the bank lending survey for the euro area

OCCAS IONAL PAPER SER IESNO. 23 / FEBRUARY 2005

THE BANK LENDINGSURVEY FOR THE EURO AREA

by Jesper BergAdrian van RixtelAnnalisa FerrandoGabe de Bondt and Silvia Scopel

In 2005 all ECB publications will feature

a motif taken from the

€50 banknote.

OCCAS IONAL PAPER S ER I E SNO. 23 / F EBRUARY 2005

THE BANK LENDINGSURVEY FOR THE

EURO AREA*

by Jesper Berg 1

Adrian van Rixtel 2

Annalisa Ferrando 2

Gabe de Bondt 2

and Silvia Scopel 2

This paper can be downloaded from the ECB’s website (http://www.ecb.int).

* The authors would like to thank Francesco Drudi, Hans-Joachim Klöckers and Philippe Moutot for providing useful comments atvarious stages of the project as well as an anonymous referee for helpful suggestions. The views expressed in this paper are

those of the authors and do not necessarily reflect those of the European Central Bank or the Danmarks Nationalbank.1 Danmarks Nationalbank.2 European Central Bank.

© 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. Reproduction foreducational and non-commercialpurposes is permitted provided thatthe source is acknowledged.

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. 23February 2005

C O N T E N T SEXECUTIVE SUMMARY 5

1 INTRODUCTION 6

2 THEORETICAL BACKGROUND OF THE BANKLENDING SURVEY FOR THE EURO AREA 7

2.1 Credit literature from a monetarypolicy transmission perspective 7

2.1.1 Interest rate channel:pass-through of official interestrates to bank lending rates 7

2.1.2 Balance sheet channel: balancesheet position of borrowers 8

2.1.3 Bank lending channel: balancesheet position of banks 8

2.2 Credit literature from a businesscycle perspective 10

2.2.1 Euro area 10

2.2.2 United States 10

2.2.3 Lessons learned 12

2.3 Synthesis of views on the banklending process 13

2.4 Other available sources ofinformation 13

2.5 Conclusions 15

3 EXPERIENCES OF OTHER CENTRAL BANKSWITH BANK LENDING SURVEYS 16

3.1 The experiences of the FederalReserve 16

3.1.1 Development and structure 16

3.1.2 Publication of the surveyresults by the Federal Reserve 17

3.2 The experiences of the Bankof Japan 18

3.2.1 Development and structure 18

3.2.2 Publication of the surveyresults by the Bank of Japan 19

3.3 Surveys on bank lending conductedin the euro area 20

4 THE STRUCTURE OF THE EURO AREA BANKLENDING SURVEY 21

4.1 Theory and practice: the designof the questionnaire 21

4.2 The sample group 24

4.3 The method of aggregation 25

4.4 The compilation guide 26

4.5 The presentation of the data 26

4.6 How to read the results of the banklending survey 26

5 THE FIRST RESULTS OF THE EURO AREABANK LENDING SURVEY 28

5.1 The results of the bank lendingsurvey 28

5.1.1 Enterprises 28

5.1.2 Loans for house purchase 32

5.1.3 Loans for consumer credit 35

5.2 Relationship with other indicators 39

5.2.1 Credit standards and realGDP growth 40

5.2.2 Credit standards and MFIloan growth 41

5.2.3 Factors affecting creditstandards and financialspreads 42

5.2.4 Factors affecting creditstandards and industrialconfidence 42

5.2.5 Conditions and terms forapproving loans and corporatebond spreads 42

5.2.6 Factors affecting loan demandand gross fixed capitalformation 43

5.2.7 Factors affecting loan demandand consumer confidence 44

5.3 Conclusions 45

4ECBOccasional Paper No. 23February 2005

6 CONCLUDING REMARKS 46

ANNEXES

1 The questionnaire of the banklending survey for the euro area 47

2 The compilation guide of the banklending survey for the euro area 55

3 The October 2004 Federal ReserveSenior Loan Officer Opinion Surveyon Bank Lending Practices 58

4 The October 2004 Bank of JapanSenior Loan Officer Opinion Surveyon Bank Lending Practices at LargeJapanese Banks 71

REFERENCES 82

5ECB


EXECUTIVESUMMARY

In 2003 the Eurosystem developed a banklending survey for the euro area in order toobtain more detailed information on creditmarkets, and therefore on the role of credit inbusiness cycles and in the transmission processof monetary policy. The survey consists of a setof qualitative questions which, at the beginningof each quarter, will be put to a predefinedsample group of banks located in the 12countries of the euro area. The first survey wasconducted in early 2003, and its results werepublished for the first time in May 2003.

This occasional paper explains why the banklending survey was developed and describes itsmain features. It discusses the importance ofcredit developments for both the economy andthe functioning of monetary policy, and furtherclarifies why the survey was introduced.Furthermore, the paper demonstrates that thevalue added of implementing a bank lendingsurvey for the euro area lies in particular inthe way it provides greater insight intodevelopments in credit standards, non-interestrate credit conditions and terms, the riskperception of banks and the willingness ofbanks to lend. Credit standards are the internalguidelines or criteria of a bank which reflectthe bank’s loan policy. The terms andconditions of a loan refer to the specificobligations agreed upon by the lender and theborrower. This occasional paper also considerssimilar surveys conducted by the FederalReserve System in the US and by the Bank ofJapan.

The three main reasons for the introduction ofthe bank lending survey for the euro area can besummarised as follows. First, it should providemonetary policy-makers in the euro area withmore specific information related to creditconditions such as information on changes incredit standards, credit conditions and terms,and loan demand, for both enterprises andhouseholds. Second, the bank lending surveyshould also provide specific information thatcan only be obtained directly from the lenders –i.e. the banks, for example with regard to thequestion as to whether banks are relying more

EX E CU T I V E S UMMARYheavily on non-price rationing of loans, and ifso why. Third, information derived from thebank lending survey, in particular regardingchanges in credit standards, should help policy-makers to gain a better insight into futureeconomic developments.

Finally, this occasional paper presents theresults of the first eight rounds of the euro areabank lending survey and compares them withinformation collected from other sources. Theanalysis carried out shows that overall, even atthis early stage, it is possible to identify somesystematic patterns in the results from the banklending survey that prove to be in line withindicators retrieved from other data sources.


1 I N T RO D U C T I O NWith all peoples, in all eras, commercial crises were always accompanied by monetary crises.The institution of credit alone thus leads, by its abuse, to the commercial crises. Credit is theprimary and principal motor of this whole mechanism.Clément Juglar, 1889

The Eurosystem – the European Central Bank(ECB) and the national central banks (NCBs)of the 12 EU Member States that have adoptedthe euro – has developed a survey on banklending developments in the euro area, whichwas published for the first time in April 2003.This survey helps the Governing Council ofthe ECB to assess monetary and economicdevelopments in the euro area, and offersimportant input into the monetary policydecision-making process. The surveycomplements existing quantitative statistics onbank retail interest rates and credit withqualitative information on supply and demandconditions in the euro area credit markets andon the bank lending policies of euro area banks.

Credit developments are an importantdeterminant of economic developments, andconditions in credit markets may affect theway monetary policy has an impact on theeconomy. The bank lending survey enhancesthe understanding of the Eurosystem and thepublic on conditions existing for bank lending.

The bank lending survey for the euro areaaddresses issues such as credit standardsfor approving loans as well as credit termsand conditions applied to enterprises andhouseholds. It also asks banks for anassessment of the conditions affecting creditdemand. The survey is addressed to senior loanofficers of a representative sample of euro areabanks, and is conducted four times a year. Thesample group participating in the surveycomprises around 90 banks from all euroarea countries, and takes into account thecharacteristics of their respective nationalbanking structures. This occasional paper aimsto provide detailed insights into this tool andalso to serve as a reference work for thoseinterested in obtaining a deeper understandingof the euro area bank lending survey.

This occasional paper is organised asfollows. Chapter 2 reflects the theoreticalunderpinnings behind the decision to establisha bank lending survey for the euro area.Chapter 3 provides an overview of theexperiences of other central banks with banklending surveys. Chapter 4 describes thestructure of the bank lending survey for theeuro area. Chapter 5 follows with a discussionof the results to date of the survey based on thefirst eight survey cycles (January 2003 –October 2004). Chapter 6 concludes with somefinal remarks. Annexes 1 and 2 contain theactual questionnaire together with thecompilation guide that accompanies it. Finally,Annexes 3 and 4 contain similar surveysconducted by the US Federal Reserve and theBank of Japan respectively.

7ECB


2 THEORETICALBACKGROUNDOF THE BANK

LENDINGSURVEY FOR

THE EURO AREA

This chapter justifies the decision to implementa bank lending survey for the euro area. Themain objective of such a loan survey is toimprove information on the lending behaviourof banks in order to enhance our understandingof the role of credit in the monetarytransmission mechanism and in businesscycles. A better understanding of thesignificance and relevance of creditdevelopments is important for the ECB’smonetary policy. Credit developments mayhave a different implication for the conductof monetary policy, depending on theirdeterminants, especially whether supply ordemand factors are the dominant factors inplace. Changes in the quantity and price ofbank credit depend on a wide range of supplyand demand factors that are not always directlyobservable, such as the prevailing competitiveforces in credit markets, the availability ofalternative sources of finance, the cost andavailability of loanable funds, the value ofcollateral, financing needs in the economy, etc.

This chapter is organised as follows. First, abrief review is provided of the literature on theimportance of credit from the perspective ofmonetary policy transmission (Section 2.1) andfrom a business cycle perspective (Section2.2). Both sections simply outline the mainissues at stake and do not aim at providing acomplete and comprehensive overview oftheoretical and empirical bank lending studies.The interested reader can examine theempirical findings for euro area (countries),particularly regarding the role of banks in thetransmission mechanism and the existence offinancial friction affecting firms’ investmentsas presented in ECB Working Papers 91-114,ECB (2002), Angeloni et al. (2003a and b), andAngeloni and Ehrmann (2003). Section 2.3syntheses both views on the bank lendingprocess. Section 2.4 examines other availablesources of credit information apart from thebank lending survey, focusing on thequalitative nature and subjective views andexpectations of the responses to the banklending survey. Finally, Section 2.5 concludeswith some summarising remarks.

2 TH EOR E T I C A L B A CKGROUND O F T H E B ANKL END I NG S U RV E Y F OR TH E E URO A R E A

2.1 CREDIT LITERATURE FROM A MONETARYPOLICY TRANSMISSION PERSPECTIVE

2.1.1 INTEREST RATE CHANNEL:PASS-THROUGH OF OFFICIAL INTERESTRATES TO BANK LENDING RATES

Monetary policy decisions are transmittedthrough the economy in a variety of ways, all ofwhich eventually affect the evolution of pricesand output. The monetary policy transmissionmechanism is a combination of all theeconomic channels through which, over time,monetary policy affects the economy.1

The traditional macroeconomic textbookpicture of the monetary policy transmissionmechanism contends that central banksinfluence the intertemporal allocation ofresources by changing policy interest rates.However, a central bank controls only the shortend of the yield curve. The reaction of interestrates at longer maturity depends on marketexpectations about the future stance ofmonetary policy.

Commercial banks play a key role intransmitting changes in official rates to banklending rates. The pass-through of interest ratechanges to bank lending rates depends on theinterplay of supply and demand for creditand on the structure of banking markets.Competition in the financial servicesindustry, bank-customer relations, preferencesregarding the maturity of credit contracts orvariability of interest rates, risk premia and theadministrative cost of effectively changinginterest rates are all likely to influence theeffectiveness of monetary policy actions viatheir influence on the bank lending rate pass-through.

1 A more detailed overview of these monetary policytransmission mechanisms can be found in ECB (2000) and(2002).


2.1.2 BALANCE SHEET CHANNEL: BALANCESHEET POSITION OF BORROWERS

In the last few decades, many studies havefocused on credit markets, which play a criticalrole in the transmission of monetary policyactions to the real economy (Angeloni et al.(2003a)).2 The credit view departs from thetraditional macroeconomic textbook picture ofmonetary policy transmission by stressing thatfinancial markets are characterised byimperfections. Agency costs associated withimperfect information between lenders andborrowers or costly monitoring create a wedgebetween the cost of external and internal funds(the external finance premium) and increasethe sensitivity of investment to variables suchas net worth or cash flow. The credit viewdistinguishes between different non-monetaryassets, either along the dimension of bankversus non-bank sources of funds, or along themore general dimension of external versusinternal financing. It also highlights the factthat borrowers are heterogeneous, stressingthat some may be more vulnerable to changes incredit conditions than others. A rise in interestrates may have a much stronger contractionaryimpact on the economy if the balance sheets ofborrowers are already weak, introducing thepossibility of distributional effects of monetarypolicy.

The balance sheet channel, which is alsodenoted as a financial accelerator or “broad”credit channel, emphasises the role of theborrowers’ financial structure in thepropagation of financial and real shocks.3 Afirm’s financial position, which can be derivedfrom its balance sheet, is likely to be animportant factor with regard to the possibilityof obtaining external finance. This is reflectedin the size of the external finance premium. Achange in borrowers’ net worth or collateralmodifies the external finance premium and theoverall terms of credit faced by borrowers.Monetary policy actions can affect borrowers’net worth in different indirect ways. Anexpansionary monetary policy strengthensborrowers’ net worth by a rise in equity, house,

land or other asset prices, or by an increase infirms’ cash flow caused by the decline innominal interest rates.

2.1.3 BANK LENDING CHANNEL: BALANCESHEET POSITION OF BANKS

The bank lending channel, also known as the“narrow” credit channel, is based on twoconditions. First, it assumes that monetarytightening drains liquidity from the bankingsystem: as a result, bank liabilities areshrinking. Because of this decline in totalliabilities, banks would have to adjust totalassets accordingly. Given the imperfectsubstitutability between loans and other assets,the result would be a decline in the supply ofloans. Second, it assumes that the decline in thesupply of loans affects borrowers. Householdsand small firms in particular would lack accessto other forms of credit apart from bank loans.This condition of bank dependence means thatthere is no perfect substitute for bank loans onthe liability side of certain types of borrowers.

Both conditions are still subject toconsiderable debate in the economic literature.Turning to the first condition, Romer andRomer (1990) argue that following a monetarypolicy tightening, banks may issue moneymarket liabilities such as large certificates ofdeposits to offset any drop in deposits. In thisway, monetary policy authorities have limitedcontrol on the liability side of bank balancesheets. Furthermore, if there is indeed a declinein bank liabilities following a monetarytightening, banks may sell liquid assets instead

2 Among many others, see Bernanke and Blinder (1988),Gertler and Gilchrist (1993), Bernanke and Gertler (1995),Hall (1999), de Bondt (2000) and Kakes (2000).

3 For example, a theoretical study by Kiyotaki and Moore(1997) shows how credit constraints interact with aggregateeconomic activity over the business cycle. In particular, thedynamic interaction between credit limits and asset prices ascollateral for loans is a powerful transmission mechanism bywhich the effects of shocks persist, amplify and spill over toother sectors. Other well-known theoretical studies employingthis mechanism in a dynamic context are Bernanke andGertler (1989) and Greenwald and Stiglitz (1993). Thesemodels are able to generate a propagation and amplificationmechanism through which small monetary and realdisturbances have persistent real effects.

9ECB



LENDINGSURVEY FOR

THE EURO AREA

of reducing loan supply to realise the decline intotal assets. Overall, the issuance activity ofmoney market liabilities by banks and theholding of liquid assets may act as a buffer thatinsulates loan supply from monetary policychanges. Regarding the second condition, onemay argue that the dependence on bankfinancing in the euro area economy is decliningbecause of the growing importance of security-based financing. Although only a limitednumber of (large) enterprises have access tofinancing sources based on securities instead ofloans, these firms may help those firms withoutdirect access to capital markets through tradecredit (Kohler, Britton and Yates (2000)). Inother words, financing flows between firmsmay weaken the degree of dependence on bankfinancing in an economy.

Banks also play a key role in observed periodsof sharply increased non-price credit rationing,also known as “credit crunches”. The creditrationing theory contends that some of theborrower’s demand for credit is turned down,even if the borrower is willing to pay the priceof the loan contract (Stiglitz and Weiss(1981)). As a result, bank lending rates do notalways equilibrate the supply of and demandfor bank credit, and the aggregate amount ofcredit is constrained by factors such as non-interest rate credit terms. The underlying factorof a credit crunch may be a change in the riskperception of banks (internal constraint),which in turn could be triggered by a change inmonetary policy, or a shortage of bank capital(external constraint). The latter phenomenon isalso called a capital crunch or squeeze(Bernanke and Lown (1991) and Woo (1999)).Another indication of the independent role ofthe supply of credit in a capital crunch contextis that poorly capitalised banks reduce theirlending much more than better capitalisedbanks (Peek and Rosengren (1995), Altunbaç etal. (2004)).

Another important ingredient of the banklending channel is the willingness of banks tolend, which is closely related to their riskperception as mentioned before. The amount of

credit provided by banks depends on the degreeof uncertainty about the creditworthiness ofborrowers and on the state of bankexpectations. In this respect, it is noteworthythat in post-Keynesian credit view models, notonly asymmetric information between lendersand borrowers is essential, but also asymmetricexpectations between lenders and borrowersand the general state of confidence (Wolfson(1996)). Although borrowers may know morethan lenders and vice versa (“asymmetricinformation”), both lenders as well asborrowers are subject to fundamentaluncertainty about the future. It is notnecessarily the case that both will come to thesame conclusion about the future profitabilityof any particular project. Due to theseasymmetric expectations, borrowers will berationed when projects are deemed safe by theborrower but too risky by the lender. Thecertainty or confidence with which theseassessments are made is also important.

Two states of confidence can therefore bedistinguished. The first is the belief byconsumers and producers in their role asborrowers about prospective returns fromlabour and financial investments andinvestment projects respectively (i.e. thewillingness to consume and the willingness toinvest). The second is the so-called “state ofcredit”, which is governed by the confidencethat lenders have in financing consumption orinvestment expenditures (this is termed thewillingness to lend). In a crisis, a suddenchange in one of these states of confidencefrom strong to very weak is quite typical. Theeconomy only recovers from the crisis whenboth states of confidence have recovered.


2.2 CREDIT LITERATURE FROM A BUSINESSCYCLE PERSPECTIVE

This section reviews the empirical evidence ofthe role of credit in the business cycle for theeuro area and the United States.4 On the onehand, in the business cycle literature, in earlyversions of real business cycle (RBC) models,credit and monetary policy play no role at all.5

On the other hand, monetarist interpretations ofthe business cycle put forward a key role forcredit. One of the originators of credit cycles isJuglar (Niehans (1992)), whose main idea wasthat economic fluctuations originate in creditmarkets. Prices and their interaction with bankcredit were assigned a dominant role in thecyclical mechanism, which consists of threephases. In the “prosperity phase” of thebusiness cycle, which may last from seven tonine years, banks that are confident about thefuture (compare with Keynes’ state of credit –see Keynes (1936)) tend to extend creditrelatively easily, and are unworried about thegradual decline in their liquidity. As aconsequence, prices will rise. During the short“crisis phase”, credit growth shows a steep,virtually vertical drop. After this crisis period,the overblown credit structure is graduallyadjusted to more normal proportions, pricessubside, confidence returns and the stage is setfor a new upswing. This is the so-calledliquidation phase, which lasts two to four yearsand in any case is much shorter than theprosperity stage.

2.2.1 EURO AREA

An in-depth analysis of the role of bank loans tohouseholds and enterprises in business cyclesfor the euro area as a whole has not yet, to thebest of our knowledge, been performed.6 Incontrast, the literature about business cycles inindividual European countries is extensive.These country studies, however, focus eitheron the interaction between national businesscycles and international business cycles or on aparticular theoretical framework, e.g. the RBCtheory.7 Although some studies have addressedthe cyclical properties of money, prices and

interest rates, there is a surprising absence offocus on credit. Of course, national studies forindividual euro area countries have appearedthat provide anecdotal evidence of strongcredit cycles. For example, the real estatelending crisis in the late 1980s and early 1990sin France and the banking crisis of the early1990s in Finland have been investigated indepth by Pazarbasioglu (1997) and Vihriälä(1997) respectively.

