JSP | Jurnal Studi Pemerintahan AFFILIATION: Atma Jaya Catholic University of Indonesia CORRESPONDENCE: (rosdiana.sijabat@atmajaya.ac.id) DOI: https://doi.org/10.18196/jgp.111109 CITATION: Sijabat, Rosdiana. (2020). Disentangling the Association Between Government Debt and Economic Growth: A Granger Causality Approach from Indonesia. Jurnal Studi Pemerintahan. 11(1). 1-23. ARTICLE HISTORY: Received: January 16th, 2020 Accepted: January 30th, 2020 DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA Rosdiana Sijabat ABSTRACT: This paper aims to examine the possible Granger-causality relationship between public debt and economic growth in Indonesia between 1998 and 2018. To accomplish this aim, a time series regression approach as well as diagnostic tests such as the Augment Dickey–Fuller test and Johansen cointegration test (which provides evidence of a long-term relationship between external debt and economic growth) were conducted. A VECM Granger causality approach was chosen to investigate the causal link between government and economic growth. The VECM estimation provides new evidence that, over the long term, domestic debt has significantly and positively affected economic growth; at the same time, external debt has significantly and negatively affected economic growth. Meanwhile, Granger-causality analysis shows that economic growth has a unidirectional causal relationship with external public debt, but does not have such a relationship with domestic public debt. For this study, a series of 20 data points per variable were analyzed, covering 1998 through 2018. This sample size is rather small, and as such its findings are not perfect. The use of a much larger data set would enhance any similar studies in the future. Nonetheless, this study illuminates the role of government debt in the economy by highlighting the importance of domestic markets as sources of public debt to promote economic growth in Indonesia, and recommends that the government do so. Keywords: Economic growth; Government debt; Johansen Cointegration; Granger causality. ABSTRAK: Penelitian ini menguji kemungkinan adanya Granger-causality antara hutang pemerintah dan pertumbuhan ekonomi Indonesia dari tahun 1998 hingga 2018. Dengan pendekatan time series regression, diagnostic tests seperti uji Augment Dickey–Fuller dan uji Johansen cointegration, lalu ditetapkan pendekatan VECM Granger causality untuk meneliti hubungan kausalitas hutang pemerintah dan pertumbuhan ekonomi. Hasil uji kointegrasi Johansen memberikan bukti adanya hubungan jangka panjang antara hutang luar negeri (eksternal) dan pertumbuhan ekonomi. Selanjutnya, dari analisis VECM ditemukan bahwa dalam jangka panjang, hutang domestik berpengaruh signifikan dan positif terhadap pertumbuhan ekonomi, namun hutang eksternal berpengaruh negatif dan signifikan terhadap pertumbuhan ekonomi. Berdasarkan analisis Granger-casality, pertumbuhan ekonomi memiliki undirectional causality dengan hutang eksternal pemerintah, namun tidak memiliki hubungan kausalitas dengan hutang domestik pemerintah. Implikasi kebijakan dari studi ini adalah pentingnya memanfaatkan sumber- sumber hutang pemerintah dari pasar domestik dan lebih berhati-hati dalam melakukan hutang pemerintah dari sumber eksternal, jika hutang pemerintahan menjadi bagian dari strategi mendorong pertumbuhan ekonomi. Faktor ketersediaan data menyebabkan data runtut waktu yang dianalisis hanya 20 titik waktu per variabel mulai tahun 1998 hingga 2018, jumlah ini relatif kecil, sehingga temuan dalam studi ini tidak sepenuhnya sempurna. Periode waktu yang lebih panjang untuk penelitian serupa di masa mendatang dapat bermanfaat. 1 mailto:rosdiana.sijabat@atmajaya.ac.id DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 2 INTRODUCTION Theoretical and practical debate on the association between public debt and economic growth is ongoing (Eberhardt & Presbitero, 2015; Bökemeier & Greiner 2015). Public debt is understood as resulting from limited fiscal responsibility within public policy (Pečarić, Slišković, & Kusanović, 2018). At the same time, however, public debt can help promote economic growth and development (Pegkas, 2018; Korkmaz, 2015). This is particularly true in developing nations where economic growth is constrained by limited financial resources, such as Indonesia (Barrett, 2018; Muhdi & Sasaki, 2009). The policy foundation on the structure of government debt in Indonesia is regulated in Law number 17 of 2003 concerning State Finance. Article 12 paragraph 3 stipulates that the amount of debt is limited to a maximum of 60 percent of GDP and a budget deficit that is allowed a maximum of 3 percent of GDP. This law adopts one of the classifications regarding budget deficit limits and debt limits that are used as an international reference, the Maastricht Treaty. According to the Directorate General of Financing and Risk Management (DJPPR), the Ministry of Finance of the Indonesian Government, the government incurred debt because the government made the debt as one source of financing in the context of development and improving the welfare of the people in Indonesia. Debt is such an instrument pursued by the government to meet shortages of government spending in the current fiscal year. Another argument from the Indonesian government that tends to have to do debt is because the ratio of government revenue, especially domestic income, is tax which is still not optimal. For example in 2018, Indonesia's tax ratio1 was only 11.5% of GDP, this ratio is relatively low compared to some neighboring countries such as Malaysia (15%), the Philippines (14.6%), and Singapore (13.6%) (Ministry of Finance of the Indonesian Government, 2018). The government revenue from tax which is still relatively low is partly due to the level of tax compliance2 in Indonesia which only reached an average of 60.1% over the last 10 years (2010- 2018), where the government is aiming to reach the level of tax compliance at 80% every year. Indonesia has maintained a high level of public debt over the past twenty years, maintaining an average debt–GDP ratio of 43.7% between 1998 and 2018. Public debt has grown annually as deficits have occurred in the national budget. As Figure 1 depicts, between 1998 and 2004, Indonesia's public debt was always more than 50% of its GDP. Since 2011, the debt–GDP ratio has remained below 30%, reaching an all-time low (23%) in 2012. 1 Tax ratio is the nation's tax revenue relative to its gross domestic product (GDP) which is used as an indicator to assess the performance of tax revenue. Such ratio indicates the government's ability to collect taxes from the total economy (GDP) (Ministry of Finance of the Republic of Indonesia, www.kemenkeu.go.id, 24/03/2019). 2 Tax compliance refers to the fulfillment of all tax obligations as determined by applicable law which includes tax returns within a specified period, correctly declaring income and deductions, paying taxes assessed at the due date and paying taxes collected (Doran, 2009). http://www.kemenkeu.go.id/ DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 3 350 330,1 300 250 200 150 100 50 0 2011 2012 2013 2014 2015 2016 2017 2018 2019 Figure 1. Total debt-to-GDP Ratios in Indonesia, 1998-2018 (%) Source: illustrated by the author based on data from the Ministry of Finance of Indonesia. Figure 2 shows that, over the past nine years, Indonesia has experienced an average deficit of 228.2 trillion rupiah, with the highest deficit (330.1 trillion rupiah) occurring in 2017. Average: Rp. 228,2 Trillion 296,7 326 297 241,4 224,1 222,5 44,2 72,3 Figure 2. State Budget Deficit (Trillion Rupiah) Source: illustrated by the author based on data from the Ministry of Finance of Indonesia. Owing to the continued government debt and budget deficits in Indonesia, as elaborated above, this paper investigates association between public debt and economic growth. First, it is important to understand whether public debt has had a positive effect on economic growth in Indonesia. Second, to design flexible economic policy, it is paramount to determine whether 28,3 30,0 26,0 23,0 24,9 24,7 27,4 28,3 85 Average: 43.7% 35 29,8 29,2 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 33 39 47 57 58 61 67 77 89 DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 4 maintaining public debt is sound fiscal policy. Third, this study offers a real answer to recent criticism of the government's fiscal policies as well as the rapid increase in public debt these have precipitated. As such, the significance of this study lies in its understanding of the relationship between public debt and economic growth in Indonesia. To do so, this study employs a VECM Granger causality test, with its results providing several novel contributions. First, empirical analysis is conducted by using the Granger causality test to understand two related phenomena: the association between domestic public debt and economic growth and the link between external public debt and economic growth. Second, analysis was conducted utilizing the most recent available data on public debt and economic growth, covering the period between 1998 and 2018. Third, it is hoped that this analysis of the Indonesian government's fiscal policy and reliance on public debt can contribute a new means of evaluating such debt's importance in the Indonesian economy. LITERATURE