorequirements on (only) domestic subsidiaries.The conditions for this waiver are aimed atensuring that the parent guarantees thecommitments of the subsidiary that has beenexempted from the solo capital requirements.

(b) Range of available approachesThe full spectrum of approaches for thecalculation of capital requirements asenvisaged by the Basel II Framework, from thesimple methods to those based on internal

19 This is reflected in the different Consultation Papers (CP)released by the CEBS. See in particular the CP3 (supervisoryreview process), CP4 (solvency reporting framework), CP5(supervisory disclosure), CP7 (recognition of ECAIs), CP9(supervisory cooperation) and CP10 (validation of advancedcalculation methods).

20 Investment firms are f irms authorised under the Markets inFinancial Instruments Directive (MiFID), 2004/39/EC.


models, will be available in the EU. Theapplication of the New Framework in the EU toall banks and investment firms, irrespective oftheir level of complexity or sophistication,indeed requires the full spectrum of approachesto be available to all institutions. In thiscontext, it should be noted that, in the US, theapplication of Basel II will only encompass themost advanced methods (see Box 3).

The comprehensive scope of application in theEU was also the basis for the development ofrules on “permanent partial use”. In the contextof credit risk, partial use refers to thepossibility of using the simpler standardisedapproach for certain exposure classes, whileapplying the more sophisticated IRBapproaches for the remaining classes. UnderBasel II, a bank using the IRB for any exposuremay only permanently apply the standardisedapproach for so-called non-material exposureclasses and business lines. All other exposureshave to be treated under the IRB within a timeframe agreed with the respective supervisorupfront (“roll out” plan). This rule aims to limitthe potential for “cherry-picking” betweenapproaches with different risk sensitivities.

For small EU banks, it may be particularly hard,or even impossible, to roll out the IRB approach

to certain exposure classes where they havea very limited number of counterparties.Nevertheless, the exposure to thesecounterparties may still be material in relation tothe bank’s overall exposures. For this reason,there is the possibility for exposures to banks,investment firms, sovereigns and certain otherpublic sector bodies to qualify for a permanentpartial use. Such exposures could then remain onthe standardised approach, independent of theirmateriality, while the bank otherwise uses theIRB approach.

The broad application in the EU of the capitalrequirements framework to institutions thatrange from very sophisticated internationallyactive banks to less complex investment firms,together with the potentially burdensomeoverall impact of the new rules on certaininvestment firms, requires that the prudentialstandards be adapted. Therefore, investmentfirms falling within certain categories may beexempted by the competent authorities fromcalculating capital requirements foroperational risk. Where this exemption is used,the currently used “expenditure-based” capitalcharge, that is otherwise abolished, will beretained. Under this approach, investmentfirms are required to hold minimum capital inrelation to their overhead costs.

Box 3

IMPLEMENTATION OF BASEL II IN THE US AND ITS IMPLICATIONS FOR EU BANKS

The US authorities decided to confine the spectrum of approaches available in the Basel IIFramework.1 Large internationally active banks (consolidated assets over USD 250 billion andforeign exposure of at least USD 10 billion), including subsidiaries of foreign banks, will berequired to use only the advanced methodologies to calculate capital requirements. Thesemethods include the advanced IRB approach for credit risk and the AMA for operational risk.Approximately ten banks fall into this category. In addition, other banks that meet therequirements for the use of the advanced approaches will be allowed to opt into Basel II. Thiswill be the case for approximately another ten banks. Overall, the banks which are expectedto apply Basel II are those that are active in cross-border banking. They account forapproximately 99% of the foreign assets held by the top fifty US banking organisations and forabout two-thirds of the assets of US banks. The remaining banks (approximately 6,500) in the

1 Federal Reserve Board et al. (2003).

25ECB


3 THE NEWFRAMEWORK

IN THE EUUS and subsidiaries of foreign banks that donot meet the criteria to adopt Basel II willremain under the current Basel I rules.2

The implementation of Basel II in the US mayhave implications for EU banks operating in theUS, especially those with a significant presence(see chart for some examples). These banksmay be required to operate under rules thatdiffer from those applying to the rest of thebanking group, in particular the parentcompany in the respective home country. Thiscan be costly and inefficient and potentiallyraises negative competitive effects forEuropean banks. A US subsidiary of a foreignbank has to comply with US rules, which willentail the need for these subsidiaries to eithermeet the prerequisites in order to apply theadvanced approaches of Basel II or to remain under the current Basel I based rules. Hence,subsidiaries which would have targeted the foundation IRB will continue to apply the Basel Irules, irrespective of whether or not the rest of the banking group applies these approaches on aconsolidated basis.

The US authorities have publicly conveyed that they would be prepared to explore thepossibility of allowing foreign subsidiaries to use conservative estimates for a transitionperiod.3 However, further work to develop concrete proposals still needs to be pursued. In thiscontext, the work of the BCBS’ Accord Implementation Group (AIG) in maintaining a levelplaying-field across countries and in achieving an acceptable level of consistency in theimplementation of Basel II is of the utmost importance.

More recently, the results of the fourth quantitative impact study (QIS4) have shown a muchhigher than expected capital reduction (of around 15% on aggregate and by more than 26% inmore than half of the participating banks) vis-à-vis the current level of the 26 large US financialinstitutions replying to the QIS4, and also a very wide dispersion across banks (ranging from adecrease of 47% to an increase of 56%).4

These results have prompted the US authorities to delay the scheduled publication of the Notice ofProposed Rulemaking (NPR). They now plan to make the US Basel II proposal available in thefirst quarter of 2006 and to introduce additional prudential safeguards to address the concernsresulting from the QIS4. In particular, a one-year delay in the implementation will be proposed.Further, the capital floors tied to the current Basel I rules will extend to 2011, after which they willbe reconsidered on an institution-by-institution basis.5 Given the significant implications that thedecisions by the US authorities may have in maintaining a level playing-field internationally, thisis an issue that will have to be closely monitored in the EU.

2 These rules will be revised somewhat and made more risk-sensitive. This regime is sometimes called “Basel IA”.3 Ferguson (2003).4 Schmidt Bies (2005) and Powell (2005).5 Federal Reserve Board et al. (2005).

European banking groups with a s igni f icantpresence in the US

(US assets as % of total assets)

Source: ECB calculations on the basis of the 2004 annualreports of the respective banking groups.Note: Figures refer to North America for ABN Amro, BNPParibas, Deutsche Bank and HSBC; f igures refer to loans/commitments for ABN Amro and BNP Paribas.

0

5

10

15

20

25

30

35

0

5

10

15

20

25

30

35

HSBC ABNAmro

DeutscheBank

Royal Bank ofScotland

BarclaysBank

BNPParibas


(c) External credit assessment institutionsExternal credit assessment institutions(ECAIs) play an important role in the NewFramework as their assessments (ratings) willbe used for the calculation of the capital chargefor credit risk under the standardised approach.In order to obtain regulatory recognition forcapital purposes, ECAIs need to comply withminimum requirements.21 A high degree ofconsistency of Member States’ practicesregarding the recognition of ECAIs will beindispensable. In this way, comparability ofratings and a level playing-field for ECAIs canbe ensured and the potential risks forregulatory arbitrage reduced. Moreover,supervisory co-operation will be crucial toreduce the compliance costs for those ECAIsthat seek recognition in more than one MemberState.

The possibility of applying the principle ofmutual recognition to ECAIs should in thatrespect be very helpful. However, whilst theCRD only includes an option for mutualrecognition, one could argue that this shouldbe the general rule for ECAIs within the EU. Ifa competent authority in a Member Stateassesses an ECAI as complying with theeligibility criteria, this evaluation could thenbe used automatically by the competentauthorities in other Member States withoutconducting any further assessment. The samerationale applies when a competent authorityhas developed a mapping for an ECAI’s ratingson supervisory risk weights. Competentauthorities in other Member States could thenaccept and use this mapping withoutdetermining their own process.

The recognition of ECAIs for solvencypurposes is expected to influence theavailability of external ratings. This, in turn,will have an impact on the assessment of theeligibility of collateral for the credit operationsof the Eurosystem, which comprises the ECBand the twelve national central banks of theeuro area, given the need for marketable debtinstruments to have a high credit standing to beincluded in the list of collateral accepted.

Although the Eurosystem uses its ownassessment criteria, the eligibility criteria andthe process of recognition for supervisorypurposes will influence the quality of theratings and the level playing-field in theratings’ market.

In this context, it is important to achieve aprudent and fair approach to supervisors’assessment of ECAIs, especially to avoidcreating entry barriers for new market players.For the latter, the focus should therefore be onthe evaluation of the robustness and soundnessof assessment methodologies rather than on,for example, market acceptance or “trackrecord”.

(d) Supervisory disclosureThe EU capital framework introduces specificdisclosure requirements for the Member States’competent authorities.22 These requirements,which should be distinguished from the Pillar IIIrequirements that target banks’ disclosures,were introduced with the aim of enhancingsupervisory convergence and transparency. Thisinnovative aspect, not mirrored in the Basel IIFramework, is expected to contribute to themaintenance of a level playing-field andthe fostering of financial integration. Thedisclosures encompass a minimum set ofrequirements that range from the publication ofthe legal texts and rules to the exercise of certaindiscretionary measures or the choice of certainoptions available in the new rules. With regard tothe exercise of effective supervision, both thegeneral criteria and the methodologies usedduring examinations of the supervised entitiescovered under the supervisory review processshould be made publicly available. Finally,aggregate statistical data on some keyimplementation aspects have to be published aswell.

(e) Private equity and venture capitalIn striving to improve the competitive positionof the European economy, concerns were raisedthat the new capital rules would have a negative

21 On this, see also the CP7 of the CEBS.22 On this, see also the CP5 of the CEBS.

27ECB


3 THE NEWFRAMEWORK

IN THE EUimpact on the private equity and venture capitalindustry, which is seen as a key driver foreconomic growth. Banks in the EU play a crucialrole as the main source for this kind of financing.Banks provide on average approximately aquarter of the total funds raised, although thereare important differences across Member States(see Chart 6).