2.2.2 UNITED STATES

In contrast to the literature for the euro area,many Federal Reserve studies haveinvestigated the role of credit in businesscycles in the United States. The main findingsfrom these studies are summarised below.These findings are, however, notuncontroversial. For instance, Poole (1993)questions the role of credit in business cyclesand concludes that inflation surprises andrevisions in expectations about future incomeflows drive business cycles.

The first main finding is that credit crunchesare strongly related to fluctuations in economicactivity. Looking at post-war business cyclesin the United States, Eckstein and Sinai (1986)show that every recession since the mid-1950s(that is, six out of the eight post-warrecessions) was preceded and triggered by acredit crunch. In another paper, Sinai (1993)emphasises that credit crunches are part of an

4 For an extensive review of cyclical theories of f inancial crises,see Wolfson (1994). The main differences between the variousf inancial crisis theories concern i) the reasons for thedevelopment of f inancial diff iculties in the business sector,ii) the factor influencing the supply and demand of credit, andiii) the def ining patterns of a f inancial crisis.

5 For instance, see Van Els (1995) for a survey of RBC modelsand the role of money in RBC theory.

6 A study on the properties of business cycles in the euro area byDöpke (1999) ignores the role of credit. Agresti and Mojon(2003) do look, among many other variables, at total privatesector loans. They f ind that loans lag economic activity by upto four quarters.

7 Among many other references, references for the first groupof studies are Ortega (1998) (Spain, 1970-1996), Bergman etal. (1992) (Nordic countries, 1870-1988) and, for the secondgroup of studies, Stanca and Gallegati (1998) (Italy 1861-1992), Fiorito and Kollintzas (1992) (G-7 countries) andKarras (1994) (France, Germany and United Kingdom).

11ECB



LENDINGSURVEY FOR

THE EURO AREA

endogenous process and that each credit crunchdiffers from earlier ones only in its superficialfeatures. Wojnilower (1980 and 1985) alsocontends that credit crunches have been acrucial determinant of the post-war recessionsin the United States. More recently,Wojnilower (1997) argues that even thoughcredit crunches have not triggered a recessionin the United States for a long time, disruptionsin credit markets have continued to play animportant role in business cycles. Schreft andOwens (1995) define a period of a sharpincrease in non-price credit rationing as acredit crunch. They identify four creditcrunches in the United States during the period1960-1992, whereby each credit crunchepisode was preceded by a period of strongcredit demand, relatively rapid growth incredit extensions and rising inflationaryexpectations.

An overview study of the literature by Sharpe(1995) assesses whether a credit crunch in theUnited States in the early 1990s was caused byincreased capital requirements, more stringentregulatory practices, or the widespreaddeterioration of bank balance sheets. At thedisaggregated level, a robust link is foundbetween loan growth and loan performance andbank profitability, though the interpretation ofsuch findings remains ambiguous. Changes incapital standards or regulatory behaviour alsofail to explain convincingly the drop inaggregate lending.

The second main finding is that creditstandards help to predict loan and real GDPgrowth. Asea and Blomberg (1998) show, bylooking at two million commercial andindustrial loans granted by 580 US banksbetween 1977 and 1993, that bankssystematically change their lending standardsfrom tightness to laxity over the business cycle.Their findings suggest that lax lendingstandards, which tend to occur duringexpansions, exert considerable influence onthe dynamics of aggregate business cyclefluctuations.

The notion of a lending standards cycle is alsoanalysed in depth by Weinberg (1995). Hismain point is that there is a natural tendency forlending standards to vary inversely with thelevel of activity in the credit markets.

Lown, Morgan and Rohtagi (2000) find that anet tightening of credit standards – as forexample reported in the Federal Reserve’s“Senior Loan Officer Opinion Survey on BankLending Practices” – is highly negativelycorrelated with aggregate commercial loangrowth. Furthermore, it is also correlated withother measures of credit availability such asloan spreads, the relative importance ofcommercial paper with respect to bank loans,and the spread between interest rates on non-financial commercial paper and Treasury bills.They also find that reported changes in creditstandards can predict narrower measures ofbusiness activity, including inventoryinvestment and industrial production. Finally,using the vector autoregression (VAR)approach, these authors conclude that banks settheir credit standards based largely on theirown lending capacity coupled with theirexpectations, so that credit standards seem tobe relatively exogenous compared with othermacroeconomic variables included in thesystem.8

In later studies, Lown and Morgan also use theVAR methodology to investigate the causalitybetween credit standards, bank lending andeconomic output (Lown and Morgan (2002 and2004)). The results indicate that shocks tocredit standards have a significant impact onboth commercial loan volumes and real output– in other words, there is a strong correlationbetween loan officers’ reports of tighter creditstandards, as evidenced by the results fromthe Senior Loan Officer Opinion Survey, and

8 The VAR approach is based on the assumptions that all theeconomic and f inancial variables which are used in theempirical investigation are endogenous and that each can bewritten as a linear function of its own lagged values and thelagged values of all the other variables in the system, wherethe number of lags is to be determined somehow (Kennedy(1998), p.168). It is nowadays widely used for empiricalmacroeconomic research.


observed slowdowns in commercial lendingand output. Changes in these standards lead tochanges in output and bank loans, while pastvalues of bank loans cause changes in creditstandards. And importantly, when creditstandards are taken into account, thesignificance of changes in the Fed’s policyinterest rate – i.e. the federal funds rate – onoutput is reduced.

All but one of the recessions in the US werepreceded by a high net percentage of loanofficers reporting a tightening of creditstandards. The only exception – the 1982recession – was preceded by a sharp shiftupward in the net percentage of banks reportinga tightening of credit standards, from a neteasing of credit standards toward a nettightening. Tighter credit standards are usuallyfollowed by a slowdown in the growth ofcommercial loans.

This confirms the earlier finding of Schreft andOwens (1991). They constructed an indicatorthat measured the net (un)willingness of banksto lend, which showed that a tightening ofcredit standards occurs before or duringrecessions. Changes in commercial creditstandards also help to predict narrowermeasures of economic activity such asindustrial production and inventoryinvestment, of which the latter is a notoriouslyunpredictable variable. Following a tighteningof credit standards, loans decline sharply andoutput falls.

The third and final main finding is that banks’willingness to lend has an impact on economicactivity. Lown (1990) explores the bankingindustry’s role in the economy and findsevidence to support the idea that changes in thecomposition of banks’ asset holdings do tend topredict changes in economic activity. Forexample, the ratio of security holdings to totalassets strongly predicts economic growth.These relationships can be explained either bychanges in banks’ willingness to lend or bychanges in firms’ willingness to borrow. Ducaand Garrett (1995) find that a proxy for non-

interest rate credit conditions significantlyaffects bank consumer lending. The proxy usedis an index of the change in banks’ willingnessto offer consumer instalment loans, based onthe Federal Reserve Board’s bank lendingsurvey. This proxy for the willingness to lendsubstantially affects durable consumerspending. Schreft and Owens (1991) show thata decreased willingness to lend, proxied againby a measure based on the Federal ReserveBoard’s bank lending survey, occurs eitherbefore or during recessions.

2.2.3 LESSONS LEARNED

What can be learned from the US findings? Ingeneral, two main conclusions can be drawnfrom the above.

The first is that credit crunches can be veryharmful for the economy. This implies that allpotentially useful information about whethersupply or demand factors play a key role ingranting bank credit should be closelymonitored and assessed. It is exactly in thiscontext that bank lending survey responses,which improve insight into bank creditstandards and conditions and the demand forbank loans as perceived by banks, might help tosolve the credit supply versus demand puzzle.For example, Suzuki (2004) uses the lendingattitude of banks to identify the demand for andsupply of bank loans in Japan.

The second lesson is that the US evidenceshows that responses to the bank lendingsurvey contain valuable information aboutfuture economic growth and loan growth. Thisapplies especially to the tightness in creditstandards. More generally, the US findingsillustrate that subjective survey measures ofbanks’ views and expectations could provideaccurate and meaningful indicators for loanand economic growth. They confirm thepotential usefulness of qualitative vis-à-visquantitative data, as for instance shown byManski (2004).

13ECB



LENDINGSURVEY FOR

THE EURO AREA

2.3 SYNTHESIS OF VIEWS ON THE BANKLENDING PROCESS

This section provides an overview of theemerging picture of the relevant determinantsof bank lending, together with an overview ofthe different monetary transmission channelsand the relationship between creditdeterminants.

Figure 1 above illustrates in a simplifiedway the interaction between the maindeterminants of the supply of and demand forbank credit. For the sake of simplicity, itabstracts from many interactions between thekey determinants of credit cycles. To take butone example, the interactions between thewillingness to consume/invest and to lend arenot included.

The supply of bank lending depends on theability and willingness of banks to lend, which,in turn, is affected by various monetary

transmission channels. In this respect,essential factors are bank lending rates(interest rate channel) and non-price creditconditions and terms other than the interest rate(bank lending and balance sheet channel).More generally, the different monetarytransmission channels affect the creditstandards for approving credit and the non-interest rate credit conditions and terms. Inturn, financing needs of consumers andproducers and thus the demand for banklending depend on the ability and thewillingness to consume and invest. Thedemand of bank lending also depends on, forinstance, bank lending rates.

2.4 OTHER AVAILABLE SOURCES OFINFORMATION

Table 1 provides an overview of the relationsbetween credit determinants and monetarytransmission channels and, to the best of

Credit standards:Price and non-priceconditions and terms

Bank lendingchannel

Interest rate channel

Ability andwillingness to lend

Bank lending supply

Bank lending demand

Balance sheetchannel

Banks

Households

Ability and willingnessto consume

Ability and willingnessto invest

Enterprises

Monetary policy

Figure Determinants of bank lending


our knowledge, all available sources ofinformation on credit determinants apart fromthe bank lending survey.

With regard to credit standards, insight intocompetition faced by banks from other banks,non-banks or capital markets and the riskperception of banks is not readily availablefrom other sources, or if so, only in the form ofindirect indicators. The potential value addedof qualitative responses from the bank lendingsurvey is therefore generally high.

Turning to credit conditions and terms, data arein particular available on the interest ratechannel. For example, several bank lending ratesare available for the euro area, whereas

information on the maturity structure can beextracted from the monetary financial institution(MFI) balance sheet data. Data on non-interestrate credit conditions and terms are less widelyavailable, in particular information on the sizeand non-interest rate charges of credit lines andthe risk perception of banks. Since late 2003,though, statistics on MFI interest rates (i.e. MFIinterest rate statistics) have provided insightsinto the size of non-interest rate charges appliedto loans to households for house purchase and forconsumer credit in the euro area (ECB (2003)).Current insight into expected bank lendingbehaviour is typically only based on anecdotalevidence. The potential value added of the banklending survey is generally high, particularly fornon-interest rate credit conditions and terms and

Table 1 Overview of relat ionships between credit determinants, monetary transmiss ionchannels and sources of credit information

Credit determinants Transmission channel Sources of creditinformation other than thebank lending survey

Credit standards

Competition from other banks, non-banks or from market Interest rate and bank lending Indirect indicators offinancing channel competition

Balance sheet position of banks (costs related to bank Bank lending channel Bank balance sheetscapital position, bank liquidity position, access to marketfinancing)

Degree of bank dependence Bank lending channel Relative importance of bankloans

Risk perception/tolerance of banks (risk to general, Bank lending and balance Risk proxies, e.g. corporatehousehold or firm-specific economic outlook, risk sheet channel bond spreads, loan-to-valueon collateral demanded) ratio

Credit conditions and terms

Maturity structure Interest rate channel Bank balance sheetsLoan rates, bank margins Interest rate and Retail bank rates/MIRExternal finance premium Balance sheet channel statistics

Proxies for rates on externaland internal finance

Balance sheet position of borrowers (creditworthiness, Balance sheet channel Net worth of borrowers basedcollateral requirements and quality) on asset prices, income, cash

flow, etc.

Size and non-interest rate charges of credit lines Bank lending channel MIR statistics, anecdotalevidence

Expected lending behaviour

Willingness to lend, confidence of banks Bank lending and balance Anecdotal evidencesheet channel

(Expected) demand for loans

Ability and willingness to consume and invest Interest rate channel Real sector variables andconfidence indicators

15ECB



LENDINGSURVEY FOR

THE EURO AREA

expected bank lending behaviour. The banklending survey provides timely additionalqualitative data from a lender and from a loansupply perspective.

Finally, with regard to demand for loans, awide range of real sector variables, capturingthe ability to consume or invest, are available.Confidence indicators for the real sector, whichreflect the willingness to consume or invest,are also available. Two widely-usedconfidence indicators for the euro area are theEC consumer and producer confidenceindicators and the Purchasing Managers’Index, which are all based on monthly surveys.The potential value added of the bank lendingsurvey here is the timely availability of insightsinto expected loan demand from a lenderperspective.

2.5 CONCLUSIONS

This section has provided a theoreticalunderpinning for the bank lending survey in theeuro area. Ample empirical evidence for theUnited States shows that most post-warrecessions have been preceded or triggered by asharp fall in credit or by a credit crunch.Furthermore, credit standards and terms andthe willingness of banks to lend tend to affecteconomic activity. However, similar stylisedfacts of credit cycles in the euro area could notbe detected, mainly due to a lack of adequatedata that would permit the identification ofbank loan supply and demand effects.

An overview of the relationship between creditdeterminants and monetary transmissionchannels and the available sources ofinformation on credit determinants other thanthe bank lending survey shows that the lattercould potentially improve insight into thelending behaviour of banks in the euro area and,through this, into the role of credit in themonetary transmission process and in businesscycles as well. The potential value added of aeuro area bank lending survey is that it providesimproved insight into credit standards, non-

interest rate credit conditions and terms, therisk perception of banks and the willingness ofbanks to lend. Responses to the euro area banklending survey might particularly help inidentifying bank loan supply versus demandand in more effectively extracting informationfrom qualitative data, i.e. the subjective viewsand expectations of the major euro area banksvis-à-vis quantitative data.


Given the important role that credit plays in theeconomy, as discussed in the previous chapter,the bank lending survey for the euro area hasbeen developed to provide more specificinformation on credit conditions for bothenterprises and households. Related to this,two of the world’s other leading central bankshave developed their own surveys with rathersimilar objectives. Experiences with banklending surveys conducted by the FederalReserve System in the United States and by theBank of Japan have shown that they canprovide important additional information forthe assessment of past and future developmentsin credit markets. This chapter predominantlyfocuses on the experiences with bank lendingsurveys in the United States, and brieflydiscusses the much more recent correspondingsurvey in Japan. It ends with a short assessmentof the use of bank lending surveys in individualeuro area countries.

3.1 THE EXPERIENCES OF THE FEDERALRESERVE

The Federal Reserve (or Fed for short)publishes a quarterly bank lending survey thatis officially entitled the “Senior Loan OfficerOpinion Survey on Bank Lending Practices”.As the Federal Reserve is the central bank withthe longest experience in conducting a banklending survey, particular attention will be paidto its experiences. This section will discuss thedevelopment and structure of the survey andthe publication of its results.

3.1.1 DEVELOPMENT AND STRUCTURE

The Senior Loan Officer Opinion Survey wasintroduced for the first time in 1967. Sincethen, the basic structure of the survey in termsof regular questions and possible answers hasbeen changed several times, reflectingchanging concerns about specific credit issues.During the 1980s, the focus of the surveyshifted from conjectural analysis to morestructural issues. This direction was reversedin the early 1990s, apparently because concerns

about a credit crunch had become more acute.Thus, most of the present questions weredeveloped at that time and have provided ratheruseful time series since then, and indeedconstitute the core framework of the currentsurvey. In practice, since the survey haschanged a number of times, the analysis basedon its results is somewhat constrained. Forinstance, in 1981 the original sample of bankswas reduced and the set of regular questionscut from 22 to six, to allow for the inclusionof more ad hoc questions on currentdevelopments. Moreover, in 1984, the questionon commercial credit standards for businessloans was dropped and only reintroduced with adifferent wording at the beginning of 1990.This means that the most suitable informationthat can be used for empirical studies, i.e.covering the longest time horizon and availablefor a stable sample group, therefore relates tothe questions on credit standards.

The survey is generally conducted on aquarterly basis and in such a way that its resultsare available for the January, May, August andNovember meetings of the Federal OpenMarket Committee. The Federal Reserve hasthe authority to conduct up to two additionalsurveys during the year. For example, in March2001 the Fed conducted a supplementarysurvey to assess changes in lending conditionssince the beginning of that year.9

The Fed’s bank lending survey is a combinationof a set of regular questions and variousadditional ad hoc questions, which addresscurrent topics of interest regardingdevelopments and trends in credit markets.These topical questions have covered a varietyof issues in the past, such as banks’participation in secondary loan markets,changes in the credit quality of commercial realestate loans, and activities of participatingbanks in the credit default swap market.Overall, the Fed adopts a rather flexibleapproach with respect to the structure of its

3 EXP ER I EN C E S O F OTH ER C EN TR A L B ANK SW I TH B ANK L END I NG S U RV E Y S

9 Senior Loan Off icer Opinion Survey, March 2001; seewebpage Federal Reserve Board (http: www.federalreserve.gov/boarddocs/SnLoan-Survey/200103/default.htm).

17ECB


3 EXPERIENCESOF OTHER

CENTRAL BANKSWITH BANK

LENDING SURVEYS

bank lending survey, which allows it to acquirerather detailed information on specific issues,if deemed necessary. In contrast to the morequantitative survey on loan rates that the Fedalso publishes, the bank lending survey isqualitative.

The current sample size of the survey is about60 domestic banks, usually the largest in eachof the 12 Federal Reserve Districts. Banks areadded or replaced as needed. As a result ofmerger activities involving some of the largestUS banks, the sample has been adjusted ratherfrequently in recent decades. In addition, anumber of branches and agencies of foreignbanks located in the United States are coveredas well (results for the latter are publishedseparately).

The implementation of the survey is theresponsibility of the Research Departments ofthe district Federal Reserve Banks involved.The district banks send the questionnaire to theindividual banks and provide explanations andsupport whenever there are new questions orparticipants. After receiving the individualresponses, they send them to the FederalReserve Board.

The Federal Reserve’s bank lending surveycovers a wide range of types of loans, suchas commercial real estate loans, residentialmortgage loans and credit card loans. All inall, the survey seems to play an importantpart in forming the Fed’s understanding ofdevelopments in US credit markets, and itsresults are widely reported in the US and in theinternational financial media. Based on theresults of various research papers, it isgenerally considered a good early leadingindicator and is the prime indicator for thepossible occurrence of credit crunch-typesituations (see sub-Section 2.2.2).

3.1.2 PUBLICATION OF THE SURVEY RESULTS BYTHE FEDERAL RESERVE

The Federal Reserve publishes a full report onthe survey’s results on the internet, and hard

copies can be obtained as well.10 For questionson changes “over the past three months”, whererespondents can choose among variouscategories (tightened considerably, tightenedsomewhat, remained basically unchanged,eased somewhat, and eased considerably),the Fed publishes the absolute number ofrespondents as well as percentage values foreach category. The presentation of these resultsfollows a classification that distinguishesbetween “all respondents”, “large banks” and“other banks”.

For questions where banks are required tochoose among various factors to explainchanges in their lending policies as well aschanges in loan demand, the Federal Reservereports the simple averages of the totalresponses after having assigned a numberbetween 1 and 5 to some questions (usinga scale where 1 = “tightened considerably”and 5 = “eased considerably”), and for otherquestions a number between 1 and 3 (where1 = “not important” and 3 = “very important”).

