Mose, Na aly; Fumey, Michael P o ide
A icle
Public and p i a e in es men s and economic g ow h in
Ghana and Kenya
Financial In e ne Qua e ly
P o ided in Coope a ion wi h:
Uni e si y o In o ma ion Technology and Managemen , Rzeszów
Sugges ed Ci a ion: Mose, Na aly; Fumey, Michael P o ide (2024) : Public and p i a e in es men s
and economic g ow h in Ghana and Kenya, Financial In e ne Qua e ly, ISSN 2719-3454, Sciendo,
Wa saw, Vol. 20, Iss. 3, pp. 29-41,
h ps://doi.o g/10.2478/ iq -2024-0017
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/329878
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-nc-nd/3.0/
10.2478/ iq -2024-0017
Abs ac A gene al concep ion is ha in es men induces economic g ow h, bu he e is s ill deba e o e
which ype o in es men con ibu es mo e o economic g ow h. The disagg ega ion o in es -
men in o public and p i a e componen s allows es ima ion o he impac o he wo ypes o
in es men s on economic g ow h. This esea ch, he e o e, empi ically es ima es he ela ion-
ship be ween each in es men componen agains economic g ow h by cons uc ing panel da a
o Ghana and Kenya om 1991 o 2022. The empi ical s a egy adop ed in his s udy can be
di ided in o h ee majo s ages. Fi s , he LLC uni oo es in he panel se ies is unde aken.
Second, i in eg a ed in he same o de , a Kao co-in eg a ion es is conduc ed. Finally, i he se-
ies is co-in eg a ed, he ec o o coin eg a ion in he long un is es ima ed using he dynamic
o dina y leas squa es (DOLS) me hod. Ou es ima ion esul s, based on he panel coin eg a ion
app oach con i m a long- un ela ionship be ween he s udy a iables. Fu he analysis shows
ha public in es men can p omo e economic g ow h in he long un. In con as , he esul s
indica e ha p i a e in es men can obs uc g ow h. The s udy has shown ha p i a e in es -
men did no always inc ease economic g ow h in Ghana and Kenya. The s udy indings indica e
ha public in es men is mo e e icien ly alloca ed in Ghana and Kenya han p i a e in es men ,
sugges ing he bes economic s a egy is o p i a e in es men o be complemen a y and p o-
mo e highe public in es men o imp o e public sec o p oduc i i y. The e o e, policymake s
should ocus on c ea ing a a ou able in es men clima e, p o iding iscal s imulus and p o-
mo ing public-p i a e pa ne ships o enhance in as uc u e de elopmen and s imula e p i a e
-sec o in es men , which can sus ain long- e m economic g ow h.
JEL classi ica ion: E22, O47, R42
Keywo ds: Economic G ow h, In es men P ojec Financing, P i a e In es men , Public In es men
Recei ed: 29.03.2024 Accep ed: 05.06.2024
Ci e his:
Mose, N. & Fumey, M.D. (2024). Public and p i a e in es men s and economic g ow h in Ghana and Kenya. Financial In e ne Qua e ly 20(3),
pp. 29-41.
© 2024 Na aly Mose & Michael P o ide Fumey, published by Sciendo. This wo k is licensed unde he C ea i e Commons A ibu ion-
NonComme cial-NoDe i a i es 3.0 License.
1 Uni e si y o Eldo e , Kenya, e-mail: [email protected], ORCID: h ps://o cid.o g/0000-0003-0467-235X.
2 No hwes e n Poly echnical Uni e si y, China, e-mail: [email protected], ORCID: h ps://o cid.o g/0009-0000-3213- 2233.
g ow h in A ica, i needs o be accompanied by ins i u-
ional imp o emen s and s abili y o maximize i s e ec-
i eness. I is impo an o no e ha public in es men
can ha e a c owding-in e ec on p i a e in es men in
he long un, s imula ing economic g ow h indi ec ly.
Supplemen ing p i a e and public in es men s wi h
o eign di ec in es men (FDI), echnology, human
capi al, and public consump ion is essen ial o eco-
nomic g ow h. G oss capi al o ma ion, labou g ow h,
FDI and go e nmen expendi u e explain economic
g ow h in mos de eloping coun ies (Ya emenko,
2022). FDI b ings in ex e nal capi al and expe ise, s im-
ula ing economic ac i i y and c ea ing jobs (Popescu,
2014; Si bulescu e al., 2022). I is essen ial o de elop-
ing coun ies, wi h o eign di ec in es men (FDI) be-
ing a signi ican capi al in low. FDI has been ound o
posi i ely impac economic g ow h and de elopmen ,
imp o ing mac oeconomic indica o s and he coun y's
image on he wo ld s age (Si bulescu e al., 2022). High
ins i u ional quali y, including obus go e nance,
anspa en egula o y amewo ks, e ec i e legal sys-
ems, and educed co up ion le els, is c ucial in
a ac ing FDI (Lan e al., 2011). A he same ime, pub-
lic expendi u e has mixed e ec s on economic g ow h.
