Corporate responsibility and corporate misbehavior: are CSR reporting firms indeed responsible?

Rei maie , Ch is ine; Schul ze, Wol gang; Vollme , Julia
A icle — Published Ve sion
Co po a e esponsibili y and co po a e misbeha io : a e
CSR epo ing i ms indeed esponsible?
Re iew o Accoun ing S udies
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Rei maie , Ch is ine; Schul ze, Wol gang; Vollme , Julia (2024) : Co po a e
esponsibili y and co po a e misbeha io : a e CSR epo ing i ms indeed esponsible?, Re iew o
Accoun ing S udies, ISSN 1573-7136, Sp inge US, New Yo k, NY, Vol. 30, Iss. 2, pp. 1804-1872,
h ps://doi.o g/10.1007/s11142-024-09850-8
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/330628
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Vol:.(1234567890)
Re iew o Accoun ing S udies (2025) 30:1804–1872
h ps://doi.o g/10.1007/s11142-024-09850-8
Co po a e esponsibili y andco po a e misbeha io : a e
CSR epo ing i ms indeed esponsible?
Ch is ineRei maie 1· Wol gangSchul ze1 · JuliaVollme 1
Accep ed: 11 July 2024 / Published online: 28 Oc obe 2024
© The Au ho (s) 2024
Abs ac 
We in es iga e whe he i ms ha p oclaim a commi men o co po a e social
esponsibili y (CSR) by CSR epo ing indeed in e nalize such a commi men and
beha e mo e esponsibly. We analyze he associa ion o he issuance and quali y o
olun a y CSR epo s wi h he occu ence, numbe , and se e i y o co po a e mis-
beha io s, bo h p eceding and subsequen o CSR epo ing. We ind a signi ican ly
posi i e associa ion o CSR epo ing wi h ou measu es o p io and u u e misbe-
ha io . The esul s a e co obo a ed by a quasi-na u al expe imen a ound he Rana
Plaza disas e whe e we ind ha he signa o ies o an acco d o be e wo king
condi ions ha e signi ican ly highe p io and u u e misbeha io ela i e o non-
signa o ies and i ms una ec ed by he exogenous shock. Ou esul s a e in line wi h
legi imacy heo y implying ha , on a e age, he i ms’ p oclaiming commi men o
CSR is no a signal o in e nalized commi men bu mo e likely se es g eenwashing
and imp ession managemen pu poses.
Keywo ds Co po a e social esponsibili y (CSR)· Co po a e misbeha io · CSR
epo ing· Real e ec s· Signaling heo y· Legi imacy heo y
JEL Classi ica ion G18· G32· K38· K42· M41· M48· Q01
* Wol gang Schul ze
W[email p o ec ed]g.de
1 Depa men o Accoun ing andCon ol, Uni e si y o Augsbu g, Uni e si ae ss . 16,
86135Augsbu g, Ge many
1805
Co po a e esponsibili y andco po a e misbeha io : a e…
1 In oduc ion
Whe he i ms ha publicly p oclaim commi men o co po a e social espon-
sibili y (CSR) indeed in e nalize such commi men s and ac esponsibly is an
un esol ed esea ch ques ion.1 Recen ini ia i es o egula e CSR epo ing, such
as he Eu opean Non-Financial Repo ing Di ec i e 2014/95/EU, conside CSR
epo ing a means o be e align manage ial decision-making wi h s akehold-
e s’ in e es s, ha is, o al e i m beha io owa d mo e esponsibili y (e.g.,
GRI 2011; Dechow 2023). Howe e , p io empi ical e idence on whe he CSR
epo ing is indeed associa ed wi h mo e esponsible beha io is inconclusi e
(see, e.g., Ch is ensen e al. (2021) o a e iew).
While mo e and mo e i ms olun a ily disclose CSR in o ma ion, he e is
no indica ion ha he numbe o cases o co po a e i esponsibili y is dec easing.
Media epo s e eal misbeha io such as human igh s abuse o en i onmen al
damage daily. Disclosu e heo y sugges s ha i ms ha signal commi men o CSR
should “walk hei alk” and educe misbeha io (He malin 2013; Hoi e al. 2013),
in line wi h s ewa dship heo y and he ela ed p inciple o “wha ge s measu ed ge s
done” (Donaldson and Da is 1991; Da is e al. 1997). Socio-poli ical legi imacy
heo y (Dowling and P e e 1975), in con as , a gues ha i ms p ima ily p oclaim
a commi men o CSR o legi imize misbeha io and enhance hei public epu a-
ion (Mone a e al. 2006). Which heo y mo e accu a ely explains i m beha io is
la gely an empi ical ques ion, which we e isi om a no el pe spec i e.
P io s udies commonly measu e CSR- ela ed ac i i ies as an agg ega e meas-
u e o CSR pe o mance. These measu es agg ega e esponsible and i esponsible
beha io and hus cap u e whe he one o he wo p edomina es (Hughes e al. 2001;
Co mie e al. 2004). This measu emen and agg ega ion issue has been ound o
be one o he main easons why p io s udies ind con adic o y e idence abou he
ela ion be ween CSR epo ing and co po a e (mis)beha io (e.g., Pa en 2002;
Ma ingly and Be man 2006; S ike e al. 2006; Semeno a and Hassel 2015; Ch is-
ensen e al. 2021).
When i ms do posi i e hings in some a eas o compensa e o i esponsi-
ble beha io in o he s, agg ega e measu es o CSR pe o mance may indica e
imp o emen s despi e inc eases in misbeha io (Ko chen and Moon 2012; Adams
and Abhayawansa 2022). Such agg ega ion is highly p oblema ic because i e-
sponsible ac i i ies such as co up ion, aud, and child labo a e associa ed
wi h high economic and social cos s, whe eas esponsible ac i i ies such as CO2
educ ion and de elopmen aid do no ha e bene i s on he same scale and can-
no compensa e o he i esponsible beha io (Hail e al. 2018). Responsible and
i esponsible ac i i ies a e in ac conside ed dis inc concep s ha do no simply
desc ibe opposi e e lec ions o CSR bu a e la gely independen o each o he
and ollow di e en pu poses (e.g., McGui e e al. 2003; S ike e al. 2006). We
1 P ac i ione s and academics in e changeably e e o “co po a e esponsibili y,” “co po a e social
esponsibili y (CSR),” “en i onmen al, social, and go e nance (ESG) issues,” and “sus ainabili y” o
desc ibe commi men o sus ainable economic de elopmen and e hical s anda ds (e.g., VanMa ewijk
2003; Ch is ensen, Hail, and Leuz 2021).
1806
C.Rei maie e al.
he e o e do no examine an agg ega e measu e o CSR pe o mance, bu ocus
on i esponsible beha io o analyze i i ms ha p oclaim commi men o CSR
subsequen ly a oid misbeha io , in line wi h F iedman’s (1970, 126) demand on
i ms o s ay “wi hin he ules o he game” as well as he EU Taxonomy p inci-
ple o “do no signi ican ha m” (EU 2019, 4).2
To da e, o he bes o ou knowledge, only one s udy (Ch is ensen 2016) has
analyzed he ela ion o olun a y CSR epo ing and a non-agg ega ed measu e o
misbeha io . I inds a decline in CSR- ela ed lawsui s subsequen o issuing a CSR
epo . These esul s may imply ha i ms imp o e hei CSR pe o mance subse-
quen o CSR epo ing as hey in e nalize he p oclaimed commi men s, consis en
wi h he in en ions o hose who emphasize he s ewa dship ole o disclosu e, like
he Global Repo ing Ini ia i e (GRI) (Ch is ensen 2016). An al e na i e explana ion,
howe e , is ha CSR epo ing p o ides a mechanism o success ul image imp o e-
men , such ha legal p osecu ion dec eases while co po a e beha io is no imp o ed.
Mo eo e , p osecu ed cases o co po a e misbeha io a e a e (Ma k-Unge ich
and Weiskop 2007), so many cases o misbeha io go unno iced when using such
measu es, pa icula ly hose ha occu in subsidia ies unde o eign ju isdic ion. In
ac , legal p osecu ion is no he only o m o sanc ions o co po a e i esponsibil-
i y, as i ms ace epe cussions om cus ome s and o he s akeholde s when hei
public image is ha med. The public media plays a i al ole in e ealing misbeha io
and shaping s akeholde s’ pe cep ions (Einwille e al. 2010; Dube and Zhu 2021).
We hus s udy a b oade measu e o misbeha io ha cap u es he i ms’ accoun -
abili y o he public, based on c i ical media epo s. Mo eo e , p io li e a u e has
only s udied he dicho omous decision whe he o olun a ily issue a CSR epo .
We ex end his by also conside ing CSR disclosu e quali y based on Bloombe g’s
ESG disclosu e sco e o p o ide a deepe unde s anding o he ela ion be ween
CSR epo ing and misbeha io .
We use he independen Co po a e C i ic Resea ch Da abase (CCRD), which
includes a b oad ange o summa ies o public in o ma ion on ( esponsible and
i esponsible) co po a e beha io (CCRD 2020a). The CCRD is pa o he B i -
ish E hical Consume Ini ia i e, which p o ides e hical a ings based on i s esea ch
da abase. I collec s i m-speci ic CSR- ela ed news om all o e he wo ld. The
da abase co e s all i ms ha sell p oduc s o se ices in Eu ope. We iden i y cases
and e alua e he se e i y o misbeha io based on his in o ma ion. We de ine
misbeha io as iola ions o he p inciples o he Uni ed Na ions Global Compac
(UNGC), which may be conside ed a consensus on global e hical s anda ds (e.g.,
Rasche and Waddock 2014).3 Ou measu es o co po a e i esponsibili y hus do no
2 “The e is one and only one social esponsibili y o business— o use i s esou ces and engage in ac i -
i ies designed o inc ease i s p o i s so long as i s ays wi hin he ules o he game, which is o say,
engages in open and ee compe i ion wi hou decep ion o aud” (F iedman 1970, 126). “Responsible
businesses enac he same alues and p inciples whe e e hey ha e a p esence, and know ha good
p ac ices in one a ea do no o se ha m in ano he ” (UNGC 2020a).
3 The UNGC is a wo ldwide olun a y compac be ween he Uni ed Na ions, i ms, and o he pa ies
(e.g., NGOs and business associa ions) (UNGC 2020b). The en UNGC p inciples encompass human
igh s, labo , en i onmen , and an i-co up ion ha e lec esponsibili y o people, he plane , and long-
e m co po a e success (UNGC 2020a).
1807
Co po a e esponsibili y andco po a e misbeha io : a e…
ocus on only one dimension o CSR, bu comp ehensi ely include en i onmen al,
social, and go e nance (ESG) issues. Fo obus ness, we also use he ESG con o-
e sy sco e om Re ini i as an al e na i e measu e o co po a e i esponsibili y.
Ano he ea u e ha dis inguishes ou s udy om p io esea ch is ha we con-
side lead-lag ela ions (G ange 1969), i.e., co po a e i esponsibili y be o e and
a e CSR epo ing. One migh expec ha i ms wi h (mo e se e e) p io misbe-
ha io con inue o engage in (mo e se e e) misbeha io in he u u e (T e iño 2005;
Wu 2014). Howe e , i CSR epo ing is indeed a c edible signal o inc eased com-
mi men o CSR, i should os e change and b eak a di ec ela ion be ween pas and
u u e misbeha io (e.g., Linnenluecke and G i i hs 2010).
Ou main sample comp ises he i ms lis ed in he Fo une Global 500 be ween
2003 and 2013. Fo obus ness, we also analyze he pe iod un il 2018. We ocus on
olun a y s and-alone CSR epo s based on he GRI guidelines, which ha e become
he de ac o s anda d o CSR epo ing (DeVillie s and Alexande 2014). We also
conside CSR epo s ha do no apply he GRI guidelines in obus ness analyses.
The yea 2003 ep esen s he onse o a pe iod o in ense CSR epo ing (F eundlieb
and Teu ebe g 2013). Ou main sample pe iod ends in 2013 o a oid po en ial biases
due o (1) ea ly adop ion o he EU di ec i e 2014/95/EU, which was issued in 2014
and ende s CSR epo ing manda o y o la ge Eu opean i ms as o Janua y 2017;
and (2) adop ion o he In e na ional In eg a ed Repo ing Council’s amewo k,
published in Decembe 2013, which commonly leads o CSR epo s being com-
bined wi h annual epo s and o he epo s in a single in eg a ed epo ha is no
cap u ed by he GRI measu e. We collec da a on misbeha io o he ime be o e
and a e CSR epo ing, co e ing he pe iod 1999–2018. Compa ed o p io CSR
s udies, we use a la ge da ase ha allows o b oade conclusions.
To co obo a e ou indings, we analyze a second sample ha comp ises he i ms
a ec ed by an exogenous shock o ex ile- ela ed indus ies. In a quasi-na u al expe -
imen , we in es iga e a di e en o m o p oclaiming commi men o CSR. A e he
Rana Plaza disas e in Bangladesh in 2013, many i ms olun a ily signed an acco d
o be e wo king condi ions. We compa e he signa o ies o non-signa o ies wi hin
ex ile- ela ed indus ies and o a con ol g oup o i ms ha we e no a ec ed by he
disas e o he pe iod 2000–2018.
Signaling heo y implies ha i ms ha p oclaim commi men o CSR a e com-
mi ed o educing misbeha io . In con as , legi imacy heo y sugges s ha p io
misbeha io inc eases he likelihood o i ms subsequen ly p oclaiming CSR com-
mi men on he su ace bu no educing misbeha io . In bo h samples, we ind e i-
dence consis en wi h legi imacy heo y. We ind a signi ican ly posi i e ela ion
be ween all measu es o CSR epo ing and p io as well as u u e misbeha io .
Bo h he numbe and se e i y o p io misbeha io s a e posi i ely associa ed wi h
he decision o disclose a CSR epo as well as i s quali y as cap u ed by Bloomb-
e g’s ESG disclosu e sco e. On he o he hand, he numbe and se e i y o u u e
misbeha io s inc ease subsequen o CSR epo ing. This indica es ha i ms
expend highe e o on CSR epo ing when hey ha e a his o y o pas misbe-
ha io and expec u u e misbeha io . These indings a e obus o di e en me h-
ods o add essing endogenei y and o al e na i e a iable measu emen s, model
speci ica ions, and sample composi ions. Likewise, he quasi-na u al expe imen

1808
C.Rei maie e al.
indica es ha he signa o ies ha e signi ican ly highe misbeha io p e and pos
2013, compa ed o bo h pee g oups. Signa o ies show an inc ease in misbeha io
a e signing he acco d, compa ed o non-signa o ies o he same indus y and a
con ol g oup o non- ex ile i ms.
Ou con ibu ion is wo old: Fi s , we con ibu e o he li e a u e on he conse-
quences o CSR epo ing and g eenwashing (e.g., Cla kson e al. 2008; He lin and
Wallace 2017) by ocusing on i esponsible beha io iden i ied by c i ical media
epo s o esol e measu emen issues in p io esea ch. Ou esul s a e consis en
wi h he explana ion ha he i ms’ p oclaiming hei commi men o CSR by olun-
a y CSR epo ing o signing an acco d o be e wo king condi ions is no a signal
o in e nalized esponsibili y bu mo e likely se es g eenwashing pu poses. These
esul s a e impo an because in many pa s o he wo ld he e is an ongoing con o-
e sial deba e abou manda ing CSR epo ing (e.g., Ch is ensen e al. 2021). Ou
analyses show ha olun a y CSR epo ing is no su icien o al e i m beha io
owa ds mo e esponsibili y and is linked o an inc ease in he numbe and se e i y
o u u e misbeha io s. Howe e , whe he manda o y CSR epo ing does be e and
achie es he goal o indi ec ly egula ing i m beha io h ough public p essu e is
up o u u e esea ch. Fi s suppo i e e idence is limi ed o speci ic con ex s like
hyd aulic ac u ing (Bone i e al. 2023b), g eenhouse gas emissions (e.g., Downa
e al. 2021; G ewal e al. 2023; Toma 2023), and clima e change isk disclosu es
unde he SEC 2010 ule (Kim e al. 2023), o o posi i e CSR ac i i ies (Fiech e
e al. 2022).
Second, we con ibu e o he li e a u e on he de e minan s o CSR epo -
ing (e.g., Co mie e al. 2005; Tho ne e al. 2014) by iden i ying p io misbeha -
io ac oss he b oad ange o CSR dimensions as an addi ional de e minan o CSR
epo ing. Knowledge abou wha mo i a es olun a y CSR epo ing is inc easing
bu s ill limi ed (e.g., Hahn and Luel s 2014; Huang and Wa son 2015). E idence on
p io misbeha io as a p eceden o CSR epo ing is limi ed o a ew case s udies
(e.g., Bone i e al. 2023a). O he ela ed s udies ei he es ic misbeha io measu e-
men o he en i onmen al dimension o measu e pe o mance and epo ing a he
same poin in ime (e.g., Neu e al. 1998; Haddock 2005).
2 Rela ed esea ch andhypo heses de elopmen
The e is a la ge li e a u e on he consequences o olun a y CSR epo ing, including
capi al ma ke e ec s and eal e ec s on i m beha io (e.g., Dhaliwal e al. 2011;
Ma sumu a e al. 2014; see Ch is ensen e al. 2021 o a e iew). These p io s udies
la gely ely on signaling heo y o legi imacy heo y.4 Acco ding o signaling heo y,
esponsible i ms will signal hei quali y o he ma ke o di e en ia e hemsel es
om compe i o s (Ake lo 1970; Spence 1973; Mo is 1987). By p o iding addi-
ional in o ma ion ha is ha d o mimic, i ms wi h supe io CSR pe o mance can
signal hei conce n abou and compliance wi h s akeholde s’ expec a ions (Cla kson
4 F ynas and Yamahaki (2016) p o ide an o e iew o heo ies used in p io s udies o explain CSR.
1809
Co po a e esponsibili y andco po a e misbeha io : a e…
e al. 2008; Plumlee e al. 2015). In line wi h he s ewa dship heo y o disclosu e,
addi ional in o ma ion imp o es s akeholde s’ moni o ing and hence he alignmen
o manage ial decisions wi h s akeholde s’ in e es s (Lambe e al. 2007). The use
o measu emen and con ol sys ems o CSR epo ing inc eases i ms’ awa eness o
misbeha io and p omo es in e nal sel - egula ion and esponsible co po a e go e n-
ance (Topping 2012; Wu 2014; Adams and Abhayawansa 2022).
S anda d se e s conside disciplining mechanisms he main eason o manda -
ing CSR epo ing (e.g., EC 2021). Ye , i is unclea whe he epo ing can indeed
achie e he aspi ed eal e ec s. Fi s suppo i e e idence on manda o y epo ing
is limi ed o speci ic cases o CSR. Howe e , i is unclea whe he olun a y CSR
epo ing would no also be su icien o achie e he desi ed eal e ec s (Ch is ensen
e al. 2021; Fiech e e al. 2022).
