Mladeno ska, Suzana Mak eshanska; Tashe ska, Biljana
A icle
Fiscal decen aliza ion and economic g ow h: Empi ical
e idence om Eu opean coun ies
Economy, Business and De elopmen : An In e na ional Jou nal (EB&D)
P o ided in Coope a ion wi h:
Ss. Cy il and Me hodius Uni e si y in Skopje, Facul y o Economics
Sugges ed Ci a ion: Mladeno ska, Suzana Mak eshanska; Tashe ska, Biljana (2024) : Fiscal
decen aliza ion and economic g ow h: Empi ical e idence om Eu opean coun ies, Economy,
Business and De elopmen : An In e na ional Jou nal (EB&D), ISSN 2671-3470, Ss. Cy il and
Me hodius Uni e si y in Skopje, Facul y o Economics, Skopje, Vol. 5, Iss. 1, pp. 44-56,
h ps://doi.o g/10.47063/ebd.00017
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/314350
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/4.0/
44
Economy, Business & De elopmen (2024) 5(1), 44-56
DOI: 10.47063/ebd.00017
RESEARCH PAPER
ISSN
2671-3462 (p in )
2671-3470 (online)
Jou nal homepage: h ps://jou nals.ukim.mk/index.php/ebd
FISCAL DECENTRALIZATION AND ECONOMIC GROWTH: EMPIRICAL EVIDENCE
FROM EUROPEAN COUNTRIES
Suzana Mak eshanska Mladeno ska
Associa e P o esso , Ss. Cy il and Me hodius Uni e si y in Skopje,
Facul y o Economics – Skopje
[email p o ec ed]
Biljana Tashe ska
Associa e P o esso , Ss. Cy il and Me hodius Uni e si y in Skopje,
Facul y o Economics – Skopje
[email p o ec ed]
Abs ac
This pape empi ically explo es he ela ionship be ween iscal decen aliza ion and economic g ow h,
using panel da ase o 31 Eu opean coun ies, o e he pe iod 1972-2012. The empi ical esul s indica e
ha iscal decen aliza ion, quan i a i ely measu ed as he sha e o local go e nmen expendi u es and
e enues in o al go e nmen expendi u es and e enues has a a o able impac on economic g ow h
in he Eu opean coun ies, and also ha e enue decen aliza ion is shown o be mo e e ec i e han
expendi u e decen aliza ion in e ms o s imula ing economic g ow h. Fu he , ou empi ical esul s also
sugges ha he ela ionship be ween decen aliza ion and economic g ow h is nonlinea , i.e., he e is
some op imal le el o decen aliza ion ha maximizes economic g ow h. This means ha iscal
decen aliza ion is expec ed o ha e a mo e p onounced posi i e impac on g ow h in less decen alized
coun ies, while in mo e decen alized coun ies, a u he inc ease in decen aliza ion s a s o hinde
economic g ow h.
In addi ion, he g ow h-enhancing e ec o iscal decen aliza ion is e en s ongly con i med in ou
subsample o ad anced Eu opean coun ies, when he long- e m e ec o decen aliza ion was
examined. The e o e, compa ed wi h ou p e ious indings on he impac o iscal decen aliza ion on
g ow h in Cen al and Eas e n Eu opean coun ies, we ind ha iscal decen aliza ion has a di e en
impac on economic g ow h in ad anced Eu opean coun ies is-à- is Cen al and Eas e n Eu opean
coun ies, i.e., while i enhances g ow h in he o me , i hinde ed g ow h in he la e .
Keywo ds: Fiscal decen aliza ion, economic g ow h, Eu opean coun ies.
JEL classi ica ion: H71, H72, H77
45
In oduc ion
Fiscal decen aliza ion has been a opic o conside able in e es in ecen economic li e a u e, pa ly due
o i s po en ial implica ions o economic g ow h. T ans e ing iscal esou ces and esponsibili ies o
local go e nmen s is p ima ily aimed a achie ing a mo e e icien deli e y o public goods and se ices.
This enhanced e iciency wi hin he public sec o is expec ed o os e economic g ow h and o imp o e
he wel a e o ci izens. The e o e, o e ecen decades, many de eloping coun ies ha e adop ed iscal
decen aliza ion ini ia i es o s eng hen hei public sec o e ec i eness. Acco ding o Ga man e al.
(2001) mo e han 80% o 75 de eloping coun ies had implemen ed some o m o iscal decen aliza ion
e o m in he p eceding wo decades. Ano he epo s a es ha mo e han 120 de eloping coun ies
ha e conduc ed some ype o decen aliza ion since 2008 (I anyna and Shah, 2014). In he case o
de eloped coun ies, he pic u e is qui e he same. The index o egional au ho i y by Hooghe e al.