The concern that the increase in risk weightsfor equity investments may lead to a significantreduction in the bank financing of investmentsin private equity and venture capital triggeredadditional research. The impact study carriedout in the EU on the New Framework raised theissue that for investments in venture capitalthrough a diversified venture capital fund, therisk weightings under the IRB approachproposed by the European Commission in itsThird Consultative Proposals (CP3) wererather high (PricewaterhouseCoopers, 2004).This report recommended that the treatment ofequity exposures should consider the profile ofthe investments, taking into account the degreeof diversification. This recommendation wasincorporated in the CRD. For investments inprivate equity firms that constitute sufficientlydiversified portfolios, the risk weightsassigned under the IRB rules are now lowerthan under the Basel II Framework. Forexample, under the “simple risk weightapproach” for equity exposures, the capitalcharge under the EU rules may in certaininstances be approximately 24% lower thanunder Basel II.

(f) Financing of small and medium-sizedenterprisesSmall and medium-sized enterprises (SMEs)play an important role in fostering economicgrowth and entrepreneurship, and often act assub-contractors to larger firms. They userelatively more bank financing than largerenterprises and many of them have credit lineswith only one bank. SMEs represent more than99% of all EU enterprises by number and about70% of total employment (EuropeanCommission, 2003b). These figures arecomparable to the situation in the US and

Chart 6 Share of banks in new funds raisedfor private equity

(percentages; end 2004)

Source: ECB calculations on the basis of European PrivateEquity and Venture Capital Association (2005).1) LU is missing.

Japan, although in the US, SMEs account for amuch lower share of employment.

In the context of the new capital rules, therehave been concerns that quantitative ratingmethods such as the IRB approach might putmore emphasis on financial ratios than onqualitative factors such as the entrepreneur’sability, product ideas or business plans. Inaddition, start-up companies might be at adisadvantage as they frequently lack a ratinghistory. This shift in credit assessment bybanks could therefore raise the cost of credit forSMEs and restrict the availability of theirfinancing. Given the importance of these firmsfor the EU economy, this might ultimately havea negative effect on economic growth,employment and innovation. A number ofauthors have studied this issue in detail, e.g.Fabi et al. (2004) or Dietsch and Petey (2004).

In its finalisation of the new capital rules, theBCBS made several adjustments to alleviatethese concerns. As explained in greater detailin Section 2.2.1, under certain conditions,exposures to small businesses can under thestandardised and the IRB approaches qualifyfor the preferential treatment of retailexposures. In addition, the IRB approach alsoprovides for lower risk weights for SMEs thatcontinue to fall under the corporate treatment.

0

10

20

30

40

50

60

70

80

90

0

10

20

30

40

50

60

70

80

90

AT PT BE ES IT FR NL EU15 1) DE DK UK IE FI SE GR


The potentially negative effect on SMEs wasa major reason behind the 2002 EuropeanCouncil request to the European Commissionto present a report on the consequences ofthe Basel II Framework for all sectors ofthe European economy, with particularattention to SMEs. In this “Barcelonareport”, it was concluded that the new capitalrules would present an overall beneficialoutcome for the vast majority of SMEs(PricewaterhouseCoopers, 2004). Therefore,the proposed EU rules for the treatment ofSMEs do not deviate from those adopted bythe BCBS. Overall, the discussion on thepotentially adverse effects of Basel II on smalland medium-sized enterprises has now lost itsearlier intensity as the most significantconcerns have been addressed through specificmodifications to the Framework.

(g) ProcyclicalityProcyclicality in the context of Basel II refersto the exacerbation of the economic cycle as aconsequence of the new capital requirements.In a period of strong economic growth, banksmay make more financing available at moregenerous conditions, possibly resulting in anasset boom, while in an economic downturn thereverse could take place, resulting in a creditcrunch. Procyclicality is not a typical Europeanconcern, but the EU rules provide for a specificmonitoring arrangement, as well as a role forthe ECB. The topic of procyclicality isdiscussed in greater detail in Section 4.

(h) The consolidating supervisorThe increase of cross-border banking in the EUand the emergence of pan-Europeaninstitutions have increased the need tostrengthen cooperation and convergence insupervisory practices and requirements. Tomeet this need, the EU capital framework hasenhanced the role of the “consolidatingsupervisor”, which is the authority responsiblefor the supervision of a banking group on aconsolidated basis.

This is an important innovative aspect of theEU capital framework that should foster

supervisory convergence, financial integrationand financial stability. It has also beenidentified as crucial by major cross-borderinstitutions, which have been vociferous inpublicly stating that the current provisions aretoo modest and that they would like to see aneven more prominent role for the consolidatingsupervisor. This issue will be more extensivelydiscussed in Section 5.

(i) Real estate lendingThe rules regarding the treatment of residentialand commercial real estate lending have beenadapted to the EU context. The aim is to have aconsistent treatment for both types of lendingunder the different approaches. Accordingly,the same eligibility criteria for the recognitionof real estate collateral will be used both underthe standardised and the foundation IRBapproach. In addition, minimum requirementsto ensure reliable collateral management andprudent valuation of mortgage property areapplicable to all approaches. This issue will beaddressed more extensively in Section 6.

(j) Covered bondsCovered bonds have a key role in Europeancapital markets and in the funding of mortgageand public sector lending. They encompassdebt securities issued by EU banks that aresubject to particular collateral arrangementsunder public regulation and supervision. As yetanother distinct feature of the Europeanlandscape, they have a specific treatment in thenew EU rules, whereas they are not addressedunder Basel II. This topic will be addressed ingreater detail in Section 7.

29ECB


4 PROCYCLICALITY

PART II – ISSUES OF SPECIFIC RELEVANCE INTHE EU CONTEXT

4 PROCYCLICALITY

4.1 DEFINITION

During the development of the NewFramework many observers pointed outthat the new rules might lead to increasedprocyclicality in the financial system. TheECB has also repeatedly commented on thisissue.23 Procyclicality refers to the empiricalobservation that banks’ loan business tends tofollow the same cyclical pattern as that of thereal economy. Hence, loans typically showstrong growth in an economic upturn and slowgrowth or even contraction in an economicdownturn. In general, banks’ lending activitiesshow procyclical characteristics regardlessof the design of capital requirements. Thisarises, for instance, on account of the existenceof asymmetric information or marketimperfections. In the context of Basel II, thediscussion focuses on the potential foradditional procyclicality as a result of the newcapital rules.

Basel II may give rise to procyclical effectsowing to the fact that the three main inputparameters of the IRB approach are themselves– albeit to different degrees – influenced bycyclical movements. As regards the role ofPDs, an economic downturn may lead to anincrease in banks’ estimates of borrower PDs ifbanks use a short-term assessment horizon.Such a “point-in-time” rating changes due tovariation in the credit quality over the course ofthe business cycle. By contrast, a “through-the-cycle” rating requires a longer-term analysis ofborrowers’ default risk on the basis of ascenario which takes into account the effect of,for example, an economic slowdown. In thecase of the LGD, the losses that occur in theevent of a default may increase in an economicdownturn, because there is some evidence thatrecoveries from defaulted debt are lower in arecession. Finally, the exposure at default of aloan may also increase as borrowers make more

use of their loan commitment limits during adownturn.

These arguments indicate that, in an economicdownturn, the higher risk sensitivity of banks’rating systems may lead to rising capitalrequirements. In the event of a pronouncedrecession, banks’ capital ratios may even fallclose to the 8% minimum level. If banks areunable to adjust their capital level directly,they may have to reduce their lending to a levelbelow that caused by the weaker demand in arecessionary environment. This potentiallyreduced availability of new bank lending couldthen exacerbate the downturn.

4.2 EMPIRICAL EVIDENCE

As regards research on procyclicality, theanalysis has so far mainly concentrated on thebehaviour of bank lending during pastrecessions, such as the US slowdown in the1990s. Lowe (2002) and Allen and Saunders(2004) survey the literature on procyclicality.Key empirical studies on procyclicality areKashyap and Stein (2003), Catarineu-Rabell etal. (2003), and Gordy and Howells (2004).

Kashyap and Stein (2003) documentcomprehensive evidence for procyclical effectsin capital requirements during the 1998-2002economic downturn. In their analysis,particular attention is given to methodologicalproblems. The paper applies three alternativecredit risk approaches, namely Standard andPoor’s credit ratings, Moody’s KMV model24

and a major international bank’s internal creditrisk model. All three methods lead toeconomically significant increases in capitalrequirements due to the time-varying riskweights. Rating agency estimates for creditrisk, which should in principle take a longerhorizon, lead to increases of between 30% and45%, whereas more “point-in-time” models

23 ECB (2005b).24 Moody’s KMV default risk forecast is an established

indicator to calculate a f irm’s credit risk on the basis of itsstock price, balance sheet information and an option pricingmodel.


produce capital increases of between 70% and90%. The authors also discuss possible policymeasures to mitigate these effects.

Catarineu-Rabell et al. (2003) use both atheoretical general equilibrium model of thebanking system and an empirical analysis toevaluate how the choice of a particular creditrisk assessment system affects the likelihood ofsharply increasing capital requirements inrecessions. The authors demonstrate that a“point-in-time” approach could substantiallyincrease procyclicality. The results alsoindicate that not all banks might use “throughthe cycle” rating systems, therefore raisingimportant policy conclusions for theimplementation of the Framework. Hence, theauthors argue that banks should receiveincentives to choose more stable credit riskassessment systems. According to the authors,few banks have sufficient capital to cover theincreased capital requirement in a recession.

Gordy and Howells (2004) undertake acomprehensive simulation to investigate theimpact of three measures to mitigateprocyclicality. They point out that most of theliterature focuses on procyclicality in Pillar I andthe different ways to smooth it. According to theauthors, smoothing necessarily affects risksensitivity. Hence, investors will not be able toinfer changes in portfolio risk from changes incapital ratios. In order to comply with Pillar III,they suggest an alternative solution whichconsists of dampening the output instead of theinput. Here, the suggestion focuses on the designof an autoregressive rule that only allows for apartial adjustment of regulatory capital from oneperiod to the other. Therefore, sudden changes ofcapital requirements would be smoothed andprocyclicality would be alleviated. This ruleshould be made public to investors, who wouldthen be able to infer the level of currentunsmoothed capital requirements.

