Ndjokou, I choko Mo ande Mondjeli Mwa; Asaloko, P ince Pi a
A icle
Empi ical e i ica ion o he link be ween he digi al di ide
and women's economic pa icipa ion in A ica
The Eu opean Jou nal o Compa a i e Economics (EJCE)
P o ided in Coope a ion wi h:
Uni e si y Ca lo Ca aneo (LIUC), Cas ellanza
Sugges ed Ci a ion: Ndjokou, I choko Mo ande Mondjeli Mwa; Asaloko, P ince Pi a (2024) : Empi ical
e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica,
The Eu opean Jou nal o Compa a i e Economics (EJCE), ISSN 1824-2979, Uni e si y Ca lo Ca aneo
(LIUC), Cas ellanza, Vol. 21, Iss. 1, pp. 123-164,
h ps://doi.o g/10.25428/1824-2979/030
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/320198
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-nd/4.0/
The Eu opean Jou nal o Compa a i e Economics
Vol. 21, no. 1, pp. 123-164
ISSN 1824-2979
h p://dx.doi.o g/10.25428/1824-2979/030
Fi s published online: 06/09/2024
Empi ical e i ica ion o he link be ween he digi al
di ide and women's economic pa icipa ion in A ica
I choko M.M. Mwa Ndjokou*, P ince Pi a Asaloko**
Abs ac
This a icle examines he link be ween he digi al di ide and women's economic pa icipa ion. We use a
c oss-sec ional model based on a panel o 45 A ican coun ies. Ou esul s show ha he digi al di ide
(cell phone, bandwid h and In e ne ) wo sens women's economic pa icipa ion. The nega i e in luence o
he digi al di ide on women's economic pa icipa ion ends o be ampli ied in coun ies wi h low
democ acy and in he indus ial sec o . Ou esul s emain s able when we add cul u al a iables and when
we use he al e na i e measu e o he digi al di ide. Howe e , a e using he quan ile eg ession app oach,
we ind ha hese in luences a y a di e en in e als along he dis ibu ion o women's economic
pa icipa ion.
JEL classi ica ion: I21, J21, O55
Keywo ds: Digi al di ide, Women’s economic pa icipa ion, OLS, Quan ile eg ession
1. In oduc ion
Fo se e al decades, he issue o women’s economic pa icipa ion has been a he
hea o he conce ns o go e nmen s and in e na ional bodies, due o i s impac on
inc easing economic g ow h (Pimkina and De la Flo , 2020) and educing po e y and
inequali y (Ve ick, 2018). Indeed, be ween 1970 and 1990, se e al a emp s we e made
o imp o e issues ela ed o women’s economic pa icipa ion, and his pe iod is known
as he “pionee ing” pe iod. The con e ences in Mexico Ci y in 1975, Copenhagen in
1980, Nai obi in 1985 and Beijing in 1995 a e among he key e en s o his pe iod
(Uni ed Na ions, 1996). Du ing his pe iod, women in de eloping coun ies acqui ed
* Resea che a he Uni e si y o Ma oua, Facul y o Economics and Managemen , E-mail:
mo ande@yahoo.
** Uni e si y o Yaoundé 2 (Came oon), Uni e si y o Kisangani (DRC), Applied Economics and
De elopmen Resea ch Labo a o y (LAREAD), E-mail: p incepi [email protected], O cid:
0009-0002-6394-8610
Acknowledgemen s. The au ho s would like o hank he edi o and he anonymous e iewe s o hei
commen s and sugges ions o imp o emen . They would also like o hank Ms. Lydie Dzou O ou, M .
Gilles Limnyuy and D Ch is ian Tsopmo o hei ou s anding wo k in e iewing his documen .
Howe e , he au ho s emain solely esponsible o any esidual e o s.
Con ibu ions o he au ho s. I choko M.M. Mwa Ndjokou and P ince Pi a Asaloko con ibu ed o he
concep ion and de elopmen o his s udy. Ma e ial p epa a ion, da a collec ion and analysis we e
ca ied ou . All au ho s ead and app o ed he e ised manusc ip .
Con lic o in e es . The au ho s decla e no
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
124
skills a an unp eceden ed a e. A he same ime, hey had o en e he labou ma ke
(UN Women, 2020). Un o una ely, mo e han wen y yea s a e he Beijing
con e ence, he s a us o women s ill seems dep essing (Klasen e al., 2021).
The Uni ed Na ions epo (2021) s a es ha digi al inclusion is one o he
p io i ies o gende equali y and women’s empowe men . This is pa icula ly ue in
some coun ies, whe e mobile phones and he in e ne a e seen as p omising ways o
ackling majo challenges such as equal access o he labou and poli ical ma ke s
(S amps, 1990; Raja e al., 2013; Dencik and Leis e , 2015) and inancial inclusion
(Asongu and Odhiambo, 2018). Indeed, ICTs p o ide women wi h access o
in o ma ion on job o e s, aining and p o essional de elopmen oppo uni ies. They
can use online pla o ms, social ne wo ks and sea ch engines o sea ch and apply o
jobs (McKinsey Global Ins i u e, 2015; In e na ional Labou O ganisa ion, 2017). In
addi ion, ICTs enable women o se up hei own businesses and become en ep eneu s.
They can use e-comme ce pla o ms, social ne wo ks and digi al inancial se ices o
s a and de elop hei business. (Wo ld Bank, 2019a).
Al hough he eme gence o ICTs in some coun ies has o e ed many
oppo uni ies o ma ginalised women, in A ica he digi al e olu ion has esul ed in
unequal access o new echnologies, leading o wha is now known as he “digi al
di ide” (Moolman e al., 2007). Acco ding o Gu umu hy (2004), he digi al di ide is a
pe sis en eali y, wi h s uc u al ba ie s limi ing access o echnological ools, o digi al
educa ion and socio-economic oppo uni ies. Cas ells (2002) de ines he “digi al di ide”
as he di e ence in access o and use o ICTs be ween indi iduals.
Wi h his in mind, se e al au ho s, including Tiem o é (2008), asse ha a
signi ican p opo ion o he popula ion is excluded om he app op ia ion o ICTs.
No e e yone has access o hese digi al ools, and hei dis ibu ion be ween social
g oups is highly unequal (G anjon e al., 2009). Acco ding o a epo by he Uni ed
Na ions De elopmen P og amme (UNDP, 2012), women in A ica ha e mo e
di icul y han men in accessing and using hese echnologies in he long e m because
o a pa ia chal social s uc u e ha encou ages hese p ac ices. Indeed, he digi al di ide
can exclude women om employmen oppo uni ies ha equi e digi al skills o access
o echnology. This can con ine hem o low-paid, low-skilled jobs (Wo ld Economic
Fo um, 2017). The digi al di ide can hinde women’s en ep eneu ship by limi ing
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
125
women’s access o he online in o ma ion, ools and esou ces needed o s a and un a
business (In e na ional Cen e o Resea ch on Women, 2016). In addi ion, he digi al
di ide can limi women’s access o online aining and p o essional de elopmen ,
hinde ing hei ca ee ad ancemen (In e na ional Telecommunica ion Union, 2019).
Wha ’s mo e, esea ch on he gende digi al di ide (Mo in-Sylla, 2005; Ta an,
2011) es ablishes ha women ha e a lowe o e all p obabili y o bene i ing om he
in o ma ion socie y han men. Women’s inabili y o ake ad an age o he oppo uni ies
o e ed by ICTs is hampe ed by a numbe o ac o s, including illi e acy, po e y,
language ba ie s and inadequa e access o echnological ools due o he low
ep esen a ion o women in posi ions o esponsibili y in ICT de elopmen p og ammes
(Me alopulos, 2008). E en when women ha e he necessa y skills o use ICTs,
“pe sis en cul u al ba ie s, such as s e eo ypical de ini ions o he oles o women and
men in A ica, con inue o hinde hei ull pa icipa ion in he in o ma ion age” (UN,
2005).
Gende dispa i ies a ound he wo ld a e also e lec ed in he digi al domain.
