Institutions and credit risk in banking system the case of emerging economies

PoliciesandSustainableEconomicDevelop ment |1InstitutionsandCreditRiskinBankingSys tem:TheCaseofEmergingEconomies SUDINHTHANHUniversityofEconomicsHCMC-dinhthanh@ueh.edu.vn NGUYENPHUCCANH

PoliciesandSustainableEconomicDevelop ment |1

InstitutionsandCreditRiskinBankingSys tem:TheCaseofEmergingEconomies

SUDINHTHANHUniversityofEconomicsHCMC-dinhthanh@ueh.edu.vn

NGUYENPHUCCANHUniversityofEconomicsHCMC- canhnguyen@ueh.edu.vn

Abstract

Theextantliteraturehasdocumentedthedeterminantsofbankcreditrisk,butdoesnot paymuchattentiontoinstitutions.Thisstudyfillsthisgapbyinvestigatingtheeffectsofins titutionsonbankcreditrisk.Ourdatacovers33emergingeconomiesovertheperiodof200 2-

2013.ApplyingsystemGMMestimatorforunbalancedpaneldata,thestudyfindsintrigui ngfindings.Givencreditsupply,theeffectsofbankliquidityandbankprofitabilityareneg ative,whiletheeffectsofbankcapitalandbanksizeandbankconcentrationaresignifican tlypositive.Ontheotherhand,intermsof creditdemandtheeffectofrealG D P grow th r ateisnegative.Ourmainfindingisthatinstitutionshavenegativeeffectsonbankcreditri sk.Particularly,inthepresentofinstitutions,theinteractioneffectsofbankliquidity,bank capitalandbanksize,bankprofitability,andrealG D P growthratearestrengthened.The seresultsimplythatimprovementsininstitutionsaresignificantforcontroloverbankcre ditriskinemergingcountries.

Keywords:institution;bankingsystem;creditrisk; emergingeconomies

1 Introduction

Therehasbeenagrowingliteratureon determinantsofbankc re dit riskin the

re c e nt years,e s p e c i a l l y sincethe2008globalfinancial crisis.Severalstudiesblamethecrisisonexcessiverisktakingo f bankingsystem(Agnello& Sousa,2012;Hoque,Andriosopoulos,Andriosopoulos,& Douady,2015).Adeepunderstandingofcreditriskdeterminantsiscrucialimportancefortheproperevaluationofbankingsystemriskandhasadirectlinkwiththedevelopmentofsuitableregulationandprudentialtools

Infact,therearenumerousworksfocusingonthe

determinantsofcreditriskinbanking

systemi n c l u d i n g microeconomicfactorssuchasbankliquidity,bankcapital,banksize,bankcompetition,c r e d i t derivatives,internalratingsystems,collateral,relationshipbetweenlenderandcustomer(e.g.,I m b i e r o w i c z &Rauch,2014;Agarwalet al.,2016;DeLiset al.,2001;Mandalae t al.,2012);andm a c r o

e c o n o m i c factorssuchasinflation,unemployment,houseprice,creditcycles,businesscycle,andeconomicgrowth(e.g.,Hoqueetal.,2015;Castro,2013;

Chen,

2007;Jiménezetal.,2005;Tajike t al.,2015;Marcucci&Quagliariello,2009;Jiménezetal.,2005)

Thecreditactivitiesofbankingsystemarethefunctionoftheasymmetricinformationproblem(Lindsetetal.,2014;Miller,2015;Neyer,2004).Thisproblemmay

bereduced ifthe institutionofacountryisimproved.Thequalityofinstitutionandfinancialregulationaredefinitelydeterminantso f bankc r e d i t r i s k sincetheimprovementsi n institutionandfi n a n c i a l regulationr e d u c e theasymmetricinformationproblems,thusinducebankingsystemtoprovidebankloanswithbetterquality.Ontheotherha

nd,financialinsuranceregulationmaystimulatebanksintakingmorerisks.Indeed,theroleofpublicinstitutionsorpublicgovernanceinreduc ing theoverallriskinthee c o n o m y isexploredinseveralstudies(Cohenetal.,1983;DalBó&Rossi,2007;Duttaetal.,2013;Ho&Michaely,1988;Williamson,1981).However,notmuchisknownaboutpublicinstitutionsindeterminingc r e d i t r i s k i

n bankingsystemi n e m e r g i n g countries.I n addition,thea s s o c i a t i o n s betweenpublicinstitutionsandotherdeterminantsofcreditrisksuchasbankcapital,bankliquidity,banksize,bankcompetition,andeconomicgrowthareignored.Therefore,thisstudygoestofillthisgapbyinvestigatingtheeffectsofpublicinstitutionsandtheassociationsbetweenthemwithbankcharacteristicsincludingbankcapital,bankliquidity,banksize,andbankcompetition,whichreflectthedeterminantsofbankcreditsupply,onbankcreditrisk.Thispaperisnotthefirsttoexplorethedeterminantsofcreditriskinbankingsystem,butitisinnovativei n severalways

