External Debt and Economic Growth in the Western Balkan Countries,
with Special Focus to Albania, Kosovo and North Macedonia in the
course of the Pandemic COVID-19
BARDHYL DAUTI
Department of Economics, Faculty of Economics, University of Tetovo
Ilindenska road, nn, 1200, Tetovo,
REPUBLIC OF NORTH MACEDONIA
ISMET VOKA
Department of Economics, Faculty of Economics, Business and Development,
European University of Tirana,
Xhanfize Keko road, no. 56, Tirana, Albania
ALBANIA
Abstract: - The objective of this study is to offer an empirical valuation of the relationship between external
debt and economic growth in the Western Balkan (WB) countries, focusing specifically on the countries like
Albania, Kosovo and North Macedonia, combined with other WB countries like Bosnia and Herzegovina,
Montenegro and Serbia. The empirical model provides the impact of external debt and other control variables
like total investments, population growth, inflation, literacy ratio, trade openness on economic growth in the
Western Balkan countries, using a panel level data for 6 Western Balkan countries, covering a yearly time span:
2000-2022. Different estimation methodologies like Fixed Effects with Driscol and Kraay standard errors,
robust LSDV and GMM estimates, were employed for the purpose of the research. The findings of the research
confirm growth-deteriorating effect of external debt for target group of countries like Albania, Kosovo and
North Macedonia and growth enhancement effect of external debt for the second group of countries like Bosnia
and Herzegovina, Montenegro and Serbia. Other control variables like total investments, trade openness,
inflation and population growth are found as crucial factors on explaining growth performance of the WB
countries. In addition, COVID-19 interacted with external debt and financial crisis interacted with external
debt, appears as crucial factors explaining growth pattern of the WB countries.
Key-Words: - Economic Growth, Fiscal Policy, Western Balkan Countries, Public Finances, COVID-
19.
Received: August 8, 2021. Revised: June 7, 2022. Accepted: June 24, 2022. Published: July 25, 2022.
1 Introduction
The association between external debt and economic
growth is of essential importance for the Western
Balkan (WB hereafter) countries, once having
regard the general public debt limit that these
countries should impose for fulfilling economic
prerequisites for EU adherence criteria, in line with
fiscal policy framework of the European Union [16].
Moreover, the debt component of the WB countries
is heavily dependent upon external debt. The last
two decades of the transition period, covering the
years of 90th and 20th, WB countries possessed low
level of capital accumulation, due to different
problematic political and economic circumstances
the region went through in the near past, thus,
making these countries very much likely to finance
their investment needs at their early stage of
development using external debt. Albania, Kosovo
and North Macedonia, subjected as specific
countries in the study, constitute extreme cases with
respect to debt cyclical component in relation to
GDP. While Albania and Kosovo have almost
constantly recorded debt to GDP ratio at thrilling
values, representing the highest values for Albania
and lowest one for Kosovo, North Macedonia on the
other hand, follows debt rule limit specified by the
Commission of the European Union, which is 60 %
of GDP [16]. Therefore, the study gives a special
priority to these three countries, within the sample
of the Western Balkan countries, once having regard
the heterogeneous nature of the debt cyclical
component in the sample countries of WB-6.
Many scholars have empirically tested the growth
effect of external debt, by confirming a growth
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1303
Volume 19, 2022
deteriorating effect of external debt, due to costly
investments raised by costly servicing activities [12,
24] and growth enhancement effect of external debt,
which mainly arise in cases when the domestic
capital is insufficient to finance growth [15, 18].
Furthermore, the relationship between external debt
and economic growth in terms of practical
institutional life is also crucial in the academic
debate. External debt has a catalyst impact on
investments; savings and capital inflow, implying
that foreign savings complement domestic savings,
thus, satisfying the investment demand, especially in
transition countries who face limited financial
resources for financing investment needs [18].
However, due to ‘’debt overhang’’
1
and ’crowding
out’
2
effect of external debt on investment
activities, a deteriorating effect of external debt on
economic growth is foreseen, making foreign capital
inflow to drop down, due to macroeconomic
instability, which on the other hand can utilize
further growth adverse effects [30; 26]. The
empirical evidence on growth deteriorating effect of
external debt advocates that countries with lack of
institutional efficiency are more likely to experience
growth adverse effects from the external debt, in
cases when external debt approaches to 15-30
percent of GDP [11] or is in the range of 20 percent
of GDP [13]. However, other studies suggest that
growth-deteriorating effect of external debt becomes
more severe even in cases when the threshold level
of debt limit reaches on average, 35-40 percent of
GDP [37].
The main motivation of the study is to empirically
examine the external debt-growth nexus model’’
in a sample of the six WB countries, with a special
focus on the selected WB countries like Albania,
Kosovo and North Macedonia, using a panel
regression analysis, during the yearly period from
2000 to 2022
3
. There are three-research questions
addressed in the study: What is the nature of the
impact of external debt in the two group of WB
1
Exists in the cases where actual debt overcomes the
anticipated debt, thus, making the countries repaying
ability problematic.
2
If the external debt is serviced mainly through the
foreign capital, a little room is left for enhancement effect
of investments on growth, on the second cycle of the
economic activity. In this case, the cost of servicing the
public debt, via external debt can crowd out public
investment expenditures, thus, reducing the total
investments and complementing the private investments
[25]
3
The data for the year of 2022 are projected based on a
three years moving average, covering in principle the
years of 2019, 2020 and 2021.
countries, whether it is growth declining or growth
enhancement? What is the impact of the pandemic
COVID-19 on external debt in both group of WB
countries, once having regard that external debt
during the COVID-19 crisis went through positive
cyclical movement, with the aim of keeping stable,
the sustainability of public finances and what is the
impact of external debt on economic growth during
the financial crisis turmoil. The outlined results of
the study are likely to offer an intuition for the
policy makers of these countries regarding whether
or not the accumulation of additional external debt
should be stimulated or depressed. Furthermore, the
results of the study suggest growth-declining effect
of external debt for the WB countries like Albania,
Kosovo and North Macedonia and growth
enhancing effect of external debt for countries like
Bosnia and Herzegovina, Montenegro and Serbia.
The findings of the paper also suggests growth-
deteriorating effect of external debt during COVID-
19 period and growth enhancement effect of
external debt during financial crisis period, for the
whole sample of the WB 6 countries. The
structure of the paper is organized as follows: The
coming section present a review of literature on debt
nexus-growth relationship. Third section stylizes
some facts regarding the debt cyclical behaviour in
relation to GDP and debt policies in the WB
countries. Section four presents the research
methodology as well as the empirical models of the
panel data followed by hypothesis. Section five
presents the results of the study and the last section
concludes the study.
2 Literature Review
The growth nexus debt based relationship has
produced ambiguous results with respect to
estimated impact of external debt on economic
growth. The growth enhancement effect of external
debt in the developing countries is mainly supported
in cases where borrowed funds take place in
profitable projects, subjected by lasting
macroeconomic stability of these countries [37].
The growth adverse effect of external debt is
explained through ‘’debt overhang effect’’, which
