13021 FACTA UNIVERSITATIS Series: Economics and Organization Vol. 21, No 4, 2024, pp. 257 - 270 https://doi.org/10.22190/FUEO240918017H © 2024 by University of Niš, Serbia | Creative Commons Licence: CC BY-NC-ND Original Scientific Paper THE IMPACT OF PUBLIC EXPENDITURES ON THE ECONOMIC GROWTH OF ALBANIA1 UDC 336.1/.5:338.1(496.5) Marsida Harremi "Fan S. Noli" University, Economy Faculty, Korçë, Albania ORCID iD: Marsida Harremi https://orcid.org/0009-0008-2042-072X Abstract. The main purpose of this paper is to analyze public expenditures and their impact on economic growth in Albania. It is widely recognized that an increase in public expenditures translates into an increase in GDP level. The analysis of the impact of public expenditures is associated with elements that affect economic growth both positively and negatively. Therefore, this is a topic that requires continuous study, not only for governance but also to understand the impact they have on each individual and the economy as a whole. Albania is a small country with an open economy, so the study of the impact of public expenditures on the economy is very important to understand their use as an instrument of fiscal policy and to predict trends in the future. In the conditions of change and reformation of fiscal policies, the structure of government expenditures will likely also change. To study the level of expenditures helps to understand in which functions the government has mostly directed the revenue it has received from different sources. We also highlight which government functions are well covered by spending and which are at low levels and require more attention. This paper takes into study health public expenditures, defense public expenditures, education public expenditures and total public expenditures. These variables are analyzed based on the econometric model. These variables have the highest impact on the level of GDP. Keywords: public expenditures, GDP, Albania. JEL Classification: H51, H56, E62 1. INTRODUCTION Fiscal policy has an important impact on the growth of a country’s economy. Changing the level of public expenditures is one of the instruments of this policy. At a theoretical level, an increase in public expenditure could have positive, negative or no effects on growth. Received September 18, 2024 / Revised November 11, 2024 / Accepted November 14, 2024 Corresponding author: Marsida Harremi "Fan S. Noli" University, Economy Faculty, Bulevardi Rilindasit 11 Korçë, Albania | E-mail: mharremi@yahoo.com https://orcid.org/0009-0008-2042-072X mailto:mharremi@yahoo.com 258 M. HARREMI a. The purpose of the paper The purpose of this paper is to study and analyze the distribution of public expenditures and the impact of each component on economic growth. Albania is pursuing fiscal consolidation policies, which may lead to changes in the structure of government expenditures. Often, restrictive fiscal policies mean a reduction in public expenditure. In these cases, the government must make decisions to limit and reduce consumption and public investment. But which components of public expenditure should be cut? The answer depends on the contribution that each component of expenditures has to economic growth and this contribution varies from one country to another. This study will be based on the analysis of the public expenditures of Albania for the period 2000-2022. Through an econometric analysis, we will reach conclusions on the impact they have on the economy. b. Objectives 1. To analyze the theories and models developed by different researchers for public spending and the impact they have on economic growth. 2. To determine the relationship between public expenditures and economic growth through the econometric model. 3. To analyze the impact these variables have on the study carried out based on the relevant tests. 4. To reach a conclusion and recommendations for effective public expenditures in the country. 