How does ESG rating disagreement influence analyst forecast dispersion?

Spi a, Robin
A icle
How does ESG a ing disag eemen in luence analys
o ecas dispe sion?
Junio Managemen Science (JUMS)
P o ided in Coope a ion wi h:
Junio Managemen Science e. V.
Sugges ed Ci a ion: Spi a, Robin (2024) : How does ESG a ing disag eemen in luence analys
o ecas dispe sion?, Junio Managemen Science (JUMS), ISSN 2942-1861, Junio Managemen
Science e. V., Planegg, Vol. 9, Iss. 3, pp. 1769-1804,
h ps://doi.o g/10.5282/jums/ 9i3pp1769-1804
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/305315
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/
Junio Managemen Science 9(3) (2024) 1769-1804
Junio Managemen Science
www.jums.academy
ISSN: 2942-1861
Edi o :
DOMINIK VAN AAKEN
Ad iso y Edi o ial Boa d:
FREDERIK AHLEMANN
JAN-PHILIPP AHRENS
THOMAS BAHLINGER
MARKUS BECKMANN
CHRISTOPH BODE
SULEIKA BORT
ROLF BRÜHL
KATRIN BURMEISTER-LAMP
CATHERINE CLEOPHAS
NILS CRASSELT
BENEDIKT DOWNAR
RALF ELSAS
KERSTIN FEHRE
MATTHIAS FINK
DAVID FLORYSIAK
GUNTHER FRIEDL
MARTIN FRIESL
FRANZ FUERST
WOLFGANG GÜTTEL
NINA KATRIN HANSEN
ANNE KATARINA HEIDER
CHRISTIAN HOFMANN
SVEN HÖRNER
KATJA HUTTER
LUTZ JOHANNING
STEPHAN KAISER
NADINE KAMMERLANDER
ALFRED KIESER
NATALIA KLIEWER
DODO ZU KNYPHAUSEN-AUFSESS
SABINE T. KÖSZEGI
ARJAN KOZICA
CHRISTIAN KOZIOL
MARTIN KREEB
TOBIAS KRETSCHMER
WERNER KUNZ
HANS-ULRICH KÜPPER
MICHAEL MEYER
JÜRGEN MÜHLBACHER
GORDON MÜLLER-SEITZ
J. PETER MURMANN
ANDREAS OSTERMAIER
BURKHARD PEDELL
MARCEL PROKOPCZUK
TANJA RABL
SASCHA RAITHEL
NICOLE RATZINGER-SAKEL
ASTRID REICHEL
KATJA ROST
THOMAS RUSSACK
FLORIAN SAHLING
MARKO SARSTEDT
ANDREAS G. SCHERER
STEFAN SCHMID
UTE SCHMIEL
CHRISTIAN SCHMITZ
MARTIN SCHNEIDER
MARKUS SCHOLZ
LARS SCHWEIZER
DAVID SEIDL
THORSTEN SELLHORN
STEFAN SEURING
VIOLETTA SPLITTER
ANDREAS SUCHANEK
TILL TALAULICAR
ANN TANK
ORESTIS TERZIDIS
ANJA TUSCHKE
MATTHIAS UHL
CHRISTINE VALLASTER
PATRICK VELTE
CHRISTIAN VÖGTLIN
STEPHAN WAGNER
BARBARA E. WEISSENBERGER
ISABELL M. WELPE
HANNES WINNER
THOMAS WRONA
THOMAS ZWICK
Volume 9, Issue 3, Sep embe 2024
JUNIOR
MANAGEMENT
SCIENCE
Anna Sme diagina,Los in T ansc ip ion: Expe imen al
Findings on E hnic and Age Biases in AI Sys ems
Jaqueline Domnick, Au hen ici y and B and Ac i ism –An
Empi ical Analysis
A hanasios Kons an inos Kallinikidis, The Employees’
En ep eneu ial Mindse : The In luence o Pe cei ed
Supe iso E o on he Employees’ En ep eneu ial
Passion
Leonie Böhm, Mo i a ions and Ou comes o he An i-
Consump ion P ac ice ‘S ooping’
Lukas Hilke, When Does Ma ke ing & Sales Collabo a ion
A ec he Pe cei ed Lead Quali y? –The
Mode a ing E ec s o IT Sys ems
Hannah F anziska Gundel, Accele a o Impac on Pee
Ne wo king - Examining he Fo ma ion, Use, and
De elopmen o In e -O ganiza ional Ne wo ks
Among Ea ly-S age S a -Ups
Anna Simon, De eloping and Main aining a S ong Co po a e
Cul u e, While Coping Wi h a Wo k o ce G owing
Signi ican ly: A Quali a i e Analysis onCo po a e
Cul u e De elopmen o Fas -G owing S a -Ups
Robin Spi a, How Does ESG Ra ing Disag eemen In luence
Analys Fo ecas Dispe sion?
Ch is oph Weebe , De elopmen o a Cos Op imal
P edic i e Main enance S a egy
Alexand a Hanna James, The Munich En ep eneu ial
Ecosys em in he Heal h Sec o : Cu en S a e and
Imp o emen A eas
1591
1609
1634
1665
1681
1700
1733
1769
1805
1836
Published by Junio Managemen Science e.V.
This is an Open Access a icle dis ibu ed unde he e ms o he CC-BY-4.0
(A ibu ion 4.0 In e na ional). Open Access unding p o ided by ZBW.
ISSN: 2942-1861
How Does ESG Ra ing Disag eemen In luence Analys Fo ecas Dispe sion?
Robin Spi a
Uni e si y o Bay eu h
Abs ac
The p ac ice o esponsible and sus ainable in es ing has led o he inco po a ion o en i onmen al, social and go e nance
(ESG) in o ma ion in o in es men decisions. The ole o ESG a ing agencies has been o acili a e decision-making by ag-
g ega ing uns uc u ed ESG in o ma ion in o a single a ing. Ma ke pa icipan s, such as inancial analys s, ely on hese
a ings as pa o hei esea ch. Howe e , ESG a ing agencies a ely ag ee in hei assessmen o a company’s ESG pe o -
mance, leading o di e gen ESG a ings. This pape uses an OLS eg ession model based on a la ge sample o i m da a o
in es iga e whe he ESG a ing agency disag eemen inc eases analys s’ o ecas dispe sion. I builds on p e ious esea ch
by Kimb ough e al. (2022). The esul s do no p o ide su icien e idence o suppo a signi ican ela ionship be ween ESG
disc epancies and analys o ecas dispe sion. This calls in o ques ion he impo ance o non- inancial ESG in o ma ion in
analys s’ assessmen o a company‘s inancial pe o mance.
Keywo ds: analys o ecas ; disag eemen ; ESG a ing agencies; ESG sco e; in e media ies
1. In oduc ion
In he las en yea s, he expanding p ac ice o sus ain-
able and esponsible in es ing has esul ed in he inco po a-
ion o en i onmen al, social, and go e nance (ESG) in o -
ma ion in o in es men decisions. An es ima ed US$ 35 il-
lion in asse s unde managemen a e now in es ed wi h ESG
in o ma ion in mind (Global Sus ainable In es men Alliance
(GSIA), 2021, p. 9). Meanwhile, he pa allel inc ease in de-
mand om s akeholde s o accu a e in o ma ion on i ms’
ESG pe o mance, has led o he o ma ion o ESG a ing
agencies. ESG a ing agencies a e hi d pa y in o ma ion in-
e media ies ha p o ide quan i a i e e alua ions o a i m’s
ESG pe o mance (Scale & Kelly, 2010, p. 71). The concep
o ESG pe o mance in en s o desc ibe how well a i m man-
ages i s ESG isks and oppo uni ies (MSCI, 2022b, p. 3).
The inal esul o his e alua ion is hen compiled in o an
ESG a ing sco e. In 2018 alone, in es o s spen $ 500 mil-
I wan o hank Jan Sei z, who was my ad iso a he Uni e si y o
Bay eu h, o gi ing me he oppo uni y o w i e my mas e hesis, always
ha ing g ea sugges ions, p o iding new and g ea ideas o imp o e my
hesis and gene ally suppo ing me h oughou he ime o w i ing.
lion on ESG a ings, highligh ing hei impo ance o guiding
in es men decisions (Gilbe , 2021).
Howe e , he e is conside able disag eemen abou wha
makes an in es men sus ainable and esponsible. ESG a ing
agencies a ely ag ee in hei assessmen o a i m’s ESG pe -
o mance. This is ema kable conside ing how o en c edi
a ing agencies align in hei assessmen (Sind eu & Ken ,
2018). Consequen ly, egula o s and he media ha e aised
conce ns abou whe he ESG a ings can e ec i ely guide in-
es men decisions (Ch is ensen e al., 2021, p. 147). I he e
is no ag eemen among a ing agencies, ESG a ings migh
mislead ma ke pa icipan s. In es o s need o unde s and
wha he me hodology chosen by ESG a ing agencies ac u-
ally measu es and why. O he wise, ESG a ings isk “c ea ing
a alse sense o con idence among in es o s who don’ eally
unde s and wha lies behind he numbe s – and he e o e
don’ eally unde s and wha hey’ e buying” (Allen, 2018).
One impo an g oup ha elies on ESG a ings a e inan-
cial analys s. Financial analys s a e p o essionals who pe -
o m inancial analyses on behal o hei clien s o help hem
make in es men decisions. To conduc hose analyses, i-
nancial analys s use a ous ypes o in o ma ion abou i ms,
DOI: h ps://doi.o g/10.5282/jums/ 9i3pp1769-1804
© The Au ho (s) 2024. Published by Junio Managemen Science.
This is an Open Access a icle dis ibu ed unde he e ms o he CC-BY-4.0
(A ibu ion 4.0 In e na ional). Open Access unding p o ided by ZBW.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041770
including ESG in o ma ion (Wansleben, 2012, p. 407-410).
Non- inancial ESG- ela ed in o ma ion a e aluable because
hey p o ide insigh s in o i m- ela ed isks and oppo uni-
ies. Bu , due o he inconsis encies in he way di e en i ms
epo ESG in o ma ion and a lack o s anda diza ion, inan-
cial analys s inc easingly ely on ESG a ing agencies o an-
alyze ESG in o ma ion (Ko san onis and Se a eim 2019, p.
53; Doyle 2018, p. 8 .).
O e he pas i e yea s, he e has been conside able
p og ess in he li e a u e on why ESG a ing agencies dis-
ag ee ha much. Fo ins ance, esea ch shows ha scope,
weigh ing and measu emen (Be g e al., 2022, p. 1335
.), he use o di e en da a impu a ion me hods (Ko san-
onis & Se a eim, 2019, p. 54), and g ea e ESG disclosu e
(Ch is ensen e al., 2021, p. 34 .) lead o g ea e ESG dis-
ag eemen .
Though, he e a e ew s udies ha examine he impac
o ESG a ing disag eemen on analys s’ o ecas dispe sion.
P e ious s udies ha e examined he ela ionship be ween
c edi a ings and analys o ecas dispe sion (A amo e
al., 2009, p. 101), o he empi ical associa ion be ween CSR
and in o ma ion asymme y (Cho e al., 2013, p. 81 .).
Dispe sion is o en in e p e ed as a measu e o unce ain y
and in o ma ion asymme y (Ba on e al., 2010, p. 333).
O he s udies ha e examined how manda o y ESG disclo-
su e a ec s he accu acy and dispe sion o analys s’ ea nings
o ecas s (K uege e al., 2021, p. 35), o he ela ionship
be ween ESG disag eemen and analys o ecas dispe sion
o US i ms (Kimb ough e al., 2022, p. 29 .). Howe e ,
no s udy has ye examined he ela ionship be ween ESG
disag eemen and analys o ecas dispe sion globally.
Wi h his hesis, I a emp o ill his esea ch gap by em-
pi ically in es iga ing he in luence o ESG a ing disag ee-
men on analys o ecas dispe sion in an in e na ional se -
ing. Fo ecas dispe sion migh e lec he amoun o in-
o ma ion commonly a ailable o analys s (Han & Man y,
2000, p. 119). When analys s sha e a common o ecas -
ing model and obse e he same i m-p o ided disclosu es
bu ha e di e en p i a e in o ma ion, hey will place less
weigh on hei p i a e in o ma ion as he in o ma i eness o
i m-p o ided disclosu e inc eases, dec easing o ecas dis-
pe sion. The mo e ESG- ela ed in o ma ion a i m is disclos-
ing, he lowe he dispe sion o analys s’ ea nings o ecas
should be (Lang & Lundholm, 1996, p. 471). Con a y, a
high dispe sion migh sugges a lack o public in o ma ion
and hence analys s ely mo e on hei own p i a e in o ma-
ion. Al e na i ely, g ea e dispe sion could also indica e less
ag eemen among analys s due o he inabili y o unwilling-
ness o some analys s o ully and objec i ely ga he and p o-
cess ESG- ela ed in o ma ion (Behn e al., 2008, p. 330). I
analys s ha e he same i m-p o ided and p i a e in o ma-
ion bu pu di e en weigh s on he componen s o i m-
p o ided disclosu e in o ecas ing ea nings, addi ional dis-
closu e may inc ease he dispe sion o analys o ecas s (Lang
& Lundholm, 1996, p. 471 .). I p edic ha he dispe sion
o analys s’ o ecas s is no due o a lack o ESG- ela ed dis-
closu e, bu a he due o disc epancies in he e alua ion o
ESG in o ma ion. Howe e , analys s o en ely on ESG a ing
agencies o make sense o ESG- ela ed in o ma ion. Ra ing
agencies ha di e in he scope, weigh ing, and measu e-
men o ESG- ela ed in o ma ion (Ko san onis & Se a eim,
2019, p. 53). Consequen ly, he disag eemen be ween ESG
a ing agencies should inc ease analys s’ o ecas dispe sion.
The emainde o his hesis is s uc u ed as ollows:
Chap e wo desc ibes he cha ac e is ics o inancial an-
alys s and hei p ac ices. Chap e h ee ocuses on he
in eg a ion o ESG c i e ia in o in es men decision making.
A e wa ds, chap e ou in es iga es he ESG a ing agencies
and hei disag eemen . Chap e i e de elops he hypo he-
sis o he associa ion be ween ESG a ing disag eemen and
analys o ecas dispe sion. A e ha , chap e six ou lines
he empi ical s udy. Chap e se en and eigh in e p e he
empi ical esul s. Chap e nine highligh s he limi a ions o
his s udy and u u e esea ch oppo uni ies. Finally, chap e
en concludes.
2. Financial Analys s
2.1. His o ical Backg ound o he P o ession
This chap e b ie ly in oduces he eade o he eme -
gence o inancial analys s as a p o ession. Be o e he wen i-
e h cen u y he p ac ice o inance was no ye associa ed wi h
p o essional s a us. Only a e ha he p o ession o inancial
analys eme ged (Wansleben, 2012, p. 408). A de ining mo-
men o he inancial analys p o ession was he in oduc ion
o he s ock icke in 1867 (P eda, 2006, p. 754). P io o i s
in oduc ion, p ice in o ma ion would be deli e ed by mes-
senge s and s ocks may ade using nume ous icke symbols
and some imes e en di e en p ices. Thus, he s ock icke
enabled ma ke pa icipan s o moni o i m p ices mo e e -
icien ly (Fishe , 2019). Wi h he in oduc ion o he s ock
icke , a subse o inancial analys s known as echnical an-
alys s eme ged. Technical analysis es s on he assump ion
o epe i i e p ice beha io han can be analyzed by ocus-
ing on ends in s ock p ices (Wansleben, 2012, p. 418 .).
The o he subse o inancial analys s known as undamen al
analys s eme ged much la e in he 1930s. Thei p edeces-
so s we e s a is icians and accoun an s in banks, no echni-
cal analys s. The eason o he la e appea ance o undamen-
al analys s was ha hey encoun e ed se ious obs acles, as
nei he i ms no inancial inside s sha ed in o ma ion abou
co po a e undamen als be o e 1929 (Kno -Ce ina, 2011, p.
429). Al hough analys s had de eloped p ac ices o in e p e
i ms be o e he 1930s, hey simply lacked eliable da a. This
changed wi h he 1933 and 1934 Ac in he US. While he
1933 Ac es ablished laws o new issuances, including egis-
a ion and disclosu e equi emen s, he 1934 Ac ocused on
annual, biennial, and e en - ela ed epo ing equi emen s
o aded i ms (Bens on, 1973, p. 133). The disclosed in o -
ma ion allowed undamen al analys s o accu a ely in e p e
i ms’ ea nings powe and alue (Jacobson, 1997, p. 25).
Equally impo an o he ise o he inancial analys p o-
ession was he ongoing inancializa ion o he US economy
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1771
and public. Du ing he 1950s sha e owne ship doubled (Ja-
cobson, 1997, p. 109). This su ge in s ock owne ship no
only c ea ed a demand o in es men ad ice, bu also os-
e ed public legi imacy o analys s. This can be seen as a
c i ical p ocess in which inancial analys s ul ima ely we e in
a posi ion o ask ques ions and execu i es had o answe (Ja-
cobson, 1997, p. 7). Ano he c i ical de elopmen has been
he de elopmen o he ce i ied inancial analys (CFA) ex-
amina ion as well as i s wo ldwide accep ance. The s anda d-
ized cu iculum p o ided a sou ce o legi imacy o he ana-
lys p ac ice. As Ke chum (1967, p. 35) poin s ou , knowl-
edge and i s applica ion builds he “keys one o a p o ession”.
(Wansleben, 2012, p. 411 .)
The nex chap e ocuses on he p ac ices o inancial
analysis and ea nings o ecas ing commonly used by inan-
cial analys s. The pu pose is o de elop an unde s anding
o how inancial analys s e alua e he pe o mance o i ms
and o show he eade wha ypes o in o ma ion a e used
in hei e alua ion.
2.2. Analys P ac ices
2.2.1. P ocess o Financial Analysis
Collec ing and O ganizing In o ma ion
Financial analysis desc ibes he p ocess o collec ing, p o-
cessing, and e alua ing undamen al in o ma ion abou i ms
and de i ing in es men ecommenda ions o clien s based
on he analysis. The e o e, he i s s ep is o collec and o -
ganize all ele an in o ma ion abou he i m.
The p ima y sou ce o in o ma ion is i m da a. Such as
inancial s a emen s, annual and qua e ly esul announce-
men s, p ess eleases, and o he ela ed news (Ba ke , 1998,
p. 10). Wi h hese sou ces o in o ma ion, hough, analys s
mus always be cau ious and ques ion he eliabili y o he
disclosed in o ma ion. A e all, i ms a e pu suing hei own
sel -in e es and may engage in c ea i e accoun ing, window
d essing o down igh alsi ica ion o hei books (B. G aham
& Dodd, 2009, p. 68). Besides ha , inancial analys s a end
analys con e ences, main ain in ensi e con ac wi h in es o
ela ions ep esen a i es, and isi co po a e headqua e s
and p oduc ion acili ies o ill he gaps le by disclosed i m
in o ma ion (Ma s, 1998, p.86-111). In addi ion o i m in-
o ma ion, analys s also d aw on o he sou ces o in o ma ion
o hei analysis. In p inciple, any kind o in o ma ion ha
can e en ually a ec u u e ma ke de elopmen s is ele an .
This can include all kinds o newspape s, business epo s,
books o s udies, o o he in o ma ion sou ces on mac oe-
conomic, poli ical and social ends. In addi ion, pe sonal
con ac s o sell-side analys s, ex e nal hink anks, i m ep-
esen a i es and people om academia as well as ex books
on inancial analysis play an impo an ole (Leins, 2018, p.
75-77). Hence, he e is a wide ange o inancial in o ma ion
sou ces ha analys s d aw on.
In he las i e yea s, non- inancial ESG in o ma ion has
become an inc easingly impo an sou ce o in o ma ion o
analys s. Acco ding o he CFA Ins i u e, 85% o hei mem-
be s now conside E, S, and/o G ac o s when making in-
es men decisions (CFA Ins i u e, 2020, p. 4). This change
is based on he iew ha in eg a ing ESG ac o s in o inan-
cial analysis allows o a mo e ho ough assessmen o bo h
idiosync a ic and ma ke -wide isk, as well as g ow h oppo -
uni ies, which can imp o e long- e m isk-adjus ed e u ns
(CFA Ins i u e 2020, p. 27, MSCI 2022b, p. 2). Financial
analys s d aw om a mix o in e nal and ex e nal ESG in-
o ma ion. On he one hand, hey e alua e ESG in o ma ion
published di ec ly by i ms in hei inancial and s a u o y e-
po ing. Howe e , he consis ency and compa abili y o ESG
in o ma ion om i ms is poo because egula ions on disclo-
su e and epo ing s anda ds a e s ill in de elopmen (CFA
Ins i u e, 2020, p. 37 .). On he o he hand, hey d aw on
ESG a ings om a ing agencies such as MSCI and Sus aina-
ly ics. 63% o inancial analys s use hem o hei i m anal-
ysis. S ill, a majo p oblem wi h hese a ings is ha hey
a y widely ac oss di e en a ing p o ide s. S a e S ee
Global Ad iso s epo s a co ela ion o only 0.53 be ween
he a ings o MSCI and Sus ainaly ics o i ms in he MSCI
Wo ld Index. These a ing disc epancies esul om di e -
ences in he collec ed da a, conduc ed esea ch, and models
used o gene a e a ings, including alua ion me hodologies
and weigh ing o a ious ESG in o ma ion. (CFA Ins i u e,
2020, p. 40)
Ye no all sou ces o in o ma ion a e equally aluable.
