The short-run relationship between the financial system and economic growth: New evidence from regional panels

International Review of Financial Analysis 29 (2013) 70–78

Contents lists available at SciVerse ScienceDirect

International Review of Financial Analysis

The short-run relationship between the financial system and economic growth: New evidence from regional panels Paresh Kumar Narayan a,⁎, Seema Narayan b a b

Centre for Financial Econometrics, Deakin University, Melbourne, Australia School of Economics, Finance, and Marketing, Royal Melbourne Institute of Technology University, Melbourne, Australia

a r t i c l e

i n f o

Article history: Received 1 March 2012 Received in revised form 2 March 2013 Accepted 25 March 2013 Available online 2 April 2013 Keywords: Financial system Economic growth Panel data Bank credit

a b s t r a c t In this paper, we examine the impact of the financial system on economic growth for a panel of 65 developing countries. The novelty of our paper is that we examine these relationships for various regional panels. Our main findings are that while for the full panel of 65 countries there is evidence of financial sector-led growth, bank credit has a negative effect on economic growth. At the regional level, for the Middle Eastern countries evidence suggests that neither the financial sector nor the banking sector contributes to growth. Except for Asia, the role of financial sector development on economic growth is relatively weak. Finally, except for the Middle Eastern countries, clear evidence is found in favour of bank credit having a statistically significant and negative effect on economic growth. © 2013 Elsevier Inc. All rights reserved.

1. Introduction The role of financial sector development on economic growth was first identified over 100 years ago by Bagehot (1873), who argued that the financial system played a crucial role in stimulating industrialisation in England by facilitating the mobilisation of capital. A related observation was later made by Schumpeter (1911). His idea was based on the relationship between the financial intermediary sector and the resulting allocation of savings to firms, which he perceived as having implications for productivity growth and technological change (see also Schumpeter, 1934). There are several empirical studies that examine the relationship between financial sector development and economic growth. This literature can be divided into two branches. 12 One strand of this literature examines the impact of stock market developments, namely, market capitalisation, turnover ratio, and stocks traded on economic growth. The second strand of this literature focuses on the relationship between banking sector developments, namely, private credit and liquid liabilities, and economic growth. In the next section, we

⁎ Corresponding author. Tel.: +61 3 9244 6180; fax: +61 3 9244 6034. E-mail addresses: [email protected] (P.K. Narayan), [email protected] (S. Narayan). 1 Bekaert, Harvey, and Lundblad (2001) examine the relationship between financial liberalisation and economic growth for a panel of emerging markets and find that financial liberalisation stimulates economic growth. 2 In a recent study, Amable and Chatelain (2001) showed how financial infrastructure fosters economic growth. They argued that financial infrastructure decreases depositor transaction costs and changes consumer welfare through increasing the proximity of financial services, which, they argue, increase savings and endogenous growth. 1057-5219/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.irfa.2013.03.012

review these two strands of the literature. The main message from this literature survey is that there is strong support for the hypothesis that financial sector and banking sector (commonly referred to as the financial system in the literature) development promote economic growth. Within this branch of the literature, a sub-set of studies show that trade openness and export growth contribute to economic growth (see Lucas, 2009; Wacziarg & Welch, 2008). Motivated by these findings, some studies have begun to examine the relationship between finance and trade (see, for example, Baltagi, Demetriades, & Law, 2009; Bordo & Rousseau, 2012; Demetriades & Rousseau, 2011). Our study contributes to this literature by examining the relationship between financial and banking sector developments and economic growth for a panel of 65 developing countries. Our study is different from the extant literature in three ways. First, we focus only on developing countries. While Anwar and Sun (2011), contrary to the literature, do not find evidence in favour of financial sector-led economic growth, their study is based on one developing country, Malaysia. Therefore, we extend the Anwar and Sun (2011) study by considering no fewer than 65 developing countries. Moreover, unlike previous studies (see Section 2), we do not form a panel representing a combination of developed and developing countries. Our motivation for departing from the literature on this approach is as follows. Including both developed and developing countries in cross-section or panel data analysis of the impact of financial sector development on economic growth can lead to biased results in the sense that the developed countries on the panel may be responsible for the positive relationship between financial/banking sector development and economic growth. Thus, to generalise that financial/banking sector development stimulates economic growth in a panel including both developed and developing countries can be misleading because the

