A a-Ankomah, Richmond; Adjei-Man ey, Kwame; Amankwah, Aku o
A icle
Digi al inancial se ices and li elihood di e si ica ion in
u al Ghana
Cogen Economics & Finance
P o ided in Coope a ion wi h:
Taylo & F ancis G oup
Sugges ed Ci a ion: A a-Ankomah, Richmond; Adjei-Man ey, Kwame; Amankwah, Aku o (2024) :
Digi al inancial se ices and li elihood di e si ica ion in u al Ghana, Cogen Economics & Finance,
ISSN 2332-2039, Taylo & F ancis, Abingdon, Vol. 12, Iss. 1, pp. 1-20,
h ps://doi.o g/10.1080/23322039.2024.2330434
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/321457
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/
Cogen Economics & Finance
ISSN: 2332-2039 (Online) Jou nal homepage: www. and online.com/jou nals/oae 20
Digi al financial se ices and li elihood
di e sifica ion in u al Ghana
Richmond A a-Ankomah, Kwame Adjei-Man ey & Akuffo Amankwah
To ci e his a icle: Richmond A a-Ankomah, Kwame Adjei-Man ey & Akuffo Amankwah (2024)
Digi al financial se ices and li elihood di e sifica ion in u al Ghana, Cogen Economics &
Finance, 12:1, 2330434, DOI: 10.1080/23322039.2024.2330434
To link o his a icle: h ps://doi.o g/10.1080/23322039.2024.2330434
© 2024 The Au ho (s). Published by In o ma
UK Limi ed, ading as Taylo & F ancis
G oup
View supplemen a y ma e ial
Published online: 05 Ap 2024.
Submi you a icle o his jou nal
A icle iews: 2410
View ela ed a icles
View C ossma k da a
Ci ing a icles: 6 View ci ing a icles
Full Te ms & Condi ions o access and use can be ound a
h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode=oae 20
DEVELOPMENT ECONOMICS | RESEARCH ARTICLE
Digi al inancial se ices and li elihood di e si ica ion in u al Ghana
Richmond A a-Ankomah
a
, Kwame Adjei-Man ey
b
and Aku o Amankwah
c
a
Ins i u e o S a is ical, Social and Economic Resea ch, Uni e si y o Ghana, Legon, Ghana;
b
Depa men o Applied
Economics, Uni e si y o En i onmen and Sus ainable De elopmen , Somanya, Ghana;
c
Li ing S anda ds Measu emen
S udy, De elopmen Da a G oup, Washing on, DC, USA
ABSTRACT
The eme gence o digi al inancial sys ems, especially mobile money, has signi ican ly
changed he inancial se ices space in sub-Saha an A ica in ways ha p omo e inan-
cial inclusion among he poo , including hose in u al communi ies. Focusing on
mobile money, which is he main a enue o digi al inancial se ices o u al house-
holds in Ghana, his s udy in es iga es whe he digi al inance a ec s u al house-
holds’li elihood di e si ica ion, using a na ionally ep esen a i e da a on Ghana.
Employing se e al econome ic me hods, including ins umen al a iable echniques
o add ess po en ial endogenei y bias, he s udy inds mobile money o be posi i ely
associa ed wi h households’choice o non- a m business as a li elihood op ion as
well as hei engagemen in li es ock p oduc ion. Howe e , we ind a nega i e ela-
ionship be ween mobile money and di e si ica ion in c op p oduc ion including he
ex en o which households engaged in he c op sec o a e able o di e si y c op p o-
duc ion. The implica ion is ha when u al households ha e access o mobile money
se ices, hey p oduce ewe numbe o di e en c ops while explo ing non- a m
ac i i ies and li es ock p oduc ion as addi ional li elihood op ions. The esul s unde -
sco e he need o add ess cons ain s ha limi u al households’access o digi al
inancial ools, especially mobile money, in Ghana and in simila de eloping coun y
con ex s.
IMPACT STATEMENT
This s udy employs a na ionally ep esen a i e mic o-da a om u al Ghana o in es i-
ga e he e ec o digi al inance on li elihood di e si ica ion using se e al econome -
ic me hods including ins umen al a iables o add ess po en ial endogenei y bias.
The s udy ep esen s a signi ican a emp a add essing an impo an gap in he
exis ing li e a u e which is abou whe he digi al inance b ings any alue addi ion o
li elihood di e si ica ion e o s by u al households. The key inding is ha when u al
ag icul u al households ha e access o digi al inancial se ices h ough mobile
money, hey p oduce ewe di e en c ops, while explo ing non- a m ac i i ies and
li es ock p oduc ion as addi ional li elihood op ions. The inding implies ha access
o digi al inance may in luence u al households’abili y o ealloca e p oduc i e
esou ces o a eas ha yield highe e u ns o hei li elihoods and/o mi iga e he
isk o income loss. Expanding upon his inding wi h an explo a ion o he key co e-
la es o di e si ica ion and access o digi al inance, he s udy shows ha add essing
cons ain s o access o digi al inance o mobile money (pa icula ly, he limi ed
mobile phone connec i i y, elec ici y access o supply challenges, and low educa ion
o digi al illi e acy in u al a eas) is impo an o deepening li elihood di e si ica ion
in u al Ghana.
ARTICLE HISTORY
Recei ed 7 No embe 2023
Re ised 6 Feb ua y 2024
Accep ed 10 Ma ch 2024
KEYWORDS
Digi al inance; mobile
money; li elihood
di e si ica ion; c ops;
li es ock; non- a m
en e p ises
REVIEWING EDITOR
Goodness Aye, Uni e si y o
Ag icul u e, Maku di Benue
S a e, NIGERIA
SUBJECTS
Economics and
De elopmen ; Economics;
De elopmen S udies;
Finance; Ru al De elopmen
JEL CLASSIFICATIONS
G50; I31; Q00; O39
1. In oduc ion
The ad en o digi al inance (DF), such as mobile money, has signi ican ly changed he inancial se ices
space in sub-Saha an A ica (SSA), especially o he poo and he his o ically unbanked, allowing
CONTACT Richmond A a-Ankomah [email p o ec ed] Ins i u e o S a is ical, Social and Economic Resea ch, Uni e si y o Ghana,
Legon, Ghana
Supplemen al da a o his a icle can be accessed online a h ps://doi.o g/10.1080/23322039.2024.2330434.
ß2024 The Au ho (s). Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/4.0/), which
pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. The e ms on which his a icle has been
published allow he pos ing o he Accep ed Manusc ip in a eposi o y by he au ho (s) o wi h hei consen .
COGENT ECONOMICS & FINANCE
2024, VOL. 12, NO. 1, 2330434
h ps://doi.o g/10.1080/23322039.2024.2330434
coun ies o use inno a i e echnologies o leap og he inancial de elopmen s ages ha de eloped
coun ies wen h ough (A on, 2018; Ba a, 2013). Mobile money has become an impo an inno a i e
ool o acili a ing inancial inclusion and enhancing inancial ma ke pa icipa ion (Ahmad e al., 2020;
Amoah e al., 2020; Bongomin e al., 2018;N’d i & Kakinaka, 2020; Ouma e al., 2017) especially in SSA.
Acco ding o he 2021 Global Findex epo , SSA is leading he mobile money e olu ion, wi h he sha e
o adul s owning mobile money accoun s inc easing om 12% in 2014 o 33% in 2021, compa ed wi h
he global sha e o 2% and 10% in 2014 and 2021, espec i ely (Asli e al., 2022). In addi ion, di e ences
in inancial accoun owne ship in SSA ac oss weal h quin iles ha e been hal ed o e he pas decade,
d i en mainly by mobile money (Asli e al., 2022).
Aside om acili a ing inancial inclusion, he e is g owing empi ical e idence linking digi al inance
o a ious li elihood and wel a e ou comes (A a-Ankomah, 2022; Lee e al., 2021; Ouma e al., 2017;
Pep ah e al., 2020). Fo ins ance, in Bangladesh, Lee e al. (2021) ound ha mobile money inc eases
household consump ion, educes ex eme po e y and bo owing, and sho e-up sa ings. Simila ly,
Pep ah e al. (2020) ound a posi i e ela ionship be ween mobile money adop ion and household ou -
comes, including a m ou pu , consump ion expendi u e, and asse alue, using da a om h ee dis ic s
in Ghana. Ouma e al. (2017) obse ed ha access o digi al inance enhances sa ings beha io and he
ex en o sa ing among households in Kenya, Malawi, Uganda, and Zambia. Recen s udies using a
na ionally ep esen a i e da a on Ghana also obse ed a posi i e co ela ion be ween mobile money
accoun owne ship and access o c edi , heal h insu ance and emi ances (A a-Ankomah, 2022; A a-
Ankomah e al., 2024). In addi ion, ecen ad ances in digi al inancial echnology ha e allowed poo
households wi h mobile money accoun s o ecei e in e na ional emi ances (Ahmad e al., 2020;
Amoah e al., 2020). In a mo e ecen s udy, Okye e e al. (2024a) show ha adop ion o digi al inancial
se ices has posi i e e ec on he esilience capaci y o a m households in No he n Ghana.