Globally, a ound 52% o women a e no connec ed, compa ed wi h 44% o men
(In e na ional Telecommunica ion Union, 2020). Ye he digi al gende gap is e en mo e
p onounced in de eloping coun ies, whe e women a e 7% less likely o own a cell
phone and 18% less likely o own a sma phone (Global Sys em o Mobile
Communica ion, 2022). Access o compu e s emains limi ed in hese egions, wi h only
54% o u ban households equipped, compa ed wi h jus 17% in u al a eas (Ambe &
Chichaibelu, 2023). The e a e ma ked egional di e ences when i comes o digi al
inequali ies be ween men and women. A ica is he egion whe e his digi al gende gap
is mos p onounced, pa icula ly in e ms o cell phone owne ship and mobile in e ne
use. This gap in luences women's pa icipa ion in he labo ma ke in A ica, whe e hei
pa icipa ion a e s ands a 57.56%, and 71.8% o hem a e mo e dependen on he
lowe -paid in o mal sec o (Global En ep eneu ship Moni o , 2020/2021).
Howe e , in de eloping coun ies in gene al, and in A ica in pa icula , he e is a
g ea deal o in e es in women's economic pa icipa ion. Un o una ely, e y ew
s udies ha e been published on he de e minan s o women's economic empowe men ,
including he link be ween he digi al di ide and women's economic empowe men in
he mac o case. Howe e , he wo k o Omo oso e al (2020) which analyzes gende
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
126
di e ences in b oadband in e ne use and i s e ec s on women's labo ma ke
pa icipa ion. Using an ins umen al a iable app oach, he esul s sugges ha
exogenous b oadband in e ne use leads o inc eases o a ound 14.1 and 10.6 pe cen age
poin s in labo ma ke pa icipa ion o single and ma ied women wi h some le el o
educa ion in Sou h A ica. Fo Ambe & Chichaibelu (2023) using he Pakis an Social
and Li ing S anda ds Measu emen Su ey (2019-2020), hey examine he digi al di ide
be ween gende s and be ween u al and u ban a eas. They use an ins umen al a iable
app oach o in es iga e he e ec o cell phone o sma phone owne ship on women's
labo ma ke pa icipa ion. The esul s indica e ha ins i u ional and socio-cul u al
no ms explain mos o he gap in cell phone o sma phone owne ship be ween men
and women. The ins umen al a iable app oach demons a es ha cell phone o
sma phone owne ship inc eases women's pa icipa ion in he labo ma ke .
Howe e , while ou esul s con i m ce ain aspec s o p e ious s udies, ou s udy
appea s o complemen exis ing wo k by o e ing a mo e con ex ual and nuanced
analysis o he e ec s o he digi al di ide on women's economic pa icipa ion in A ica,
while using econome ic me hods ha e eal complex dynamics ha p e ious
app oaches ha e been unable o cap u e. Fi s ly, we ha e op ed o bo h pa ame ic and
non-pa ame ic analysis. The choice o he c oss-sec ional panel-based model is
app op ia e, as i allows us o cap u e in a- and in e -coun y a ia ions, as well as he
e ec o poli ical and sec o al a iables, which has no been he case in p e ious s udies,
and also he use o quan ile eg ession as a means o be e unde s anding how he
e ec o he nume ical di ide a ies a di e en le els o women's economic
pa icipa ion. This esponds o me hodological limi a ions in p e ious s udies ha
would no ha e cap u ed his he e ogenei y in e ec s.
Secondly, his s udy p o ides new insigh s, no ably by showing how he e ec s
a y acco ding o he poli ical con ex and women's sec o o ac i i y. Finally, his pape
makes a con ibu ion o he li e a u e on he digi al di ide and women's economic
pa icipa ion in A ica. A e es ima ing a dynamic panel da a model using he OLS
me hod o e he pe iod 2000-2021, ou esul s clea ly show ha he digi al di ide
dec eases women's economic pa icipa ion in A ica. The emainde o he pape is
s uc u ed as ollows: he ollowing sec ion gi es a b ie o e iew o he li e a u e
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
127
e iew (2); sec ion (3) desc ibes he da a used and he es ima ion s a egy; sec ion (4)
p esen s he empi ical esul s.
2. Li e a u e e iew
Women’s pa icipa ion in economic li e has a ac ed conside able a en ion in
economic li e a u e since Mince ’s (1962) seminal s udy o ma ied women’s
pa icipa ion in he labou ma ke . This a emp o ein e p e he s a ic analysis o
labou supply was ollowed by se e al heo ies, including Becke ’s (1965) “ ime
alloca ion heo y”, conce ning he household p oduc ion model and he alloca ion o
women’s ime. The aim o hese heo ies was o iden i y he ac o s associa ed wi h
women’s choices o economic pa icipa ion. Howe e , mos empi ical analyses o he
mic o and mac o de e minan s o women’s economic pa icipa ion ha e ocused on
digi al echnology.
The empi ical li e a u e ha has eme ged on he e ec s o he digi al di ide and
women’s pa icipa ion in he wo k o ce can be di ided in o wo pa s. The i s
highligh s ha women can heo e ically bene i om he p oli e a ion o ICTs in a
numbe o ways, including making labou ma ke s mo e anspa en , lexible, inno a i e
and inclusi e, making i easie o women o ind jobs and o employe s o ind skilled
wo ke s among women (Nikulin, 2017; Ha eez e al., 2020), as be e access o labou
ma ke in o ma ion imp o es he job sea ch p ocess (Raja e al., 2013). In he Uni ed
S a es, o example, De ling, (2017), shows ha b oadband in e ne use is associa ed
wi h inc eased labou ma ke pa icipa ion o ma ied women, bu ound no
co esponding e ec o single women. Focusing on de eloping coun ies, Omo oso
and Obembe (2016) and Nikulin (2017), ind a posi i e e ec o in e ne and mobile
phone pene a ion on women's labou ma ke pa icipa ion. In he Indo-Paci ic, Wa son
e al (2018), also ound a posi i e co ela ion be ween in e ne use and women’s labou
ma ke pa icipa ion. In A ica, Nkoumou and Song (2021) ound ha ICTs inc ease
women’s pa icipa ion in he labou ma ke . These esul s a e consis en wi h hose
ob ained by Mainuddin e al. (2015); E obi e al. (2018) and Asongu and Odhiambo
(2020).
Secondly, no all indi iduals ha e access o and bene i om he po en ial o ICTs
(Bou ilie , 2019; Cummings & O'Neil, 2015). These inequali ies indica e ha he
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
128
po en ial e ec s o ICT can be o se by he ‘digi al di ide’ (Ragnedda & Musche ,
2013; Ma iscal, e al., 2020). The digi al di ide can ha e a signi ican impac on women’s
di e en ac i i ies, and in pa icula on hei in eg a ion in o he labou ma ke
(McKinsey Global Ins i u e, 2015; Alozie and Akpan-Obong, 2017). Coen-Pi ani and
Lugaue (2008), using mic oda a om he US, ound ha owning echnological de ices
did no inc ease he labou ma ke pa icipa ion o single women compa ed o ma ied
women, ye hey ep esen ed a ound a hi d o he obse ed popula ion. Simila ly,
Ca dia (2007) concluded ha inc eased adop ion o hese echnologies had no
signi ican e ec on emale labou o ce pa icipa ion a es in he US, bu did inc ease
hei pa icipa ion in he p o essions. The gende gap in economic pa icipa ion is
signi ican in A ica (Fab izio e al., 2020). S udies by Jiang and Luh (2017) and An onio
and Tu ley (2014) show ha women in hese coun ies ha e a much lowe a e o
access o echnology han men, which jus i ies hei low ep esen a ion in he labou
ma ke .
The lack o consensus in he li e a u e on he e ec s o ICTs on women’s
empowe men , as well as he lack o measu emen o ICTs, mo i a e u he esea ch on
he e ec s o he digi al di ide on women’s economic pa icipa ion. Gi en ha A ican
coun ies a e s ill in he ea ly s ages o ICT adop ion, and based on he a o emen ioned
li e a u e. No ha e p e ious s udies examined he e ec s o he digi al di ide on
women’s sec o al ac i i ies. This esea ch aims o ill hese gaps in he li e a u e. The
hypo hesis o his s udy is as ollows: The digi al di ide handicaps women’s economic
pa icipa ion in A ican coun ies.
3. Me hodology
3.1. Va iable and da a
Ou sample is composed o 45 A ican coun ies. We chose he pe iod 2000-2021
because o he a ailabili y o con inuous da a o a la ge numbe o coun ies. Table A1
in he Appendix lis s he coun ies, while Tables A2 and A3 de ine he desc ip i e
s a is ics o all a iables, and p esen he co ela ion ma ix espec i ely. Fou main
sou ces we e used o ob ain he da a. The dependen a iables, including women’s
economic pa icipa ion and women’s economic ac i i y sec o s (ag icul u e, indus y and
se ices) we e ob ained om he ILO da abase, he digi al di ide indica o s we e
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
129
ob ained om he Wo ld Bank’s WID (2022), as we e some economic con ol a iables
such as women’s educa ion, GDP pe capi a and na u al esou ces. Global go e nance
indica o s a e aken om he Poli ical Ins i u ions da abase and Poli y IV. De ini ions
and sou ces a e p o ided in Appendix A4.
