The pricing of sustainability-linked bonds on the primary and secondary bond markets

Poggensee, Jannis
A icle — Published Ve sion
The p icing o sus ainabili y-linked bonds on he p ima y
and seconda y bond ma ke s
Jou nal o Asse Managemen
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Poggensee, Jannis (2025) : The p icing o sus ainabili y-linked bonds on he
p ima y and seconda y bond ma ke s, Jou nal o Asse Managemen , ISSN 1479-179X, Palg a e
Macmillan UK, London, Vol. 26, Iss. 4, pp. 411-431,
h ps://doi.o g/10.1057/s41260-024-00390-z
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323680
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/
ORIGINAL ARTICLE
The p icing o sus ainabili y-linked bonds on he p ima y
and seconda y bond ma ke s
Jannis Poggensee
1
Re ised: 13 No embe 2024 / Accep ed: 17 No embe 2024 / Published online: 4 Feb ua y 2025
ÓThe Au ho (s) 2025
Abs ac This pape in es iga es he yield di e en ials
be ween sus ainabili y-linked bonds (SLBs)—a no el ixed
income ins umen whose coupon paymen s a e linked o
he achie emen o p ede ined sus ainable pe o mance
a ge s—and con en ional bonds issued by he same
company, bo h on he p ima y and seconda y bond ma -
ke s. By i ing yield cu es wi h he Nelson–Siegel
S ensson me hod on he SLB’s p icing day and applying
panel eg essions o examine yield di e en ials he ea e ,
he s udy assesses whe he SLBs ade a a p emium,
indica ing highe p ices and consequen ly lowe yields.
Fixed-e ec s panel eg ession is u ilized o isola e he
unobse ed ime-in a ian yield di e en ial be ween SLBs
and a ma ched syn he ic con en ional bond wi h he same
esidual ma u i y. The esul s show a s a is ically signi i-
can bu economically small p emium o SLBs in bo h he
p ima y and seconda y ma ke . The sus ainabili y p emium
is no signi ican ly d i en by he SLB’s penal y s uc u e
and luc ua es o e ime. This esea ch con ibu es o he
li e a u e by applying a no el me hodological amewo k
o examine he e ol ing na u e o SLB p emiums and hei
implica ions o bo h issue s and in es o s.
Keywo ds Sus ainable inance G een p emium Bond
p icing Sus ainabili y-linked bonds Socially esponsible
in es ing
In oduc ion
Clima e change is one o he mos challenging issues o ou
ime. Th ough physical isks, which co e en i onmen al
disas e s and ansi o y isks, which a e business- ela ed
isks ha e ol e due o socie al and economical shi s
owa d a low-ca bon u u e, clima e change poses a isk o
he economy and he inancial sys em. This pushes o a
undamen al con e sion owa d mo e sus ainable p oduc-
ion and consump ion p ocesses ha equi e subs an ial
in es men s.
Asse ma ke s will play a key ole in inancing he g een
ansi ion and di ec ing cash lows o companies wi h he
bes s a egies and e o s o comba clima e change and i s
consequences. Consequen ly, new inancing ins umen s
eme ged in deb ma ke s in ecen yea s. A e he i s
emission by he Eu opean In es men Bank (EIB) in 2007,
he g ow h o g een bonds soa ed, and his end is likely o
hold in he u u e (Flamme 2021). Mo eo e , addi ional
ypes o sus ainable deb in e ms o social and sus ainable
bonds (GSS; g een, social and sus ainable bonds in he
ollowing) ha e been issued un il oday. In 2022, GSS
bonds had an issuance olume o 795 billion USD
acco ding o Re ini i ’s ESG Bond Guide Da abase.
Re lec ing he g owing impo ance o ESG- ela ed deb ,
he academic li e a u e s a ed ocusing on he p icing o
g een bonds, pa icula ly he ex en o which g een bonds
ade a a p emium—which is e e ed o as ‘‘g eenium’’
whe e p ices o g een asse s a e highe (and consequen ly
yields lowe ) compa ed o con en ional bonds. E idence in
his ield in a o o g een p emiums—‘‘g eenium’’—has
been mixed so a (e.g., Kap aun e al. 2021). Al hough he
inc eased anspa ency and accep ance o GSS bonds migh
lead o lowe bond yields on he p ima y and on he sec-
onda y ma ke , he lack o a uni o m de ini ion o eligible
&Jannis Poggensee
[email p o ec ed]
1
QBER - Ins i u u
¨ Quan i a i e Be iebs- und
Volkswi scha liche Fo schung, Ch is ian-Alb ech s-
Uni e si a
¨ zu Kiel, Olshausens aße 40, 24098 Kiel,
Ge many
Jou nal o Asse Managemen (2025) 26:411–431
h ps://doi.o g/10.1057/s41260-024-00390-z
p ojec s would dampen he c edibili y o he ma ke lea -
ing i s ne e ec on bond e u ns unclea (Bundesbank
2019). Besides, pa icula ly g een bonds some imes lack
he c i e ion o ‘‘addi ionali y,’’ because hey e inance
exis ing p ojec s and asse s bu no ep esen ing an inno-
a i e beyond ‘‘business as usual’’ ajec o y o educe
g eenhouse gas (GHG) emissions (Maino 2022). Finally,
g een bonds axonomies do no ex ensi ely include some
‘‘ha d- o-aba e’’ sec o s like ene gy o manu ac u ing
which, howe e , ha e high deb capi al equi emen s o
p e inance he sus ainable con e sion o he p oduc ion
p ocess.
In esponse o hese limi a ions, sus ainabili y-linked
bonds, SLBs in he ollowing, ha e eme ged as no el and
al e na i e ixed income ins umen o e ing a di e en
app oach o sus ainable inance. Unlike GSS bonds, which
es ic inancing o speci ic en i onmen ally bene icial
p ojec s (e.g., enewable ene gy, ene gy e iciency o pol-
lu ion p e en ion),
1
SLBs o e issue s he lexibili y o
inance gene al-pu pose ini ia i es while commi ing o
measu able sus ainabili y a ge s. Those a ge s, which can
ela e o en i onmen al, social o go e nance ou comes,
c ea e inancial incen i es o issue s o mee p ede ined
sus ainabili y goals. Nonachie emen o hose a ge s leads
o a ying inancial cha ac e is ics, mos commonly a
coupon s ep-up (ICMA 2020).
This makes SLBs pa icula ly a ac i e o companies
aiming o in eg a e sus ainabili y ac oss hei ope a ions,
while p o iding in es o s wi h a amewo k ha holds
issue s accoun able. Gi en he ma u i y o he g een bond
ma ke and he mixed e idence ega ding he exis ence o a
‘‘g eenium,’’ he shi o in es iga ing SLBs is essen ial.
Thei lexible app oach equi es empi ical examina ion o
unde s and how inancial ma ke s p ice SLBs compa ed o
con en ional bonds.
The In e na ional Capi al Ma ke Associa ion (ICMA),
which se es as he sec e a ia o he SLB-p inciples, made
wo key p o isions o how SLBs can add ess he c i ics
g een bonds a e con on ed wi h (Vul u ius e al. 2022).
Fi s , SLBs a e in ended o inance gene al-pu pose p o-
jec s ha ha e company-wide sus ainabili y objec i es,
p io i ized by he sus ainable a ge a company selec s. In
pa icula , sus ainabili y pe o mance a ge s (SPT) should
be beyond a ‘‘business as usual’’ ajec o y, ela ed o sci-
ence-based scena ios (like he Science-Based Ta ge Ini-
ia i e, SBTi) o policy a ge s (Pa is Ag eemen ). Pas
pe o mance o he issue agains a Key Pe o mance
Indica o should be epo ed o e a pe iod o a leas h ee
yea s in addi ion o he ele ance and posi ioning o he
sus ainable pe o mance a ge agains indus y pee s.
Secondly, he no el SLB-penaliza ion scheme should
incen i ize he issue ’s e o o each i s a ge s and signal
in es o s ha he issue c edibly a emp s o dampen he
nega i e consequences a ising om clima e change.
2
Despi e hei po en ial, ecen indus y epo s s ess
g owing in es o skep icism owa d SLBs.
