Financial returns of going green: evidence from MSCI indices

Heldmann, Jan; B ückne , Thomas; Dang, Huong Dieu
A icle — Published Ve sion
Financial e u ns o going g een: e idence om MSCI
indices
Jou nal o Asse Managemen
Sugges ed Ci a ion: Heldmann, Jan; B ückne , Thomas; Dang, Huong Dieu (2025) : Financial e u ns
o going g een: e idence om MSCI indices, Jou nal o Asse Managemen , ISSN 1479-179X,
Palg a e Macmillan UK, London, Vol. 26, Iss. 7, pp. 768-787,
h ps://doi.o g/10.1057/s41260-025-00404-4
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/333384
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 p://c ea i ecommons.o g/licenses/by/4.0/
Vol:.(1234567890)
Jou nal o Asse Managemen (2025) 26:768–787
h ps://doi.o g/10.1057/s41260-025-00404-4
ORIGINAL ARTICLE
Financial e u ns o going g een: e idence omMSCI indices
JanHeldmann1· ThomasB ückne 1· HuongDieuDang2
Re ised: 5 Ma ch 2025 / Accep ed: 17 Ap il 2025 / Published online: 7 June 2025
© The Au ho (s) 2025
Abs ac
This s udy examines how he g een c i e ia (GCE) used by MSCI o c ea e g een equi y indices in luence hei inancial pe -
o mance. We analyze he Clima e Change (CC), Pa is-Aligned Benchma k (PAB), Socially Responsible In es men (SRI),
and SRI Fil e ed PAB (SRI PAB) index a ian s in compa ison wi h hei s anda d non-g een coun e pa in each o he ou
egions: he Wo ld, he USA, Eu ope, and Eme ging Ma ke s (EM). O e all, he g een indices o en ma ched o exceeded he
e u ns o hei s anda d index wi hou adding signi ican isk. Wi h ew excep ions in he EM, he g een indices exhibi ed
be e long- e m inancial pe o mance han hei s anda d index. O e 2015–2023, he CC, PAB, SRI, and SRI PAB indices
espec i ely deli e ed cumula i e excess e u ns o 4.7%, 5.8%, 13.7%, and 7.5% ela i e o he s anda d index. Thei e u ns
co-mo ed closely wi h he ma ke and he s anda d index’s e u ns. The GCEs s a is ically and signi ican ly con ibu ed o
g een indices’ ela i e inancial ou pe o mance.
Keywo ds MSCI· G een index· Cumula i e e u n di e en ial· Weal h ela i e
In oduc ion
The g owing need o in es men solu ions add essing
clima e, social, and go e nance conce ns has d awn g ea
in es o s’ in e es in g een p oduc s.1 A he end o 2023,
global sus ainable unds managed abou US$2.2 illion and
con inued o ou g ow con en ional unds in e ms o in lows
(Co ina and Phyu 2024). Many o hese p oduc s use g een
indices eplica ed h ough Exchange T aded Funds (ETFs);
howe e , i emains unce ain whe he such g een indices/
ETFs consis en ly ou pe o m hei con en ional non-g een
coun e pa s. Some s udies epo posi i e ela i e inancial
pe o mance (Fio delisi e al. 2023; Rompo is 2023; Pa lo a
and de Boy ie 2022), g ea e esilience du ing inancial c i-
ses (O as e al. 2014) and supe io pe o mance du ing he
COVID-19 pandemic (Lin and Swain 2024). Con e sely,
o he esea ch inds g een indices/ETFs pe o m compa a-
bly o adi ional benchma ks, a y ac oss egions, o show
no signi ican di e ence in isk-adjus ed e u ns (Bolognesi
e al. 2024; Jonwall e al. 2024; Dumi escu e al. 2023;
Cunha e al. 2020; Jain e al. 2019; Benson e al. 2010).
Some s udies e en sugges unde pe o mance o g een indi-
ces/ETFs ela i e o adi ional inancial p oduc s (Lean and
Nguyen 2014; O as e al. 2012). E idence on e u n co ela-
ion is also mixed, wi h some au ho s obse ing decoupling
(Ang 2015; Lean and Nguyen 2014) and o he s documen ing
high co ela ions o co-mo emen s (Rompo is 2023; Managi
e al. 2012). Thus, i emains unclea whe he g een indices/
ETFs exhibi be e inancial pe o mance han hei non-
g een coun e pa s. Fu he mo e, e y li le e o has been
made o di ec ly examine i and how c i e ia employed o
cons uc g een indices ela e o hei inancial pe o mance.
I g een c i e ia do no a ec a g een index’s pe o mance,
any obse ed ou pe o mance ( ela i e o i s s anda d non-
g een index) is likely due o luck. The exis ing e idence,
* Huong Dieu Dang
huong.dang@can e bu y.ac.nz
1 Uni e si y o Bay eu h, Bay eu h, Ge many
2 Uni e si y o Can e bu y, Ch is chu ch, NewZealand
1 In his s udy, we add ess a ious sus ainable in es men s, o exam-
ple, En i onmen al, Social & Go e nance (ESG), Social Responsible
In es men (SRI), sus ainable, e hical, g een as g een in es men s.
769Financial e u ns o going g een: e idence omMSCI indices
mos ly o ETFs, implying a mixed ela ionship (Aba e e al.
2021; Papa hanasiou and Kou sokos as 2024).
Ou s udy is mo i a ed by he mixed epo ed esul s on
he inancial pe o mance o g een indices/ETFs and he
unexplo ed (mixed) ela ionship be ween g een c i e ia and
he inancial pe o mance o g een indices (ETFs). In his
s udy, we choose o explo e Mo gan S anley Capi al In e -
na ional (MSCI) indices as hey ha e been widely used as
e e ence indices o g een equi y ETFs.2 We add ess wo
main esea ch ques ions: (1) How did MSCI g een indices
pe o m inancially compa ed wi h hei espec i e s anda d
(pa en ) index? and (2) How did he g een c i e ia employed
by MSCI o c ea e each g een index a ec i s inancial
pe o mance?
We ocus on indices a he han ETFs o se e al easons.
Fi s , indices a e mo e sui able o pe o mance analysis, as
he impac o o al expense a io, axes,3 dis ibu ion pol-
icy,4 and eplica ion s a egy5 a ies ac oss p o ide s and
can obscu e e u n analyses. Second, he exac benchma k
an ETF acks can change o e ime, while he ETF keeps
he same In e na ional Secu i ies Iden i ica ion Numbe s
(ISIN), which can dis o benchma k-adjus ed e u n analy-
ses.6 Thi d, ETF p o ide s use sligh ly modi ied benchma ks
ega ding exclusion h esholds, compa ed wi h ou bench-
ma k being he s anda d non-g een index. Tha is, p o ide s
aim o achie e a pe o mance ad an age in hei espec i e
pee g oup by o e ing ETFs ha do no ack exac ly he
compe i o ’s benchma k bu he same g een exposu e.7
We examine a se o 20 MSCI indices spanning ou majo
egions: he Wo ld, he USA, Eu ope, and eme ging ma ke s
(EM). Wi hin each egion, we in es iga e he pe o mance o
each o he ou g een indices— he Clima e Change (CC),
he Pa is-Aligned Benchma k (PAB), he Socially Respon-
sible In es men (SRI), and he SRI il e ed PAB (SRI PAB)
index— ela i e o he s anda d (pa en ) MSCI index. Ou
p ima y da ase on index e u ns ex ends om Janua y 2015
o Decembe 2023, p o iding 2160 mon hly index obse -
a ions. Conside ing he a ailabili y o i m-le el da a and
Tobin’s Q measu e, ou sample o 1,076,340 i m-mon h
obse a ions is limi ed o Decembe 2022, and includes
5267 unique cons i uen s o 50 coun ies and 11 sec o s.
The ich his o ical da a o MSCI index cons i uen s enables
us o ca y ou mo e ho ough and eliable analyses han
using compa able ETFs.8
Ou s udy di e s om ecen esea ch on ETFs’ inancial
pe o mance, such as ElBannan (2024). Fi s , we ocus on
indices ins ead o ETFs. By doing so, we a oid he added
complexi y o axes, dis ibu ion policies, and eplica ion
me hods ha can in luence ETF e u ns. Second, ou index
cons i uen da a allows us o con ol o di e ences in sec o s
and le els o de elopmen ac oss coun ies. Thi d, we ana-
lyze ou dis inc ypes o g een indices a he han ea ing
all ESG o SRI p oduc s as one ca ego y. This includes gi -
ing special a en ion o he Pa is-Aligned Benchma k (PAB)
and SRI PAB indices which bo h ha e no been s udied in
he li e a u e.
Ou s udy is ela ed o Kossen ini e al. (2024) and Jacob
and Wilkens (2021). The o me ocus on ou MSCI ESG
Leade indices co e ing di e en pa s o he Eu opean
egion: Eu ope, he Economic and Mone a y Union (EMU),
Eme ging Ma ke s Eu ope, and he Middle Eas and Eu ope,
o he yea s om 2007 o 2020. The la e analyze ou
ESG and ou ca bon- ocused MSCI Wo ld Indices om
2011 o 2019. Ou o ou 16 g een indices, only he Wo ld
SRI o e laps wi h Jacob and Wilken’s (2021) sample. We
2 As o Oc obe 2024, asse s unde managemen in ETFs linked o
MSCI equi y indexes eached app oxima ely $1.72 illion. See:
h ps:// i . msci. com/ aum- e s- linked- msci- index es.
3 Wi hholding ax bene i s on US di idends, depending on he und
domicile o an ETF, can ha e a subs an ial impac on und pe o -
mance. Physically eplica ing ETFs wi h US exposu e domiciled in
I eland ha e a educed wi hholding ax bu den o 15% compa ed o,
o example, 30% o ETFs domiciled in Luxembou g.
4 Fo example, accumula ing ETFs ein es hei di idends while dis-
ibu ing ETFs pay ou di idends.
5 The e a e wo main eplica ion s a egies, physical and syn he ic
eplica ion. They spli in o ull o physically op imized eplica-
ion and swap eplica ion (which can be un unded, unded and ully
unded). The eplica ion s a egy di ec ly in luences he pe o mance
o an ETF. Physical eplica ion bene i s om secu i ies lending while
hose using swaps a e, depending on he quali ica ion o he index
unde Sec ion871(m) o he US In e nal Re enue Code, no bu dened
by ax egula ions. Gene ally, ESG ETFs use physical eplica ion.