Marcelo and Scheicher (2005) use Moody’sKMV model and a hypothetical portfolio ofloans to 6,000 large, non-financial EU firmsto estimate the foundation IRB capital

requirements. The paper aims to replicate, asclosely as possible, the implementation of afoundation IRB approach using both a “point-in-time” approach and a “through-the-cycle”perspective. Two main points emerge in theresults (see Chart 7). First, banks’ regulatorycapital requirements under the IRB approachremain below 8% for the hypothetical EU15corporate portfolio, the median capitalrequirement being 4.9% and the standarddeviation 0.4%. In this context, it is importantto bear in mind that the estimated decline inregulatory minimum capital requirementsrelative to Basel I does not automatically meana decline in the overall capital held by banks tocover the risks. Second, the “through-the-cycle” approach indeed overall producesrelatively low capital volatility.

Other studies arrive at similar conclusionsabout the potential weight of procyclicality.25

Altman et al. (2005) undertake an extensivesimulation exercise with annual ratingstransition matrices over the period 1981-2000.Allowing for positive correlation betweenLGDs and PDs, the cyclicality of capitalcharges is increased. Hence, if banks can alsouse internal estimates of LGD, their sensitivityto economic cycles might rise. According to

Chart 7 Difference in sensitivity of capitalrequirements for exposures on EU firms in a“point-in-time” and “through-the-cycle” approach(percentages)

Source: Marcelo and Scheicher (2005).

25 See also Carling et al. (2002), Segoviano and Lowe (2002),Illing and Paulin (2004), Amato and Furf ine (2004),Goodhart and Segoviano (2004), Roesch (2005) or Hoffmann(2005).

012345678

012345678

1992 1994 1996 1998 2000 2002 2004

“through-the-cycle” capital ratio“point-in-time” capital ratio

31ECB


4 PROCYCLICALITY

Bangia et al. (2002), US banks may requirefrom 25% to 30% more capital in a recessionthan in a growth period. Finally, Berger andUdell (2003) argue that problems in loanperformance are partly caused by changes inbank behaviour during economic expansions.In these periods, higher risk loans are providedand the problems then materialise during thefollowing downturn. The cyclical variation inloan officers’ credit standards would be a keyfactor in explaining this phenomenon. Theauthors test the empirical implications of thishypothesis using a US sample over the period1980-2000 and find empirical support for it.

4.3 MITIGATING MEASURES

In order to address procyclicality concerns, theBasel II Framework includes a number ofspecific measures, some of which were alreadyincluded in the CP3.26 In particular, underPillar II, banks are asked to evaluate their riskbearing capacity with respect to scenarioswhich would particularly affect their creditexposures. In addition, banks using the IRBapproach are required to implement a morespecific credit risk “stress test” to evaluate howcertain specific events affect their capitalrequirements. For this purpose, the banks’analysis should take into consideration at leastthe impact of mild economic downturns such astwo quarters of zero GDP growth. In general,the development of reliable stress tests forcredit risk under the IRB approach is still thefocus of analytical work. It is thereforeadvisable that both banks and supervisorsdevote further efforts to developing suitablemethodologies.

Given the potentially cyclical behaviour ofminimum capital requirements, banks couldpre-emptively set aside sufficient capital in theform of buffers over and above the regulatoryminimum. Such buffers, which are encouragedunder Basel II, can diminish the potentiallynegative macroeconomic effects of a downturn.In some EU countries (e.g. Spain), the practiceof “dynamic provisioning” is activelyencouraged by the supervisor. Under this

approach, the possible loss over the whole lifeof the loan is taken into account in theprovisioning process, thereby giving dueconsideration to the loan’s full risk profile overthe whole business cycle.

Finally, banks should in their assessment of theborrowers’ credit risk under the IRB approachdraw on a longer time horizon (“through thecycle” approach). This is particularly relevantin the case of banks lending to firms incyclically sensitive sectors, which should bemore conservative in their assessment of thedefault risk in periods of upturn. Banks are alsorequired to use for their own LGDs and EADsvalues that have been estimated under thescenario of an economic downturn.

Overall, it can be concluded that procyclicalityconcerns were extensively addressed when theNew Framework was being developed. Ascompared with earlier drafts of the Framework,the scope for such effects has been clearlyreduced. The potential tools to reduce thecreation of additional procyclicality in thefinancial system, such as forward-looking creditrisk assessments, stress tests and dynamicprovisioning, should in this context be seen ascomplementary measures and exploited to themaximum possible extent. EU supervisoryauthorities have a common interest in consideringappropriate ways to reduce the risk of increasedprocyclicality, since macroeconomic conditionsare gradually becoming more closely interwoven,particularly in the euro area.

4.4 MONITORING IN THE EU

From the discussion above, it should be clear thatprocyclicality resulting from the new capitalrules is not a typical EU concern, but relates tothe general design of the Framework. The above-mentioned “Barcelona report”, which is animpact study for the EU of the new capital rules,concluded that the New Framework is unlikely to

26 The CP3 reduced the slope of the risk weight curve, whichexpresses the relationship between the PD and the riskweight.


significantly amplify the economic cycle.However, it also said that the EuropeanCommission and national authorities shouldkeep this issue under review, during the initialtransition stage and in the medium term, andstand ready to amend the Framework should thisassessment prove too optimistic.

To this end, the CRD provides for specificmonitoring arrangements to tackle procyclicalityonce the New Framework is fully implemented.The European Commission, together with theMember States, is required to periodicallymonitor the possible effects of the CRD on theeconomic cycle. In this process, a contributionfrom the ECB has to be taken into account. TheEuropean Commission has then to report on abiennial basis to the European Parliament and theCouncil and, if deemed necessary, proposemeasures to address the concerns.

5 HOME-HOST ISSUES AND THECONSOLIDATING SUPERVISOR

5.1 INTERNATIONAL AND EUROPEANCONTEXT

Banks typically perform their foreign activitiesvia branches and subsidiaries,27 which canresult in complicated group structures. TheNew Framework will increase the need for

cooperation between the supervisors of thesedifferent group entities because the capitalrules apply at each level of the banking group.The implementation of the new rules maytherefore require the group to obtain approvalfor its use of certain approaches from hostsupervisors on a solo or sub-consolidated basis,as well as from the home supervisor for the(top) consolidated level.28 By indicating thenumber of different countries where a selectedset of EU banking groups has subsidiaries,Table 2 provides an indication of the potentialcomplexity involved in the cooperationbetween home and host authorities.

In order to assist national authorities in thiscomplex issue, the BCBS has developedguidance on how the new capital rules should

27 In addition, banks can also directly provide services on across-border basis; this third possibility is not discussed anyfurther in the text. Branches, in contrast to subsidiaries, arenot separate legal entities; in the EU they benef it from the“single passport” and “home country” control.

28 The supervisor of a foreign establishment of a bankinggroup, be it a branch or subsidiary, will be referred to in thetext as the host supervisor. In the event that it is necessary tomake clear the type of local establishment, the terms hostbranch supervisor or host subsidiary supervisor may be used.This terminology is not consistent with that used in the EUdirectives, where the term host supervisor is reserved for thesupervisor of a foreign branch, but which is clearer toexplain the practical arrangements for the supervision of abanking group.

Totalconsolidated

Group name assets EU 15 EU 25 World

Deutsche Bank 804 8 10 18BNP Paribas 783 6 9 20Royal Bank of Scotland 638 2 2 5ABN Amro 560 5 6 17Fortis 425 5 6 10Dexia 349 7 8 11Banco Santander Central Hispano 346 5 5 18Nordea 262 4 5 6Banca Intesa 259 4 6 8Danske Bank 243 2 3 3

Table 2 Number of countries where selected EU banking groups have a banking subsidiary

(EUR billlions; end 2003)

Source: Bankscope.Note: Because of limitations of the database, the table might not be complete.

33ECB


5 HOME-HOSTISSUES AND THECONSOLIDATING

SUPERVISOR

be implemented on a cross-border basis.29

These state that the legal responsibilities ofnational supervisors remain unaffected by theNew Framework, although the home supervisorhas a key role in overseeing the implementationprocess. The principles further stress the needfor enhanced cooperation and due recognitionof the host supervisors’ concerns. Finally, theBCBS is also sensitive to the issue of increasedcompliance costs as supervisors have to avoidredundant and uncoordinated approvals.

These considerations are even more valid in theEU, where the Single Market and the singlecurrency have given a significant impetus tocross-border banking. Here, the existingarrangements between home-host authoritieshave come under increased pressure and willeven more so under the new capital rules.Moreover, the rules also introduce newrequirements in terms of risk management andreporting systems, which result in highercompliance costs for banks. To alleviate someof these concerns, the CRD streamlines theinteraction of the group with its differentsupervisors, in particular for prudentialreporting and the approval of internal methodsto calculate capital requirements. WhereasBasel II has kept the legal responsibilities ofnational supervisors unchanged, the CRDprovides for an enhanced role of the“consolidating supervisor”, the supervisorresponsible for the consolidated supervision ofa banking group.

However, giving a more prominent role to theconsolidating supervisor, which is as a rulefrom the Member State where the bank headingthe group is based, raises delicate questionsabout host authorities continuing to havesufficient powers and information regardingtheir banking systems. Chart 8 illustrates thatthis is particularly relevant for the new MemberStates. In most of them a large share of the localbanking market is foreign-controlled and someof these foreign presences can even bequalified as systemically relevant. For the timebeing, this presence mainly takes the form ofsubsidiaries, where the primary supervisory

responsibility continues to rest with the hostauthority, even under the CRD. However, ifthese subsidiaries were to be transformed intobranches, large segments of local bankingsectors would become directly supervised byauthorities from other EU Member States. Sucha prospect has become nearer with Nordea’splans to create a one-bank structure, based onthe European Company Statute, and conductbusiness in local markets through branches.30

5.2 POWERS AND RESPONSIBILITIES OF THECONSOLIDATING SUPERVISOR

The consolidating supervisor, also known asthe consolidated supervisor, is responsible forthe group-wide supervision of a bankinggroup.31 The debate surrounding theconsolidating supervisor is sometimesobfuscated by the use of similar terms whichcover very different meanings (see Box 4). Allthese terms refer to the supervisor who takes agroup-wide perspective, but the extent of his orher responsibilities and powers, and the waythey interact with the solo supervision of groupentities, differ strongly. At the one end of thespectrum, the supervisor only coordinates thecollection and dissemination of information

29 Basel Committee on Banking Supervision (2003 and 2004b).30 Nordea (2003).31 The CRD does not explicitly use these terms, but rather

refers to “the competent authority responsible for exercisingsupervision on a consolidated basis”.