3
Some a e
accused o g eenwashing, migh i be due o unambi ious
sus ainable pe o mance a ge s o a coupon s ep-up being
no puni i e enough. Fo ins ance, Tesco’s SLB, linked o
he educ ion o g eenhouse gas emissions, only co e s 2%
o i s o al emissions. US-based Le el 3 Communica ion’s
$900 million SLB ma u ing in 2029 can be edeemed by
he in es o in Janua y 2024 be o e any penal y o no
achie ing he sus ainabili y a ge s kicks in. Al oge he ,
such loopholes unde mine he c edibili y o he ma ke and
weaken in es o con idence.
Conside ing hese challenges and gi en he lexibili y
and po en ial o SLBs o incen i ize company-wide sus-
ainabili y, his pape aims o in es iga e how SLBs a e
p iced in compa ison wi h con en ional bonds. I is
assessed whe he SLBs exhibi a sus ainabili y p emium—
like g een bond’s ‘‘g eenium’’—and how SLB-speci ic
ea u es like he penal y scheme migh a ec he p emium.
The pape aims o ill a signi ican esea ch gap by p o-
iding a comp ehensi e analysis o SLB-p icing beha io
in bo h p ima y and seconda y ma ke s. Unde s anding he
p icing dynamic o SLBs is c i ical o bo h issue s ha
ha e a high need o inancing due o he equi emen s o
he g een ans o ma ion o a low-emission economy and
he e o e seek o op imize inancing cos , and in es o s,
who aim o balance he inancing o i ms wi h a c edible
commi men owa d comba ing clima e change wi h
e u ns.
To he bes o my knowledge, no o he s udy has sys-
emically compa ed SLB yields o con en ional bonds
om he same issue ac oss bo h he p ima y and sec-
onda y ma ke s. This s udy also con ibu es o he b oade
discussion o whe he sus ainable asse s, such as SLBs,
exhibi a p emium (i.e., ade a highe p ices and lowe
yields) due o in es o s’ as e o such asse s, which may
bene i issue s by lowe ing cos o deb (i.e., Ze bib 2019;
Bachele e al. 2019; Hachenbe g and Schie eck 2018).
Addi ionally, his s udy adds o he discussion abou
de e minan s o yield di e en ials be ween sus ainable and
con en ional asse s (Bachele e al. 2019; Kap aun e al.
1
Fo an o e iew o eligible p ojec s, see he g een bond p inciples
o he ICMA h ps://www.icmag oup.o g/asse s/documen s/Sus ain
able- inance/2022-upda es/G een-Bond-P inciples_June-2022-
280622.pd
2
Pa icula ly by se ing ambi ious and science-based pe o mance
a ge s, se ing su icien ly high penal ies and issuing SLBs wi h no
embedded call op ions o minimize he pe iod in which he coupon
s ep-up has o be paid issue s os e he c edibili y o SLBs.
3
h ps://www.en i onmen al- inance.com/con en /analysis/slbs-a -
in lec ion-poin -in-2023-as-sus ainable-bond-ma ke - ebounds.h ml
412 J. Poggensee
2021; La cke and Wa s 2020) and which SLB ea u es
migh a ec he p icing di e en ials in he c oss sec ion
(Ko
¨lbel and Lambillon 2022; E landsson and Mielnik
2022). Finally, he s udy is pa icula ly imely gi en he
ise in g eenwashing conce ns in he SLB ma ke . Se e al
con ibu ions (Libe a o e2021; Haq and Doumbia 2022;
Reznick e al. 2022) blame SLBs being embedded wi h
weak penal y s uc u es, sus ainabili y a ge s ha a e oo
easy o achie e o ea ly call op ions so ha he SLB can be
edeemed be o e he penal y mechanism ma e ializes.
To da e, Ko
¨lbel and Lambillon (2022) o e he mos
comp ehensi e s udy o SLBs; howe e , hei esea ch is
limi ed o yield compa isons a he issuance s age (p ima y
ma ke ). This pape goes u he by ex ending he analysis
o bo h p ima y and seconda y ma ke s. Theo e ically, he
analysis is based on Pede sen e al. (2021) and Pas o e al.
(2021) de eloping ESG-adjus ed asse p icing models and
con ibu es o he deba e whe he in es o de i es u ili y
om sus ainable in es men s and hence would be willing
o pay highe secu i y p ices, i.e., accep lowe e u ns.
Gi en he lexibili y and po en ial o SLBs o incen i ize
company-wide sus ainabili y, expanding he li e a u e o
SLBs is c ucial o p ac ical implica ions in sus ainable
inance. This aises impo an ques ions in e ms o how
SLBs a e p iced and how SLB-speci ic ea u es such as he
penal y s uc u e migh impac he p emium, u he jus i-
ying he need o empi ical esea ch.
Fi s , a p ima y ma ke analysis is applied by i ing
yield cu es wi h he Nelson–Siegel–S ensson me hod
(NSS) o cons uc yield cu es o con en ional bonds
issued by he same company. A e wa d, he SLB yield is
o e laid, o examine whe he he SLB p iced below (would
imply a sus ainabili y p emium) in line o abo e (would
imply ha SLBs ade a highe yields and ca y highe cos
o deb o he issue ) i s yield cu e. Mo eo e , he SLB
p icing in he seconda y is in es iga ed by applying a ixed-
e ec s panel eg essions which compa es he yields o
SLBs and a ma ched syn he ic bond o in es iga e whe he
a po en ial sus ainabili y p emium would pe sis on he
seconda y ma ke . Finally, he ea u es ha may d i e yield
di e en ials be ween he wo bond ypes a e examined.
The esul s p o ide se e al empi ical indings. Fi s , I
p o ide an o e iew o he nascen SLB ma ke , i s
p omising amewo k and some s uc u al and inancial
cha ac e is ics making his ixed income ins umen p one
o g eenwashing isks. Empi ically, I show ha SLBs ade
a a sus ainabili y p emium on he p ima y ma ke which
implies ha issue s bene i om lowe cos o capi al and
in es o s accep lowe yields o holding a sus ainabili y-
linked asse . The p emium o 4.68 basis poin s on a e age
declines o 3 basis poin s when mo ing o he seconda y
ma ke . Mo eo e , he p emium is ola ile o e ime and
anishes du ing he second hal o 2022. Finally, c oss
sec ionally, he e a e di e ences ega ding he p emia wi h
some SLBs exhibi highe yields compa ed o con en ional
bonds. Companies wi h he highes ESG sco e and com-
panies ha ollow science-based a ge s p iced igh e , i.e.,
hose can expec lowe cos o deb . Con a y, he cumu-
la i e size o he s ep-up in ela ion o he coupon o he
bond is no a signi ican d i e o he SLB p emium,
implying ha his op ional s ep-up is no a p ima y d i e
o SLB p icing.
The emainde o he pape is s uc u ed as ollows:
Sec ion ‘‘Rela ed li e a u e and hypo heses’’ e iews he
ela ed li e a u e and posi s hypo heses. Sec ion ‘‘The
sus ainabili y-linked bond ma ke ’’ p o ides a ma ke
o e iew. Sec ion ‘‘Da a selec ion and me hodology’’
highligh s he sample selec ion p ocess and he applied
me hodologies o he empi ical pa in sec ion ‘‘Empi ical
esul s’’. Sec ion ‘‘Robus ness’’ p o ides obus ness es s.
Finally, sec ions ‘‘Discussion’’ and ‘‘Conclusion’’ discuss
he indings and conclude.
Rela ed li e a u e and hypo heses
The academic esea ch so a p ima ily ocused on egula
g een bonds, because o hei his o ical dominance in e ms
o issuance olume. Leading con ibu ions in he ield
mainly p o ided mixed e idence whe he g een bonds
ade a a ‘‘g eenium’’ implying lowe yields o g een
bonds compa ed o hei con en ional bond win (Ze bib
2019; Bake e al. 2018; Hachenbe g and Schie eck 2018;
Gian a e and Pe i 2019) o , in con as , ei he a highe
yields as epo ed in Ka p and Mandel (2018) and
Bachele e al. (2019) o a no di e ence as in (La cke and
Wa s 2020; Flamme 2021). The mixed e idence is p e-
sumably a ibu able o di e en ime pe iods, issue ypes,
ma ke s and empi ical app oaches. Fo example, some
s udies apply a ma ching app oach o ind a g een bond’s
‘‘ win’’ and a e wa d eg ess i s yield sp ead on emaining,
no exac ly ma ched explana o y a iables (Ze bib 2019;
Bachele e al. 2019), while o he s udies—ins ead o il-
e ing ou mos o he con en ional bonds h ough a
ma ching p ocedu e— eg ess he bond yields on a se o
con ols and a g een dummy a iable, which is he e ec o
being g een on he bond yield sp ead (Bake e al. 2018;
Fa ica e al. 2021). A ecen s udy by Kap aun e al.