(2010) e eals ha 70% o hese coun ies ha e expe ienced an ac i e decen aliza ion p ocess since
1950.
Schola s s udying he heo e ical and p ac ical dimensions o iscal ede alism ha e long deba ed on he
e iciency bene i s o iscal decen aliza ion (Tiebou , 1956; Musg a e, 1959; Oa es, 1972; Rubin eld
1987). The i s gene a ion o iscal decen aliza ion heo ies, pionee ed by Oa es (1972), while no
di ec ly add essing he issue o economic g ow h, unde line ha in he case o di e se local p e e ences
and needs local go e nmen ’s p o ision o public good and se ices will lead o g ea e e iciency and
will imp o e he ci izen’s wel a e. Oa es’s well known decen aliza ion heo em s a es ha local
go e nmen s a e be e posi ioned o iden i y ci izens’ di e se p e e ences o public goods and se ices,
and secondly by ailo ing se ices o local needs and managing cos s mo e e ec i ely, hey can p o ide
public goods and se ices a lowe cos s han cen al go e nmen s. Fu he , iscal decen aliza ion
places local go e nmen s in compe i ion wi h one ano he and he e o e p omo es iscal discipline by
encou aging hem o p io i ize spending and gene a e e enue mo e e ec i ely (Tiebou , 1956; B ennan
and Buchanan, 1980). Consequen ly, he p oponen s o iscal decen aliza ion a gue ha i can os e
economic g ow h by mo e e icien esou ce alloca ions, g ea e esponsi eness o local p e e ences
and enhanced accoun abili y o local go e nmen s (Oa es, 1993; Ma inez-Vazquez and McNab, 2006).
On he o he hand, nume ous au ho s ha e shi ed hei a en ion o he in e sed side o he coin: he
po en ial isks associa ed wi h decen aliza ion in e ms o economic pe o mance (P ud’Homme, 1995;
Tanzi, 1996; Rod iguez-Pose and Gill, 2005). C i ics a gue ha decen aliza ion could po en ially
unde mine economic pe o mance by causing agmen a ion, ine iciencies om o e lapping unc ions
and duplica ing cos s, iscal dispa i ies among egions, e c. Fo example, local go e nmen s may be
disposed o bigge co up ion because o icials a he local le el a e mo e suscep ible o demands o
local in e es g oups (P ud’homme, 1995; Shah 2004). Mo eo e , excessi e decen aliza ion can make
iscal policy coo dina ion mo e complex and ha m mac oeconomic s abili y (Oa es, 2005). In some
cases, as in less de eloped coun ies, local go e nmen s may lack he capaci y o e ec i ely manage
decen alized esponsibili ies, esul ing in ine ec i e ou come o iscal decen aliza ion (P ud’Homme,
1995, Rod iguez-Pose and K oije , 2009, Ode o, 2004). In his espec , P ud’homme (1995) and Tanzi
(1996) i s ly ecognized ha poo ly designed decen aliza ion sys ems, whe e subna ional go e nmen s
a e allowed o bo ow wi hou con ol and he cen al go e nmen co e s any de aul s, lead o
mac oeconomic ins abili y and wo se o e all economic pe o mance.
An impo an ex ension o he iscal ede alism li e a u e is he second-gene a ion heo ies ini ia ed by
Weingas (1995) ha b ing a di e en pe spec i e by assuming he p esence o sel -in e es ed
go e nmen o icials wi h hei own agenda, opposed o he bene olen go e nmen o icials assumed
in he p e ious li e a u e. Acco ding o Ha ield (2006) economic policy is no decided by bene olen
social planne s, bu by sel -in e es ed go e nmen o icials wi h a leas one eye on hei eelec ion
p ospec s. In his con ex , he second-gene a ion heo ies ex end and adap he old decen aliza ion’s
lessons acco ding o his new pe spec i e o he ole o he go e nmen o icials (Qian and Weingas
1997; Ga za elli 2004; Oa es, 2005).
The la e s udies on iscal ede alism emphasize he c i ical impo ance o local go e nmen e enue
gene a ion ha makes local go e nmen s mo e esponsi e o ci izens, educes co up ion and inc eases
he incen i es o e icien p o ision o public goods and se ices. In o he wo ds, i local go e nmen s
a e gi en g ea e au onomy o e hei iscal e enues, as a esul , hey become mo e di ec ly
46
accoun able o he ci izens (Rodden, 2003). Howe e , some au ho s a gue ha decen aliza ion can
lead o iscal compe i ion be ween local go e nmen s o engage in a ace o he bo om on he axa ion
o mobile ac o s, hence unde p o iding p oduc i e public expendi u e (Wilson, 1986; Zod ow and
Mieszkowski, 1986; B ueckne 2004). On he o he hand, he in e go e nmen al iscal ans e s by
concen a ing axing powe a cen al go e nmen and unde mining iscal discipline o local go e nmen s
can lead o a la ge and less e ec i e public sec o (G ossman, 1989; Bahl and Linn, 1992).