3.1.1. Measu e o digi al di ide
Acco ding o Song e al (2020), h ee ca ego ies o indica o s can be used o
unde s and he digi al di ide. They ela e o: (i) access (compu e pene a ion, mobile
pene a ion, In e ne se ice p o ide s pe capi a, In e ne access p ices), (ii) use
(In e ne use s pe capi a, b oadband subsc ibe s pe capi a, ime spen online, In e ne
bandwid h pe capi a) and (iii) ou come (economics o e-comme ce, bene i s o online
shopping, e-lea ning ou comes, e-go e nmen ). Due o he lack o da a in some A ican
coun ies, h ee main measu es o he digi al di ide a e used: he mobile di ide (MD),
he In e ne di ide (ID) and he bandwid h di ide (BD). The economic li e a u e uses
wo app oaches o measu e he digi al di ide: absolu e and ela i e.
3.1.1.1.1. The Absolu e App oach
The absolu e app oach e alua es he gap o each selec ed indica o , he gap being
calcula ed as he di e ence be ween 100% ( he o al pene a ion a e) and he
pe cen age le el o he indica o (Rice & Ka z, 2003).
3.1.1.2. The Rela i e App oach
In his app oach, he digi al di ide indica o is calcula ed as he di e ence in
digi al pene a ion be ween he coun y assumed o be a he echnological on ie and
hose o o he coun ies o e he cou se o a yea . To his end, he coun y wi h he bes
pe o mance in digi al pene a ion is de e mined each yea . Fo ou sample case, his is
Sou h A ica because o i s abo e-a e age digi al pene a ion compa ed o he o he
A ican coun ies in ou sample. In addi ion, o his a icle we use he absolu e
app oach as he base a iable and we e i y ou obus ness using he ela i e app oach.
3.1.2. The dependen a iable: Measu ing women’s economic
pa icipa ion
The dependen a iable in his s udy is women’s economic pa icipa ion, cap u ed
by he women’s economic pa icipa ion index, acco ding o he new ILO (2018)
measu e, his a iable is calcula ed by women’s indi idual employmen and labou
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
130
ma ke pa icipa ion (on a scale o 0-100). ha women’s economic empowe men
implies Acco ding o he MDGs (2015), labou ma ke pa icipa ion is likely o p omo e
women’s economic independence. The choice o his a iable is consis en wi h he
con empo a y li e a u e on gende -sensi i e de elopmen (Achuo e al. 2022).
3.1.3. Con ol a iables
Based on he indings o he li e a u e, we ha e selec ed he ollowing con ol
a iables: con ol o co up ion, GDP pe capi a g ow h, seconda y educa ion o
women, na u al esou ce en s, emale popula ion and in la ion. Fo he con ol o
co up ion (con -co ), he use o public unds o pe sonal en ichmen , including pe y
and g and co up ion, and he “cap u e” o he s a e by eli es and p i a e in e es s.
Values ange om -2.5 (low) o +2.5 (high).
GDP pe capi a: his a iable is cap u ed by he GDP g ow h a e. Economic
g ow h is an impo an sou ce o income o women (Du lo, 2012). Jonson (2015) inds
ha an inc ease in GDP is associa ed wi h highe emale labou o ce pa icipa ion. This
inding is consis en wi h Anyanwu (2016) who demons a es ha g ow h in GDP pe
capi a inc eases gende equali y in women’s employmen . In line wi h he wo k o Du lo
(2012) we expec a posi i e sign om his indica o on women's economic pa icipa ion
in A ica.
Women’s seconda y educa ion (Edusec) cap u ed by he a e o women en olled
in seconda y educa ion, and conside ed in he li e a u e as a a iable o women’s
empowe men (Du lo, 2012; Asongu e al. 2020). Educa ion ends o b oaden poin s o
iew, educe e hnocen ism and he e o e inc ease lexibili y in accep ing new cus oms
and no ms. The le el o educa ion a ained by he gene al popula ion hus plays an
impo an ole in he g owing accep ance o he concep o gende equali y and
women’s empowe men (Anyanwu, 2012). Indeed, women wi h a high le el o
educa ion a e mo e likely o en e he labou ma ke in u ban a eas wo ldwide (Ogawa
and Ak e ,2007).
Na u al esou ce en s, he sum o oil en s, na u al gas en s, coal en s (ha d and
so ), mining en s and o es y en s. This a iable shows how he dis ibu ion o
na u al esou ce en s in luences women’s pa icipa ion in he labou o ce. The
expec ed sign o his a iable is nega i e. The emale popula ion (Pop emm) as a % o
he o al popula ion could be an indica o o women’s empowe men i hey ep esen
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
137
Table 1: E ec o he digi al di ide on women’s economic pa icipa ion
Women’s economic pa icipa ion index
OLS
(1)
(2)
(3)
(4)
(5)
(6)
Mobile phone
di ide
-0.077***
(0.012)
-0.104***
(0.018)
In e ne di ide
-0.298***
(0.032)
-0.373***
(0.047)
Bandwid h di ide
-0.094***
(0.025)
-0.159***
(0.041)
GDP_pe capi al
0.030***
(0.008)
0.034***
(0.008)
0.036***
(0.009)
Women’s educa ion
0.131***
(0.087)
0.223***
(0.090)
0.174***
(0.093)
Co up ion Con ol
0.203***
(0.987)
0.082**
(0.994)
0.148***
(1.101)
To al emale
popula ion
0.103***
(0.028)
0.078***
(0.025)
0.062**
(0.030)
To al income om
na u al esou ces
-0.148***
(0.015)
-0.116***
(0.017)
-0.174***
(0.015)
In la ion
0.639***
(0.147)
0.546***
(0.143)
0.657***
(0.167)
Cons an
42.824***
(1.601)
41.824***
(4.603)
30.707***
(2.425)
28.161***
(5.261)
43.624***
(2.544)
45.144***
(5.918)
Obse a ions
975
596
958
585
863
525
R-squa ed
0.033
0.200
0.068
0.243
0.014
0.183
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho based on WDI and OTI da a (2022)
4.3. Sensi i i y and obus ness analysis
4.3.1. Sensi i i y analysis o cul u al a iables
Sensi i i y analysis wi h cul u al a iables (such as alues, belie s, social no ms) is
c ucial o unde s anding how hese ac o s in luence di e en economic ou comes.
Adding cul u al a iables o he model allows us o check whe he ou esul s emain
s able and o co ec he p oblem o omi ed a iables. Ou esul s in Table (2) e eal
ha , by adding cul u al a iables o he model, he digi al di ide educes women's
economic pa icipa ion. Mo e speci ically, in his able, we ind ha cul u al a iables
such as linguis ic agmen a ion and Ca holic eligion signi ican ly inc ease women's
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
138
economic pa icipa ion. This can be explained by he ac ha colonial policies and
p ac ices a emp ed o ake in o accoun belie s, cus oms, language and cul u e. Fo
example, a F ench legal sys em sys ema izes he use o adi ional alues and ein o ces
social no ms. This conclusion is consis en wi h An onio and Tu ley (2014) and
Cummings and O'Neil (2015).
Howe e , he esul s also show ha in A ica, social iden i y in he o m o e hnic
agmen a ion and Muslim and P o es an eligious agmen a ion educes women's
abili y o access economic oppo uni ies. Indeed, Islam has been accused o educing
A ican women's sphe e o ac i i y o he home and he amily. Especially as a icula ed
by Islamic sages, i p opaga es an ideology o emale domes ici y. This con ibu es in no
small measu e o he "menagiza ion" o women in Muslim households ac oss he
con inen . This nega i e inding is consis en wi h ha o Suda kasa (1986) and Ongo e
al. (2022), who ound ha he Muslim eligion hinde s women's empowe men .
In addi ion, P o es an ism, like many eligions, has his o ically p omo ed
adi ional gende oles, whe e women we e o en seen as esponsible o he home and
child en, while men we e he main economic p o ide s. This can limi women’s
oppo uni ies o engage in economic ac i i ies ou side he home (Webe , 1905, No is &
Ingleha , 2011). Conse a i e in e p e a ions o P o es an eligious ex s may p omo e
he idea ha women should submi o male au ho i y, including in he economic
con ex . This pe spec i e can limi women's oppo uni ies o leade ship and economic
au onomy (Woodhead, 2012).