Additionally, the Federal Reserve provides sixindicators based on the results of the survey(see Box 1). The indicators are constructedon the basis of balances, i.e. net percentages.For instance, in the case of the net percentageof respondents reporting tightening standardsin loans, the balance is calculated as thedifference between the number of respondentsreporting “tightened considerably” or“tightened somewhat” and those reporting“eased considerably” or “eased somewhat” as apercentage of all respondents. Therefore, thelarger the difference, the greater the nettightening of credit standards. For theindicators representing the results forcommercial and industrial loans, the netpercentages of answers for both large andmedium-sized firms and small firms are shown.With regard to households, the correspondingindicators show the results for various types ofloans, including credit cards and other loans,residential mortgages and consumer loans.

10 See http://www.federalreserve.gov/boarddocs/SnLoanSurvey/.


With respect to the use of the survey’s results,the Federal Reserve has started to link theresponses of individual banks to specificquestions, which are not available to thegeneral public, to various other bank-specificfinancial indicators. One recent example of theuse of individual responses is the study byBassett and Carlson (2002) on the developmentof US commercial bank profits in 2001. Thisseems to be the first publication of the FederalReserve to contain an analysis based onindividual bank responses to the Senior LoanOfficer Opinion Survey that have been linkedto other reported data on individual banks. Inparticular, the authors identified the sourcesbehind the slowdown in commercial andindustrial loans at the individual bank level bydistinguishing between supply and demandfactors.

3.2 THE EXPERIENCES OF THE BANKOF JAPAN

The bank lending survey implemented by theBank of Japan (BoJ) is called the “Senior LoanOfficer Opinion Survey on Bank LendingPractices at Large Japanese Banks”. Thissection will discuss its development andstructure and the publication of its results. Asthe BoJ bank lending survey is a relativelyrecent product, little information is publicly

available with respect to its use for analyticalpurposes.

3.2.1 DEVELOPMENT AND STRUCTURE

The BoJ’s bank lending survey was introducedin March 2000 and is, as has beenacknowledged publicly, broadly modelled onthe Federal Reserve’s Senior Loan OfficerOpinion Survey (see Hida et al. (2002)). TheBoJ’s survey aims to measure the views ofsenior loan officers at large Japanese banksregarding loan market developments by meansof a multiple choice questionnaire. The surveyis conducted quarterly in January, April, Julyand September (Bank of Japan (2000)). TheBoJ has the authority to implement additionalsurveys and last used this possibility in October2004 when, in addition to the regular survey, itdistributed an ad hoc survey on loans byrespondents to firms that had been downgradedor upgraded according to the banks’ internalcredit rating systems during the past threemonths, following the categorisation of firmsaccording to size.

Compared with the Fed’s survey, differencesexist in terms of the specific items that arecovered by individual questions, and the timedimension. For example, with regard to thelatter, the BoJ’s survey is somewhat moreforward-looking. Furthermore, the regular BoJ

Box 1

INDICATORS REPRESENTING THE RESULTS OF THE FEDERAL RESERVE’S BANK LENDING SURVEY

Measures of supply and demand for commercial and industrial (C&I) loans, by size offirm seeking loanNet percentage of domestic respondents tightening standards for C&I loansNet percentage of domestic respondents increasing spreads of loan rates over banks’ costs offundsNet percentage of domestic respondents reporting stronger demand for C&I loans

Measures of supply and demand for loans to householdsNet percentage of domestic respondents tightening standards on consumer loansNet percentage of domestic respondents reporting stronger demand for loans to householdsNet percentage of domestic respondents tightening standards for mortgages to individuals

19ECB


3 EXPERIENCESOF OTHER

CENTRAL BANKSWITH BANK

LENDING SURVEYS

bank lending survey has fewer questions (only13) than its counterpart at the Federal Reserve.These questions are relatively evenly splitbetween questions on the demand for loans andquestions on the supply of loans or lendingpolicies. The latter cover in particular itemssuch as changes in credit standards forapplications from firms and households,changes in the terms and conditions of lending,and changes in the spreads of loan rates over thebanks’ costs of funds.

The survey sample consists of 50 large privatebanks in terms of lending volume. The BoJ haspublicly announced that it revises the list ofbanks included in the sample group every threeyears and did so for the first time in April 2003(Bank of Japan (2003)). This is important,given the recent important changes in thestructure of the Japanese banking system(see Van Rixtel et al. (2004)). The aggregatedamount of loans covered by these banksaccounted for approximately 74% of theaverage amount of outstanding domestic loansof Japanese private banks in 2003, comprisingloans by city banks, long-term credit banks,trust banks, regional banks, second tierregional banks and shinkin banks (see VanRixtel (2002)).11 The participation of the latterthree groups of banks, which consist ofrelatively small banks, is particularlyimportant, as their responses are a goodindication for credit developments related tosmall and medium-sized firms, since thesefinance their activities predominantly throughbank loans.

The BoJ sends the questionnaire by mail at theend of the month prior to the survey, andcompiles the results by the end of the surveymonth (Bank of Japan (2000)). The BoJ’sFinancial Markets Department is in charge ofimplementing the survey and sends thequestionnaire directly to the participatingbanks.

The BoJ’s bank lending survey enables it tocompare changes in bank lending policies withinformation obtained from other surveys that

reflect changes in the financing needs of thenon-financial corporate sector, particularly itsshort-term economic survey of enterprises inJapan or Tankan (Hida et al. (2002)). Thus, bycombining the results from its bank lendingsurvey and the Tankan survey, the BoJ hasinformation from both the providers andborrowers of funds, reflecting developments inthe demand for and supply of loans from bothsides.

3.2.2 PUBLICATION OF THE SURVEY RESULTS BYTHE BANK OF JAPAN

The BoJ publishes the results of its banklending survey both on the internet and inhard copy format.12 For questions whererespondents can choose among five categoriesof answers (such as “substantially stronger”,“moderately stronger”, “about the same”,“moderately weaker” and “substantiallyweaker”), the absolute number of respondents,the percentage values for each category and thevalue of the diffusion index (DI) are presented.With the release of the April 2001 survey, theBoJ started to present the results of the surveyalso in the form of DI values for both thecurrent and previous survey (Bank of Japan(2001)). Diffusion indices give a weightedaverage of the results of a particular question,whereby different answers are weighteddifferently. For example, for several questionson loan demand the DI value on a scale from-100 to 100 is calculated as “(the percentage ofrespondents selecting “substantially stronger”+ the percentage of respondents selecting“moderately stronger” * 0.5) – (the percentageof respondents selecting “substantiallyweaker” + the percentage of respondentsselecting “moderately weaker” * 0.5)”.

11 Shinkin banks are regional f inancial institutions operating as aco-operative f inancial institution, relatively similar to co-operative savings banks.

12 See http://www.boj.or.jp/en/stat/stat_f.htm.


For questions where banks are required tochoose among various factors to explaintheir lending behaviour, the tables are similarto those used by the Federal Reserve, andreport simple averages based on a three-pointscale where 1 = “important”, 2 = “somewhatimportant” and 3 = “not important”.

3.3 SURVEYS ON BANK LENDING CONDUCTEDIN THE EURO AREA

According to the individual countries thatcollectively form the Eurosystem (i.e. the ECBand the 12 euro area NCBs), no euro area NCBhas yet conducted a full-scale and formalqualitative bank lending survey, in the form ofa fixed set of questions which are asked withina certain time period to a fixed group of banks.

In the case of the Eurosystem, such informationfor the euro area as a whole has been lacking inthe past. While the information channels at thenational level are still in place, these aredifficult to utilise in the Eurosystem in theabsence of a standardised structure for thecollection of information. In addition, thenumber of banks that operate on a cross-borderbasis and see the whole euro area as theirdomestic market seems to be increasing. Forthis purpose, it was deemed extremely useful tostipulate more formal procedures for thecollection of this type of information in theform of a bank lending survey submitted to euroarea banks.

21ECB


4 THE STRUCTUREOF THE EURO

AREA BANKLENDING SURVEY

This chapter describes the actual design ofthe euro area bank lending survey, which to alarge extent was based on the theoreticalconsiderations presented in Chapter 2 and thepractical experiences described in Chapter 3.The most important part of the bank lendingsurvey is the questionnaire. Other importantelements include the sample of banks thatparticipate in the survey, the method ofaggregation of the answers, the protection ofconfidentiality, the compilation guide, thepresentation of the results and the way to readthe results.

4.1 THEORY AND PRACTICE: THE DESIGN OFTHE QUESTIONNAIRE

From the start of the single monetary policy inJanuary 1999 up until the introduction of theeuro area bank lending survey in 2003, themonetary policy analyses of the Eurosystemconcerning credit conditions in the euro areahad been predominantly based on a set ofquantitative data included in the officialMoney and Banking Statistics, reflectingoverall developments in the euro area creditmarkets. These data provide the Eurosystemwith information on aggregate credit marketdevelopments. However, they do not permit anyinvestigation of supply and demand conditionsin the euro area credit markets in general and ofthe bank lending policies of individual banks inthe euro area in particular, for which qualitativesurvey data would be more appropriate.

With regard to the availability of data on creditmarkets in the euro area, the assessment of thebanking sector’s general lending behaviourforms a very relevant part of monetary policydecision-making, given its important role inthe transmission process of monetary policy.This process has been described in detail inChapter 2. Information on issues related tobanks’ lending policies, such as creditstandards for approving loans, the willingnessof banks to lend and credit conditions andterms, is of considerable importance inassessing the relevance of the various

transmission channels. Credit standards are theinternal guidelines or criteria of a bank whichreflect the bank’s loan policy. The terms andconditions of a loan refer to the specificobligations agreed upon by the lender and theborrower.

Furthermore, as also discussed in Chapter 2,some general euro area-wide indicators existwith respect to the demand for creditconcerning household and industrialconfidence that can give some indication ofcredit demand conditions. However, banks’assessments of demand conditions, notably thepurposes for which enterprises seek credit,are not covered specifically in any otherinformation sources apart from the banklending survey. Since banks monitor andassess credit markets on a continuous basis,they should be qualified to analyse thedevelopments and reasons behind them.Indeed, for the monetary policy analyses ofdevelopments in the euro area credit aggregatesand credit markets, it would be instrumental toobtain more information on the possible factorsthat underlie these developments.

It is clear that the Eurosystem is well informedabout the behaviour of its respective bankingsystem, and that it uses this information in itsregular assessments of the economic andmonetary situation. In many cases, thisinformation is gained from regular contactswith the national banking sectors in the euroarea. These information links are in many casesinformal, but they remain valuable in thatthey provide the NCBs with information abouttheir banks’ assessments of developments incredit markets. The euro area bank lendingsurvey adds value to this process by providinga more formalised instrument to underpinthe discussions in the Eurosystem ondevelopments in credit markets at the euroarea-wide level.

Thus, overall, based on the theoretical insightspresented in Chapter 2, the main motivation forthe actual design of the bank lending survey canbe summarised into three main points. First, the

4 TH E S T RU C TUR E O F T H E E URO A R E A B ANKL END I NG S U RV E Y


bank lending survey should provide euro areamonetary policy-makers with more specificinformation related to credit conditions, suchas information on changes in credit standards,credit conditions and terms, and loan demand,for both enterprises and households. Thisinformation is of crucial importance inimproving understanding of the transmissionprocess of monetary policy. For example, thevalue added of the bank lending survey withrespect to banks’ credit standards shouldimprove knowledge of the balance sheet andbank lending channels. The survey should alsogenerate more specific information on creditconditions and terms, which had been lesswidely available before the introduction of thesurvey, particularly regarding non-interest ratecredit conditions and terms such as the size andnon-interest rate charges of credit lines,collateral requirements and loan covenants.Second, the bank lending survey should alsoprovide specific information that can only beobtained directly from the lenders – i.e. thebanks, for example related to the question as towhether banks are relying more heavily on non-price rationing of loans and if so, why. Third,information derived from the bank lendingsurvey, in particular regarding changes incredit standards, should help provide policy-makers with greater insight into futureeconomic developments.

The actual design of the euro area bank lendingsurvey takes these considerations fully intoaccount. The questionnaire covers both loandemand and supply, but focuses more on loansupply developments in order to fill the gapbetween the quantitative data on credit marketsalready available and the data requirementsobserved in Chapter 2. Particular attention ispaid to issues such as credit standards, non-interest rate conditions and terms, and thewillingness of banks to lend, an approach thatis comparable to the orientation of the loansurveys of the Federal Reserve and the Bank ofJapan. This also facilitates internationalcomparisons. Furthermore, a limited numberof questions are formulated in terms of

expectations, given the need for monetarypolicy to be forward-looking.

The actual questionnaire of the euro area banklending survey is included in Annex 1. Itconsists of 18 regular questions (including oneopen-ended question), and is classifiedaccording to loans to the two specific sectorsthat are the central focus of the survey, i.e.“loans or credit lines to enterprises” and “loansto households”.

The specific classification of the questionnairehas resulted in the following structure (seeBox 2). There are ten questions on creditstandards applied to new loans or credit linesto enterprises and households, and sevenquestions on demand for loans or credit lines toenterprises and households. For enterprises,the general questions on developments in pastand future credit standards and demand aresubdivided into a sectoral breakdown (smalland medium-sized enterprises versus largeenterprises) and a maturity breakdown (short-term loans versus long-term loans). Thequestions on why and how standards havechanged for enterprises do not contain thesesubdivisions. For households, the questions aresplit between similar questions on loans forhouse purchase and on new loans for consumercredit and other lending. This split has beenmade to capture the important contribution thathousing loans make in the overall creditprovision by banks in many euro area countries.

Thus, there are ten questions that pay attentionto supply factors: seven on credit standardsand three on conditions and terms of newloans. Seven questions focus on loan demand.Both backward-looking and forward-lookingquestions are included, in order to capture bothdevelopments that have taken place andexpectations regarding future developments incredit markets (13 questions are backward-looking and four forward-looking).

23ECB




Box 2

THE STRUCTURE OF THE QUESTIONNAIRE

• Questions on credit standards applied to loans or credit lines to enterprises and households(ten questions), which can be broken down as follows:• Changes in credit standards over the past three months (two questions). These questions

cover, in the case of loans or credit lines to enterprises, both small and medium-sizedenterprises and large enterprises, and short and long-term loans. In the case ofhouseholds, they address loans for house purchase and consumer credit and other lending.

• Factors which have affected changes in credit standards over the past three months (threequestions). The factors taken into account are the cost of funds and balance sheetconstraints, pressure from competition, perception of risk and the risk on the collateraldemanded (the latter only in the question on loans or credit lines to enterprises).

• Conditions and terms for approving loans or credit lines over the past three months (threequestions). These questions investigate changes in pricing, such as margins on averageand riskier loans, and other conditions and terms, which include non-interest ratecharges, collateral requirements, maturity, the size of the loan or credit line and loancovenants (the latter two for enterprises only) and the loan-to-value ratio (the question onloans to households for house purchase only).

• Expected changes in credit standards over the next three months (two questions). Similarcoverage as the questions addressing developments in credit standards over the past threemonths.

• Questions on demand for loans or credit lines to enterprises and households (sevenquestions), which can be broken down as follows:• Demand for loans or credit lines over the past three months by enterprises and households

(two questions). These questions cover, in the case of loans or credit lines to enterprises,both small and medium-sized enterprises and large enterprises, and short and long-termloans. In the case of households, they address loans for house purchase and consumercredit and other lending.

• Factors which have affected changes in demand over the past three months (threequestions). The factors taken into account are financing needs, debt restructuring (only inthe question on enterprises) and the use of alternative finance.

• Expected changes in demand for loans or credit lines to enterprises and households (twoquestions). Similar coverage as the questions addressing developments in demand forloans or credit lines over the past three months.

• One open-ended question

The questionnaire generally uses a scale withfive possible answers. As an example, theoptions in the first question (“Over the past threemonths, how have your bank’s credit standardsas applied to the approval of loans or credit linesto enterprises changed?”) range from “creditstandards eased considerably” to “tightenedconsiderably”. The more moderate options are

“eased somewhat” or “tightened somewhat”,while the neutral option is “remained basicallyunchanged”. For questions related to demand,“tightened” and “eased” are replaced by“decreased” and “increased”. For questions onwhy standards or demand have changed, theanswers relate to whether a specific factor hascontributed “considerably” or “somewhat” to a


particular development. It should be noted thatthese questions relate to how a specific factor,say costs related to a bank’s capital position,have contributed to developments in creditstandards. Thus, one could envisage that costsrelated to a bank’s capital position havecontributed considerably to tightening in asituation where a bank’s credit standards haveeased considerably owing to other factors.13

The questions on why credit standards ordemand have changed and the questions on howcredit standards have changed contain thesame possible general factors independentof whether the questions relate to loans toenterprises, or to households for housepurchases or consumer credit and otherlending. Perception of risk, for example, is apossible explanatory factor for why creditstandards have changed, independent of whatthe loan is intended for. However, the specificrisk factors can differ. The industry or firm-specific outlook is a factor for loans toenterprises, but not for loans to households.Likewise, housing market prospects representa very relevant risk factor for loans tohouseholds for house purchase.

4.2 THE SAMPLE GROUP

In setting up the sample of banks participatingin the bank lending survey, the Eurosystemtook into consideration the qualitative natureof the information it provides, the voluntarybasis of the participation of banks and theneed to capture sufficiently the specifics ofthe banking system in each Member State.Another crucial concern was to ensure theconfidentiality of the information provided bythe individual commercial banks.

As the bank lending survey is an opinion surveyaimed at collecting qualitative data, it does notneed to obey the same statistical requirementsfor coverage and representativity as otherstatistics. Nevertheless, the results of the banklending survey need to reflect broadly thesituation in the euro area as a whole.

As a consequence, the sample group forthe bank lending survey reflects pragmaticconsiderations, rather than a more sophisticatedsampling theory. The banking structures acrossindividual countries in the euro area differgreatly. In some countries, the market isdominated by just one bank or a handful ofbanks, whereas in others there are hundreds, andin some cases even thousands, of banks.Furthermore, it was also deemed useful toproduce survey results at a national level.Against this background, two steps wereenvisaged: first, the NCBs selected theirnational samples, and second, the nationalresults were adjusted to obtain the euro arearesults. The sample group consists of varioustypes of banks that are deemed representative oftheir national banking systems by the NCBs.For example, certain savings banks are includedin a number of countries. To protect theconfidentiality of the banks participating in thesurvey, more detailed information about thesample group’s characteristics cannot beprovided. Since the start of the survey in 2003,the sample group has consisted of 86 banks andcovers approximately 40 percent of euro areabank lending. The coverage is greater than thatof the US survey, but less than that ofthe Japanese survey. The distribution acrosscountries in the euro area is shown in Table 2.While the distribution in some cases issomewhat different from the share of lending,the ranking of countries is more or less similarfor the lending share and for the share of thesample.

13 In some of the questions, an additional column “NA” (notapplicable) has been added, which respondents can tick if theydo not know whether a specif ic factor contributed to changesin credit standards (questions 2, 9 and 10) or in demand forloans or credit lines (questions 5, 14 and 15), and whetherspecif ic conditions and terms changed (questions 3, 11 and12). There is an important difference between using the “NA”option and the “unchanged” option. The “unchanged” optionimplies that a bank believes that a particular factor, say costsrelated to the bank’s capital position, has contributed tobasically unchanged credit standards. The “NA” option, on theother hand, implies that the bank does not know what impactthis may have had. The purpose of the “NA” factor is thereforeto avoid distorting the distribution of answers.

25ECB




The sample group of banks participating in thesurvey is subject to adjustments either relatedto changes in the credit markets or to thecharacteristics of the individual commercialbanks, e.g. in the case of mergers or takeovers.The sample group is also conditioned by theneed to maintain a certain degree ofrepresentation with regard to bank lendingmarkets and lending categories at the euro arealevel. The Eurosystem monitors developmentsin the national banking sectors and creditmarkets in order to identify changes which maynecessitate an appropriate adjustment of thesample group of banks.

