An e, Lenna ; Meye , And é
A icle — Published Ve sion
C oss-lis ings o blockchain-based okens issued
h ough ini ial coin o e ings: Do liquidi y and speci ic
c yp ocu ency exchanges ma e ?
Decisions in Economics and Finance
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: An e, Lenna ; Meye , And é (2021) : C oss-lis ings o blockchain-based okens
issued h ough ini ial coin o e ings: Do liquidi y and speci ic c yp ocu ency exchanges ma e ?,
Decisions in Economics and Finance, ISSN 1129-6569, Sp inge In e na ional Publishing, Cham, Vol.
44, Iss. 2, pp. 957-980,
h ps://doi.o g/10.1007/s10203-021-00323-0
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/286840
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
Decisions in Economics and Finance (2021) 44:957–980
h ps://doi.o g/10.1007/s10203-021-00323-0
1 3
C oss‑lis ings o blockchain‑based okens issued
h oughini ial coin o e ings: Do liquidi y andspeci ic
c yp ocu ency exchanges ma e ?
Lenna An e1,2 · And éMeye 1
Recei ed: 18 Decembe 2019 / Accep ed: 16 Feb ua y 2021 / Published online: 17 Ma ch 2021
© The Au ho (s) 2021
Abs ac
Ini ial coin o e ings (ICOs) ep esen a no el unding mechanism whe e digi al
okens a e issued on he blockchain and sold o in es o s. One majo eason o he
success o his inancing model is he ac ha he issued okens can immedia ely
be aded on seconda y ma ke s. This e en s udy analyzes 250 exchange c oss-lis -
ings o 135 di e en okens issued h ough ICOs on 22 c yp ocu ency exchanges.
We ind signi ican abno mal e u ns o 6.51% on he lis ing day and 9.97% o e a
se en-day window a ound he e en . Fu he analysis shows ha he esul s clea ly
di e o indi idual c yp ocu ency exchanges, as lis ings on indi idual exchanges
yield e u ns o up o 34% on he e en day, while o he s a e negligible. An in es i-
ga ion o liquidi y- ela ed me ics shows ha lowe p io ading olume and asse
ma ke capi aliza ion ha e posi i e e ec on lis ing e u ns. In es o s use phases o
high ma ke liquidi y o sell o posi ions a ound he pe iod o c oss-lis ing e en s.
The esul s on he c oss-lis ing e ec s o ICOs may be o ele ance o in es o s/ ad-
e s, ICO p ojec s, c yp ocu ency exchanges and egula o s.
Keywo ds C yp ocu ency· C yp ocu ency exchanges· Blockchain· E he eum·
E en s udy· Token sales
JEL Classi ica ion G10· G12· G14
* Lenna An e
an e@blockchain esea chlab.o g
1 Blockchain Resea ch Lab, Max-B aue -Allee 466, 22765Hambu g, Ge many
2 Facul y o Business Economics & Social Sciences, Uni e si y o Hambu g, Von-Melle-Pa k
5, 20146Hambu g, Ge many
1 In oduc ion
Ini ial coin o e ings (ICOs) p o ide en u es wi h an al e na i e ool o co -
po a e inancing. Many p ojec s ha e al eady inanced hemsel es by selling
958
L.An e, A.Meye
1 3
blockchain-based digi al okens ha ep esen some o m o ( u u e) alue, and
many ICO in es o s ha e al eady made a o une o o al loss. One o he main
easons o he success o ICOs is he ac ha okens can be aded on seconda y
ma ke s as soon as a c yp ocu ency exchange ag ees o lis hem. A special ea u e
o c yp ocu ency ma ke s is ha he e is i ually no o de ou ing h ough b o-
ke s; ins ead, he asse s a e aded on a la ge numbe o di e en exchanges, each
o which has i s own speci ic cha ac e is ics and use base. I is no uncommon o
ICO okens o be successi ely c oss-lis ed on mul iple exchanges, which means ha
c oss-lis ing e en s occu mo e equen ly o okens han o s ocks. One speaks o
a c oss-lis ing when an asse , which is al eady adable on a leas one seconda y
ma ke , is lis ed on ano he seconda y ma ke —in he case o ICOs okens c yp o-
cu ency exchanges. Acco dingly, c oss-lis ings ne e ep esen he i s oppo uni y
bu a u he op ion o ade an asse .
Fo s ock ma ke s, ma ke segmen a ion, ma ke liquidi y, in o ma ion disclosu e
and in es o p o ec ion ha e been iden i ied as easons o c oss-lis ings (Roosen-
boom and Dijk 2009). Ye s ock ma ke s a e di e en om c yp ocu ency ma ke s.
Fo example, c yp ocu encies a e aded 24/7, and hey a e bu decen alized com-
pu e p o ocol ha in heo y any exchange can lis wi hou he consen o he p o-
jec i sel . Some c yp ocu ency exchanges a e ully cen alized and esemble s ock
exchanges, while o he , decen alized exchanges solely ope a e as compu e code on
a blockchain in as uc u e, allowing he use s o ade pee - o-pee wi h each o he
(Wa en and Bandeali 2017). The le el o egula ion and anspa ency also di e s
ac oss exchanges. Some o hem manipula e o in la e hei own ma ke s in o de o
signal liquidi y o po en ial use s and p ojec s and o he eby ea n mo e ading and
lis ing ees (Alameda Resea ch 2019; Bi wise Asse Managemen 2019).
The e ec s o c oss-lis ings a e a a he new esea ch ield o he c yp ocu ency
sec o . To ou knowledge, wo s udies ha e speci ically analyzed such e ec s, bu
so a no s udy has ocused solely on ICOs. Benede i (2019) inds posi i e lis ing
e u ns o a sample o 3,625 okens on 108 exchanges and iden i ies an inc ease in
ading olume and cha ac e is ics like on-blockchain ans e olumes o E he eum-
based okens. An e (2019) analyzes 327 c oss-lis ings on 22 exchanges and inds
abno mal e u ns o 5.7% on he lis ing day and 9.2% o e a 7-day e en window
a ound he lis ing. The s udy also inds signi ican di e ences be ween indi idual
exchanges, as well as e idence o in o med ading. Fo ini ial lis ing e ec s o ICOs,
Benede i and Kos o e sky (2018) ind signi ican unde p icing in a sample o ICOs,
wi h abno mal buy-and-hold e u ns o 48% wi hin he i s 30 ading days, while
Mom az (2019a) inds a e age unde p icing o 8.2% in ICO campaigns o e $1 mil-
lion. Fu he mo e, Mom az (2019b) iden i ies an a e age abno mal e u n o 14.8%
on digi al cu encies wi hin he i s ading day a e hei ini ial exchange lis ing.
The e u ns a e also d i en by key igu es like liquidi y and ma ke capi aliza ion.
This is he i s c oss-lis ing s udy o ocus on ICO okens. The goal o his a icle
is o p o ide a undamen al scien i ic basis o c oss-lis ing e ec s o ICO okens. I
shall be iden i ied i c oss-lis ings ep esen posi i e ma ke (quali y) signals, simila
o adi ional ma ke s, which a e exp essed by abno mal e u ns. We aim o explo e
how ICO okens pe o m in exchange c oss-lis ings and whe he speci ic c yp ocu -
ency exchanges o liquidi y- ela ed cha ac e is ics—in line wi h signaling heo y
959
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
(Spence 1973)—ha e any e ec s on lis ing e u ns. This can help o be e unde -
s and he e iciency o ICO oken ma ke s and o de i e co esponding implica ions
o heo y (i.e., academic esea ch) and p ac ice (i.e., c yp o ma ke pa icipan s). To
his end, we subjec a da a se o 250 lis ing e en s o 135 di e en c yp ocu encies
issued h ough ICOs o an e en s udy o abno mal e u ns. Po en ial e ec s o indi-
idual exchanges a e es ed h ough sub-samples, and we build eg ession models o
abno mal e u ns o es o any e ec s o liquidi y me ics.