2. LITERATURE REVIEW According to Ansari (1993), the relationship between public expenditures and economic growth has been addressed in two different areas of economic policies: the public finance literature and the macroeconomic models literature (p. 31). The relationship between public expenditure and national revenue has been treated in a characteristically dissimilar manner in two major areas of economic analysis. While public finance studies have generally postulated that growth in public expenditure over time is caused by growth in national revenue, most macroeconomic models have tended to take the opposite view. The divergent views on the causative relation between the two variables, in turn, rest on more basic differences in assumptions (Singh & Sahni, 1984, p. 630). The public sector and production will increase with economic development (Goffman, 1968, p. 59). Government expenditures do not play a significant role in promoting economic growth in the four countries in our study (the Philippines is the exception). This is surprising because it is widely believed that government has played an important role in the development of these countries (Dogan & Tang, 2006, p. 55). The state must manipulate the levels of aggregate demand to avoid insufficient or excessive demand by adjusting the level of expenditures and tax revenues to achieve full employment. The growth of public expenditure in the case of Turkey is not directly dependent on and determined by economic growth, as Wagner’s law states. Of course, public expenditure is the outcome of many decisions in the light of changing economic circumstances (Demirbas, 1999, pp. 18-19). According to Huang, (2006, p. 144), empirically, Wagner’s Law investigates the long-run relations between government size and the economy. These different views result from The Impact of Public Expenditures on the Economic Growth of Albania 259 different assumptions that have been made about the relationship between public expenditures and economic growth. Classical economists think that adjustments in price levels can automatically lead demand to reach the level of full employment, but Keynes argues that the process of self- regulation is impossible without state intervention because it would lead to a decline in employment and also a decline of the national product (Demirbas, 1999). On the one hand, Singh & Sahni (1984), Ram (1986) and Holmes & Hutton (1990) conclude that government expansion through increased public expenditures has a positive effect on economic growth. Public infrastructure, education and health expenditures can in principle be complementary to private activities and therefore have positive effects on GDP. For example, new transport infrastructure saves travel time and therefore, will bring positive effects to private agents. In that regard, Li & Huang (2009) studied the relationship between per capita real GDP growth and physical capital, human capital and health investment in the production function. Panel data models were used in the estimation based on the provincial data from 1978 to 2005. The empirical evidence showed that both health and education have positive significant effects on economic growth. Wang( 2011) studied the international total healthcare expenditure data of 31 countries from 1986 to 2007 to explore the causality between an increase in healthcare expenditure and economic growth. Panel regression analysis and quantile regression analysis were used. The estimation of the panel regression reveals that health expenditure growth will stimulate economic growth; however, economic growth will reduce health expenditure growth. Concerning the estimation of quantile regression, in countries with low levels of growth, health expenditure growth will reduce economic growth. Mehrara & Musai (2011) examined the stationary and co-integration relationship between health expenditure and GDP based on the panel co-integration analysis for a sample of 13 Middle East and North Africa (MENA) countries, using data from 1995 to 2005. The findings indicated that the share of health expenditures to GDP decreases with GDP. This implied that healthcare is not a luxury good in MENA countries. Elmi & Sadeghi (2012) investigated the causality and co-integration relationships between economic growth and healthcare expenditures in developing countries from 1990 to 2009. Their findings indicated that revenue is an important factor across developing countries in the level and growth of healthcare expenditure in the long-run. Additionally, the health-led growth hypothesis in developing countries is confirmed. Taban (2006) investigated the relationship between health and economic growth in Turkey within the context of causality, using data from 1980 to 2000. According to the empirical results, a two-way causality relationship was seen between life expectancy at birth and economic growth, no causal relationship was found between health expenditures and economic growth. Mankiw, Romer, & Weil (1992), found a positive relationship between education and economic growth, by considering an extended Solow growth model. Barro & Lee (1993) investigated that there is a positive relationship between education and economic growth by