makes the country’s debt repaying ability costly in
relation to the benefit of the earlier borrowings,
thus, discouraging further domestic and foreign
investments [30]. Furthermore, as concern to
developing and transition countries, as it is the case
of the Western Balkan countries, external debt is
valued in a foreign currency, being exposed to risks
associated with exchange rate fluctuations, which
increases the likelihood of debt adverse effect of
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1304
Volume 19, 2022
external debt into these countries [6]. Growth
enhancement effect of external debt is grounded on
capital accumulation and productivity growth that
external debt usually causes on transition countries,
who typically do not have sufficient domestic
capital and investment opportunities at early stage
of development. Further one, in this section as
outlined in the following table, we show the results,
which confirm growth deteriorating and growth
enhancement effect of external debt and growth
neutral effect of external debt on economic growth,
through different channels.
Table 1. Some of the late empirical studies on the
external debt nexus economic growth
Author
and
year
Sample
and
period
Findings
Growth deteriorating effect of external debt
[12]
CEE:
period:
1995-
2003
Growth adverse
effect of external
debt by decreasing
the investments, due
to allocation of
savings on debt
servicing activities
and increases of the
overall
macroeconomic
risks for domestic
and foreign
investors leading to
further reduction in
investments.
[24]
20 high
external
debt
countrie
s
selected
from
Asia
and
Latin-
Americ
a: e
period
1991-
2004
External debt as a
share of GDP is
negatively associated
to growth, signifying
that excessive debt is
unfavourable to
economic growth.
[28]
Oman:
period:
1990-
2015.
The study exposes a
negative effect of
external debt on
economic growth in
Oman.
[41]
five
Sub-
Sahara
The findings
indicate that
external debt is
n
African
(SSA)
countri
es:
period
1990~2
015
OLS and
dynamic
OLS
negatively and
significantly
associated to
economic growth.
[1]
23
Low-
income
countrie
s. Using
data
over the
period
2000-
2017
SUR model
External debt
meaningfully
declines investment
and economic growth
for both, the total
sample and the sub-
samples.
Growth enhancement effect of external debt
[2]
111
countri
es;
period
1971-
2010
FE and
2SLS
estimations
071-879-
009
Continental countrie
s (AUT, BEL, FRA,
GER, ITA, and
NETH) face more
growth decreasing
public debt effects
than
mainly Liberal state
s (AUS, CAN, IRL,
NZ, CH, USA, UK).
Public debt
seemingly utilises
unbiased or positive
growth effects,
while
for Nordic states
(DEN, FIN, NOR,
SWE) a non-linear
relationship is
exposed, with
negative debt
effects, by around
60% of GDP.
[17]
10
former
commu
nist
countri
es,
membe
r
countri
es of
the EU
Quadratic
regression
equation
A non-linear
relationship between
government debt to
GDP ratio and the
per capita GDP
growth rate is
found. Moreover,
the authors found
that turning point of
government debt is
50%. If the
government debt to
GDP ratio surpasses
this level, it could
generate a negative
impact on the GDP
growth rate.
[41]
32
Dynamic
Public debt has
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1305
Volume 19, 2022
states
from
1993 to
2012
enhancement impact
on investments,
which in turn
stimulates growth,
suggesting that a
production based
generation of public
debt.
[36]
Ghana;
period
1970 to
2012
The study claim
positive long-run
relationship between
public debt and
economic growth. In
addition, in the short
run a bidirectional
Granger causality
link exists between
public debt and
economic growth.
Growth effect of external debt based on threshold level
[35]
38
African
countri
es:
period
1980-
2010
a low (high) level of
debt does not have a
significant effect on
growth,
[43]
Vietna
m:
period
2000-
2013
1% increase of
external debt,
increases growth by
1.29%; above
(below) 21.5%,
threshold level of
external debt it
deteriorates
(stimulate) growth.
[44]
10
countri
es:
period
2005-
2015
External debt below
(above) the
threshold level of
33.17% is positively
(negatively)
associated to
growth. However, at
the threshold of
33.17%, a 1%
increase in external
debt decreases GDP
growth by 0.02%.
Note: Summary papers with empirical studies.
As concern to the relationship between external debt
and economic growth, at panel level, some of the
studies, as presented on table 1, confirm growth-
enhancing effect of external debt, mainly driven by
public investment, financed by the debt component
[42]. In line with these findings, at country level, is
the study of the relationship between public debt
and economic growth in Ghana, which confirm
growth-enhancing effect of external debt, using
VECM and Johansen Cointegration technique [36].
Some other studies confirm non-linear relationship
between external debt and economic growth,
depending from the threshold level of external debt
[44; 17] and some other studies outline growth
deteriorating effect of external debt driven by
macroeconomic risks associated by debt financing
component of private investments [12]. For the
Western Balkan countries, at panel level, the
empirical literature on external debt growth nexus
relationship is scare, being in general of a
descriptive nature. The paper will add value on the
tested hypothesis related to the impact of external
debt as a crucial fiscal sustainability factor on
economic growth for the WB countries and hence
contribute to maintaining a healthy fiscal
convergence policy for the EU economic
approximation path of the WB countries.
3 Data and Stylized Facts
Global development finance (2000) and World Bank
reports, gives a special insight on defining countries
indebtedness level, based on the ratio of the stock of
external debt (ED) to Gross National Income (GNI).
For less indebted countries, this ratio is below 48
percent; for countries that are more indebted it is in
between 48 and 80 percent and highly indebted
countries have a ratio of ED/GNI above 80 percent.
As viewed from figure 1 the higher average
intendedness level for WB-6 is recorded in 2020,
generally accepted as a pandemic year, which
provoked debt level due to borrowing from
international financial institutions to finance
liquidity concerns of the private sector of the
respective WB countries [16].
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1306
Volume 19, 2022
Fig. 1: External debt stock as a share of Gross
National Income in the Western Balkan Countries.
Note: Total external debt is the sum of public,
publicly guaranteed, and private nonguaranteed
long-term debt, use of IMF credit, and short-term
debt. GNI Gross National Income (formerly GNP
Gross National Product) is the sum of value added
by all resident producers plus any product taxes
(less subsidies) not included in the valuation of
output plus net receipts of primary income
(compensation of employees and property income)
from abroad.
Source: World Bank, World Development Indicator,
2022 and author’s calculations.
The leading country in terms of debt level during
the last decade, observed in two periods (2011-
2018) and (2019-2022), on average, is Montenegro,
with average recorded external debt in terms of
Gross National Income by 144.03 and 171.32
percent, respectively, and followed by North
Macedonia and Serbia. Among the WB countries,
the lowest debt level sourced from external sources
was recorded in Kosovo, which is an example of
country, being less likely to finance the economic
activities from foreign sources.
4 Methodology and Econometric
Framework
4.1 Unit Root Test
The unit root tests for the variables employed in the
model are performed in order to examine the
stationary trend of the data, and therefore avoid
spurious regression and incorrect inferences. We
employee panel unit root test, to allow for fixed
effects and unit specific time trends, since their test
includes a degree of heterogeneity [32]
4
. The
4
Following Leving et al. (2002), null and alternative
hypotheses are , where and
variables in levels (integrated at order zero) and first
difference (integrated at order 1), for 0 and 1 time
lag are found to be significant by the definition
panel unit root test, implying the rejection of the
null hypothesis for the presence of a unit root in the
data, making the panels stationary [32].
Table 2. Unit root tests 2000-2020
Lin Levin and Chu unit root test: Ho:
Panels contain unit roots; Ha: Panels are
stationary
Variab
les
Adjusted t
statistic [p
value]
Adjusted t
statistic [p
value]
ADF
regressi
on
I (0) levels
I (1) First
difference
Nr of
lags