Finally, acco ding o some au ho s, he ela ionship be ween decen aliza ion and economic g ow h is
no necessa ily linea . Thiessen (2003) ound ha a low le el o iscal decen aliza ion may hinde long
e m economic g ow h because local go e nmen s lack su icien mo i a ion o enhance alloca i e and
p oduc ion e iciency. Equally, excessi e decen aliza ion can also lead o economic ine iciencies and
social wel a e losses, a ec ing mac oeconomic s abili y and exace ba ing income inequali y. In his
espec , coun ies ha s a wi h a lowe ini ial le el o public sec o decen aliza ion a e likely o
expe ience mo e signi ican posi i e e ec s o decen aliza ion on economic g ow h, in con as o
coun ies ha ha e al eady achie ed a highe le el o iscal decen aliza ion (Blochlige and Ege ,
2013). As a esul , decen aliza ion has he po en ial o os e economic g ow h by enhancing public
sec o e iciency, bu i s impac emains ambiguous and depends on mul iple ac o s and con ex -speci ic
ci cums ances (Li ack e al., 1998).
The p ima y objec i e o his s udy is o empi ically explo e he link be ween iscal decen aliza ion and
economic g ow h in Eu opean coun ies and o obse e whe he he impac o decen aliza ion on g ow h
is consis en ac oss ad anced Eu opean coun ies compa ed o de eloping coun ies in Cen al and
Eas e n Eu ope (explo ed in ou p e ious empi ical s udy). Fu he mo e, we examine whe he he e is
a nonlinea associa ion be ween iscal decen aliza ion and economic g ow h ac oss hese coun ies.
The emaining sec ions o he pape a e o ganized as ollows i.e. he second sec ion o e s a concise
o e iew o he empi ical li e a u e ela ed o his opic, he hi d sec ion elabo a es he me hodology
employed and he da a u ilized in he s udy, he ou h sec ion p esen s and discusses he indings om
he empi ical in es iga ion and he inal sec ion concludes wi h closing ema ks.
Re iew o Empi ical Li e a u e
Nume ous empi ical s udies ha e a emp ed o examine and quan i y he ela ionship be ween iscal
decen aliza ion and economic g ow h, bu hei indings ha e been inconsis en . Some s udies ind
posi i e co ela ions, indica ing ha iscal decen aliza ion os e s economic g ow h. On he o he hand,
some s udies ind no signi ican ela ionship o e en nega i e co ela ions be ween iscal
decen aliza ion and economic g ow h.
The majo i y o he empi ical s udies a e c oss-coun y analyses ha use ex ensi e samples om a ied
coun ies (Oa es, 1995; Da oodi and Zou, 1998; Yilmaz, 1999; Iimi, 2005; Ma inez-Vazquez and
McNab, 2006; Cana i e-Baca eza e al., 2020). Howe e , mos o he hese s udies a e dominan ly
ocused on ad anced OECD coun ies due o be e a ailabili y o da a (Thiessen, 2003; Elle , 2004;
Tho n on, 2007; Rod iguez-Pose and Ezcu a, 2011; Baska an and Feld, 2013; Gemmel e al., 2013;
Filippe i and Sacchi, 2016), whe eas only a smalle ac ion o s udies a e ocused on de eloping
coun ies (Enikolopo and Zhu a skaya, 2007; A is o nik, 2012; Mak eshanska-Mladeno ska and
Tashe ska, 2019; Hani e al., 2020; Ko o un e al., 2020). On he o he hand, nume ous s udies also
explo e he ela ionship be ween decen aliza ion and g ow h wi hin speci ic coun ies, p ima ily China
(Zhang and Zou, 1998; Lin and Liu, 2000; Jin and Zou, 2005; Qiao e al., 2008; Yang, 2016; Song e al.,
2019; Ding e al., 2019) o USA (Xie e al., 1999; Akai and Saka a, 2002; S ansel, 2005).