REVIEW Public debt occurs as a result of budget deficits (Das, 2016), which may be caused by numerous factors, including investment in major infrastructure and development projects, war financing, natural disasters, economic crises, and increased government expenditures (Aybarc, 2019). All of these can cause fiscal gaps, in which the government budget is incapable of covering its expenditures (Kiminyei, 2018; Kharusi & Ada, 2018). When responding to such fiscal gaps and budget deficits, governments often utilize public debt (Thilanka & Ranjith, 2018). As such, public debt is an important fiscal resource and strategy through which governments can fund their projects and make up for any shortcomings in their budgets (Kiminyei, 2018). Public debt is defined as direct government fixed-term contractual obligations towards third parties that are not covered by foreign currency reserves or non-stock securities (Aybarc, 2019). Such debt may be categorized as external or domestic. According to Aybarc (2019), external public debt refers to the money borrowed by governments, companies, and individuals from foreign banks, governments, or financial institutions. Meanwhile, Avdjiev, Binder, and Sousa (2017) define external debt as debt owed to non-citizens that is paid in foreign currency. Domestic public debt, meanwhile, refers to government obligations to local money lenders (Avdjiev, Binder & Sousa, 2017; Shkolnyk & Koilo, 2018) or markets (Bua, Pradelli, & Presbitero, 2014). A government may choose to fund its public debt through domestic means for three reasons: to cover budget deficits, to implement monetary policies through open market operations, or to develop the domestic financial market (Alison, 2003). Although many empirical studies have examined the effect of public debt on economic growth, the association between public debt and economic growth (both in developed and developing nations) remains widely debated. Studies have shown that public debt may have a positive or negative effect on economic growth (Dombi & Dedák, 2019). Some have suggested that, while public debt can promote short-term economic growth, over time these positive effects become increasingly deleterious (Shkolnyk & Koilo, 2018; Aybarc, 2019). DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 5 The Negative Association between Government Debt and Economic Growth In 2002, Patillo, Helene, and Luka examined the link between public debt and economic growth in 93 developing countries between 1969 and 1998. Employing panel data analysis, they found evidence that a non-linear Laffer curve explained the association between public debt and economic growth. They also found that the influence of public debt on economic growth extent is informed by its extent, concluding that external debt has a deleterious effect on economic growth when its amount surpasses 35–40% of the GDP and 160–170% of exports. Reinhart and Rogoff (2010) examined the association between high public debt, economic growth, and inflation in 44 countries (both developing and developed). They found that, when public debt is higher than 90% of the GDP, it has a negative effect on the national economy. This finding was supported by Blake (2015), who found that public debt exceeding 100% of the GDP (with more than 55% coming from foreign sources) has a deleterious effect on the economy. Blake reached this conclusion by analyzing quarterly data from Jamaica that covered the period between 1990 and 2014 using an autoregressive distributed-lag (ADRL) model to identify public debt's short- and long-term effects on the economy. Shah and Pervin (2012) analyzed the short- and long-term effects of public debt on economic growth in Bangladesh between 1974 and 2010, finding that external public debt has had deleterious effects in the period studied. They showed that Bangladesh has experienced a debt overhang, wherein the state accumulated a significant amount of debt but lacked the ability to repay said debt. Shah and Pervin also found that this external debt has had a crowding-out effect, wherein private investment became less common owing to the extent of government borrowing. Meanwhile, Puente-Ajovin and Sanso-Navarro (2014) employed a Granger causality test to examine the effect of public debt in OECD countries between 1980 and 2009, finding that economic growth is negatively affected by public debt. Woo and Kumar (2015) similarly showed that, in developed countries, public debt has had a deleterious effect on economic growth, concluding that a 10% increase in public debt would reduce economic growth by 0.2%. Gómez- Puig and Sosvilla-Rivero (2016) investigated members of the European Economic and Monetary Union (EEMU) using a data time series spanning from 1960 through 2012 to analyze the short- and long-term effects of public debt on economic growth, finding that such debt has deleterious consequences over the long term. More recently, Kharusi and Ada (2018) examined the link between government external borrowing and economic growth in Oman using a data time series covering the period from 1990 to 2015. Employing an ADRL cointegration approach, they investigated the short-term dynamics of external debt and economic growth, finding that such debt has had a negative effect on Oman's economic growth. In Pakistan, another developing country, Butt and Hassan (2008) used ARDL and co- integration analysis to investigate the association between external public debt and economic growth between 1975 and 2005. They found that public debt has had no effect on Pakistan's short- and long-term economic growth. Another recent study, conducted by Kum and Öktem (2018), employed autoregressive (AR3) panel data analysis to understand the link between economic 3 The autoregressive (AR) model is a form of stationary time series model, the p-ordered AR model states that the value of the t observation depends on the values of the observation over the previous p period. In DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 6 growth and public debt in 15 EU member states between 2000 and 2016. They found that increases in these countries' GDPs reduced public debt over the short term, but increased it over the long term. They thus found a negative correlation between public debt and long-term GDP growth, which they attributed to public debt's negative effects on economic growth. Public debt may negatively affect economic growth owing to the crowding-out effect (see, Picarelliy, Vanlaer, & Marneffe, 2019; Huang, Panizza & Varghese, 2018), in which private investment is reduced owing to high levels of government borrowing in the domestic market. As the government experiences high levels of fiscal deficits and public debt, the repayment of this debt absorbs available capital resources and increases interest rates (Picarelliy, Vanlaer & Marneffe, 2019). The crowding-out effect occurs when public debts redirect national savings that could otherwise be used for investment. This is exacerbated by the fact that increased interest rates reduce private actors' willingness to borrow money, thereby limiting their investments. According to Picarelliy, Vanlaer and Marneffe (2019), external debt can cause the appreciation of domestic currency, which can reduce competitiveness, stimmy investments, and hinder economic growth (Buaa, Pradelli & Presbitero, 2014; Picarelliy, Vanlaer & Marneffe, 2019). The Positive Association between Government Debt and Economic Growth On the other hand, some empirical studies present a positive association between government debt and economic growth. Al-Zeaud (2014) employed an Ordinary Least Squares (OLS) approach to investigate the link between public debt and economic growth in Jordan between 1991 and 2010, finding that public debt has had a positive effect on the national economy. Meanwhile, Spilioti and Vamvoukas (2015) examined public debt in Greece between 1970 and 2009, finding that it has had positively affected economic growth. More broadly, Karagoz and Caglar (2016) employed a panel data model to investigate the effect of public debt in 17 OECD member states, concluding that public debt and economic growth were positively correlated in these countries. Thao (2018) examined the effects of public debt on economic growth in six ASEAN countries (Indonesia, Malaysia, Filipina, Singapura, Thailand, and Vietnam) between 1995 and 2015 by employing a General Method of Moments (GMM) approach and considering Foreign Direct Investment (FDI), with gross fixed capital and real exchange rates as variables. Thao found that public debt has a significant positive effect on GDP per capita in these countries, which could be attributed in part to their governments' effective use of investments to promote long-term economic growth. Meanwhile, Daud (2016) analyzed the direct link between federal public debt, capital accumulation, and investment levels in Malaysia, finding a positive correlation between public debt and economic growth. This study took investment as representative of economic growth, federal public debt as a proxy for increased productivity, and degree of openness as illustrative of government policy. Employing OLS estimation and an ARDL cointegration bound test, Daud (2016) concluded that the accumulation of federal public debts has been positively correlated with the concept of AR, an event is