-7.67[0.00]***
-15.67[0.00]***
0

-2.78[0.02]**
-3.23[0.00]***
1

-3.47[0.00]***
-13.69[0.00]***
0

-1.70 [0.04]**
-5.63[0.00]***
1

-1.851[0.03]**
-5.751[0.00]***
0

-0.81 [0.20]
-4.24[0.00]***
1


-3.06[0.00]***
-8.52[0.00]***
0


-1.41[0.07]*
-5.56[0.00]***
1

-4.64[0.00]***
-11.06[0.00]***
0

-1.72[0.04]**
-4.79[0.00]***
1

-3.35[0.00]***
-9.86[0.00]***
0

-3.25[0.00]***
-6.02[0.00]***
1

-1.32[0.09]*
-10.36[0.00]***
0

-0.48[0.31]
-1.37[0.08]*
1
Note: ***, ** and *, indicate rejection of the unit-
root hypothesis at the significance level of 1%, 5%
and 10%, respectively.
All the variables are stationary in levels, with
exception to population growth variable, which is
becoming stationary at the first difference.
Therefore, the stationary variables in the regression
equation are specified in levels, with exception of
population growth, which is considered in the first
difference, as a non-stationary variable.
4.2 Econometric Framework
The paper will try to shed light on the impact of
external debt on economic growth, as a crucial fiscal
sustainability factor, in the Western Balkan
countries, relying on a yearly panel data set for the
period 2000-2022. The reduced form of the growth
equation for the estimation purpose is as follows:
 
  

 󰇛󰇜
<1, respectively. The alternative hypothesis
assumed the same degree of stationarity across countries.
29,94
59,16
47,16
42,16
69,53
23,15
65,66
70,48
68,59
144,03
73,96
31,15
62,88
65,54
74,73
148,79
66,81
30,05
74,19
72,13
89,93
74,39
39,33
67,44
67,92
78,66
164,52
68,89
32,80
68,17
68,53
81,11
171,32
70,03
34,06
A L B B I H M K D M N E S R B K O S
2000-2010 2011-2018 2019 2020 2021 2022
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1307
Volume 19, 2022
Where  is the dependent variable denoting the
growth level of the WB countries,  is the focus
variable of external debt, 
 is the vector of
control variables, is the regional dummy
variables denoting the geographical location of the
southern WB countries. is the dummy variable
denoting the pandemic COVID-19 period. is the
dummy variable denoting the financial crisis. is
the constant. is country dummy, is year
dummy and  is usual error term. Extending the
approach of Abdelaziz et al (2019), the equation for
estimating the impact of external debt on economic
growth in the Western Balkan countries is the
following:
 
  
  


 󰇛󰇜
Where  is the country index,
 is the time index, denoting the years from
2000 to 2022. The empirical model assumes that
growth level of the WB countries is a function of
initial level of growth, external debt
5
and the control
variables
6
like: inflation, total investment, trade
openness, schooling, population growth, the
interaction terms between external debt with
regional dummy, COVID dummy and
financial crisis dummy, , as well as the
constitutive terms of the respective dummy
variables. With respect to the role of the interaction
terms with the regional dummy, the aim of the study
is to differentiate the debt impact on economic
growth across two group countries, the southern
group of the WB countries, like Albania, Kosovo
and North Macedonia and the north group of the
WB countries, like Bosnia and Herzegovina,
Montenegro and Serbia. In the same way, as
concern to the interaction term between external
debt and, denoting covid dummy, the aim of the
study is to differentiate the impact of external debt
on economic growth across two periods, the
pandemic period and the normal period, which is
considered as a benchmark category of period in
relation to the covid period. Following this logic, the
inclusion of the interaction term between external
debt and financial crisis dummy in the model,
5
This variable constitute the main interest of the study.
6
The control (regime) variables are included in the model
to augment the regression model of the growth equation,
which explain the variation of growth level of the WB
countries that may affect the relationship between growth
and external debt.
serves for differentiating the impact of external debt
on economic growth across two periods, the
financial crisis period, which occurred during the
year 2008 and the normal period. For the static
panel model, we rely on fixed effects with Driscoll
and Kraay standard errors (FEDK). The FEDK
estimates are asymptotically efficient in the panel
samples where time series, ‘Texceeds the number
of panels ‘N’ [15, 23]. By relying on large T
asymptotic, FEDK estimates are robust to general
forms of cross-sectional as well as temporal
dependence as well as to heteroscedasticity and
autocorrelation [23]. We also control for time and
country fixed effects FEDK estimates. As a
robustness check to the FEDK estimates, we also
use Least Square Dummy Variable estimates, in
order to evaluate the net effect of each regressor,
accounting also for unobserved heterogeneity [8,
22]. We address the issue of the lagged dependent
variables, as well as the concerns with respect to
unobserved fixed effects and endogenous
independent regressors, accounting also for
heteroskedastic and auto correlated standard errors
across panel members, by using General Method of
Moments (GMM) [3; 5; 7; 39; 40]. The dynamic
panel data model can be expressed as follows:
 
