Do credit ratings affect capital structure?

Do#Credit#Ratings#Affect#

Capital#Structure?#

!

!

!

!

!

!

!

Yasmin'Buijs''

10647066'

Economics'and'Business'

Finance'&'Organization'

S.R.'Arping''

June'28,'2016'

'

!

!

!

Abstract.# This'thesis'investigates'the'effect'of'credit'rating'changes'on'capital'structure' decisions'for'firms'near'an'upgrade'or'a'downgrade.'The'financial'literature'recognizes'that' there'is'an'effect,'but'there'is'not'a'discrete'answer'on'this'question.'Kisgen'(2006)'suggests' that'firms'near'an'upgrade'or'a'downgrade'issue'relatively'less'net'debt'relative'to'equity' than'firms'not'near'a'credit'rating'change.'This'thesis'does'not'fully'support'these'findings.' This'thesis'suggests'that'most'firms'near'a'downgrade'and'firms'at'the'boundary'of'losing'or' achieving'the'investment'grade'rating'issue'less'net'debt.' ' '

Statement#of#Originality

# # This'document'is'written'by'Student'Yasmin'Buijs,'who'declares'to'take'full' responsibility'for'the'contents'of'this'document.' ' I'declare'that'the'text'and'the'work'presented'in'this'document'is'original'and' that'no'sources'other'than'those'mentioned'in'the'text'and'its'references'have' been'used'in'creating'it.' ' The'Faculty'of'Economics'and'Business'is'responsible'solely'for'the'supervision' of'completion'of'the'work,'not'for'the'contents.' ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

!

Introduction#

'

Nowadays,'every'company'with'public'debt'has'a'credit'rating,'and'these' credit'ratings'have'real'importance'for'many'observers'of'financial'markets'(Boot,' Milbourn'&'Schmeits,'2016).'Most'of'them'have'credit'ratings'of'the'three'most' prominent'credit'rating'agencies:'Moody’s,'Standard'&'Poor'and'Fitch.''For'many' investors'credit'ratings'are'important,'therefore'managers'take'those'credit'ratings' into'account.'For'example,'Bloomberg'(2016)'reported'that'PPC,'South'Africa’s' biggest'cement'producer,'is'planning'a'capital'raise'of'254'million'dollars'to'prevent' that'they'will'be'downgraded.'Furthermore,'a'2005'issue'of'the'Economist)concluded' that'credit'ratings'are'among'the'most'powerful'factors'in'today’s'capital'markets,' and'that'this'influence'is'underestimated.'Kisgen'(2006)'finds'that'credit'rating' changes'directly'affect'capital'structure'decisions,'but'Kemper'(2013)'does'not' support'this'analysis'and'says'that,'except'for'B\'rated'firms,'no'other'rating'groups' limit'their'debt'financing'when'facing'a'credit'rating'change.'In'addition,'Brealey,' Myers'and'Mohanty'(2012)'argue'that'corporate'managers'pay'too'much'attention'to' credit'ratings,'and'that'the'influence'is'exaggerated.'Factors'as'‘tax'advantages'of' interest'deductibility’'and'‘business'risk’'are'seen'as'the'more'important'ones'that' affect'the'capital'structure.''So,'there'is'not'a'concrete'answer'on'the'question'if' credit'rating'changes'directly'affect'the'capital'structure'of'a'firm.'' '' I'expect'that'credit'ratings'affect'the'capital'structure'of'a'firm'because' different'credit'ratings'are'associated'with'some'costs'and'benefits.'For'example,' credit'ratings'act'as'a'signal'of'firm'quality'(Kisgen,'2006),'credit'ratings'can'provide' information'about'the'quality'of'a'firm'to'investors'that'is'not'public.'Therefore,'a' rating'change'can'change'the'cost'of'capital'of'a'firm.'Furthermore,'some'regulations' are'based'directly'on'credit'ratings,'which'means'that'these'rules'influence'investors' groups'that'are'only'allowed'to'invest'in'particular'bonds'or'stocks.'An'upgrade'can' result'in'access'to'investors'who'are'otherwise'restricted'from'investing'in'a' particular'bond.'There'are'also'some'other'events'that'happen'when'a'firm'is'up\'or' downgraded,'such'as'a'change'in'coupon'rate,'change'in'capital'reserve' requirements,'change'in'third\party'relationships,'or'a'loss'of'access'to'a'commercial' paper'market.''

' On'the'basis'of'an'empirical'test,'I'want'to'analyse'if'credit'rating'changes' affect'the'capital'structure'decisions'of'a'firm.'I'want'to'investigate'what'the'role'of' credit'ratings'is'on'the'capital'structure'of'firms,'and'if'CFOs'are'influenced'in'their' decisions'by'credit'ratings'changes.'In'this'thesis'I'also'want'to'examine'if'due'to'bad' credit'ratings,'firms'issue'more'debt'than'equity'and'is'it'for'good'credit'ratings'firms' the'other'way'around.'Furthermore,'do'expected'changes'in'credit'ratings'result'in'an' other'capital'structure?'Do'firms'which'expect'a'change,'act'differently?'To'measure' this'effect,'I'will'use'a'dummy'variable'that'makes'a'distinction'between'firms'near'a' credit'rating'downgrade'or'upgrade'and'firms'not'near'a'credit'rating'downgrade'or' upgrade.'The'Credit'Ratings'that'I'will'use'are'from'Standard'&'Poor’s,'and'their' long\term'credit'rating'includes'a'plus'or'a'minus.'Firms'near'an'upgrade'has'a'plus' in'their'credit'rating'(e.g.'AA+)'and'firms'near'a'downgrade'has'a'minus'in'their'credit' rating'(e.g.'AA\).'I'test'the'effect'of'firms'near'a'credit'rating'change'on'the'issuance' of'net'debt'versus'net'equity'from'January'1973'to'April'2016.''I'find'that'firms'are' influenced'by'credit'rating'changes'in'their'capital'structure'decisions.'Firms'at'the' boundary'of'being'downgraded'from'an'investment'grade'rating'to'a'speculative' grade'rating'are'more'likely'to'reduce'debt'and'less'likely'to'issue'debt'than'other' firms.'Likewise,'firms'at'the'threshold'of'being'upgraded'to'investment'grade'rating' also'issue'less'net'debt'relative'to'equity.'On'the'other'hand,'where'Kisgen'(2006)' find'a'negative'effect'of'a'potential'downgrade'or'upgrade'on'net'issuance'of'debt' versus'equity,'I'only'find'this'result'for'most'firms'near'a'downgrade.'' The'remainder'of'the'thesis'is'organized'as'follows.'In'the'next'chapter,'the' literature'review,'I'relate'my'thesis'to'the'existing'literature'and'my'contribution'to'it.' The'methodology'chapter'include'my'methodology'to'test'the'effect'of'credit'rating' changes'on'capital'structure'decisions.'My'results'of'this'test'are'presented'in'the' results'chapter.'Finally,'in'the'last'chapter,'I'summarize'and'discuss'my'findings'and' draw'a'conclusion.' ' ' ' ' ' '

