Chen, Han eng; Hänsel, Ma hias; Nguyen, Hiep
A icle
Mone a y policy ansmission, cen al bank digi al
cu ency, and bank ma ke powe
Jou nal o Economics and S a is ics
P o ided in Coope a ion wi h:
De G uy e B ill
Sugges ed Ci a ion: Chen, Han eng; Hänsel, Ma hias; Nguyen, Hiep (2025) : Mone a y policy
ansmission, cen al bank digi al cu ency, and bank ma ke powe , Jou nal o Economics and
S a is ics, ISSN 2366-049X, De G uy e Oldenbou g, Be lin, Vol. 245, Iss. 4/5, pp. 527-576,
h ps://doi.o g/10.1515/jbns -2024-0008
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/333321
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/
Han eng Chen*, Ma hias Hänsel and Hiep Nguyen
Mone a y Policy T ansmission, Cen al Bank
Digi al Cu ency, and Bank Ma ke Powe
h ps://doi.o g/10.1515/jbns -2024-0008
Recei ed Janua y 7, 2024; accep ed Decembe 10, 2024
Abs ac : In e es a es on new cen al bank digi al cu encies (CBDCs) can be
expec ed o en e he mone a y policy oolki soon. Using an ex ended Sid auski
(1967) model ea u ing an oligopsonis ic banking sec o , we s udy he complex
ansmission o in e es a es on CBDC, which gene ally in ol e bo h di ec and
indi ec effec s. This is because a CBDC a e cu does no only affec he a e on he
CBDC i sel , bu also induces he non-compe i i e deposi p o ide s o adjus hei
sp eads, as he new subs i u e o hei p oduc s becomes ela i ely less a ac i e. A
calib a ion exe cise sugges s ha he indi ec effec s depend s ongly on he sou ces
o deposi ma ke powe : I d i en by high concen a ion, hey subs an ially ampli y
he agg ega e effec s o he CBDC policy a e, bo h in esponse o ansi o y shocks as
well as ega ding i s long- un wel a e effec s. This con as s hem wi h policies
di ec ed a he banking sec o which a e weakened by a less compe i i e deposi
ma ke .
Keywo ds: CBDC; digi al cu ency; bank ma ke powe ; mone a y ansmission
JEL Classifica ion: E42; E43; E52
A icle No e: This a icle is pa o he special issue “Cen al Bank Digi al Cu ency”published in he
Jou nal o Economics and S a is ics. Access o u he a icles o his special issue can be ob ained a
www.deg uy e .com/jbns .
We a e g a e ul o Mikael Bask, Ch is oph Be sch, Mikael Ca lsson, Da ia Finocchia o, Paul Klein, Pe
K usell, Ku Mi man, La s Ljungq is , and Yimei Zou o hei use ul eedback and commen s. Financial
suppo by he Jan Wallande s och Tom Hedelius Founda ion, Sweden, is g a e ully acknowledged by
Hänsel. The opinions exp essed in his a icle a e he sole esponsibili y o he au ho s and do no
necessa ily eflec he iews o he Na ional Ins i u e o Economic Resea ch.
*Co esponding au ho : Han eng Chen, Uppsala Uni e si y, Uppsala, Sweden; Cen e o Mone a y
Policy and Financial S abili y, S ockholm Uni e si y, S ockholm, Sweden; and Na ional Ins i u e o
Economic Resea ch, S ockholm, Sweden, E-mail: [email p o ec ed]
Ma hias Hänsel, S ockholm School o Economics, S ockholm, Sweden; and Cen e o Mone a y Policy
and Financial S abili y, S ockholm Uni e si y, S ockholm, Sweden
Hiep Nguyen, Uppsala Uni e si y, Uppsala, Sweden
Jou nal o Economics and S a is ics 2025; 245(4–5): 527–576
Open Access. © 2025 he au ho (s), published by De G uy e . This wo k is licensed unde he
C ea i e Commons A ibu ion 4.0 In e na ional License.
1 In oduc ion
Mone a y au ho i ies a ound he wo ld a e explo ing he possibili y o issuing a new
digi al paymen ins umen widely accessible o he public. As o May 2024, 134
coun ies and cu ency unions, which accoun o 98 % o he global GDP, a e
conside ing a cen al bank digi al cu ency (CBDC) (A lan ic Council 2024). Mo i a-
ions o such a new paymen ins umen include ensu ing adequa e public money,
educing sys emic isk and imp o ing financial s abili y, inc easing compe i ion in
paymen s, and p omo ing financial inclusion (Enge and Ben Siu-Cheong 2017).
One o he less-discussed aspec s o CBDC is i s po en ial o enable a di ec
implemen a ion o mone a y policy (e.g., Aue e al. 2022; Bank o In e na ional
Se lemen 2020). In e es on CBDC could become a new policy ins umen , p o iding
policy-make s wi h g ea e flexibili y o influence he eal e u ns o money asse s
and sides ep financial in e media ies. Howe e , banks will no idly s and by i he
cen al bank makes i mo e a ac i e o hold an asse p o iding simila se ices as
hei deposi s. In u n, he ac ual equilib ium impac o CBDC a es should depend
bo h on households’liquidi y p e e ences and he esponse o financial sec o agen s
wi h subs an ial ma ke powe .
A key con ibu ion o ou wo k compa ed o he exis ing li e a u e is ha
ega ding he la e channel, we explici ly dis inguish be ween diffe en sou ces o
bank ma ke powe , ma ke concen a ion and diffe en ia ion: Cu en heo ies end
o ocus ei he on one o he o he (see below o u he discussion), bu clea ly, bo h
aspec s a e ele an in eali y: The ma ke o bank deposi s does no only display
subs an ial concen a ion (see e.g. Co bae and D’E asmo 2020) bu he espec i e
deposi s also diffe in p ac ice due o egionally diffe ing b anch ne wo ks, bundling
wi h diffe en paymen ca ds, e c. Indeed, we find he ac ual sou ce o bank ma ke
powe o ha e impo an implica ions o he effec o a CBDC a e as a policy ool,
which sugges s assump ions on he o m o impe ec bank compe i ion o be c ucial
o he ou comes o quan i a i ely modeling CBDC policy.
In pa icula , we conduc ou analysis using an ex ended Sid auski (1967) model
building on he amewo k p oposed by Niepel (2024): In he model, households gain
u ili y om holding diffe en o ms o liquid asse s which allows us o cap u e
a ious ela ed aspec s such as CBDC design and he p i a esec o ’s need o deposi s
o diffe en banks in a pa simonious way close o ex book heo ies. Howe e , i
should be no ed ha such a model has he implica ion ha CBDC is no “special”
compa ed o al e na e o ms o go e nmen -p o ided liquidi y (in he sense o
explici ly modeled design ea u es). In u n, ou heo e ical analysis would equally
apply o o he liquidi y ypes as long as cen al banks can affec hei e u ns flexibly
enough.
528 H. Chen e al.
C ucially, in ou amewo k, we allow o a banking sec o in which bank ma ke
powe is de i ed bo h om ma ke concen a ion and households’impe ec abili y
o subs i u e be ween banks. We assume a common deposi ma ke in which a se o
non-compe i i e banks compe e by offe ing diffe en ia ed deposi s, bu do no
es ic i o be ei he monopsonis ic (as e.g. in Niepel 2024) o monopsonis ically
compe i i e (as e.g. in Bacche a and Pe azzi 2022). Ra he , such se ings a e nes ed as
limi cases, allowing us o a y he deg ee o deposi ma ke concen a ion in o de o
demons a e i s impo ance o he ansmission o CBDC a es. We also do no
es ic diffe en banks’deposi s o be undamen ally he same o households as in
Chiu e al. (2023), so ha oligopsonis ic se ings emain cha ac e ized by diffe en
sou ces o ma ke powe .
Addi ionally, we assume limi ed subs i u abili y be ween CBDC and bank deposi s,
wi h pe ec subs i u abili y asa limi case. We belie e he e a e good easons oexpec
ha in p ac ice, CBDC would no be almos pe ec ly subs i u able wi h bank deposi s.
Fo example, bank deposi s a e ypically bundled wi h o he financial se ices such as
c edi lines (e.g. o e d a acili ies, c edi ca ds), while CBDC may be pe cei ed as
offe ing mo e p i acy and secu i y. Addi ionally, o he ea u es such as he in e op-
e abili y be ween CBDC and deposi s and he abili y o conduc in e na ional ans-
ac ions migh also limi p ac ical subs i u abili y (Bacche a and Pe azzi 2022).
Fo he pu pose o his pape , we conside he in e es a es on CBDC as he main
policy ins umen o in e es bu also discuss implica ions o ese e a es. In ou
model, bo h can be shown o affec he eal alloca ion h ough he a e age cos o
liquidi y, bu he influence o he CBDC sp ead consis s o bo h a di ec and an indi ec
effec . Clea ly, an inc ease in he CBDC sp ead ( ela i e o a isk- ee a e) di ec ly
inc eases he households’cos o liquidi y. A he same ime, he ising sp ead also
enables banks o widen he sp eads on he deposi s hey offe , as he al e na i e
sou ce o liquidi y becomes compa a i ely less a ac i e. This in oduces he indi-
ec effec , eminiscen o he deposi channel o mone a y policy p oposed by
D echsle e al. (2017).
While he quan i a i e magni ude o he di ec effec depends simply on he
amoun o CBDC households will choose o hold gi en i s design, he s eng h o he
indi ec effec is mo e nuanced, depending c ucially on he sou ce o ma ke powe
in he deposi ma ke . In ui i ely, i deposi ma ke concen a ion is low, indi idual
banks a e small and canno affec he amoun o CBDC households will choose o
hold. In u n, changes in he CBDC sp ead ha e li le impac on he equilib ium
deposi sp ead and he indi ec effec is small. On he o he hand, i he deposi
ma ke is highly concen a ed, banks can p ac ically compe e wi h CBDC and adjus
hei sp eads mo e, making he indi ec effec ela i ely la ge. Ou calib a ion
exe cises sugges he indi ec effec o subs an ially ampli y he agg ega e esponse
o CBDC a e changes in se ings wi h high ma ke concen a ion bu less so i
CBDC Policy T ansmission 529
concen a ion is lowe and diffe en ia ion cons i u es a ela i ely mo e impo an
sou ce o deposi ma ke powe .