The ICO phenomenon can be con empla ed om a ious angles, such as he pe -
spec i e o p ojec s seeking unding (An e e al. 2018), in es o s looking o inan-
cial e u n (Fisch & Mom az 2020), socie y bene i ing om a new sou ce o inance
o s a -ups (Fisch e al. 2020), o speci ic indus y playe s, such as c yp ocu ency
exchanges (Benede i 2019). Mos o he exis ing s udies assume he pe spec i e o
p ojec s, in es iga ing he de e minan s o success. The p esen s udy mos ly ocuses
on he pe spec i e o in es o s, whom lis ing e u ns may a o d conside able (sho -
e m) gains. Howe e , he analysis may also be o in e es o all o he in ol ed pa -
ies: (1) The p ojec s can be e unde s and wha in es o s and he ma ke look o
and how o espond o in es o expec a ions; (2) egula o s may lea n whe he spe-
ci ic a eas o he ma ke equi e in e en ion, o ins ance i lis ings a e p eceded by
signi ican e u ns, which could indica e ma ke manipula ion; (3) c yp ocu ency
exchanges may be able o iden i y ‘p omising’ p ojec s o lis ; and (4) socie y and
egula o s can imp o e hei unde s anding o he ma ke o ICOs and assess i s (in)
e iciency.
In ollowing sec ion, h ee hypo heses a e de i ed on he basis o li e a u e. We
will p o ide an o e iew o c oss-lis ings, whe e s ock ma ke s ha e ecei ed much
mo e esea ch a en ion han c yp ocu encies and blockchain-based inancing such
as ICOs. In addi ion, special cha ac e is ics o he ma ke o ICO okens, c yp o-
cu ency exchanges and liquidi y me ics, o which explici hypo heses e e , a e
explained. The e iew o he li e a u e also in o ms ou selec ion o a iables. Sec-
ion3 desc ibes he da a se and me hods while Sec .4 is dedica ed o some desc ip-
i e s a is ics. Sec ion5 p esen s he empi ical indings, ela ing i s o he gene al
ma ke e ec s o lis ing e en s and hen o indi idual c yp o exchanges and ma ke
cha ac e is ics. Leading up o a b ie conclusion, Sec .6 o e s a discussion o he
esul s, some limi a ions o he pape , implica ions and a enues o u u e esea ch.
2 Rela ed li e a u e and esea ch ques ions
2.1 Exchange (c oss‑)lis ings
The e ec s o c oss-lis ings ha e been s udied ex ensi ely o s ock ma ke s. By
c oss-lis ing s ocks, some companies a e able o inc ease hei alua ion (Foe s e
and Ka olyi 1999; Mille 1999). The p ice o s ocks and c yp ocu encies ises in
an icipa ion o c oss-lis ings and declines a e he e en s (An e 2019; Dha an and
Ikenbe y 1995). Acco ding o signaling heo y (Spence 1973), p ojec s can show
hei quali y by cos ly signals. This applies o c oss-lis ings, which equi e ime and
esou ces, o en including lis ing ees. Acco dingly, B is e al. (2007) conclude ha
960
L.An e, A.Meye
1 3
c oss-lis ings cons i u e posi i e quali y signals. S ocks end o expe ience nega i e
d i s in he days and mon hs ollowing he lis ing e en s (Alexande and Janaki a-
manan 1988; Dha an and Ikenbe y 1995; Foe s e and Ka olyi 1999; Howe and
Kelm 1987). Howe e , Mille (1999) and Jaya aman e al.(1993) ail o con i m such
a ec s. Fo c yp ocu ency c oss-lis ings, An e (2019) shows signi ican nega i e
CAARs o − 1.5% o e a h ee-day window a e he e en .
The li e a u e has iden i ied a leas ou majo easons o c oss-lis an asse : (1)
ma ke segmen a ion, (2) liquidi y, (3) in o ma ion disclosu e and (4) in es o p o-
ec ion (Roosenboom and Dijk 2009), each o which we will discuss in some mo e
de ail. Fi s , c oss-lis ings can educe ba ie s o in e na ional in es o s and hus
educe ma ke segmen a ion (E unza and Losq 1985). Me on (1987) in oduces
he in es o ecogni ion hypo hesis, which s a es ha he a i al o new in es o s
educes he expec ed e u ns o exis ing in es o s. Mo i a ions o c oss-lis include
exposing he asse o new in es o s, which may educe he cos o capi al (Bake
e al. 2002), and a ac ing new use s o cus ome s o he p oduc (Bancel and Mi -
oo 2001; Mi oo 1992; Pagano e al. 2002). An e (2019) sugges s ha he c yp-
ocu ency exchange ma ke is highly segmen ed, due o he deg ee o egula ion
o o ganiza ional di e ences (e.g., he in eg a ion o ia cu encies e sus so-called
s ablecoins ha digi ally ep esen ia cu ency). Fu he mo e, liquidi y conce ns
a e a majo c oss-lis ing mo i a ion. The in oduc ion o new ading ma ke s by
means o c oss-lis ings imp o es liquidi y. T ading olume ises wi h c oss-lis ings
bo h o s ocks (Foe s e and Ka olyi 1999) and c yp ocu encies (Benede i 2019).
Exposing an asse o mul iple ma ke s a ac s liquidi y ade s who compe e o
o de low (Chowdh y and Nanda 1991). Thi d, in o ma ion asymme y e e s o
he si ua ion whe e some pa ies ha e p i a e in o ma ion which hey may exploi a
he expense o o he s (Mille and Rock 1985). By disclosing addi ional in o ma ion,
such asymme y can be educed. Can ale (1996) sugges s ha he lis ing o asse s in
ma ke s ha equi e mo e comp ehensi e disclosu e can se e as a signal o qual-
i y. Las ly, acco ding o bonding heo y (Co ee 2002), i m alue inc eases a e
lis ing in a coun y wi h s onge in es o p o ec ion. Resul s om s ock and c yp-
ocu ency ma ke s show ha c oss-lis ings ha e a posi i e e ec on e u ns in US
ma ke s (Doidge e al. 2004; An e 2019) and elsewhe e (T oge 2007), which can be
explained by highe o di e en egula ion and hus in es o p o ec ion.
Gi en ha c oss-lis ings lead o posi i e abno mal e u ns o s ocks and c yp-
ocu encies and can he e o e be in e p e ed as a sign o quali y, i is easonable o
assume ha his is also he case o ICO okens.
Hypo heses 1 C oss-lis ings e en s o ICO okens ep esen quali y signals ha
lead o posi i e abno mal e u ns.
2.2 Blockchain‑based inancing
Blockchain echnology, which enables he secu e ansmission o alue on he
In e ne , was i s in oduced as he mechanism unde lying he Bi coin whi epape
(Nakamo o 2008). Tokens a e he digi al and ans e able ep esen a ion o alues

961
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
on blockchains. S a -ups in he blockchain ecosys em ha e been using oken sales
since 2013 o aise capi al by issuing okens ha p o ide a speci ic ( u u e) alue
o in es o s. In ICOs, he okens usually ep esen cu encies o speci ic alues ha
complemen he p ojec s’ p oduc s o se ices (e.g., ouche s o so wa e licenses)
(An e e  al. 2018). The okens a e usually “min ed” by uploading decen alized
compu e code on a blockchain in as uc u e (called sma con ac s, o in he
special case o okens: oken con ac s), like E he eum. Any ype o logic can be
p og ammed in o a oken con ac , including ans e es ic ions o in la ion. Once
uploaded o he blockchain, he code is execu ed in a decen alized manne and use s
can ans e he okens ac oss he ne wo k (An e 2020).