In mos A ican coun ies, go e nmen consump ion,
and public in es men do no posi i ely in luence eco-
nomic g ow h in mos cases (Ndiaye, 2018). Acco ding
o Aydin and Demi öz (2023), human capi al has been
associa ed wi h long- e m p oduc i i y gains, wi h
a mo e signi ican e ec om quali y imp o emen s.
The wo coun ies unde conside a ion, Ghana and
Kenya, albei being in Sub-Saha an A ica, ha e some-
wha di e se economic sys ems and policies. Uppe -
middle-class Kenya is a g owing ma ke wi h a mo e
di e se economy and a obus manu ac u ing and ou -
ism indus y; Ghana has a la ge ag icul u al sec o and
a bu geoning se ices sec o . Conce ning ade policies,
Ghana has o en main ained a mo e p o ec ionis
s ance wi h sligh ly highe a i ba ie s. In con as ,
Kenya has adop ed a mo e open ade policy, mainly
due o i s membe ship in egional economic bodies like
he Eas A ican Communi y (EAC) (Wasike, 2022). No -
wi hs anding hese di e ences, bo h coun ies ha e
ecen ly implemen ed iscal s imulus p og ams and
public in es men plans o p omo e economic g ow h
and de elopmen . While Ghana's go e nmen has o-
cused on in as uc u e de elopmen , pa icula ly in
he ene gy and anspo a ion sec o s, Kenya has made
la ge expendi u es in a ious a eas, including housing,
oads, and heal hca e acili ies (Ndiaye, 2018). These
wo coun ies we e chosen o panel analysis because
eliable da a we e a ailable and because hey had di-
e se economic s a egies and de elopmen al s ages
wi hin he same geog aphic con ex .
In es men is c ucial in economic de elopmen ,
con ibu ing o job c ea ion, echnological ad ance-
men , skill de elopmen , and p oduc i i y imp o e-
men s (Abdulka im, 2023). The e olu ion o in es men
p ojec inancing has led o a mo e holis ic iew ha
conside s he ole o he p i a e sec o in achie ing
global inancial and economic goals (Reichel e al.,
2023). Measu es o o ganizing a ou able in es men ,
inancial, and economic condi ions include economic
s imula ion, ha moniza ion o business law, and access
o ma ke s (Sibi skaya e al., 2015). Public inancial e-
sou ces alone a e insu icien o add ess sus ainable
change, so public ac o s a e conside ing how o edi-
ec p i a e sec o in es men owa d sus ainable eco-
nomic- ela ed ac i i ies (Vydobo a, 2022).
Public and p i a e in es men s con ibu e o eco-
nomic g ow h, wi h public in es men ha ing a mo e
subs an ial posi i e impac (Ezzahid & Ra ik, 2023). The
impac o public in es men on economic g ow h may
a y depending on he le el o de elopmen and he
deg ee o p i a e-public capi al subs i u abili y
(Fa hadi, 2015). The empi ical li e a u e p o ides con-
adic o y esul s, wi h some s udies showing a posi i e
impac o public in es men on g ow h (Ocolișanu e
al., 2022), while o he s ind a de imen al e ec
(Me ga, 2022). In he case o eme ging Eu opean Union
and Cen al Eu opean coun ies, he long- e m impac
o a public capi al shock on G oss Domes ic P oduc
(GDP) is es ima ed o be nega i e (Me ga, 2022). In
ad anced economies, public in es men spending may
ha e a ela i ely coun e p oduc i e e ec on GDP
g ow h due o high le els o c owding ou (Ocolișanu e
al., 2022). This is likely because o he ad anced posi-
ion o hese coun ies in e ms o ansi ional dynam-
ics (A i & Koc, 2020). P i a e in es men posi i ely in-
luences economic g ow h in de eloping coun ies, as
e idenced by mul iple empi ical s udies (Me ga, 2022).
In A ica, public in es men di ec ly impac s ou pu
and con ibu es o he g ow h o o al ac o p oduc i -
i y (Ezzahid & Ra ik, 2023). Howe e , he e ec i eness
o public in es men in s imula ing g ow h can be in lu-
enced by a ious ac o s. The quali y o ins i u ions is
an impo an de e minan , as imp o emen s in ins i u-
ional quali y can posi i ely in luence public in es men
and c owd-in p i a e in es men (Adeosun e al., 2020).