not always influenced by factors that occur at the same time, but also previously known as lag (time difference). AR model is written: 𝑍𝑡 = ∅1𝑍𝑡−1 + ∅2𝑍𝑡−2 + ⋯ + ∅𝑝𝑍𝑡−𝑝 + 𝑎𝑡 (Greene, 2018). DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 7 economic growth in the country. Importantly, this study indicates that there is an optimal level of public debt for economic growth. Only if this level of debt is exceeded will economic growth be stymied. The positive impact of public debt on a country's economic growth can occur if the increase in external debt or total outstanding debt is less than the interest rate on the total debt accumulation (Todaro & Smith, 2015; Dombi & Dedák, 2019). This concept is known as basic transfer, which is the difference between foreign exchange inflow and foreign exchange outflow at international borrowing or is the calculation of the difference between capital inflow minus capital outflow against interest payments on the accumulation of external debt (Islamov, 2011). This basic transfer concept can be written as follows: BT = dD - rD = (d-r) D. BT is the basic transfer value, r is the interest payment for external debt accumulation, rD is the total interest payment per year, d is the amount of increase in external debt, and dB is the total external debt outstanding. If d> r, then basic transfers will be positive, meaning that countries that have external debt experience economic growth as a result of external debt. Conversely, if r> d, it means that a country that has external debt loses foreign exchange. Government debt, especially foreign debt if allocated to productive investment will encourage economic growth, because productive investment has a rate of return> r. Government Debt and Economic Growth in Indonesia Within an Indonesian context, studies of the link between public debt and economic growth have had mixed results. Swastika, Ginanjar, and Masih (2013) combined a Wavelet Coherence (WTC) approach, Maximal Overlap Discrete Wavelet Transform (MODWT), and multivariate non-linear regression technique of Hansen Threshold (2000). Wavelet analysis was used to examine co-movement and causality, while the Hansen Threshold was employed to investigate non-linearity in the debt-economic growth ratio. This study, which focused on the period between 2003 and 2012, found a complex and dynamic lead–lag relationship between the external debt–GDP ratio and economic growth in Indonesia. It also suggested that public debt is inversely related with short-term economic growth, while contributing minimally to long-term economic growth. Another study was conducted by Cholifihani (2008), who analyzed the short- and long- term effect of external public debt on GDP growth between 1980 and 2005, finding a negative correlation in which every 1% increase in public debt reduced the GDP by 0.13%. Muhdi and Sasaki (2009) investigated the effects of government borrowing on Indonesia's macroeconomy to determine whether the crowding-out effect was occurring in the country. They found that, although external public debt has promoted investment and economic growth in Indonesia, it has also caused the rupiah to depreciate as the government has had to repay its debts. This has been particularly prominent since the Southeast Asian financial crisis of 1997, when the Indonesian government accumulated significant domestic debts. Ultimately, they find that external public debt has resulted in a crowding-out effect. A relatively recent study was conducted by Djulius (2018), who examined the role of FDI, foreign borrowings, and domestic savings in short- and long-term economic development. Using data from the World Bank and Bank Indonesia, this study used an error correction model (ECM) DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 8 to investigate the link between several variables and economic growth. Djulius found that external debt and domestic savings have significantly influenced Indonesia's short-term economic growth; however, over the long-term, FDI has a significant negative effect on economic growth. Meanwhile, although foreign borrowings have negatively affected Indonesia's short-term economic growth, they do not significantly affect growth over the long term. Table 1. Association of Public Debt and Economic Growth Author Sample Result Al-Zeaud (2014) Jordan There is a positive role for government debt to the economy. Blake (2015) Jamaica If the ratio of government debt to GDP > 100%, and external debt> 55% of GDP, government debt will have a negative impact on the economy Butt & Hassan (2008) Pakistan Government debt has no impact on economic growth both in the short term and long term. Cholifihani (2008) Indonesia There is a negative relationship between external debt and GDP. Daud (2016) Malaysia If the government debt does not exceed the optimal debt limit, the debt will have a positive impact on the economy, conversely if the debt exceeds the optimal level, it will have a negative impact on the Malaysian economy. Djulius (2018) Indonesia In the long run, foreign debt will not significantly affect economic growth. Gómez-Puig & Sosvilla-Rivero (2016) European Union Countries Government debt has a negative impact on the economic performance of EU countries in the long run. Karagoz & Caglar (2016) 17 OECD countries There is a positive relationship between government debt and economic growth. Kharusi & Ada (2018) Oman External debt has a negative impact on economic growth. Kum & Öktem (2018) 15 European Union Countries There is a negative correlation between government debt and GDP growth in the long run. Muhdi & Sasaki (2009) Indonesia External debt has a positive effect on investment and economic growth in Indonesia. Patillo, Helene, & Luka (2002) 93 Developing Countries External debt has a negative association with economic growth. Puente-Ajovin & Sanso-Navarro (2014) OECD Countries Economic growth and government debt are negatively correlated. Reinhart & Rogoff (2010) 44 countries If the ratio of government debt to GDP exceeds 90%, then government debt will have a negative impact on the economy. Shah & Pervin (2012) Bangladesh The accumulation of large amounts of debt with weak ability to pay causes a crowding out effect. Spilioti & Vamvoukas (2015) Greek Government debt has a positive impact on economic growth. Swastika, Ginanjar & Masih (2013) Indonesia There is an inverse relationship between debt and economic growth in the short run and the contribution of debt to the Indonesian economy is relatively small in the long run. DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 9 Thao (2018) 6 countries ASEAN There is a positive and significant impact of public debt on GDP growth per capita. Woo & Kumar (2015) 38 advanced and emerging economies An increase in government debt will has deteriorate economic growth. Source: Author’s summary based on existing literature. RESEARCH METHOD Model Specification This study takes public debt, both external and domestic, and real GDP growth. For this, it employs a data time series covering 1998 through 2018. Data regarding Indonesia's GDP and debts were obtained from the Indonesian Ministry of Finance. To determine whether a causal relationship exists between public debt and economic growth, a bivariate model of public debt and economic growth was examined, as written in the following equation: GDPt = α0 + α1 GDDt + α2 GEDt + µt (1) In this equation, 𝛼0 is the constant, 𝛼1 and 𝛼2 are the estimated parameters, t = 1, 2 … 20 are indices of the years covered by this study (1998 through 2018), and µ is the stochastic error term. Meanwhile, GDP refers to Indonesia's economic growth (as indicated by real GDP), GDD refers to gross domestic debt, and GED refers to its gross external (foreign) debt. To understand Indonesia's economic growth, data were collected regarding its real economic growth; meanwhile, to understand its public debts, data were collected regarding its outstanding domestic and external debts. Descriptive analysis was employed to examine the distribution of observed data and avoid biased results (Greene, 2018). The Jarque–Bera normality test was further used to examine the normality of the data time-series, with the zero hypothesis that data would be normally distributed. Statistical testing referred to Jarque (2011): Where: JB = amount of observations; n = amount of data; S = skewness; K = kurtosis To ensure the normality of the data, Gross Domestic Debt (GDD) and Gross External Debt (GED) were converted to logarithmic values. Consequently, Equation (1) may be written as: 𝐿𝐺𝐷𝑃𝑡 = 𝛼0 + 𝛼1𝐿𝐺𝐷𝐷𝑡 + 𝛼2𝐿𝐺𝐸𝐷𝑡 + 𝜇𝑡 (3) As seen in Table 2, before the variable GDP was converted to logarithmic form, it did not meet the Jarque–Bera criteria for normality, 129.03 > Prob. 0.00 (Jarque, 2011; Abdellatif, Moutaouakilb & Satori, 2018; Jarque, 2011)—and thus regression produced biased results. Given this finding, GDP was converted to logarithmic form, then it is named as LGDP, producing the results 0.46 < Prob. 0.79. This indicated that normal distribution was achieved through LGDP. As such, bias in regression results was minimal (Jarque, 2011; Mantalos, 2010). 