As p e iously s a ed, so a he e is no empi ical consensus ega ding he impac o iscal
decen aliza ion on economic g ow h. Empi ical s udies deli e di e gen esul s based on a ying
da ase s, es ima ion echniques and speci ica ions and di e en indica o s o decen aliza ion. Hence,
many empi ical s udies con i m ha he e is a posi i e ela ionship be ween iscal decen aliza ion and
economic g ow h (Lin and Liu, 2000; Akai and Saka a, 2002; Iimi, 2004; S ansel, 2005; Qiao e al., 2008;
Buse , 2011; Blochlige and Ege , 2013; Gemmell e al., 2013; Filippe i and Sacchi, 2016; Sla inskai e,
47
2017; Hani e al., 2020; Sla inskai e e al., 2020). O he s udies sugges ha iscal decen aliza ion has
a nega i e impac on economic g ow h (Zhang and Zou, 1998; Xie e al., 1999; Enikolopo and
Zhu a skaya, 2007; Jin and Zou, 2005; Rod iguez-Pose and Ezcu a, 2011; Baska an and Feld, 2013)
o ha i has no clea impac on g ow h (Da oodi and Zou, 1998; Bodman and Fo d, 2006; Tho n on,
2007; Baska an and Feld, 2013; Asa yan and Feld, 2015; Ko o un e al., 2020).
I we analyze sepa a ely he e enue e sus expendi u e decen aliza ion, ecen s udies (Rod iguez-
Pose and K oije , 2009; Blochlige and Ege , 2013; Gemmell e al., 2013) mainly ind ha e enue
decen aliza ion has mo e p onounced s imula ing e ec s on economic g ow h, while expendi u e
decen aliza ion is associa ed wi h lowe economic g ow h; howe e o he s udies ind no signi ican
impac o e enue decen aliza ion (Tho n on, 2007) o s onge e ec o expendi u e decen aliza ion
(Cana i e-Baca ezza e al., 2020). In addi ion, in e go e nmen al iscal ans e s ha e a mo e
p onounced ad e se impac on g ow h (Rod iguez – Pose and K oije , 2009).
Unlike mos s udies ha explo e he linea ela ionship, Thiessen (2003) indica es a non-linea
ela ionship be ween iscal decen aliza ion and economic g ow h. Acco ding o him, his ela ionship is
in e ed “U” shaped (bell-shaped), meaning ha he e is some op imal deg ee o iscal decen aliza ion
ha maximizes g ow h, beyond which highe decen aliza ion s a s o educe he economic g ow h a e.
The diminishing e u ns on decen aliza ion a e also con i med by Elle (2004), Blochlige and Ege
(2013), Song e al., (2019) who ind ha coun ies o egions wi h lowe le els o decen aliza ion end
o expe ience mo e signi ican posi i e e ec s on economic g ow h. F om ecen empi ical esea ch on
he opic, he bell-shaped ela ionship be ween iscal decen aliza ion and g ow h was con i med by
Belko icso a and Boo (2021) on a sample o 29 OECD coun ies and by Ca ni i e al. (2019) on a panel
da ase o 25 Eu opean coun ies ha is mo e ela ed o ou empi ical wo k.
Fu he , some s udies ha e ound a ying impac s o iscal decen aliza ion on g ow h in ad anced
e sus de eloping coun ies. Fo example, s udies by Da oodi and Zou (1998), Im (2010) and
Sla inskai e (2017) did no iden i y a signi ican ela ionship in ad anced coun ies, while esul s we e
mixed o de eloping coun ies. In con as , Cana i e-Baca eza e al. (2020) obse ed a posi i e e ec
o decen aliza ion in ad anced coun ies bu no signi ican e ec in de eloping coun ies. In line wi h
his, he impac o decen aliza ion on economic g ow h may a y also upon he quali y o ins i u ional
and poli ical ac o s wi hin a coun y. Fo example, as sugges ed by Enikolopo and Zhu a skaya (2003);
Iimi (2004), Buse (2011), A is o nik (2012), Hani e al. (2020), in de eloping coun ies, he impac o
decen aliza ion on economic g ow h a ies based on he quali y o poli ical go e nance wi hin he
coun y. Consequen ly, A is o nik (2012) a ibu es he limi ed success o iscal decen aliza ion in
Eas e n Eu opean economies o he absence o such a o able ins i u ional condi ions. Mak eshanska-
Mladeno ska and Tashe ska (2019) also con i med on he sample o CEE coun ies ha
decen aliza ion may be pa icula ly ha m ul o economic g ow h in he coun ies in ea ly s ages o
economic de elopmen , whe e he adminis a i e capabili y o local go e nmen s is insu icien , and hey
may no be esponsi e o p e e ences o he local ci izens.
In conclusion, he impac o iscal decen aliza ion on economic g ow h con inues o be a subjec o
ongoing deba e. Despi e con lic ing esul s om nume ous empi ical s udies, his pape aims o explo e
he connec ion be ween iscal decen aliza ion and economic g ow h ac oss a comp ehensi e da ase
co e ing ad anced and de eloping Eu opean coun ies du ing he ex ended pe iod om 1972 o 2012.