Legi imacy heo y, in con as , a gues ha i ms use olun a y disclosu es
o legi imize co po a e ac i i ies (Pa en 1991; Cho and Pa en 2007). In pa icu-
la , i ms wi h poo p io CSR pe o mance migh ace s ong public p essu e and,
hence, epo on CSR e o s o p e en o educe epu a ional and legi imacy dam-
ages (Lindblom 1994). Manage s may use CSR epo ing oppo unis ically o
imp ession managemen and g eenwashing pu poses despi e he po en ial cos s o
ge ing caugh (Me kl-Da ies and B ennan 2007; Mishina e al. 2010). Nume ous
s udies ha e shown ha CSR epo ing is o en unbalanced and sel -lauda o y and
neglec s nega i e inciden s (Wiseman 1982; Holde -Webb e al. 2009).
P io empi ical s udies a e inconclusi e abou he ela ion be ween olun a y
CSR epo ing and CSR pe o mance (Ch is ensen e al. 2021). To measu e CSR
pe o mance, p io s udies ocus on en i onmen al aspec s (e.g., Neu e al. 1998;
Kim and Lyon 2011), agg ega e esponsible and i esponsible beha io (e.g.,
Schul ze and T omme 2012; Mahoney e al. 2013; Lee and Max ield 2015), and
measu e a iables o in e es a he same poin in ime wi hou conside ing lead-lag
ela ions. The li e a u e on he consequences o CSR epo ing on misbeha io has
ocused on he issuance o a CSR epo o cap u e CSR epo ing, wi hou consid-
e ing he quali y o he epo ing (e.g., Ch is ensen 2016). O he s udies, howe e ,
ha e shown ha he quali y o CSR disclosu es is in o ma i e in es ablishing a ela-
ion be ween CSR epo ing and, o ins ance, i m alue (Plumlee e al. 2015). To
signal hei CSR commi men o , al e na i ely, o legi imize hei beha io , i ms
may no only decide o issue a olun a y CSR epo bu also expend highe e o
o be e CSR disclosu e quali y. We hence also conside CSR disclosu e quali y in
ou analyses.
Some s udies ind suppo o signaling heo y (e.g., Al-Tuwaij i e  al. 2004;
Mahoney e al. 2013); o he s suppo legi imacy heo y (e.g., Kim and Lyon 2011;
Cho e  al. 2012). E en wi hin single s udies, he esul s a e no conclusi e. Fo
ins ance, Cla kson e al. (2008) ind a posi i e ela ion be ween CSR disclosu es and
CSR pe o mance, which suppo s signaling heo y, bu hey also ind ha legi i-
macy heo y explains pa e ns in hese disclosu es. Explana ions o he inconclu-
si e esul s include inadequa e sample selec ion, omi ed o con ounding a iables
measu ing CSR pe o mance (Pa en 2002; Ch is ensen e al. 2021), and, in pa -
icula , agg ega ing esponsible and i esponsible beha io s (Ma ingly and Be man
2006; S ike e al. 2006; Semeno a and Hassel 2015). To measu e esponsibili y in
1810
C.Rei maie e al.
he sense o a oiding misbeha io , i is impo an o dis inguish i esponsible om
esponsible beha io (McGui e e al. 2003; S ike e al. 2006; K üge 2015).
To he bes o ou knowledge, only Ch is ensen (2016) does no use an agg e-
ga e measu e o CSR pe o mance; he ins ead uses CSR- ela ed lawsui s o cap u e
misbeha io . He inds ha olun a ily issuing a CSR epo educes he likelihood
o CSR- ela ed lawsui s in he ollowing yea , and concludes ha a e CSR epo -
ing, i ms imp o e hei CSR pe o mance due o an in e naliza ion o p oclaimed
commi men s. Howe e , he nega i e associa ion o CSR epo ing and CSR- ela ed
lawsui s is no an unambiguous indica ion ha CSR epo ing indeed imp o es he
i ms’ beha io . An al e na i e explana ion, which is no uled ou in his s udy, is
ha he olun a y issuance o a CSR epo p o ides a mechanism o success ul
image imp o emen , such ha legal p osecu ion dec eases wi hou eal imp o e-
men s in co po a e beha io .
In a simila ein, Raghunandan and Rajgopal (2023) s udy he ela ion be ween
i ms’ signing he Business Round able’s “S a emen on he Pu pose o a Co po-
a ion” and iola ions o US ede al laws. As in Ch is ensen (2016), hei depend-
en a iable equi es a legal in es iga ion, bu such in es iga ions a e a e (Ma k-
Unge ich and Weiskop 2007) and may be biased by he i ms’ poli ical connec ions
(McCa en e al. 2022). Mos cases o co po a e misbeha io will no esul in a
ede ally epo ed iola ion. In pa icula , cases o misbeha io ha occu in o eign
subsidia ies ha a e no subjec o US laws will go unno iced.
In con as , he public media plays a i al ole in e ealing co po a e misbeha io
and shaping s akeholde s’ pe cep ions (Einwille e al. 2010; Dube and Zhu 2021).
In i s desi e o publish a en ion-p o oking pieces, he media has he incen i es,
esou ces, and sc u iny o e eal misbeha io . The e o e, o cap u e i ms’ o e all
accoun abili y o he public, we measu e co po a e misbeha io based on c i ical
media epo s ac oss he b oad ange o CSR dimensions. We conside misbeha io
be o e and a e CSR epo ing o di ec ly add ess lead-lag ela ions, which helps o
iden i y causal ela ions (G ange 1969).
Rega ding he ela ion be ween CSR epo ing and p io misbeha io , signal-
ing heo y implies ha esponsible i ms bene i om CSR epo ing by signaling
ha hey ha e lea ned om pas mis akes. S akeholde s a e pa icula ly igilan in
obse ing he ac i i ies o p e iously i esponsible i ms, such ha s a egic cos s
inc ease (DeTienne and Lewis 2005). To educe such cos s, esponsible i ms ha e
incen i es o signal hei ype. Legi imacy heo y a gues ha i ms use CSR epo ing
a e he e ela ion o misbeha io o espond o inc eased public p essu e. To legi i-
mize hemsel es, i ms may expend highe e o o issue a CSR epo o imp o e
he epo ’s quali y as cap u ed by he public’s eye. E idence o his is limi ed o a
ew case s udies. Jan adej and Ken (1999) show ha he mining i m B oken Hill
published mo e en i onmen al in o ma ion a e he disas e a Ok Tedi Mine. O h-
e s ind simila e ec s a e he Exxon Valdez oil spill (Pa en 1992), he BP oil
spill (He lin and Wallace 2017), and he Fukushima nuclea disas e (Bone i e al.
2023a, b).5 Bo h heo ies imply a posi i e ela ion be ween p io misbeha io and
5 Ch is ensen (2016) includes p io misconduc , measu ed as publicized lawsui s, as a con ol a iable in
his analysis and inds insigni ican esul s.
1811
Co po a e esponsibili y andco po a e misbeha io : a e…
CSR epo ing. Howe e , i esponsible i ms may no ha e su icien a o able CSR
in o ma ion o be willing o issue a CSR epo o imp o e hei CSR disclosu e
quali y, implying a nega i e ela ion. Based on signaling and legi imacy heo y, we
hypo hesize:
H1: P io co po a e misbeha io is posi i ely associa ed wi h CSR epo ing.
Rega ding he e ec s o CSR epo ing on subsequen i m beha io , he a gumen s
o signaling and legi imacy heo y di e . Signaling heo y implies ha i ms expec
capi al ma ke bene i s om imp o ed CSR disclosu e and be e CSR in o ma ion,
which, in u n, imp o es moni o ing and hence i m beha io . This is he a ionale ha
s anda d se e s a e ollowing (e.g., Ch is ensen 2016; EC 2021) and sugges s ha i ms
ha olun a ily issue CSR epo s o spend highe e o s o imp o e CSR disclosu e
quali y will in e nalize hei CSR commi men and imp o e hei beha io .
Howe e , i ms ha e no s ong incen i es o change hei beha io . Sanc ions
on euphemis ic epo ing and misbeha io a e la gely absen (Ma k-Unge ich
and Weiskop 2007). P io e idence sugges s ha in es o s only show e y lim-
i ed eac ions o olun a y CSR epo s (Yoon andSe a eim 2020; Bu zillo e al.
2023). Fi ms may hence use CSR epo ing o imp o e hei legi imacy and epu a-
ion wi hou changing hei ac i i ies (Lindblom 1994). E en highe CSR disclo-
su e quali y, as measu ed by publicly a ailable ESG sco es, does no e lec in e nal
imp o emen s in he esponsibili y o ope a ions bu la gely ep esen s he ul ilmen
o in o ma ion equi emen s by he a ing agencies (Ahmed e  al. 2023). I CSR
epo ing helps i ms o success ully imp o e hei public image, i may lead o less
legal sc u iny and in es o ac i ism bu no necessa ily o less misbeha io . Based
on legi imacy heo y, we hypo hesize:
H2: CSR epo ing is posi i ely associa ed wi h u u e co po a e misbeha io .
3 Measu emen o CSR epo ing andco po a e misbeha io
We measu e CSR epo ing (CSRRi, ) in wo ways: Fi s ly, we use an indica o a i-
able o issuing a olun a y s and-alone GRI-based CSR epo (CSRR_Ii, ), as in
p io esea ch (e.g., Simne e al. 2009; Dhaliwal e al. 2011; Ch is ensen 2016).6
We collec da a om he GRI Da abase, which co e s GRI epo s since 1999. We
use Co po a eRegis e , he i ms’ websi es, and in e ne sea ches o e i y he da a
6 The GRI guidelines p o ide a CSR epo ing amewo k o high quali y (e.g., Ch is ensen 2016) and
a e he mos equen ly used (e.g., KPMG 2022). The GRI does no ocus solely on inancial ma e iali y
bu ex ends o co po a e impac s on sus ainable de elopmen . Such double ma e iali y is cen al o cu -
en deba es and de elopmen s in ol ing manda o y non- inancial epo ing (Adams and Abhayawansa
2022). Despi e ecen a emp s o in eg a e CSR opics in o annual epo s (e.g., he in oduc ion o he
In eg a ed Repo ing F amewo k), s and-alone epo s a e s ill he mos common o m o CSR epo ing
(e.g., KPMG 2008, 2011, 2017, 2022). We di e en ia e nei he be ween he GRI e sions G2 o G4 no
be ween he di e en applica ion le els (A + o C, undecla ed, GRI- e e enced).
1818
C.Rei maie e al.
Rus icus 2010). Se e al s udies use i m age as a alid IV (Ha jo o and Jo 2011;
Ch is ensen 2016). O he s use i m age as a p oxy o epu a ion (e.g. Da a e al.
1999). Fi ms wi h a highe epu a ion end o disclose mo e o main ain epu a ion
(Rei maie and Schul ze 2017). Olde i ms ha e mo e expe ience and lowe epo -
ing cos s (Cam e man and Cooke 2002; Alsaeed 2006), sugges ing a posi i e ela-
ion o AGEi, o CSR epo ing. In con as , he e is no ela ion o u u e misbe-
ha io . Misbeha io depends on ac o s ha de elop independen ly o i m age, like
co po a e cul u e o compe i ion. Li e a u e on accoun ing aud suppo s he i el-
e ance o i m age wi h espec o misbeha io since aud i ms do no di e in age
compa ed o non- aud i ms (Fa be 2005; Chen e al. 2006).
Ou second IV is FFLOATi, . High ee loa comes wi h high agency cos s (Fama
and Jensen 1983; P encipe 2004). To educe hese cos s and o announce compli-
ance, manage s use olun a y disclosu es, implying a posi i e ela ion o FFLOATi,
wi h CSR epo ing (B amme and Pa elin 2006). On he o he hand, a di ec ela-
ion be ween FFLOATi, and co po a e misbeha io is no in ui i e. Chen e  al.
(2006) ind ha boa d oom cha ac e is ics a e mo e ele an in explaining aud han
owne ship cha ac e is ics. O he s udies ind no signi ican ela ion be ween inan-
cial aud and blockholde s (La cke e al. 2007; Johnson e al. 2009).
We use wo u he IVs in addi ional analyses ha ha e been used in p io
esea ch23: The dicho omous a iable DJSIi, -1 indica es DJSI lis ing in -1, which
posi i ely ela es o non inancial disclosu es (Cho e al. 2012). Lis ed i ms a e mo e
likely o issue a CSR epo , as ela ed disclosu es a e equi ed o con inued lis ing
(Dhaliwal e al. 2012). In con as , he e is no e idence o a ela ion be ween DJSI
lis ing in -1 and co po a e misbeha io in + 1. The dicho omous a iable CROSSi,
indica es c oss-lis ing a h ee o mo e s ock exchanges in . Addi ional lis ings
imply addi ional egula o y and epo ing equi emen s, which inc ease olun a y
disclosu es. In con as , c oss-lis ing does no subjec i ms o he CSR egula ions
o laws o ha coun y and hence does no a ec misbeha io h ough legal bonding
(Shi e al. 2018).
4.2.3 Second s age SR model
The SR me hod does no equi e a comple e se o IVs and i s nonlinea mod-
els.24 In compa ison o he 2SLS model, i equi es an addi ional eg esso (V) in
he second s age wi h he ollowing cha ac e is ics (Lewbel 2014): V is exogenous
and con inuously dis ibu ed (Lewbel e al. 2012). V has a posi i e coe icien and
23 Fo obus ness pu poses, we epea ou analyses wi h all combina ions o h ee o wo o he ou IVs
and also analyze al e na i e IVs (Sec .7.2).
24 Lewbel (2014, 38) explains he SR me hod as ollows: “Fo example, suppose an obse ed bina y a -
iable D sa is ies D = I (V + W* ≥ 0), whe e V is he obse ed special eg esso , and W* is an unobse ed
la en a iable. […] Special eg esso me hods wo k by exploi ing he ac ha i V is independen o W*
ime. Likewise, we do no use indus y agg ega es since he endogenous pa o he epo ing decision
a ies no only ac oss indus ies bu also wi hin indus ies. The use o anked endogenous eg esso s
would only ans e endogenei y om he o iginal eg esso s o he anked ones.
Foo no e 22 (con inued)

1819
Co po a e esponsibili y andco po a e misbeha io : a e…
is mono onous (Dong and Lewbel 2015).25 We include he a e age USD-EUR
exchange a e (EXCHANGE ) as V in model (2). We use all di e en measu es o
u u e and p io misbeha io and CSR epo ing, as ou lined in Sec .326:
USD and EUR a e he mos impo an cu encies wo ldwide (e.g., No lo 2009).
The US and Eu ope a e he mos s ongly ep esen ed egions in ou sample. The
USD/EUR exchange a e is exogenous.27 While i does no in luence CSR epo ing,
i migh in luence u u e misbeha io since exchange a es in luence he compe i i e
en i onmen in which a i m ope a es. Exchange a e luc ua ions induce addi ional
isk o in e na ionally ope a ing i ms (Ko abe and Mu ay 2004). The yea ly a e -
age exchange a e is con inuously dis ibu ed.
5 Sample
Ou main sample is based on all 814 i ms lis ed in he Fo une Global 500 in he
pe iod om Janua y 1, 2003, o Decembe 31, 2013. The Fo une Global 500 lis s
he wo ld’s la ges i ms in e ms o o al sales and is equen ly used in CSR s ud-
ies because o i s global impac (e.g., Mulle and Whi eman 2009; Lee and Max-
ield 2015).28 The sample pe iod s a s in 2003, which ma ks he onse o in ense
olun a y CSR epo ing due o ising public and go e nmen al in e es in CSR
(F eundlieb and Teu ebe g 2013). The sample pe iod ends in 2013 o wo easons.
Fi s , he Eu opean Pa liamen and he Council issued EU di ec i e 2014/95/EU in
Oc obe 2014, which ende ed CSR epo ing manda o y as o Janua y 2017 o
la ge Eu opean i ms. Ea ly adop e s migh bias ou analyses ha explici ly ocus
(4)
Fu u eCIR
i, +1
=𝛽
0
+𝛽
1
CSRR
i,
+𝛽
2
UNGC
i, +1
+𝛽
3
PO
i
+𝛽
4
HO
i
+𝛽
5
GLOBAL
i, +
1
+𝛽6P io CIRi, −3∕Σ3y+𝛽7SIZEi, +1+𝛽8ROEi,
+𝛽
9
EXCHANGE
+IND +YEAR +𝜀
i, +1
25 Dong and Lewbel (2015) elax he addi ional equi emen o g ea suppo o V.
26 Fu u eCIRi, +1 = CIR_Ii, +1, CIR_Ni, +1, o CIR_Si, +1; P io CIRi, -3/∑3y = CIR_Ii, -3/∑3y, CIR_Ni, -3/∑3y,
o CIR_Si, -3/∑3y; and CSRRi, = CSRR_Ii, o CSRR_Qi, . The i s s age eg ession emains basically
unchanged. Since i includes all de e minan s and con ol a iables o he second s age eg ession, i also
includes V in he SR app oach.
27 IVs and he SR need o be exogenous in he espec i e model. Tha is, he IVs in model (3) need
o co ela e wi h CSRRi, bu no Fu u eCIRi, +1, and he SR in model (4) needs o co ela e wi h
Fu u eCIRi, +1 bu no CSRRi, . Polynomial eg essions o u u e misbeha io (CIR_Ii, +1, CIR_Ni, +1,
o CIR_Si, +1) on he SR con i m a posi i e coe icien and mono onic g aphical appea ance (Dong and
Lewbel 2015; Bon emps and Nauges 2016).
28 Analyzing only he la ges i ms educes po en ial conce ns abou he disp opo iona e media co e -
age o la ge i ms and imp o es he likelihood ha all cases o misbeha io a e de ec ed by public media
and hence he CCRD.
(ei he uncondi ionally o a e condi ioning on co a ia es) hen a ia ion in V changes he p obabili y
ha D = 1 in a way ha aces ou he dis ibu ion o W* (ei he he uncondi ional dis ibu ion o he dis-
ibu ion condi ional on co a ia es).”.
Foo no e 24 (con inued)
1820
C.Rei maie e al.
on olun a y CSR epo ing. Second, he In e na ional In eg a ed Repo ing Coun-
cil published he In eg a ed Repo ing F amewo k in Decembe 2013. Adop ing
i ms no longe issue s and-alone CSR epo s bu gene ally in eg a e CSR epo s,
annual epo s, and o he epo s in o one epo , which ou GRI measu e does no
cap u e.
We exclude 127 i ms wi h missing da a in Da as eam o Eikon, 96 i ms no
co e ed by he CCRD, and 104 i ms wi h missing da a in o he sou ces (UNGC
homepage, DJSI o In e b and anking), esul ing in a sample o 487 unique
i ms. We u he exclude 858 i m-yea obse a ions due o bank up cy o me g-
e s and acquisi ions. We do no exclude hose i ms comple ely o a oid su i-
o ship biases. Ou inal main sample includes 4,499 i m-yea obse a ions. We
collec da a on misbeha io o he ime be o e and a e olun a y CSR epo -
ing (2000–2014). As Bloombe g’s ESG disclosu e sco e does no co e all i ms
o e he en i e pe iod, he analyses e e o a educed sample o 3,150 i m-yea
obse a ions.