Addi ionally, he le el and s abili y o public in es men
a e key ac o s. High and s able le els o public in es -
men a e mo e likely o p omo e economic g ow h,
while low le els and high ola ili y can hinde g ow h
(Wasike, 2022). On he o he hand, p i a e in es men
has been ound o posi i ely impac economic g ow h
in bo h he sho and long un (Ghani & Din, 2006).
While public in es men is essen ial o economic
g ow h depends on he sou ce o unds by he na ional
go e nmen . I public in es men is inanced by highe
di ec ax, he ne e ec on economic g ow h may be
nega i e despi e a posi i e e ec o ma ginal p oduc i -
i y o p i a e capi al. I public in es men is inanced by
bo owing o deb , he economic agen s will sa e he
ex a income due o non- axa ion o oday, o pay u-
u e axes, and hus dep ess consump ion (Gup a,
2020). Thus, inc eased go e nmen expendi u e is o -
se by he low le el o consump ion and he impac o
iscal policy educ ion (Yo o, 2017). Acco ding o Gup a
(2020), an inc ease in income ax e enue can be u i-
lized o inance in es men in public capi al, posi i ely
a ec ing he endogenous g ow h a e and he a e o
sa ings. I also emphasizes he ole o ins i u ions, go -
e nmen policies, and incen i es in p omo ing and sup-
po ing hese endogenous ac o s.
The In es men -led g ow h heo y, as p oposed by
Ha od (1956), emphasizes he impo ance o in es -
men in s imula ing g ow h. I sugges s ha highe le -
els o in es men , pa icula ly in inno a ion, can con-
ibu e o bo h sho - e m eco e y and long- e m
s uc u al ans o ma ion. The main hemes and as-
sump ions o he in es men -led g ow h heo y can be
summa ized as ollows. Fi s , capi al g ow h heo y
ocuses on s udying op imal inancial in es men s o
long- e m g ow h (E s ignee e al., 2015). Second,
in as uc u al in es men is conside ed a necessa y
bu insu icien condi ion o economic eco e y and
indus ial g ow h. Thi d, he model emphasizes he
condi ions o capi al g ow h and i s ela ionship o
o he long- un in es men models (Li & Hoi, 2014).
Fou h, he impac o g oss and o eign in es men s on
economic g ow h is analyzed, wi h a posi i e co ela-
ion be ween in es men and GDP g ow h (Bjelić,
2021).
Table 1 p esen s he summa y o empi ical s udies
on he impac o in es men componen s on economic
g ow h ac oss he globe.
The i s hypo hesis builds on he Keynesian iew,
i a gues ha he egula ion o economic ac i i ies by
he go e nmen passes h ough coun e cyclical p o-
cesses. This leads he go e nmen o be in ol ed in he
ma ke by suppo ing he agen ac i i ies when he
demand is dep essed and o slow ac i i ies when in la-
ion ea s se in. In he sho un, go e nmen expendi-
u e can be employed o s imula e agg ega e demand
and s imula e g ow h (Yo o, 2017). The a gumen in
a ou o public in es men is ha some expendi u es,
especially public in es men , such as elec ici y, oads,
sewe s, s ee ligh s, wa e sys ems, educa ion and
heal h gene a e ex e nali ies ha enhance he p oduc-
i i y o p i a e ac o s and hus boos economic
g ow h (Bleje & Khan, 1984; Aschaue , 1989; Calde ón
& Se én, 2010; Yo o, 2017). The heo y sugges s ha
coun ies can os e sus ainable and long- e m eco-
nomic g ow h by in es ing in educa ion, heal h, and
esea ch and de elopmen (Alam e al., 2021). The
model a gues ha p oduc i i y decline had been
caused by a decline in public expendi u e on in as uc-
u e (Aschaue , 1989).
The second hypo hesis based on he neoclassical
iew, a gues ha accele a ed public expendi u e a e
ha m ul o economic g ow h in he long un. The eco -
e y policies by public expendi u e may e en ha e
a dep essi e impac on he economy mainly because o
he c owding-ou e ec s hey exe on in es men and
p i a e consump ion. As a esul , hese nega i e e ec s
in luence economic agen s' an icipa ion o u u e con-
sequences o iscal policy and lead hem o adjus hei
beha iou acco dingly o consump ion and sa ings hus
making iscal policy ools impo en (Ba o, 1990; Yo o,
2017).