 
󰆒 󰇛󰇜
Where,
 is the dependent variable,
 is the
lagged dependent variable, 
󰆒 is the set of
explanatory variables and  is the standard error.
Dynamic panel data regression using Arrellano-
Bover/Blundell/Bond estimation procedure [3; 5] is
considered as a robustness check to LSDV
estimates. Following Roodman’s approach, we have
employed the stata command xtdpdsys. The new
xtdpdsys jointly offer most of xtabond2’s features,
while moving somewhat towards its syntax and
running significantly faster [39; 40]. The lagged
dependent variable and the variables that potentially
show high inertia with the dependent variable are
treated as endogenous components, like population
growth, trade openness as a share of GDP and
investments as a share of GDP. We use only one lag
for the dependent variable in the GMM and exclude
the dummy variables employed in static panel
models, like regional dummy, covid dummy
and financial crisis dummy.
4.3 Variable Description and Hypothesis
The dependent variabledenotes economic
growth of WB countries, calculated as a percentage
change of real GDP growth, and sourced from IMF,
world economic outlook (WEO) database of January
2022. Lagged dependent variable is introduced
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1308
Volume 19, 2022
in the GMM model to control for initial level of
growth and potential endogeneity problem.
External debt variable denoted by  is the
external debt as a share of gross national income,
sourced from WEO. The empirical literature suggest
twofold relationship between external debt and
economic growth. This variable is lagged by one
period to allow the external debt the grace period
before it starts impacting growth. We expect a
growth heightening effect of external debt, due to
capital accumulation and productivity growth that
external debt usually causes on transition countries,
which typically do not possess sufficient domestic
capital and investment opportunities at early stage
of development [9; 10]
7
. External debt is considered
as a catalyst factor for investments, savings and
capital inflow, implying that foreign saving
complement domestic savings, thus, satisfying the
investment demand [18]. However, due to ‘’debt
overhang’
8
and ‘’crowding out’’
9
effect of public
debt on investment activities, we expect a
deteriorating effect of public debt variable on
economic growth [30; 26]. The specified null
hypothesis is that the coefficient of external debt is
zero; i.e. external debt has no impact on economic
growth 󰇛 󰇜 and the alternative
hypothesis is that the external debt impacts
economic growth and the coefficient of external
debt is statistically different from zero󰇛
󰇜. By studying the relationship between external
debt and economic growth, we test the hypothesis
that indebted countries due to low capital
accumulation at early stage of development are
expected to record lower economic growth.
Inflation denoted by  is the percentage change
of the average consumer prices, sourced from WEO.
Inflation rate is the first control (regime) variable
7
WB countries usually possessed low level of capital
accumulation at the late years of 2000, due to different
political and economic circumstances they went through,
in the late years of 90th, like devastating wars and
conflicts, which caused significant macroeconomic
turbulences in these countries.
8
Exists in the cases where actual debt overcomes the
anticipated debt, thus, making the countries repaying
ability problematic.
9
If the external debt is serviced mainly through the
foreign capital, a little room is left for enhancement effect
of investments on growth, on the second cycle of the
economic activity. In this case, the cost of servicing the
public debt, via external debt can crowd out public
investment expenditures, thus, reducing the total
investments and complementing the private investments
[25].
employed in the model
10
. The empirical literature
support growth-deteriorating effect of inflation rate,
growth enhancement effect of inflation rate and
non-linear relationship between inflation and
economic growth [21; 25; 27; 33;]. We expect
bidirectional relationship between inflation and
economic growth for the WB countries, once having
regard the heterogeneous nature of the WB
countries, with respect to macroeconomic
performance. The null hypothesis in this case is that
inflation rate has no impact on economic growth;
i.e. 󰇛  󰇜 and the alternative hypothesis is
that inflation rate impacts economic
growth󰇛  󰇜. Based on the relationship
between inflation and economic growth, the
developed hypothesis is that high inflation is
expected to be associated with less growth in the
WB countries.
Total investment denoted by is the total value
of gross fixed capital formation and changes in
inventories and acquisitions less disposal of
valuable for a unit or sector, as a percent of GDP
11
,
sourced from IMF, World Economic Outlook. The
Gross Fixed Capital Formation (GFCF) is consisted
from the investments components, which mainly
come from private, public and government sector.
The empirical evidence regarding the impact of each
investment category within GFCF on economic
growth is mostly positive. Private investment is
considered to have growth enhancement effect due
to the increase of productivity from technology
spillover effect [4]. In addition, public investment
increases productivity of the private sector, which in
turn rises the economic growth [4]. Public
investments applied by governments may enhance
growth in the long run through positive spillover
effects provided by the value added from the public
goods, in terms of positive externalities that public
investments in education, physical infrastructure and
research and development contribute to growth [38].
The variable  is included in the model in its
lagged form, in order to avoid endogeneity problem
between the growth and investments, due to the high
inertia that both variables expose to each other
subjected by the two-way interactions in both
10
Control variables are included for increasing the
explanatory power of the model and chose the best fit of
the data that minimizes the error sum of square as
mention by Hansen (2000). vc
11
More specifically, Gross fixed capital formation is a
flow value who measures net investments resulting from
the difference of acquisition and disposals in fixed capital
assets by enterprises, government and households within
the domestic economy, during an accounting period.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1309
Volume 19, 2022
directions
12
. Following the empirical evidence
regarding the nexus between investments and
growth, we expect growth enhancement effect of
investments in WB countries. The null hypothesis is
that total investments have no impact on economic
growth; i.e. 󰇛  󰇜 and the alternative
hypothesis is that total investments impact economic
growth󰇛  󰇜. Accordingly, we develop
the hypothesis of a growth enhancement effect of
total investments.
Trade openness denoted by, is trade openness
measured by the sum of exports and imports over
GDP, data sourced from UNCTAD. This variable is
included in the model to capture trade liberalization
progress in the WB countries. The empirical
literature support positive association between trade
