Finance-Growth Nexus: Evidence from Systematically Important Islamic Finance Countries

Transcription

1. Finance-Growth Nexus : Evidence from Systematically Important Islamic Finance Countries Edib Smolo Department of Economics and Management, International University of Sarajevo (IUS), Sarajevo, Bosnia and Herzegovina, Hrasnička cesta 15, 71210 Ilidža, Bosnia and Herzegovina, edib.smolo@gmail.com ABSTRACT This paper examines the relationship between bank concentration and economic growth in countries with systematically important Islamic finance. For this purpose, we use a number of estimation methods using a sample of 25 countries and covering the period of 2000-2019. Our analysis reveals that financial development has an insignificant impact on economic growth. These results are robust across different estimation specifications and methods used. JEL classification: C23, F43, G2, O1, O4, P52, Kew words: financial development, economic growth, Islamic finance, finance-growth nexus. 1.0 INTRODUCTION Although the well–functioning financial structure is, in general, a key to long–term sustainable economic growth and overall stability, the debate on the relationship between financial development and economic growth remains non–fading. The theoretical literature provides startlingly different and sometimes conflicting views on the finance – growth nexus. In brief, there are four major views that prevail in the literature with regards to the causal relationship between the overall financial development and economic growth, namely: (1) supply–leading – where financial development affects economic growth (Bittencourt, 2012; Goldsmith, 1969; Gurley & Shaw, 1955; McKinnon, 1973; Patrick, 1966; Porter, 1966; Schumpeter, 1912); (2) demand–following – where financial development follows economic growth (Al-Yousif, 2002; Robinson, 1952); (3) bidirectional causal relationships between finance and economic growth (Demetriades & Hussein, 1996; Greenwood & Smith, 1997); and (4) no relationship – no causal relationship between finance and economic growth (Lucas, 1988).1 Empirically, the importance of banking sector development as a catalyst for economic growth has been given predominant emphasis, in great part due to the fact that the banking sector forms a main component of the financial sector especially in developing countries. Banks provide a number of financial services needed for well–functioning of a financial system such as mobilization and allocation of financial resources, identification of investment opportunities, facilitating trading and diversification of risk, and information processing that may lead to a decrease in the transaction cost and asymmetric information problems as well as improvement in the corporate governance mechanisms, which are reasons underlying the supply–leading 1 Al–Yousif (2002) finds support for all these views. He used both time series and panel data and the results varied due to use of different variables. Similar findings are reported by Demetriades and Hussein (1996). A short summary is provided by Prochniak and Wasiak (2017). 1
2. view (Shumpeter, (1912); Gurley and Shaw, (1955); Goldsmith, (1969); and McKinnon, (1973). Although the view has received much support since the publication of King and Levine’s (1993) paper, the global financial crisis has brought in the fore the skepticism of at best the qualified view on the positive effect of banking sector development on economic growth. Such studies as Favara (2003), Rioja & Valev (2004), Shen & Lee (2006), Arcand, Berkes, & Panizza (2012) and Law & Singh (2014) suggest that there might be limits to benefits of finance. Given its relative resilience to the global financial crisis, the Islamic finance industry (IFI) attracted a lot of attention in last decade. Consequently, the global assets of the IFI grew to about US$ 2.5 trillion in 2018 and it is projected to grow to about $3.47 trillion by 2024 with a projected growth of 6%, a growth rate that is far higher than the conventional financial industry's annual growth rate (Deloitte, 2019). Nevertheless, the share of IFI in the global finance is insignificant (ICD-REFINITIV, 2018; IFSB, 2019). According to the Islamic Financial Services Board (IFSB), there are 13 systemically important Islamic finance jurisdictions2 although, in our study we are going to include more jurisdictions. However, despite overwhelming literature on finance–growth nexus in general, this issue has not been adequately addressed within the countries with a growing market share of IFI. To the best of our knowledge, we did not come across a study focusing on this sample countries that may offer unique characteristics. This motivated us to investigate this relationship within these countries. Besides, majority of studies are primarily focused on developed economies while ignoring developing ones. The sample countries with their Islamic finance exposure offer new avenue for investigation of this relationship. Furthermore, recent studies indicate no significant impact of financial development on economic growth as this finance-growth nexus is fading in recent years (Ali, Ibrahim, & Shah; Rousseau & Wachtel, 2011). We want to see whether the financial liberalization and introduction of Islamic finance practices in recent years added value to the overall economic growth via financial development improvement (if any). Having said that, the main objective of this paper is to remedy these issues through analysis of the existing literature and contribution to this growing area of research by exploring the impact of financial development on economic growth within countries with systematically important Islamic finance. Using a panel dataset of 25 countries and covering the period of 2000-2019, we will therefore test the following hypotheses: !0: There is a positive relationship between financial development and economic growth !1: There is a negative or no relation between financial development and economic growth The remainder of the paper is organized as follows: Section 2 provides a literature review; Section 3 describes the data and methodology used; Section 4 analyses empirical results; and Section 5 is left for concluding remarks. 