Rica dian equi alence heo y says ha inancing
go e nmen spending ou o cu en o u u e axes
and de ici s will ha e equi alen e ec s on he o e all
economy. The impac o public expendi u e on g ow h
Table 1: Re iew o In es men -G ow h li e a u e
Re e ences Coun y (Pe iod) Me hods Resul s
Khan & Reinha (1990) De eloping coun ies
(1970-1979) OLS Public in es men has no impac
P i a e in es men is bene icial
Ghani & Din (2006) Pakis an
(1973-2004) VAR Public in es men has no impac
P i a e in es men is bene icial
Makuyana & Odhiambo (2018) Sou h A ica
(1970-2017) ARDL Public in es men is ha m ul
P i a e in es men is bene icial
Nguyen & T inh (2018) Vie nam
(1990-2016) ARDL Public in es men has no impac
P i a e in es men is bene icial
Doménech & Sicilia (2021) OECD coun ies
(1960-2017) Desc ip i e Public and p i a e in es men s a e
bene icial
Ahamed (2022) De eloping coun ies
(1990-2019) ARDL Public and p i a e in es men s a e
bene icial
es men challenges a e he inancial mel down o
2007-2008 and he COVID-19 pandemic pe iod o 2020-
2023 (Mose, 2021). Fu he mo e, Kenya and Ghana
ha e expe ienced low economic g ow h a es and in-
su icien in es men p ojec inancing o s imula e
in as uc u e de elopmen . The s udy used he dy-
namic o dina y leas squa es (DOLS) es ima ion ech-
nique du ing analysis. The seconda y da a was ob-
ained om he Wo ld Bank online da abase (Wo ld
De elopmen Indica o s). The dependen a iable, eco-
nomic g ow h, was measu ed as eal G oss Domes ic
P oduc (GDP) pe capi a, whe eas he independen
a iables we e p i a e and public in es men s. Table
2 shows he desc ip ion o he a iables used in he
cu en s udy.
The empi ical li e a u e e iew in Table 1 has con-
i med ha he ela ionship be ween in es men com-
ponen s and economic g ow h is a ied and inconclu-
si e. Fu he mo e, e y ew s udies ha e been ca ied
ou in Kenya and Ghana using he dynamic OLS (DOLS)
es ima ion app oach and he abo e s udies ha e used
di e en es ima ion me hods and sample sizes.
The cu en s udy will adop a quan i a i e e-
sea ch design o analyze he end and ole o di e en
ypes o in es men s on economic g ow h in Ghana
and Kenya o he pe iod 1991-2022. The pe iod has
been chosen since he wo coun ies aced se e al in-
Re e ences Coun y (Pe iod) Me hods Resul s
Jahan e al. (2022) Asian coun ies
(2020) ANOVA Public in es men has no impac
P i a e in es men is ha m ul
Abdulka im (2023) Nige ia
(1981-2020) ARDL Public in es men is ha m ul
P i a e in es men is bene icial
No es: Au o eg essi e Dis ibu ed Lag (ARDL); O dina y Leas Squa es eg ession (OLS); Vec o au o eg ession (VAR)
Sou ce: Au ho ’s own wo k.
Table 2: Desc ip ion o s udy a iables
Va iable Measu e P oxy Da a sou ce Expec ed sign
Dependen a iable
Economic g ow h
(GDP)
Cons an US Dolla
(2015) GDP pe capi a WDI No p edic ed
Abdulka im (2023)
Independen a iables
Public in es men
(PUB) Pe cen G oss capi al
o ma ion (% GDP) WDI Posi i e (Ahamed,
2022)
P i a e in es men
(PRV) Pe cen Domes ic c edi o p i a e
sec o (% GDP) WDI Posi i e (Ahamed,
2022)
Fo eign di ec
in es men (FDI) Pe cen Fo eign di ec in es men ,
ne in lows (% GDP) WDI Posi i e (Nguyen
& T inh, 2018)
Public consump ion
(PC) Pe cen Go e nmen consump ion
expendi u e (% GDP) WDI Nega i e (Ghani
& Din, 2006)
Human capi al (HC) Numbe Labo o ce ( o al) WDI Posi i e (Makuyana
& Odhiambo, 2018)
Sou ce: Au ho ’s own wo k.
(1)
The econome ic baseline model measu ing he
e ec o selec ed a iables on economic g ow h is e-
w i en as ollows.
(2)
Whe e: GDPi, - Economic g ow h, measu ed by GDP
pe capi a g ow h, PUBi, - Public in es men , measu ed
by g oss capi al o ma ion, PRVi, - P i a e in es men ,
measu ed by domes ic c edi o he p i a e sec o ,
Acco ding o he neoclassical g ow h model o
Solow ou pu depends on capi al and labou inpu s
(Solow, 1956). As hypo hesized by Solow, an ex ension
de eloped by Aschaue (1989); Ba o (1990) as well as
Yo o (2017) in he empi ical li e a u e is adop ed. In
he cu en s udy, i is hypo hesized ha economic
g ow h depends on he in es men used o he capi al
s ock and on he labou o ce. The s udy adop s he
empi ical wo k o Yo o (2017) as well as Makuyana and
Odhiambo (2018) o build a simple heo e ical baseline
es ima ion g ow h model as p esen ed below.