liberalization and economic growth, mainly due to
the gains that trade liberalization provides to
economic growth, like providing a steady state level
of income, reduction of corruption and smuggling,
greater economies of scale and scope, knowledge
and technology spillovers and stimulation of export
platform FDI [31; 20]. Positive relationship between
trade liberalization and economic growth is
expected for the case of the WB countries [19]. The
null hypothesis is that trade openness have no
impact on economic growth; i.e. 󰇛 󰇜
and the alternative hypothesis is that trade openness
impact economic growth󰇛 󰇜. In this
case, we test the hypothesis of a positive association
between trade openness and economic growth.
Schooling, measured in terms of the
percentage of total working-age population with
advanced education, data sourced from the World
Bank, is included in the model to account for the
impact of human capital development on economic
growth of the WB countries. There is growing
empirical literature related to positive association of
human capital with the economic growth, which is
mainly supported by the hypothesis that human
capital developments through raising of the
marginal product of physical capital, induces further
accumulation of human capital, influencing the raise
of output [4]. Both, microeconomic and
macroeconomic research approach on the relations
between education and productivity appear quite
consistent with each other and are strongly recalling
of a causal interpretation of Barro's finding of a
positive effect of educational investments on
economic growth. Therefore, it is expected that
12
In the theoretical premises, supported by many
empirical evidences, investment is potential source of
growth. In addition, Gross Capital Formation derives
Economic growth of the country (IMF, 2012).
human capital developments in the WB countries to
be positively related to economic growth. The null
hypothesis in this case is that schooling has no
impact on economic growth; i.e. 󰇛  󰇜
and the alternative hypothesis is that schooling
affects economic growth󰇛  󰇜. On the
grounds of the relationship between schooling and
economic growth, we test the hypothesis that
schooling has a positive impact on economic
growth.
Population growth  is the percentage change
of population on yearly basis, sourced from WEO.
This variable is used on the model on behalf of the
theoretical reflection of a Solow - standard neo-
classical growth model, in the steady state. It is
expected to reduce income per capita and therefore
reduce growth via second round impact, due to the
increase of the likelihood of an economy to use
scare savings and resources. In a rapidly growing
population, it becomes costly to satisfy public needs
through extending of services [29]. On the other
hand, population growth is regarded as a growth of
labor force and production process, which for the
WB countries is an important input of growth
prospects. Therefore, the expected impact of
population growth on economic growth is
ambiguous. The null hypothesis is that population
growth has no impact on economic growth; i.e.
󰇛  󰇜 and the alternative hypothesis is
that population growth impacts economic
growth󰇛  󰇜. With respect to the
direction of the impact of population growth and
economic growth, we test the hypothesis of a
growth enhancement effect of population growth.
Dummy variables; denotes the regional dummy,
where , stands for south group of the WB
countries
13
and captures the benchmark
category of north group of the WB countries
14
.
denotes the dummy variable capturing the outlier
effect of the pandemic COVID-19, where ,
stands for the COVID-19 pandemic years of 2020,
2021 and 2022, and captures the benchmark
category of the normal years without pandemic. D3
denotes the dummy variable capturing the outlier
effect of the financial crisis, where, D3 = 1 stands for
the financial crisis year of 2008 and D3 = 0 captures
the benchmark category of the normal years without
financial crisis. The interaction term between the
variable of interest, external debt and the dummy
variables, and are included in the model
to estimate the difference in the effects of external
debt on economic growth between two groups of
13
Albania, Kosovo and North Macedonia.
14
Bosnia and Herzegovina, Montenegro and Serbia.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1310
Volume 19, 2022
countries
15
, two periods (COVID and non-COVID
period) and two periods (financial crisis and non-
financial crisis), respectively. With respect to the
relationship between interaction terms and
economic growth, the hypothesis is that economic
growth may to a certain extent, be independent of
the country-specific determinants and therefore be
related to the geographical region of the WB
countries, which have been plagued by political
instability in the near past and to both crisis,
pandemic COVID-19 and financial crisis.
Therefore, the specified countries in the study
(Albania, Kosovo and North Macedonia) may be
more likely to finance their growth potentials
relying on external resources, once, outlining the
insufficiencies of the domestic capital that these
countries possessed during the early transition
period. In addition, a negative effect of the
pandemic COVID-19 and financial crisis on
economic growth in the addressed countries
(Albania, Kosovo and North Macedonia) is
expected due to the reasons that these countries
were more likely to finance the consequences of
these two crisis with debt from International
Financial Institutions.
5 Results and Discussion of the
Results
In this section, we present the estimated coefficients
of the augmented growth model using Fixed Effects
with Driscoll and Kraay standard errors with year
and country dummies (column 1) and robust LSDV
estimate (column 2-4). Among LSDV estimates, to
interpret the results we consider robust LSDV
estimates, accounting for time and country dummies
(column 3 and 4). Moreover, the LSDV estimates
with time and country dummies fit the data much
better, with an R-square of 68.3 per cent and 70.5
percent, respectively. We discuss the economic
interpretation of the models summarized in table 3,
bearing in mind that significant coefficients from
the FEDK estimates and LSDV estimates,
accounting for country and year dummies will be
interpreted and discussed. These specifications are
robust to heteroscedasticity and serial correlation.
To distinguish the effect of external debt on
economic growth, with respect to regional
differentials of the WB countries, the pandemic
COVID-19 and the financial crisis, we have
15
WB countries that belong to the southern part of the
WB region and WB countries that belong to the northern
part of the WB region.
included the interaction terms between external debt
and dummy variable
16
, external debt and
dummy variable
17
and external debt and dummy
variable
18
, respectively. By these interactions, we
test the hypothesis that the effect of external debt on
the economic growth of the WB countries is
different among the WB countries, based on
regional differentials, pandemic differentials and
financial crisis differentials, respectively.
Table 3. Results from static panel estimation
techniques
Dep variable
(1)
(2)
(3)
(4)