2.0 LITERATURE REVIEW The interaction and relationship between financial development and economic growth has a 2 According to the report, “This report considers the Islamic financial sector as being systemically important when the total Islamic banking assets in a country comprise more than 15% of its total domestic banking sector assets. The report uses the Islamic banking segment as the criterion for systemic importance of Islamic finance, since about 76% of Islamic financial assets are held within the banking sector. A recognition of systemic importance is also considered for jurisdictions that are within one percentage point of the 15% benchmark, provided they have active involvement (among the top 10) in the other two sectors of the IFSI – Islamic capital markets and takāful.” These countries are Iran, Sudan, Brunei, Saudi Arabia, Kuwait, Malaysia, Qatar, UAE, Bangladesh, Djibouti, Jordan, Palestine and Bahrain (IFSB, 2019, pp. 10-11). 2
3. long history . A discussion on the topic was initiated by Schumpeter (1912) and further elaborated theoretically by Robinson (1952), Goldsmith (1969), Shaw (1973), and Lucas (1988). This led to a number of empirical studies that investigated this relationship and looked for empirical evidences in order to support original economic theories. Two papers by Levine (1997, 2003) provide extensive overviews of the finance–growth nexus literature. In the first paper, Levine (1997) found a strong positive link between the functioning of the financial system and long–run economic growth. In the second review paper, Levine (2003) found that countries with better–developed financial systems (i.e. the size of banking system and the liquidity of stock markets) tend to grow faster. Based on theoretical foundations laid down by Goldsmith (1969) and Shaw (1973), a number of authors found a positive contribution of financial development on economic growth and this is the most commonly accepted and confirmed view in the literature (Ahmed & Ansari, 1998; Beck, Degryse, & Kneer, 2014; Beck, Levine, & Loayza, 2000; Christopoulos & Tsionas, 2004; King & Levine, 1993; Levine, Loayza, & Beck, 2000; Rajan & Zingales, 1998; Seetanah, Ramessur, & Rojid, 2009). Not only that there is a positive relationship between financial development and economic growth but also financial depth and economic growth (Christopoulos & Tsionas, 2004; Guiso et al., 2004; Seetanah et al., 2009). Furthermore, studies also indicate that a well–developed legal system and accounting standards contribute to the development of financial intermediaries (Levine et al., 2000) while others point out to the significance of other factors such as investment, openness, education, macroeconomic stability and institutional development (Bittencourt, 2012; Seetanah et al., 2009). Although a large theoretical and empirical literature shows the importance and impact of financial development on economic growth, its benefits are not equally shared by all countries. In particular, region, income level and a type of economy affect the impact of financial deepening (Barajas, Chami, & Yousefi, 2013). Contrary to general view about the finance–growth relationship that can be find in the literature, there are a number of studies showing different results. For example, De Gregorio and Guidotti (1995), Fernandez and Galetovic (1994), Luitel and Khan (1999), Ram (1999), Andersen and Tarp (2003), Favara (2003), Naceur and Ghazouani (2007), Hsueh et al. (2013) and (Carré & L’œillet, 2018) show that the direction of causality between financial development and economic growth is sensitive to the financial development indicators, econometric techniques and sample period used.3 Thus, a negative relationship is found in number of studies (Andersen & Tarp, 2003; Khan, 2001; Luintel & Khan, 1999). In addition, according to some empirical results, the level of financial development is beneficial to growth only up to a certain threshold, beyond which it has negative effects suggesting that we need an “optimal” level of financial development for growth rather than simply “more” finance (Law & Singh, 2014; Prochniak & Wasiak, 2017). As if the current discussion did not create enough confusion, there are also studies that find more than one relationship in their samples (Halkos & Trigoni, 2010; Hassan, Sanchez, & Yu, 2011; Marques, Fuinhas, & Marques, 2013). 3 For example, the study by Hsueh et al. (2013) offer a number of interesting findings. First, their study shows that the direction of causality between financial development and economic growth among ten Asian countries surveyed during period 1980 to 2007 is sensitive to the financial development variables used. Consequently, a one–way Granger causality from financial development to economic growth is found in case of Malaysia, Indonesia, Korea, Singapore, Thailand, Taiwan and China, thus supporting the supply–leading hypothesis. On contrary, when the M1 variable is used the study shows one–way Granger causality from economic growth to financial development in case of Malaysia. However, no Granger causality is found in case of the Philippines, India and Japan. 3
4. 3 .0 DATA, MODEL AND METHODOLOGY This study employs panel data techniques using annual time series data of the selected macroeconomic indicators of 25 countries with systematically important Islamic finance industry. The countries covered are Afghanistan, Bahrain, Bangladesh, Bosnia and Herzegovina, Brunei Darussalam, Djibouti, Egypt, Indonesia, Iran, Iraq, Jordan, Kuwait, Malaysia, Maldives, Oman, Pakistan, Qatar, Saudi Arabia, Senegal, Sudan, Tunisia, Turkey, United Arab Emirates, West Bank and Gaza (Palestine) and Yemen.4 All data are sourced from the World Bank’s database and cover the period of 2000-2019, making it a total of 20 years. As we are covering 20-year period and 25 countries (crosssectional units), there should be 500 observation in total (20 x 25). However, as a number of countries reported no data the total number of observations for each country is not the same. Hence, we are dealing with an unbalanced panel. Following the existing literature, this study uses the real per capita GDP growth (GDP) as a measure