(2021)—equipped wi h he la ges da ase o da e—con-
cludes ha in es o s became skep ical abou he en i on-
men al impac o g een bonds and hose only exhibi a
g een bond p emium (o -4 bps), i hey a e ex e nally
ce i ied as well as ha ing a high sus ainable epu a ion.
O e all, he mixed indings on g eenium in g een bonds
aise ques ions abou whe he SLBs, despi e hei s uc u al
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 413
di e ences, migh exhibi a simila sus ainabili y p emium.
Acco dingly, his s udy’s i s hypo hesis is:
1. SLBs exhibi a sus ainabili y p emium in bo h p ima y
and seconda y bond ma ke s, simila o he g eenium
concep in g een bonds.
Coun e o he abo e su eyed li e a u e on g een bonds
esea ch on SLB bonds is sca ce. Mos o he exis ing
li e a u e is quali a i e, discussing he mechanism o SLBs
and how SLBs could o e come he g eenwashing-conce ns
GSS bonds a e con on ed wi h (Vul u ius e al. 2022;
Gi aldez and Fon ana 2022; Maino 2022). The ew
empi ical s udies a e as ollows: Libe adzki e al. (2021)
ind ha he SLB issued by he B i ish g oce y and e aile
Tesco p iced lowe han con en ional bonds o he same
issue om F ench Ca e ou and Ge man Me o du ing he
i s hal o 2021 e en unde he coupon s ep-up scena io.
Howe e , hey ma ch bonds wi h la ge di e en ial in
ma u i y. Ko
¨lbel and Lambillon (2022) ind ha SLBs p ice
29.2 bps lowe han ma ched con en ional bonds o he
same issue a issuance da e on he p ima y ma ke . Since
he a e age s ep-up in hei sample is 26.6 bps, issue s
would bene i om a sus ainabili y p emium, e en i hey
do no achie e hei p ede ined pe o mance a ge . They
conclude SLBs a e a ‘‘ ee lunch’’ o issue s. Be ada e al.
(2022) c ea e a concep ual amewo k ha examines unde
which condi ions SLBs incen i ize company manage s o
exe e o , which will occu when he penal y sa ing is
highe han he mone a y cos o exe ing e o o achie e
he pe o mance a ge . Mo eo e , en i onmen al-con-
ce ned in es o s ha obse e manage ’s e o , de i e a
bene i om he imp o ed pe o mance and a e willing o
pay highe bond p ices. The e o e, companies wi h a mo e
c edible ESG-s a egy could exhibi highe SLB p emia as
hypo hesis 2 s a es:
2. Building on indings ha issue cha ac e is ics, such as
ESG sco es, in luence bond p icing, issue -speci ic
cha ac e is ics a e associa ed wi h an SLB p emium.
While SLBs hold po en ial o add ess ce ain limi a ions
o g een bonds, empi ical esea ch on hei p icing emains
limi ed, especially in compa ison wi h g een bonds. As
mos s udies ocus on he p ima y ma ke , hey may no
accoun o changing ma ke condi ions, which could a ec
SLB p icing o e ime. Fo ins ance, Ko
¨lbel and Lambillon
(2022) only compa e p ima y ma ke issuance yields, while
Be ada e al. (2022) explo e SLB misp icing by se ing
uppe and lowe p icing bounds bu do no examine yield
beha io o e a longe ime span. The ole o SLB-speci ic
ea u es, pa icula ly penal y mechanisms, is also an a ea
needing u he explo a ion. Exis ing esea ch sugges s ha
issue s could bene i om lowe cos s o capi al by se ing
ambi ious sus ainabili y a ge s, subs an ial coupon s ep-
ups and longe pe iods o achie e hese a ge s (E landsson
and Mielnik 2022). These au ho s a gue ha g eenium in
SLBs should accoun o he op ion alue o he s ep-up
clause, as i can c ea e op ionali y o ecei e highe pay-
men s i a ge s a e missed. The e o e, hypo hesis 3 posi s:
3. Gi en he ole o penal y s uc u es in SLBs, i s
speci ic ea u es, such as penal y magni ude o
missed sus ainabili y a ge s, a ec he p emium.
Despi e he apid g ow h and impo ance o SLBs, sig-
ni ican gaps emain in unde s anding hei p icing, pa -
icula ly in seconda y ma ke s whe e changing condi ions
could impac yields. Gi en he s uc u al di e ences
be ween SLBs and g een bonds, i is essen ial o in es iga e
whe he SLBs demons a e simila p emium and which
unique ac o s migh d i e hei p icing dynamics.
The sus ainabili y-linked bond ma ke
The inaugu al bond was launched by I alian u ili y Enel
S.p.A. in Sep embe 2019. The single anche, o aling 1.5
billion USD, ma u ing in Sep embe 2024 and paying a a e
o 2.65% is subjec o he pe cen age o ins alled capaci y
in enewable ene gy. I a 55 pe cen age o ins alled
enewable gene a ion capaci y (as a sha e o o al consol-
ida ed ins alled capaci y) had no been eached un il
Decembe 31s , 2021, he coupon would ha e s epped up
by ?25 bps o 2.9% p.a. un il he bonds’ expi y da e. In
2021, he pe o mance a ge has been success ully
eached; he e o e, he a e emained a a 2.65% p.a. le el.
4
Acco ding o Enel, he ansac ion was success ul, enabling
he I alian ene gy company o ob ain a inancial ad an age
equal o 20 bps compa ed o a con en ional bond issuance
(Enel 2019).
5
In he wake o he publica ion o he ICMA guidelines in
June 2020, i e co e componen s we e es ablished o ensu e
a ce ain s anda d including: selec ion o a Key Pe o -
mance Indica o , calib a ion o a sus ainabili y pe o -
mance a ge , de ini ion o he a ying bond cha ac e is ics
(i.e., mos commonly he magni ude o he coupon s ep-
up), epo ing s anda ds and hi d-pa y e i ica ion. The
p io s agnan issue olume o SLBs sky ocke ed. In
Sep embe 2020, he B azilian pulp and pape company
4
Enel epo ed o ha e eached 57.5% o ins alled new enewable
gene a ion capaci y by he end o 2021.
5
h ps://www.enel.com/con en /dam/enel-common/p ess/en/2019-
Sep embe /SDG%20bond%20ENG%20(003).pd
414 J. Poggensee

Suzano issued he second $750 million SLB. As Fig. 1
shows, he las qua e in 2020 saw an issuance olume o
$8 billion ollowed by a soa ing g ow h in 2021. As o all
(g een) deb asse s, he issuances in 2022 declined, a ec ed
by ising in e es a es, a highe economic unce ain y, a
wo sening isk pe cep ion and he wa in he Uk aine.
Appendix Fig. 3shows ha hal o SLBs a e denomi-
na ed in Eu o, ollowed by he USD. Toge he , he wo
cu encies accoun o 86% o he issued olume. Nex ,
Appendix Fig. 4plo s he issuance olume ac oss sec o s in
pe cen . S ikingly, ca bon in ensi e ‘‘ha d o aba e sec-
o s’’ like u ili ies, indus ials o basic ma e ials ha a e
some imes no compa ible wi h he g een bond s anda ds
(Maino 2022), ep esen a high po ion o SLB issuance.
Abou hal o he bond’s pe o mance indica o s a e
ela ed o a g eenhouse gas emission a ge as Fig. 5shows.