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
139
Table 2: cul u al a iables
Women’s economic pa icipa ion index
OLS
(1)
(2)
(3)
Mobile phone di ide
-0.071***
(0.012)
In e ne di ide
-0.213***
(0.028)
Bandwid h di ide
-0.106***
(0.022)
GDP_pe capi al
0.017*
(0.010)
0.011
(0.008)
0.012
(0.009)
Women’s educa ion
0.064***
(0.042)
0.118***
(0.045)
0.065***
(0.046)
Co up ion Con ol
1.013*
(0.967)
-1.464
(0.961)
-1.179
(1.101)
To al emale popula ion
0.114***
(0.032)
0.134***
(0.030)
0.143***
(0.032)
To al income om na u al esou ces
-0.128***
(0.013)
-0.116***
(0.015)
-0.136***
(0.016)
In la ion
0.540***
(0.117)
0.506***
(0.118)
0.504***
(0.128)
Ca holic
0.489***
(0.665)
0.605***
(0.656)
0.809***
(0.825)
Muslim
-0.036***
(0.083)
-0.071***
(0.818)
-0.070***
(0.098)
P o es an
-0.037***
(0.063)
-0.498***
(0.385)
-0.628***
(0.568)
E hnic agmen a ion
-0.162
(0.569)
-0.513*
(0.629)
-0.278*
(0.900)
Linguis ic agmen a ion
0.098***
(0.025)
0.068***
(0.048)
0.051***
(2.974)
Cons an
61.847***
(5.115)
55.507***
(4.905)
0.247***
(0.180)
Obse a ions
572
561
504
R-squa ed
0.640
0.653
0.638
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho based on WDI and OTI da a (2022)
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
140
4.3.2. Sensi i i y analysis based on coun y poli ical sys ems
The aim o his analysis is o assess whe he he ela ionship be ween he digi al
di ide and women's economic pa icipa ion is pe cei ed in he same way in di e en
coun ies, depending on he poli ical sys em. To his end, he analysis is conduc ed a
wo le els. On he one hand, we analyzed he ela ionship be ween he digi al di ide and
women's economic pa icipa ion in a democ a ic coun y and, on he o he , in an
au oc a ic coun y. In hese di e en g oups o coun ies, he esul s show ha in
democ a ic coun ies, he cell phone digi al di ide and bandwid h ha e no signi ican
consequence on women's economic pa icipa ion, whe eas he In e ne digi al di ide
hampe s women's economic pa icipa ion.
These esul s also show ha e o s o imp o e eedom o all in A ica since he
2000s ha e con ibu ed o he inclusion o gende in he economic sphe e, ha
democ acy is an essen ial elemen in ad ancing gende equali y, and ha women's access
o ICT inc eases hei pa icipa ion in he wo k o ce, as highligh ed by (S oup, 2008).
On he o he hand, in au oc a ic coun ies, all measu es o he digi al di ide signi ican ly
hinde women's economic pa icipa ion. In such a con ex , women-led sec o s o
economic ac i i y a e handicapped, po en ially hinde ing inclusi e de elopmen (S oup,
2008).
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
141
Table 3: The e ec o he digi al di ide on women’s economic pa icipa ion: by poli ical sys em
Women’s economic pa icipa ion index
OLS
Democ a ic sys em
Au oc a ic egime
Mobile phone
di ide
-0.064
(0.051)
-0.322***
(0.044)
In e ne di ide
-0.273*
(0.063)
-0.976***
(0.087)
Bandwid h
di ide
0.287
(0.069)
-0.462***
(0.058)
GDP_pe capi al
0.268**
(0.113)
0.143**
(0.067)
0.148**
(0.074)
0.005
(0.005)
0.018*
(0.006)
0.026*
(0.007)
Women’s
educa ion
0.303***
(0.269)
0.137***
(0.280)
0.190***
(0.310)
0.158
(0.179)
0.318**
(0.219)
0.596***
(0.243)
Co up ion
Con ol
0.013***
(3.690)
0.303***
(3.365)
0.888***
(2.929)
-0.805***
(4.035)
-0.612***
(3.746)
-0.040***
(4.410)
To al emale
popula ion
-0.067
(0.155)
-0.071
(0.131)
-0.050
(0.130)
-0.020
(0.074)
0.068
(0.080)
0.034
(0.101)
To al income
om na u al
esou ces
-0.230***
(0.063)
-0.219***
(0.065)
-0.225***
(0.061)
-0.002***
(0.053)
-0.063***
(0.058)
-0.075***
(0.060)
In la ion
-0.050
(0.203)
-0.065
(0.196)
-0.007
(0.198)
0.094***
(0.086)
0.013***
(0.061)
0.086***
(0.355)
Cons an
76.249***
(23.069)
63.438***
(17.776)
52.975***
(18.005)
16.344**
(7.551)
-14.721*
(8.600)
19.602**
(8.761)
Obse a ions
113
110
103
95
94
89
R-squa ed
0.450
0.490
0.456
0.699
0.713
0.660
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho based on WDI and OTI da a (2022)
4.4. Analysis o he obus ness o he esul s
The e a e se e al aspec s o he obus ness analysis o his s udy. Fi s , we
in oduce addi ional digi al di ide a iables o accoun o he in luence o he digi al
di ide on women’s economic pa icipa ion. Second, he esul s may be in luenced by he
measu e o women’s economic pa icipa ion han he o e all index, so we check he
obus ness by sec o al measu es o women’s economic pa icipa ion. Thi d, he esul s
may also be in luenced by he es ima ion echnique. The e o e, o es he obus ness o
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
142
ou main esul s, we pe o m wo signi ican obus ness es s in his subsec ion. Fi s ,
we check he boundedness o he dependen a iable wi h he TOBIT, censo ed
Poisson and unca ed nega i e binomial es ima o s and second, we pe o m he
obus ness es wi h he quan ile eg ession (QR) me hod o make ou esul s mo e
obus . O e all, in all obus ness es s, we ind speci ica ion esul s equi alen o hose
in he base able.
4.4.1. Al e na i e measu es o b idge he digi al di ide
The p e ious sec ion used he absolu e digi al di ide. This measu e may gi e ise
o a limi ed examina ion o he link be ween he digi al di ide and women's economic
pa icipa ion. This sec ion he e o e conside s he ela i e digi al di ide. The ole o his
indica o is o ein o ce he esul o he digi al di ide on women's economic
pa icipa ion. O e all, he esul s alida e ha he digi al di ide sys ema ically hinde s
women's economic pa icipa ion in A ica. The ampli ude is highe wi h he b oadband
channel and lowe wi h he cell phone channel. So, while he e ogenei y pe sis s o
o he digi al channels, A ican coun ies seem o be con e ging owa ds cell phone use,
which jus i ies he low magni ude o i s digi al di ide. The low dispa i y in cell phone
use ends o inc ease women's economic pa icipa ion. On he one hand, i cell phones
we e he only digi iza ion channel, A ican coun ies would end o ha e high emale
economic pa icipa ion a es.
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
143
Table 4: Robus ness check using ela i e digi al di ide componen s
2
Women’s economic pa icipa ion index
OLS
(1)
(2)
(3)
Mobile phone di ide (R)
-0.040**
(0.018)
In e ne di ide (R)
-0.161***
(0.041)
Bandwid h di ide (R)
-0.513**
(0.071)
GDP_pe capi al
0.024**
(0.010)
0.036***
(0.009)
0.021**
(0.010)
Women’s educa ion
0.063***
(0.086)
0.174***
(0.093)
0.012***
(0.090)
Co up ion Con ol
0.792***
(1.046)
0.138***
(1.094)
0.672***
(0.988)
To al emale popula ion
0.047*
(0.027)
0.062**
(0.030)
0.048*
(0.028)
To al income om na u al esou ces
-0.173***
(0.014)
-0.173***
(0.015)
-0.165***
(0.017)
In la ion
0.564***
(0.145)
0.626***
(0.158)
0.620***
(0.140)
Cons an
60.232***
(3.206)
47.567***
(5.385)
-14.541
(36.179)
Obse a ions
589
531
597
R-squa ed
0.160
0.184
0.158
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho based on WDI and OTI da a (2022)
4.4.2. Acco ding o women’s economic ac i i y sec o s
This obus ness analysis is impo an because i allows us o check whe he he
ela ionships obse ed be ween he digi al di ide and women’s economic pa icipa ion
a e solid and consis en , e en when di e en sec o s o women’s economic ac i i y a e
2
(R) Rep esen s he di e en measu es o he ela i e digi al di ide
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
144
aken in o accoun . In o he wo ds, examining he obus ness o he esul s ensu es ha
he ela ionships be ween he digi al di ide and women's sec o al ac i i y a iables a e
no simply he esul o coincidence o s a is ical bias. This s eng hens he alidi y o
he conclusions d awn om he analysis and p o ides a mo e solid basis o poli ical o
economic decision-making. Thus, in his sec ion, we es he e ec s o he digi al di ide
on sec o al economic pa icipa ion using he h ee sub-indices o women’s economic
pa icipa ion (ag icul u e, indus y and se ices) as dependen a iables in he dynamic
model, again wi h OLS es ima ion.