Such conside a ions do no only apply o empo a y changes in in e es a es,
bu also o how cen al bank policy can affec long- un wel a e h ough inducing a
mo e efficien liquidi y mix. Indeed, we no only demons a e ha in a se ing wi h
mo e deposi ma ke concen a ion, he effec s o he CBDC a e on deposi sp eads
and wel a e become mo e p onounced, bu also h ough a decomposi ion ha he
o me is indeed he cause o he la e . This also implies ha he mone a y policy
chosen by a Ramsey planne wi h limi ed ins umen s will depend on deposi ma ke
concen a ion.
In con as o CBDC policy, he impac o ese e a es, which we b iefly analyze
as s and-in o mone a y policy ins umen s di ec ed a he banking sec o , dec eases
wi h highe deposi ma ke concen a ion. Unde he assump ion ha he shock
makes i mo e expensi e o banks o p o ide deposi s, i causes banks o inc ease
hei deposi sp eads and households’cos o liquidi y. This effec is smalle i ma ke
concen a ion is high, as his makes he sp eads cha ged by banks ela i ely mo e
dependen on hei demand schedule, which is o he wise no di ec ly affec ed by he
policy.
Ou wo k ela es o he g owing and ecen li e a u e on CBDC, which has
s udied hese po en ial new paymen ins umen s om a a ie y o pe spec i es. Fo
example, Agu e al. (2022) analyze he ade-offs associa ed wi h CBDC design gi en
he e ogeneous household p e e ences o e paymen ins umen s and ne wo k
effec s ega ding hei use. They conclude ha cen al banks should indeed issue
in e es -bea ing CBDCs and choose hei a e so ha o he paymen ins umen s
emain in use. Simila ly, Keis e and Sanches (2022) highligh ade-offs associa ed
wi h CBDC design choices. In pa icula , hey a gue ha a CBDC wi h a deposi -like
design would ha e posi i e effec s by inc easing paymen - and exchange efficiency,
bu may also dec ease in es men by inducing highe unding cos s o banks.
Piazzesi and Schneide (2022), in u n, wa n ha CBDC c owding ou bank deposi s
may dec ease efficiency in financial in e media ion due o a complemen a i y
be ween offe ing bo h deposi s and c edi lines. O he wo k has s udied he impac o
CBDC adop ion on financial s abili y wi h diffe ing findings, i.e. ha CBDC may ei he
imp o e financial s abili y (Fe nández-Villa e de e al. 2021) o encou age banking
panics (Williamson 2022).
Gi en ha we s udy CBDC in a se -up wi h non-compe i i e banks, ou wo k is
pa icula ly ela ed o Andol a o (2021); Bacche a and Pe azzi (2022) as well as Chiu
e al. (2023), which all s udy he impac o CBDC in oduc ion in he p esence o a non-
compe i i e banking sec o . Andol a o (2021) ocuses on he impac o CBDC in o-
duc ion on bank lending and economic ac i i y and finds ha a CBDC may no
impede ei he . In ac , non-compe i i e banks o ced o inc ease hei deposi a es
530 H. Chen e al.
will be subjec o an addi ional inflow o deposi s due o he mo e a ac i e a es and,
in u n, con e his addi ional unding in o lending. Chiu e al. (2023) ob ain simila
esul s in a diffe en se -up allowing o diffe ing deg ees o bank ma ke powe . In
con as o ou wo k, hese pape s only conside concen a ion as a sou ce o deposi
bank ma ke powe and ocus on he long- un effec s o CBDC on bank lending and
gene al economic ac i i y and conside , while we also analyze he ansmission o
sho - e m shocks. While Bacche a and Pe azzi (2022) also sha e he long- un ocus,
hey conside a monopolis ic compe i i e deposi ma ke on which deposi ma ke
powe is only d i en by diffe en ia ion.
Jiang and Zhu (2021) and Ga a e al. (2022) sha e ou ocus by s udying mon-
e a y pass- h ough in se ings wi h impe ec ly compe i i e o he e ogeneous banks,
espec i ely. Jiang and Zhu (2021) s udy he pass- h ough o bo h ese e and CBDC
a es in a amewo k simila o Chiu e al. (2023). In he p esence o a non-compe i i e
banking sec o , he in oduc ion o CBDC is shown o po en ially weaken he ese e
pass- h ough, as pe ec subs i u abili y o ces banks o ma ch he CBDC a e on he
deposi ma ke . CBDC can essen ially “dic a e” he economy. The CBDC a e, in u n,
may ha e a pa icula ly s ong pass- h ough o deposi a es, while i s effec s on loan
a es depend on he ese e a e in an ambiguous way. Majo diffe ences be ween
he wo k o Jiang and Zhu (2021) and ou s a e ha hey also only conside one ma gin
o banking compe i ion, and, due o he assump ion o pe ec subs i u abili y
be ween bank deposi s and CBDC, ule ou he p esence o he indi ec effec s
discussed abo e, as he CBDC a e will ei he de e mine he deposi a e comple ely
o no affec i all. Ga a e al. (2022) conside a amewo k wi h diffe ing bank ypes
(“la ge”and “small”) compe ing o deposi s om wo ke s ha ing he e ogeneous
p e e ences o e he non-mone a y benefi s (e.g. ex ensi e b anch ne wo ks) hey
offe . They find ha he pass- h ough o he CBDC a e o he deposi a e is s onge i
he CBDC a e is high compa ed o he ese e a e, which, howe e , hu s he “small”
bank. In con as o ou wo k, hei ocus is on bank he e ogenei y, om which we
abs ac . Also, in hei se up, no one ac ually ends up holding CBDC ( he digi al
cu ency can again pe ec ly subs i u e o bank deposi s and is ou -compe ed by
banks), so hei model canno p o ide o indi ec effec s o he CBDC a e on
households’liquidi y cos s ei he .
Fu he mo e, ou esea ch is ela ed o se e al s udies analyzing he in e ac ion
o bank ma ke powe and mone a y policy ansmission. In pa icula , D echsle
e al. (2017) p opose a deposi channel o mone a y policy. As in e es a e inc eases
aise he oppo uni y cos s o holding cash, non-compe i i e banks a e able o
inc ease he deposi sp ead in esponse o igh e mone a y policy, consequen ly
educing he o e all amoun o deposi s. This, in u n, can affec bo h he liquidi y
p emium and bank lending. Choi and Roche eau (2023) s udy his channel
CBDC Policy T ansmission 531
heo e ically in a sea ch- heo e ic model o deposi ma ke s. Addi ionally, es ima ing
a s uc u al model o he banking sec o , Wang e al. (2022) simila ly find bank
ma ke powe o ha e impo an effec s on he ansmission o a e changes o
deposi a es. In addi ion o he D echsle e al. (2017) mechanism, hei model also
explici ly conside s an oligopolis ic lending ma ke , whe e banks addi ionally
espond by adjus ing hei lending a e ma kups.
The es o he pape is o ganized as ollows: Sec ion 2 desc ibes he elemen s o
he model economy and cha ac e izes i s equilib ium. Sec ion 3 analyzes he
ansmission mechanisms o he in e es a es on CBDC and ese es by quali a i ely
cha ac e izing he channels h ough which he in e es a es affec eal alloca ion.
Sec ion 4 calib a es he model o conduc nume ical exe cises. Then, Sec ion 5 quan-
i a i ely demons a es he ex en o which deposi ma ke powe affec s policy
ansmissions unde sho - un mone a y policy shocks. Sec ion 6 in es iga es he
implica ions o deposi ma ke powe o he efficacy o CBDC policy in he long un.
Nex , we conduc obus ness es s in Sec ion 7. Finally, Sec ion 8 summa izes he
esul s and concludes.
2 Model
We s udy an ex ended Sid auski (1967) model, building on Niepel (2024), in which
bo h he go e nmen and banks p o ide liquidi y o households ha gain u ili y om
holding i . Households subs i u e impe ec ly be ween a go e nmen -issued o m o
liquid asse ha we in e p e as CBDC and comme cial bank deposi s. Banks und
hemsel es by bo owing deposi s om he households and in es in capi al and
ese es which a e used o “back up”deposi issuance. We ollow D echsle e al.
(2017) and assume ha banks a e non-compe i i e in he deposi ma ke . Banks ha e
ma ke powe due o bo h ma ke concen a ion and impe ec subs i u abili y
be ween banks’deposi se ices. Neoclassical fi ms p oduce a common consump ion
good using capi al and labo , and a consolida ed go e nmen /cen al bank issues
CBDC and ese es.