Table 1 panel A shows he dis ibu ion o ou main sample i ms pe ICB
supe sec o and he p opo ion o i ms wi h misbeha io (CIR_Ii, ) pe supe sec-
o . CIR_Ii, is highes in he supe sec o s Food & Be e age (ICB code 3500),
Pe sonal & Household Goods (3700), and Technology (9500), whe e mo e han
50 pe cen o he i ms engage in misbeha io . A 17.65 pe cen , CIR_Ii, is low-
es in he supe sec o Cons uc ion & Ma e ials (2300). Panel B p esen s he
i ms’ coun y o o igin and shows ha mo e han one- hi d o he i ms a e om
he US (36.96 pe cen ), ollowed by Japan (14.78), he UK (8.42), F ance (7.19),
and Ge many (5.13). The emaining 27.5 pe cen a e in 22 di e en coun ies.
The pe cen age o i m-yea obse a ions wi h misbeha io is highes in Tu -
key (100 pe cen , one i m wi h misbeha io in each yea ) and Taiwan (71.43
pe cen ). I is lowes in Aus ia and Colombia (0 pe cen each, one/ wo i ms
wi hou misbeha io ).
Table 2 p esen s he de elopmen o he numbe o i ms (no ) issuing a GRI
epo in a pa icula yea (CSRR_Ii, = 0/1) o e ou sample pe iod. The hi d
( ou h) column p esen s he co esponding numbe o i ms wi h (wi hou ) mis-
beha io in he ollowing yea (CIR_Ii, +1 = 0/1). The numbe o i ms issuing a
GRI epo inc eased om 58 in 2003 o 262 in 2013. The numbe o i ms wi h
subsequen misbeha io does no ollow a sys ema ic end; he le els a e simila
in he i s and las sample yea (145 in 2003 and 157 in 2013). This sugges s ha
he inc ease in CSR epo ing does no come wi h a sys ema ic educ ion in u u e
misbeha io .
Table 3 p o ides he desc ip i e s a is ics. Fo ease o in e p e a ion, he alues
o he numbe (CIR_Ni, -3) and se e i y o cu en (CIR_Si, ), p io (CIR_Ni, -3, CIR_
Ni,∑3y, CIR_Si, -3, o CIR_Si,∑3y), and u u e misbeha io s (CIR_Ni, +1 o CIR_Si, +1) a e
aw da a be o e log ans o ma ion.29 Fo y- h ee pe cen o he obse a ions a e GRI
epo e s in . Fi y- h ee pe cen issue some kind o CSR epo in . The a e age i m
29 To adjus o i s s ongly skewed dis ibu ion, we log ans o m ou measu es o he numbe and se e -
i y o misbeha io s in ou empi ical analyses (see Sec .3 o de ails).
1821
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 1 Classi ica ion o sample i ms
Indus y Supe sec o Numbe o
Fi ms %o Fi ms pe CIR_Ii,
Supe sec o Indus y No Misbeha io Misbeha io %
Panel A: Sample i ms pe ICB supe sec o
0001 Oil & Gas 0500 Oil & Gas 41 8.42 8.42 247 116 31.96
1000 Basic Ma e ials 1300 Chemicals 12 2.46 6.37 76 31 28.97
1700 Basic Resou ces 19 3.90 116 52 30.95
2000 Indus ials 2300 Cons uc ion & Ma e ials 21 4.31 18.48 182 39 17.65
2700 Indus ial Goods & Se ices 69 14.17 519 180 25.75
3000 Consume Goods 3300 Au omobiles & Pa s 29 5.95 13.55 226 62 21.53
3500 Food & Be e age 15 3.08 67 76 53.15
3700 Pe sonal & Household Goods 22 4.52 111 114 50.67
4000 Heal h Ca e 4500 Heal h Ca e 22 4.52 4.52 122 77 38.69
5000 Consume Se ices 5300 Re ail 43 8.83 13.14 255 130 33.77
5500 Media 9 1.85 61 18 22.78
5700 T a el & Leisu e 12 2.46 70 30 30.00
6000 Telecommunica ions 6500 Telecommunica ions 15 3.08 3.08 91 32 26.02
7000 U ili ies 7500 U ili ies 38 7.80 7.80 247 86 25.83
8000 Financials 8300 Banks 48 9.86 18.48 326 107 24.71
8500 Insu ance 31 6.37 215 67 23.76
8600 Real Es a e 0 0 0 0 -
8700 Financial Se ices 11 2.26 46 26 36.11
9000 Technology 9500 Technology 30 6.16 6.16 136 143 51.25
To al 487 100 100 3.113 1.386 30.81
1822
C.Rei maie e al.
Table 1 (con inued)
Na ion Numbe o
Fi ms % o Fi ms CIR_Ii,
No Misbeha io Misbeha io %
Panel B: Sample i ms pe coun y
Aus alia 10 2.05 71 19 21.11
Aus ia 2 0.41 20 0 0.00
B azil 3 0.62 17 4 19.05
Canada 18 3.70 128 19 12.93
Colombia 1 0.21 1 0 0.00
Denma k 2 0.41 15 1 6.25
Finland 5 1.03 38 11 22.45
F ance 35 7.19 236 109 31.59
Ge many 25 5.13 160 102 38.93
Hong Kong 3 0.62 27 6 18.18
India 7 1.44 71 4 5.33
I eland 3 0.62 26 5 16.13
I aly 11 2.26 82 12 12.77
Japan 72 14.78 532 142 21.07
Mexico 2 0.41 16 4 20.00
Ne he lands 11 2.26 56 49 46.67
Po ugal 2 0.41 6 1 14.29
Singapo e 2 0.41 17 1 5.56
Sou h Ko ea 17 3.49 87 32 26.89
Spain 12 2.46 106 21 16.54
Sweden 6 1.23 56 10 15.15
Swi ze land 11 2.26 80 40 33.33
Taiwan 4 0.82 10 25 71.43
1823
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 1 (con inued)
Thailand 1 0.21 8 1 11.11
Tu key 1 0.21 0 6 100.00
Uni ed Kingdom 41 8.42 140 228 61.96
Uni ed S a es 180 36.96 1.107 534 32.54
To al 487 100 3.113 1.386 30.81
P esen s he dis ibu ion o sample i ms ac oss ICB codes and geog aphic egions. Panel A shows he absolu e and pe cen al dis ibu ion o sample i ms ac oss ICB
supe sec o s (SP) and he pe cen al dis ibu ion o sample i ms ac oss ICB indus ies (IN). In addi ion, i shows he absolu e numbe o i m-yea obse a ions pe ICB
supe sec o wi hou (CIR_Ii, = 0) and wi h misbeha io (CIR_Ii, = 1) as well as he pe cen age o i m-yea obse a ions wi h misbeha io pe ICB supe sec o (CIR_
Ii, = 1). Panel B shows he absolu e and pe cen al dis ibu ion o sample i ms ac oss coun ies. In addi ion, i shows he absolu e numbe o i m-yea obse a ions pe
coun y wi hou (CIR_Ii, = 0) and wi h misbeha io (CIR_Ii, = 1) as well as he pe cen age o i m-yea obse a ions wi h misbeha io pe coun y (CIR_Ii, = 1)

1824
C.Rei maie e al.
has a Bloombe g ESG disclosu e sco e o 40.32 pe cen . On a e age, almos one- hi d
o he sample i ms conduc misbeha io (CIR_Ii, ). The mean o 0.94 o CIR_Ii,∑3y
indica es ha i ms engage in misbeha io in abou one o he pas h ee yea s on a e -
age. The a e age i m conduc s 0.54 cases o misbeha io pe yea ; he maximum
numbe in and + 1 is 23 (CIR_Ni, , CIR_Ni, +1). The maximum numbe o misbe-
ha io s in -3 is 30 (CIR_Ni, -3). The a e age i m conduc s 1.71 cases o misbeha io
o e he p io h ee yea s; he maximum numbe is 44 (CIR_Ni,∑3y). The maximum
se e i y sco e o misbeha io is 93.00; he mean is 2.37 in (CIR_Si, ), 2.36 in + 1
(CIR_Si, +1), and 2.61 in -3 (CIR_Si, -3). The maximum sum o e he p io h ee yea s
is 226.00 (CIR_Si,∑3y). The se e i y measu e based on he Re ini i con o e sy sco e
anges om ze o o one wi h a mean o 0.33 in (CIR_Ri, ), 0.32 in + 1 (CIR_Ri, +1),
and 0.35 in -3 (CIR_Ri, -3). No ably, a ound one- hi d o he obse a ions show DJSI
lis ing (mean o dicho omous a iable DJSIi, = 0.33) and UNGC membe ship in
(mean = 0.31).
Table 4 p esen s he esul s o di e ence in means es s. Panel A p esen s he
de e minan s o GRI epo ing o epo ing s. non- epo ing i ms. The di e ence
is signi ican o all a iables excep ROEi, . The mean o all measu es o p io mis-
beha io is signi ican ly highe o GRI epo e s han non- epo e s. This indica es
ha CSR epo ing inc eases a e misbeha io , consis en wi h H1. Panel B shows
he de e minan s o u u e misbeha io o i ms wi h and wi hou misbeha io in
+ 1. The di e ence in means is signi ican o all a iables. The mean o CSRR_
Ii, and CSRR_Qi, is signi ican ly highe o i ms wi h u u e misbeha io han o
i ms wi hou u u e misbeha io . Consis en wi h H2, his indica es a posi i e ela-
ion be ween CSR epo ing and u u e misbeha io . Table5 shows signi ican co -
ela ions be ween ou measu es o CSR epo ing and p io (panel A) as well as
u u e co po a e misbeha io (panel B), consis en wi h H1 and H2.
Table 2 Sample i ms wi h/wi hou GRI epo s and u u e misbeha io pe yea
P esen s he dis ibu ion o sample i ms wi h and wi hou GRI epo s pe yea (CSRR_Ii, ) and he dis-
ibu ion o i ms wi h and wi hou a leas one case o misbeha io in he ollowing yea + 1 (CIR_Ii, +1)
Yea CSRR_Ii, CIR_Ii, +1
Repo e Non- epo e Misbeha io No misbeha io
2003 58 332 145 245
2004 88 306 184 210
2005 106 301 143 264
2006 119 283 132 270
2007 147 251 111 287
2008 195 216 100 311
2009 213 204 117 300
2010 228 187 130 285
2011 254 171 101 324
2012 263 160 100 323
2013 262 155 157 260
To al 1,933 2,566 1,420 3,079
1825
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 3 Desc ip i e s a is ics
Va iable N Mean SD Min Median Max
CSR epo ing
CSRR_Ii, 4,499 0.43 0.50 0.00 0.00 1.00
CSRR_I2i, 4,499 0.53 0.50 0.00 1.00 1.00
CSRR_Qi, 3,150 40.32 11.31 1.98 39.93 74.33
CSR epo ing his o y
P io GRI_Ii, 4,499 0.47 0.50 0.00 0.00 1.00
P io GRI_Ni, 4,499 1.84 2.68 0.00 0.00 14.00
CSRR_Ii, -3 4,499 0.28 0.45 0.00 0.00 1.00
Co po a e misbeha io
CIR_Ii, 4,499 0.31 0.46 0.00 0.00 1.00
CIR_Ni, 4,499 0.54 1.25 0.00 0.00 23.00
CIR_Si, 4,499 2.37 5.38 0.00 0.00 93.00
CIR_Ri, 3,986 0.33 0.37 0.00 0.16 1.00
Fu u e co po a e misbeha io
CIR_Ii, +14,499 0.32 0.46 0.00 0.00 1.00
CIR_Ni, +14,499 0.55 1.24 0.00 0.00 23.00
CIR_Si, +14,499 2.36 5.30 0.00 0.00 93.00
CIR_Ri, +13,982 0.32 0.36 0.00 0.16 1.00
P io co po a e misbeha io
CIR_Ii, -3 4,499 0.31 0.46 0.00 0.00 1.00
CIR_Ii,∑3y 4,499 0.94 1.14 0.00 0.00 3.00
CIR_Ni, -3 4,499 0.58 1.41 0.00 0.00 30.00
CIR_Ni,∑3y 4,499 1.71 3.45 0.00 0.00 44.00
CIR_Si, -3 4,499 2.61 5.97 0.00 0.00 93.00
CIR_Si,∑3y 4,499 7.68 15.84 0.00 2.00 226.00
CIR_Ri, -3 3,194 0.35 0.40 0.00 0.16 1.00
Fu he a iables
AGEi, 4,499 35.00 26.99 0.00 30.00 156.00
BRANDi, 4,499 0.14 0.35 0.00 0.00 1.00
CROSSi, 4,499 0.41 0.49 0.00 0.00 1.00
DJSIi, 4,499 0.33 0.47 0.00 0.00 1.00
DJSIi, -1 4,499 0.33 0.47 0.00 0.00 1.00
EXCHANGE 4,499 0.77 0.05 0.68 0.76 0.88
FFLOATi, 4,499 80.01 21.71 0.00 88.00 100.00
GLOBALi, 4,499 41.03 32.42 -73.47 39.48 689.70
GLOBALi, +14,499 41.53 30.70 -73.47 41.13 215.62
HOi4,499 3.98 0.38 3.18 4.17 4.96
LEGALi4,499 0.46 0.50 0.00 0.00 1.00
LEVi, 4,499 26.04 15.08 0.00 24.75 98.92
MSHAREi, 4,499 -7.26 7.55 -59.00 -4.89 -0.00
POi4,499 4.32 0.25 3.58 4.36 4.94
RDi, 4,487 712,730.20 1,583,548.00 0.00 1,029.00 1.36e + 07
1826
C.Rei maie e al.
6 Empi ical esul s
6.1 P io co po a e misbeha io andCSR epo ing
Table6p esen s he esul s o model (1), which es s H1 ( ha p io co po a e mis-
beha io and olun a y CSR epo ing a e posi i ely associa ed). Panel A shows
he esul s o a logis ic eg ession on CSRR_Ii, based on ou six measu es o
P io CIRi, -3/∑3y, displayed in column (A) ough (F). In all columns, he McFad-
den pseudo-R2 is abou 0.26, which is simila o ela ed s udies (e.g., Dhaliwal e al.
2011; Ch is ensen 2016) and indica es ha he eg ession is well-speci ied. The
goodness o i (GOF) es is no signi ican , indica ing ha he model i s he da a
well (Kleinbaum and Klein 2010). The a ea unde he ROC cu e is highe han 82
pe cen . The model classi ies abou 75 pe cen o epo ing and non- epo ing i ms
co ec ly and is highly signi ican (p < 0.01). O e all, he quali y o ou model speci-
ica ions is e y simila o ela ed s udies (B amme and Pa elin 2006; Legend e and
Code e 2013).30 All a iance in la ion ac o s (VIFs) a e below he conse a i e
h eshold o 5. Robus s anda d e o s con ol o he e oscedas ici y. Indus y and
yea ixed e ec s con ol o indus y and ime di e ences.
All measu es o p io misbeha io a e signi ican ly posi i ely ela ed o CSRR_
Ii, . The odds a ios indica e ha CSR epo ing is be ween 6 and 22 pe cen
mo e likely o i ms wi h p io misbeha io han o hei pee s (CIR_Ii, -3: 1.197
(p < 0.05), CIR_Ii,∑3y: 1.084 (p < 0.05), CIR_Ni, -3: 1.223 (p < 0.05), CIR_Ni,∑3y:
1.165 (p < 0.01), CIR_Si, -3: 1.096 (p < 0.05), CIR_Si,∑3y: 1.064 (p < 0.1)). The e ec
Table 3 (con inued)
Va iable N Mean SD Min Median Max
ROAi, 4,488 5.12 5.64 -38.96 4.38 62.35
ROEi, 4,499 15.24 64.94 -662.10 13.57 3,821.40
SIZEi, 4,499 16.86 1.15 9.32 16.85 20.04
SIZEi, +14,499 16.91 1.13 9.88 16.90 20.29
STATEi, 4,499 2.24 10.48 0.00 0.00 93.50
UNGCi, 4,499 0.31 0.46 0.00 0.00 1.00
UNGCi, +14,499 0.33 0.47 0.00 0.00 1.00
VOLAi, 4,499 26.06 8.14 6.65 25.01 73.02
P esen s he desc ip i e s a is ics o all a iables used in ou main and obus ness analyses o ou main
sample o Fo une Global 500 i ms o e he pe iod 2003–2013. Values o he numbe and se e i y o
cu en (CIR_Ni, , CIR_Si, ), u u e (CIR_Ni, +1, CIR_Si, +1), and p io misbeha io s (CIR_Ni, -3, CIR_
Ni,∑3y, CIR_Si, -3, CIR_Si,∑3) a e aw da a be o e log ans o ma ion. Since hese da a a e s ongly skewed,
hey a e log ans o med in ou empi ical es s (Tables4, 5, 6, 7and 8). All a iables a e de ined as in
Appendix3
30 Some s udies calcula e o he quali y measu es like he Nagelke ke R2 o he Cox & Snell R2. Values
o abou 0.41 and 0.30 o all ou model speci ica ions a e in an unc i ical ange.
1827
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 4 Di e ence in means es s
P esen s he di e ence in means es s. Panel A shows he di e ence in means es s o he de e minan s
o CSR epo ing (CSRR_Ii, ). Panel B shows he di e ence in means es s o he de e minan s o u u e
misbeha io (CIR_Ii, +1). Values o CIR_Ni, -3, CIR_Ni,∑3y, CIR_Si, -3, and CIR_Si,∑3y a e log ans o med
as in all empi ical es s. All a iables a e de ined as in Appendix3
* , **, *** indica e ha he es ima ed di e ences in means a e s a is ically signi ican a he 10 pe cen , 5
pe cen , and 1 pe cen le el, espec i ely, using a wo- ailed es
Panel A: Di e ence in means es s o he de e minan s o CSR epo ing (CSRR_Ii, = 0/1)
All GRI epo e Non- epo e
Va iable Mean Mean Mean Di . in means
CIR_Ii, -3 0.3136 0.3678 0.2728 -0.0950 -6.8341***
CIR_Ii,∑3y 0.9438 1.0879 0.8352 -0.2528 -7.4151***
CIR_Ni, -3 0.2932 0.3550 0.2466 -0.1083 -7.4254***
CIR_Ni,∑3y 0.6305 0.7344 0.5523 -0.1821 -7.9941***
CIR_Si, -3 0.7107 0.8306 0.6203 -0.2103 -7.4962***
CIR_Si,∑3y 1.3064 1.4653 1.1867 -0.2786 -7.4268***
DJSIi, 0.3350 0.4744 0.2300 -0.2445 -17.7888***
UNGCi, 0.3121 0.4904 0.1777 -0.3127 -23.7712***
AGEi, 34.9953 39.5820 31.5401 -8.0419 -10.0016***
CROSSi, 0.4072 0.4775 0.3542 -0.1232 -8.3924***
BRANDi, 0.1445 0.1945 0.1068 -0.0877 -8.3481***
MSHAREi, -7.2645 -7.6841 -6.9484 0.7357 3.2397***
STATEi, 2.2394 2.6453 1.9336 -0.7117 -2.2568**
LEGALi0.4614 0.5732 0.3772 -0.1960 -13.3036***
FFLOATi, 80.0113 83.3963 77.4614 -5.9349 -9.1604***
VOLAi, 26.0618 25.4587 26.5162 1.0575 4.3194***
LEVi, 26.0360 26.7908 25.4675 -1.3233 -2.9154***
SIZEi, 16.8573 17.1238 16.6565 -0.4673 -13.8026***
ROEi, 15.2401 14.0574 16.1310 2.0736 1.0603
Panel B: Di e ence in means es o he de e minan s o u u e co po a e misbeha io (CIR_Ii, +1 = 0/1)
All Fu u e
misbeha io No u u e
misbeha io
Va iable Mean Mean Mean Di . in means
CSRR_Ii, 0.4297 0.4908 0.4014 -0.0894 -5.6496***
CSRR_Qi, 40.3241 13.1035 39.1184 -3.9851 -9.2032***
UNGCi, +10.3343 0.3697 0.3180 -0.0518 -3.4240***
POi4.3190 4.3290 4.3144 -0.0146 -1.8455*
HOi3.9799 3.9367 3.9998 0.0631 5.2324***
GLOBALi, +141.5294 49.2299 37.9781 -11.2518 -11.5956***
CIR_Ii, -3 0.3136 0.5648 0.1977 -0.3670 -26.5095***
CIR_Ii,∑3y 0.9438 1.8183 0.5404 -1.2779 -40.9976***
CIR_Ni, -3 0.2932 0.5751 0.1631 -0.4120 -28.6581***
CIR_Ni,∑3y 0.6305 0.3637 1.2090 -0.8453 -40.3836***
CIR_Si, -3 0.7107 1.2714 0.4521 -0.8193 -29.8303***
CIR_Si,∑3y 1.3064 2.2136 0.8880 -1.3255 -37.8636***
SIZEi, +116.9086 17.4152 16.6750 -0.7402 -21.4503***
ROEi, 15.2401 17.6670 14.1208 -3.5462 -1.7027*
1834
C.Rei maie e al.