A speci ic o m o ICOs is so-called ini ial exchange o e ings (IEOs), whe e
okens a e di ec ly sold on a c yp o exchange. IEOs gained momen um in 2019, o e -
ing se e al ad an ages o e egula ICOs. Fi s , he p ojec s can a ge he exis ing
and e i ied use base o an exchange, which p omises wide dis ibu ion and lowe
ma ke ing cos s, and second, in es o s know ha he oken will be adeable on he
exchange a e he IEO. Leading exchanges such as Binance ha a e associa ed wi h
high c oss-lis ing e u ns (An e 2019) may se e as a signal o quali y o in es o s.
Howe e , he signaling bene i s o IEOs s ongly depend on he quali y o he selec ed
exchange. The e is clea ly a isk o asymme ic in o ma ion and ad e se selec ion,
no only be ween p ojec s and in es o s bu also be ween exchanges and hei use s.
Finally, secu i y oken o e ings (STOs) a e a special ype o oken sale whe e he
okens ep esen secu i ies like s ocks o bonds. An e and Fiedle (2020) lis a ious
ad an ages o STOs o e adi ional secu i ies o e ings: Tokens can immedia ely
be ans e ed and aded 24/7, clea ing and se lemen p ocesses can be mo e e i-
cien , cus ody and b oke s can become obsole e and he blockchain in as uc u e
o e s a ully cus omizable le el o anspa ency. In es o s may p e e STOs because
he le el o in es o p o ec ion is highe han in ICOs. The au ho s analyze 151
STOs and ound ha cheap ( o ake) signals, like social media ne wo k size o he
s a ed amoun o human capi al, can p omo e unding success and can he e o e be
exploi ed by he p ojec s.
While he e a e, as desc ibed abo e, a a ie y o di e en o ms o blockchain-
based oken unding, his s udy e e s exclusi ely o ICOs and no o he speci ic
subg oup o IEOs o he ela ed g oup o STOs.
2.2.1 ICO des ina ion ma ke s
The selec ed exchange clea ly cons i u es a c ucial cha ac e is ic o an ICO. C yp-
ocu ency exchanges ha e been associa ed wi h ma ke manipula ion (Alameda
Resea ch 2019; Bi wise Asse Managemen 2019) and ha e also been hacked (An e
2018). Thei le el o egula o y o e sigh a ies s ongly, om compa a i ely s ic
egula ion (e.g., US-based exchanges like Coinbase o K aken) o exchanges loca ed
in ax ha ens ha lack o e sigh o amoun o no hing mo e han a blockchain-based
compu e p o ocol wi hou an unde lying legal en i y. A lack o egula ion opens
he doo o se e al ypes o mo al haza d: (1) misbeha io by he exchanges hem-
sel es (e.g., in la ing hei ading olume), (2) exchange employees using o leaking
962
L.An e, A.Meye
1 3
con iden ial in o ma ion, i.e., abou u u e oken lis ings, and (3) ICO p ojec eams
using inside knowledge abou u u e lis ings.
Exchanges also di e wi h espec o hei use o ia cu encies. Only some
exchanges o e deposi s and wi hd awals in ia cu encies, while o he s use so-
called s ablecoins o mimic ia cu encies. S ablecoins a e digi al okens backed
by ia cu encies, i.e., a digi al dolla is issued o e e y dolla in a bank accoun
(An e e al. 2020). On s ablecoin exchanges, use s can only ade and ans e hese
digi al coins bu canno easily deposi o wi hd aw hem om o o hei pe sonal
bank accoun . This ep esen s a ma ke ba ie bu allows he exchanges o o e ia -
like ading wi hou he need o banking pa ne s. The e o e, ICO p ojec s p e e o
be lis ed on exchanges ha o e a ia ga eway. An e (2019) shows ha US-based
c yp ocu ency exchanges ea u e highe e u ns on c oss-lis ings, likely due o hei
s ic e egula ion.
The abo e easons p incipally sugges ha ICO p ojec s will wan o c oss-lis
hei oken as widely as possible. Ye , i s ly, c oss-lis ings usually do no come o
ee, and secondly, he choice o some exchanges, which may be weakly egula ed o
p one o manipula ion, may indeed be pe cei ed as a nega i e quali y signal by ma -
ke pa icipan s. The size o an exchange should also make a di e ence. The e o e,
we expec ha he e u ns o c oss-lis ing e en s will di e ac oss he 22 c yp ocu -
ency exchanges in ou sample.
Hypo heses 2 C oss-lis ings lead o e ec s o a ying s eng h based on he
espec i e c yp ocu ency exchanges hey ake place on.
The goal is o iden i y whe he s a is ically signi ican di e ences a e p esen on
he basis o di e en exchanges. This can be a basis o u u e in es iga ions, which,
e.g., could analyze exchange-speci ic cha ac e is ics, such as egula o y s a us, ju is-
dic ion o secu i y mechanisms mo e p ecisely.
2.2.2 Liquidi y‑ ela ed cha ac e is ics
Liquidi y can be desc ibed as he ease o selling an asse . The e a e a ious ea-
sons o , o cos s o , illiquidi y, such as exogenous ansac ion cos s (e.g., ading
ees), demand p essu e (e.g., a ailabili y o buye s and selle s) and coun e pa y
isks (e.g., p i a e in o ma ion), all o which will ha e an e ec on asse p ices
(Amihud e al. 2005). Se e al s udies s a e ha o inc ease liquidi y is he p ima y
mo i a ion o c oss-lis s ocks (Fan o and Ka mel 1997; Mi oo 1992), and Sil a and
Chá ez (2008) ind ha c oss-lis ed i ms ha e lowe liquidi y cos s han non-c oss-
lis ed i ms. Fo his eason, we also expec o ind e ec s o liquidi y on c oss-lis -
ing e u ns o ICOs. In he ollowing, a ious liquidi y- ela ed cha ac e is ics a e
discussed.
Ma ke capi aliza ion The i s candida e indica o o liquidi y is he ma ke
capi aliza ion o c yp ocu encies. Da a p o ide s like coinma ke cap.com o
coingecko.com ank c yp ocu encies based on hei ma ke capi aliza ion. The e-
o e, g ea e ma ke capi aliza ion a o ds he p ojec s mo e a en ion. Ye ma ke
963
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
capi aliza ion as a me ic is p oblema ic as i can be in la ed by only dis ibu ing a
small sha e o he okens publicly, while ma ke da a p o ide s s ill calcula ed he
me ic based on he ull sha e o okens. Addi ionally, implied ma ke capi aliza-
ion is only ele an i ma ke s a e liquid, which many c yp ocu ency ma ke s
a e no .
T ading olume is ano he po en ial indica o o he liquidi y o c yp ocu en-
cies, bu i may likewise be lawed, due o manipula ion on he pa o he c yp o-
cu ency exchanges (Alameda Resea ch 2019; Bi wise Asse Managemen 2019).
An e (2019) inds ha p e-lis ing ma ke capi aliza ion and ading olume, as
well as he ading olume o he ma ke p oxy Bi coin, a ec e u ns om c oss-
lis ings o c yp ocu encies. A high p e-lis ing ading olume could concei ably
educe lis ing e u ns, as he c oss-lis ing will yield less addi ional liquidi y.
Ra io be ween ma ke capi aliza ion and ading olume Since he ading
olume and ma ke capi aliza ion o ICO okens a e no oo s ongly co ela ed
( = 0.67 in ou sample), he a io o he wo me ics migh be use ul o analy-
sis. Tha a io could p o ide an indica ion o unde alua ion o o e alua ion o
asse s. I an asse wi h high ma ke capi aliza ion has a low ading olume, his
can be in e p e ed as a sign o illiquidi y— he asse is po en ially o e alued. On
he o he hand, i may also simply signal ha he e a e no ele an exchanges
on which he asse can be aded. Fo example, in es o s may no be willing o
pu chase okens a exis ing exchanges due o in es o p o ec ion o secu i y con-
ce ns. Low ma ke capi aliza ion combined wi h a high ading olume may indi-
ca e an unde alued asse , as he cos o liquidi y is low.