for the economic growth. This is a standard indicator of the economic growth used in the literature. The literature, however, does not offer a consensus concerning a financial development measure. As pointed out briefly in the introduction, the results differ significantly due to indicator(s) used as proxy for financial development. Consequently, many authors use several proxies for the financial development measure or opt for a sort of index based on a number of proxies (Carré & L’œillet, 2018). Some authors use different measures for robustness results, while others are forced to use certain measures due to availability of data especially in panel data analysis covering countries with different levels of financial development (for instance, markets dominated by banks as opposed to those with a welldeveloped stock markets) (Valickova, Havranek, & Horvath, 2015). In line with this brief explanation and the existing literature, we use three indicators of financial development: (i) a ratio of credit to the private sector provided by financial intermediaries as a percentage of GDP (PR). This measures the efficiency of funds channeling to the private sector (Al-Malkawi & Abdullah, 2011; Levine, 1997); (ii) a ratio of liquid liabilities to GDP (LL) that measures the financial sector size and depth. It also represents banks’ ability to mobilize funds (Compton & Giedeman, 2011; King & Levine, 1993; Law & Singh, 2014; Levine, 1997); (iii) a ratio of broad money supply (M3) to GDP that measures financial sector size. The use of M3 is seen more appropriate for countries in which money is primarily used as store of value (Hassan et al., 2011; Valickova et al., 2015).5 As for control variables, we use a number of macroeconomic variables to control their impacts on economic growth as the literature point out to their significance in determining economic growth. In particular, we use gross capital formation (GCF), trade openness (TO), human capital (HC), bank cost to income ratio (BCI), inflation rate (I), and the financial crisis dummy (C). The GCF is a control variable that reflects the overall economic development of a country (Levine & Renelt, 1992). The TO, measured as the sum of exports and imports of goods and services as a share of GDP, represents the current economic activities of a country (Beck et al., 4 Although, the IFSB report indicates 12+1 countries (Iran, Sudan, Brunei, Saudi Arabia, Kuwait, Malaysia, Qatar, UAE, Bangladesh, Djibouti, Jordan and Palestine + Bahrain) that are withing 15% share in the total banking assets in respective countries, we included the other countries from the report that have more than 3% share in the total banking assets for a sake of this study. We will, however, run robustness tests on the original 13 countries later on. 5 Due to non-availability of sufficient data on stock market development in majority of countries we did not include it as an indicator of financial development in above model. 4
5. 2014 ; Bist, 2018). The HC, proxied by the total labor force comprising of people ages 15 and older, measures the human capital accumulation and it is one of the pillars of the economic development theory. The BCI measures overhead costs relative to gross revenues with higher ratios indicating lower levels of cost efficiency. It is argued that bank efficiency and its stability promote economic growth (Beck, Demirgüç-Kunt, & Merrouche, 2013). The financial crisis dummy (C) is used as an indicator of macroeconomic development. It takes the value of one for the year 2008 and 2009 and zero otherwise to capture the effect of the global financial crisis on economic growth. During a financial crisis, banks are faced with a number of challenges that make them fragile. This brings about uncertainty in the market and increases overall risk. Likewise, inflation is used as a proxy for monetary (in)stability. It affects not only growth but also financial activities within a country through its direct impact on the interest rates. Countries with high inflation tend to have financial systems that are generally underdeveloped and prone to financial crises (Bist, 2018). A summary of these variables and their expected impact on our dependent variable is presented in Table 1 below. Table 1: Summary of All Variables VARIABLES SYM DEFINITION DEPENDENT VARIABLE Economic Growth GDP The real per capita GDP growth rate (annual %). INDEPENDENT VARIABLE(S) Private credit PR A ratio of private credit by deposit money banks and other financial institutions to GDP. Liquid liabilities LL A ratio of liquid liabilities to GDP. Broad money M3 This is the sum of currency outside banks. CONTROL VARIABLES Trade openness TO The sum of exports and imports of goods and services measured as a share of GDP. Gross capital GCF The net increase in physical assets (investment formation minus disposals) within the measurement period and it can be measured as a ratio of GDP. a b c d e Bank cost to income BCI Human capital Inflation HC I Financial crisis C It measures overhead costs relative to gross revenues. Labor force, total (people ages 15 and older) Inflation adjusted by the GDP deflator. A dummy variable to capture the effect of the global financial crisis 2008-2009. SIGN SOURCE WDIa +|- IFSb +|+|- IFS IFS + WDI + WDI - GFDc + - ILOd IMFe - GFD The World Development Indicators (WDI). The World Bank. International Financial Statistics (IFS), International Monetary Fund (IMF) The Global Financial Development (GFD). The World Bank. International Labour Organization (ILO), ILOSTAT database International Monetary Fund, International Financial Statistics and data files using World Bank data on the GDP deflator. To assess the impact of the financial development on economic growth in systematically important Islamic finance countries we employ the following panel data model: "#$!" = &'#!" + )*!" + +! + ," + -!" Eq. 1 Where GDPit is the real per capita GDP of country i at time t, and where i denotes the cross– sectional dimension (i.e. country) and t denotes the time dimension (i.e. year); FDit represents a measure of financial development of country i at time t as measured by one of the financial development indicators (PR, LL, or M3); Xit is a vector of all control variables (TO, GCF, BCI, 5