Indeed, he nega i e ela ionship be ween he digi al di ide and women's
pa icipa ion in he indus ial sec o may e lec he ac ha A ica has expe ienced an
access and usage gap in new p oduc s and p ocesses, a low a e o inno a ion in a ious
coun ies ha e a knock-on e ec on indus ies. Fu he mo e, he ac ha he digi al
di ide is less likely o ha e an impac on women's economic pa icipa ion in he
ag icul u al sec o may be jus i ied by he high isibili y o women in his sec o . These
esul s a e close o hose o Ndubuisi e al (2021). Finally, economic a iables such as
GDP g ow h, emale popula ion, co up ion con ol and in la ion a e almos all
posi i ely and signi ican ly associa ed wi h women's pa icipa ion in he h ee economic
sec o s, while women's educa ion and na u al esou ce en s a e nega i ely associa ed
wi h women's economic ac i i y.
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
145
Table 5: The e ec o he digi al di ide on women’s sec o al economic pa icipa ion
OLS
VARIABLES
Ag icul u e
Indus y
Se ices
Ag icul u e
Indus y
Se ices
Ag icul u e
Indus y
Se ices
(1)
(2)
(3)
(5)
(6)
(7)
(8)
(9)
(10)
Mobile phone di ide
0.024***
(0.093)
-0.116***
(0.574)
-10.665***
(2.333)
In e ne di ide
-0.500
(0.212)
-0.764***
(0.529)
-28.586***
(7.191)
Bandwid h di ide
0.036***
(0.053)
-0.350
(0.985)
-5.477
(4.046)
GDP_pe capi al
0.555**
(0.726)
0.022
(0.545)
1.034
(2.154)
0.350*
(0.740)
0.020
(0.560)
1.011
(2.211)
0.087**
(0.776)
-0.007
(0.473)
0.748
(1.835)
Women’s educa ion
0.758
(0.399)
0.286***
(0.861)
0.282***
(7.051)
0.869*
(0.834)
3.500***
(2.086)
8.997***
(7.867)
0.959*
(6.592)
6.528**
(2.115)
23.990***
(6.993)
Co up ion Con ol
0.840**
(0.168)
0.730***
(0.899)
0.084***
(0.984)
0.061***
(0.066)
0.049***
(0.082)
0.055***
(0.079)
0.629**
(0.859)
0.612***
(0.775)
0.176***
(0.054)
To al emale popula ion
0.292
(0.508)
-1.096
(0.877)
-6.853**
(3.401)
-1.228
(2.497)
-0.958
(0.867)
-5.713*
(3.340)
1.535
(2.615)
0.567
(0.857)
-0.510
(3.180)
To al income om
na u al esou ces
-0.051***
(0.015)
-0.957***
(0.646)
-0.287***
(0.602)
-0.224***
(0.461)
-4.615***
(0.724)
-0.402***
(0.037)
-0.578***
(0.072)
-3.276***
(0.571)
-0.822***
(0.274)
In la ion
0.034***
(0.014)
2.838
(2.671)
0.592*
(0.772)
0.095***
(0.409)
0.384
(0.735)
0.207**
(0.014)
0.048***
(0.907)
3.845
(2.976)
0.585**
(0.515)
Cons an
1,735.686***
(432.570)
983.980***
(179.413)
4,115.365***
(752.273)
2,668.751***
(574.676)
1,160.252***
(214.375)
4,911.830***
(971.104)
1,429.218***
(450.597)
526.878***
(167.363)
2,385.525***
(711.308)
Obse a ions
596
596
596
585
585
585
525
525
525
R-squa ed
0.106
0.095
0.101
0.104
0.104
0.110
0.106
0.085
0.083
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
146
4.4.3. Checking he limi ed na u e o he dependen a iable
Since he dependen a iable is limi ed o he in e al [0-100], he use o OLS o
simila me hods could lead o biased esul s. This is because OLS is no sui able o
limi ed dependen a iables wi h a high deg ee o di e si y. A dependen a iable may
be con inuous o e one o mo e in e als, bu may also ake on speci ic alues wi h a
de ined p obabili y. Models o limi ed dependen a iables a e designed o handle
samples ha a e ei he unca ed o censo ed. To co ec o his bias, speci ic
es ima o s can be used. In his s udy, we use TOBIT, censo ed Poisson and unca ed
nega i e binomial models.
These models a e called “coun ing models” because hey coun he occu ences o
an e en . Mo e p ecisely, hey ake in o accoun he p oblems o censo ing and
unca ion. A sample is conside ed unca ed i ce ain obse a ions ha should ha e
been included ha e been sys ema ically omi ed. Con e sely, a sample is said o be
censo ed when no obse a ions ha e been excluded, bu pa o he in o ma ion has
been masked. These si ua ions can occu wi h he ex eme alues (0 and 100) o
women's economic pa icipa ion. The esul s p esen ed in Table 6, whe he o no
egional ixed e ec s a e in ol ed, con i m he p e ious obse a ions: he digi al di ide
in A ica is a eali y.
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
153
Table 8: IV-LEWBEL, (2012)
(1)
(2)
(3)
VARIABLES
WEP
Mobile phone digi al di ide
-0.00928***
(0.00139)
In e ne digi al di ide
-0.0250***
(0.00428)
Bandwid h digi al di ide
-0.0139***
(0.00284)
GDP_pe capi al
0.431***
(0.101)
0.448***
(0.0994)
0.516***
(0.109)
Women’s educa ion
-0.00127
(0.0333)
0.0468
(0.0323)
0.0647*
(0.0348)
Co up ion Con ol
0.000100
(0.00158)
0.000256
(0.00159)
-0.000529
(0.00181)
To al emale popula ion
2.624***
(0.226)
2.426***
(0.228)
2.476***
(0.261)
To al income om na u al esou ces
-0.0200**
(0.00777)
-0.00851
(0.00714)
-0.0153*
(0.00894)
In la ion
-8.11e-05
(0.00505)
-0.00290
(0.00407)
-0.00236
(0.00449)
Cons an
-0.861***
(0.120)
-0.746***
(0.122)
-0.786***
(0.139)
Obse a ions
637
633
561
R-squa ed
0.993
0.993
0.993
Kleibe gen-Paap k Wald F s a is ic
144.3***
156.4***
127.3***
Sande son-Windmeije
144.3
156.4
127.3
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho s based on WDI and OTI da a (2022)
4.4.5.2. IV-2SLS LIML and GMM2S app oach
Second, in con as o he simple IV-2SLS app oach, we apply he IV-2SLS and
IV-GMM2S Limi ed In o ma ion Maximum Likelihood (LIML), espec i ely. The IV-
GMM2S app oach yields mo e e icien es ima ion coe icien s wi h s anda dised e o s
based on he wo-s age gene alised me hod o momen s. In IV-GMM2S, he e icien o
op imal weigh ing ma ix is he in e se o an es ima e o he co a iance ma ix o he
o hogonali y condi ions. The e iciency gains o his es ima o compa ed wi h he
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
154
adi ional IV-2SLS es ima o come om he use o he op imal weigh ing ma ix, he
es ic ions on o e -iden i ica ion o he model and he elaxa ion o he o hogonali y
condi ions. Fo an exac ly iden i ied model, he GMM2S and adi ional IV-2SLS
es ima o s coincide, and unde he assump ions o condi ional homoscedas ici y and
independence, he GMM2S es ima o also coincides wi h he adi ional IV-2SLS
es ima o (Da idson and MacKinnon, 1993; Baum, 2006; Hayashi, 2011). In con as ,
he IV-2SLS-LIML app oach con ols o he p oblem o weak ins umen s.