taking 129 countries as their sample. In contrast to such a positive relationship, some empirical studies explain that education and economic growth are not significantly related. Bils & Klenow (2000) viewed that there might be a positive correlation between education and economic growth, but the relationship between education and economic growth does not necessarily explain the educational influence on economic growth. As 260 M. HARREMI far as their views, both education and economic growth can be affected by the total factor productivity. Karagol & Palaz (2004) study shows that there is a long‐run equilibrium relationship between GNP and defence expenditures. Furthermore, the short-run causality test indicates that there is a unidirectional causality between variables, from defence expenditure to economic growth. To see the effect of a shock, we employed impulse response analyses. The results show that GNP decreased during the period then output finally recovered from the initial shock to defence expenditures. An investigation of the relationship between defence expenditures and economic growth in South Africa by Mosikari & Matlwa( 2014) concludes that there is long run relationship between defence expenditure and economic growth. Also, for causal analysis military expenditure seems to granger cause gross domestic product per capita at a 5 percent significance level. In their study, Yilgör, Karagöl, & Saygili (2014) analyze GDP and defence expenditures of the developed countries with cross-sectional ADF and SURADF unit root tests using annual data for the years 1980–2007. They conclude that in the long term, according to the Pedroni cointegration test, a relationship exists between defence expenditure and economic growth. Furthermore, by utilizing the Granger causality test, we find that defence expenditure is a factor in economic growth. In other words, our study validates the hypothesis that defence expenditures by economically developed countries positively contribute to their economies. 3. METHODOLOGY The type of study is descriptive-correlational, which consists of a dependent variable, which is GDP economic growth, and independent variables, which are Exp Edu, education expenditures, Exp Health, health expenditures, Exp Defence, defence expenditures. It tries to find the relationship between the variables and describe it. Among many influencing factors, we have chosen those because there are typical factors for Albania as a developing country. The scientific research question is: what are some of the components of government expenditures that affect sustainable economic growth? The methodology that followed is detailed in every step with the correct processing of data, the verification of every statistical test, the clear raising of hypotheses and the realization of the analysis of the influencing factors of public expenditure on economic growth. The data sources are secured from the World Bank (2023) and represent the period from 2020-2022 with annual data. The Eviews- 12 program was used for the construction of the econometric model, data analysis and statistical tests. a. Model specification GDP Growth (Y) - dependent variable Exp Edu (X1) - independent variable Exp Health (X2) - independent variable Exp Defence (X3) - independent variable This paper intends to perform the multiple linear regression of Y against X1, X2 and X3, according to the equation: The Impact of Public Expenditures on the Economic Growth of Albania 261 (1) For 22 years this paper estimates the linear multivariable equation for the choice, based on the observed data that we have available, and which are presented as a time series. Thus, through the estimators of the ordinary least squares method, we will find the values of the estimated parameters of the model b0, b1, b2 and b3 and we will interpret them. Let us start with the dependent variable Yt to see if the series is stationary, to make predictions for future periods. Fig. 1 GDP growth of Albania (2000-2022) Source: Own calculation, 2023 i. Figure 1 shows that the values of the series fluctuate towards the average value (4.15) sustainably, with a somewhat more exaggerated fluctuation around the year 2020. An important factor in the economic decline in 2020 is Covid-19. 