2.1 Households
We conside an economy consis ing o many iden ical and infini ely-li ed house-
holds, wi h he measu e no malized o one. The ep esen a i e household alues
consump ion, c
and liquidi y se ices, z
+1
, acco ding o a pe iod u ili y unc ion o
he o m
532 H. Chen e al.
u(c ,z +1)= (1− )c1−ψ
+ z1−ψ
+1
()
1−σ
1−ψ
1−σ,
whe e ∈(0,1)is he ela i e weigh o liquidi y se ices in u ili y, ψ∈(0, 1) is he
in e se in a empo al elas ici y o subs i u ion be ween consump ion and liquidi y,
and σ> 0 is he in e se in e empo al elas ici y o subs i u ion be ween consump ion-
liquidi y bundles ac oss ime. We assume ha CBDC and deposi s a e impe ec sub-
s i u es: Liquidi y se ices a e de i ed om eal holdings o CBDC, m
+1
, and deposi s,
n
+1
, acco ding o a cons an elas ici y o subs i u ion (CES) agg ega o
z +1=(1−γ)m1−ϵ
+1+γn1−ϵ
+1
()
1
1−ϵ,
whe e γ∈(0, 1) is he ela i e liquidi y weigh o bank deposi s, and ϵ∈(0, 1) is he
in e se elas ici y o subs i u ion be ween CBDC and deposi s. The liquidi y weigh
pa ame e , γ, cap u es how use ul deposi s a e o he pu pose o holding liquidi y
ela i e o he same quan i y o CBDC. We ollow D echsle e al. (2017) in assuming
ha deposi s a e hemsel es a composi e good issued by a se o Nnon-compe i i e
banks. Each bank ihas mass 1/Nand p oduces deposi s o a quan i y ni
+1/N. The
household alues deposi s a diffe en banks such ha
n +1=1
N∑
N
i=1
ni
+1
()
1−η
()
1
1−η
,(1)
whe e ηdeno es he in e se elas ici y o subs i u ion be ween banks. The ep esen-
a i e household can be hough o as an agg ega ion o many indi idual households
who may ha e di e se p e e ences o holding deposi s a diffe en banks. The e o e,
he ep esen a i e household subs i u es deposi s impe ec ly ac oss banks, which
implies ha 0 < η<1.
In ou amewo k, in addi ion o deposi s and CBDC, households can in es
di ec ly in capi al. This is necessa y o he model o ea u e ealis ic amoun s o
capi al and liquid asse s, as he agg ega e amoun o he o me ypically a exceeds
he amoun o he la e in mode n economies.
1
Since we ound i o be no c ucial o
ou esul s, we abs ac om a labo supply choice o simpli y he analysis and
ins ead assume he ep esen a i e agen o inelas ically supply a cons an amoun o
labo l. The household’s budge cons ain is hen gi en by
1No e ha ou assump ion is isomo phic o al e na i ely assuming ha households do no hold
capi al di ec ly bu also p o ide unding o banks h ough a compe i i e asse ma ke no p o iding
liquidi y se ices. I banks we e no only he only agen s able o hold capi al and only ob ain unding
h ough deposi s, he model would ei he ea u e way oo much deposi s, way oo li le capi al, o way
oo low bank le e age a ios.
CBDC Policy T ansmission 533
c +kh
+1+m +1+1
N∑
N
i=1
ni
+1+τ =w l+π +kh
Rk
+m Rm
+1
N∑
N
i=1
ni
Rn,i
,(2)
whe e kh
+1a e di ec holdings o capi al, τ
is he lump-sum ax ne o go e nmen
ans e , w
is he wage a e, π
is he di idends om fi ms and banks, Rk
is he e u n
on capi al, Rm
+1is he eal g oss in e es a e on CBDC, and Rn,i
+1is he eal g oss in e es
a e on deposi s a bank i. We assume ha he e u ns on CBDC and deposi s a e
isk- ee, i.e. Rm
+1and Rn,i
+1a e known a ime . The household, aking p ices, p ofi s,
and axes as gi en, sol es
max
c ,kh
+1,m +1,ni
+1
{}
∞
=0
E0∑
∞
=0
β u(c ,z +1)
s. .c +kh
+1+m +1+1
N∑
N
i=1
ni
+1+τ =w l+π +kh
Rk
+m Rm
+1
N∑
N
i=1
ni
Rn,i
,
kh
+1,m +1,ni
+1≥0.
We now u n o he fi s -o de op imali y condi ions o he household p og am.
De ailed de i a ions a e p o ided in he Appendix A.1. Fi s , he household op imally
alloca es esou ces be ween deposi s a indi idual banks acco ding o
ni
+1=n +1
χn,i
+1
χn
+1
()
−1
η
,(3)
which closely esembles demand equa ions o diffe en ia ed consump ion goods
commonly de i ed in New Keynesian models. The ela i e sha e o deposi s a bank i,
ni
+1/n +1, mus ela e nega i ely o i s co esponding ela i e cos , χn,i
+1/χn
+1. He e, χn,i
+1
is he in e es - a e diffe en ial be ween he isk- ee a e, R
+1, and he deposi a e
offe ed by bank i
χn,i
+1=1−Rn,i
+1
R
+1
,
which ep esen s he oppo uni y cos o holding deposi s a bank iand which we
he ea e e e o as deposi sp ead. The isk- ee a e is defined in he s anda d way
as he in e se o he expec ed alue o he household’s s ochas ic discoun ac o ,
Λ
+1
,
R
+1=1
E [Λ +1],(4)
whe e he s ochas ic discoun ac o is defined as ollows:
534 H. Chen e al.
c +k +1−k (1−δ)=a kα
l1−α−Q ,
whe e
Q =m +1μ+n +1ν +ζ +1ρ
()
.
The esou ce cons ain has he s anda d in e p e a ion ha a ailable ou pu in he
economy is spli be ween consump ion, c
, and in es men , k
+1
−k
(1 −δ). Howe e ,
he e a e esou ce cos s associa ed wi h he p o ision o liquidi y o he household,
summa ized by he e m Q:μpe uni o CBDC and ν
+ζ
+1
ρpe uni o deposi . The
esou ce cos o deposi s has wo e ms because he banking sec o incu s a cos o
deposi issuance, ν
, and he go e nmen incu s a cos o issuing ese es used by he
banking sec o , ζ
+1
ρ, o“back up”deposi issuance. As he household demands
liquidi y se ices in p opo ion o consump ion, we can combine he e ms c
and Q
,
and ew i e he esou ce cons ain as
c Ω c
+k +1−k (1−δ)=a kα
l1−α,(25)
whe e Ω c
≥1 is gi en by
Ω c
=1+
1−
1
χz
+1
()
1
ψ
(1−γ)χz
+1
χm
+1
()
1
ϵ
μ+γχz
+1
χn
+1
()
1
ϵ
ω+ϕζ1−φ
+1+ζ +1ρ
()
⎛
⎝⎞
⎠.(26)
2.6 Policy and Equilib ium
The consolida ed go e nmen se s he in e es a es on CBDC and ese es and
elas ically supplies hese asse s o households and banks o mee demand. A policy
consis s o Rm
+1,R
+1,τ
{}
≥0and an equilib ium condi ional on policy consis o
–a se o posi i e p ices, w ,Rk
+1,R
+1,χm
+1,χn
+1,χz
+1,χ
+1
{}
≥0;
–a posi i e alloca ion, {c ,k +1} ≥0;
–and posi i e CBDC, deposi s and ese es holdings, {m +1,n +1,z +1, +1} ≥0,
such ha (4), (6)–(11), (14)–(16), (22), (23) and (25) a e sa isfied.
3 Mone a y Policy T ansmission
In his sec ion, we elabo a e on he ansmission mechanisms o he in e es a es on
CBDC and analy ically cha ac e ize he channels h ough which hey affec he
alloca ion. We also b iefly discuss he ansmission o ese e a es, gi en ha we
will con as hei effec s wi h CBDC below. O e all, his analysis builds he
CBDC Policy T ansmission 541
ounda ion o he quan i a i e exe cise in he nex sec ions whe e we s udy how
mone a y policy affec s he eal economy.
3.1 Real Effec s o Mone a y Policy
The wo key condi ions ha cha ac e ize he equilib ium alloca ion, he Eule
equa ion (11) and he esou ce cons ain (25), all closely pa allel he condi ions o a
ex book RBC model. The diffe ences ela i e o an RBC model a e he quan i ies Ωc
+1
and Ω c
+1. Impo an ly, he di ec impac o liquidi y on he household’s consump ion/
sa ings decision, cap u ed by Ωc
+1, depends solely on he a e age cos o liquidi y,
χz
+1. So we will mos ly ocus on he effec s o policy on χz
+1when s udying ans-
mission below. Fo his pu pose, i is ins uc i e o fi s lay down how he a e age
cos o liquidi y wo ks h ough ou model economy.
The Eule equa ion (11) shows ha he household’s consump ion/sa ings choice
depends on liquidi y h ough he ma ginal u ili y o consump ion, which changes
wi h he a e age cos o liquidi y acco ding o
∂uc,
∂χz
+1
=c−σ
∂Ωc
∂χz
+1
,whe e ∂Ωc
∂χz
+1
∝σ−ψ
ψ.
We see ha he sign o he impac on he ma ginal u ili y o consump ion depends on
he ela i e magni udes o ψand σ. I he household’s in a empo al elas ici y o
subs i u ion be ween consump ion and liquidi y is smalle han he in e empo al
elas ici y o subs i u ion, i.e. ψ>σ, an inc ease in he cos o liquidi y leads o a
dec ease in he ma ginal u ili y o consump ion. This is d i en by he ac ha an
inc ease in he cos o liquidi y, acco ding o (6), educes he household’s demand o
i . A dec ease in he le el o liquidi y hen dec eases he ma ginal u ili y o con-
sump ion, and hence he e is consump ion–liquidi y complemen a i y. On he o he
hand, when ψ<σ, an inc ease in he cos o liquidi y leads o an inc ease in he
ma ginal u ili y o consump ion. In he case whe e ψ=σ, he household’s u ili y is
sepa able in consump ion and liquidi y and he cos o liquidi y has no di ec impac
on consump ion/sa ings choices.
Mo eo e , he sp eads on CBDC and deposi s also show up in he agg ega e
esou ce cons ain (25) h ough he e m Ω c
. This eflec s he esou ce cos s asso-
cia ed wi h liquidi y p o ision, incu ed by he go e nmen and he banking sec o .
In he special case whe e he household does no alue liquidi y se ices, i.e. →0,
bo h Ωc
+1and Ω c
+1con e ge o one. A his “cashless limi ”, he cos o liquidi y has no
impac on he household’s consump ion/sa ings since no liquid asse s a e held.