Ba ke (1998, p. 11) su eyed analys s abou hei p io i-
ized sou ces o in o ma ion. He inds ha pe sonal con ac s
a e pa icula ly impo an o analys s (see Table 11 in he
annex). By speaking o i m ep esen a i es, analys s seek o
gain in o ma ion ad an ages ha goes beyond he disclosed
in o ma ion. These can be, o example, cla i ica ions o i-
nancial s a emen no es, opinions on he i ms’ economic po-
si ioning ela i e o compe i o s o p ojec ions o nex qua -
e ly sales in a segmen . Ye , he s udy does no include
non- inancial in o ma ion. In addi ion o he sou ce o in-
o ma ion, he e a e ou in o ma ion a ibu es ha ma e
o inancial analys s. The in o ma ion i sel mus be ei he
imely, applicable, c edible, o o iginal o be o alue (see
Table 12 in he annex). Fi s , he imeliness o he in o ma-
ion ma e s. A e inancial analys s ha e analyzed a speci ic
piece o in o ma ion, and a widely accep ed in e p e a ion
has aken hold among pa icipan s in he inancial ma ke ,
he da a is deemed o be inco po a ed in o he p ice. Conse-
quen ly, he in o ma ion loses i s ele ance o inancial an-
alys s. (Leins, 2018, p. 78 .). Weekly newspape s, such as
he Economis , se e as a good example. By he ime he
inancial analys eads he newspape , he in o ma ion has
al eady been p iced in o a ew days. Consequen ly, weekly
magazines a e no eally use ul o he analys in e ms o he
imeliness o hei in o ma ion (Leins, 2018, p. 80). Second,
he applicabili y o he in o ma ion also plays an impo an
ole. Applicabili y in his con ex means he use ulness o
he in o ma ion o he ma ke o ecas s. A highly applica-
ble in o ma ion o en al eady con ains in o ma ion on how
i could in luence inancial ma ke s and links he in o ma-
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041772
ion o speci ic i ms, economic sec o s o ma ke egions.
This is e y help ul because iden i ying he po en ial impac
o in o ma ion on inancial ma ke s is one o he mos chal-
lenging asks o inancial analys s (Leins, 2018, p. 84 .).
An example o an applicable in o ma ion sou ce is Ba on’s.
The magazine e alua es ma ke ends and d aws up implica-
ions o i ms and indus ies. The hi d c i e ia inancial ana-
lys s use when e alua ing in o ma ion is c edibili y. C edible
sou ces help inancial analys s in c a ing in en i e na a i es
while concu en ly s eng hening hei posi ion as expe s in
inance. Academic esea ch, in pa icula , is o en conside ed
a highly c edible sou ce o in o ma ion (Leins, 2018, p. 88).
The ou h c i e ia is o iginali y. Analys s can p omo e hei
o ecas s as unique and in en i e i hey employ in o ma ion
ha has no al eady been used by o he analys s. An seem-
ing unique ma ke pe spec i e gi es in es o s he imp ession
ha hey ha e been p o ided esou ces o help hem na i-
ga e he unce ain ies o inancial ma ke s. This is impo an
because in es o s ha e many ways o assessing inancial ma -
ke da a. Wi hin his con ex , analys s mus gene a e unique
s a emen s o cap u e hei audience’s in e es . (Leins, 2018,
p. 91-94).
Fo ecas ing and Valua ion
A e collec ing and o ganizing all ele an in o ma ion,
he nex s ep o inancial analys s is o make p ojec ions and
o e alua e whe he a i m is a good o a bad in es men
based on i s cu en sha e p ice. Fo his, inancial analys s
need o e alua e a i m in e ms o i s unde lying in insic
alue. Acco ding o B. G aham and Dodd (1934) his in in-
sic alue “is unde s ood o be ha alue which is jus i ied
by he ac s, e.g., he asse s, ea nings, di idends, de ini i e
p ospec s, as dis inc , le us say, om ma ke quo a ions es-
ablished by a i icial manipula ion o dis o ed by psycholog-
ical excesses” (B. G aham & Dodd, 2009, p. 64). Financial
analys s es ima e he in insic alue o a i m a e e alua -
ing all ele an in o ma ion a hei disposal. In es o s can
p o i om hei e alua ion when he in insic alue de ia es
om he ma ke alue o a i m. This occasionally happens
because he p ice o he sha es is based on wha in es o s
belie e hose sha es a e wo h (Kolle e al., 2020, p. 80).
Ha ing said ha , inancial analysis is by na u e no an exac
science (B. G aham & Dodd, 2009, p. 61). Financial analys s
can only calcula e he in insic alue o a i m o he bes o
hei abili y and he knowledge a ailable o hem.
To calcula e in insic alue, inancial analys s need o
know how alue is c ea ed. The concep o alue has been
in oduced by Al ed Ma shall in 1890 and has p o en o be
bo h las ing in i s alidi y and di icul in i s applica ion. In
sho , he wo main d i e s o alue a e g ow h and e u n
on in es ed capi al (ROIC). G ow h can be achie ed ei he
o ganically h ough gene al ma ke expansion o by gaining
ela i e ma ke sha e, o ino ganically h ough me ge s and
acquisi ions (Kolle e al., 2020, p. 260). ROIC, by con as ,
is he esul o a compe i i e ad an age ha allows he i m
o ei he command p emium p ices o o enhance he e i-
ciency o i s p oduc ion p ocess (Kolle e al., 2020, p. 224).
Fi ms c ea e alue when hey g ow, and ea n a ROIC g ea e
han hei oppo uni y cos o capi al (Kolle e al., 2020, p.
53). Fi ms ha in es in e enue g ow h and imp o ing hei
ROIC will gene a e highe discoun ed alues o u u e cash
lows. Howe e , he e is one ca ea . G ow h alone is no
enough o ealize highe discoun ed u u e cash lows (see
Figu e 2 in he annex). In cases whe e he e u n on capi-
al is below he i m’s cos o capi al, highe g ow h ac ually
leads o a educ ion in he discoun ed alue o u u e cash
lows (Kolle e al., 2020, p. 94 .). Hence, i ms should y
o ind he combina ion o e enue g ow h and ROIC ha
p oduces he highes discoun ed alue o u u e cash lows.
Non- inancial ac o s such as ESG can also be a alue
d i e o i ms. Acco ding o Henisz e al. (2019), ESG c e-
a es alue in i e ways. Fi s , i acili a es e enue g ow h.
Regula o s a e mo e inclined o g an access, pe mi s and li-
censes o i ms wi h a s ong ESG posi ion. Hence c ea ing
new oppo uni ies o g ow h. Cus ome s a e also willing o
pay and addi ional 5% o a g een p oduc . Second, ESG e-
duces cos s. Among o he s, a s ong ESG posi ion can help
o inc ease esou ce e iciency and hus educe ope a ing ex-
penses such as aw-ma e ial cos s and he ue cos o wa e
o ca bon. Resou ce e iciency can boos ope a ing p o i s as
much as 60%. Thi d, ESG educes egula o y and legal in-
e en ions. A s ong ESG posi ion can educe a i m’s isk
o ha m ul s a e in e en ion. Acco ding o he s udy, one-
hi d o co po a e p o i s a e a isk om s a e in e en ions.
Fou h, a s ong ESG posi ion may boos employee p oduc i -
i y. I allows i ms o a ac and keep alen ed s a , boos em-
ployee mo i a ion by p o iding hem a sense o pu pose and
enhance o e all p oduc i i y. Fi h, ESG can imp o e long-
e m e u ns on in es men and capi al alloca ion. Fo ex-
ample, by alloca ing capi al o mo e sus ainable in es men
oppo uni ies, which educes he isk o u u e w i e downs
and di es men s (Henisz e al., 2019, p. 3-8).
The nex s ep o inancial analys s is o use one o a ious
alua ion me hods o es ima e he alue o a i m. The mos
commonly used alua ion me hod is he discoun ed cash low
(DCF) me hod. This me hod discoun s u u e cash lows by
he oppo uni y cos o capi al. The idea behind i is ha u-
u e cash lows a e wo h less because o he ime alue o
money and he iskiness o u u e cash lows and hus need
o be adjus ed (Kolle e al., 2020, p. 86). The discoun ed
p esen alue o u u e cash lows in his case ep esen s he
in insic alue o he i m. By cap u ing he u u e pe o -
mance o a i m in a single numbe , inancial analys s can
de e mine whe he a i m is unde alued o o e alued el-
a i e o i s ma ke p ice. They can also compa e di e en
i ms wi h each o he . The adi ional DCF me hod includes
only inancial numbe s. Bu , non- inancial ESG ac o s can
be in eg a ed in o he DCF me hod wi h li le e o . This
is because ESG ac o s a e o en ma e ial and in luence he
i m’s long- e m cash lows (Wild, 2017, p. 54 .). One sho -
coming o he DCF me hod, howe e , is ha each yea ’s cash
low p o ides li le in o ma ion abou he i m’s compe i i e
posi ion and economic pe o mance. Declining ee cash low

R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1773
may indica e ei he poo pe o mance o in es men in he
u u e (Kolle e al., 2020, p. 305).
Fo he DCF me hod o wo k, inancial analys s need o
make p ojec ions abou u u e cash lows. Ye , he u he
cash lows a e in he u u e, he less accu a e he p ojec ions
become (Asqui h & Weiss, 2016, p. 359). G aham and Dodd
poin ou his p oblem in hei book Secu i y Analysis. They
w i e, "some ma e s o i al signi icance, e.g., he de e mi-
na ion o he u u e p ospec s o an en e p ise, ha e ecei ed
li le space, because li le o de ini e alue can be said on he
subjec .” (B. G aham & Dodd, 1934, p. ii). This leads o he
p oblem o deciding how many yea s in o he u u e o o e-
cas and how de ailed he o ecas should be. Depending on
he du a ion o he o ecas , he inancial analys will a i e a
di e en DCFs. In addi ion, he e is also he p oblem o se -
ing app op ia e g ow h a es, in e es a es, axes, e c. Con-
sequen ly, calcula i e app oaches such as he DCF me hod
can ne e p oduce p ecise esul s. They a e always app oxi-
ma ions o he u u e which a e p one o e o s (Leins, 2018,
p. 72). To compensa e o hese unce ain ies, some inan-
cial analys s c ea e se e al cash low scena ios (Win o h e
al., 2010, p. 10). O he s adjus hei numbe s acco ding o
he analys consensus. S ill o he s ely on hei gu eeling
o weak he numbe s o hei liking (Leins 2018, p. 11 .
Wansleben 2012, p. 417 .). As a as ESG ac o s a e con-
ce ned, hey usually ha e an impac o e a longe pe iod o
ime. Assessing ESG ac o s and hei impac can he e o e
p o ide essen ial insigh s in o u u e alue d i e s and hus
imp o e long- e m o ecas ing capabili ies (Wild, 2017, p. 55
.).
In he pas , inancial analys s and in es o s used ea nings
a he han DCF o calcula e he in insic alue o a i m. To
use ea nings as a measu e o alue c ea ion is in p inciple
no a bad idea, since i ms ha c ea e alue o en also ha e
a ac i e ea nings and ea nings g ow h. Mo eo e , ea nings
equals cash low o e he li e ime o he i m (Kolle e al.,
2020, p. 195 .). Howe e , p ac i ione s ha e mo ed away
om his me hod. The eason o his is ha no all ea n-
ings c ea e alue. Ma gin imp o emen s ha come pu ely
om cos cu ing, e.g. esea ch and ma ke ing expenses, hu
alue c ea ing in he long e m (Kolle e al., 2020, p. 195).
Fu he mo e, ea nings can be accoun ing ic ion (B. G aham
& Dodd, 2009, p. xxx). Almos all i ms need o in es in
plan , equipmen , o wo king capi al. F ee cash low is wha ’s
le o in es o s once in es men s ha e been sub ac ed om
ea nings (Kolle e al., 2020, p. 92). Fo simplici y, inan-
cial analys s and academics ha e some imes assumed ha all
i ms ha e he same ROIC. I his we e he case, di e ences
in he i ms’ cash lows would only esul om di e ences in
g ow h, making ea nings g ow h a sui able measu e o di e -
en ia ion (Kolle e al., 2020, p. 87 .). Though, some imes
sho - e m ea nings a e he only eliable da a a ailable o i-
nancial analys s. In pa icula , when he unce ain y abou
he i m is so g ea ha he cash low canno be accu a ely
calcula ed. In his case, ea nings a e o g ea impo ance o
he inancial analys (Kolle e al., 2020, p. 204).
In addi ion o he DCF and ea nings me hod, he e a e
se e al o he alua ion me hods wo h men ioning. How-
e e , I will con ine mysel o alua ion mul iples and liqui-
da ion alue, because I conside hese o be he mos impo -
an . Valua ion mul iples assume ha simila asse s should
ade o a simila p ice. Fi ms in he same indus y and
wi h simila pe o mance should ade a he same mul iple.
The mos popula alua ion mul iple is he p ice- o-ea nings
(P/E) mul iple, which is simply he equi y alue o he i m
di ided by i s ne income (Kolle e al., 2020, p. 559). The
ad an age o hese mul iples is ha hey do no ace he p ob-
lem o inpu s based on es ima es, because only he ma ke
p ice and inancial s a emen s a e needed o he calcula-
ion (Wansleben, 2012, p. 416). One majo p oblem, ne -
e heless, is whe he he i ms a e compa able a all. This
equi es a close look a he inancial s a emen s. Fo exam-
ple, a i m wi h mo e deb ela i e o equi y should ade a
a lowe P/E a io han a i m wi h no deb , because mo e
deb means highe isk o sha eholde s and a highe cos
o equi y (Kolle e al., 2020, p. 559 .). Also, compa isons
o di e en a ios ac oss di e en indus ies and among di -
e en i ms migh be misleading. Ano he p oblem is ha
he ma ke alua ion migh be in la ed by a specula i e bub-
ble o es ima es o ea nings, book alue, and so o h can
be w ong (Wansleben, 2012, p. 416 .). Occasionally, DCF
and alua ion mul iples may be inapp op ia e. This is he
case, o ins ance, when he i m is expec ed o cease ope a-
ions. Then i makes mo e sense o use he liquida ion alue
(Asqui h & Weiss, 2016, p. 354). The choice o he igh
alua ion me hod he e o e depends on he ci cums ances o
he i m. In some cases, he use o se e al alua ion me hods
may e en ha e complemen a y bene i s.
In es men Recommenda ion
A e ha ing de e mined he alue o i m, inancial an-
alys s make in es men ecommenda ions o hei clien s
based on hei inancial analysis. To unde line hei e-
po s, inancial analys s use pe suasi e cha s, ables, and
illus a ions (Riles (2006,2011) in Leins (2018, p. 12)). An-
alys s’ ecommenda ions a e in luen ial, as is e iden om
he changes in he p ice o a i m’s s ock a e hei elease.
Especially i he ecommenda ions a e widely publicized
h ough he media o a e issued by analys s wi h high c e-
den ials (Secu i ies and Exchange Commission (SEC) 2010;
B own e al. 2009, p. 107). Ryan and Ta le (2004, p. 51)
ind ha analys ac i i ies such as issuing ea nings o ecas s
and in es men ecommenda ions a e associa ed wi h a 17%
change in he ma ke -adjus ed p ice o s ocks on he London
S ock Exchange.
In gene al, one can dis inguish be ween i e di e en
ypes o in es men ecommenda ions. These a e Sell, Un-
de pe o m, Hold, Buy and S ong Buy, whe eby he in-
be ween le els Unde pe o m and Buy indica e a weake
con ic ion o he analys . Tha said, no all ecommenda-
ions ca y he same weigh . In ac , ecommenda ion ha e
o be assessed ela i e o he analys ’s p e ious ecommen-
da ion and he consensus opinion. Fo example, i a inancial
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041774
analys jus ei e a es he same a ing, i ca ies less weigh .
O , i he analys me ely issues a ecommenda ion in line wi h
he consensus. Con a y, i he inancial analys s eleases a
ecommenda ion ou o line wi h consensus, i ca ies mo e
weigh , because he analys s ands aside om he sa e y o
he he d and akes a g ea e epu a ional isk (B own e al.,
2009, p. 92).
Ul ima ely, howe e , i is he cus ome who decides wha
o do wi h he in o ma ion. The analys s’ epo only p o ides
in o ma ion ega ding he cos o bene i o in es ing in a ce -
ain s ock. Whe he he cus ome can ul ima ely p o i om
his in o ma ion is, ne e heless, an open ques ion. A e all,
inancial analys s o en ha e a con lic o in e es when i
comes o hei ecommenda ion. Cus ome s should he e o e
c i ically sc u inize and compa e he in o ma ion (Secu i ies
and Exchange Commission (SEC), 2010). Acco ding o Win-
o h e al. (2010), he mo e sophis ica ed inancial clien s a e
mo e in e es ed in discussing ac s, unde lying assump ions
and a gumen s han in ecommenda ions hemsel es. This is
because ins i u ional in es o s ypically use in o ma ion om
se e al analys s, compa ing hei assessmen s (Win o h e al.,
2010, p. 10 .).
The nex chap e ocuses on he p ac ice o ea nings o e-
cas ing and o ecas dispe sion. The aim o his chap e is
o show he eade he di e ences be ween inancial analy-
sis and ea nings o ecas s. I also aims o build a heo e ical
ounda ion o he dependen a iable o his mas e hesis.
2.2.2. P ac ice o Ea nings Fo ecas ing
Fo ecas Es ima es
Ea nings o ecas s a e ubiqui ous in oday’s inancial
ma ke s. In es o s ely hea ily on ea nings o ecas s when
making in es men decisions (Gi oly & Lakonishok, 1980, p.
221). Gi oly and Lakonishok (1983) men ion ha „Ea nings
pe sha e eme ge o m a ious s udies as he single mos im-
po an accoun a iable in he eyes o he in es o s“ (Gi oly
and Lakonishok (1983) in Jennings (1985, p. 1)). This iew
con adic s sha ply wi h he no ion ha he alue o a i m is
equal o i s discoun ed long- e m cash lows. Howe e , due
o he unp edic abili y o u u e cash lows, p ac i ione s use
ea nings as a easonable p oxy o DCF. Accoun ing ea nings
a e well de ined, and public i ms’ ea nings s a emen s a e
subjec o ho ough audi s be o e hey a e published. As a
esul , in es o s conside ea nings o be ai ly eliable and
con enien measu e o alue public i ms (McClu e, 2022).
The economic impo ance o ea nings o ecas s can also
be seen in he amoun o esou ces de o ed o he p epa-
a ion and analysis o such in o ma ion by he in es men
communi y. La ge b oke age i ms employ la ge amoun s o
inancial analys s o p oduce ea nings o ecas s. These sell-
side analys s dissemina e hei in o ma ion o o he ma ke
pa icipan . In doing so, he b oke age i ms hope o ea n
ading commissions. As a esul , buy-side analys s ace he
po en ial con lic s o wo king o in es men banking i ms
and he need o gene a e commissions. In addi ion o sell-
side analys s, he e a e also buy-side analys s and indepen-
den analys s who p epa e ea nings o ecas s. Independen
analys s p o ide hei esea ch o a selec g oup o indi idu-
als on a con ac basis. Buy-side analys s ypically wo k o
mu ual unds o pension unds o o he non-b oke age i ms
and p o ide esea ch exclusi ely o hose i ms (Gell, 2011,
p. 10 .). This aises he ques ion o whe he ins i u ional in-
es o s ha e an in o ma ion ad an age o e o he in es o s.
Acco ding o G oysbe g e al. (2008), he e is no such ad-
an age. G oysbe g e al. (2008) ind ha he o ecas s o
buy-side analys s a e in ac mo e op imis ic and less accu-
a e han hose o sell-side analys s. They a ibu e his o
he highe e en ion a e o low-quali y analys s and he ac
ha buy-side i ms do no measu e he pe o mance o hei
analys s agains each o he and sell-side analys s (G oysbe g
e al., 2008, p. 37 .). They u he men ion ha buy-side
analys s a e less able o communica e di ec ly wi h i m ep-
esen a i es (G oysbe g e al., 2008, p. 26).
To o ecas ea nings, inancial analys s build inancial
models ha es ima e p ospec i e e enues and cos s o i ms.
The model e alua es in o ma ion abou he gene al economy,
he indus y and he speci ic i m and hen gene a es an es-
ima e o he i m’s ea nings. The weigh ing o he h ee
sou ces o in o ma ion, howe e , di e s be ween analys s. I
he inancial analys belie es ha he i m is no able o accu-
a ely o ecas ea nings, he is mo e likely o ely on indus y
and economic da a. (Jennings, 1985, p. 2).
When alking abou ea nings o ecas s, wha is mean
is usually he consensus ea nings es ima e. The consensus
ea nings es ima e e e s o he mean o median o he o e-
cas s o a g oup o inancial analys s. Typically, inancial ana-
lys s es ima e a i m’s qua e ly o annual ea nings pe sha e
(EPS). The mo e inancial analys s p o ide a o ecas es i-
ma e, he mo e accu a e he consensus es ima e becomes, as
ex eme and unin o med es ima es ca y less weigh (Ba on
e al. (1998) in Bya d e al. (2011, p. 94)). The accu acy
o he o ecas also inc eases wi h he amoun o in o ma-
ion a ailable o analys s, hei o ecas ing expe ience, and
hei epu a ion (Kle ke, 2013, p. 2). A he beginning o
he pe iod, analys s ha e a highe o ecas e o compa ed
o igh be o e he ea nings elease (Caps a e al., 1995, p.