𝐽𝐵 = 𝑛 [ + 6 24 (2) DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 10 Table 2. Summary Statistics Description Description (After converting GDP to logarithmic form) Mean 4,86E-16 Mean 2,75E-15 Median 0,800865 Median 0,168612 Maximum 3,031828 Maximum 4,245945 Minimum -15,31601 Minimum -4,981990 Std. Deviation 3,840424 Std. Deviation 2,027994 Skewness -3,171624 Skewness -0,260217 Kurtosis 13,35519 Kurtosis 3,506524 Jarque-Bera 129,0335 Jarque-Bera 0,461491 Probability 0,000000 Probability 0,793942 Source: Authors’ calculation using EVIEWS 10. RESULT AND DISCUSSION Unit Root: Augmented Dickey–Fuller Stationarity Test When conducting time series analysis, it is necessary to ensure that the time series being analyzed is stationary, i.e. is consistent and constant. The most widely used method for checking stationary of data is the Augmented Dickey–Fuller (ADF) test (Dickey & Fuller, 1981; Paparoditis & Politis, 2018; Salles et al., 2019). This test utilizes unit root tests to determine whether the analyzed time series is stationary, thereby avoiding spurious regression. ADF testing employs the following equation: Yt  a0  zt  a1Yt  1  aiYt  1  t i1 (4) with a0 being the constant and t being the deterministic trend. If the autoregressive (AR) representation Yt contains a unit root, the t-ratio for a1 must be consistent with the hypothesis a1 = 0 (MacKinnon, 2010). According to this test, if the p-value > 5%, it may be determined that the analyzed variable includes a unit root; conversely, if the p-value < 5%, no unit root exists. Afterwards, the best lag duration for the variable was chosen to enable a Granger causality analysis of public debt and economic growth. As this approach examines the association between variables, it requires the lagged value of the variable being analyzed to be linked with the values of other independent variables. For time series with n iterations, the analysis follows the equation yt = (y1t, y2t, y3t, …, ynt). The degree of  (VAR ()) was determined through the following equation: 𝑦𝑡 = 𝐴0 + 𝐴1𝑦𝑡−1 + 𝐴2𝑦𝑡−2 + … + 𝐴𝑝𝑦𝑡−𝑝 + (5) With  being the lag amount, n being the number of variables, yt being the vector (n,1) of in the VAR model; A0 being the vector (n.1) of the intercept; A1 being the vector (n.n) of the coefficient matrix, and εt being the vector (n.1) of the error terms. As shown in Table 3, the ADF test shows that two variables are not stationary at the 5% level, namely domestic public debt (LGDD) and external public debt (LGED). Meanwhile, the variable GDP was stationary, with a t-statistic value of -2.342166. ADF testing was thus conducted at the first difference (Kulaksizoglu, 2015; Paparoditis & Politis, 2018). At the first difference, the p DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 11 variable LGDD was not stationary at the 5% level, while LGDP and LGED were stationary, with t- statistic values of -12.45130 and -4.319236 (Table 3). ADF testing was thus conducted at the second difference, a higher level of integration. This test found that LGDP was stationary, with a t-statistic value of -9.609003; LGDD was stationary, with a t-statistic value of -6.424901; and LGED was stationary, with a t-statistic value of -4.440034. As such, all variables were determined to be stationary at the second difference. Table 3. The Augmented Dickey Fuller Test Variables ADF t-statistics: Level Values ADF t-statistics: first- order difference ADF t-statistics: second- order difference LGDP -2.342166 (0.02190) -12,45130 (0.0001) -9.60900 (0.0000) LGDD 1.756847 (0.97520) -0.0694642 (0.3999) -6.424901 (0.000) LGED 2.181299 (09902) -4.3019236 (0.0002) -4.440034 (0.0002) Notes: the null hypothesis: GDP; LGDD; LGED has a unit root. Number in brackets are p value, the significance value of 5%. Source: Authors’ calculation using EVIEWS 10. Further analysis of the time series required the application of a VAR or VECM model (Rothe & Sibbertsen, 2006; Kim & Choi, 2017). To determine which model was appropriate, a cointegration test was conducted. If cointegration was found, the VECM model would be used. Conversely, if no cointegration was found, the VAR model would be employed. The model was chosen after the stability of the model and the optimal lag length were determined (Jian et al., 2019). Estimation of Vector Autoregressive (VAR) Autocorrelation is most commonly a problem in VAR systems, and as such determining the optimal lag length was necessary (Greene, 2018). There are many methods that can determine the optimal lag period for the VAR model, among other is to see the polynomial AR roots, including by inspecting the AR roots polynomial value (Jian et al., 2019; Lutkepohl, 2017; Morano, 2012). If the value of the AR roots polynomial < 1, the model may be identified as stable. As reported in Table 4, through the third lag no value was higher than one; as such, the model was declared stable. Table 4. Root of Characteristic Polynomial Lag 3 Root Modulus 0,944489 – 0,159699i 0,957895 0,944489 + 0,159699i 0,957895 -0,734370 0,734370 -0,239267 – 0,653343i 0,695777 -0,239267 + 0,653343i 0,695777 0,656430 0,656430 0,272507 – 0,456090i 0,531299 0,272507 + 0,456090i 0,531299 0,275579 0,275579 Source: Authors’ calculation using EVIEWS 10. To determine the optimal lag length within the VAR or VECM model, the lowest value of the Akaike Information Criterion (AIC), Final Prediction Error (FPE), Schwarz Information DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 12 𝑖=𝑟+1 Criterion (SIC), and Hannan-Quinn Criterion (HQ) were sought, as was the highest LR value (Cernat-Cernat, 2009; Muruganandan, 2017; Niedzwiecki & Ciolek, 2017). As seen in Table 4, lag one was optimal, as it had the lowest FPE and SIC values as well as the highest LR value. This indicated that the variables within the model are not only mutually influential in the current period but were also mutually influential in the previous period. The VAR lag order results also showed that lag one was most appropriate for cointegration, VAR, VECM, and Granger causality testing. The decision to use lag one was further supported by the fact that LR had a statistical value of 38.33275 and FPE had a value of 2.07e-06, as seen in Table 5 (see, Niedzwiecki & Ciołek, 2017), as well as the fact that the SIC had a statistical value of -4.012209 and HQ value of - 4.541896 (both significant at the 5% level). Table 5. Vector Autoregressive Criteria Lag Order Lag Logl LR FPE AIC SC HG 0 26,03933 NA 1,34e-05 -2,710509 -2,563472 -2,695894 1 51,10305 38,33275* 2,07e-06* -4,600359* -4,012209* -4,541896* 2 55,67869 5,383099 3,97e-06 -4,079846 -3,050582 -3,977535 3 65,20070 7,841658 5,37-06 -4,141259 -2,670882 -3,995101 Notes: * significant at 5% significance level. Source: Authors’ calculation using EVIEWS 10. Cointegration Test A Johansen cointegration test was conducted to ascertain the long-term correlation of the values being analyzed, a necessary precursor for employing the vector error correction model (VECM) (Menegaki, 2019; Jian et al., 2019; Zou, 2018). The Johansen cointegration test involves two stages. First, the VAR order is determined, then the cointegration vector is tested, with the null hypothesis being that no cointegration exists. This null hypothesis may be rejected after maximum eigenvalue and trace testing, both of which are conducted with the null hypothesis that no cointegration exists. The Johansen test was conducted by checking , with     (with  being the amount of cointegration vectors within the VECM equation and β being the cointegration vector). To test the cointegration of the variables being analyzed, their rank within  was examined through their eigenvalues (Menegaki, 2019). Different null hypotheses were used in eigenvalue and trace testing. For the trace test, the null hypothesis that the amount of cointegration vectors ≤ r. Meanwhile, for the maximum eigenvalue test, a separate test was used for individual eigenvalues, with the null hypothesis being that the number of cointegration vectors was r, with the alternative being (r + 1). The following equations were used for the maximum eigenvalue test and for the trace test: 𝜆𝑇𝑟𝑎𝑐𝑒(𝑟) = −𝑇 ∑ 𝑔 𝑖𝑛(1 − �̂�𝑖) (6) 𝜆𝑀𝑎𝑥(𝑟, 𝑟 + 1) = −𝑇 ln(1 − �̂�𝑟+1) (7) As shown in Table 6, the Johansen cointegration test resulted in a trace statistic value of 47.52813 > critical value 29.79707 and a maximum eigenvalue of 31.03284 > critical value 21.13162. As this indicated that the analyzed variables were stably cointegrated over the long term, further analysis DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 13 was conducted using the vector error correction model (VECM) (Jian et al., 2019; Liang & Schienle, 2019). Table 6. Johansen Test for Cointegration Unrestricted Cointegration Rank Test (Trace) Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0,05 Critical Value Prob.** None* 0,804717 47,52813 29,79707 0,0002 At most 1* 0,536756 16,49529 15,49471 0,0352 At most 2 0,093961 1,874776 3,841466 0,1709 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) Hypothesized No. of CE(s) Eigenvalue Max-Eigen Stat. 0,05 Critical Value Prob.