Pa ame e in e ence is based on he consis en es ima ion o a maximum o ou
e ms o he asymp o ic a iance, while he speci ica ion es is based on he
asymp o ically no mal (o asymp o ically equi alen , op ionally adjus ed chi2)
dis ibu ion o he o e iden i ying es o he pa ame e s a is ic (Fulle , 1977; Da idson
and MacKinnon, 1993; Baum, 2006; Ana olye , 2019). The esul s o his analysis a e
shown in Table 9 below. Analysis o his able allows us o see ha he coe icien s
es ima ed by he IV-LIML and GMM2S me hods a e ela i ely di e en om hose o
he adi ional IV-2SLS which may e lec he in ui ion ha he baseline esul s
ollowing he IV-2SLS app oach ha e been co ec ed o he low ins umen a ion
p oblem while p o iding mo e e icien coe icien s. Ou esul s con i m he baseline
esul s in he sense ha he digi al di ide educes women's economic pa icipa ion. The
esul s conce ning he quali y o he ins umen s a e sa is ac o y. Rega ding he
ele ance o he ins umen s, he Kleibe gen -Paap Wald k F-s a is ic is used o es he
weakness o he ins umen s (Kleibe gen and Paap, 2006). The Kleibe gen -Paap Wald
k F-s a is ic mus be a leas 10 o weak iden i ica ion no o be conside ed a p oblem
(Tadadjeu e al., 2023b; Nguena and Pi a, 2023; Mignamissi e al., 2024).
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
155
Table 9: IV-LEWBEL, (2012) LIML and GMM2S
(1)
(2)
(3)
(4)
(5)
(6)
LIML
GMM2S
VARIABLES
WEP
Mobile
phone digi al
di ide
-0.0136***
(0.00216)
-0.00898***
(0.000509)
In e ne
digi al di ide
-0.0285***
(0.00523)
-0.0259***
(0.00202)
Bandwid h
digi al di ide
-0.0144***
(0.00325)
-0.0140***
(0.00113)
GDP_pe
capi al
0.303***
(0.106)
0.412***
(0.101)
0.510***
(0.111)
0.375***
(0.0497)
0.346***
(0.0538)
0.462***
(0.0487)
Women’s
educa ion
-0.0475
(0.0369)
0.0397
(0.0327)
0.0634*
(0.0351)
0.0182
(0.0145)
0.0327**
(0.0146)
0.0611***
(0.0190)
Co up ion
Con ol
-0.000148
(0.00153)
0.000197
(0.00157)
-0.000544
(0.00180)
0.000154
(0.000686)
0.00126
(0.000841)
0.00112
(0.000808)
To al emale
popula ion
2.691***
(0.226)
2.419***
(0.225)
2.473***
(0.261)
2.333***
(0.113)
2.144***
(0.117)
2.440***
(0.117)
To al income
om na u al
esou ces
-0.0199***
(0.00745)
-0.00688
(0.00704)
-0.0150*
(0.00892)
-0.0233***
(0.00591)
-0.00963*
(0.00532)
-0.0158**
(0.00632)
In la ion
-0.00111
(0.00637)
-0.00357
(0.00426)
-0.00252
(0.00458)
0.00251
(0.00386)
-0.00462*
(0.00274)
-0.00228
(0.00312)
Cons an
-0.888***
(0.120)
-0.738***
(0.121)
-0.784***
(0.139)
-0.712***
(0.0593)
-0.595***
(0.0625)
-0.767***
(0.0619)
Obse a ions
637
633
561
637
633
561
R-squa ed
0.993
0.993
0.993
0.993
0.993
0.993
Kleibe gen-
Paap k Wald
F s a is ic
144.3
156.4
127.3
144.3
156.4
127.3
Sande son-
Windmeije
144.3
156.4
127.3
144.3
156.4
127.3
Robus s anda d e o s in pa en heses *** p<0.01, ** p<0.05, * p<0.1
Sou ce: Au ho s based on WDI and OTI da a (2022)
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
156
5. Conclusion
This a icle aims o in es iga e he ela ionship be ween he digi al di ide and
women's economic pa icipa ion. Based on da a co e ing 45 A ican coun ies and using
pa ame ic and non-pa ame ic app oaches, we ound e idence o he digi al di ide
cu se on women's economic pa icipa ion. Speci ically, we ound ha he digi al di ide
ends o educe women's economic pa icipa ion, bu ha his e ec di e s acco ding o
poli ical sys em, ypes o economic ac i i y and di e en digi al di ide indices. Indeed,
by s udying he he e ogenei y o esul s be ween coun ies, we ound ha he digi al
di ide educes economic pa icipa ion mo e in mo e ac u ed egions and coun ies
wi h weak democ acies.
These esul s emain b oadly obus when we use o he measu es o he digi al
di ide. To pu he ela ionship be ween he digi al di ide and women's economic
pa icipa ion in o pe spec i e, we employed a non-pa ame ic app oach using he
quan ile eg ession echnique. The esul s show ha he in luence o he digi al di ide
a ies a di e en in e als o women's economic pa icipa ion.
On he basis o ou analysis, wo main ecommenda ions, among o he s, can be
iden i ied: (i) he main ecommenda ion would be o educe he global digi al di ide by
imp o ing digi al inclusion o all, alongside inancial inclusion, in o de o popula ise
he digi al echnologies and se ices needed o business de elopmen and women's
en ep eneu ship; (ii) imp o e gende equali y in he wo k o ce, by making labou
ma ke s mo e inclusi e and imp o ing access o ICTs in o de o educe in o ma ion
asymme ies and elimina e he gende gap in decision-making.
Limi a ion: his s udy may be limi ed by a lack o eliable da a. In addi ion, he
a ailable da a may no e lec all dimensions o he digi al di ide, such as connec ion
quali y and digi al skills. Fu he mo e, he me hod used does no ake in o accoun a
causal ela ionship be ween he digi al di ide and women's economic pa icipa ion.
Fu u e wo k: Ca y ou longi udinal s udies acking changes in he digi al di ide
and hei impac on women's economic pa icipa ion o e a long pe iod. Compa ing
hese dynamics be ween di e en coun ies o egions could also p o ide aluable
insigh s in o he ac o s ha p omo e o hinde his ela ionship. Fo example,
compa ing he impac o he digi al di ide on women's economic pa icipa ion in
coun ies wi h ad anced digi al policies and de eloping coun ies.
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
157
Appendices
Table A1: lis o coun ies
Alge ia, Angola, Benin, Bo swana, Bu kina-Faso, Bu undi, Came oon, Cen al A ican
Republic, Chad, Como os, Democ a ic Republic o Congo, Congo, Djibou i, Gabon,
Equa o ial Guinea, Cô e d'I oi e, E i ea, E hiopia, Gambia, Ghana, Guinea, Guinea-
Bissau, Kenya, Leso ho, Libe ia, Libya, Madagasca , Malawi, Mali, Mo occo, Mau i ius,
Mau i ania, Mozambique, Namibia, Nige , Nige ia, Rwanda, Sao Tome-and-P incipe,
Senegal, Sie a-Leone, Swaziland, Tanzania, Togo, Tunisia, Uganda, Zambia, Zimbabwe.
Table A2: Desc ip i e s a is ics
Va iable
Obs.