74). Pa o his change in o ecas e o is due o manage s
in luencing analys s’ o ecas s by p o iding hem wi h addi-
ional in o ma ion. (Chop a, 1998, p. 36). Manage s ha e
an incen i e o e ise ea nings es ima es downwa d, because
missing he consensus ea ning es ima e is associa ed wi h a
signi ican d op in s ock p ice (J. R. G aham e al., 2005, p.
3 ). Thus, he consensus ea nings es ima e a ies o e he
yea .
Fo ecas Biases
Financial analys s a e consciously o unconsciously sub-
jec o biases when making hei ea nings o ecas s. The wo
mos p ominen o ecas biases in he li e a u e a e op imism
and he ding. Op imism desc ibes he pe sis en endency o
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1775
inancial analys s o issue o e ly posi i e ea nings o ecas s.
I is measu ed as he di e ence be ween he consensus ea n-
ings o ecas and he la e ealized ea nings (Becke s e al.,
2004, p. 75). The endency o op imism has been docu-
men ed as ea ly as he 1970s and pe sis s un il oday (Mc-
Donald 1973, p. 509; Ba e ield and Comiskey 1975, p. 244).
D eman and Be y (1995, p. 39) ind ha he op imism bias
is pe sis en ac oss indus ies and economic cycles. Financial
analys s ha e incen i es o issue mo e op imis ic o ecas s.
Hong and Kubik (2003, p. 345 .) no e ha inancial analys s
who a e mo e op imis ic han he consensus a e mo e likely
o expe ience posi i e ca ee de elopmen s. The eason o
his is ha in es men banks and b oke age houses wan an-
alys s o p omo e s ocks in o de o gene a e unde w i ing
business and ading commissions. A hanassakos and Kalim-
ipalli (2003, p. 59) u he poin ou ha o ecas op imism is
he la ges a he beginning o he yea . As mo e in o ma ion
becomes a ailable du ing he yea , inancial analys s canno
a o d o con inue being o e ly op imis ic wi hou damaging
hei epu a ion.
He ding, on he o he hand, desc ibes he social phe-
nomenon o inancial analys s o con o m and he e o e no
o de ia e oo much om he consensus. Scha s ein and
S ein (1990) ind ha he ding owa d he consensus is less
likely caused by undamen al in o ma ion, bu a he a lack
o in o ma ion. Financial analys s who ha e li le o no in o -
ma ion end o he d mo e (Welch, 2000, p. 371). In addi ion,
he eluc ance o de ia e om he consensus has been shown
o inc eases wi h he numbe o es ima es ha a e close o
he consensus and he inaccu acy o analys s’ p e ious es i-
ma es (J. R. G aham (1999) and S ickel (1990) in Becke s
e al. (2004, p. 75)). This migh be explained by he ac
ha in es o s iew ag eemen wi h he consensus as an in-
dica ion o o ecas eliabili y (De Bond & Fo bes, 1999, p.
144 .). By simply endo sing he consensus opinion, inancial
analys s ake less epu a ional isk. As Keynes said, "wo ldly
wisdom eaches ha i is be e o epu a ion o ail con en-
ionally han o succeed uncon en ionally“ (Keynes, 2018, p.
138). De Bond and Fo bes (1999, p. 146) u he men ions
ha he ding in ensi ies wi h he di icul y and ambigui y o
he ask. He ding may also be explained by ca ee conce ns.
Hong e al. (2000, p. 123) ind ha olde analys s a e mo e
likely o p oduce o ecas s ha de ia e om he consensus,
while younge analys s end o be less bold. Simila , Zwiebel
(1995, p 2 .) no e ha younge inancial analys s a e mo e
likely o unc ion as opinion leade s due o lowe epu a ional
isks. S ill, hey also no e ha he ea nings e isions om
olde inancial analys s ecei e mo e weigh due o highe
epu a ional capi al.
In he li e a u e he e a e di e en explana ions o o e-
cas biases. Gell (2011) dis inguishes six ca ego ies o ex-
plana ions o biases: cogni i e bias, s a egic bias, selec ion
bias, news bias, skewed ea nings dis ibu ion bias and man-
agemen bias explana ion. Fi s , unde he cogni i e bias
explana ion, inancial analys s a e supposed o be i a ional
and o sys ema ically make mis akes when p ocessing pub-
licly a ailable ea nings in o ma ion. Second, he s a egic
bias explana ion assumes ha inancial analys s a e a ional,
bu p oduce biased o ecas s due o s a egic incen i es. Fo
example, inancial analys s publish posi i e ea nings epo s
o please a i m’s managemen in o de o main ain a good
ela ionship wi h he i m. Thi d, he selec ion bias explana-
ion s a es ha inancial analys s make op imis ic o ecas s
only o hose i ms abou which hey a e uly op imis ic be-
cause hese i ms a e mo e likely o b ing in ading com-
missions. Fo i ms ha unde pe o m, analys s s op mak-
ing o ecas s, esul ing in ou da ed and hence biased o e-
cas s. Fou h, he news bias explana ion a ibu es o ecas
op imism o he asymme ic imeliness o ea nings due o ac-
coun ing conse a ism. Good news a e simply e lec ed in
o ecas s in a mo e imely manne han bad news. Fi h,
he skewed ea nings dis ibu ion bias explana ion assumes
ha inancial analys s a e u h ul, unselec i e, and a ional.
Though, hey can choose whe he o o ecas he mean o
median o an ea nings dis ibu ion and hus bias ea nings
o ecas s. Las , he managemen bias a ibu es o ecas bias
o he managemen p ac ices o accoun ing disc e ion and
guiding analys s’ expec a ions (Gell, 2011, p. 13 .).
Fo ecas Dispe sion
Analys o ecas dispe sion measu es he a ia ion in an-
alys s’ ea nings o ecas o a ce ain i m and pe iod. Dispe -
sion hus e lec s he di e gence in analys s’ opinion abou a
i m’s u u e ea nings (Han & Man y, 2000, p. 99). Theo e -
ical esea ch shows ha o ecas dispe sion may e lec bo h
unce ain y and in o ma ion asymme y (Ba y and Jennings
1992, p. 175 . Ba on e al. 1998, p. 422).
Unce ain y a ises because inancial analys s do no ha e
he exac ea nings numbe s and ins ead need o es ima e
ea nings. When ea nings a e announced, unce ain y de-
c eases. As a esul , he dispe sion o analys s’ ea nings es-
ima es ypically dec eases as well (Ba on e al., 2010, p.
332). Likewise, Imho and Lobo (1992, p. 437) in e p e
o ecas dispe sion as a p oxy o ex an e ea nings unce -
ain y. Chop a (1998, p. 38) inds ha he dispe sion o
ea nings es ima es declines o e he yea . He a ibu es he
decline in dispe sion o qua e ly ea nings eleases and e-
sul ing imp o ed isibili y o he i m’s p ospec s. Fu he ,
Acke (1997, p. 264 .) no es ha inancial analys s issue
mo e op imis ic o ecas s when unce ain y a ound he i m
is high. Howe e , i he unce ain y is low, inancial analys s
may hesi a e o issue op imis ic o ecas s due o epu a ional
conce ns. Fo he same eason, inancial analys s may also
a oid issuing con a ian o ecas s when unce ain y is low.
In o ma ion asymme y e e s o he di e ences in in o -
ma ion a ailable o inancial analys s. In his con ex , in-
o ma ion can be di ided in o public and p i a e in o ma-
ion. On he one hand, public in o ma ion comp ises all
i m- ela ed in o ma ion ha is eely accessible o all inan-
cial analys s. P i a e in o ma ion, on he o he hand, is only
a ailable o he indi idual analys . F om he pe spec i e o
in o ma ion asymme y, o ecas dispe sion esul s om he
di e en le el o in o ma ion a ailable o inancial analys s
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041776
(Ba on e al., 2010, p. 332). Ajinkya e al. (1991, p. 393)
a gue ha analys o ecas dispe sion is in pa due o di e -
en ial in o ma ion a ailable o inancial analys s a di e en
imes. No all inancial analys s p epa e and submi hei EPS
upda es a exac ly he same ime, so he e may be a ime lag
be ween EPS es ima es.
The ela ionship be ween public disclosu e and o ecas
dispe sion is hus no so ob ious. The e ec o disclosu e
depends on whe he a ia ions in o ecas s a e due o di -
e ences in in o ma ion o di e ences in o ecas ing models.
I analys s sha e a common o ecas ing model and obse e
he same i m-p o ided in o ma ion bu ha e di e en p i-
a e in o ma ion, hey will a ach less weigh o hei p i-
a e in o ma ion as he in o ma i eness o he i m-p o ided
in o ma ion inc eases, he eby educing o ecas dispe sion.
Con a y, i analys s ha e he same i m-p o ided and p i a e
in o ma ion, bu assign di e en weigh s o he cons i uen s
o he i m-p o ided in o ma ion when o ecas ing ea nings,
addi ional disclosu e may inc ease inancial analys s’ o ecas
dispe sion. Hence an obse ed posi i e associa ion be ween
ea nings disclosu e and o ecas dispe sion implies ha i-
nancial analys s di e in hei o ecas ing models, so ha
hey d aw di e en conclusions om he same obse ed dis-
closu es. Wi h mo e disclosu es, hei ea nings o ecas s be-
come mo e dispe sed. By con as , an obse ed nega i e ela-
ionship be ween ea nings disclosu e and o ecas dispe sion
implies ha inancial analys s a y p ima ily in hei p i a e
in o ma ion (Lang & Lundholm, 1996, p. 471 .). In ad-
di ion o he disclosu e i sel , i s quali y is also impo an .
Ea lie esea ch has shown ha poo quali y disclosu e o i-
nancial in o ma ion is associa ed wi h high analys o ecas
dispe sion. Dechow e al. (1996, p. 27) ind ha o ecas
dispe sion inc eases a e he disclosu e o alleged ea nings
manipula ions. Swamina han (1991, p. 40) shows ha o e-
cas dispe sion dec eased ollowing he elease o Secu i ies
and Exchange Commission (SEC)-manda ed segmen da a.
Mo eo e , (Ba on e al., 2010, p. 331) ind ha le els
and changes in o ecas dispe sion e lec unce ain y and in-
o ma ion asymme y o a ying deg ees. Acco ding o hem,
le els o o ecas dispe sion be o e ea nings announcemen s
mainly e lec s he a ia ion in unce ain y and no in in-
o ma ion asymme y. Recip ocally, changes in dispe sion
a ound ea nings announcemen s e lec a ia ion in in o ma-
ion asymme y a he han a ia ion in unce ain y. This
means ha when looking a le els o o ecas dispe sion,
i.e. EPS es ima es by analys s p io o ea nings announce-
men s, i is p ima ily unce ain y ha is esponsible o o e-
cas dispe sion. Bu , he e is also esea ch ha disag ees wi h
he p oposi ion ha o ecas dispe sion e lec s unce ain y.
Imho and Lobo (1992, p. 437) s udy he dispe sion o ana-
lys s’ o ecas s p io o ea nings announcemen s and sugges
ha he inc eased o ecas dispe sion is due o noise in inan-
cial s a emen s a he han o unce ain y.
Las , i ms wi h high o ecas dispe sion expe ience ce -
ain eal e ec s. Han and Man y (2000, p. 119-121) ind
ha i ms wi h high o ecas dispe sion ace high cos s o cap-
i al and low ea nings pe sis ence. Also, Die he e al. (2002,
p. 2135-2137) and Johnson (2004, p. 1975 .) demons a e
ha in es o s pay a p emium o s ocks wi h a high dispe -
sion o analys s’ o ecas s, which leads o lowe u u e s ock
e u ns, i.e., he deg ee o dispe sion is nega i ely associa ed
wi h u u e s ock e u ns. Die he e al. (2002, p. 2137-
2139) explain his nega i e associa ion wi h ma ke ic ion.
In pa icula , highe dispe sion induces a s onge op imis ic
bias in s ock p ices, as op imis ic in es o s d i e up p ices,
while pessimis ic iews a e no e lec ed in s ock p ices due
o sho -selling es ic ions, causing s ocks wi h high dispe -
sion o be o e alued.
2.3. Limi s o Ma ke Fo ecas ing
Financial analys s analyze p esen i m in o ma ion and
make es ima es abou he u u e. These u u e o ecas s a e
hen used by ma ke pa icipan s o ou pe o m he o e all
s ock ma ke . I is he e o e assumed ha he ac i i ies o
inancial analys s add alue o inancial ma ke s. Economic
heo y hough exp esses a g ea deal o skep icism abou i-
nancial analys s’ abili y o o ecas ma ke de elopmen s. In
1933, Al ed Cowles empi ically es ed he a emp o p e-
dic he de elopmen o s ock p ices. A e analyzing 7500
s ock ma ke o ecas s om inancial se ice p o ide s, he
concluded ha “s a is ical es s o he bes indi idual eco ds
ailed o demons a e ha hey exhibi ed skill, and indica ed
ha hey mo e p obably we e esul s o chance” (Cowles,
1933, p. 323). To es whe he his esul s we e due o a lack
o skill, he epea ed his es wi h he hen edi o o he Wall
S ee Jou nal. Cowles came o he same conclusion. O 90
o ecas s, hal we e success ul, and hal we e no (Cowles,
1933, p. 323). Kendall and Hill (1953, p. 11) la e ali-
da ed Cowles’ (1933) indings by showing ha s ock p ices
mo e andomly a he han p edic ably. Acco ding o eco-
nomic heo y, inancial analys s should hence no be able o
p edic ma ke mo emen s.
The mos well-known economic heo y is Eugene Fama’s
e icien ma ke hypo hesis. The e icien ma ke hypo he-
sis s a es ha all in o ma ion ha is publicly a ailable abou
a i m is ins an ly e lec ed in a i m’s s ock p ice (Malkiel
& Fama, 1970, p. 383), which makes long- e m p edic ion
o s ock ma ke mo emen s impossible. Acco ding o Fama,
“The E idence in suppo o he e icien ma ke model is ex-
ensi e, and [...]con adic o y e idence is spa se” (Malkiel
& Fama, 1970, p. 416). Fo he wo k o inancial analys s,
his means ha he e is no a chance o sys ema ically iden i-
ying unp iced in o ma ion ha will be e lec ed in he s ock
p ice a some poin in he u u e. O he wise, acco ding o he
logic o e icien ma ke s, he sha e p ice would ha e al eady
isen (Leins, 2018, p. 21). Acco ding o Fama, inancial an-
alys s can only p edic s ock p ice mo emen in an e icien
ma ke i hey ha e access o inside in o ma ion no acces-
sible o he gene al public. Since inancial analys s do no
egula ly possess inside in o ma ion, he scope o p edic
ma ke mo emen s appea s o be limi ed (Malkiel & Fama,
1970, p. 413).
S ill, esea ch sugges s ha he ma ke is no always e -
icien . Fo example, Jones and Li zenbe ge (1970, p. 147
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1783
ble ac oss indus ies (Nagy e al., 2020). S ill, no all inpu s
may be equally ma e ial o i ms’ ESG pe o mance. As a
esul , ESG a ing agencies o en apply di e en weigh s o
di e en a iables and ca ego ies and some imes e en o
di e en pilla s based on inancial ma e iali y (La cke e al.,
2022, p. 4). Those weigh ings a e mos ly based on sub-
jec i e judgmen s, e en hough a ious ESG a ing agencies
a ionalize hei decisions (Bo o & Pa alano, 2020, p 31).
Ano he app oach is o apply an op imized weigh ing based
on his o ical da a. In his app oach, he weigh s a e adjus ed
o mi o he bes inancial pe o mance based on a collec-
ion o his o ical da a. Fo example, a esea ch epo om
MSCI inds ha weigh s o 25% E pilla , 5% S pilla and 70%
G pilla yield he bes inancial esul s. Ye ano he app oach
is o apply indus y-speci ic weigh s. The ad an age o his
app oach is ha i mo e p ecisely e lec s indus y exposu es
o E, S, and G isks. Bu , he disad an age o his app oach
is ha i leads o mo e complexi y and is less compa able
ac oss indus ies. The same esea ch epo om MSCI no es
ha he E pilla weigh ing a ies om 5.8% o he commu-
nica ions se ices sec o o 62.1% o u ili ies. The S pilla
weigh ing a ied be ween 16.3% o he ene gy sec o and
59.8% o he inancial sec o . In he sho e m, Nagy e
al. (2020) ind ha bo h equal-weigh ed and op imized ap-
p oaches demons a ed supe io pe o mance, a ibu ed o
inc eased exposu e o go e nance issues. In he longe e m,
howe e , he indus y-speci ic weigh ed app oach showed
he s onges inancial pe o mance (Nagy e al., 2020). Be-
cause ESG measu es he long- e m isks and oppo uni ies o
a i m’s inancial pe o mance, he indus y-speci ic weigh ed
app oach hence appea s supe io .
ESG a ing agencies some imes also inco po a e con o-
e sies su ounding a ed i ms in o hei ESG a ings. Con-
o e sies a e e en s ha cause epu a ional damage and
demons a e a i m’s lack o p epa edness and/o inabili y
o deal wi h eme ging e en s and isks. Ha ing said ha , no
all ESG a ing agencies include con o e sies in o hei ESG
a ings. Some p o ide con o e sies as a s and-alone a ing
ha exis s alongside he pilla s and con ibu es o a combined
o e all ESG a ing, while o he s do no conside con o e -
sies a all (Bo o & Pa alano, 2020, p 30). In summa y, he e
a e me hodological pa ame e s ha allow ESG a ing agen-
cies o p oduce di e en ESG a ings. In addi ion, ESG a ing
agencies a e no ully anspa en abou how hei a ings a e
p oduced.
4.2.3. ESG Ra ings and Ra ing Biases
A a ing is an e alua ion p o ided by a hi d pa y. I
is an in o ma ion p oduc , a s a emen c ea ed wi h he ex-
plici pu pose o being communica ed ou wa d (Poon, 2012,
p. 460). In he case o ESG a ings, ESG a ing agencies p o-
ide an e alua ion o a i m’s ESG pe o mance, which hey
communica e o in es o s and o he s akeholde s.
ESG a ings a e usually exp essed in he o m o le e s
o numbe s. Some ESG a ing agencies use a se en-poin
scale om AAA o CCC. O he s use a wel e-poin scale om
A+ o D, simila o g ades in he Anglo-Ame ican educa ion
sys em. Ye o he s publish sco es on a pe cen ile basis using
a scale o 1 o 100, whe e 100 can ei he ep esen high ESG
quali y (posi i e) o high ESG isk (nega i e). In addi ion,
many ESG a ing agencies claim o measu e indus y- ela i e
ESG pe o mance, while some claim o measu e absolu e ESG
pe o mance.
Indus y-adjus ed a ings enable in es o s o compa e
ESG pe o mance among i ms ope a ing wi hin he same
indus y. In his way, i ms can be compa ed agains hei
indus y-pee s in hei abili y o manage inancial ma e ial
ESG isks. Howe e , ESG a ings based on indus y c i e ia
hinde he abili y o compa e i ms ac oss di e en indus-
ies and a e highly dependen on he assigned indus y. In
con as , absolu e ESG a ings can be compa ed ac oss indus-
ies. Al hough a ings may a y depending on he indus y
o which i ms a e assigned. Fi ms in mo e sus ainable in-
dus ies end o ecei e highe a ings, while i ms in less
sus ainable indus ies end o ecei e lowe a ings (La cke
e al., 2022, p. 3 ).
Mo eo e , ESG a ings a e expensi e. Ins i u ional in-
es o s spend on a e age $ 487,000 pe yea on ex e nal ESG
a ings, da a, and consul an s. Many use mo e han one ESG
sou ce in hei in es men p ocess (The Sus ainAbili y Ins i-
u e, 2022, p. 5). A 2021 su ey inds ha mo e han hal
o ins i u ional in es o s use mo e han one ESG da a and
esea ch sou ce, wi h 25% an icipa ing o use six o mo e
sou ces in he nex wo o h ee yea s (Capi al G oup, 2021,
p. 29). The e has also been discussion abou whe he ESG
pe o mance can be dis illed in o a single a ing. Some in-
es o s hide behind ESG a ings and use hem as a subs i u e
o in-dep h ESG esea ch and analysis. They may see ESG
a ings as a quick ix. This happens because some in es o s
may lack he esou ces o undamen al esea ch o simply
wan o check a box (The Sus ainAbili y Ins i u e, 2020, p.
31). O he in es o s s ess ha ESG pe o mance can no be
agg ega ed in o a single a ing and ha addi ional in-house
esea ch is needed o make sense o ESG a ings (The Sus ain-
Abili y Ins i u e, 2022, p. 30). These in es o s iew ESG a -
ings as a s a ing poin o help hem unde s and he b oade
landscape and o benchma k i ms agains each o he . Fo in-
s ance, a poo a ing may signal he need o u he esea ch.