6. HC , I, and C); µi is a country-specific effect; ηt is a time-specific effect; and εit is a random error term that captures all other variables. Due to the presence of non-linearity in growth equations, and in line with prevailing literature, we transfer all variables (except financial crisis) in logarithmic form (Naceur, Blotevogel, Fischer, & Shi, 2017). The literature is overwhelmed with various estimation methods that have been used in analysing the finance–growth nexus. Different methods, as well as different indicators and samples, led to different conclusions. In this particular case, we are going to apply a number of estimation methods. First, following the literature (Seetanah et al., 2009; Shen & Lee, 2006) and due to the availability of data, the study will apply the fixed effects (FE) and the random effects (RE) models. From Equation 1 above, the FE model refers to the case when µi and ηt are treated as fixed parameters. On the contrary, when they are treated as random variables with zero means and constant variance, then we are referring to the RE model. Furthermore, to choose which model, the FE or the RE estimator, fits the data better, we use the Hausman test which assumes that there is no difference between the two models’ estimates (H0). Rejecting the null hypothesis is rejected, the FE estimator is the appropriate one, otherwise we should go for the RE estimator. Second, as number of studies indicated biasness in the FE and RE estimation methods (Nickell, 1981), we also provide estimation based on the Least Squares Dummy Variables estimator (LSDV). Nevertheless, Kiviet (1995), Bruno (2005a, 2005b) and Bun and Carree (2006) showed that LSDV and GMM panel data estimators suffer from inconsistency and finitesample biases. Hence, we turn to the bias-corrected LSDV estimator (LSDVC) initialized by Anderson and Hsiao (1982), as our preferred estimation strategy, where it uses the second lag of dependent variable (in differenced or level form) as an instrument. Finally, the generalized methods of moments (GMM) method will be used for robustness results. 4.0 4.1 EMPIRICAL RESULTS AND DISCUSSION Descriptive Analysis: An Overview First of all, it is important to note here that due to a potentially non-linear relationship between economic growth and control variables, and in line with prevailing literature, we transform all control variables (except crisis) into natural logarithm forms. Hence, we will use these variables in natural logarithm forms throughout the study (Naceur et al., 2017). Table 2 presents the descriptive statistics of the variables used in this study. The average economic growth of the sample countries is positive and relatively stable. Private credit, as a proxy of financial development, has a relatively higher variations as compared to the other two proxies, namely liquid liability and broad money. Table 2: Summary of Descriptive Statistics Variable GDP per capita Private credit Liquid liabilities Broad money Trade openness Gross capital formation Bank cost to income Human capital Inflation Crisis (dummy) Obs 457 457 412 454 432 417 396 457 457 457 Mean 1.881 41.123 54.392 59.414 86.803 24.356 47.825 16114739 7.124 .105 Std. Dev. 4.872 26.533 27.515 27.992 46.991 7.046 14.83 26755709 10.25 .307 6 Min -15.397 1.266 8.968 11 19.101 2.855 22.683 90256 -26.1 0 Max 50.236 127.232 135.12 140.092 304.329 48.869 139.468 1.348e+08 52.924 1
7. Table 3 : The impact of financial development on economic growth: Private credit VARIABLES Private credit Trade openness Gross capital formation Bank cost to income Human capital Inflation (1) POLS (2) POLS (3) RE (4) RE (5) FE (6) FE (7) LSDV (8) LSDV (9) LSDVC (10) LSDVC -0.061** (0.024) 0.058 (0.046) -0.001 (0.056) 0.129** (0.059) 0.035*** (0.012) 0.065* (0.034) -0.062*** (0.024) 0.060 (0.045) 0.004 (0.056) 0.122** (0.059) 0.036*** (0.012) 0.053 (0.034) -0.156*** (0.050) -0.075** (0.033) 0.078 (0.063) -0.014 (0.065) 0.024 (0.078) 0.032* (0.020) 0.062* (0.034) -0.079** (0.033) 0.087 (0.063) -0.006 (0.064) -0.001 (0.078) 0.034* (0.020) 0.047 (0.033) -0.163*** (0.047) -0.021 (0.059) 0.222* (0.115) -0.039 (0.073) -0.159 (0.105) -0.166 (0.102) -0.010 (0.042) -0.021 (0.059) 0.222* (0.115) -0.039 (0.073) -0.159 (0.105) -0.166 (0.102) -0.010 (0.042) -0.338*** (0.103) -0.053** (0.024) 0.046 (0.045) -0.019 (0.056) 0.151** (0.059) 0.038*** (0.012) -0.006 (0.041) -0.053** (0.024) 0.046 (0.045) -0.019 (0.056) 0.151** (0.059) 0.038*** (0.012) -0.006 (0.041) -0.354*** (0.104) 0.039 (0.069) 0.124 (0.112) -0.022 (0.096) -0.162 (0.134) -0.102 (0.089) -0.019 (0.057) 0.279*** (0.056) 0.039 (0.069) 0.124 (0.112) -0.022 (0.096) -0.162 (0.134) -0.102 (0.089) -0.019 (0.057) -0.284* (0.148) 0.279*** (0.056) YES 348 YES 348 23 23 Crisis (dummy) GDP per capita (t-1) Constant 1.553*** (0.480) NO 367 0.089 23 5.875*** 1.613*** (0.475) NO 367 0.114 23 6.569*** 2.016*** 2.108*** (0.635) (0.636) NO NO 367 367 0.309 0.267 23 23 13.945** 25.660*** 12.984*** 14.998*** Year dummies Observations R-squared No of countries Significance of model Breusch-Pagan LM test Hausman test Model degrees of freedom 6 7 6 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.10 7 5.339*** (1.564) YES 367 0.320 23 2.303*** 5.339*** (1.564) YES 367 0.320 23 2.303*** 1.835*** (0.487) YES 367 0.177 23 3.202*** 1.835*** (0.487) YES 367 0.177 23 3.202*** 20.131*** 6 16.172** 7 23 23 7