Chart 8 Market share of foreign-ownedbanks in the new EU Member States


Source: Own calculations on the basis of ECB (2005c).Note: Market shares are calculated on the basis of totalbanking assets.

0

20

40

60

80

100

0

20

40

60

80

100

EE SK LT CZ HU PL MT LV SI CY


(“coordinating supervisor”), thus leaving thepowers of host subsidiary supervisors intact.At the other end, the supervisor assumes almostall the powers of the host supervisors (“leadsupervisor”). The CRD approach regarding theextended function of the consolidatingsupervisor lies somewhere in between thesetwo models.

Under the CRD, the consolidating supervisorcontinues to check compliance with supervisoryrequirements on a consolidated basis, as underthe CBD. The consolidating supervisor has alsobeen allocated responsibilities similar to thoseof the coordinating supervisor under theFinancial Conglomerates Directive. Moreover,

additional tasks not explicitly listed may beconferred upon him or her, which will have to bespecified in the written cooperation agreementthat the consolidating supervisor has toconclude with the host subsidiary authorities.Finally, and more controversially, in certaincases, the decisions of the consolidatingsupervisor will be binding for the hostsupervisors.

In more detail, the tasks of the consolidatingsupervisor are as follows:

– Supervisory overview and assessment ofcompliance with supervisory requirements.This covers the supervision of a parent

Box 4

OTHER SUPERVISORY ARRANGEMENTS FOR FINANCIAL GROUPS

Coordinating supervisor

The concept of a coordinating supervisor or coordinator is rooted in the FinancialConglomerates Directive.1 Under this the coordinating supervisor is allocated the followingtasks:

– to perform a supervisory overview and assessment of compliance with supervisoryrequirements in areas such as capital adequacy, risk concentration and intra-grouptransactions;

– to coordinate the gathering and dissemination of information;– to plan and coordinate supervisory activities;– to take certain technical decisions (e.g. relating to the identification of the conglomerate,

supervisory reporting); and– to undertake any additional tasks not mentioned in the Directive and conferred upon him or

her.

Lead supervisor and college of supervisors

The concept of lead supervisor is associated with the proposals by the European FinancialServices Round Table (EFR).2 The EFR starts from the concepts of coordinating andconsolidating supervisors, but brings them further by recommending that the principle of homecountry control is extended from branches to subsidiaries. In this way, the lead supervisorbecomes responsible for the supervision of the banking group at the different (sub)consolidated

1 Directive 2002/87/EC. The Financial Conglomerates Directive is to a large extent based on earlier work by the Joint Forum, see inparticular the coordinator paper, Joint Forum (1999).

2 European Financial Services Round Table (2004 and 2005). The EFR was formed in 2001 to provide an industry voice on Europeanpolicy issues related to financial services. Its members comprise the chairpersons or chief executives of 20 leading Europeanf inancial institutions. In its report, the EFR mentions that the lead supervisor may also be called the consolidating supervisor.

35ECB


5 HOME-HOSTISSUES AND THECONSOLIDATING

SUPERVISOR

and solo levels. The lead supervisor would be the single point of contact of the banking group withits supervisors and would decide on all reporting schemes at the different levels of the group. Inthe area of capital requirements, he or she would decide on the supervisory review process (“PillarII”) at the different levels, validate and authorise internal models for capital requirements andapprove the allocation of capital in the group. Host subsidiary supervisors in the EFR proposalwould still be involved via the college of supervisors that would act as a forum for informationexchange, discussion and provision of advice. However, it would not be a decision-making body,as only the lead supervisor has the power to take decisions.

To achieve this model, the EFR suggests making extensive use of the delegation of supervisorypowers that is already in the CBD, by formalising it in a memorandum of understanding andgiving it a sound legal basis (preferably) in an EU regulation. The CBD already provides for thepossibility of delegating supervisory powers and tasks. However, it seems that up to now thispossibility has not often been used, in particular for subsidiaries. This may be due to the legaluncertainty regarding the liability of supervisors in case of such delegation. A complicatingfactor is also that cost sharing arrangements must be agreed for the performance of tasks onbehalf of another supervisor.

The terms “lead supervisor” and “college of supervisors” were already used for the supervisionof insurance groups although the content of the terms differ somewhat from the EFR concepts.The Insurance Groups Directive3 is the legal basis for the cooperation between insurancesupervisors, and the “Helsinki Protocol” (2000) gives guidelines on how such cooperationshould work in practice. The Protocol provides for a college of supervisors, called acoordination committee, and the lead supervisor. The latter will in practice carry out most ofthe work of the group-wide supervision by pulling together all relevant information, analysingit and sharing the findings with the coordination committee. Such a lead supervisor can only beappointed if there is unanimity within the committee; in general, the lead supervisor will be thesupervisor of the country where the group has its dominant insurance undertaking. Thecoordination committee itself is composed of supervisors of all countries in which the grouphas undertakings and who are involved in its day-to-day supervision.

3 Directive 98/78/EC.

bank’s compliance with supervisoryrequirements on a consolidated basis. Areascovered are: (i) the minimum level of ownfunds under Pillar I, (ii) limits regardingshareholdings held outside the financialsector, (iii) the process to assess thatadequate “internal” capital is available, (iv)large exposures requirements, (v) disclosurerequirements, (vi) robust governancearrangements.

– Coordination of the gathering anddissemination of information. Thecoordinating supervisor has to provide thehost subsidiary supervisors with all therelevant information they need for the

supervision of these subsidiaries. Thecoordinating supervisor also has a particularinformation duty in the event that anemergency situation occurs that threatensthe stability of the financial system, when heor she has to alert the central banks andMinistries of Finance concerned.

– Planning and coordination of supervisoryactivities. Particular reference is made to thereview of compliance with the CRDrequirements and the evaluation of risks bycompetent authorities under Pillar II andinspections of foreign branches andsubsidiaries in the Member States.


The parent and the subsidiaries of a bankinggroup will be able to jointly apply to use IRBapproaches to calculate capital requirementsfor credit risk, AMA for operational risk andthe recognition of internal models for marketrisk. In that case, the competent authoritieshave to work together on whether or not to grantpermission, and if so under what conditions.Such an application has to be submitted only tothe consolidating supervisor. The competentauthorities have to decide on the applicationwithin six months in a single document. In theabsence of an agreement among the authoritiesinvolved, the consolidating supervisor shalltake its own decision. This last element inparticular has caused controversy as it shiftspowers from the supervisors of subsidiaries tothe consolidating supervisor.

5.3 ASSESSMENT OF THE CONSOLIDATINGSUPERVISOR

The more prominent role of the consolidatingsupervisor in combination with the enhancedinformation exchange between supervisorsprovided for by the CRD, is beneficial tofinancial stability and financial integration.First, the arrangements ensure that theconsolidating supervisor has a comprehensivegroup-wide view, while the host subsidiarysupervisors will have easier access to groupinformation that might be relevant for theentities they supervise.

Second, for banking groups, the newarrangements lead to a more streamlinedinteraction with their different supervisors andlower compliance costs. The risk of different,inconsistent or even conflicting, supervisoryapproaches is therefore significantly reduced.Since large cross-border groups typicallyorganise some of their key business functionson a group-wide basis, it also leads to a bettermatch between the way they are supervised andmanaged. Finally, it contributes to the creationof a level playing-field for European financialinstitutions in the international market. A pan-European group that has to deal with multiplesupervisors, for example, to get its internal risk

models approved, has a competitivedisadvantage compared with its US counterpartthat needs to submit only one such application.

Notwithstanding these potential benefits, theissue of the consolidating supervisor alsoraises complex implementation questions.First, at present, the powers, resources andexperience of national supervisors differsubstantially. These differences are rooted infactors such as the size of the financial system,the way the supervisor is financed, politicalchoices and supervisory approach. Not allcountries therefore have the resources or know-how to act as the consolidating supervisor ofmajor international banking groups.

Second, notwithstanding the ultimate decision-making power of the consolidating supervisorregarding the approval of group-wide methods,it is important that any disagreements with hostsubsidiary supervisors are adequatelyaddressed so as not to undermine the powers ofthe host supervisors. For example, although theconsolidating supervisor would approve group-wide advanced calculation methods for capitalrequirements, it is not clear how this wouldrelate to the responsibility of the ongoingsupervision of these methods.

In this context, it is advisable that guidance isdeveloped for the cooperation between theconsolidating supervisor and the hostsubsidiary supervisors, an area in which theCEBS is already active.32 The CEBS isdeveloping its guidance along a risk-based andproportional approach. Hence, the degree ofinformation exchange and cooperationbetween supervisors should be related to theimportance of the subsidiaries, both in relationto the host local market and the group as awhole. In order to avoid uncertainty, it isimportant that such arrangements areformalised via memoranda of understanding(MoU). A certain degree of flexibility isneeded in such MoUs to accommodate thespecific features of an individual group. At the

32 See in particular the CP9 of the CEBS.

37ECB


6 REAL ESTATELENDINGsame time, standardisation and supervisory

convergence should be aimed at ensuring alevel playing-field.

Third, in the event that the consolidatingsupervisor takes a different stance to that ofthe host subsidiary supervisor on certainprudential issues, this could lead to acompetitive distortion within national markets.In particular, institutions that are not part of awider group (e.g. local cooperative banks orsavings banks) could be disadvantagedcompared with institutions that belong to agroup in case the consolidating supervisortakes a more lenient stance. Similarly, banksthat are part of a group with a differentconsolidating supervisor may be treateddifferently, which calls for supervisoryconvergence through the CEBS.

Fourth, the granting of more powers andresponsibilities to the consolidating supervisorshould not lead to a disconnection between theresponsibilities for prudential supervision andother closely related areas of public policy.These include financial stability monitoring,the management of deposit insurance,emergency liquidity assistance and tax payersupport. Most of these areas continue to beorganised on a national basis and theinvolvement of a consolidating supervisor inthe supervision of locally licensed institutionsmight create difficulties.

Banking supervisors have a national mandateand are accountable to national political bodiessuch as ministries of finance and parliaments.The latter are primarily concerned about theimpact on the national financial system orbudget. The commitment of a foreignsupervisor to entities in the host country maytherefore not be fully credible. In particular,this is the case when the parent bank is notsystemically relevant in its home country, butthe subsidiary in the host country is. In thisscenario, it seems unlikely that the homesupervisor would have an equally strongincentive to monitor or intervene in the localsubsidiary as the host authority. Moreover, in

the event of any public financial intervention tosupport the group, in relative terms, it wouldbenefit the host country more than the homecountry.