They ely on hei own hinking and use ESG a ings o he
unde lying da a a he han he sco es hemsel es. They de-
elop a s ong sense o which ESG ac o s a e he mos im-
po an o a pa icula indus y and hen pe o m hei own
e alua ion o a i m’s ESG pe o mance (The Sus ainAbili y
Ins i u e, 2022, p. 23 .).
ESG a ings a e also egula ly biased. The mos p e a-
len biases a e i m size, geog aphical bias, and indus y a -
ilia ion. One pa e n is ha ESG a ings a e biased owa ds
la ge -sized i ms. Fi ms wi h highe ma ke capi aliza ion
o ee loa a e mo e likely o be co e ed by a e s, and
hei a ings a e mo e likely o be eassessed. Recen ini ial
public o e ings a e unlikely o be a ed in hei i s yea o
lis ing (B ackley e al., 2022). Unlis ed i ms a e o en ex-
cluded om ESG a ings comple ely (Zhang, 2021). La ge
i ms also end o ecei e highe a e age a ings compa ed

R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041784
o smalle i ms (Giese e al., 2019, p. 77). The eason
o his migh be ha la ge i ms disclose mo e ESG da a
due o mo e designa ed employees o he adop ion o be -
e sus ainable managemen ools (D empe ic e al., 2020,
p. 153). A second pa e n is ela ed o he i m’s loca-
ion. Fi ms lis ed on exchanges in No h Ame ica and Eu ope
a e a mo e likely o ge p ope ly a ed han hose ading
elsewhe e, pa icula ly in eme ging ma ke s (B ackley e al.,
2022). Mo eo e , i ms in Eu ope egula ly achie e highe
a ings on a e age han i ms in he US. This pa e n is no
due o highe quali y ESG p ac ices by Eu opean i ms, bu
a he o manda o y epo ing equi emen s. Fi ms in he
EU a e equi ed by law o epo on a ious en i onmen al
and social opics unde he Non- inancial Repo ing Di ec i e
and Co po a e Sus ainabili y Repo ing Di ec i e. As a esul ,
he e is g ea e a ailabili y o non- inancial in o ma ion (La-
Bella e al., 2019, p. 5). The hi d pa e n is indus y-based.
ESG a ings ha a e no indus y-adjus ed, i.e., do no assign
sco es based on indus y pee s, may assign highe a e age
sco es o ce ain indus ies (such as banking and elecom-
munica ions) and lowe sco es o o he s (such as obacco
and gambling) (La cke e al., 2022, p. 5). Fu he mo e,
indus y-weigh ed ESG a ings assume ha i ms in he same
indus y ha e simila business models and a e he e o e ex-
posed o simila ESG isks and oppo uni ies. Howe e , his
app oach can cause o e simpli ica ion in cases whe e i ms
a e no compa able. While i is impo an o s anda dize
me hodologies, wi hou indi idualized weigh ings, ESG a -
ings migh be skewed (Sipiczki, 2022, p. 8).
In addi ion, esea ch shows ha ESG a ings ha e mo ed
upwa d o e ime. D. E. Shaw (2022) analyze MSCI’s agg e-
ga e ESG sco es o all Russel 1000 i ms be ween 2015 and
2021, and ind ha sco es ha e imp o ed by 18% o e his
pe iod. S ill, s uc u al changes, such as changes in index
composi ion, changes in componen weigh ing, and g ea e
disclosu e by i ms, accoun o only 6% o his imp o emen .
The emaining 12 a e no explained by MSCI. D. E. Shaw
(2022) a ibu e his gap o g ade in la ion (D. E. Shaw,
2022, p. 6). O he esea ch shows ha low sco ing i ms
ha e seen g ea e imp o emen in hei ESG a ings han
high-sco ing i ms. They a ibu e his g ea e sco e imp o e-
men o inc eased in es o s sc u iny (Bo o & Pa alano, 2020,
p. 43).
Fu he mo e he e a e a numbe o issues ha a ec he
quali y o ESG a ings. Fi s , he e is a con lic o in e es a is-
ing om he p o ision o consul ing se ices o a ed i ms.
The p ac ice o o e ing paid se ices o a ed i ms aises se-
ious conce ns abou he independence o hose ESG a ings
(La cke e al., 2022, p. 7). Tang e al. (2022, p. 29) ind
ha i ms a ilia ed wi h ESG a ing agencies ecei e highe
ESG a ings han i ms no a ilia ed wi h hem. Second, ESG
a ings a e mos ly backwa d-looking, i.e., hey e alua e pas
pe o mance, while in es o s ac ually look o indica o s o
u u e pe o mance (The Sus ainAbili y Ins i u e, 2020, p.
28). As a esul , in es o s ha e s a ed ha hey would like
o ha e mo e imely upda es (The Sus ainAbili y Ins i u e,
2020, p. 43). Thi d, ecen esea ch claims ha ESG a -
ings do no eliably p edic u u e sus ainabili y pe o mance
and do no co ela e wi h ESG isk managemen capabili ies
(B ackley e al., 2022). Fou h, in es o s epo ha ESG
a ing agencies o en do no espond o complain s abou in-
accu a e in o ma ion om a ed i ms. ESG a ing agencies
a e o en no su icien ly s a ed o p o ide comp ehensi e
suppo (The Sus ainAbili y Ins i u e, 2020, p. 28).
4.2.4. Compa ison o ESG Ra ing Agencies’ Me hodologies
Sus ainaly ics
In e ms o me hodology o ESG a ings, Sus ainaly ics
assesses a i m’s ESG pe o mance by measu ing he ex en
o which a i m’s economic alue is exposed o unmanaged
ma e ial ESG isks (Sus ainaly ics, 2021, p. 4). The analysis
is based on da a collec ed om a i m’s public disclosu e, he
media, and NGO epo s. The model includes be ween 70-
90 ESG indica o s o la ge and mid cap i ms and be ween
20-30 o small cap i ms. Indica o s a e selec ed based on
hei ele ance o he assigned pee g oup and o he i m’s
pa icula business model. A he momen , Sus ainaly ics dis-
inguishes be ween 138 pee g oups, which a e ca ego ized
in o 42 dis inc indus ies. Sus ainaly ics uses building blocks
ha s a wi h co po a e go e nance, conside ma e ial ESG
issues, and hen look o idiosync a ic ESG issues. Be as a e
hen used by embedding he impac o e en s on inancial
pe o mance in o he p ocess (Sus ainaly ics, 2021, p. 5-8).
Once he analysis is done, i ms ha e wo weeks o p o ide
eedback and submi addi ional in o ma ion. The inal e-
sul compiled in o a sco e be ween 0 and 100, wi h a lowe
sco e being be e as i means less exposu e o unmanaged
ESG isks (Sus ainaly ics, 2020, p. 7). The a ing is abso-
lu e, meaning i is compa able ac oss all pee g oups co -
e ed. (Sus ainaly ics, 2021, p. 4). In addi ion, Sus ainaly -
ics p o ides indi idual E/S/G clus e sco es and con o e sy
esea ch. Those a e no used o calcula e he ESG Risk Ra -
ing bu p o ide in es o s wi h addi ional in o ma ion on ESG
pe o mance (Sus ainaly ics, 2021, p. 12-14). The a ings
a e upda ed annually, while con o e sy esea ch is upda ed
as e en s occu (Sus ainaly ics, 2020, p. 5).
MSCI ESG Resea ch
MSCI’s a ing me hodology is as ollows. Fi s , MSCI
collec s mac o da a, i m disclosu es and da a om media,
NGOs, and o he s akeholde s (MSCI, 2022b, p. 14). Then,
MSCI measu es a i m’s exposu e o ma e ial ESG isks and
he quali y o a i m’s isk managemen (MSCI, 2022b, p. 6).
This is done by analyzing he indi idual E/S/G pilla s based
on a selec ion o 35 key issues. Fi m-speci ic excep ions a e
allowed o i ms wi h di e si ied business model, acing con-
o e sies, o based on indus y ules (MSCI, 2022b, p. 3 .).
Figu e 4 in he annex shows an example o chosen key mea-
su es o he Coca Cola. Each en i onmen al and social key
issue ypically accoun s o 5% o 30% o he o al ESG a -
ing. The weigh ings ake in o accoun he indus y’s con i-
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1785
bu ion, ela i e o all o he indus ies, o nega i e o posi i e
en i onmen al o social impac s, as well as he ime ame in
which he isk o oppo uni y is expec ed o ma e ialize. The
weigh o he go e nance pilla is se a a minimum alue
o 33% (MSCI, 2022b, p. 5 .). Con o e sies a e di ec ly in-
cluded in he a ing o indica e s uc u al p oblems in a i m’s
isk managemen (MSCI, 2022b, p. 9). MSCI is p oac i ely
eaching ou o i ms o eedback. Bu , hey do no issue
su eys o ques ionnai es o conduc gene al in e iews wi h
i ms. Nei he a e in o ma ion ha is no publicly a ailable o
s akeholde s accep ed and aken in o accoun (MSCI, 2022b,
p. 14).To a i e a he inal ESG a ing, he weigh ed a -
e age o he E/S/G pilla is compu ed and hen no malized
ela i e o indus y pee s. The bes possible sco e is AAA and
he wo s CCC. The a ing is in ended o be in e p e ed ela-
i e o a i m’s pee s and no absolu e (MSCI, 2022b, p. 10
.). A e he a ing is published, i ms a e moni o ed on a
sys ema ic and ongoing basis. Con o e sies a e moni o ed
on a daily basis and new in o ma ion is e lec ed in epo s on
a weekly basis. Signi ican changes o sco es igge a e iew
and e a ing (MSCI, 2022b, p. 14).
Re ini i
Re ini i ESG sco es a e designed o anspa en ly and ob-
jec i ely measu e a i m’s ela i e ESG pe o mance, commi -
men and e ec i eness (Re ini i , 2022b, p. 3). The me hod-
ology is as ollows. Re ini i ’s model is ully au oma ed, da a-
d i en, and anspa en , making i ee om subjec i i y and
hidden calcula ions and inpu s (Re ini i , 2022b, p. 6). The
analysis is based exclusi ely on publicly a ailable da a om
annual epo s, i m websi es, NGO websi es, s ock exchange
ilings, CSR epo s and news sou ces (Re ini i , 2022b, p. 4).
The model cap u es and calcula es o e 630 i m le el ESG
measu es, o which a subse o 186 o he mos compa a-
ble and ma e ial a e used o he i m alua ion and sco ing
p ocess (Re ini i , 2022b, p. 6). Indica o s ha a e i ele-
an o a pa icula sec o a e excluded (Re ini i , 2022b,
p. 9). No epo ing on imma e ial da a poin s has no sig-
ni ican in luence on a i m’s a ing, howe e , no epo ing
on highly ma e ial da a poin s has a nega i e impac on a
i m’s a ing (Re ini i , 2022b, p. 3). The ESG measu es a e
hen agg ega ed in o ca ego ies. En i onmen al and social
ca ego ies a e benchma ked agains o he i ms in he same
indus y, whe eas go e nance ca ego ies a e benchma ked
agains o he i ms in he same coun y o inco po a ion. Ca -
ego ies a e hen compiled in o weigh ed E/S/G pilla s om
which he inal ESG sco e is calcula ed (Re ini i , 2022b, p.
8 .). In es iga ed i ms a e no asked o eedback, al hough
hey may eques upda es a any ime (Deloi e, 2021). Re-
ini i has wo di e en sco es. The egula ESG sco e and
he ESGC sco e, which discoun s o ESG con o e sies im-
pac ing he i m. The inal a ing is issued bo h in poin s
om 0-100 and in le e g ades om D- o A+, wi h a highe
sco e o g ade indica ing be e ESG pe o mance. ESG da a
and sco es a e ecalcula ed on an ongoing basis o align wi h
co po a e epo ing pa e ns (Re ini i , 2022b, p. 3 .).
The nex chap e del es in o he issue o disag eemen
among ESG a ing agencies. In pa icula , he ex en o which
ESG a ing agencies disag ee and he easons o hei dis-
ag eemen . The chap e aims o build a heo e ical ounda-
ion o he independen a iable o his mas e hesis.
4.3. Disag eemen among ESG Ra ing Agencies
ESG a ing agencies can disag ee signi ican ly wi h e-
spec o hei ESG a ings. In a ecen s udy, Be g e al. (2022)
examine he disag eemen be ween he ESG a ings o i e
majo ESG a ing agencies (KLD, Sus ainaly ics, Moody, Re-
ini i and S&P Global). They ind an a e age co ela ion o
only 54% be ween he ESG a ings (see Table 1), which is
su p ising since hese ESG a ings a e supposed o measu e
he same isk cons uc . A he pilla le el, he disag eemen
is e en highe wi h co ela ions o 0.53, 0.42, and 0.30 o E,
S, and G, espec i ely. ESG a ing agencies appea o disag ee
he mos on go e nance issues, wi h some ESG a ing agen-
cies e en exhibi ing nega i e co ela ions. The nega i e co -
ela ions indica e ex eme disag eemen among ESG a ing
agencies. Fi ms ha we e conside ed o ha e good ESG pe -
o mance by one ESG a ing agency, we e conside ed o ha e
bad ESG pe o mance by he o he ESG a ing agency. The
esul s indica e ha he in o ma ion in es o s ecei e om
ESG a ing agencies is ela i ely noisy.
O he s udies suppo he no ion ha he e is a signi i-
can disag eemen among ESG a ing agencies. P all (2021)
analyses he co ela ions be ween six majo ESG a ing agen-
cies (MSCI, S&P, Sus ainaly ics, CDP, ISS and Bloombe g).
He inds e en lowe co ela ions be ween hose ESG a ing
agencies, wi h an a e age co ela ion o jus 35%. MSCI’s
co ela ion wi h bo h Sus ainaly ics and S&P is below 50%
(see Table 17 in he annex). The es o he co ela ions
ange om 0.74 (be ween S&P and Bloombe g) o 0.07 (be-
ween ISS and CDP). S a e S ee Global Ad iso s (2019,
p. 2) assesses c oss-sec ional co ela ions be ween ou ma-
jo a ing agencies (Sus ainaly ics, MSCI, RobeccoSAM and
Bloombe g). The esul s show an a e age co ela ion o 60%.
The co ela ion be ween Sus ainaly ics and MSCI is only 53%
(see Table 18 in he annex), which is consis en wi h he ind-
ings o P all (2021). Bo o and Pa alano (2020, p. 28) exam-
ine he ESG a ing a ia ion among h ee majo ESG a ing
agencies (Bloombe g, MSCI, and Re ini i ) o he compo-
nen s o he S&P 500 and STOXX 600 indices. They ind la ge
di e ences, wi h an a e age R-squa ed o 0.21 o he S&P
500 and 0.18 o he STOXX 600. F om a co ela ion pe -
spec i e, hese alues co espond o 46% and 42% o he
wo indices, espec i ely.
In ano he analysis, Bo o and Pa alano (2020, p. 29)
compa e he disag eemen be ween ESG a ing agencies and
c edi a ing issue s. Fo his pu pose, hey selec ed lis ed
i ms by la ges ma ke capi aliza ion o ep esen a ious
indus ies. The esul s show ha ESG a ings agencies dis-
ag ee signi ican ly in hei ESG a ings, while c edi a ing
issue s mos ly ag ee (see Figu e 1). Be g e al. (2022, p. 6
.) e en epo a co ela ion be ween c edi a ings o 99%.
P all (2021) ind ha he c edi a ings o he i ms in hei
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041786
Table 1: Co ela ions be ween ESG a ings (Sou ce: Be g e al. (2022, p. 30))
No e: Co ela ions be ween ESG a ings a he agg ega e a ing le el (ESG) and a he le el o he en i onmen al dimension (E), he social dimension (S),
and he go e nance dimension (G). SA, SP, MO, RE, KL, and MS a e sho o Sus ainaly ics, S&P Global, Moody’s ESG, Re ini i , KLD, and MSCI, espec i ely.
sample ha e a co ela ion be ween 94% and 96%. The e o e,
he disag eemen seems o be unique o non- inancial a ing
agencies.
This aises he ques ion why ESG a ing agencies disag ee
ha much. As men ioned in he p e ious chap e , ESG a ing
agencies use e y di e en a ing me hodologies o collec ,
measu e and analyze ESG in o ma ion. The disag eemen
can be mos ly a ibu ed o di e en ESG a ing me hodolo-
gies. Because ESG a ing agencies compe e wi h each o he
o ma ke sha e, he e is no single app oach o ESG a ing
me hodologies. Each ESG a ing agency uses i s own p o-
p ie a y me hodology o di e en ia e i sel om hei pee s
and o mee in es o s’ needs (B ackley e al., 2022). In addi-
ion, ESG a ing me hodologies a e o en no ully anspa -
en (Doyle, 2018, p. 8). Be g e al. (2022) seek o unde -
s and which ac o s con ibu e o he disag eemen among
ESG a ing agencies. They decons uc ESG a ings in o h ee
ac o s: scope ( he a ibu es ha he ESG a ing agencies a -
emp o measu e), measu emen ( he measu es used o as-
sess he a ibu es), and weigh ing ( he ela i e impo ance
assigned o he a ibu es). They ind ha he majo i y o
disag eemen be ween ESG a ing agencies can be a ibu ed
o di e ences in measu emen (56%) and scope (38%), wi h
weigh ing di e ences accoun ing o only 6% o he disag ee-
men . The one excep ion o he s udy is MSCI, whe e he
scope, a he han he measu emen , accoun s o mos o
he disag eemen due o he i m-speci ic weigh s (Be g e
al., 2022, p. 16 .).
A he scope le el, ESG a ing agencies di e in he
amoun and ype o inpu a iables. While se e al ESG
a ing agencies use ESG amewo ks, such as GRI, SASB, and
TCFD o selec inpu a iables, o he s do no . Inpu a iables
a e also selec ed o some deg ee based on da a a ailabili y
o ensu e ha each indica o can be accu a ely measu ed
o e ime. In cases whe e i ms do no p o ide di ec in o -
ma ion, app oxima ions a e used, which may o may no be
accu a e. Pe haps con a y o expec a ions, Ch is ensen e al.
(2021, p. 5 .) ind ha inc eased i m disclosu e does no
lead o mo e consis en ESG a ings. Ins ead, hey ind ha
i ac ually inc eases he disag eemen be ween ESG a ing
agencies. This is because he subjec i e na u e o ESG in o -
ma ion allows o di e en in e p e a ions o he disclosed
in o ma ion, leading o g ea e disag eemen among ESG
a ing agencies. Inpu a iables can also di e signi ican ly
ac oss indus ies o i ms o accoun o inancial ma e iali y
(Bo o & Pa alano, 2020, p 31). ESG a ing agencies may
also eplace inpu a iables h ough ime, making i di icul
o compa e ESG a ings o e ime o e en leading o changes
in pas ESG a ings (Esc ig-Olmedo e al., 2019, p. 14).
A he measu emen le el, disag eemen be ween ESG
a ing agencies can a ise due o di e ences in he in e p e-
a ion o ESG in o ma ion. Fo ins ance, ESG a ing agen-
cies use expe judgemen o de e mine which inpu ac o s
a e ma e ial o a ious indus ies, how o in e p e a ious
inpu ac o s, o how o handle da a gaps (Bo o and Pa a-
lano 2020, p. 31; Ko san onis and Se a eim 2019, p. 54).
Be g e al. (2022, p. 18) ind ha he ESG a ing agencies’
assessmen o a i m in indi idual ca ego ies can in luence
hei o e all iew o he i m, a phenomenon hey called he
a e e ec . When a a e had a posi i e iew o a i m’s
pa icula indica o , hey we e mo e likely o ha e a posi-
i e iew o he i m’s o he indica o s as well. Be g e al.
(2022, p. 17) u he ind ha ce ain ca ego ies a e mo e
p one o disag eemen . ESG a ing agencies mos ly disag ee
on clima e isk managemen , p oduc sa e y, co po a e go e -
nance, co up ion and en i onmen al managemen sys ems.
O he ca ego ies, such as en i onmen al ines, clinical ials,
employee u no e , HIV p og ams and non-g eenhouse gas
ai emissions a e less p one o disag eemen . Ano he ac o
ha in luences he in e p e a ion o ESG in o ma ion is expe-
ience. Many in es o s c i icize he insu icien senio i y and
enu e o esea ch analys s who de elop ESG a ings, s a ing
ha esea ch eams a e s e ched oo hin and do no ha e
a deep enough unde s anding o he issues and sec o s (The
Sus ainAbili y Ins i u e, 2020, p. 29). In any case, he le el
o expe ience a ec s he quali y o ESG a ings and hus he
disag eemen be ween ESG a ing agencies.
Finally, he e a e weigh s. ESG a ing agencies o en as-
sign di e en weigh s o di e en inpu a iables and ca e-
go ies. These weigh s can be ei he de e mined by expe
judgmen o based on quan i a i e da a-d i en app oaches
(Bo o & Pa alano, 2020, p 31). Inpu a iables and ca e-
go ies ha ha e a g ea e impac on he i m’s inancial pe -
o mance o en ecei e a highe weigh ing (La cke e al.,
2022, p. 4). Since weigh s a e assigned by he indi idual
ESG a ing agencies, he e may be di e ences in weigh ings
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1787
Figu e 1: Compa ison o disag eemen be ween ESG a ings and c edi a ings
(Sou ce: Bo o and Pa alano (2020, p. 29))
ha can lead o disag eemen be ween ESG a ing agencies.