8. Table 4 : The impact of financial development on economic growth: Liquid liabilities VARIABLES Liquid liabilities Trade openness Gross capital formation Bank cost to income Human capital Inflation Crisis (dummy) (1) POLS (2) POLS (3) RE (4) RE (5) FE (6) FE (7) LSDV (8) LSDV (9) LSDVC (10) LSDVC -0.033 (0.040) 0.017 (0.049) -0.021 (0.055) 0.142** (0.059) 0.030** (0.012) 0.062* (0.034) -0.043 (0.040) 0.023 (0.048) -0.018 (0.054) 0.134** (0.058) 0.031** (0.012) 0.048 (0.034) -0.154*** (0.049) -0.090 (0.057) 0.054 (0.067) -0.037 (0.063) 0.024 (0.077) 0.027 (0.020) 0.054 (0.034) -0.113* (0.058) 0.074 (0.067) -0.030 (0.062) -0.002 (0.077) 0.029 (0.020) 0.037 (0.034) -0.168*** (0.047) -0.038 (0.102) 0.138 (0.102) -0.004 (0.072) -0.173* (0.103) -0.085 (0.074) 0.045 (0.036) -0.069 (0.100) 0.183* (0.100) 0.007 (0.070) -0.206** (0.101) -0.089 (0.072) 0.024 (0.035) -0.185*** (0.046) -0.023 (0.041) 0.005 (0.049) -0.038 (0.055) 0.167*** (0.059) 0.033*** (0.012) -0.009 (0.041) -0.023 (0.041) 0.005 (0.049) -0.038 (0.055) 0.167*** (0.059) 0.033*** (0.012) -0.009 (0.041) -0.065 (0.099) -0.042 (0.153) 0.174** (0.081) -0.003 (0.097) -0.145 (0.108) -0.101 (0.138) -0.017 (0.036) 0.291*** (0.071) -0.042 (0.153) 0.174** (0.081) -0.003 (0.097) -0.145 (0.108) -0.101 (0.138) -0.017 (0.036) -0.273*** (0.086) 0.291*** (0.071) YES 344 YES 344 23 23 GDP per capita (t-1) Constant 1.754*** 1.831*** 2.385*** 2.513*** 4.239*** 4.421*** 1.982*** 1.982*** (0.482) (0.477) (0.641) (0.641) (1.157) (1.132) (0.488) (0.488) Year dummies NO NO NO NO YES YES YES YES Observations 361 361 361 361 361 361 361 361 R-squared 0.074 0.099 0.161 0.113 0.234 0.269 0.159 0.159 No of countries 23 23 23 23 23 23 23 23 Significance of model 4.718*** 5.547*** 9.610 22.433*** 1.797* 3.874*** 2.778*** 2.778*** Breusch-Pagan LM test 13.435*** 16.425*** Hausman test 10.396 13.670* Model degrees of freedom 6 7 6 7 6 7 23 23 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.10 8
9. Table 5 : The impact of financial development on economic growth: Broad money VARIABLES Broad money Trade openness Gross capital formation Bank cost to income Human capital Inflation Crisis (dummy) (1) POLS (2) POLS (3) RE (4) RE (5) FE (6) FE (7) LSDV (8) LSDV (9) LSDVC (10) LSDVC -0.050 (0.040) 0.037 (0.048) -0.033 (0.056) 0.142** (0.060) 0.034*** (0.013) 0.065* (0.035) -0.056 (0.039) 0.041 (0.047) -0.029 (0.055) 0.135** (0.059) 0.035*** (0.012) 0.052 (0.035) -0.156*** (0.050) -0.107* (0.058) 0.082 (0.066) -0.039 (0.063) 0.037 (0.078) 0.034* (0.020) 0.059* (0.034) -0.123** (0.059) 0.098 (0.067) -0.030 (0.063) 0.011 (0.078) 0.036* (0.021) 0.043 (0.034) -0.167*** (0.047) -0.068 (0.100) 0.198* (0.102) -0.002 (0.072) -0.144 (0.103) -0.078 (0.069) 0.049 (0.036) -0.083 (0.098) 0.241** (0.100) 0.008 (0.071) -0.183* (0.102) -0.085 (0.068) 0.029 (0.036) -0.186*** (0.047) -0.032 (0.040) 0.021 (0.047) -0.048 (0.055) 0.167*** (0.059) 0.036*** (0.012) -0.006 (0.041) -0.032 (0.040) 0.021 (0.047) -0.048 (0.055) 0.167*** (0.059) 0.036*** (0.012) -0.006 (0.041) -0.346*** (0.105) -0.001 (0.169) 0.150 (0.122) -0.016 (0.104) -0.160 (0.143) -0.102 (0.098) -0.022 (0.056) 0.278*** (0.055) -0.001 (0.169) 0.150 (0.122) -0.016 (0.104) -0.160 (0.143) -0.102 (0.098) -0.022 (0.056) -0.273** (0.121) 0.278*** (0.055) YES 348 YES 348 23 23 GDP per capita (t-1) Constant 1.690*** 1.761*** 2.168*** 2.282*** 3.861*** 4.052*** 1.929*** 1.929*** (0.488) (0.483) (0.645) (0.647) (1.122) (1.099) (0.493) (0.493) Year dummies NO NO NO NO YES YES YES YES Observations 367 367 367 367 367 367 367 367 R-squared 0.077 0.101 0.210 0.155 0.237 0.271 0.166 0.166 No of countries 23 23 23 23 23 23 23 23 Significance of model 5.018*** 5.791*** 11.893* 24.223*** 2.114* 4.125*** 2.975*** 2.975*** Breusch-Pagan LM test 15.732*** 18.458*** Hausman test 10.047 10.047 Model degrees of freedom 6 7 6 7 6 7 23 23 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.10 9
10. 4 .2 Results and Discussion Table 3, Table 4, and Table 5 present the estimated results of Eq(1) using private credit, liquid liabilities and broad money as proxies for the financial development measure, respectively. In each table, the pooled ordinary least square (POLS), the rendom effect (RE), the fixed effect (FE), the least square dummy variable (LSDV) and the bias-corrected least square dummy variable (LSDVC) methods were used to estimate the results. Each estimator consists of two estimations. The first column is considering the main independent variable and all control variables except the crisis dummy. The second collumn is including the crisis dummy to see the effect of the recent global financial crisis (GFC) 2008-2009 on economic growth. Given the fact that the sample is covering the countries with a significant share of the Islamic finance industry (IFI) and that it has been argued that the IFI showed a degree of resilience to the GFC, it would be interesting to see how these countries performed during these trouble years. Ever since the GFC, this topic has been scrutinised by researchers, and the results are far from being conclusive (Ibrahim, 2015). In general, based on the Breusch-Pagan Lagrange multiplier (LM) test, it is evident that RE model is to be preferred to POLS as there is evidence of significant differences across countries. To determine whether to use FE or RE, we run the Hausman test and its results depend on the proxy used for the financial development measure. FE is preferred in case of private credit only while RE is preferred in case of liquid liabilities and borad money at 5% significance level. Finally, given the overall superiority of the LSDVC estimator, we will base our interpration primarily on these results, although for a comparative purposes we may refer to other estimators as well. In particular, Table 3 shows conflicting results. Our main focus variable, financial development as measured by private credit, is decreasing economic growth when POLS, RE, FE and LSDV estimators are used, although it is insignificant under FE model. In contrast, it is insignificant with a positive sign in case of LSDVC. These results remain consistant even after adding the financial crisis dummy variable. The results from Table 4 and Table 5 where liquid liabilities and broad money are used as proxies for the financial development measure are more consistent. In both cases, results show that majority of our main variables are not significantly associated with economic growth in the sample countries. As a reason for insignificance of financial development on economic growth, Rousseau and Wachtel (2011) mentioned the overall weakening of the finance-growth nexus. It seems that financial liberalization taking place in developing countries (to which majority of our sample countries belong) is not accompanied with the required expertise to make this process smooth. Consequently, this untackled issue is causing instabilities in the market making financial development less effective in generating desired