( , , , , )GDP PUB PRV FDI PC HC=
, 0 1 , 2 , 3 ,
4 , 5 , ,
i i i i
i i i
GDP PUB PRV FDI
PC HC
= + + +
+ + +
es is Kao (1999), which is based on he Engel-G ange
app oach and con ols o homogenei y on uni s in he
panel se (Alam e al., 2021). I he coin eg a ion is con-
i med, his will sugges ha he selec ed a iables
sha e he same s ochas ic end, and hus, hey can be
combined in he long un. I he se ies is co-in eg a ed,
he panel ec o e o co ec ion model is conduc ed o
con i m he long- un causali y o con e gence o he
a ge a iables. In he analysis, he e o co ec ion
e m (ECT) alue is expec ed o be nega i e and signi i-
can a 1%. Once he panel se ies is es ed o causali y
using he ec o o coin eg a ion, he model is es ima -
ed using DOLS o check o long- un ela ionships be-
ween coe icien s.
In analysis, se e al es ima ion op ions a e a ailable
when using panel da a o be ween g oup da a, o in-
s ance, panel o dina y leas squa es (POLS), pooled
mean g oup (PMG), ully-modi ied OLS (FMOLS), and
dynamic OLS (DOLS) es ima ion echnique (Ped oni,
2001). The panel o dina y leas squa es (POLS) es ima-
ion echnique is he mos common analysis app oach
when combined wi h ixed o andom e ec s. One ad-
an age o POLS is ha i can con i m he p esence o
con e gence in panel da a; howe e , some imes POLS
leads o biased es ima es. Thus, o con i m he es i-
ma es ha a e no biased, he s udy can conduc analy-
sis using FMOLS o DOLS es ima ion ha a ou s small
sample da a. The DOLS model is obus o a ious de-
pa u es om s anda d eg ession assump ions in
e ms o esidual co ela ion, he e oscedas ici y, mis-
speci ica ion o unc ional o m and non-no mali y o
esiduals (S ock & Wa son, 1993). The e o e, his s udy
u ilized he DOLS es ima o pionee ed by S ock and
Wa son (1993) o es ima e he long- un ela ionship.
Pos -es ima ion panel diagnos ic es s (he e osceda-
s ici y, au oco ela ion, c oss-sec ional dependence,
and no mali y) we e conduc ed du ing he eg ession
analysis o a oid misleading in e ences.
Vec o au o eg ession (VAR) lag selec ion c i e ia
we e applied o de e mine he lag leng h and bes
model es ima o . The esul o he VAR selec ion c i e-
ia is p esen ed in Table 3.
FDIi, - Fo eign di ec in es men , measu ed by o eign
di ec in es men ne in low, PCi, - Public consump ion,
measu ed by go e nmen consump ion expendi u e,
HCi, - Human capi al, measu ed by Labou g ow h a e,
β - he eg ession coe icien , εi, - is he e o e m and
he subsc ip s i and ep esen coun y and ime
dimensions espec i ely.
The econome ic baseline model was ew i en in
loga i hm o m (ln), o inc ease a iance s abili y and
emo e ou lie s, as shown below.
(3)
To analyze he long- un e ec o in es men a ia-
bles on economic g ow h using DOLS equa ion 3,
abo e, was u ilized du ing eg ession analysis.