6 Fo example, he Amundi Index MSCI Eme ging Ma ke s (EM)
SRI PAB UCITS ETF (ISIN: LU1861138961) was launched in Janu-
a y 2019 and acked he MSCI EM SRI Index. In Oc obe 2019, he
e e ence index was swi ched o MSCI EM SRI 5% Issue Capped.
F om Decembe 2020 onwa d, he ETF acked he MSCI EM SRI
Fil e ed PAB. Ano he example is he iSha es MSCI Wo ld SRI
UCITS ETF (ISIN: IE00BYX2JD69). I was launched in Oc o-
be 2017 wi h MSCI Wo ld SRI Selec Index being he benchma k.
In No embe 2019, i s a ed acking he MSCI Wo ld SRI Selec
Reduced Fossil Fuel Index.
7 Please see he g een p oduc ange (SFDR A icle 8 and 9) om
Amundi, www. amund ie . nl/ en/ p o e ssion al/ e - p odu c s/ sea ch,
iSha es, www. isha es. com/ us/ p odu c s/ e - in es men s and X acke s,
www. e . dws. com/ en- us/ e - p odu c s.
8 MSCI p o ides cu en da a on i s indices in he ac shee s and ia
he cons i uen download unc ion a h ps:// www. msci. com/ cons
i uen s. His o ical da a is no eely a ailable, bu ou da a se ex ends
om he beginning o 2015 o he end o 2023. The index pe o -
mance o 12 indices launched a e 2015 has been econs uc ed (by
MSCI) using he iden ical me hods and p inciples o MSCI. This his-
o ical da a a ailabili y makes ou s udy unique, as we a e no awa e
o any o he s udy ha analyses MSCI PAB and SRI PAB indices
o e such a long pe iod o ime, including he Co ona c isis and he
Uk aine wa .
770 J.Heldmann e al.
ex end hese wo s udies by co e ing a mo e ecen pe iod
ha includes signi ican global e en s such as he onse o
he wa in Uk aine. Ou s udy encompasses wo addi ional
majo egions, he USA, and Eme ging Ma ke s. Impo -
an ly, we place pa icula emphasis on indices aligned wi h
he Pa is Clima e Ag eemen , including he PAB and SRI
PAB indices, which a e o pa icula in e es o clima e-
conscious in es o s. A ecen s udy conduc ed by Bolognesi
e al. (2024) in es iga es he MSCI SRI and MSCI PAB indi-
ces, bu i s sample is limi ed o he USA o a sho pe iod,
Augus 2021–May 2024. We employ a iche da ase ha
co e s ou egions o e a longe pe iod (2015–2023). Mos
impo an ly, we ca y ou index-le el eg ession analyses
ha con ol o sec o - and coun y-speci ic pa ame e s. This
allows us o di ec ly e alua e he e ec i eness o he g een
c i e ia (GCE) employed by MSCI o c ea e each g een
index ype.
To assess he inancial pe o mance o a g een index, we
use cumula i e e u n di e en ial (CRD) as ou p ima y
measu e.9 Fo obus ness es s, we u ilize Weal h Rela i e
(WR), Tobin’s Q,10 he Sha pe a io, and he T eyno a io.
CRD and WR cap u e he pe o mance o a g een index,
ela i e o i s s anda d index, o e an in es men ho izon;
each is compu ed di ec ly a he index le el. O he me ics
a e no ela i e measu es; each is es ima ed using an index’s
mon hly cons i uen weigh s and cons i uen s’ me ics. To
examine an index’s e u n co-mo emen wi h he ma ke
index, we es ima e he ma ke ac o be a coe icien using
Fama-F ench models. We also assess he dynamic condi-
ional co ela ions be ween g een and s anda d index e u ns
o e ime using a GARCH(1,1) model wi h a Gaussian dis-
ibu ion on daily e u n da a. To explo e he ela ionship
be ween g een c i e ia and a g een index’s inancial pe -
o mance, we conduc OLS eg essions and employ ou
dummy a iables, each ep esen s he impac o MSCI’s
g een c i e ia (GCE) used o c ea e one g een index ype.
Ou analysis documen s s ong e idence suppo ing he
a ac i eness o g een indices. On a e age, o e he pe iod
Janua y 2015–Decembe 2023, an in es o would ha e
espec i ely ea ned a 4.7%, 5.8%, 13.7%, and 7.5% g ea e
cumula i e e u n i s/he had in es ed in he g een CC, PAB,
SRI, and SRI PAB index ins ead o he s anda d index. Tak-
ing in o accoun he expense a ios o he la ges ETFs which
eplica e he g een and s anda d indices, an in es o would
ha e espec i ely ea ned a ne cumula i e e u n di e en ial
o 4%, 5.1%, 13%, and 6.8% du ing 2015–2023. Ac oss he
ou egions, mos g een indices demons a ed long- e m
ou pe o mance (excep in EM, whe e only he SRI ou pe -
o med he s anda d index). On a e age, g een indices ac oss
egions achie ed g ea e Sha pe and T eyno a ios and
deli e ed be e isk-adjus ed e u ns han hei espec i e
s anda d index. Fu he mo e, g een indices’ e u ns mo ed
closely wi h hose o he ma ke index and hei espec i e
s anda d index. We ind s ong e idence ha MSCI’s GCEs
s a is ically and signi ican ly con ibu ed o g een indices’
be e inancial pe o mance, ela i e o he s anda d index.
The posi i e e ec s o GCEs we e mo e p onounced on he
ou pe o mance o he SRI and SRI PAB indices; bo h a e
aligned wi h he Pa is Clima e Ag eemen .
Ou main esul s a e consis en ac oss obus ness es s,
including analyses wi h one-, wo-, and h ee-mon h lagged
con ol a iables, al e na i e e u n measu es, and a ious
subsamples. Ou key inding indica es ha g een indices
can deli e compe i i e e u ns and, in mos cases, ou pe -
o m hei s anda d non-g een coun e pa s o e a ela i ely
long ime pe iod. By emphasizing he ole o g een c i e-
ia in index cons uc ion, ou s udy con ibu es o ongoing
discussions on he inancial a ac i eness o g een equi y
in es men s.
The es o ou pape is s uc u ed as ollows.The “B ie
o e iew o MSCI’s me hods o c ea ing g een equi y
indices” sec ion p o ides a b ie o e iew o he p o-
cess by whichMSCI selec s cons i uen s and cons uc s
he examined g een indices.The “Li e a u e e iew and
esea ch ques ions” sec ion e iews he li e a u e and p o-
poses esea ch ques ions. The“Da a and me hods” sec ion
desc ibes da a and p esen s he me hod. The“Empi ical
esul s” sec ion discusses he empi ical esul s. The“Con-
clusion” sec ion concludes he key indings.
B ie o e iew o MSCI’s me hods
o c ea ing g een equi y indices
The MSCI S anda d (pa en ) Index se es as he s a ing
uni e se o de e mining he eligible uni e se o a g een
index.11 The pa en indices in ou sample we e launched
in Ma ch 1986, excep o EM in Janua y 2001. The g een
indices a e cons uc ed by applying exclusion c i e ia o
9 As we compa e MSCI g een indices o hei co esponding MSCI
s anda d index, we do no need o conside index p o ide ’ cha ac e -
is ics ha a ec an index’s inancial pe o mance.
10 We a oid using accoun ing igu es om indi idual companies
o compu e index-le el e u n on equi y (ROE) and e u n on asse s
(ROA) because se e al ac o s can dis o hese accoun ing e u n
measu es. Di e ences in accoun ing s anda ds and legal amewo ks
ac oss coun ies can lead o inconsis en inancial epo ing. In addi-
ion, manage s o en ha e signi ican lexibili y, which may allow
hem o adjus o smoo h ea nings o e ime. Fu he mo e, du ing
pe iods o high in la ion, dep ecia ion expenses may be unde s a ed
since hey do no e lec he ue eplacemen cos s o equipmen ,
po en ially leading o a i icially in la ed ea nings.
11 See h ps:// www. msci. com/ index- me ho dology o an o e iew o
he cons uc ion me hodology o all MSCI Indices.
771Financial e u ns o going g een: e idence omMSCI indices
he pa en index, leading o a smalle eligible uni e se. The
deg ee o exclusion a ies ac oss he g een indices, wi h
s ic e c i e ia esul ing in ewe componen s compa ed o
he pa en index. Among he indices analyzed, he MSCI
Clima e Change (CC) Index applies he mildes exclusions,
while he MSCI Socially Responsible In es ing (SRI) il-
e ed PAB Index en o ces he s ic es exclusions. Bo h he
PAB and SRI PAB indices a e consis en wi h he Pa is
Ag eemen ’s objec i es, wi h he SRI PAB excluding he
mos secu i ies.
The MSCI CC index, and heMSCI clima e PAB index
To o m he MSCI CC Index (launched in June 2019) and
he MSCI Clima e PAB Index (launched in Oc obe 2020),
a nega i e sc eening is applied o he pa en index in he
i s s ep. Companies a e excluded om he pa en index
i hey iola e ei he he UN Global Compac P inciples
o sec o policies, such as gene a ing e enue abo e se
h esholds om con o e sial weapons, obacco, o ossil
uels. Nex , he sus ainabili y o he emaining compa-
nies is e alua ed using he Low Ca bon T ansi ion (LCT)
sco e, which conside s h ee componen s: a company’s
ca bon oo p in , i s clima e- ela ed isks, and i s abili y
o manage hose isks. Companies a e ca ego ized in o
i e g oups based on hei LCT sco es: Solu ions, Neu-
al, Ope a ional T ansi ion, P oduc T ansi ion, and Asse
S anding. Companies classi ied as Solu ions (wi h he
highes LCT sco e o h ee) ecei e a highe weigh ing in
he inal index. Weigh adjus men s a e applied ac oss all
i e ca ego ies ela i e o he secu i y weigh s in he pa -
en index, o ming he inal eligible uni e se o he CC
Index. Fo ins ance, in he “Wo ld” egion, his p ocess
ans o ms he MSCI Wo ld pa en index in o he MSCI
Wo ld CC Index.
The MSCI Clima e PAB Index, howe e , applies s ic e
c i e ia wi h a PAB o e lay aligned wi h he goals o he
Pa is Ag eemen . Speci ically, companies mus educe hei
Weigh ed A e age Ca bon In ensi y (WACI) by 50% and
lowe hei annual ca bon oo p in by 7% compa ed o he
pa en index. Addi ionally, he Clima e PAB Index excludes
companies de i ing e enue om ossil uel ac i i ies, such
as coal, oil, na u al gas explo a ion, p ocessing, o powe
gene a ion wi h excessi e g eenhouse gas in ensi y. Finally,
he weigh ing o each company in he MSCI Clima e PAB
Index is de e mined using i s LCT sco e.