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

positive relationship may simply be driven by the developed markets of the panel. A second way our study is different is as follows. We, for the first time in this literature, divide the sample of 65 developing countries into regions. Thus, we form regional panels. For example, we have an Asian panel, a European panel, an African panel, a South American panel, and a Middle Eastern panel. The formation of regional panels is motivated by Narayan, Mishra, and Narayan (2011), who show that regional panels of countries have relatively more homogeneous financial indicators. The advantages of forming regional panels are twofold: (1) we are able to test the finance–growth relationship for a more homogenous group of countries; and (2) we are able to compare the finance–growth experiences of different regions. Third, our study examines the short-run relationship between financial systems and economic growth. There are very few studies which have considered the short-run relationship; exceptions are Loayza and Ranciere (2006) and Kaminsky and Reinhart (1999). There are three factors that motivate us to undertake a short-run investigation: (a) data limitations, (b) the concern that averaging data leads to loss of information and prevents the estimation of a more flexible model capable of allowing parameter heterogeneity across countries (see Loayza & Ranciere, 2006), and (c) in the short-run the banking sector development, if over-liberalised, can have negative effects on economic growth, hence a short-run analysis allows us to examine this possibility. The balance of our paper proceeds as follows. In Section 2, we provide a brief overview of the literature on the finance-economic growth nexus. In Section 3, we discuss the theoretical motivation, and in Section 4, we discuss the data, the empirical model, and the results. In the final section, we provide some concluding remarks. 2. Literature review There is a large volume of studies on this topic. In this section, we only review selected recent studies that share some common features with the present study. Levine, Loayza, and Beck (2000) examined the relationship between financial intermediary development and economic growth for a panel of 74 developed and developing countries, and for a crosssection of 71 developed and developing countries. For the panel data, they used the Arellano and Bond (1991) panel-GMM estimator. They found that financial intermediary variables, namely, liquid liabilities and private credit, have a statistically significant and positive effect on economic growth in both cross-sectional and panel data models. King and Levine (1993a) examined the relationship between economic growth and financial sector indicators (ratio of liquid liabilities of the financial system to GDP, ratio of deposit money bank deposit assets to deposit money bank domestic assets plus central bank domestic assets, private sector credit, and ratio of claims on the non-financial private sector to GDP) using cross-sectional data for 80 developed and developing countries. They found that their four measures of the financial system, namely financial depth, the relative importance of banks vis-a-vis the central bank, the percentage of credit allocated to nonfinancial private firms, and credit to private sector, all have a statistically significant and positive effect on growth indicators. Levine (1998) examined the impact of the banking sector development, proxied by credit allocated by deposit taking banks to the private sector divided by GDP, on economic growth, capital stock accumulation, and productivity growth. His empirical analysis was based on 42 developed and developing countries over the period 1976–1993. He used a panel GMM estimator and found that banking sector development has a statistically significant positive effect on economic growth. Levine and Zervos (1998) examined the impact of stock market and banking sector development on economic growth for a cross-section of 45 developed and developing countries using data for the period

71

1976–1993. They found that banking sector development and stock market liquidity were both good predictors of economic growth, capital accumulation, and productivity growth. Cole, Moshirian, and Wu (2008) examined the impact of banking sector stock returns on economic growth for 38 countries (both developed and developing). They used data for the period 1973 to 2001 and their empirical analysis is based on the Arellano and Bond GMM estimator. They found that bank stock returns have a statistically significant and positive effect on economic growth, and bank stock returns have a larger impact on economic growth in a panel of emerging markets compared with a panel of developed markets. Shen and Lee (2006) examined the relationship between financial development and economic growth for a panel of 48 developed and developing countries. They used data for the period 1976–2001 and their estimation was based on the ordinary least squares and two-stage least squares procedures. They found that only stock market development has a positive impact on economic growth. Beck and Levine (2004) examined a panel of 40 developed and developing countries over the period 1976–1998, and estimated the impact of stock market and banking sector developments on economic growth using the Arellano and Blundell system-GMM estimator. They found that stock market and banking sector developments both have statistically significant and positive effects on economic growth. The relatively more recent studies have also documented evidence that the financial system leads to economic growth. In a panel data study based on 31 Chinese provinces, Hasan, Wachtel, and Zhou (2009) used the GMM estimator and found that the development of financial markets promoted economic growth at the provincial level. Similar findings were reported by Zhang, Wang, and Wang (2012) for a data set consisting of 286 Chinese cities over the 2001–2006 period. Bittencourt (2012) used time series and panel data models to estimate the relationship between financial development and economic growth for four Latin American countries. He found strong evidence that financial development contributes to economic growth. Using historical data (1896 to 2000), Campos, Karanasos, and Tan (2012) found that financial development contributes to economic growth in Argentina. While the bulk of this literature confirms that the financial sector contributes to economic growth, some recent studies on developing countries cast doubt on this positive relationship; see Anwar and Sun (2011). The main message emerging from this brief literature review is that the financial system contributes to economic growth. Therefore, before we propose our empirical framework, in the next section, we consider some key theoretical issues that motivate our empirical framework. 3. Theoretical considerations In this section, we discuss the short-run theoretical association between financial and banking sector developments and the other determinants of economic growth, such as inflation, openness, and capital stock, considered in this study. The relationships discussed here are obviously also possible in the long-run. 3.1. Finance and economic growth Generally, the literature recognises four functions of the financial sector which are perceived to be growth-enhancing. First, financial intermediaries facilitate pooling and trading of risk. The idea is simple: in the absence of financial markets, investors constrained by liquidity shocks are forced to withdraw funds invested in long-term investment projects. Withdrawal of investment funds hurts economic growth. Financial markets are a remedy to liquidity constraints since they provide lenders immediate access to funds. At the same time, financial markets offer borrowers a long-term supply of capital. Stock markets also offer investors an opportunity to diversify their