The disag eemen be ween ESG a ing agencies can be
bo h unin en ionally and in en ionally. Unin en ional dis-
ag eemen in ESG a ings o en occu s a he le el o speci ic
inpu ac o s o da a poin s. Unin en ional di e gence in ESG
a ings may occu when di e en a e s e alua ing he same
i m ha e di e en access o da a o in e p e he same in o -
ma ion di e en ly, esul ing in di e gen conclusions abou
he i m’s ESG pe o mance. In en ional disag eemen in ESG
a ings ypically occu s a he composi e ESG sco e le el, and
is he esul o he a e ’s comp ehensi e analysis o he i m’s
ESG pe o mance based on i s own me hodology. This dis-
ag eemen e lec s he di e ing pe spec i es and app oaches
used by he di e en ESG a ing agencies in e alua ing a
i m’s ESG pe o mance (B ackley e al., 2022).
The disag eemen be ween ESG a ing agencies, which is
o en caused by he lack o consis ency and s anda diza ion
in a ing me hodologies, can limi he use ulness o ESG a -
ings in p o iding eliable and meaning ul in o ma ion abou
a i m’s long- e m esilience and non- inancial pe o mance
(B ackley e al., 2022). Wi hou a consis en and s anda d-
ized app oach o ESG a ings, i can be di icul o compa e
and e alua e he ESG pe o mance o di e en i ms, mak-
ing i challenging o use ESG a ings as a ool o in o med
decision-making (La cke e al., 2022, p. 6). Howe e , while
g ea e consis ency in ESG a ings may be desi able in e ms
o p o iding mo e eliable and meaning ul in o ma ion abou
a i m’s pe o mance, i is no clea whe he in es o s nec-
essa ily wan g ea e consis ency in a ing me hodologies.
G ea e egula ion o ESG a ings may help o s anda dize
he in o ma ion inpu and a ing p ocess, esul ing in mo e
consis en a ings and educing he disag eemen be ween
ESG a ing agencies. On he one hand, g ea e consis ency
may educe he amoun o con lic ing o con adic o y ESG
a ings, making i easie o in es o s o compa e and e al-
ua e he ESG pe o mance o di e en i ms. On he o he
hand, he inclusion o mul iple pe spec i es and app oaches
in he ESG a ing p ocess may p o ide a mo e comp ehensi e
and nuanced iew o a i m’s pe o mance, and may be seen
as a posi i e cha ac e is ic by some in es o s (B ackley e al.
2022; The Sus ainAbili y Ins i u e 2020, p. 44 .).
The nex chap e ocuses on he de elopmen o he hy-
po hesis. This chap e aims o p o ide a heo e ical ame-
wo k ha can be used o make p edic ions abou he associ-
a ion be ween ESG disag eemen and he dispe sion o ana-
lys s’ o ecas s.
5. Hypo hesis De elopmen : In luence o ESG Ra ing Dis-
ag eemen on Analys Fo ecas Dispe sion
In his mas e hesis, I seek o unde s and he ela ion-
ship be ween ESG a ing disag eemen and analys o ecas
dispe sion. ESG a ings a e measu es o a i m’s pe o mance
in ela ion o ESG c i e ia. These a ings aim o measu e
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041788
a i m’s exposu e o ESG isks and oppo uni ies, and how
hose isks and oppo uni ies may impac he i m’s inancial
pe o mance (MSCI, 2022b, p. 3). ESG a ing disag eemen
e e s o he deg ee o a ia ion in hese a ings among di -
e en ESG a ing agencies. Analys o ecas dispe sion e e s
o he deg ee o disag eemen among analys s in hei o e-
cas s o a i m’s u u e EPS pe o mance. In he accoun ing
and inance li e a u e, analys o ecas dispe sion is widely
ecognized as an impo an measu e, and is o en used as
a p oxy o he unce ain y and he di e gence in analys s’
belie s and he lack o consensus o ag eemen (Ba y and
Jennings 1992, p. 172; Aba banell e al. 1995, p. 32; Ba on
e al. 2010, p. 422).
The e is ongoing deba e in he li e a u e abou why ESG
a ing agencies may disag ee in hei a ings o a i m’s pe -
o mance and how ESG in o ma ion is ele an o ma ke pa -
icipan s. Ch is ensen e al. (2021, p. 4-6) examine whe he
a i m’s ESG disclosu e impac s he disag eemen be ween
ESG a ing agencies. They ind ha g ea e ESG disclosu e
leads o g ea e ESG a ing disag eemen . They u he ind
ha ESG disag eemen is associa ed wi h highe s ock e u n
ola ili y and la ge absolu e p ice mo emen s, and is he e-
o e ele an o ma ke pa icipan s. K uege e al. (2021, p.
35) s udy how manda o y ESG disclosu e a ec s he dispe -
sion o analys s’ ea nings o ecas s. They ind ha as manda-
o y ESG disclosu e imp o es, analys ea ning o ecas s be-
come less dispe sed. They also ind ha manda o y ESG dis-
closu e signi ican ly educes he amoun o nega i e ESG in-
ciden s in a i m-yea (K uege e al., 2021, p. 49). Cho e
al. (2013, p. 81 .) in es iga e whe he CSR pe o mance e-
duces he bid-ask sp ead, a p oxy o in o ma ion asymme y.
They ind ha bo h posi i e and nega i e CSR pe o mance
seem o educe in o ma ion asymme y. In o ma ion asym-
me y i sel is o en in e p e ed as a cons i uen o unce ain y
(Ba on e al., 2010, p. 333). Ha ing said ha , he li e a u e
on he ela ionship be ween a ings and analys o ecas dis-
pe sion is sca ce. A amo e al. (2009, p. 85) examine a
dispe sion-based ading s a egy. They ind ha a po olio
s a egy based on buying low dispe sion s ocks and selling
high dispe sion s ocks yields a s a is ically signi ican e u n.
They u he ind ha ecen c edi a ing downg ades lead
o highe analys o ecas dispe sion (A amo e al., 2009, p.
99 .). The e a e e en ewe s udies when i comes o he ela-
ionship be ween ESG a ing disag eemen and analys o e-
cas dispe sion. In ac , du ing my esea ch I we e only able
o ind one s udy ha add essed his ela ionship. Kimb ough
e al. (2022, p. 48) examine whe he ESG a ing disag ee-
men is associa ed wi h disag eemen among ma ke pa ic-
ipan s. They ind ha ESG a ing disag eemen is posi i ely
associa ed wi h analys o ecas dispe sion, bid-ask sp ead
and u u e s ock e u n ola ili y. Though, he ela ionship
be ween ESG a ing disag eemen and analys o ecas dis-
pe sion is only s a is ically signi ican a he 10% le el, indi-
ca ing a weak link be ween he wo a iables and ha he e-
la ionship may no be causal. Also, Kimb ough e al. (2022)
analyzed he ela ionship be ween ESG a ing disag eemen
and analys o ecas dispe sion in he US. The e o e, u he
esea ch may be needed o con i m o e u e he ela ionship
be ween he wo a iables. This mas e hesis aims o ill
his esea ch gap by conduc ing an empi ical analysis on he
ela ionship be ween ESG a ing disag eemen and analys
o ecas dispe sion in an in e na ional se ing (Kimb ough e
al., 2022, p. 48). Since esea ch on he associa ion be ween
ESG disag eemen and analys o ecas dispe sion is spa se,
he hypo hesis de elopmen is discussed in mo e de ail. Di -
e en a gumen s o a posi i e, nega i e and no associa ion
a e p esen ed. A decision is hen made in a o o one di ec-
ion o he o he based on he s onges a gumen s.
The e a e se e al a gumen s ha could be made in a-
o o a posi i e ela ionship be ween ESG a ing dispe sion
and analys o ecas dispe sion. Fi s , i analys s use di e -
en ESG a ings, his could lead o di e ences in hei EPS
o ecas s, as each ESG a ing agency p o ides di e en in o -
ma ion and pe spec i es. The access o ESG a ings can be
cos ly, wi h ins i u ional in es o s on a e age spending on $
487,000 pe yea on ESG a ings, da a and consul an s (The
Sus ainAbili y Ins i u e, 2022, p. 5). This means ha some
analys s may no ha e he esou ces o access paid ESG a -
ings se ices o may choose o use ewe o hem in hei
e alua ions. This could lead o di e ences in he ESG a -
ings used by analys s, esul ing in a ia ions in hei o e-
cas s. Addi ionally, he selec ion o ESG a ings by indi idual
analys s may be a ac o , as some a ing agencies a e mo e
likely o disag ee wi h o he s (see Table 17 and 18 in he an-
nex). Acco ding o Capi al G oup (2021, p. 29), he majo i y
o in es o s use be ween wo and i e di e en ESG a ings
(57%), while some use only one (24%) o none a all (7%).
This means ha i is possible ha analys s a e no using he
same ESG a ing agencies in hei assessmen s, which could
con ibu e o he dispe sion in hei o ecas s. In he u u e, i
is expec ed ha he numbe o ESG a ings used by in es o s
will inc ease, which may lead o a dec ease in he e ec o
ESG disag eemen on analys o ecas dispe sion as he a i-
a ions in ESG a ings a e a e aged ou .
Second, e en hough analys s may use he same ESG a -
ings, hei in e p e a ions and esul ing EPS o ecas s can
a y signi ican ly. This is because some analys s may sim-
ply iew ESG a ings as a o m o box-checking exe cise and
do no del e deepe in o how ESG a ing agencies a i e a
hei ESG a ings (The Sus ainAbili y Ins i u e, 2020, p. 31).
O he s may use ESG a ings as a s a ing poin o u he
esea ch, sc u inizing he measu emen , scope, and weigh s
o he a ings in hei analysis. High le els o disag eemen
among ESG a ing agencies in pa icula can be seen as a
eason o a mo e in-dep h analysis (Bo o & Pa alano, 2020,
p. 29 .). As a esul , analys s may de elop di e en p i-
a e knowledge abou ESG a ings, leading o dispe sion in
analys EPS o ecas s. This iew is consis en wi h Lang and
Lundholm (1996, p. 471 .), who a gues ha ha as public
in o ma ion becomes less in o ma i e, analys s place mo e
emphasis on hei p i a e in o ma ion. I is also consis en
wi h Behn e al. (2008, p. 330) who a gues ha g ea e
dispe sion may e lec a lack o ag eemen among analys s,
po en ially due o some analys s’ inabili y o eluc ance o

R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1789
ully and objec i ely ga he and in e p e ESG- ela ed in o -
ma ion.
Thi d, analys s may disag ee abou whe he ESG a ings
ac ually e lec a i m’s non- inancial pe o mance, as he e
seems o be no consensus e en among ESG a ing agencies.
Assuming ha ESG a ing agencies obse e he same i m-
disclosed ESG in o ma ion, and a e i ms based on hei non-
inancial isks and oppo uni ies, he e should be no dispe -
sion in ESG a ings. Bu , ESG a ing agencies seem o be
no su e wha cons i u es as good o bad ESG pe o mance,
esul ing in widely di e gen ESG a ings (Bo o & Pa alano,
2020, p. 64 .). This aises ques ions abou he c edibili y and
eliabili y o hese a ings as a measu e o i ms’ non- inancial
pe o mance (La cke e al., 2022, p. 6). As a esul , analys s
ely mo e on p i a e in o ma ion in addi ion o ESG a ings
(Lang & Lundholm, 1996, p. 471 .), leading o di e gen
EPS o ecas s.
The e a e also wo a gumen s ha could be made in a-
o o a nega i e ela ionship be ween ESG a ing dispe sion
and analys o ecas dispe sion. Fi s , ESG a ing dispe sion
could se e as a p oxy o he disclosu e o he e ogenous ESG
in o ma ion, which in u n can lead o a educ ion in ana-
lys o ecas dispe sion. ESG a ing agencies ac as in o ma-
ion in e media ies by ga he ing, agg ega ing and e alua ing
a i m’s public non- inancial in o ma ion. Some ESG a ing
agencies e en conduc hei own su eys, he e o e p oduc-
ing and acili a ing hei own disclosu e o ESG in o ma ion
(Scale & Kelly, 2010, p. 71). Unde he p emise ha an-
alys o ecas dispe sion e lec s he amoun o in o ma ion
commonly a ailable o analys s, o ecas dispe sion should
dec ease wi h mo e ESG in o ma ion being a ailable (Han
& Man y, 2000, p. 119). This is because i analys s sha e
a common o ecas ing model and obse e he same ESG in-
o ma ion bu ha e di e en p i a e in o ma ion, hey will
a ach less weigh o hei p i a e in o ma ion as he in o -
ma i eness o ESG in o ma ion inc eases, he eby educing
o ecas dispe sion (Lang & Lundholm, 1996, p. 471). This
iew is consis en wi h K uege e al. (2021, p. 9) who a -
gues ha as mo e and be e ESG in o ma ion is made a ail-
able, he di e si y o opinions may dec ease, and EPS o ecas
dispe sion should dec ease. Nex o he quan i y o disclo-
su e, he quali y also seems o be impo an . Swamina han
(1991, p. 40) ind ha o ecas dispe sion dec eases ollow-
ing he elease o newly manda ed segmen in o ma ion by
he SEC. Dechow e al. (1996, p. 3) ind ha o ecas dis-
pe sion inc eases ollowing alleged iola ion o gene ally ac-
cep ed accoun ing p inciples. Because ESG in o ma ion is
la gely uns anda dized, equen ly uns uc u ed, di icul o
compa e and ends o be mo e subjec i e han inancial dis-
closu es (Sipiczki, 2022, p. 6), one could a gue ha h ough
he agg ega ion and e alua ion o uns anda dized and un-
s uc u ed ESG in o ma ion, ESG a ing agencies inc ease he
quali y o ESG disclosu es, he eby educing analys o ecas
dispe sion.
Second, ESG a ing disag eemen may e lec s di e en
pe spec i es and app oaches o ESG a ing agencies, allow-
ing o a mo e comp ehensi e and nuanced unde s anding o
a i m’s ESG pe o mance, and hus educing he dispe sion
o analys s’ o ecas s. When ESG a ing agencies ha e di e -
en pe spec i es and app oaches o e alua ing a i m’s ESG
pe o mance, i leads o a mo e comp ehensi e and nuanced
unde s anding o he i m. This is because he ESG a ings be-
come mo e dispe sed, meaning hey e lec a wide ange o
iewpoin s and a g ea e amoun o unde lying da a (Scale
and Kelly 2010, p. 72; The Sus ainAbili y Ins i u e 2020, p.
44 .). As a esul , analys s ha e access o mo e in o ma ion
and can o m a mo e in o med opinion abou a i m’s inan-
cial p ospec s. This ul ima ely leads o a dec ease in o ecas
dispe sion and inc eased ag eemen among analys s. Fo his
o hold ue, hough, analys s would ha e o ha e access o
he same ESG a ings and in e p e hem in he same way
(Lang & Lundholm, 1996, p. 471 .).
In addi ion, he e a e se e al a gumen s why he e may
no be a signi ican ela ionship be ween ESG a ing dispe -
sion and analys o ecas dispe sion. Fi s , ESG disag eemen
may no ha e an e ec on analys o ecas dispe sion i ESG
a ings e lec a i m’s long- e m ESG pe o mance, while an-
alys o ecas s e lec a i m’s sho - e m p o i abili y. In his
iew, ESG a ings p o ide analys s wi h in o ma ion abou a
i m’s long e m isks and oppo uni ies (Bo o & Pa alano,
2020, p. 14). Fo example, a poo en i onmen al pe o -
mance can lead o nega i e consequences such as ines, legal
ac ion, and damage o a i m’s epu a ion, which in u n can
a ec inancial pe o mance. Whe eas, a good en i onmen-
al pe o mance can imp o e a i m’s epu a ion and mi i-
ga es he isk o egula o y sc u iny (Henisz e al., 2019, p. 3-
8). Howe e , i is di icul o p edic when hese ESG isks will
ma e ialize in he u u e. In con as , analys EPS o ecas s
a e p ojec ions o a i m’s sho - e m inancial pe o mance,
wi h a ime ho izon ypically limi ed o he nex qua e o is-
cal yea . The e o e, mos ESG isks a e unlikely o be ele an
o analys s’ EPS o ecas s and may no be used when making
EPS o ecas s. S ill, some ESG a ing agencies include con o-
e sies in o hei ESG a ings. Con o e sies a e sho - e m
epu a ional isks ha a ise om nega i e media a en ion
(Bo o & Pa alano, 2020, p. 30). Because hese con o e sies
a ec a i m’s sho - e m pe o mance, analys may conside
ESG a ings when making hei EPS o ecas s. As a esul , he
ela ionship be ween ESG a ing disag eemen and analys
o ecas dispe sion depends on whe he ESG a ings e lec
bo h sho - e m and long- e m ESG pe o mance.
Second, ESG a ings dispe sion may no a ec analys
o ecas dispe sion due o a lack in he anspa ency o ESG
a ings. ESG a ing agencies end o no ully disclose hei
ESG a ing me hodologies. In es o s lack an clea unde -
s anding abou which me ics, inpu s and weigh s ESG a -
ing agencies use in hei e alua ion, as well as he deg ee o
subjec i i y ha in hei assessmen s (B ackley e al., 2022).
This lack o anspa ency makes i di icul o analys s o
use ESG a ings as a eliable sou ce o in o ma ion o in o m
hei ea nings o ecas s. As a esul , analys s may use o he
sou ces o ESG in o ma ion beyond ESG a ings o in o m
hei ea nings o ecas s such as i m-p o ided disclosu es,
ma ke and indus y ends and speci ic news and e en s.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041790
This eliance on o he sou ces o ESG in o ma ion educes
he signi icance o ESG a ings and hei disag eemen , caus-
ing analys s o igno e ESG a ings.
Thi d, he dispe sion o ESG a ings may no be ele-
an o analys s’ o ecas s because o he backwa d-looking
da a used in ESG a ings. ESG a ing agencies mos ly use
publicly a ailable in o ma ion o access a i m’s ESG pe o -
mance. The e o e, hey can p oduce an accu a e assessmen
o a i m’s pas ESG pe o mance. Bu , analys s a e in e es ed
in o ecas ing a i m’s u u e inancial pe o mance (The Sus-
ainAbili y Ins i u e, 2020, p. 28). Sheng and The eno
(2012, p. 21) a gue ha analys s’ EPS o ecas s ep esen
ma ke pa icipan s‘ expec a ions o a i m’s u u e ea nings
p io o he elease o accoun ing da a. Pas ESG in o ma ion
may al eady be p ized in by he ma ke (Malkiel & Fama,
1970, p. 383). Also, pas pe o mance is no necessa ily a
eliable indica o o u u e pe o mance, which is why ana-
lys s use es ima es and co ec hei o ecas s on an ongoing
basis (Caps a e al., 1995, p. 74). Thus, ESG a ings may be
o limi ed use o u u e in es men decisions and a e he e-
o e no conside ed by analys s in hei o ecas s. As a esul ,
he e would be no signi ican ela ionship be ween he dis-
pe sion o ESG a ings and he dispe sion o analys s’ o e-
cas s.
Ha ing conside ed all he a gumen s, I belie e ha ana-
lys o ecas dispe sion is d i en by di e ences in he in e -
p e a ion and use o ESG a ings. Acco dingly, a posi i e as-
socia ion be ween ESG a ing dispe sion and analys o ecas
dispe sion is conside ed he mos likely hypo hesis. The e-
o e, I hypo hesize:
H1: ESG a ing disag eemen is posi i ely asso-
cia ed wi h analys o ecas dispe sion
This means ha as he dispe sion be ween ESG a ings in-
c eases, he dispe sion in analys o ecas also inc eases. To
es whe he he e is a posi i e ela ionship be ween he dis-
pe sion o ESG a ings and he dispe sion o analys o ecas s,
I conduc an empi ical analysis.
6. Empi ical S udy
6.1. Sample
I s a wi h an ini ial sample o 7,186 global public i ms
ob ained om Re ini i Eikon. The i ms a e cons i uen s o
he Ma ke WD index. The ini ial sample consis s o 71,860
i m-yea obse a ions anging om 2012 o 2022. The nec-
essa y i m da a and he in-house ESG a ings we e collec ed
om Re ini i Eikon. The ime pe iod o 10 yea s is chosen
so ha he ea nings ola ili y o he las 5 yea s can be cal-
cula ed co espondingly o each ESG a ing obse a ion. In
a i s s ep, I make su e ha he sample does no con ain
duplica es, i.e., does no con ain mo e han one obse a ion
belonging o he same i m-yea . In a second s ep, I ensu e
ha all i m-yea obse a ions a e dis inc ly a ibu able o a
single i m and a single iscal yea . Then, wi h he excep ion
o he disag eemen be ween ESG a ing agencies, I calcula e
all a iables equi ed o he empi ical analysis and emo e
missing obse a ions om he da ase .
This subsample is hen used o collec he espec i e o he
ESG a ings. In o al, I hand-collec ESG a ings om 4
p ominen ESG a ing p o ide s: MSCI, S&P, ISS, Sus ainaly -
ics. When a ESG a ing agency eleased mul iple ESG a ings
o a gi en i m yea , I collec ed he las ESG a ing p o ided
o a gi en yea . The ESG a ings collec ed a y in da a a ail-
abili y. Fo some ESG a ings, such as Sus ainaly ics and ISS,
only he la es ESG a ings o 2022 a e a ailable, while MSCI
and S&P, o example, p o ide ESG a ings co e ing a pe iod
om 2018 o 2022. Also, no all ESG a ing agencies pub-
lish hei co esponding E/S/G pilla sco es o he a ings. In
o de o ob ain a su icien ly la ge da a basis, S&P and ISS
we e included in he empi ical analysis. The ini ial in en ion
was o include only MSCI, Sus ainaly ics and Re ini i Eikon.