The MSCI SRI index, and heMSCI SRI il e ed PAB
(SRI PAB) index
The MSCI SRI Index (launched in June 2011, excep o
EM in Ma ch 2014) and he MSCI SRI il e ed PAB Index
(launched in June 2020) apply s ic e exclusion c i e ia han
he Clima e Indexes discussed abo e. Bo h indices imple-
men mo e sec o -speci ic exclusions wi h low h esholds,
esul ing in a signi ican ly smalle eligible uni e se com-
pa ed o he pa en index. A de ailed compa ison o he
exclusion c i e ia and h esholds can be ound in “Appendix
A.”
The MSCI SRI il e ed PAB Index is mo e exclusion-
a y han he MSCI SRI Index due o i s alignmen wi h he
Pa is Ag eemen goals. I applies addi ional h esholds o
companies gene a ing e enue om oil and gas ac i i ies.
Fu he mo e, companies wi h se ious iola ions o sus ain-
able in es men objec i es a e lagged as “Red,” ecei ing
an ESG Con o e sies Sco e o 0, and a e excluded om
he index. The emaining companies a e a ed using MSCI
ESG Ra ings, which ange om AAA o CCC. Fo inclu-
sion, exis ing index cons i uen s mus ha e a minimum a -
ing o BB and an ESG Con o e sies Sco e o a leas 1,
while new addi ions equi e a minimum a ing o A and an
ESG Con o e sies Sco e o 4.
Bo h indices employ a bes -in-class app oach, selec -
ing only he op 25% o companies wi hin each sec o . The
SRI il e ed PAB Index also inco po a es he PAB o e lay,
equi ing companies o educe hei WACI by 50% and hei
ca bon oo p in by 7% annually ela i e o he pa en index.
To ensu e di e si ica ion, he weigh o any single company
in he index is capped a 5%.
Li e a u e e iew and esea ch ques ions
F om a po olio heo y pe spec i e, bo h nega i e sc een-
ing and bes -in-class app oaches used o cons uc g een
po olios limi he in es men uni e se, po en ially educ-
ing di e si ica ion and inc easing isk compa ed o b oade
po olios (Ba ne and Salomon 2006; Renneboog e al.
2008). Ne e heless, se e al s udies indica e ha g een
s ocks o hose wi h high ESG sco es exhibi a o able
cha ac e is ics, such as lowe unsys ema ic isk (Giese
e al. 2019; Hong and Kacpe czyk 2009) and educed
capi al cos s (G ego ye al. 2021; Un uh e al. 2016; Ng
and Rezaee 2015). Mos esea ch on g een indices o ETFs
inds minimal di e ences in e u ns compa ed o bench-
ma ks, and any obse ed di e ences a e ypically aligned
wi h isk-adjus ed e u ns.
S udies on SRI and ESG indices epo ei he be e o
simila pe o mance compa ed o con en ional benchma ks.
Those inding be e pe o mance include S a man (2005),
Cunha e al. (2020), Jain e al., (2023), and Kossen ini e al.
(2024). In con as , s udies epo ing simila pe o mance
include Sch öde (2007), Collison e al. (2008), Consolandi
e al. (2009), Benson e al. (2010), Jain e al. (2019), Jacob
and Wilkens (2021), and Bolognesi e al. (2024). Se e al
s udies also documen ha g een indices/ETFs demons a e

772 J.Heldmann e al.
g ea e esilience du ing c ises (O as e al. 2014, 2013;
Omu a e al. 2021; Lin and Swain 2024; Huang 2024;
ElBannan 2024). An i-ESG ETFs, howe e , end o unde -
pe o m (Rompo is 2024).
In e ms o e u ns co-mo emen , Jain e al. (2019) ind
ha g een indices closely ack hei benchma ks ac oss he
USA, E.U., EM, and global ma ke s du ing 2013–2017.
Simila ly, Managi e al. (2012) obse e no dis inc cha ac-
e is ics be ween g een indices and hei benchma ks, no ing
high co-mo emen ac oss a ious ma ke condi ions. These
indings alone sugges ha g een indices a e no ma kedly
dis inc in es men ehicles compa ed wi h hei adi ional
benchma ks.
Despi e he g owing popula i y o g een in es men s,
e y ew empi ical s udies di ec ly in es iga e how g een
indices’ cons uc ion c i e ia a ec hei inancial pe o -
mance. This is likely due o he challenge in ob aining
indices’ cons i uen weigh ing da a o e a easonably long
pe iod. Rela ed s udies mainly examine unds ca ego ized
by sus ainabili y/ESG a ings and epo mixed indings.
Using 634 Eu opean mu ual unds du ing 2014–2019,
Aba e e al. (2021) epo be e pe o mance o unds wi h
high Mo nings a Sus ainabili y Ra ings. In con as , Papa-
hanasiou and Kou sokos as (2024) ind ha ESG unds
wi h low Mo nings a a ings ou pe o m among 235 ESG
mu ual unds du ing 2010–2022. Folge -La onde e al.
(2022) obse e ha highe - a ed ETFs wi h an Eco-Fund
Label ha e lowe e u ns du ing he COVID-19 ma ke
c ash. Rompo is (2022a, b) ind no signi ican isk-adjus ed
e u n di e ences o ESG ETFs and a nega i e ela ion-
ship be ween e u ns and ESG me ics, espec i ely. Gi en
he abo e mixed indings, i emains unclea whe he sus-
ainabili y/ESG a ings a e linked o ETFs’ be e inan-
cial pe o mance. In eali y, he c i e ia used o cons uc
a g een index a e complex and ex end a beyond ESG a -
ings, as we b ie ly discussed in “B ie o e iew o MSCI’s
me hods o c ea ing g een equi y indices” sec ion. The
unexplo ed ela ionship be ween g een c i e ia and g een
index’s inancial pe o mance mo i a es us o in es iga e
how he g een c i e ia used by MSCI a ec i s g een indi-
ces’ inancial pe o mance.
Ou li e a u e e iew highligh s se e al gaps ha ha e
no been well add essed in p io s udies. Mos s udies
ocus on gene ic ESG/SRI ETFs/indices wi hou analyz-
ing hose aligned wi h he Pa is Ag eemen ’s emissions
educ ion goals. Re u n compa isons o en in ol e mul iple
p o ide s agains a single benchma k, igno ing p o ide -
speci ic cha ac e is ics. Fu he mo e, he impac o g een
c i e ia used o c ea e an index and i s inancial pe o -
mance emains unexplo ed. Ou s udy add esses hese gaps
by conduc ing a comp ehensi e analysis o g een MSCI
indices, u ilizing cons i uen weigh ing da a ac oss ou
majo egions, assessing hei alignmen wi h he Pa is
Ag eemen , and benchma king hem agains hei espec-
i e MSCI s anda d indices. We aim o add ess he esea ch
gaps by answe ing he ollowing wo esea ch ques ions:
1. How did g een MSCI indices pe o m inancially com-
pa ed wi h i s s anda d non-g een index?
2. How did an index’s G een C i e ia E ec i eness (GCE)
a ec i s inancial pe o mance?
Da a andme hods
Da a
Ou s udy makes use o bo h index-le el and i m-le el da a.
Ou mon hly index da ase includes 20 MSCI equi y indices
om Janua y 2015 o Decembe 2023. The i e examined
indices (one s anda d and ou g een-o ien ed a ian s) a e
analyzed in each o he ou egions: he Wo ld, he USA,
EU, and EM. We collec daily Ne To al Re u ns (in USD)
om Bloombe g, which include di idend ein es men s and
conside wi hholding axes. To cap u e an index’s exposu e
o he ma ke index, we use mon hly Fama-F ench ac o
e u ns and isk- ee a es om F ench’s (2024) online da a
lib a y.12
Fo he i m-le el analysis, MSCI gene ously p o ided
us wi h mon hly da a on each index’s cons i uen s and hei
weigh s om Janua y 2015 o Decembe 2022. This da ase
includes 5267 unique i ms om 11 sec o s ac oss 50 coun-
ies. We a e able o me iculously map 5034 o hese i ms
(co e ing mo e han 99.9% o he o al indices’ weigh s) o
hei ISINs. O he i m-speci ic da a is ob ained om Re ini-
i Eikon’s Da as eam, which co e s nea ly ou en i e sam-
ple. To ou knowledge, his is one o he i s s udies ha
employ such a ich MSCI index da a, including wo Pa is
Ag eemen -aligned indices, o e a ela i ely long pe iod.
Me hods
Index’s inancial pe o mance measu es
We employ cumula i e e u n di e en ial (CRD) as ou p i-
ma y measu e ha cap u es he inancial pe o mance o
a g een index gi, ela i e o i s pa en index pi. The e u n
di e en ial dgi, o g een index gi ela i e o i s pa en index
pi a ime is compu ed as:
(1)
dgi, = gi, − pi,
12 See h ps:// mba. uck. da m ou h. edu/ pages/ acul y/ ken. ench/ da a_
lib a y. h ml.
773Financial e u ns o going g een: e idence omMSCI indices
The
CRDgi,T
o g een index gi o e he ime ho izon T
is gi en by:
The a e age CRD,
ACRD
, o a g een index gi ac oss
ou egions k a ime , is gi en by:
In addi ion, we calcula e he ne CRD and ne ACRD by
deduc ing one wel h o he co esponding ETFs’ a e age
annual cos s13 om he mon hly e u ns o an index in ou
sample.
In obus ness analyses, we use wo o he pe o mance
measu es namely he weal h ela i e (WR) and Tobin’s Q.
The WR is he a io o he compounded e u ns be ween a
g een index gi and i s pa en index pi a ime , es ima ed as
ollows:
A WR alue g ea e han one indica es ha an in es o
would be be e o in es ing in he g een index gi ins ead o
i s s anda d index pi o e he ime ho izon T.
We es ima e an index’s Tobin’s Q as he weigh ed sum o
i m j’s weigh
𝜔
in index i and i m j’s Tobin’s Q a ime ,
whe e J is he o al numbe o i ms in he index:14
In a simila app oach, we also es ima e he Sha pe a io
and T eyno a io o each index.