Finally, one should note the views of theindustry, which is pursuing an enhanced rolefor the consolidating supervisor. The EuropeanBanking Federation has already expressed itsdisappointment and continues to plead for anextension of the responsibilities of theconsolidating supervisor to Pillars II and III,turning his or her role into that of a real“lead supervisor”. The supervisory communitycontinues to be reluctant to support this model,advocating instead increased supervisorycooperation in combination with supervisoryconvergence. In general, supervisors do notsee a need for any change in the legalresponsibilities between home and host.However, there are other supervisors who takea rather sympathetic view towards the leadsupervisor (Brouwer, 2004). Given the above-mentioned considerations, it seems preferablethat before moving towards the path of the leadsupervisor, more experience is first gained withthe actual working of the extended role of theconsolidating supervisor and the potentialitiesoffered by the Lamfalussy framework.

6 REAL ESTATE LENDING

6.1 OVERVIEW

The differences between Basel II and the EUrules are particularly visible in the treatment ofresidential real estate (RRE) and commercialreal estate lending (CRE) under the standardisedapproach. The regulatory treatment of suchlending is of great importance given the highshare of bank loans it accounts for in the EU.Chart 9 gives an indication of this and, since itdoes not include commercial real estate lending,reflects an even larger significance than issuggested by the figures. Moreover, CRElending has been a recurring cause of problems inthe banking industry over the past decades (e.g.the savings and loans crisis in the US, the


Swedish banking crisis), warranting aparticularly prudent approach.

The BCBS provides for the preferential capitaltreatment of certain types of real estate lendingin case it is secured by mortgages. Thispreferential treatment applies to both thestandardised approach and the foundation IRB,subject to stringent conditions to assure that thelending exhibits low risk.

The EU rules are not a direct transposition ofBasel II. From the outset, the EuropeanCommission aimed to be as consistent aspossible in its treatment of RRE and CRElending across the different approaches. TheCommission therefore extended the Basel IIconditions for preferential treatment under thefoundation IRB approach to the standardisedapproach. Hence, the differences betweenBasel II and the EU rules are especiallymanifest in the standardised approach, onwhich this paper focuses.

Although the Commission’s proposalsreceived criticism from the industry becausethey were more stringent than Basel II incertain respects,33 the Commission maintainedits original position. The EU rules thereforehave the merit of achieving a high degree ofconsistency between the standardised approachand the foundation IRB. As explained below, tosome extent this comes at the expense of a lessconservative treatment of CRE. However, inthe specific field of CRE that produces incometo service the debt, the EU rules apply the samecriteria as Basel II. This is important from afinancial stability point of view, given thelower additional protection provided by thattype of collateral.

6.2 RESIDENTIAL REAL ESTATE

The preferential risk weights for secured RRElending under the standardised approach areidentical under Basel II and the CRD.However, the difference lies in the criteria to bemet for these lower risk weights (see Table 3).The main difference is that the CRD imposestwo independence criteria.34 By contrast, BaselII does not impose these criteria under thestandardised approach, but only under thefoundation IRB. Hence, in this respect the EUrules are more conservative.

The first independence criterion (borrower riskdoes not influence property value) is typicallygiven for RRE. The exception might be the case

Table 3 Requirements for a preferentia l r isk weighting of res idential real estate lending

Basel II CRD

35% risk weight for RRE lending instead of 100%1. Over-collateralisation Yes, but not quantified Yes, but not quantified2. Property value not materially depending on borrower credit quality No Yes3. Borrower risk not materially depending on collateral value No Yes, but may be waived if

loss rates in the jurisdictionare “sufficiently low”

33 European Mortgage Federation (2003).34 Where independence of collateral value and borrower default

risk is not given, the additional protection from the collateralis limited. This is often referred to as “wrong-way” risk.

Chart 9 Loans for house purchases as shareof total loans to non-banks


Source: ECB.Note: Non-banks comprise non-monetary f inancialinstitutions.

0

10

20

30

40

50

60

0

10

20

30

40

50

60

DK NL FI DE PT BE ES IE EuroArea

FR SE GR AT IT LUUK

39ECB


6 REAL ESTATELENDINGof very special property that is tailored to

individual needs and is therefore difficult toresell.

The second independence criterion (propertyvalue does not influence borrower risk) is aconstraining factor when the borrower has torely on the collateral’s rental income to servethe loan. This criterion would exclude from thepreferential treatment firms that specialise inowning and letting residential real estate. Inthese cases, the collateral offers only limitedadditional protection compared to anunsecured loan because the default of theborrower will typically result from insufficientrental income. Under such conditions, it is verylikely that the property value as derived fromthe discounted net rental income will also havebeen negatively affected.

The fact that the borrower relies on the incomefrom the collateral does not necessarily meanthat the credit risk of such exposure is similarto that of an unsecured exposure. Morespecifically, the credit risk will depend on howthe loss expectation in RRE lending compareswith the loss expectation of other loans. In itsproposal, the Commission assumed that theloss expectation in RRE lending varies amongnational markets. It also allows nationalauthorities to dispense with the secondindependence criterion within their nationalmarket under certain conditions.

The condition for such a waiver is that thenational authorities have evidence of loss ratesthat are sufficiently low to justify suchtreatment. Given that the CRD offers noguidance as to how to assess this,implementation efforts should focus on whatconstitutes sufficiently low loss rates. This isone of the fields where supervisors, through theCEBS, could achieve a harmonisedimplementation. Starting with an analysis ofnational loss rates and comparing them withloss rates for other asset classes, it should bepossible to define a quantitative threshold.This common threshold would facilitate themutual recognition of preferential risk weights

and thereby promote the integration of aEuropean market for RRE lending.

While data on loss rates for RRE lending arenot readily available, the volatility ofresidential property prices can be assumed tobe an important driver. In this respect, it isinteresting to note that there are pronounceddifferences in the development of housingprices across Member States (see Chart 10). Itis likely that these differences also translateinto differences in loss rates.

Finally, a more subtle difference betweenBasel II and the EU rules is that, under thelatter, the collateral must be used or rented-outby the owner, who may not be the borrower.Basel II, by contrast, requires this to be theborrower. In both cases, purely speculativeholdings of property would be excluded.However, the CRD would allow for situationswhere a loan to a small firm (i.e. the borrower)is secured by a mortgage on the residenceof the firm’s owner. Such situations may beespecially relevant for SMEs.

Chart 10 Di f ferences in res identia l propertyprice changes across EU countries

(percentages)

Source: ECB, based on non-harmonised national data.Note: EU average is GDP-weighted (value: 9%). Bars arecalculated as average annual percentage changes over theindicated period and the non-horizontal line measuresthe difference relative to EU151) average for the period2000-2003.1) Data for DK and LU are missing.

-10

-5

0

5

10

15

20

25

30

AT DE PT FI BE SE IT FR NL GR IE UK ES-10

-5

0

5

10

15

20

25

30

1994-19981999-20033 years difference with EU15 1) averageEU15 1) average (2000-03): 9%


6.3 COMMERCIAL REAL ESTATE

CRE lending is a particularly sensitive area as,in the past, banking crises have sometimescoincided with a crisis in the commercial realestate market. Strict regulatory criteria havetherefore to be met before such lending canbenefit from a preferential capital treatment.

Under Basel II, the preferential treatment forCRE is only made available in national marketswhere the loss rates observed in the past do notexceed certain thresholds (see Table 4). TheCRD, by contrast, is more flexible since itimposes the requirement on loss rates only incases where the property value materiallyinfluences the borrower risk; this is basicallywhere the borrower has to rely on thecollateral’s rental income to serve the loan. Asa result, in the EU, jurisdictions where thethresholds are exceeded can also apply for apreferential CRE treatment.35 This extendedpreferential treatment reflects the fact that aloan to a firm that does not rely on real estateincome is less risky than an otherwise identicalunsecured loan.

Basel II specifically lists offices and multi-purpose or multi-tenanted properties as eligibletypes of property. The CRD, by contrast, doesnot include a restriction on the type ofpremises, but imposes an abstractindependence criterion for the property value

Table 4 Requirements for a preferentia l r isk weighting of commercial real estate lending

Basel II CRD

50% risk weight for CRE lending instead of 100%1. Over-collateralisation (eligible lending may exceed neither Yes Yes

50% of market value nor 60% of mortgage lending value)2. Property value not materially depending on borrower credit quality Yes, implicitly: Yes

property must be officesor multi-purpose/multi-tenanted

3. Borrower risk not materially dependent on collateral value No Yes, but may be waived ifpoint 4 is met.

4. Loss rates (as percent of outstanding CRE) in the jurisdiction Yes No (i.e. not generally,not in excess 0.3% on eligible loans and 0.5% overall see point 3.)

not to depend on the borrower risk. Thisbasically serves the same purpose as listing theproperty types, but gives more leeway forprudential judgement. With a view to achievinga level playing-field, it is important thatEuropean supervisors strive for convergence inthis area.

Another distinguishing feature of the EUframework is the preferential treatment forCRE leasing. Such leasing can obtain the lowerrisk weight when the conditions for therecognition of mortgage lending are met. Theonly specific requirement is that the lessor, i.e.the party which rents out the property, retainsfull ownership in the same way as a mortgagelender. However, depending on the structureof the lease, a lessor may also be exposedto specific risks that are not incurred by amortgage lender. For example, a lessor mayincur a price risk when having to sell theproperty if it is not fully amortised. While suchleasing-specific risks are not addressed in theCRD, a prudent implementation should onlyextend the preferential treatment to thoseleases under which the lessor bears risks thatare not larger than those of a mortgage lender.

35 Under the foundation IRB, Basel II also allows forrecognition of this loss-reducing effect. In this way, theCommission’s proposal achieves higher consistency acrossapproaches.

41ECB


7 COVERED BONDS7 COVERED BONDS

7.1 CONCEPT OF COVERED BOND

The EU rules provide for a specific treatment ofcovered bonds, which is not present in Basel II.A covered bond is a secured bond issued by abank or other financial institution. However,the security for the bond is not merely over theinstitution’s assets generally but a designatedpool of assets, typically mortgage loans orpublic sector loans, is specifically designatedas collateral. Covered bonds differ fromsecuritisation bonds in a number of importantrespects. First, bondholders have recourse bothto the issuing institution and to the assets thatprovide cover for the bonds. Second, the bondsdo not achieve off-balance sheet treatment asthe asset pool that provides cover remains onthe institution’s balance sheet.