Table 6 (con inued)
ROEi, ? 0.001 1.45 0.001 1.25 0.001 1.37
Cons an -18.260 -3.78*** -13.760 -2.87*** -13.583 -2.82***
Indus y ixed e ec s Included Included Included
Yea ixed e ec s Included Included Included
N 3,150 3,150 3,150
Adjus ed R20.395 0.393 0.393
Highes VIF 2.45 2.48 2.52
Mean VIF 1.51 1.49 1.51
P esen s he eg ession esul s o model (1). Panel A uses CSRR_Ii, as he dependen a iable and hence e e s o he associa ion be ween p io co po a e misbeha io and
he olun a y issuance o a s and-alone GRI epo in . I shows he odds a ios, coe icien s, and z-s a is ics om pooled logis ic eg essions wi h indus y and yea ixed
e ec s. Columns (A) o (F) di e in he measu es o p io misbeha io : Column (A) includes CIR_Ii, -3. Column (B) includes CIR_Ii,∑3y. Column (C) includes CIR_Ni, -3.
Column (D) includes CIR_Ni,∑3y. Column (E) includes CIR_Si, -3. Column (F) includes CIR_Si,∑3y. Panel B uses CSRR_Qi, as dependen a iable and hence e e s o he
associa ion be ween p io co po a e misbeha io and CSR disclosu e quali y in . I shows he coe icien s and -s a is ics om OLS eg essions wi h indus y and yea
ixed e ec s. We do no un logis ic eg essions he e since, in con as o CSRR_Ii, , CSRR_Qi, is no bina y. As in panel (A), columns (A) o (F) di e in he measu es o
p io misbeha io . In bo h panels, alues o CIR_Ni, -3, CIR_Ni,∑3y, CIR_Si, -3, and CIR_Si,∑3y a e log ans o med as in all empi ical es s. All a iables a e de ined as in
Appendix3
*, **, *** indica e ha he es ima ed coe icien s a e s a is ically signi ican a he 10 pe cen , 5 pe cen , and 1 pe cen le el, espec i ely, using a wo- ailed es

1835
Co po a e esponsibili y andco po a e misbeha io : a e…
Bloombe g.32 O e all, ou indings suppo H1 and imply ha i ms use CSR epo ing
o signal ha hey ha e lea ned om pas mis akes, o , al e na i ely, ha i ms g een-
wash pas misbeha io o educe epu a ional damages acco ding o legi imacy heo y.
6.2 CSR epo ing and u u e co po a e misbeha io
Table 7 p esen s he esul s o model (2), which es s H2 on he ela ion o CSR
epo ing wi h u u e misbeha io .33 Panel A columns (A) h ough (F) show he esul s
o he decision o issue a CSR epo (CSRR_Ii, ) and ou h ee measu es o u u e
misbeha io CIRi, +1, wi h wo al e na i e p io pe iod measu es o misbeha io as
a con ol. The Du bin-Wu-Hausman es indica es endogenei y ela ed o CSRR_Ii,
(p < 0.01/0.05) in all cases. The cen e ed R2 is be ween 0.10 and 0.49. The models a e
highly signi ican (p < 0.01). AGEi, and FFLOATi, ul ill he c i e ion o ele ance,
as simple co ela ions (un abula ed) and mul i a ia e es s (Tables6and 7) indica e
signi ican ela ions o CSRR_Ii, .34 Rega ding exogenei y, AGEi, and FFLOATi, show
no economically meaning ul co ela ions wi h any measu e o Fu u eCIRi, +1 (be ween
0.03 and 0.08).35 The Kleinbe gen-Papp F s a is ic o he i s s age is be ween 46
and 48, which is abo e he c i ical alue o 19.93 o a model wi h one endogenous
eg esso and wo IVs (S ock and Yogo 2002; Bascle 2008).36 We can ejec weak
iden i ica ion and associa ed asymp o ic biases based on hese esul s (Bascle 2008)
and con i m he ele ance and sui abili y o he IVs. We ind a signi ican ly posi i e
associa ion o CSRR_Ii, wi h all measu es o Fu u eCIRi, +1 (p < 0.01/0.05), consis en
wi h H2 and legi imacy heo y. In all columns, we documen a posi i e associa ion
be ween pas and u u e misbeha io . Column (G) analyzes he inc emen al e ec o
he se e i y o p io misconduc in mo e de ail. We include he in e ac ion o CIR_
Ii,∑3y and CIR_Si,∑3y in model (2) wi h u u e misconduc CIR_Si, +1 as he depend-
en a iable. We con inue o ind a signi ican ly posi i e associa ion o CSRR_Ii, wi h
CIR_Si, +1 (0.263; p < 0.05). Mo eo e , while we ind ha he coe icien on CIR_Si,∑3y
is posi i e (0.352; p < 0.01), he coe icien on CIR_Ii,∑3y is nega i e (-0.256; p < 0.01),
which indica es ha p io inciden s o misbeha io in he absence o high se e i y a e
associa ed wi h less se e e u u e misconduc . Howe e , he in e ac ion is posi i ely
signi ican (0.114; p < 0.01), implying ha he se e i y o u u e misbeha io inc eases
in he equency o p io misbeha io when he inciden s a e mo e se e e.
32 This is in line wi h li e a u e on ESG a ing co e age, such as by Bloombe g, ha inds an inc ease in
CSR disclosu es o mo e-co e ed i ms (Bikme o a and Pi insky 2022).
33 We use he e oscedas ici y- obus s anda d e o s ( ) since he Pagan-Hall es is signi ican wi h
p < 0.01 and ejec s homoscedas ici y.
34 This also holds o he addi ional IVs DJSIi, -1 and CROSSi, .
36 We canno use he Ande son-Rubin o C agg-Donald es s (which also se e o alida e ele ance)
since we use he e oscedas ici y- obus s anda d e o s (e.g., Baum e al.2007; Cheng e al.2014).
35 Co ela ions o DJSIi, -1 and CROSSi, a e sligh ly highe (be ween 0.16 and 0.18), bu he C s a is ic
es o exogenei y o DJSIi, -1 and CROSSi, inds insigni ican esul s in ou addi ional analyses (p > 0.1,
un abula ed), suppo ing hei exogenei y (Bascle 2008). P io s udies ha e alida ed he exogenei y o
i m age (e.g., Ha jo o and Jo 2011; Ch is ensen 2016); hence, he p e equisi e o pe o m he C s a is ic
es o a leas one exogenous ins umen (Woold idge 2013) may be conside ed ul illed.
1836
C.Rei maie e al.
Table 7 CSR epo ing and u u e co po a e misbeha io (2SLS model)
Panel A: Volun a y issuance o a s and-alone GRI epo and u u e misbeha io
Fu u
eCIRi, +1 = β0+ β1CSRR_Ii, + β2UNGCi, +1 + β3POi+ β4HOi+ β5GLOBALi, +1 + β6P io CIRi, -3/∑3y + β7SIZEi, +1+ β8ROEi, + IND + YEAR + εi, +1 (2)
Column (A)Column (B)Column (C)Column (D)Column (E)Column (F)Column (G)
Dep. a
. 2nd s age CIR_I +1 CIR_I +1 CIR_N +1 CIR_N +1 CIR_S +1 CIR_S +1 CIR_S +1
Coe . zCoe .z Coe . zCoe .z Coe . zCoe .z Coe . z
CS
RR_Ii, (ins umen .) 0.4485.84 *** 0.2773.93 *** 0.324 4.56***0.170 2.57***0.509 3.99 ***0.251 2.17 ** 0.263 2.29**
UN
GCi, +1 -0.116 -4.67*** -0.078 -3.52*** -0.083 -3.58*** -0.053 -2.52**-0.145-3.54 ***-0.101-2.75 ***-0.102 -2.85***
PO
i0.1042.98*** 0.0571.84* 0.074 2.35** 0.0291.010.142 2.60***0.065 1.31 0.071 1.46
HO
i-0.081 -3.78*** -0.049 -2.53**-0.065 -3.34*** -0.028 -1.58-0.100-2.83 ***-0.024-0.74 -0.041 -1.33
GLOBAL
i, +1 0.0001.450.000 0.81 0.001 1.97** 0.0001.260.000 0.06 -0.000 -0.83-0.000 -0.97
CIR_
Ii, -30.25714.71***
CIR_
Ii,∑3y0.17726.71 ***-0.256 -9.17***
CIR_
Ni, -30.336 18.40***
CIR_
Ni,∑3y0.30326.84***
CIR_
Si, -30.44727.73***
CIR_
Si,∑3y0.42539.84***0.352 18.47***
CIR_
Ii,∑3y*CIR_Si,∑3y0.11411.39***
SIZE
i, +1 0.0698.65*** 0.0476.62*** 0.075 9.79***0.053 7.79***0.150 11.27*** 0.1089.14*** 0.095 8.24***
RO
Ei, 0.0000.410.000 0.65 0.000 0.29 0.0000.640.000 3.06***0.000 3.78***0.000 3.62***
Cons an
-1.162 -5.77*** -0.699 -3.89*** -1.230 -6.50*** -0.790 -4.61*** -2.491 -7.61*** -1.797 -6.14*** -1.468 -5.11***
Indus y ixed e ec
sIncluded IncludedIncludedIncludedIncludedIncludedIncluded
Ye
a ixed e ec sIncluded IncludedIncludedIncludedIncludedIncludedIncluded
N4
,499 4,4994,499 4,4994,499 4,49
94
,499
F s a i
s ic 39.69 *** 75.68 ***34.18 *** 51.34 *** 78.77 ***116.50 *** 131.04 ***
Cen e ed R²
0.0950.27540.22890.35830.35920.475
10
.4936
Highes
VIF9.739.879.7810.01 9.61 9.64 9.88
Mean VI
F2.342.362.352.382.352.3
62
.76
Du bin-W
u-Hausman
Chi2 (endog
enei y)
36.5096 ***14.988 *** 20.144 *** 5.917 ** 18.417 *** 5.083 ** 6.066 **
Kleinbe gen
-Paap Wald
k F s a is ic
(weak
iden i ica ion)
47.11746.28246.94045.70247.99047.62
64
6.345
Hans
en J s a is ic
(o e iden i ica ion
)
10.213 ** 7.973 ** 11.742 *** 8.843 ** 16.769 *** 11.174 ** 12.973 ***
1837
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 7 (con inued)
1s s age esul s wi h dependen a iable: CSRR_I
i,
Coe . Coe . Coe . Coe . Coe . Coe . Coe .
AGEi, 0.002 6.29***0.002 6.30 ***0.002 6.28 ***0.002 6.27***0.002 6.37 ***0.002 6.37 ***0.002 6.31***
FFLOATi, 0.001 2.42 ** 0.0012.40**0.001 2.42** 0.001 2.36** 0.001 2.42 ** 0.0012.42**0.001 2.44**
UNGCi, +1 0.21112.58***0.211 12.59*** 0.21012.55***0.210 12.55*** 0.21112.62***0.211 12.60*** 0.21112.59***
POi-0.191 -6.37*** -0.190 -6.35*** -0.192 -6.40*** -0.192 -6.40*** -0.190 -6.35*** -0.189 -6.33*** -0.189 -6.32***
HOi0.054 2.51** 0.0542.52**0.054 2.50 ** 0.055 2.57***0.054 2.52** 0.0542.50**0.052 2.41**
GLOBALi, +1 0.001 4.59***0.001 4.58 ***0.001 4.51 ***0.001 4.45***0.001 4.60 ***0.001 4.62 ***0.001 4.55 ***
CIR_Ii, -30.028 1.92 *
CIR_Ii,∑3y0.0101.59-0.012 0.84
CIR_Ni, -30.0322.24**
CIR_Ni,∑3y0.022 2.25**
CIR_Si, -30.013 1.71 *
CIR_Si,∑3y0.0061.05-0.008-0.86
CIR_Ii,∑3y*CIR_Si,∑3y0.0101.89*
SIZEi, +1 0.018 2.66***0.018 2.58 ***0.017 2.49 ** 0.016 2.31 ** 0.017 2.43 ** 0.0182.52**0.017 2.36**
ROEi, -0.000 -1.22-0.000-1.22 -0.000 -1.21-0.000 -1.19-0.000 -1.22-0.000-1.23 -0.000 -1.25
Cons an 0.700 3.82***0.699 3.80 ***0.726 3.95 ***0.734 3.96***0.716 3.87 ***0.696 3.75 ***0.727 3.91***
Indus y ixed e ec sIncluded IncludedIncludedIncludedIncludedIncludedIncluded
Yea ixed e ec sIncluded IncludedIncludedIncludedIncludedIncludedIncluded
Panel B: CSR disclosu
e quali y and u u e misbeha io
Fu u
eCIRi, +1 = β0+ β1CSRR_Qi, + β2UNGCi, +1+ β3POi+ β4HOi+ β5GLOBALi, +1 + β6P io CIRi, -3/∑3y + β7SIZEi, +1+ β8ROEi, + IND + YEAR + εi, +1 (2)
Column (A)Column (B)Column (C)Column (D)Column (E)Column (F)Column (G)
Dep. a
. 2nd s age CIR_I +1 CIR_I +1 CIR_N +1 CIR_N +1 CIR_S +1 CIR_S +1 CIR_S +1
Coe . zCoe .z Coe . zCoe .z Coe . zCoe .z Coe . z
CS
RR_Qi, (ins umen .) 0.0214.11*** 0.0153.37*** 0.0142.97*** 0.0081.88* 0.0313.72*** 0.0192.65*** 0.0212.97***
UN
GCi, +1 -0.045 -2.01**-0.039 -1.97**-0.036-1.81 *-0.033-1.85 *-0.076-2.03 ** -0.078 -2.33**-0.075-2.29 **
PO
i0.0932.28**0.079 2.22** 0.0411.170.025 0.82 0.1301.98**0.077 1.31 0.0841.46
HO
i-0.014 -0.51-0.012 -0.49-0.017-0.73 -0.010 -0.470.018 0.41 0.0340.910.025 0.66
GLOBAL
i, +1 -0.000 -0.22-0.000 -0.450.000 0.20 0.0000.09-0.001 -1.76* -0.001 -2.06**-0.001 -2.08**
CIR_
Ii, -30.25812.10***
CIR_
Ii,∑3y0.17621.19***-0.256 -7.20***
CIR_
Ni, -30.321 15.45 ***
CIR_
Ni,∑3y0.28821.73***
CIR_
Si, -30.417 21.79***
CIR_
Si,∑3y0.40832.03***0.316 14.19***
CIR_Ii,∑3y
*
CIR_Si,∑3y
0.1219.98***
1838
C.Rei maie e al.
Table 7 (con inued)
SIZE
i, +1 0.0241.320.010 0.61 0.0422.48**0.031 2.04** 0.0893.03*** 0.0622.42**0.041 1.65*
ROEi, -0.000 -1.93* -0.000 -0.84-0.000 -1.57-0.000 -0.320.000 1.41 0.000 3.07***0.000 3.06***
Cons an -1.240 -5.17*** -0.790 -3.68*** -1.085 -5.01*** -0.690 -3.53*** -2.851 -7.08*** -1.960 -5.48*** -1.623 -4.59***
Indus y ixed e ec sIncludedIncludedIncludedIncludedIncludedIncludedIncluded
Yea ixed e ec sIncludedIncludedIncludedIncludedIncludedIncludedIncluded
N3,150 3,150 3,150 3,150 3,1503,150 3,150
F s a is ic 332.84*** 263.54 *** 51.17 *** 62.42*** 98.48*** 125.63 ***138.93***
Cen e ed R² 0.1016 0.2678 0.2490 0.3673 0.3213 0.4559 0.4711
Highes VIF6.356.326.276.256.316.2
78
.13
Mean VIF2.452.472.452.472.462.4
83
.46
Du bin-Wu-Hausman
Chi2 (endogenei y)
18.385 *** 10.649 *** 8.728 *** 3.073 *15.108*** 7.437*** 9.333***
Kleinbe gen-Paap Wald
k F s a is ic (weak
iden i ica ion)
35.172 34.399 34.87533.72235.46834.58334.112
1
s s age esul s wi h dependen a iable: CSRR_Qi,
Coe . Coe . Coe . Coe . Coe . Coe . Coe .