72

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

risk. For example, investors can purchase shares in a number of firms. This diversification allows one to invest in relatively more risky productive technology. Subsequently, this investment, and indeed returns, are a source of economic growth (Obstfeld, 1994). Second, financial intermediaries ensure an efficient allocation of funds over investment projects by acquiring information ex-ante. King and Levine (1993b) developed an endogenous growth model, showing the connections between finance, entrepreneurship and economic growth. In their model, financial systems are perceived as influencing decisions to invest in productivity-enhancing activities through evaluating prospective entrepreneurs and funding the most promising ones. They argue that financial institutions are able to deliver evaluation and monitoring services more efficiently than individual investors. This lowers the cost of investing in productivity-enhancing activities, thus stimulating economic growth; for a lucid discussion, see Levine (2005). Third, stock markets provide better corporate control. Equity capital introduces a new possibility of aligning interests between the management and the ownership of the firm (Nieuwerburgh, Buelens, & Cuyvers, 2006). Fourth, financial markets mobilise savings in an efficient way. Savings are channeled into long-term assets that are more productive than short-term assets (Bencivenga & Smith, 1991). Moreover, the small denomination of securities allows greater individual participation in the stock market (see Nieuwerburgh et al., 2006). A potential negative effect of rapid stock market growth may result from speculative pressures (Singh, 1997). One avenue for these pressures may be transactions induced by the euphoria created by financial liberalisation, which essentially rewards speculators with short-term horizons and punishes those with a long-term view (Keynes, 1936). The result is that the economy may experience a greater degree of risk in the face of financial liberalisation than without it, putting upward pressure on interest rates due to the high risk factor (Federer, 1993). Higher interest rates are likely to retard investment and, as a result, hurt economic growth. In related work, Devereux and Smith (1994) argued that greater risk sharing in stock trading activities could potentially reduce the savings rate, hence hampering economic growth. De Long, Shleifer, Summers, and Waldmann (1989) pointed out that excessive stock trading by virtue of creating “noise” in the market can lead to misallocation of resources. A second view suggests that the inverse relationship between economic growth and financial institutions results from the relatively underdeveloped equity markets in developing countries (Singh, 1997). Equity markets are considered to be relatively weak in facilitating the process of financial intermediation between households and the business sector (Fry, 1997). Compared with the workings of the financial market, commercial banks are different in that equity funding is repaid contingent on firm performance. A potential negative effect from the banking sector on economic growth exists as well. In the McKinnon–Shaw model, for instance, banks allocate credit according to transaction costs and perceived risk of default and not on the expected productivity of the investment project. The average efficiency of investment is thus reduced. The loan rate ceiling is lowered as investments with lower returns now become profitable. Low interest rates lead to a bias in favour of current consumption and against future consumption. The implication is that the savings rate may fall below the socially optimum level. 3.2. Inflation, openness, capital stock, and economic growth There are two opposing views on the relationship between inflation and economic growth. One view perceives inflation as having a positive effect on economic growth, resulting from inflation's positive influence on capital accumulation. This has commonly come to be known as the Mundell–Tobin effect. Mundell (1965) and Tobin (1965)