Re e ence ma ke (Bi coin ading olume and ma ke capi aliza ion) A high
ading olume o he e e ence ma ke Bi coin can be a signal o gene al ma ke
liquidi y, which could a ec he e u ns on c oss-lis ings in wo opposi e ways.
On he one hand, ac i e Bi coin ading could mean ha ade s ha e li le a en-
ion o spa e o a c oss-lis ing. On he o he hand, high o e all ma ke liquidi y
could mean a highe in low in o a new ading pai . Bi coin’s ma ke capi aliza-
ion may in u n se e as a p oxy o he size and/o sen imen o he o e all ma -
ke . A high alue may signal ha c yp ocu ency in es o s ha e ecen ly achie ed
highe e u ns, i.e., ha ma ke sen imen is good (a bull ma ke ), which may lead
o a highe in low o capi al o new c oss-lis ings. Howe e , ha sen imen migh
only apply o Bi coin, in which case o he , newly lis ed c yp ocu encies will no
bene i . Scien i ic e idence has shown co-jumping e ec s o Bi coin and o he
c yp ocu encies (e.g., Bou i e al. 2020), and Bi coin is connec ed o o he c yp-
ocu encies ia e u n shocks (Ji e al. 2019). I can he e o e be expec ed ha
Bi coin as he la ges and mos ele an c yp ocu ency also has co esponding
in luences on he ICO okens unde conside a ion.
Based on he p io explana ions, we posi ha cha ac e is ics such as ading
olume and ma ke capi aliza ion, as well as he e e ence ma ke Bi coin, ha e
an e ec on he e u ns on ICO c oss-lis ings.
Hypo heses 3 Liquidi y- ela ed cha ac e is ics ep esen ele an ac o s ha
in luence e u ns om c oss-lis ings o ICO okens.
964
L.An e, A.Meye
1 3
Since he e a e no compa able s udies o c oss-lis ing e ec s in c yp ocu ency
ma ke s, his s udy aims o iden i y whe he liquidi y- ela ed cha ac e is ics a e gen-
e ally ele an o he de elopmen o c oss-lis ing e u ns. This may in u n p o-
ide a basis o u u e (mo e speci ic) in es iga ions. The nex sec ion desc ibes he
empi ical se up we use o es hese expec a ions.
3 Da a andme hodology
3.1 Sample
We collec ed da a on lis ing e en s o ICO okens o 22 c yp ocu ency exchanges,
e aining only ICOs ha had a leas 31days o ading his o y p io o he lis ing, so
ha we we e able o calcula e e u ns o he 30-day pe iod leading up o he e en s.
This equi emen implies ha all e en s a e c oss-lis ings, al hough he exac num-
be o exis ing seconda y ma ke s whe e a oken is al eady lis ed is unknown due o
a lack o anspa ency and da a a ailabili y. The lis ing e en s we e cap u ed using
a Teleg am bo (@c yp oe en bo ) ha connec s o exchange APIs and epo s new
ading pai s, as well as h ough he c yp ocu ency ma ke da a p o ide block.cc,
which epo s new oken lis ings o c yp ocu ency exchanges. In line wi h McWil-
liams and Siegel (1997), any e en s o a gi en ICO oken wi h o e lapping es ima-
ion and/o e en pe iods we e d opped. This le 250 e en s e e ing o 135 okens.
Fo each e en , we collec ed daily p ices, ading olumes and ma ke capi aliza ion
bo h o he oken i sel and o Bi coin om coinma ke cap.com o 31days p io o
he exchange lis ing and 10days a e he e en . As c yp ocu ency ma ke s ope a e
24/7, we use daily close p ices, which a e ce ain o cap u e any lis ing e ec s on he
pa icula day and a e u he mo e less likely o be ou lie s han daily highs o lows.
3.2 E en s udy me hodology
We choose a 7-day (− 3, + 3) e en window a ound he lis ing day o cap u e any
e ec s bo h be o e and a e he lis ing. Bo h subpe iods will be o in e es o c yp-
ocu ency ade s, who may y o de elop ading s a egies based on he esul s. In
line wi h An e (2019), who no es ha he high ola ili y o c yp ocu encies means
ha long es ima ion pe iods bea he isk o dis o ing ma ke e u ns, we choose a
a he sho 21-day es ima ion window (− 30, − 10). Ini ially, we use wo al e na i e
models o calcula e expec ed e u ns. Fi s , in he cons an mean e u n model, he
expec ed e u n o an asse is simply equal o i s mean e u n o e he es ima ion
pe iod. Second, he ma ke model ep esen s an ex ension, as a e e ence ma ke is
addi ionally used o accoun o he e ec s o o e all ma ke mo emen s. We use
Bi coin as he e e ence ma ke , as i accoun s o o e hal o he c yp ocu ency
ma ke , i ades on all he exchanges co e ed by he sample, and i s p ice clea ly
a ec s o he c yp ocu encies (e.g., Bu nie 2018; Gkillas e  al. 2018; Hu e  al.
2019). The e o e, he ma ke model is a sa e basis and is used o u he analysis
in his a icle. By also (ini ially) es ing he cons an mean e u n model, we can
971
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
Table 3 OLS eg essions o CAAR on liquidi y me ics
***, **, * indica es signi icance a he 1%, 5% and 10% le el, espec i ely. Cons an e m and yea dummy a iables a e included in each model bu no epo ed. He e -
oskedas ici y- obus s anda d e o s in pa en heses
Window N− 3 o + 3 − 3 o − 2 − 1 0 + 1 + 2 o + 3
Me ic Coe . R2Coe . R2Coe . R2Coe . R2Coe . R2Coe . R2
ATV 250 − 0.040***
(0.012)
0.116 − 0.001
(0.006)
0.025 − 0.005
(0.003)
0.041 − 0.026**
(0.011)
0.074 − 0.004
(0.003)
0.031 − 0.004
(0.004)
0.012
AMC 245 − 0.039***
(0.014)
0.093 0.002
(0.008)
0.026 − 0.009
(0.006)
0.044 − 0.024**
(0.011)
0.064 − 0.004
(0.004)
0.027 − 0.004
(0.004)
0.001
AMC/ATV 245 0.056**
(0.023)
0.092 0.007
(0.009)
0.028 0.003
(0.011)
0.035 0.042**
(0.020)
0.069 0.003
(0.005)
0.023 0.002
(0.008)
0.001
BTV 250 − 0.087*
(0.050)
0.087 − 0.037*
(0.022)
0.036 − 0.002
(0.019)
0.035 − 0.001
(0.039)
0.052 0.004
(0.019)
0.024 − 0.051***
(0.019)
0.047
BMC 250 − 0.101
(0.070)
0.084 − 0.023
(0.029)
0.028 − 0.003
(0.024)
0.035 − 0.002
(0.057)
0.052 0.008
(0.024)
0.025 − 0.081***
(0.023)
0.063

972
L.An e, A.Meye
1 3
ma ke capi aliza ion and he CAAR o hose wo pe iods. Howe e , hese models
ha e lowe explana o y powe (9.3%/6.4%), which sugges s ha ading olume is
he mo e ele an me ic. As explained abo e, he a io o asse ma ke capi aliza ion
o asse ading olume (AMC/ATV) pe mi s conclusions abou po en ial o e alua-
ion o unde alua ion. We ind posi i e and signi ican e ec s bo h o he (− 3, + 3)
window (0.056, p < 0.05) and o he e en day (0.042; p < 0.05).
Besides asse -speci ic me ics, we also es o any in luence o he o e all ma ke ,
which we p oxy by Bi coin. Bi coin ading olume (BTV) has a signi ican nega i e
e ec ac oss he ull e en window. Unlike wi h he p e ious asse me ics, howe e ,
his e ec is no a ibu able o he day o he e en o he day be o e o a e , bu
a he o he (− 3, − 2) pe iod (− 0.037; p < 0.1) and he (+ 2, + 3) pe iod (− 0.051;
p < 0.01). Fo Bi coin ma ke capi aliza ion (BMC), we iden i y a highly signi ican
nega i e e ec o (+ 3, + 2) as he only ele an e ec .