4.3 THE METHOD OF AGGREGATION

As noted above, since the choice of the samplegroups for each country gives rise to somedifferences between the share of a country inthe sample and the share of the country in banklending, country weights are used to aggregatethe national results at the euro area level. Theseweights are the shares of the total lendingaggregates for each country in the total lendingaggregate for the euro area. More specifically,they are derived from the total amountoutstanding of euro area lending to euro arearesidents in each quarter, taking into accountonly the market segments considered in the

survey (i.e. loans to enterprises (non-financialcorporations), consumer credit, lending tohouseholds for house purchase and otherlending to households). This method ensures anappropriate statistical representation, given theheterogeneous characteristics of the nationalbanking systems and national sample groups.Table 3 reports the country weights used tocalculate the euro area results of the banklending survey in the first quarter of 2004.Since the beginning of the survey, the countryweights have varied slightly across countries,reflecting both structural and cyclical changes.

Weighting is not applied at the national level,as the national sample groups are sufficientlyrepresentative, and also because the survey is aqualitative exercise, whereby smaller banks arerepresentative of a much larger group of similarbanks.

The responses by individual banks to thesurvey can in some instances containinformation that the bank in question wouldnot like the public to have access to. Theconfidentiality of individual responses istherefore an important aspect that in turn alsocontributes to the high quality of responses tothe survey. The Eurosystem has therefore putconsiderable emphasis on protecting theconfidentiality of individual responses.

Table 3 Country weights

Source: ECB.

Countryweights at BE DE GR ES FR IE IT LU NL AT PT FI Total

2002 Q4 2.9% 35.9% 1.3% 11.1% 17.4% 1.8% 13.2% 1.0% 8.4% 3.1% 2.6% 1.3% 100.0%

2004 Q2 2.8% 33.1% 1.6% 12.9% 17.2% 2.3% 13.7% 0.9% 8.6% 3.0% 2.6% 1.4% 100.0%

Source: ECB.

Table 2 The distr ibution of banks across countries in the sample group

BE DE GR ES FR IE IT LU NL AT PT FI Total

4 17 3 10 15 5 7 5 6 5 5 4 86


The confidentiality of the results has beenprotected in a manner that allows maximumpossible access to the results of the survey bothwithin and outside the Eurosystem, withouthowever ever compromising theconfidentiality of the individual responses.

4.4 THE COMPILATION GUIDE

The questionnaire is accompanied by a shortcompilation guide (see Annex 2). The purposeof the latter is to help individual respondents byproviding basic information on how to fillout the survey. The compilation guide statesthe purpose of the survey and summarisesits structure. It is deliberately kept short,reflecting the spirit of the survey (i.e. that itis not intended to be a precise statisticalexercise). The compilation guide also statesthe intended respondent. This is a seniorloan officer, e.g. the chairman of the creditcommittee at or just below Board level.

Additionally, the compilation guide specifiesthat the time horizon in the backward-lookingquestions is three months, which is in principlealso the case in the forward-looking questions.However, given the different time horizonsused in the formulation of credit policies andexpectations regarding credit demand, thequestionnaire recognises that the forward-looking horizon is more flexible. It should benoted that the questions are formulated in termsof changes occurring between the end of thelast month of the quarter three months earlierand the end of the quarter that has finished or isabout to finish. Thus, in principle the answersshould reflect the situation at the end of thequarter, rather than the average change over thequarter.

The compilation guide includes a shortexplanation of 17 of the possibly more complexterms used in the questionnaire. However, theguide again emphasises that these definitionspurely aim at providing general guidance.Respondents should be able to answer the

questionnaire without a detailed knowledge ofstatistical terms, and should not have to resortto statistical information sources to reply to thequestions.

4.5 THE PRESENTATION OF THE DATA

The ECB has published the results of the banklending survey ever since the second survey inApril 2003. The publication of results reflectsthe fact that the survey represents a major data-gathering exercise of the Eurosystem whichyields important statistics on credit conditionsin the euro area.

The results of the bank lending survey arepublished in two different formats in thesecond month of every quarter. The MonthlyBulletin contains a summary of the results; themore detailed results are attached to a pressrelease and can be accessed on the ECB’swebsite. Some NCBs also publish nationalresults of the bank lending survey after the ECBhas published the euro area results.

Since the beginning of the survey, the overallresponse rate to the survey has been very high,ranging from 95% to 100%. This ensures a veryacceptable degree of representativeness. Thenumber of banks responding to a specificquestion can however differ depending on thetype of the question, reflecting the fact thatsome banks only operate in certain segments,and therefore only answer questions related tothose segments.

4.6 HOW TO READ THE RESULTS OF THE BANKLENDING SURVEY

The detailed results of the bank lending surveythat are published by the ECB include thepercentage distribution of the weightedanswers to the questionnaire for all the banksresponding, the net percentage and the numberof banks responding. The net percentage forchanges in credit standards is calculated as

27ECB




the difference between the percentage ofrespondents answering that they tightenedconsiderably or somewhat, minus thepercentages responding that they easedconsiderably or somewhat. Thus, in thereported tables, positive figures indicate a nettightening and negative figures a net easing ofcredit standards. Similarly, the net percentagesfor the questions on demand for loans aredefined as the difference between thepercentage of respondents answering that thedemand for loans has increased considerably orsomewhat, minus the percentage respondingthat demand has decreased somewhat orconsiderably. Thus, positive figures indicate anet increase in demand for loans, whilenegative figures reflect a net decrease indemand for loans. As a consequence, there is nodistinction in this calculation as to the degreeof tightening/easing of credit standards or anyincrease/decrease in demand in the replies.


The main objective of the bank lending surveyis to enhance knowledge of the role ofdevelopments in banks’ lending policies in themonetary transmission process. In this respect,the qualitative results obtained from the banklending survey should enable monetary policy-makers to assess credit developments moreaccurately and should represent an additionalsource of information in the monetary policy-making process.

This chapter presents the results of the firsteight rounds of the euro area bank lendingsurvey and compares them with informationcollected from other sources. The final sub-section concludes.

5.1 THE RESULTS OF THE BANK LENDINGSURVEY

This section focuses on the results of the euroarea bank lending survey conducted fromJanuary 2003 until October 2004.14 First of all,a general overview of changes in credit

standards and loan demand for both enterprisesand households is presented. The reasons forthese changes are then described in more detail.The section also provides an overview of thefactors affecting demand.

5.1.1 ENTERPRISES

In the first six survey rounds, banks tightenedtheir credit standards applied to the approval ofloans to enterprises, but with declining netpercentages (see Chart 1). For the last twoavailable quarters, banks reported a slight neteasing of credit standards. Banks were alsoasked about their expectations regarding thedevelopment of credit standards, which havegenerally been in line with the actual outcomes.

Chart 1 shows that the net tightening of creditstandards generally moved opposite to changes

5 TH E F I R S T R E S U LT S O F T H E E URO A R E A B ANKL END I NG S U RV E Y

Chart 1 Credit standards and demand for loans or credit l ines to enterprises

Note: “Realised” values refer to the period in which the survey was conducted. “Expected” values are the net percentagescalculated from the responses given by the banks in the previous survey. For instance, “expected” values for the fourth quarter of2004 were reported by banks in the October 2004 survey.

-40

-20

0

20

40

60

80

-40

-20

0

20

40

60

80

realisedexpected

Q42002

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q42002

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q42003 2004 2003 2004

Changes in credit standards(net percentages of banks tightening standards)

Changes in demand(net percentages of banks reporting an increase in demand)

14 Because the survey questions are phrased in terms of changesover the past three months or expectations of changes over thenext three months, the illustrative analysis will consider atime span that ranges from the fourth quarter of 2002 to, whereappropriate, the fourth quarter of 2004.

29ECB


5 THE FIRSTRESULTS

OF THE EUROAREA BANK

LENDING SURVEY

Chart 2 Factors af fect ing credit standards for loans to enterprises

(net percentage of banks contributing to tightening)

Source: ECB.Note: The net percentage is def ined as the difference between the sum of “contributed considerably to tightening” and “contributedsomewhat to tightening” and the sum of “contributed somewhat to easing” and “contributed considerably to easing”.

-30

10

30

50

70

90

-30

-10 -10

10

30

50

70

90

Pressure from competition

Cost of funds and balance sheet constraints

Costs relatedto bank’scapital

position

Bank’s ability toaccess market

financing

Bank’s liquidityposition

Expectationsof generaleconomicactivity

Industry orfirm-specific

outlook

Risk oncollateraldemanded

Perception of risk

Competition fromother banks

Competition fromnon-banks

Competition from marketfinancing

2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003

2002 20042003 2002 20042003 2002 20042003-30

-10

10

30

50

70

90

-30

-10

10

30

50

70

90

in the net percentages of banks reporting anincrease in the demand for loans to enterprises.The results of the first four surveys indicatethat the net tightening of credit standards wasaccompanied by a reported net decrease indemand for loans (i.e. a negative value for the

net percentage of banks reporting an increase inloan demand). Furthermore, the declining netpercentages of banks tightening creditstandards were reflected in the declining netpercentages of banks reporting a decrease inloan demand for the first four surveys, and a


Chart 3 Condit ions and terms for approving loans or credit l ines to enterprises

(net percentage of banks reporting tightening standards)

Note: The net percentage is def ined as the difference between the sum of “tightened considerably” and “tightened somewhat” andthe sum of “eased somewhat” and “eased considerably”.

-20

0

20

40

60

80

100

-20

0

20

40

60

80

100

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Margins on average loans

Price

Margins on riskier loans

reported net increase in loan demand in thefourth quarter of 2003. Since the first quarter of2004, this relationship has become somewhatmore ambivalent: although banks reported acontinuous decrease in the net tightening ofcredit standards, culminating in a net easing in

the last two quarters of 2004, loan demand (ona net basis) did not structurally improve,according to the banks (see the right panel ofChart 1).

Other conditions and terms

2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003

MaturitySize of loan orcredit line

Collateralrequirements

Loanscovenants

Non-interestrate charges

-20

0

20

40

60

80

100

-20

0

20

40

60

80

100

31ECB


5 THE FIRSTRESULTS


LENDING SURVEY

Chart 4 Factors af fect ing demand for loans and credit l ines to enterprises

(net percentages of banks reporting an increase in demand)

Note: The net percentage is def ined as the difference between the sum of “contributed considerably to higher demand” and“contributed somewhat to higher demand” and the sum of “contributed somewhat to lower demand” and “contributed considerablyto lower demand”.

Loans from otherbanks

Issuance of debtsecurities

Financing needs

2002 20042003 2002 20042003 2002 20042003 2002 20042003

2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003

-60

-40

-20

0

20

40

60

-60

-40

-20

0

20

40

60M&A and corporate

restructuring

Issuance ofequity

Debt restructuringFixed investment Inventories and workingcapital

Use of alternative finance

Internalfinancing

Loans from non-banks

-60

-40

-20

0

20

40

60

-60

-40

-20

0

20

40

60

Considering the reasons behind the tighteningof credit standards, Chart 2 shows the factorsaffecting credit standards for approving loansor credit lines to enterprises.

Since the beginning of the survey, the threemost important factors reported by banks whenexplaining credit standard developments werethose related to the perception of risk:“expectations regarding general economic


activity”, “industry or firm-specific outlook”and “risk on collateral demanded”. The last twofactors in particular explain the developmentsin the net tightening and net easing of creditstandards reported in the eight rounds of thesurvey. For the last two available quarters, thefactor that contributed the most to the neteasing of credit standards was pressure fromcompetition, particularly “competition fromother banks” and “competition from marketfinancing”.

In terms of how credit standards weretightened, this mostly took place through thesetting of price conditions (i.e. particularly inthe form of “margins on riskier loans”).Basically all conditions and terms werecontinuously tightened to a lesser extent oreven eased during the eight rounds of thesurvey (see Chart 3).

Turning to loan demand from enterprises, alarge number of factors contributed to theoverall development of the results of the firsteight rounds of the survey (see Chart 4). For

example, “debt restructuring” has so far alwaysbeen reported as the factor contributing themost frequently to the overall net increase inloan demand. Others, such as “fixedinvestment”, were also always seen ascontributing to lower demand, although in thesecond quarter of 2004 a substantialimprovement was reported. Financing needsrelated to “M&A and corporate restructuring”showed a positive trend, resulting in theJanuary 2004 survey citing for the first timethis factor as having contributed to an increasein loan demand. Since then, financing needsrelated to M&A activity have always positivelysupported loan demand. Conversely, in themost recent available quarters the increase inthe use of internal financing as a source offunds for enterprises has considerablycontributed to lower demand for bank loans.

5.1.2 LOANS FOR HOUSE PURCHASE

After having declined for the first three surveyrounds, the net percentage of banks reporting anet tightening of credit standards applied to the

Chart 5 Credit standards and demand for loans or credit l ines to households for housepurchase


realisedexpected

Q42002

Q42002

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q420032003 2004 2004

-40

-30

-20

-10

0

10

20

30

40

-40

-30

-20

-10

0

10

20

30

40Changes in credit standards

(net percentages of banks tightening standards)Changes in demand


33ECB


5 THE FIRSTRESULTS


LENDING SURVEY

approval of loans for house purchase slightlyincreased in both the October 2003 and theJanuary 2004 surveys, before stabilising in theApril 2004 survey. In the last two quarters of2004, this trend reversed and, for the first timesince the survey began, banks reported aneasing of credit standards (see Chart 5). Thischange in the application of credit standardswas not reflected on the demand side, wherebanks have always reported, although atdifferent levels, an increase in the demand forloans to households. Unfortunately, surveyrespondents have not been very successful inpredicting developments: in five survey roundsout of eight, banks actually expected a decreasein the demand for loans for house purchase byhouseholds.

Turning to the factors related to the nettightening of credit standards, a very similarpicture to what has been observed for loans andcredit lines to enterprises emerges. Tighteningwas mostly explained by the perception of risk(both related to “general economic activity”,

Chart 6 Factors af fect ing credit standards appl ied to the approval of loans to households forhouse purchase(net percentage of banks contributing to tightening)

Note: The net percentage is def ined as the difference between the sum of “contributed considerably to tightening” and “contributedsomewhat to tightening” and the sum of “contributed somewhat to easing” and “contributed considerably to easing”.

and specifically to “housing marketprospects”), whereas pressure from othercompetitors generally contributed to easingcredit standards (see Chart 6). For example, inthe October 2004 survey round, the reported neteasing of credit standards reflects opposingeffects, with higher risk perceptions related tothe housing market on the one hand, set againstclear improvements with respect to pressurefrom competition on the other.

With regard to conditions and terms forapproving loans to households for housepurchase, changes in the lending policies ofbanks mainly occurred through priceconditions (see Chart 7).

2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003


Cost of fundsand balance

sheet constraints

Competitionfrom other

banks

Expectations ofgeneral economic

activity

Competitionfrom non-banks

Housing marketprospects

Perception of risk

-40

-20

0

20

40

60

-40

-20

0

20

40

60


Regarding the factors reported in explanationfor the development of loan demand for housepurchase, “housing market prospects” hasalways contributed, although at differentlevels, to an increase in demand. In contrast,

“consumer confidence” and “non-housingrelated consumption expenditure” contributedin all survey rounds to a net decrease in demand(see Chart 8).

Chart 7 Condit ions and terms for approving loans to households for house purchase

(net percentages of banks reporting tightening standards)


-40

-20

0

20

40

60

-40

-20

0

20

40

60

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3

2002 20042003 2002 20042003 2002 20042003 2002 20042003

Q4 Q1 Q22004

Q32003

Margins on average loans


Price

Margins on riskier loans


Loan-to-value ratio Maturity Non-interestrate charges

-40

-20

0

20

40

60

-40

-20

0

20

40

60

35ECB


5 THE FIRSTRESULTS


LENDING SURVEY

5.1.3 LOANS FOR CONSUMER CREDIT

The results for loans for consumer credit,which constitute the other main part of themarket for loans to households, do not show aspecific pattern over time. For example, the lastquarter of 2003 recorded a significant drop inthe net percentage of banks tightening theircredit standards (see Chart 9). This had alreadybeen anticipated in October 2003, when bankseven expected an overall net easing of creditstandards for the last quarter of 2003.

Demand for consumer credit has alsofluctuated, but in a manner sometimes notdirectly related to credit standards: a lesserdegree of net tightening did not alwaystranslate into a net increase in demand. At thesame time, especially in 2004, banks have beenmore optimistic in their expectations aboutfuture developments in loan demand.

Unfortunately, the reality has not always livedup to expectations, although for the most recentavailable quarters, banks seem to have moresuccessfully predicted at least the trend of loandemand, if not the correct levels.

Chart 8 Factors af fect ing demand for loans to households for house purchase



-40

-20

0

20

40

-40

-20

0

20

40

2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003

Financing needs

Housing marketprospects

Consumerconfidence

Non-housingrelated

consumptionexpenditure

Householdsavings

Other sources offinance

Loans from otherbanks



In all survey rounds, factors related to riskperceptions mostly explained the trendreported of a net tightening of credit standardsfor consumer credit (see Chart 10). On the otherhand, competition from other banks and otherfinancial institutions always played a minorcounterbalancing effect in easing (on a netbasis) credit standards.

Chart 9 Credit standards and demand for loans or credit l ines to households for consumercredit


Q42002

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q42003

-30

-20

-10

0

10

20

30

40

-30

-20

-10

0

10

20

30

40

realisedexpected

Changes in credit standards(net percentages of banks tightening standards)

Changes in demand(net percentages of banks reporting an increase in demand)

2004Q4

2002Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

2003 2004

37ECB


5 THE FIRSTRESULTS


LENDING SURVEY

Chart 10 Factors af fect ing credit standards appl ied to the approval of loans to households forconsumer credit(net percentages of banks reporting tightening standards)

Note: The net percentage is def ined as the difference between the sum of “contributed considerably to tightening” and “contributedsomewhat to tightening” and the sum of “contributed somewhat to easing” and “contributed considerably to easing”.

As was the case with loans to households forhouse purchase, the tightening of conditionsand terms for approving loans to households forconsumer credit mostly took place through awidening of margins, particularly the margins

on riskier loans. However, in the last fivesurvey rounds, margins on average loans werereported as contributing to easing creditstandards (see Chart 11).

Cost of funds and balance sheet constraints


Competition from other banks Competition from non-banks

Risk on collateral demandedCreditworthiness of consumersExpectations ofgeneral economic activity

Perception of risk

Q42002

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

-20

-10

0

10

20

30

40

50

-20

-10

0

10

20

30

40

50

-20

-10

0

10

20

30

40

50

-20

-10

0

10

20

30

40

50


Turning to the demand for consumer credit, andin particular looking at the forces driving itsmovement, it appears that over the observedperiod, the use of alternative finance did nothave a significant impact on the overalldevelopment of this credit component (seeChart 12). On the contrary, factors related tofinancing needs generally contributed to a netdecrease in demand. The only exception is“spending on durable consumer goods”, whichcontributed to lower demand for consumercredit during the first six survey rounds, butthen became a factor contributing to anincrease in demand in the last two. Thepercentages of factors contributing todevelopments in demand for consumer creditvary considerably: for example, althoughconsumer confidence contributed to a decreasein demand for all available survey cycles, thesize of this contribution has fallen considerablyover time.

Chart 11 Condit ions and terms for approving loans to households for consumer credit

(net percentages of banks reporting tightening standards)

-20

-10

0

10

20

30

40

50

-20

-10

0

10

20

30

40

50Price

Margins onaverage loans

Margins onriskier loans


Maturity Non-interestrate charges


2002 20042003 2002 20042003 2002 20042003 2002 20042003 2002 20042003


39ECB


5 THE FIRSTRESULTS


LENDING SURVEY

5.2 RELATIONSHIP WITH OTHER INDICATORS

The value added of the bank lending survey isthat it provides “qualitative” information thatis directly derived from the lenders themselves.Moreover, these data allow us to consider

supply and demand-side factors of loandevelopments separately. This is particularlyuseful, as other sources of information tend tobe of a quantitative nature and often representthe interaction of both supply and demandfactors.