Me hodology and Da a
Model
Ou empi ical s udy ocuses on examining he impac o iscal decen aliza ion on economic g ow h in
Eu opean coun ies. The dependen a iable in ou model is he annual eal g ow h a e o GDP pe
capi a and ou p ima y subjec o in e es is he coe icien on he iscal decen aliza ion a iable, which
is expec ed o be posi i e and signi ican gi en he con en ional a gumen s in a o o iscal
decen aliza ion.
48
Figu e 1 plo s he a e age GDP pe capi a g ow h a e in Eu opean coun ies on he a e age sha e o
local go e nmen expendi u es in o al go e nmen expendi u es o e he s udy pe iod and gi es an
ini ial assessmen o he main esea ch opic. Fi s , he simple linea associa ion be ween bo h a iables
seems o show he exis ence o a sligh ly nega i e ela ionship be ween economic g ow h and he le el
o iscal decen aliza ion. Second, he ela ionship be ween economic g ow h and iscal decen aliza ion
is be e explained by a quad a ic han a linea unc ion o m, indica ing ha he e is some op imal le el
o iscal decen aliza ion in e ms o economic g ow h.
Figu e 1: The link be ween economic g ow h and iscal decen aliza ion in Eu opean coun ies
Sou ce: Au ho s’ calcula ions.
As we can see om he simple sca e plo analysis, he ela ionship be ween economic g ow h and
iscal decen aliza ion is mo e likely o be non-linea han linea . The nega i e sign o he coe icien o
quad a ic e m o he iscal decen aliza ion sugges s an inc easing slope o he economic g ow h
unc ion a lowe le el and dec easing slope a highe le el o decen aliza ion. This implies ha , as
sugges ed by Thiessen (2003), he e is some op imal le el o iscal decen aliza ion ha maximizes
economic g ow h. The e o e, in ou empi ical model we also es o he non-linea ela ion be ween
decen aliza ion and g ow h.
Howe e , he in o ma ion p o ided by he g aph should be in e p e ed wi h cau ion, as economic g ow h
does no depend exclusi ely on he deg ee o iscal decen aliza ion o a coun y and omi ed a iables
may a ec he obse ed ela ionship. The e o e, in o de o es whe he iscal decen aliza ion ma e s
o economic g ow h, we es ima e he econome ic model.
In speci ica ion o ou econome ic model, we ha e ollowed he mos common used analy ical
amewo k in oduced by Da oodi and Zou (1998). The model is de i ed om Ba o’s economic g ow h
model (Ba o, 1990), whe e economic g ow h is based on a Cobb-Douglas p oduc ion unc ion o
physical capi al, human capi al, and public spending by di e en le els o go e nmen . In addi ion,
ollowing Da oodi and Zou (1998), Akai and Saka a (2002), Ma inez-Vazquez and McNab (2003),
Filippe i and Sacchi (2016), e c., we also include se e al con ol a iables o accoun o o he ac o s
ha could in luence economic g ow h, such as demog aphic a iables and mac oeconomic policy
a iables.
The e o e, he empi ical model is ep esen ed as ollows:
𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑔𝑟𝑜𝑤𝑡ℎ𝑖𝑡
= 𝛽0 + 𝛽1 𝑝ℎ𝑦𝑠𝑖𝑐𝑎𝑙 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑡 + 𝛽2ℎ𝑢𝑚𝑎𝑛 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑡 + 𝛽3𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑖𝑡
+ 𝛽4𝑚𝑎𝑐𝑟𝑜𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠𝑖𝑡+ 𝛽5𝑑𝑒𝑐𝑒𝑛𝑡𝑟𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑖𝑡 + 𝑢𝑖𝑡
y = -0.003x2 + 0.1353x + 1.109
R² = 0.1591
y = -0.0079x + 2.4392
R² = 0.0056
0
1
2
3
4
5
6
0 5 10 15 20 25 30 35 40 45 50
Economic g ow h
Fiscal decen aliza ion
49
We employ an unbalanced panel da a eg ession model ha spans mul iple coun ies and pe iods,
enabling us o le e age a g ea e olume o obse a ions. As ecommended by he Hausman es , we
u ilized a ixed e ec s model. To co ec o he e oscedas ici y and se ial co ela ion, we applied Whi e
c oss-sec ion weigh s and a i s o de au o eg ession componen (AR1), espec i ely.
Da a and Va iables
In ou empi ical s udy we use a comp ehensi e panel da ase co e ing 31 Eu opean coun ies (including
27 EU membe coun ies, along wi h No way, Iceland, Swi ze land, and G ea B i ain) o e he pe iod
o 1972-2012. Addi ionally, we explo e he long- e m e ec o a subsample o ad anced coun ies. In
o de o do his, we a e aged he da a o i e-yea pe iods o smoo h he da a o e he mac oeconomic
cycle. In a p e ious empi ical s udy, we ha e examined he impac o iscal decen aliza ion on economic
g ow h in he Cen al and Eas e n Eu opean coun ies using he same app oach (Mak eshanska-
Mladeno ska and Tashe ska, 2019), hus enabling he explo a ion o po en ial a ia ions in he in luence
o iscal decen aliza ion on economic g ow h be ween ad anced and de eloping coun ies.