contended, on the assumption that money and capital are substitutes, that a rise in inflation will increase the cost of holding money and induce a portfolio shift from money to capital. The resulting fall in interest rate stimulates investment and economic growth. The opposing view treats inflation as a tax on investment, leading to an increase in the effective cost of investment (De Gregorio, 1993; Fischer, 1993; Stockman, 1981). Inflation creates an uncertain macroeconomic policy environment, reducing the efficiency of the price mechanism. This disrupts economic decision-making and retards productivity growth (Fischer, 1993). Macroeconomic uncertainty resulting from inflation can also delay investment. If investors perceive the uncertainty to be temporary, they adopt a wait-and-see approach (see Pindyck & Solimano, 1993). As argued by Fischer (1993), an additional channel through which inflation negatively affects economic growth is when inflation uncertainty induces capital flight. 3 The theoretical literature on trade openness and economic growth is inclusive, in that trade openness is seen as a stimulus for economic growth as well as an impediment to economic growth. For example, Grossman and Helpman (1992) argued that trade openness stimulates technological change, by increasing domestic imports of goods and services, which includes new technology. New technology may make the production process more efficient and raise productivity, which in turn will have a positive effect on a country's economic growth. Levine and Renelt (1992) similarly argued that if trade openness exposes countries to more investment goods, then one source of economic growth could be through the investment channel. Other studies, such as Batra (1992), and Batra and Slottje (1993), amongst others, argued that trade openness to the extent that it reduces tariffs has a negative effect on economic growth. The idea is that a reduction in tariffs reduces the relative price of domestic manufacturing, making it less attractive relative to foreign goods. The capital stock and economic growth relationship is well established in the economic growth literature. Capital stock has a positive effect on economic growth, both in the Cobb–Douglas production function and in various growth models. 4. Data, descriptive statistics, model specification, and results 4.1. Data Our panel data set includes 65 developing countries. 4 The time component of the data is for the period 1995 to 2011, ensuring 17 observations per country. The following variables are used in the estimation: economic growth rate, gross fixed capital formation (GFCF), inflation, trade openness measured as exports plus imports as a percentage of GDP, market capitalisation of listed companies as a percentage of GDP, domestic credit provided by the banking sector as a percentage of GDP, and stocks traded as a percentage of GDP. All data are extracted from the World Development Indicators published by the World Bank. In our study, the banking sector is proxied by domestic credit provided by the banking sector, consistent with the literature. This is a commonly used proxy because consistent time series data on domestic credit is available for a large number of countries. 3 Similarly, money growth in cash-in-advance models with production (see Cooley & Hansen, 1989 and Stockman, 1981) generates a pure inflation tax effect, which discourages market activities requiring cash. As a result, consumption, work effort, output, and the capital stock all decline with the inflation rate (Laing, Li, & Wang, 1998). 4 These countries are Argentina, Bahrain, Bangladesh, Bolivia, Botswana, Brazil, Bulgaria, Chile, China, Colombia, Costa Rica, Cote d'Ivoire, Croatia, Cyprus, Ecuador, Egypt, El Salvador, Estonia, Ghana, India, Indonesia, Iran, Jamaica, Jordan, Kazakhstan, Kenya, South Korea, Kuwait, Latvia, Lebanon, Lithuania, Malaysia, Malta, Mexico, Moldova, Mongolia, Morocco, Nigeria, Oman, Pakistan, Panama, Papua New Guinea, Paraguay, Peru, Philippines, Poland, Romania, Russia, Saudi Arabia, Serbia and Montenegro, Singapore, the Slovak Republic, Slovenia, South Africa, Sri Lanka, Tanzania, Thailand, Trinidad and Tobago, Tunisia, Uganda, Ukraine, the United Arab Emirates, Uruguay, Venezuela, and Zambia.

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

The financial sector, on the other hand, is proxied by market capitalisation and stocks traded. This choice is motivated by a recent study by Narayan et al. (2011), which shows the speed of convergence in market capitalisation and stocks traded for 120 countries. In Table 1, we report some commonly used descriptive statistics of the data. We organise the descriptive statistics — namely, mean, standard deviation, coefficient of variation, kurtosis, and skewness of the data by regional panels, including for a panel of all 65 countries. The 65 countries are categorised into an Asian panel, an African panel, a Middle Eastern panel, a European panel, and a Central and South American panel. We observe disparities in all variables across regions. Beginning with market capitalisation as a percentage of GDP, the average over the 1995–2011 period varies from a low of 22% in the European panel of our sample to around 59% in the Middle Eastern panel. Stocks traded as a percentage of GDP average around 3.9% in the Central and South American panel to 44% in the Asian panel. The standard deviation follows a similar pattern. Domestic credit offered by the private