O he exis ing e idence also sugges s a posi i e e ec o digi al inance on household in es men s
and p oduc ion ac i i ies. Speci ically, o ag icul u e, which is he main means o li elihood o he u al
poo (Da is e al., 2017), Mille (2019) documen s ou channels h ough which digi al inance could
impac he sec o ’s de elopmen : (1) sa ings o help smoo hen consump ion o e ime; (2) insu ance
agains possible c op loss; (3) paymen and ansac ions, including cash ans e s o u al households;
and (4) c edi o a me s. A ecen s udy by Okye e e al. (2024b) ound ha mobile money adop ion
inc eases a m pe o mance (including p oduc i i y) and wel a e o households in No he n Ghana. A.
Islam e al. (2018) ound ha mobile money inc eased small- and medium-scale en e p ise pu chases o
ixed asse s in Uganda, Tanzania, and Kenya, wi h educed ansac ion cos s, inc eased liquidi y, and
c edi wo hiness being he main enable s. Ahmad e al. (2020) and A on (2018) p o ide mo e de ailed
e iews o he ex an li e a u e on he mic o-economic implica ions o digi al inance.
Digi al inance has he po en ial o e olu ionize ag i- ood sys ems and li elihood di e si ica ion o
u al households (Finge , 2023), bu s udies in his a ea a e lacking. In pa icula , he ole o digi al
inance (mobile money) in in luencing u al households’li elihood di e si ica ion, and he ex en o o m
o di e si ica ion has emained la gely unexplo ed in he li e a u e especially hose on Ghana, al hough
many empi ical s udies ha e es ablished a linkage be ween mobile money and household wel a e
(Amoah e al., 2020; A a-Ankomah & Okye e, 2022; Jack & Su i, 2014; Lee e al., 2021; Mahama &
Maha jan, 2017;N’d i & Kakinaka, 2020; Pep ah e al., 2020), as no ed abo e. Also, his e idence gap
exis s in spi e o a g owing li e a u e on he d i e s o u al li elihood di e si ica ions (Dagunga e al.,
2022; Loison, 2015,2019; Ma in & Lo enzen, 2016) and he impo ance o di e si ica ion o household
economic wellbeing including esilience o a ious shocks (see Akaakohol & Aye, 2014; As aw e al.,
2019; Asmah, 2011; Dagunga e al., 2020; Mohammed e al., 2021). On he d i e s o li elihood di e si i-
ca ion, Dagunga e al. (2022), o example, ound a posi i e linkage be ween sa ings and you h income
di e si ica ion in i e cocoa g owing egions o Ghana. Also, access o inance has been highligh ed as
an impo an d i e o di e si ica ion in se e al o he s udies (see Ba ool & Jamil, 2019; Kha un & Roy,
2012; Loison, 2015,2019). Using panel da a, Loison (2019) del ed in o he egional and gende di e en-
ces as d i e s o li elihood di e si ica ion in wo ag icul u al egions o u al Kenya. While hese s udies
a e insigh ul, none explo ed he link be ween digi al inance and u al li elihood di e si ica ion.
Agains his backg ound, his s udy examines he linkage be ween digi al inance (mobile money) and
u al household li elihood di e si ica ion by ocusing on h ee impo an issues in Ghana. Fi s , we
2 R. ATTA-ANKOMAH ET AL.
examine he d i e s o mobile money adop ion among u al households, including he gende dimen-
sions o hese de e minan s. Second, we examine he e ec o mobile money on di e en li elihood
di e si ica ion op ions (c op p oduc ion, li es ock p oduc ion, and non- a m business ope a ions) while
explo ing whe he access o digi al inance is associa ed wi h a de-emphasis on mo e adi ional (ag icul-
u al) p oduc ion and emphasis on less adi ional al e na i es in he non- a m sec o . Thi d, we disag-
g ega e he sample by gende o he household head o explo e whe he mobile money adop ion has a
gende ed-di e en ial impac on li elihood di e si ica ion.
This s udy is i al om bo h policy and li e a u e pe spec i es. As no ed abo e, while s udies on he
mic o impac o mobile money ha e expanded ecen ly, none o he exis ing s udies explo e he linkage
be ween mobile money and u al li elihood di e si ica ion in Ghana. Thus, he cu en s udy will con ib-
u e uniquely o he g owing li e a u e on digi al inance and u al li elihood di e si ica ion in A ica. This
s udy’s policy ele ance emana es om he gende dimension i explo es, he di e en li elihood
op ions conside ed, and he ac ha u al households in Ghana engage in a po olio o li elihood ac i -
i ies (Da is e al., 2017). In addi ion, he go e nmen o Ghana is implemen ing a numbe o li elihood
imp o emen p og ams and policies o s eng hen mobile money se ice deli e y in he coun y, making
he cu en s udy ele an in ha ega d. The policy implica ions o his s udy a e ele an o o he SSA
coun ies wi h a con ex simila o ha o Ghana.
This s udy uses na ionally ep esen a i e mic o-da a om u al Ghana and employs econome ic
me hods (including ins umen al a iables) o con ol o he po en ial endogenei y be ween mobile
money and he li elihood di e si ica ion a iables. Speci ically, we used he pene a ion o mobile
phones a he household le el as an ins umen . We ind ha when u al ag icul u al households ha e
access o digi al inancial se ices h ough mobile money, hey p oduce ewe di e en c ops, while
explo ing non- a m ac i i ies and li es ock p oduc ion as addi ional li elihood op ions. In o he wo ds,
hey can di e si y in o less adi ional sou ces o li elihood. The implica ion is ha access o digi al
inance may in luence u al households’abili y o ealloca e p oduc i e esou ces o a eas ha yield
highe e u ns o hei li elihoods and/o mi iga e he isk o income loss. These e ec s pe ain o bo h
male-headed and emale-headed households, excep in he case o he de-emphasis on c op p oduc ion,
which mainly pe ains o male-headed households. These indings, oge he wi h he key esul s on
o he impo an co ela es o di e si ica ion and access o digi al inance, sugges ha cons ain s o
access o digi al inance o mobile money, such as limi ed mobile phone connec i i y, access o elec i-
ci y, and low educa ion o digi al illi e acy in u al a eas, need policy a en ion.
The es o he pape is o ganized as ollows. Sec ion 2 p o ides a b ie con ex ual backg ound o
mobile money and li elihood di e si ica ion in Ghana. Sec ion 3 p esen s he da a and empi ical s a egy
used o he s udy, and Sec ion 4 discusses he esul s. Finally, in Sec ion 5, we p esen he conclusions
and policy implica ions o he s udy.
2. Mobile money and li elihood ou comes in Ghana
Ghana is an in e es ing coun y o his s udy o se e al easons. Fi s , mobile phone subsc ip ions ha e
inc eased om 13 in 2005 o app oxima ely 123 pe 100 people by 2021 (Wo ld Bank Mic oda a
Ca alog, 2023). Simila ly, he numbe o indi iduals wi h mobile money accoun s soa ed om abou
350,000 in 2012, h ee yea s a e he launch o mobile money in Ghana o app oxima ely 17.1 million in
2020 (Dabalen & Mensah, 2023; Ma e n, 2018). In ac , Ghana is one o he coun ies in SSA ha has
seen a su ge in mobile money usage in ecen yea s, and he sha e o adul s aged 15þyea s wi h
mobile money accoun s inc eases by 47 pe cen age poin s be ween 2014 and 2021 (13% o 60%), while
he sha e o accoun s a inancial ins i u ions inc eased om 35% o 39% o e he same pe iod (Asli
e al., 2022; Ma e n, 2018). In addi ion, he join owne ship o bank accoun s and mobile money
accoun s has seen a subs an ial inc ease om 23% in 2014 o abou 30% in 2021, d i en mainly by
mobile money (Asli e al., 2022).
This su ge in mobile money usage in he coun y could be explained by he egula o y policies o he
Bank o Ghana and he inancial sec o in e ope abili y sys em in oduced join ly by he go e nmen o
Ghana, cen al and comme cial banks, and he Ghana Chambe o Telecommunica ions (Ox o d Business
G oup, 2018). In addi ion, he go e nmen o Ghana, in collabo a ion wi h o he de elopmen pa ne s,
COGENT ECONOMICS & FINANCE 3
has in oduced a numbe o policies, including he Na ional Financial Inclusion and De elopmen
S a egy, he Digi al Financial Se ices Policy, and he Cash-Li e Roadmap, in an e o o imp o e mobile
money se ices deli e y and expand he geog aphic each (Minis y o Finance, n.d.).