The dependen a iable in he model is he economic g ow h i.e. he annual eal g ow h a e o GDP pe
capi a, wi h da a sou ced om he Wo ld De elopmen Indica o s Da abase. The main explana o y
a iable o in e es in he model is iscal decen aliza ion. To measu e iscal decen aliza ion, we use
wo mos con en ional measu es: expendi u e decen aliza ion (local go e nmen expendi u es o o al
go e nmen expendi u es a io) and e enue decen aliza ion (local go e nmen e enues o o al
go e nmen e enues a io). These wo measu es ela e o di e en aspec s o decen aliza ion, so hey
may esul in di e en ou comes. Expendi u e decen aliza ion may no always imply e ec i e
decen aliza ion, while e enue decen aliza ion al hough is ha de o achie e, i may e e o highe
accoun abili y and mo e e icien expendi u es a local le el o go e nmen . The da a se ies o he iscal
decen aliza ion indica o s a e de i ed om Fiscal Decen aliza ion Da abase o he Wo ld Bank
1
.
Physical capi al is ep esen ed by wo a iables: he g oss sa ings o GDP a io and he annual change
o g oss ixed capi al o ma ion o GDP a io, while selec ing sui able indica o s o human capi al was
challenging, as mos es ed a iables we e s a is ically insigni ican . Howe e , i is di icul o dismiss he
impo ance o human capi al o economic g ow h. The lack o s a is ical signi icance o hese a iables
is p obably due o limi a ions and da a se ies discon inui y. F om he a ious es ed a iables, we op ed
o include he seconda y school en ollmen a io (in addi ion, we also es ed he public educa ion
expendi u es, public e enues o science and esea ch, uni e si y en ollmen a e, he numbe o pa en
applica ions and esea che s pe million ci izens).
As o he o he con ol a iables, mac oeconomic policy a iables include public sec o size (gene al
go e nmen expendi u es o GDP a io), budge balance o GDP a io, in la ion (annual g ow h a e o
CPI) and ade openness (impo s and expo s o goods and se ices o GDP a io). To enhance he
model’s explana o y capaci y, we also inco po a e a ious demog aphic a iables ha a e equen ly
obse ed in exis ing economic g ow h s udies: popula ion g ow h a e, u ban popula ion sha e, and he
dependency a io. The da a o hese a iables a e ex ac ed om he Wo ld De elopmen Indica o s
Da abase.
Resul s and Discussion
Table 1 displays he empi ical esul s on he impac o iscal decen aliza ion on economic g ow h ac oss
a comp ehensi e sample o 31 Eu opean coun ies and a subsample o 17 ad anced Eu opean
coun ies. Reg ession equa ions (1) and (5) e e o expendi u e decen aliza ion o he en i e sample
and he ad anced coun ies’ subsample, espec i ely. In con as , eg ession equa ions (2) and (6) a e
associa ed wi h he e enue decen aliza ion in he espec i e samples. Equa ions (3) and (4) e e o
he esul s o he non-linea ela ionship es be ween decen aliza ion and g ow h o he en i e sample
o 31 Eu opean coun ies, discussed in he nex sec ion.
1
Gi en ha he analyzed ime se ies spans 40 yea s, we e ained om ex ending he da a using addi ional iscal
decen aliza ion da abases. Ou decision was d i en by he desi e o main ain da a consis ency and esul s in eg i y.
Ins ead, we sou ced he en i e da ase om he Fiscal Decen aliza ion Da abase o he Wo ld Bank, which
un o una ely p o ides decen aliza ion da a only up un il 2012.
50
Ou empi ical esul s indica e ha iscal decen aliza ion in gene al has a a o able impac on economic
g ow h in he Eu opean coun ies, as he linea ela ionship be ween expendi u e and e enue
decen aliza ion and GDP pe capi a g ow h a e appea s o be posi i e in all equa ions. Fu he , as
expec ed, he e enue decen aliza ion is shown o be mo e e ec i e in e ms o s imula ing economic
g ow h. Howe e , o he whole sample, he posi i e ela ionship be ween decen aliza ion and g ow h
is ela i ely weak. Namely, he coe icien o expendi u e decen aliza ion, al hough posi i e, is a he
small and insigni ican (equa ion 1), while he coe icien o e enue decen aliza ion is posi i e and
s a is ically signi ican a he 10% con idence le el (equa ion 2). On he o he hand, in he subsample o
ad anced Eu opean coun ies, when he long- e m e ec s o decen aliza ion on g ow h a e examined
using 5-yea a e ages, bo h, expendi u e and e enue decen aliza ion appea o ha e a s a is ically
signi ican posi i e impac on economic g ow h (equa ions 5 and 6).