Chart 12 Factors af fect ing demand for loans to households for consumer credit



Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

-40

-30

-20

-10

0

10

20

30

-40

-30

-20

-10

0

10

20

30Financing needs

Household savings

Spending on durableconsumer goods

Consumer confidence

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q42002

Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Securities purchases


Other sources of financeLoans from other banks

-40

-30

-20

-10

0

10

20

30

-40

-30

-20

-10

0

10

20

30


Following the structure of the bank lendingsurvey, this section compares some of thereported variables in the survey (creditstandards, factors and conditions affectingstandards and factors affecting demand) withinformation from other sources (e.g. loangrowth, corporate spreads, industrial andconsumer confidence, etc.).

The purpose of this analysis is twofold. First, itseeks to assess the quality of the bank lendingsurvey data by comparing them with otheravailable statistics. Second, it attempts toinvestigate the information content of creditstandards in relation to other economic andfinancial variables. However, because of thelow number of observations available, it is tooearly to draw any firm conclusions about theleading or lagging properties between thevariables.

5.2.1 CREDIT STANDARDS AND REAL GDPGROWTH

Credit standards, among other factors, may belinked to economic activity, given that to theextent that credit availability depends onlenders’ standards, a tightening of standardsshould cause a decline in spending by firms thatdepend on banks for credit. To support thisview, researchers at the US Federal Reservehave shown that there is a strong correlationbetween loan officers’ reports of tightercredit standards and slowdowns in output.15

Moreover, using the VAR methodology, theyshow that changes in credit standards lead tochanges in output and bank loans.

At the euro area level, it is not yet possible tomake similar analyses because the number ofobservations currently available is still low.Nevertheless, Chart 13 presents developmentsin real activity alongside those in creditstandards.

As already indicated, given the small numberof observations and, in particular, the absenceof turning points in both series, it is not yetpossible to draw any firm conclusions. 15 See Chapter 2, in particular Lown et al. (2000).

Nonetheless, it can be argued that theprogressive easing of credit standards observedbetween the fourth quarter of 2002 and the thirdquarter of 2004 appears to lead the upswing inactivity observed between the second quarterof 2003 and the second quarter of 2004.However, this analysis remains tentative,particularly as one must bear in mind the factthat many factors may impact on economicactivity, including interest rates, exchangerates and consumer and business confidence.Furthermore, some of these factors may impacton both activity and credit standards as appliedby loan officers.

However, even if this is the case, informationon credit standards may still provide someuseful insights as such information is availableon a timelier basis than quantitativeinformation on economic activity.

Chart 13 Comparison of bank lending survey(BLS) data on credit standards and real GDPgrowth

Source: ECB.Note: For credit standards, the net percentage is def ined asthe difference between the sum of “contributed considerablyto tightening” and “contributed somewhat to tightening” andthe sum of “contributed somewhat to easing” and“contributed considerably to easing”.

-10

0

10

20

30

40

50

60

70 0.0

0.5

1.0

1.5

2.0

2.5Q2 Q3

2002Q4 Q1 Q2 Q3 Q4 Q1 Q2

2004Q3

2003

credit standards to enterprises (left-hand scale)credit standards to households for house purchase (left-hand scale)credit standards to households for consumer credit (left-hand scale)real GDP growth (right-hand scale, inverted)

net percentage tightening year-on-year growth rate

41ECB


5 THE FIRSTRESULTS


LENDING SURVEY

5.2.2 CREDIT STANDARDS AND MFI LOANGROWTH

One aim of the survey is to complementexisting sources of information on creditdevelopments. In the United States, Lown andMorgan (2002 and 2004) show that a high net

percentage of tightening is typically associatedwith low (if not negative) credit growth. Theresults of the bank lending survey suggest thatfor the euro area this relationship is not alwaysapparent. For instance, the declining nettightening of credit standards applied to loansto enterprises over the first six survey rounds

Chart 14 Comparison of BLS data on credit standards and MFI loan growth

b) households

Source: ECB.Notes: For credit standards, the net percentage is def ined as the difference between the sum of “contributed considerably totightening” and “contributed somewhat to tightening” and the sum of “contributed somewhat to easing” and “contributedconsiderably to easing”. For demand, the net percentage is def ined as the difference between the sum of “contributed considerablyto higher demand” and “contributed somewhat to higher demand” and the sum of “contributed somewhat to lower demand” and“contributed considerably to lower demand”.

a) enterprises

-15

-10

-5

0

5

10

15

20

25

30

35

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

11.0

credit standards applied to the approval of loans to households (left-hand scale)demand for loans to households (left-hand scale)year-on-year rate of growth in loans to households (right-hand scale)

House purchase Consumer credit

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

net percentage tightening percentage changes

-40

-20

0

20

40

60

80

2.0

2.5

3.0

3.5

4.0

4.5

5.0

credit standards applied to the approval of loans to enterprises (left-hand scale)demand for loans to enterprises (left-hand scale)year-on-year rate of growth in loans to enterprises (right-hand scale)

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

net percentage tightening percentage changes


has not been associated with an increase inloan growth (see Chart 14a). Only in the lasttwo available quarters, when banks easedstandards, was a slight pick-up in loan growthrecorded. A very similar picture emergeswhen looking at the household sector (seeChart 14b). However, this may be related tothe aforementioned difficulty of studyingthe lagged impact of changes in creditstandards on loan developments. The demandfor loans as reported by survey respondents isalso presented in Chart 14, as this representsanother factor that helps to determine the actualdevelopment of loan growth. The chart showsthat the reported net percentages of demand forloans are difficult to reconcile consistentlywith actual loan growth, especially in the caseof enterprises.

5.2.3 FACTORS AFFECTING CREDIT STANDARDSAND FINANCIAL SPREADS

Turning to the factors affecting creditstandards, Chart 15 shows that the improvedability of banks to access market financingcontributed to a general decline in the nettightening of credit standards and evencontributed to an easing of credit standards. Asthe same chart reveals, this trend is confirmedby the initial increase and subsequent declinein the spread of BBB-rated bonds issued bybanks in the capital markets, which may indeedbe interpreted as an improvement in banks’access to market financing.16

5.2.4 FACTORS AFFECTING CREDIT STANDARDSAND INDUSTRIAL CONFIDENCE

Another reason reported for the tightening ofcredit standards is the risk perception related tothe industry or firm-specific outlook. Chart 16compares the significant drop in the netpercentage tightening of credit standardsapplied to loans or credit lines to enterpriseswith the industrial confidence indicator asreported by the European Commission’sBusiness and Consumer Surveys.17 Since thesecond quarter of 2003, the industrialconfidence indicator has become less negative

and, at the same time, fewer banks reported thisfactor as contributing to a tightening of creditstandards.

5.2.5 CONDITIONS AND TERMS FOR APPROVINGLOANS AND CORPORATE BOND SPREADS

One of the conditions and terms that is affectedby the reported tightening of credit standardsfor enterprises during the first six surveyrounds is banks’ margins on riskier loans. Thedeclining net percentages reported for thisfactor suggest that banks perceived lowercredit risk in loan markets. This is in linewith the information provided by corporatebonds and credit default swap spreads (see

16 According to Standard and Poor’s, an obligation rated BBBexhibits adequate protection parameters. However, adverseeconomic conditions or changing circumstances are morelikely to lead to weakened capacity on the part of the obligor tomeet its f inancial commitment on the obligation.

17 The industrial conf idence indicator is the average of balancesof the answers to the questions on production expectations,order books and stocks of f inished products. Balances areseasonally adjusted.

Chart 15 Comparison of BLS data and BBBfinancial spreads

Sources: ECB and Datastream.Note: The net percentage reported for banks’ ability to accessmarket f inancing is def ined as the difference between thesum of “contributed considerably to tightening” and“contributed somewhat to tightening” and the sum of“contributed somewhat to easing” and “contributedconsiderably to easing”.

-10

-5

0

5

10

15

20

0

50

100

150

200

250

300

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

bank’s ability to access market financing (left-hand scale)BBB financial bond spreads (right-hand scale)

net percentage tightening basis points

43ECB


5 THE FIRSTRESULTS


LENDING SURVEY

Chart 17).18 Both indicators tend to berelatively sensitive to changes in theperception of credit risk, and their low level inthe course of 2004 suggests a fairly positivecredit risk outlook.

5.2.6 FACTORS AFFECTING LOAN DEMAND ANDGROSS FIXED CAPITAL FORMATION

Turning to the demand side, the bank lendingsurvey provides information on the reasonsdriving the demand for loans from bothenterprises and households. Regarding demandfor loans from enterprises, the survey resultsindicate that financing needs related to fixedinvestment have improved substantially sincethe first survey round, although they are stillreported with negative percentages, indicatingthat fixed investment still contributes to a netdecrease in demand for loans by enterprises.Chart 18 compares this information fromthe bank lending survey with the growth rateof the GDP component that is mostly relatedto investment, namely gross fixed capitalformation.

For most reported survey rounds, a decline inthe contribution of fixed investment to lowerloan demand from enterprises is related to someextent to a positive increase in the quarter-on-quarter rate of growth in gross fixed capitalformation.

18 A company’s credit default swap spread is the cost per annumfor protection against a default by the company. Credit defaultspread indices have been found to be good indicators of creditrisk. See for instance Hull et al. (2000).

Chart 17 Comparison of BLS data and BBBcorporate bond spreads

Sources: ECB, Datastream, Merrill Lynch and Credit Trade.Note: The net percentage reported for margins on riskierloans refers to the difference between the sum of thepercentages for “tightened considerably” and “tightenedsomewhat” and the sum of the percentages for “easedsomewhat” and “eased considerably”.

0

10

20

30

40

50

60

70

80

90

100

0

50

100

150

200

250

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

margins on riskier loans to enterprises (left-hand scale)BBB credit default swap spreads (right-hand scale)BBB corporate bond spreads (right-hand scale)

net percentage tightening basis points

Chart 16 Comparison of BLS data andindustr ia l conf idence

Sources: ECB and European Commission Business andConsumer Surveys.Note: The net percentage reported for the industry or f irm-specif ic outlook is def ined as the difference between the sumof “contributed considerably to tightening” and “contributedsomewhat to tightening” and the sum of “contributedsomewhat to easing” and “contributed considerably toeasing”.

0

10

20

30

40

50

60

70

80

90 -20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

industry or firm-specific outlook (left-hand-scale)industrial confidence (right-hand scale, inverted)

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

net percentage tightening percentage balances


5.2.7 FACTORS AFFECTING LOAN DEMAND ANDCONSUMER CONFIDENCE

Looking at loan demand from households, onefactor that has been mentioned as contributingto a slight net decrease in demand is consumerconfidence. Chart 19 compares this factor withthe consumer confidence indicator constructedand published by the European Commission. 19

This chart shows that the drop in the consumerconfidence indicator in 2002 was followed by asteady improvement in 2003. Although the netpercentages are still negative, indicating thatconsumer confidence contributed to a netdecrease in loan demand from households, thedata obtained from the bank lending survey arein line with this upward trend.

19 The consumer conf idence indicator is the average of balancesof the answers to the questions on the f inancial situation ofhouseholds, the general economic situation, unemploymentexpectations (with inverted sign) and savings, all over the next12 months. Balances are seasonally adjusted.

Chart 18 Comparison of BLS data and grossf ixed capital formation

Source: ECB.Note: The net percentage reported for f ixed investment isdef ined as the difference between the sum of “contributedconsiderably to higher demand” and “contributed somewhatto higher demand” and the sum of “contributed somewhat tolower demand” and “contributed considerably to lowerdemand”.

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

-60

-40

-20

0

20

40

60

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

fixed investment affecting demand for loans to enterprises (left-hand scale)quarter-on-quarter percentage change in gross fixed capital formation – seasonally adjusted(right-hand scale)

net percentage percentage per annum

Chart 19 Comparison of BLS data and consumer conf idence

Sources: ECB and European Commission Business and Consumer Surveys.Note: The net percentage reported for consumer conf idence is def ined as the difference between the sum of “contributedconsiderably to higher demand” and “contributed somewhat to higher demand” and the sum of “contributed somewhat to lowerdemand” and “contributed considerably to lower demand”.

-40

-35

-30

-25

-20

-15

-10

-5

0

-23

-21

-19

-17

-15

-13

-11

-9

-7

House purchase

consumer confidence affecting loans to households (left-hand scale)consumer confidence (right-hand scale)

Consumer credit

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

Q2 Q32002

Q4 Q1 Q2 Q3 Q4 Q1 Q22004

Q32003

net percentage percentage balances

45ECB


5 THE FIRSTRESULTS


LENDING SURVEY

5.3 CONCLUSIONS

This chapter presented the first results of thebank lending survey for the euro area, whichhas been conducted since January 2003, andcompared them with information derived fromother sources. The graphical analysis carriedout shows that overall, even at this early stageof actually conducting the survey, it is possibleto identify some systematic patterns in theresults from the bank lending survey that proveto be in line with indicators retrieved fromother data sources. This could indicate that thesample is indeed very representative of the euroarea. However, with only eight observationsavailable, these results should be interpretedwith a high degree of caution. More time andexperience will be needed to assess howexactly to interpret these results and how tojudge the relationship between them and actualeconomic and financial developments.


When more than three years ago theEurosystem decided to assess whether or not itshould implement a bank lending survey for theeuro area, one of the main purposes of such asurvey was to provide the ECB GoverningCouncil with new input into the monetarypolicy decision-making process. Almost twoyears after the first survey, the assessment ofthe whole exercise is positive. The results ofthe survey are used in a regular way tocomplement information derived from existingquantitative statistics on bank retail interestrates and credit. In this respect, the surveyprovides useful information on developmentsin supply and demand conditions in the euroarea credit markets and on the bank lendingpolicies of euro area banks.

The analysis presented in this occasional papershows that overall, even at this early stage, theresults of the survey are not only consistentwith other statistics, but could also provideadded value by improving our understandingof actual economic and financial developmentsin the euro area. The graphical analysisundertaken in Chapter 5 should, of course, bereinforced with more sophisticated tools,which will be possible when longer times serieshave become available.

High-quality representative results from thebank lending survey should serve the interestsof both policy-makers and market participants.By taking part in the bank lending survey,banks contribute to generating information thatis particularly useful for the euro area bankingsector. At the moment, qualitative informationsources on euro area-wide developments inbank lending behaviour, which would allowindividual banks to assess their own behaviourvis-à-vis their competitors, are still scarce.There are also a number of additional benefitsrelated to obtaining a better insight intopossible reasons for structural changes in theeuro area banking sector, developing a moresuitable framework for analysing the economicsituation in the euro area, and interpretingcertain irregular events in the euro areafinancial markets.

6 CONC LUD I NG R EMARK S

47ECB


ANNEX 1

A NN EX I

T H E QU E S T I ONNA I R E O F T H E B ANK L END I NGSURV E Y F OR TH E E URO A R E A

1. Over the past three months, how have your bank’s credit standards as applied to the approval of loans or credit lines to enterprises changed?

Overall Loans to small and medium-sized enterprises

Loans to large enterprises

Short-term loans

Long-term loans

Tightened considerably Tightened somewhat Remained basically unchanged

Eased somewhat Eased considerably 2. Over the past three months, how have the following factors affected your bank’s credit standards as applied to the approval of loans or credit lines to enterprises (as described in question 1 in the column headed “Overall”)? Please rate the contribution of the following factors to the tightening or easing of credit standards using the following scale: – – = contributed considerably to tightening of credit standards – = contributed somewhat to tightening of credit standards � = contributed to basically unchanged credit standards + = contributed somewhat to easing of credit standards + + = contributed considerably to easing of credit standards NA = not applicable A) Cost of funds and balance sheet constraints � Costs related to your bank’s capital position � Your bank’s ability to access market financing (e.g. money or bond market financing) � Your bank’s liquidity position B) Pressure from competition � Competition from other banks � Competition from non-banks � Competition from market financing C) Perception of risk � Expectations regarding general economic activity � Industry or firm-specific outlook � Risk on the collateral demanded

– – – � + ++ NA

D) Other factors, please specify


3. Over the past three months, how have your bank’s conditions and terms for approving loans or credit lines to enterprises changed? Please rate each factor using the following scale: – – = tightened considerably – = tightened somewhat � = remained basically unchanged + = eased somewhat + + = eased considerably NA = not applicable A) Price � Your bank’s margin on average loans (wider margin = tightened, narrower margin = eased) � Your bank’s margin on riskier loans B) Other conditions and terms � Non-interest rate charges � Size of the loan or credit line � Collateral requirements � Loan covenants � Maturity

– – – � + + + NA

C) Other factors, please specify 4. Over the past three months, how has the demand for loans or credit lines to

enterprises changed at your bank, apart from normal seasonal fluctuations? Overall Loans to small and

medium-sized enterprises


Short-term loans

Long-term loans

Decreased considerably Decreased somewhat Remained basically unchanged

Increased somewhat Increased considerably

49ECB


ANNEX 1

5. Over the past three months, how have the following factors affected the demand for loans or credit lines to enterprises (as described in question 4 in the column headed “Overall”)? Please rate each possible factor using the following scale: – – = contributed considerably to lower demand – =contributed somewhat to lower demand � =contributed to basically unchanged demand + =contributed somewhat to higher demand + + =contributed considerably to higher demand NA = not applicable A) Financing needs � Fixed investment � Inventories and working capital � Mergers/acquisitions and corporate restructuring � Debt restructuring B) Use of alternative finance � Internal financing � Loans from other banks � Loans from non-banks � Issuance of debt securities � Issuance of equity

– – – � + + + NA

C) Other factors, please specify 6. Please indicate how you expect your bank’s credit standards as applied to the

approval of loans or credit lines to enterprises to change over the next three months.



Short-term loans

Long-term loans

Tighten considerably Tighten somewhat Remain basically unchanged Ease somewhat Ease considerably 7. Please indicate how you expect demand for loans or credit lines to enterprises

to change at your bank over the next three months (apart from normal seasonal fluctuations)



Short-term loans

Long-term loans

Decrease considerably Decrease somewhat Remain basically unchanged Increase somewhat Increase considerably


8. Over the past three months, how have your bank’s credit standards as applied to the approval of loans to households changed?