In 2022, he go e nmen o Ghana, h ough an ac o pa liamen , in oduced an elec onic le y (E-
Le y) o gene a e e enue, adop ing models ha appea o ha e wo ked in o he coun ies in he
egion. This policy was no popula among Ghanaians (Amoah e al., 2023), p obably because o i s less
p o-poo design. Cu en ly, he e a e h ee main elecommunica ion companies in Ghana (MTN Ghana,
Ai el/Tigo, and Voda one Ghana) p o iding mobile money se ices, wi h MTN Ghana ha ing he la ges
sha e.
Ag icul u e ends o be he majo sou ce o income, and hence, he li elihood o he u al poo in
SSA and Ghana in pa icula . Da is e al. (2017) show ha abou 44% o households in Ghana ha e a m-
ing as hei main sou ce o income. In addi ion, di e si ica ion in o non- a m income-gene a ing ac i i ies
is common among he u al poo in he coun y. Mo e han 23.5% o u al households in Ghana ha e
di e se income po olios (Da is e al., 2017). Despi e se ing as he main means o li elihood o he
u al poo and con ibu ing subs an ially o he coun y’s GDP, ag icul u e in Ghana is domina ed by
smallholde a me s, and is bese wi h a my iad o challenges, including low adop ion o sus ainable
ag icul u al p ac ices, low le el o comme cializa ion, and clima e change (As a o e al., 2022). Reso ing
o and di e si ying in o al e na i e sou ces o li elihood, such as non- a m businesses and li es ock p o-
duc ion, is e y common in u al Ghana (Mahama & Maha jan, 2017; Mahama & Nkegbe, 2021).
3. Me hodology
3.1. Da a
This s udy uses seconda y mic o da a om he se en h and mos ecen ound o he Ghana Li ing
S anda ds Su eys (GLSS 7), a na ionally ep esen a i e su ey conduc ed by he Ghana S a is ical
Se ice (GSS) in 2016/2017.
1
The su ey was designed o co e 15,000 households, selec ed h ough a
wo-s age p obabili y sampling app oach. In he i s s age, 1000 enume a ion a eas (EAs) we e andomly
selec ed om all egions o Ghana based on he p obabili y p opo ional o size (PPS) o each egion.
Following he household lis ing o he selec ed EAs, he second s age in ol ed andom selec ion o 15
households om each EA. O he 15,000 selec ed households, 14,009 we e success ully in e iewed. O
he households in e iewed, 7991 (which accoun s o abou 57%) we e loca ed in u al a eas. This s udy
ocuses on he u al subsample o unde s and he po en ial e ec o access o digi al inance on di e -
en o ms o li elihood di e si ica ion in he u al economy.
GLSS 7 is a mul i- opic household su ey ha p o ides comp ehensi e da a on household socioeco-
nomic cha ac e is ics and engagemen in many o ms o economic ac i i ies. We ely on in o ma ion on
households’owne ship o mobile money accoun s, which a e he dominan means o digi al inancial
se ices a ailable o u al households in Ghana, as well as in o ma ion on hei in ol emen in c op p o-
duc ion, li es ock p oduc ion, and non- a m business ac i i ies.
3.2. Reg ession model, measu emen o a iables and es ima ion echniques
To in es iga e he ela ionship be ween access o digi al inance and li elihood di e si ica ion, we i s
explo ed he key co ela es o each o hese wo a iables in sepa a e eg ession models. Howe e , in
he model o access o digi al inance, measu es o li elihood di e si ica ion a e included as independ-
en a iables; simila ly, access o digi al inance en e s he models o li elihood di e si ica ion as an
independen a iable. We speci y wo main models as ollows:
Momoi¼dDi iþXbþ i(1)
Di i¼aMomoiþXhþei(2)
Equa ion (1) is he model o access o digi al inance, which is measu ed by he owne ship o mobile
money accoun s by he household, while Equa ion (2) models he co ela es o li elihood di e si ica ion.
Xin bo h equa ions ep esen s a ec o o con ol a iables o o he co ela es o access o digi al
4 R. ATTA-ANKOMAH ET AL.
inance and li elihood di e si ica ion. Momo is a dicho omous measu e ha akes a alue o 1 i a leas
one membe o he household has a mobile money accoun , bu ze o o he wise. Di cap u es household
li elihood di e si ica ion. Sec ion 3.2.1 explains how li elihood di e si ica ion and o he a iables a e
measu ed.
3.2.1. Measu es o li elihood di e si ica ion and o he a iables
The measu es o di e si ica ion used in he s udy cap u e di e si ica ion wi hin a gi en li elihood sou ce
in he u al economy, o which h ee sou ces a e conside ed –c op p oduc ion, li es ock p oduc ion and
non- a m en e p ises. The di e si ica ion wi hin c op p oduc ion, which we e e o as c op di e si ica-
ion, is measu ed in wo ways. Fi s , we measu e c op di e si ica ion as he o al numbe o di e en
ypes o c ops cul i a ed by households, which means ha o u al households ha a e no engaged in
c op p oduc ion, he sco e o c op di e si ica ion will be ze o. Second, we used he en opy measu e o
di e si ica ion o c ea e an index o c op di e si ica ion, which p o ides a composi e measu e o di e si i-
ca ion ac oss and wi hin six di e en g oups o c ops (i.e. cash c ops, ce eals, ui s, legumes, oo s, and
ege ables). The en opy measu e o di e si ica ion has mainly been applied o measu e he dis ibu ion
o i ms’ac i i ies ac oss di e en p oduc s o ma ke segmen s (see Palepu, 1985; Raghuna han, 1995;
Theil, 1967). We ollow his app oach o gene a e a di e si ica ion index o households’engagemen in
di e en ypes o c op p oduc ion. The index is o mally de i ed as ollows:
CDi¼X
n
i¼1
½Piln 1
Pi
=ln nðÞ (3)
Whe e CD
i
is he di e si ica ion sco e o household i. P
i
is he sha e o a pa icula c op g oup, say
legumes, in he o al numbe o c ops g own by he household. In o he wo ds, he numbe o di e en
legume c ops g own by a household is exp essed as a p opo ion o he o al numbe o c ops g own
by he household. nis he numbe o c op g oups, and he e a e six c op g oups conside ed in his
s udy, as men ioned ea lie . Acco ding o Raghuna han (1995), he nume a o is e e ed o as he
en opy alue, whe eas he denomina o p o ides he maximum en opy alue. The alues o CD ange
be ween ze o and one, whe e ze o co esponds o no di e si ica ion ( ha is, he household g ows only
a single c op om only one o he six g oups), and one means ha he dis ibu ion is e en ac oss he
six di e en g oups o c ops ( ha is, he numbe o c ops g own wi hin each o he six g oups a e all
equal). By de ini ion, he en opy sco e exis s only o households ha g ow a leas one c op; hence, CD
ac ually measu es he ex en o c op di e si ica ion among households ha a e in o c op p oduc ion.
Fo di e si ica ion wi hin li es ock p oduc ion, we use he o al numbe o di e en ypes o li es ock
aised by he household as a measu e. Addi ionally, we examine li es ock di e si ica ion using a measu e
ha explo es he ex en o di e si ica ion among households ha epo ed aising li es ock. Howe e ,
we used p incipal componen analysis (PCA) o gene a e his measu e ins ead o he en opy me hod
o wo easons: (1) he numbe o di e en ypes o li es ock ha he households we e asked o epo
on was much smalle (24 di e en li es ock) compa ed o he 42 di e en c ops; (2) o 10 o hese di -
e en li es ock, he da a showed ha he numbe o households aising each o he 10 cons i u ed less
han a pe cen age o all u al households, lea ing only 14 o classi ica ion in o subg oups. As a esul ,
we pe o med PCA on he emaining 14 di e en li es ock ypes o gene a e a sco e (based on he i s
p incipal componen ) ha measu es he ex en o li es ock di e si ica ion among households ha aise
li es ock. Each o he 14 li es ock ypes en e ed ou PCA as a bina y a iable, whe e one means he
household aised a pa icula li es ock, and ze o i o he wise.