73

sector averages around 45% in the African, European, and Central and South American panels, while it is twice as much in the Middle Eastern panel. Even in terms of economic growth, there are wide differences across regional panels. The average growth rate in the Asian panel is the highest at around 5.4%, followed by the African panel (4.6%), the Middle Eastern panel (4.4%), and the European and the Central and South American panels (around 3.6%). The coefficient of variation suggests that economic growth rate is the most volatile for the European panel of countries followed by countries in the Central and South American panel. Economic growth is found to be least volatile for countries in the Asian and African panels. Finally, the kurtosis and skewness statistics suggest that most of the data series are non-normal. 4.2. Model specification Based on the conceptual framework presented earlier, we have a dynamic panel specification. Our preference, given the relatively small panel that we have, is to use the Arellano and Bond (AB, 1991) generalised method-of-moments (GMM) estimator. Our proposed model is of the following form:

Table 1 Descriptive statistics. GGDP

DCBY

INF

GFCF

MCAPY OPEN

STOCKSY

Panel A: Full-sample of countries Mean 4.206 56.169 10.695 SD 4.002 44.279 36.547 Skewness −1.049 1.700 22.688 Kurtosis 6.497 8.008 623.997 CV 0.951 0.788 3.417

2.13E + 13 39.750 1.37E + 14 47.045 11.888 2.383 170.557 9.793 6.468 1.184

87.696 16.298 53.327 36.556 2.967 4.152 17.176 27.078 0.608 2.243

Panel B: Asian countries Mean 5.449 73.245 SD 3.691 40.384 Skewness −1.235 0.709 Kurtosis 8.174 2.416 CV 0.677 0.551

6.313 6.729 4.064 27.221 1.066

8.33E + 13 57.137 2.94E + 14 59.374 5.579 1.662 37.402 5.600 3.532 1.039

106.018 44.450 96.576 51.285 2.050 1.545 6.611 4.793 0.911 1.154

Panel C: Middle Eastern countries Mean 4.358 83.453 4.646 SD 3.059 67.933 6.546 Skewness 0.423 1.462 3.181 Kurtosis 4.855 4.760 18.139 CV 0.702 0.814 1.409

2.78E + 13 59.337 1.01E + 14 50.810 4.531 1.635 24.315 6.563 3.634 0.856

94.872 30.234 33.463 57.064 0.361 3.743 2.624 20.567 0.353 1.887

Panel D: European countries Mean 3.574 48.651 19.398 4.66E + 11 21.153 SD 4.762 31.260 73.359 1.58E + 12 19.547 Skewness −1.390 1.425 11.807 4.873 1.962 Kurtosis 5.556 4.957 161.552 29.044 8.137 CV 1.332 0.643 3.782 3.382 0.924

102.845 4.172 37.393 8.961 0.350 4.526 2.166 25.481 0.364 2.148

Panel E: Central and South American Countries Mean 3.602 45.403 10.228 6.32E + 12 31.915 SD 4.081 23.553 12.787 1.91E + 13 33.071 Skewness −0.201 0.703 3.863 4.544 1.415 Kurtosis 4.301 2.489 22.916 25.219 4.429 CV 1.133 0.519 1.250 3.021 1.036

67.889 3.887 34.571 7.543 0.848 3.304 3.554 15.090 0.509 1.941

Panel E: African countries Mean 4.611 45.058 SD 2.955 48.602 Skewness −0.528 0.912 Kurtosis 4.972 4.489 CV 0.641 1.079

67.524 9.714 18.024 24.804 0.045 3.616 2.601 16.758 0.267 2.554

9.665 9.525 3.011 16.259 0.985

1.66E + 12 38.866 4.35E + 12 56.899 5.895 2.691 49.827 9.842 2.622 1.464

This table reports some commonly used descriptive statistics; namely, mean, standard deviation (SD), skewness, Kurtosis, and coefficient of variation (CV). These statistics are reported for each of the seven variables — GDP growth rate (GGDP), domestic credit by the banking sector (DCBY), inflation rate (INF), gross fixed capital formation (GFCF), stock market capitalisation as a percentage of GDP (MCAPY), trade openness (OPEN) and value of stocks traded as a percentage of GDP (STOCKSY). The statistics are reported in various panels representing each of the regional panels; namely, Asian, Middle Eastern, European, Central and South American, and African country panels. Panel A reports results for all 65 countries in our data set.