Among hose, o e 80% ocus on scope 1 and scope 2
emissions,
6
whe eas ype 3 emissions, which a e ha de o
measu e bu o en ep esen ing mos o a company’s
emissions,
7
) a e a ely included. Emission- ela ed KPIs
ha e he ad an age ha hey can be aligned wi h science-
based scena ios mo e easily and some issue s had a his o y
o epo ing be o e he issuance.
Since he issuance o SLBs is based on olun a y
guidelines and p inciples es ablished by he ICMA—lea -
ing oom o di e en in e p e a ions—and no public
au ho i y se ing a legal amewo k ha moni o s he
issuance, SLBs could be p one o he isk o g eenwashing
p ima ily due o a small size o he (cumula i e) coupon
s ep-up ha is no ma e ial, callable ea u es o edeem he
SLB be o e o sho ly a e he s ep-up ma e ializes o
unambi ious pe o mance a ge s achie ed easily by he
issue . Figu e 6plo s he size o he coupon s ep-ups. The
igu es indica e ha mo e han 50% o he SLBs a e
embedded wi h 25 bps penal y pe annum ei he solely o
agg ega ed i a bond ca ies mo e han one KPI. Rega dless
o he issue s a ing i s size o i s coupon- a e, i is common
p ac ice o se he ee o 25 bps pe annum.
8
In summa y, SLB ha e eme ged as p omising ins u-
men o sus ainable inance, o e ing lexibili y and
inancial incen i es o issue s o achie e he s e ed pe -
o mance a ge s. Howe e , he ma ke could be p one o
g eenwashing, as issue s choose he a ge s independen ly
and se he penal y.
Da a selec ion and me hodology
The nex sec ions ocus on he inal sample selec ion p o-
cess and he applied me hodologies ha in es iga e he
p icing o SLBs on he p ima y and seconda y bond
ma ke s.
Sample selec ion
Mos o he da a o igina es om Re ini i ’s ESG Bond
Guide Da abase, which p o ides s a ic GSS and SLB
bonds, including size, ma u i y, issuance day, senio i y and
yield a issuance. Since he issuance o he inaugu al SLB
bond in 2019, Re ini i has lis ed 526 SLBs as o he
epo ing da e Decembe 31
s
, 2022. Secondly, Re ini i
Eikon’s bond iewe app displays SLB e ms, o ins ance,
he KPI desc ip ion, he magni ude o he coupon s ep-up
o he coupon paymen da es. Since he SLB de ails a e
pa ly no ye comp ehensi e, e ms we e also collec ed
manually, based on company p ess eleases, in es o ela-
ions, he o e ing memo andum and bond p ospec uses.
Addi ionally, he bond iewe app p o ides he deb
s uc u e o he issue including all o he ac i e con en-
ional bonds ading on he deb ma ke and gi es access o
p icing and yield da a. I download he ime se ies da a o
all issue s’ ac i e bonds since he issuance day o he SLB
un il Sep embe 30 h, 2022. Time se ies da a includes bid
and ask yields, he yield o ma u i y, swap sp eads, ask and
bid p ices, modi ied du a ion and he emaining ime o
edemp ion.
F om he ini ial 526 bonds, 76 bonds a e d opped ha
exhibi a bi u ca ed s uc u e.
9
Secondly, 70 bonds wi h a
4,2
8,6
97,42
71,32
2019 2020 2021 2022
VOLUME
YEAR
SLB ISSUANCE VOLUME IN
BILLION USD
Fig. 1 Issuance olume. This igu e plo s he amoun o issued SLBs
in billion USD be ween 2019 and 2022 acco ding o da a om
Re ini i .
6
The g eenhouse gas p o ocol di e en ia es be ween di e en scopes
o emissions. Scope 1 emissions a e di ec emissions a company
causes du ing i s p oduc ion p ocess. Scope 2 emissions a e indi ec
emissions om he gene a ion o pu chased elec ici y, hea o s eam.
Scope 3 emissions a e
consequences o a company’s indi ec ac i i ies in i s alue chain.
7
Acco ding o Deloi e, Scope 3 is nea ly always he bigges one and
accoun s o mo e han 70% o i ms ca bon oo p in (h ps://www2.
deloi e.com/uk/en/ ocus/clima e-change/ze o-in-on-scope-1-2-and-3-
emissions.h ml
8
S ep-ups make up app oxima ely 75% o he penaliza ion scheme.
Besides s ep-ups, he o he common inancial incen i es associa ed
wi h KPIs a e coupon s ep-downs, edemp ion p emia whe e he
issue pays a p ede e mined p emium on he edemp ion p ice a
ma u i y, dona ions o ounda ions o o ganiza ions o he issue ’s
choice, ea ly edemp ion whe e he bond is edeemed ea lie a a
p ede ined p ice and he pu chase o ca bon c edi s p opo iona e o
he p incipal amoun
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 415
a ying coupon- ype and 109 bonds ha ha e ei he issued
no o he o a single bond a e le ou . Re ini i Eikon lacks
con inuous yield and p icing da a o some Chinese issue s,
so hose a e also no conside ed. Addi ionally, only SLBs
ha ha e been issued be o e June 30 h, 2022 a e included
wi hin he sample and all bonds ha ha e a callable ea u e
a e excluded because hey could a ec he p icing. Fol-
lowing Be ada e al. (2022), I keep bonds wi h a clean-up
call op ion (whe e he bond can be called wi hin he las
h ee mon hs o ma u i y) and wi h a make- whole callable
ea u e ( edeeming he bond is associa ed wi h highe cos s
o he company). Table 1shows he sample selec ion
p ocedu e in mo e de ail. To ensu e compa abili y be ween
SLBs and con en ional bonds, a ma ching app oach was
applied based on speci ic c i e ia such as issue , cu ency,
coupon s uc u e and c edi a ing. These condi ions we e
essen ial in iden i ying compa able bonds o he inal
sample o 45 SLBs.
Me hodology o p ima y ma ke analysis
The p ima y goal o he analysis is o de e mine whe he
SLBs exhibi a sus ainabili y p emium on he issuance day
by compa ing hei yields o hose o ma ched con en ional
bonds. This is done by applying he Nelson–Siegel–
S ensson (NSS) me hod o cons uc yield cu es o each
issue and o e laying he SLB’s yield o assess i s ela i e
posi ioning.
In he i s s ep, he p esence o yield di e en ials
be ween SLBs and ma ched con en ional bonds—based on
he issue , coupon s uc u e, cu ency and a ing s uc-
u e—is in es iga ed on he p icing da e o he SLB, whe e
i s yield, coupon and p ice a e de e mined subsequen ly o
he ini ial p ice alks, he book-building p ocess and he
alloca ion o in es o s. The SLB’s yield a issuance ep-
esen ing he yield he in es o ecei es i he bond is held
ill ma u i y and pe o mance a ge s a e me (and he
penal y mechanism does no ma e ialize)
10
—is compa ed
o he yield o ma u i y o con en ional bonds al eady
ading in he seconda y ma ke . The compa ison allows o
iden i y i SLBs a e p iced a lowe yields.
The e o e, he me hod o he Clima e Bond Ini ia i e
(CBI 2017) is ex ended by applying he Nelson–Siegel–
S ensson (NSS) me hod (Nelson and Siegel 1987; S ens-
son 1994). Typically, cen al banks es ima e so e eign
yield cu es o ob ain an empi ical ep esen a ion o he
e m s uc u e o in e es a es, se ing as a key e e ence
poin o many o he ma ke s (Ande sen 2018). The
Appendix discusses he NSS and i s applica ion mo e
de ailedly.
The NSS me hod in e s ( heo e ical) yields o e he
bonds’ emaining ime o ma u i y spec um isualized by
i ing a yield cu e o each o he 45 SLB-bond issue s on
SLB’s p icing day. A e wa d, he yield o he SLB is
o e layed o de e mine whe he i s yield a issuance is
below, in line o abo e i s cu e. I an SLB bond is below
i s yield cu e i sugges s a sus ainabili y p emium, whe e
in es o s ecei e a lowe yield, and issue s ace lowe cos
o deb as hey would ha e expec ed o ecei e i hey
would had held (issued) a con en ional bond which is
in e p e ed as e idence o sus ainabili y p emium in es-
o s pay.