Wi h ega ds o di e si ica ion in ol ing non- a m ac i i ies, we used only a dicho omous measu e
which akes a alue o 1 i he household owns a leas a non- a m en e p ise and ze o i o he wise. We
ely only his dicho omous measu e because while he su ey asked households o epo on up o ou
non- a m en e p ises, he da a showed ha only six pe cen o he households had wo o mo e non-
a m en e p ises. Those wi h no non- a m en e p ises and hose wi h only one non- a m en e p ise con-
s i u ed 67% and 27% o he sample, espec i ely.
COGENT ECONOMICS & FINANCE 5
Table 1 p o ides u he de ails ega ding he a iable labels and desc ip ions o li elihood di e si ica-
ion a iables. Simila ly, Table 1 p o ides in o ma ion on how all he o he a iables including he con ol
a iables a e de ined as well as wi h hei summa y s a is ics.
3.2.2. Es ima ion echniques and s a egy o add essing po en ial endogenei y
Table 2 p o ides he de ails o he speci ic es ima ion echniques applied o Equa ions (1) and (2). Each
es ima ion echnique was selec ed based on i s sui abili y o he na u e o he dependen a iable
being modeled. Fo dicho omous a iables, logi eg ession me hod was applied while Poison eg ession
me hod was applied o coun dependen a iables and he O dina y Leas Squa es me hod was applied
o models wi h con inuous dependen a iables. Thus, he models o access o digi al inance and
Table 1. Desc ip i e s a is ics on a iables used.
Desc ip ion Mean SD Min Max
momo equals 1 i a leas a household (HH) membe has a mobile money accoun ,
o he wise ze o
0.17 0.38 0 1
c op_ o al is he numbe o di e en ypes o c ops HH cul i a ed ( he e we e o al o
42 di e en ypes o c ops he HH was equi ed o epo on).
2.51 2.28 0 13
o al_di e se is an en opy measu e o c op di e si ica ion ac oss 6 di e en g oups o
c ops - i cap u es he ex en o di e si ica ion ac oss he indi idual c op ypes as
well as he 6 di e en g oups o c ops
0.38 0.23 0 0.92
n_li es ock is he numbe o di e en ypes o li es ock aised by he HH ( he e we e
o al o 24 di e en ypes o li es ock he HH was equi ed o epo on).
1.16 1.78 0 11
li es ock_di is a PCA sco e based on he i s p incipal componen using owne ship o
14 di e en ypes o li es ock.
−0.41 1.37 −2.26 6.73
own_en e p2 akes a alue o 1 i he HH owns a non- a m en e p ise; o he wise ze o 0.33 0.47 0 1
emale equals 1 i he gende o he HH head is emale, o he wise ze o. 0.28 0.45 0 1
age is he HH head’s age in comple ed yea s 47.41 16.34 15 99
age_squa ed is he squa ed o age 2515 1707 225 9801
ma ied_con equals 1 i HH head is in ma i al o consensual ela ionship, o he wise ze o. 0.69 0.46 0 1
Religion o HH head a h ee le els, measu ed as dummies:
no_ eligion 0.07 0.25 0 1
ch is ians 0.63 0.48 0 1
o he _ eli 0.30 0.46 0 1
Educa ion o HH head a ou le els, measu ed as dummies:
no_educa ion (HH head has no educa ion) 0.40 0.49 0 1
p ima y (HH head comple ed p ima y educa ion) 0.17 0.38 0 1
JHS_JSS_Middle (HH headed comple ed junio high school o equi alen ) 0.30 0.46 0 1
sec_ oc_plus (HH head comple ed a leas senio high school, oca ional school o
equi alen )
0.08 0.26 0 1
hhsize is he numbe o pe sons who make up he HH. 4.69 3.11 1 28
any_shock equals 1 i a leas a HH membe abo e i e yea s was ill o inju ed in he
pas 12 mon hs, o he wise ze o
0.44 0.5 0 1
el_ng id equals 1 i HH communi y is connec ed o na ional powe g id, o he wise ze o 0.81 0.39 0 1
in_ins equals 1 o a ailabili y o o mal inancial ins i u ion in HH communi y,
o he wise ze o.
0.29 0.46 0 1
p i_com equals 1 o a communi y wi h p ima y school, o he wise ze o 0.87 0.34 0 1
JHS_com equals 1 o a communi y wi h Junio High School, o he wise ze o 0.73 0.44 0 1
SHS_com equals 1 o a communi y wi h Senio High School, o he wise ze o 0.30 0.46 0 1
Ecological zone whe e HH is loca ed a h ee le els, measu ed as dummies:
u al_coas al 0.14 0.35 0 1
u al_ o es 0.38 0.49 0 1
u al_sa anna 0.47 0.5 0 1
m_pene a ion is mobile phone pene a ion a e a he communi y/EA 0.49 0.18 0.02 0.94
p op_mm is HH membe s owning phone as a p opo ion o all HH membe s 0.39 0.33 0 1
ln a msize is na u al log o o al a m size owned o ope a ed by HH 1.25 1.01 0 9.21
ln o al_cos is na u al log o he o al ag icul u al inpu cos o he HH 4.11 2.93 0 9.05
ln_wel a e is na u al log o annual eal pe capi a adul equi alen consump ion
expendi u e
7.66 0.88 3.66 11.05
Dummies o 10 egions o Ghana:
2
wes e n 0.10 0.30 0 1
cen al 0.09 0.28 0 1
g ea e _acc a 0.02 0.13 0 1
ol a 0.11 0.32 0 1
eas e n 0.10 0.30 0 1
ashan i 0.09 0.28 0 1
b ong_aha o 0.09 0.29 0 1
no he n 0.12 0.33 0 1
uppe _eas 0.14 0.34 0 1
uppe _wes 0.15 0.35 0 1
No e: ‘SD’s ands o s anda d de ia ion, and ‘min’and ‘max’ espec i ely, s and o minimum and maximum.
6 R. ATTA-ANKOMAH ET AL.
owne ship o non- a m en e p ises we e es ima ed using logi eg essions, whe eas hose o he coun
measu es o di e si ica ion (ie C op_ o al and n_li es ock) we e es ima ed using Poisson eg essions and
hose o con inuous measu es o di e si ica ion (ie To al_di e se and li es ock_di ) we e es ima ed
using OLS.
By de ini ion, Equa ions (1) and (2) indica e a po en ial simul anei y p oblem, which could be a majo
sou ce o endogenei y bias, be ween each o he di e si ica ion a iables and he mobile money a i-
able. Wi hou add essing his p oblem, Equa ions (1) and (2) only help us lea n abou he associa ion
be ween access o digi al inance and ou measu es o li elihood di e si ica ion. We he e o e add ess
his po en ial endogenei y bias using ins umen al a iable models. Gi en ou in e es in he e ec o
access o digi al inance (mobile money) on li elihood di e si ica ion, we sough o an ins umen o
mobile money, and no o li elihood di e si ica ion. Following ecen s udies (such as Ake & Mbi i,
2010; A a-Ankomah, 2024; Buka i & Koomson, 2020), we conside ed wo candida es as po en ial ins u-
men s o mobile money: he p opo ion o households in a communi y/enume a ion a ea ha owns
mobile phones (labelled in Table 1 as m_pene a ion) and he p opo ion o household membe s ha
own mobile phone (labelled in Table 1 as p op_mm). While hese wo a iables a e measu es o mobile
phone pene a ion, m_pene a ion p o ides a measu e a he le el o he communi y/enume a ion a ea
le el, whe eas p op_mm measu es mobile phone pene a ion a he household le el. In ou p elimina y
analysis, bo h a iables passed he unde -iden i ica ion es ; howe e , only p op_mm passed he weak
iden i ica ion es . Hence, he esul s epo ed in his s udy used p op_mm as he ins umen o mobile
money.
Ou ins umen al a iable models we e es ima ed using he limi ed-in o ma ion maximum likelihood
(LIML) es ima o . Ou choice o LIML was in o med by he a gumen ha he LIML es ima o ends o
p oduce less biased es ima es, is mo e obus o weak ins umen s, and p oduces con idence in e als
wi h be e co e age a es han wo-s age leas squa es (Poi, 2006; S ock e al., 2002; S ock & Yogo,
2005). In addi ion, he LIML es ima o pe o ms be e han bo h wo-s age leas squa es and gene alized
me hods o momen s in ini e samples bu is equi alen o a wo-s age leas squa es me hod in la ge
samples (Came on & T i edi, 2010).
4. Resul s and discussion
This sec ion p esen s and discusses he indings o he s udy. We i s p esen b ie desc ip i e esul s on
he ela ionship be ween ou di e si ica ion a iables and access o digi al inance, a e which we p e-
sen he esul s o he econome ic analysis. Unde he econome ic esul s, he co ela es o access o
digi al inance in u al Ghana a e i s p esen ed and discussed, ollowed by he co ela es o li elihood
di e si ica ion. The sec ion hen discusses he esul s o he e ec o access o digi al inance on li eli-
hood di e si ica ion in models ha accoun o po en ial endogenei y.