The e o e, he e a e also no esou ce cos s associa ed wi h liquidi y p o ision. Then,
he model collapses in o a s anda d RBC model.
542 H. Chen e al.
To conclude, we ha e seen ha he household’s consump ion/sa ings decision
only depends on he a e age cos o liquidi y, which in u n is a unc ion o he
sp eads on CBDC and deposi s. As hese a e also he sole endogenous de e minan s o
he liquidi y cos e m Ω c
+1, he go e nmen can affec he alloca ion only inso a as i
affec s hese sp eads. While he go e nmen con ols he CBDC sp ead di ec ly
h ough he CBDC a e, he deposi sp ead is de e mined by he banking sec o . Bu as
we will see below, he go e nmen can influence i s beha io h ough he in e es
a es on bo h ese es and CBDC.
3.2 In e es on CBDC
We now explain he channels h ough which he household’s a e age cos o
liquidi y can be influenced by he CBDC a e. Suppose he go e nmen lowe s he
CBDC a e so ha he CBDC sp ead widens.
2
Diffe en ia ing he a e age cos o
liquidi y, gi en by (7), wi h espec o he CBDC sp ead yields
∂χz
+1
∂χm
+1
=(1−γ)1
ϵχm
+1
χz
+1
()
−1
ϵ
⏟⏞⏞⏟
di ec e ec
+γ1
ϵχn
+1
χz
+1
()
−1
ϵ∂χn
+1
∂χm
+1
⏟⏞⏞⏟
indi ec e ec
.(27)
This exp ession shows ha he CBDC sp ead wo ks h ough wo channels: Fi s ly, i
di ec ly inc eases he cos o liquidi y by he fi s e m. The s eng h o his di ec
effec is inc easing in he ela i e liquidi y weigh o CBDC, 1 −γ, and dec easing in
how cos ly CBDC is ela i e o he a e age cos o liquidi y, χm
+1/χz
+1. Compa ing he
di ec effec wi h he household’s demand o CBDC (9), we see ha i is jus he sha e
o CBDC in he o al s ock o liquidi y, m
+1
/z
+1
. In ui i ely, he mo e impo an
CBDC is as a sou ce o liquidi y o he household, he la ge he impac o i s cos on
liquidi y’s a e age cos .
Secondly, he CBDC sp ead affec s he cos o liquidi y h ough he deposi side,
gi en by he second e m. The s eng h o his indi ec effec is inc easing in he
ela i e liquidi y weigh o deposi s, γ, and dec easing in how cos ly deposi s a e
ela i e o he a e age, χn
+1/χz
+1. Compa ing he indi ec effec wi h he household’s
demand o deposi s (10), we see ha i is equal o he p oduc o he sha e o deposi s
in he o al s ock o liquid, n
+1
/z
+1
, and he change in he deposi sp ead caused by a
change in he CBDC sp ead, ∂χn
+1/∂χm
+1. Analogous o he di ec effec , he mo e
impo an deposi s a e as a sou ce o liquidi y he la ge is his indi ec effec .
Howe e , he sign and he magni ude o he second effec also depend on how he
banking sec o esponds o an inc easing CBDC sp ead, cap u ed by ∂χn
+1/∂χm
+1.
2Fo simplici y, we assume he e ha he ese e sp ead is cons an .
CBDC Policy T ansmission 543
In his ega d, he op imali y condi ion (16) shows ha he CBDC sp ead can
influence he deposi sp ead h ough he bank’s ma ginal benefi o deposi issuance
(le -hand side). Specifically, CBDC sp ead affec s he elas ici y o demand o deposi s
ha he bank aces, gi en by (18). As we discussed p e iously, he demand elas ici y
depends on a weigh ed a e age o he household’s elas ici ies o subs i u ion o
consump ion, 1/ψ, and CBDC, 1/ϵ. The CBDC sp ead de e mines his a e age h ough
he ela i e weigh s
, gi en by (19). Taking he pa ial de i a i e o he demand
elas ici y (18) wi h espec o he CBDC sp ead, we ge
1
N
∂s
∂χm
+1
()
1
ψ−1
ϵ
()wi h ∂s
∂χm
+1
=−1−ϵ
ϵ
s (1−s )
χm
+1
<0.(28)
The pa ial de i a i e shows ha he ma ginal impac o CBDC sp ead is non-ze o
only i ψ≠ϵ. In ui i ely, banks collec i ely ace compe i ion om CBDC and
consump ion o he household’s esou ces. The e o e, any ou flow om deposi s
depends on he household’s elas ici ies o subs i u ion o CBDC and consump ion. The
CBDC sp ead only influences he ela i e impo ance o hese wo sou ces o deposi
ou flow, indica ed by s
. I he household finds i as easy o subs i u e om deposi s o
consump ion as i does o CBDC, i.e. ψ=ϵ, hen he wo sou ces o compe i ion o he
banks a e equally impo an and he CBDC sp ead does no influence he banks’
deposi sp ead. In such a case, he equilib ium sp ead is se equal o he ma ginal cos
o deposi issuance plus a cons an ma kup, simila o he case whe e banks a e
monopsonis ically compe i i e.
In gene al, i seems easonable o expec ha deposi s will be mo e subs i u able
wi h CBDC han wi h consump ion, i.e. ψ>ϵ. Then, an inc ease in he CBDC sp ead
makes he demand elas ici y o deposi s (18) less nega i e in alue and, in u n,
dec eases he ma ginal benefi o deposi issuance. The in ui ion is ha when i s
sp ead widens, CBDC becomes a compa a i ely expensi e sou ce o liquidi y and a
la ge ac ion o po en ial subs i u ion ou o deposi s will go o consump ion
(indica ed by a dec ease in s
and mo e weigh being pu on 1/ψ). The elas ici y o
demand mo es close o 1/ψ, which is smalle han 1/ϵ, and hus dec eases in absolu e
alue. The e o e, an inc ease in he CBDC sp ead makes he household’s demand o
deposi s less elas ic. Fo banks wi h ma ke powe , a less elas ic demand means ha
in o de o a ac addi ional deposi s om he household, he sp ead needs o be
lowe ed by mo e han be o e. Tha is, he ma ginal benefi o deposi issuance
dec eases. Gi en a fixed ma ginal cos , his implies ha he equilib ium deposi
sp ead inc eases. In o he wo ds, an inc ease in he CBDC sp ead is akin o gi ing
banks mo e ma ke powe . Banks ake ad an age o his and cha ge a highe sp ead
on deposi s in equilib ium.
As we alluded o p e iously, ma ke condi ions in he deposi ma ke also play a
cen al ole. I deposi s a diffe en banks a e pe ec subs i u es o he deposi
544 H. Chen e al.
ma ke is pe ec ly dispe sed, he equilib ium deposi sp ead is de e mined wi hou
he influence o he CBDC sp ead. I he household does no diffe en ia e be ween
banks, each indi idual bank’s choice o how much deposi s o issue does no ma e
o he equilib ium sp ead, which will equal he ma ginal cos o deposi issuance
(20): he ma ke is compe i i e. Simila ly, i he deposi ma ke is pe ec ly dispe sed
and he only sou ce o ma ke powe is diffe en ia ion, he impac o each indi idual
bank’s sp ead on he agg ega e deposi sp ead goes o ze o. The deposi ma ke
becomes monopsonis ically compe i i e wi h a cons an ma kup o e ma ginal cos
solely depending on he subs i u abili y be ween banks, gi en by (21). In bo h cases,
he go e nmen canno use he CBDC sp ead o influence he banking sec o .
To sum up, when he go e nmen dec eases he CBDC a e and widens he CBDC
sp ead, i di ec ly inc eases he household’s a e age cos o liquidi y and affec s
alloca ion. Mo eo e , a highe CBDC sp ead inc eases he sp ead on bank deposi s,
p o ided ha banks ha e sufficien ma ke powe , which aises he household’s cos
o liquidi y u he . The ansmission o he CBDC a e h ough he banking sec o is
simila o he deposi channel o mone a y policy p oposed by D echsle e al. (2017).
The au ho s desc ibe a si ua ion in which he household holds cash issued by he
go e nmen and deposi s issued by banks wi h ma ke powe . Policy-make s can
induce an inc ease in he deposi sp ead by inc easing he household’s oppo uni y
cos o holding cash, cap u ed by he nominal in e es a e on isk- ee bonds. In ou
model, ins ead, he al e na i e o bank deposi s is CBDC. The go e nmen can
simila ly affec banks’deposi sp ead by changing he household’s oppo uni y cos
o holding his al e na i e, i.e. i s sp ead.
3.3 In e es on Rese es
While in e es a es on cen al bank ese es ha e adi ionally no been a pa ic-
ula ly salien policy ool ( o example, he Fed only s a ed o emune a e ese es in
2008), we b iefly analyse hem as a s and-in o mone a y policy ins umen s
di ec ed a he banking sec o : Since he ese e a e shock effec i ely inc eases he
banks’cos o deposi p o ision, we expec ha o he (unmodeled) mone a y policy
shocks ha affec consume s only h ough he banking sys em o be affec ed by
deposi ma ke powe in a quali a i ely simila way.
In pa icula , he in e es on ese es affec s he household’s a e age cos o
liquidi y only h ough i s impac on he deposi sp ead. Suppose he go e nmen
dec eases he ese e a e so ha he ese e sp ead inc eases.
3
Taking he fi s
de i a i e o he a e age cos o liquidi y wi h espec o he ese e sp ead, we ge
3Fo simplici y, we assume he e ha he CBDC sp ead is cons an .
CBDC Policy T ansmission 545
∂χz
+1
∂χ
+1
=γ1
ϵχn
+1
χz
+1
()
−1
ϵ∂χn
+1
∂χ
+1
.