6 Discussion
One p ima y goal o his s udy is o gain a basic unde s anding o whe he ICO
oken c oss-lis ings ep esen —simila o c oss-lis ings o o he asse s, such as socks
– posi i e ma ke signals ha esul in abno mal e u ns. Rega dless o he du a ion
o he e en windows o he choice o he cons an mean e u n model e sus he
ma ke model, c oss-lis ing e en s p oduce excess e u ns o c yp ocu encies issued
h ough ICOs. We hus accep Hypo heses 1. Fo 3-, 5-, and 7-day windows a ound
he lis ing e en s, we iden i y posi i e and highly signi ican CAARs, which inc ease
wi h he leng h o he e en window. The posi i e e ec s o c oss-lis ing e en s ma ch
he ou come o p io li e a u e on c yp ocu ency c oss-lis ings (An e 2019), as well
as he s ock ma ke li e a u e (Foe s e and Ka olyi 1999; Mille 1999). While An e
(2019) iden i ied CAARs o 5.7% o = 0 and 9.2% o (− 3, + 3) o c yp ocu ency
c oss-lis ings, we ind CAARs o 6.51% o he lis ing day and 9.97% o he ull
e en window, which sugges s ha ma ke s o ICOs a e e en less e icien han c yp-
ocu encies in gene al. This con i ms ha c oss-lis ing e en s a e mos ly posi i e
news o ICO okens ha lead o posi i e p ice co ec ions in he sho e m.
The posi i e e u ns sugges ha ICO p ojec s can use c oss-lis ings o signal
hei quali y o he ma ke . Simila esul s apply o c oss-lis ings o s ocks (B is e al.
2007). In line wi h signaling heo y (Spence 1973), a c oss-lis ing is a cos ly sig-
nal, equi ing coo dina ion, lis ing ees o disclosu e o in o ma ion. A c oss-lis ing
may a o d enewed ecogni ion o p ojec s wi h li le liquidi y, which we e pe haps
al eady conside ed a ailu e. Acco dingly, an implica ion o his s udy is ha ICO
p ojec s should cons an ly wo k o lis hei okens a new exchanges. This will lead
o abno mal e u ns, which can sa is y he exis ing in es o base and a ac new
in es o s o oken use s. Howe e , gi en ha we conside abno mal e u ns only un il
h ee days a e he c oss-lis ing, we canno say wha he longe - e m e ec s a e.
The posi i e impac o he c oss-lis ing e en s may be explained by he educ ion
o in es men ba ie s o new in es o s om o he ju isdic ions, bu also he p o-
jec s’ inc eased isibili y (An e 2019; Bake e al. 2002; E unza and Losq 1985).
973
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
In line wi h he in es o ecogni ion hypo hesis (Me on 1987), c yp ocu ency p o-
jec s may seek o a ac new in es o s in o de o educe he expec ed e u ns o
exis ing in es o s. Al hough he expec ed e u ns do no ac ually decline, he addi-
ional demand om new use s allows exis ing in es o s o ealize p o i s. E en he
p ojec s hemsel es, which usually e ain a po ion o hei okens, gain an addi-
ional oppo uni y o sell hose okens and hus o ob ain cheape inance. Jaya aman
e al. (1993) make an analogous a gumen o s ock ma ke c oss-lis ings.
We ind signi ican nega i e CAARs o − 1.4% o he pos -lis ing pe iod (+ 2, + 3).
A close look a he indi idual days a e he lis ing e en s shows ha his e ec is
especially a ibu able o day wo, wi h a signi ican nega i e AAR o − 1.21%. These
esul s a e in line wi h he li e a u e on s ocks (Foe s e and Ka olyi 1999; Howe and
Kelm 1987) and c yp ocu encies (An e 2019) and sugges ha exis ing in es o s use
he inc eased liquidi y o c oss-lis ing e en s o sell all o pa o hei holdings, and
his inc eased supply o e compensa es he demand om new ade s.
While posi i e lis ing announcemen e ec s ha e been iden i ied o s ock ma -
ke s (e.g., Mille 1999; Roosenboom and Dijk 2009), we in u n ind a highly signi i-
can CAAR o 4.97% o e he h ee p e-lis ing days, which sugges s ha he ma ke s
knew abou he upcoming lis ing e en s be o ehand and adjus ed he p ices acco d-
ingly. An e (2019) e en inds a signi ican posi i e AAR o 6% on he day be o e he
announcemen s, poin ing o he possibili y ha exchanges ha p e-announce lis ings
leak in o ma ion. We simila ly iden i y posi i e e u ns o 2.66% o he (− 3, − 2)
window ha indica e ha hese e ec s canno only be explained by announcemen s
(as hey a e only in = − 1 o = 0 in ou sample) bu ha p i a e in o ma ion leaks
in o he ma ke and igge s in o med ading. Based on he indings, wo impli-
ca ions can be de i ed o c yp ocu ency exchanges. On he one hand, i may be
wo hwhile o c oss-lis ICO okens, since he co esponding abno mal p ice e ec
shows ha in es o in e es exis s despi e al eady exis ing ma ke access. Howe e ,
p ima y lis ings po en ially may lead o e en highe in e es and ini ial e u ns – his
s udy canno answe his ques ion. The second implica ion is ha in o ma ion abou
upcoming lis ings should be be e secu ed o p o ec unin o med in es o s om
exploi a ion by in o med ade s. This does no only apply o exchanges bu is also
ele an o he supe iso y au ho i ies, which should c ea e legal ce ain y, i.e., eg-
ula ion, and ul ima ely in es iga e iola ions.
We also accep Hypo heses 2, as e ec s o c oss-lis ings di e signi ican ly ac oss
he 22 c yp ocu ency exchanges we examined. Only ou exchanges deli e signi i-
can posi i e esul s o he ull e en window, while i e exchanges yield nega i e
esul s. We ind e idence o unloading e ec s in he o m o signi ican nega i e
abno mal e u ns a e he c oss-lis ings, hough o a ying deg ees o he indi id-
ual exchanges. In ac , we do no ind any signi ican posi i e e ec s a e he e en
day. I he e o e appea s ha owne s o ICO okens, i.e., in es o s, should sell hei
okens no la e han he day o he lis ing, and po en ial buye s should e ain om
a has y pu chase bu wai a leas o h ee days. This p o ides a s a ing poin o
u u e s udies o longe ex-pos pe iods, since i is qui e concei able ha e ec s will
con inue o la en ou o possibly e en e e se.
Ano he in e es ing ac is ha Bi ex, which was based in he US when he
c oss-lis ing e en s ook place (now legally based in Liech ens ein), achie ed he
974
L.An e, A.Meye
1 3
highes CAAR o he sample. The US a e gene ally conside ed o o e good in es-
o p o ec ion. The success o p ojec s c oss-lis ed on Bi ex is consis en wi h
bonding heo y (Co ee 1999, 2002), which sugges s ha i ms om coun ies wi h
weake in es o p o ec ion can inc ease hei alua ion by c oss-lis ing in coun ies
wi h s onge secu i ies egula ion. This inding indica es ha p ojec s may ha e an
incen i e o lis hei okens on mo e igh ly egula ed exchanges in o de o signal
quali y and legal ce ain y.
Rega ding liquidi y- ela ed cha ac e is ics, we iden i y signi ican e ec s o all
o he i e selec ed me ics. We he e o e accep Hypo heses 3. High ma ke capi ali-
za ion and high ading olumes, i.e., by implica ion high liquidi y, o ICO okens
ha e nega i e e ec s on e u ns. Clea ly p e iously illiquid asse s bene i mo e om
a c oss-lis ing, as expec ed. Ma ke capi aliza ion o ading olume may co ela e
wi h he numbe o exchanges on which he asse is al eady adable. Lis ing an asse
on an addi ional exchange will hen only ha e ma ginal e ec s. These esul s on he
ma ke capi aliza ion o ICO okens may indica e ha p ojec s may ini ially sell hei
okens a a lowe alue, esul ing in a low alua ion. Wi h new ma ke s, his alua-
ion inc eases and ma ke liquidi y inc eases. This would con i m he ma ke liquid-
i y hypo hesis (Agga wal e al. 2002).