Re u ns co‑mo emen
We es ima e he Fama–F ench (FF) i e- ac o model o
examine he co-mo emen s o a g een index’s e u ns ela-
i e o he ma ke index’s e u ns in each egion. In his anal-
ysis, we ocus on he coe icien es ima e o he ma ke ac o
(2)
CRD
gi,T=
T
∑
=1
dgi,
(3)
ACRD
gi, =
1
4
4
∑
k=1
CRDgi, ,k
(4)
WR
gi, =∏
T
=1(1+ gi,
)
∏
T
=1
(1+
pi, )
(5)
Tobin
�sQi, =
J
∑
j
=
1
𝜔i,j, ∗Tobin�sQj,
(
𝛽1)
. I
𝛽1
is close o one, i indica es ha he g een index i’s
e u ns mo e closely wi h he ma ke index’s e u ns. Ou
baseline FF i e- ac o model is speci ied as ollows:
whe e
i,
ep esen s an index i’s e u n a ime , and
, is
he
isk- ee a e a ime .
The independen a iables include he ma ke excess
e u n (
m,
–
,
), small minus big (SMB ), high minus low
(HML ), obus minus weak (RMW ), and conse a i e
minus agg essi e (CMA ) ac o s a ime .
In addi ion o he baseline FF i e- ac o model (Fama
and F ench 2015), we also es ima e he FF h ee- ac o
model (Fama and F ench 1993), he Ca ha Fama-F ench
ou - ac o model (Ca ha 1997), and he Ca ha Fama-
F ench six- ac o model (Fama and F ench 2018) o obus -
ness es s.
To examine he co-mo emen s be ween he e u ns o a
g een index and i s espec i e s anda d index (in he same
egion), we apply he GARCH (1,1) model wi h a Gaussian
dis ibu ion o ou daily e u ns. This enables us o unde -
s and he co ela ion dynamics o e ime be ween g een
indices and hei pa en index in each egion.
OLS index‑le el eg essions
We ca y ou o dina y leas squa es (OLS) eg essions a he
index le el o analyze how he e ec i eness o g een c i e ia,
deno ed as GCE, a ec s an index’s inancial pe o mance.
Fo each pe o mance me ic ou lined ea lie (CRD, WR,
Tobin’s Q), we es ima e he ollowing eg ession model:
whe e
Yi,
: A pe o mance measu e (CRD, WR, Tobin’s Q) o
index
i
a ime
, as discussed abo e.
GCE_Indexi
: A dummy a iable ha akes he alue o 1
i index i was cons uc ed acco ding o he c i e ia o a ce -
ain index. This means ha GCE_PAB equals o 1 only o
he PAB index, while i is ze o o all o he s, including he
s anda d, CC, SRI, and SRI PAB indices. This logic applies
o all o he g een index ypes acco dingly.
(6)
i, − , =𝛼+𝛽1
(
m, − ,
)
+𝛽2SMB +𝛽3HML
+𝛽
4
RMW
+𝛽
5
CMA
+𝜖
(7)
Y
i, =𝛼+
4
∑
k=1
𝛽i×GCE_Indexi+
4
∑
g=1
𝛾g×HDI_G oupg, −12
+
10
∑
s=1
𝛿s×Sec o s, −1+𝜓1Pos _Launchi, −1+𝜓2log MVi, −
1
+𝜓
3
Co id
−1
+𝜓
4
Uk aine
−1
+𝜖
13 Fo his pu pose, we calcula e he a e age o al expense a ios
o he la ges ETFs which eplica e g een and s anda d indices. The
a e age annual o al expense a ios we ha e calcula ed o he s and-
a d (g een) index o he Wo ld is 13 (21), o he USA 11 (20), o
Eu ope 15 (19) and o EM 17 (25) basis poin s espec i ely.
14 Al hough Tobin’s Qs a he company le el a e annual alues, he
Tobin’s Qs a he index le el a e compu ed mon hly as he weigh s o
cons i uen s in an index change mon hly and he composi ion o an
index also changes o e he cou se o a yea .
774 J.Heldmann e al.
HDI_G oupg, −12
: Pe cen age o i ms in a Human De el-
opmen Indica o (HDI) g oup
g
a ime
( −12)
.15
Sec o s, −1
: Pe cen age o i ms in sec o
s
a ime (
−1)
.16
Pos _Launchi, −1
: Dummy a iable indica ing i index
i
had al eady been launched p io o o was launched a ime
( −1)
.
lMV −1∶
Loga i hm o he a e age weigh ed ma ke alue
o index
i
a ime (
−1)
.
Co id −1∶
One-mon h lagged dummy a iable o he
COVID-19 pe iod.
Uk aine −1
: One-mon h lagged dummy a iable o he
Uk aine wa pe iod.
𝜖
: E o e m a ime .
Unlike ElBannan (2024), which uses a single dummy
a iable o ep esen a ious ESG unds, we analyze ou
g een index ypes sepa a ely. Ins ead o g ouping hem
in one ca ego y, we c ea e ou dis inc dummy a iables
(GCE_Indexi). In equa ion (7), he s anda d index ac s as
he baseline ca ego y and is cap u ed by he in e cep (
𝛼
).
I he coe icien o GCE_Indexi is posi i e (
𝛽i>0)
and
s a is ically signi ican , i sugges s ha MSCI’s c i e ia o
c ea e his speci ic g een index ype has a posi i e impac
on i s pe o mance measu e
Yi,
, e en a e con olling o
ele an index-, sec o -, coun y- and ime-speci ic a iables.
Ou da ase includes 50 coun ies. Using mul iple coun-
y-speci ic a iables ( o example, coun y dummy a ia-
bles) leads o mul icollinea i y and would obscu e he e ec s
o ou main explana o y a iables, g een c i e ia e ec i e-
ness (GCE). We a oid mul icollinea i y by using he Human
De elopmen Indica o (HDI) as a p oxy o coun y-speci ic
e ec s, and sec o composi ion, which cap u es b oad c oss-
sec ional di e ences.17
Ou p ima y inancial pe o mance measu e is he cumu-
la i e e u n di e en ial (CRD), which e lec s he long- e m
pe spec i e o in es o s. We use OLS eg ession as i o e s
clea and easily in e p e able coe icien es ima es which
allow us o di ec ly assess he in luence o he GCEs on an
index’s ela i e e u n. Fu he mo e, we obse e no signi i-
can di e ences in esul s be ween cumula i e (CRD) and
non-cumula i e e u n measu e (WR). The usage o cumula-
i e alues does no a ec he main insigh s o conclusions
o he s udy, sugges ing ha he OLS app oach is sui able
o ou analysis.
To u he alida e ou model, we conduc Residual e -
sus Fi ed Plo s o ou baseline eg ession o CRD. The
esiduals appea easonably cen e ed a ound ze o ac oss he
ange o i ed alues, indica ing no se e e iola ions o lin-
ea i y. We also examine added- a iable plo s, and he pa ial
eg ession lines show a linea pa e n o ou GCE dummies.
Ou model u he includes con ols o sec o composi ion,
coun y cha ac e is ics (HDI), index’s ma ke capi aliza ion,
and majo e en s (COVID-19, Uk aine wa ).
Mos impo an ly, we use Newey–Wes s anda d e o s in
all analyses o add ess po en ial au oco ela ion and he e -
oskedas ici y.18 This ensu es ha any omi ed a iable bias
o iola ions o classical assump ions a e minimized. We
acknowledge ha no model is pe ec ; howe e , he absence
o s ong e idence o non-linea i y o se e e speci ica ion
e o s suppo s he alidi y o ou OLS analysis esul s.
In addi ion o ou baseline OLS analysis o cumula i e
e u n di e en ial, we conduc se e al obus ness es s using
al e na i e inancial pe o mance me ics (WR, Tobin’s Q).
We u he analyze subsamples, including speci ic sub-indi-
ces, sub- egions, pe iods o heigh ened ma ke unce ain y,
and samples wi h wo-mon h/ h ee-mon h lags. The esul s
emain consis en ac oss hese es s, ein o cing he eliabil-
i y o ou main indings.
Empi ical esul s
S a is ics
Table1 p esen s he s a is ics o e u ns o he 20 MSCI
indices in ou egions o e he pe iod Janua y 2015–Decem-
be 2023 (columns 1–5), and wo sub-pe iods o di e en
ma ke condi ions: he p e-pandemic (ma ke s abili y)
pe iod Janua y 2015–Janua y 2020 (columns 6–10), and
he pandemic pe iod Feb ua y 2020–Janua y 2022 (columns
11–15).19 Ac oss all pe iods, he g een indices gene ally
15 Each yea , cons i uen s’ coun ies in ou sample a e g ouped in o
i e HDI ca ego ies based on hei lagged nume ical alues: ≤0.6,
0.6–0.7, 0.7–0.8, 0.8–0.9, and ≥0.9. The highes ca ego y (≥0.9)
se es as he e e ence g oup. Since i m weigh s wi hin an index
change mon hly and index cons i uen s also e ol e o e ime, he HDI
g oup alues a e upda ed each mon h. The HDI da a is sou ced om
he Uni ed Na ions De elopmen P og am (UNDP).
16 The ele en sec o s include Heal h Ca e, Consume Disc e iona y,
Ma e ials, Indus ials, In o ma ion Technology, Financials, Consume
S aples, Ene gy, U ili ies, Communica ion Se ices, and Real Es a e
which we use as he e e ence ca ego y.
17 Subsequen empi ical analyses show ha he g een indices ha e
be as close o 1 and hey all main ain high co ela ions (gene ally
abo e 0.95) wi h hei espec i e s anda d indices. Tha is, he g een
indices ha e simila sys ema ic isk as he ma ke index, and mac o-
economic ac o s, such as in e es a es o in la ion, likely a ec bo h
g een and s anda d indices simila ly. Thus, we do no employ coun-
y-speci ic mac oeconomic con ols.
18 Addi ionally, we compu e Whi e s anda d e o s o all models and
ob ain consis en esul s.
19 We choose Feb ua y 2020 as he s a ing poin o he pandemic,
as he Wo ld Heal h O ganiza ion (WHO) decla ed i a Public Heal h
Eme gency o In e na ional Conce n on Janua y 30 (WHO 2020).
The pe iod ends in Janua y 2022, aligning wi h he onse o he
Uk aine Wa in Feb ua y 2022. We obse e ha by Ma ch/ Ap il
2022, he s ic measu es p e iously enac ed in esponse o he Co id-
775Financial e u ns o going g een: e idence omMSCI indices
demons a e be e pe o mance han he s anda d index, as
e idenced by g ea e o qui e compa able mean e u ns. On
a e age, he e u ns a e highes in he USA and lowes in
EM.
Panel A o Fig.1 depic s he a e age cumula i e e u n
di e en ials (ACRD) o All Regions, and Panels B-E
espec i ely show he cumula i e e u n di e en ials (CRD)
o he Wo ld, he USA, Eu ope, and EM.