AG
Ei, -0.014 -2.14**-0.014 -2.15**-0.014 -2.16**-0.015-2.17** -0.014 -2.05**-0.014-2.08** -0.014 -2.12**
FFLOA
Ti, 0.0848.36*** 0.0848.26*** 0.0848.32*** 0.0838.17*** 0.0848.41*** 0.0848.30*** 0.0838.23***
UN
GCi, +1 2.659 6.61***2.653 6.60***2.637 6.54***2.620 6.49***2.678 6.67***2.653 6.61***2.652 6.60***
PO
i-3.260 -4.10*** -3.242 -4.08*** -3.297 -4.14*** -3.277 -4.12*** -3.267 -4.11*** -3.281 -4.13*** -3.265 -4.10***
HO
i-2.740 -4.58*** -2.720 -4.54*** -2.744 -4.59*** -2.695 -4.50*** -2.746 -4.58*** -2.708 -4.51*** -2.710 -4.51***
GLOBAL
i, +1 0.0547.81*** 0.0537.72*** 0.0537.71*** 0.0537.61*** 0.0547.85*** 0.0547.76*** 0.0537.71***
CIR_
Ii, -30.848 2.26**
CIR_
Ii,∑3y0.354 2.09** 0.055 0.14
CIR_
Ni, -31.0022.94***
CIR_
Ni,∑3y0.681 2.77***
CIR_
Si, -30.3681.93*
CIR_
Si,∑3y0.3292.10**0.121 0.52
CIR_
Ii,∑3y*CIR_Si,∑3y0.0750.56
SIZE
i, +1 2.745 14.30*** 2.719 13.70***2.705 13.89***2.672 13.32***2.720 13.58*** 2.683 12.99***2.673 12.93***
RO
Ei, 0.0031.460.003 1.48 0.0031.490.003 1.51 0.0031.480.003 1.51 0.0031.50
Cons an
14.0322.78*** 14.2442.80*** 14.985 2.93***15.1742.95*** 14.5162.81*** 14.9522.88*** 15.086 2.89***
Indus y ixed e ec
sIncludedIncludedIncludedIncludedIncludedIncludedIncluded
Yea ixed e ec s
IncludedIncludedIncludedIncludedIncludedIncludedIncluded
1839
Co po a e esponsibili y andco po a e misbeha io : a e…
P esen s he i s and second s age coe icien s and -/z-s a is ics om a 2SLS eg ession o model (2) wi h indus y and yea ixed e ec s. Panel A uses CSRR_Ii, as CSR
epo ing a iable and hence e e s o he associa ion be ween he olun a y issuance o a s and-alone GRI epo in and u u e co po a e misbeha io in + 1. Columns
(A) o (G) di e in he measu e o p io and u u e misbeha io : Column (A) includes CIR_Ii, -3 and CIR_Ii, 1. Column (B) includes CIR_Ii,∑3y and CIR_Ii, +1. Column
(C) includes CIR_Ni, -3 and CIR_Ni, +1. Column (D) includes CIR_Ni,∑3y and CIR_Ni, +1. Column (E) includes CIR_Si, -3 and CIR_Si, +1. Column (F) includes CIR_Si,∑3y
and CIR_Si, +1. Column (G) includes CIR_Ii,∑3y, CIR_Si,∑3y, hei in e ac ion, and CIR_Si, +1. Panel B uses CSRR_Qi, as CSR epo ing a iable and hence e e s o he
associa ion be ween CSR disclosu e quali y in and u u e co po a e misbeha io in + 1. As in panel A, columns (A) o (G) di e in he measu es o p io and u u e mis-
beha io . In bo h panels, alues o CIR_Ni, -3, CIR_Ni,∑3y, CIR_Ni, +1, CIR_Si, -3, CIR_Si,∑3y, and CIR_Si, +1 a e log ans o med as in all empi ical es s. All a iables a e
de ined as in Appendix3
* , **, *** indica e ha he es ima ed coe icien s a e s a is ically signi ican a he 10 pe cen , 5 pe cen , and 1 pe cen le el, espec i ely, using a wo- ailed es
Table 7 (con inued)

1840
C.Rei maie e al.
Panel B shows he esul s o CSRR_Qi, . The endogenei y esul s a e simila o
CSRR_Ii, . In columns (A) h ough (G), we ind a signi ican ly posi i e associa ion
o CSRR_Qi, wi h all measu es o Fu u eCIRi, +1 (p < 0.01/0.1), consis en wi h H2
and legi imacy heo y. Volun a y CSR epo ing does no seem o os e beha io al
change and b eak he di ec ela ion be ween p io and u u e misbeha io . The in e -
ac ion esul s in column (G) a e la gely equi alen o panel A. O e all, he esul s
imply ha mo e ex ensi e olun a y CSR epo ing is no a signal o c edible CSR
commi men . Ins ead, mo e ex ensi e olun a y CSR epo ing is associa ed wi h an
inc ease in he occu ence, numbe , and se e i y o u u e misbeha io s; hence, i
likely se es as an ins umen o imp ession managemen and g eenwashing.
Table8 p esen s he esul s o he SR model, whe e we im ou lie s in all con in-
uous a iables a he 2.5 h and 97.5 h pe cen iles (panel A wi h CSRR_Ii, : N = 4,278,
panel B wi h CSRR_Qi, : N = 2,996) and de-mean V o ensu e ha i s mean is ze o,
consis en wi h he unde lying me hodology (Dong and Lewbel 2015; Bon emps and
Nauges 2016). The gene al Whi e (1980) es nega es he e oscedas ici y. We p esen
ma ginal e ec s since eg ession coe icien s in bina y models ha e less explana-
o y powe . The esul s a e in e en ially equi alen o he 2SLS esul s. All measu es
o Fu u eCIRi, +1 a e posi i ely associa ed wi h CSRR_Ii, (p < 0.01/0.05/0.1) and
CSRR_Qi, (p < 0.01/0.05). All SR model speci ica ions show adequa e es s a is ics.
7 Addi ional analyses and obus ness
In addi ional and obus ness analyses, we add ess conce ns ha ac o s such as a
i m’s CSR epo ing his o y, al e na i e IVs, a iable measu emen , model spec-
i ica ion, o sample composi ion may a ec ou esul s. All es s s ongly suppo
ou main in e ences. Table9 summa izes he esul s o models (1) and (2) o ou
measu es o p io and u u e misbeha io based on CIR_Ii, . Resul s wi h ou o he
measu es o misbeha io (based on CIR_Ni, , CIR_Si, ) a e in e en ially equi alen .
7.1 Repo ing his o y
Fi s , we add ess he conce n ha he i m’s his o y o CSR epo ing may a ec
ou esul s. Following G ange (1969), we include an addi ional a iable in model
(1) ha accoun s o p io CSR epo ing. We use ei he a dicho omous a iable ha
indica es whe he i m i has issued a GRI epo in any yea p io o (P io GRI_Ii, )
o a ca ego ical a iable ha coun s he numbe o p io GRI epo s be o e (P i-
o GRI_Ni, ). Fo bo h a ian s, we ind a signi ican ly posi i e associa ion wi h CSR
epo ing in (CSRR_Ii, o CSRR_Qi, ) (p < 0.01, un abula ed). As in ou main anal-
yses, we ind a signi ican ly posi i e associa ion be ween each measu e o p io mis-
beha io (CIR_Ii, -3, CIR_Ii,∑3y, CIR_Ni, -3, CIR_Ni,∑3y, CIR_Si, -3, o CIR_Si,∑3y) and
CSRR_Qi, (p < 0.01/0.05). Howe e , he e is no signi ican ela ion wi h CSRR_Ii, .
Subsequen o misbeha io , i ms wi h a (mo e expe ienced) his o y o CSR epo -
ing imp o e he quali y o hei epo s a he han inc ease hei likelihood o issu-
ing ano he CSR epo , which is no su p ising since hey ha e al eady been issuing
1841
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 8 CSR epo ing and u u e co po a e misbeha io (SR model)
Panel A: Volun a y issuance o a s and-alone GRI epo and u u e misbeha io
Fu u eC
IRi, +1 = β0+ β1CSRR_Ii, + β2UNGCi, +1 + β3POi+ β4HOi+ β5GLOBALi, +1+ β6P io CIRi, -3/∑3y+ β7SIZEi, +1 + β8ROEi, + β9EXCHANGE
+ IND +
YEAR + εi, +1 (4)
Column (A)Column (B)Column (C)Column (D)Column (E)Column (F)
Dep. a iable
2nd s ageCIR_I +1 CIR_I +1 CIR_N +1 CIR_N +1 CIR_S +1CIR_S +1
Coe . zMa g.
E ec zCoe .z Ma g.
E ec zCoe .z Ma g.
E ec zCoe .z Ma g.
E ec zCoe .Z Ma g.
E ec zCoe .z Ma g.
E ec z
CSR
R_Ii, (ins umen .)0.018 3.81***0.104 3.85***0.015 3.21***0.092 2.39**0.017 3.27***0.108 3.86***0.019 4.28***0.118 7.41***0.026 3.32*** 0.2043.99*** 0.0151.92* 0.1311.80*
UNGC
i, +1-0.004 -2.44**-0.021-3.31***-0.003-2.34** -0.021 -1.61-0.004-2.52** -0.026 -1.80* -0.005 -3.53*** -0.030 -3.15*** -0.010 -4.13*** -0.079 -4.83***-0.008-3.42***-0.072-3.02***
PO
i0.008 3.83***0.048 4.90***0.006 2.60***0.034 2.15**0.008 3.34***0.050 3.50***0.006 3.08***0.039 2.23** 0.0030.830.023 0.64 0.0010.420.013 0.36
HO
i-0.006 -4.16*** -0.033 -2.63*** -0.006 -4.39*** -0.036 -3.33*** -0.007 -4.38*** -0.042 -4.59*** -0.005 -3.87*** -0.031 -4.29*** -0.006 -2.49**-0.045 -2.56**-0.004-1.85*-0.036-1.78*
GLOBAL
i, +10.000 1.71*0.000 1.38 0.0001.140.000 0.66 0.0000.960.000 1.15 0.0000.440.000 0.41 0.0000.080.000 0.08 -0.000 -0.33-0.000-0.27
CIR_
Ii, -30.005 4.59***0.026 2.35**
CIR_
Ii,∑3y0.0049.41***0.024 4.30***
CIR_
Ni, -30.0087.43***0.049 5.19***
CIR_
Ni,∑3y 0.0069.87*** 0.0376.81***
CIR_
Si, -30.01113.51***0.089 6.40***
CIR_
Si,∑3y 0.01218.20***0.102 9.71***
SIZE
i, +10.002 3.52***0.010 1.95*0.001 2.51**0.007 1.57 0.0023.84***0.013 2.65*** 0.0013.14*** 0.0093.21*** 0.0556.73*** 0.0444.05*** 0.0055.77***0.041 5.91***
RO
Ei, -0.000 -0.42-0.000-0.11 0.0000.440.000 0.08 0.0000.440.000 0.30 0.0000.640.000 0.13 0.0000.560.000 0.31 0.0000.750.000 0.43
Cons
an -0.055 -4.46*** -0.310 -3.77*** -0.031 -2.61*** -0.191 -1.83* -0.060 -4.48*** -0.375 -3.94*** -0.049 -4.28*** -0.303 -3.17*** -0.102 -5.17*** -0.808-4.55***-0.083-4.30***-0.737-6.13***
Sp
ecial eg esso IncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded IncludedIncluded
Indus
y ixed e ec s IncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded
Ye
a ixed e ec s IncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded
872,4872,4872,4872,4872,4872,4872,4872,4872,4872,4872,4872,4
N
Ch
i21,075.14 ***1,253.10 ***1,026.23 ***1,168.12 ***900.43 ***1,119.98 ***
340.0440.0520.0030.0720.0720.0ESM
R
43.662.698.505.551.669.5amgi
S
Wh
i e es (homoskedas ici y)241.70 *** 220.54 *** 246.79 *** 222.65 ***237.41 ***230.21 ***
5.25.25.25.25.25.2le elgnimmi
T
010101010101selpmas pa s oo
B
Sa gan
s a is ic (o e id. es ic ions es )-346.252-354.041-329.438-270.521-487.05
4-
474.606
Basm
ann s a is ic (
o e id. es ic ions es
)-4,293.60-4,294.78 -4,290.86-4,278.61 4,309.19 -4,308.18
1842
C.Rei maie e al.
Table 8 (con inued)
Panel B: CSR disclosu e quali y and u u e misbeha io
Fu u eC
IRi, +1 = β0+ β1CSRR_Qi, + β2UNGCi, +1 + β3POi+ β4HOi+ β5GLOBALi, +1 + β6P io CIRi, -3/∑3y+ β7SIZEi, +1 + β8ROEi, + β9EXCHANGE
+ IND +
YEAR + εi, +1 (4)
Column (A)Column (B)Column (C)Column (D)Column (E)Column (F)
Dep. a iable
2nd s ageCIR_I +1 CIR_I +1 CIR_N +1 CIR_N +1 CIR_S +1CIR_S +1
Coe . zMa g.
E ec zCoe .z Ma g.
E ec zCoe .z Ma g.
E ec zCoe .z Ma g.
E ec zCoe .Z Ma g.
E ec zCoe .z Ma g.
E ec z
CS
RR_Qi, (ins umen .)0.001 6.73***0.005 9.79***0.001 6.09***0.005 8.37***0.002 6.26***0.005 5.30***0.002 6.31***0.005 4.26*** 0.0024.97*** 0.0135.67*** 0.0014.00***0.012 2.03**
UNGC
i, +1-0.002 -2.01**-0.008-1.21 -0.003 -2.16**-0.010-2.47** -0.004 -2.87*** -0.012 -2.21**-0.005-3.76***-0.017-1.78*-0.007-3.83***-0.057-3.96*** -0.006 -3.31*** -0.056 -4.16***
PO
i0.001 0.42 0.0030.270.004 4.52 0.0131.14-0.002-0.93 -0.007 -0.67-0.002-0.77 -0.006 -0.98-0.004-1.24 -0.033 -3.33*** -0.003 -1.13-0.034-0.70
HO
i-0.001 -0.84-0.004-0.29 -0.002 -1.39-0.008-0.93 -0.000 -0.13-0.001-0.08 -0.001 -0.62-0.004-0.51 0.0010.370.007 0.26 0.0020.940.020 0.41
GLOBAL
i, +1-0.000 -4.31***-0.000-3.61***-0.000-3.73***-0.000-3.01***-0.000-4.62***-0.000-8.79*** -0.000 -5.19*** -0.000 -5.35*** 0.000-3.06***-0.001 -3.48*** 0.000-2.85***-0.001-1.99**
CIR_
Ii, -30.004 3.87***0.013 1.07
CIR_
Ii,∑3y0.0036.54*** 0.0122.75***
CIR_
Ni, -30.0064.71***0.017 2.67***
CIR_
Ni,∑3y 0.0067.37*** 0.0203.66***
CIR_
Si, -30.0089.20*** 0.0643.70***
CIR_
Si,∑3y 0.00812.98***0.081 2.44**
SIZE
i, +1-0.002 -1.99**-0.005-0.67 -0.002 -2.35**-0.007-2.14** -0.002 -1.68* -0.005 -1.42-0.002-2.32** -0.007 -1.210.001 0.38 0.0040.330.000 0.32 0.004 0.11
RO
Ei, 0.000 0.57 -0.000 -0.11-0.000-0.41 -0.000 -0.16-0.000-0.24 -0.000 -0.07-0.000-0.16 -0.000 -0.05-0.000-0.27 -0.000 -0.09-0.000-0.15 -0.000 -0.02
Cons
an -0.074 -3.18***-0.141-0.92 -0.044 -3.26***-0.169-1.84*-0.047-2.89***-0.137-1.48 -0.037 -2.31**-0.118-1.48 -0.077 -3.61*** -0.600 -3.30***-0.069-3.64***-0.683-1.59
Sp
ecial eg esso Included Included IncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded
Indus
y ixed e ec sIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded
Ye
a ixed e ec sIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncludedIncluded
699,2699,2699,2699,2699,2699,2699,2699,2699,2699,2699,2699,2
N
Ch
i2622.97 *** 715.05 *** 564.50 ***638.46 ***539.93 ***674.04 ***
430.0930.030.030.0520.0620.0ESM
R
40.68.516.647.662.681.6amgi
S
Wh
i e es (homoskedas ici y)203.61 *** 210.28 *** 198.64 *** 211.76 *** 206.27 *** 207.70 ***
5.25.25.25.25.25.2le elgnimmi
T
010101010101selpmaspa s oo
B
Sa gan s a is
ic (o e id. es ic ions es )-231.035-202.186-173.809-153.478-375.77
4-
344.216
Basm
ann s a is ic (
o e id. es ic ions es
)-2,993.20 -2,984.61-2,980.32-2,973.48 -3,008.4
1-
3,006.17
1843
Co po a e esponsibili y andco po a e misbeha io : a e…
P esen s he coe icien s and ma ginal e ec s wi h z-s a is ics om a SR eg ession o model (4) wi h indus y and yea ixed e ec s. Panel A uses CSRR_Ii, as CSR
epo ing a iable and hence e e s o he associa ion be ween he olun a y issuance o a s and-alone GRI epo in and u u e co po a e misbeha io in + 1. Columns
(A) o (F) di e in he measu es o p io and u u e misbeha io : Column (A) includes CIR_Ii, -3 and CIR_Ii, +1. Column (B) includes CIR_Ii,∑3y and CIR_Ii, +1. Column (C)
includes CIR_Ni, -3 and CIR_Ni, +1. Column (D) includes CIR_Ni,∑3y and CIR_Ni, +1. Column (E) includes CIR_Si, -3 and CIR_Si, +1. Column (F) includes CIR_Si,∑3y and
CIR_Si, +1. Panel B uses CSRR_Qi, as CSR epo ing a iable and hence e e s o he associa ion be ween CSR disclosu e quali y in and u u e co po a e misbeha io in
+ 1. As in panel (A), columns (A) o (F) di e in he measu es o p io and u u e misbeha io . In bo h panels, alues o CIR_Ni, -3, CIR_Ni,∑3y, CIR_Ni, +1, CIR_Si, -3,
CIR_Si,∑3y, and CIR_Si, +1 a e log ans o med as in all empi ical es s. All con inuous a iables a e immed a he 2.5 h and 97.5 h pe cen ile, consis en wi h he unde ly-
ing me hodology (Dong and Lewbel 2015; Bon emps and Nauges 2016). S anda d e o s (un abula ed) o he ma ginal e ec s a he mean a e boo s apped. All a iables
a e de ined as in Appendix3
* , **, *** indica e ha he es ima ed coe icien s a e s a is ically signi ican a he 10 pe cen , 5 pe cen , and 1 pe cen le el, espec i ely, using a wo- ailed es
Table 8 (con inued)
1850
C.Rei maie e al.
highe public sc u iny on CSR and ela ed (mis-) epo ing due o he s onge egula-
ions in Eu ope. Mos speci ica ions o model (2) con i m a signi ican ly posi i e asso-
cia ion o CSR epo ing wi h u u e misbeha io (p < 0.01/0.05/0.1).44
7.6 Al e na i e sample composi ion: ex ended ime ame
Finally, we analyze a b oade sample o Fo une Global 500 i ms un il 2018.45 We com-
pa e ou main sample pe iod (2003–2013) o he pe iods 2014–2016 (a e he issuance
o he EU di ec i e and he In eg a ed Repo ing F amewo k) and 2017–2018 (a e
he e ec i e da e o he EU di ec i e). Fo 2014–2016, we ind a signi ican ly nega i e
associa ion be ween p io misbeha io and CSRR_Ii, in model (1) (p < 0.01/0.05/0.1).
Fo 2017–2018, he associa ion is insigni ican o CIR_I -3, CIR_I∑3y, and CIR_N∑3y
and signi ican ly nega i e o CIR_N -3, CIR_S -3, and CIR_S∑3y (p < 0.05/0.1, un abu-
la ed). In model (2), he associa ion o CSRR_Ii, wi h u u e misbeha io is insigni i-
can o CIR_Ii, -3, CIR_Ii,∑3y, and CIR_Ni, -3 and nega i ely signi ican o CIR_Ni,∑3y,
CIR_Si, -3, and CIR_Si,∑3y (p < 0.05/0.1, un abula ed) o 2014–2016 and insigni ican
o 2017–2018. Howe e , his ex ended ime pe iod is cha ac e ized by a ansi ion
om olun a y o manda o y CSR epo ing. We obse e in he da a collec ion p o-
cess ha many i ms ha e eplaced GRI-based s and-alone CSR epo s wi h in eg a ed
epo s o manda o y epo s which a e no cap u ed by ou measu e o CSRR_Ii, . The
CSRR_Ii, measu e is he e o e no sui able o his ime pe iod, and he esul s canno
be in e p e ed as e idence o a dec ease in misbeha io .