FEDK
LSDV
LSDV
LSDV

0.035**
0.049***
0.022**
0.035***
[2.93]
[2.75]
[2.05]
[2.74]

-0.022**
-0.063***
-0.021
-0.022*
[-2.74]
[-3.26]
[-0.10]
[-1.94]

0.094*
0.041
0.134***
0.094***
[2.07]
[0.68]
[3.99]
[2.63]

0.020
0.119**
-0.025
0.020
[0.67]
[2.61]
[-1.49]
[0.79]

0.004
-0.084
-0.048
0.004
[0.10]
[-1.11]
[-1.38]
[0.07]

0.326
0.209
0.328**
0.326**
[1.77]
[1.56]
[2.09]
[2.11]

-0.040
-0.125***
-0.031**
-0.040*
[-1.79]
[-4.26]
[-2.01]
[-1.76]

-0.076***
-0.035*
-0.079***
-0.076***
[-7.73]
[-1.68]
[-4.44]
[-4.57]

0.055**
0.088**
0.052
0.055
[3.84]
[2.14]
[0.87]
[1.06]
2.211
0.777
-1.056
16
Since the interaction indicates that the effect of external
debt on economic growth is different for two different
values of regional dummy, the unique effect of external
debt is not limited to B1, but also depends on the values of
dummy variable, which captures the regional
indicator. stands for countries belonging to the
southern part of the WB region (Albania, Kosovo and
North Macedonia) and captures the benchmark
category of the northern part of the WB region (Bosnia
and Herzegovina, Montenegro and Serbia).
17
In the same way, here, the presence of a significant
interaction indicates that the effect of external debt on
economic growth is different for two different values of
dummy ( =1 for the years of 2020, 2021 and 2022,
0=otherwise; capturing the benchmark category of the
years within the period of the sample, without COVID.
18
In the same way, here, the presence of a significant
interaction indicates that the effect of external debt on
economic growth is different for two different values of
dummy (=1 for the year of 2008, 0=otherwise;
capturing the benchmark category of the years within the
period of the sample, without financial crisis.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1311
Volume 19, 2022
[0.88]
[0.80]
[-0.59]
0.054
1.132
-0.286
0.054
[0.05]
[1.06]
[-0.33]
[0.06]
-0.255
-2.846
-0.255
[-0.36]
[-1.27]
[-0.09]
Year
dummies
Yes
No
Yes
Yes
Country
dummies
Yes
Yes
No
Yes
Constant
0.210
4.379
4.409*
0.272
[0.06]
[0.86]
[1.90]
[0.08]
Observations
132
132
132
132
R-squared
0.697
0.357
0.683
0.705
Number of
groups
6
Note: Note: ***, ** and *, indicate rejection of the
null hypothesis for B coefficients at the significance
level of 1%, 5% and 10%, respectively. Model 1 is
the FEDK estimates with country and year
dummies. Models 2 4 are the robust LSDV
estimates.
Focusing on the results of the robust LSDV
estimates accounting for year dummies (column 4),
the estimated coefficient of external debt for the WB
countries that belong in the southern region,
(WBSR, hereafter), in the equation of growth is
󰇛  󰇜 percent. For the WB
countries that belong to the northern region
(WBNR, hereafter), namely Bosnia and
Herzegovina, Montenegro and Serbia, the
coefficient of external debt is 󰇛
 󰇜 percent. The difference of 
percentage point lower for WBSR countries is
statistically significant at 10 percent level of
significance (column 4). Thus, we conclude that
there is sufficient evidence against the hypothesis
that the size of economic growth between two
groups of countries, does not vary with respect to
the level of external debt. These results indicate that
a considerable  percent increase in the level of
external debt in the WBSR (WBNR) countries,
decreases (increases) economic growth by 
(󰇜 percent, ceteris paribus. Positive (negative)
association of the external debt with economic
growth for the case of the WBNR (WBSR)
countries confirm the growth enhancement effect
(growth deteriorating effect) of external debt,
although in terms of magnitude the size of the
economic impact is very small in both cases. The
positive relationship between external debt and
economic growth for the sample of WBNR
countries, namely Bosnia and Herzegovina,
Montenegro and Serbia, in general, is an indication
that WBNR countries, due to insufficient level of
domestic capital, at early stages of development rely
pretty much on a foreign resources to finance the
domestic productive activities, mainly via external
debt. The negative relationship between external
debt and economic growth for countries like
Albania, Kosovo and North Macedonia can be
explained by the crowding out effect of the debt on
private investments, thus, diminishing growth
prospects in these countries, in the long run. The
same estimated elasticities of the coefficients of
external debt for the WBSR and WBNR countries
are confirmed in other estimates also, including both
FEDK estimates (column 1) and GMM estimates
(see table 4, columns 1-2). The growth enhancement
effect of external debt favours the Keynesian
approach of fiscal policy by endorsing the
governmental intervention, by using external debt as
a valid resource to increase the public and private
investments [15].
The interaction term of external debt with COVID
dummy is statistically significant at 1 percent level
of significance. Focusing on the LSDV estimates
(column 4), regarding this interaction term, the
estimated coefficient of external debt for the
COVID years is 󰇛  󰇜
percent, whereas, for the normal years, without
COVID, it is󰇛  󰇜 percent.
The statistically significant difference of
percentage points in favor of normal years,
without COVID-19, means that the size of economic
growth between two periods vary with respect to the
level of external debt. Hence, 10 percent increase in
the level of external debt, decreases (increases) the
economic growth in the COVID period (normal
period), on average by  and 0.34 percent,
respectively, ceteris paribus. The explanation of the
growth deteriorating effect of external debt in the
pandemic COVID-19 years can be attributed to the
fact that public spending during the COVID-19
period went through rapid restructuring in all WB
countries. Public spending’s were mainly focused
on unproductive activities, like maintaining the
service sector with government subsidies to save
jobs and liquidity in the private sector and not on
productive sectors of the economy where the value
added activity is generated. These government
subsidies during the COVID-19 years were financed
mostly from external resources via external debt,
although the subsidies from the World Bank were
not missing (World Bank, 2021). On the other hand,
the growth enhancement effect of the external debt
in the normal years (without COVID-19), is a signal
that WB countries rely their development level on
external resources, due to insufficient domestic
capital to finance growth. Contrary to expectations,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1312
Volume 19, 2022
the results confirm a growth enhancement effect of
external debt during the financial crisis period,
although its impact on economic sense is confirmed
to be very small. The estimated elasticity of growth
enhancement effect of external debt during the crisis
period is almost 0.09 percent,󰇛
 󰇜 whereas, during the normal period the
estimated elasticity of growth enhancement effect of
external debt is󰇛  󰇜. The
statistically significant difference of 0.055
percentage points in favor of financial crisis year
means that the size of economic growth between
two periods, vary with respect to the level of
external debt. Hence, based on these results, a
considerable increase of external debt during the
financial crisis period (normal period), by 10
percent, increases economic growth of the WB
countries by average 1.0 (0.3) percent, respectively,
ceteris paribus. The explanation behind the scope of
this result can be attributed to the fact that short
recession should not affect the pace of growth,
leaving potential growth unharmed in the longer run
[19].
The coefficient of investment is statistically
significant at one percent level of significance, at
the FEDK (column 1) and LSDV (column 4)
specifications which confirm that investments are
positively associated to growth level, as expected,
although in magnitude the size of the investment
coefficient is relatively small. The growth
enhancement element of investments may be
originated from the public investments that the
region of WB countries have undertaken in the late
years, to accomplish the target of converging their
economies with the EU standards [16]. These public
investments in the form of government investments
manifested mainly on service sector, contributed to
the efficiency of public goods, like education and
physical infrastructure. In addition, private
investments may lay behind the scope of growth
enhancement effect of total investments, due to the
increase of productivity spillover effect [4]. Based
on the estimated coefficient of the total investments
as a share of GDP, a considerable increase of the
investments, say by 10 percent increases growth
level by average 1 percent, ceterus paribus (LSDV
and FEDK estimates).
The coefficient of the population growth is
statistically significant and positively associated to
growth level in the LSDV specification, accounting
for year and country dummies (column 4). Hence,
10 percent increase on population growth, increases
growth level of the WB countries by average 3.2
percent, holding other variables constant. In
addition, the coefficient of inflation is statistically
significant at 5 percent level of significance in the
LSDV estimates, accounting for year and country
dummies, laying on a negative relationship with
growth, although in magnitude its size is very small.
Therefore, average growth-deteriorating effect of
inflation differential per 10 per cent change is just
0.22 per cent, ceterris paribus. We also present the
results from the dynamic panel models: General
Method of Moments (GMM), using Arrellano-Bond
(using xtabond stata command) on column (1),
Arrellano-Bover/Blundell/Bond estimation
procedure (using xtpdsys stata command), shown on
column (2).
Table 4. Results from dynamic panel estimation
techniques
Dep variable
1
2

GMM
GMM

-0.272***
-0.188***
[-3.12]
[-2.61]

0.044*
0.052***
[1.86]
[2.63]

-0.085***
-0.069***
[-3.97]
[-4.83]

0.042
0.086
[0.45]
[1.06]

0.247***
0.168***
[5.48]
[4.55]

0.129
0.088
[0.97]
[0.92]

0.245
0.224
[0.82]
[0.78]

-0.223***
-0.213***
[-4.15]
[-4.89]

-0.046***
-0.041***
[-4.95]
[-4.89]