As in es o s in e es ed in sus ainabili y a e ypically long-
e m o ien ed, ou discussion below ocuses on he ela i e
pe o mance o e he medium- and long- e m ho izon. Fo
All Regions, he Wo ld, and he USA, all he g een indi-
ces exhibi posi i e CRDs (Panel A, B, and C). The ACRD
(Panel A) highligh s ha he wo indices wi h he mos
es ic i e il e s (SRI and SRI PAB) gain momen um and
ou pe o m hei less g een coun e pa s in he la e yea s
Table 1 Desc ip i e s a is ics o index e u ns, Janua y 2015–Decembe 2023
Panels A, B, C, D o his able espec i ely show he e u n s a is ics o he i e indices examined o he Wo ld, he USA, Eu ope, and Eme ging
Ma ke s (EM) egion. We epo he mean e u n, median e u n, he s anda d de ia ion (SD) o e u ns, he minimum and maximum e u n o
he o al s udy pe iod (Janua y 2015–Decembe 2023), he p e-Co id-19 pe iod (Decembe 2015–Janua y 2020), and he Co id-19 pe iod (Feb-
ua y 2020–Janua y 2022). The i e examined indices consis o he MSCI S anda d (pa en ) Index, he MSCI Clima e Change Index (CC), he
MSCI Clima e Pa is-Aligned Index (PAB), he MSCI Socially Responsible In es men index (SRI) and he MSCI Socially Responsible In es -
men il e ed Pa is-Aligned Benchma k Index (SRI PAB). The as e ixis nex o he mean and median e u ns espec i ely indica e he p alue
o he es and Wilcoxon ank sum es showing whe he he e u n s a is ics o an index a e s a is ically di e en om ze o. ***p alue ≤ 1%;
**1% < p alue ≤ 5%; *5% < p alue ≤ 10%
Panel A: Wo ld To al: 01/15–12/23 P e-Co id-19: 01/15–01/20 Co id-19: 02/20–01/22
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
Mean Median SD Min Max Mean Median SD Min Max Mean Median SD Min Max
S anda d 0.82* 1.4** 4.49 −13.2 12.8 0.73* 1.15** 3.35 −7.6 7.9 1.41 2.53 5.80 −13.2 12.8
Clima e Change 0.91** 1.34** 4.62 −12.3 12.8 0.81* 1.09** 3.36 −7.6 7.6 1.53 2.49 5.86 −12.3 12.8
PAB 0.88** 1.54** 4.55 −12.7 12.7 0.85** 1.15** 3.30 −7.4 7.7 1.4 2.47 5.87 −12.7 12.7
SRI 0.91** 1.3** 4.52 −10.8 11.9 0.81* 1.14** 3.31 −7.6 7.7 1.59 2.78 5.70 −10.8 11.9
SRI PAB 0.9** 1.12** 4.54 −11.1 11.9 0.83* 1.02** 3.27 −7.7 7.7 1.57 2.6 5.66 −11.1 11.9
Panel B: USA To al: 01/15–12/23 P e-Co id-19: 01/15–01/20 Co id-19: 02/20–01/22
Mean Median SD Min Max Mean Median SD Min Max Mean Median SD Min Max
S anda d 1** 1.4*** 4.63 −12.7 13.1 0.92** 1.27*** 3.45 −9.1 8.2 1.69 2.66 5.96 −12.7 13.1
Clima e Change 1.11** 1.78*** 4.79 −11.7 13.3 1.01** 1.54*** 3.48 −8.9 8.1 1.85 2.85 6.06 −11.7 13.3
PAB 1.09** 1.59*** 4.73 −11.9 13.2 1.1** 1.58*** 3.42 −8.6 8.1 1.67 2.83 6.04 −11.9 13.2
SRI 1.13** 1.4*** 4.73 −10.3 12.7 0.99** 1.36*** 3.45 −8.6 7.8 2.03 3.38* 5.97 −10.3 12.7
SRI PAB 1.06** 1.21*** 4.74 −11.4 12.3 0.93** 1.13*** 3.45 −8.7 7.8 1.96 2.25 5.96 −11.4 12.3
Panel C: Eu ope To al: 01/15–12/23 P e-Co id-19: 01/15–01/20 Co id-19: 02/20–01/22
Mean Median SD Min Max Mean Median SD Min Max Mean Median SD Min Max
S anda d 0.55 0.7 4.85 −14.6 17.0 0.43 0.59 3.67 −7.8 6.7 0.96 2.77 6.32 −14.6 17.0
Clima e Change 0.55 0.44 4.92 −14.5 16.4 0.44 0.23 3.69 −8.2 6.2 0.94 2.99 6.35 −14.5 16.4
PAB 0.62 0.54 4.92 −14.6 16.1 0.52 0.38 3.64 −8.0 6.2 1.04 3.11 6.33 −14.6 16.1
SRI 0.68 0.75 4.75 −10.9 15.9 0.61 0.68 3.58 −7.8 7.2 1.04 3.27 6.09 −10.9 15.9
SRI PAB 0.71 1.02 4.81 −11.7 15.4 0.67 0.88 3.53 −7.7 6.9 1.18 3.17 6.10 −11.7 15.4
Panel D: EM To al: 01/15–12/23 P e-Co id-19: 01/15–01/20 Co id-19: 02/20–01/22
Mean Median SD Min Max Mean Median SD Min Max Mean Median SD Min Max
S anda d 0.39 0.33 5.09 −15.4 14.8 0.47 0.24 4.57 −9.0 13.2 0.87 1.43 5.65 −15.4 9.2
Clima e Change 0.37 0.23 5.15 −14.7 15.6 0.49 0.26 4.60 −8.9 13.1 0.85 1.26 5.58 −14.7 9.7
PAB 0.39 0.16 5.10 −15.8 13.3 0.47 0.13 4.56 −9.1 13.3 0.97 1.78 5.77 −15.8 9.7
SRI 0.54 0.21 5.44 −18.1 18.2 0.55 0.28 4.13 −8.7 13.6 1.46 2.05 6.86 −18.1 13.4
SRI PAB 0.36 0.18 5.12 −19.1 14.5 0.43 0.44 4.07 −8.6 13.2 1.01 2.03 6.75 −19.1 12.8
Foo no e 19 (con inued)
19 ou b eak had s a ed o ease, e u ning o p e-Ma ch 2020 le els,
al hough his a ied by coun y.
782 J.Heldmann e al.
PAB indices (Column 4–6). Nex , conside ing he domi-
nance o US cons i uen s, we analyze wo subsamples o
egions (Panel B): one excludes he Wo ld (Column 1–3),
and one excluding he USA (Column 4–6). In panel C, we
di ide ou sample in o wo sub-pe iods and conduc sep-
a a e analyses o he ma ke unce ain y (Column 1–3),
and he ma ke s abili y pe iod (Column 4–6). Finally, o
minimize conce ns o e e se causali y, we es ima e wo
OLS eg essions using wo samples (Panel D): one wi h
wo-mon h lagged con ol a iables (Column 1–3), and one
wi h h ee-mon h lagged con ol a iables (Column 4–6).
We obse e a qui e consis en esul ha he GCEs s a is i-
cally and signi ican ly con ibu e o g een indices’ ela i e
ou pe o mance (CRD, WR) and, excep o he GCE o CC,
g ea e alua ion (Tobin’s Q). The mos es ic i e index,
he SRI PAB, exe s he s onges impac on CRD, WR and
Tobin’s Q.
While he GCE o he CC index does no show a signi i-
can posi i e e ec on Tobin’s Q, his does no impac he
o e all implica ions o ou s udy. Rela i e e u n measu es
CRD and WR a e de i ed pu ely om ma ke da a, whe eas
Tobin’s Q is de i ed based on bo h ma ke and accoun ing
da a which, o some ex en , is subjec o ea nings manage-
men . Tobin’s Q shows whe he an index is o e - alued o
unde - alued ( ela i e o he eplacemen cos o i s asse s).
Tobin’s Q is a alua ion me ic and does no cap u e he
e u ns o g een indices ela i e o hei s anda d indi-
ces. Thus, i is no su p ising ha he esul o he Tobin
Q’s analysis sligh ly di e s om hose o CRD and WR’
analyses. The di e ence is mainly o he CC index ( he
leas es ic i e o he ou examined g een indices in each
egion). Wha ma e s mos is ha he signi ican e ec s o
GCEs on Tobin’s Q o he o he h ee g een indices indi-
ca e ha ma ke pa icipan s do ecognize and ewa d he
s ic e g een c i e ia employed—especially in he case o
he SRI and SRI PAB indices, which a e closely aligned wi h
he Pa is Ag eemen . O e all, he obus esul s ob ained
om CRD and WR analyses, and he signi ican e ec s o
GCEs o SRI/SRI PAB ( he wo es ic i e g een indices
ha aligned wi h he Pa is Clima e Ag eemen ) ac oss all
h ee measu es (CRD, WR, Tobin’s Q) emain he key d i -
e s o ou conclusions.
Conclusion
Ou s udy analyses he inancial pe o mance 20 MSCI
equi y indices om 2015 o 2023 and aims a examining
how he g een c i e ia (GCE) used by MSCI o c ea e g een
indices in luence hei inancial pe o mance. We compa e
he MSCI s anda d (pa en ) non-g een index wi h ou MSCI
g een indices in each o he ou examined egions, including
wo ha align wi h he Pa is Ag eemen ’s objec i es.
Ou analyses e eal se e al indings. Fi s , a e aged
ac oss he ou egions, he ou g een indices (CC, PAB,
SRI, and SRI PAB) achie e be e inancial pe o mance
han he s anda d index, as demons a ed by long- e m posi-
i e cumula i e e u n di e en ials (CRD) and he weal h
ela i es (WR) being g ea e han 1 o e he s udy pe iod,
pa icula ly o he SRI indices. While we do see sho - e m
unde pe o mance, his is no unexpec ed conside ing he
well-known ade-o be ween in es ing in sus ainabili y and
achie ing immedia e inancial e u ns. Mos impo an ly,
all, excep one, g een indices deli e g ea e isk-adjus ed
e u ns (Sha pe and T eyno a ios), and highe Tobin’s Qs
indica ing g ea e ma ke alua ion han he s anda d index.
We a e, howe e , cau ious o he EM egion as only he
SRI pe o m subs an ially be e han he s anda d index.
Second, he e u ns o he ou g een indices a e closely
ela ed o he e u ns o he ma ke index and he e u ns o
he s anda d index. The sys ema ic isks o hese g een indi-
ces, as e lec ed by hei be a coe icien es ima es de i ed
om Fama-F ench models, a e simila o hose o s anda d
indices, sugges ing ha hey do no comp omise on di e si-
ica ion. Thi d, MSCI’s GCEs used o cons uc g een indi-
ces a e signi ican ly associa ed wi h hei ela i e inancial
ou pe o mance.