This app oach o e ing a new pe spec i e compa ed o
he adi ional g een bond li e a u e, which ypically
eg ess a panel o issuance yields o g een and con en-
ional bonds on a g een dummy a iable along wi h issue
and bond-speci ic da a (Kap aun e al. 2021; Bake e al.
2018; Fa ica e al. 2021).
Me hodology o seconda y ma ke analysis
To analyze he yield di e en ials be ween SLBs and con-
en ional bonds on he seconda y ma ke subsequen ly, a
s ic e ma ching app oach was applied as in Ze bib (2019),
Bachele e al. (2019), Kap aun e al. (2021), La cke and
Wa s (2020), Flamme (2021), o Gian a e and Pe i
(2019). Unlike adi ional eg ession me hods, eg essing
he bond yields on hei cha ac e is ics and a ‘‘g een’’
Table 1 Final sample
Sample educ ion Numbe o SLBs
Ini ial sample 526
Less issued a e June 30 h, 2022 -89
Less bonds wi h bo h RegS and 144A o e ing -76
Less loa ing coupon s uc u e -56
Less Chinese issue s/missing p icing da a -63
Less no CB -56
Less 1 o 2 CB -53
Less ma u i y ma ching -35
Less o he ma ching c i e ia -53
Sample size 45
Table 1Summa izes he selec ion p ocess, showing he inal sample
o 45 SLBs a e he exclusions.
9
152 SLBs a e issued in a bi u ca ed s uc u e ha ing bo h a 144A
o e ing, whe e he secu i ies p ima y o e ed o US in es o s and a
Regula ionS o e ing, ha co e s an in es o base ou side he US. I is
common o companies o issue bo h 144A and Regula ionS
secu i ies. Re ini i lis s bo h he RegS and he 144A po ion o he
bond wi hin i s da abase, bu o a oid double-coun ing he o als o
he RegS o e ing ype a e no included and coun ed a second ime.
10
I s indus y p ac ice o quo e SLB yields wi hou conside ing he
po en ial (coupon) penal y (Be ada e al. 2022). All he collec ed
SLB da a om Re ini i is quo ed wi hou he possible s ep-up.
416 J. Poggensee
dummy a iable o he sus ainable label, a ma ched-pai s
app oach is applied ha allows o isola ing he ‘‘sus ain-
abili y’’ label e ec di ec ly by con olling o bond cha -
ac e is ics ha impac bo h SLBs and con en ional bonds
simila ly. A ma ched-pai s app oach o e s ad an ages by
con olling o a iables ha a ec yield sp eads simila ly.
This me hod isola es he speci ic e ec o he sus ainable
label on bond yields by c ea ing syn he ic bond wi h he
same esidual ma u i y as he SLB. As no ed by Ze bib
(2019) and Bachele e al. (2019), his app oach is mo e
app op ia e o dis inguishing yield di e en ials because i
minimizes he noise ha could esul om unobse ed
di e ences be ween bond cha ac e is ics, which a e no
cap u ed by dummy a iables. I has become a widely
accep ed me hod in he g een bond li e a u e o measu ing
he p icing e ec o sus ainabili y labels (Kap aun e al.
2021).
To achie e a highe deg ee o accu acy, only he wo
closes and mos simila bonds, ega ding edemp ion da e,
amoun ou s anding and coupon size, a e selec ed o
compa ison. Via linea in e - and ex apola ion, a syn he ic
bond is cons uc ed wi h he same esidual ma u i y as he
SLB. By compa ing hese syn he ic bonds wi h he SLBs,
he ma ched-pai s app oach can mo e p ecisely measu e
any sus ainabili y p emium.
Following Ze bib’s (2019) a gumen a ion, only di e -
ences in liquidi y and ma u i y which bo h canno exac ly
be ma ched should explain yield di e en ials be ween
bonds o he same issue . The e o e, a syn he ic bond is
c ea ed by selec ing wo con en ional bonds wi h he
closes ma u i y, spanning an in e al o ?/- 3 yea s o he
SLB again ha ing he same cha ac e is ics (same issue ,
cu ency, senio i y (paymen ank), callable o bulle
s uc u e, a ing) excep he sus ainabili y label. On each
espec i e day o he emaining SLB’s eno , he ask yield
o he syn he ic bond is linea ly in e - o ex apola ed o
accoun o ma u i y di e ences. The in e ed heo e ical
yields o he syn he ic bond ma ch he SLB’s esidual
ma u i y.
Beyond he ma ching app oach and he c ea ion o he
syn he ic bond, liquidi y di e en ials emain a po en ial
d i e o ask yield di e en ials. The e o e, he ask yield
di e en ial o he bond pai s Dy
i,
is eg essed on he bond
pai s bid–ask sp ead ia ixed-e ec s panel eg ession
ollowing he me hodology used by Ze bib (2019),
Kap aun e al. (2021), Bachele e al. (2019), o ex ac he
sus ainable e ec as:
Dyi; ¼piþDLiquidi yi; þei; ð1Þ
wi h pi being he ixed e ec ha cap u es he ime-in-
a ian unobse ed ‘‘sus ainable’’ e ec o each issue .
Hence, pi is he esidual di e ence be ween a SLB and he
syn he ic con en ional bond ha is a ibu able o he
sus ainable label o he bond, a e con olling o liquidi y
di e ences. To measu e liquidi y sp eads DLiquidi y
i,
bid–
ask sp eads a e used, one o he mos widesp ead p oxies
o bond liquidi y (Fong e al. 2017).
Finally, a c oss-sec ional eg ession is conduc ed o
in es iga e he de e minan s o he a ying sus ainabili y
p emium ac oss issue s, he sus ainabili y ixed e ec p
i
o
each o he 45 issue s is eg essed on a se o con ol
a iables.
pi¼aoþCon olsiþei:ð2Þ
Con ol a iables include quan i a i e a iables, i.e., he
ma u i y, he issued amoun and quali a i e a iables: a -
ing and he cu ency. Ra ing is a scaled a iable which is
assigned in ascending o de , while cu ency is a dummy
a iable equal o 1 i he SLB is issued in USD. Mo eo e ,
SLB-speci ic con ols in e ms o he coupon s ep-up and
he commi men o science-based a ge s a e added.
Empi ical esul s
The nex sec ion p esen s he esul s o he yield cu e
i ing o he p ima y ma ke in 5.1 and he esul s o he
ixed-e ec panel eg ession a e applying a ma ching
p ocedu e.
P ima y ma ke esul s
Appendix Fig. 7plo s he es ima ed yield cu es o each
o he 45 SLB issue s. The blue do s ep esen he con-
en ional bonds yield o ma u i y—gi en i s emaining
li e ime—while he g een do s show he yield a issuance
o he SLB.
O e all, he yield cu es demons a e he ypical s uc-
u e, posi i ely slopped wi h highe long- e m deb yields.
Besides, he SLB yield and he in e ed con en ional bond
yield di e om each o he . While some bonds (Snam,
A2A, Ca e ou , No a is) exhibi a new issue concession
implying ha he SLBs ade abo e he yield cu e ( o he
bene i o in es o s), mos o he SLBs ade a sligh ly
(Dee e, London Quad an ) o clea ly lowe yields (i.e.,
Suzano, Ana Holdings, Gene al Mills). O e all, his con-
i ms he i s pa o hypo hesis 1 ha SLBs ade a a
p emium on he p ima y ma ke . Mo eo e , his inding is
consis en wi h Bake e al. (2018), Fa ica e al. (2021)o
Kap aun e al. (2021) who ind ha g een municipal and
g een co po a e bonds ade a a ‘‘g eenium’’ a issuance
implying lowe yields compa ed o con en ional bonds.
The di e ence in he mean o yields be ween is -4.68
basis poin s, which implies ha issue s on a e age can
expec lowe cos o capi al i hey issue SLBs, p o ided
ha he penal y does no ma e ialize p ospec i ely. This is
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 417
consis en wi h Be ada e al. (2022), concluding ha 20%
o he SLBs a e o e p iced ( o he bene i o he issue and
i s sha eholde s) and wi h Ko
¨lbel and Lambillon (2022),
al hough he yield di e ence is much lowe han hei s.
The p alue o a pai ed wo-sided es and a nonpa a-
me ic Wilcoxon es , p o ided in Table 2, con i m ha he
yields a e signi ican ly di e en om each o he .