growth. Another reason could be attributed to the low level of financial development in the sample countries as the avearge is 41.13%. In short, based on the estimation results from all three tables, it is evident that financial development – proxied by private credit, liquid liabilities and broad money – has no significant impact on economic growth in sample countries with systematically important share of Islamic finance indutry. Significantly negative impact of financial development on economic growth that is reported under POLS, RE and LSDV estimators in Table 3 and RE estimator in Table 4 and Table 5 need to be taken with a great caution as these estimatiors are produce biased and inconsistent results as mentioned earlier. Hence, we rely on the results provided by the LSDVC estimators which show insignificant effect of financial development on economic growth. The lagged dependent variable, is the only variable that is significant in all three tables across all the specifications. The coefficient has a relatively stable impact on economic growth 10
11. ranging from 0278 in case of broad money to 0 .291 in case of liquid liabilities. Similarly, the trade opennes is significantly contributing to the economic growth only in Table 4 where liquid liability proxy is used. This is in line with findings reported by Rajan and Zingales (1998) and Law and Singh (2014). However, when it comes to other control variables, it seems that majority of them have no significant impact on economic growth under the LSDVC estimator.6 As for crisis dummy, its results are consistent under all estimators and it shows that the crisis decreases economic growth significantly in the case of our sample countries. This may be caused by a fact that these economies opened their financial markets and underwent a great deal of financial liberalization to cater for different and growing needs of its customers. This led to the development of Islamic finance products and services without a proper supervision and regulation. This financial liberalization and opennes led to a greater internationalization of their markets that made them vulnerable to financial crises. At the same time, the inherent stability of IFI may not realized yet and the current market share of IFI in these countries may not offer diversification and stability benefits as it is theoretically argued. This is in line with study by Shahzad, Ferrer, Ballester, and Umar (2017) who reject the decoupling hypothesis as the Islamic stock market offer no diversification benefits to investors. 4.3 Robustness Tests As for robustness tests, we opt for the GMM estimators. Table 6 provide estimation results for difference and system GMM estimators using private creadit, liquid liabilities and broad money as the main focus variables. Our earlier results, using LSDVC estimator as the most suitable method for the estimation in this case, are confirmed using the GMM estimators. All our focus variables are found to have negative, yet insignificant effect on economic growth using system GMM which is considered superior as compared to difference GMM estimators under which these variables are found to be significant. Lagged dependent and crisis variables are found to have significant effect on economic growth – positive and negative respectively. All other control variables are in line with previously reported results. All in all, using a number of esimation methods and in particular the LSDVC estimator that is found to be the superior for the dynamic panel data such is the current one, this study find no evidence of significant effect of financial development on economic growth. These results are supported by using three proxies for financial development – private credit, liquid liability and broad money – that lead to the same results. In addition, robustness tests using GMM estimators are aslo in conformity with the main results. It seems that finance-growth nexsus link is fading as pointed out by Rousseau and Wachtel (2011). In addition, given the rapid growth of financial industry in the sample countries and recent growth of Islamic finance industry, its services and products coupled with financial deregulation that has not been sufficiently addressed, the effects of financial development may have been diluted . At the same time, we may want to explore new proxies for financial development that may explain the finance-growth nexus in a better and more efficient way. 6 The results are pretty much same with all other estimators as well in most cases. 11
12. Table 6 : The impact of financial development on economic growth: Robustness tests VARIABLES GDP per capita (t-1) Private credit (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Diff GMM Diff GMM Sys GMM Sys GMM Diff GMM Diff GMM Sys GMM Sys GMM Diff GMM Diff GMM Sys GMM Sys GMM 0.185 (0.182) -0.272* (0.142) 0.203 0.454*** 0.577*** (0.182) (0.155) (0.162) -0.295** -0.024 -0.008 (0.142) (0.019) (0.018) Liquid liabilities 0.361* (0.195) 0.224* (0.132) -0.622*** -0.678*** (0.206) (0.186) 0.469*** 0.560*** (0.172) (0.149) -0.018 (0.038) Broad money Trade openness -0.143 (0.171) 0.185 (0.152) 0.060 (0.151) -0.014 (0.147) 0.048 (0.037) -0.043 -0.005 0.036 (0.172) (0.047) (0.079) Gross capital formation 0.210 0.129** 0.105 (0.152) (0.062) (0.075) Bank cost to income 0.001 0.024 0.159 (0.151) (0.079) (0.100) Human capital -0.009 0.015 0.017 (0.147) (0.017) (0.020) Inflation 0.039 0.046 0.062 (0.037) (0.029) (0.044) Crisis (dummy) -0.198*** -0.163*** (0.048) (0.060) Constant 0.774** -0.316 (0.378) (1.314) Observations 324 324 348 348 No. of instruments 21 22 23 24 No. of groups 23 23 23 23 AR1 p-value 0.037 0.037 0.037 0.037 AR2 p-value 0.254 0.254 0.254 0.254 Hansen p-value 0.415 0.415 0.415 0.415 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.10 -0.264 (0.195) 0.232 (0.168) 0.148 (0.171) 0.140 (0.146) 0.031 (0.042) 320 21 23 0.037 0.254 0.415 12 -0.174 (0.178) 0.246 (0.151) 0.048 (0.152) 0.089 (0.122) 0.013 (0.039) -0.221*** (0.048) 320 23 23 0.037 0.254 0.415 -0.012 (0.059) 0.106* (0.056) 0.043 (0.075) 0.013 (0.016) 0.050* (0.030) 0.768 (0.471) 344 23 23 0.037 0.254 0.415 0.353* (0.194) 0.114 (0.107) 0.455*** 0.551*** (0.165) (0.160) -0.027 (0.043) -0.757*** -0.733*** (0.204) (0.164) 0.022 -0.296 -0.190 (0.047) (0.189) (0.162) 0.066 0.260 0.230* (0.049) (0.162) (0.135) 0.116* 0.109 -0.004 (0.069) (0.164) (0.139) 0.009 0.177 0.069 (0.011) (0.144) (0.104) 0.059 0.025 0.011 (0.046) (0.041) (0.035) -0.161*** -0.209*** (0.061) (0.044) 0.290 (0.532) 344 324 324 24 21 23 23 23 23 0.037 0.037 0.037 0.254 0.254 0.254 0.415 0.415 0.415 -0.014 (0.034) -0.020 (0.054) 0.106* (0.057) 0.039 (0.078) 0.013 (0.017) 0.049* (0.028) 0.849* (0.457) 348 23 23 0.037 0.254 0.415 -0.007 (0.043) 0.005 (0.087) 0.073 (0.095) 0.126 (0.138) 0.009 (0.023) 0.057 (0.047) -0.162** (0.063) 0.247 (1.798) 348 24 23 0.037 0.254 0.415