Ou esea ch adds aluable insigh s in o he sus ainable
inance li e a u e in se e al ways. Fi s , we ocus on he
PAB and SRI PAB indices, making his one o he i s
s udies o explo e he inancial a ac i eness o Pa is
Ag eemen -Aligned Benchma k in es men s. Second, we
analyze he inancial pe o mance o MSCI g een indices
compa ed o hei s anda d MSCI non-g een coun e pa s,
using a ich MSCI da ase wi h mon hly weigh ing da a
om 2015 o 2022; he eby elimina ing conce ns o po en-
ial e ec s o an index p o ide ’s cha ac e is ics. Thi d, we
con ibu e o he discussion on g een equi y in es men s,
showing hei po en ial as ools o di e si ica ion and
imp o ed e u ns.
Ou indings o e impo an p ac ical insigh s o in es-
o s, index p o ide s, und manage s, and policymake s.
Gi en he obse ed long- e m ou pe o mance, in es o s
seeking o align hei po olios wi h he Pa is Ag eemen
goals could conside ETFs acking he MSCI PAB and
SRI PAB indices (SRI PAB only o Eme ging Ma ke s).
Policymake s can also use hese Pa is Ag eemen -aligned
indices as e e ence s anda ds o sus ainable inance ini ia-
i es, o example, unde he Sus ainable Finance Disclosu e
Regula ion (SFDR) A icle 9 o he EU Taxonomy. They can
help o e i y ha inancial p oduc s mee de ined sus ain-
abili y c i e ia and hus inc ease anspa ency and educe

783Financial e u ns o going g een: e idence omMSCI indices
g eenwashing. Using hese indices as benchma ks o go -
e nmen pension unds o o he public in es men ehicles
can u he help o di ec capi al owa d p ojec s which aim
o mi iga e clima e change. This is inc easingly impo an
conside ing ha New Zealand Supe annua ion Fund and
No ges Bank In es men Managemen , which manages No -
way’s so e eign weal h und, ha e di es ed om oil compa-
nies,27 and o he so e eigns will ollow sui . Ou es ablished
e idence o he ou pe o mance o he wo Pa is Ag eemen -
aligned indices o e s so e eign weal h unds impo an ools
o pu sue sus ainable in es men s a egies and accele a e
go e nmen s’ e o s o suppo he ansi ion o a low-ca -
bon economy.
Ou s udy has ce ain limi a ions as i ocuses on 20 MSCI
indices o e a speci ic pe iod, which includes he COVID-19
pandemic and he ongoing wa in Uk aine, po en ially in lu-
encing he esul s due o hese unusual economic condi ions.
The limi ed a ailabili y o his o ical cons i uen weigh ing
da a ac oss egions also es ic s he scope o ou analysis,
making i less applicable o o he indices o ime pe iods.
Fu u e esea ch could add ess hese limi a ions by exam-
ining a wide se o indices o e a longe pe iod. I could
also explo e how ESG/g een c i e ia used by di e en index
p o ide s in luence pe o mance unde di e en economic
condi ions.
Appendix A: Compa ison o exclusion
h esholds ing een MSCI’s indices
The MSCI SRI and SRI Fil e ed Pa is-Aligned Bench-
ma k (PAB) indices a e designed o e lec socially
esponsible in es ing p inciples while aligning wi h sus-
ainabili y and clima e- ocused goals. These me hods
impose s ic h esholds on e enues de i ed om con o-
e sial o sensi i e ac i i ies o de e mine company (con-
s i uen s) eligibili y. The h esholds a y be ween he SRI
Index and he s ic e SRI Fil e ed PAB Index, ensu ing
ha in es men s mee highe en i onmen al, social, and
go e nance (ESG) s anda ds. The ollowing summa y
ou lines he key h esholds o hese ac i i ies based
on he MSCI SRI Indexes Me hodology (MSCI 2024a)
and he MSCI SRI Fil e ed PAB Indexes Me hodology
(MSCI 2024b). Fo Con o e sial Weapons, companies
a e excluded en i ely om SRI and SRI il e ed PAB
indices, wi h a 0% h eshold o all e enues. Fo Con-
en ional Weapons, he SRI index allows up o 5% o e -
enues om p oduc ion and 15% om componen s, while
he SRI il e ed PAB educes his h eshold o 5% o bo h
p oduc ion and componen s. Ci ilian Fi ea ms a e also
es ic ed, wi h p oduc ion e enues capped a 0% and
dis ibu ion e enues capped a 5% o bo h he SRI and
SRI il e ed PAB indices. Simila ly, o Nuclea ac i i ies,
e enues om weapons p oduc ion a e es ic ed o 0%,
powe gene a ion is capped a 5%, and nuclea supplie s
a e limi ed o 15% unde he SRI index. Howe e , he SRI
il e ed PAB educes he supplie h eshold o 5%. The -
mal Coal is excluded en i ely om he SRI and he SRI
PAB index while bo h allow up o 5% o e enues om
powe gene a ion. Fo Tobacco, p oduc ion is no allowed
(0% h eshold) in bo h SRI indices, while dis ibu ion is
capped a 5%. Alcohol- ela ed ac i i ies ha e mo e lenien
h esholds. P oduc ion e enues a e capped a 5% o bo h
SRI indices, and dis ibu ion e enues a e capped a 15%.
Simila ly, Gambling is limi ed o 5% o owne ship e -
enues and 15% o se ices in bo h SRI and SRI il e ed
PAB indices. Fo Adul En e ainmen , p oduc ion e -
enues a e limi ed o 5%, and dis ibu ion is capped a 15%
unde bo h SRI indices. Gene ically modi ied o ganisms
(GMOs) a e es ic ed o a 5% h eshold o p oduc ion
e enues. Oil and Gas ac i i ies ace s ic e limi a ions.
Re enues om con en ional and uncon en ional oil and
gas ac i i ies a e capped a 0% o bo h SRI indices. How-
e e , powe gene a ion om oil and gas is allowed up o
30% unde he SRI il e ed PAB index.
27 See h ps:// www. g een peace. o g/ ao ea oa/ p ess- elea se/ supe -
unds- 950m- ossil- uel- di es men - an- aha- momen - o - nz- econo my/
and h ps:// www. hegu a dian. com/ busin ess/ 2017/ no / 16/ oil- and- gas-
sha es- dip- as- no wa ys- cen al- bank- ad is es- oslo- o- di es .
784 J.Heldmann e al.
Appendix B: OLS obus ness analysis u ilizing a ious subsamples
Panel A: Subsamples o indices
Va iables Sample: S anda d, CC and PAB indices Sample: S anda d, SRI and SRI PAB indices
P o ide (1) (2) (3) (4) (5) (6)
CRD WR Tobin’s Q CRD WR Tobin’s Q
GCE_CC 0.082*** 0.114*** 0.23*** NA NA NA
(0.005) (0.005) (0.053) NA NA NA
GCE_PAB 0.11*** 0.136*** 0.322*** NA NA NA
(0.007) (0.006) (0.059) NA NA NA
GCE_SRI NA NA NA 0.074*** 0.08*** 0.37***
NA NA NA (0.004) (0.005) (0.024)
GCE_SRI PAB NA NA NA 0.123*** 0.137*** 0.618***
NA NA NA (0.007) (0.008) (0.051)
Sec o composi e YES YES YES YES YES YES
HDI composi e YES YES YES YES YES YES
Log o ma ke alue YES YES YES YES YES YES
Pos Launch dummy YES YES YES YES YES YES
Uk aine dummy YES YES YES YES YES YES
Co id dummy YES YES YES YES YES YES
Obse a ions 1140 1140 1140 1140 1140 1140
Adj. R20.733 0.83 0.922 0.66 0.664 0.895
Panel B: Subsamples o egions
Va iables Sample excluding “Wo ld” Region Sample excluding “US” Region
P o ide (1) (2) (3) (4) (5) (6)
CRD WR Tobin’s Q CRD WR Tobin’s Q
GCE_CC 0.036*** 0.038*** −0.083** −0.013** 0.013* −0.084**
(0.007) (0.008) (0.038) (0.006) (0.007) (0.037)
GCE_PAB 0.063*** 0.066*** 0.115** 0.012 0.033*** 0.017
(0.009) (0.01) (0.048) (0.008) (0.009) (0.046)
GCE_SRI 0.071*** 0.077*** 0.332*** 0.068*** 0.068*** 0.336***
(0.005) (0.006) (0.021) (0.004) (0.005) (0.021)
GCE_SRI PAB 0.094*** 0.104*** 0.645*** 0.076*** 0.092*** 0.526***
(0.009) (0.01) (0.043) (0.008) (0.009) (0.043)
Sec o composi e YES YES YES YES YES YES
HDI composi e YES YES YES YES YES YES
Log o ma ke alue YES YES YES YES YES YES
Pos Launch dummy YES YES YES YES YES YES
Uk aine dummy YES YES YES YES YES YES
Co id dummy YES YES YES YES YES YES
Obse a ions 1425 1425 1425 1425 1425 1425
Adj. R20.56 0.564 0.901 0.603 0.598 0.888
785Financial e u ns o going g een: e idence omMSCI indices
Panel C: Subsamples o ime pe iods
Va iables Sample ma ke unce ain y Sample ma ke s abili y
P o ide (1) (2) (3) (4) (5) (6)
CRD WR Tobin’s Q CRD WR Tobin’s Q
GCE_CC 0.054*** 0.009* 0.002 0.008 0.017 −0.219***
(0.006) (0.005) (0.073) (0.01) (0.011) (0.048)
GCE_PAB 0.136*** 0.034*** 0.146 0.038*** 0.044*** −0.069
(0.012) (0.009) (0.125) (0.01) (0.011) (0.046)
GCE_SRI 0.112*** 0.04*** 0.377*** 0.025*** 0.026*** 0.292***
(0.007) (0.005) (0.070) (0.004) (0.004) (0.02)
GCE_SRI PAB 0.152*** 0.043*** 0.515*** 0.041*** 0.047*** 0.395***
(0.008) (0.006) (0.081) (0.012) (0.014) (0.057)
Sec o composi e YES YES YES YES YES YES
HDI composi e YES YES YES YES YES YES
Log o ma ke alue YES YES YES YES YES YES
Pos Launch dummy YES YES YES YES YES YES
Obse a ions 700 700 700 1200 1200 1200
Adj. R20.775 0.487 0.857 0.481 0.474 0.847
Panel D: Sample wi h wo- and h ee-mon h lagged con ol a iables