Howe e , he da a basis would hen ha e been oo small. Fo
he empi ical analysis, a o al o 9,577 ESG a ings om bo h
MSCI and S&P we e accessed, esul ing in 3,785 and 3,888
ESG a ings espec i ely. In he case o Sus ainaly ics and ISS,
385 ESG a ings we e accessed, esul ing in 329 and 284 a -
ings, espec i ely. In addi ion, he pilla sco es o S&P and
Sus ainaly ics o he yea 2022 we e also collec ed.
A e collec ing he ESG a ings, he wo da ase s a e
me ged and subsequen ly adjus ed o missing alues in he
ESG disag eemen calcula ion. The inal sample consis s o
3,968 i m-yea obse a ions anging om 2018 o 2022.
Table 19 in he annex shows he espec i e sample selec ion
p ocedu e. As men ioned p e iously, he sample is an in-
e na ional sample. All a ailable coun y obse a ions we e
collec ed, wi h he excep ion o he US. In ac , he inal
sample consis s o 54 unique coun ies. The h ee la ges
posi ions a e Japan, India and he Uni ed Kingdom, which
accoun o 19.5%, 8.1% and 5.2% o he sample, espec-
i ely. Table 20 in he annex shows he composi ion o he
sample by coun ies. The sample di e s om he s udy by
Kimb ough e al. (2022) in wo impo an ways. Fi s , Kim-
b ough e al. (2022, p. 5) ocus only on i ms in he US due o
he olun a y na u e o ESG in o ma ion epo ing, whe eas
ou sample includes all coun ies excluding he US1. Second,
Kimb ough e al. (2022, p. 2) collec ESG a ing in o ma-
ion om KLD (now MSCI), ASSET4 (now Re ini i Eikon),
and Vigeo Ei is (now Moody’s). The e o e, his s udy di e s
om Kimb ough in ha he ype and quan i y o di e en
ESG a ings and he coun y choice di e s. The nex chap e
add esses he esea ch design o his empi ical s udy.
6.2. Resea ch Design
To es he hypo hesis whe he he e is a posi i e associ-
a ion be ween analys o ecas dispe sion and ESG disag ee-
men , I pe o m an empi ical analysis based on an OLS e-
g ession,
AF_DISPi, =β0+β1ESG_Disag eemen i,
+βkCon olsi, +ϵi,
(1)
1All coun ies e e s o all he coun ies included in he Ma ke WD index.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1791
whe e AF_DISPi, is he dependen a iable,
ESG_Disag eemen i, he independen a iable, Con olsi,
he con ol a iables and ϵi, he e o e m. Table 21 in he
annex epo s all a iables used in he eg ession analysis.
AF_DISPi, e e s o he ela i e dispe sion be ween analys s’
o ecas s. I is calcula ed as he na u al loga i hm o he s an-
da d de ia ion o analys s’ o ecas dispe sion o annual EPS
scaled by he absolu e alue o he mean analys s’ o ecas o
i m iin yea . The absolu e alue is impo an o be ma h-
ema ically co ec since in a na u al loga i hm one canno
di ide by a nega i e numbe . O he wise, obse a ions would
be los du ing he analysis. Kimb ough e al. (2022, p. 39)
and Cui e al. (2018, p. 21) scale analys dispe sion using he
absolu e alue o he mean. Howe e , while Kimb ough e al.
(2022) use he na u al loga i hm, Cui e al. (2018) do no . In
addi ion, K uege e al. (2021, p. 51) de ine analys dispe -
sion as he s anda d de ia ion o analys s’ o ecas s di ided
by he s ock p ice o i m i in yea . Ini ially, I wan ed o use
he s anda d de ia ion o analys s’ o ecas s o calcula e ana-
lys dispe sion (AF_Dispe sion_0i, ). Bu , as can be seen in
he his og am in Figu e 5 in he annex, he obse a ions a e
no -no mally dis ibu ed using his measu e. Fo his eason,
I used he na u al loga i hm o ans o m analys o ecas
dispe sion. A e ha , he sample obse a ions o analys
dispe sion a e no mally dis ibu ed as indica ed by he bell
cu e (see Figu e 6). ESG_Disag eemen i, is he a iable o
in e es . I is calcula ed as he na u al loga i hm o he s an-
da d de ia ion o ESG a ings scaled by he absolu e alue o
he mean ESG o ecas o i m i in yea . This measu e is
used o make AF_Dispe sioni, and ESG_Disag eemen i,
compa able. In con as , Kimb ough e al. (2022, p. 39) use
he absolu e alue o he di e ence be ween he pe cen ile
ank o ESG a ings as a measu e o ESG dispe sion. They
also use he s anda d de ia ion o he pe cen ile anks o ESG
a ings as a measu e o ESG dispe sion in hei s udy, bu no
when examining he in luence on analys o ecas dispe sion
(Kimb ough e al., 2022, p. 48). Ch is ensen e al. (2021,
p. 39) use ESG disag eemen as he dependen a iable and
calcula e i using he s anda d de ia ion o ESG a ings. In
he empi ical analysis, he s anda d de ia ion o ESG a ings
is calcula ed in such a way ha i an ESG a ing is missing,
he s anda d de ia ion is s ill calcula ed o he a ailable
ESG a ings. Apa om his, a leas h ee ESG a ings a e
equi ed. To a i e a ESG_Disag eemen i, , ESG a ings
hemsel es mus i s be made compa able. Each ESG a ing
p o ide uses i s own a ing scale, which makes i di icul o
compa e ESG a ings. Re ini i (2022b, p. 3) and S&P Global
(2022, p. 3) use a pe cen ile ank sco es be ween 0 and 100,
whe e 100 ep esen s he bes sco e. Sus ainaly ics (2020,
p. 7) also uses a pe cen ile ank sco e. Bu , he pe cen ile
anks ange om 1 o 100, wi h 0 being he bes and 100
he wo s . MSCI (2022b, p. 12) uses a le e -based a ing
sys em wi h 12 ca ego ies, whe e AAA ep esen s he bes
sco e and CCC he wo s . ISS also uses a le e -based a ing
sys em. Howe e , ISS ESG (2022, p. 2) uses only 7 le e s,
wi h D- ep esen ing he wo s and A+ ep esen ing he bes
sco e .To make he ESG a ings o Re ini i Eikon, MSCI,
S&P, ISS and Sus ainaly ics compa able, I i s change he
di ec ion o Sus ainaly ics’ ESG sco e so ha 100 ep esen s
he bes and 1 he wo s . Then I s anda dize Sus ainaly ics’
ESG sco e so ha 0 ep esen s he wo s sco e. Then I di ide
he ESG sco es o he h ee ESG p o ide s by 10 o a i e a
a 10-poin a ing ank scale, which seems mo e app op ia e
gi en he lowe numbe o sco e g ades om MSCI and ISS.
A e ha , I con e he le e -based sco es om MSCI and
ISS in o nume ic sco es. Since one le e equals ze o, I di-
ide he highes possible sco e en by n−1 o a i e a he
espec i e nume ical sco es o MSCI and ISS (See Equa ion
2).
0+10
(n−1)=nume ic sco e ank (2)
I also cons uc h ee al e na i e measu es o ESG dis-
ag eemen . The i s al e na i e measu e is ESG_Disag eem-
en _3. Simila o ESG_Disag eemen , i is compu ed as he
na u al loga i hm o he s anda d de ia ion o ESG a ings
scaled by he absolu e alue o he mean ESG o ecas o i m
i in yea . The indi idual ESG a ings a e also made com-
pa able in he same way as o ESG_disag eemen . The
di e ence is ha o ESG_Disag eemen _3 only he ESG
a ings o Re ini i Eikon, MSCI and S&P a e used o calcula e
he s anda d de ia ion. Simila ly, ESG_Disag eemen _4 is
calcula ed using he ou ESG a ings om Re ini i Eikon,
MSCI, S&P and Sus ainaly ics. ESG_Disag eemen _5 uses
all i e ESG a ings. I an ESG a ing is no a ailable in a pa -
icula i m yea , he al e na i e measu es a e no calcula ed
o his pa icula i m yea . I is he e o e equi ed ha all
ESG a ings necessa y o he calcula ion a e a ailable.
In addi ion, I cons uc h ee measu es o analyse he
disag eemen among ESG a ing agencies on he E/S/G pil-
la sco es. The measu es only include he pilla sco es o
Re ini i Eikon, S&P and Sus ainaly ics, as he o he pilla
sco es a e no publicly a ailable ee o cha ge. Simila o
ESG_Disag eemen , E/S/G_disag eemen i is compu ed as
he na u al loga i hm o he s anda d de ia ion o he E/S/G
pilla sco es scaled by he absolu e alue o he mean E/S/G
o ecas o i m i in yea . E_Disag eemen cap u es he
disag eemen among ESG a ing agencies abou en i onmen-
al issues. S_Disag eemen cap u es he disag eemen While
S_Disag eemen cap u es he disag eemen be ween ESG a -
ing agencies on social issues and G_Disag eemen on go e -
nance issues. One issue is o make he pilla sco es compa a-
ble. Re ini i and S&P c ea e pilla sco es and subsequen ly
weigh hem o a i e a hei ESG a ings. The pilla sco es
o Re ini i and S&P a e di ec ly compa able. This is because
he subsequen weigh ing does no a ec he indi idual pilla
sco es. In he case o Sus ainaly ics, he sum o he indi idual
pilla sco es equals he inal ESG a ing. I is no en i ely clea
om Sus ainaly ics’ a ing me hodology how he indi idual
E/S/G pilla s a e weigh ed. To pe o m an empi ical anal-
ysis, an equal weigh ing is assumed. The pilla sco es om
Sus ainaly ics a e he e o e mul iplied by h ee o a i e a a
compa able pilla sco e.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041792
To con ol o he in luence o o he a iables, I include
se e al con ol a iables in my empi ical analysis. The con-
ol a iables a e chosen based on simila p e ious s udies.
Con ols consis s o i m size, book- o-ma ke - a io, ana-
lys ollowing, ea nings su p ise, o ecas ho izon, ea nings
ola ili y, indica o a iable o nega i e ea nings, le e age
and Zmijewski inancial dis ess sco e (Ch is ensen e al.
2021, p. 40; Behn e al. 2008, p. 333 . Hope 2003, p. 25;
Kimb ough e al. 2022, p. 38). The calcula ions o he con-
ol a iables a e gi en in Table 21 in he annex. Fi m size
is included because la ge i ms would be expec ed o ha e
a smalle dispe sion (Behn e al., 2008, p. 333). Analys
ollowing is included based on Lang and Lundholm (1996,
p. 482), who ind a posi i e associa ion be ween analys ol-
lowing and o ecas cha ac e is ics. Ea nings su p ise is also
based on Lang and Lundholm (1996, p. 489), who ind ha
la ge changes in ea nings a e ela ed o less accu a e o e-
cas s. Fo ecas ho izon is conside ed based on Chop a (1998,
p. 37), who inds ha a o ecas u he away om o he ac-
ual ea nings announcemen da e is less accu a e and mo e
dispe sed han a o ecas close o he announcemen da e.
Howe e , because many i m-yea obse a ions a e missing
o calcula e he a iable, he con ol a iable is ul ima ely no
included in he s udy. Ea nings ola ili y is included based
on K oss e al. (1990, p. 465) who ind ha i ms wi h la ge
his o ical ea nings a ia ions ha e less accu a e analys ’s
ea nings o ecas s. Va iabili y in ea nings should inc ease
he di icul y o o ecas ing, esul ing in la ge dispe sion.
The indica o a iable o nega i e ea nings, le e age and
Zmijewski inancial dis ess sco e a e included o con ol o
unce ain ies a ising om s ained inancial condi ions and
bank up cy isk. The indica o a iable o nega i e ea nings
is included based on Hwang e al. (2014, p. 29) who ind
ha analys s’ o ecas s o i ms wi h nega i e ea nings a e
on a e age less accu a e han o i ms wi h posi i e ea n-
ings. Le e age is included based on Hope (2003, p. 11)
who men ions ha highly le e ed i ms end o ha e mo e
a iable ea nings. Zmijewski (1984, p. 65-69)‘s inancial
dis ess sco e is included based on Behn e al. (2008, p.
333) who no e ha inancially dis essed i ms end o ha e
less accu a e o ecas s. The book- o-ma ke a io is included
based on he Kimb ough e al. (2022, p. 19) o con ol o
g ow h oppo uni ies ela ed o ESG. In addi ion, I u he
include indus y and yea ixed e ec s. The a iables a e
winzo ized a bo h ails a he 1% le el.
7. Empi ical Resul s
7.1. Desc ip i e S a is ics
Table 2 epo s he desc ip i e s a is ics o he indi idual
ESG a ings. As can be seen in Table 2, Re ini i , MSCI and
S&P a e ep esen ed in he sample wi h a ound 4000 ESG a -
ings each, while ISS and Sus ainaly ics a e only ep esen ed
wi h jus a ound 300 a ings.
Fu he mo e, i can be seen ha he la ges obse a ion
o ISS has a alue o 6.36, and no close o en. This is due o
he ac ha no ESG a ings be e han B ha e been assigned
o he i ms in he sample. Acco dingly, no ESG a ings om
ISS o i ms wi h excellen ESG pe o mance a e ep esen ed
in he sample. In addi ion, i can be seen ha he smalles ob-
se a ion o Sus ainaly ics has a alue o 4.63, and no close
o ze o. Thus, Sus ainaly ics is dis o ed o i ms wi h pa ic-
ula ly poo ESG pe o mance. This is because Sus ainaly ics
assigns i ms o he wo s ca ego y a a alue abo e 40. The
assigned nominal alue, hough, goes beyond 40. To a oid
dis o ions and make Sus ainaly ics compa able, one could
se he maximum obse ed alue as he uppe limi and hen
adjus he o he ESG a ings acco dingly. Bu , due o he sub-
o dina e ole o Su ainaly ics in he sample and o he obus -
ness checks, his app oach was no applied he e. S ill, i is im-
po an o be awa e o his bias o he u he cou se o his
empi ical analysis. I is also no iceable ha he ESG a ings
o S&P and ISS ha e a ela i ely low mean o 3.84 and 2.91.
Toge he wi h he also low median alues, his indica es ha
S&P and ISS gene ally assign lowe ESG a ings han Re ini-
i , MSCI and Sus ainaly ics. I is also wo h no ing ha he
s anda d de ia ion o MSCI and S&P wi h 2.68 and 2.48 a e
highe han hose o he o he ESG a ing p o ide s. This in-
dica es ha he ESG a ings o MSCI and S&P a e mo e dis-
pe sed a ound he mean. Thus, a g ea e a iabili y in ESG
a ings.
Table 3shows he co ela ion be ween he ESG a ings
o di e en ESG a ing p o ide s. The co ela ions be ween
he ESG a ings a e low. This is consis en wi h he obse a-
ions o P all and S a e S ee Global Ad iso s (see Table 18
and 19). Hence, ESG a ing p o ide s gene ally do no ag ee
abou he ESG pe o mance o i ms. The e o e, esul ing in
high le els o disag eemen among ESG a ing agencies. The
highes le els o disag eemen a e ound be ween Sus aina-
ly ics and o he ESG a ing p o ide s. Ye , some o he co -
ela ions a e no empi ically signi ican a he 1% le el. The
highes le els o ag eemen a e ound be ween Re ini i and
S&P, and S&P and ISS wi h 0.55 and 0.55, espec i ely.
Table 4and 5p esen he desc ip i e s a is ics o he em-
pi ical analysis. Table 4shows he desc ip i e s a is ics o
analys o ecas dispe sion and ESG disag eemen be o e he
ans o ma ion wi h he na u al loga i hm. Bo h a iables
a e calcula ed as he s anda d de ia ion (See Table 21). The
mean and median o AF_DISP_0 a e 49.38 (0.43). The s an-
da d de ia ion o AF_DISP_0 is 342.99. These s a is ics in-
dica e ha he e a e subs an ial a ia ions in o ecas s made
by inancial analys s.
The eason why I ans o m analys o ecas dispe sion is
ha he a iable is highly dispe sed a ound he mean, highly
skewed, and exhibi s a high posi i e ku osis. All o his can
be p oblema ic o accu acy o he hypo hesis es . Fi s , he
s anda d de ia ion is g ea e han he mean. Hence he co-
e icien o a ia ion2is mo e han one. This means ha an-
alys o ecas dispe sion exhibi s a g ea deg ee o ela i e
a iabili y. A g ea deg ee o a iabili y in he da a se is
2The coe icien o a ia ion is de ined as he a io o he s anda d de ia-
ion o he mean and is a s anda dized measu e o dispe sion.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1799
Table 9: Al e na i e measu es o ESG disag eemen (Sou ce: Own illus a ion)
(1) (2) (3) (4) (5) (6)
AF_DISP AF_DISP AF_DISP AF_DISP AF_DISP AF_DISP
ESG_Disag eemen _3 0.0517** 0.0440* 0.0349 0.0180
(0.029) (0.053) (0.107) (0.391)
ESG_Disag eemen _4 -0.0227
(0.823)
ESG_Disag eemen _5 -0.0548
(0.799)
Size 0.0458*** 0.0515*** -0.106*** -0.0641 -0.0805
(0.000) (0.000) (0.000) (0.446) (0.385)
NANA -0.0832** 0.0790** 0.139 0.107
(0.012) (0.021) (0.415) (0.642)
BTM 307.6*** 223.4*** 232.0*** 292.1* 251.8
(0.000) (0.000) (0.000) (0.105) (0.213)
Ea nings_VOL 1.14e-11 3.34e-10** 2.35e-10 3.52e-11
(0.906) (0.041) (0.856) (0.981)
Ea nings_Su p ise 1.27e-4 1.09e-4* 7.16e-4* 6.60e-4
(0.111) (0.085) (0.063) (0.138)
Le e age -4.078*** -4.495*** -1.987 -2.247
(0.000) (0.000) (0.256) (0.268)
ZMIJ 0.202*** 0.746*** 0.799*** 0.445* 0.468
(0.000) (0.000) (0.000) (0.100) (0.138)
LOSS 1.054*** 0.882*** 1.155*** 1.114***
(0.000) (0.000) (0.000) (0.001)
Yea -Fixed E ec s No No No Yes Yes Yes
Coun y Fixed E ec s No No No Yes Yes Yes
N 3,783 3,783 3,783 3,783 141 127
R-Squa e 0.001 0.096 0.228 0.392 0.428 0.405
Adjus ed R-Squa e 0.001 0.095 0.227 0.382 0.344 0.303
No e: P- alues a e below he coe icien s in b acke s. The signi icance le els a e ma ke wi h s a s: * p<0.10, ** p<0.05, *** p<0.01.
in es iga e he in luence o non- inancial disclosu e egula-
ions on he associa ion be ween ESG disag eemen and an-
alys o ecas dispe sion. Resea che s could use a di e ence-
in-di e ence design o con ol whe he he in oduc ion o a
non-disclosu e egula ion is associa ed wi h g ea e analys
o ecas dispe sion. The e a e wo non- inancial disclosu e
egula ions ha a e o pa icula in e es . On is he Eu o-
pean Union’s NFRD, which equi es all i ms co e ed by he
di ec i e o epo o he i s ime o he 2017 inancial
yea on non- inancial issues (Hankampe -Vandebulcke, 2021,
p. 4). The o he is an amendmen o he Financial Ins u-
men s Exchange Ac o Japan, which equi es lis ed i ms in
Japan wi h a cu en iscal yea -end o epo on ESG issues
by Ma ch 2023 (Tomoko & Kyoko, 2022). Bo h a e o in e -
es o a di e ence-in-di e ence design. Un o una ely, due
o he chosen ime pe iod o his sample, i is no possible o
apply such a di e ence-in-di e ence design o his empi ical
s udy.