0 1 2 3 4 5 6 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 Series: Sample 2000 2022 Observations 23 Mean 4.148962 Median 4.019346 Maximum 8.908528 Minimum -3.302082 Std. Dev. 2.716534 Skewness -0.560172 Kurtosis 3.782374 Jarque-Bera 1.789475 Probability 0.408715 Fig. 2 Distribution of normality of yt values Source: Own calculation, 2023 262 M. HARREMI Ho: Distribution of normal values The probability value of the J-B test statistic was greater than 0.05, therefore Ho is true. The distribution value of the Yt slights bias (suppression) from the left, is also shown by the negative value of Skewness. ii. In this paper we use the ADF Test to test the stationarity of the series formally. From Figure 1, it seems that the most likely form of your series is a "Random walk with drift, without trend". Performing the ADF unit root test resulted in Table 1. Table 1 GDP unit root test Null Hypothesis: GDP_GROWTH has a unit root Exogenous: Constant Lag Length: 0 (Automatic - based on SIC, maxlag=4) t-Statistic Prob.* Augmented Dickey-Fuller test statistic -3.611533 0.0141 Test critical values: 1% level -3.769597 5% level -3.004861 10% level -2.642242 *MacKinnon (1996) one-sided p-values. Augmented Dickey-Fuller Test Equation Dependent Variable: D(GDP_GROWTH) Method: Least Squares Date: 12/19/23 Time: 20:15 Sample (adjusted): 2001 2022 Included observations: 22 after adjustments Variable Coefficient Std. Error t-Statistic Prob. GDP GROWTH (-1) -0.764993 0.211820 -3.611533 0.0017 C 3.054207 1.044363 2.924468 0.0084 R-squared 0.394731 Mean dependent var -0.095545 Adjusted R-squared 0.364468 S.D. dependent var 3.380234 S.E. of regression 2.694732 Akaike info criterion 4.906983 Sum squared resid 145.2316 Schwarz criterion 5.006168 Log likelihood -51.97681 Hannan-Quinn criter. 4.930348 F-statistic 13.04317 Durbin-Watson stat 2.112255 Prob(F-statistic) 0.001741 Source: Own calculation, 2023 The probabilistic value of the ADF statistic was lower than 0.05, verifying Ho, which indicates that the series does not have a unit root, so it is stationary. Consequently, based on the observed values of the series, we can predict the future. b. Regress Y on X1, X2 and X3 The results of the analysis are presented in Table 2. The Impact of Public Expenditures on the Economic Growth of Albania 263 Table 2 Least Squares Dependent Variable: GDP_GROWTH Method: Least Squares Date: 12/19/23 Time: 18:54 Sample: 2000 2022 Included observations: 23 Variable Coefficient Std. Error t-Statistic Prob. EXP_EDU 2.198381 2.424048 0.906905 0.3758 EXP__HEALTH -6.805401 1.964927 -3.463437 0.0026 EXP__DEFENCE -5.159503 3.046904 -1.693359 0.1067 C 45.60279 13.59728 3.353818 0.0033 R-squared 0.429822 Mean dependent var 4.148962 Adjusted R-squared 0.339793 S.D. dependent var 2.716534 S.E. of regression 2.207268 Akaike info criterion 4.578159 Sum squared resid 92.56865 Schwarz criterion 4.775637 Log likelihood -48.64883 Hannan-Quinn criter. 4.627824 F-statistic 4.774300 Durbin-Watson stat 2.597844 Prob(F-statistic) 0.012076 Source: Own calculation, 2023 The multiple regression model is statistically significant, but before interpreting its parameters, we must test: Phase One Functional form, Phase Two Multicollinearity, Phase Three The normality of the waste, Phase Four Autocorrelation, Phase Five Heteroskedasticity. Phase One: Functional form Ho: The linear form is suitable Ha: The functional form is not suitable The basic hypothesis is tested with Ramsey's RESET test (table 3) Table 3 The Ramsey RESET Test Ramsey RESET Test Equation: UNTITLED Specification: GDP_GROWTH EXP_EDU EXP_HEALTH EXP_DEFENCE C Omitted Variables: Squares of fitted values Value df Probability t-statistic 0.327957 18 0.7467 F-statistic 0.107556 (1, 18) 0.7467 Likelihood ratio 0.137023 1 0.7113 F-test summary: Sum of Sq. df Mean Squares Test SSR 0.549841 1 0.549841 Restricted SSR 92.56865 19 4.872034 Unrestricted SSR 92.01881 18 5.112156 Unrestricted SSR 92.01881 18 5.112156 264 M. HARREMI LR test summary: Value df Restricted LogL -48.64883 19 Unrestricted LogL -48.58032 18 Unrestricted Test Equation: Dependent Variable: GDP_GROWTH Method: Least Squares Date: 12/19/23 Time: 19:01 Sample: 2000 2022 Included observations: 23 Variable Coefficient Std. Error t-Statistic Prob. EXP_EDU 3.207334 3.953525 0.811259 0.4278 EXP__HEALTH -10.08976 10.21487 -0.987752 0.3364 EXP__DEFENCE -7.836632 8.739370 -0.896704 0.3817 C 67.23231 67.40703 0.997408 0.3318 FITTED^2 -0.058650 0.178835 -0.327957 0.7467 R-squared 0.433208 Mean dependent var 4.148962 Adjusted R-squared 0.307255 S.D. dependent var 2.716534 Source: Own calculation, 2023 The probabilistic value of the test statistic is greater than 0.05. This is evidence for the validity of the basic hypothesis. Therefore, the linear form of the specified model is suitable. Phase two: Multicollinearity test Table 4 Multicollinearity test EXP__DEFENCE EXP__HEALTH EXP_EDU EXP__DEFENCE 