0.071
0.108
[1.06]
[1.59]
3.820
[1.56]
2.007**
2.574***
[2.29]
[3.22]
-3.129
-4.703
[-1.01]
[-1.57]
Year dummies
No
No
Country dummies
No
No
Constant
-10.921
[-1.50]
Observations
132
132
R-squared
Number of groups
6
6
Wald test
102.87
102.87
p-value
0.000
0.000
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1313
Volume 19, 2022
Note: ***, ** and *, indicate rejection of the null
hypothesis for B coefficients at the significance
level of 1%, 5% and 10%, respectively. Model 1
shows one-step results from system GMM using
Arrellano Bond dynamic panel estimation
technique. Model 2 shows one-step results from the
system GMM using Bover/Blundell Bond
estimation technique, using stata command xtpdys.
Z-statistics in brackets, ***, ** and * indicate
significance of coefficients at 1, 5 and 10 per cent,
respectively. For the GMM results (column 1-2),
internal instruments are used for endogenous
variables (population growth, trade openness and
total investments). Lag limits are ½ for lagged
dependent variable and 2/3 for endogenous
components.
The coefficient of COVID dummy is economically
large and statistically significant in GMM estimates.
This coefficient measures the external debt
differentials on economic growth, between WB
countries with respect to the pandemic COVID-19,
assuming that the interacted regressors are zero.
Since the interacted regressors of the external debt
are continuous variables, it is unlikely they are equal
to zero. The positive coefficient of COVID
dummy, in both system GMM estimates shows
that the level of growth enhancement of external
debt during the pandemic COVID-19 period, is
higher in magnitude in comparison to the normal
period, whereas, the coefficient of regional
dummy, and financial crisis dummy, are
insignificant in all estimates.
The lagged dependent variable, captured by the
initial growth level of the WB countries is
statistically significant, laying on a negative
relationship with current growth, confirming that the
persistence effects subject the current growth level
of the WB countries. One of the crucial features of
the neoclassical growth model is the prediction of a
low coefficient of elasticity of the initial growth
(less than one), which predicts conditional
convergence. The negative sign of the initial growth
level, referring to neoclassical theory, means that
holding other variables constant, the WB economies
are tending to not approach to their long run
position at the rate indicated by the magnitude of the
coefficient [34]. The variable of trade openness
results significant in the dynamic specifications. A
considerable increase of trade openness, say, by 
percent, increases economic growth of the WB
countries by 2.4 percent, on average, ceteris paribus.
The explanatory variables of schooling resulted
insignificant in all specifications. The fact that this
insignificant regressor reported in both
specifications, static (FEDK and LSDV) and
dynamic panel models (GMM) estimates, suggest
that the expounding power of the lagged dependent
variable in the dynamic specification is being
originally ascribed to other variables in the static
specifications.
6 Conclusion
The research has identified some of the
determinants of economic growth in the Western
Balkan countries. Using different estimation
methodologies from the static and dynamic panel
models, we focused the research mainly on the
importance of external debt which largely explain
the size of the growth level in the WB countries,
once having regard the insufficiency of the domestic
capital that these countries possessed during the
long transition period to finance the initial
development stages. The results of the paper proved
the hypothesis that external debt alongside with
other control variables, like: total investments as a
share of GDP, population growth and inflation
differentials measured by the average consumer
price index, have significant impact on growth level
in both group of the WB countries. The growth
enhancement effect of external debt is confirmed for
northern countries of the WB region, like Bosnia
and Herzegovina, Montenegro and Serbia, whereas,
the growth deteriorating effect of external debt is
confirmed for southern countries of the WB region,
like Albania, Kosovo and North Macedonia. From
the control variables, investments and population
growth are found to be growth-enhancing factors,
although in terms of economic impact their effect is
relatively small, whereas inflation rate is confirmed
as a growth-deteriorating factor for the WB
countries. The findings of the paper also confirm a
growth deteriorating impact of COVID-19, due to
the increase of emergency spending on unproductive
sectors in order to save jobs and liquidity in the
private sector and growth enhancing impact of
financial crisis, which is mainly a result of short run
expansionary fiscal policies.
The economic importance of the findings of this
research paper are on providing an analytical
foundation for the evaluation of the economic
policies of the WB countries aimed at increasing
growth level in the region. The paper contributes to
the literature review in the field of the nexus model
between growth and external debt as well as on the
determinants of growth level in the transition
countries, especially Western Balkan countries,
relying on different methodologies. The limitations
of the study are pertaining to the institutional control
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1314
Volume 19, 2022
variables, which can be regarded as growth
enhancement factors, considering the fact that
institutional performance of the WB region is
subjected target by the governments, for being
considered as a crucial force for leading the growth
performance of the WB region. Due to these
conditions, a permanent institutional approach is
needed in Public Financial Management through a
fiscal risk assessment instrument, where all fiscal
parameters will be recorded, analysed and managed.
Therefore, the need for intervention and medium-
system is highly recommended. The intervention
measures may include designing an effective system
for managing public finances in difficult times, with
various actions from actively monitoring
macroeconomic developments, establishing a
regular fiscal risk assessment, considering the
possibilities for flexibility and fiscal space and
managing the public investments that will positively
affect the acceleration of economic recovery. The
mixed evidence with respect to the impact of
external debt on economic growth at the WB
countries implies that both group of WB countries,
WBNR and WBSR countries, shall focus on their
macroeconomic performance, at country level, in
order to have a better view with respect to using
debt component for stimulating growth.
References:
[1] Abdelaziz, H., Rim, B.O., & Majdi, K.,
External debt, investment, and economic
growth. Journal of Economic Integration, Vol.
34, No. 4, 2019, pp. 725-745.
[2] Ahlborn, M., & Schweickert, R., Public debt
and economic growtheconomic systems
matter. International Economics and Economic
Policy, Vol. 15, No.2, 2018, pp. 373-403.
[3] Arellano, M., & Bover, O., Another look at the
instrumental variable estimation of error-
components models, Journal of econometrics,
Vol. 68, No.1, 1995, pp. 29-51.
[4] Barro, R. J., Human capital and economic
growth. In Policies for long-run economic
growth, a symposium sponsored by the Federal
Reserve Bank of Kansas City, Jackson Hole,
WY. Kansas City, MO: Federal Reserve Bank
of Kansas C, Wyoming, 1992, pp. 99-216.
[5] Blundell, R., & Bond, S., Initial conditions and
moment restrictions in dynamic panel data