10. Conclusion
Non- inancial ESG in o ma ion has become an inc eas-
ingly impo an sou ce o in o ma ion o he in es men
communi y, as i allows o a mo e ho ough assessmen
o a i m’s long- e m isks and oppo uni ies. One impo -
an g oup ha elies on non- inancial ESG in o ma ion a e
inancial analys s. Financial analys s use non- inancial in o -
ma ion alongside adi ional inancial in o ma ion o in o m
hei o ecas s. Howe e , ESG in o ma ion o en lack s an-
da diza ion and a e di icul o compa e. Fo his eason,
inancial analys s inc easingly ely on ESG a ing agencies
as hi d-pa y in o ma ion in e media ies o make sense o
a ailable ESG in o ma ion. ESG a ing agencies agg ega e
he a ailable ESG in o ma ion and p oduce ESG a ings by
assessing a i m’s ESG pe o mance. Those ESG a ings in-
end o in o m in es o s abou a i m’s abili y o cope wi h
long- e m isks and oppo uni ies. Howe e , ESG a ing
agencies disag ee on wha cons i u es as good ESG pe -
o mance. This leads o some imes widely di e gen ESG
a ings. The eason ESG a ing agencies end o disag ee is

R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041800
Table 10: E/S/G Disag eemen (Sou ce: Own illus a ion)
(1) (2) (3) (4) (5) (6)
AF_DISP AF_DISP AF_DISP DISP DISP AF_DISP
E_Disag eemen -0.0340 0.0224
(0.734) (0.839)
S_Disag eemen -0.0813 -0.0308
(0.536) (0.777)
G_Disag eemen -0.0893 -0.0206
(0.465) (0.862)
Size 0.0436 -0.00984 0.0360 -0.0187 0.0336 -0.0137
(0.441) (0.931) (0.541) (0.872) (0.552) (0.904)
NANA -0.277 -0.381 -0.282 -0.403 -0.281 -0.403
(0.441) (0.343) (0.248) (0.306) (0.250) (0.306)
BTM 467.1*** 386.8 423.8** 351.9 438.5*** 367.1
(0.001) (0.105) (0.014) (0.165) (0.004) (0.133)
Ea nings_VOL -6.58e-10 -3.84e-10 -4.29e-10 -2.65e-10 -4.54e-10 -3.05e-10
(0.319) (0.796) (0.541) (0.862) (0.525) (0.842)
Ea nings_Su p ise 0.0004** 0.0003 0.0004** 0.0003 0.0004** 0.0003
(0.024) (0.155) (0.019) (0.151) (0.029) (0.157)
Le e age -2.142 -1.314 -2.356 -1.390 -2.238 -1.350
(0.470) (0.585) (0.432) (0.565) (0.444) (0.575)
ZMIJ 0.322 0.328 0.354 0.329 0.327 0.322
(0.479) (0.389) (0.443) (0.387) (0.465) (0.398)
LOSS 0.926*** 0.968* 0.929*** 0.984* 0.910*** 0.977**
(0.006) (0.093) (0.004) (0.088) (0.008) (0.090)
Yea -Fixed E ec s No Yes No Yes No Yes
Coun y Fixed E ec s No Yes No Yes No Yes
N 83 79 83 79 83 79
R-Squa e 0.236 0.413 0.241 0.414 0.241 0.413
Adjus ed R-Squa e 0.142 0.262 0.148 0.262 0.148 0.262
No e: P- alues a e below he coe icien s in b acke s. The signi icance le els a e ma ke wi h s a s: * p<0.10, ** p<0.05, *** p<0.01.
because hey di e in scope, weigh ing, and measu emen o
ESG in o ma ion. Because ESG a ing agencies compe e wi h
each o he o ma ke sha e, he e is no single app oach o
ESG a ing me hodologies. In addi ion, ESG a ing me hod-
ologies a e no ully anspa en . As inancial analys s seek
o unde s and ESG a ings and hei unde lying da a, hey
ob ain hei own p i a e in o ma ion, leading o di e gen
opinions abou a i m’s long- e m isks and oppo uni ies.
The objec i e o his mas e hesis was o empi ically in-
es iga e he in luence o ESG a ing disag eemen on an-
alys o ecas dispe sion in an in e na ional se ing. P io
esea ch based on Kimb ough e al. (2022) ound a posi-
i e associa ion be ween ESG a ing disag eemen and an-
alys o ecas dispe sion o i ms in he US. The i s eg es-
sion model wi hou con ol a iables shows ha he e is in-
deed a s a is ically signi ican ela ionship be ween ESG dis-
ag eemen and analys o ecas dispe sion. The coe icien o
ESG disag eemen is 0.0557 and is s a is ically signi ican a
he 5% le el. Because bo h ESG disag eemen and analys
o ecas dispe sion we e ans o med wi h he na u al loga-
i hm, a 1% inc ease in ESG disag eemen is associa ed wi h
a 5.57% inc ease in analys o ecas dispe sion. Howe e ,
he i s model has a low R-squa ed alue and he e o e does
no p oduce p edic ions ha a e easonably p ecise. The in-
oduc ion o con ol a iables inc eases he p edic abili y o
he empi ical model. The second ( hi d) model wi h h ee
(eigh ) con ol a iables a e also s a is ically signi ican a he
5% (10%) le el. A 1% inc ease in ESG disag eemen is as-
socia ed wi h a 5,34% (4,08%) inc ease in analys o ecas
dispe sion. Howe e , he inclusion o yea and coun y ixed
e ec s wi hin he eg ession model leads o a no able shi in
he na u e o he ob ained esul s, yielding s a is ically non-
signi ican indings. To ensu e he alidi y and eliabili y o
hese indings, I employ se e al obus ness checks. Fi s , I
add ess he p esence o skewed dis ibu ions in some o he
con ol a iables by applying a na u al loga i hm ans o -
ma ion. This ans o ma ion helps o con ol o ou lie s ha
migh in luence he eg ession model. A e implemen ing
his adjus men , he esul s emain consis en wi h he main
indings, p o iding addi ional con idence in he obus ness
o he indings. Second, I exclude inancial i ms and u ili-
ies om he analysis due o hei undamen ally dis inc na-
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1801
u e om p i a e i ms. Addi ionally, eal es a e i ms a e
excluded due o hei unusually high le els o le e age. De-
spi e hese exclusions, he esul s emain consis en wi h he
main indings, ein o cing he s abili y o he obse ed ela-
ionships. Thi d, I examine he ime consis ency o he ela-
ionship be ween ESG disag eemen and analys o ecas dis-
pe sion. Howe e , he e is a de ia ion om he main esul s
in he yea s 2018 and 2019. This incon-sis ency p omp s u -
he in es iga ion in o he po en ial ac o s d i ing he a i-
a ion and unde sco es he need o cau ious in e p e a ion
o he mo e dis an esul s. Fou h, al e na i e measu es o
ESG disag eemen a e employed o assess hei impac on he
esul s. Despi e hese a ia ions in measu emen , he main
indings emain unchanged, indica ing obus ness in he e-
la ionship be ween ESG disag eemen and analys o ecas
dispe sion. Fi h, I explo e he indi idual in luence o en i-
onmen al, social, and go e nance ac o s on analys o ecas
dispe sion. Howe e , due o he small sample size, he esul s
a e no s a is ically signi ican and canno be conside ed ep-
esen a i e. O e all, he empi ical esul s emain obus a e
pe o ming se e al obus ness checks. Hence, no de ini i e
conclusion can be d awn ega ding he in luence o ESG dis-
ag eemen on he dispe sion o analys s’ o ecas s.
These indings hold signi ican implica ions o p ac i ion-
e s, pa icula ly hose in ol ed in he in es men indus y, as
hey challenge he ele ance o non- inancial ESG in o ma-
ion p o ided by ESG a ing agencies in in o ming inancial
analys s’ o ecas s. This mas e hesis also p esen s oppo -
uni ies o u he esea ch in he ield. Po en ial a enues
include in es iga ing he in luence o en i onmen al, social,
and go e nance (ESG) c i e ia on analys o ecas dispe sion,
o employing a di e ence-in-di e ence design o s udy he
e ec s o new non- inancial disclosu e equi emen s. Fo in-
s ance, esea che s could explo e he impac o egula o y
amewo ks like he Eu opean Non-Financial Repo ing Di-
ec i e (NFRD), which p eda es he sample pe iod co e ed in
his s udy, o he ecen amendmen o he Financial Ins u-
men s Exchange Ac o Japan which manda es lis ed i ms in
Japan o include ESG in o ma ion in hei cu en iscal yea
epo ing by Ma ch 2023.
Re e ences
Aba banell, J. S., Lanen, W. N., & Ve ecchia, R. E. (1995). Analys s’ o ecas s
as p oxies o in es o belie s in empi ical esea ch. Jou nal o
Accoun ing and Economics,20(1), 31–60.
Acke , L. F. (1997). P io Unce ain y, Analys Bias, and Subsequen Abno -
mal Re u ns. The Jou nal o Financial Resea ch,20(2), 263–273.
Ajinkya, B. B., A iase, R. K., & Gi , M. J. (1991). Volume o T ading and he
Dispe sion in Financial Analys s’ Ea nings Fo ecas s. The Accoun -
ing Re iew,66(2), 389–401.
Allen, K. (2018). Lies, damned lies and ESG a ing me hodologies. Financial
Times. Re ie ed Sep embe 30, 2022, om h ps://www. .com/
con en /2e49171b-a018-3c3b-b66b-81 d7a170ab5
Asqui h, P., & Weiss, L. A. (2016). Lessons in Co po a e Finance. New Je sey.
A hanassakos, G., & Kalimipalli, M. (2003). Analys Fo ecas Dispe sion and
Fu u e S ock Re u n Vola ili y. Qua e ly Jou nal o Business and
Economics,42(1 & 2), 57–78.
A amo , D., Cho dia, T., Jos o a, G., & Philipo , A. (2009). Dispe sion in
analys s’ ea nings o ecas s and c edi a ing. Jou nal o Financial
Economics,91(1), 83–101.
Ba e ield, R. M., & Comiskey, E. E. (1975). The Accu acy o Analys s’ Fo e-
cas s o Ea nings pe Sha e. Jou nal o Business Resea ch,3(3),
241–252.
Ba ke , R. G. (1998). The ma ke o in o ma ion—e idence om inance
di ec o s, analys s and und manage s. Accoun ing and Business
Resea ch,29(1), 3–20.
Ba on, O. E., Kim, O., Lim, S. C., & Douglas, E. S. (1998). Using Analys s’
Fo ecas s o Measu e P ope ies o Analys s’ In o ma ion En i on-
men . The Accoun ing Re iew,73(4), 421–433.
Ba on, O. E., S and o d, M. H., & Yu, Y. (2010). Fu he E idence on he Re-
la ion be ween Analys s’ Fo ecas Dispe sion and S ock Re u ns.
Con empo a y Accoun ing Resea ch,26(2), 329–357.
Ba y, C. B., & Jennings, R. H. (1992). In o ma ion and Di e si y o Ana-
lys Opinion. The Jou nal o Financial and Quan i a i e Analysis,
27(2), 169–183.
Becke s, S., S elia os, M., & Thomson, A. (2004). Bias in Eu opean Analys s’
Ea nings Fo ecas s. Financial Analys s Jou nal,60(2), 74–85.
Behn, B. K., Choi, J.-H., & Kang, T. (2008). Audi Quali y and P ope ies o
Analys Ea nings Fo ecas s. The Accoun ing Re iew,83(2), 327–
349.
Belnap, A. (2022). The e ec o in e media y co e age on disclosu e: E -
idence om a andomized ield expe imen [Co ec ed p oo in
p ess].Jou nal o Accoun ing and Economics.
Bens on, G. J. (1973). Associa ionRequi ed Disclosu e and he S ock Ma -
ke : An E alua ion o he Secu i ies Exchange Ac o 1934. The
Ame ican Economic Re iew,63(1), 132–155.
Be g, F., Kölbel, J. F., & Rigobon, R. (2022). Agg ega e Con usion: The Di-
e gence o ESG Ra ings. Re iew o Finance,26(6), 1315–1344.
Be k, J., & an Binsbe gen, J. H. (2021). The Impac o Impac In es ing
[Wo king Pape No. 3981, S an o d Uni e si y G adua e School
o Business].
Beunza, D., & Ga ud, R. (2007). Calcula o s, Lemmings o F ame-Make s?
The In e media y Role o Secu i ies Analys s. The Sociological Re-
iew,55(2), 13–39.
Bloombe g. (2022). ESG May Su pass $41 T illion Asse s in 2022, Bu No
Wi hou Challenges, Finds Bloombe g In elligence. P ess Release.
Re ie ed Decembe 6, 2022, om h ps : / / www. bloombe g .
com/company/p ess/esg- may- su pass- 41- illion- asse s- in-
2022-bu -no -wi hou -challenges- inds-bloombe g-in elligence/
BNP Pa ibas. (2019). The ESG Global Su ey 2019. Re ie ed Decembe 5,
2022, om h ps : / / secu i ies. cib .bnppa ibas /app /uploads/
si es/3/2019/08/esg-global-su ey-en-2019.pd
Bo o, R., & Pa alano, R. (2020). ESG In es ing: P ac ices, P og ess and
Challenges. Re ie ed Decembe 5, 2022, om www.oecd.o g/
inance/ESG-In es ing-P ac ices-P og ess-and-Challenges.pd
B ackley, A., B ock, E. K., & Nelson, J. (2022). Ra ing he Ra e s Ye Again:
Inc easing ESG Sc u iny Makes Cu en Ra e he Ra e s S udy
E en Mo e C ucial. Sus ainabili y. Re ie ed Decembe 15, 2022,
om h ps : / / www. sus ainabili y. com / hinking / a ing - he -
a e s- ye - again- inc easing- esg- sc u iny- makes- cu en - a e-
he- a e s-s udy-e en-mo e-c ucial/
B own, R., Chan, H. W. H., & Ho, Y. K. (2009). Analys s’ ecommenda ions:
om which signal does he ma ke ake i s lead? Re iew o Quan-
i a i e Finance and Accoun ing,33, 91–111.
Bya d, D., Li, Y., & Yu, Y. (2011). The E ec o Manda o y IFRS Adop ion on
Financial Analys s’ In o ma ion En i onmen . Jou nal o Accoun -
ing Resea ch,49(1), 69–96.
Capi al G oup. (2021). ESG Global S udy 2021. Re ie ed Janua y 12,
2023, om h ps : / / www. capi alg oup . com /con en /dam /
cgc/ enan s /eacg /esg/global- s udy /esg- global- s udy - ull -
epo (en).pd
Caps a , J., Paudyal, K., & Rees, W. (1995). The Accu acy and Ra ionali y o
Ea nings Fo ecas s by UK Analys s. Jou nal o Business Finance &
Accoun ing,22(1), 67–85.
CFA Ins i u e. (2020). Fu u e o Sus ainabili y in In es men Managemen :
F om Ideas o Reali y. Re ie ed No embe 5, 2022, om h ps:
//www.c ains i u e.o g/-/media/documen s/su ey/ u u e-o -
sus ainabili y.pd
Cha e ji, A., & Le ine, D. (2006). B eaking Down he Wall o Codes: E alu-
a ing Non-Financial Pe o mance Measu emen . Cali o nia Man-
agemen Re iew,48(2), 29–51.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041802
Cho, S.-Y., Lee, C., & P ei e , R. J. (2013). Co po a e Social Responsibili y
Pe o mance In o ma ion and In o ma ion Asymme y. Jou nal o
Accoun ing and Public Policy,32(1), 71–83.
Chop a, V. K. (1998). Why So Much E o in Analys s’ Ea nings Fo ecas s?
Financial Analys s Jou nal,54(6), 35–42.
Ch is ensen, D., Se a eim, G., & Sikochi, A. (2021). Why is Co po a e Vi ue
in he Eye o The Beholde ? The Case o ESG Ra ings. Accoun ing
Re iew,97(1), 147–175.
Cision. (2020). Mo nings a o Acqui e Sus ainaly ics and Expand Access
o ESG Resea ch, Da a, and Analy ics o In es o s Wo ldwide.
PR Newswi e. Re ie ed No embe 25, 2022, om h ps : / /
www.p newswi e.com/news- eleases/mo nings a - o-acqui e-
sus ainaly ics- and- expand- access- o- esg- esea ch- da a- and-
analy ics- o -in es o s-wo ldwide-301044196.h ml
Collischon, M., & Ebe l, A. (2020). Le ’s Talk Abou Fixed E ec s: Le ’s Talk
Abou All he Good Things and he Bad Things. KZ SS Kölne
Zei sch i ü Soziologie und Sozialpsychologie,72(2), 290–299.
Co nell, B. (2020). ESG p e e ences, isk and e u n. Eu opean Financial
Managemen ,27(1), 12–19.
Cowles, A. (1933). Can S ock Ma ke Fo ecas e s Fo ecas ? Econome ica,
1(3), 309–324.
Cui, J., Jo, H., & Na, H. (2018). Does Co po a e Social Responsibili y A ec
In o ma ion Asymme y? Jou nal o Business E hics,148(3), 549–
572.
D. E. Shaw. (2022). Keep he Change: Analyzing he Inc ease in ESG Ra -
ings o U.S. Equi ies. Re ie ed Janua y 3, 2023, om h ps :
//www.deshaw.com/asse s/a icles/DESCO_Ma ke _Insigh s_
ESG_Ra ings_20220408.pd
De Bond , W. F. M., & Fo bes, W. P. (1999). He ding in analys ea nings
o ecas s: e idence om he Uni ed Kingdom. Eu opean Financial
Managemen ,5(2), 143–163.
Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1996). Causes and Conse-
quences o Ea nings Manipula ion: An Analysis o Fi ms Subjec
o En o cemen Ac ions by he SEC. Con empo a y Accoun ing Re-
sea ch,13(1), 1–36.
Deloi e. (2021). ESG Ra ings: do hey add alue? How o ge p epa ed?
Deloi e. Re ie ed No embe 28, 2022, om h ps://www2.
deloi e.com/ce/en/pages/abou -deloi e/a icles/esg- a ings-
do- hey-add- alue.h ml
Deloi e. (2022). Wha is ESG? Deloi e. Re ie ed Decembe 5, 2022,
om h ps://www2.deloi e.com/hu/en/pages/ene gy- and-
esou ces/a icles/e.h ml
Die he , K. B., Malloy, C. J., & Sche bina, A. (2002). Di e ences o Opinion
and he C oss Sec ion o S ock Re u ns. The Jou nal o Finance,
57(5), 2113–2141.
Dillenbu g, S., G eene, T., & E ekson, H. (2003). App oaching Socially Re-
sponsible In es men wi h a Comp ehensi e Ra ings Scheme: To-
al Social Impac . Jou nal o Business E hics,43(3), 167–177.
Doyle, T. M. (2018). Ra ings Tha Don’ Ra e. Washing on D.C. Re ie ed
No embe 4, 2022, om h ps://acc co pgo .o g/wp-con en /
uploads/2018/07/ACCF_Ra ingsESGRepo .pd
D eman, D. N., & Be y, M. A. (1995). Analys Fo ecas ing E o s and Thei
Implica ions o Secu i y Analysis. Financial Analys s Jou nal,
51(3), 30–41.
D empe ic, S., Klein, C., & Zwe gel, B. (2020). The In luence o Fi m Size on
he ESG Sco e: Co po a e Sus ainabili y Ra ings Unde Re iew.
Jou nal o Business E hics,167(2), 333–360.
Dyck, A., Mo se, A., & Zingales, L. (2006). Who Blows he Whis le on Co -
po a e F aud? The Jou nal o Finance,65(6), 2213–2253.
Esc ig-Olmedo, E., Fe nández-Izquie do, M. A., Fe e o-Fe e o, I., Ri e a-
Li io, J. M., & Muñoz-To es, M. J. (2019). Ra ing he Ra e s:
E alua ing how ESG Ra ing Agencies In eg a e Sus ainabili y
P inciples. Sus ainabili y,11(3), 1–16.
Eu opean Pa liamen and o he Council. (2014). Di ec i e 2014/95/EU o
he Eu opean Pa liamen and o he Council o 22 Oc obe 2014.
L 330, 1–9. O icial Jou nal o he Eu opean Union.
Fama, F. E., & F ench, K. R. (1992). The C oss-Sec ion o Expec ed S ock
Re u ns. The Jou nal o Finance,47(2), 427–465.
Fish, A. J. M., Kim, D. H., & Venka aman, S. (2019). The ESG Sac i ice.
Re ie ed Decembe 5, 2022, om h p://dx.doi.o g/10.2139/
ss n.3488475
Fishe , K. (2019). No embe 1867: The In en ion o he S ock Ticke .
Ken Fishe . Re ie ed No embe 4, 2022, om h ps://www.
ken ishe .com/ma ke - his o y/no embe -1867- he- in en ion-
o - he-s ock- icke
F ankel, R., & Li, X. (2004). Cha ac e is ics o a i m’s in o ma ion en i-
onmen and he in o ma ion asymme y be ween inside s and
ou side s. Jou nal o Accoun ing and Economics,37(2), 229–259.
Gell, S. (2011). De e minan s o Ea nings Fo ecas E o , Ea nings Fo ecas Re-
ision and Ea nings Fo ecas Accu acy [Doc o al disse a ion, Uni-
e si y o Cologne].
Giese, G., Lee, L.-E., Melas, D., Nagy, Z., & Nishikawa, L. (2019). Founda-
ions o ESG In es ing: How ESG A ec s Equi y Valua ion, Risk,
and Pe o mance. The Jou nal o Po olio Managemen ,45(5),
69–83.
Gilbe , M. (2021). The Rising Cos o In es ing Responsibly. Bloombe g
Opinion. Re ie ed No embe 4, 2022, om h ps : / / www.
bloombe g.com/opinion/a icles/2019-01-11/ he- ising-cos -
o -esg-and-socially- esponsible-in es ing?leadSou ce=u e i y%
20wall
Gi oly, D., & Lakonishok, J. (1980). Financial Analys s’ Fo ecas s o Ea n-
ings. Jou nal o Banking and Finance,4(2), 221–233.