First,thisisthefirststudyusingthreedimensionsofpublicinstitutionsincludingg

overnmenteffectiveness,regulatoryquality,andruleoflawtoinvestigatetheeffectsofinstitutionalqualityoncreditriskinbankingsysteminemergingcountries

Second,weuseinteractiontermsbetweeneachinstitutionalindicatorandbankcr

editsupply

factorsinordertoexaminetheeffectofc r e d i t supply’sdeterminantsonbankcreditriskwheninstitutionsarepresent.Inaddition,weuseinteractiontermsbetweeneachinstitutionalindicatorandrealeconomicgrowthinordertoexaminetheeff

present.Thesemethodologiesensure

thatourmodelcatchesallthema in driversofc re dit activitiesre la ting toc redit riskin bankingsystem.

Withthesestrategies,webelievethatourstudyhassignificantcontributiontobothscholarandpractice.Infact,ourempiricalresultsshowthatinstitutionalqualityhassignificantnegativeeffecto n creditriskinbankingsystem,indicatingthatimprovementsininstitutions(suchas

governmenteffectiveness,ruleoflaw,andregulatoryquality)havegoodeffectsonbankingsystemstabilityandr e d u c i n g systematicrisk.Thisresulthasstrongcontributiontotheliteratureofinstitution,wheretheinstitutionalqualityimpactsnotonlyoneconomicgrowth(e.g.,Acemoglu&Robinson,2008;D o l l a r

&Kraay,2003;Fatás&Mihov,2005),butalsoonthestabilizationoffinancialmarket.Mostnotably,wefind thatimprovementsingovernmenteffectivenessandruleoflawinlinewiththehigherofliquidity,capital,size,competition,andlowerofprofitabilityinbankingsystemincreasebankcreditrisk.Theinteractiontermsbetweenregulatoryqualitywithbankcreditsupplycharacteristicshavesameeffectsasgovernmenteffectivenessandruleoflawexcludingtheprofitabilityandcompetitioninbankingsystem,whichhaveoppositeimpactsonbankcreditrisk.Moreprecisely,w e fi n d thattheinteractiontermsbetweenimprovementsi n institutionsw i t h e c o n o m i c growthincreasebankcreditrisk.Thisresultsshowtherisk-

takingactivitiesofbankingsystemwithhigherliquidity,capital,size,competition,andlowerprofitabilityinemergingeconomiesi n thecaseofbettergovernmenteffectivenessandruleoflaw.Betterregulationsalsosimulatebankstakingmorerisksexcluding

bankingsystemwithlowerprofitabilityandhighercompetition,whichmeansthatbetterregulationshelpbankingsystemmorestabilizationandmoresafety

Therestofthepaperproceedsasfollowingmanner.Section2reviewsrelatedbankcreditriskdeterminantsthroughthedriversofcreditdemandandcreditsupply.Section3introducesmethodologyanddata.S e c t i o n 4 presentsourempiricalevidencesf r o m threedimensionso f institutionsincludinggovernmenteffectiveness,ruleoflaw,andregulatoryquality.Finalsectionc o n c l u d e s

2 Literaturereview

Manypreviousempiricalstudiesanalyzedeterminantso f bankc r e d i t r i s k o rnonperformingloans,includingmacroeconomicfactorsandspecificbankingsectorcharacteristics(Castro,2013).Thebankcreditriskisgenerallydefinedastheriskthataloanisnotbeingpaidbyborrowerstobankpartiallyortotally.Theanalysisofbankcreditriskdeterminantsinbankingsystemisessentialforpolicymakerssinceitprovidesearlyalarmsformacroeconomicmanagementinfrontofshocksandpreventingfi n a n c i a l systemf r o m a possiblec r i s i s (Agnelloe t al.,2011;Agnello& Sousa,2012;Beltratti&Stulz,2012)