Literature#Review#

! ' That'credit'ratings'influence'the'capital'structure'policy'is'recognized'in'most' financial'literature'(Damodaran,'1997c'Brigham,'Gapenski'and'Ehrhardt,'1998c' Moyer,'McGuigan,'Rao'and'Kretlow,'2011),'but'a'concrete'answer'on'this'subject'is' missing.'Graham'and'Harvey'(2002)'find'that'credit'ratings'are'the'second'highest' concern'for'CFOs'when'determining'their'capital'structure.'57.1%'of'the'CFOs'said' that'credit'ratings'were'important'or'very'important'in'their'capital'structure'decision.' Kisgen'(2006)'conclude,'on'the'basis'of'an'empirical'test,'that'managers'are'directly' affected'by'credit'ratings'in'their'capital'structure'decisions'and'concluded'that'the' effect'of'credit'ratings'on'capital'structure'can'be'seen'as'complementary'to'existing' trade\off'theory'of'capital'structure.'He'used'his'CR\CS'model'(Credit'Rating'–' Capital'Structure)'and'finds'that'firms'near'a'credit'change'issue'1%'less'net'debt' relative'to'net'equity'annually'as'a'percentage'of'total'assets'than'firms'not'near'a' credit'rating'change.'This'means'that'firms'with'a'plus'or'minus'in'their'credit'rating,' issue'relatively'less'net'debt'than'net'equity.'The'motivation'for'this'finding'is'that' firms'near'an'upgrade'will'not'take'the'risk'to'miss'this'opportunity'by'issuing'debt' and'for'firms'near'a'downgrade'do'not'want'to'take'the'risk'that'the'credit'ratings' agencies'lower'their'credit'rating'as'a'result'of'new'debt'issues.'Kemper’s'(2013)' analysis'is'not'very'supportive'on'Kisgen’s'analysis.'He'argues'that'the'effect'of' credit'rating'changes'on'capital'structure'only'holds'for'particular'ratings'and'does' not'hold'for'all'credit'ratings.'Especially,'he'only'finds'that'the'effect'holds'for'B\' ratings'firms.'He'argued'that'in'that'case'the'financing'behaviour'of'avoiding'debt' may'be'more'an'indication'of'lack'of'access'to'debt'than'an'indication'of'an'attempt' to'decrease'debt'financing'for'credit'rating'reasons.'Therefore,'it'is'time'to' reinvestigate'the'model,'but'now'with'some'additional'information,'data'and' variables.' '' Although'Kisgen'(2006)'provide'evidence'that'there'is'a'relation'between' credit'ratings'and'capital'structure,'he'does'not'make'a'distinction'of'the'effect'by' rating.'The'effect'can'be'different'across'some'firms.'I'believe'that'for'every'firm' credit'rating'are'important'and'that'they'take'these'into'account,'but'I'also'believe' that'for'some'firms'these'ratings'are'more'important.'' '

Every'firm'has'the'desire'to'achieve'or'maintain'a'certain'credit'rating,'but'for'some' firms'this'desire'is'more'important'to'achieve.'Kisgen'(2006)'argues'that'every'firm' cares'about'their'credit'rating'regardless'of'their'credit'rating.'So,'this'assumes'a'firm' with'A\'rating'cares'as'much'about'their'rating'as'a'firm'with'a'BBB\'rating,'and'is'as' likely'to'change'his'financing'behaviour.'I'think'that'there'may'be'a'difference' between'the'different'credit'ratings.'Especially'for'firms'that'are'at'the'boundary'of' being'downgraded'from'investment'grade'rating'to'junk'bond'rating.'Likewise,'a'firm' just'below'the'investment'grade'rating'is'more'likely'to'change'his'financing' behaviour'to'perceive'that'upgrade.'A'change'in'credit'ratings'for'these'firms'has'a' greater'impact'on'their'discrete'costs'than'for'other'credit'rating'changes.'Cantor'and' Packer'(1997)'states'that'regulations'for'investments'in'bonds'are'related'to'credit' ratings.'Institutional'investors'are'restricted'from'owning'speculative'grade'bonds' (i.e.'junk'bonds).'Thus,'a'firm'that'is'downgraded'to'a'speculative'grade'rating'will'no' longer'attract'institutional'investors.'This'can'be'a'great'obstacle'and'can'be'costly.' Banks'are'also'restricted'from'holding'speculative'grade'bonds,'because'they'have' to'be'a'stable'factor'and'should'not'disturb'the'efficiency'of'the'market.'Insurance' companies'have'the'same'regulations'and'are'also'restricted'from'investing'in' speculative'grade'bonds'(Kisgen,'2006).'In'addition,'credit'ratings'are'also'related'to' the'capital'requirements'for'investing'in'a'firm’s'bonds,'this'is'dictated'by'the' Securities'and'Exchange'Commission'(Cantor'and'Packer,'1997).' '' Therefore,'my'expectations'are'that'there'is'relation'between'credit'ratings' and'capital'structure'decisions'for'every'firm,'but'this'effect'is'especially'applicable' for'firms'at'the'boundary'of'losing'or'obtaining'their'investment'grade'rating.'' ' The)trade0off)theory) '' The'traditional'trade\off'theory'states'that'a'value\maximizing'firms'determine' their'optimal'capital'structure'by'balancing'the'related'costs'and'benefits.'To' determine'how'much'debt'and'how'much'equity'is'optimal,'the'cost'of'debt'and' benefits'of'debt'are'offset.'The'advantages'of'financing'with'debt'are'the'tax'shield,' and'the'costs'of'financing'with'debt'are'the'cost'of'financial'distress,'which'includes' bankruptcy'costs'and'non\bankruptcy'costs.'The'firm’s'optimal'capital'structure' involves'the'trade\off'of'this'costs'and'benefits'(Bradley,'Jarrel'&'Kim,'1984).'' '

Kisgen’s'(2006)'CR\CS'model'states'that'there'are'also'costs'and'benefits'related'to' credit'rating'changes.'If'this'cost'or'benefit'is'material,'a'firm'balance'this'cost'or' benefit'against'the'cost'of'benefit'that'is'related'with'the'traditional'trade\off'theory.' This'will'result'in'different'financial'behaviour,'in'some'cases'the'cost'or'benefit'from' the'credit'rating'changes'will'outweigh'traditional'trade\off'theory'factors'and'in'some' cases'it'will'be'the'other'way'around.'' ' The)pecking)order)theory) '' The'pecking'order'theory'focuses'on'asymmetrical'information.'This'theory' focuses'on'prioritize'their'financing'strategy'based'on'the'path'of'least'resistance.' Internal'financing'is'the'first'preferred'method,'followed'by'debt'and'only'when' internal'funds'have'been'used'up'and'a'firms'has'reached'his'debt'capacity,'it'will' issue'equity'(Myers.'1984).'Myers'argues'that'firms'only'issue'new'equity'when'it'is' overvalued,'so'they'can'take'an'advantage'of'this.'This'is'assumed,'because' managers'know'more'about'the'firm'and'the'true'condition'of'the'firm'(information' asymmetry).'This'implies'that'debt'will'increase'if'the'investment'exceeds'the' internally'generated'funds,'and'debt'decrease'if'the'investment'is'lower'than'the' internally'generating'funds.' The'CR\CS'model'of'Kisgen'(2006)'states'that'for'a'change'in'debt,'a'discrete' cost'or'benefit'will'occur'due'to'credit'rating'change.'Again,'assuming'that'these'CR\ CS'effects'and'pecking'order'effects'are'material,'managers'will'face'a'trade\off' between'the'cost'of'issuing'equity'or'the'cost'that'are'associated'with'a'potential' credit'rating'change.'This'is'especially'important'for'firms'near'a'credit'rating' upgrade'or'downgrade.'Thus,'in'some'cases,'firms'will'do'the'contrary'of'what'is' stated'by'the'pecking'order'theory.'Firms'facing'an'upgrade'may'choose'to'issue' equity'instead'of'debt'to'enhance'the'chance'of'begin'upgraded'and'obtain'the' benefits'of'a'higher'credit'rating.'Likewise,'firms'near'a'downgrade'may'reduce'their' debt'in'order'to'avoid'a'downgrade'and'avoid'the'cost'associated'with'a'downgrade.'' ' ' ' ' '

Methodology#

!