I seems concei able ha he ading olume o okens wi h lowe p io olume
inc eases mo e s ongly wi h c oss-lis ings. The e ec s a e g ea e o smalle p o-
jec s, which may be due o example o educed cos o capi al (Bake e al. 2002),
new cus ome s (Bancel and Mi oo 2001; Mi oo 1992; Pagano e al. 2002) o bond-
ing (An e 2019; Doidge e al. 2004). The in es iga ion o he a io be ween he wo
me ics o oken liquidi y con i ms his e ec . A high a io o ma ke capi aliza ion
o ading olume has a posi i e e ec on c oss-lis ing e u ns.
The analysis o he o e all ma ke , i.e., by using Bi coin as a p oxy, pe mi s some
conclusions ega ding he e ec s o ma ke liquidi y and ma ke sen imen . We iden-
i y a nega i e e ec o Bi coin’s ading olume, which may indica e ha in imes o
high ma ke liquidi y, c oss-lis ings ecei e less a en ion. The bulk o hese e ec s
occu s in (− 3, − 2) and in (+ 2, + 3), which indica es ha high ma ke liquidi y is
used o unloading. Exis ing in es o s may use p e-lis ing phases o en e posi ions
and he pos -lis ing phase o liquida e hei posi ions, as sugges ed by he li e a u e
on c oss-lis ings (Foe s e and Ka olyi 1999; Howe and Kelm 1987). Wi h ega d
o Bi coin’s ma ke capi aliza ion, we only ind a signi ican nega i e e ec o he
pos -lis ing window (+ 2, + 3). As wi h Bi coin ading olume, his may indica e
ha posi i e o e all ma ke sen imen encou ages exis ing in es o s o liquida e hei
posi ions and ha es he p o i s. In p inciple, e e se causali y is also concei able:
The sale o okens causes an inc ease in Bi coin ading olume. Howe e , any such
e ec s a e likely o be minimal, gi en he ela i e sizes o he asse s in ol ed.
Acco dingly, he ma ke his o y o ICO okens p o ides a basis o in es o s o
ade s o assess he po en ial e ec s o c oss-lis ings and o mula e po en ial in es -
men s a egies. In addi ion, he conside a ion o liquidi y- ela ed cha ac e is ics p o-
ides a basis o c yp ocu ency exchanges o iden i y p omising ICO okens o o
op imize he iming o c oss-lis ings. The implica ion o op imiza ion o c oss-lis ing
iming also applies o ICO p ojec s, which a e, howe e , dependen on exchanges
and he e o e only ha e limi ed in luence.
975
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
7 Limi a ions and u u e esea ch
Being among he i s in i s ield, his s udy is subjec o se e al limi a ions. Fi s ly,
sample size: Since he c yp ocu ency ma ke is s ill ela i ely young, he numbe
o ICOs and associa ed c oss-lis ing e en s is limi ed. We ha e been unable o iden-
i y and eco d all ICO c oss-lis ing e en s du ing he selec ed pe iod, and u he -
mo e had o exclude a la ge numbe o c oss-lis ings ha ook place in o e lapping
ime in e als. We use ma ke da a om coinma ke cap.com, which co e s he mos
impo an exchanges bu misses o he s. Ou da a se o 250 obse a ions gene ally
su ices o analyses, ye some o he sub-samples o indi idual exchanges a e on
he small side. A la ge sample would be use ul o con i m ou indings.
Secondly, he numbe o exchanges on which he p ojec s p e iously aded may
also a ec he abno mal e u ns o c oss-lis ing e en , as i does in s ock ma ke s
(You e al. 2013). Since we we e unable o ob ain his in o ma ion, he e is a isk o
omi ed a iable bias. I such da a could be collec ed ex-pos , i would be in e es -
ing o es he hypo hesis o he ‘dec easing ma ginal u ili y o c oss-lis ings’, i.e.,
whe he p ojec s ha a e al eady lis ed on a numbe o exchanges bene i less om
ye ano he c oss-lis ing. Thi dly, e en s udies in he s ock ma ke li e a u e o en
use a ious o he asse p icing models o signi ican ly longe es ima ion windows.
Applying hese me hods o he c yp ocu ency ma ke equi ed us o ely on a much
sho e p edic ion window and compa a i ely simple asse p icing models.
The e y limi ed li e a u e on c oss-lis ings o c yp ocu encies o ICOs and he
simul aneous weal h o analogous s udies on adi ional inancial ma ke s sugges
a numbe o a enues o u u e esea ch. Fi s , while we chose an e en window o
se en days and daily p ices, u u e s udies may conside comple ely di e en pe i-
ods and ime in e als. Ou esul s essen ially show ha he wide he window, he
highe he abno mal e u n: 6.51% o = 0, 8.7% o (− 1, + 1), 9.12% o (− 2, + 2)
and 9.97% o (− 3, + 3). This begs he ques ion as o how a his end may con-
inue. Also, he posi i e AAR o he e en day o e compensa es he nega i e CAAR
o he pos -lis ing pe iod (+ 1, + 3). Fu u e esea ch migh asce ain whe he lis ing
e en s also ha e a long- e m posi i e e ec .
Second, while we ha e p o ided some e idence on how c oss-lis ing e ec s di -
e be ween exchanges, he explana ion hese di e ences mus be le o u u e
esea ch. Possible de e minan s include he anspa ency o an exchange, o e -
all liquidi y, access o de i a i es, ia deposi s and wi hd awals, and loca ion. I
will be in e es ing o see whe he he e u ns depend mo e on he epu a ion o he
exchanges o on hei undamen al p ope ies. A ela ed ques ion is whe he he di -
e ences in e u ns a e e en a ibu able o he exchanges hemsel es o a he o he
p ojec s ha selec hese exchanges o c oss-lis ing.
Thi d, se en o he se en een exchanges ha ha e a posi i e bu insigni ican CAAR
o e he ull e en window could be called p e-e en gaine s, in ha mos o hei e ec s
acc ue be o e he ac ual e en . On hese exchanges and he ou signi ican ones, in o -
ma ion abou he c oss-lis ing appea s o ha e ound i s way in o he p ices be o e he
e en i sel . Howe e , since we do no know when he c oss-lis ings we e announced,
we a e unable o iden i y any in o ma ion leakage and po en ially in o med ading.
976
L.An e, A.Meye
1 3
Fou h, excess cash holdings o c oss-lis ed companies a e alued mo e highly
han hose o companies lis ed on a single s ock exchange (F ésa d and Sal a 2010).
The ma ke can o en moni o he amoun o c yp ocu ency collec ed, as he block-
chain echnology p o ides a anspa en iew o all asse s based on i . Fu u e in es-
iga ions could analyze he ela ionship be ween excessi e cash holdings and p ojec
alua ion. Fi h, p ojec s o en e ain some o hei own c yp ocu ency o bene i
om u u e app ecia ion. By moni o ing he blockchain and add esses wi h hese
speci ic okens, i should be possible o de e mine i eam okens a e ans e ed
immedia ely be o e he announcemen o new c oss-lis ings.
Six h, u u e esea ch could in es iga e o wha ex en abno mal e u ns su ound-
ing c oss-lis ings depend on in o ma ion abou he p ojec ha was al eady a ailable
a he ime o he ICO and on he basis o which in es o s assess he quali y o a p o-
jec , including human capi al, business model quali y, business plans o whi epape s,
unding goals and a a ie y o o he me ics (An e e al. 2018). Fo example, i may
be possible o iden i y whe he ax ha en-based p ojec s a e mo e likely o expe ience
abno mal e u ns be o e he announcemen o c oss-lis ings (in o med ading) o
whe he he p opo ion o okens e ained by he eam has an e ec on c oss-lis ings.