F om ou empi ical esul s we can implici ly conclude ha iscal decen aliza ion has a mo e a o able
e ec on g ow h in mo e de eloped Eu opean coun ies, p obably due o hei highe go e nmen
spending e iciency compa ed o he less de eloped Cen al and Eas e n Eu opean coun ies (e.g.
Halasko a e al., 2022). This inding is also con i med by ou p e ious empi ical s udy on he e ec s o
iscal decen aliza ion on he economic g ow h in 11 Cen al and Eas e n Eu opean coun ies, o e he
pe iod om 1992-2012, whe e we ound ha iscal decen aliza ion has a no ably ad e se e ec on
economic g ow h (Mak eshanska-Mladeno ska and Tashe ska, 2019).
The e o e, he empi ical indings om bo h ou s udies a e in line wi h he p e iously s a ed con en ional
a gumen ha he mac oeconomic e ec s o decen aliza ion do no solely depend on he size o he
local go e nmen expendi u e o e enue decen aliza ion. On con a y, decen aliza ion is expec ed o
ha e mo e p onounced posi i e e ec on economic g ow h in economies whe e ce ain ins i u ional and
poli ical p econdi ions a e me , such as: ins i u ional quali y, iscal au onomy o subcen al go e nmen s,
le el o democ a iza ion o he coun y, e c. (see Buse , 2011). The e o e, iscal decen aliza ion, when
quan i a i ely measu ed as in ou case, is expec ed o ha e a ying mac oeconomic implica ions ac oss
di e en coun y g oups due o di e ences in he quali a i e aspec s o he decen aliza ion p ocess.
Table 1: Empi ical esul s
Dependen a iable: Annual g ow h a e o GDP pe capi a
Independen a iables
1
2
3
4
5
6
Exp decen aliza ion
0.0099
0.0589
**
0.0361
**
0.0116
0.0291
0.0178
Exp decen aliza ion ^2
-0.0006
**
0.0003
Re decen aliza ion
0.0260
*
0.1078
***
0.0457
**
0.0154
0.0352
0.0222
Re decen aliza ion ^2
-0.0010
***
0.0004
Go e nmen
expendi u es
-0.0495
**
-0.0520
**
-0.0558
**
-0.0575
***
-0.0708
***
-0.0622
***
0.0235
0.0227
0.0223
0.0212
0.0200
0.0178
Budge balance
0.0967
**
0.1032
**
0.0947
**
0.1068
**
0.0451
0.0446
0.0444
0.0444
In la ion
0.0015
0.0015
0.0015
0.0016
0.0448
***
0.0451
***
0.0017
0.0017
0.0017
0.0017
0.0152
0.0164
Sa ings
0.0097
0.0099
0.0095
0.0120
-0.0286
-0.0166
0.0331
0.0340
0.0339
0.0348
0.0333
0.0350
Capi al (g ow h)
0.2354
***
0.2365
***
0.2369
***
0.2379
***
0.3007
***
0.2994
***
0.0275
0.0275
0.0272
0.0271
0.0200
0.0196
Pa en s
0.1011
0.1105
0.0981
0.1213
-0.2807
-0.2770
0.1303
0.1265
0.1308
0.1269
0.1143
0.1155
School
0.0105
0.0094
0.0079
0.0062
0.0083
***
0.0086
***
0.0077
0.0077
0.0075
0.0076
0.0025
0.0024
Openness
0.0238
**
0.0224
**
0.0214
**
0.0202
**
-0.0055
-0.0039
0.0096
0.0093
0.0098
0.0093
0.0046
0.0047
Popula ion
-1.4607
***
-1.5029
***
-1.4945
***
-1.5199
***
-1.9966
***
-1.9766
***
0.2011
0.1974
0.2032
0.1926
0.1235
0.1228
U baniza ion
-0.1416
***
-0.1380
***
-0.1414
***
-0.1350
***
-0.0716
***
-0.0691
***
0.0429
0.0423
0.0431
0.0427
0.0128
0.0123
Dependency
-0.0603
***
-0.0563
***
0.0225
0.0209
51
Cons an
10.2681
***
9.8653
***
10.2416
***
8.9538
***
15.1846
***
13.5782
***
3.4600
3.3563
3.4616
3.3263
1.3168
1.7182
AR(1)
0.1619
0.1492
0.1546
0.1444
-0.1504
**
-0.1802
**
0.1018
0.0990
0.1001
0.0987
0.0683
0.0757
R-squa ed
0.7935
0.7962
0.7961
0.7992
0.9344
0.9315
R-squa ed adjus ed
0.7778
0.7807
0.7801
0.7836
0.9047
0.9006
F-s a is ic
50.4924
***
51.4460
***
50.0172
***
51.1039
***
31.5289
***
30.1164
***
Du bin-Wa son s a is ics
1.9127
1.9154
1.9141
1.9174
2.3012
2.3325
In e ed AR Roo s
0.16
0.15
0.15
0.14
-0.15
-0.18
C oss - sec ion
31
31
31
31
17
17
Sample
1972-
2012
1972-
2012
1972-
2012
1972-
2012
1976-
2012
1976-
2012
Obse a ions
595
596
595
596
91
91
No e: Whi e he e oskedas ici y consis en s anda d e o s a e p o ided below he coe icien s.