1.000000 -0.683595 -0.205613 EXP__HEALTH -0.683595 1.000000 0.502726 EXP_EDU -0.205613 0.502726 1.000000 None of the values of the correlation coefficients between the factors is greater than 0.8. Defense expenditures are inversely proportional to health and education expenditures, even the negative linear relationship between defense expenditures and health expenditures is relatively strong (0.68). Expenditures on health and education have a positive correlation. Also, the VIF (variance inflation factor) values were smaller than 5, an indicator of the absence of multicollinearity. Table 5 Variance inflation factor Variance Inflation Factors Date: 12/19/23 Time: 19:16 Sample: 2000 2022 Included observations: 23 Coefficient Uncentered Centered Variable Variance VIF VIF EXP_EDU 5.876009 315.1755 1.405498 EXP__HEALTH 3.860939 685.2356 2.526904 EXP__DEFENCE 9.283625 87.25526 1.971625 C 184.8860 872.8135 NA Source: Own calculation, 2023 The Impact of Public Expenditures on the Economic Growth of Albania 265 Phase three: Test of normality of residuals In this phase, tests of normality of the residuals are performed through the Jarque-Berra statistic. Ho: The waste distribution is normal Ha: The distribution of waste is not normal Fig. 3 Test of normality of residuals Source: Own calculation, 2023 The probabilistic value (p-value) of the J-B statistic (0.56) was greater than 0.05, and the Ho hypothesis is confirmed. Only a small negative bias is observed. Phase Four: Serial Correlation LM Test Ho: There is no autocorrelation of residuals Ha: There is autocorrelation of residuals Ho testing was done with the Breusch-Godfrey test statistic with the LM approach. The probability value of this statistic was greater than 0.05, Ho is confirmed. Therefore, residuals of the model do not "suffer" from autocorrelation (table 6). Table 6 Serial Correlation LM Test Breusch-Godfrey Serial Correlation LM Test: F-statistic 1.336606 Prob. F(2,17) 0.2890 Obs*R-squared 3.125259 Prob. Chi-Square(2) 0.2096 Test Equation: Dependent Variable: RESID Method: Least Squares Date: 12/19/23 Time: 19:29 Sample: 2000 2022 Included observations: 23 Presample missing value lagged residuals set to zero. 266 M. HARREMI Variable Coefficient Std. Error t-Statistic Prob. EXP_EDU 1.059533 2.480590 0.427129 0.6746 EXP__HEALTH -0.699347 1.979602 -0.353277 0.7282 EXP__DEFENCE 0.037290 2.998062 0.012438 0.9902 C 0.670641 13.37757 0.050132 0.9606 RESID(-1) -0.407315 0.252244 -1.614768 0.1248 RESID(-2) -0.177638 0.255624 -0.694921 0.4965 R-squared 0.135881 Mean dependent var 8.72E-15 Adjusted R-squared -0.118272 S.D. dependent var 2.051260 S.E. of regression 2.169174 Akaike info criterion 4.606028 Sum squared resid 79.99034 Schwarz criterion 4.902244 Log likelihood -46.96932 Hannan-Quinn criter. 4.680525 F-statistic 0.534642 Durbin-Watson stat 1.895733 Prob(F-statistic) 0.747305 Source: Own calculation, 2023 The residuals were not affected by any of the regressors (explanatory variables). Phase Five: Heteroskedasticity Test Ho: The distribution of residuals is homoscedastic Ha: The distribution of residuals is heteroskedastic Table 7 Heteroskedasticity Test Heteroskedasticity Test: Breusch-Pagan-Godfrey F-statistic 0.934858 Prob. F(3,19) 0.4432 Obs*R-squared 2.958334 Prob. Chi-Square(3) 0.3981 Scaled explained SS 2.036801 Prob. Chi-Square(3) 0.5648 Test Equation: Dependent Variable: RESID^2 Method: Least Squares Date: 12/19/23 Time: 19:36 Sample: 2000 2022 Included observations: 23 Variable Coefficient Std. Error t-Statistic Prob. C 7.405639 36.17120 0.204739 0.8400 EXP_EDU -10.18861 6.448404 -1.580021 0.1306 EXP__HEALTH 4.909617 5.227060 0.939269 0.3594 EXP__DEFENCE 0.600766 8.105313 0.074120 0.9417 R-squared 0.128623 Mean dependent var 4.024724 Adjusted R-squared -0.008963 S.D. dependent var 5.845594 S.E. of regression 5.871731 Akaike info criterion 6.534947 Sum squared resid 655.0673 Schwarz criterion 6.732424 Log likelihood -71.15189 Hannan-Quinn criter. 