4.1. Desc ip i e analysis
The a e age p opo ion o u al households owning mobile phones ac oss he a ious communi ies/enu-
me a ion a eas was 49%, while only 17% o he households owned a leas one mobile money accoun
(see Table 1). This sugges s ha he e we e many households in u al Ghana wi h mobile phones ha
did no own o ope a e a mobile money accoun as o 2016/2017, he su ey yea . The cha in he bo -
om igh o Figu e 1 shows ha he mobile money accoun owne ship a e di e s sligh ly be ween
Table 2. Es ima ion s a egy.
Dependen a iable Va iable name Na u e o dependen a iable Rele an equa ion Es ima ion model
Access o digi al inance Momo Dicho omous Equa ion (1) Logi
C op di e si ica ion C op_ o al Coun Equa ion (2) Poisson
To al_di e se Con inuous Equa ion (2) OLS
Li es ock di e si ica ion n_li es ock Coun Equa ion (2) Poison
li es ock_di Con inuous Equa ion (2) OLS
Non- a m en e p ise Own_en e 2 Dicho omous Equa ion (1) Logi
No e: Ins umen al a iable (IV) es ima ions a e also pe o med o all dependen a iables (see Sec ion 3.2.1 o de ails).
COGENT ECONOMICS & FINANCE 7
endogenei y. Thus, we conclude ha access o digi al inance is endogenous in models (1), (2), and (3).
This also implies ha ins umen al a iable models a e mo e app op ia e o de e mining he e ec o
digi al inance on c op di e si ica ion and li es ock di e si ica ion. In he case o he model o owne -
ship o an NFE and he ex en o li es ock di e si ica ion (using he PCA sco e), we do no ejec he null
hypo hesis o exogenei y. The exogenei y o digi al inance in non- a m en e p ises and he in ensi y o
li es ock di e si ica ion models imply ha he esul s in Columns (4) and (5) o Table 4 hold. The es s
o unde -iden i ica ion, wi h he null hypo hesis ha he model is unde -iden i ied, yield a p alue o
less han .01 o all he models, and hence, we ejec he null hypo hesis and conclude ha he models
a e no unde -iden i ied. In o he wo ds, he e is a high co ela ion be ween he ins umen and
endogenous a iable, con i ming ha ou ins umen is ele an . Simila ly, he weak iden i ica ion es ,
wi h he null hypo hesis ha he model is weakly iden i ied, is ejec ed o all he models because he
F-s a is ic o each o he models is a g ea e han 10, he con en ional h eshold o de e mining weak
iden i ica ion.
The IV LIML esul s in Table 5 show ha access o digi al inance ends o inc ease he numbe o di -
e en ypes o li es ock ea ed. The esul s in Table 5 Column 4 indica es ha he e ec disappea s wi h
espec o he ex en o li es ock di e si ica ion measu ed using PCA sco es, which es ic s he sample
o households ha own a leas one ype o li es ock. This means ha he signi ican e ec in he model
o he numbe o di e en li es ock (see Column 3 o Table 5) may be la gely d i en by he a ia ion
be ween households wi h no li es ock and hose wi h any numbe o li es ock. In con as o he esul s
on di e si ica ion owa d li es ock, we ind ha access o digi al inance educes di e si ica ion be ween
c ops and wi hin c op g oups. In o he wo ds, households wi h access o digi al inance a e less likely o
ha e di e si ied c op p oduc ion. Thus, access o digi al inancial se ices is associa ed wi h di e si ica-
ion in o non-c op p oduc ion a enues and his may be d i en by alloca i e e iciency ha may be
linked o access o digi al inancial se ices, including access o c edi , sp eading isk, o bo h. In pa icu-
la , i is easonable o belie e ha u al ag icul u al households specialize in he p oduc ion o pa icula
c ops i , h ough access o digi al inance, hey a e able o engage in NFEs ha gene a e mo e income
han when hey g ow only c ops. Indeed, e idence sugges s ha inancial inclusion, especially access o
ag icul u al c edi , leads a me s o in ensi y he cul i a ion o speci ic c ops and lowe di e si ica ion
wi hin c op p oduc ion (Geb eselassie, 2006; Loison, 2019; Xie e al., 2019).
4.2.4. Digi al inance and NFEs - sub-sample esul s o only u al ag icul u al households
To ein o ce ou indings on he e ec o digi al inance on di e si ica ion h ough NFEs, we es ic he
sample o u al ag icul u al households only and e-examine his ela ionship. The esul s a e p esen ed
in Table 6.
4
These esul s se e as u he obus ness check. Column (1) is o households ha a e in a
leas one ype o ag icul u al ac i i y (ie c op p oduc ion, o li es ock p oduc ion), while column (2) is
o households ha a e in bo h ypes o ag icul u al ac i i ies. The indings show ha ha ing access o
digi al inance has a posi i e and signi ican e ec on di e si ying li elihood sou ces among u al
Table 6. Logi eg ession esul s on e ec o digi al inance on owne ship o NFEs (sub-sample esul s o only u al
ag icul u al households).
Independen a iables
(1) (2)
Household is in o ei he c op o li es ock
p oduc ion
Household is in o bo h c op and li es ock
p oduc ion
momo 0.280 0.411
(0.094) (0.128)
Cons an −6.048 −5.730
(0.653) (0.826)
Con ol a iables and egional ixed e ec s Yes Yes
Obse a ions 5944 2841
Pseudo R-squa ed 0.102 0.0987
Numbe o clus e s 549 432
Chi squa e s a is ics 474.9 248.0
pValue o chi squa e 0 0
No e: (1) Robus (clus e ed) s anda d e o s a e in pa en heses; (2) p<.01, p<.05, p<.1; and (3) endogenei y es esul s om IV
LIML eg essions show ha mobile money was no endogenous –see Table A4 in he Online Appendix.
14 R. ATTA-ANKOMAH ET AL.
ag icul u al households ( ha is, when non-ag icul u al households a e excluded om he analysis), and
he e ec is s onge o households ha engage in bo h ypes o ag icul u al ac i i ies. I is possible
ha hose households ha a e in ol ed in bo h c op and li es ock p oduc ion ha e ga ne ed he expe i-
ence o di e si ying e en hough his di e si ica ion was limi ed o he ag icul u al sec o . The e o e,
hey ind i easie o di e si y in o NFEs when hey ha e access o digi al inance and oppo uni ies o
inclusion in he inancial sys em. These esul s show ha ou o e all inding o he s ong e ec o
digi al inance in p omo ing li elihood di e si ica ion among u al households is consis en o all u al
households and u al ag icul u al households.
4.2.5. Gende he e ogenei y in he e ec o digi al inance on li elihood di e si ica ion
Table 7 summa izes he he e ogenei y in he e ec o digi al inance on li elihood di e si ica ion by he
gende o he household head.
5
Whe e he endogenei y es yields an endogenous esul o access o
digi al inance, he IV esul s a e in e p e ed, whe eas he OLS/logi /Poison esul s a e in e p e ed o
cases whe e he endogenei y es showed ‘no endogenous’ esul s. The esul s show a signi ican e ec
on owne ship o NFE o bo h male- and emale-headed households, al hough he e ec appea s s on-
ge among emale-headed households, which a e consis en wi h he indings o Loison (2019) and Lay
e al. (2008).
While he esul s demons a e no signi ican e ec o digi al inance on c op di e si ica ion among
emale-headed households, we obse e a nega i e ela ionship among male-headed households, imply-
ing ha he nega i e e ec o digi al inance on c op di e si ica ion obse ed in he p e ious sec ion is
mainly d i en by male-headed households. Rega ding li es ock p oduc ion, he esul s show signi ican
di e si ica ion in he numbe o di e en li es ock aised among male-headed households while digi al
inance has no signi ican e ec on li es ock di e si ica ion among households headed by emales. An
implica ion o hese indings is ha inc eased access o digi al inance among emales in u al Ghana
will lead o g ea e deg ee o li elihood di e si ica ion away om ag icul u e. P e ious esea ch has
shown ha FHHs in u al a eas end o be mo e ulne able in e ms o access o esou ces han o he
households (Adjei-Man ey e al., 2022). Thus, suppo o such households h ough imp o ed access o
mobile money and o he digi al inancial se ices can help expand oppo uni ies o economic pa icipa-
ion by emales, and lead o inc eased emale economic empowe men and imp o emen in li ing
s anda ds. This is in line wi h a ecen s udy by M. S. Islam e al. (2022) ha ound ha pa icipa ion in
NFEs has signi ican and posi i e e ec s on u al women’s income. NFE as an impo an a enue o
women’s economic pa icipa ion is u he unde pinned by exis ing s uc u al challenges (such as he
pa ia chal na u e o land owne ship) in Ghana ha p e en women om ea ning maximum e u ns
om a m wo k.