13. 5 .0 CONCLUSION The present study was designed to determine the effect of financial development on economic growth in countries with systematically important Islamic finance. Our findings suggest that financial development plays insignificant role on economic growth of the sample countries. These results, although counterintuitive, are consistent across different specifications and estimators. Based on our findings, financial development – at least as understood from the literature so far – may not be a significant factor in determining economic growth of a country. As such, there is a need to find better measures for financial development as the most commonly used ones (private credit, liquid liabilities and broad money) are shown to be insignificant. In addition, there might be other factors more important for economic growth than financial development that need to be properly addressed. In line with this, it would be interesting to investigate the impact of different aspects of financial development on economic growth. Besides, using updated dataset with additional or new set of variables may shed extra light on the topic and provide additional recommendations. REFERENCES Ahmed, S. M., & Ansari, M. I. (1998). Financial Sector Development and Economic Growth: The South-Asian Experience. Journal of Asian Economics, 9(3), 503-517. Al-Malkawi, H.-A., & Abdullah, N. (2011). Finance-Growth Nexus: Evidence from a Panel of MENA Countries. International Research Journal of Finance and Economics, 63, 129-139. Al-Yousif, Y. K. (2002). Financial development and economic growth: Another look at the evidence from developing countries. Review of Financial Economics, 11(2), 131-150. doi:https://doi.org/10.1016/S1058-3300(02)00039-3 Ali, M., Ibrahim, M. H., & Shah, M. E. Impact of non-intermediation activities of banks on Economic Growth and Volatility: An Evidence from OIC. The Singapore Economic Review, 0(0), 1-16. doi:10.1142/s0217590820420023 Andersen, T. B., & Tarp, F. (2003). Financial liberalization, financial development and economic growth in LDCs. Journal of international development, 15(2), 189-209. doi:10.1002/jid.971 Anderson, T. W., & Hsiao, C. (1982). Formulation and estimation of dynamic models using panel data. Journal of Econometrics, 18(1), 47-82. doi:https://doi.org/10.1016/03044076(82)90095-1 Arcand, J. L., Berkes, E., & Panizza, U. (2012). Too Much Finance? IMF Working Paper, WP/12/161. Barajas, A., Chami, R., & Yousefi, S. R. (2013). "The Finance and Growth Nexus ReExamined: Do All Countries Benefit Equally?," IMF Working Paper No. 13/130. Retrieved from Washington: Beck, T., Degryse, H., & Kneer, C. (2014). Is more finance better? Disentangling intermediation and size effects of financial systems. Journal of Financial Stability, 10, 50– 64. Beck, T., Demirgüç-Kunt, A., & Merrouche, O. (2013). Islamic vs. conventional banking: Business model, efficiency and stability. Journal of Banking & Finance, 37(2), 433-447. doi:https://doi.org/10.1016/j.jbankfin.2012.09.016 Beck, T., Levine, R., & Loayza, N. (2000). Finance and the Sources of Growth. Journal of Financial Economics, 58(1-2), 261-300. Bist, J. P. (2018). Financial development and economic growth: Evidence from a panel of 16 African and non-African low-income countries. Cogent Economics & Finance, 6(1), 1449780. doi:10.1080/23322039.2018.1449780 13
14. Bittencourt , M. (2012). Financial development and economic growth in Latin America: Is Schumpeter right? Journal of Policy Modeling, 34, 341–355. Bruno, G. S. F. (2005a). Approximating the bias of the LSDV estimator for dynamic unbalanced panel data models. Economics Letters, 87(3), 361-366. doi:https://doi.org/10.1016/j.econlet.2005.01.005 Bruno, G. S. F. (2005b). Estimation and Inference in Dynamic Unbalanced Panel-data Models with a Small Number of Individuals. The Stata Journal, 5(4), 473-500. doi:10.1177/1536867x0500500401 Bun, M. J. G., & Carree, M. A. (2006). Bias-corrected estimation in dynamic panel data models with heteroscedasticity. Economics Letters, 92(2), 220-227. doi:https://doi.org/10.1016/j.econlet.2006.02.008 Carré, E., & L’œillet, G. (2018). The Literature on the Finance–Growth Nexus in the Aftermath of the Financial Crisis: A Review. Comparative Economic Studies, 60(1), 161-180. doi:10.1057/s41294-018-0056-6 Christopoulos, D. K., & Tsionas, E. G. (2004). Financial development and economic growth: evidence from panel unit root and cointegration tests. Journal of Development Economics, 73(1), 55– 74. Compton, R. A., & Giedeman, D. C. (2011). Panel evidence on finance, institutions and economic growth. Applied Economics, 43(25), 3523-3547. doi:https://doi.org/10.1080/00036841003670713 De Gregorio, J., & Guidotti, P. E. (1995). Financial development and economic growth. World Development, 23(3), 433-448. doi:https://doi.org/10.1016/0305-750X(94)00132-I Deloitte. (2019). Sharia Economy: Boosting Financial Inclusion through Sharia Economy in Indonesia. Jakarta: PT Deloitte Consulting. Demetriades, P. O., & Hussein, K. A. (1996). Does financial development cause economic growth? Time series evidence from 16 countries. Journal of Development Economics, 51(2), 387-411. Favara, G. (2003). An Empirical Reassessment of the Relationship Between Finance and Growth. IMF Working Paper, WP/03/123. Fernandez, D., & Galetovic, A. (1994). Schumpeter might be right-but why? Explaining the relation between finance, development, and growth. Johns Hopkins University SAIS Working Paper in International Economics. Goldsmith, R. W. (1969). Financial Structure and Development. New Haven, CT: Yale University Press. Greenwood, J., & Smith, B. D. (1997). Financial markets in development, and the development of financial markets. Journal of Economic Dynamics and Control, 21(1), 145-181. doi:https://doi.org/10.1016/0165-1889(95)00928-0 Guiso, L., Jappelli, T., Padula, M., Pagano, M., Martin, P., & Gourinchas, P.