To'test'my'hypothesis,'I'will'use'the'same'model'as'Kisgen'(2006)'used,'the' CR\CS'model,'but'with'some'adaptions.'His'empirical'test'was'based'on'a'dummy' for'a'plus'or'minus'in'a'credit'rating,'to'identify'if'a'firm'is'near'an'upgrade'or'a' downgrade.'To'test'if'there'was'a'relation'between'credit'ratings'and'capital' structure,'he'used'net'debt'versus'net'equity'issuance'as'dependent'variable.'He' tested'if'firms'at'the'boundary'of'being'downgraded'or'upgraded,'issuing'less'debt' relative'to'equity.'Firms'near'a'downgrade'would'issue'less'debt'relative'to'equity'to' avoid'a'downgrade.'Likewise,'firms'near'an'upgrade'would'not'issue'new'debt'to' increase'the'chance'of'being'upgraded.'The'model'used'three'control'variables:'one' for'leverage,'one'for'profitability'and'one'for'size.'The'control'variable'profitability'is' included,'because'firms'that'are'more'profitable'face'lower'financial'distress'costs.' Furthermore,'for'these'firms'interest'tax'shield'are'worth'more.'Thus,'the'perspective' of'less'financial'distress'costs'and'more'advantage'of'the'interest'tax'shield'predict' that'more'profitable'firms'issue'more'debt'(Frank'&Goyal,'2009).'Size'is'included'as' control'variable,'because'larger'firms'are'more'diversified'and'face'lower'default'risk.' Thus,'firms'that'have'more'sales'issue'more'new'debt'than'firms'that'have'less'sales' (Frank'&'Goyal,'2009).'The'control'variable'leverage'is'included,'because'the'debt' ratio'will'influence'capital'structure'decisions.'Assumed'is'that'firms'with'more'debt,' face'higher'debt'costs,'and'higher'financial'distress'costs.'Thus,'firms'with'more' leverage'will'issue'less'new'debt'(Frank'&'Goyal,'2009)' '' In'Kisgen’s'model'different'dummies'are'used'for'a'potential'upgrade'or' downgrade.'One'combined'dummy'that'equals'one'if'a'firm'has'a'plus'or'a'minus'in' their'credit'rating,'and'a'separate'dummy'which'makes'a'distinction'for'a'plus'and'a' minus:'a'dummy'for'a'credit'rating'with'a'plus'and'a'dummy'for'a'credit'rating'with'a' minus.'If'firms'do'not'have'a'plus'or'minus'in'their'credit'rating,'they'are'not'near'a' credit'rating'change.'The'value'of'the'dummy'will'then'be'zero.'' '' The'plus,'minus'or'no'sign'should'precisely'reflect'the'nearness'of'a'credit' rating'change,'because'these'are'used'for'categorize'different'firms.'Nevertheless,' this'might'be'a'too'broad'distinction'and'this'could'affect'the'precision'of'the'test.'For' instance,'a'strong'AA\'firms'may'be'not'near'a'downgrade'and'a'weak'AA+'may'be' not'near'an'upgrade.'This'could'cause'a'weaker'test,'because'the'effect'can'be'

underestimated.'In'this'test'is'also'assumed'that'firms'are'only'concerned'about'an' upgrade'or'downgrade'to'a'different'rating'class,'for'example'from'BBB+'to'A\.'Thus,' it'is'assumed'that'firms'do'not'care'about'the'rating'change'within'a'rating,'for' example'from'BBB'to'BBB+.'Thus,'in'this'case'the'effect'can'also'be' underestimated,'I'think'firms'are'concerned'with'a'rating'change'of'any'type,'also'the' change'within'a'Broad'Rating.' '' Another'implication'of'this'model'is'that'debt'and'equity'issuance'and' reductions'are'not'fast'events.'They'are'going'hand'in'hand'with'transaction'costs,' therefore'these'issuance'and'reductions'are'lump'and'sporadic.'So,'it'may'take' some'time'before'the'issuances'or'reductions'take'place.'Furthermore,'there'is'a'lag' between'the'time'the'decision'is'made'and'the'time'it'occurs'in'the'data.'These' implications'likely'result'in'more'noise'in'the'model,'which'may'cause'an' underestimation'of'the'true'effect'of'credit'ratings'on'capital'structure'(Kisgen,'2006).' '' To'get'a'more'realistic'idea'of'the'effect'of'credit'ratings'on'capital'structure,' large'debt'offerings'are'excluded'(i.e.'debt'offerings'greater'than'10%'of'total' assets).'Small'debt'offerings'would'result'in'a'downgrade'for'a'firms'near'a' downgrade,'but'would'not'result'in'a'downgrade'for'a'firms'not'near'a'downgrade.' But'any'firm'that'does'a'large'debt'offering'might'be'downgraded,'even'though'it'is' not'near'a'downgrade.'Large'debt'offerings'are'frequently'associated'with' reorganisations,'changes'in'management'or'acquisitions'(Kisgen,'2006).'Thus,' assumed'is,'that'in'that'case,'credit'rating'changes'will'be'significant,'and'can'be' excluded'from'the'sample.'' '' The'CR\CS'model'has'net'issuance'of'debt'versus'equity'as'dependent' variable,'to'predict'the'effect'on'capital'structure.'To'see'if'there'is'an'effect'on'this' dependent'variable'in'the'subsequent'period,'it'is'related'to'the'credit'rating'situation' at'a'particular'point'in'time.'The'credit'rating'situation'is'measured'quarterly'and'the' capital'structure'effect'is'measured'over'the'subsequent'three'months.'This'is'based' on'book'values,'since'Standard'&'Poor’s'also'focus'on'book'values.' ' The'regression'models'look'like'this:' NetDIssit)=)α)+)β0CRPOM)+)φKit)+)εit) ) NetDIssit)=)α)+)β1CRPlus)+)β2CRMinus)+)φKit)+)εit) )

)The'variable'definitions'are'as'follows:' Dit' ='book'long\term'debt'plus'book'short\term'debt'for'firm'i'at'time)t'' ΔDit'' ='long\term'debt'issuance'minus'long\term'debt'reduction'plus'changes'' '''in'current'debt'for'firm)i'from'time't'to't)+)1)' ΔLTDit'='long\term'debt'issuance'minus'long\term'debt'reduction'for'firm)i'' ''''''''''''''from'time)t'to't)+)1)' Eit'' ='book'value'of'shareholders’'equity'for'firm)i'at'time') ΔEit)) ='sale'of'common'and'preferred'stock'minus'purchases'of'common'and' '''preferred'stock'for'firm'i'from'time't)to't)+)1)' Ait'' ='beginning\of\year'total'assets'for'firm'i'at'time't) CRPlus)='dummy'variable'(equal'to'1)'for'firms'having'a'plus'credit' ''''Rating'at'the'beginning'of'the'period.' CRMinua='dummy'variable'for'firms'having'a'minus'credit'rating'at'the' ''''beginning'of'the'period.'

CRPOM''=''CRPlus'+''CRMinus'='dummy'variable'for'firms'having'a'minus'

''''or'plus'credit'rating'at'the'beginning'of'the'period.'

Kit'' '''='set'of'control'variables,'including'leverage:'Di,t−1/('Di,t−1'+Ei,t−1),'

''''''profitability:''EBITDAi,t−1/'Ai,t−1'and'size:'ln(Salesi,t−1)' ' NetDIssit='(ΔDi,t)−'ΔEi,t)/'Ai,t.)