5. Conclusion
The eme gence o digi al inancial sys ems, especially mobile money, has signi ican ly changed he inan-
cial se ices space in SSA in ways ha p o ide oppo uni ies o p omo e inancial inclusion among he
poo , especially hose in u al a eas. Focusing on mobile money, which is he main a enue o digi al
Table 7. He e ogeneous e ec s o momo on li elihood di e si ica ion by gende o household head.
Type o a iable Gende o household head
IV esul s
OLS/logi / PoisonEndogenei y es esul coe icien
c op_ o al Male Endogenous −3.218 −0.027
Female No endogenous −0.789 −0.000
o al_di e se Male Endogenous −0.206 −0.006
Female No endogenous −0.010 −0.006
n_li es ock Male Endogenous 1.292 −0.011
Female No endogenous 0.688−0.011
li es ock_di Male No endogenous 0.648 0.084
Female No endogenous 0.030 0.136
own_en e p2 Male No endogenous 0.034 0.296
Female No endogenous 0.364 0.465
No e. (1) Robus (clus e ed) s anda d e o s a e in pa en heses; (2) p<.01, p<.05, p<0.1; and (3) all models passed he iden i ica-
ion es s ( e e o Table A3 in he Online Appendix o de ailed esul s).
COGENT ECONOMICS & FINANCE 15
inancial se ices o u al households in Ghana, his s udy in es iga es whe he digi al inance in luences
u al households’li elihood di e si ica ion, using na ionally ep esen a i e da a om Ghana. This s udy
also del es in o he e ec on di e en o ms o li elihood di e si ica ion among u al households. Using
se e al econome ic me hods, including ins umen al a iable echniques, he s udy ound mobile
money o be posi i ely associa ed wi h households’choice o non- a m en e p ises as a li elihood
op ion, as well as engagemen in li es ock p oduc ion. Howe e , we ind a nega i e ela ionship
be ween mobile money and di e si ica ion in c op p oduc ion, as well as he ex en o c op di e si ica-
ion by households wi hin he c op sec o . In o he wo ds, when ag icul u al households ha e access o
mobile money se ices, hey end o p oduce ewe numbe o di e en c ops while explo ing non- a m
ac i i ies and li es ock p oduc ion as addi ional li elihood op ions. Gene ally, he esul s show ha
when u al households become mo e included in he inancial sys em h ough mobile money, hey end
o di e si y in o less adi ional sou ces o li elihood. This e ec pe ains o bo h male-headed and
emale-headed households, excep in he case o di e si ica ion in c op p oduc ion, whe e we ound he
nega i e e ec o be mainly d i en by male-headed households. In addi ion o mobile money, o he
impo an co ela es o di e si ica ion wo h highligh ing a e he educa ion o he household head, he
expe ience o a heal h shock by he household, expendi u e on ag icul u al inpu s, and he geog aphic
loca ion o he household. Educa ion was also ound o be an impo an co ela e o access o digi al
inance in addi ion o o he ac o s such as ag icul u al inpu cos , access o he na ional elec ici y g id,
and he p opo ion o household/communi y membe s owning a mobile phone.
These indings ha e se e al impo an policy implica ions. The di e si ica ion away om c op p oduc-
ion and he di e si ica ion in o non- a m business ac i i ies and li es ock p oduc ion sugges s ha u al
households may be ealloca ing p oduc i e esou ces in o a eas ha yield highe e u ns o hei li eli-
hoods and/o mi iga e he isk o income losses, and ha access o digi al inance acili a es his p ocess.
While his shi may enhance u al li elihoods, i may also ee land esou ces o mo e comme cialized
ag icul u al p oduc ion beyond subsis ence ag icul u e p ac iced by many u al households in de elop-
ing coun ies such as Ghana. This means ha he Go e nmen o Ghana should implemen policies o
suppo he adop ion and use o digi al inancial se ices by u al households as pa o b oade s a -
egies o e olu ionize ag icul u e and he u al economy a he imposing an E-le y on mobile money
ansac ions. Such suppo i e policies could also lead o inc eases in women’s economic pa icipa ion
and empowe men by acili a ing hei engagemen in NFE ac i i ies. The esul s also highligh he need
o add ess cons ain s ha limi u al households’access o digi al inancial ools, especially mobile
money, o which ‘non- o mal’educa ion o he less educa ed on how o use hese digi al ools is c u-
cial. To do his, we ecommend a collabo a i e e o be ween go e nmen and majo elcos ope a ing
mobile money se ices in Ghana. In addi ion, go e nmen should enhance in es men in he p o ision
o complemen a y in as uc u e equi ed o b oadening access o mobile money se ices in u al com-
muni ies, such as making elec ici y mo e accessible as well as incen i izing elcos o make mobile
phones and ne wo ks mo e accessible in u al communi ies.
While he da a used in his s udy came om he la es GLSS (a na ionally ep esen a i e household
su ey in Ghana), i is impo an o acknowledge ha he da a was collec ed in 2016/2017, hence, may
no cap u e ecen de elopmen s wi hin he mobile money space. The e o e, while his s udy o e s sig-
ni ican insigh s in o he socioeconomic impac o mobile money in he u al economy, i may be wo h-
while o eplica e he s udy when a new na ionally ep esen a i e da a becomes a ailable in o de o
lea n abou he impac o ecen de elopmen s in he mobile money space. Also, he measu es o di e -
si ica ion used in his s udy la gely ocused on di e si ica ion wi hin b oad li elihood ac i i ies a eas and
no ac oss hese a eas, which is an a ea u he s udies can explo e. Addi ionally, u he s udies may di -
ec ly explo e impac o mobile money on income di e si ica ion, which can encapsula e o he sou ces o
income such as emi ances and asse s.
No es
1. The au ho s ob ained an anonymized e sion o he da a om GSS and did no di ec ly engage any o he
pa icipan s o he su ey. Hence, he esea ch ac i i ies conduc ed by he au ho s o his s udy a e e y less
likely o lead o a iola ion o any esea ch e hics.
16 R. ATTA-ANKOMAH ET AL.
2. This s udy uses he en egions ha exis ed a he ime o he GLSS 7 su ey. The egional bounda ies ha e
since been e-dema ca ed o c ea e addi ional six egions in he coun y.
3. The ull eg ession esul s a e displayed in Table A2 in he Online Appendix.
4. The ull esul s a e p esen ed in Table A3 in he online appendix.
5. The ull esul s a e shown in he Online Appendix in Tables A4 and A5.
Au ho con ibu ions
Richmond A a-Ankomah: Concep ion, me hodology, da a managemen and analysis, w i ing d a and discussion;
e ising and edi ing he in ellec ual
Kwame Adjei-Man ey: Li e a u e e iew, w i ing d a and discussion, e ising and edi ing he in ellec ual
con en s
Aku o Amankwah: Li e a u e e iew, w i ing d a and discussion, e ising and edi ing he in ellec ual con en s
Disclosu e s a emen
No po en ial con lic o in e es was epo ed by he au ho (s).
Funding
This s udy ecei ed no unding.
Abou he au ho s
Richmond A a-Ankomah is a de elopmen economis and Senio Resea ch Fellow a he Ins i u e o S a is ical,
Social and Economic Resea ch, Uni e si y o Ghana. His esea ch in e es is di e se bu la gely con e ges a ound
de elopmen issues Conce ning households and i ms in sub Saha an A ica.
Kwame Adjei-Man ey is esea che a he Depa men o Applied Economics, Uni e si y o En i onmen and
Sus ainable De elopmen , Somanya, Ghana. His esea ch in e es s a e in he a eas o ene gy and en i onmen al eco-
nomics, de elopmen economics and beha io al economics.
Aku o Amankwah is an Economis o he Li ing S anda ds Measu emen S udy (LSMS), he Wo ld Bank’s lagship
household su ey p og am housed a he De elopmen Da a G oup. His p ima y a eas o esea ch a e po e y, labo ,
aquacul u e, ag icul u e and u al de elopmen , and me hodological s udies o imp o e household su eys.
ORCID
Richmond A a-Ankomah h p://o cid.o g/0000-0001-9982-9377
Da a a ailabili y s a emen
The main da a unde lying his s udy a e a ailable upon eques om he Ghana S a is ical Se ice (h ps://mic oda-
a.s a sghana.go .gh/index.php/ca alog/97/s udy-desc ip ion).