Intheliterature,bankcreditriskisdividedintosystematiccreditriskandunsystematiccreditrisk(e.g.,Ahmad&Ariff,2007;Aver,2008;Fryeetal.,2000;Dietsch&Petey,2002).Thedriversofsystematiccreditriskinclude:

(i)macroeconomicfactorssuchasunemployment,economicgrowth,

changesineconomicpoliciessuchasmonetarypolicy,fiscalpolicy,e c o n o m i clegislationchanges,ortradepolicy,

(iii)andpoliticalchanges.Thesefactorsmayinfluencetheprobabilityofborrower

spayingtheirloans.Forinstance,Aver(2008)findsthatemployment,long-terminterestrates,andthe

valueofstockmarkethaveimportantimpactsoncreditriskoftheS l o v e n i a n bankingloanportfolio.Kattai(2010)andFainsteinandNovikov(2011)showthatbankingsystemshighlighttheimportanceofeconomicgrowthinEstonia,Latvia,andLithuania.SalasandS a u r i n a (2002)andJakubík(2007)alsopointoutthatGDPgrowthandinterestratesarethemainm a c r o e c o n o m i c factorsaffectingbankcreditriskfortheSpanishandCzech,respectively

borrowerrelationship,andtheeffectso f characteristicso f theborrowersando ftheloano n bankc r e d i t r i s k Theyfi n d thatc o l l a t e r a l i z e d loanshaveahigherprobabilityofdefault,andthatwhileloansgrantedbysavingsbanksareriskier,a c l o s e bank-

borrowerrelationshipi n c r e a s e s thewillingnessf o r r i s k

-t a k i n g behaviorofbanks

Nevertheless,thereisamultitudeofempiricalstudieslookingatmaindriversofbankcreditriskandhighlightingthatm a c r o e c o n o m i c andmicroeconomicv a

r i a b l e s shouldbei n c l u d e d intotheanalysisduetotheirconsiderableinfluence.Thevastmajorityoftheseempiricalworksconsiderthem a c r o e c o n o m i c andmicroeconomicfactorsasthemostimportantdriversinthedeterminantsofbankc r e d i t r i s k basedo n a singlecountryanalysis.S o m e providea m u l t i -

c o u n t r y c o m p a r a t i v e analysiswithoutconcerningtotheeffectsofinstitutionanditsassociationswithmicroeconomicfactorsandmacroeconomicfactors.I nthisstudy,w e i n s p e c t thedeterminantso f c r e d i t riski n bankingsystemundertheaspectsofcreditsupplyandcreditdemandfactorstobetterunderstandingtheinfluenceofinstitutiononcreditfunctionandrisk-

takingbehaviorofoverallbankingsystem

Theinstitutionofacountrywhichisdefinedastherulesofthegameinasociety(North,1990),

includesthreefeatures:

(i)“humanlydevised”whichcontrastswithothereconomicfundamentals;

(ii)“therulesofthegame”toset“constraints”onhumanbehavior;and(iii)theirmajoreffectwillbethroughincentives(seeNorth,1981;Acemoglu&Robinson,20

08).Severalworkshavestudiestheeffectso f institution,whichi s n a m e d asthen

e w institutionale c o n o m i c s , buttheeffectso f institutiononcreditriskinbankingsystemarestillignored.Therefore,wetrytobuildourargumentstoexplainforthisissuebyfourdirections

cescreditriskinbankingsystem.Intheliteratureofeconomicgrowth,institutionisconsideredthedifferencesinresidualsofeconomicgrowth.Previousstudiesagreethatbetterinstitutionalqualityispositivelyandsignificantlycorrelatedwithecono

micgrowth(Young&Sheehan,2014).Forinstance,Djankovetal.(2002)fi n d s thatthelowercosto f o p e n i n g a medium-

sizebusinessintheUnitedStatesinc o m p a r i s o n

w i t h Nigeria,Kenya,Ecuador,andDominicanRepublicishighlycorrelatedwithe

c o n o m i c growth.Meanwhile,B e e k m a n e t al.( 2 0 1 4 ) fi n d thatcorruptionreducesincentiveso f individualsinbothvoluntarycontributionsandinvestmentsinLiberia,andthusimpactsone c o n o m i c growth.Therefore,betterinstitutionalqualitywillbeinlinewithhighereconomicgrowthandhenceimprovingthefinancialsituationofborrowersandreducingcreditriskinb