Va iables Sample wi h wo-mon h lagged con ol a iables Sample wi h h ee-mon h lagged con ol a iables
P o ide (1) (2) (3) (4) (5) (6)
CRD WR Tobin’s Q CRD WR Tobin’s Q
GCE_CC 0.024*** 0.042*** −0.034 0.027*** 0.046*** −0.058
(0.007) (0.008) (0.035) (0.007) (0.008) (0.036)
GCE_PAB 0.06*** 0.073*** 0.14*** 0.064*** 0.077*** 0.105**
(0.009) (0.01) (0.047) (0.008) (0.011) (0.048)
GCE_SRI 0.063*** 0.068*** 0.331*** 0.065*** 0.071*** 0.352***
(0.005) (0.005) (0.023) (0.005) (0.005) (0.024)
GCE_SRI PAB 0.081*** 0.095*** 0.653*** 0.085*** 0.099*** 0.629***
(0.009) (0.01) (0.045) (0.008) (0.01) (0.045)
Sec o composi e YES YES YES YES YES YES
HDI composi e YES YES YES YES YES YES
Log o ma ke alue YES YES YES YES YES YES
Pos launch dummy YES YES YES YES YES YES
Uk aine dummy YES YES YES YES YES YES
Co id dummy YES YES YES YES YES YES
Obse a ions 1880 1880 1880 1860 1860 1860
Adj. R20.524 0.542 0.896 0.522 0.54 0.892
This “Appendix” epo s he coe icien es ima es o he G een C i e ia E ec i eness (GCE) o each g een index ( he MSCI Clima e Change
Index-CC, he MSCI Clima e Pa is-Aligned Index-PAB, he MSCI Socially Responsible In es men Index-SRI, and he MSCI Socially Respon-
sible In es men il e ed Pa is-Aligned Benchma k Index-SRI PAB) ob ained om he OLS index-le el analysis o he cumula i e e u n di e -
en ial (CRD, columns 1 and 4), weal h ela i e (WR, columns 2 and 5) and Tobin’s Q (columns 3 and 6) o each subsample in Panels A–D
F om he en i e sample co e ing he pe iod Janua y 2015–Decembe 2022, we c ea e 8 di e en subsamples. In Panel A, we examine wo
subsamples o which one consis s o he s anda d and wo less s ic CC and PAB indices (columns 1–3), and he o he including he s anda d
and wo mo e s ic SRI and SRI PAB indices (columns 4–6). In Panel B, we analyze wo subsamples o which one excludes he “Wo ld” egion
(columns 1–3), and one exclude he “US” egion (columns 4–6). In Panel C, we look a wo di e en sub-pe iods: ma ke unce ain y (columns
1–3) and ma ke s abili y (columns 4–6). The ma ke unce ain y pe iod co e s he Uk aine wa and he Co id pandemic while he ma ke s abil-
i y pe iod co e s he emaining mon hs o ou s udy pe iod. In Panel D, we explo e a sample wi h wo-mon h lagged con ol a iables (columns
1–3) and a sample wi h h ee-mon h lagged con ol a iables (columns 4–6). In bo h samples, he index-le el HDI composi e is lagged by a yea
We es ima e all models wi h he Newey–Wes obus s anda d e o s. The coe icien es ima es o GCEs a e epo ed i s , ollowed by he
s anda d e o s (in pa en heses). Fo b e i y eason, he coe icien es ima es o he in e cep and con ol a iables a e no epo ed. ***, ** and *
ep esen s p alue signi icance a he 0.01, 0.05 and 0.10 le els espec i ely
786 J.Heldmann e al.
Acknowledgemen We would like o hank MSCI o gene ously p o-
iding us wi h his o ical mon hly index cons i uen da a. We a e g a e-
ul o help ul commen s om an anonymous e iewe , Tom Coupe,
and Klaus Schä e . All e o s a e ou s. Huong Dang would like o
hank he E skine G an O ice and he Depa men o Economics and
Finance a he Uni e si y o Can e bu y, he Bay eu h Humbold Cen e
and he Chai o Finance and Banking a he Uni e si y o Bay eu h
(UBT) o he g ea suppo h oughou he isi o UBT, du ing which
his p ojec was ini ia ed.
Funding Open Access unding enabled and o ganized by CAUL and
i s Membe Ins i u ions.
Open Access This a icle is licensed unde a C ea i e Commons A i-
bu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a-
ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce,
p o ide a link o he C ea i e Commons licence, and indica e i changes
we e made. The images o o he hi d pa y ma e ial in his a icle a e
included in he a icle’s C ea i e Commons licence, unless indica ed
o he wise in a c edi line o he ma e ial. I ma e ial is no included in
he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will
need o ob ain pe mission di ec ly om he copy igh holde . To iew a
copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Re e ences
Aba e, G., I. Basile, and P. Fe a i. 2021. The le el o sus ainabili y
and mu ual und pe o mance in Eu ope: An empi ical analysis
using ESG a ings. Co po a e Social Responsibili y and En i-
onmen al Managemen 28 (5): 1446–1455. h ps:// doi. o g/ 10.
1002/ cs . 2175.
Ang, W.R. 2015. Sus ainable in es men in Ko ea does no ca ch a cold
when he Uni ed S a es sneezes. Jou nal o Sus ainable Finance
& In es men 5 (1–2): 16–26. h ps:// doi. o g/ 10. 1080/ 20430 795.
2015. 10427 37.
Ba ne , M.L., and R.M. Salomon. 2006. Beyond dicho omy: The cu -
ilinea ela ionship be ween social esponsibili y and inancial
pe o mance. S a egic Managemen Jou nal 27 (11): 1101–1122.
h ps:// doi. o g/ 10. 1002/ smj. 557.
Benson, C.C., N. Gup a, and R. Ma e i. 2010. Does isk educ ion
mi iga e he cos s o going g een? An empi ical s udy o sus ain-
able in es ing. Sou he n Jou nal o Business and E hics 2: 7–25.
Bolognesi, E., A. D eassi, M. Miglia acca, and A. Pal inie i. 2024.
Mapping sus ainable in es ing: Explo ing ambigui ies and con-
sis encies among sus ainable indices. Jou nal o En i onmen al
Managemen 367: 122081. h ps:// doi. o g/ 10. 1016/j. jen m an.
2024. 122081.
Ca ha , M.M. 1997. On pe sis ence in mu ual und pe o mance. The
Jou nal o Finance 52 (1): 57–82. h ps:// doi. o g/ 10. 1111/j. 1540-
6261. 1997. b038 08.x.
Collison, D.J., G. Cobb, D.M. Powe , and L.A. S e enson. 2008. The
inancial pe o mance o he FTSE4Good indices. Co po a e
Social Responsibili y and En i onmen al Managemen 15 (1):
14–28. h ps:// doi. o g/ 10. 1002/ cs . 144.
Consolandi, C., A. Jaiswal-Dale, E. Poggiani, and A. Ve celli. 2009.
Global S anda ds and E hical S ock Indexes: The Case o he Dow
Jones Sus ainabili y S oxx index. Jou nal o Business E hics 87
(S1): 185–197. h ps:// doi. o g/ 10. 1007/ s10551- 008- 9793-1.
Co ina, H., and M. Phyu. 2024. Sus ainable unds con inue o ou -
g ow pee s in 2023. ISS Insigh s. Re ie ed om h ps:// insig h s.
issgo e na nce. com/ pos s/ sus a inable- unds- con i nue- o- ou g
ow- pee s- in- 2023/.
Cunha, F.A.F.D.S., E.M. De Oli ei a, R.J. O sa o, M.C. Klo zle, F.L.
Cy ino Oli ei a, and R.G.G. Caiado. 2020. Can sus ainable in es -
men s ou pe o m adi ional benchma ks? E idence om global
s ock ma ke s. Business S a egy and he En i onmen 29 (2):
682–697. h ps:// doi. o g/ 10. 1002/ bse. 2397.
Dumi escu, A., J. Jä inen, and M. Zak iya. 2023. Hidden Gem o
Fool’s Gold: Can passi e ESG ETFs ou pe o m he benchma ks?.
In e na ional Re iew o Financial Analysis 86: 102540. h ps://
doi. o g/ 10. 1016/j. i a. 2023. 102540.
ElBannan, M.A. 2024. Re u ns beha io o ESG ETFs in he COVID-
19 ma ke c ash: A e g een unds mo e esilien ?. Jou nal o Co -
po a e Accoun ing & Finance 35 (2): 187–223. h ps:// doi . o g/ 10.
1002/ jca . 22680.
Fama, E.F., and K.R. F ench. 1993. Common isk ac o s in he e u ns
on s ocks and bonds. Jou nal o Financial Economics 33 (1):
3–56. h ps:// doi. o g/ 10. 1016/ 0304- 405X(93) 90023-5.
Fama, E.F., and K.R. F ench. 2015. A i e- ac o asse p icing model.
Jou nal o Financial Economics 116 (1): 1–22. h ps:// doi. o g/ 10.
1016/j. j ine co. 2014. 10. 010.
Fama, E.F., and K.R. F ench. 2018. Choosing ac o s. Jou nal o
Financial Economics 128 (2): 234–252. h ps:// doi. o g/ 10. 1016/j.
j ine co. 2018. 02. 012.
Fio delisi, F., G. Galloppo, G. La anzio, and V. Paimano a. 2023.
Looking a socially esponsible in es men s a egies h ough
he lenses o he global ETF indus y. Jou nal o In e na ional
Money and Finance 137: 102917. h ps:// doi. o g/ 10. 1016/j. jimon
in. 2023. 102917.
Folge -La onde, Z., S. Pashang, L. Feo , and A. ElAl y. 2022. ESG
a ings and inancial pe o mance o exchange- aded unds du -
ing he COVID-19 pandemic. Jou nal o Sus ainable Finance &
In es men 12 (2): 490–496. h ps:// doi. o g/ 10. 1080/ 20430 795.
2020. 17828 14.
F ench, K.R. 2024. Da a Lib a y. Re ie ed om h ps:// mba. uck.
da m ou h. edu/ pages/ acul y/ ken. ench/ da a_ lib a y. h ml.
Giese, G., L.-E. Lee, D. Melas, Z. Nagy, and L. Nishikawa. 2019.
Founda ions o ESG in es ing: How ESG a ec s equi y alua-
ion, isk, and pe o mance. The Jou nal o Po olio Manage-
men 45 (5): 69–83. h ps:// doi. o g/ 10. 3905/ jpm. 2019. 45.5. 069.