Vec o au o eg ession (VAR) lag selec ion c i e ia
we e applied o con i m he lag leng h and bes model
es ima o . The selec ion c i e ia a e essen ial in educ-
ing esidual co ela ion p oblems. Coun ies like Ghana
and Kenya can in e ac h ough impo s, expo s, hu-
man capi al ans e , and economic in eg a ion, hus
c ea ing c oss-sec ional dependence by expe iencing
s anda d shocks. The e o e, his s udy's c oss-sec ional
dependence es is essen ial o check i he da a exhibi
c oss-sec ional dependency (CD). The s udy adop ed
he Pesa an CD de eloped by Pesa an (2004) o check
o c oss-sec ional dependency. Panel da a se ies a e
usually cha ac e ized by s ochas ic ends ha a e easi-
ly emo ed h ough di e encing. The selec ion o he
uni oo es depends upon c oss-sec ional depend-
ence (Phillips & Sul, 2003). Va ious panel uni oo
es s, such as IPS, ADF, Phillips Pe on, and Le in-Lin-
Chu (LLC) es s, a e a ailable. LLC uni oo es is com-
monly used in c oss-sec ional dependence (Le in-Lin-
Chu 2002). I he uni oo o he panel se ies is in e-
g a ed in he same o de , a coin eg a ion es is pe -
o med. Hausman (1987) es was conduc ed o sup-
po he long- un es ima ion me hod used in he s udy
based on OLS when combined wi h Fixed o Random
e ec . In his s udy, wo coin eg a ion es s ha e been
sugges ed o use in in es iga ing he long- un ela ion-
ship. The i s es is Ped oni (2001), and he second
, 0 1 , 2 , 3 ,
4 , 5 , ,
i i i i
i i i
lnGDP lnPUB lnPRV lnFDI
lnPC lnHC
= + + +
+ + +
Table 3: Lag leng h selec ion c i e ia
Lag LogL LR FPE AIC SC HQ
0 249.3659 NA 6.77e-12 -8.691639 -8.474637 -8.607508
1 443.7425 340.15910 2.38e-14 -14.347950 -12.828930 -13.759030
2 775.4439 509.39850 6.46e-19 -24.908710 -22.087680* -23.815000*
3 816.4799 54.2262* 6.06e-19* -25.088570* -20.965530 -23.490080
4 850.4569 37.61733 8.26e-19 -25.016320 -19.591270 -22.913040
No e: * = lag o de selec ed by he c i e ion; SIC = Schwa z in o ma ion c i e ion; AIC = Akaike in o ma ion c i e ion;
HQ: Hannan-Quinn in o ma ion c i e ion; FPE: Final p edic ion e o ; LR: sequen ial modi ied LR es s a is ic (each
es a 5% le el)
Sou ce: Au ho ’s own wo k.
To check i he da a exhibi c oss-sec ional depend-
ence (CD), he Pesa an CD es de eloped by Pesa an
(2004) was adop ed by he s udy. The Pesa an CD e-
sul is epo ed in Table 4 below.
Acco ding o he a ious lag leng h selec ion c i e-
ia, such as AIC and SIC, a lag o 4 is chosen o he VAR
as i epo ed he minimum alue. A lag o ou was
chosen o educe he se ial co ela ion p oblem. In ad-
di ion, he AIC es ima ion model was p e e ed since i
had he lowes alue (-25.01632) compa ed o o he
es ima ion c i e ia.
Table 4: Pesa an CD esul
Tes S a is ic D P ob
B eusch-Pagan LM -2.797984 1 0.0944
Pesa an scaled LM -1.271367 - 0.2036
Bias-co ec ed scaled LM -1.239109 - 0.2153
Pesa an CD -1.672718 - 0.0944
Sou ce: Au ho ’s own wo k.
A uni oo es was conduc ed o asce ain he
o de o in eg a ion and a oid spu ious co ela ions in
DOLS es ima ion. The s udy in es iga ed he s a iona i-
y o s udy a iables using he Le in-Lin-Chu (LLC) es .
The s a iona i y esul is epo ed in Table 5.
Table 4 epo s ou Pesa an CD esul ; he es indi-
ca es he absence o c oss-sec ional dependence in ou
eg ession a iables, con i med by he P- alue 0.0944
> 0.0500. Since he p- alue o 0.0944 o he Z-s a is ic
is mo e han 0.05, he null hypo hesis should no be
ejec ed. This con i ms ha c oss-sec ional dependence
is no a p oblem. As a esul , he s udy applied he Le -
in-Lin-Chu (LLC) es in uni oo analysis.
Table 5: S a iona i y Tes
Va iables LLC a Le el O de LLC a i s di e ence O de
S a is ic P ob. S a is ic P ob.
GDP 4.88497 1.0000 I(1) -4.09049*** 0.0000 I(0)
PUB -0.01595 0.4936 I(1) -7.86468*** 0.0000 I(0)
PRV 0.82469 0.7952 I(1) -6.33159*** 0.0000 I(0)
FDI 0.62478 0.7339 I(1) -2.48678*** 0.0064 I(0)
PC -0.63709 0.2620 I(1) -6.91668*** 0.0000 I(0)
HC 68.57610 1.0000 I(1) -25.40740*** 0.0000 I(0)
No e: *** Deno es signi icance a a 1% le el o signi icance, H0: Panels con ain uni oo s, GDP - Economic g ow h;
PUB - Public in es men ; PRV - P i a e in es men ; FDI - Fo eign di ec in es men ; PC - Public consump ion; HC -
Human consump ion
Sou ce: Au ho ’s own wo k.
o de and con ain uni oo . This alidi y o e s he
ou e o es o se e al coin eg a ion ela ionships. To
ca y ou a coin eg a ion analysis based on OLS es ima-
ion, he Hausman es was conduc ed o choose be-
ween a ixed o andom e ec model. The Hausman
es esul is shown in Table 6 below.