Fo CSRR_Qi, , we ind no signi ican associa ions wi h p io o u u e misbeha -
io in ei he ime pe iod. Again, he in e p e a ion o hese esul s equi es cau ion, as
many o ou sample i ms a e ope a ing in ju isdic ions ha a e a ec ed by he Eu o-
pean and o he ini ia i es o manda e he disclosu e o CSR- ela ed in o ma ion. The
i ms’ epo ing and hence ou measu e o CSRR_Qi, will he e o e be a ec ed by
hese egula ions. O e all, ou measu es o CSRRi, do no well e lec olun a y CSR
epo ing du ing he la e ime pe iods, which is why we excluded hese pe iods om
ou main analyses. Mo e u u e esea ch is needed o analyze he e ec s o manda o y
CSR epo ing on misbeha io in a clean se ing.
8 Quasi‑na u al expe imen
To u he alida e ou esul s, we use a quasi-na u al expe imen and analyze an
exogenous shock o ex ile- ela ed indus ies. In 2013, he Rana Plaza disas e in
Bangladesh led many i ms o sign an acco d o be e wo king condi ions. We
analyze he olun a y signa u e as an addi ional o m o p oclaiming CSR commi -
men , o shed mo e ligh on he ela ion be ween public CSR commi men and ac ual
45 A e excluding i ms no co e ed by he da abases used, he sample includes 632 i ms.
44 The associa ion is insigni ican when using he combina ions CSRR_Ii, and CIR_Ii,∑3y o CSRR_Qi,
and CIR_Si,∑3y.

1851
Co po a e esponsibili y andco po a e misbeha io : a e…
beha io .46 The gene al expec a ion is ha ca as ophes like he Rana Plaza disas e
lead o be e wo king condi ions in he ela ed indus ies, since public sc u iny ol-
lowing he e en s is high and i ms ha e incen i es o ix hei sho comings. We
compa e he i ms a ec ed by he disas e o a con ol g oup o una ec ed i ms
om o he indus ies in a di e ence-in-di e ences design. In addi ion, we di ide he
g oup o a ec ed i ms in o signa o ies and non-signa o ies o he acco d. Signal-
ing heo y implies a pa icula educ ion in misbeha io o signa o ies i he signa-
u e cons i u es a c edible signal o CSR commi men . Howe e , he esul s o ou
analyses abo e imply ha many i ms p oclaim CSR commi men o legi imiza ion
a he han signaling. Legi imacy heo y implies ha p io misbeha io inc eases he
likelihood o i ms signing he acco d and ha signa o ies do no educe, and may
e en inc ease, hei misbeha io a e wa ds. We hus expec ha signa o ies beha e
less esponsibly han non-signa o ies p e and pos acco d.47
We analyze 528 i ms in i e ICB subsec o s: Clo hing & Accesso ies (ICB code
3763), Foo wea (3765), Food Re aile s & Wholesale s (5337), Appa el Re aile s
(5371), and B oadline Re aile s (5373). The 176 i ms ha signed he acco d in
2013 (Bangladeshacco d 2013) a e concen a ed in hese i e ICB subsec o s, excep
o h ee i ms ha we exclude since hei main ac i i ies a e in inadequa ely ep e-
sen ed subsec o s (Food P oduc s (3577), Fu nishing (3726), and Toys (3747)). We
exclude 95 i ms ha signed he acco d a e 2013 o signed he ollow-up acco d
ha became e ec i e in 2018. We u he exclude 139 signa o ies and 171 non-sig-
na o ies ha a e no co e ed by he da abases used (Da as eam, Eikon, and CCRD).
The inal sample includes 34 signa o ies and 89 non-signa o ies. As in Paik e al.
(2017), he size o hese wo subsamples is ela i ely small bu su icien (Ba nes
2017). We analyze misbeha io om 2000 o 2018. As a con ol sample, we analyze
he Fo une Global 500 i ms desc ibed in Sec . 7.6 (632 i ms). Table10 shows
he esul s. Panel A p o ides he means and s anda d de ia ions o CIR_Ii, o he
pe iod p e (2000–2012) and pos he disas e (2014–2018) o he h ee samples.
Linea eg essions es o di e ences (panel B). Figu e1 illus a es he esul s.
46 In Ap il 2013, he Rana Plaza building in Dhaka collapsed, killing mo e han 1,100 people and badly
inju ing housands mo e (Bangladeshacco d 2019). This was an unexpec ed shock o he ma ke and
a ec ed ex ile- ela ed i ms all o e he wo ld ha commonly ou sou ce hei manu ac u ing in Bang-
ladesh o sa e cos s (Quinlan and Sheldon 2011; Ba nes 2017). In esponse o an inc ease in ex e nal
p essu e a e he disas e , ou sou cing i ms, ade unions, and wi nesses ini ia ed a sa e y ag eemen
( he Acco d on Fi e and Building Sa e y in Bangladesh) e ec i e in May 2013 – only one mon h a e
he disas e (Bangladeshacco d 2019). The aim was o ga he money and esou ces o imp o e sa e y and
wo king condi ions and os e CSR o e economic a ge s (Paik, Lee, and K umwiede 2017).
47 Ano he , mo e ecen inciden occu ed in Dhaka in 2017. An old hea ing boile exploded in a ac o y
despi e inspec ions in acco dance wi h he acco d (Bangladeshacco d 2017). The ac o y was a supplie
o se e al i ms ha had signed he acco d in 2013. Akba and Deegan (2021) highligh ha a main
eason o inc eased disclosu es on wo kplace sa e y a e he Rana Plaza disas e is o espond o ou side
p essu e. Sinko ics e al. (2016) u he ind ha ou side p essu e a e he disas e has led i ms o p io -
i ize measu able s anda ds implemen a ion abo e he needs o hei employees and socie y as a whole. In
addi ion, i ms seek g ea e e iciency in hei p ocesses o cope wi h he inc easing cos o compliance,
and neglec ha m ul side e ec s ha des oy social alue. O e all, his indica es ha imp o emen s a e
he disas e do no necessa ily add ess all dimensions o CSR.
1852
C.Rei maie e al.
Table 10 Quasi-na u al expe imen : Bangladesh Acco d
Panel A: Means (s anda d de ia ion) o CIR_Ii, by i m g oups and pe iods FIRMS TOTAL
F500 Signa o ies Non-signa o ies
POST P e shocke en 0.2807 0.3710 0.1780 0.2726
(0.4494) (0.4836) (0.3827) (0.4453)
N = 8,216 N = 442 N = 1,157 N = 9,815
Pos shock e en 0.2066 0.3882 0.2449 0.2193
(0.4050) (0.4888) (0.4305) (0.4139)
N = 3,160 N = 170 N = 445 N = 3,775
TOTAL 0.2601 0.3758 0.1966 0.2578
(0.4387) (0.4847) (0.3976) (0.4375)
N = 11,376 N = 612 N = 1,602 N = 13,590
1853
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 10 (con inued)
Panel B: Linea eg ession esul s o he dependen a iable CIR_Ii,
Wi hou de e minan s and con ol a iables
(N = 13,590) Wi h de e minan s and con ol a iables
(N = 8,966)
Va iable Coe icien Coe icien
POST -0.0740 -8.13*** -0.0633 -6.37***
FIRMS
 Signa o ies 0.0904 4.25*** 0.1015 3.26***
 Non-signa o ies -0.1026 -7.51*** 0.0222 0.97
POST * FIRMS
 F500 Signa o ies 0.0912 2.26** 0.0874 1.87*
 F500 Non-signa o ies 0.1409 5.44*** 0.0358 1.37
UNGCi, 0.0027 0.27
POi0.0495 2.93***
HOi-0.0620 -4.67***
GLOBALi, 0.0006 4.07***
CIR_Ii, -1 0.3847 39.57***
SIZEi, 0.0593 16.92***
ROEi, -0.0114 -1.79*
Cons an 0.2807 58.47*** -1.1928 -10.98***
Indus y ixed e ec s Included
P esen s he esul s o a quasi-na u al expe imen su ounding he Rana Plaza disas e . Panel A shows means and s anda d de ia ions o CIR_Ii, as well as he numbe o
es ima ions o each i m g oup and ime pe iod. Panel B shows he esul s o linea eg essions excluding and including he de e minan s and con ol a iables o model
(2). All a iables a e de ined as in Appendix3
* , **, *** indica e ha he es ima ed coe icien s a e s a is ically signi ican a he 10 pe cen , 5 pe cen , and 1 pe cen le els, espec i ely, using a wo- ailed es
1854
C.Rei maie e al.
O e all, CIR_Ii, has signi ican ly declined o he a e age i m in he o al sample
om p e o pos 2013. The coe icien o -0.0740 (p < 0.01) o POST indica es ha
he a e age una ec ed Fo une Global 500 i m has educed misbeha io by 0.0740
om p e o pos 2013 (0.2807 s. 0.2066). The mean o CIR_Ii, is highe o sig-
na o ies han o una ec ed i ms by 0.0904 p e 2013 (p < 0.01; 0.3710 s. 0.2807)
and by 0.1816 pos 2013 (p < 0.01; 0.3882 s. 0.2066). This di e ence has hus
inc eased by (0.1816–0.0904 =) 0.0912 om p e o pos 2013 (p < 0.05). CIR_Ii, o
non-signa o ies is lowe han ha o una ec ed i ms by 0.1026 p e 2013 (p < 0.01;
0.1780 s. 0.2807). Pos 2013, i is highe by 0.0383 (p < 0.1; 0.2449 s. 0.2066).
This di e ence has hus inc eased by (0.0383–0.1026 =) 0.1409 (p < 0.01). The
main in e ences hold a e including he de e minan s and con ol a iables o model
(2) and using indus y ixed e ec s.48 O e all, he esul s indica e ha misbeha io
is mo e likely o signa o ies compa ed o bo h non-signa o ies and una ec ed i ms
p e and pos 2013 and ha signa o ies do no educe hei misbeha io a e signing
he acco d. Fi ms seem o sign he acco d o g eenwash misbeha io a he han o
signal ue CSR commi men .
Fig. 1 Mean co po a e misbeha io p e and pos 2013 by i m g oups. Displays he mean co po a e mis-
beha io (CIR_Ii, ) p e and pos 2013 by i m g oups (c ea ed in STATA). On a e age, co po a e misbe-
ha io is signi ican ly lowe o he o e all sample o i ms and o he con ol g oup o una ec ed Fo -
une Global 500 i ms in he pe iod a e he inciden compa ed o he pe iod be o e he inciden . While
o bo h non-signa o ies and signa o ies co po a e misbeha io is signi ican ly highe a e he inciden ,
he a e age co po a e misbeha io emains signi ican ly highe o signa o ies compa ed o non-signa o-
ies as well as o he una ec ed Fo une Global 500 i ms
48 The di e ence in he mean o CIR_Ii, p e 2013 and he change o CIR_Ii, om p e o pos 2013 o
non-signa o ies compa ed o una ec ed i ms is no longe signi ican . All o he di e ences, pa icula ly
hose o signa o ies, emain signi ican . POi, HOi, GLOBALi, , CIR_Ii, -1, and SIZEi, a e signi ican a
p < 0.01; ROEi, is signi ican a p < 0.1. UNGCi, is no signi ican . We measu e p io misbeha io (CIR_
Ii, -1) in he yea be o e he dependen a iable CIR_Ii, in o de o con ol o he di ec ela ion be ween
pas and u u e misbeha io . We measu e all o he a iables a he same ime as CIR_Ii, , i.e., in (com-
pa able o model (2) whe e Fu u eCIRi, +1 and u he a iables a e measu ed a he same ime, + 1). We
do no use yea ixed e ec s since POST cap u es ime di e ences.
1855
Co po a e esponsibili y andco po a e misbeha io : a e…
9 Conclusion
We analyze co po a e misbeha io p e and pos olun a y CSR epo ing o
add ess he ecip ocal ela ion o p oclaimed and implemen ed CSR commi -
men o e ime. P io li e a u e on he ela ion be ween CSR epo ing and CSR
pe o mance o en agg ega es esponsible and i esponsible beha io s. To a oid
measu emen p oblems a ising om his agg ega ion, we measu e i esponsible
beha io only, bu ac oss all CSR dimensions and p eceding as well as sub-
sequen o CSR epo ing. We ind signi ican ly posi i e associa ions be ween
CSR epo ing ( he olun a y issuance o a GRI epo and CSR disclosu e
quali y) and bo h p io and u u e misbeha io (occu ence, numbe , and se e -
i y). This implies ha p io misbeha io inc eases he e o s expended o CSR
epo ing. Howe e , ou esul s imply ha CSR epo ing, e en o be e qual-
i y, is no an indica ion o imp o ed ope a ions bu is ela ed o inc eased u u e
misbeha io , bo h in e ms o numbe and se e i y.
We subs an ia e ou indings using he Rana Plaza disas e in Bangladesh in
2013 as a quasi-na u al expe imen . We analyze he olun a y decision o sign an
acco d o be e wo king condi ions as an addi ional o m o p oclaiming CSR
commi men . We ind ha misbeha io is mo e likely o signa o ies be o e and
a e he signa u e, compa ed o non-signa o ies and i ms om indus ies no
a ec ed by he exogenous shock. Signa o ies show no sign o imp o emen . Con-
sis en wi h legi imacy heo y, ou esul s imply ha olun a ily p oclaiming CSR
commi men is no a c edible signal o he la e bu se es imp ession manage-
men and g eenwashing pu poses.
An ine i able limi a ion o ou s udy is ha we canno en i ely ule ou endo-
genei y issues. S ill, mul iple app oaches ha add ess hese issues suppo ou
in e ences. Also, due o a lack o da a, we had o exclude some i ms om coun-
ies wi h appa en human igh s iola ions like China. Ou analyses ocus on he
pe iod un il 2013 when CSR epo ing was gene ally olun a y and be o e some
ju isdic ions had begun o manda e i . Whe eas we ind an inc ease in misbeha -
io subsequen o olun a y CSR epo ing du ing ou main sample pe iod, we
ind no such e idence o he ex ended ime pe iod o 2014–2018. Howe e , his
la e ime pe iod is cha ac e ized by a ansi ion o manda o y epo ing du ing
which ou measu es do no well e lec olun a y CSR epo ing, which is why
we excluded i om ou main analyses. Whe eas some ju isdic ions, like Eu ope,
ha e decided o manda e he disclosu e o CSR in o ma ion, o he s, like he US,
emain undecided abou whe he and how o manda e CSR epo ing. The esul s
om ou main sample pe iod a e una ec ed by CSR epo ing manda es and p o-
ide aluable insigh s o deba es o e wha olun a y CSR epo ing can and
canno achie e. Ne e heless, u u e esea ch is needed o in es iga e whe he
manda o y CSR epo ing is mo e e ec i e in al e ing i m beha io owa d
mo e esponsibili y han olun a y CSR epo ing.

1856
C.Rei maie e al.
Table 11 UNGC p inciples and co esponding CCRD (sub-)ca ego ies
UNGC ca ego ies CCRD ca ego ies CCRD subca ego ies Examples o misbeha io pe CCRD sub-
ca ego y
Human igh s
P inciple 1: Businesses should suppo and espec he p o-
ec ion o in e na ionally p oclaimed human igh s; and
P inciple 2: make su e ha hey a e no complici in human
igh s abuses
People Human igh s Racial disc imina ion, gende disc imina-
ion, dispossession, human igh s iola-
ions, sexual ha assmen , in imida ion,
iolence
Labo
P inciple 3: Businesses should uphold he eedom o
associa ion and he e ec i e ecogni ion o he igh o
collec i e ba gaining;
P inciple 4: he elimina ion o all o ms o o ced and
compulso y labou ;
P inciple 5: he e ec i e aboli ion o child labou ; and
P inciple 6: he elimina ion o disc imina ion in espec o
employmen and occupa ion
Wo ke s’ igh s Unnecessa y wo ke inju ies, po e y wages,
excessi e o e ime, o ced labou , poo
wo king condi ions
Supply chain managemen Poo supply chain policy, unsus ainable
sou cing, poo condi ions in supplie
companies
En i onmen
P inciple 7:Businesses should suppo a p ecau iona y
app oach o en i onmen al challenges;
P inciple 8: unde ake ini ia i es o p omo e g ea e en i-
onmen al esponsibili y; and
P inciple 9: encou age he de elopmen and di usion o
en i onmen ally iendly echnologies
En i onmen Clima e change Des uc ion o ain o es /de o es a ion,
elease o g eenhouse gas emissions
Pollu ion & oxics Ai pollu ion, wa e pollu ion, oil spills,
widesp ead use o pes icides
Habi a s & esou ces Habi a des uc ion, h ea s o endange ed
species (biodi e si y), illegal logging,
displacemen o local communi ies, wa e
deple ion
An i-co up ion
P inciple 10:Businesses should wo k agains co up ion in
all i s o ms, including ex o ion and b ibe y
Poli ics Poli ical ac i i ies Poli ical dona ions (e.g., agains en i on-
men al laws)
An i-social inance B ibe y, ax a oidance, p ice- ixing
Appendix1
1857
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 12 Misbeha io se e i y sco e
Panel A: Sco ing scheme
Sco ing E ec s Range & Magni ude
C i e ia Desc ip ion
1 poin limi ed e ec s Numbe o ca ego ies Only one ca ego y a ec ed
Impac Li le: only one pe son//li ing being a ec ed
Heal h se e i y Ra ely a di ec h ea o li e (no signi ican magni ude and consequences)
Social se e i y Typically esul s in li le incon eniences
En i onmen al se e i y Ra ely a di ec h ea o he en i onmen
2 poin s mino e ec s Numbe o ca ego ies One o wo ca ego ies a ec ed
Impac Mode a e: impac on a small g oup (≤ 500) o people/li ing beings
Heal h se e i y Tempo a y discom o , a ely a di ec h ea o li e
Social se e i y Typically esul s in an incon enience o daily li e
En i onmen al se e i y Ra ely a di ec h ea o he en i onmen
3 poin s mode a e e ec s Numbe o ca ego ies Mo e han one ca ego y a ec ed
Impac G ea : la ge g oup o people/li ing beings is a ec ed
Heal h se e i y Tempo a y inju ies, o en h ea ening o li e, some damage una oidable
Social se e i y Typically esul s in dis up ions o daily li e
En i onmen al se e i y O en h ea ening o he en i onmen , some damage una oidable
4 poin s majo e ec s Numbe o ca ego ies Mo e han wo ca ego ies a ec ed
Impac G ea : la ge g oup o people/li ing beings is a ec ed
Heal h se e i y Pe manen inju ies, ex ensi e damage o communi ies likely
Social se e i y Resul s in majo dis up ions o daily li e
En i onmen al se e i y Ex ensi e damage o he en i onmen likely
Appendix2
1858
C.Rei maie e al.