Loans for house purchase Consumer credit and other lending

Tightened considerably Tightened somewhat Remained basically unchanged Eased somewhat Eased considerably 9. Over the past three months, how have the following factors affected your bank’s credit standards as applied to the approval of loans to households for house purchase (as described in question 8)? Please rate the contribution of the following factors to the tightening or easing of credit standards using the following scale: – – = contributed considerably to tightening of credit standards – = contributed somewhat to tightening of credit standards � = contributed to basically unchanged credit standards + = contributed somewhat to easing of credit standards + + = contributed considerably to easing of credit standards NA = not applicable A) Cost of funds and balance sheet constraints B) Pressure from competition � Competition from other banks � Competition from non-banks C) Perception of risk � Expectations regarding general economic activity � Housing market prospects

– – – � + + + NA


51ECB


ANNEX 1

10. Over the past three months, how have your bank’s conditions and terms for approving loans to households for house purchase changed? Please rate each factor using the following scale: – – = tightened considerably – = tightened somewhat � = remained basically unchanged + = eased somewhat + + = eased considerably NA = not applicable A) Price � Your bank’s margin on average loans (wider margin = tightened, narrower margin = eased) � Your bank’s margin on riskier loans B) Other conditions and terms � Collateral requirements � “Loan-to-value” ratio � Maturity � Non-interest rate charges

– – – � + + + NA

C) Other factors, please specify 11. Over the past three months, how have the following factors affected your bank’s credit standards as applied to the approval of consumer credit and other lending to households (as described in question 8)? Please rate the contribution of the following factors to the tightening or easing of credit standards using the following scale: – – = contributed considerably to tightening of credit standards – = contributed somewhat to tightening of credit standards � = contributed to basically unchanged credit standards + = contributed somewhat to easing of credit standards + + = contributed considerably to easing of credit standards NA = not applicable A) Cost of funds and balance sheet constraints B) Pressure from competition � Competition from other banks � Competition from non-banks C) Perception of risk � Expectations regarding general economic activity � Creditworthiness of consumers � Risk on the collateral demanded

– – – � + + + NA



12. Over the past three months, how have your bank’s conditions and terms for approving consumer credit and other lending to households changed? Please rate each factor using the following scale: – – = tightened considerably – = tightened somewhat � = remained basically unchanged + = eased somewhat + + = eased considerably NA = not applicable A) Price � Your bank’s margin on average loans (wider margin = tightened, narrower margin = eased) � Your bank’s margin on riskier loans B) Other conditions and terms � Collateral requirements � Maturity � Non-interest rate charges

– – – � + + + NA

C) Other factors, please specify 13. Over the past three months, how has the demand for loans to households

changed at your bank, apart from normal seasonal fluctuations? Loans for house purchase Consumer credit and other

lending Decreased considerably Decreased somewhat Remained basically unchanged Increased somewhat Increased considerably

53ECB


ANNEX 1

14. Over the past three months, how have the following factors affected the demand for loans to households for house purchase (as described in question 13)? Please rate each factor using the following scale: – – = contributed considerably to lower demand – = contributed somewhat to lower demand � =contributed to basically unchanged demand + = contributed somewhat to higher demand + + = contributed considerably to higher demand NA = not applicable A) Financing needs � Housing market prospects � Consumer confidence � Non-housing-related consumption expenditure B) Use of alternative finance � Household savings � Loans from other banks � Other sources of finance

– – – � + + + NA

C) Other factors, please specify 15. Over the past three months, how have the following factors affected the demand for consumer credit and other lending to households (as described in question 13)? Please rate each factor using the following scale: – – = responsible for considerable decrease – = responsible for decrease � = responsible for neither decrease nor increase + = responsible for increase + + = responsible for considerable increase NA = not applicable A) Financing needs � Spending on durable consumer goods, such as cars, furniture, etc. � Consumer confidence � Securities purchases B) Use of alternative finance � Household savings � Loans from other banks � Other sources of finance

– – – � + + + NA

C) Other factors, please specify


16. Please indicate how you expect your bank’s credit standards as applied to the approval of loans to households to change over the next three months. Loans for house purchase Consumer credit and other

lending Tighten considerably Tighten somewhat Remain basically unchanged Ease somewhat Ease considerably 17. Please indicate how you expect demand for loans to households to change over the next three months at your bank (apart from normal seasonal fluctuations). Loans for house purchase Consumer credit and other

lending Decrease considerably Decrease somewhat Remain basically unchanged Increase somewhat Increase considerably 18. Over the past three months, have there been any other issues of importance for bank lending behaviour in the euro area or in your country which are not covered by this survey?

55ECB


ANNEX 2

The questionnaire is accompanied by a shortcompilation guide. The purpose of thecompilation guide is to help the respondentwith basic information on how to fill out thesurvey. The compilation guide states thepurpose of survey and summarises the structureof the survey. It is deliberately kept short,reflecting the spirit of the survey as not being aprecise statistical exercise. The compilationguide also states the intended respondent. Thisis a senior loan officer, e.g. the chairman of thecredit committee at or just below Board level.

Additionally, the compilation guide spells outthat the time horizon in the backward lookingquestions is three months, and in the forwardlooking questions in principle also threemonths. However, given the different timehorizons used in the formulation of creditpolicies and expectations regarding creditdemand, the questionnaire recognises that theforward-looking horizon is more flexible. Itshould be noted that the questions areformulated in terms of changes occurringbetween the end of the last month of the quarterthree months earlier and the end of the quarterthat is finished or about to finish. Thus, inprinciple the answers should reflect thesituation at the end of the quarter, rather thanthe average change over the quarter.

The compilation guide includes an overview ofthe possible more complex terms used in thequestionnaire. A total of 17 terms areexplained. However, it is also here emphasisedthat this is for guidance and that therespondents should be able to answer thequestionnaire without a detailed knowledge ofstatistical terms. Thus respondents are notexpected to have to resort to statisticalinformation systems to answer the questions.

The most important term explained is probably“credit standards”. Credit standards includenot only the written standards a bank has, butalso the unwritten practice and theirapplication. The two latter components make itmore likely that changes occur over a quarter.

ANNEX 2

TH E COMP I L AT I ON GU I D E O F T H E B ANKL END I NG S U RV E Y F OR TH E E URO A R E A

GUIDELINES FOR THE COMPLETION OF THEBANK LENDING SURVEY QUESTIONNAIRE

In the backward looking questions (allquestions except 6, 7, 16 and 17), the timehorizon is three months. For instance, inJanuary the survey relates to changes betweenthe end of September and the end of December.

In the forward-looking questions (6, 7, 16 and17), the time horizon is in principle also threemonths (including the survey month), but someflexibility is given in view of the different timehorizons used in the formulation of creditpolicies and expectations regarding creditdemand.

In questions 2, 3, 5, 9, 10, 11, 12, 14 and 15,an answer should be given for all factors. Ifyou do not have information about a specificfactor, please use the option “not applicable”(column NA in the questionnaire). Should youjudge that other factors or a specific marketsegment had a significant impact on overalldevelopments, please specify under the option“Other factors”.

TERMS USED IN THE BANK LENDING SURVEYQUESTIONNAIRE

Capital (question 2)Defined in accordance with the Basel capitaladequacy requirements; includes both tier 1capital (core capital) and tier 2 capital(supplementary capital).

Collateral (questions 2, 3, 10, 11 and 12)The security given by a borrower to a lenderas a pledge for the repayment of a loan. Thiscould include certain financial securities,such as equity or debt securities, real estateor compensating balances. A compensatingbalance is the minimum amount of a loan thatthe borrower is required to keep in an account atthe bank.


Consumer confidence (questions 14 and 15)Consumers’ assessments of economic andfinancial trends, in a particular country and/orin the euro area. They include assessments ofthe past and current financial situation ofhouseholds and resulting prospects for thefuture, the past and current general economicsituation and resulting prospects for the futureand the advisability of making major purchasesof durable consumer goods.

Covenant (question 3)A covenant is an agreement or stipulationexpressed in loan contracts, particularlycontracts with enterprises, by which theborrower pledges to take certain action (anaffirmative covenant) or refrain from takingcertain action (a negative covenant), and isconsequently part of the terms and conditionsof a loan.

Credit line (questions 1-7)A credit line is a facility with a stated maximumamount, which an enterprise is entitled toborrow from a bank at any given time. In thesurvey, developments regarding credit linesshould be interpreted as changes in the netamount drawn under either an existing or a newcredit line.

Credit standards (questions 1, 2, 6, 8, 9, 11 and16)Credit standards are the internal guidelines orcriteria, which reflect a bank’s loan policy.They are the written and unwritten criteria, orother practices related to this policy, whichdefine the types of loan a bank considersdesirable and undesirable, the designatedgeographic priorities, the collateral deemedacceptable and unacceptable, etc. In thesurvey, changes in written loan policies shouldbe considered together with changes in theirapplication.

Credit terms and conditions (questions 3, 10and 12)The terms and conditions of a loan refer to thespecific obligations agreed upon by the lenderand the borrower. In the context of this bank

lending survey, they consist of the direct priceor interest rate, the maximum size of the loanand the access conditions, and other terms andconditions in the form of non-interest ratecharges (i.e. fees), collateral requirements(including compensating balances), loancovenants and maturity (short versus long-term).

Enterprises (questions 1, 4, 6 and 7)Enterprises refer to non-financial corporations,i.e. all private and public institutional units,whatever their size and legal form, which arenot principally engaged in financialintermediation but rather in the production ofgoods and non-financial services.

Enterprise size (questions 1, 4, 6 and 7)The distinction between large and small andmedium-sized enterprises is based on annualsales. A firm is considered large if its annualnet turnover is more than €50 million.

Expectations regarding general economicactivity (question 11)This includes changes in the unemploymentoutlook. Any other relevant changes in socio-economic factors can be inserted under theoption “Other factors”.

Households (questions 8-17)Households are individuals or groups ofindividuals acting as consumers or asproducers of goods and non-financial servicesexclusively intended for their own finalconsumption and small-scale marketproducers.

Housing market prospects (question 9)This includes the risk on collateral demanded.

LoansThe loans covered by the bank lending surveyare those granted to euro area residents bydomestic branches, including loans or creditlines to enterprises, loans to households forhouse purchase, and consumer credit and otherlending to households.

57ECB


ANNEX 2

The definition of loans is that given inRegulation (EC) No. 2423/2001 of theEuropean Central Bank of 22 November 2001concerning the consolidated balance sheet ofthe monetary financial institutions (MFI)sector (ECB/2001/13). However, interbankloans should be excluded. Following thisdefinition, financial (but not operating) leasesgranted by an MFI are to be recorded as loans.For the purposes of the survey, factoring, ifprovided by an MFI, should also be treated as aloan. Financial leasing and factoring offered byinstitutions other than MFIs should not beincluded.

Loan-to-value ratio (question 10)The ratio of the amount borrowed to theappraisal or market value of the underlyingcollateral, usually taken into consideration inrelation to loans used for real estate financing.

Maturity (questions 1, 4, 6 and 7)The concept of maturity used in the banklending survey is original maturity, and onlytwo different types are used, i.e. short-term andlong-term. Short-term loans are loans with anoriginal maturity of one year or less and,consequently, long-term loans are loans thathave an original maturity of more than oneyear.

Non-banks (questions 2, 5, 9 and 11)In general these are non-monetary financialcorporations. More specifically, they includeinsurance corporations and pension funds,financial auxiliaries and other financialintermediaries.

Non-interest rate charges (questions 3, 10and 12)These are various kinds of fees which can bepart of the pricing of a loan, such ascommitment fees on revolving loans,administration fees (e.g. document preparationcosts), and charges for enquiries, guaranteesand credit insurance.


The detailed structure of the October 2004Federal Reserve Senior Loan Officer OpinionSurvey on Bank Lending Practices is as follows(Federal Reserve Board (2004)). The surveycovers both businesses (questions 1-11), whichare for a number of questions separated intolarge and middle-market firms (annual sales of$50 million or more) and small firms (annualsales of less than $50 million), and households(questions 12-19). Loans to businesses areseparated into so-called commercial andindustrial (C&I) loans (questions 1-9) andcommercial real estate loans (questions 10-11).The total number of questions in the October2004 survey is slightly lower than in previoussurveys in 2004 and in the October 2003survey. The October 2004 survey does paymore attention to commercial and industrialloan developments compared with previoussurveys, particularly regarding the identity andnature of increased competition from othersources of business credit (such as capitalmarkets, other banks and other financialintermediaries such as hedge funds andinsurance companies) and banks’ outlook forbusiness loan credit quality. Loans tohouseholds are separated into residentialmortgage loans (questions 12-13) andconsumer lending (questions 14-19).

Of the 19 questions in the October 2004 survey,13 focus on more supply-oriented factors, suchas credit standards, terms of loans, willingnessto lend and securitisation of loans, and six ondemand for loans. Thus, the Fed’s bank lendingsurvey is more oriented towards loan supplythan loan demand. The time horizon of thesurvey is clearly backward-looking, with themain focus on developments “over the pastthree months”, although a few questions arerelated to the current situation. All in all, noneof the questions in the October 2004 survey askrespondents about future developments. Themost recent question of a forward-lookingnature was in the April 2004 survey, whenrespondents were asked about the likelyaverage rate of increase in home prices duringthe next twelve months. This question was asupplementary one.

ANNEX 3

TH E O C TOB ER 2 0 0 4 F ED ER A L R E S E RV ES EN I O R L O AN O F F I C E R O P I N I ON S U RV E YON B ANK L END I NG P R A C T I C E S

A separate set of 11 questions in the October2004 survey focuses on bank lending practicesat selected branches and agencies of foreignbanks in the US. The results are based on theanswers of 20 respondents. The structure ofthese questions is similar to the core set ofquestions of the questionnaire which is sent tothe US banks in the sample group.

59ECB


ANNEX 3

(Status of policy as of October 2004)

1 As published by the Federal Reserve Board on its website. SeeFederal Reserve Board (2004).

2 The sample is selected from among the largest banks in eachFederal Reserve District. In the table, large banks are def inedas those with total domestic assets of $20 billion or more as ofJune 30, 2004. The combined assets of the 35 large bankstotaled $3.62 trillion, compared to $3.84 trillion for the entirepanel of 57 banks, and $7.04 trillion for all domesticallychartered, federally insured commercial banks.

applications for C&I loans or credit lines –other than those to be used to finance mergersand acquisitions – to large and middle-marketfirms and to small firms changed? (If your bankdefines firm size differently from thecategories suggested below, please use yourdefinitions and indicate what they are.)

All Respondents Large Banks Other Banks

Banks Pct Banks Pct Banks Pct

Tightened considerably 0 0.0 0 0.0 0 0.0Tightened somewhat 0 0.0 0 0.0 0 0.0Remained basically unchanged 45 78.9 27 77.1 18 81.8Eased somewhat 12 21.1 8 22.9 4 18.2Eased considerably 0 0.0 0 0.0 0 0.0

Total 57 100.0 35 100.0 22 100.0

A. Standards for large and middle-market f irms

(annual sales of $50 million or more)




Total 55 100.0 34 100.0 21 100.0

B. Standards for smal l f irms

(annual sales of less than $50 million)

Questions 1-3 ask about changes in your bank’scommercial and industrial (C&I) lendingpolicies over the past three months. If yourbank’s lending policies have not changed overthe past three months, please report them asunchanged even if the policies are eitherrestrictive or accommadtive relative to longer-term norms. If your bank’s policies havetightened or eased over the past three months,please so report them regardless of how theystand relative to longer-term norms. Also,please report changes in enforcement ofexisting policies as changes in policies.

1. Over the past three months, how have yourbank’s credit standards for approving

S EN I O R L O AN O F F I C E R O P I N I ON S U RV E Y ONBANK L END I NG P R A C T I C E S AT S E L E C T ED L A RG EBANK S I N T H E UN I T ED S TAT E S 1 , 2


2. For applications for C&I loans or credit lines– other than those to be used to finance mergersand acquisitions – from large and middle-market firms and from small firms that yourbank currently is willing to approve, how havethe terms of those loans changed over the past

three months? (Please assign each term anumber between 1 and 5 using the followingscale: 1 = tightened considerably, 2 = tightenedsomewhat, 3 = remained basically unchanged,4 = eased somewhat, 5 = eased considerably.)


Mean Mean Mean

Maximum size of credit lines 3.13 3.09 3.19Costs of credit lines 3.24 3.24 3.24Spreads of loan rates over your bank’s cost of funds 3.40 3.38 3.43(wider spreads = tightened, narrower spreads = eased)Premiums charged on riskier loans 3.11 3.12 3.10Loan covenants 3.18 3.21 3.14Collateralization requirements 2.98 2.97 3.00Other 3.50 3.50 3.50

Number of banks responding 55 34 21

B. Terms for smal l f irms



Mean Mean Mean

Maximum size of credit lines 3.31 3.27 3.36Costs of credit lines 3.35 3.39 3.27Spreads of loan rates over your bank’s cost of funds(wider spreads = tightened, narrower spreads = eased) 3.52 3.56 3.45Premiums charged on riskier loans 3.21 3.24 3.18Loan covenants 3.16 3.18 3.14Collateralization requirements 3.00 2.94 3.09Other 4.00 4.00 0.00


A. Terms for large and middle-market f irms


61ECB


ANNEX 3

3. If your bank has tightened or eased its creditstandards or its terms for C&I loans or creditlines over the past three months (as described inquestions 1 and 2), how important have beenthe following possible reasons for the change?

(Please respond to either A, B, or both asappropriate and rate each possible reasonusing the following scale: 1 = not important,2 = somewhat important, 3 = very important.)


Mean Mean Mean

Improvement in your bank’s current or expected capital position 1.23 1.35 1.00More favorable or less uncertain economic outlook 1.87 1.90 1.80Improvement in industry-specific problems 1.36 1.42 1.22More aggressive competition from other banks or nonbank lenders(other financial intermediaries or the capital markets) 2.47 2.43 2.55Increased tolerance for risk 1.43 1.45 1.40Increased liquidity in the secondary market for these loans 1.33 1.45 1.10Reduction in defaults by borrowers in public debt markets 1.31 1.37 1.20Other 2.00 2.00 0.00


B. Possible reasons for easing credit standards or loan terms


Mean Mean Mean

Deterioration in your bank’s current or expected capital position 1.13 1.00 1.33Less favorable or more uncertain economic outlook 1.38 1.00 2.00Worsening of industry-specific problems 1.86 1.80 2.00Less aggressive competition from other banks or nonbank lenders(other financial intermediaries or the capital markets) 1.00 1.00 1.00Reduced tolerance for risk 1.75 1.40 2.33Decreased liquidity in the secondary market for these loans 1.00 1.00 1.00Increase in defaults by borrowers in public debt markets 1.00 1.00 1.00Other 1.00 1.00 0.00


A. Possible reasons for t ightening credit standards or loan terms


In the most recent three surveys, respondentbanks overall have reported an easing ofbusiness lending standards and terms despite apickup in loan demand. Banks that have easedstandards or terms have indicated that theyhave done so primarily in response to increasedcompetition from other sources of businesscredit. Questions 4 and 5 ask about the identityand nature of this competition. Question 6 asksabout your bank’s outlook for business loancredit quality over the next year.

4. If your bank has eased standards or termssince the beginning of the year as a result ofgreater competitive pressures in the C&I loanmarket, how has the degree of competitionfrom the following alternative sources of fundschanged during that period? (Please assign, forthose entities listed that your bank views as apotential source of credit for your C&Icustomers, a number between 1 and 5 using thefollowing scale: 1 = increased considerably,2 = increased somewhat, 3 = has beenlittle changed, 4 = decreased somewhat,5 = decreased considerably.)



Primarily temporary, reflectingcurrent economic conditions 11 21.6 6 18.2 5 27.8Primarily permanent, reflecting a changein the structure of the C&I loan market 15 29.4 10 30.3 5 27.8Not clear at this point 25 49.0 17 51.5 8 44.4

Total 51 100.0 33 100.0 18 100.0


Mean Mean Mean

Capital markets (commercial paper, bonds, equity) 2.54 2.48 2.67Special purpose investment vehicles(for example, collaterized loan obligations) 2.57 2.43 2.83Insurance companies 2.76 2.71 2.86Investment banks 2.46 2.29 2.77U.S. commercial banks 1.74 1.74 1.75Foreign banks 2.52 2.54 2.44Hedge funds 2.72 2.62 3.00Other 2.00 0.00 2.00


5. Does your bank view this increasingcompetition from other sources of businesscredit as primarily temporary, reflecting

current economic market conditions, or as amore permanent change in the structure of theC&I loan market?

63ECB


ANNEX 3



Loan quality is likely to continue toimprove 16 28.1 10 28.6 6 27.3Loan quality is likely to stabilizearound current levels 37 64.9 24 68.6 13 59.1Loan quality is likely to begin to decline 4 7.0 1 2.9 3 13.6

Total 57 100.0 35 100.0 22 100.0

6. Over the past two years, C&I loandelinquencies and chargeoffs have improvedsubstantially. Looking ahead over the nextyear, and assuming that economic activity

progresses in line with consensus forecasts,what is your bank’s outlook for these measuresof C&I loan quality?