* 10%, ** 5%, *** 1% le el o signi icance.
Sou ce: Au ho s’ calcula ions.
Fu he , in addi ion o he linea ela ionship be ween decen aliza ion and g ow h, we also es o he
nonlinea e ec s o decen aliza ion on economic g ow h. Speci ically, equa ions (3) and (4) examine
he quad a ic ela ionship be ween decen aliza ion and economic g ow h ac oss he en i e sample.
These equa ions inco po a e he quad a ic o ms o he iscal decen aliza ion indica o s, namely he
quad a ic sha e o local expendi u es and local e enues in gene al go e nmen expendi u e and
e enues, espec i ely. The esul s e eal a s a is ically signi ican quad a ic ela ionship be ween iscal
decen aliza ion and economic g ow h, when conside ing bo h expendi u e and e enue
decen aliza ion. The nega i e sign associa ed wi h he decen aliza ion a iable indica es an in e ed
“U” (pa abolic) shape, implying ha inc easing decen aliza ion ini ially suppo s economic g ow h, bu
beyond a ce ain “op imal” le el o decen aliza ion, u he decen aliza ion s a s o decele a e g ow h.
In his con ex , i has been no ed in he li e a u e ha excessi e decen aliza ion can make policy
coo dina ion mo e complex and ha m mac oeconomic s abili y ia iscal policy coo dina ion p oblems.
Rega ding he o he es ed a iables, he public sec o size nega i ely impac s he GDP pe capi a
g ow h a e o he en i e sample and o he subsample o ad anced coun ies. The posi i e in luence
o he budge balance on he g ow h a e is also con i med o he en i e sample. The le el o ade
openness posi i ely impac s g ow h in he en i e sample, while in he subsample o ad anced coun ies
i loses signi icance. In e ms o physical capi al, he g ow h a e o g oss ixed capi al o ma ion has a
s a is ically signi ican posi i e e ec bo h o he en i e sample and o he sample o ad anced
coun ies, con i ming he g ow h-inducing e ec o in es men s in capi al. On he o he hand, in la ion
appea s o be a ele an ac o only o he subsample o ad anced coun ies, wi h a s imula ing e ec
on g ow h. Seconda y school en ollmen seems o be a signi ican con ibu ing ac o o g ow h only in
he equa ions o ad anced coun ies as well. Rega ding he demog aphic a iables in he model, he
popula ion g ow h a e and he sha e o u ban popula ion con i m he s a is ically signi ican nega i e
impac on g ow h, in bo h he en i e sample and he subsample o ad anced coun ies. Fo he
subsample o ad anced coun ies, he impac o he dependency a io was also conside ed, and i had
a s a is ically signi ican nega i e coe icien , unlike he case o he CEE coun ies, whe e i had a
signi ican ly posi i e coe icien (Mak eshanska-Mladeno ska and Tashe ska, 2019). This migh be
because he dependency a io is a g ea e in he ad anced coun ies, and hey a e on a highe s age
o demog aphic ansi ion compa ed o he new membe s a es. Empi ical e idence shows ha on a
lowe le el o demog aphic ansi ion, popula ion ageing can ha e bene icial e ec s on economic ac i i y
and g ow h, as mo e people en e he wo k o ce, howe e a la e s ages he inc easing sha e o olde
(inac i e) popula ion lead o diminishing esul s. The o he con ol a iables, as he a io o sa ings o
GDP and pa en s, did no ha e s a is ically signi ican coe icien s.
Conclusion
The empi ical analysis o he ela ionship be ween iscal decen aliza ion and economic g ow h in he
Eu opean coun ies shows a posi i e o e all impac o e enue and expendi u e decen aliza ion on
g ow h a e o GDP pe capi a. Fu he mo e, he ad anced Eu opean coun ies expe ience a mo e