Table 12 (con inued)
5 poin s se e e e ec s Numbe o ca ego ies Se e al ca ego ies a ec ed
Impac G ea : la ge g oup o people/li ing beings is a ec ed
Heal h se e i y Li e sa ing ac ions needed, dea h occu s, ex ensi e and widesp ead se e e dam-
age o communi ies
Social se e i y Resul s in ex eme dis up ions o daily li e
En i onmen al se e i y Ex ensi e and widesp ead se e e damage o he en i onmen
Panel B: Examples
Ca ego y Sco e Examples
E 1 poin Poo oxic chemicals policy
2 poin s Small gas leak; unau ho ised elease o dus ; b each o license o using mo e wa e han pe mi ed
3 poin s Illegal discha ge o was e wa e , wa e con amina ion, i e pollu ion; emi ing illegal le els o ai pollu ion
4 poin s Des uc ion o ain o es , clea -cu land
5 poin s La ge-scale oil spill, ex eme pollu ion
S 1 poin Un ai dismissal o a wo ke
2 poin s Di e en ypes o disc imina ion (gende , ace, disabili y e c.)
3 poin s Wages below he subsis ence le el; excessi e o e ime
4 poin s ‘Sla e-like condi ions’; inc ease in landlessness
5 poin s Fac o y collapse killing wo ke s
G1 poin Excessi e di ec o s’ pay
2 poin s B ibe y scheme; co up ion
3 poin s Tax e asion; money launde ing schemes
4 poin s P o ound in il a ion o a na ional go e nmen
5 poin s (No cases iden i ied)
1859
Co po a e esponsibili y andco po a e misbeha io : a e…
Table 13 Lis o a iables
Va iable Measu emen Sou ce
CSR epo ing
CSRRi, A placeholde o one o he h ee CSR epo ing a iables (below) in yea
CSRR_Ii, Dicho omous a iable: 1 indica es he olun a y issuance o a s and-alone GRI epo by i m i in
yea ; 0 indica es no s and-alone GRI epo in yea
GRI da abase, Co po a eRegis e ,
i m websi es, in e ne
CSRR_I2i, Dicho omous a iable: 1 indica es he olun a y issuance o any kind o s and-alone CSR epo
by i m i in yea ; 0 indica es no s and-alone CSR epo in yea
Co po a eRegis e , i m websi es,
in e ne
CSRR_Qi, CSR disclosu e quali y measu ed by he Bloombe g’s ESG disclosu e sco e o i m i in yea (in
pe cen )
Bloombe g
CSR epo ing his o y
P io GRI_Ii, Dicho omous a iable: 1 indica es he olun a y issuance o a leas one s and-alone GRI epo by
i m i be o e ; 0 indica es no s and-alone GRI epo be o e
GRI da abase, Co po a eRegis e ,
i m websi es, in e ne
P io GRI_Ni, Ca ego ical a iable: coun s he numbe o GRI epo s issued by i m i be o e , i.e., i measu es
in how many yea s i m i had issued a GRI epo un il -1
GRI da abase, Co po a eRegis e ,
i m websi es, in e ne
P io CSRR_Qi, CSR disclosu e quali y measu ed by he Bloombe g’s ESG disclosu e sco e o i m i in he las
a ailable yea p io o -3, i.e. in -4 o , i CSRR_Qi, is no a ailable in -4, hen in -5 o , i
CSRR_Qi, is also no a ailable in -5, hen in -6, e c. (in pe cen )
Bloombe g
CSRR_Ii, -3 Dicho omous a iable: 1 indica es he olun a y issuance o a s and-alone GRI epo in yea
-3; 0 indica es no s and-alone GRI epo in yea -3
GRI da abase, Co po a eRegis e ,
i m websi es, in e ne
Co po a e misbeha io
CIRi, A placeholde o one o he ou a iables o co po a e i esponsibili y (below) in yea
Appendix3
1866
C.Rei maie e al.
Ch is ensen, H. B., L. Hail, and C. Leuz. 2021. Manda o y CSR and sus ainabili y epo ing: Economic
analysis and li e a u e e iew. Re iew o Accoun ing S udies 26(3): 1176–1248.
Ch is odoulou, D., and S. McLeay. 2014. The double en y cons ain , s uc u al modeling and economic
es ima ion. Con empo a y Accoun ing Resea ch 31(2): 609–628.
Cla kson, P. M., Y. Li, G. D. Richa dson, and F. P. Vas a i. 2008. Re isi ing he ela ion be ween en i-
onmen al pe o mance and en i onmen al disclosu e: An empi ical analysis. Accoun ing, O gani-
za ions and Socie y 33(4–5): 303–327.
Co mie , D., I. M. Go don, and M. Magnan. 2004. Co po a e en i onmen al disclosu e: Con as ing man-
agemen ’s pe cep ions wi h eali y. Jou nal o Business E hics 49(2): 143–165.
Co mie , D., M. Magnan, and B. VanVel ho en. 2005. En i onmen al disclosu e quali y in la ge Ge man
companies: Economic incen i es, public p essu es o ins i u ional condi ions? Eu opean Accoun -
ing Re iew 14(1): 3–39.
Co po a e C i ic Resea ch Da abase (CCRD). 2020b. Publica ions Re e enced by E hical Consume .
Re ie ed Ma ch 4, 2020, om h p:// www. co po a ec i ic. o g/ in o/ / publi ca io ns. aspx.
Co po a e C i ic Resea ch Da abase (CCRD). 2020a. Resea ch & Ra ings. How is Co po a e C i ic com-
piled? Re ie ed Ma ch 4, 2020, om h p:// www. co po a ec i ic. o g/ in o/ / compi led. aspx
Co po a e C i ic Resea ch Da abase (CCRD). 2020c. Gul o Mexico oil spill, habi a s, pollu ion and
GHG impac s. Re ie ed Janua y 5, 2020, om h p:// www. co po a ec i ic. o g/ abs ac s. aspx?%
20IDs= 545454.
Co po a e C i ic Resea ch Da abase (CCRD). 2021a. E hical Consume : Abou E hical Consume .
Re ie ed Sep embe 27, 2021, om h ps:// www. e hic alcon sume . o g/ abou - us.
Co po a e C i ic Resea ch Da abase (CCRD). 2021b. E hical Consume Resea ch & Consul ancy: Co -
po a e Resea ch Da abase. Re ie ed Sep embe 27, 2021, om h ps:// esea ch. e hic alcon sume .
o g/ co po a e- esea ch- da ab ase.
Co po a e C i ic Resea ch Da abase (CCRD). 2021c. E hical Consume Resea ch & Consul ancy: Ou
E hical Ra ings Sys em. Re ie ed Sep embe 27, 2021, om h ps:// esea ch. e hic alcon sume . o g/
co po a e- esea ch- da ab ase/ ou - e hic al- a in gs- sys em.
Co po a e C i ic Resea ch Da abase (CCRD). 2021d. E hical Consume Resea ch & Consul ancy: Pa -
ne ships. Re ie ed Sep embe 27, 2021, om h ps:// esea ch. e hic alcon sume . o g/ esea ch- consu
l ancy/ pa n e shi ps.
Co po a e C i ic Resea ch Da abase (CCRD). 2021e. E hical Consume Resea ch & Consul ancy: Global
Di ec o y o E hical Consump ion O ganisa ions. Re ie ed Sep embe 27, 2021, om h ps:// esea
ch. e hic alcon sume . o g/ esea ch- hub/ global- di ec o y- e hic al- consu mp ion- o gan isa i ons.
Co po a e C i ic Resea ch Da abase (CCRD). 2021 . E hical Consume : Ou E hical Ra ings. Re ie ed
Sep embe 27, 2021, om h ps:// www. e hic alcon sume . o g/ abou - us/ ou - e hic al- a in gs.
Da a, S., M. Iskanda -Da a, and A. Pa el. 1999. Bank moni o ing and he p icing o co po a e public
deb . Jou nal o Financial Economics 51(3): 435–449.
Da is, J. H., F. D. Schoo man, and L. Donaldson. 1997. Towa d a s ewa dship heo y o managemen .
The Academy o Managemen Re iew 22(1): 20–47.
Dechow, P. M. 2023. Unde s anding he sus ainabili y epo ing landscape and esea ch oppo uni ies in
accoun ing. The Accoun ing Re iew 98(5): 481–493.
DeTienne, K. B., and L. W. Lewis. 2005. The p agma ic and e hical ba ie s o co po a e social esponsi-
bili y disclosu e. The Nike case. Jou nal o Business E hics 60(4): 359–376.
DeVillie s, C. J., and D. Alexande . 2014. The ins i u ionalisa ion o co po a e social esponsibili y
epo ing. The B i ish Accoun ing Re iew 46(2): 198–212.
Dhaliwal, D. S., O. Z. Li, A. Tsang, and Y. G. Yang. 2011. Volun a y non inancial disclosu e and he cos
o equi y capi al: The ini ia ion o co po a e social esponsibili y epo ing. The Accoun ing Re iew
86(1): 59–100.
Dhaliwal, D. S., S. Radhak ishnan, A. Tsang, and Y. G. Yang. 2012. Non inancial disclosu e and ana-
lys o ecas accu acy: In e na ional e idence on co po a e social esponsibili y disclosu e. The
Accoun ing Re iew 87(3): 723–759.
Donaldson, L., and J. H. Da is. 1991. S ewa dship heo y o agency heo y: CEO go e nance and sha e-
holde e u ns. Aus alian Jou nal o Managemen 16(1): 49–64.
Donaldson, T., and T. W. Dun ee. 1994. Towa ds a uni ied concep ion o business e hics: In eg a i e
social con ac s heo y. The Academy o Managemen Re iew 19(2): 252–284.
Donaldson, T., and T. W. Dun ee. 1999. Ties ha bind: A social con ac s app oach o business e hics.
Ha a d Business School P ess.

1867
Co po a e esponsibili y andco po a e misbeha io : a e…
Dong, Y., and A. Lewbel. 2015. A simple es ima o o bina y choice models wi h endogenous eg es-
so s. Econome ic Re iews 34(1–2): 82–105.
Dowling, J., and J. P e e . 1975. O ganiza ional legi imacy: Social alues and o ganiza ional beha io .
The Paci ic Sociological Re iew 18(1): 122–136.
Downa , B., J. E ns be ge , S. Reichels ein, S. Schwenen, and A. Zaklan. 2021. The Impac o ca bon dis-
closu e manda es on emissions and inancial ope a ing pe o mance. Re iew o Accoun ing S udies
26(3): 1137–1175.
Dube, S., and C. Zhu. 2021. The disciplina y e ec o social media: E idence om i ms’ esponses o
Glassdoo e iews. Jou nal o Accoun ing Resea ch 59(5): 1783–1825.
Einwille , S. A., C. E. Ca oll, and K. Ko n. 2010. Unde Wha Condi ions Do he News Media In lu-
ence Co po a e Repu a ion? The Roles o Media Dependency and Need o O ien a ion. Co po a e
Repu a ion Re iew 12(4): 299–315.
Eu opean Commission (EC). 2021. Co po a e sus ainabili y epo ing. Re ie ed May 31, 2021, om
h ps:// ec. eu opa. eu/ in o/ busin ess- econo my- eu o/ compa ny- epo ing- and- audi ing/ compa ny- epo
ing/ non- inan cial- epo ing_ en.
Eu opean Union (EU). 2019. Regula ion (EU) 2019/2088 o he Eu opean Pa liamen and o he Coun-
cil o 27 No embe 2019 on sus ainabili y‐ ela ed disclosu es in he inancial se ices sec o .
Re ie ed Janua y 14, 2022, om h ps:// eu - lex. eu opa. eu/ eli/ eg/ 2019/ 2088/ oj? msclk id= 55 05
e62b1 8111e c851a 1b441 213b6 c1& locale= en.
Fama, E. F., and M. C. Jensen. 1983. Sepa a ion o owne ship and con ol. The Jou nal o Law & Eco-
nomics 26(2): 301–325.
Fa be , D. B. 2005. Res o ing us a e aud: Does co po a e go e nance ma e ? The Accoun ing
Re iew 80(2): 539–561.
Fiech e , P., J. -M. Hi z, and N. Lehmann. 2022. Real e ec s o a widesp ead CSR epo ing manda e:
E idence om he Eu opean Union’s CSR Di ec i e. Jou nal o Accoun ing Resea ch 60(4):
1499–1549.
Flamme , C. 2013. Co po a e social esponsibili y and sha eholde eac ion: The en i onmen al awa e-
ness o in es o s. Academy o Managemen Jou nal 56(3): 758–781.
F eundlieb, M., and F. Teu ebe g. 2013. Co po a e social esponsibili y epo ing – a ansna ional analy-
sis o online co po a e social esponsibili y epo s by ma ke -lis ed companies: Con en s and hei
e olu ion. In e na ional Jou nal o Inno a ion and Sus ainable De elopmen 7(1): 1–26.
F iedman, M. 1970. The social esponsibili y o business is o inc ease i s p o i s. The New Yo k Times
Magazine, 119(41): 122–126.
F ynas, J. G., and C. Yamahaki. 2016. Co po a e social esponsibili y: Re iew and oadmap o heo e ical
pe spec i es. Business E hics: A Eu opean Re iew 25(3): 258–285.
Game schlag, R., K. Moelle , and F. Ve bee en. 2011. De e minan s o olun a y CSR disclosu e: Empi -
ical e idence om Ge many. Re iew o Manage ial Science 5(2–3): 233–262.
Global Repo ing Ini ia i e (GRI). 2011. A new phase: The g ow h o sus ainabili y epo ing. GRI’s Yea
in Re iew 2010/11. Re ie ed Ma ch 4, 2019, om h ps:// www. globa l epo ing. o g/ esou celi
b a y/ GRI- Yea - In- Re iew- 2010- 2011. pd .
G ange , C.W.J. 1969. In es iga ing causal ela ions by econome ic models and c oss-spec al me hods.
Econome ica 37 (3): 424–438.
G ewal, J., G. D. Richa dson, and J. Wang. 2023. E ec s o Manda o y Ca bon Repo ing on Un ep e-
sen a i e En i onmen al Disclosu es. Re ie ed Augus 16, 2023, om h ps:// ss n. com/ abs ac =
41661 84.
Hacks on, D., and M. J. Milne. 1996. Some de e minan s o social and en i onmen al disclosu es in New
Zealand companies. Accoun ing, Audi ing & Accoun abili y Jou nal 9(1): 77–108.
Haddock, J. 2005. Consume in luence on in e ne -based co po a e communica ion o en i onmen al
ac i i ies: The UK ood sec o . B i ish Food Jou nal 107(10): 792–805.
Hahn, R., and M. Kuehnen. 2013. De e minan s o sus ainabili y epo ing: A e iew o esul s, ends,
heo y, and oppo uni ies in an expanding ield o esea ch. Jou nal o Cleane P oduc ion
59:5–21.
Hahn, R., and R. Luel s. 2014. Legi imizing nega i e aspec s in GRI-o ien ed sus ainabili y epo ing: A
quali a i e analysis o co po a e disclosu e s a egies. Jou nal o Business E hics 123(3): 401–420.
Hail, L., A. Tahoun, and C. Wang. 2018. Co po a e scandals and egula ion. Jou nal o Accoun ing
Resea ch 56(2): 617–671.
Ha jo o, M. A., and H. Jo. 2011. Co po a e go e nance and CSR nexus. Jou nal o Business E hics
100(1): 45–67.
1868
C.Rei maie e al.
Ha jo o, M. A., and H. Jo. 2015. Legal s. no ma i e CSR: Di e en ial impac on analys dispe sion,
s ock e u n ola ili y, cos o capi al, and i m alue. Jou nal o Business E hics 128(1): 1–20.
He, J. 2006. Pollu ion ha en hypo hesis and en i onmen al impac s o o eign di ec in es men : The
case o indus ial emission o sul u dioxide (SO2) in Chinese p o inces. Ecological Economics
60(1): 228–245.
He lin, F., and D. Wallace. 2017. The BP oil spill: Sha eholde weal h e ec s and en i onmen al dis-
closu es. Jou nal o Business Finance & Accoun ing 44(3–4): 337–374.
He malin, B. E. 2013. Leade ship and co po a e cul u e. In The handbook o o ganiza ional econom-
ics, ed. R. Gibbons and J. Robe s, 432–478. P ince on Uni e si y P ess.
Hoi, C. K., Q. Wu, and H. Zhang. 2013. Is co po a e social esponsibili y (CSR) associa ed wi h
ax a oidance? E idence om i esponsible CSR ac i i ies. The Accoun ing Re iew 88(6):
2025–2059.
Holde -Webb, L., J. R. Cohen, L. Na h, and D. Wood. 2009. The supply o co po a e social esponsi-
bili y disclosu es among U.S. i ms. Jou nal o Business E hics 84(4): 497–527.
House, R. J., P. J. Hanges, M. Ja idan, P. W. Do man, and V. Gup a. 2004. Cul u e, leade ship, and
o ganiza ions. The GLOBE s udy o 62 socie ies. SAGE Publica ions.
Huang, X., and L. Wa son. 2015. Co po a e social esponsibili y esea ch in accoun ing. Jou nal o
Accoun ing Li e a u e 34:1–16.
Huang, Q., Y. Li, M. Lin, and G. A. McB aye . 2022. Na u al disas e s, isk salience, and co po a e
ESG disclosu e. Jou nal o Co po a e Finance 72:102152.
Hughes, S. B., A. Ande son, and S. Golden. 2001. Co po a e en i onmen al disclosu es: A e hey use-
ul in de e mining en i onmen al pe o mance? Jou nal o Accoun ing and Public Policy 20(3):
217–240.
In e b and. 2017. Bes Global B ands. Re ie ed May 30, 2017, om h p:// in e b and. com/ bes -
b ands/ bes - global- b ands/ me ho dology/.
Ioannou, I., and G. Se a eim. (2017). The consequences o manda o y co po a e sus ainabili y epo -
ing.The Ox o d Handbook o Co po a e Social Responsibili y, Vol. 2, Ox o d Uni e si y P ess.
Jan adej, P., and P. Ken . 1999. Co po a e en i onmen al disclosu es in esponse o public awa eness
o he Ok Tedi Coppe Mine disas e : A legi imacy heo y pe spec i e. Accoun ing Resea ch
Jou nal 12(1): 72–88.
Ja idan, M. 2004. Pe o mance o ien a ion. In Cul u e, leade ship, and o ganiza ions The GLOBE
s udy o 62 socie ies, ed. R.J. House, P.J. Hanges, M. Ja idan, P.W. Do man, and V. Gup a,
239–281. SAGE Publica ions.
Johnson, S. A., H. E. Ryan J ., and Y. S. Tian. 2009. Manage ial incen i es and co po a e aud: The
sou ces o incen i es ma e . Re iew o Finance 13(1): 115–145.
Kabasakal, H., and M. Bodu . 2004. Humane o ien a ion in socie ies, o ganiza ions, and leade a ib-
u es. In Cul u e, leade ship, and o ganiza ions. The GLOBE s udy o 62 socie ies, ed. R.J.
House, P.J. Hanges, M. Ja idan, P.W. Do man, and V. Gup a, 564–601. SAGE Publica ions.
Kim, E. -H., and T. P. Lyon. 2011. S a egic en i onmen al disclosu e: E idence om he DOE’s ol-
un a y g eenhouse gas egis y. Jou nal o En i onmen al Economics and Managemen 61(3):
311–326.