6.584612 F-statistic 0.934858 Durbin-Watson stat 1.511577 Prob(F-statistic) 0.443207 Source: Own calculation, 2023 The Impact of Public Expenditures on the Economic Growth of Albania 267 The probability value of the Breusch-Pagan-Godfrey test statistic was greater than 0.05 (Table 7). This indicates the lack of heteroscedasticity of the residuals in the evaluated model. The distribution of residuals was not affected by the regressors. The three independent variables linearly explain about 34% of the total variance of Y (Adjusted R-squared). Multivariable regression model equation: GDP_GROWTH = 2.19838056765*EXP_EDU - 6.80540082074*EXP__HEALTH - 5.15950295442*EXP__DEFENCE + 45.6027915016 Fig. 4 Cusum Source: Own calculation, 2023 4. RESULTS In summary, we can say that in the multivariable regression model specified and evaluated through least squares method estimators, none of its main assumptions were violated. Also, the functional form of the model (linear form) is suitable. Returning once again to the results of Table 2, we see that only one of the three explanatory variables (X2, health expenditures) affects Y in a statistically significant way, for a significance level of alpha=0.05. This influence is negative. The value of the regression coefficient b2 next to this variable shows that, if X2 increases by 1 unit, Y is expected to decrease by 6.4 units, provided that we keep the other two variables under control (unchanged). The value -6.805 measures the estimated change in average GDP as a result of a 1 percent change in health expenditures. From this model, we understand that if health expenditures increase by 1 percent, then GDP decreases by 6.805 percent. The negative sign before the coefficient indicates an inverse relationship between these variables. As Mehrara & Musai (2011) conclude, the share of health expenditures to GDP decreases with GDP. This implied that healthcare is not a luxury good in MENA countries. The value 2.198 measures the estimated change in average GDP as a result of a one percentage unit change in education expenditures. From this model, we understand that if the expenditures made by the government in education increase by 1 percent, then the GDP increases by 2.198 percent. The positive sign indicates a direct relationship between these two variables. As agreed by Mankiw, Romer, & Weil (1992), Barro & Lee (1993) 268 M. HARREMI and Bils & Kleno, (2000), both education and economic growth can be affected by the total factor productivity At the equation of regression, we can consider the defence expenditures as statistically significant (with 10% significance) as the p-value is approximately 0.10. The value -5.159 measures the estimated average change in GDP as a result of a 1 percent change in defense expenditures. From this model, we understand that if the public expenditures on social protection and public order increase by 1 percent, then GDP decreases by 5.159 percent. The value is negative indicating an inverse relationship between them. 5. CONCLUSION Health expenditure has a negative impact on economic growth, but indirectly it has a positive impact on the growth of health care and the increase of the well-being of the individual. Defense expenditure also has a negative impact on economic growth, but we can consider significant, so the use of defence resources should be done in a balanced way between the three main categories of expenses (personnel, operations and maintenance as well as investments for modernization and infrastructure), to ensure harmonious and integrated development of all areas of defence. The government should channel expenditures towards the most productive sectors of the economy, targeting projects that increase the level of health and defence services. It is important to first ensure universal access in both sectors. In this way, the cost of business development can be reduced, the standard of living for the country's poor can be increased and economic growth can be promoted in the long term. REFERENCES Ansari, M. I. (1993, July). Testing the Relationship Between Government Expenditure and National Income in Canada, Employing Granger Causality and Cointegration Analysis. Managerial Finance, 19(7), 31-46. https://doi.org/10.1108/eb013733 Barro, R. J., & Lee, J.-W. (1993, December). International comparison of educational attainment. Journal of Monetary Economics, 32(3), 363-394. https://doi.org/10.1016/0304-3932(93)90023-9 Bils, M., & Klenow, P. J. (2000, December). Does Schooling Cause Growth?. American Economic Association, 90(5), 1160-1183. https://doi.org/10.1257/aer.90.5.1160 Demirbas, S. (1999, May). Co-integration Analysis-causality Testing and Wagner’s Law: The Case of Turkey, 1950-1990. University of Leincester Discussion Papers, 1-27. Dogan, E., & Tang, T. C. (2006, October). Government Expenditure And National Income: Causality Tests For Five South East Asian Countries. International Business & Economics Research Journa, 5(10), 49-58. Elmi, Z. M., & Sadeghi, S. (2012). Health care expenditures and economic growth in developing countries: panel co-integration and causality. Middle-East Journal of Scientific Research, 12(1), 88-91. Goffman, I. J. (1968). Causality Between Public Expenditure and National Income Author(s): Balvir Singh and Balbir S. Sahni Source: The Review of Economics and Statistics, Nov., 1984, Vol. 66, No. 4 (Nov., 1984), pp. 630-644 Published by: The MIT Press. Holmes, J. M., & Hutton, P. A. (1990, February). On the Casual Relationship between Government Expenditures and National Income. The Review of Economics and Statistics, 72(1), 87-95. Huang, C.