Finally, he dis ibu ion o yield di e ences is plo ed in
Fig. 2 ep esen ing he ke nel densi y o he 45 bond pai s
in basis poin s. A la ge mess is cen e ed sligh ly le om
ze o yield di e en ial, skewed o he le wi h hin ails
indica ing ha SLBs ade a lowe yields and some exhibi
pa icula ly la ge di e en ials. The blue dashed line plo s
he median o he dis ibu ion which is also nega i e (-2,56
basis poin s).
Seconda y ma ke esul s
The p ima y ma ke analysis examines yield di e en ials
solely on he issuance day which can be in luenced by a
gap be ween deb supply and demand ( o ins ance, a e
hikes could occu on he issuance day).
Table 3 epo s summa y s a is ics. The a e age SLB
yield in he sample is 2.10 % (210 bps) and below he
syn he ic bond yield which is 2.12%. Mo eo e , he syn-
he ic bond is mo e ola ile han he SLB and he bid–ask
sp ead be ween he bond pai s is cen e ed a ound ze o wi h
a low s anda d de ia ion.
Be o e es ima ing he yield di e en ials, a bunch o
s a is ical es s is applied. Appendix Table 7lis s he
esul s. A Hausman es indica es ha he ixed-e ec s
eg esso is mo e e icien han he andom e ec s es i-
ma o . Since he B eusch–Pagan es implies ha he e o
a iances a e he e oscedas ic equa ion (1) is es ima ed wi h
obus s anda d e o s. Mo eo e , Beck–Ka z obus es i-
ma ion is used o add ess he p esence o se ial co ela ion
Table 2 Pai ed es and Wilcoxon es .
S a is ic Value
Yield SLB (bps) 141.96
Yield CB (bps) 146.64
Yield di e en ial (bps) -4.68
p alue es 0.01
p alue Wilcoxon es 0.05
Do SLBs ade a lowe yields on he p ima y ma ke ? Table 2 epo s
he mean di e ences, pa ame ic and nonpa ame ic es s be ween
yields a issuance o SLBs ‘‘Yield G een’’ and ma ched con en ional
bonds ‘‘Yield B own’’ as desc ibed in sec ion ‘‘Me hodology o
p ima y ma ke analysis’’.
0.00
0.01
0.02
0.03
0.04
-50 -25 025 50
YieldDi
densi y
Yield Di e en ial Densi y Cu e
Fig. 2 Densi y cu e o yield
di e en ials in basis poin s. This
igu e plo s he densi y o he
SLB- and con en ional bond
yield di e en ial ac oss 45
SLBs. The SLB yield is he
yield a issuance, while he
con en ional bond’s yield is he
es ima ed NSS yield a he same
ma u i y by i ing a yield cu e.
The x-axis plo s he yield
di e en ial, whe eby a nega i e
alue is an indica ion ha a
p emium exis s. The blue
dashed line is he median o he
yield di e ence.
418 J. Poggensee
-0.1%
0.2%
0.5%
0.8%
0612
YIELD
MATURITY
A2A
0.0%
0.6%
036
YIELD
MATURITY
AeonMall27
0.0%
0.6%
1.2%
0369
YIELD
MATURITY
Ana Holdings
0.0%
0.3%
0.6%
036
YIELD
MATURITY
Aeon Mall 26
0.0%
1.5%
3.0%
01020
YIELD
MATURITY
Analog De ice
0.0%
1.5%
3.0%
048
YIELD
MATURITY
Ca e ou 26+29
0.0%
1.0%
2.0%
YIELD
MATURITY
CPI
2.0%
3.0%
4.0%
0510
YIELD
MATURITY
Dee e
-0.4%
0.0%
0.4%
0.8%
0510
YIELD
DURATION
Eni
1.0%
2.0%
3.0%
49
YIELD
MATRUTIY
Gene al Mills
0.0%
0.5%
1.0%
036
YIELD
MATURTIY
Helsingbo g 0.75
0.0%
0.7%
1.4%
02.55
YIELD
MATURITY
Helsingbo g 0,875%
-0.3%
0.1%
0.5%
0.9%
-3 2 7 12
YIELD
MATURITY
Holcim
0.0%
0.5%
1.0%
04812
YIELD
MATURITY…
HolcimHel e ia
-0.2%
0.1%
0.4%
0 5 10 15
YIELD
MATURITY
Leg and
1.5%
2.5%
41016
YIELD
MATURITY
London Quad an
-0.2%
0.4%
01020
YIELD
MATURITY
No a is
0.015
0.04
0612
YIELD
MATURITY
NWD -4bp
Fig. 7 Yield Cu es es ima ed by Nelson–Siegel–S ensson me hod.
Plo s yield cu es o 45 issue s on hei espec i e p icing da e. The
blue do s a e he obse ed yields o he issue ’s con en ional bonds
ha ade on he seconda y ma ke . A yield cu e is i ed wi h he
Nelson- Siegel S ensson me hod. Finally, he SLB is o e laid. A SLB
below i s cu e implies a sus ainable p emium.
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 425

0.0%
0.3%
0.5%
048
YIELD
MATURITY
Obayashi
0.0%
1.5%
3.0%
0510
YIELD
MATURITY
Op us
0.0%
1.0%
2.0%
0510
YIELD
MATURITY
Pe no Rica d
-1.0%
1.0%
0612
YIELD
MATURITY
Repsol
1.5%
2.5%
0612
YIELD
DURATION
Sempco b
0.00%
5.00%
46810
YIELD
DURATION
S&P
Y…
3.0%
4.0%
0510
YIELD
MATURITY
Sembco p22
0.0%
0.5%
1.0%
1.5%
0 5 10 15
YIELD
MATURITY
Snam
0.0%
0.2%
0.4%
310
YIELD
MATURITY
TDK
0.2%
0.4%
0.6%
0.8%
0 5 10 15
YIELD
MATURITY
Toda
1.5%
3.0%
0102030
YIELD
MATURITY
Telus
0.5%
1.5%
048
YIELD
DURATION
Sano i
0.2%
0.3%
0.4%
0.5%
0.6%
0612
YIELD
MATURITY
Hulic
0.0%
0.3%
0.6%
0612
YIELD
MATURITY
Fuyo
Fig. 7 con inued
426 J. Poggensee
Appendix: Tables
See Tables 7,8,9,10.