Based on he esul s p esen ed in Table 5, econom-
ic g ow h, public in es men , p i a e in es men , o -
eign di ec in es men , public consump ion, and hu-
man capi al we e all non-s a iona y and in eg a ed in
he o de I(1). Howe e , hey we e ans o med by he
i s di e ence o become s a iona y. The LLC es has
con i med ha he a iables a e in eg a ed in he same
Table 6: Hausman Tes ou pu
Tes Summa y Chi-Sq. S a is ic Chi-Sq. d. . P ob.
Pe iod andom 26.729513*** 5 0.0001
No e: * * * Deno es signi icance a a 1% le el o signi icance
Sou ce: Auh o ’s own wo k.
Since all selec ed a iables we e in eg a ed in o he
o de I(1), a non-s a iona y coin eg a ion es was pe -
o med o asce ain he long- un equilib ium and se e -
al coin eg a ion ela ionships. G ange (1988) submi s
ha i ac o s a e in eg a ed, he esiduals will be in e-
g a ed a he le el, and i no , hen i s -o de in eg a-
ion will be ound. The s udy adop ed he Kao co-
in eg a ion es o check o he long- un ela ionships
be ween selec ed a iables. The ixed e ec model is
used du ing coin eg a ion analysis. Table 7 shows he
Kao coin eg a ion es esul s.
The Hausman es esul suppo ed using he ixed
e ec es ima ion me hod o e he andom me hod as
ein o ced by p- alue (0.0001 < 0.0500). The ixed
e ec model is accep ed because he chi-squa e s a is-
ic o he Hausman es ejec ed he null hypo hesis o
andom e ec . The panel da a was es ima ed using he
ixed e ec model o he panel es ima ion echnique,
which is gea ed a con olling o ime-in a ian and
unobse able coun y e ec s.
Table 7: Kao Residual Coin eg a ion Tes
- S a is ic P ob.
ADF 2.369214 0.0089
Residual a iance 0.001695 -
HAC a iance 0.000182 -
Sou ce: Au ho ’s own wo k.
The es ima ion o he ec o e o co ec ion mod-
el (VECM) was o asce ain he di ec ion o causa ion
and adjus i owa ds he equilib ium. VECM is conduc -
ed o check o he long- un causali y o con e gence
o he a ge a iables. Table 8 shows he VECM analy-
sis coe icien s.
Table 7 con i ms he coin eg a ion ela ionship
since he ADF s a is ic is signi ican a a 1 % signi icance
le el. Hence, we ejec he null hypo hesis. Once he
long- un ela ionship was con i med, he s udy checked
o long- un causali y using ec o e o co ec ion
models.
Table 8: Value o coe icien s
ECT Coe icien S d. E o -S a is ic P ob.
C(1) -0.846009 0.210856 -4.012263 0.0015
C(2) -2.037543 0.491794 -4.143087 0.0012
C(3) 0.042720 0.831401 0.051383 0.9598
C(4) -0.161763 0.203908 -0.793312 0.4418
C(5) -0.214576 0.149145 -1.438708 0.1739
C(6) 1.900558 0.631102 3.011490 0.0100
C(7) 1.016450 0.494544 2.055329 0.0605
C(8) 0.388778 0.517284 0.751575 0.4657
C(9) -0.033721 0.506931 -0.066520 0.9480
C(10) 0.106619 0.079259 1.345198 0.2016
C(11) -0.088031 0.054159 -1.625418 0.1281
C(12) 0.226852 0.672490 0.337332 0.7413
C(13) -0.494261 0.743985 -0.664343 0.5181
C(14) -97.035220 44.565890 -2.177343 0.0485
C(15) -96.836900 46.206620 -2.095737 0.0562
C(16) 1.343803 0.628481 2.138177 0.0521
Sou ce: Au ho ’s own wo k.
signi ican esul also con i ms long- un causali y be-
ween in es men componen s and economic g ow h.
Mo ing o wa d, he s udy es ima ed he DOLS model
o ob ain he long- un coe icien s
Based on he esul in Table 8, he e o co ec ion
e m (ECT) is nega i e and signi ican a 1 % (-0.846).
This con i ms he con e gence; he economy can eco -
e by 84.6% du ing he cu en yea . The nega i e and
p i a e in es men , public in es men , and economic
g ow h. Table 9 p esen s he long- un es ima e coe i-
cien s.