-J. (2006, December). Government expenditures in china and taiwan: Do they follow wagner’s law?. Jurnal of Economic Development, 31(2), 139-149. Karagol, E., & Palaz, S. (2004). Does defence expenditure deter economic growth in Turkey? A cointegration analysis. Defence and Peace Economics, 15(3), 289-298. https://doi.org/10.1080/10242690320001608908 Li, H., & Huang, L. (2009, September ). Health, education and economic growth in China: Empirical fidings and impications. China Economic Review, 20(3), 374-387. https://doi.org/10.1016/j.chieco.2008.05.001 https://doi.org/10.1108/eb013733 https://doi.org/10.1016/0304-3932(93)90023-9 https://doi.org/10.1257/aer.90.5.1160 https://doi.org/10.1080/10242690320001608908 https://doi.org/10.1016/j.chieco.2008.05.001 The Impact of Public Expenditures on the Economic Growth of Albania 269 Mankiw , G., Romer, D., & Weil, D. N. (1992). A Contribution to the Empirics of Economic Growth. NBER Working Paper Series. Mehrara, M., & Musai, M. (2011, August). Granger causality between Health and Economic Growth in oil exporting coutries. Interdisciplinary Journal of Research in Busness, 1(8), 103-108. Mosikari, T. J., & Matlwa, K. (2014, September). An Analysis of Defence Expenditure and Economic Growth in South Africa. Mediterranean Journal of Social Sciences, 5(20), 2769-2776. http://dx.doi.org/10.5901/ mjss.2014.v5n20p2769 Ram, R. (1986, Mars). Government Size and Economic Growth: A New Framework and Some Evidence from Cross-Section and Time-Series Data. The American Economic Review, 76(1), 191-203. Singh, B., & Sahni, B. S. (1984, November). Causality Between Expenditure and National Income. The Review of Economics and Statistics, 66(4), 630-644. https://doi.org/10.2307/1935987 Taban, S. (2006). Türkiye’de sağlık ve ekonomik büyüme arasındaki nedensellik ilişkisi [The Causality Relationship between Health and Economic Growth in Turkey]. Sosyo Ekonomi, 4(4), 31-46. Wang, K. M. (2011, July). Health care expenditures and economic growth: Quantile panel-type analysis. Economic Modelling, 28(4), 1536-1549. https://doi.org/10.1016/j.econmod.2011.02.008 World Bank. (2023). Retrieved from https://databank.worldbank.org/reports.aspx?source=2&country=ALB Yilgör, M., Karagöl, E. T., & Saygili, Ç. A. (2014). Panel Causality Analysis Between Defence Expenditure and Economic Growth in Developed Countries. Defence and Peace Economics, 25(2), 193-203. https://doi.org/ 10.1080/10242694.2012.724879 6. APPENDIX Model data http://dx.doi.org/10.5901/%0bmjss.2014.v5n20p2769 http://dx.doi.org/10.5901/%0bmjss.2014.v5n20p2769 https://doi.org/10.2307/1935987 https://doi.org/10.1016/j.econmod.2011.02.008 https://databank.worldbank.org/reports.aspx?source=2&country=ALB https://doi.org/%0b10.1080/10242694.2012.724879 https://doi.org/%0b10.1080/10242694.2012.724879 270 M. HARREMI UTICAJ JAVNIH RASHODA NA EKONOMSKI RAST ALBANIJE Osnovna svrha ovog rada je analiza javnih rashoda i njihovog uticaja na ekonomski rast u Albaniji. Opšte je poznato da povećanje javnih rashoda dovodi do povećanja nivoa BDP-a. Analiza uticaja javnih rashoda povezana je sa elementima koji utiču na ekonomski rast i pozitivno i negativno. Dakle, ovo je tema koja zahteva kontinuirano proučavanje, ne samo za upravljanje, već i za razumevanje uticaja koji oni imaju na svakog pojedinca i privredu u celini. Albanija je mala zemlja sa otvorenom ekonomijom, pa je proučavanje uticaja javnih rashoda na privredu veoma važno za razumevanje njihove upotrebe kao instrumenta fiskalne politike i predviđanje trendova u budućnosti. U uslovima promene i reformisanja fiskalnih politika, verovatno će se promeniti i struktura državnih rashoda. Proučavanje nivoa rashoda pomaže da se razume u koje funkcije je vlada uglavnom usmeravala prihode koje je primila iz različitih izvora. Takođe ističemo koje funkcije vlade su dobro pokrivene potrošnjom, a koje su na niskom nivou i zahtevaju više pažnje. U ovom radu razmatraju se javni rashodi u zdravstvu, javni rashodi za odbranu, javni rashodi za obrazovanje i ukupni javni rashodi. Ove varijable se analiziraju na osnovu ekonometrijskog modela. Ove varijable imaju najveći uticaj na nivo BDP-a. Ključne reči: javni rashodi, BDP, Albanija