No ice ha he ma ginal impac o he ese e sp ead is e y simila o he indi ec
effec o he CBDC sp ead in (27). This is no su p ising since bo h effec s wo k
h ough he banking sec o . The impac o he ese e sp ead is he p oduc o he
sha e o deposi s in he o al s ock o liquid, n
+1
/z
+1
, and he change in he deposi
sp ead caused by he change in he ese e sp ead, ∂χn
+1/∂χ
+1. Again, he mo e
impo an deposi s a e as a sou ce o liquidi y, he la ge is his effec . Bu , i s sign
and he magni ude also depend on how he banking sec o esponds o an inc easing
ese e sp ead, ∂χn
+1/∂χ
+1. This key e m is in u n de e mined by how banks eac o
he highe cos o deposi p o ision induced by χ and depend o wha ex en
po en ial ou flows o CBDC a e conside ed in banks’deposi a e se ing.
4 Calib a ion
To gauge he impo ance o deposi ma ke concen a ion o he efficacy o a CBDC
a e as a policy ins umen , we calib a e ou model o conduc a ious nume ical
exe cises below. A de ailed desc ip ion o ou p ocedu e is p o ided in Appendix A.5.
A pe iod is in e p e ed as a qua e . Following Niepel (2024), we adop he case
wi h a monopsonis bank as a benchma k bu will compa e i wi h he case N=3
below: Gi en he symme ic banks, his alue implies a Hi schman-He findahl-Index
(HHI) o 1/3, close o he a e age coun y-le el HHI o 0.35 es ima ed by D echsle e al.
(2017) o he U.S. o e he pe iod om 1994 o 2013.
We s a wi h exogenously se ing se e al pa ame e s o alues om he li e -
a u e: In line wi h s anda d con en ion, we choose he household’s isk a e sion
pa ame e σ o be 2, he capi al sha e α o be 1/3 and he dep ecia ion a e o be 2.5 %.
We no malize l=1/3. Addi ionally, we ollow Bacche a and Pe azzi (2022) and
assume o ou benchma k exe cises ha CBDC is designed so ha i s elas ici y o
subs i u ion wi h espec o he deposi agg ega e is ϵ= 1/6. Gi en he unce ain y
abou whe he his will be he p ac ically ele an magni ude o any ac ual CBDC,
we also conside diffe en alues o ϵin he obus ness exe cises in Sec ion 7.
We also exogenously se he ini ial s eady s a e’s policy: Fi s ly, we assume ha
he cen al bank chooses o pay a nominal in e es a e o 0 on he CBDC, i.e. i has he
same e u n as cash. This is in line wi h many cen al seeming eluc an o emu-
ne a e a po en ial CBDC, and, assuming 2 % end infla ion, implies a eal annual
g oss e u n o 0.98 and Rm= 0.981/4. We u he mo e assume he annual g oss e u n
on ese es o amoun o R = 0.991/4, implying a nominal ese e a e o 1 % annually:
546 H. Chen e al.
The Fed s a ed o pay nominal in e es a es on ese es only in 2008 and hey
a e aged oughly 1 % in he ime since.
Now, o be able o clea ly iden i y he effec o deposi ma ke concen a ion in
ou model, we es ic he model e sions wi h N= 1 and N= 3 o be iden ical in all
o he dimensions: In pa icula , he CBDC is designed so ha agg ega e CBDC hold-
ings amoun o jus 12 % o deposi holdings while we induce he consump ion
eloci y c/z o be equal o 1.2. The o me a ge eflec s ha in many ju isdic ions
conside ing he implemen a ion o a CBDC, policymake s seem eluc an o induce
subs an ial disin e media ion o he banking sec o .
4
I also implies ha he o al
amoun o CBDC held is in line wi h jus physical cu ency being eplaced, which
ypically amoun ed o app ox. 12 % o agg ega e deposi holdings in he pos -wa US.
5
In con as , he c/z a ge ensu es ha e en a e he in oduc ion o a CBDC, he
o e all liquidi y eloci y is simila o cu en le els.
6
The a ge s a e achie ed by
se ing γ= 0.5938 and =0.0252. Finally, as in Niepel (2024), we aim o induce a
deposi ma kdown o 1.5, which we also induce by choosing ψacco dingly in he N=1
e sion. To achie e he same in he model e sion wi h N= 3, we addi ionally use he
pa ame e ηgo e ning he subs i u abili y be ween diffe en banks’deposi s. This
esul s in ψ= 0.3774 and η= 0.33, espec i ely.
7
No e ha he e ψ<σ, which implies consump ion and liquidi y se ices o be
subs i u es. This calib a ion esul is in line wi h Niepel (2024), who calib a es a e y
simila alue o ψ. While subs i u abili y be ween liquidi y se ices and con-
sump ion may seem coun e in ui i e i he o me a e aken o solely ep esen
ansac ion se ices, i is pe haps less so i one conside s addi ional benefi s o
liquidi y. Fo example, liquid asse s may also be use ul o insu ing agains idio-
sync a ic isk, as o example in Hugge (1993).
8
4Fo example, Fede al Rese e Go e no Michelle Bowman oiced ela ed conce ns, s a ing ha ”a
CBDC, i no p ope ly designed, could dis up he banking sys em and lead o disin e media ion,
po en ially ha ming consume s and businesses, and could p esen b oade financial s abili y isks”
(Bowman 2023).
5This s a emen is based on he se ies MBCURRCIR and DPSACBW027SBOG om FRED.
6Acco ding o FRED (Se ies: M2V), M2 eloci y in he US is ypically be ween 1.4 and 1.8, while
agg ega e consump ion is usually a ound 60–70 % o ou pu . Thus, in he cu en si ua ion wi hou
CBDC, a c/za ound 1.2 seems easonable.
7No e ha wi h N= 1, he pa ame e ηplays no ole.
8The calib a ion o ψis limi ed by he pa ame e es ic ions on he class o u ili y unc ions used by
us and Niepel (2024). Fo he model o ea u e a well-defined deposi sp ead in he monopsonis ic
case (N= 1), i is necessa y ha ψ< 1 (see e.g., equa ion (9) in Niepel (2024)). Al hough comple-
men a i y could s ill be achie ed by se ing 1 > ψ>σ, we chose o p io i ize ha ing a alue o he
in e empo al elas ici y o subs i u ion ha is in line wi h he s anda d con en ion σ≥1 in he
li e a u e. Besides, subs i u abili y is also necessa y o he model o con o m wi h he con en ional
wisdom ha highe policy a es educe agg ega e consump ion demand.
CBDC Policy T ansmission 547
Rega ding he cos s o p o iding diffe en o ms o liquidi y, Niepel (2024)
discusses a ious e idence sugges ing ha he annual cos o deposi - and ese e
p o ision may amoun o up o 1.2 % and 0.05 %, espec i ely. We hus choose
ω= 0.003 and ρ= 1.3 ×10−4 o ou qua e ly calib a ion. We u he mo e es ic
μ=ω+ρ, implying ha he cos s o p o iding CBDC a e equi alen o he go e nmen
ope a ing a na ow bank. Again ollowing Niepel (2024), we se φ= 1.5: No e ha his
pa ame e will effec i ely only ma e o exe cises wi h changing ese e a es, as
we always choose ϕ o induce a ese e- o-deposi a io o 0.1945, he midpoin o he
ange conside ed in said pape . This is achie ed by ϕ= 0.0021. Table 1 p esen s he ull
lis o calib a ed pa ame e s.
5 Sho -Run Analysis
A med wi h he calib a ed model, we can now assess he implica ions o deposi
ma ke powe o he CBDC a e as a policy ool, and, in pa icula , i s ole in shaping
Table :Baseline calib a ion.
Pa ame e Value Sou ce/Ta ge
Household
β.–/Annual R =%
σS anda d
l
/No maliza ion
ν. c/z=.
ψ. See ex
ϵ/Bacche a and Pe azzi ()
η. See ex
γ. m/n=.
Banks
ω. Niepel ()
φ.Niepel ()
ϕ. ζ=. (Niepel )
Fi ms
α/S anda d
δ. S anda d
Go e nmen
ρ.×−Niepel ()
μ. μ=ω+ρ
R ./% nom. e u n
Rm./% nom. e u n
548 H. Chen e al.
he ansmission o policy h ough he di ec and indi ec effec s ou lined abo e. We
s a by analyzing o wha ex en i ma e s i a cen al bank aims o use he CBDC
a e as a ool o influence business cycle fluc ua ions, o which we compu e line-
a ized impulse esponse unc ions (IRFs) o he economy o shocks o he CBDC and
ese e a es unde he diffe en assump ions on deposi ma ke powe (i.e. he cases
wi h N= 1 and N= 3). No e ha acco ding o he Blancha d–Kahn c i e ion, all model
e sions ea u e locally unique equilib ia.
5.1 Policy Shocks
When analyzing he impac o a CBDC a e shock nume ically below, we assume i o
ollow a log AR (1) p ocess
log Rm
+1
()
=(1−ρm)log(Rm)+ρmlog Rm
()
+em
,
whe e ρmis he pe sis ence pa ame e , Rmis he s eady s a e CBDC a e, and em
is he
exogenous shock. The exogenous shock is non-ze o in he fi s pe iod o he simu-
la ion and e u ns o ze o a e wa d. In o de o p ope ly isola e he effec , when
analyzing he CBDC a e, we assume ha he ese e a e is se so ha he ese e
sp ead is cons an a i s s eady s a e le el, i.e. i ulfills
R
+1=R
+1(βR ),
whe e R is he s eady s a e ese e a e. When we discuss he case o he ese e
a es, equi alen assump ions a e made, effec i ely in e changing he p ocesses o
R and Rm.