Questions 7-9 deal with changes in demand forC&I loans over the past three months.

7. Apart from normal seasonal variation, howhas demand for C&I loans changed over thepast three months? (Please consider only fundsactually disbursed as opposed to requests fornew or increased lines of credit.)



Substantially stronger 0 0.0 0 0.0 0 0.0Moderately stronger 21 36.8 16 45.7 5 22.7About the same 30 52.6 16 45.7 14 63.6Moderately weaker 6 10.5 3 8.6 3 13.6Substantially weaker 0 0.0 0 0.0 0 0.0

Total 57 100.0 35 100.0 22 100.0

A. Demand for C&I loans from large and middle-market f irms





Total 55 100.0 34 100.0 21 100.0

B. Demand for C&I loans from smal l f irms



8. If demand for C&I loans has strengthened orweakened over the past three months (asdescribed in question 7), how important havebeen the following possible reasons for the

change? (Please respond to either A, B, or bothas appropriate and rate each possible reasonusing the following scale: 1 = not important,2 = somewhat important, 3 = very important.)


Mean Mean Mean

Customer inventory financing needs decreased 1.50 1.67 1.40Customer accounts receivable financing needs decreased 1.63 1.67 1.60Customer investment in plant or equipment decreased 2.13 2.33 2.00Customer internally generated funds increased 2.13 2.33 2.00Customer merger or acquisition financing needs decreased 2.00 2.33 1.80Customer borrowing shifted from your bank to other bank ornon-bank credit sources because these other sources becamemore attractive 2.13 2.33 2.00Other 3.00 3.00 0.00


B. I f weaker loan demand (answer 4 or 5 to questions 7A or 7B), possible reasons


Mean Mean Mean

Customer inventory financing needs increased 1.93 1.85 2.14Customer accounts receivable financing needs increased 1.96 1.90 2.14Customer investment in plant or equipment increased 1.85 1.80 2.00Customer internally generated funds decreased 1.08 1.05 1.14Customer merger or acquisition financing needs increased 1.62 1.58 1.71Customer borrowing shifted to your bank from other bank ornon-bank sources because these other sources became less attractive 1.81 1.90 1.57Other 1.00 1.00 0.00


A. I f stronger loan demand (answer 1 or 2 to question 7A or 7B), possible reasons

65ECB


ANNEX 3



The number of inquiries has increasedsubstantially 1 1.8 1 2.9 0 0.0The number of inquiries has increasedmoderately 21 38.2 13 38.2 8 38.1The number of inquiries has stayedabout the same 32 58.2 20 58.8 12 57.1The number of inquiries has decreasedmoderately 1 1.8 0 0.0 1 4.8The number of inquiries has decreasedsubstantially 0 0.0 0 0.0 0 0.0

Total 55 100.0 34 100.0 21 100.0

9. At your bank, how has the number ofinquiries from potential business borrowersregarding the availability and terms of newcredit lines or increases in existing lines

changed over the past three months? (Pleaseconsider only inquiries for additional C&I linesas opposed to the refinancing of existingloans.)





Total 57 100.0 35 100.0 22 100.0




Total 57 100.0 35 100.0 22 100.0

Questions 10-11 ask about commercial realestate loans at your bank, includingconstruction and land development loans andloans secured by nonfarm nonresidential realestate. Question 10 deals with changes in yourbank’s standards over the last three months.Question 11 deals with changes in demand. Ifyour bank’s lending standards or terms havenot changed over the relevant period, pleasereport them as unchanged even if they are eitherrestrictive or accommodative relative to

longer-term norms. If your bank’s standards orterms have tightened or eased over the relevantperiod, please so report them regardless of howthey stand relative to longer-term norms. Also,please report changes in enforcement ofexisting standards as changes in standards.

10. Over the past three months, how have yourbank’s credit standards for approvingapplications for commercial real estate loanschanged?

11. Apart from normal seasonal variation, howhas demand for commercial real estate loanschanged over the past three months?

67ECB


ANNEX 3




Total 53 100.0 33 100.0 20 100.0




Total 53 100.0 33 100.0 20 100.0

Questions 12-13 ask about residential mortgageloans at your bank. Question 12 deals withchanges in your bank’s credit standards overthe past three months and Question 13 dealswith changes in demand over the same period.If your bank’s credit standards have notchanged over the relevant period, please reportthem as unchanged even if the standards areeither restrictive or accomodative relative tolonger-term norms. If your bank’s creditstandards have tightened or eased over the

relevant period, please so report themregardless of how they stand relative to longer-term norms. Also, please report changes inenforcement of existing standards as changesin standards.

12. Over the past three months, how have yourbank’s credit standards for approvingapplications from individuals for mortgageloans to purchase homes changed?

13. Apart from normal season variation, howhas demand for mortgages to purchase homeschanged over the past three months? (Please

consider only new originations as opposed tothe refinancing of existing mortgages.)




Much more willing 1 1.9 1 2.9 0 0.0Somewhat more willing 6 11.1 4 11.8 2 10.0About unchanged 47 87.0 29 85.3 18 90.0Somewhat less willing 0 0.0 0 0.0 0 0.0Much less willing 0 0.0 0 0.0 0 0.0

Total 54 100.0 34 100.0 20 100.0




Total 34 100.0 18 100.0 16 100.0




Total 52 100.0 33 100.0 19 100.0

Questions 14-19 ask about consumer lending atyour bank. Question 14 deals with changes inyour bank’s willingness to make consumerloans over the past three months. Questions15-18 deal with changes in credit standards andloan terms over the past three months, pleasereport them as unchanged even if the policiesare either restrictive or accomodative relativeto longer-term norms. If your bank’s policies

have tightened or eased over the past threemonths, please so report them regardless ofhow they stand relative to longer-term norms.Also, please report changes in enforcement ofexisting policies as changes in policies.

14. Please indicate your bank’s willingness tomake consumer installment loans now asopposed to three months ago.


applications for credit cards from individualsor households changed?


applications for consumer loans other thancredit card loans changed?

69ECB


ANNEX 3


Mean Mean Mean

Credit limits 2.96 2.93 3.00Spreads of interest rates charged on outstanding balancesover your bank’s cost of funds (wider spreads = tightened,narrower spreads = eased) 2.96 2.93 3.00Minimum percent of outstanding balances required to berepaid each month 2.96 3.00 2.92Minimum required credit score (increased score = tightened,reduced score = eased) 2.93 3.00 2.85The extent to which loans are granted to some customers thatdo not meet credit scoring thresholds (increased = eased,decreased = tight-ened) 2.89 3.00 2.77Other 0.00 0.00 0.00



Mean Mean Mean

Maximum maturity 3.08 3.12 3.00Spreads of loan rates over your bank’s cost of funds(wider spreads = tightened, narrower spreads = eased) 3.06 3.06 3.05Minimum required downpayment 3.02 3.03 3.00Minimum required credit score (increased score = tightened,reduced score = eased) 3.04 3.09 2.95The extent to which loans are granted to some customers thatdo not meet credit scoring thresholds (increased = eased,decreased = tight-ened) 3.00 3.00 3.00Other 2.00 0.00 2.00


17. Over the past three months, how has yourbank changed the following terms andconditions on new or existing credit cardaccounts for individuals or households?(Please assign each term a number between

1 and 5 using the following scale: 1 = tightenedconsiderably, 2 = tightened somewhat,3 = remained basically unchanged, 4 = easedsomewhat, 5 = eased considerably.

18. Over the past three months, how has yourbank changed the following terms andconditions on consumer loans other than creditcard loans? (Please assign each term a number

between 1 and 5 using the following scale:1 = tightened considerably, 2 = tightenedsomewhat, 3 = remained basically unchanged,4 = eased somewhat, 5 = eased considerably.)





Total 54 100.0 34 100.0 20 100.0

19. Apart from normal seasonal variation, howhas demand for consumer loans of all typeschanged over the past three months?

71ECB


ANNEX 4

The detailed structure of the Bank of Japan’sSenior Loan Officer Opinion Survey on BankLending Practices as implemented in October2004 is as follows (Bank of Japan (2004)).Unlike the Federal Reserve’s survey, the Bankof Japan (BoJ) survey is organised into a groupof questions covering demand for loans (sixquestions) and another group of questions onbank lending policies (seven questions). Thus,the survey is relatively evenly broken downinto questions covering supply and demandfactors. In addition, the October 2004 surveyincludes one ad hoc question on the share ofloans extended to firms that have beendowngraded or upgraded in the past threemonths according to the banks’ internal creditratings.

The survey covers three groups of borrowers,namely firms, local government andhouseholds. Of these, local governments areonly included in two questions on loan demand.The first group is subdivided into large firms,medium-sized firms and small firms, wherebylarge firms are defined as corporations withcapital of one billion yen or more and with morethan 300 regular employees, while small firmshave capital of 300 million yen or less and haveup to 300 regular employees. In addition, abreakdown of firms is provided by industry(manufacturing and non-manufacturing, whichis segmented into construction and real estate,finance and insurance and other non-manufacturing). Thus, the BoJ survey gives arather detailed overview of developments inbanks’ lending policies regarding the Japanesecorporate sector. Loans to households areseparated into housing loans and consumerloans. In total, the survey includes sevenquestions on loans only to firms, two questionson loans only to households, two questions onloans to both firms and households, and twoquestions on loans to all three sectors, i.e.firms, households and local government. Thus,the BoJ bank lending survey primarilyconcentrates on the business sector, which iscovered in 11 of the 13 regular questions.

ANNEX 4

TH E O C TOB ER 2 0 0 4 B ANK O F J A PAN S EN I O RLOAN O F F I C E R O P I N I ON S U RV E Y ON B ANKL END I NG P R A C T I C E S AT L A RG E J A PAN E S E B ANK S

Regarding the time horizon of the survey, theBoJ survey covers both the past and the future,as four questions are forward-looking andcover the next three months. With respect toother particular characteristics, the surveyincludes two questions on changes in thespread of loan rates over the banks’ cost offunds, which differentiate firms according totheir ratings (high, medium or low).


S EN I O R L OAN O F F I C E R O P I N I ON S U RV E Y ONBANK L END I NG P R A C T I C E S AT L A RG E J A PAN E S EB ANK S ( O C TOBER 2 0 0 4 ) 1

DEMAND FOR LOANS (QUESTIONS 1-6)

1. How has demand for loans from borrowers (firms, local governments, and households)changed over the past three months (apart from normal seasonal fluctuations)?

Note: DI for demand for loans = (percentage of respondents selecting “substantially stronger” + percentage of respondents selecting“moderately stronger” × 0.5 ) - (percentage of respondents selecting “substantially weaker” + percentage of respondents selecting“moderately weaker” × 0.5) (same for question 2 and 4).

Substantiallystronger

Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

2 4 38 5 1

(4%) (8%) (76%) (10%) (2%)

1 4 41 4 0

(2%) (8%) (82%) (8%) (0%)

2 12 33 2 0

(4%) (25%) (67%) (4%) (0%)

Number of respondents selecting each option(percentage of respondents selecting each option)

Total

50

50

49

(July 2004)

Firms 1 -18

DI for demand forloans(% point)

Local governments 2 -10

Households 14 8

*

Demand for loans from borrowers: c lass i f iedby borrower type

(DI, % points)

Note: See question 6. for the outlook.

firms local governmentshouseholds

-30

-20

-10

0

10

20

30

-30

-20

-10

0

10

20

30

Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct.

Stro

nger

Wea

ker

Outlook

2001 2002 2003 20042000

1 As published by the Bank of Japan on its website. See Bank ofJapan (2004).

73ECB


ANNEX 4

2. How has demand for loans from firms changed over the past three months? Please give abreakdown by industry and firm size.

Large firms 3 -9

All industriesDI for demand for

loans(% point)(July 2004)

Medium-sized firms 4 -11

Small firms 0 -17


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

1 8 35 5 1

(2%) (16%) (70%) (10%) (2%)

1 4 43 2 0

(2%) (8%) (86%) (4%) (0%)

2 2 41 4 1

(4%) (4%) (82%) (8%) (2%)

Total

50


50

50


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

2 6 34 7 1

(4%) (12%) (68%) (14%) (2%)

1 5 41 3 0

(2%) (10%) (82%) (6%) (0%)

2 6 37 4 1

(4%) (12%) (74%) (8%) (2%)

Total

50


50

50

Large firms 2 -9

ManufacturingDI for demand for



Small firms -4 -10


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

0 9 35 5 1

(0%) (18%) (70%) (10%) (2%)

1 4 42 3 0

(2%) (8%) (84%) (6%) (0%)

0 3 41 5 1

(0%) (6%) (82%) (10%) (2%)

Total

50


50

50

Large firms 1 -9

NonmanufacturingDI for demand for



Small firms 4 -16

Demand for loans from f irms: c lass i f ied byf irm size

(DI, % points)

large firms medium-sized firmssmall firms


Stro

nger

Wea

ker

2001 2002 2003 20042000

-30

-20

-10

0

10

-30

-20

-10

0

10


Of which:

3.a If demand for loans from firms has increased at your bank (that is, the answer to question 2,“All industries” is either “Substantially stronger” or “Moderately stronger”), to what factorsdo you attribute this increase? (Please rate each possible reason using the following scale:3 = important, 2 = somewhat important, 1 = not important.)

Note: Average is calculated by multiplying theshare of respondents selecting each option bythe scale of each option, then adding up theresult (same for question 3b, 5a and 5b).

Large firms 2 -9

Constructionand real estate


(July 2004)


Small firms 3 -17

Large firms 1 -1

Financeand insurance


(July 2004)


Small firms 4 -4

Large firms 1 -7

Other nonmanufacturingDI for demand for



Small firms -3 -13


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

3 1 42 3 1

(6%) (2%) (84%) (6%) (2%)

5 1 38 5 1

(10%) (2%) (76%) (10%) (2%)

2 2 44 1 1

(4%) (4%) (88%) (2%) (2%)


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

3 4 35 7 1

(6%) (8%) (70%) (14%) (2%)

1 1 44 3 0

(2%) (2%) (90%) (6%) (0%)

3 3 38 5 0

(6%) (6%) (78%) (10%) (0%)


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

2 5 36 6 1

(4%) (10%) (72%) (12%) (2%)

0 8 39 3 0

(0%) (16%) (78%) (6%) (0%)

0 4 40 5 1

(0%) (8%) (80%) (10%) (2%)

Total

50


50

50

Total

50


49

49

Total

50


50

50

Large firmsMedium-

sized firmsSmall firms

Average Average Average

(1) Customers' sales increased 2.00 1.80 1.50

1.89 2.00 1.50

1.00 1.00 1.25

1.11 1.00 1.25

1.11 1.00 1.25

(6) Decline in interest rates 1.33 1.60 1.50

(7) Other factors 1.22 1.00 1.50


(2) Customers' fixed investment increased

(3) Customers' funding from other sourcesbecame difficult to obtain(4) Customers' internally-generated fundsdecreased(5) Customers' borrowing shifted from othersources to your bank

75ECB


ANNEX 4

3.b If demand for loans from firms has decreased at your bank (that is, the answer to question 2,“All industries” is either “Substantially weaker” or “Moderately weaker”), to what factorsdo you attribute this decrease? (Please rate each possible reason using the following scale:3 = important, 2 = somewhat important, 1 = not important.)

4. How has demand from households for housing and consumer loans changed?

Large firmsMedium-



(1) Customers' sales decreased 1.33 1.50 1.80

1.67 2.00 1.80

2.17 1.50 1.40

2.00 2.00 1.60

2.17 2.00 1.40

(6) Rise in interest rates 1.00 1.00 1.00

(7) Other factors 1.00 1.00 1.00


(2) Customers' fixed investment decreased

(3) Customers' funding from other sourcesbecame easy to obtain(4) Customers' internally-generated fundsincreased(5) Customers' borrowing shifted from yourbank to other sources

Housing loans 16 11


(July 2004)

Consumer loans -6 -6


Moderatelystronger

Aboutthe same

Moderatelyweaker

Substantiallyweaker

3 12 32 2 0

(6%) (25%) (65%) (4%) (0%)

0 0 42 4 1

(0%) (0%) (89%) (9%) (2%)

Total

49


47

Demand for loans from households:c lass i f ied by type of loan

(DI, % points)

housing loansconsumer loans


Stro

nger

Wea

ker

2001 2002 2003 20042000

-15

-10

-5

0

5

10

15

20

25

-15

-10

-5

0

5

10

15

20

25


5.a If demand for loans from households has increased at your bank (that is, the answer to question4 is either “Substantially stronger” or “Moderately stronger”), to what factors do you attributethis increase? (Please rate each possible reason using the following scale: 3 = important,2 = somewhat important, 1 = not important.)

5.b If demand for loans from households has decreased at your bank (that is, the answer toquestion 4 is either “Substantially weaker” or “Moderately weaker”), to what factors doyou attribute this decrease? (Please rate each possible reason using the following scale:3 = important, 2 = somewhat important, 1 = not important.)

6. How are demand for loans from borrowers (firms, local governments, and households) likelyto change over the next three months (apart from normal seasonal fluctuations)?

Note: DI for demand for loans = (percentage of respondents selecting “increase substantially” + percentage of respondents selecting“increase somewhat” × 0.5 ) - (percentage of respondents selecting “decrease substantially” + percentage of respondents selecting“decrease somewhat” × 0.5).

Housing loans Consumer loans

Average Average

(1) Increase in housing investment 1.60 n.a.

1.27 n.a.

(3) Decrease in income 1.07 n.a.

(4) Decline in interest rates 2.33 n.a.

(5) Other factors 1.40 n.a.

Number of banks responding 15 0

(2) Increase in household consumption

Housing loans Consumer loans

Average Average

(1) Decrease in housing investment 2.00 1.00

1.50 2.40

(3) Increase in income 1.00 1.00

(4) Rise in interest rates 1.00 1.00

(5) Other factors 1.50 1.00

Number of banks responding 2 5

(2) Decrease in household consumption

(July 2004)

Firms 5 5


0

Households 6 6

Local governments 1

Increasesubstantially

Increasesomewhat

Remainabout

the same

Decreasesomewhat

Decreasesubstantially

0 7 41 2 0

(0%) (14%) (82%) (4%) (0%)

0 1 49 0 0

(0%) (2%) (98%) (0%) (0%)

1 6 40 2 0

(2%) (12%) (82%) (4%) (0%)


Total

50

50

49

77ECB


ANNEX 4

LENDING POLICIES (QUESTIONS 7-13)

7. Over the past three months, how have your bank’s credit standards for approving applicationsfrom firms and households changed?

Note: DI for credit standards = (percentage of respondents selecting “eased considerably” + percentage of respondents selecting“eased somewhat” × 0.5 ) - (percentage of respondents selecting “tightened considerably” + percentage of respondents selecting“tightened somewhat” × 0.5).

(July 2004)

Large firms 14 13

DI for creditstandards(% point)

Medium-sized firms 26 24

35

39Households

Small firms 35

37

Easedconsiderably

Easedsomewhat

Remainedbasically

unchanged

Tightenedsomewhat

Tightenedconsiderably

2 10 38 0 0

(4%) (20%) (76%) (0%) (0%)

4 18 28 0 0

(8%) (36%) (56%) (0%) (0%)

6 23 21 0 0

(12%) (46%) (42%) (0%) (0%)

11 16 22 0 0

(22%) (33%) (45%) (0%) (0%)


Total

50

50

50

49

Credit standards for approving appl icationsfor loan: c lass i f ied by f irms and households

(DI, % points)


large firms medium-sized firmssmall firmshouseholds

Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct.2001 2002 2003 20042000

0

10

20

30

40

50

60

0

10

20

30

40

50

60

Eas

ed Outlook


8.a If your bank has eased its credit standards for loans to firms over the past three months (asdescribed in questions 7), what were the important factors that led to the change? (Please rateeach possible reason using the following scale: 3 = important, 2 = somewhat important, 1 = notimportant.)