Once coin eg a ion is con i med among he selec -
ed a iables, he s udy used he panel DOLS es ima ion
echnique o iden i y he long- un ela ionship be ween
Table 9: DOLS Reg ession coe icien s
Va iable Coe icien S anda d e o -S a is ics p- alue
PUB 0.098096 0.045578 2.152270 0.0405**
PRV -0.252063 0.042296 -5.959449 0.0000***
FDI 0.041242 0.013686 3.013395 0.0056***
PC -0.282784 0.062964 -4.491214 0.0001***
HC 0.904220 0.058087 15.566760 0.0000***
Goodness o Fi R2 = 0.985673 Adjus ed R2 = 0.969223
F = 44.557460 P- alue(F) = 0.000000
B eusch-Pagan F = 0.861208 P ob > F = 0.523759
B eusch-God ey χ2 = 3.733390 P ob > χ2 = 0.058310
Pesa an CD Z = - 1.672718 P ob > Z = 0.094400
Ja que-Be a χ2 = 0.955339 P- alue(χ2) = 0.620227
Du bin-Wa son DW = 2.111040 -
No e: * p < 0.1, ** p < 0.05, *** p < 0.01 a e signi icance le els in which he null hypo hesis is ejec ed. GDP - Econo-
mic g ow h; PUB - Public in es men ; PRV - P i a e in es men ; FDI - Fo eign di ec in es men ; PC - Public consump-
ion; HC - Human consump ion
Sou ce: Au ho ’s own wo k.
én, 2010). The empi ical li e a u e p o ides con adic-
o y esul s, wi h some s udies showing a posi i e im-
pac o public in es men on g ow h (A i & Koc, 2020;
Ahamed 2022), while o he s ind a de imen al e ec
(Makuyana & Odhiambo, 2018; Nguyen & T inh, 2018).
The s udy indings suppo he simila s udy by Ghani
and Din (2006) in Pakis an and Ahamed (2022) o de-
eloping coun ies, whe e he s udies a gued he sig-
ni icance o public in es men in inducing g ow h
h ough posi i e spillo e o he p i a e sec o (A i
& Koc, 2020). Howe e , hese indings con adic Ma-
kuyana and Odhiambo's (2018) empi ical s udy in Sou h
A ica, whe e he s udy epo ed a nega i e ela ion-
ship be ween public in es men and g ow h. This was
a ibu ed o he c owding e ec o p i a e in es men
by public in es men . Finally, Nguyen and T inh (2018)
ound he ela ionship insigni ican in Vie nam.
The e ec o p i a e in es men on g ow h is nega-
i e and signi ican a a 1 pe cen signi icance le el.
This means ha a 1 pe cen inc ease in p i a e in es -
men will cause economic g ow h o decline by 0.25
pe cen . Based on he empi ical esul , p i a e in es -
men slows economic g ow h in Ghana and Kenya. This
is agains heo e ical expec a ions, whe e p i a e in-
es men in luences g ow h by exhibi ing inc easing
e u ns o scale (Doménech & Sicilia, 2021). Some-
imes, public in es men can c owd ou p i a e in es -
men , hus comp omising g ow h in de eloping coun-
ies. Fu he mo e, nega i e esul s could be a ibu ed
Based on he esul o Table 9, he coe icien o
public in es men is posi i e and s a is ically signi ican
a a 5 pe cen signi icance le el, as pe he p- alue o
he coe icien . Since he p- alue o 0.0405 o he -
s a is ic is less han 0.05, he null hypo hesis should be
ejec ed, sugges ing he a iable is signi ican . I means
ha a 1 pe cen inc ease in public in es men will lead
o abou a 0.09 pe cen inc ease in economic g ow h,
as expec ed in Ghana and Kenya. This implies ha pub-
lic in es men enables economic g ow h in bo h coun-
ies h ough inc eased e iciency and p oduc i i y.
In es men in co e in as uc u e s imula es o al ac o
p oduc i i y g ow h in co e sec o s and p i a e sec o
o ganiza ions, hus p omo ing g ow h (Aschaue ,
1989). Bo h coun ies ha e di e en budge a y alloca-
ions owa ds go e nmen expendi u e, bu conside -
ing bo h ha e excess labou supply, he abili y o he
public sec o o in luence g ow h is highe han ha o
he p i a e sec o . Acco ding o mac oeconomic li e a-
u e, public in es men can s imula e economic ac i i-
ies ia sho - e m e ec s on agg ega e demand by
aising he p oduc i i y o he p i a e sec o (Ghani
& Din, 2006). Mos de eloping coun ies employ expan-
siona y iscal policy ools and iscal s imulus packages
o g ow he public sec o and hus encou age p oduc-
i i y and e iciency (Doménech & Sicilia, 2021). Public
in es men gene a es posi i e spillo e by p o iding
public goods such as educa ion, heal h, and in as uc-
u e de elopmen (Ghani & Din, 2006; Calde ón & Se -