Mean
S anda d
de ia ion
Min
Max
WEP
989
53.302
18.698
12.335
87.123
Mobile phone di ide
976
134.556
44.486
0
185.559
In e ne di ide
959
76.567
16.249
0
88.13
Hea y Band di ide
864
101.215
23.635
0
115.675
GDP_pe capi a
949
10.399
31.195
-24.847
558.56
Women’s educa ion
919
11.316
11.505
.001
67.89
Co up ion Con ol
935
-.771
.633
-2.45
1.161
Female Popula ion
729
97.66
24.237
16.626
151.314
To al na u al
esou ces
850
70.916
40.059
.785
347.997
In la ion
921
4.424
8.138
-11.199
103.337
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
158
Table A3 : Co ela ions
Va iables
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(1) WEP
1.000
(2) Mobile di ide (A)
-0.182
(0.000)
1.000
(3) Mobile di ide (R)
-0.048
(0.140)
-0.305
(0.000)
1.000
(4) In e n di ide (A)
-0.261
(0.000)
0.727
(0.000)
-0.248
(0.000)
1.000
(5) In e n di ide (R)
-0.114
(0.001)
0.722
(0.000)
-0.167
(0.000)
0.669
(0.000)
1.000
(6) Bandep di ide (A)
-0.117
(0.001)
0.749
(0.000)
-0.175
(0.000)
0.669
(0.000)
1.000
(0.000)
1.000
(7) Bandep di ide (R)
-0.235
(0.000)
0.025
(0.445)
-0.137
(0.000)
0.109
(0.001)
-0.050
(0.137)
-0.047
(0.165)
1.000
(8) GDP_pe capi a
0.014
(0.660)
0.069
(0.034)
0.009
(0.776)
0.058
(0.079)
-0.024
(0.479)
-0.023
(0.503)
0.078
(0.016)
1.000
(9) Women’s educa ion
-0.024
(0.468)
0.083
(0.012)
0.056
(0.092)
0.242
(0.000)
0.215
(0.000)
0.210
(0.000)
-0.307
(0.000)
0.115
(0.001)
1.000
(10) Co up ion con ol
0.112
(0.001)
-0.038
(0.244)
0.034
(0.299)
-0.029
(0.387)
-0.046
(0.181)
-0.055
(0.115)
0.135
(0.000)
-0.052
(0.122)
-0.231
(0.000)
1.000
(11) Female popula ion
0.145
(0.000)
-0.265
(0.000)
0.036
(0.328)
-0.174
(0.000)
-0.167
(0.000)
-0.170
(0.000)
0.182
(0.000)
0.000
(0.996)
-0.096
(0.010)
0.103
(0.007)
1.000
(12) Na u al esou ces
-0.253
(0.000)
-0.246
(0.000)
0.016
(0.645)
-0.333
(0.000)
-0.070
(0.055)
-0.073
(0.048)
-0.266
(0.000)
0.004
(0.904)
0.037
(0.289)
0.164
(0.000)
0.090
(0.023)
1.000
(13) In la ion
0.035
(0.285)
-0.001
(0.973)
0.123
(0.000)
0.026
(0.443)
0.065
(0.061)
0.063
(0.071)
-0.111
(0.001)
0.011
(0.732)
0.059
(0.074)
0.077
(0.023)
0.020
(0.601)
0.376
(0.000)
1.000
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
159
Table A 4: Va iable de ini ions
Va iables
Desc ip ion
Sou ces
Women’s
pa icipa ion in
he labou ma ke
Employmen /popula ion a io, 15–64-yea -olds,
women (%) (ILO modelled es ima e).
OIT (2022)
In e ne digi al
di ide
In e ne access and usage pe capi a gap indica o
(calcula ed om WDI da a)
WDI (2022)
Mobile elephone
di ide
Pe capi a elephone access and use indica o
(calcula ed om WDI da a)
WDI (2022
GDP_pe capi al
G oss domes ic p oduc di ided by popula ion a
mid-yea
WDI (2022)
Women’s
educa ion
Numbe o gi ls en olled in seconda y school.
WDI (2022)
Financial
de elopmen
Volume o inancial esou ces p o ided o he
p i a e sec o by inancial ins i u ions.
WDI (2022)
Female
popula ion
P opo ion o women in he o al popula ion.
WDI (2022)
Human capi al
Mean yea s o schooling
UNDP
Con ol o
co up ion
cap u es pe cep ions o he ex en o which public
powe is exe cised o o ms o co up ion
GTI (2022)
Na u al Rescue
annui y
Sum o oil en s, na u al gas en s, coal en s (ha d
and so ), mining en s and o es y en s.
WDI (2022)
In la ion
e lec ing he annual pe cen age change in he
cos o he a e age consume o acqui ing a
baske o goods and se ices ha may be ixed o
modi ied a speci ied in e als, o example each
yea
WDI (2022)
Digi al di ide in
he panhandle
Pe capi a elephone access and use indica o
(calcula ed om WDI da a)
WDI (2022)
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
160
Re e ences
Achuo E. D., Asongu S., Tchamyou V. (2022), ‘Women empowe men and en i onmen al
sus ainabili y in A ica’, AGDI Wo king Pape , WP/22/004.
Alozie N. O., Akpan‐Obong P. (2017), ‘The digi al gende di ide: Con on ing obs acles o
women's de elopmen in A ica’, De elopmen Policy Re iew, 35(2), 137-160.
Ambe H., Chichaibelu B. B. (2023), ‘Na owing he gende digi al di ide in Pakis an: Mobile
phone owne ship and emale labou o ce pa icipa ion’, Re iew o De elopmen Economics, 27(3),
1354-1382.
Ana olye S. (2019), ‘Many ins umen s and/o eg esso s: A iendly guide’, Jou nal o Economic
Su eys, 33(2), 689-726.
Ang J. B. (2013), ‘A e mode n inancial sys ems shaped by s a e an iqui y?’, Jou nal o Banking &
Finance, 37, 4038–4058. A ailable a : h ps://doi.o g/10.1016/j.jbank in.2013.07.023
An onio A., Tu ley D. (2014), ‘The gende digi al di ide in de eloping coun ies’, Fu u e
In e ne , 6(4), 673-687.
Anyanwu J. C. (2012), ‘Accoun ing o Female Employmen in A ica’, Eu opean Economic Le e s,
1(1), 14–26.
Anyanwu J. C. (2016), ‘Analysis o gende equali y in you h employmen in A ica’, A ican
De elopmen Re iew, 28, 397–415.
A ellano M., Bond S. (1991), ‘Some es s o speci ica ion o panel da a: Mon e Ca lo e idence
and an applica ion o employmen equa ions’, The e iew o economic s udies, 58(2), 277-297.
Asongu S. A., Odhiambo N. M. (2020), ‘Inequali y and gende inclusion: Minimum ICT policy
h esholds o p omo ing emale employmen in Sub-Saha an A ica’, Telecommunica ions Policy,
44(4).
A om D., Ndoya H. (2022), ‘Does coun y s abili y spu economic complexi y? E idence om
panel da a’, Applied Economics Le e s, 1-7.
Awoa P. A., Ondoa H. A., Tabi H. N. (2022), ‘Women's poli ical empowe men and na u al
esou ce cu se in de eloping coun ies’, Resou ces Policy, 75, 102442.
Baum A., Poplawski-Ribei o M., Webe A. (2012), ‘Fiscal Mul iplie s and he S a e o he
Economy’, IMF Wo king Pape , WP/12/286, In e na ional Mone a y Fund
Baum T. (2006), ‘Re lec ions on he na u e o skills in he expe ience economy: Challenging
adi ional skills models in hospi ali y’, Jou nal o Hospi ali y and Tou ism Managemen , 13(2), 124-
135.
Becke G. S. (1965), ‘A Theo y o he Alloca ion o Time’, The economic jou nal, 75(299), 493-517.
Blundell R., Bond S. (1998), ‘Ini ial condi ions and momen es ic ions in dynamic panel da a
models’, Jou nal o econome ics, 87(1), 115-143.
Bou ilie S. (2019), ‘G owing Up Malala: Is Today’s Educa ed Gi l Tomo ow’s Neolibe al
Woman?’, J. In . De , 32(2), 186–206. A ailable a : h ps://doi.o g/10.1002/jid.3442
Ca lsson-Kanyama A., Lind´en, A. L. (2007), ‘Ene gy e iciency in esidences—challenges o
women and men in he No h’, Ene gy Policy, 35(4), 2163–2172. A ailable a :
h ps:// doi.o g/10.1016/j.enpol.2006.06.018
Cas ells M. (2002), The In e ne galaxy: Re lec ions on he In e ne , business, and socie y, Ox o d
Uni e si y P ess, USA.
Coen-Pi ani D., León A., Lugaue S. (2008), ‘The e ec o household appliances on emale
labou o ce pa icipa ion: E idence om mic o da a’, Teppe School o Business Pape , 54.
Co den W.M., Nea y J.P. (1982), ‘Booming sec o and de-indus ialisa ion in a small open
economy’, The economic jou nal, 92(368), 825-848.
Empi ical e i ica ion o he link be ween he digi al di ide and women's economic pa icipa ion in A ica
A ailable online a h ps://ejce.liuc.i
161
Cummings C., O’Neil T. (2015), Do digi al in o ma ion and communica ions echnologies inc ease he oice
and in luence o women and gi ls?, O e seas De elopmen Ins i u e.
Cza necka M. e al. (2022), ‘Social media engagemen in shaping g een ene gy business models’,
Ene gies, 15, 1727. h ps://doi.o g/10.3390/en15051727
Dao M.C., Liu L.Q. (2017), Finance and employmen in de eloping coun ies: The wo king capi al channel,
In e na ional Mone a y Fund.