Note: Average is calculated by multiplying theshare of respondents selecting each option bythe scale of each option, then adding up theresult (same for question 8b).

9. Over the past three months, how have the terms and conditions of loans to firms changed?

Note: DI for terms and conditions of loans =(percentage of respondents selecting “easedconsiderably” + percentage of respondentsselecting “eased somewhat” × 0.5) - (percentage ofrespondents selecting “tightened considerably”+ percentage of respondents selecting “tightenedsomewhat” × 0.5).

Large firmsMedium-



1.42 1.50 1.48

1.75 1.45 1.31

1.67 1.45 1.34

2.58 2.36 2.28

1.92 1.50 1.31

1.92 1.36 1.10

1.33 1.27 1.24

(8) Others 1.17 1.18 1.28


(2) A more favorable or less uncertain economicoutlook

(7) An increased tolerance for risk

(3) An improvement in industry- or firm-specificproblems

(4) More aggressive competition from other banks

(5) More aggressive competition from non-banks

(6) More aggressive competition from capitalmarkets

(1) An improvement in your bank's asset portfolio

8.b If your bank has tightened its credit standards for loans to firms over the past three months (asdescribed in questions 7), what were the important factors that led to the change? (Please rateeach possible reason using the following scale: 3 = important, 2 = somewhat important, 1 = notimportant.)

Large firmsMedium-



n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

n.a. n.a. n.a.

(8) Others n.a. n.a. n.a.


(2) A less favorable or more uncertain economicoutlook(3) A worsening of industry- or firm-specificproblems

(7) A reduced tolerance for risk

(4) Less aggressive competition from other banks

(5) Less aggressive competition from non-banks

(6) Less aggressive competition from capitalmarkets

(1) An deterioration in your bank's asset portfolio

Large firmsMedium-


(1) Maximum size of credit lines 9 10 11

6 8 4

(3) Premiums charged on riskier loans 2 -1 -2

(4) Collateralization requirements 3 5 6

(5) Others 0 0 0


(2) Spreads of loan rates over your bank's cost offunds (wider spreads = tightened, narrower spreads =eased)

DI for terms and conditions of loans (% point)

79ECB


ANNEX 4

10. Over the past three months, how has your bank changed the spreads of loan rates over yourbank’s cost of funds?

Note: DI for spreads of loan rates = percentage of respondents selecting “increased” - percentage of respondents selecting“decreased”.

11. Over the next three months, how are your bank’s credit standards for firms and householdslikely to change?

Note: DI for credit standards = (percentage of respondents selecting “ease considerably” + percentage of respondents selecting“ease somewhat” × 0.5) - (percentage of respondents selecting “tighten considerably” + percentage of respondents selecting “tightensomewhat” × 0.5).

Medium ratings -8 -8

For firms withDI for spreads of

loan rates(% point)(July 2004)

Low ratings 18 16

High ratings -20 -20

Increased Remainedbasically

unchanged Decreased

1 38 11

( 2% ) ( 76% ) ( 22% )

2 42 6

( 4% ) ( 84% ) ( 12% )

9 41 0

( 18% ) ( 82% ) ( 0% )

50

50

50


Total

DI for creditstandards(% point)

Large firms 15 12

(July 2004)

Medium-sized firms

Small firms 42 35

Households 39 33

2330

Easeconsiderably

Easesomewhat

Remainbasically

unchanged

Tightensomewhat

Tightenconsiderably

4 7 39 0 0

( 8% ) ( 14% ) ( 78% ) ( 0% ) ( 0% )

5 20 25 0 0

( 10% ) ( 40% ) ( 50% ) ( 0% ) ( 0% )

10 22 18 0 0

( 20% ) ( 44% ) ( 36% ) ( 0% ) ( 0% )

12 14 23 0 0

( 25% ) ( 29% ) ( 47% ) ( 0% ) ( 0% )

50

Total

50


49

50

Spreads of loan rates over reporting banks’cost of funds: c lass i f ied by rating ofborrower f irm(DI, % points)


high ratings medium ratingslow ratings

Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct. Apr. Oct.2001 2002 2003 20042000

-40

-20

0

20

40

60

80

100

-40

-20

0

20

40

60

80

100

Incr

ease

dD

ecre

ased

Outlook


Note: DI for terms and conditions of loans = (percentage of respondents selecting “ease considerably” + percentage of respondentsselecting “ease somewhat” × 0.5) - (percentage of respondents selecting “tighten considerably” + percentage of respondentsselecting “tighten somewhat” × 0.5).

Note: DI for spreads of loan rates = percentage of respondents selecting “increase” - percentage of respondents selecting “decrease”.

Large firmsMedium-


(1) Maximum size of credit lines 11 12 13

5 4 0

(3) Premiums charged on riskier loans 3 0 0

(4) Collateralization requirements 3 4 6

(5) Others 0 0 0


(2) Spreads of loan rates over your bank's cost offunds (wider spreads = tightened, narrower spreads =eased)

DI for terms and conditions of loans (% point)

12. Over the next three months, how are your bank’s terms and conditions of loans to firms likelyto change?

Medium ratings 2 8

For firms withDI for spreads of

loan rates(% point)

High ratings -18

(July 2004)

-10

Low ratings 22 32

Increase Remainthe same

Decrease

1 39 10

( 2% ) ( 78% ) ( 20% )

7 37 6

( 14% ) ( 74% ) ( 12% )

12 37 1

( 24% ) ( 74% ) ( 2% )


50

Total

50

50

13. Over the next three months, how does your bank intend to change the spreads of loan rates?

81ECB


ANNEX 4

AD HOC SURVEY ON LOANS TO DOWNGRADED/UPGRADED FIRMS

What is the share (by value) of loans extended to firms that were downgraded or upgraded in thepast three months in your bank’s internal credit ratings as a percentage of the total loans in eachfirm size category?

Respondents were asked to select one option for each firm size category.

1. Loans to downgraded firms

2. Loans to upgraded firms

1)The aggregated loan amount of the f ifty reporting banks accounts for approximately 74 percent of average amount of outstandingof the domestic loans of Japanese private banks (city banks, regional banks, regional banksII, trust banks, long-term credit banks,and shinkin banks) in the year 2003.2) Households do not include small f irms owned and run by individuals.3) Local governments include administrations of prefectures and cities, as well as their directly managed businesses such as publictransportation, water, electricity and gas utilities, hospitals, and others.4) Def initions of f irm size are as follows. Large: Corporations with capital of ¥1 billion and over with more than 300 regularemployees (wholesaling and services f irms capitalized at ¥1 billion and over with more than 100 regular employees; and retailing,food and beverage services f irms capitalized at ¥1 billion and over with more than 50 regular employees). Small: Corporations withcapital of ¥300 million or less or with 300 regular employees or fewer (wholesaling firms capitalized at ¥100 million or less with 100regular employees or fewer; retailing, food and beverage services and other services firms capitalized at ¥50 million or less with 50regular employees or fewer (100 or fewer for services firms)). Medium-sized: Corporations which fall between the above twocategories.5) Rating in questions 10 and 13 refers to the banks’ internal ratings. These are broad ratings as indicated below. High: AAA-BBB fordomestic ratings of long-term corporate bonds. Medium: BB-B for domestic ratings of long-term corporate bonds. Low: CCC orlower for domestic ratings of long-term corporate bonds.

50% or more 25% or more 10% or more 5% or moreLess than

5%

0 0 5 14 30

(0%) (0%) (10%) (29%) (61%)

0 5 12 11 21

(0%) (10%) (25%) (22%) (43%)

0 0 6 18 25

(0%) (0%) (12%) (37%) (51%)


Total

Large firms 49

Medium-sized firms 49

Small firms 49

50% or more 25% or more 10% or more 5% or moreLess than

5%

0 3 12 12 22

(0%) (6%) (25%) (25%) (45%)

0 0 18 9 22

(0%) (0%) (37%) (18%) (45%)

0 0 7 17 25

(0%) (0%) (14%) (35%) (51%)


Total

Large firms 49

Medium-sized firms 49

Small firms 49

83ECB


REFERENCESR E F E R E N C E SAgresti, A.-M. and B. Mojon (2003), “Some stylised facts on the euro area business cycle”, in

Angeloni, I., A. Kasyhap and B. Mojon (eds) (2003), Monetary Policy Transmission in theEuro Area, Cambridge University Press.

Altunbaç, Y., G. de Bondt and D. Marqués (2004), “Bank capital, bank lending, and monetarypolicy in the euro area”, Kredit und Kapital, 37(4), 443-464.

Angeloni, I., A. K. Kashyap and B. Mojon (eds) (2003a), Monetary Policy Transmission in theEuro Area, Cambridge University Press.

Angeloni, I., A. K. Kashyap, B. Mojon and D. Terlizzese (2003b), “The output compositionpuzzle: a difference in the monetary transmission mechanism in the euro area and U.S.”,Journal of Money, Credit and Banking, 35(6), 1265-1306.

Angeloni, I. and M. Ehrmann (2003), “Monetary transmission in the euro area: early evidence”,Economic Policy, 37, 469-492.

Asea, P. K. and S. B. Blomberg (1998), “Lending cycles”, Journal of Econometrics, 83(1-2),March-April, 89-128.

Bank of Japan (2000), Introduction of the “Senior Loan Officer Opinion Survey on Bank LendingPractices at Large Japanese Banks” (“Loan Survey”), 21 April.

Bank of Japan (2001), Introduction of diffusion indexes to the “Senior Loan Officer OpinionSurvey on Bank Lending Practices at Large Japanese Banks”, 26 April.

Bank of Japan (2003), Revising the list of surveyed banks for the “Senior Loan Officer OpinionSurvey on Bank Lending Practices at Large Japanese Banks”, 24 April.

Bank of Japan (2004), “Senior Loan Officer Opinion Survey on Bank Lending Practices at LargeJapanese Banks – October 2004”, 26 October.

Bassett, W. F. and M. Carlson (2002), “Profits and balance sheet developments at U.S.commercial banks in 2001”, Federal Reserve Bulletin, June, 259-288.

Berger, A. N., and G. F. Udell (2002), “The institutional memory hypothesis and theprocyclicality of bank lending behavior”, Board of Governors of the Federal Reserve System,Finance and Economics Discussion Series, No 2003-02, December 2002.

Bergman, M., S. Gerlach and L. Jonung (1992), “External influences in Nordic business cycles:1870-1988”, Open Economies Review, 3(1), 1-22.

Bernanke, B. S. and A. S. Blinder (1988), “Credit, money, and aggregate demand”, AmericanEconomic Review, 78(2), May, 435-439.

Bernanke, B. S. and M. Gertler (1989), “Agency costs, net worth, and business fluctuations”,American Economic Review, 79(1), March, 14-31.


Bernanke, B. S. and M. Gertler (1995), “Inside the black box: the credit channel of monetarypolicy transmission”, Journal of Economic Perspectives, 9(1), 27-48.

Bernanke, B. S. and C. S. Lown (1991), “The credit crunch”, Brookings Papers on EconomicActivity, 2, 204-239.

Bondt, G. J. de (2000), Financial Structure and Monetary Transmission in Europe: A Cross-country Study, Edward Elgar.

Brunner, A. D. and W. B. English (1993), “Profits and balance sheet developments at U.S.commercial banks in 1992”, Federal Reserve Bulletin, 79, July, 649-673.

Döpke, J. (1999), “Stylized facts of Euroland’s business cycles”, Jahrbücher fürNationalökonomie und Statistik, 591-610.

Duca, J. and B. Garret (1995), “Credit availability, bank consumer lending, and consumerdurables”, Federal Reserve Bank of Dallas Working Paper, No 9514.

Eckstein, O. and A. Sinai (1986), “The mechanisms of the business cycle in the post-war era”, inR. Gordon (ed.) (1986), The American Business Cycle: Continuity and Change, University ofChicago Press, 39-122.

European Central Bank (2000), “Monetary policy transmission in the euro area”, MonthlyBulletin, July, 43-58.

European Central Bank (2002), “Recent findings on monetary policy transmission in the euroarea”, Monthly Bulletin, October, 43-53.

European Central Bank (2003), “New ECB statistics on MFI interest rates”, Monthly Bulletin,December, 23-25.

Els, P. J. A. van (1995), “Real business cycle models and money: a survey of theories and stylizedfacts”, Weltwirtschaftliches Archiv, 131(2), 223-264.

Federal Reserve Board (2004), “The October 2004 Senior Loan Officer Opinion Survey on BankLending Practices”, 15 November.

Fiorito, R. and T. Kollintzas (1992), “Stylized facts of business cycles in the G7 from a realbusiness cycles perspective”, CEPR Discussion Paper, No 681.

Gertler, M. and S. Gilchrist (1993), “The role of credit market imperfections in the monetarytransmission mechanism: arguments and evidence”, Scandinavian Journal of Economics,109(2), 309-340.

Greenwald, B. and J. E. Stiglitz (1993), “Financial market imperfections and business cycles”,Quarterly Journal of Economics, 108(1), 77-114.

Hall, S. (1999), “Theories and concepts of the credit channel”, Bank of England, mimeo, October.

85ECB


REFERENCES

Hamdani, K., A. P. Rodrigues and M. Varvatsoulis (1994), “Survey evidence on credit tighteningand the factors behind the recent credit crunch”, Federal Reserve Bank of New York, Studieson Causes and Consequences of the 1989-1992 Credit Slowdown, New York, 211-257.

Hida, N., K. Fujita, M. Ihara and N. Baba (2002), “What do we learn about current developmentsin the bank lending market in Japan from BoJ’s Senior Loan Officer Opinion Survey?”, Bankof Japan Market Review, No 2002-E-5, December.

Hull, J., M. Predescu and A. White (2004), “The relationship between credit default swap spreads,bond yields and credit rating announcements”, University of Toronto, Joseph L. RotmanSchool of Management Working Paper, January.

Juglar, C. [1862] (1889), Des crises commerciales et de leur retour périodique en France, enAngleterre et aux Etats-Unis, 2nd ed. Paris, Guillaumin.

Kakes, J. (2000), Monetary Transmission in Europe: The Role of Financial Markets and Credit,Edward Elgar.

Karras, G. (1994), “Sources of business cycles in Europe: 1960-1988. Evidence from France,Germany, and the United Kingdom”, European Economic Review, 38(9), 1763-1778.

Kashyap, A., J. Stein and D. Wilcox (1993), “Monetary policy and credit conditions: evidencefrom the composition of external finance”, American Economic Review, 83(1), March, 79-98.

Keeton, W.R. (1999), “Does faster loan growth lead to higher loan losses?”, Federal Reserve ofKansas City Economic Review, second quarter, 57-75.

Kennedy, P. (1998), A Guide to Econometrics, Blackwell.

Keynes, J. M. (1936), The General Theory of Employment, Interest, and Money.

Kiyotaki, N. and J. Hardman-Moore (1997), “Credit cycles”, Journal of Political Economy,105(2), April, 211-249.

Kohler, M., E. Britton and T. Yates (2000), “Trade credit and the monetary transmissionmechanism”, Bank of England Working Paper Series, No 115, June.

Lown, C. S. (1990), “Banking and the economy: what are the facts?”, Federal Reserve Bank ofDallas Economic Review, September, 1-14.

Lown, C., D. P. Morgan and S. Rohatgi (2000), “Listening to loan officers: the impact ofcommercial credit standards on lending and output”, Federal Reserve Bank of New YorkEconomic Policy Review, 6(2), July, 1-16.

Lown, C. S. and D. P. Morgan (2002), “Credit effects in the monetary mechanism”, FederalReserve Bank of New York Economic Policy Review, 8(1), 217-235.


Lown, C. S. and D. P. Morgan (2004), “The credit cycle and the business cycle: new findingsusing the Loan Officer Opinion Survey”, Stockholm Institute for Financial ResearchResearch Report Series, No 27.

Manski, C. F., (2004), “Measuring expectations”, Econometrica, 72(5), 1329-1376.

Niehans, J. (1992), “Juglar’s credit cycles”, History of Political Economy, 24(3), 545-569.

Nilsen, J. H. (2002), “Trade credit and the bank lending channel”, Journal of Money, Credit andBanking, 34(1), 226-253.

Ortega, E. (1998), “The Spanish business cycle and its relationship to Europe”, Bank of SpainWorking Paper, No 9819.

Pazarbasioglu, C. (1997), “A credit crunch? Finland in the aftermath of the banking crisis”, IMFStaff Papers, 44(3), 315-327.

Peek, J. and E. Rosengren (1995), “The capital crunch: neither a borrower nor a lender be”,Journal of Money, Credit, and Banking, 27(3), 625-638.

Poole, W. (1993), “Credit veils and credit realities”, Federal Reserve Bank of New York QuarterlyReview, 18(1), 83-85.

Rixtel, A. van (2002), Informality and Monetary Policy in Japan: The Political Economy of BankPerformance, Cambridge University Press.

Rixtel, A. van, Y. Wiwattanakantang, T. Souma and K. Suzuki (2004), “Banking in Japan: willtoo big to fail prevail?”, in B. Gup (ed.) (2004), Too Big to Fail: Policies and Practices inGovernment Bailouts, Praeger, 253-283.

Romer, C. D. and D. H. Romer (1990), “New evidence on the monetary transmission mechanism”,Brookings Papers on Economic Activity, 1990(1), 149-213.

Schreft, S. L. and R. E. Owens (1991), “Survey evidence of tighter credit conditions: what does itmean?”, Federal Reserve Bank of Richmond Working Paper Series, No 91-5.

Schreft, S. L. and R. E. Owens (1995), “Identifying credit crunches”, Contemporary EconomicPolicy, 13(2), 63-76.

Sharpe, S. A. (1995), “Bank capitalization, regulation, and the credit crunch: a critical review ofthe research findings”, Board of Governors of the Federal Reserve System Finance andEconomics Discussion Series, No 95/20.

Sinai, A. (1993), “Financial and credit cycles: generic or episodic?”, Federal Reserve Bank ofNew York Quarterly Review, 18(1), 40-45.

Stanca, L. and M. Gallegati (1998), “The historical properties of Italian macroeconomicfluctuations: are business cycles all alike?”, Rivista Internazionale di Scienze Sociali, 106(2),April-June, 119-137.

87ECB


REFERENCES

Stiglitz, J. E. and A. Weiss (1981), “Credit rationing in markets with imperfect information”,American Economic Review, 71(3), June, 393-410.

Suzuki, T. (2004), “Credit channel of monetary policy in Japan: resolving the supply versusdemand puzzle”, Applied Economics, 36(21), 2385-2396.

Vihriälä, V. (1997), “Banks and the Finnish credit cycle 1986-1995”, Bank of Finland Studies,No E:7.

Weinberg, J. A. (1995), “Cycles in lending standards”, Federal Reserve Bank of RichmondEconomic Quarterly, 81(3), Summer, 1-18.

Wojnilower, A. M. (1980), “The central role of credit crunches in recent financial history”,Brookings Papers on Economic Activity, 1980(2), 277-340.

Wojnilower, A. M. (1985), “Private credit demand, supply and crunches: how different are the1980s?”, American Economic Review, 75(2), May, 351-356.

Wojnilower, A. M. (1997), Business cycles in a financially deregulated America, conferencepaper, “Monetary Theory as a Basis for Monetary Policy”, September 4-7, Trento, Italy.

Wolfson, M. H. (1994), Financial Crises: Understanding the Postwar U.S. Experience, secondedition, Sharpe.

Wolfson, M. H. (1996), “A post Keynesian theory of credit rationing”, Journal of Post KeynesianEconomics, 18(3), 443-470.

Woo, D. (1999), “In search of ‘capital crunch’: supply factors behind the credit slowdown inJapan”, IMF Working Paper, No 99/03.


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.

89ECB


L IST OFREFERENCES

15 “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.

the bank lending survey for the euro area

Documents