ankingsystem

Second,betterinstitutionalqualityreducesasymmetricinformationproblems(

Cohenetal.,1983;Ho&Michaely,1988; Willia ms on, 1981) ; hence,banksw il

l belessprobabilityofm a k in g wrong decisionsinlendingtobadborrowers

Third,betterinstitutionalqualityreducesoverallriskintheeconomy(Cohene

tal.,1983;Ho&Michaely,1988;Williamson,1981),therebyinducingalowersystematicriskandreducingcreditrisko f bankcreditportfolio.In

fact,betterinstitutionalqualityhelpseconomicagentsmoretrustableinbusinesstransactions.Thus,institutionalquality

positivelyimpactstheefficiencyofbusinesses(DalB ó &Rossi,2007),andmitigatessystematicriskinoveralleconomy.Duttaetal

(2013)find thatworsecorruptionsituationleadstohighinequality,poverty,andemploymentandthus underminestheeffectivenessofeconomicgrowthinIndia

Four,betterinstitutionalqualityreducestransactioncostineconomicactiviti

esingeneralandc r e d i t activitiesinparticularly.Asaresult,betterinstitutionalqualityinduceshigherefficiency ofc r e d i t activitiesandbettercontrolofcreditriskinbankingsystem

Althoughthe viewsofsystematicand unsystematiccreditriskdominateintheliteratureofbankc r e d i t r i s k , w e c a n examinethedeterminantso f c r e d i t r i

s k i n bankingsystemu n d e r thebasicdeterminantsondemandandsupplysidesofcreditfunction.Theapproachofthesystematicandunsystematiccreditriskissuitableforbanklevelstudiesduetodefiningfactorsrelatingtosystematico r unsystematicrisk,buttheapproachofdeterminingthedriversofcreditriskfromtheviewsofsupplyanddemandsideincreditfunctionisbetteratthebankingsystemlevelorcountrylevel,sincei t iseasytodefinethedriversofcreditriskintheoverallbankingsystem

Intheliterature,thefunctionofcreditisimpactedbysupplyanddemandsidefactors.Onthedemandside,i n c o m e levelandgrowthr a t e o f GDPperc a p i t a ar

em a i n determinantso f c r e d i t demand(Backé&Wójcik,2008).Forinstance,Duprey(2012)usesrealGDPgrowthinexplainingthedifferentpatternofbankbehaviorsovermacroeconomicfluctuationsat459publicbanksin93countries.Similarly,ElekdagandHan(2015)findthatdomesticfactorssuchaseconomicgrowthandmonetarypolicyshocksaremoredominantthanexternalfactorsindrivingrapidcreditgrowthinemergingAsia

PoliciesandSustainableEconomicDevelop ment |7

Indeed,economicgrowthincreasestheexpectedincomeandprofitinlinewithbetterfinancialconditionsofprivatesector,thusallowingforhigherlevelsofindebtedness(Kiss,etal.,2006),whilehouseholdsmaywanttoincrease debtlevelstosmoothconsumption in currenttimesincetheyexpecthigherincomeinthefuture(Backé&Wójcik,2008).Inaddition,highereconomicgrowthi n c r e a

s e s thedisposableincomeofpeopleandthusstimulatestheminconsumptionandinvestment.F i r m s arealsoinduc e d toexpandtheiroperations toca tc hup

with increasingdemand,therebystimulatingtheinc re as e d credit demand.Even thoughtheprofi t offirmsmaybehigherduetoe c o n o m i c growth,thusstimulatingthemtorelymoreoninternalfundsinsteadofbankloans(Kisse t al.,2006),thiseffectmaybenotstrongasthepositiveeffectsofeconomicgrowthoncreditde

mandi n developingoremergingeconomiessincehigheconomicgrowthneedsa

bundleofcapital

Accordingly,fluctuationsineconomicgrowthdefinitelychangecreditdemandandthen,inturn,i m p a c t oncreditriskinbankingsystem,whichexplainstherelationbetweencreditriskandbusinesscycle(Marcucci& Quagliariello,2009).For example,us ing laggedpercentagechangeinG DP toexplainforthebankingcrisisintheNordiccountriesintheperiodof1980s-

1990s,Pesola(2001)fi n d s thattheshortfallsofGDPgrowthbelowforecastcontributetotheirbankingcrises.SalasandS a u r i n a (2002)finds thattheGDPgrowthrateinlinewithothermicroeconomic factorssuchasfi r m s , andfamilyindebtedness,rapidpastcreditorbranchexpansion,inefficiency,portfoliocomposition,size,netinterestmargin,capitalratio,andmarketpoweraredriversofcreditriskinS p a n i s h c o m m e r c i a l andsavingsbanksi n thep e r i o d 1985-