Kim, J. -B., C. Wang, and F. Wu. 2023. The eal e ec s o isk disclosu e: E idence om clima e
change epo ing in 10-Ks. Re iew o Accoun ing S udies 28(4): 2271–2318.
Kleinbaum, D. G., and M. Klein. 2010. Logis ic eg ession. A sel -lea ning ex , 3 d ed. Sp inge .
Kolk, A., and P. Pe ego. 2010. De e minan s o he adop ion o sus ainabili y assu ance s a emen s:
An in e na ional in es iga ion. Business S a egy and he En i onmen 19(3): 182–198.
Kolk, A., S. Walhain, and S. VanDeWa e ingen. 2001. En i onmen al epo ing by he Fo une Global
250: Explo ing he in luence o na ionali y and sec o . Business S a egy and he En i onmen
10(1): 15–28.
Ko abe, M., and J. Y. Mu ay. 2004. Global sou cing s a egy and sus ainable compe i i e ad an age.
Indus ial Ma ke ing Managemen 33 (1): 7–14.
Ko chen, M., and J. J. Moon, 2012. Co po a e social esponsibili y o i esponsibili y. The B.E. Jou -
nal o Economic Analysis & Policy, 12(1), A icle 55.
KPMG. (1999). KPMG In e na ional Su ey o En i onmen al Repo ing 1999. Re ie ed Ma ch 4, 2020,
om h ps:// www. esea chga e. ne / publi ca ion/ 25479 6996_ KPMG_ In e na io nal_ su ey_ o _ en i
onmen al_ epo ing_ 1999.
1869
Co po a e esponsibili y andco po a e misbeha io : a e…
KPMG. 2002. KPMG In e na ional Su ey o Co po a e Sus ainabili y Repo ing 2002. Re ie ed Ma ch
4, 2020, om h ps:// www. esea chga e. ne / publi ca ion/ 25474 6739_ KPMG_ In e na io nal_ Su ey_
o _ Co po a e_ Sus a inabi li y_ Repo ing_ 2002.
KPMG. 2008. KPMG In e na ional Su ey o Co po a e Responsibili y Repo ing 2008. Re ie ed Oc o-
be 21, 2020, om h ps:// www. in. kpmg. com/ secu eda a/ aci/ iles/ sus c o po a e e spons ibili y ep
o in gsu ey2008. pd .
KPMG. 2011. KPMG In e na ional Su ey o Co po a e Responsibili y Repo ing 2011. Re ie ed Oc o-
be 21, 2020, om h ps:// asse s. kpmg/ con e n / dam/ kpmg/ pd / 2012/ 02/ Co po a e- espo nsibl i y-
epo ing- 2012- eng. pd .
KPMG. 2017. The KPMG Su ey o Co po a e Responsibili y Repo ing 2017. Re ie ed Oc obe 21,
2020, om h ps:// asse s. kpmg/ con e n / dam/ kpmg/ xx/ pd / 2017/ 10/ kpmg- su ey- o - co po a e-
espo nsibi li y- epo ing- 2017. pd .
KPMG 2022. KPMG Su ey o Sus ainabili y Repo ing 2022. Re ie ed Oc obe 10, 2022, om h ps://
asse s. kpmg. com/ con e n / dam/ kpmg/ sg/ pd / 2022/ 10/ ss - small- s eps- big- shi s. pd .
K üge , P. 2015. Co po a e goodness and sha eholde weal h. Jou nal o Financial Economics 115(2):
304–329.
Lambe , R., C. Leuz, and R. E. Ve ecchia. 2007. Accoun ing in o ma ion, disclosu e, and he cos o
capi al. Jou nal o Accoun ing Resea ch 45(2): 385–420.
La cke , D. F., and T. O. Rus icus. 2010. On he use o ins umen al a iables in accoun ing esea ch.
Jou nal o Accoun ing and Economics 49(3): 186–205.
La cke , D. F., S. A. Richa dson, and I. Tuna. 2007. Co po a e go e nance, accoun ing ou comes, and
o ganiza ional pe o mance. The Accoun ing Re iew 82(4): 963–1008.
La cke , D. F., L. Pomo ski, B. Tayan, and E. M. Wa s, 2022. ESG a ings – a compass wi hou di ec-
ion.Rock Cen e o Co po a e Go e nance a S an o d Uni e si y Wo king Pape Fo hcoming.
Re ie ed Augus 9, 2024 om h ps:// pape s. ss n. com/ sol3/ pape s. c m? abs ac _ id= 41796 47.
Lee, J., and S. Max ield. 2015. Doing Well by Repo ing Good: Repo ing Co po a e Responsibili y and
Co po a e Pe o mance. Business and Socie y Re iew 120(4): 577–606.
Legend e, S., and F. Code e. 2013. De e minan s o GRI G3 applica ion le els: The case o he Fo une
Global 500. Co po a e Social Responsibili y and En i onmen al Managemen 20(3): 182–192.
Lewbel, A. 2014. An o e iew o he special eg ession me hod. In The Ox o d handbook o applied non-
pa ame ic and semipa ame ic econome ics and s a is ics, ed. J.S. Racine, L. Su, and A. Ullah,
38–61. Ox o d Uni e si y P ess.
Lewbel, A., Y. Dong, and T. T. Yang. 2012. Compa ing ea u es o con enien es ima o s o bina y
choice models wi h endogenous eg esso s. Canadian Jou nal o Economics 45(3): 809–829.
Li, X., Y. Lou, and L. Zhang, 2023. Do comme cial ies in luence ESG a ings? E idence om Moody’s
and S&P. h ps:// pape s. ss n. com/ sol3/ pape s. c m? abs ac _ id= 41902 04.
Lindblom, C. K. 1994. The implica ions o o ganiza ional legi imacy o co po a e social pe o mance
and disclosu e. Con e ence Pape , C i ical Pe spec i es on Accoun ing Con e ence, New Yo k, NY.
Linnenluecke, M. K., and A. G i i hs. 2010. Co po a e sus ainabili y and o ganiza ional cul u e. Jou nal
o Wo ld Business 45(4): 357–366.
Mahoney, L. S., L. Tho ne, L. Cecil, and W. LaGo e. 2013. A esea ch no e on s andalone co po a e
social esponsibili y epo s: Signaling o g eenwashing? C i ical Pe spec i es on Accoun ing
24(4–5): 350–359.
Maignan, I., and D. A. Rals on. 2002. Co po a e social esponsibili y in Eu ope and he U.S.: Insigh s
om businesses’ sel -p esen a ions. Jou nal o In e na ional Business S udies 33(3): 497–514.
Ma k-Unge ich , B., and R. Weiskop . 2007. Filling he emp y shell. The public deba e on CSR in Aus ia
as a pa adigma ic example o a poli ical discou se. Jou nal o Business E hics 70(3): 285–297.
Ma sumu a, E. M., R. P akash, and S. C. Ve a-Munoz. 2014. Fi m- alue e ec s o ca bon emissions and
ca bon disclosu es. The Accoun ing Re iew 89(2): 695–724.
Ma ingly, J. E., and S. L. Be man. 2006. Measu emen o co po a e social ac ion – disco e ing axonomy
in he KLD a ings da a. Business & Socie y 45(1): 20–46.
McCa en, M., I. Diaz-Rainey, H. Robe s, and E. K. M. Tan. 2022. Poli ical connec ions, aci powe and
co po a e misconduc . Jou nal o Business Finance & Accoun ing 49(9–10): 1530–1552.
McGui e, J., S. Dow, and K. A gheyd. 2003. CEO incen i es and CSR. Jou nal o Business E hics 45(4):
341–359.
Meek, G. K., C. B. Robe s, and S. J. G ay. 1995. Fac o s in luencing olun a y annual epo disclosu es
by U.S., U.K. and Con inen al Eu opean mul ina ional co po a ions. Jou nal o In e na ional Busi-
ness S udies 26(3): 555–572.
1870
C.Rei maie e al.
Me kl-Da ies, D. M., and N. M. B ennan. 2007. Disc e iona y disclosu e s a egies in co po a e na -
a i es: Inc emen al in o ma ion o imp ession managemen ? Jou nal o Accoun ing Li e a u e
26:116–194.
Mishina, Y., B. J. Dykes, E. S. Block, and T. G. Pollock. 2010. Why “good” i ms do bad hings: The
e ec o high aspi a ion, high expec a ions, and p ominence on he inciden o co po a e illegali y.
The Academy o Managemen Jou nal 53(4): 701–722.
Mone a, J. M., P. A chel, and C. Co ea. 2006. GRI and he camou laging o co po a e unsus ainabili y.
Accoun ing Fo um 30(2): 121–137.
Mo is, R. D. 1987. Signalling, agency heo y and accoun ing policy choice. Accoun ing and Business
Resea ch 18(69): 47–56.
Mulle , A., and G. Whi eman. 2009. Explo ing he geog aphy o co po a e philan h opic disas e
esponse: A s udy o Fo une Global 500 i ms. Jou nal o Business E hics 84(4): 589–603.
Mu akin, M. B., and A. Khan. 2014. De e minan s o co po a e social disclosu e: Empi ical e idence
om Bangladesh. Ad ances in Accoun ing 30(1): 168–175.
Neu, D., H. Wa same, and K. Pedwell. 1998. Managing public imp essions: En i onmen al disclosu es in
annual epo s. Accoun ing, O ganiza ions and Socie y 23(3): 265–282.
No lo , C. 2009. Key cu ency compe i ion. The Eu o e sus he Dolla . Coope a ion and Con lic 44(4):
420–442.
Paik, G. H., B. Lee, and K. R. K umwiede. 2017. Co po a e social esponsibili y pe o mance and ou -
sou cing: The case o he Bangladesh agedy. Jou nal o In e na ional Accoun ing Resea ch 16(1):
59–79.
Pa en, D. M. 1991. Exposu e, legi imacy, and social disclosu e. Jou nal o Accoun ing and Public Policy
10(4): 297–308.
Pa en, D. M. 1992. In a-indus y en i onmen al disclosu es in esponse o he Alaskan oil spill: A no e
on legi imacy heo y. Accoun ing, O ganiza ions and Socie y 17(5): 471–475.
Pa en, D. M. 2002. The ela ion be ween en i onmen al pe o mance and en i onmen al disclosu e: A
esea ch no e. Accoun ing, O ganiza ions and Socie y 27(8): 763–773.
Plumlee, M., D. B own, R. M. Hayes, and R. S. Ma shall. 2015. Volun a y en i onmen al disclosu e
quali y and i m alue: Fu he e idence. Jou nal o Accoun ing and Public Policy 34(4): 336–361.
P encipe, A. 2004. P op ie a y cos s and de e minan s o olun a y segmen disclosu e: E idence om
I alian lis ed companies. Eu opean Accoun ing Re iew 13(2): 319–340.
Qianqian, D., and S. Rui. 2018. Pee pe o mance and ea nings managemen . Jou nal o Banking &
Finance 89:125–137.
Quinlan, M., and P. Sheldon. 2011. The en o cemen o minimum labo s anda ds in an e a o neo-libe al
globaliza ion: An o e iew. The Economic and Labo Rela ions Re iew 22(2): 5–31.
Raghunandan, A., and S. Rajgopal, 2023. Do socially esponsible i ms walk he alk? Jou nal o Law
and Economics, o hcoming.
Raghunandan, A., and S. Rajgopal. 2022. Do ESG unds make s akeholde - iendly in es men s? Re iew
o Accoun ing S udies 27(3): 822–863.
Rasche, A., and S. Waddock. 2014. Global sus ainabili y go e nance and he UN Global Compac : A
ejoinde o c i ics. Jou nal o Business E hics 122 (2): 209–216.
Re ini i . 2022. En i onmen al, social and Go e nance sco es om Re ini i . Re ie ed Sep embe 29,
2022, om h ps:// www. e in i i . com/ con e n / dam/ ma ke ing/ en_ us/ docum en s/ me ho dology/ e in
i i - esg- sco es- me ho dology. pd .
Reimann, F., J. Raue , and L. Kau mann. 2015. MNE subsidia ies’ s a egic commi men o CSR in
eme ging economies: The ole o adminis a i e dis ance, subsidia y size, and expe ience in he
hos coun y. Jou nal o Business E hics 132(4): 845–857.
Rei maie , C., and W. Schul ze. 2017. Enhanced business epo ing: Value ele ance and de e minan s o
alua ion- ela ed disclosu es. Jou nal o In ellec ual Capi al 18(4): 832–867.
Robe s, R. W. 1992. De e minan s o co po a e social esponsibili y disclosu e: An applica ion o s ake-
holde heo y. Accoun ing, O ganiza ions and Socie y 17(6): 595–612.
Schul ze, W., and R. T omme . 2012. The concep o en i onmen al pe o mance and i s measu emen in
empi ical s udies. Jou nal o Managemen Con ol 22:375–412.
Semeno a, N., and L. G. Hassel. 2015. On he alidi y o en i onmen al pe o mance me ics. Jou nal o
Business E hics 132(2): 249–258.
Shi, H., X. Zhang, and J. Zhou. 2018. C oss-lis ing and CSR pe o mance: E idence om AH sha es.
F on ie s o Business Resea ch in China 12(11): 126–140.
1871
Co po a e esponsibili y andco po a e misbeha io : a e…
Simne , R., A. Vans aelen, and W. Chua. 2009. Assu ance on sus ainabili y epo s: An in e na ional
compa ison. The Accoun ing Re iew 84(3): 937–967.
Sinko ics, N., S. Fe dous Hoque, and R. R. Sinko ics. 2016. Rana Plaza collapse a e ma h: A e CSR
compliance and audi ing p essu es e ec i e? Accoun ing, Audi ing & Accoun abili y Jou nal
29(4): 617–649.
Skinne , D. J. 1994. Why i ms olun a ily disclose bad news. Jou nal o Accoun ing Resea ch 32(1):
38–60.
Skouloudis, A., N. Jones, C. Malesios, and K. E angelinos. 2014. T ends and de e minan s o co po a e
non- inancial disclosu e in G eece. Jou nal o Cleane P oduc ion 68:174–188.
Spence, M. 1973. Job ma ke signaling. The Qua e ly Jou nal o Economics 87(3): 355–374.
S ock, J. H., and M. Yogo, 2002. Tes ing o weak ins umen s in linea IV eg ession. Na ional Bu eau
o Economic Resea ch Technical Wo king Pape Se ies 284. Re ie ed May 30, 2017, om h p://
www. nbe . o g/ pape s/ T0284.
S ock, J. H., and M. W. Wa son. 2012. In oduc ion o econome ics, 3 d ed. Pea son.
S ike, V. M., J. Gao, and P. Bansal. 2006. Being good while being bad: Social esponsibili y and he
in e na ional di e si ica ion o U.S i ms. Jou nal o In e na ional Business S udies 37(6): 850–862.
Taylo , M. S. 2005. Unbundling he pollu ion ha en hypo hesis. Ad ances in Economic Analysis & Policy
4(2): 1–28.
Tho ne, L., L. S. Mahoney, and G. Mane i. 2014. Mo i a ions o issuing s andalone CSR epo s: A su -
ey o Canadian i ms. Accoun ing, Audi ing & Accoun abili y Jou nal 27(4): 686–714.
Toma , S. 2023. G eenhouse Gas Disclosu e and Emissions Benchma king. Jou nal o Accoun ing
Resea ch 61(2): 451–492.
Topping, N. 2012. How does sus ainabili y disclosu e d i e beha io change? Jou nal o Applied Co po-
a e Finance 24(2): 45–48.
T e iño, L. K. 2005. Ou o ouch – The CEO’s ole in co po a e misbeha io . B ooklyn Law Re iew
70(4): 1195–1212.
Tschopp, D. J. 2005. Co po a e social esponsibili y: A compa ison be ween he Uni ed S a es and he
Eu opean Union. Co po a e Social Responsibili y and En i onmen al Managemen 12(1): 55–59.
Uni ed Na ions Global Compac (UNGC). 2020a. Uni ed Na ions Global Compac . The en p inciples
o he UN Global Compac . Re ie ed Oc obe 21, 2020, om h ps:// www. unglo balco mpac . o g/
wha - is- gc/ missi on/ p inc iples.
Uni ed Na ions Global Compac (UNGC). 2020b. Who we a e. Ou Pa icipan s. Re ie ed Oc obe 21,
2020, om h ps:// www. unglo balco mpac . o g/ wha - is- gc/ pa i cipan s.
Uni ed Na ions Global Compac (UNGC). 2020c. Repo ing on he SDGs. Re ie ed Oc obe 21, 2020,
om h ps:// www. unglo balco mpac . o g/ ake- ac ion/ ac ion/ sdg- epo ing.
VanMa ewijk, M. 2003. Concep s and de ini ions o CSR and co po a e sus ainabili y: Be ween agency
and communion. Jou nal o Business E hics 44(2–3): 95–105.
Wa son, A., P. Sh i es, and C. Ma s on. 2002. Volun a y disclosu e o accoun ing a ios in he UK. The
B i ish Accoun ing Re iew 34(4): 289–313.
Wel o d, R. 2004. Co po a e social esponsibili y in Eu ope and Asia. Jou nal o Co po a e Ci izenship
13:31–47.
Wel o d, R. 2005. Co po a e social esponsibili y in Eu ope, No h Ame ica and Asia. Jou nal o Co po-
a e Ci izenship 17:33–52.
Whi e, H. 1980. A he e oskedas ici y–consis en co a iance ma ix es ima o and a di ec es o he e -
oskedas ici y. Econome ica 48(4): 817–838.
Wicke , C., A. G. Sche e , and L. J. Spence. 2016. Walking and alking co po a e social esponsibili y:
Implica ions o i m size and o ganiza ional cos . Jou nal o Managemen S udies 53(7): 1169–1196.
Williams, O. F. 2014. The Uni ed Na ions Global Compac : Wha did i p omise? Jou nal o Business
E hics 122(2): 241–251.
Wiseman, J. 1982. An e alua ion o en i onmen al disclosu es made in co po a e annual epo s.
Accoun ing, O ganiza ions and Socie y 7(1): 53–63.
Woold idge, J. M. 2003. Solu ions manual and supplemen a y ma e ials o Econome ic analysis o
c oss sec ion and panel da a. MIT P ess.
Woold idge, J. M. 2013. In oduc o y econome ics. A mode n app oach, 5 h ed. Mason.
Wu, J. 2014. The an eceden s o co po a e social and en i onmen al i esponsibili y. Co po a e Social
Responsibili y and En i onmen al Managemen 21(5): 286–300.

1872
C.Rei maie e al.
Yoon, A. and Se a eim, G. 2020. S ock p ice eac ions o ESG news: The ole o ESG a ings and disa-
g eemen . h ps:// schol a spa ce. manoa. hawaii. edu/ se e / api/ co e/ bi s eams/ b3b44 b92- 706b- 4474-
aab2- a20a6 e51bd 2c/ con e n .Accessed 1 Jul 2024.
Young, S., and M. Ma ais. 2012. A mul i-le el pe spec i e o CSR epo ing: The implica ions o
na ional ins i u ions and indus y isk cha ac e is ics. Co po a e Go e nance: An In e na ional
Re iew 20(5): 432–450.
Zhao, M. 2012. CSR-based poli ical legi imacy s a egy: Managing he s a e by doing good in China and
Russia. Jou nal o Business E hics 111(4): 439–460.
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.