Such financing instruments are of greatimportance in the EU capital markets, althoughtheir significance differs across Member States(see Chart 11). Several Member States have inthe recent past adopted legislation to allow theissuance of such bonds. It is therefore justifiedthat the CRD takes account of this typicalEuropean instrument. Since the most commonform (the German “Pfandbrief”) is well-knownoutside European capital markets, even aspecific treatment under Basel II itself mighthave been desirable.

The treatment of covered bonds under the EUrules is based on the legal definition of thisinstrument in the UCITS Directive(Undertakings for Collective Investment inTransferable Securities),36 which is very broad.It does not refer to an identical instrument in allMember States, but rather to a range ofinstruments that are not subject to harmonisedrules at the EU level. In this way, the definitionaccommodates for the wide differences thatexist in national laws on covered bonds in areassuch as assets used as collateral, restrictionsplaced the issuing bank’s activities, the degreeof “over-collateralisation”, and the way

bankruptcy remoteness is achieved. Over-collateralisation refers to the extent to whichthe value of the collateral has to be higher thanthe nominal value of the bond and bankruptcyremoteness ensures that the bond holders’access to the collateral is truly superior to thatof all other creditors.

In the light of this diversity, the definition ofcovered bonds for the purposes of capitalrequirements is more restrictive as onlyspecific assets – in some cases subject to over-collateralisation – may be used as collateral.The CRD lists in this context mainly:

– exposures that qualify for a 0% risk weightunder the standardised approach. Thisincludes exposures on central governments,central banks, multilateral developmentbanks and international organisations.Exposures to other public sector entities,regional governments and local authoritiesmay also be included if they are consideredto carry the same credit risk as centralgovernments;

– loans secured by residential real estatemortgages which, including prior liens, areno more than 80% of the value of theproperty; and

36 Directive 85/611/EEC, art. 22(4).

Chart 11 Share of covered bonds in theissues of private sector bonds outstanding


Sources: ECB and European Mortgage Federation.

0

20

40

60

80

100

0

20

40

60

80

100

DK ES SE EuroArea

LU IE FR ATDE


– loans secured by commercial real estatemortgages which, including prior liens, are nomore than 60% of the value of the property.The competent authorities may also recogniseloans where this “loan-to-value” ratio is up to70%; the condition is that the value of thetotal assets pledged (i.e. not only themortgaged assets) exceeds the outstandingamount of the bonds by at least 10%;

Further, up to 10% of the outstanding coveredbonds may be covered by exposures to banksand investment firms. In this way, the issuingbank may to a certain extent replace eligiblecollateral by deposits with banks of a highcredit standing when collateral is scarce.

The additional restrictiveness under the CRDis not expected to lead to a large scaleineligibility of instruments that are issuedunder national legislations. Rather, the presentrange of covered bonds in the EU will also beeligible for the preferential capital treatment.For example, the “loan-to-value” limits formortgage loans in the CRD reflect the highestthresholds present in these legislations, thougha number of them are more conservative. Forexample, the 10% limit for the substitution ofeligible assets by other assets such as bankdeposits, is a possible constraint for someMember States that currently allow higherlimits.

7.2 CAPITAL TREATMENT

The covered bonds that meet the CRDrequirement are treated as exposures to banks.The risk weighting is based on the creditstanding of the issuing bank, while at the sametime recognising the effects of the collateral.The collateral is recognised in the form of

reduced risk weights under the standardisedapproach or in the form of reduced LGDs underthe IRB approaches.

Under the standardised approach, coveredbonds receive reduced risk weights based onthe weights of senior exposures to the issuer inthe manner described in Table 5.

As regards treatment under the IRBapproaches, the EU rules are fully consistentwith Basel II, since a bank’s internal ratingsystem needs to comprise both a borrower anda facility dimension. Based on the borrowerdimension, PDs are assigned to exposures,while the facility dimension underlies theassignment of LGDs. The collateral to whichthe bondholders have a preferential claimaffects the facility dimension. While Basel IIdoes not encompass any specific rules forcovered bonds, the collateral of the bond wouldlead to a reduced LGD if the bank were able toget supervisory approval for an estimate of thiscollateral effect under the advanced IRB.Under the foundation IRB, such covered bondsmay receive a reduced supervisory LGD of12.5%.37 The advanced IRB would require theinvesting bank to use its own LGD estimatesfor covered bonds. Under both the foundationand advanced IRB the risk weights continue todepend also on the PD of the issuer.

Chart 12 depicts the risk weights for allapproaches. It shows that in many situations,the standardised approach will deliver the

Risk weight of senior exposure to issuer 20 50 100 150

Covered bond risk weight 10 20 50 100

Table 5 Risk weights for the issuer and covered bonds under the standardised approach

(percentages)

37 Compare this with the supervisory LGD of 45% for seniorclaims and 75% for subordinated claims. For certain coveredbonds, the LGD may further be reduced to 11.25% until theend of 2010.

43ECB


CONCLUDINGREMARKS

38 This is the “Option 1” of the standardised approach; bycontrast, under “Option 2”, the risk weights for exposures onbanks are directly determined on the basis of the banks’ratings and not on the basis of the sovereign’s rating. See alsoSection 2.2.1.

39 With an LGD of less than 7%, capital requirements forcovered bonds under the advanced IRB (M = 5 years) wouldbe consistently lower than under the foundation IRB.

lowest risk weights for covered bonds injurisdictions where risk weights for exposuresto banks are based on the ratings of thesovereign38 and the sovereign itself receives a0% risk weight. Only for issuers with very lowPDs would the IRB approaches lead to lowerrisk weights. Default experience for corporateborrowers published by rating agenciessuggests that this will only be the case forissuers whose rating for their senior debt iscomparable to an S&P rating of “AA” or better.

How the outcome under the advanced IRBrelates to the other approaches depends verymuch on the assumptions for the M and LGDrisk factors. For M, one can assume themaximum possible value of five years given theusually long maturities of covered bonds. ForLGD, Chart 12 assumes the same value as thesupervisory LGD (i.e. 12.5%) because there isa lack of default experience, a difficulty thatbanks under the advanced IRB will also face.On the basis of these assumptions, theadvanced IRB appears unattractive comparedwith the foundation IRB because, under thelatter, M will in most cases be fixed at 2.5years. However, this can change if a bankreceives supervisory approval for lowerestimates of LGD.39

Another observation from this comparison isthat, in jurisdictions where risk weights forbanks are directly based on their own ratingsrather than on those of their sovereign, theresulting weights for covered and non-coveredissues will often be the same. For example,when an issuer’s senior debt has an S&P ratingof “A” under the standardised approach ofBasel II it would receive a 50% risk weight. Forthe same issuer’s covered bonds, a 20%weighting would apply under the EU rules. Butthis 20% would often apply anyway, evenwithout the specific EU treatment, given thatcovered bonds have typically issue specificratings of “AA” or better that results in thesame risk weight.

CONCLUDING REMARKS

This paper provides an overview of the NewBasel Capital Framework from an EUperspective and discusses certain importantissues relating to the implementation of BaselII in the EU. Despite the similarities betweenthe draft EU framework and the BCBSproposal, a number of different treatments arejustified by EU specificities.

The Basel II Framework is expected to fosterstability in the EU financial system. Thisobjective will be achieved by strengtheningincentives for banks to conduct sound riskmodelling and management, and by involvingsupervisors more directly in the review ofbanks’ risk profiles, risk management practicesand risk-bearing capacity. Furthermore, thenew disclosure requirements will enablemarket participants to have better informationon banks, which will support the functioning ofmarket discipline. These beneficial effects onthe financial system will also translate into a

Chart 12 Risk weights for covered bonds

(percentages)

Source: ECB.

0

5

10

15

20

25

30

0

5

10

15

20

25

30

0.00 0.10 0.20 0.30 0.40 0.50

Risk weight

Probability of default

advanced IRB approach (LGD = 12.5%, M = 5 years)standardised approach, option 2 foundation IRB approach (LGD = 12.5%, M = 2.5 years)standardised approach, option 1, sovereign of best credit quality


positive contribution to the EU economy atlarge.

Notwithstanding, there are still a number ofkey challenges in the EU implementationprocess which need to be tackled in order tofully reap the potential benefits of Basel II.First, it is important to preserve a level playing-field across the EU. A consistentimplementation of the EU rules into nationallegislation and a coherent application of theframework are therefore necessary. In thiscontext, reducing the number of options in thedraft EU capital requirements framework is achallenging task. The CEBS has already madeprogress here, which needs to be kept up. TheEU will also have to follow-up closely on howBasel II will be implemented in the US and howthe decisions by US authorities will interactwith the activity of EU banks in the US anddecisions by EU supervisors.

Second, the debate on the optimal design of theconsolidating supervisor’s functions continuesas some of the largest firms want to move to the“lead supervisor” model. Although pursuing anenhanced role for the consolidating supervisoroffers advantages from the perspectives offinancial integration and financial stability,there remain a number of compleximplementation issues that need to be takeninto account.

Third, the potentially adverse effects of capitalrequirements on the EU macroeconomicenvironment require ongoing monitoring. Thisissue and the related question of the structuralimpact of the new rules on the financial systemare of particular interest to manyconstituencies, including central banks. Thisexplains the reference to the role of the ECB inthe monitoring of possible procyclical effectsin the CRD.

In the medium-term, a number of issues maypotentially achieve increasing prominence.First, the impact of the new accountingmeasures on the capital definition may turn outto be quite important. Here, banking

supervisors have already considered theinterplay between the implementation ofcertain International Financial ReportingStandards (IFRS) and the definition ofregulatory capital. There are also plans to carryout a more fundamental review of the definitionof regulatory capital.