1997.MeyerandYeager(2001)andGambera(2000)find thatmacroeconomi

cvariablesaregoodpredictorsforthenon-performing loansintheUS.MarcucciandQuagliariello(2008)usedatafromItalianbanksandfind thatthec r e d i t riskisincreasedineconomicdownturns.Similarly,Hoggarthetal.(2005)providethesameevidenceforthecaseoftheUK

Therearesomepossibleexplanationstothecyclicalityofcreditrisksuchasdisastermyopia,over-

optimism,herdbehaviors,andinsufficientm a r k e t disciplines(Marcucci& Quagliariello,2008).MarcucciandQuagliariello(2008)verifythecyclicalityofcreditriskfromthefollowingperspectives:

(i)e c o n o m i c growthincreasesthep r o fi t o f fi r m , w h i c h r a i s e s a s s e t p r

i c e s r i s e a n d customers’expectationsatthebeginningofexpansionaryphaseandthenincreasesaggregatedemand.Asaresult,theincreasingofaggregat

edemandinducesarapidgrowthinbankcreditportfolioandine c o n o m y ' s indebtedness,wherebanksusuallyunderestimatetheirr is k exposureandre la x theirc r e d i t standardsduetoo v e r -

quacydeteriorateconsequentlyascyclicality

optimism,herdbehaviors,andinsufficientmarketdisciplines,whichdependmuchoninstitutionalenvironmentofacountry.Sincebetterinstitutionalqualityreducesasymmetricinformationproblems,transactioncost,andrisk,whileimprovingtheefficiencyofassetallocations,andpropertyrightprotection(Cohenetal.,1983;Ho&Michaely,1988;Williamson,1981),betterinstitutioni n association

w i t h highere c o n o m i c growthm a y stretchcyclicalityo f c r e d i t r i s k i n bankingsystemundertheimpactofrisk-takingbehaviorofbanks

Theliteratureofbankingactivitieshaveconcludedthattherearemanydeterminantsofbankrisk-

takingbehaviorsuchascapitalrequirement,banksize,bankleverageandfinancialliberalization(Ashrafetal.,2016;Bhagatetal.,2015;Blaško&SinkeyJr.,2006;Borio&Zhu,2012;Cubillas&G o n z á l e z , 2014;Efingetal.,2015;Gallowayetal.,1997;García-Marco&Robles-Fernández,2008).Ashrafetal

(2016)find

strongevidencethatbankrisk-takingissignificantlyhigherincountries,whichhavehighindividualism,lowuncertaintyavoidance,andlowpowerdistanceculturalvalues.B u c h andDeLong(2008)examinecross-

borderbankmergersandfindthatthesupervisorys t r u c t u r e s ofthepartners’countriesinfluence changesinpost-

mergertotalriskandstronghostregulatorslimitingtheriskyactivitiesoftheirlocalbanks

Asasummary,thehighereconomicgrowthmakesbothpeopleandbanksmoreconfidenceandm o r e optimistic,whilebetterinstitutionalqualitybooststhisconfidenceandoptimistichigherduetobetterqualityofgovernment,regulation,lawsystem,propertyrightprotection,andalsoreducescorruptionproblem,thereforebanksmaytakeriskieractivitiesintheircreditportfolio.Therefore,w e arguethatbetterinstitutionalqualityinlinewithhighereconomicgrowthwillsimulaterisk-

takingactivitiesofbanks,henceincreasingthecreditriskofoverallbankingsystem

Onthesupplyside,banksareassumedtomakedecisionsontheircreditportfoliosdependingonallavailableinformationandtheirinternalfinancialconditionssuchascapital,liquidity,profitability,assets,thereforetheseinternalcharacteristicsofbankshavestrongimpactsoncreditrisk.Theextentliteraturehasexaminedthedriversofcreditsupplyofbanks(Auel&deMendonça,2011;Aysun,2016;B h a u m

i k e t al.,2011;Ciccarellie t al.,2015;Liu& Tallada,2015;Yagihashi,2011).MarcucciandQuagliariello(2008),forinstance,documentthatbanksmayreactstrongertoexternalshockssuchasrecessionineconomybycuttingtheircreditsupplyhigheriftheyf a c e toconstraintssuchascapitalintheirbalancesheet.While,BernankeandGertler(1995)arguethatbankscanadjusttheircreditsupplyinrespondingwithexternalshockssuchasacontractionaryo r expansionarymonetarypolicy,whereacontractionarymonetarypolicyleadstoadecreasinginavailablefundsuchasdepositsorotherfunds,andthusbanks