** None* 0,804717 31,03284 21,13162 0,0015 At most 1* 0,536756 14,62051 14,26460 0,0439 At most 2 0,093961 1,874776 1,874776 0,1709 Source: Authors’ calculation using EVIEWS 10. To confirm the relevance of the VECM, the stationarity of the residual (e) at the level through ADF is assesed. As shown in Table 6, the residual (e) is stationary at the level, with lag one not containing any unit roots. It may thus be concluded that cointegration was found amongst all the model's variables, and as such VECM analysis was appropriate. Table 7. ADF Test for the Residual Variable ADF t- statistics Critical Value Conclusion 1 % 5% 10% e -2,923095 -2,685718 -1,959071 -1,607458 Stationary Source: Authors’ calculation using EVIEWS 10. VECM Estimation and Granger Causality As alluded earlier, after determining that all the variables investigated contained cointegration, it was possible to employ VECM Estimation and Granger causality. The equations used for VECM analysis are presented below: 𝛥𝐿𝐺𝐷𝑃𝑡 = 𝐴1 ∑ 𝑘−1 𝐵1𝑖𝛥𝐿𝐺𝐷𝑃𝑡−1 + ∑ 𝑘−1 𝐶1𝑖𝛥𝐿𝐺𝐷𝐷𝑡−1 + ∑ 𝑘−1 𝐷1𝑖𝛥𝐿𝐺𝐸𝐷𝑡−1 + 𝜇1𝐸𝐶𝑇𝑡−1 + 𝑒𝑡 (8) 𝑖=1 𝑖=1 𝑖=1 𝛥𝐿𝐺𝐷𝐷𝑡 = 𝐴2 ∑ 𝑘−1 𝐵2𝑖𝛥𝐿𝐺𝐷𝑃𝑡−1 + ∑ 𝑘−1 𝐶2𝑖𝛥𝐿𝐺𝐷𝐷𝑡−1 + ∑ 𝑘−1 𝐷2𝑖𝛥𝐿𝐺𝐸𝐷𝑡−1 + 𝜇2𝐸𝐶𝑇𝑡−1 + 𝑒𝑡 (9) 𝑖=1 𝑖=1 𝑖=1 𝛥𝐿𝐺𝐸𝐷𝑡 = 𝐴3 ∑ 𝑘−1 𝐵2𝑖𝛥𝐿𝐺𝐷𝑃𝑡−1 + ∑ 𝑘−1 𝐶2𝑖𝛥𝐿𝐺𝐷𝐷𝑡−1 + ∑ 𝑘−1 𝐷3𝑖𝛥𝐿𝐺𝐸𝐷𝑡−1 + 𝜇3𝐸𝐶𝑇𝑡−1 + 𝑒𝑡 (10) 𝑖=1 𝑖=1 𝑖=1 As discussed above, cointegration testing was used to show the long-term association between economic growth and government debt. To determine causality, meanwhile, it was necessary to employ VECM Estimation (Ahad & Dar, 2018; Zou, 2018). If the economic growth variable (LGDP) is predictive of changes in government debt (LGDD, LGED), regression can be shown through these two variables. If this occurs, economic growth may be identified as the Granger cause of changes in government debt. Conversely, if this is not the case, it may be concluded that no Granger causality exists (Menegaki, 2019; Jian et al., 2019). A variable may be understood as significantly influential if its t statistics < t table. In the model used, the t tables were 2.10092 (5%) and 1.73406 (10%). Meanwhile, with the degree of DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 14 freedom being n - k - 1, this model had a degree of freedom of 21 - 2 - 1 at the 5% level. As such, it DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 15 was concluded that domestic government debt (LGDD) and external government debt (LGED) have significant long-term effects on economic growth (LGDP), as shown by the t-statistics being 2.08297 and 4.48951. Domestic debt is positively correlated with economic growth, while external debt is negatively correlated. Based on Table 8, the cointegration equation may be written: GDPt-1 = 99,407667 + 7,074042GDDt-1 - 14,60334GED t-1 It was found that every percentage point increase in domestic government debt was followed by a 7.074042 percentage point increase in economic growth; conversely, every 1 percentage point increase in external government debt reduced economic growth by 14.6033 percentage points. This outcome highlights the long-run inverse relationship between external debt and economic growth in Indonesia. Table 8. Estimation Results of Cointegration Equation Cointegrating Eq: CointEq1 LGDP(-1) 1,000000 LGDD(-1) 7,074042 (2,08297*) [3,39613] LGED(-1) -14,60334 (4,48951*) [-3,25277] C 99,40766 Notes: numbers in square brackets and in parenthesis are the corresponding standard errors and the t -statistics respectively. Source: Authors’ calculation using EVIEWS 10. VECM estimation results are summarized in the table below, showing that the VECM model had a relative goodness of fit model except for Equation (10), as the R2 value > 0.8 and the AIC and SIC were relatively low. For Equation (8), ECTt-1 was negative and statistically significant at the 5% level. The short-term VECM outcome indicates that changes in domestic debt (LGDD) in one to two previous time periods did not significantly influence changes in economic growth (LGDP) as evidenced by the respective t-values, namely [1.24105] and [-0, 56581] t-table 2.10092. The LGED coefficient value (-1) of -9.295100 explains that when there is an increase in changes in external debt in the previous period of 1 unit, then the change in economic growth (LGDP) will decrease by -9.295100 percent. Furthermore, the LGED coefficient (LGED (-2)) of -4.468089 explains that when there is an increase in external debt changes in the previous two periods of 1 unit, the change in LGDP economic growth will decrease by -4.468089 percent. Based on the value of Adj. R-squared in the short-term VECM equation, changes in the domestic government debt (LGDD) and external government debt (LGED) are able to explain variations in changes in the economic growth (LGDP) of 0.702621 or 70.26 percent, while the remaining 29.74 percent explained by changes in other variables outside the VECM being estimated. DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 16 𝑗=1 Table 8. Estimation Results of VECM: Short Run Estimation Error Correction: D(LGDP) D(LGDD) D(LGED) CointEq1 -0,441880 0,027797 0,056320 (0,15209) (0,00709) (0,03083) [-2,90544] [3,91980] [1,82703] D(LGDP(-1)) -0,116651 0,007063 -0,039686 (0,22536) (0,01051) (0,04568) [-0,51763] [0,67218] [-0,86884] D(LGDP(-2)) 0,110138 -0,010809 -0,035735 (0,14499) (0,00676) (0,02939) [0,75963] [-1,59883] [-1,21601] D(LGDD(-1)) 5,136255 0,584710 -0,524458 (4,13862) (0,19298) (0,83884) [1,24105] [3,02997] [-0,62522] D(LGDD(-2)) -1,007766 -0,049512 0,483000 (1,78111) (0,08305) (0,36101) [-0,56581] [-0,59617] [1,33793] D(LGED(-1)) -9,295100 0,154685 0,015508 (2,13084) (0,09936) (0,43189) [-4,36217] [1,55686] [0,03591] D(LGED(-2)) -4,468089 0,270916 -0,166922 (1,89684) (0,08845) (0,38446) [-2,35555] [3,06308] [-0,43417] C 0,366087 0,026643 0,062809 (0,21800) (0,01017) (0,04419) [1,67926] [2,62102] [1,42146] R-squared 0,825071 0,943989 0,556897 Adj. R-squared 0,702621 0,904781 0,246725 F-statistics 6.738029 24.07642 1.795446 Akaike information criterion 1.396101 -4.735008 -1.796091 Schwarz criterion 1.791822 -4.229287 -1.400370 Notes: Number in parentheses are the standard errors; Number in square bracket are the t-statistics. Source: Authors’ calculation using EVIEWS 10. From the results of the VECM estimation in the table above, the following VECM is generated: ∆𝐺𝐷𝑃𝑡 −0,441880 −0,116651 0,007063 −0,039686 ∆𝐺𝐷𝑃𝑡−1 [∆𝐿𝑛𝐺𝐷𝐷𝑡] = [ 0,027797 ] + [ 5,136255 0,584710 −0,524458] [∆𝐿𝑛𝐺𝐷𝐷𝑡−1] ∆𝐿𝑛𝐺𝐸𝐷𝑡 0,056320 −9,295100 0,154685 0,015508 ∆𝐿𝑛𝐺𝐸𝐷𝑡−1 0,110138 −0,010809 −0,035735 ∆𝐺𝐷𝑃𝑡−2 + [−1,007766 −0,049512 0,483000 ] [∆𝐿𝑛𝐺𝐷𝐷𝑡−2] −4,468089 0,270916 −0,166922 ∆𝐿𝑛𝐺𝐸𝐷𝑡−2 Granger causality testing is used to examine the causal association between variables. Two variables may be said to have a causal relationship if their values are mutually predictive. For example, Variable X may be identified as the Granger cause of Variable Y if its past value can be used to predict the latter's current value (see Zou, 2018). The Granger causality test may be DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 17 𝑖=1 expressed mathematically as follows: 𝑦𝑡 = ∑𝑝 𝑎11 𝑦𝑡 − 𝑗 ∑𝑝 𝑎12 𝑦𝑡 − 𝑗 𝜀1𝑡 (11) DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 18 𝑗=1 𝑗=1 𝑋𝑡 = ∑𝑝 𝑎21 𝑦𝑡 − 𝑗 ∑𝑝 𝑎22 𝑦𝑡 − 𝑗 𝜀2𝑡 (12) With p being the maximum lag and a being the coefficient of the model. For this study, Granger causality was tested by examining the association between LGDP, LGDD and LGED as variables in the VECM. As seen in Table 10, the Granger causality test found that domestic public debt (LGDD) was not the Granger cause of economic growth (LGDP) (p-value > 0.05); similarly, economic growth (LGDP) was not the Granger cause of domestic public debt (LGDD). However, external public debt (LGED) was statistically found to be a Granger cause of economic growth (LGDP), while economic growth (LGDP) was identified as a Granger cause of external public debt (LGED), with a -value < 0.05. Table 10. Estimation Results of VECM Granger Causality Dependent variable: D(GDP) Excluded Chi-sq df Prob. D(LGDD) 2,421513 2 0,2980 D(LGED) 19,14891 2 0,0001 All 32,95436 4 0,0000 Dependent variable: D(LGDD) Excluded Chi-sq df Prob. D(GDP) 3,793536 2 0,1501 D(LGED) 9,400302 2 0,0091 All 13,41062 4 0,0094 Dependent variable: D(LGED) Excluded Chi-sq df Prob. D(GDP) 1,815510 2 0,4034 D(LGDD) 2,756232 2 0,2521 All 2,757619 4 0,5992 Notes: the null hypothesis: existence of Granger cause. Source: Authors’ calculation using EVIEWS 10. DISCUSSION This study has examined the causal relationship between government debt (both domestic and external) and economic growth in Indonesia between 1998 and 2018. A stationary data time series was analyzed using ADF testing. At the level, economic growth (GDP) was stationary, while domestic government debt (LGDD) and external government debt (LGED) were not. At the first difference, meanwhile, government domestic debt (LGDD) was not stationary, but LGDP and LGED were stationary at the 5% level. Further ADF testing found that, at the second difference, all variables were stationary. This enabled the data time series to be analyzed using a VAR or VECM model. To determine which analytical model was appropriate, Johansen cointegration testing was conducted. This test found a stable long-term association between economic growth (LGDP), domestic public debt (LGDD), and external public debt (LGED). External public debt was found to significantly and negatively influence economic growth, while domestic public debt had a significant and positive effect. The positive effect of public debt on economic growth supports the findings of Thao (2018), Spilioti and Vamvoukas (2015), and Daud (2016). The Granger causality test found unidirectional causality between external public debt and economic growth. This reinforces earlier findings (see Cholifihani, 2008; and Djulius, 2018) that external public debt is DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 19 negatively associated with economic growth in Indonesia, and refutes findings (see Muhdi and Sasaki, 2009) that external debt and economic growth are negatively correlated. This suggests that the Indonesian government should focus on domestic debt, rather than external debt, as the former is positively associated with economic growth. This reflects the findings of Buaa, Pradelli, and Presbitero (2014), who found that domestic public debt poses fewer risks to national currency, reduces vulnerability, and enables the government to implement countercyclical monetary policy to mitigate the deleterious effects of outside phenomena. To optimize its domestic sources of debt, the Indonesian government must improve its institutional frameworks and promote a healthy macroeconomy, particularly as related to the debt market and public debt management (Abbas & Christensen, 2010). CONCLUSION In managing the state budget, the Indonesian government encounters one of the important challenges, namely the budget deficit, both domestic and foreign sources of income cannot meet development funding. This condition causes the government to make debt as an effort to cover the budget deficit. The average budget deficit of 330 trillion Rupiah from 2011 to 2018. Meanwhile, for 20 years from 1998 to 2018, the ratio of government debt to Indonesian GDP averaged 43.7%. The role and impact of a country's government debt on development and economic growth remains a topic of theoretical and practical debate among academics and policymakers. Thus, this study was motivated by the theoretical and empirical discussions of the association between public debt and economic growth. It investigated the relationship between government debt and economic growth in Indonesia over the period 1998 to 2018. Data regarding public debt and economic growth from the Indonesian Ministry of Finance, covering the period 1998 and 2018. Empirical analysis was carried out with three main procedures. First, the stationarity properties of the time series were checked, so that the stationarity test was performed using the ADF test. Results from the ADF test demonstrate that all the variable series were stationary at the second difference. Second, Johansen cointegration test is performed to check for the long run cointegration relationship between the time series. The Johansen cointegration test demonstrate that series are cointegrated based on the trace test and the maximum Eigenvalue. Cointegration test results also revealed that in the long run, domestic government and economic growth have a positive and significant association. On the other hand, government debt from external sources is negatively associated with Indonesia's economic growth. This finding corroborates the neoclassical view which posits that external debt may impede economic growth due to a crowding- out effect of government debt. Third, based on the long-term association between government debt and economic growth, the VECM is adopted for empirical analysis purposes. Granger causality and VECM estimations were used to establish the causal relationship between government debt and economic growth. The empirical results of this analysis suggested that economic growth, domestic public debt, and external public debt are co-integrated over the long term. Testing further found that, although a causal association existed between external public debt and economic growth, no such association existed between domestic public debt and economic growth. IMPLICATION This study has some policy implication to guide policymakers in the area of government debt management in Indonesia. Its findings reaffirm that domestic government debt has a DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 20 significant and positive relationship with economic growth in Indonesia. As such it underscores the importance of such debt for economic development in Indonesia. For a policy perspective, policymakers need to be better prepared instruments for developing government debt in the country for example by providing more effective debt instruments for mobilizing debt that are attractive to the Indonesian and foreign investors. In contrast, external debt appears to dictate long-term GDP growth; consequently, the government must be more cautious in dealing with such debt. Future studies could focus on the reason why domestic public debt positively influences economic growth while external public debt has a negative influence, as this study did not consider such issues. One could also examine the source of economic growth, such as domestics and foreign investments, export, imports and domestic interest rate. ACKNOWLEDGMENT This work was partly funded by the Department of Business Administration, Atma Jaya Catholic University of Indonesia through Research Grant of 2018. REFERENCES Abbas, A., & Christensen, J. (2010). The Role of Domestic Debt Markets Ineconomic Growth: An Empirical Investigation for Low-income Countries and Emerging Markets. 209–255. IMF Staff Papers 57. Abdellatif, D., Moutaouakil, K. E., & Satori, K. (2018). Clustering and Jarque-Bera Normality Test to Face Recognition. Procedia Computer Science, 127, 246-255. doi:10.1016/j.procs.2018.01.120 Ahad, M., & Dar, A. A. (2018). Dynamic Relationship between Financial Development and Import Demand for Bangladesh: An Evidence from Combine Cointegration and Granger Causality Approach. Global Business Review, 19(3), 543-555. doi:10.1177/0972150917713863 Alison, J. (2003). Key Issues for Analyzing Domestic Debt Sustainability. Debt Relief International Publication. Al-Zeaud, H. A. (2014). Public Debt and Economic Growth: An Empirical Assessment. European Scientific Journal, 10(4), 148-158. Avdjiev, S., Binder, S., & Sousa, R. (2017). External Debt Composition and Domestic Credit Cycles. Bank for International Settlements, Monetary and Economic Department . Basel, Swiss: BIS Working Papers 627. Retrieved May 7, 2019, from https://www.bis.org/publ/work627.pdf Aybarc, S. (2019). Theory of Public Debt and Current Reflections. (B. Acikgoz, Ed.) Public Economics and Finance, IntechOpen, 1-12. doi:10.5772/intechopen.82730 Barrett, P. (2018). Interest-Growth Differentials and Debt Limits in Advanced Economies. IMF Working Paper, 18(82), 1-55. Retrieved from https://www.imf.org/en/Publications/WP/Issues/2016/12/31/The-Role-of-Domestic-Debt-Markets-in-Economic-Growth-An-Empirical-Investigation-for-Low-20973 https://www.imf.org/en/Publications/WP/Issues/2016/12/31/The-Role-of-Domestic-Debt-Markets-in-Economic-Growth-An-Empirical-Investigation-for-Low-20973 https://www.researchgate.net/publication/323841300_Clustering_and_Jarque-Bera_Normality_Test_to_Face_Recognition https://www.researchgate.net/publication/323841300_Clustering_and_Jarque-Bera_Normality_Test_to_Face_Recognition https://journals.sagepub.com/doi/abs/10.1177/0972150917713863 https://journals.sagepub.com/doi/abs/10.1177/0972150917713863 https://journals.sagepub.com/doi/abs/10.1177/0972150917713863 https://eujournal.org/index.php/esj/article/view/2688 https://www.bis.org/publ/work627.htm https://www.bis.org/publ/work627.htm http://www.bis.org/publ/work627.pdf http://www.bis.org/publ/work627.pdf https://www.intechopen.com/books/public-economics-and-finance/theory-of-public-debt-and-current-reflections https://www.imf.org/en/Publications/WP/Issues/2018/04/11/Interest-Growth-Differentials-and-Debt-Limits-in-Advanced-Economies-45794 DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 21 https://www.imf.org/en/Publications/WP/Issues/2018/04/11/Interest-Growth- Differentials-and-Debt-Limits-in-Advanced-Economies-45794 Blake, T. (2015). Investigating the Impact of Public Debt on Economic Growth in Jamaica. Working Paper. bank of Jamaica. Retrieved December 10, 2018, from http://boj.org.jm/uploads/pdf/papers_pamphlets/papers_pamphlets_Investigating_th e_Impact_of_Public_Debt_on_Economic_Growth_in_Jamaica.pdf Bökemeier, B., & Greiner, A. (2015). On the Relation between Public Debt and Economic Growth: An Empirical Investigation. Economics and Business Letters , 4, 137–50. Bua, G., Pradelli, J., & Presbitero, A. F. (2014). Domestic public debt in Low-Income Countries: Trends and structure. Review of Development Finance, 1-19. doi:http://dx.doi.org/10.1016/j.rdf.2014.02.002 Butt, S. A., & Hassan, A. (2008, July). Role of Trade, External Debt, Labor Force and Education in Economic Growth: Empirical Evidence from Pakistan by Using ARDL Approach. European Journal of Scientific Research, 20(4). Cernat-Cernat, B. (2009). Some Empirical Evidence on The Relationship between Acquisitions, Disposals and The Stock Market. The Romanian Economic Journal, 7(33), 1-19. Cholifihani, M. (2008). A Co-integration Analysis of Public Debt Service and GDP in Indonesia. Journal of Management and Social Sciences , 4(2), 68-81. Das, P. (2016). Debt Dynamics, Fiscal Deficit, and Stability in Government Borrowing in India: A Dynamic Panel Analysis. Asian Development Bank Institute. Tokyo: ADBI Working Paper. Retrieved December 8, 2018, from https://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf Daud, S. N. (2016). The Real Effect of Government Debt: Evidence from The Malaysian Economy. Journal of Economic Coopertaion and Development, 37(3), 57-86. Dickey, D. A., & Fuller, W. A. (1981). Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root. Econometrica, 49(4), 1057-1072. doi:10.2307/1912517 Djulius, H. (2018). Foreign Direct Investment or External Debt and Domestic Saving: Which Has Greater Impacts on Growth. Etikonomi: Jurnal Ekonomi, 17(1), 37-44. doi:10.15408/etk.v17i1.7120 Dombi, A., & Dedak, I. (2019). Public Debt and Economic Growth: What do Neoclassical Growth Model Teach Us? Applied Economics, 51(29), 3104-3121. doi:10.1080/00036846.2018.1508869 Doran, P. T., & Zimmerman, M. K. (2009). Examining The Scientific Consensus on Climate Change. Eos Trans, AGU, 90(3), 22-23. doi:10.1029/2009EO030002 http://www.imf.org/en/Publications/WP/Issues/2018/04/11/Interest-Growth- http://www.imf.org/en/Publications/WP/Issues/2018/04/11/Interest-Growth- http://www.imf.org/en/Publications/WP/Issues/2018/04/11/Interest-Growth- http://boj.org.jm/uploads/pdf/papers_pamphlets/papers_pamphlets_Investigating_the_Impact_of_Public_Debt_on_Economic_Growth_in_Jamaica.pdf http://boj.org.jm/uploads/pdf/papers_pamphlets/papers_pamphlets_Investigating_the_Impact_of_Public_Debt_on_Economic_Growth_in_Jamaica.pdf http://boj.org.jm/uploads/pdf/papers_pamphlets/papers_pamphlets_Investigating_th https://d-nb.info/1150851724/34 https://d-nb.info/1150851724/34 https://www.sciencedirect.com/science/article/pii/S1879933714000037 https://www.sciencedirect.com/science/article/pii/S1879933714000037 http://dx.doi.org/10.1016/j.rdf.2014.02.002 https://www.semanticscholar.org/paper/Some-Empirical-Evidence-on-the-Relationship-Between-Cernat-Gruici/6c71fc81849dc5a2b05f9040e54e5b302c8a4e7d https://www.semanticscholar.org/paper/Some-Empirical-Evidence-on-the-Relationship-Between-Cernat-Gruici/6c71fc81849dc5a2b05f9040e54e5b302c8a4e7d http://ibt.edu.pk/assets/custom/journal/jbs/4.2/1.%20A%20Cointegration%20Analysis%20of%20Public%20Debt%20Service%20and%20GDP%20in%20Indonesia.pdf https://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf https://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf http://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf http://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf http://www.adb.org/sites/default/files/publication/181400/adbi-wp557.pdf http://www.sesric.org/pdf.php?file=ART15012803-2.pdf https://www.jstor.org/stable/1912517 https://www.jstor.org/stable/1912517 http://journal.uinjkt.ac.id/index.php/etikonomi/article/view/7120 http://journal.uinjkt.ac.id/index.php/etikonomi/article/view/7120 https://www.tandfonline.com/doi/full/10.1080/00036846.2018.1508869 https://www.tandfonline.com/doi/full/10.1080/00036846.2018.1508869 https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2009EO030002 https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2009EO030002 DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 22 Eberhardt, M., & Presbitero, A. F. (2015, September). Public Debt and Growth: Heterogeneity and Non-Linearity. Journal of International Economics, 97(1), 45-58. doi:10.1016/j.jinteco.2015.04.005 Gómez-Puig, M., & Sosvilla-Rivero, S. (2016). Debt-Growth Linkages in EMU Across Countries and Time Horizons. IREA Working Papers . University of Barcelona, Research Institute of Applied Economics. Retrieved January 8, 2019, from https://econpapers.repec.org/paper/irawpaper/201610.htm Greene, W. H. (2018). Econometric Analysis (8th ed.). United States: Pearson. Gupta, P., & Singh, A. (2016). Causal Nexus between Foreign Direct Investment and Economic Growth: A Study of BRICS Nations using VECM and Granger Causality Test. Journal of Advances in Management Research, 13(2), 179-202. doi:10.1108/JAMR-04-2015-0028 Huang, Y., Panizza, U., & Varghese, R. (2018). Does Public Debt Crowd Out Corporate Investment? International Evidence. Working Paper No. HEIDWP08-2018. Retrieved March 21, 2018, from https://repository.graduateinstitute.ch/record/296011/files/HEIDWP08-2018.pdf Islamov, B. (2001, March). Central Asia: Problems of External Debt and Its Sustainability. Discussion Paper Series A No.413. The Institute of Economic Research, Hitotsubashi University. Retrieved December 11, 2018, from http://www2.ier.hit- u.ac.jp/Common/publication/DP/DP413.pdf Jarque, C. M. (2011). Jarque-Bera Test. (M. Lovric, Ed.) Berlin, Heidelberg: Springer. doi:10.1007/978-3-642-04898-2_319 Jian, J., Fan, X., He, P., Xiong, H., & Shen, H. (2019). The Effects of Energy Consumption, Economic Growth and Financial Development on CO2 Emissions in China: A VECM Approach. Sustainability, 11(4850), 1-16. doi:10.3390/su11184850 Karagoz, M., & Caglar, M. (2016). Does Debt Really Crack the Whip? Evidence from a Panel of Selected OECD Countries. Economics and Finance, 38, 430-437. Kharusi, S. A., & Ada, M. S. (2018). External Debt and Economic Growth: The Case of Emerging Economy. Journal of Economic Integration, 33(1), 1141-1157. doi:10.11130/jei.2018.33.1.1141 Kim, J. H., & Choi, I. (2017). Unit Roots in Economic and Financial Time Series: A Re-Evaluation at The Decision-Based Significance Levels. Econometries, 5(41), 1-23. doi:10.3390/econometrics5030041 Kiminyei, F. K. (2018). The Nexus between Tax Revenue and Government Expenditure in Kenya. International Journal of Economics & Management Sciences, 7(5), 1-5. doi:10.4172/2162-6359.1000546 https://econpapers.repec.org/article/eeeinecon/v_3a97_3ay_3a2015_3ai_3a1_3ap_3a45-58.htm https://econpapers.repec.org/article/eeeinecon/v_3a97_3ay_3a2015_3ai_3a1_3ap_3a45-58.htm http://www.ub.edu/irea/working_papers/2016/201610.pdf http://www.ub.edu/irea/working_papers/2016/201610.pdf https://www.amazon.com/Econometric-Analysis-8th-William-Greene/dp/0134461363 https://econpapers.repec.org/article/emejamrpp/v_3a13_3ay_3a2016_3ai_3a2_3ap_3a179-202.htm https://econpapers.repec.org/article/emejamrpp/v_3a13_3ay_3a2016_3ai_3a2_3ap_3a179-202.htm http://repec.graduateinstitute.ch/pdfs/Working_papers/HEIDWP08-2018.pdf http://repec.graduateinstitute.ch/pdfs/Working_papers/HEIDWP08-2018.pdf http://www2.ier.hit-u.ac.jp/Common/publication/DP/DP413.pdf http://www2.ier.hit-u.ac.jp/Common/publication/DP/DP413.pdf http://www2.ier.hit-/ https://www.mdpi.com/2071-1050/11/18/4850 https://www.mdpi.com/2071-1050/11/18/4850 https://www.mdpi.com/2071-1050/11/18/4850 https://www.sciencedirect.com/science/article/pii/S2212567116302143 https://www.sciencedirect.com/science/article/pii/S2212567116302143 https://www.e-jei.org/upload/JEI_33_1_1141_1157_2013600154.pdf https://www.e-jei.org/upload/JEI_33_1_1141_1157_2013600154.pdf https://www.mdpi.com/2225-1146/5/3/41 https://www.mdpi.com/2225-1146/5/3/41 https://www.researchgate.net/publication/330318004_The_Nexus_between_Tax_Revenue_and_Government_Expenditure_in_Kenya DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 23 Korkmaz, S. (2015). The Relationship Between External Debt and Economic Growth in Turkey. Proceedings of the Second European Academic Research Conference on Global Business, Economics, Finance and Banking (EAR15 Swiss Conference), (pp. 1-9). Kulaksizoglu, T. (2015). Lag Order and Critical Values of The Augmented Dickey-Fuller Test: A Replication. Journal of Applied Econometrics, 30(6), 1010. doi:10.1002/jae.2458. Kum, H., & Öktem, Ö. (2018). Government Debts and Economic Growth: The Case for Selected European Union Countries. International Journal of Economics and Financial Issues, 8(6), 192-196. doi:10.32479/ijefi.7399 Liang, C., & Schienle, M. (2019). Determination of Vector Error Correction Model in High Dimensions. Journal of Econometrics, 208(2), 418-441. doi:10.1016/j.jeconom.2018.09.018 Lutkepohl, H. (2017). Estimation of Structural Vector Autoregressive Models. Communications for Statistical Applications and Methods, 24(5), 421-441. doi:10.5351/CSAM.2017.24.5.421 MacKinnon, J. G. (2010). Critical Values for Cointegration Tests. Queen’s University, Department of Economics. Canada: Queen’s Economics Department Working Paper No. 1227. Retrieved January 6, 2019, from http://qed.econ.queensu.ca/working_papers/papers/qed_wp_1227.pdf Mantalos, P. (2010). Robust Critical Values for The Jarque-Bera Test for Normality. Jönköping University, Jönköping International Business School. Jönköping, Sweden: JIBS Working Papers. Retrieved November 14, 2018, from http://web.ipac.caltech.edu/staff/fmasci/home/astro_refs/NormalityTests.pdf Menegaki, A. N. (2019). The ARDL Method in The Energy-Growth Nexus Field: Best Implementation Strategies. Economies, 7(105), 1-16. doi:10.3390/economies7040105 Morana, C. (2012). PC-VAR Estimation of Vector Autoregressive Models. Open Journal of Statistics, 2, 251-259. doi:10.2139/ssrn.2006847 Muhdi, & Sasaki, K. (2009). Roles of External and Domestic Debt in Economy: Analysis of a Macroeconometric Model for Indonesia. Interdisciplinary Information Sciences, 15(2), 251-265. doi:10.4036/iis.2009.251 Muruganandan, S., Vijayakumar, N., & Dharani, M. (2017). Impact of US Stock Market and Oil Price on BRIC Stock Markets and Their Interdependency. Journal of Contemporary Research in Management, 12(4), 79-87. Niedzwiecki, M., & Ciolek, M. (2017). Akaike's Final Prediction Error Criterion Revisited. 