Re e ences
Adjei-Man ey, K., Kwakwa, P. A., & Adusah-Poku, F. (2022). Un a eling he e ec o gende dimensions and wood
uel usage on household ood secu i y: E idence om Ghana. Heliyon,8(11), e11268. h ps://doi.o g/10.1016/j.heli-
yon.2022.e11268
Ahmad, H. A., G een, C., & Jiang, F. (2020). Mobile money, inancial inclusion and de elopmen : A e iew wi h e e -
ence o A ican expe ience. Jou nal o Economic Su eys,34(4), 753–792. h ps://doi.o g/10.1111/joes.12372
Ahmed, M. T., Bhanda i, H., Go doncillo, P. U., Quicoy, C. B., & Ca naje, G. P. (2018). Fac o s a ec ing ex en o u al
li elihood di e si ica ion in selec ed a eas o Bangladesh. SAARC Jou nal o Ag icul u e,16(1), 7–21. h ps://doi.o g/
10.3329/sja. 16i1.37419
Akaakohol, M. A., & Aye, G. C. (2014). Di e si ica ion and a m household wel a e in Maku di, Benue S a e, Nige ia.
De elopmen S udies Resea ch. An Open Access Jou nal,1(1), 168–175. h ps://doi.o g/10.1080/21665095.2014.
919232
Ake , J. C., & Mbi i, I. M. (2010). Mobile phones and economic de elopmen in A ica. Jou nal o Economic
Pe spec i es,24(3), 207–232. h ps://doi.o g/10.1257/jep.24.3.207
COGENT ECONOMICS & FINANCE 17
Amoah, A., Ko le, K., & Asiama, R. K. (2020). Mobile money as a inancial inclusion ins umen : Wha a e he de e mi-
nan s? In e na ional Jou nal o Social Economics,47(10), 1283–1297. h ps://doi.o g/10.1108/IJSE-05-2020-0271
Amoah, A., Kwablah, E., Amoah, B., & Adjei-Man ey, K. (2023). Willingness o pay o elec onic ansac ion le y:
Empi ical e idence om Ghana. A ican Jou nal o Economic and Managemen S udies,14(4), 663–679. h ps://doi.
o g/10.1108/AJEMS-09-2022-0359
A on, J. (2018). Mobile money and he economy: A e iew o he e idence. The Wo ld Bank Resea ch Obse e ,33(2),
135–188. h ps://doi.o g/10.1093/wb o/lky001
As aw, S., Scognamillo, A., Di Cap e a, G., Si ko, N., & Ignaciuk, A. (2019). He e ogeneous impac o li elihood di e si-
ica ion on household wel a e: C oss-coun y e idence om Sub-Saha an A ica. Wo ld De elopmen ,117, 278–295.
h ps://doi.o g/10.1016/j.wo ldde .2019.01.017
Asli, D., Klappe , L., Singe , D., & Ansa , S. (2022). The Global Findex Da abase 2021: Financial inclusion,digi al pay-
men s, and esilience in he age o COVID-19. Wo ld Bank. h ps://doi.o g/10.1596/978-1-4648-1897-4
Asmah, E. E. (2011). Ru al li elihood di e si ica ion and ag icul u al household wel a e in Ghana. Jou nal o
De elopmen and Ag icul u al Economics,3(7), 325–334.
As a o , R. K., Boakye, A. N., & Essuman, J. (2022). Adop ion and in ensi y o use o mobile money among smallholde
a me s in u al Ghana. In o ma ion De elopmen ,38(2), 204–217. h ps://doi.o g/10.1177/0266666921999089
A a-Ankomah, R. (2022). C edi o households in Ghana: Has mobile money (momo) imp o ed inclusi e access?
Scien i ic A ican,16, e01230. h ps://doi.o g/10.1016/j.scia .2022.e01230
A a-Ankomah, R. (2024). Mobile money and in a-household employmen di e si ica ion in Ghana. A ican Jou nal o
Science, Technology, Inno a ion and De elopmen ,1–17. h ps://doi.o g/10.1080/20421338.2024.2317657
A a-Ankomah, R., Asan e-Addo, C., Okye e, C. Y., & Asan e-Poku, N. A. (2024). Does digi al inancial echnology
imp o e access o inance among households expe iencing idiosync a ic heal h shocks in Ghana? Unpublished
Manusc ip , Las modi ied, Janua y 9, 2024.
A a-Ankomah, R., & Okye e, Y. C. (2022). Wel a e e ec s o inancial inclusion se ices in Ghana: A compa a i e ana-
lysis o mobile money and o he inancial se ices. Global Social Wel a e,10(1), 83–92. h ps://doi.o g/10.1007/
s40609-022-00234-x
Ba ou , P. T., Rahaman, W. A., & Mohammed, I. (2021). Impac o mobile money access on in e na emi ances, con-
sump ion expendi u e and household wel a e in Ghana. Jou nal o Economic and Adminis a i e Sciences,37(3),
337–354. h ps://doi.o g/10.1108/JEAS-04-2020-0045
Ba a, A. (2013). Mobile money o inancial inclusion: Policy and egula o y pe spec i e in Zimbabwe. A ican Jou nal
o Science, Technology, Inno a ion and De elopmen ,5(5), 345–354. h ps://doi.o g/10.1080/20421338.2013.829287
Ba ool, S., & Jamil, F. (2019). Ru al employmen and income di e si ica ion in Pakis an. Pakis an Jou nal o
Ag icul u al Sciences,56(02), 503–510. h ps://doi.o g/10.21162/PAKJAS/19.7484
Bongomin, O. C. G., N ayi, M. J., Munene, C. J., & Malinga, A. C. (2018). Mobile money and inancial inclusion in Sub-
Saha an A ica: The mode a ing ole o social ne wo ks. Jou nal o A ican Business,19(3), 361–384. h ps://doi.o g/
10.1080/15228916.2017.1416214
Buka i, C., & Koomson, I. (2020). Adop ion o mobile money o heal hca e u iliza ion and spending in u al Ghana.
In S. A. Chu chill (Ed.), Mo ing om he millennium o he sus ainable de elopmen goals (pp. 37–60). Palg a e
Macmillan.
Came on, A. C., & T i edi, P. K. (2010). Mic oeconome ics using S a a ( e . ed.). S a a P ess.
Dabalen, A., & Mensah, J. T. (2023). Ten ac s abou digi al echnology adop ion in Ghana. A ica Can End Po e y.
Dagunga, G., Ayamga, M., & Danso-Abbeam, G. (2020). To wha ex en should a m households di e si y?
Implica ions on mul idimensional po e y in Ghana. Wo ld De elopmen Pe spec i es,20, 100264. h ps://doi.o g/
10.1016/j.wdp.2020.100264
Dagunga, G., Azumah, S. B., Zaka ia, A., Boa eng, A. N., Mensah, B. K., Boa eng, S. E., Tsekpo, M. E., & Kankam, P.
(2022). Sa ings and economic di e si ica ion among you h in Ghana: Implica ions o policy and p ac ice. Cogen
Economics & Finance,10(1), 2123887. h ps://doi.o g/10.1080/23322039.2022.2123887
Da is, B., Di Giuseppe, S., & Zezza, A. (2017). A e A ican households (no ) lea ing ag icul u e? Pa e ns o house-
holds’income sou ces in u al Sub-Saha an A ica. Food Policy,67, 153–174. h ps://doi.o g/10.1016/j. oodpol.2016.
09.018
Di ibsa, S., & Tassew, B. (2015). Analysis i li elihood di e si ica ion s a egy o u al households: A case s udy o
Ambo Dis ic , Oomiya Regional S a e, E hiopia. In e na ional Jou nal o Cu en Resea ch and Academic Re iew,
3(8), 406–426.
Finge , R. (2023). Digi al inno a ions o sus ainable and esilien ag icul u al sys ems. Eu opean Re iew o Ag icul u al
Economics,50(4), 1277–1309. h ps://doi.o g/10.1093/e ae/jbad021
Geb eselassie, S. (2006). In ensi ica ion o smallholde ag icul u e in E hiopia: Op ions and scena ios (Fu u e
Ag icul u es Conso ium Discussion Pape 007). Fu u e Ag icul u es.
Geb u, G. W., Ichoku, H. E., & Phil-Eze, P. O. (2018). De e minan s o li elihood di e si ica ion s a egies in Eas e n
Tig ay Region o E hiopia. Ag icul u e & Food Secu i y,7(1), 62. h ps://doi.o g/10.1186/s40066-018-0214-0
G zybowski, L. (2015). The ole o ne wo k e ec s and consume he e ogenei y in he adop ion o mobile
phones: E idence om Sou h A ica. Telecommunica ions Policy,39(11), 933–943. h ps://doi.o g/10.1016/j. elpol.