5.2 Impulse Responses
5.2.1 Response o a CBDC Ra e Shock
Figu e 1 shows he IRFs, as de ia ions om he non-s ochas ic s eady s a e, o a
nega i e 10 basis poin s shock o he qua e ly CBDC a e. Na u ally, he dec ease in
he CBDC a e immedia ely widens he CBDC sp ead by essen ially he same
magni ude as he agg ega e capi al s ock and he isk- ee a e changes li le. The
inc easing CBDC sp ead aises he household’s a e age cos o liquidi y in bo h he
baseline N= 1 and he al e na i e N= 3 cases so ha households choose o enjoy less
liquidi y se ices. This dec eases he household’s demand o liquidi y se ices bu
inc eases he household’s cu en ma ginal u ili y o consump ion, eflec ed in a
highe Ωc
+1. This is due o ou calib a ion ea u ing ψ<σ. In o he wo ds, he
CBDC Policy T ansmission 549
household’s oppo uni y cos o sa ing, in u ili y e ms, goes up. The household is
incen i ized o sa e less and inc ease cu en consump ion.
9
O e all, he effec o he
cu is no o e ly s ong in ei he case, eflec ing i s size and he assumed scena io o
limi ed CBDC adop ion.
Ne e heless, he e is a subs an ial diffe ence in he ela i e magni udes o he
esponses, wi h e.g. he o e all impac o he shock on agg ega e consump ion being
almos wice as la ge in he baseline case wi h N= 1. As ou lined abo e, in ou model,
he eal effec s o a CBDC a e a e ansmi ed ia i s effec on he liquidi y cos χzand
he impac o he CBDC sp ead on his e m can be decomposed in o a di ec effec
and an indi ec effec , shown in (27). Now, Figu e 2 displays his decomposi ion o he
esponses o he liquidi y cos : The g een dashed lines show he di ec effec s and he
ed solid lines show he indi ec effec s. The sum o he lines equals he o iginal
impulse esponses o he cos o liquidi y in Figu e 1.
Figu e 1: Impulse esponses o 10 basis poin s dec ease in CBDC a e. Figu e shows he impulse
esponses o key economic a iables o a 10 basis poin s dec ease in he CBDC a e unde wo diffe en
ma ke s uc u es (monopsony N= 1 and oligopsony N= 3).
9No e ha in he absence o a labo supply ma gin, he esponse o ou pu is de e mined by he
esponse o he capi al shock, implying ha he a e shocks do no induce he posi i e como emen o
consump ion and ou pu ypical o e.g. New Keynesian models.
550 H. Chen e al.
ha banks’deposi sp eads a e kep fixed a he le el calib a ed in he baseline
s eady s a e wi h Rm= 0.98: The diffe ence be ween he ue and coun e ac ual CE
gains hen isola es he con ibu ion o he indi ec effec s o wel a e.
15
Figu e 6 illus a es hese CE wel a e effec s o he by now amilia cases o a
monopsonis ic bank (N= 1) and he oligopsonis ic en i onmen wi h h ee banks
(N= 3). Clea ly, he model sugges s ha paying a highe eal e u n on CBDC has
wel a e benefi s, which eflec s he esul s by Niepel (2024) who finds CBDC o be a
e y efficien means o liquidi y p o ision: As he highe a e on CBDC induces he
ep esen a i e agen o hold ela i ely mo e o i , he agg ega e cos s o liquidi y
p o ision a e educed. O e all, he con ibu ion o he indi ec effec o agg ega e
wel a e is subs an ial and can accoun o almos a hi d o he o e all CE gains wi h
N= 1. Wi h N= 3, he sha e is smalle bu s ill no iceable. O cou se, his is again
because wi h diffe en ia ion being he mo e impo an sou ce o deposi ma ke
powe , he smalle banks eac less o he dec easing CBDC sp eads.
Figu e 5: Deposi sp ead agains diffe en CBDC a e changes. Figu e shows how he deposi sp ead
esponds o diffe en CBDC a es. The esponses a e shown o wo ma ke s uc u es (monopsony N=1
and oligopsony N= 3).
15 No e ha his decomposi ion diffe s om he one used in he p e ious Sec ion 5, which applied o
he cos o liquidi y χz.
CBDC Policy T ansmission 557
Conside ing he esul s abo e, i seems ha indi ec effec s h ough he
banking sec o can be quan i a i ely impo an o he wel a e gains a cen al
bank can achie e by paying posi i e in e es on CBDC. This is pa icula ly he case
i deposi ma ke powe isde e minedbyconcen a ion and ei e a es he poin
ha he sou ce o bank ma ke powe is an impo an de e minan o he model
effec s o CBDC policy. In e es ingly, gi en ha wel a e gains a e highe o N=1
han o N= 3, he exe cise abo e also sugges s ha o gi en policy a es, mo ing
om low o high banking concen a ion can po en ially inc ease agg ega e effi-
ciency as he diffe ing deposi a es change he composi ion o he agg ega e
liquidi y mix.
7 Robus ness Tes s
Whilewedisciplinemos pa ame e so ou modelinlinewi h heli e a u e,akey
unce ain y is he subs i u abili y be ween CBDC and deposi s, cap u ed by ϵ.Since
Figu e 6: CE wel a e gain e sus Coun e ac ual wel a e gain. Figu e compa es he consump ion-
equi alen (CE) wel a e gains and coun e ac ual wel a e gains gi en diffe en CBDC a es. The
compa isons a e shown o wo ma ke s uc u es (monopsony N= 1 and oligopsony N= 3).
558 H. Chen e al.
o mos coun ies, CBDC s ill la gely emains a heo e ical possibili y, we a e le
o specula e ega ding some aspec s o he ela ionship. The e o e, we es he
obus ness o he main insigh s in ou pape by changing he subs i u abili y
be ween CBDC and deposi s. The main specifica ions assume a “medium”deg ee o
subs i u abili y be ween CBDC and deposi s, i.e. ϵ= 1/6, ollowing Bacche a and
Pe azzi (2022). The es is hen changing he deg ee o subs i u abili y o ϵ= 1/4 and
ϵ= 1/10 and epea ing he exe cises abo e: No e ha his en ails e-calib a ing
pa ame e s such as ψand γ o achie e he same s eady s a e a ge s as in he
baseline model, ensu ing ha he esul ing diffe ences eflec only he diffe ing
choice o ϵ.
7.1 Sho -Run Analysis
Figu es A1 and A2 in Appendix C show he impulse esponses o a 10 basis poin s
dec ease in he CBDC a e wi h he al e na i e specifica ions desc ibed abo e. We see
ha he main akeaways om Sec ion 5 s ill s and: An inc ease in he CBDC sp ead
inc eases consump ion and lowe s capi al in es men . The ex en o which deposi
ma ke powe is shaped by concen a ion s ill has a no iceable impac : Highe
ma ke concen a ion no iceably amplifies he impulse esponses, al hough bo h he
agg ega e impac o he shock as well as he ela i e con ibu ion o concen a ion
dec ease (inc ease) o highe (lowe ) ϵ. Na u ally, i he ep esen a i e agen finds i
ha de o subs i u e deposi s wi h CBDC, banks need o be less conce ned wi h he
impac o i s e u n Rmon deposi demand, which dampens he indi ec effec and in
u n he s eng h o he o e all esponse. I he e is hus less scope o e all o indi ec
effec s due o lowe subs i u abili y, he impac o deposi ma ke concen a ion on
he effec s o he shock is lowe .
We also conside how he esponse o he ese e shock is shaped by η, o which
Appendix Figu es A3 and A4 display he IRFs o again he cases ϵ= 1/4 and ϵ= 1/10,
espec i ely. In con as o CBDC, lowe (highe ) subs i u abili y now inc eases
(dec eases) he agg ega e effec s o he shock bu also dampens (s eng hens) he
impac o concen a ion N. Wi h less scope o subs i u ion be ween CBDC and
deposi s, banks ha e mo e scope o eac o hei inc eases in cos s due o he lowe
ese e a es, ampli ying hei agg ega e effec s.
O e all, he abo e esul s imply ha he a en ion cen al banks will need o pay
o deposi ma ke powe and i s sou ces is going o depend on how hey design hei
po en ial CBDC: In case i becomes a he easy o consume s o swi ch be ween
deposi s and he digi al cu ency, i is going o ma e mo e o he effec s o
mone a y policy.
CBDC Policy T ansmission 559
7.2 Long-Run Analysis
Simila conside a ions as o he sho - un analysis abo e apply o he long- un
effec s o he CBDC a e: Figu es A5 and A6 illus a e he CE wel a e gains o diffe en
le els o he elas ici y o subs i u ion be ween CBDC and bank deposi s o he cases
o a monopsonis ic bank (N= 1) and he oligopsonis ic se ing (N= 3), espec i ely. In
ei he case, a highe elas ici y o subs i u ion (lowe alue o ϵ) co esponds o
o e all s onge wel a e effec s o he CBDC a e in he long un, as i ends up
affec ing deposi sp eads and hence he agg ega e liquidi y mix mo e.
8 Conclusions
This pape has s udied he ansmission and effec s o in e es a es on CBDC, a
po en ial new policy le e a cen al banke s’disposal. The analysis ocused on bank
ma ke powe and highligh ed ha i he deposi ma ke is concen a ed, a CBDC a e
will in gene al affec he agg ega e economy h ough bo h di ec and indi ec effec s:
A highe e u n o holding he digi al cu ency no only affec s households’sa ing-
and po olio decisions by i sel , bu also incen i izes he non-compe i i e banks o
adjus hei deposi sp eads, which may in u n s eng hen he agg ega e effec o he
policy inno a ion.
Ou simple heo e ical model implies ha he la e indi ec channel has he
po en ial o subs an ially ampli y he gene al equilib ium consequences o CBDC
policy, bu only i concen a ion is he key de e minan o deposi ma ke powe . I
deposi ma kdowns a e ins ead due o banks p o iding diffe en ia ed liquidi y
se ices and concen a ion is lowe , he di ec channel ypically domina es. We
es ablish hese insigh s bo h o he sho un, by s udying policy shocks, as well as
he long un, by compa ing s eady s a es. Rega ding he la e , we also ex end he
insigh s o Niepel (2024) on op imal policy, which depends on deposi ma ke con-
cen a ion and he po en ial scope o indi ec effec s pa icula ly i he go e nmen
canno subsidize banks di ec ly. Mo eo e , deposi ma ke concen a ion is also
ele an o o he mone a y policy ools di ec ed a he banking sec o (such as
ese e a es) bu ends o affec hem in he opposi e way, weakening hei efficacy.