Another medium-term issue is the nexus ofmarket and credit risk, which may bring thecurrent market risk treatment under furtherscrutiny. As things stand, the border betweenthe two risk categories is not very clear. Due tothe improved abilities to trade, hedge and valuecredit risk related instruments, they play anincreasingly significant role in the tradingbooks of financial institutions. This, however,raises concerns over the prudence of thepresent simple approach to the specific marketrisk of debt instruments in the trading book. Anexample of this is the impressive developmentof credit derivatives and Collateralised DebtObligations (CDOs). At this stage, the BCBSdoes not plan to issue new rules on the existingboundary between trading book and bankingbook, and only a limited number of adaptationsto the current regime were recently adopted.However, in the future, the financial industrymay call for a harmonised treatment of thesetwo main risk categories.40

Finally, and in relation to the preceding issue,the use of credit portfolio management isundergoing significant changes. Banks havestarted to apply increasingly sophisticatedportfolio approaches to their loan books. Themodelling of credit correlations and portfoliodiversification effects is a very active area ofresearch, both among practitioners andacademics. Hence, the BCBS’s decision tolimit recognition for regulatory capitalpurposes to the single risk factor IRB approachmay come under increased scrutiny in thefuture.

40 Duff ie and Singleton (2003) provide a detailed analysis ofcredit derivatives and also discuss the integrated modeling ofmarket and credit risk.

45ECB


GLOSSARY

GLOSSARY

Advanced (calculation) approaches: methods available to banks to calculate their regulatorycapital requirements based on own risk estimates. Includes the foundation and advanced internalratings-based (IRB) approach for credit risk, the advanced measurement approaches (AMA) foroperational risk, and the internal models approach for market risk.

Asymptotic single risk factor (ASRF) model: theoretical model underlying the risk weightfunction (which relates risk factors to risk weights) of the internal ratings-based approach.

Banking book: the bank portfolio consisting of financial instruments that are not held for trading.

Capital ratio: ratio of regulatory own funds (core capital and supplementary capital) to total risk-weighted assets. Also called solvency ratio.

Consolidating supervisor: the supervisor responsible for the supervision on a consolidated basisof a banking group. As a rule, this is the supervisor of the Member State where the parent bank ofthe group is based.

Core capital: regulatory capital that consists of own funds components of the highest quality,such as fully-paid capital and disclosed reserves from post-tax retained earnings. Also called Tier1 capital.

Covered bond: a secured bond issued by a bank or other financial institution. The security of thebond is not merely over the institution’s assets generally, but also over a designated pool of assets,typically mortgage loans or public sector loans.

Credit risk: risk of losses in on and off-balance sheet positions resulting from the failure of acounterparty to perform according to a contractual arrangement.

Double default: refers to the fact that the risk of both a borrower and a guarantor defaulting on thesame obligation may be substantially lower than the risk of only one of the parties defaulting.

Economic capital: capital held and allocated by the bank internally as a result of its ownassessment of risk. It can differ from regulatory capital, which is determined according tosupervisory rules.

Effective maturity: remaining time before a borrower is scheduled to fulfil his obligation; one ofthe requested input parameters to derive the risk weight under the internal ratings-basedapproach.

Expected loss: (credit) losses that are expected to occur within a certain time period (generallyone year).

Exposure at default (EAD): the exposure a bank is likely to have on a borrower at the moment theborrow defaults; one of the requested input parameters to derive the risk weight under the internalratings-based approach.


External credit assessment institution (ECAI): institution recognised by supervisors andwhose assessments may be used to calculate regulatory capital requirements for credit risk underthe standardised approach.

Internal ratings-based (IRB) approach: advanced approach by which a bank can use its owncredit assessments to calculate its regulatory capital requirements for credit risk. Depending onthe risk factors the bank is allowed to estimate, a distinction is made between a foundation IRBand an advanced IRB approach.

Lead supervisor: supervisory model under which the consolidating supervisor would be thesingle supervisory contact point of a banking group, both for its branches and subsidiaries. In thearea of capital requirements, the lead supervisor’s tasks and responsibilities would extend to thethree pillars of the New Framework.

Loss given default (LGD): the loss, measured as a percentage of the exposure at default, which islikely to occur in case a borrower defaults; one of the required input parameters to derive the riskweight under the internal ratings-based approach.

Madrid compromise: the agreement reached by the Basel Committee in January 2004 tocalibrate the risk weight functions used in the internal ratings-based approach on the basis ofunexpected credit losses only.

Market risk: risk of losses in on and off-balance sheet positions arising from movements inmarket prices and volatilities.

Operational risk: risk of losses resulting from inadequate or failed internal processes, people orsystems, or from external events. It includes legal, but not reputational or legal risk.

Permanent partial use: in the context of credit risk, this refers to the combined use of thestandardised approach for certain exposure classes whilst applying the more sophisticated IRBapproaches for the remaining classes on a non-temporary basis.

Probability of default (PD): the likelihood that a borrower will default within a certain timeperiod (generally one year); one of the required input parameters to derive the risk weight underthe internal ratings-based approach.

Procyclicality: exacerbation of the economic cycle as a result of changes in regulatory capitalrequirements.

Quantitative Impact Study (QIS): data collection exercise organised by supervisors to assessthe impact of the new capital rules on banks.

Regulatory capital (own funds): own funds that are eligible to meet the regulatory capitalrequirements; consist of core capital and supplementary capital, after a number of deductions.Regulatory capital is the numerator of the capital ratio.

Risk-weighted assets: risk measure that consists of multiplying each asset value by a factor (riskweight) that is a proxy of the (credit) risk related to the asset class. Risk-weighted assets are the

47ECB


GLOSSARY

denominator of the capital ratio. For operational risk and market risk, the risk-weighted assetsthat enter into the capital ratio are derived from the directly calculated capital requirements.

Solvency ratio: see capital ratio.

Standardised approach: method by which a bank can use external ratings (if available) byexternal credit assessment institutions to calculate its regulatory capital requirements for creditrisk.

Supplementary capital: regulatory capital that consists of own funds of a lower quality than corecapital, such as certain types of subordinated debt; also called Tier 2/Tier 3 capital.

Trading book: the bank portfolio consisting of financial instruments that are in principle held forshort-term trading purposes, i.e. they are held intentionally for short-term resale and/or with theintent of benefiting from short-term price movements or to lock in arbitrage profits.

Unexpected loss: (credit) losses which may occur above the expected loss within a certain timeperiod (e.g. one year) and within a high confidence level (e.g. 99.9%).


REFERENCES

Allen L. and Saunders A. (2004), “Incorporating systemic influences into risk measurements: asurvey of the literature”, Journal of Financial Services Research, vol. 26, No. 2, pp. 161-191.

Altman E., Brady B., Resti A. and Sironi A. (2005), “The link between default and recovery rates:implications for credit risk models and procyclicality”, Journal of Business, forthcoming.

Amato J. D. and Furfine C. H. (2004) “Are credit ratings procyclical?”, Journal of Banking andFinance, 28(11), pp. 2641-2677.

Bangia A., Diebold F., Kronimus A., Schagen C. and Schuermann T. (2002), “Ratings migrationand the business cycle, with application to credit portfolio stress testing”, Journal of Bankingand Finance, No. 26, pp. 445-474.

Basel Committee on Banking Supervision (2005a), “An explanatory note on the Basel II IRB riskweight functions”, July.

Basel Committee on Banking Supervision (2005b), “The Application of Basel II to tradingactivities and the treatment of double default effects”, July.

Basel Committee on Banking Supervision (2005c), Newsletter No. 4, January.

Basel Committee on Banking Supervision (2004a), “International convergence of capitalmeasurement and capital standards”, June.

Basel Committee on Banking Supervision (2004b), “Principles for the home-host recognition ofAMA operational risk capital”, January.

Basel Committee on Banking Supervision (2003), “High-level principles for the cross-borderimplementation of the New Accord”, August.

Basel Committee on Banking Supervision (1996), “Amendment to the Capital Accord toincorporate market risks”, January.

Basel Committee on Banking Supervision (1988), “International convergence of capitalmeasurement and capital standards”, July.

Berger A. and Udell G. (2003), “The institutional memory hypothesis and the procyclicality ofbank lending behaviour”, BIS, Working Paper No. 125.

Brouwer H. (2004), “Financial integration in the EU – some food for thought”, Speech at theboard meeting of the European Banking Federation, 5 November 2004.

Carling K., Jacobson T., Lindé J. and Roszbach K. (2002), “Capital charges under Basel II:corporate credit risk modelling and the macro economy”, Sveriges Riksbank Working PaperSeries, No. 142, September.

49ECB


REFERENCES

Catarineu-Rabell E., Jackson P. and Tsomocos D. (2003), “Procyclicality and the new BaselAccord – banks’ choice of loan rating system”, Bank of England - Working Paper No. 181.

De Fontnouvelle P., Rosengren E. and Jordan J. (2004), “Implications of alternative operationalrisk modeling techniques”, Mimeo.

Deutsche Bundesbank (2004), “New capital requirements for credit institutions”, MonthlyBulletin, September, pp. 73-94.

Dietsch M. and Petey J. (2004) “Should SME exposures be treated as retail or corporateexposures? A comparative analysis of default probabilities and asset correlations in Frenchand German SMEs”, Journal of Banking and Finance, 28(4), pp. 773-788.

Duffie D. and Singleton K. (2003), “Credit risk: pricing, measurement and management”,Princeton University Press.

European Banking Federation (2004), “ECOFIN general approach welcome but does not go farenough, says FBE”, 7 December 2004.

European Central Bank (2005a), “Eurosystem collateral framework: inclusion of non-marketableassets in the Single List”, Press release, 22 July 2005.

European Central Bank (2005b), “Opinion of 17 February 2005 at the request of the Council of theEuropean Union on a proposal for directives of the European Parliament and of the Council re-casting Directive 2000/12/EC of the European Parliament and of the Council of 20 March 2000relating to the taking up and pursuit of the business of credit institutions and Council Directive93/6/EEC of 15 March 1993 on the capital adequacy of investment firms and credit institutions(CON/2005/4)”.

European Central Bank (2005c), “Banking structures in the new EU Member States”, January.

European Commission Services (2004), “Proposal for new capital requirements regime for creditinstitutions and investment firms”, COM(2004)486, 14 July 2004.

European Commission (2003a), “Review of the capital requirements for credit institutions andinvestment firms – Third quantitative impact study: EU results”, July.

European Commission (2003b), “SMEs in Europe 2003”, Observatory of European SMEs, No. 7.

European Financial Services Round Table (2005), “On the lead supervisor model and the future offinancial supervision in the EU”, June.

European Financial Services Round Table (2004), “Towards a lead supervisor for cross-borderfinancial institutions in the European Union”, June.

European Mortgage Federation (2003), “Position Paper on November 18 Commission WorkingDocument on capital requirements for credit institutions and investment firms”, Mimeo,January.