gyi;t ¼ α 0 gyi;t−1 þ βX ′ i;t þ ψF ′ i;t þ υi þ ε i;t ;

i ¼ 1; …N;

t ¼ 1; …; T

where gy stands for the economic growth rate of country i, computed from real GDP at time t, α0 and β are parameters to be estimated; X is a vector of core explanatory variables, namely, inflation, gross fixed capital formation, and trade openness (exports plus imports divided by GDP) used to model economic growth; F is our measure of financial/banking sector development, proxied by stocks traded as a percentage of GDP, market capitalisation as a percentage of GDP, and banking sector credit available to the private sector as a percentage of GDP; υ is country specific effects; and ε is the error term. By comparison, using the panel OLS estimator (with fixed and random effects) is problematic because the lagged dependent variable is correlated with the error term. By first differencing Eq. (1), the Arellano and Bond (1991) GMM estimator solves this problem and eliminates country specific effects. E(εi,t − εi,t − 1) = 0 but (gyi,t − 1 − gyi,t − 2) is not independent of (εi,t − εi,t − 1). The AB method solves this problem by using two or more lags of the first difference of the growth rate as instruments. With respect to (Xi,t − Xi,t − 1) and (Fi,t − Fi,t − 1), we assume that the financial and banking sector variables and the control values are predetermined in the sense that E(Xi,t,εi,s) ≠ 0 and E(Fi,t,εi,s) ≠ 0 for s b t but zero for s ≽ t. For the predetermined variables, one or more period lagged levels of the variables are orthogonal to the differenced error term and thus form valid instruments for respective first differenced right-hand side variables. It should be noted that to examine the robustness of our results, we also estimate the model using the system dynamic panel data estimator of Arellano and Bover (1995) and Blundell and Bond (1998). The results were very similar to the Arellano and Bond (1991) estimator so to conserve space we do not report all the results here. We only report results for the full sample panel of 65 countries. The rest of the results are, however, available from the author upon request. A final issue is the choice between the one-step estimator and the two-step estimator for the first-differenced GMM estimator and the system-GMM estimator. It has been shown that the two-step estimator is asymptotically more efficient than the one-step estimator. However, in subsequent work, Blundell and Bond (1998) have shown that the asymptotic inferences based on the one-step estimator are more reliable in that they have correct empirical size distributions. Given this, we use the one-step estimator. Our approach is also consistent with recent studies using this methodology; see, for instance, Cole et al. (2008).

74

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

4.3. Results 4.3.1. Overall results The results from the full panel of 65 developing countries are reported in Table 2. The results are reported from two estimators: columns 2–4 include results from the differenced-GMM estimator, while results in columns 5–7 are from the system-GMM estimator. In column 2 of Table 2, the results from model 1, where the impact of market capitalisation on economic growth is estimated, are reported. We find that an increase in market capitalisation as a percentage of GDP has a statistically significant and positive effect on economic growth. Consistent with theory, GFCF has a statistically significant (at the 1% level) and positive effect and inflation has a statistically significant (at the 1% level) and negative effect on economic growth. Finally, we find that openness has a statistically significant (at the 1% level) and positive effect on economic growth in the developing countries considered here. In column 3 of Table 2, results from the model where we estimate the impact of stocks traded as a percentage of GDP are reported. We find that stocks traded as a percentage of GDP have a statistically significant (at the 1% level) and positive effect on economic growth. The GFCF has a statistically insignificant effect on economic growth, while inflation has a statistically significant (at the 1% level) and negative effect on economic growth. Finally, we find that trade openness has a statistically significant and positive effect on economic growth. In column 4 we present results from model 3. In model 3, we estimate the impact of domestic credit provided by the banking sector as a percentage of GDP; the other regressors are the same as those used in models 1 and 2, namely, GFCF, inflation, and trade openness. The results reveal that domestic credit has a statistically significant (at the 1% level) negative effect on economic growth. In terms of the impact of the other variables on economic growth, we find results consistent with models 1 and 2. The GFCF appears to be statistically insignificant, while inflation and trade openness have a negative and positive effect on economic growth, respectively. All results are statistically significant at the 1% level.