Das G.G., D ine I. (2020), ‘Dis ance om he echnology on ie : How could A ica ca ch-up
ia socio-ins i u ional ac o s and human capi al?’, Technological Fo ecas ing and Social Change, 150,
119755.
Da idson R., MacKinnon J.G. (1993), Es ima ion and in e ence in econome ics, Ox o d Uni e si y
p ess.
Dencik L., Leis e O. (Eds.), (2015), C i ical pe spec i es on social media and p o es : Be ween con ol and
emancipa ion, Rowman & Li le ield.
De ling L. J. (2017), ‘B oadband in he Labo Ma ke : The impac o esiden ial high-speed
in e ne on ma ied women’s labo o ce pa icipa ion’, In e na ional Labo Re iew, 70, 451–482.
Domguia E. N. e al. (2022), Does en i onmen al ax kill employmen ? E idence om OECD
and non-OECD coun ies, Jou nal o Cleane P oduc ion, 380, 134873.
Doumbia A. G., Kuepie M. (2008), ‘Ni eau e de e minan s de l’inse ion des emmes su le
ma ch´e du a ail au Mali’, S a eco, 103, 61–75.
Du lo E. (2012), ‘Women empowe men and economic de elopmen ’, Jou nal o Economic
li e a u e, 50(4), 1051-1079.
Edo S., Okodua H., Odebiyi J. (2019), ‘In e ne adop ion and inancial de elopmen in Sub‐
Saha an A ica: E idence om Nige ia and Kenya’, A ican De elopmen Re iew, 31(1), 144-160.
E obi U.R., Tanaken B.V., Asongu S.A. (2018), ‘Female economic pa icipa ion wi h
in o ma ion and communica ion echnology ad ancemen : E idence om Sub-Saha an
A ica’, Sou h A ican Jou nal o Economics, 86(2), 231–246.
Fab izio S. e al. (2020), ‘Women in he labo o ce: The ole o iscal policies’, IMF S a
Discussion No e, Feb ua y, In e na ional Mone a y Fund.
Fallon K.M., Bou ilie S. (2022), ‘The digi al di ide wi hin he women’s mo emen in Ghana:
Implica ions o oice and inclusion’, Social Mo emen S udies, 21(5), 677-696.
Galpe in H., A cidiacono M. (2021), ‘Employmen and he gende digi al di ide in La in
Ame ica: A decomposi ion analysis’, Telecommunica ions Policy, 45(7), 102166.
G anjon F. (2009), ‘Inégali és numé iques e econnaissance sociale : Des usages populai es de
l’in o ma ique connec ée’, Les Cahie s du numé ique, 5(1), 19-44.
Gu umu hy A. (2004), GENDER and ICTs, Ins i u e o De elopmen S udies.
Ha eez M. e al. (2020), ‘The impac o ICT on women empowe men : E idence om selec ed
Sou h Asian coun ies’, Nus Jou nal o Social Sciences and Humani ies, 6(1), 45–61.
Hayashi F. (2011), Econome ics. P ince on Uni e si y P ess.
Ince M. (2010), ‘How he educa ion a ec s emale labo o ce? Empi ical e idence om
Tu key’, P ocedia Social and Beha io al Sciences, 2, 634–639.
Jiang W. J., Luh Y.H. (2017), ‘Gende digi al di ide in a pa ia chal socie y: wha can we lea n
om Blinde –Oaxaca decomposi ion?’, Quali y & Quan i y, 51(6), 2555-2576.
Kanjilal-Bhadu i S., Pas o e F. (2018), ‘Re u ns o educa ion and emale pa icipa ion nexus:
E idence om India’, The Indian Jou nal o Labou Economics, 61, 515-536.
Klasen S., Pie e s J. (2012), ‘Push o pull? D i e s o emale labo o ce pa icipa ion du ing
India’s economic boom’, IZA Discussion Pape , 6395.
EJCE, ol. 21, no. 1 (2024)
A ailable online a h ps://ejce.liuc.i
162
Klasen S. e al. (2021), ‘Wha d i es emale labou o ce pa icipa ion? Compa able mic o-le el
e idence om eigh de eloping and eme ging economies’, The Jou nal o De elopmen
S udies, 57(3), 417-442.
Kleibe gen F., Paap R. (2006), ‘Gene alized educed ank es s using he singula alue
decomposi ion’, Jou nal o econome ics, 133(1), 97-126.
Koenke R., Basse J G. (1978), ‘Reg ession quan iles’, Econome ica: jou nal o he Econome ic
Socie y, 33-50.
Lewbel A. (2012), ‘Using he e oscedas ici y o iden i y and es ima e mismeasu ed and
endogenous eg esso models’, Jou nal o Business & Economic S a is ics, 30(1), 67-80.
Liu X., Liu Z. (2022), ‘Explo ing he impac o media discou se on social pe cep ions owa ds
biomass ene gy u iliza ion in China’, Ene gy S a egy Re iews, 42, A icle 100896. A ailable a :
h ps://doi.o g/10.1016/j.es .2022.100896
Mainuddin A. K. M. e al. (2015), ‘Women empowe men and i s ela ion wi h heal h seeking
beha io in Bangladesh’, Jou nal o amily & ep oduc i e heal h, 9(2), 65.
Mankiw N.G. (2001), ‘The inexo able and mys e ious adeo be ween in la ion and
unemploymen ’, The Economic Jou nal, 111(471), 45-61.
Mazzone A. (2022), ‘Gende and ene gy in in e na ional de elopmen : Is he e a e u n o he
‘ eminiza ion’ o po e y discou se?’, De elopmen , 65(1), 17–28. A ailable a :
h ps:// doi.o g/10.1057/s41301-022-00330-7
McKinsey Global Ins i u e (2015), Digi al globaliza ion: The new e a o global lows, Ma ch. McKinsey
and Company.
Me alopulos P. (2008), De elopmen communica ion sou cebook: B oadening he bounda ies o
communica ion, Wo ld Bank Publica ions.
Mignamissi D., Kue e Y. F. M. (2021), ‘Resou ce en s and happiness on a global pe spec i e:
The esou ce cu se e isi ed’, Resou ces Policy, 71, 101994.
Mignamissi D. e al. (2024), ‘Does ade openness inc ease CO2 emissions in A ica? A
e alua ion using he composi e index o Squalli and Wilson’, En i onmen Sys ems and Decisions,
1-29.
Mince J. (1962), ‘Labo o ce pa icipa ion o ma ied women: A s udy o labo supply’,
in Aspec s o labo economics (pp. 63-105). P ince on Uni e si y P ess.
Moolman J., P imo N., Shackle on S. J. (2007), ‘Taking A by e o echnology: Women and
ICTs’, Agenda: Empowe ing Women o Gende Equi y, 21(71), 4–14. A ailable a :
h ps://www.js o .o g/s able/27739232
Mo in-Sylla M.H. (2005), The gende digi al di ide in F ancophone A ica: A ha sh eali y, ENDA
Thi d Wo ld, Daka , ISBN 92-95049-11.
Ndubuisi G., O iom C., Te eh G. K. (2021), Digi al in as uc u e and employmen in se ices: E idence
om Sub-Saha an A ican coun ies. In Fo hcoming (Ed.). Telecommunica ions Policy.
Ngoa G. B. N., Song, J. S. (2021), ‘Female pa icipa ion in A ican labo ma ke s: The ole o
in o ma ion and communica ion echnologies’, Telecommunica ions Policy, 45(9), 102174.
Ngounou B. A. e al. (2023), ‘Inclusi e g ow h in he ace o inc easing u baniza ion: Wha
expe ience o A ican coun ies?’, Re iew o De elopmen Economics.
Nguena C. L. L., Asaloko P. P. (2023), ‘Financial inclusion, g ow h and po e y: E idence om
A ica in COVID-19 e a’, Région e Dé eloppemen , (57), 81-100.
Nikulin D. (2017), ‘The impac o ICT on women’s economic empowe men ’, in H. Kau , E.
Lechman, A. Ma szk, e al. (Eds.), (di ), ca alyzing de elopmen h ough ICT adop ion: The de eloping
wo ld expe ience (pp. 15–24). Sp inge .
Njoh A. J., Akiwumi F. A. (2012), ‘The impac o eligion on women empowe men as a
millennium de elopmen goal in A ica’, Social indica o s esea ch, 107.