We iew ou findings as ele an o bo h p ac ice and heo y: Rega ding he
o me , i means ha using in e es a es on CBDC as a policy ool necessi a es a
de ailed unde s anding o he deposi ma ke and compe i ion he eon. Fo modeling,
in u n, hey e eal ha specific assump ions on bank compe i ion can impo an ly
affec esul s o models o CBDC- and mone a y policy, e en i consis ency wi h he
same agg ega e momen s is ensu ed. I hus seems in e es ing o u u e esea ch o
560 H. Chen e al.
explo e his u he by inco po a ing oligopsonis ic banks in o quan i a i e models o
CBDC allowing o iche ic ions and shocks.
Appendix A De i a ions
A.1 Households
The household, aking p ices, p ofi s and axes as gi en, sol es
max
c ,kh
+1,m +1,ni
+1
{}
∞
=0
E0∑
∞
=0
β u(c ,z +1)
s. .c +kh
+1+m +1+1
N∑
N
i=1
ni
+1+τ =w l+π +kh
Rk
+m Rm
+1
N∑
N
i=1
ni
Rn,i
,
kh
+1,m +1,ni
+1≥0.
Focusing on he in e io solu ion, he fi s -o de condi ions wi h espec o
capi al, CBDC and deposi s a e
kh
+1:1=E Λ +1Rk
+1
[] (A.1)
m +1:
uz, zm, +1
uc,
=χm
+1(A.2)
ni
+1:
uz, zni, +1
uc,
=1
Nχn,i
+1(A.3)
whe e a, deno es he pa ial de i a i e o unc ion wi h espec o i s a gumen a,
Λ
+1
is he household’s s ochas ic discoun ac o
Λ +1=βuc, +1
uc,
,
χm
+1and χn,i
+1a e he CBDC sp ead and deposi sp ead a bank i, espec i ely,
χm
+1=1−Rm
+1
R
+1
,χn,i
+1=1−Rn,i
+1
R
+1
,
and he isk- ee a e is defined as
R
+1=1
E [Λ +1].
CBDC Policy T ansmission 561
A.1.1 Demand o Indi idual Bank Deposi s
Household’sfi s -o de condi ion (A.3) wi h espec o deposi s a any bank ican be
w i en as
uz, zn, +1
uc,
n +1
ni
+1
()
η
=χn,i
+1.(A.4)
Since he las exp ession holds o any bank, i means ha o any wo banks i
and j
χn,i
+1
ni
+1
n +1
()
η
=χn,j
+1
nj
+1
n +1
()
η
,
om which we find he demand o bank deposi s j
nj
+1=χn,i
+1
χn,j
+1
()
1
η
ni
+1.(A.5)
Le Tdeno e he sum o deposi sp eads ha he household incu s, and inse
(A.5) in o he exp ession,
T=1
N∑
N
i=1
ni
+1χn,i
+1=1
N∑
N
j=1
χn,i
+1
χn,j
+1
()
1
η
ni
+1χn,j
+1,
o find an exp ession o ni
+1
ni
+1=NT χn,i
+1
()
−1
η
∑N
j=1χn,j
+1
()
η−1
η
.(A.6)
We plug equa ion (A.6) in o he defini ion o agg ega e deposi , gi en by (1),
n +1=Nη
η−1T∑
N
i=1
χn,i
+1
()
η−1
η
()
η
1−η
.(A.7)
Le χn
+1be he sp ead associa ed wi h one uni o agg ega e deposi , n
+1
.By
se ing n
+1
= 1, we see om equa ion (A.7) ha
χn
+1=1
N∑
N
i=1
χn,i
+1
()
η−1
η
()
η
η−1
.(A.8)
Gi en equa ion (A.8), we see ha equa ion (A.6) can be w i en as
562 H. Chen e al.
ni
+1=T
χn
+1
χn,i
+1
χn
+1
()
−1
η
,(A.9)
and inse ing he esul ing exp ession in o (1), we ge
n +1=1
N∑
N
i=1
T
χn
+1
χn,i
+1
χn
+1
()
−1
η
⎛
⎝⎞
⎠1−η
⎛
⎜
⎜
⎝⎞
⎟
⎟
⎠
1
1−η
=T
χn
+1
.(A.10)
Combining he exp ession o Tand (A.10) we see ha
T=1
N∑
N
i=1
ni
+1χn,i
+1=n +1χn
+1.(A.11)
Las ly, inse ing equa ion (A.11) in o (A.9), we ge he household’s demand o
deposi s a bank i
ni
+1=n +1
χn,i
+1
χn
+1
()
−1
η
.(A.12)
Combining he household’s demand schedule wi h he fi s -o de condi ion
(A.4), we see ha (A.4) can be exp essed as
uz, zn, +1
uc,
=χn,i
+1
n +1
ni
+1
()
−η
=χn
+1.(A.13)
A.1.2 Op imali y Condi ions
Gi en he unc ional o m assump ions, he household’sfi s -o de condi ions (A.2)
and (A.13) become
m +1:
z−ψ
+1
(1− )c−ψ
(1−γ)z +1
m +1
()
ϵ
=χm
+1(A.14)
ni
+1:
z−ψ
+1
(1− )c−ψ
γz +1
n +1
()
ϵ
=χn
+1.(A.15)
We combine (A.14) and (A.15) o ge he a io
m +1
n +1
=(1−γ)χn
+1
γχm
+1
()
1
ϵ
.(A.16)
We plug equa ion (A.16) in o CES unc ion o z
+1
and sol e o he a io o z
+1
o n
+1
CBDC Policy T ansmission 563
z +1
n +1
=
(1−γ)χn
+1
()
1−ϵ
()
1
ϵ+γχ
m
+1
()
1−ϵ
()
1
ϵ
()
ϵ
1−ϵ
γχm
+1
⎛
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎝⎞
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎠
1
ϵ
.(A.17)
Inse ing (A.17) in o equa ion (A.15) and sol e o z
+1
, we ge he household’s
op imal demand o liquidi y
z +1=c
1−
1
χz
+1
()
1
ψ
,(A.18)
whe e χz
+1is he a e age cos o liquidi y aced by he household
χz
+1=χm
+1χn
+1
(1−γ)1
ϵχn
+1
()
1−ϵ
ϵ+γ1
ϵχm
+1
()
1−ϵ
ϵ
()
ϵ
1−ϵ
.
Gi en household’s op imal demand o z
+1
,wefind he household’s demand o
m
+1
and n
+1
m +1=z +1(1−γ)χz
+1
χm
+1
()
1
ϵ
n +1=z +1γχz
+1
χn
+1
()
1
ϵ.
(A.19)
Plugging op imal z
+1
, gi en by (A.18), in o he fi s -o de condi ion (A.1), we find
he household’s Eule equa ion
c−σ
Ωc
=βE Rk
+1c−σ
+1Ωc
+1
[] (A.20)
whe e Ωc
is gi en by
Ωc
=(1− )1−σ
1−ψ1+
1−
()
1
ψχz
+1
()
1−1
ψ
()
ψ−σ
1−ψ
.
A.2 Banks
The da e- p og am o a ypical bank is
max
i
+1,Rn,i
+1
−ni
+1νi
+E Λ +1ki
+1Rk
+1+ i
+1R
+1−ni
+1Rn,i
+1
()[]
s. .ni
+1=n +1
χn,i
+1
χn
+1
()
−1
η
ki
+1=ni
+1− i
+1,
whe e
564 H. Chen e al.
νi
ζi
+1
()
=ω+ϕζ
i
+1
()
1−φ,ζi
+1= i
+1
ni
+1
.
The fi s -o de condi ions o bank iwi h espec o i s deposi a e and ese e
holdings a e, espec i ely,
Rn,i
+1:χn,i
+1+χn,i
+1
en,i
+1
=νi
−νi
ζ, ζi
+1(A.21)
i
+1:−νi
ζ, =χ
+1,(A.22)
whe e χ
+1=1−R
+1/R
+1and en,i
+1deno es he elas ici y o demand o deposi s a
bank iwi h espec o i s deposi sp ead, χn,i
+1, which in a symme ic indus y equi-
lib ium can be shown o be
en,i
+1=∂ni
+1
∂χn,i
+1
χn,i
+1
ni
+1
.
Gi en unc ional o m assump ions, he fi s -o de condi ion (A.21) becomes
χn,i
+11+1
en,i
+1
()
=ω+φϕ ζi
+1
()
1−φ,
whe e bank i’s op imal ese es- o-deposi s a io is gi en by he fi s -o de condi ion
(A.22)
ζi
+1=χ
+1
ϕ(φ−1)
()
−1
φ
.
To find he demand elas ici y, en,i
+1,wediffe en ia e he household’s demand o
deposi a bank i, equa ion (A.12) wi h espec o χn,i
+1and mul iply i wi h he a io
χn,i
+1/ni
+1
en,i
+1=−
1
η
n +1
χn,i
+1
χn
+1
χn,i
+1
()
1
η
+1
η
n +1
χn
+1
χn
+1
χn,i
+1
()
1
η∂χn
+1
∂χn,i
+1
+χn
+1
χn,i
+1
()
1
η∂n +1
∂χn
+1
∂χn
+1
∂χn,i
+1
⎛
⎝⎞
⎠χn,i
+1
ni
+1
=−1
η+1
η
χn,i
+1
χn
+1
∂χn
+1
∂χn,i
+1
+χn,i
+1
χn
+1
()
1
ηχn,i
+1
nn,i
+1
∂n +1
∂χn
+1
∂χn
+1
∂χn,i
+1
(A.23)
In a symme ic indus y equilib ium, whe e χn,i
+1=χn,j
+1and ni
+1=nj
+1 o any
bank iand j,
CBDC Policy T ansmission 565
χn
+1=1
N∑
N
i=1
χn,i
+1
()
η−1
η
()
η
η−1
=χn,i
+1
n +1=1
N∑
N
i=1
ni
+1
()
1−η
()
1
1−η
=ni
+1
Then, equa ion (A.23) educes o
en,i
+1=1
N
∂n +1
∂χn
+1
χn
+1
n +1
()
−1−1
N
()
1
η.