0.5%
1.5%
2.5%
0612
YIELD
MATURITY
Woolwo h
2.5%
4.0%
5.5%
0102030
YIELD
MATURITY
Suzano 2020
0.0%
0.6%
0246
YIELD
MATURITY
Axpo
2.0%
3.0%
4.0%
051015
YIELD
MATURITY
Suzano 2032
0.2%
0.4%
0.6%
0510
YIELD
MATURITY
Enel 2019 Eu o
0.5%
1.5%
2.5%
0612
YIELD
MATURITY
Enel2021 USD
1.0000%
0 5 10 15
YIELD
DURATION
Suzano 2028
YTM
Fig. 7 con inued
Es ima ed Yield Di e en ial
4
3
2
1
0
-1
-2
-3
22 22 Ma -22 22 May-22 22 22 22 Sep-22
YIELD
Fig. 8 Es ima ed Seconda y Ma ke P emium o e ime. Re e s o a
ime end in he sus ainabili y p emium. Ins ead o es ima ing he
baseline-equa ion Dy
i,
=p
i
?DLiquidi y
i,
?e
i,
o e all ime pe iods,
he sample is di ided in o mon hly subsamples o measu e he
sus ainabili y p emium o each mon h. The es ima ed a e age ixed
e ec p
i
(i.e., he sus ainabili y p emium) is plo ed o each mon h
om Janua y 2022 o Sep embe 2022
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 427
Table 7 S a is ical es s
Tes S a is ic p
alue
In e p e a ion Conclusion
F es o indi idual e ec s 133.86 0 Rejec he null hypo hesis ha a model wi h no
independen a iables i s be e
P esence o indi idual
e ec s
Hausman es 8.1111 0.0044 Rejec he null hypo hesis o no co ela ion be ween
e o s and eg esso s
Use o ixed e ec s
B eusch–Pagan es o
he e oscedas ici y
70.191 0 Rejec he null hypo hesis ha he e o a iances a e
all equal
P esence o
he e oscedas ici y
B eusch–God ey/Woold idge es o
se ial co ela ion
8379 0 Rejec he null hypo hesis o no au oco ela ion
be ween he esiduals
P esence o se ial
co ela ion
Pesa an es o c oss-sec ional
dependence
14.088 0 Rejec he null hypo hesis o no c oss-sec ional
dependence
P esence o c oss-sec ional
dependence
Table 8 Es ima ed p emia on he p ima y and on he seconda y
ma ke
Issue P ima y ma ke
es ima ion
Seconda y ma ke
es ima ion
A2A 4.34 8.61
AeonMall26 -23.76 -12.74
AeonMall27 -2.56 -3.02
AnaHoldings -40.53 -5.94
AnalogDe ice -2.31 5.55
Axpo25 -8.66 -5.17
Axpo27 4.19 1.93
Ca e ou 26 11.43 7.03
Ca e ou 29 8.09 9.77
CPI -5.79 3.75
Dee e -4.93 1.8
Enel24 -4.21 0.8
Enel27 5.41 5.1
Enel26USD -7.56 -9.19
Enel28USD 0.52 -6.19
Eni 5.26 3.07
Fuyo -2.29 3.23
Gene alMills -10.73 -11.43
Helsingbo g0.75 -20.07 -7.98
Helsingbo g
0.875
-22.87 -2.13
Holcim -2.62 -1.7
HolcimHel e ia -4.23 -1.4
Hulic -11.52 3.25
Leg and 7.63 -9.55
LondonQuad an 1.07 -0.9
No a is 5.98 6.69
NWD -4.53 7.03
Obayashi -11.25 -6.84
Table 8 con inued
Issue P ima y ma ke
es ima ion
Seconda y ma ke
es ima ion
Op us 5.02 6.90
Pe nod 2.49 -3.18
Repsol 1.26 15.84
S?P 9.17 -14.11
Sano i 0.46 -1.89
Sembco p -7.84 0.24
Sembco p22 -1.95 -5.32
Snam29 10.04 -16.57
Snam34 7.91 -9.76
Suzano20 -27.67 -18.09
Suzano28 -16 -33.60
Suzano32 -39.41 -7.87
TDK26 -4.62 -4.75
TDK28 -1.08 -1.03
Telus -3.86 -9.09
Toda 7.67 -12.53
Woolwo h -14.72 -2.81
This able shows he es ima ed p emia on he p ima y and on he
seconda y ma ke . The p ima y ma ke es ima ion is he di e ence
be ween he yield o he SLB a issuance and he yield cu e es i-
ma ed wi h he NSS me hod. A nega i e alue implies a
sus ainabili y p emium.
Table 9 Fixed-e ec s eg ession wi h ola ili y con ol
S a is ic N Mean S . De . MinMedian Max
p
i
45 -0.024 0.085 -0.327-0.018 0.159
This able shows he dis ibu ion o he es ima ed SLB p emium. The
p emium is de ined as he ixed e ec p
i
o he eg ession Dy
i,
=p
i
?
DLiquidi y
i,
?DVola ili y
i,
?e
i,
. Vola ili y is he 10-day olling
annualized ola ili y o he SLB and he syn he ic bond calcula ed ex
pos .
428 J. Poggensee
Table 8: Compa ison o Es ima ed Sus ainabili y
P emia on he P ima y and on he Seconda y Bond
Ma ke
The p ima y ma ke es ima ion is conduc ed by i ing yield
cu es o each issue . A e wa d, he yield o he SLB is
compa ed wi h he in e ed yield wi h he same esidual
ma u i y. The seconda y ma ke p emia a e es ima ed ia a
ixed-e ec panel eg ession. A nega i e alue is e idence
o a sus ainable p emium; hence, SLBs would ade a
lowe yields compa ed o hei ma ched con en ional
bonds.
Appendix: The Nelson–Siegel–S ensson Me hod
A pa ame ic model, he Nelson–Siegel–S ensson (NSS
mode), is applied o in e heo e ical p ices (and yields)
o e he ma u i y spec um isualized by a yield cu e o
each o he 45 SLB-bond issue s.
Nelson and Siegel es ablished a pa simonious nonlinea
op imiza ion model whe e pa ame e s e lec he ypical
shapes o yield cu e (Nelson and Siegel 1987).
By in eg a ing he solu ion o a second o de -di e en ial
equa ion, hey speci y he spo a e (ze o-coupon bond
yield) as a unc ion o ma u i y and on he below de ined
be a pa ame e s o he ollowing o m:
zb;mðÞ¼b0þb1
1exp m
k1

m
k1
0
@1
A
þb2
1exp m
k1

m
k1
exp m
k1

0
@1
A
þb3
1exp m
k2

m
k2
exp m
k2

0
@1
Að5Þ
The NSS-model, which is ex ensi ely used by cen al
banks,
13
desc ibes he e m s uc u e o in e es a es (spo
a es) by six pa ame e s ha ep esen long- un le el o
in e es , he slope and he cu a u e.
b0 ep esen s he long- e m in e es a e and is always
g ea e ze o, b1is he sp ead be ween he long- and he
sho - e m in e es a e and gene a es a mono onically
inc easing (i b1is nega i e) o dec easing unc ion (i b1is
posi i e), while he b2pa ame e gene a es a hump-shape
(i posi i e) o U-shaped (i nega i e) unc ion. k1speci ies
he posi ion o he hump o U-shape (decay ac o ).
The las e m b3wi h an addi ional k-pa ame e adds an
addi ional u ning poin and is an ex ension o he NS-
model made by S ensson o be e cap u e mo e complex
e m s uc u es in he sho and in he long- e m (S ensson
1994).
F om he spo a e, S ensson de i es he discoun
unc ion as o he o m:
dm;bðÞ¼exp z ;m;bðÞ
100 m
 ð6Þ
This equa ion discoun s each paymen low o i s p esen
alue wi h a ying ma u i y- ela ed spo ) a es.
Finally, he discoun unc ion is used o de i e heo e -
ical p ices and yields o e he ma u i y spec um and hose
a e compa ed wi h he ac ual obse ed ones ha ade in
he ma ke by sol ing an op imiza ion p oblem o he o m
Table 10 Es ima ion o he sus ainable p emium
Dependen a iable Model1 Yield Model2
Model (1) (2)
Va iables
SLB
-0.0284
***
-0.0237
***
(0.0055) (0.0050)
Ma u i y 0.1057
***
0.0388
***
(0.0053) (0.0040)
BidAskSp ead 0.4298
***
0.4042
***
VIX (0.0202) (0.0326)
0.0332
***
(0.0006)
Sho T easu y 0.2037
***
(0.0055)
LongT easu y 1.145
***
(0.0067)
Fixed-e ec s
Issue Yes Yes
Ra ing Yes Yes
Cu ency Yes Yes
Da e Yes
Sec o
R2 Yes
0.94 0.96
Obse a ions 24,405 24,101
Table 10 e e s o equa ion 4and es s whe he SLBs exhibi a p e-
mium. Each SLB’s and i s con en ional bond’s ‘‘ win’’ yield a e
eg essed on a se o dummy a iables including issue , a ing, cu -
ency, sec o and ime ixed e ec s. Con ols o liquidi y (bid–ask
sp ead) and ma u i y a e added and a ‘‘g een’’ dummy speci ies he
yield di e en ial be ween he bond pai s. The second speci ica ion
adds mac o a iables, i.e., he VIX, he sho - e m 3-mon h US- ea-
su y a e and he long 10-yea US- easu y a e. He e oscedas ici y–
obus s anda d e o s a e in pa en heses. *** implies signi icance a
he 1% le el, ** a he 5% le el and * a he 10% le el
13
The ECB es ima es daily yield cu es o he eu o a ea and de i es
o wa d and pa yield cu es based on he NSS me hod
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 429
PN
i¼1ðyi; -~
yi; ðb ))
2
ha is de e mined o minimize he sum
o squa ed yield de ia ions. The inal b es ima es a e used
o plo he yield cu es o each issue .
Since de ia ions be ween ac ual and NSS es ima ed
yields ha e been minimized, he cu e will app oxima ely
i he obse ed yields.