models. Journal of econometrics, Vol.87, No.1,
1998, pp. 115-143.
[6] Bordo, M. D., Meissner, C. M., & Stuckler, D.,
Foreign currency debt, financial crises and
economic growth: A long-run view. Journal of
international Money and Finance, Vol. 29, No.
4, 2010, pp. 642-665.
[7] Bowsher, C. G., On testing overidentifying
restrictions in dynamic panel data
models. Economics letters, Vol. 77, No. 2,
2002, pp. 211-220.
[8] Bun, M. J., & Kiviet, J. F., On the diminishing
returns of higher-order terms in asymptotic
expansions of bias. Economics Letters, Vol. 79,
No. 2, 2003, pp. 145-152.
[9] Chowdhury, K., A structural analysis of
external debt and economic growth: some
evidence from selected countries in Asia and
the Pacific, Applied Economics, Vol. 26, No.
12, 1994, pp.1121-1131.
[10] Chowdhury, A. R., External debt, growth and
the HIPC initiative: is the country choice too
narrow? In Debt relief for poor countries,
Palgrave Macmillan, London. 2004, pp. 158-
180.
[11] Cordella, T., Ricci, L. A., & Ruiz-Arranz, M,
Debt Overhang Or Debt Irrelevance?:
Revisiting the Debt-growth Link. IMF Working
Paper, WP/05/223, 2005.
[12] Ciftcioglu, S., & Begovic., The relationship
between economic growth and selected
macroeconomic indicators in a group of Central
and East European countries: a panel data
approach. Problems and Perspectives in
Management, Vol. 6, No. 3, 2008, pp. 24-30.
[13] Clements, B., Bhattacharya, R., & Nguyen, T.
Q., External debt, public investment, and
growth in low-income countries. International
Monetary Fund. No 3-249, 2003.
[14] Dauti, B., Determinants of foreign direct
investment in transition economies, with
special reference to Macedonia. Evidence from
gravity model. The South East European
Journal of Economics and Business, Vol.10,
No. 2, 2015, pp. 7-28.
[15] Dauti, B., & Elezi, Sh., Economic growth in
the Central East European Union and the
Western Balkan countries in the course of
Stability and Growth Pact and COVID-19,
Zbornik radova Ekonomskog fakulteta u Rijeci:
časopis za ekonomsku teoriju i praksu, Vol.40,
No.1, pp.29-61.
[16] Dauti, B., Fiscal Policy of the European Union,
in Qorraj, G. & Hashi, I. (eds), European
Union and the Western Balkans, 2021, ISBN /
978-83-65020-33-8/159-178.
[17] Dinca, G., & Dinca, M.S., Public debt and
economic growth in the EU post-communist
countries, Romanian Journal of Economic
Forecasting, Vol.18, No.2, 2015, pp. 119-132.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1315
Volume 19, 2022
[18] Eaton, J., Sovereign debt: A primer, The World
Bank Economic Review, Vol. 7, No. 2, 1993,
pp. 137-172.
[19] European Commission., Impact of the Current
Economic and Financial Crisis on Potential
Output, Occasional Papers No. 49, 2009.
[20] Falvey, R., Foster, N., & Greenaway, D., Trade
liberalization, economic crises, and
growth. World Development, Vol. 40, No. 11,
2012, pp.2177-2193.
[21] Gillman, M., & Nakov, A., Granger causality
of the inflationgrowth mirror in accession
countries, Economics of Transition, Vol.12,
No.4, 2004, pp. 653-681.
[22] Greene, W. H., Econometric Analysis,Upper
Saddle River, N.J.: Prentice Hall, 2013.
[23] Hoechle, D., Robust standard errors for panel
regressions with cross-sectional dependence,
The stata journal, Vol. 7, No. 3, 2007, pp. 281-
312.
[24] Hwang, J. T., Chung, C. P., & Wang, C. H.,
Debt overhang, financial sector development
and economic growth, Hitotsubashi Journal of
Economics, Vol. 51, No.1, 2010, pp. 13-30.
[25] Iqbal, Nasir, & Saima Nawaz., Investment,
inflation and economic growth nexus, The
Pakistan Development Review, Vol.48, No.4,
2009,pp. 863-874.
[26] Karagol, E., The causality analysis of external
debt service and GNP: The case of Turkey,
Central Bank Review, Vol. 2, No. 1, 2012, pp.
39-64.
[27] Khan, M. S., & Ssnhadji, A. S., Threshold
effects in the relationship between inflation and
growth, IMF Staff papers, Vol. 48, No.1, 2001,
pp. 1-21.
[28] Kharusi, S. A., & Ada, M. S., External debt and
economic growth: The case of emerging
economy. Journal of economic
integration, Vol. 33, No. 1, 2018, pp. 1141-
1157.
[29] Klasen, S., & Lawson, D., The impact of
population growth on economic growth and
poverty reduction in Uganda, No. 133.
Diskussionsbeiträge, 2007.
[30] Krugman, P., 1988, Financing vs. forgiving a
debt overhang, Journal of development
Economics, Vol. 29, No. 3, 1988, 253-268.
[31] Lee, J. W., Capital goods imports and long-run
growth, Journal of development
economics, Vol. 48, No. 1, 1995, pp. 91-110.
[32] Levin, A., Lin, C. F., & Chu, C. S. J., Unit root
tests in panel data: asymptotic and finite-
sample properties, Journal of
econometrics, Vol. 108, No. 1, 2002, pp. 1-24.
[33] Mallik, G., & Chowdhury, A., Inflation and
economic growth: evidence from four south
Asian countries, Asia-Pacific Development
Journal, Vol. 8, No. 1, 2001, pp. 123-135.
[34] Moral-Benito, E., Panel growth regressions
with general predetermined variables:
likelihood-based estimation and Bayesian
averaging, CEMFI WP No. 1006, 2010.
[35] Ndoricimpa, A., Threshold effects of debt on
economic growth in Africa, African
Development Review, Vol. 29, No. 3, 2017, pp.
471-484.
[36] Owusu‐Nantwi, V., & Erickson, C., Public debt
and economic growth in Ghana. African
Development Review, Vol. 28, No. 1, pp. 116-
126.
[37] Pattillo, C. A., Poirson, H., & Ricci, L. A.,
External debt and growth, Available at SSRN
879569, 2002.
[38] Qayyum, A., Trade liberalization, financial
development and economic growth. Pakistan
Institute of Development Economics, Working
Paper, 2007, 19, 2007, Islamabad.
[39] Roodman, D., How to do xtabond2: An
introduction to difference and system GMM in
Stata, The stata journal, Vol. 9, No. 1, 2009a,
pp. 86-136.
[40] Roodman, D., A note on the theme of too many
instruments. Oxford Bulletin of Economics and
statistics, Vol. 71, No. 1, 2009b, pp. 135-158.
[41] Sánchez-Juárez, I., & García-Almada, R.,
Public debt, public investment and economic
growth in Mexico, International Journal of
Financial Studies, Vol. 4, No. 2, 2016, pp. 1-
14.
[42] Shittu, W. O., Hassan, S., & Nawaz, M. A.,
The nexus between external debt, corruption
and economic growth: evidence from five SSA
countries, African Journal of Economic and
Management Studies, Vol. 9, No. 3, 2018, pp.
319-334.
[43] Tho, L. P. T. D., & Trường, N. X., The Impact
of External Debt to Economic Growth in Viet
Nam: Linear and Nonlinear Approaches.
In International Econometric Conference of
Vietnam, Springer, Cham, 2019, pp. 952-967.
[44] Vu, Y. H., Nguyen, N. T., Nguyen, T. T., &
Pham, A. T., The threshold effect of
government’s external debt on economic
growth in emerging countries. In International
Econometric Conference of Vietnam, Springer,
Cham, 2019, pp. 440-451.
[45] World Bank., Western Balkans Regular
Economic Report, The Economic and Social
Impact of COVID-19. World Bank,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1316
Volume 19, 2022
Washington, DC. World Bank.
https://openknowledge.worldbank.org/handle/1
0986/33670 License: CC BY 3.0 IGO, No. 17,
2020.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
-Bardhyl Dauti carried out the conceptualization of
the study and was responsible for the econometric
assessment, methodology development and design
of the hypothesis and estimations.
-Ismet Voka was responsible for the investigation
process, methodology development, execution of
literature review part and the conclusion part.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
The authors of the paper funded the study.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2022.19.117
Bardhyl Dauti, Ismet Voka
E-ISSN: 2224-2899
1317
Volume 19, 2022