' '' For'this'data'the'Compustat'database'is'used'with'data'from'1973'to'April' 2016'for'firms'that'have'a'credit'rating'in'this'database.'The'data'about'the'capital' structure'of'firms'come'also'from'Compustat.'I'exclude'firms\years'which'have' missing'data'for'the'relevant'variables.'The'credit'rating'that'is'used'is'the'credit' rating'Standard'&'Poor’s'Long\Term'Domestic'Issuer'Credit'Rating,'this'is'the'same' credit'ratings'Kisgen'(2006)'used,'but'he'only'used'the'credit'ratings'from'1985'till' 2001.'Because'I'use'data'from'1973'till'2016,'I'include'a'dummy'for'the'financial' crisis'of'2007\2008.'In'addition,'to'test'if'firms'act'differently'when'they'are'at'the' threshold'of'being'downgraded'to'non\investment'grade'rating'a'dummy'for'BBB\'is' added.'Likewise,'when'firms'are'at'the'threshold'of'being'upgraded'to'an'investment' grade'rating'a'dummy'for'BB+'is'added.'' ' '

'' Therefore,'my'regression'is'an'extended'form'of'the'regression'of'Kisgen.' ' NetDIssit)=)α)+)β0CRPOM)+)β3Crisis+)β4BBBPlus+)β5BBMinus+)φKit)+)εit) ) NetDIssit)=)α)+)β1CRPlus)+)β2CRMinus)+)β3Crisis+)β4BBBPlus+)β5BBMinus+)φKit)))) +)εit) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )

Results#

!

'' In'this'section'is'evaluated'if'firms'near'an'upgrade'or'downgrade'issue' relatively'less'net'debt'relative'to'net'equity'and'if'firms'at'the'threshold'of'being' upgraded'to'investment'grade'rating'or'being'downgraded'to'non\investment'grade' rating,'issue'less'net'debt'relative'to'net'equity.'The'adapted'CR\CS'tests'this' hypothesis.'' '' First'of'all,'the'three'control'variable'are'all'significant'and'all'have'a'positive' effect'on'the'net'issuance'of'debt'versus'equity.'However,'the'expected'value'of'the' leverage'control'variable'was'negative,'which'implied'that'firms'with'more'leverage' issue'less'new'debt.'This'result'suggests'the'opposite.'This'might'be'the'result'of' managers'responding'to'the'pressure'to'increase'firm'value'by'increasing'debt'and' reducing'equity'(Berger,'Ofek'&'Yermack,'1997).'' '' The'results'in'table'1'suggest'that'firms'at'the'edge'of'losing'or'achieving'their' investment'grade'rating'issue'less'net'debt'relative'to'equity.'The'second'regression' shows'that'firms'with'a'BBB\'rating'issue'0.5%'less'net'debt'relative'to'equity'than' other'firms'and'the'fourth'regression'shows'that'they'issue'0.3%'less'net'debt' relative'to'equity.'For'the'BB+'firms'the'same'effect'is'revealed,'firms'with'a'BB+' rating'issue'0.4%'or'0.6%'respectively'less'net'debt.'Therefore,'firms'at'the'threshold' of'being'downgraded'to'non\investment'grade'rating'or'upgraded'to'investment' grade'ratings'issue'less'net'debt'relative'to'equity'as'a'percentage'of'total'assets'(or' more'net'equity'relative'to'debt'as'a'percentage'of'total'assets).'' '' Table'1'reveals'also'that'during'the'crisis'firms'issued'more'net'debt'relative' to'equity,'in'both'regressions'firms'issued'1.1%'more'net'debt'relative'to'net'equity.' That'can'be'related'to'that'during'the'crisis'stock'is'frequently'undervalued.'Graham' and'Harvey'(2002)'concluded'that'two\third'of'the'CFOs'agreed'that'‘the'amount'by' which'our'stock'is'undervalued'or'overvalued'was'an'important'or'very'important' consideration’'in'issuing'equity.'Furthermore,'firms'are'more'likely'to'repurchase' stock'when'the'market'values'are'low'(Baker'&'Wurgler,'2002).'This'indicates'that' the'coefficients'are'not'just'statistically'significant,'but'also'economically'significant.'' On'the'other'hand,'the'POM'variable'is'not'significant'in'the'second'regression'of' table'1'and'the'sign'is'different'in'the'both'regressions.'This'might'be'caused'by'the' case'that'firms'act'differently'to'an'upgrade'and'a'downgrade,'because'in'the'fourth'

regression'the'dummy'plus'is'significant.'From'that'regression'can'be'concluded'that' firms'with'a'plus'in'their'credit'ratings'issue'0.2%'more'net'debt'than'other'firms.' From'my'results'in'this'table'can'be'concluded'that'firms'near'a'downgrade'do'not' significantly'change'their'capital'structure'decisions,'this'is'in'contrast'with'Kisgen’s' findings'(2006).' ' Table#1) Effect)Credit)Ratings)on)Capital)Structure)Decisions)0)Plus)or)Minus)Test0)adapted) version)of)Kisgen’s)CR0CS)model.) This'table'shows'the'coefficients'and't\statistics'of'the'pooled'time'series'regression'on'the' net'debt'raised'for'the'year'minus'the'net'equity'raised'for'the'year'divided'by'beginning'of' year'total'assets'on'the'dummy'variables'for'credit'ratings'and'on'control'variables' measured'at'the'beginning'of'the'year.'POM'is'dummy'variable'that'equals'one'if'a'credit' rating'has'a'plus'or'minus'in'their'credit'ratings.'The'dummy'variables'‘Plus’'or'‘Minus’'equal' one'if'they'have'a'Plus'or'Minus'respectively'in'their'credit'rating,'otherwise'they'have'a' value'of'zero.''The'dummy'variable'BBB\'equals'one'is'the'credit'rating'is'BBB\,'likewise,'the' dummy'variable'BB+'equals'one'if'the'credit'rating'is'BB+.'The'Crisis'dummy'variable'equals' one'between'2007'and'2008.'Control'variables'are'EBITDA/total'assets,'ln(sales),'the' natural'logarithm'of'total'sales,'and'D/(D+E),'book'debt'divided'by'book'debt'+'book'value'of' shareholder’s'equity.'The'sample'contains'data'from'1973'till'April'2016'and'missing'values' for'any'variable'are'excluded.'Debt'issuance'bigger'than'10%'of'total'assets'are'also' excluded.'Furthermore,'the'sample'only'contains'D/D+E'ratios'between'zero'and'one.' ***,**,*'indicate'statistical'significance'at'the'0.01,'0.05'and'0.10'level,'respectively.' ' Dependent)variable:)NetDIss# ' ' # (1)' (2)' (3)' (4)' Intercept) 0.009***' (21.798)' \0.038***'(\24.064)' 0.009***'(21.803)' \0.038***'(\24.021)' POM) \0.002***' (\3.417)' ' 0.001' (1.142)' ' ' Plus) ' ' 0.000' (0.215)' 0.002***'(3.127)' Minus) ' ' \0.004***' (\6.116)' (\1.594)'\0.001' BBB0) ' \0.005***' (\4.558)' ' \0.003**'(\2.523)' BB+) ' \0.004***' (\3.138)' ' \0.006***'(\4.117)' Crisis) ' 0.011***' (8.454)' ' 0.011***'(8.486)' EBITDA/A) ' 0.310***' (26.575)' ' (25.348)'0.309***' Ln(sales)) ' 0.004***' (13.827)' ' (20.464)'0.004***' D/(D+E)) ' 0.023***' (1.823)' ' (13.802)'0.023***' Adj.)R2) 0.000' 0.022' 0.001' 0.022' N) 75325' 66506' 75325' 66506' F0value) 11.676***' 209.430***' 23.409***' 185.416***' ' ' '