Funding Open Access unding enabled and o ganized by P ojek
DEAL.
Open Access This a icle is licensed unde a C ea i e Commons
A ibu 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
Ande sen, P.M. Yield cu e modelling and a concep ual amewo k
o es ima ing yield cu es: E idence om he Eu opean Cen al
Bank’s yield cu es, ECB-Wo king Pape No. 27 (2018).
Bachele , M.J., L. Beche i, and S. Man edonia. 2019. The g een
bonds p emium puzzle: The ole o issue cha ac e is ics and
hi d-pa y e i ica ion. Sus ainabili y 11 (4): 1098–1120.
Bake , M., D. Be gs asse , G. Se a eim, and J. Wu gle . 2018.
Financing he Response o Clima e Change: The P icing and
Owne ship o U.S. G een Bonds, NBER Wo king Pape Se ies.
Baue , R., and D. Hann. 2014. Co po a e En i onmen al Managemen
and C edi Risk, Wo king Pape .
Be ada, T., L. Engelha d , R. Gibson, and P. K uege . 2022. The
Economics o Sus ainabili y-Linked Bonds, Swiss Ins i u e
Resea ch Pape .
Bundesbank. 2019. The sus ainable inance ma ke : a s ock ake,
Mon hly Repo .
CBI. 2017. G een Bond Highligh s 2017.
Doumbia, D., and I. Haq. 2022. S uc u al Loopholes in Sus ainabil-
i y-Linked Bonds, Wo ld Bank Policy Resea ch Pape 10200.
Eccles, R.G., I. Ioannou, and G. Se a eim. 2014. The impac o
co po a e sus ainabili y on o ganiza ional p ocess and pe o -
mance. Managemen Science 60 (11): 2835–2857.
Ehle s, T., and F. Packe . 2017. G een bond inance and ce i ica ion.
BIS Qua e ly Re iew.
Enel. 2019. Enel Launches he wo ld’s i s ‘‘gene al pu pose SDG
linked bond’’, success ully placing a 1.5 billion U.S. dolla bond
on he U.S. ma ke .
E landsson, U. and S. Mielnik. 2022. An op ion p icing app oach o
sus ainabili y-linked bonds, An h opocene Fixed Income Ins i-
u e Discussion Pape .
Fa ica, S., R. Panzica, and M. Rancan. 2021. The p icing o g een
bonds: A e inancial ins i u ions special?. Jou nal o Financial
S abili y 54: 2688–2701.
Flamme , C. 2021. Co po a e g een bonds. Jou nal o Financial
Economics 142 (2): 499–516.
Fong, K.Y.L., C.W. Holden, and C.A. T zcinka. 2017. Wha a e he
bes liquidi y p oxies o global esea ch?. Re iew o Finance 21
(4): 1355–1401.
Gian a e, G., and M. Pe i. 2019. The g een ad an age: Explo ing he
con enience o issuing g een bonds. Jou nal o Cleane
P oduc ion 219: 127–135.
Gi aldez, J., and S. Fon ana. 2022. Sus ainabili y-linked bonds: The
nex on ie in so e eign inancing. Capi al Ma ke s Law
Jou nal 17 (1): 8–19.
Hachenbe g, B., and D. Schie eck. 2018. A e g een bonds p iced
di e en ly om con en ional bonds?. Jou nal o Asse Man-
agemen 19 (6): 371–383.
Haq, I.U., and D. Doumbia. 2022. S uc u al Loopholes in Sus ain-
abili y-Linked Bonds, Wo ld Bank Policy Resea ch Pape .
Ha zma k, S.M., and A.B. Sussmann. 2019. Do in es o s alue
sus ainabili y? A na u al expe imen examining anking and und
lows. The Jou nal o Finance 74 (6): 2789–2837.
ICMA. 2020. Sus ainabili y-Linked Bond P inciples. In e na ional
Capi al Ma ke s Associa ion.
Kap aun, J., C. La ino, C. Scheins, and C. Schlag. 2021. (In)-C edibly
G een: Which Bonds T ade a a G een Bond P emium?
P oceedings o Pa is Decembe 2019 Finance Mee ing.
Ka p , A., and A. Mandel. 2018. The changing alue o he g een
label on he US municipal bond ma ke . Na u e Clima e Change
8 (2): 161–168.
Ko
¨lbel, J.F. and A.P. Lambillon. 2022. Who pays o sus ainabili y?
An analysis o Sus ainabili y-Linked Bonds. Swiss Finance
Ins i u e Resea ch Pape .
K uege , P. 2015. Co po a e Goodness and Sha eholde Weal h. The
Jou nal o Finance and Economics 115 (2): 304–329.
La cke , D.F., and E. Wa s. 2020. Whe e’s he G eenium?. Jou nal
o Accoun ing and Economics 69: 1–26.
Libe adzki, M., P. Jawo ski, and K. Libe adzki. 2021. Sp ead analysis
o he sus ainabili y-linked bonds ied o an issue ’s g eenhouse
gases emissions educ ion a ge . Ene gies 14: 1–12.
Libe o e, S.M. 2021. Sus ainabili y-Linked Bonds do no i ou
impac amewo k, Nu een.
Maino, A. 2022. Sus ainabili y-Linked Bonds and hei ole in ene gy
ansi ion, Ox o d Ins i u e o Ene gy S udies.
Nelson, C.R., and A.F. Siegel. 1987. Pa simonious modeling o yield
cu es. The Jou nal o Business 60 (4): 473–489.
Pas o , L., R.F. S ambaugh, and L.A. Taylo . 2021. Sus ainable
in es ing in equilib ium. Jou nal o Financial Economics 142
(2): 550–571.
Pede sen, L.H., S. Fi zgibbons, and L. Pomo ski. 2021. Responsible
in es ing: The ESG-e icien on ie . Jou nal o Financial
Economics 142 (2): 572–597.
Pie sch, A., and D. Salakho a. 2022. P icing o g een bonds: D i e s
and dynamics o he g eenium, ECB Wo king Pape Se ies No.
2728.
Reznick, M., R. Usson, and S. Dema e. 2022. Do Sus ainabili y-
linked bonds ha e a s ep-up p oblem? He mes In e na ional.
S ensson, L.E.O. 1994. Es ima ing and In e p e ing Fo wa d In e es
Ra es: Sweden 1992-1994, NBER Wo king Pape , Wo king
Pape No. 4871.
Tuhkanen, H., and G. Vul u ius. 2020. A e g een bonds unding he
ansi ion? In es iga ing he link be ween companies’ clima e
a ge s and g een deb inancing. Jou nal o Sus ainable Finance
& In es men 12 (4): 1194–1216.
Vul u ius, G., A. Mal ais, and K. Fo sbacka. 2022. Sus ainabili y-
Linked Bonds—Thei po en ial o p omo e issue s’ ansi ion o
ne -ze o emissions and u u e esea ch di ec ions. Jou nal o
Sus ainable Finance and In es men 14: 116–127.
430 J. Poggensee

Ze bib, O.D. 2019. The e ec o p o-en i onmen al p e e ences on
bond p ices: E idence om g een bonds. Jou nal o Banking and
Finance 98: 39–60.
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.
Jannis Poggensee is a Ph.D. candida e a he Quan i a i e Business
and Economics Resea ch (QBER) g oup a he Uni e si y o Kiel.
The QBER Ins i u e is pa o he Facul y o Economics a Kiel
Uni e si y and deals wi h cu en empi ical esea ch in economics,
business and econome ics. My esea ch is p ima ily cen e ed on
sus ainable inance, wi h a pa icula ocus on he signaling e ec s o
ESG- ela ed deb issuance in capi al ma ke s and p icing di e en ials
be ween ESG- ela ed on con en ional asse s. My wo k also in es i-
ga es he p icing mechanisms o clima e and ca bon isks in bo h
s ock and bond ma ke s. Th ough my esea ch, I aim o con ibu e o a
deepe unde s anding o how sus ainabili y conside a ions a e in e-
g a ed in o inancial decision-making. I decla e ha he e is no
con lic o in e es associa ed wi h his submi ed pape .
The p icing o sus ainabili y-linked bonds on he p ima y and seconda y bond ma ke s 431