Minford,2014;Ramos-havetoreducetheircreditportfolio.Many empirical studiesshowthat banksmaynotresponsetoexternalshocksinthe

takingactivitiesbyincreasingtheircreditsupplydespitecontractionarymoneta

lineofwhattheauthorizerswantinsomecases.Forinstance,banksmaydorisk-rypolicy(Angelonietal.,2015;Buchetal.,2014;deMoraesetal.,2016;DellʼAricciaetal.,2014;García-Kuhnertetal.,2015;Montes&Peixoto,2014).

PoliciesandSustainableEconomicDevelo

Particularly,studiesoncreditchannelarguethattheactionsofbanksintheircreditportfoliodependo n theirinternalconditionsincludingcapital,liquidity,size,profitability,andcompetitiononbankingsector

(Aiyaretal.,2016;Aysun,2016;Gunji&Yuan,2010;Imbierowicz&Rauch,2014;Khanetal.,2016;Leon,2015;Yang&Shao,2016)

Thereisusuallyanadequate requirementintheminimumcapitalratiothat

everybankneedstomeet.Whenever,thereisanexternalshocksuchasadownturnineconomicgrowthoracontractionarymonetarypolicy,banksw i t h lowerc a p i t a l buffersabovetheminimumc a p i t a l requirementwillreducetheircreditsupplymuchmore(Bliss&Kaufman,2002).Iffirms cannotsubstitutebankloansbyothersources,theymayfacetoinsufficientfundingfortheirinvestmentprojects,andinturn,increasingthecreditriskofbanks.Hence,bankswithhighercapitalwillbelessc r e d

i t risk.But,bankswithlowercapitalratiohavetoraiseadditionalcapitalbeforeexpan

dingcreditto bindrisk-based capitalrequirement, thustheymayoptimallyforgoprofitable loanstoreducether i s k offuturecapital

inadequacy(Vanden Heuvel,2002).Asaresult,theymaychoosetorejectbadprojectsforlendingtoreducecreditrisk.KonishiandYasuda(2004)

findthattheimplementationo f thecapitaladequacyrequirementreducesrisktakingatJapanesecommercialbanks,whilebanksw i t h highercapitalratiomaytakeriskieractivitiesforhigherprofits,leadingt

oincreasingcreditrisk.Theliquidityisalsoveryimportantforbankingsystemduetotheriskofbankruptc

y(Diamond

&Rajan,1999; Gatevetal.,2009;Imbierowicz &Rauch,2014).Infact,bankshavetoconsidertheirliquiditysituationwhentheymakecreditdecisions.Bankswithhigherliquiditymaybemoreflexiblei n supplyingtheirc re di t p o r

t f o l i o , thustheyarebetterm a n a g e m e n t theirc r e d i t r is k However,higherliquiditymaysimulatebankstakingriskintheircreditactivitiessuchasthestudiesofNguyenandBoateng(2013) andNguyenetal.(2015)

Thesizeofbankisusuallymentionedasthemaindriverofcreditrisk(Demsetz&Strahan,1997;K i s h a n &Opiela,2000).Atbanklevel,largerbankshavehighermarketpowerandthustheyhavehigheradvantagesinmanagingtheircreditportfolio,whilesmallerbanksmaytakeriskduetolowerm a r k e t share.However,largerbanksmaybeimpactedbythetheory“toobigtofail”(O'hara&Shaw,1990;Papachristos,2011);thus,theytakemorerisk-

takingactivitiesthansmalleronesandhavehighercreditrisk(Bhagatetal.,2015;Chenetal.,2015;Hakenes&Schnabel,2011;Stever,2007).Atcountrylevel,largerbankingsystemmeasuringundertherelativesizeofassetsinbankingsystemw

i t h GDPdefinesthehigherdominationofbanksinfinancialsystemandthemoreimportanceoftheirroleintheeconomy,whichcreatesmoreincentivesforbanksintakingriskduetothehigherprobabilityofgovernmentininterveningwhenevertheyfacetofinancialdistresstoprotectpeoplef r o m p a n i c astheoryof“toobigto fail”