2015.08.010
18 R. ATTA-ANKOMAH ET AL.
Islam, A., Muzi, S., & Meza, J. L. R. (2018). Does mobile money use inc ease i ms’in es men ? E idence om en e -
p ise su eys in Kenya, Uganda, and Tanzania. Small Business Economics,51(3), 687–708. h ps://doi.o g/10.1007/
s11187-017-9951-x
Islam, M. S., Islam, S., Fa ema, K., & Khanum, R. (2022). Ru al women pa icipa ion in a m and o - a m ac i i ies and
household income in Bangladesh. Heliyon,8(9), e10618. h ps://doi.o g/10.1016/j.heliyon.2022.e10618
Jack, W., & Su i, T. (2014). Risk sha ing and ansac ions cos s: E idence om Kenya’s mobile money e olu ion.
Ame ican Economic Re iew,104(1), 183–223. h ps://doi.o g/10.1257/ae .104.1.183
Kassegn, A., & Abdinasi , U. (2023). De e minan s o u al households’li elihood di e si ica ion s a egies: In he case
o no h Wollo zone, Amha a Na ional Regional S a e, E hiopia. Cogen Economics and Finance,11(1), 2185347.
h ps://doi.o g/10.1080/23322039.2023.2185347
Kha un, D., & Roy, B. C. (2012). Ru al li elihood di e si ica ion in Wes Bengal: De e minan s and cons ain s.
Ag icul u al Economics Resea ch Re iew,25(1), 115–124.
Lay, J., Mahmoud, T. O., & M’Muka ia, G. M. (2008). Few oppo uni ies, much despe a ion: The dicho omy o non-
ag icul u al ac i i ies and inequali y in wes e n Kenya. Wo ld De elopmen ,36(12), 2713–2732. h ps://doi.o g/10.
1016/j.wo ldde .2007.12.003
Lee, J. N., Mo duch, J., Ra ind an, S., Shonchoy, A., & Zaman, H. (2021). Po e y and mig a ion in he digi al age:
Expe imen al e idence on mobile banking in Bangladesh. Ame ican Economic Jou nal: Applied Economics,13(1),
38–71. h ps://doi.o g/10.1257/app.20190067
Loison, S. A. (2015). Ru al li elihood di e si ica ion in sub-Saha an A ica: A li e a u e e iew. The Jou nal o
De elopmen S udies,51(9), 1125–1138.
Loison, S. A. (2019). Household li elihood di e si ica ion and gende : Panel e idence om u al Kenya. Jou nal o
Ru al S udies,69, 156–172.
Mahama, A. T., & Maha jan, K. L. (2017). De e minan s o li elihood di e si ica ion in Ghana om he na ional li eli-
hood s a egies and spa ial pe spec i e. Jou nal o In e na ional De elopmen and Coope a ion,23(1), 75–90.
Mahama, A. T., & Nkegbe, K. P. (2021). Impac o household li elihood di e si ica ion on wel a e in Ghana. Scien i ic
A ican,13, e00858. h ps://doi.o g/10.1016/j.scia .2021.e00858
Ma e n, M. (2018). How Ghana became one o A ica’s op mobile money ma ke s. CGAP (Consul a i e G oup o
Assis he Poo ) Blogpos . Wha You Need o Know.
Ma in, S. M., & Lo enzen, K. (2016). Li elihood di e si ica ion in u al Laos. Wo ld De elopmen ,83, 231–243. h ps://
doi.o g/10.1016/j.wo ldde .2016.01.018
Mille , H. (2019). Wha wo ks o digi al inancial se ices in ag icul u al de elopmen ? USAID Feed he Fu u e
Ini ia i e.
Minis y o Finance. (n.d). Towa d a cash-li e Ghana –Building an inclusi e digi al paymen s ecosys em: Digi al pay-
men s oadmap. h ps://mo ep.go .gh/si es/de aul / iles/ac s/Ghana_Cashli e_Roadmap.pd
Mohammed, K., Ba ung, E., Kansanga, M., Nyan akyi-F impong, H., & Luginaah, I. (2021). Li elihood di e si ica ion
s a egies and esilience o clima e change in semi-a id no he n Ghana. Clima ic Change,164(3-4), 1–23. h ps://
doi.o g/10.1007/s10584-021-03034-y
Mo hobi, O., & G zybowski, L. (2017). In as uc u e de iciencies and adop ion o mobile money in Sub-Saha an
A ica. In o ma ion Economics and Policy,40,71–79. h ps://doi.o g/10.1016/j.in oecopol.2017.05.003
Myeni, S., Maka e, M., & Mahonye, N. (2020). Does mobile money p omo e inancial inclusion in Eswa ini?
In e na ional Jou nal o Social Economics,47(6), 693–709. h ps://doi.o g/10.1108/IJSE-05-2019-0310
N’d i, L. M., & Kakinaka, M. (2020). Financial inclusion, mobile money, and indi idual wel a e: The case o Bu kina
Faso. Telecommunica ions Policy,44(3), 101926. h ps://doi.o g/10.1016/j. elpol.2020.101926
Nguyen, D. L., Nguyen, T. T., & G o e, U. (2022). Shocks, household consump ion, and li elihood di e si ica ion: A
compa a i e e idence om panel da a in u al Thailand and Vie nam. Economic Change and Res uc u ing,56(5),
3223–3255. h ps://doi.o g/10.1007/s10644-022-09400-9
Okye e, C. Y., A a-Ankomah, R., & Asan e-Addo, C. (2024a). Does Digi al Financial Inclusion Imp o e Food Secu i y and
Household Resilience? E idence om No he n Ghana. Unpublished Manusc ip , Las Modi ied, Janua y 2, 2024.
Okye e, C. Y., A a-Ankomah, R., & Asan e-Addo, C. (2024b). Mobile money adop ion, a m pe o mance and household
wel a e in No he n Ghana. Unpublished Manusc ip , Las modi ied, Janua y 10, 2024.
Ouma, S. A., Odongo, T. M., & We e, M. (2017). Mobile inancial se ices and inancial inclusion: Is i a boon o sa -
ings mobiliza ion? Re iew o De elopmen Finance,7(1), 29–35. h ps://doi.o g/10.1016/j. d .2017.01.001
Ox o d Business G oup. (2018). Ghana olls ou mobile money in e ope abili y pla o m.h ps://ox o dbusinessg oup.
com/a icles-in e iews/ghana- olls-ou -mobile-money-in e ope abili y-pla o m.
Palepu, K. (1985). Di e si ica ion s a egy, p o i pe o mance and he en opy measu e. S a egic Managemen
Jou nal,6(3), 239–255. 10.1002/smj.4250060305
Pep ah, J. A., O eng, C., & Sebu, J. (2020). Mobile money, ou pu and wel a e among smallholde a me s in Ghana.
SAGE Open,10(2), 215824402093111. h ps://doi.o g/10.1177/2158244020931114
Poi, B. P. (2006). Jackkni e ins umen al a iables es ima ion in S a a. The S a a Jou nal: P omo ing Communica ions
on S a is ics and S a a,6(3), 364–376. h ps://doi.o g/10.1177/1536867X0600600305
Raghuna han, S. P. (1995). A e inemen o he en opy measu e o i m di e si ica ion: Towa d de ini ional and com-
pu a ional accu acy. Jou nal o Managemen ,21(5), 989–1002. h ps://doi.o g/10.1177/014920639502100510
COGENT ECONOMICS & FINANCE 19
S ock, J. H., W igh , J. H., & Yogo, M. (2002). A su ey o weak ins umen s and weak iden i ica ion in gene alized
me hod o momen s. Jou nal o Business and Economic S a is ics,20(4), 518–529. h ps://doi.o g/10.1198/
073500102288618658
S ock, J. H., & Yogo, M. (2005). Tes ing o weak ins umen s in linea IV eg ession. In D. W. K. And ews (Ed.),
Iden i ica ion and in e ence o econome ic models (pp. 80–108). Camb idge Uni e si y P ess.
Theil, H. (1967). Economics and In o ma ion Theo y. No h-Holland.
Wo ld Bank Mic oda a Ca alog. (2023). Mobile cellula subsc ip ions (pe 100 people) –Ghana. h ps://da a.wo ldbank.
o g/indica o /IT.CEL.SETS.P2?loca ions=GH
Xie, H., Huang, Y., Chen, Q., Zhang, Y., and Wu Q. (2019). P ospec s o ag icul u al sus ainable in ensi ica ion:
A e iew o esea ch. Land 8(11), 157. h ps://doi.o g/10.3390/land8110157
20 R. ATTA-ANKOMAH ET AL.