Table 2 GMM difference and system estimator based results for the full sample of 65 countries. D-GMM1 INF GFCF OPEN MCAPY STOCKSY Credit Sargan test AR(2)

D-GMM2

D-GMM3

S-GMM1

S-GMM2

−0.0054a −0.0056a −0.0061a −0.0016a −0.0031a (0.000) (0.000) (0.000) (0.000) (0.000) −0.0000a −0.000 0.000 −0.0000a −0.0000 (0.004) (0.443) (0.817) (0.004) (0.460) 0.0861a 0.1070a 0.1398a 0.0250a 0.0354a (0.000) (0.000) (0.000) (0.000) (0.000) 0.0235a 0.0282a (0.000) (0.000) 0.0162a 0.0117a (0.000) (0.000) −0.1056a (0.000) 62.849 60.693 59.536 61.4072 60.606 (1.000) (1.000) (1.000) (1.000) (1.000) −0.5976 −1.172 −1.5399 0.7165 −0.4064 (0.5501) (0.2412) (0.1236) (0.4737) (0.6844)

S-GMM3 −0.0037a (0.000) −0.000 (0.092) 0.0946a (0.000)

The results from the system-GMM estimator are very similar in magnitude and statistical significance to those from the difference-GMM estimator. Inflation has a statistically significant and negative effect on economic growth while trade openness has a statistically significant and positive effect on economic growth. Market capitalisation and stocks traded consistently appear to have a statistically significant and positive effect on economic growth, while domestic credit has a statistically significant and negative effect on economic growth. The GFCF appears to be a statistically insignificant determinant of economic growth. In summary, the results suggest two conclusions. First, across the three models, we find that the two financial sector variables, namely, market capitalisation and stocks traded, have a statistically significant and positive effect on economic growth, while the banking sector variable (domestic credit) has a statistically significant and negative effect on economic growth in the panel of 65 developing countries that we consider. The second feature of the results is that the magnitude of the impact of inflation, GFCF, and trade openness, and their statistical significance are broadly the same across the three models regardless of the estimator. This merely confirms the robustness of the impact of these variables on economic growth for the panel of 65 developing countries in our sample. At this point, we will postpone a discussion of the reasons for our findings until later in this section after we have presented the regional results. 4.3.2. Regional results In this section, we divide our full panel of 65 developing countries into various regional panels and examine the role of market capitalisation, stocks traded, and domestic credit provided by the banking sector on economic growth at the regional level. This is a crucial robustness exercise because there is a certain level of heterogeneity amongst the various geographical regions of the globe, where the level of economic growth and stock market/banking sector development are different as well. Country heterogeneity, amongst a sample of 120 countries, has been shown by Narayan et al. (2011) with respect to convergence of stocks traded and market capitalisation. Motivated by this finding, we categorise countries into regional panels. The results from the Asian panel are presented in Table 3. We find that market capitalisation and stocks traded have a statistically Table 3 GMM difference estimator based results for Asian countries. Model 1 INF GFCF OPEN MCAPY

−0.0891a (0.000) 62.968 (1.000) −1.4507 (0.1469)

This table presents estimation results. Columns 2–4 report results based on the GMM-differenced estimator, while columns 5–7 report results based on the GMM-system estimator. Each estimator consists of three regression models. These models are distinguished in terms of the measure of the financial and banking sector variables. The first two models include market capitalisation and stocks traded while the third model includes only the banking sector credit. All models have three core macroeconomic variables; namely, inflation, gross fixed capital formation, and trade openness. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. a Denotes statistical significance at the 1% level.

Model 2 a

−0.2152 (0.009) 0.000 (0.945) 0.0450a (0.000) 0.0394a (0.000)

Model 3 a

−0.2248 (0.000) −0.000 (0.988) 0.0475a (0.001)

0.0291a (0.008)

STOCKSY Credit Sargan test AR(2)

−0.3043a (0.000) −0.0000 (0.933) 0.0834a (0.005)

7.9278 (1.000) −0.4838 (0.6285)

6.9409 (1.000) −1.1935 (0.2327)

−0.0139 (0.836) 8.3944 (1.000) −1.6891 (0.0912)

This table reports the regression results based on the GMM difference estimator for the Asian panel of countries. Model 1 is based on the market capitalisation proxy for financial market development, model 2 is based on the stocks traded proxy for financial market development, while model 3 reports results based on domestic credit proxy for banking sector development. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. a Denotes statistical significance at the 1% level.