Classic key igu es om s ock ma ke s, such as he p ice- o-ea nings a io, can only
be applied o he c yp ocu ency ma ke o a e y limi ed ex en , i a all. Possibly el-
e an me ics could, howe e , be gene a ed on he basis o he co esponding numbe o
(ac i e) use s o a oken (e.g., Me cal e’s Law) o he oken’s ansac ion ac i i y (NVT
a io; he ma ke capi aliza ion in USD di ided by he daily olume (in USD) ha is
ansmi ed on he blockchain) on he blockchain. Decen alized compu e p o ocol p ob-
ably equi es special me ics, which in u n could explain e u ns om c oss-lis ings.
8 Conclusion
This e en s udy has analyzed p ice e ec s o c oss-lis ings o c yp ocu ency issued
h ough ICOs. The sample comp ises 250 lis ing e en s o 135 indi idual ICOs. On
he day o he c oss-lis ing, we ound abno mal e u ns be ween 6.51% and 6.73%,
depending on whe he he model o expec ed e u ns includes Bi coin as he e e -
ence ma ke . Ac oss a 7-day window om h ee days be o e o h ee days a e he
e en , he abno mal e u ns amoun o almos en pe cen . Thus, c oss-lis ings a e
gene ally posi i e e en s o ICO okens, as hey ep esen cos ly signals o quali y
o he ma ke . Fu he mo e, we ound ha he abno mal e u ns also depend on he
exchanges on which he lis ings ake place, as well as on liquidi y- ela ed me ics.
O he 22 indi idual exchanges analyzed, ou had posi i e signi ican e ec s o e
he 7-day pe iod. An asse ’s p io liquidi y, in he o m o ma ke capi aliza ion and
ading olume, has nega i e e ec s on lis ing e u ns.
The esul s ex end he exis ing li e a u e on c oss-lis ings o ICO okens, show-
ing ha he e u ns om such e en s sys ema ically depend on measu able ac o s.
The e ec s a e gene ally compa able o hose in s ock ma ke s, bu much s onge .
This indica es ha he ma ke s o ICO okens a e compa a i ely ine icien . The
esul s can help ade s o m hei s a egies, assis p ojec s in selec ing app op ia e

977
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
exchanges, help exchanges selec p ojec s and ime c oss-lis ings, and p o ide egu-
la o s and socie y wi h in-dep h insigh s in o he ICO ma ke . The esul s may also
inspi e a wide ange o u u e esea ch app oaches ha can b ing mo e anspa ency
o his s ill unde - esea ched and opaque ma ke .
Appendix
See Table4 and Fig.3.
Table 4 Desc ip i e s a is ics Va iable NMean SD Median Min. Max.
Binance 250 0.13 – 0 0 1
FCoin 250 0.10 – 0 0 1
Bi inex 250 0.10 – 0 0 1
Hi BTC 250 0.08 – 0 0 1
Huobi 250 0.06 – 0 0 1
Bi humb 250 0.06 – 0 0 1
Coinbene 250 0.05 – 0 0 1
Upbi 250 0.05 – 0 0 1
Bi ex 250 0.04 – 0 0 1
YoBi 250 0.04 – 0 0 1
KuCoin 250 0.04 – 0 0 1
Ga e.io 250 0.04 – 0 0 1
OKEx 250 0.04 – 0 0 1
Bibox 250 0.04 – 0 0 1
Bi o ex 250 0.02 – 0 0 1
LATOKEN 250 0.02 – 0 0 1
Ho bi 250 0.02 – 0 0 1
Liquid 250 0.02 – 0 0 1
Coinbase 250 0.02 – 0 0 1
Poloniex 250 0.01 – 0 0 1
K aken 250 0.01 – 0 0 1
STEX 250 0.00 – 0 0 1
ATV 250 15.05 2.22 15.08 8.12 20.64
AMC 245 18.33 1.77 18.34 14.14 22.70
AMC/ATV 245 3.24 1.21 3.16 0.27 7.82
BTV 250 22.35 0.58 22.29 19.60 24.00
BMC 250 25.45 0.42 25.46 23.66 26.33
978
L.An e, A.Meye
1 3
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Decla a ions
Con lic o in e es The au ho s decla e ha hey ha e no con lic o in e es .
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 Com-
mons 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
= -3 o +3 = -3 o -2
= -1 = 0
= +1 = +2 o +3
Fig. 3 CAAR dis ibu ion plo s ac oss di e en pe iods based on he ma ke model (N = 250)
979
1 3
C oss‑lis ings o blockchain‑based okens issued h ough…
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 eco mmons. o g/ licen
ses/ by/4. 0/.
Re e ences
Agga wal, R.K., K igman, L., Womack, K.L.: S a egic IPO unde p icing, in o ma ion momen um, and
lockup expi a ion selling. J. Financ. Econ. 66(1), 105–137 (2002). h ps:// doi. o g/ 10. 1016/ S0304-
405X(02) 00152-6
Alameda Resea ch. (2019). In es iga ion in o he legi imacy o epo ed c yp ocu ency exchange ol-
ume. h ps:// x. com/ olume- epo - pape . pd
Alexande , G.J., Janaki amanan, S.: In e na ional lis ings and s ock e u ns: some empi ical e idence. J.
Financ. Quan . Anal. 23(2), 135–151 (1988). h ps:// doi. o g/ 10. 2307/ 23308 77
Amihud, Y., Mendelson, H., Pede sen, L.H.: Liquidi y and asse p ices. Found. T ends Finance (2005).
h ps:// doi. o g/ 10. 1561/ 05000 00003
An e, L.: C yp ocu ency, blockchain and c ime. In: McCa hy, K. (ed.) The Money Launde ing Ma -
ke : Regula ing he c iminal economy, pp. 171–198. Agenda Publishing (2018). h ps:// doi. o g/ 10.
2307/j. c 5c g8z1. 10
An e, L.: Ma ke Reac ion o Exchange Lis ings o C yp ocu encies (2019). h ps:// doi. o g/ 10. 13140/
RG.2. 2. 19924. 76161
An e, L., Sandne , P., Fiedle , I.: Blockchain-based ICOs: pu e hype o he dawn o a new e a o s a up
inancing? J. Risk Financ. Manag. 11(4), 80 (2018). h ps:// doi. o g/ 10. 3390/ j m1 10400 80
An e, L., Fiedle , I., & S ehle, E.: The in luence o s ablecoin issuances on c yp ocu ency ma ke s.
Finance Res. Le . (2020). h ps:// doi. o g/ 10. 1016/j. l. 2020. 101867
An e, L.: Sma con ac s on he blockchain—a bibliome ic analysis and e iew. Telema . In . (2020).
h ps:// doi. o g/ 10. 1016/j. ele. 2020. 101519
Bake , H.K., No singe , J.R., Wea e , D.G.: In e na ional c oss-lis ing and isibili y. J. Financ. Quan .
Anal. 37(3), 495–521 (2002)
Bancel, F., Mi oo, U.R.: Eu opean manage ial pe cep ions o he ne bene i s o o eign s ock lis ings.
Eu . Financ. Manag. 7(2), 213–236 (2001)
Benede i, H.: The economics o digi al oken c oss-lis ings (2019). h ps:// doi. o g/ 10. 2139/ ss n. 32673 92
Benede i, H., Kos o e sky, L.: Digi al ulips? Re u ns In es . Ini ial Coin O . (2018). h ps:// doi. o g/ 10.