' '' To'test'if'capital'structure'decisions'vary'between'different'ratings,'some' regressions'are'done'on'the'basis'of'different'Broad'Ratings.'Broad'ratings'are' defined'as'the'rating'category'which'includes'the'plus,'the'middle'and'the'minus' subcategory'for'a'particular'ratingc'a'BB'Broad'Rating'includes'the'firms'with'ratings' BB+,'BB'and'BB\.'In'table'2'and'3'the'results'are'presented,'in'table'2'the'POM' dummy'is'used,'and'in'table'3'the'dummy'Plus'and'the'dummy'Minus'is'used.' Expected'is'that'firms'in'a'particular'rating'categories'are'more'concerned'with'credit' ratings'than'other'firms.'' '' Table'2'shows'that'the'dummy'POM'is'not'significant'for'most'of'the'Broad' Ratings,'except'for'the'BBB'and'A'Broad'Rating'firms'the'variable'is'positive'and' significant.'Because'of'these'insignificant'results,'I'conclude'that'firms'act'differently' for'a'potential'upgrade'and'downgrade.'Therefore,'I’m'going'to'focus'on'table'3.'That' table'shows'that'firms'near'an'upgrade'issue'relatively'more'net'debt'than'equity,' and'firms'faces'a'downgrade'issue'relatively'less'net'debt'than'equity.''''''' '' Of'particularly'focus'is'the'threshold'of'achieving'or'losing'the'investment' grade'rating,'expected'is'that'these'firms'are'more'concerned'with'credit'rating' changes'than'other'firms.'This'effect'was'already'shown'in'table'1,'but'this'table'only' shows'this'result'for'BBB\'firms.'Assumed'is'that'this'means'that'firms'with'a'BBB\' rating'issue'no'new'debt'to'avoid'a'downgrade'to'non\investment'grade'rating.'The' results'show'these'firms'issue'0.03%'less'debt'than'other'firms'in'that'rating.'' This'result'is'not'revealed'for'the'BB+'firms,'the'Plus'dummy'is'positive'and' significant.'Therefore,'the'expected'effect'that'firms'issuing'less'net'debt'than'equity' at'the'edge'of'being'upgraded'to'an'investment'grade'rating'to'improve'the'chance' of'the'upgrade,'is'not'revealed'in'this'result.' '' From'table'3'can'also'be'concluded'that'most'firms'with'a'minus'rating'issue' less'net'debt'than'equity'than'firms'that'do'not'have'a'minus'rating,'this'is'in'contrast' with'the'results'in'table'1.'Almost'every'variable'is'significant,'except'for'the'C'Broad' Rating.'The'greatest'decrease'in'net'debt'is'for'the'B\'rating'firms.'Kemper'(2013).' argued'that'this'is'because'these'firms'have'a'problem'with'achieving'new'debt.' Firms'that'have'not'much'credit'worthiness'may'have'a'lack'of'access'to'debt,'and' the'financing'behaviour'of'reducing'their'debt'may'be'not'an'indication'of'firms'to' avoid'a'downgrade.'The'financing'behaviour'of'reducing'net'debt'versus'net'equity' for'firms'in'the'other'rating'classes'(i.e.'A,'BBB,'BB'and'B)'might'be'an'indication'of'

firms'to'decrease'the'chance'of'the'downgrade.'Only'firms'with'an'AA'Broad'Rating' issue'more'net'debt'relative'to'equity:'their'Minus'Dummy'variable'is'positive.' '' The'significance'and'positive'values'of'the'Plus'Dummy'is'revealed'in'most'of' the'variables'in'table'3.'For'firms'in'the'BBB,'BB'an'B'Broad'Rating'the'values'are' positive'and'significant,'for'the'ratings'AA'and'A'the'variables'are'insignificant.'The' variable'is'only'negative'and'significant'for'the'CCC'rating'class.'Thus,'it'appears' that'most'firms'near'an'upgrade'issue'more'net'debt'relative'to'equity'than'firms'not' near'an'upgrade'in'that'ratings.'This'is'in'sharp'contrast'with'the'findings'of'Kisgen' (2006).' '' The'last'two'columns'present'the'results'of'the'regression'on'the'two'more' broad'categories:'investment'grade'and'non'investment'grade.'The'dummy'Plus'and' the'Dummy'minus'are'significant'in'both'categories,'for'the'Plus'dummy'the'variable' is'significant'and'positive'and'for'the'Minus'dummy'the'variable'is'significant'and' negative.'This'results'are'comparable'with'the'results'found'in'the'other'columns.'' '' The'analysis'I'did,'is'not'very'supportive'on'the'findings'of'Kisgen'(2006).' Where'he'argued'that'firms'near'an'upgrade'and'a'downgrade'issue'less'net'debt'to' avoid'the'upgrade'or'downgrade,'I'only'find'this'result'for'firms'near'a'downgrade,' except'for'the'AA'rated'firms.'The'occurrence'of'a'negative'BB'Plus'dummy'in'the' regression'of'Table'1,'and'the'positive'value'of'the'Plus'dummy'in'the'regression'of' the'BB'Broad'Rating'is'contradictory.'However,'when'I'do'a'regression'with'a'dummy' for'the'BB'Broad'Ratings,'table'4'and'5'show'that'the'whole'BB'Broad'Ratings'issue' less'net'debt'relative'to'equity.'The'sample'in'tables'2'and'3'contains'only'firms'with' a'BB'Broad'Rating.'From'the'positive'Plus'rating'in'table'3'can'be'concluded'that'a' BB+'firm'issue'0.5%'more'net'debt'than'firms'with'a'BB'or'BB\'rating.'Overall,'firms' with'a'BB'Broad'Rating'issue'0.8%'less'net'debt'than'firms'in'another'rating'class'is' shown'in'tables'4'and'5.'All'ratings'are'compared'in'table'1,'which'suggests'that' firms'with'a'BB+'rating'issue'less'net'debt'relative'to'equity'than'other'ratings:'0.4%' and'0.6%'respectively'in'regression'2'and'4.'Thus,'firms'with'a'BB+'rating'issue'still' less'net'debt'than'other'firms'in'different'rating'classes.'The'expectation'that'firms' with'a'BBB\'rating'issue'no'new'debt'to'avoid'the'downgrade'to'non'investment' grade'rating,'is'confirmed'by'the'results.'In'both'table'1'and'table'3'the'related' dummies'are'negativec'the'dummy'BBB\'is'significantly'negative.'Thus'firms'on'the' threshold'of'being'downgraded'to'a'non\investment'grade'rating'issue'less'net'debt' than'other'firms:'0.5%'and'0.3%'relatively'in'regression'2'and'4.''