-O. (2004). Financial Market Integration and Economic Growth in the EU. Economic Policy, 19(40), 525-577. Gurley, J. G., & Shaw, E. S. (1955). Financial Aspects of Economic Development. American Economic Review, 45(4), 515-538. Halkos, G. E., & Trigoni, M. K. (2010). Financial development and economic growth: evidence from the European Union. Managerial Finance, 36(11), 949-957. Hassan, M. K., Sanchez, B., & Yu, J.-S. (2011). Financial development and economic growth: New evidence from panel data. The Quarterly Review of Economics and Finance, 51(1), 88104. doi:https://doi.org/10.1016/j.qref.2010.09.001 Hsueh, S.-J., Hu, Y.-H., & Tu, C.-H. (2013). Economic growth and financial development in Asian countries: A bootstrap panel Granger causality analysis. Economic Modelling, 32, 294–301. 14
15. Ibrahim , M. H. (2015). Issues in Islamic banking and finance: Islamic banks, Shari’ahcompliant investment and sukuk. Pacific-Basin Finance Journal, 34, 185-191. doi:https://doi.org/10.1016/j.pacfin.2015.06.002 ICD-REFINITIV. (2018). Islamic Finance Development Report 2019: Shifting Dynamics: The Islamic Corporation for the Development of the Private Sector (ICD) & Refinitiv. IFSB. (2019). Islamic Financial Services Industry Stability Report 2019. Kuala Lumpur: Islamic Financial Services Board (IFSB). Khan, A. (2001). Financial Development and Economic Growth. Macroeconomic Dynamics, 5(3), 413-433. doi:undefined King, R. G., & Levine, R. (1993). Finance and Growth: Schumpeter Might Be Right. Quarterly Journal of Economics, 108(3), 717-738. Kiviet, J. F. (1995). On bias, inconsistency, and efficiency of various estimators in dynamic panel data models. Journal of Econometrics, 68(1), 53-78. doi:https://doi.org/10.1016/03044076(94)01643-E Law, S. H., & Singh, N. (2014). Does too much finance harm economic growth? Journal of Banking & Finance, 41, 36-44. doi:https://doi.org/10.1016/j.jbankfin.2013.12.020 Levine, R. (1997). Financial Development and Economic Growth: Views and Agenda. Journal of Economic Literature, 35(2), 688-726. Levine, R. (2003). More on Finance and Growth: More Finance, More Growth? Federal Reserve Bank of St. Louis Review, 85(July/August), 31-46. Levine, R., Loayza, N., & Beck, T. (2000). Financial Intermediation and Growth: Causality and Causes. Journal of Monetary Economics, 46(1), 31-77. Levine, R., & Renelt, D. (1992). A Sensitivity Analysis of Cross-Country Growth Regressions. The American Economic Review, 82(4), 942-963. Lucas, R. E. (1988). On the Mechanics of Economic Development. Journal of Monetary Economics, 22(1), 3-42. Luintel, K. B., & Khan, M. (1999). A Quantitative Reassessment of the Finance-Growth Nexus: Evidence from a Multivariate VAR. Journal of Development Economics, 60(2), 381405. Marques, L. M., Fuinhas, J. A., & Marques, A. C. (2013). Does the stock market cause economic growth? Portuguese evidence of economic regime change. Economic Modelling, 32, 316–324. McKinnon, R. I. (1973). Money and Capital in Economic Development. Washington, DC: Brookings Institution. Naceur, S. B., Blotevogel, R., Fischer, M., & Shi, H. (2017). Financial Development and Source of Growth: New Evidence. IMF Working Paper, WP/17/143. Naceur, S. B., & Ghazouani, S. (2007). Stock markets, banks, and economic growth: Empirical evidence from the MENA region. Research in International Business and Finance, 21(2), 297-315. doi:https://doi.org/10.1016/j.ribaf.2006.05.002 Nickell, S. (1981). Biases in Dynamic Models with Fixed Effects. Econometrica, 49(6), 14171426. doi:https://doi.org/10.2307/1911408 Patrick, H. T. (1966). Financial Development and Economic Growth in Underdeveloped Countries. Economic Development and Cultural Change, 14(2), 174-189. Porter, R. C. (1966). The promotion of the ≪ banking habit ≫ and economic development. The Journal of Development Studies, 2(4), 346-366. doi:10.1080/00220386608421196 Prochniak, M., & Wasiak, K. (2017). The impact of the financial system on economic growth in the context of the global crisis: empirical evidence for the EU and OECD countries. Empirica, 44(2), 295-337. doi:10.1007/s10663-016-9323-9 Rajan, R. G., & Zingales, L. (1998). Financial Dependence and Growth. American Economic Review, 88(3), 559-586. 15
16. Ram , R. (1999). Financial development and economic growth: Additional evidence. The Journal of Development Studies, 35(4), 164-174. doi:10.1080/00220389908422585 Rioja, F. K., & Valev, N. T. (2004). Finance and the sources of growth at various stages of economic development. Economic Inquiry, 42(1), 127-140. Robinson, J. (1952). The Generalization of the General Theory in the Rate of Interest and Other Essays. London: Macmillan. Rousseau, P. L., & Wachtel, P. (2011). What is Happening to the Impact of Financial Deepening on Economic Growth? Economic Inquiry, 49(1), 276-288. Schumpeter, J. A. (1912). The theory of economic development. Cambridge, MA: Harvard University Press. Seetanah, B., Ramessur, S. T., & Rojid, S. (2009). Financial development and economic growth: New evidence from a sample of island economies. Journal of Economic Studies, 36(2), 124-134. Shahzad, S. J. H., Ferrer, R., Ballester, L., & Umar, Z. (2017). Risk transmission between Islamic and conventional stock markets: A return and volatility spillover analysis. International Review of Financial Analysis, 52, 9-26. doi:https://doi.org/10.1016/j.irfa.2017.04.005 Shaw, E. S. (1973). Financial deepening in economic development. New York: Oxford University Press. Shen, C.-H., & Lee, C.-C. (2006). Same financial development yet different economic growth – why? Journal of Money, Credit, and Banking, 38(7), 1907-1944. Valickova, P., Havranek, T., & Horvath, R. (2015). Financial Development and Economic Growth: A Meta-Analysis. Journal of Economic Surveys, 29(3), 506-526. doi:10.1111/joes.12068 16 View publication stats