Atthelastcharacteristicmentionedinourarguments,theprofitabilityofbanksmeasuredbyr e t u r n onassetorreturnonequitymayhavegoodnessonbankcreditrisk.Thehigherprofitabilityo f banksmakesthemmoreflexibleinmanagingtheircreditportfolio,andalsoreducepressuresin

takingactivitiesw i l l belimited.

expandingcredittooquicksothattheprobabilityofwrongcreditdecisionsorrisk-Thecompetitioninbankingsectorhasstrongimpactsonbankchoicesandthusithasimpactsonbankc re dit ris kthroughtheeffectson bankris k-

ta king behaviors.Alargeamountofliteraturearguesthathighercompetitionmakesbankstakemorerisktoaimatgettinghighermarketshare(Agorakietal.,2011;Boyd&DeNicolo,2005;Jiménezetal.,2013).Houetal

(2014),forexample,fi n d thatintensem a r k e t competitionc o m p e l s Chines

ec o m m e r c i a l bankst o developadvancedtechnicalexperienceandskills,thusimprovingtheirtechnicalefficiencyandthenthetechnicale ffi c i e n c y ispositiveassociatedwiththerisktaking

However,countrieswithafewoflargebanksdominatedreducecompetitivepressuresbetweenbanks.Whenlargebankshavemoreconfidence intakingriskandthusauthorizershavetomakesuretheyaresafe(Boyd&DeNicolo,2005;Hakenes&Schnabel,2010;Jiménezetal.,2013;Wagner,2010).MatutesandVives(2000)studythelinksbetweencompetitionfordepositsandrisktakingincentives,andconcludethatthewelfareperformanceofthemarketandtheappropriatenessofalternativeregulatorymeasuresdependonthedegreeofrivalryandthedepositinsuranceregime

Asasummary,bank’sinternalconditionssucha s liquidity,capital,size,profitability,andcompetitionwithinbankingsectorhavestrongimpactsoncreditriskthroughcreditactivities.Creditactivitiestrulydependonbank’smanagerexpectationsandfinancialbehaviors.Theexpectationsarebuiltfromtheavailableinformationthatisfacedtotheproblemofasymmetricinformation,whilethefinancialbehaviorsofbanksmanagersareimpactedbythemacroeconomicinstitutionsuchasregulationsandlawsystem.Therefore,institutiondefinitelyhasassociationswiththesedriversofbankcreditrisk.Chenetal.(2015)usebank-

leveldatafrommorethan1200banksin35emerginge c o n o m i e s duringtheperiod2000-2012andfindthattheimpactofmonetarypolicyonbanks'risk-

takingbehaviorismorepronouncedwiththeincreasingseverityofcorruption.daSilvaandDivino(2013)usedatafortheBrazilianeconomyintheperiodfrom1995t

pro-cyclicalanddefaultriskdependsonstructuralfeatures

Inthefirstviewofassociationsbetweeninstitution withbankcharacteristicsandbanking

sectorcompetition,betterinstitutioninlinewithhighercapital,higherliquidity,biggersize,lowerprofitability,andlowercompetitionmaystimulatebankingsystemintakingmoreriskduetolowerasymmetricinformationproblems,transactioncost,andoverallriskintheeconomy

However,thedifferenceaspectso f institutionincludinggovernmente ff e c t i

v e n e s s , regulationquality,andruleo f lawm a y havedifferenceimpacts.S i n c

e governmente ff e c t i v e n e s s reflectsperceptionsofthequalityofpublicservicesandcivilservice,andthedegreeofitsindependencefrompoliticalpressuresinpolicyformulationandimplementation,andthecredibilityofthegovernment'scommitmenttosuchpolicies,sobettergovernmenteffectivenessmaystimulaterisk-

takingbehaviori n bankingsystemduetobetterpublicservicesandthuslowertransactioncostandbetterconditionsf o r thegrowthinprivatesector.Regulatoryqualityreflectsperceptionsoftheabilityofthe

development,thusbetterregulatoryqualitymayinducesomerisk-takingactivitiesofbetterbanksduetothelessharmfulofpoliciesfrompoliticalpressuressothatthe“toobigtofail”m a y belessimpactsonbank’smanagers.Meanwhile,ruleoflawreflectsperceptionsoftheextenttowhichagentshaveconfidenceinandabidebytherulesofsociety,andinparticularthequalityofc o n t r a c tenforcement,propertyrights,andthepolice,thereforebankshavemorecommitmentincontractagreementandthepropertyrightsprotectionsthustheymaytakeriskmore