'' Taken'this'into'account,'the'hypothesis'that'firms'at'the'threshold'of'achieving' or'losing'their'investment'grade'rating'is'confirmed.''On'the'other'hand,'the' expectation'that'all'firms'are'concerned'with'their'credit'rating'and'that'firms'near'an' upgrade'want'to'increase'the'chance'of'an'upgrade'and'firms'near'a'downgrade' want'decrease'the'chance'of'a'downgrade'is'not'fully'shown'in'my'results.'I'find'that' most'firms'near'a'downgrade'issue'relatively'less'net'debt,'but'firms'near'an'upgrade' issue'relatively'more'net'debt.'Thus,'it'looks'like'that'firms'are'more'concerned'with'a' downgrade'than'with'an'upgrade.' ' Table#2) Effect)Credit)Ratings)on)Capital)Structure)Decisions)–By)Broad)Rating) Table'2'and'3'shows'the'coefficients'and't\statistics'of'the'pooled'time'series'regression'on' the'net'debt'raised'for'the'year'minus'the'net'equity'raised'for'the'year'divided'by'beginning' of'year'total'assets'on'the'dummy'variables'for'credit'ratings'and'on'control'variables' measured'at'the'beginning'of'the'year.'POM'is'dummy'variable'that'equals'one'if'a'credit' rating'has'a'plus'or'minus'in'their'credit'ratings.'The'dummy'variables'‘Plus’'or'‘Minus’'equal' one'if'they'have'a'Plus'or'Minus'respectively'in'their'credit'rating,'otherwise'they'have'a' value'of'zero.''The'dummy'variable'BBB\'equals'one'is'the'credit'rating'is'BBB\,'likewise,'the' dummy'variable'BB+'equals'one'if'the'credit'rating'is'BB+.'The'Crisis'dummy'variable'equals' one'between'2007'and'2008.'Control'variables'are'EBITDA/total'assets,'ln(sales),'the' natural'logarithm'of'total'sales,'and'D/(D+E),'book'debt'divided'by'book'debt'+'book'value'of' shareholder’s'equity.'The'sample'contains'data'from'1973'till'April'2016'and'missing'values' for'any'variable'are'excluded.'Debt'issuance'bigger'than'10%'of'total'assets'are'also' excluded.'Furthermore,'the'sample'only'contains'D/D+E'ratios'between'zero'and'one.' ***,**,*'indicate'statistical'significance'at'the'0.01,'0.05'and'0.10'level,'respectively.' ' ' ! ! ! Dependent'variable:'NetDIss# ' ' ' ' ' ' # (AA)' (A)' (BBB)' (BB)' (B)' ' (CCC)' ' Inv' Non\inv' Intercept) \0.028***' (\8.633)' \0.036***' (\15.327)' \0.024***' (\9.379)' \0.030***' (\6.524)' \0.042***' (\6.220)' \0.103**' (\2.359)' \0.033***' (\22.039)' \0.040***' (\9.568)' POM) 0.004***' (3.031)' \0.001' (1.068)' 0.002**' (2.239)' \0.000' (0.266)' 0.003' (1.155)' \0.029' (\1.567)' 0.001' (1.541)' 0.000' (0.266)' Crisis) \0.005' (1.496)' 0.016***'(8.149)' 0.011***'(6.652)' 0.014***'(3.841)' 0.016**'(2.498)' 0.024'(\0.428)' 0.011***'(9.755)' 0.014***'(3.751)' EBITDA/A) 0.144***' (5.904)' 0.357***'(20.243)' 0.359***'(16.186)' 0.130***'(3.910)' 0.162***'(3.334)' 0.257'(0.944)' 0.336***'(28.944)' 0.192***'(6.378)' Ln(sales)) 0.003***' (9.086)' 0.003***'(11.112)' 0.001**'(1.968)' 0.000'(\0.515)' \0.002**'(\2.104)' 0.006'(1.185)' 0.003***'(15.410)' 0.001'(1.544)' D/(D+E)) 0.027***' (7.615)' 0.036***'(13.579)' 0.028***'(10.359)' 0.045***'(9.428)' 0.057***'(7.866)' 0.088*'(1.806)' 0.028***'(17.234)' 0.040***'(9.489)' Adj.)R2) 0.032' 0.042' 0.020' 0.011' 0.010' 0.008' 0.031' 0.008' N) 6127' 19060' 21369' 10377' 6943' 677' 48329' 18176' F0value) 40.388***' 166.816***' 87,533***' 22.192***' 15.612***' 1.825' 307.851***' 28.743***'

! ! Dependent'variable:' NetDIss# ' ' ' ' ' ' ' # (AA)' (A)' (BBB)' (BB)' (B)' ' (CCC)' ' Inv' Non\inv' Intercept) \0.027***' (\8.513)' \0.035***' (\15.159)' \0.022***' (\8.875)' \0.027***' (\5.916)' \0.041***' (\6.038)' \0.104**' (\2.378)' \0.033***' (\21.872)' \0.039***' (\9.440)' Plus) 0.001' (0.556)' 0.000' (0.421)' 0.006***' (6.589)' 0.005***' (2.728)' 0.009***' (2.749)' \0.035*' (\1.827)' 0.003***' (4.940)' 0.003*' (1.707)' Minus) 0.004***' (3.412)' \0.002**' (2.043)' \0.003***' (\3.392)' \0.004**' (\2.018)' \0.017***' (3.841)' \0.003' (\0.117)' \0.001**' (\2.051)' \0.004*' (1.858)' Crisis) \0.005' (\1.545)' 0.016***' (8.271)' 0.021***' (6.770)' 0.015***' (4.050)' 0.018***' (2.759)' 0.026' (0.472)' 0.012***' (9.842)' 0.014***' (3.836)' EBITDA/A) 0.146***' (5.979)' 0.350***'(19.628)' 0.347***'(15.628)' 0.126***'(3.810)' 0.112**'(2.272)' 0.252'(0.924)' 0.332***'(28.501)' 0.189***'(6.297)' Ln(sales)) 0.003***' (9.100)' 0.003***'(11.014)' 0.000'(1.298)' \0.001'(\1.619)' \0.002**'(\2.426)' 0.007'(1.439)' 0.003***'(15.250)' 0.001'(1.397)' D/(D+E)) 0.026***' (7.255)' 0.036***' (13.626)' 0.029***' (10.727)' 0.048***' (9.949)' 0.059***' (8.107)' 0.080' (1.624)' 0.028***' (17.360)' 0.040***' (9.490)' Adj.)R2) 0.031' 0.042' 0.024' 0.013' 0.016' 0.008' 0.032' 0.008' N) 6127' 19060' 21369' 10377' 6943' 677' 48329' 18176' F0value) 51.789***' 139.853***' 86.835***' 22.085***' 19.937***' 1.752' 263.217* **' 26.124***' Table!3! Effect)Credit)Ratings)on)Capital)Structure)Decisions)–By)Broad)Rating! !

! Table!4! Effect!Credit!Ratings!on!Capital!Structure!Decisions!–By!Broad!Rating!Dummy! Table&4&and&5&shows&the&coefficients&and&t3statistics&of&the&pooled&time&series&regression&on&the&net&debt&raised&for&the&year&minus&the&net&equity& raised&for&the&year&divided&by&beginning&of&year&total&assets&on&the&dummy&variables&for&credit&ratings&and&on&control&variables&measured&at&the& beginning&of&the&year.&POM&is&dummy&variable&that&equals&one&if&a&credit&rating&has&a&plus&or&minus&in&their&credit&ratings.&The&dummy&variables& ‘Plus’&or&‘Minus’&equal&one&if&they&have&a&Plus&or&Minus&respectively&in&their&credit&rating,&otherwise&they&have&a&value&of&zero.&&The&dummy& variable&BBB3&equals&one&is&the&credit&rating&is&BBB3,&likewise,&the&dummy&variable&BB+&equals&one&if&the&credit&rating&is&BB+.&The&Crisis&dummy& variable&equals&one&between&2007&and&2008.&Control&variables&are&EBITDA/total&assets,&ln(sales),&the&natural&logarithm&of&total&sales,&and& D/(D+E),&book&debt&divided&by&book&debt&+&book&value&of&shareholder’s&equity.&The&sample&contains&data&from&1973&till&April&2016&and&missing& values&for&any&variable&are&excluded.&Debt&issuance&bigger&than&10%&of&total&assets&are&also&excluded.&Furthermore,&the&sample&only&contains& D/D+E&ratios&between&zero&and&one.& ***,**,*&indicate&statistical&significance&at&the&0.01,&0.05&and&0.10&level,&respectively.& &

Dependent&variable:&NetDIss! & & & & & &

! (AA)& (A)& (BBB)& (BB)& (B)&

& (CCC)& Inv& Non3inv&

Intercept! 30.038***&

(324.115)& 30.039***&(324.478)& 30.038***&(323.990)& 30.036***&(322.444)& 30.035***&(322.006)& 30.38***&(323.548)& 30.044***&(27.366)& 30.030***&(318.157)&

DummyBroadRating!