2139/ ss n. 31821 69
Bi wise Asse Managemen .: Bi wise asse managemen - p esen a ion o he U.S. secu i ies and exchange
commission (2019). h ps:// www. sec. go / comme n s/ s - nysea ca- 2019- 01/ s nys ea ca 201901- 51648
33- 183434. pd
Bou i, E., Roubaud, D., Shahzad, S.J.H.: Do Bi coin and o he c yp ocu encies jump oge he ? Q. Re .
Econ. Finance 76, 396–409 (2020)
B is, A., Can ale, S., Nishio is, G.P.: A b eakdown o he alua ion e ec s o in e na ional c oss-lis ing.
Eu . Financ. Manag. 13(3), 498–530 (2007). h ps:// doi. o g/ 10. 1111/j. 1468- 036X. 2006. 00370.x
Bu nie, A.: Explo ing he In e connec edness o c yp ocu encies using co ela ion ne wo ks. In: The
C yp ocu ency Resea ch Con e ence 2018, pp. 1–29 (2018). h p:// a xi . o g/ abs/ 1806. 06632
Can ale, S.: The Choice o a Fo eign Ma ke as a Signal (1996)
Chowdh y, B., Nanda, V.: Mul ima ke ading and ma ke liquidi y. Re . Financ. S ud. 4(3), 483–511
(1991)
Co ee, J.C.J.: Fu u e as his o y: he p ospec s o global con e gence in co po a e go e nance and i s
implica ions. No hwes . Uni . Law Re . 93(3), 641–708 (1999)
Co ee, J.C.J.: Racing owa ds he op: he impac o c oss-lis ing and s ock ma ke compe i ion on in e -
na ional co po a e go e nance. Columbia Law Re . 102(7), 1757–1831 (2002)
Dha an, B.G., Ikenbe y, D.L.: The long- un nega i e d i o pos -lis ing s ock e u ns. J. Finance 50(5),
1547–1574 (1995)
Doidge, C., Ka olyi, G.A., S ulz, R.M.: Why a e o eign i ms lis ed in he U.S. wo h mo e? J. Financ.
Econ. 71(2), 205–238 (2004)
980
L.An e, A.Meye
1 3
E unza, V., Losq, E.: In e na ional asse p icing unde mild segmen a ion : heo y and es . J. Finance
40(1), 105–124 (1985)
Fan o, J.A., Ka mel, R.S.: Companies ega ding a U.S. lis ing. S an . J. Law Bus. Finance 51, 51–83
(1997)
Fisch, C., Meoli, M., Visma a, S.: Does blockchain echnology democ a ize en ep eneu ial inance?
An empi ical compa ison o ICOs, en u e capi al, and REITs. Econ. Inno . New Technol. (2020).
h ps:// doi. o g/ 10. 1080/ 10438 599. 2020. 18439 91
Fisch, C., Mom az, P.P.: Ins i u ional in es o s and pos -ICO pe o mance: an empi ical analysis o in es-
o e u ns in ini ial coin o e ings (ICOs). J. Co p. Finance 64, 101679 (2020)
Foe s e , S.R., Ka olyi, G.A.: The e ec s o ma ke segmen a ion and in es o ecogni ion on asse p ices:
e idence om o eign s ocks lis ing in he Uni ed S a es. J. Finance 54(3), 981–1013 (1999). h ps://
doi. o g/ 10. 1111/ 0022- 1082. 00134
F ésa d, L., Sal a, C.: The alue o excess cash and co po a e go e nance: e idence om US c oss-lis -
ings. J. Financ. Econ. 98(2), 359–384 (2010). h ps:// doi. o g/ 10. 1016/j. j ine co. 2010. 04. 004
Gkillas, K., Beki os, S., & Si iopoulos, C.: X eme Co ela ion in C yp ocu ency Ma ke s (2018).
h ps:// doi. o g/ 10. 2139/ ss n. 31809 34
Howe, J.S., Kelm, K.: The s ock p ice impac s o o e seas lis ings. Financ. Manag. 16(3), 51 (1987).
h ps:// doi. o g/ 10. 2307/ 36659 80
Hu, A., Pa lou , C.A., Rajan, U.: C yp ocu encies: s ylized ac s on a new in es ible ins umen . Financ.
Manage. 48(4), 1049–1068 (2019). h ps:// doi. o g/ 10. 1111/ ima. 12300
Jaya aman, N., Shas i, K., Tandon, K.: The impac o in e na ional c oss lis ings on isk and e u n: he
e idence om Ame ican deposi o y eceip s. J. Bank. Finance 17(1), 91–103 (1993). h ps:// doi. o g/
10. 1016/ 0378- 4266(93) 90081-n
Ji, Q., Bou i, E., Lau, C.K.M., Roubaud, D.: Dynamic connec edness and in eg a ion in c yp ocu ency
ma ke s. In . Re . Financ. Anal. 63, 257–272 (2019)
McWilliams, A., Siegel, D.: E en s udies in managemen esea ch: heo e ical and empi ical issues.
Acad. Manag. J. 40(3), 626–657 (1997)
Me on, R.C.: A simple model o capi al ma ke equilib ium wi h incomple e in o ma ion. J. Finance
42(3), 483–510 (1987). h ps:// doi. o g/ 10. 1111/j. 1540- 6261. 1987. b045 65.x
Mille , D.P.: The ma ke eac ion o in e na ional c oss-lis ings: e idence om deposi a y eceip s. J.
Financ. Econ. 51(1), 103–123 (1999). h ps:// doi. o g/ 10. 1016/ S0304- 405X(98) 00045-2
Mille , M.H., Rock, K.: Di idend policy unde asymme ic in o ma ion. J. Finance 40(4), 1031–1051
(1985)
Mi oo, U.R.: Manage ial pe cep ions o he ne bene i s o o eign lis ing: Canadian e idence. J. In .
Financ. Manag. Acc. 4(1), 40–62 (1992). h ps:// doi. o g/ 10. 1111/j. 1467- 646X. 1992. b000 21.x
Mom az, P.P.: Ini ial Coin O e ings (2019a). h ps:// doi. o g/ 10. 2139/ ss n. 31667 09
Mom az, P.P.: The p icing and pe o mance o c yp ocu ency. Eu . J. Finance (2019b). h ps:// doi. o g/
10. 1080/ 13518 47x. 2019. 16472 59
Nakamo o, S. (2008). Bi coin : A Pee - o-Pee Elec onic Cash Sys em.
Pagano, M., Röell, A.A., Zechne , J.: The geog aphy o equi y lis ing: Why do companies lis ab oad? J.
Finance 57(6), 2651–2694 (2002). h ps:// doi. o g/ 10. 1111/ 1540- 6261. 00509
Roosenboom, P., Van Dijk, M.A.: The ma ke eac ion o c oss-lis ings: Does he des ina ion ma ke ma -
e ? J. Bank. Finance 33(10), 1898–1908 (2009). h ps:// doi. o g/ 10. 1016/j. jbank in. 2009. 04. 010
Sil a, A.C., Chá ez, G.A.: C oss-lis ing and liquidi y in eme ging ma ke s ocks. J. Bank. Finance 32(3),
420–433 (2008). h ps:// doi. o g/ 10. 1016/j. jbank in. 2007. 07. 003
Spence, M.: Job ma ke signaling. Q. J. Econ. 87(3), 355–374 (1973). h ps:// doi. o g/ 10.
1055/s- 2004- 820924
T oge , T.H.: Co po a e go e nance in a iable ma ke o seconda y lis ing. Uni . Pennsyl ania J. Bus.
Employ. Law 10(2003), 89–185 (2007)
Wa en, W., & Bandeali, A.: 0x: An open p o ocol o decen alized exchange on he E he eum block-
chain (2017). h ps:// doi. o g/ 10. 1007/ s0046 49901 080
You, L., Lucey, B.M., Shu, Y.: An empi ical s udy o mul iple di ec in e na ional lis ings. Glob. Finance
J. 24(1), 69–84 (2013). h ps:// doi. o g/ 10. 1016/j. g j. 2013. 03. 004
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.