European Private Equity and Venture Capital Association (2005), “EVCA Yearbook 2005”, June.

European Private Equity and Venture Capital Association (2004), “Implementation of Basel II inthe EU”, 14 July 2004.

Fabi F., Laviola S. and Marullo-Reedtz P. (2004), “The treatment of SME loans in the new Baselcapital accord: some evaluations”, Banca Nazionale del Lavoro Quarterly Review 57(228),pp. 29-70.

Federal Reserve Board, FDIC, OCC, OTS, (2005), “Banking agencies to perform additionalanalysis before issuing Notice of Proposed Rulemaking related to Basel II”, Press Release,29 April 2005.

Federal Reserve Board, FDIC, OCC, OTS, (2003), “Banking agencies announce revised plan forimplementation of Basel II framework “, Press Release, 30 September 2003.

Ferguson R. W. (2003), “Basel II: scope of application in the United States”, Remarks before theInstitute of International Bankers, 10 June 2003.

Goodhart C. and Segoviano M. (2004), “Basel and procyclicality: a comparison of theStandardised and IRB Approaches to an improved credit risk method”, FMG Working Paper.

Gordy M. (2003), “A risk-factor model foundation for ratings-based bank capital rules”, Journalof Financial Intermediation, No. 12, pp. 199-232.

Gordy M. and Howells B. (2004), “Procyclicality in Basel II: can we treat the disease withoutkilling the patient?”, Mimeo.

Hayes S., Saporta V. and Lodge D. (2002), “The impact of the new Basel Accord on the supply ofcapital to emerging market economies”, Bank of England, Financial Stability Review,December, pp. 112-116.

Helsinki Protocol (2000), “Protocol relating to the collaboration of the supervisory authorities ofthe Member States of the European Union with regard to the application of Directive 98/78/ECon the supplementary supervision of insurance undertakings in an insurance group”,11 May 2000.

Hoffmann B. (2005) “Procyclicality: the macroeconomic impact of risk-based capitalrequirements”, Financial Markets and Portfolio Management 19(2), pp. 179-200.

Illing M. and Paulin G. (2004), “The New Basel Capital Accord and the cyclical behaviour of bankcapital”, Bank of Canada, Working Paper, 2004-30.

Jackson P., Perraudin W., and Saporta V. (2002), “Regulatory and ‘economic’ solvency standardsfor internationally active banks”, Journal of Banking and Finance, No. 26, pp. 953-976.

Joint Forum (1999), “Supervision of financial conglomerates”, February.

51ECB


REFERENCES

Kashyap A. and Stein J. (2003), “Cyclical implications of the Basel II capital standards”,University of Chicago Working Paper.

Lowe P. (2002), “Credit risk measurement and procyclicality”, BIS, Working Paper No. 116.

Marcelo A. and Scheicher M. (2005), “Capital requirements for corporate default risk”, ECBMimeo.

Nordea (2003), “Nordea reduces complexity in its legal structure by forming one Europeancompany”, Press release, 19 June 2003.

Powell D. E. (2005), “Statement of Donald E. Powell, Chairman Federal Deposit InsuranceCorporation, on the Development of the New Basel Capital Accords before the Committee onBanking, Housing, and Urban Affairs”, 10 November 2005.

PricewaterhouseCoopers (2004), “Study on the financial and macroeconomic consequences ofthe draft proposed new capital requirements for banks and investment firms in the EU,(Markt/2003/02/F), 8 April 2004.

Roesch D. (2005) “An empirical comparison of default risk forecasts from alternative creditrating philosophies”, International Journal of Forecasting, 21(1), pp. 37-51.

Schmidt Bies S. (2005), “Testimony of Governor Susan Schmidt Bies before the Subcommitteeon Domestic and International Monetary Policy, Trade and Technology and the Subcommitteeon Financial Institutions and Consumer Credit, Committee on Financial Services, U.S. Houseof Representatives”, 11 May 2005.

Segoviano M. and Lowe P. (2002), “Internal ratings, the business cycle and capital requirements:some evidence from an emerging market economy”, BIS, Working Paper No. 117.


EUROPEAN CENTRAL BANKOCCASIONAL PAPER SERIES

1 “The impact of the euro on money and bond markets” by J. Santillán, M. Bayle andC. Thygesen, July 2000.

2 “The effective exchange rates of the euro” by L. Buldorini, S. Makrydakis and C. Thimann,February 2002.

3 “Estimating the trend of M3 income velocity underlying the reference value for monetarygrowth” by C. Brand, D. Gerdesmeier and B. Roffia, May 2002.

4 “Labour force developments in the euro area since the 1980s” by V. Genre andR. Gómez-Salvador, July 2002.

5 “The evolution of clearing and central counterparty services for exchange-tradedderivatives in the United States and Europe: a comparison” by D. Russo,T. L. Hart and A. Schönenberger, September 2002.

6 “Banking integration in the euro area” by I. Cabral, F. Dierick and J. Vesala,December 2002.

7 “Economic relations with regions neighbouring the euro area in the ‘Euro Time Zone’” byF. Mazzaferro, A. Mehl, M. Sturm, C. Thimann and A. Winkler, December 2002.

8 “An introduction to the ECB’s survey of professional forecasters” by J. A. Garcia,September 2003.

9 “Fiscal adjustment in 1991-2002: stylised facts and policy implications” by M. G. Briotti,February 2004.

10 “The acceding countries’ strategies towards ERM II and the adoption of the euro:an analytical review” by a staff team led by P. Backé and C. Thimann and includingO. Arratibel, O. Calvo-Gonzalez, A. Mehl and C. Nerlich, February 2004.

11 “Official dollarisation/euroisation: motives, features and policy implications of currentcases” by A. Winkler, F. Mazzaferro, C. Nerlich and C. Thimann, February 2004.

12 “Understanding the impact of the external dimension on the euro area: trade, capital flows andother international macroeconomic linkages“ by R. Anderton, F. di Mauro and F. Moneta,March 2004.

13 “Fair value accounting and financial stability” by a staff team led by A. Enria and includingL. Cappiello, F. Dierick, S. Grittini, A. Maddaloni, P. Molitor, F. Pires and P. Poloni,April 2004.

14 “Measuring Financial Integration in the Euro Area” by L. Baele, A. Ferrando, P. Hördahl,E. Krylova, C. Monnet, April 2004.

53ECB


L I S T O FOCCAS IONAL

PAPERS15 “Quality adjustment of European price statistics and the role for hedonics” by H. Ahnert andG. Kenny, May 2004.

16 “Market dynamics associated with credit ratings: a literature review” by F. Gonzalez, F. Haas,R. Johannes, M. Persson, L. Toledo, R. Violi, M. Wieland and C. Zins, June 2004.

17 “Corporate ‘Excesses’ and financial market dynamics” by A. Maddaloni and D. Pain, July 2004.

18 “The international role of the euro: evidence from bonds issued by non-euro area residents” byA. Geis, A. Mehl and S. Wredenborg, July 2004.

19 “Sectoral specialisation in the EU a macroeconomic perspective” by MPC task force of theESCB, July 2004.

20 “The supervision of mixed financial services groups in Europe” by F. Dierick, August 2004.

21 “Governance of securities clearing and settlement systems” by D. Russo, T. Hart,M. C. Malaguti and C. Papathanassiou, October 2004.

22 “Assessing potential output growth in the euro area: a growth accounting perspective”by A. Musso and T. Westermann, January 2005.

23 “The bank lending survey for the euro area” by J. Berg, A. van Rixtel, A. Ferrando,G. de Bondt and S. Scopel, February 2005.

24 “Wage diversity in the euro area: an overview of labour cost differentials acrossindustries” by V. Genre, D. Momferatou and G. Mourre, February 2005.

25 “Government debt management in the euro area: recent theoretical developments and changesin practices” by G. Wolswijk and J. de Haan, March 2005.

26 “The analysis of banking sector health using macro-prudential indicators” by L. Mörttinen,P. Poloni, P. Sandars and J. Vesala, March 2005.

27 “The EU budget – how much scope for institutional reform?” by H. Enderlein, J. Lindner,O. Calvo-Gonzalez, R. Ritter, April 2005.

28 “Reforms in selected EU network industries” by R. Martin, M. Roma, I. Vansteenkiste,April 2005.

29 “Wealth and asset price effects on economic activity”, by F. Altissimo, E. Georgiou,T. Sastre, M. T. Valderrama, G. Sterne, M. Stocker, M. Weth, K. Whelan, A. Willman,June 2005.

30 “Competitiveness and the export performance of the euro area”, by a Task Force of theMonetary Policy Committee of the European System of Central Banks, June 2005.

31 “Regional monetary integration in the member states of the Gulf Cooperation Council(GCC)” by M. Sturm and N. Siegfried, June 2005.


32 “Managing financial crises in emerging market economies: experience with the involvementof private sector creditors” by an International Relations Committee task force, July 2005.

33 “Integration of securities market infrastructures in the euro area” by H. Schmiedel,A. Schönenberger, July 2005.

34 “Hedge funds and their implications for financial stability” by T. Garbaravicius andF. Dierick, August 2005.

35 “The institutional framework for financial market policy in the USA seen from an EUperspective” by R. Petschnigg, September 2005.

36 “Economic and monetary integration of the new Member States: helping to chart the route”by J. Angeloni, M. Flad and F. P. Mongelli, September 2005.

37 “Financing conditions in the euro area” by L. Bê Duc, G. de Bondt, A. Calza, D. MarquésIbáñez, A. van Rixtel and S. Scopel, September 2005.

38 “Economic reactions to public finance consolidation: a survey of the literature” by M. G.Briotti, October 2005.

39 “Labour productivity in the Nordic EU countries: a comparative overview and explanatoryfactors – 1998-2004” by A. Annenkov and C. Madaschi, October 2005.

40 “What does European institutional integration tell us about trade integration?” by F. P.Mongelli, E. Dorrucci and I. Agur, December 2005.

41 “Trends and patterns in working time across euro area countries 1970-2004: causesand consequences” by N. Leiner-Killinger, C. Madaschi and M. Ward-Warmedinger,December 2005.

42 “The New Basel Capital Framework and its implementation in the European Union”by F. Dierick, F. Pires, M. Scheicher and K. G. Spitzer, December 2005.

Documents

No.42 - The New Basel Capital Framework and its implementation in