! 0.004***&(3.801)& 0.009***&(13.540)& 0.001&(0.886)& (39.788)&30.008***& 30.011***&(310.908)& 30.027***&(39.561)& 0.014***&(19.748)& 30.014***&(319.748)&

POM! 0.000&

(30.530)& 0.000&(30.795)& 0.000&(30.573)& 30.000&(30.113)& 0.000&(30.175)& 0.000&(30.525)& 0.001&(1.124)& 0.001&(1.124)&

Crisis! 0.011***&

(8.499)& 0.012***&(8.743)& 0.011***&(8.356)& 0.011***&(8.554)& 0.011***&(8.597)& 0.011***&(8.371)& 0.012***&(8.941)& 0.012***&(8.941)&

EBITDA/A! 0.306***&

(24.942)& 0.301***&(24.710)& 0.314***&(25.704)& 0.310***&(24.486)& 0.300***&(24.522)& 0.307***&(25.256)& 0.286***&(23.405)& 0.286***&(23.405)&

Ln(sales)! 0.003***&

(24.942)& 0.003***&(17.889)& 0.004***&(20.254)& 0.003***&(18.524)& 0.003***&(16.055)& 0.003***&(19.122)& 0.002***&(11.244)& 0.002***&(11.244)&

D/(D+E)! 0.024***&

(14.269)& 0.025***&(15.111)& 0.023***&(13.918)& 0.024***&(14.664)& 0.028***&(16.146)& 0.024***&(14.834)& 0.032***&(18.840)& 0.032***&(18.840)&

Adj.!R2! 0.021& 0.024& 0.021& 0.022& 0.023& 0.022& 0.027& 0.027&

N! 66506& 66506& 66506& 66506& 66506& 66506& 66506& 66506&

FJvalue! 242.101***& 270.857***& 239.776***& 255.957***& 259.900***& 255.208***& 306.048***& 306.048***& &

& & &

Table!5!

Effect!Credit!Ratings!on!Capital!Structure!Decisions!–By!Broad!Rating!Dummy!

Dependent&variable:&NetDIss& & & & & & & &

! (AA)& (A)& (BBB)& (BB)& (B)&

&

(CCC)& &

Inv& Non3inv&

Intercept! 30.038***&

(324.058)& 30.039***&(324.449)& 30.038***&(323.932)& 30.036***&(322.441)& 30.035***&(321.704)& 30.037***&(323.496)& 30.044***&(27.473)& 30.029***&(18.877)&

DummyBroadRating!

! 0.005***&(4.467)& 0.009***&(13.851)& 0.000&(0.699)& (39.569)&30.008***& 30.013***&(312.106)& 30.028***&(39.822)& 0.015***&(20.452)& 30.015***&(20.452)&

Plus! 0.002**& (2.317)& 0.001**& (2.216)& 0.001*& (0.939)& 0.001**& (2.116)& 0.003***& (3.959)& 0.002**& (2.276)& 0.003***& (4.724)& 0.003***& (4.724)& Minus! 30.002***& (33.279)& 30.003***& (33.651)& 30.002***& (32.981)& 30.002**& (32.385)& 30.003***& (33.658)& 30.002***& (33.241)& 30.002***& (32.857)& 30.002***& (32.857)& Crisis! 0.011***& (8.592)& 0.012***& (8.827)& 0.011***& (8.424)& 0.11***& (8.609)& 0.11***& (8.723)& 0.11***& (8.441)& 0.12***& (9.063)& 0.012***& (9.063)& EBITDA/A! 0.303***& (24.615)& 0.298***& (24.466)& 0.312***& (25.499)& 0.308***& (25.321)& 0.294***& (24.077)& 0.305***& (25.036)& 0.281***& (23.025)& 0.281***& (23.025)& Ln(sales)! 0.003***&

(19.734)& 0.003***&(17.935)& 0.004***&(20.349)& 0.003***&(18.626)& 0.003***&(15.721)& 0.003***&(19.185)& 0.002***&(11.037)& 0.002***&(11.037)&

D/(D+E)! 0.024***& (14.254)& 0.025***& (15.029)& 0.023***& (13.810)& 0.024***& (14.558)& 0.028***& (16.354)& 0.025***& (14.755)& 0.032***& (18.934)& 0.032***& (18.934)&

Adj.!R2! 0.022& 0.024& 0.021& 0.023& 0.024& 0.023& 0.027& 0.027&

N! 66506& 66506& 66506& 66506& 66506& 66506& 66506& 66506&

Conclusion)

!

!! That!there!is!a!relation!between!credit!ratings!and!capital!structure!is! recognized!in!most!financial!literature.!Kisgen!(2006)!confirmed!this!relation!on!the! basis!of!his!CR@CS!model!and!argued!that!there!is!direct!effect!of!credit!rating! changes!on!capital!structure!decisions.!I!extended!his!CR@CS!model!with!more! recent!data,!his!data!only!contains!the!period!of!1986!till!2001,!and!I!focused!on!the! threshold!of!the!investment/non@investment!grade!rating!by!investigating!the!effect!by! Broad!Rating!and!including!a!dummy!for!BB!Plus!and!BBB!Minus.!I!also!included!a! dummy!for!the!crisis,!because!the!recent!financial!crisis!had!an!impact!on!the!capital! structure!decision!of!firms.!! !! My!findings!are!not!fully!confirming!the!findings!of!Kisgen!(2006),!who!says! that!firms!near!an!upgrade!or!downgrade!issue!1%!less!net!debt!relative!to!equity,!to! enhance!the!chance!of!maintaining!or!achieving!the!investment!grade!rating.!My! results!are!quit!similar!for!the!firms!near!a!downgrade,!but!my!results!do!not!confirm! this!relation!for!firms!near!an!upgrade.!I!find!that!firms!near!an!upgrade!issue! relatively!more!net!debt!than!firms!not!near!a!credit!rating!change.!Thus,!it!might!be! that!firms!are!more!concerned!with!losing!their!credit!rating!than!achieving!a!better! credit!rating.!The!cost!and!benefits!related!to!the!pecking!order!theory!or!the!trade@off! theory!might!in!this!case!outweigh!the!cost!or!benefits!related!with!the!upgrade.!! The!hypothesis!that!firms!are!more!concerned!with!credit!ratings!at!the!edge! of!the!investment!grate!rating/non@investment!grade!rating!is!not!rejected.!Firms!with! a!BBB@!rating!or!a!BB+!rating!issue!relatively!less!net!debt!than!firms!in!other!rating! classes!to!avoid!the!downgrade!or!to!increase!the!chance!of!the!upgrade.!! !! Thus,!taken!this!all!into!account,!I!think!that!most!CFOs!are!concerned!with! credit!rating!changes,!but!for!firms!at!the!edge!of!losing!or!achieving!their!investment! grade!rating!are!even!more!concerned!with!these!changes.!These!changes!have!a! greater!impact!on!their!discrete!costs!than!other!credit!rating!changes,!so!in!this!case! the!costs!or!benefits!of!the!credit!rating!changes!outweigh!the!costs!or!benefits! related!to!the!other!capital!structure!theories.!! !! I!want!to!note!that!my!results!not!necessarily!mean!that!firms!near!an!upgrade! do!not!take!credit!ratings!into!account!when!making!a!capital!structure!decision.!Such! a!conclusion!is!hard!to!claim,!because!Graham!and!Harvey!(2002)!argued!that!credit!

ratings!are!the!second!highest!concern!for!CFOs.!It!is!plausible!that!there!is!much! noise!in!the!data!that!can!cause!the!insignificant!results!for!the!POM!dummy!or!for! the!other!insignificant!results.!It!is!also!possible!that!firms!near!an!upgrade!or! downgrade!use!other!attributes!than!issuing!or!reducing!debt!and!equity,!for!example! asset!restructuring!and!operating!cost!changes.!Furthermore,!the!lagged!effect!of!the! capital!structure!decisions!can!also!be!a!reason!for!the!different!results.!The! coefficients!can!underestimate!the!real!effect.!Therefore,!future!research!could!take! this!lagged!effect!more!into!account!to!get!a!more!precise!idea!of!the!real!effect!of! credit!rating!changes!on!capital!structure!decision.