To find he agg ega e demand elas ici y, we diffe en ia e household’sop imal
deposi demand, equa ion (A.19), wi h espec o he liquidi y p emium on de-
posi s, χn
+1,
∂n +1
∂χn
+1
=∂z +1
∂χz
+1
∂χz
+1
∂χn
+1
γχz
+1
χn
+1
()
1
ϵ
+z +1γ
ϵχn
+1
∂χz
+1
∂χn
+1
γχz
+1
χn
+1
()
1−ϵ
ϵ
−z +1γχz
+1
ϵχ
n
+1
()
2
γχz
+1
χn
+1
()
1−ϵ
ϵ
and mul iply he las exp ession wi h he a io χn
+1/n +1
∂n +1
∂χn
+1
χn
+1
n +1
=−1
ψγ1
ϵχz
+1
χn
+1
()
1−ϵ
ϵ
−1
ϵ(1−γ)1
ϵχz
+1
χm
+1
()
1−ϵ
ϵ
.
Las ly, we w i e he op imali y condi ion as i applies o a ep esen a i e bank
(and d opping he indi idual supe sc ip i)
χn
+1+χn
+1
1
N−1−s
ψ−s
ϵ
()
−1−1
N
()
1
η
()
−1
=ω+φϕζ1−φ
+1,(A.24)
whe e
ζ +1=χ
+1
ϕ(φ−1)
()
−1
φ
(A.25)
and s
∈[0, 1] is
s =(1−γ)1
ϵχz
+1
χm
+1
()
1−ϵ
ϵ
.
A.3 Agg ega e Resou ce Cons ain
To find he agg ega e esou ce cons ain , we s a by inse ing o al p ofi , π
, in o
he household’s budge cons ain , imposing ma ke clea ing o labo and capi al
and ea anging
566 H. Chen e al.
Figu e A2: Highe CBDC-deposi s subs i u abili y ϵ=1
10. Figu e shows he impulse esponses o key
economic a iables o a 10 basis poin s dec ease in he CBDC a e when ϵ= 1/10. The esponses a e
shown o wo diffe en ma ke s uc u es (monopsony N= 1 and oligopsony N= 3).
Figu e A3: Lowe CBDC-deposi s subs i u abili y ϵ=1
4. Figu e shows he impulse esponses o key
economic a iables o a 10 basis poin s dec ease in he ese e a e when ϵ= 1/4. The esponses a e
shown o wo diffe en ma ke s uc u es (monopsony N= 1 and oligopsony N= 3).
CBDC Policy T ansmission 573
Figu e A4: Highe CBDC-deposi s subs i u abili y ϵ=1
10. Figu e shows he impulse esponses o key
economic a iables o a 10 basis poin s dec ease in he ese e a e when ϵ= 1/10. The esponses a e
shown o wo diffe en ma ke s uc u es (monopsony N= 1 and oligopsony N= 3).
Figu e A5: CE wel a e gain (loss) wi h diffe en scena ios (N= 1). Figu e shows he CE wel a e gains o
diffe en le els o subs i u abili y be ween CBDC and deposi s in a monopsonis ic bank (N= 1).
574 H. Chen e al.
Re e ences
Agu , I ai, Anil A i, and Gio anni Dell’A iccia. 2022. “Designing Cen al Bank Digi al Cu encies.”Jou nal o
Mone a y Economics 125: 62–79.
Andol a o, Da id. 2021. “Assessing he Impac o Cen al Bank Digi al Cu ency on P i a e Banks.”The
Economic Jou nal 131 (634): 525–40.
A lan ic Council. 2024. Cen al Bank Digi al Cu ency T acke . A ailable a : h ps://www.a lan iccouncil.
o g/cbdc acke (accessed May 10, 2024).
Aue , Raphael, Jon F os , Leona do Gambaco a, Cy il Monne , Ta a Rice, and Hyun Song Shin. 2022.
“Cen al Bank Digi al Cu encies: Mo i es, Economic Implica ions, and he Resea ch F on ie .”
Annual Re iew o Economics 14: 697–721.
Bacche a, Philippe, and Elena Pe azzi. 2022. CBDC as Impe ec Subs i u e o Bank Deposi s: A
Mac oeconomic Pe spec i e. Swiss Finance Ins i u e Resea ch Pape Se ies 21–81. Swiss Finance
Ins i u e.
Bank o In e na ional Se lemen . 2020. “Cen al Bank Digi al Cu encies: Founda ional P inciples and
Co e Fea u es.”Tech. ep. Bank o In e na ional Se lemen . A ailable a : h ps://www.bis.o g/publ/
o hp33.h m (accessed Ap il 15, 2022).
Bowman, Michelle W. 2023. Conside a ions o a Cen al Bank Digi al Cu ency. Speech p esen ed a he
Geo ge own Uni e si y McDonough School o Business, Psa os Cen e o Financial Ma ke s and
Policy. A ailable a : h ps://www. ede al ese e.go /newse en s/speech/bowman20230418a.h m
(accessed Decembe 1, 2023).
Chiu, Jona han, Seyed Mohammad eza Da oodalhosseini, Jane Jiang, and Yu Zhu. 2023. “Bank Ma ke
Powe and Cen al Bank Digi al Cu ency: Theo y and Quan i a i e Assessmen .”Jou nal o Poli ical
Economy 131 (5): 1213–48.
Choi, Michael, and Guillaume Roche eau. 2023. “A Model o Re ail Banking and he Deposi s Channel o
Mone a y Policy.”Jou nal o Mone a y Economics 139: 127–47.
Figu e A6: CE wel a e gain (loss) wi h diffe en scena ios (N= 3). Figu e shows he CE wel a e gains o
diffe en le els o subs i u abili y be ween CBDC and deposi s in an oligopsonis ic en i onmen (N= 3).
CBDC Policy T ansmission 575
Co bae, Dean, and Pablo D’E asmo. 2020. “Rising Bank Concen a ion.”Jou nal o Economic Dynamics and
Con ol 115: 103877.
De G auwe, Paul, and Yuemei Ji. 2023. The Ex ao dina y Gene osi y o Cen al Banks owa ds Banks:
Some Reflec ions on i s O igin. A ailable a : h ps://cep .o g/ oxeu/columns/ex ao dina y-
gene osi y-cen al-banks- owa ds-banks-some- eflexions-i s-o igin (accessed Janua y 5, 2024).
D echsle , I ama , Alexi Sa o , and Philipp Schnabl. 2017. “The Deposi s Channel o Mone a y Policy.”
Qua e ly Jou nal o Economics 132 (4): 1819–76.
Enge , Wal e , and Ben Siu-Cheong Fung. 2017. Cen al Bank Digi al Cu ency: Mo i a ions and Implica ions.
Bank o Canada S affDiscussion Pape 2017-16. Bank o Canada. A ailable a : h ps://www.
banko canada.ca/wp-con en /uploads/2017/11/sdp2017-16.pd
Fe nández- Villa e de, Jesús, Daniel Sanches, Linda Schilling, and Ha ald Uhlig. 2021. “Cen al Bank Digi al
Cu ency: Cen al Banking o All?”Re iew o Economic Dynamics 41: 225–42.
Ga a , Rodney, Jiaheng Yu, and Haoxiang Zhu. 2022. “How Cen al Bank Digi al Cu ency Design Choices
Impac Mone a y Policy Pass-Th ough and Ma ke Composi ion.”h ps://doi.o g/10.2139/ss n.
4004341.
Hugge , Ma k. 1993. “The Risk- ee Ra e in He e ogeneous-Agen Incomple e-Insu ance Economies.”
Jou nal o Economic Dynamics and Con ol 17 (5): 953–69.
Jiang, Jane Hua, and Yu Zhu. 2021. Mone a y Policy Pass-Th ough wi h Cen al Bank Digi al Cu ency. Bank o
Canada S affDiscussion Pape 2021-10. Bank o Canada.
Keis e , Todd, and Daniel Sanches. 2022. “Should Cen al Banks Issue Digi al Cu ency?”The Re iew o
Economic S udies 90 (1): 404–31.
Niepel , Di k. 2024. “Money and Banking wi h Rese es and CBDC.”The Jou nal o Finance 79 (4): 2505–52.
Piazzesi, Monika, and Ma in Schneide . 2022. C edi Lines, Bank Deposi s o CBDC? Compe i ion and
Efficiency in Mode n Paymen Sys ems. A ailable a : h ps://web.s an o d.edu/∼piazzesi/CBDC.pd
(accessed Sep embe 20, 2023).
Sid auski, Miguel. 1967. “Ra ional Choice and Pa e ns o G ow h in a Mone a y Economy.”The Ame ican
Economic Re iew 57 (2): 534–44.
Wang, Yi ei, Toni M. Whi ed, Yu eng Wu, and Kai ong Xiao. 2022. “Bank Ma ke Powe and Mone a y Policy
T ansmission: E idence om a S uc u al Es ima ion.”The Jou nal o Finance 77 (4): 2093–141.
Williamson, S ephen D. 2022. “Cen al Bank Digi al Cu ency and Fligh o Sa e y.”Jou nal o Economic
Dynamics and Con ol 142. The Economics o Digi al Cu encies, 104146. h ps://doi.o g/10.1016/j.jedc.
2021.104146.
576 H. Chen e al.