G ego y, R.P., J.G. S ead, and E. S ead. 2021. The global p icing o
en i onmen al, social, and go e nance (ESG) c i e ia. Jou nal
o Sus ainable Finance & In es men 11 (4): 310–329. h ps://
doi. o g/ 10. 1080/ 20430 795. 2020. 17317 86.
Hong, H., and M. Kacpe czyk. 2009. The p ice o sin: The e ec s o
social no ms on ma ke s. Jou nal o Financial Economics 93
(1): 15–36.h ps:// doi. o g/ 10. 1016/j. j ine co. 2008. 09. 001.
Huang, Y. 2024. Do ESG ETFs p o ide downside isk p o ec ion
du ing Co id-19? E idence om o ecas combina ion models.
In e na ional Re iew o Financial Analysis 94: 103320. h ps://
doi. o g/ 10. 1016/j. i a. 2024. 103320.
Jacob, A., and M. Wilkens. 2021. Wha d i es sus ainable indices? A
amewo k o analyzing he sus ainable index landscape. SSRN
Elec onic Jou nal. h ps:// doi. o g/ 10. 2139/ ss n. 38744 37.
Jain, M., G.D. Sha ma, and M. S i as a a. 2019. Can sus ainable
in es men yield be e inancial e u ns: A compa a i e s udy
o ESG indices and MSCI indices. Risks 7 (1): 15. h ps:// doi.
o g/ 10. 3390/ isks 70100 15.
Jain, N., V. Meh o a, and P. Mendi a a. 2023. Pe o mance analysis
o ESG indices: A s ep owa ds ul illing social esponsibili y.
Bimaques 23 (1): 22–37.
Jonwall, R., S. Gup a, and S. Pahuja. 2024. Do socially esponsible
indices ou pe o m con en ional indices? E idence om be o e
and a e he onse o Co id-19. Co po a e Social Responsibili y
and En i onmen al Managemen 31 (5): 4995–5011. h ps:// doi.
o g/ 10. 1002/ cs . 2839.
787Financial e u ns o going g een: e idence omMSCI indices
Kossen ini, H., O. Belhassine, and A. Zenaidi. 2024. ESG index
pe o mance: Eu opean e idence. Jou nal o Asse Managemen
25 (7): 653–665. h ps:// doi. o g/ 10. 1057/ s41260- 024- 00361-4.
Lean, H.H., and D.K. Nguyen. 2014. Policy unce ain y and pe o -
mance cha ac e is ics o sus ainable in es men s ac oss egions
a ound he global inancial c isis. Applied Financial Econom-
ics 24 (21): 1367–1373. h ps:// doi. o g/ 10. 1080/ 09603 107. 2014.
925063.
Lin, X., and R.B. Swain. 2024. Pe o mance o nega i ely sc eened
sus ainable in es men s du ing c isis. In e na ional Re iew o
Economics & Finance 93: 1226–1247. h ps:// doi. o g/ 10. 1016/j.
i e . 2024. 04. 001.
Managi, S., T. Okimo o, and A. Ma suda. 2012. Do socially espon-
sible in es men indexes ou pe o m con en ional indexes?.
Applied Financial Economics 22 (18): 1511–1527. h ps:// doi.
o g/ 10. 1080/ 09603 107. 2012. 665593.
MSCI. 2024a. MSCI SRI indexes me hodology. Re ie ed om:
h ps:// www. msci. com/ docum en s/ 10199/ 2 754 e5a- 310a- 0b5 -
5273- 59216 48b1a 7c.
MSCI. 2024b. MSCI SRI il e ed PAB indexes me hodology.
Re ie ed om: h ps:// www. msci. com/ docum en s/ 10199/ d2112
9da- 4 e6- 9c55- 81d0- eb42d 4931b 51.
Ng, A.C., and Z. Rezaee. 2015. Business sus ainabili y pe o mance
and cos o equi y capi al. Jou nal o Co po a e Finance 34:
128–149. h ps:// doi. o g/ 10. 1016/j. jco p in. 2015. 08. 003.
Omu a, A., E. Roca, and M. Nakai. 2021. Does esponsible in es ing
pay du ing economic down u ns: E idence om he COVID-19
pandemic. Finance Resea ch Le e s 42: 101914. h ps:// doi. o g/
10. 1016/j. l. 2020. 101914.
O as, E., R.L. Bu i , and J.M. Mone a. 2013. Socially esponsible
in es men and cleane p oduc ion in he Asia Paci ic: Does i
pay o be good?. Jou nal o Cleane P oduc ion 52: 272–280.
h ps:// doi. o g/ 10. 1016/j. jclep o. 2013. 02. 024.
O as, E., J.M. Mone a, R. Bu i , and J. Tingey-Holyoak. 2014.
Does sus ainabili y in es men p o ide adap i e esilience
o e hical in es o s? E idence om Spain. Jou nal o Busi-
ness E hics 124 (2): 297–309. h ps:// doi. o g/ 10. 1007/
s10551- 013- 1873-1.
O as, E., J.M. Mone a, and M. Sal ado . 2012. Does socially
esponsible in es men equi y indexes in eme ging ma ke s pay
o ? E idence om B azil. Eme ging Ma ke s Re iew 13 (4):
581–597. h ps:// doi. o g/ 10. 1016/j. emema . 2012. 09. 004.
Papa hanasiou, S., and D. Kou sokos as. 2024. Sus ainabili y a -
ings and und pe o mance: New e idence om Eu opean ESG
equi y mu ual unds. Finance Resea ch Le e s 62: 105095.
h ps:// doi. o g/ 10. 1016/j. l. 2024. 105095.
Pa lo a, I., and M.E. De Boy ie. 2022. ESG ETFs and he COVID-
19 s ock ma ke c ash o 2020: Did clean unds a e be e ?.
Finance Resea ch Le e s 44: 102051. h ps:// doi. o g/ 10. 1016/j.
l. 2021. 102051.
Renneboog, L., J. Te Ho s , and C. Zhang. 2008. The p ice o e h-
ics and s akeholde go e nance: The pe o mance o socially
esponsible mu ual unds. Jou nal o Co po a e Finance. 14
(3): 302–322.h ps:// doi. o g/ 10. 1016/j. jco p in. 2008. 03. 009.
Rompo is, G.G. 2022a. The ESG ETFs in he UK. Jou nal o
Asse Managemen 23 (2): 114–129. h ps:// doi. o g/ 10. 1057/
s41260- 021- 00251-z.
Rompo is, G.G. 2022b. En i onmen al, social and go e nance
esponsibili y, inancial pe o mance and asse s: A s udy o
exchange aded unds. Resea ch Pape s in Economics and
Finance 6 (2): 23–49.h ps:// doi. o g/ 10. 18559/ e . 2022.2.2.
Rompo is, G.G. 2023. The pe o mance o ESG ETFs in he US.
Capi al Ma ke s Re iew 31 (2): 89–101.
Rompo is, G.G. 2024. The pe o mance o an i-ESG ETFs in he
Uni ed S a es. Jou nal o Asse Managemen 25 (7): 700–713.
h ps:// doi. o g/ 10. 1057/ s41260- 024- 00381-0.
Sch öde , M. 2007. Is he e a di e ence? The pe o mance cha ac-
e is ics o SRI equi y indices. Jou nal o Business Finance &
Accoun ing 34 (1–2): 331–348. h ps:// doi. o g/ 10. 1111/j. 1468-
5957. 2006. 00647.x.
S a man, M. 2005. Socially esponsible indexes: Composi ion and
pe o mance. SSRN Elec onic Jou nal. h ps:// doi. o g/ 10. 2139/
ss n. 705344.
Un uh, G., D. Ki on, N. K uschwi z, M. Ree es, H. Rubel, and Meye
zum Felde, A. 2016. In es ing o a Sus ainable Fu u e. MIT Sloan
Managemen Re iew.
Wo ld Heal h O ganiza ion. 2020. A chi ed: WHO Timeline—
COVID-19. Re ie ed om: h ps:// www. who. in / news/ i em/ 27-
04- 2020- who- imel ine--- co id- 19.
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o
ju isdic ional claims in published maps and ins i u ional a ilia ions.
Jan Heldmann is a PhD candida e andResea ch Associa e a he Chai
o Finance and Banking, Uni e si y o Bay eu h, Ge many. He ea ned
bo h his Bachelo ’s deg ee in In e na ional Economics & De elop-
men (2021) and his Mas e ’s in Business Adminis a ion (specializing
in Finance, Accoun ing, Con olling, and Taxa ion) (2023) om he
Uni e si y o Bay eu h.His Mas e ’s hesis, “Empi ical Analysis o
Va ia ions in Fac o Models,” ecei ed he Bay eu h Awa d o Finan-
cial Se ices in 2024. F om 2023 o ea ly 2025, he se ed as a P ojec
Manage o he BMBF- unded ini ia i e i led “Clima e Repo ing
by SMEs” (Klimabe ich e s a ung bei KMUs). His esea ch in e es s
include ESG, isk managemen , and ac o models.
Thomas B ückne is a Mas e ’s candida e a he Chai o Finance and
Banking, Uni e si y o Bay eu h, Ge many. He ea ned his Bachelo ’s
deg ee in Finance in 2022 and will comple e his Mas e ’s s udy in
2025, wi h a hesis i led “Fac o models on he Ge man s ock ma -
ke : An empi ical compa ison.” Since 2023, he has wo ked as a Clien
Accoun Manage specializing in he Exchange-T aded Fund business
a Amundi In es men Solu ions (Ge many), a leading Eu opean asse
manage and one o he op h ee asse manage s in e ms o o ing pe -
o mance on en i onmen al and social issues in 2023. In 2024, he was
selec ed o C édi Ag icole’s DEJou ney alen de elopmen p og am
(2024–2025). His esea ch in e es s include inancial pe o mance,
ESG, and isk managemen .
Huong Dieu Dang CFA, FRM is Senio Lec u e abo e he Ba a he
Uni e si y o Can e bu y, New Zealand, and was a Visi ing P o esso
a he Uni e si y o Bay eu h, Ge many. She ea ned an MBA om he
Uni e si y o A izona (2006) and an MSc in Managemen (Finance
concen a ion) om he Uni e si y o A izona (2006), be o e ob aining
a PhD deg ee in Economics (Finance Discipline) a he Uni e si y o
Sydney (2010). Since 2020, she has se ed as an Associa e Edi o o
he Jou nal o Applied Accoun ing Resea ch published by Eme ald. He
esea ch in e es s include c edi isk, in es men , social no m, and sus-
ainabili y. He co-au ho ed esea ch has been published in espec ed
pee - e iewed jou nals such as he Jou nal o Financial Economics,
ABACUS, and Paci ic-Basin Finance Jou nal.