References((

!

Baker,!M.,!&!Wurgler,!J.!(2002).!Market!timing!and!capital!structure.!The$Journal$of$$ $$ Finance,!57(1),!1=32.! ! Berger,!P.!G.,!Ofek,!E.,!&!Yermack,!D.!L.!(1997).!Managerial!entrenchment!and!! capital!structure!decisions.!The$Journal$of$Finance,!52(4),!1411=1438.! ! Boot,!A.!W.,!Milbourn,!T.!T.,!&!Schmeits,!A.!(2006).!Credit!ratings!as!coordination!! ! mechanisms.!Review$of$Financial$Studies,!19(1),!81=118.! ! Bowker,!J.,!(2016,!May!23).!PPC!Slumps!as!Cement!Maker!Warns!of!Credit=Rating! ! Downgrade.!Retrieved!from:!http://www.bloomberg.com/news/articles/2016=!!! 05=23/ppc=plunges=as=cement=maker=prepares=256=million=fundraising! !! Bradley,!M.,!Jarrell,!G.!A.,!&!Kim,!E.!(1984).!On!the!existence!of!an!optimal!capital!! structure:!Theory!and!evidence.!The$Journal$of$Finance,!39(3),!857=878.! ! Brealey,!R.!A.,!Myers,!S.!C.,!Allen,!F.,!&!Mohanty,!P.!(2012).!Principles$of$ $$$$$$$$$$$Corporate$Finance.!New!York:!Tata!McGraw=Hill!Education.$ ! Brigham,!E.!F.,!Gapenski,!L.!C.,!&!Ehrhardt,!M.!C.!(1998).!Financial$$ $$$$$$$$$$$ManagementC$Theory$and$Practice.!Harcourt!College!Publishers.! ! Cantor,!R.,!&!Packer,!F.!(1997).!Differences!of!opinion!and!selection!bias!in!the!credit!! rating!industry.!Journal$of$Banking$&$Finance,!21(10),!1395=1417.! ! Damodaran,!A.!(2010).!Applied!corporate!finance.!New!York:!John!Wiley!!!!! !!!!!!!!!!!!!!&!Sons.! ! Frank,!M.!Z.,!&!Goyal,!V.!K.!(2009).!Capital!structure!decisions:!which!factors!are!! reliably!important?!Financial$management,!38(1),!1=37.!

Graham,!J.,!&!Harvey,!C.!(2002).!How!do!CFOs!make!capital!budgeting!and!capital!! structure!decisions?!Journal$of$applied$corporate$finance,!15(1),!8=23.! ! Moyer,!R.!C.,!McGuigan,!J.!R.,!Rao,!R.!P.,!&!Kretlow,!W.!J.!(2011).!Contemporary$$ financial$management.!Toronto:!Nelson!Education.! ! Myers,!S.!C.!(1984).!The!capital!structure!puzzle.!The$Journal$of$Finance,!39(3),!574=! 592.! ! The!Economist,!(2005,!March!23).!Regulators!need!a!new!approach!to!an!! entrenched!industry,!Retrieved!from:!http://www.economist.com/node/3789445! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Appendix((

! The!descriptive!statistics!are!shown!in!table!6.!The!mean!of!the!control!variable!sales! is!6.5!billion,!the!mean!leverage!is!54.77%!and!the!mean!profitability!ratio!is!3.3%.! Furthermore,!the!mean!of!the!net!issuance!of!debt!versus!equity!is!0.8%.!! ! Table&6&& Descriptive$statistics$ In!table!6!the!descriptive!statistics!are!shown.!The!sample!contains!data!from!1973!till!April! 2016!and!missing!values!for!any!variable!are!excluded.!The!NetDIss$is$net!debt!raised!for! the!year!minus!the!net!equity!raised!for!the!year!divided!by!beginning!of!year!total!assets!on! the!dummy!variables.!The!mean,!standard!deviation!and!the!number!of!observations!are! given!for!the!control!variables!are!EBITDA/total!assets,!ln(sales)!in!millions,!the!natural! logarithm!of!total!sales,!and!D/(D+E),!book!debt!divided!by!book!debt!+!book!value!of! shareholder’s!equity.! $ ( Mean! Median! Std! Deviation! Obs! NetDIss$ 0.008! 0.002! 0.0747! 75326! EBITDA/A$ 0.0330! 0.307! 0.0240! 75326! ! Ln(sales)$ 6,5100! 6,5274! 1.6556! 74802! D/(D+E)$ 0.5477! 0.5505! 0.1772! 66895! ! In!table!7!is!shown!that!the!debt!to!total!capitalization!ratios!have!the!expected! relation!to!ratings:!the!higher!the!credit!ratings,!the!lower!the!debt!to!total! capitalization!ratio.!Firms!in!the!AA!Broad!Rating!category!have!median!ratios! between!42%!and!50%,!where!firms!in!the!B!Broad!Rating!category!have!median! ratios!between!66%!and!72%.!! ! Table&7& Sample$Summary$Statistics—Ratings$and$Leverage$ Means,!median!and!standard!deviations!of!D/(D+E)!by!credit!rating!within!the!sample,!and! number!of!firm=years!that!had!the!particular!credit!rating!at!the!beginning!of!the!year.! D/(D+E)!is!book!debt!divided!by!book!debt!+!book!value!of!shareholder’s!equity.! The!sample!contains!data!from!1973!till!April!2016!and!missing!values!for!any!variable!are! excluded.!Debt!issuance!bigger!than!10%!of!total!assets!are!also!excluded.!Furthermore,!the! sample!only!contains!D/D+E!ratios!between!zero!and!1.!

( ! AAA! AA+! AA! AA=! A+!

Number$of$firmUyears( 2196! 829! 3310! 3006! 5904!

D/(D+E)$ ! ! ! ! ! !

$$$Mean$ ! 44.85%! 40.39%! 46.32%! 49.49%! 50.45%!

$$$Median$ ! 44.58%! 42.59%! 49.16%! 49.85%! 51.50%!

! ( ! A! A=! BBB+! BBB! BBB=! Number$of$firmUyears( 8947! 6981! 7689! 9884! 6225! D/(D+E)$ ! ! ! ! ! ! $$$Mean$ ! 52.69%! 51.91%! 53.02%! 54.39%! 54.23%! $$$Median$ ! 53.34%! 53.64%! 53.82%! 56.03%! 55.49%! $$$Std.$Dev$ ! 15.92%! 16.32%! 15.65%! 15.90%! 15.93%! ! ( ! BB+! BB! BB=! B+! B! Number$of$firmUyears( 3481! 3899! 4161! 4568! 2032! D/(D+E)$ ! ! ! ! ! ! $$$Mean$ ! 56.83%! 57.87%! 60.81%! 65.94%! 71.49%! $$$Median$ ! 57.71%! 58.25%! 61.33%! 66.33%! 73.99%! $$$Std.$Dev$ ! 17.05%! 16.34%! 17.44%! 17.30%! 18.68%! ! ! ! ( ! B=! C!or!below! Number$of$firmUyears( 1115! 1009! ! D/(D+E)$ ! ! ! ! $$$Mean$ ! 67.56%! 72.45%! ! $$$Median$ ! 71.37%! 76.18%! ! $$$Std.$Dev$ ! 22.47%! 19.49%! ! ! ! ! ! ! ! ! ! !