40th International Conference on Telecommunication and Signal Processing (TSP) (pp. 1-6). Gdanks, Poland: Gdanks University of Technology. doi:10.1109/TSP.2017.8075977 https://www.researchgate.net/publication/306359022_The_Relationship_Between_External_Debt_and_Economic_Growth_in_Turkey https://onlinelibrary.wiley.com/doi/abs/10.1002/jae.2458 https://onlinelibrary.wiley.com/doi/abs/10.1002/jae.2458 https://www.econjournals.com/index.php/ijefi/article/view/7399 https://www.econjournals.com/index.php/ijefi/article/view/7399 https://www.econstor.eu/bitstream/10419/191548/1/1047196697.pdf https://www.econstor.eu/bitstream/10419/191548/1/1047196697.pdf http://qed.econ.queensu.ca/working_papers/papers/qed_wp_1227.pdf http://qed.econ.queensu.ca/working_papers/papers/qed_wp_1227.pdf http://qed.econ.queensu.ca/working_papers/papers/qed_wp_1227.pdf https://pdfs.semanticscholar.org/7691/fb98135bbcc867170a3605a3381c13527632.pdf http://web.ipac.caltech.edu/staff/fmasci/home/astro_refs/NormalityTests.pdf https://www.researchgate.net/publication/336655690_The_ARDL_Method_in_the_Energy-Growth_Nexus_Field_Best_Implementation_Strategies https://www.researchgate.net/publication/336655690_The_ARDL_Method_in_the_Energy-Growth_Nexus_Field_Best_Implementation_Strategies https://www.researchgate.net/publication/228163671_PC-VAR_Estimation_of_Vector_Autoregressive_Models https://www.jstage.jst.go.jp/article/iis/15/2/15_2_251/_pdf https://www.jstage.jst.go.jp/article/iis/15/2/15_2_251/_pdf https://www.researchgate.net/publication/307564889_Impact_of_US_Stock_Market_and_Oil_Price_on_BRIC_Stock_Markets_and_their_Interdependence https://www.researchgate.net/publication/307564889_Impact_of_US_Stock_Market_and_Oil_Price_on_BRIC_Stock_Markets_and_their_Interdependence https://www.researchgate.net/publication/320650901_Akaike%27s_final_prediction_error_criterion_revisited DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 24 Paparoditis, E., & Politis, D. N. (2018). The Asymptotic Size and Power of The Augmented Dickey- Fuller Test for a Unit Root. Econometric Reviews, 37(9), 955-973. doi:10.1080/00927872.2016.1178887 Patillo, C., Helene, P., & Luka, R. (2002). External Debt and Growth. IMF. IMF Working Paper 02/69. Washington: International Monetary Fund. Retrieved March 10, 2019, from https://www.imf.org/external/pubs/ft/wp/2002/wp0269.pdf Pečarić, M., Slišković, V., & Kusanović, T. (2018). Public Debt in New EU Member States - Panel Analysis and Managerial Implications. Journal of Contemporary Management Issues, 23(1), 81-97. doi:https://doi.org/10.30924/mjcmi/2018.23.2.81 Pegkas, P. (2018). The Effect of Government Debt and Other Determinants on Economic Growth: The Greek Experience. Economies, 6(10), 1-16. doi:10.3390/economies6010010 Picarelli, M. O., Vanlaer, W., & Mameffe, W. (2019). Does Public Debt Produce a Crowding Out Effect for Public Investment in The EU? Luxembourg: European Stability Mechanism Working Paper 36. doi:10.2139/ssrn.3376471 Prins, N., & Kingdom, F. A. (2016). Applying The Model-Comparison Approach to Test Spesific Research Hypotheses in Psychophysical Research Using The Palamedes Toolbox. Frontiers in Psychology, 9, 1-14. doi:10.3389/fpsyg.2018.01250 Puente-Ajovin, M., & Sanso-Navarro, M. (2015). Granger Causality between Debt and Growth: Evidence from OECD Countries. International Review of Economics and Finance, 35, 66- 77. Reed, W. R. (2015). Testing for Unit Roots with Cointegrated Data. Discussion Papers 2015-57, Economics, 1-20. Retrieved June 19, 2019, from http://www.economics- ejournal.org/economics/discussionpapers/2015-57 Reinhart, C. M., & Rogoff, K. S. (2010). Growth in a Time of Debt. American Economic Review, American Economic Association, 100(2), 573-578. doi:10.3386/w15639 Rothe, C., & Sibbertsen, P. (2006). Phillips-Perron-Type Unit Root Tests in The Nonlinear ESTAR Framework. Allgemeines Statistisches Archiv, 90(3), 439-456. doi:10.1007/s10182-006- 0244-y Sabbadini, R. (2018). Overcoming The Original Sin: Gains from Local Currency External Debt. Banco Central do Brasil. Brasilia: Working Paper 484. Retrieved June 12, 2019, from https://www.bcb.gov.br/pec/wps/ingl/wps484.pdf Salles, R., Belloze, K., Porto, F., Gonzales, P. H., & Ogasawara, E. (2019). Nonstationary Time Series Transformation Methods: An Experimental Review. Knowledge-Based System, 164, 274-291. doi:10.1016/j.knosys.2018.10.041 https://gnosis.library.ucy.ac.cy/handle/7/57436 https://gnosis.library.ucy.ac.cy/handle/7/57436 https://www.imf.org/external/pubs/ft/wp/2002/wp0269.pdf http://www.imf.org/external/pubs/ft/wp/2002/wp0269.pdf http://www.imf.org/external/pubs/ft/wp/2002/wp0269.pdf http://www.imf.org/external/pubs/ft/wp/2002/wp0269.pdf https://hrcak.srce.hr/202348 https://hrcak.srce.hr/202348 https://econpapers.repec.org/article/gamjecomi/v_3a6_3ay_3a2018_3ai_3a1_3ap_3a10-_3ad_3a130247.htm https://econpapers.repec.org/article/gamjecomi/v_3a6_3ay_3a2018_3ai_3a1_3ap_3a10-_3ad_3a130247.htm https://www.esm.europa.eu/publications/does-public-debt-produce-crowding-out-effect-public-investment-eu https://www.esm.europa.eu/publications/does-public-debt-produce-crowding-out-effect-public-investment-eu https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6064978/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6064978/ https://econpapers.repec.org/article/eeereveco/v_3a35_3ay_3a2015_3ai_3ac_3ap_3a66-77.htm https://econpapers.repec.org/article/eeereveco/v_3a35_3ay_3a2015_3ai_3ac_3ap_3a66-77.htm http://www.economics-ejournal.org/economics/discussionpapers/2015-57 https://scholar.harvard.edu/files/rogoff/files/growth_in_time_debt_aer.pdf https://link.springer.com/article/10.1007/s10182-006-0244-y https://link.springer.com/article/10.1007/s10182-006-0244-y https://www.bcb.gov.br/pec/wps/ingl/wps484.pdf https://www.bcb.gov.br/pec/wps/ingl/wps484.pdf http://www.bcb.gov.br/pec/wps/ingl/wps484.pdf http://www.bcb.gov.br/pec/wps/ingl/wps484.pdf https://www.sciencedirect.com/science/article/abs/pii/S0950705118305343 https://www.sciencedirect.com/science/article/abs/pii/S0950705118305343 DISENTANGLING THE ASSOCIATION BETWEEN GOVERNMENT DEBT AND ECONOMIC GROWTH: A GRANGER CAUSALITY APPROACH FROM INDONESIA JSP | Jurnal Studi Pemerintahan, Vol. 11, No. 1 / February 2020 | 25 Shah, M., & Pervin, S. (2012, September 1). External Public Debt and Economic Growth: Empirical Evidence from Bangladesh, 1974 to 2010. Academic Research International, 3(2), 508-515. doi:10.2139/ssrn.2029277 Shkolnyk, I., & Koilo, V. (2018). The Relationship between External Debt and Economic Growth: Empirical Evidence from Ukraine and Other Emerging Economies. Investment Management and Financial Innovations, 15(1), 387-400. doi:10.21511/imfi.15(1).2018.32 Spilioti, S., & Vamvoukas, G. (2015). The Impact of Government Debt on Economic Growth: An Empirical Investigation of the Greek Market. The Journal of Economic Asymmetries, 12(1), 34-40. Swastika, P., Dewandru, G., & Masih, M. (2013). The Impact of Debt on Economic Growth: A Case Study of Indonesia. University Library of Munich. Germany: MPRA Paper 58837. Retrieved October 24, 2019, from https://mpra.ub.uni-muenchen.de/id/eprint/58837 Thao, P. T. (2018). Impacts of Public Debt on Economic Growth in Six ASEAN Countries. Ritsumeikan Annual Review of International Studies, 17, 63-88. Thilanka, H., & Ranjith, J. S. (2018). The Impact of Public Debt on Private Investment: Sri Lankan Experience. International Journal of Business and Social Research, 8(8), 1-10. doi:http://dx.doi.org/10.18533/ijbsr.v8i8.1122 Todaro, M., & Smith, S. (2015). Economic Development. New York: Oxford University Press. Walle, Y. M. (2014, June). Revisiting the Finance-Growth Nexus in Sub-Saharan Africa: Results from Error Correction-Based Panel Cointegration Tests. African Development Review, 26(2), 310-321. Woo, J., & Kumar, M. (2015). Public Debt and Growth. Economica, 82(328), 705-739. doi:10.1111/ecca.12138 Zou, X. (2018). VECM Model Analysis of Carbon Emissions, GDP, and International Crude Oil Prices. Discrete Dynamics in Nature and Society, 1-18. doi:10.1155/2018/5350308 https://www.researchgate.net/publication/255724390_External_Public_Debt_and_Economic_Growth_Empirical_Evidence_from_Bangladesh_1974_to_2010 https://www.researchgate.net/publication/255724390_External_Public_Debt_and_Economic_Growth_Empirical_Evidence_from_Bangladesh_1974_to_2010 https://www.researchgate.net/publication/324263759_The_relationship_between_external_debt_and_economic_growth_Empirical_evidence_from_Ukraine_and_other_emerging_economies https://www.researchgate.net/publication/324263759_The_relationship_between_external_debt_and_economic_growth_Empirical_evidence_from_Ukraine_and_other_emerging_economies https://www.researchgate.net/publication/270961835_The_impact_of_government_debt_on_economic_growth_An_empirical_investigation_of_the_Greek_market https://www.researchgate.net/publication/270961835_The_impact_of_government_debt_on_economic_growth_An_empirical_investigation_of_the_Greek_market https://mpra.ub.uni-muenchen.de/58837/ https://mpra.ub.uni-muenchen.de/58837/ http://www.ritsumei.ac.jp/ir/isaru/assets/file/raris/raris-17-04_PHAM_Thi_Phuong_Thao.pdf https://thejournalofbusiness.org/index.php/site/article/view/1122 https://thejournalofbusiness.org/index.php/site/article/view/1122 http://dx.doi.org/10.18533/ijbsr.v8i8.1122 https://mediasrv.aua.gr/eclass/modules/document/file.php/AOA215/Economic%20Development%20-%20Todaro%20and%20Smith.pdf https://onlinelibrary.wiley.com/doi/abs/10.1111/1467-8268.12083 https://onlinelibrary.wiley.com/doi/abs/10.1111/1467-8268.12083 https://onlinelibrary.wiley.com/doi/abs/10.1111/ecca.12138 https://www.hindawi.com/journals/ddns/2018/5350308/ https://www.hindawi.com/journals/ddns/2018/5350308/