3 Methodologyanddata

3.1 Methodology

Thissectionpresentsourmethodologyinestimatingtheeffectsofinstitutionanditsassociationsw i t h bankcharacteristicsandbankingsectorcompetitionthroughthreestepprocedures.First,weu s e thepercentagechangei n governmenteffectivenessi n d i c a t o r s f r o m theW o r l d G o v e r n a n c e i n d i c a t o r s toproxyforthechangeininstitutionalqualityasfollowingformula:

(1)Sinceinstitutionalqualityindic a tors arescaledfrom-

2.5to+2.5f o r ea c h indicator,thusw e standardizethesei n d i c a t o r s bycalculatingthepercentagechangeo f e a c h i n d i c a t o r f o r bettermeasuringchangesininstitutionalquality.Thisproxyofchangesininstitutionalqualityisusedtoevaluatetheimpactsofinstitutiononcreditriskinbankingsystemin33emergingcountries3fr om2002to2013throughthefollowingmodification

tionsof variablesare presentedinTable1

3 List

ofemergingeconomies:Brazil,Bulgaria,Chile,Colombia,China,CzechRepublic,EgyptArabRep.,Estonia,Greece,Hu ngary,India,Indonesia,Latvia,Lithuania,Malaysia,Mexico,Morocco,Nigeria,Oman,Pakistan,P e r u , Philippines,P

oland,Romania,RussianFederation,Slovenia,SouthAfrica,KoreaRep.,Thailand,Turkey,Ukraine,VenezuelaRB,and Vietnam.

Variable Definition Source

Calculatio Reguqua PercentagechangeinRegulatoryqualityindicator(%)

(2012),andAltunbase t al

(2010).Alltheexplanatoryarelaggedtocapturedatacharacteristic,wheretheyear-

enddatatobankcharacteristicsandbankcompetitionwillhaveimpactsonbank’smanagerdecisionsinfollowingyear.TherealGDPgrowthr a t e ofthisyearwillimpactontheexpectationofpeopleinthefuture,andthusimpactsonthenextyearcreditdemand

Afterexaminingtheimpactsofchangesingovernmenteffectivenessoncreditriskinbankingsystem,weuseinteractiontermsbetweeninstitutionsandthefactorsofcreditdemandsideandc r e d i t supplysidetoevaluateinteractioneffectsoncreditriskinbankingsystem.Inwhich,weadde a c h interactiontermonebyoneintoourestimationstotesttheconsistenceofourresults,whichisprovedfortherobustnessofourfindings.Thisisthesecondstepinourstudy

Then,inthethirdstep,weusechangesinruleoflaw(Rulelaw)andregulatoryqual ity(Reguqua),respectivelytoreplaceforthegovernmenteffectivenessandreplic

atealltestproceduresasmentionedabove.Withthisstrategy,webelieveourstudyisrobustnessintestingtheimpactsofinstitutionsandinteractionstermsbetweeninstitutionsandthedriversofdemandandsupplyonc r e d i t risk.WeusethesystemGMMestimatorsfollowingthestudyofArellanoandBond(1991),ArellanoandBover(1995),extendedbyBlundellandBond(1998)

andBlundellandBond(1998)foradvantageofaddressingthebiasassociatedwiththefixedeffectsinshortpanelsandsolvingtheproblemofendogeneityindynamicpaneldata,whichissuitableforthisstudyduetotheendogeneityemergingfromtherelationshipbetweencreditriskwithotherbankcharacteristicssuchasbankliquidity,bankprofitability,orbankcapital

3.2 Data

Dataforthisstudyi s collectedf r o m WorldB a n k w i t h threedatabasesi n c l u

d i n g theGlobalFinancialDevelopmentDatabase,WorldDevelopmentIndicators,andWorldGovernanceIndicators.Thedataisyearlyfor33emergingeconomiesovertheperiodof2002-

2013withsomemissingv a l u e duetotheavailableofdatainWorldBank’sdatabase,thusunbalancepaneldataisusedforthisstudy.Fortunately,theproblemofbiasestimatorsinunbalancedpaneldataissolvedbythesystemGMMestimators.AlldatadescriptionandcorrelationmatrixarepresentedinTables2and

0.287*

**

0.000

0.214

0.6855**

* -

0.5804**

* -

0.6138**

* -

0.6248**

* -

0.6101*

** -

-(Goveff*Bankcon)

(-1)

0.0295* -0.0246**