Global Sus ainable In es men Alliance (GSIA). (2021). Global Sus ainable
In es men Re iew 2020. Re ie ed No embe 4, 2022, om
h p://www.gsi- alliance.o g/wp-con en /uploads/2021/08/
GSIR-20201.pd
Google T ends. (2022). ESG. Google T ends. Re ie ed Decembe 10, 2022,
om h ps:// ends.google.com/ ends/explo e?da e= oday%
205-y&q=esg
Go e nance & Accoun abili y Ins i u e. (2022). Sus ainabili y Repo ing
in Focus. Go e nance & Accoun abili y Ins i u e. Re ie ed
Decembe 12, 2022, om h p : / / www. ga - ins i u e . com /
ileadmin/ga_ins i u e/images/FlashRepo s/2022/G_A-2022-
Sus ainabili y_T ends_Repo .pd ? go_ee= mI ei J13 n9Ym%
2F P4nyNSYFm MikCwlKFARSZoYAo%3D
G aham, B., & Dodd, D. L. (1934). Secu i y Analysis (1s ).
G aham, B., & Dodd, D. L. (2009). Secu i y Analysis (6 h).
G aham, J. R. (1999). He ding among in es men newsle e s: Theo y and
e idence. The Jou nal o Finance,54(1), 237–268.
G aham, J. R., Ha ey, C. R., & Rajgopal, S. (2005). The Economic Implica-
ions o Co po a e Financial Repo ing. Jou nal o Accoun ing and
Economics,40(1-3), 3–73.
G oysbe g, B., Healy, P., & Chapman, C. (2008). Buy-Side s. Sell-Side An-
alys s’ Ea nings Fo ecas s. Financial Analys s Jou nal,64(4), 25–
39.
Han, B. H., & Man y, D. (2000). The Implica ions o Dispe sion in Analys s’
Ea nings Fo ecas s o Fu u e ROE and Fu u e Re u ns. Jou nal
o Business Finance & Accoun ing,27(1-2), 99–125.
Hankampe -Vandebulcke, N. (2021). Non- inancial Repo ing Di ec i e. Re-
ie ed Feb ua y 3, 2023, om h ps://www.eu opa l.eu opa.eu/
RegDa a/e udes/BRIE/2021/654213/EPRS_BRI(2021)654213_
EN.pd
Healy, P. M., & Palepu, K. (2001). In o ma ion asymme y, co po a e dis-
closu e, and he capi al ma ke s: A e iew o he empi ical dis-
closu e li e a u e. Jou nal o Accoun ing and Economics,31(1-3),
405–440.
Henisz, W., Kolle , T., & Nu all, R. (2019). Fi e ways ha ESG c ea es alue.
Re ie ed No embe 5, 2022, om h p://dln.jaipu ia.ac.in:
8080/jspui/bi s eam/123456789/2319/1/Fi e- ways- ha -
ESG-c ea es- alue.pd
Hespenheide, E. J., & Koehle , D. A. (2013). Disclosu e o long- e m busi-
ness alue – Wha ma e s? Re ie ed Decembe 15, 2022, om
h ps://www2.deloi e.com/con en /dam/insigh s/us/a icles/
disclosu e- o - long- e m- business- alue/DUP150_Repo ing_
Wha _Ma e s.pd
Higgins, H. N., & Sai o, Y. (2007). Do Analys s Play an In o ma ion In e me-
dia y Role? E idence om High and Low In angible Fi ms. SSRN,
1–28.
Hi shlei e , D., Lim, S. S., & Teoh, S. H. (2009). D i en o dis ac ion: Ex a-
neous e en s and unde eac ion o ea nings news. The Jou nal o
Finance,64(5), 2289–2325.
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-1804 1803
Hong, H., & Kubik, J. D. (2003). Analyzing he Analys s: Ca ee Conce ns
and Biased Ea nings Fo ecas s. Jou nal o Finance,58(1), 313–
351.
Hong, H., Kubik, J. D., & Solomon, A. (2000). Secu i y Analys s’ Ca ee Con-
ce ns and He ding o Ea nings Fo ecas s. The RAND Jou nal o
Economics,31(1), 121–144.
Hope, O.-K. (2003). Accoun ing Policy Disclosu es and Analys s’ Fo ecas s.
Con empo a y Accoun ing Resea ch,20(2), 295–321.
Hou, J., Zhao, S., & Yang, H. (2020). Indi idual analys s, s ock e u n syn-
ch onici y and in o ma ion e iciency. In e na ional Re iew o Fi-
nancial Analysis,71, 1–39.
Hwang, L. S., Jan, C.-L., & Basu, S. (2014). Loss Fi ms and Analys s’s Ea n-
ings Fo ecas E o s. The Jou nal o Financial S a emen Analysis,
1(2), 18–30.
Imho , E. A., & Lobo, G. J. (1992). The E ec o Ex An e Ea nings Unce -
ain y on Ea nings Response Coe icien s. The Accoun ing Re iew,
67(2), 427–439.
ISS ESG. (2022). ESG Co po a e Ra ing. Re ie ed Janua y 22, 2023, om
bi .ly/3QYA iL
Jacobson, T. C. (1997). F om P ac ice o P o ession: A His o y o he Financial
Analys s Fede a ion and he In es men P o ession.
Jennings, R. H. (1985). Reac ion o Financial Analys s o Managemen Ea n-
ings Fo ecas s.
Johnson, T. C. (2004). Fo ecas Dispe sion and he C oss Sec ion o Expec ed
Re u ns. The Jou nal o Finance,59(5), 1957–1978.
Jones, C. P., & Li zenbe ge , R. H. (1970). Qua e ly Ea nings Repo s and
In e media e S ock P ice T ends. The Jou nal o Finance,25(1),
143–148.
Kemp , A., & Os ho , P. (2007). The e ec o socially esponsible in es ing on
po olio pe o mance. Eu opean Financial Managemen ,13(5),
908–922.
Kendall, M. G., & Hill, A. B. (1953). The Analysis o Economic Time-Se ies-
Pa I: P ices. Jou nal o he Royal S a is ical Socie y,116(1), 11–
34.
Ke chum, M. D. (1967). Is inancial analysis a p o ession? Financial Analys s
Jou nal,23(6), 33–37.
Keynes, J. M. (2018). The Gene al Theo y o Employmen , In e es , and Money.
Kimb ough, M., Wang, X., Wei, S., & Zhang, J. (2022). Does olun a y
ESG epo ing esol e disag eemen among ESG a ing agencies?
[Fo hcoming].Eu opean Accoun ing Re iew.
Kle ke, T. (2013). New De e minan s o Analys s’ Ea nings Fo ecas Accu acy
[Doc o al disse a ion, Uni e si y o Cologne].
Kno -Ce ina, K. (2011). Financial Analysis. In C. Camic, N. G oss, & M.
Lamon (Eds.), Social Knowledge in he Making (pp. 405–441).
Kolle , T., Goedha , M., & Wessels, D. (2020). Valua ion: Measu ing and
Managing he Value o Companies (7 h).
Ko san onis, S., & Se a eim, G. (2019). Fou Things No One Will Tell You
Abou ESG Da a. Jou nal o Applied Co po a e Finance,31(2), 50–
58.
K oss, W., Ro, B., & Sch oede , D. (1990). Ea nings Expec a ions: The Ana-
lys s’ In o ma ion Ad an age. The Accoun ing Re iew,65(2), 461–
476.
K uege , P., Sau ne , Z., Tang, D. Y., & Zhong, R. (2021). The E ec s o
Manda o y ESG Disclosu e A ound he Wo ld. (754), 1–66.
LaBella, M. J., Sulli an, L., Russel, J., & No iko , D. (2019). The De il is
in he De ails: The Di e gence in ESG Da a and Implica ions o
Responsible In es ing. Re ie ed Janua y 2, 2023, om h ps://
bi .ly/3WY6n7N
Lang, M. H., & Lundholm, R. J. (1996). Co po a e Disclosu e Policy and
Analys Beha io . The Accoun ing Re iew,71(4), 467–492.
La cke , D. F., Pomo ski, L., Tayan, B., & Wa s, E. M. (2022). ESG Ra ings:
A Compass wi hou Di ec ion [Fo hcoming].
Leins, S. (2018). S o ies o Capi alism: Inside he Role o Financial Analys s.
Lu, S. (2021). Algo i hmic Opaci y, P i a e Accoun abili y, and Co po a e
Social Disclosu e in he Age o A i icial In elligence. Vande bil
Jou nal o En e ainmen & Technology Law,23(1), 99–159.
Malkiel, B. G., & Fama, E. F. (1970). E icien capi al ma ke s: A Re iew o
Theo y and Empi ical Wo k. The Jou nal o Finance,25(2), 383–
417.
Ma s, F. (1998). Wi sind alle Sehe : Die P axis de Ak ienanalyse.
McClu e, B. (2022). Ea nings Fo ecas s: A P ime . In es opedia. Re ie ed
Decembe 18, 2022, om h ps : / / www. in es opedia . com /
a icles/s ocks/06/ea nings o ecas s.asp
McDonald, C. L. (1973). An Empi ical Examina ion o he Reliabili y o
Published P edic ions o Fu u e Ea nings. The Accoun ing Re iew,
48(3), 502–510.
McKibben, W. (1972). Econome ic Fo ecas ing o Common S ock In es -
men Re u ns: A New Me hodology using Fundamen al Ope a ing
Da a. The Jou nal o Finance,27(2), 371–380.
Mo nings a . (2021). En e p ise Sus ainabili y Repo . Re ie ed No em-
be 25, 2022, om h ps : / / asse s . con en s ack . io / 3 /
asse s /bl 4eb669caa7dc65b2 /bl 3ad3261518bad875 /2021 _
En e p ise_Sus ainabili y_Repo .pd
MSCI. (2019). MSCI o S eng hen Clima e Risk Capabili y wi h Acquisi ion
o Ca bon Del a. Re ie ed No embe 26, 2022, om h ps://i .
msci.com/node/15706/pd
MSCI. (2022a). ESG Ra ings Key Issue F amewo k. Re ie ed No embe 26,
2022, om h ps://www.msci.com/ousolu ions/esg-in es ing/
esg- a ings/esg- a ings-key-issue- amewo k
MSCI. (2022b). MSCI ESG Ra ings Me hodology. Re ie ed No embe 4,
2022, om h ps : / / www. msci . com /documen s /1296102 /
21901542/ESG-Ra ings-Me hodology-Exec-Summa y.pd
MSCI ESG Resea ch. (2022). SEC In es men Ad ise Fi m Summa y -
Pa 2 B ochu es. Re ie ed No embe 26, 2022, om h ps :
// iles.ad ise in o.sec.go /IAPD/Con en /Common/c d_iapd_
B ochu e.aspx?BRCHR_VRSN_ID=762070
Nagy, Z., Kassam, A., & Lee, L.-E. (2016). Can ESG Add Alpha? An Analysis
o ESG Til and Momen um S a egies. The Jou nal o In es ing,
25(2), 113–124.
Nagy, Z., Lee, L.-E., & Giese, G. (2020). ESG Ra ings: How he Weigh ing
Scheme A ec ed Pe o mance. MSCI Blog. Re ie ed Decembe
18, 2022, om h ps://www.msci.com/www/blog-pos s/esg-
a ings-how- he-weigh ing/01944696204
Nine y One. (2022). Nine y One su ey inds Eu opean und indus y o e ly
elian on ESG sco es. Nine y One. Re ie ed Decembe 7, 2022,
om h ps : / / nine yone . com /en /uni ed - s a es /news oom /
nine y-one-su ey- inds-eu opean- und-indus y-o e ly- elian -
on-esg-sco es
Nomu a. (2022). How A i icial In elligence is Reshaping ESG Ra ings. Re-
ie ed Decembe 16, 2022, om h ps : / / d1q wzw6aggd4h .
cloud on .ne /abou /ESG-E alua ion-using-AI-A4.pd
No e hic. (2014). O e iew o ESG Ra ing Agencies. Re ie ed No embe
25, 2022, om h ps : / / www. no e hic . / ileadmin /use _
upload/ x_ausyno e hice udes/pd _comple s/2014_O e iew-
o -ESG- a ing-agencies.pd
Poon, M. (2012). Ra ing Agencies. In K. K. Ce ina & A. P eda (Eds.), The Ox-
o d Handbook o The Sociology o Finance (pp. 407–459). Ox o d
Uni e si y P ess.
P all, K. (2021). ESG Ra ings: Na iga ing Th ough he Haze. CFA Ins i u e
Blogs. Re ie ed Janua y 4, 2023, om h ps://blogs.c ains i u e.
o g/in es o /2021/08/10/esg- a ings-na iga ing- h ough- he-
haze/
P eda, A. (2006). Socio-Technical Agency in Financial Ma ke s: The Case o
he S ock Ticke . Social S udies o Science,36(5), 753–782.
P inciples o Responsible In es men (PRI). (2022). Annual Repo . Re-
ie ed Decembe 5, 2022, om h ps : / / dw yzx6upklss .
cloud on . ne /Uploads /b/ /m/p i _ annual _ epo _ 2022 _
689047.pd
Raghunandan, A., & Rajgopal, S. (2022). Do ESG unds make s akeholde -
iendly in es men s? Re iew o Accoun ing S udies,27, 822–863.
Re ini i . (2022a). Abou Us. Re ini i . Re ie ed No embe 28, 2022, om
h ps://www. e ini i .com/en/abou -us#wha -we-do
Re ini i . (2022b). En i onmen al, Social and Go e nance Sco es om Re-
ini i . Re ie ed No embe 28, 2022, om h ps://www. e ini i .
com/con en /dam/ma ke ing/en_us/documen s/me hodology/
e ini i -esg-sco es-me hodology.pd
Re ini i . (2022c). Re ini i ESG company sco es. Re ini i . Re ie ed
No embe 28, 2022, om h ps : / / www. e ini i . com /en /
sus ainable- inance/esg-sco es
Reu e s. (2018). Thomson Reu e s closes deal wi h Blacks one. Reu e s. Re-
ie ed No embe 28, 2022, om h ps://www. eu e s.com/
R. Spi a /Junio Managemen Science 9(3) (2024) 1769-18041804
a icle/us- homson eu e s- m- a- blacks one/ homson- eu e s-
closes-deal-wi h-blacks one-idUSKCN1MB3PY
Riles, A. (2006). Documen s: A i ac s o mode n knowledge. Uni e si y o
Michigan P ess.
Riles, A. (2011). Colla e al knowledge: Legal easoning in he global inancial
ma ke s. Uni e si y o Chicago P ess.
Ryan, P., & Ta le , R. J. (2004). A e Economically Signi ican S ock Re u ns
and T ading Volumes D i en by Fi m-speci ic News Releases?
Jou nal o Business Finance & Accoun ing,31(1& 2), 49–82.
Scale , S., & Kelly, T. F. (2010). CSR Ra ing Agencies: Wha is Thei Global
Impac ? Jou nal o Business E hics,94, 69–88.
Scha s ein, D. S., & S ein, J. C. (1990). He d beha io and in es men . The
Ame ican economic e iew, 465–479.
Secu i ies and Exchange Commission (SEC). (2010). Analyzing Analys
Recommenda ions. SEC. Re ie ed No embe 5, 2022, om
h ps://www.sec.go / m/ epo spubs/in es o - publica ions/
in es o pubsanalys sh m.h ml
Secu i ies and Exchange Commission (SEC). (2022). SEC P oposes o En-
hance Disclosu es by Ce ain In es men Ad ise s and In es men
Companies Abou ESG In es men P ac ices. Re ie ed Decembe
12, 2022, om h ps://www.sec.go /news/p ess- elease/2022-
92
Sheng, X., & The eno , M. (2012). A new measu e o ea nings o ecas un-
ce ain y. Jou nal o Accoun ing and Economics,53(1-2), 21–33.
Sind eu, J., & Ken , S. (2018). Why I ’s So Ha d o Be an ‘E hical’ In es o .
The Wall S ee Jou nal. Re ie ed No embe 4, 2022, om h ps:
//www.wsj.com/a icles/why- i s- so- ha d- o- be- an- e hical-
in es o -1535799601
Sipiczki, A. (2022). A c i ical look a he ESG ma ke . Re ie ed Decembe
16, 2022, om h ps://www.ceps.eu/wp- con en /uploads/
2022/04/PI2022-15_A-c i ical-look-a - he-ESG-ma ke .pd
S&P Global. (2020). How can AI help ESG in es ing? S&P Global. Re ie ed
Decembe 16, 2022, om h ps : / / www. spglobal . com /en /
esea ch-insigh s/a icles/how-can-ai-help-esg-in es ing
S&P Global. (2022). S&P Global ESG Sco es – Me hodology. Re ie ed Jan-
ua y 22, 2023, om h ps://www.spglobal.com/esg/documen s/
sp-global-esg-sco es-me hodology-2022.pd
S ack Exchange. (2023). Exclusion o U ili ies and Financials in Magic Fo -
mula. Quan i a i e Finance S ack Exchange. Re ie ed Janua y
30, 2023, om h ps://quan .s ackexchange.com/ques ions/
41192/exclusion-o -u ili es-and- inancials-in-magic- o mula
S a e S ee Global Ad iso s. (2019). The ESG Da a Challenge. Re ie ed
Janua y 4, 2023, om h ps : / / www. ssga . com /in es men -
opics/en i onmen al- social- go e nance/2019/03/esg- da a-
challenge.pd
S ickel, S. E. (1990). P edic ing indi idual analys ea nings o ecas s. Jou -
nal o Accoun ing Resea ch,28(2), 409–417.
Sus ainaly ics. (2020). The ESG Risk Ra ing: F equen ly Asked Ques ions –
Fo Companies. Re ie ed No embe 25, 2022, om sho u l.a /
kopX3
Sus ainaly ics. (2021). ESG Risk Ra ings – Me hodology Abs ac . Re ie ed
No embe 25, 2022, om h ps://connec .sus ainaly ics.com/
esg- isk- a ings-me hodology
Sus ainaly ics. (2022). Abou Us. Sus ainaly ics. Re ie ed No embe 25,
2022, om h ps://www.sus ainaly ics.com/abou -us
Swamina han, S. (1991). The Impac o SEC Manda ed Segmen Da a on
P ice Va iabili y and Di e gence o Belie s. The Accoun ing Re iew,
66(1), 23–41.
Tang, D. Y., Yan, J., & Yao, C. Y. (2022). The De e minan s o ESG Ra ings:
Ra e Owne ship Ma e s. P oceedings o Pa is Decembe 2021 Fi-
nance Mee ing EUROFIDAI – ESSEC.
The Economis . (2019). The London S ock Exchange buys Re ini i . The
Economis . Re ie ed No embe 28, 2022, om h ps://www.
economis . com / inance - and - economics /2019 /08 /01 / he -
london-s ock-exchange-buys- e ini i
The Sus ainAbili y Ins i u e. (2020). Ra e he Ra e s 2020: In es o Su ey
and In e iew Resul s. Re ie ed Decembe 7, 2022, om h ps:
/ / www. sus ainabili y. com /globalasse s /sus ainabili y. com /
hinking/pd s/sus ainabili y- a e he a e s2020- epo .pd
The Sus ainAbili y Ins i u e. (2022). Cos s and Bene i s o Clima e-Rela ed
Disclosu e Ac i i ies by Co po a e Issue s and Ins i u ional
In es o s. Re ie ed Janua y 2, 2023, om h ps : / / www .
sus ainabili y. com /globalasse s /sus ainabili y. com / hinking /
pd s/2022/cos s- and- bene i s- o - clima e- ela ed- disclosu e-
ac i i ies- by-co po a e-issue s- and-ins i u ional- in es o s-17-
may-22.pd
Tomoko, F., & Kyoko, N. (2022). Japan In oduces Manda o y ESG Disclo-
su es Fo Public Companies. Mo gan Lewis. Re ie ed Feb ua y
3, 2023, om h ps : / / www. mo ganlewis . com /pubs /2022 /
12/japan- in oduces- manda o y- esg- disclosu es- o - public-
companies
UN PRI and ICGN. (2018). In es o Agenda Fo Co po a e Esg Repo ing.
Re ie ed Decembe 12, 2022, om h ps://www.unp i.o g/
download?ac=6181
Wansleben, L. (2012). Financial Analys s. In K. K. Ce ina & A. P eda (Eds.),
The Ox o d Handbook o The Sociology o Finance (pp. 407–459).
Ox o d.
Welch, I. (2000). He ding among secu i y analys s. Jou nal o Financial Eco-
nomics,58(3), 369–396.
Wild, D. (2017). ESG In eg a ion App oach. In C. Ins i u e, C. S. Swi ze -
land, & S. S. Finance (Eds.), Handbook on Sus ainable In es men s
(pp. 54–59).
Win o h, K., Blombe g, J., & Kjellbe g, H. (2010). Enac ing O e lapping Ma -
ke s. Jou nal o Cul u al Economy,3(1), 3–18.
Yu, F. (2008). Analys co e age and ea nings managemen . Jou nal o Finan-
cial Economics,88(2), 245–271.
Zhang, H. (2021). I ’s Time o P i a e Companies o Come Clean on ESG.
Ins i u ional In es o . Re ie ed Janua y 2, 2023, om h ps://
www. ins i u ionalin es o . com /a icle /b1 8zb72 jzcy3 /I - s -
Time- o -P i a e-Companies- o-Come-Clean-on-ESG
Zmijewski, M. E. (1984). Me hodological Issues Rela ed o he Es ima ion
o Financial Dis ess P edic ion Models. Jou nal o Accoun ing Re-
sea ch,22, 59–82.
Zwiebel, J. (1995). Co po a e Conse a ism and Rela i e Compensa ion.
Jou nal o Poli ical Economy,103(1), 1–25.