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78 Table 4 GMM difference estimator based results for Middle Eastern countries. Model 1 INF GFCF OPEN MCAPY

0.2034 (0.295) 0.000a (0.010) 0.0520 (0.375) 0.0164 (0.328)

STOCKSY

Model 2 0.0276 (0.705) 0.000a (0.004) 0.0950a (0.006)

AR(2)

0.0534 (0.719) 0.000a (0.010) 0.0694⁎⁎ (0.038)

0.0051 (0.775)

Credit Sargan test

Model 3

1.5701 (1.000) −0.9722 (0.3309)

4.7204 (1.000) 0.3082 (0.7579)

−0.0146 (0.445) 5.2519 (1.000) 0.5823 (0.5604)

This table reports the regression results based on the GMM difference estimator for the Middle Eastern panel of countries. Model 1 is based on the market capitalisation proxy for financial market development, model 2 is based on the stocks traded proxy for financial market development, while model 3 reports results based on domestic credit proxy for banking sector development. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. a Denotes statistical significance at the 1% level. ⁎⁎ Denotes 5% level of significance.

significant and positive effect on economic growth, while domestic credit has a statistically insignificant effect on economic growth. These results are similar to those obtained from the full panel of countries. In terms of the impact of the other variables on economic growth, we find that inflation across all the three models has a statistically significant (at the 1% level) and negative effect, while GFCF has a statistically insignificant effect on economic growth. Finally, we observe that across all three models, trade openness appears to have a statistically significant and positive effect on economic growth in Asia. The results from the Middle Eastern panel are presented in Table 4. We find sharply different results for the Middle Eastern panel compared with the Asian panel and the full sample panel, in that none of the banking and stock market variables is a statistically significant determinant of economic growth. Moreover, inflation is also statistically insignificant, while openness in two of the three models and GFCF in all three models have a statistically significant and positive effect on economic growth. The results from the European panel are reported in Table 5. We find that while market capitalisation has a statistically significant (at the 1% level) and positive effect on economic growth, domestic credit has a statistically significant (at the 1% level) and negative effect on economic growth. Meanwhile, stocks traded have a statistically insignificant effect on economic growth. Inflation across all the three models has a statistically significant (at the 1% level) and negative effect on economic growth, while trade openness has a statistically significant and positive effect on economic growth. On the other hand, GFCF turns out to be statistically insignificant in all three models. The results from the Central and South American panel are reported in Table 6. The results suggest that market capitalisation and stocks traded have a statistically insignificant effect on economic growth, while domestic credit has a statistically significant (at the 1% level) and negative effect on economic growth. Except for trade openness, which has a statistically significant and positive effect on economic growth across all three models, none of the other macro variables has a statistically significant effect on economic growth. The results from the African panel are reported in Table 7. Our main findings are that market capitalisation has a statistically significant and positive effect on economic growth while stocks traded have a statistically

75

insignificant effect on economic growth. Domestic credit, as was found to be the case in most of the other regional panels, appears to have a statistically significant and negative effect on economic growth. Inflation and trade openness have statistically significant negative and positive effects on economic growth, respectively, while GFCF is mainly statistically insignificant. 4.3.3. Diagnostic tests In this section, we finish of the results by examining some of the commonly used diagnostic tests to check whether or not the data are consistent with the assumptions of the Arellano and Bond (1991) estimator. In particular, we report the Sargan test statistic which examines the overidentification restrictions. It essentially tests whether the instruments are uncorrelated with the error terms in the estimated equation. The null hypothesis is that the instruments as a group are exogenous. A finding of exogenous instruments is needed for the validity of the GMM estimates. The Sargan test statistic together with its associated p-values is reported in the last two rows of the full panel and regional tables. The Sargan test statistics for all models appear with a p-value greater than 0.10, hence we are unable to reject the null hypothesis. The second test we report is the Arellano and Bond test for autocorrelation. The null hypothesis is ‘no autocorrelation’ and relates to the differenced residuals. We only report the test statistics and its associated p-values for AR(2) because it detects autocorrelation in levels. For all the estimated models, we are unable to reject the null hypothesis of ‘no autocorrelation’. There is robust evidence that all models are free from autocorrelation at the 1% level. 4.4. Discussion of results For the finance-economic growth nexus, we find two interesting results. First, when we consider the panel of 65 countries as a whole, we find that stock market development (proxied by stocks traded and market capitalisation) has a statistically significant and positive effect on economic growth. When we examine the same hypothesis by considering regional panels, there is relatively less evidence in favour of the stock market-led growth hypothesis. For

Table 5 GMM difference estimator based results for European countries. Model 1 INF GFCF OPEN MCAPY

−0.0033⁎ (0.056) −0.000 (0.517) 0.1545a (0.000) 0.0287a (0.000)

STOCKSY

Model 2

Model 3 a

−0.0022 (0.004) −0.000 (0.657) 0.1676a (0.000)

0.0919 (0.846)

Credit Sargan test AR(2)

−0.0058a (0.000) 0.000a (0.000) 0.2059a (0.000)

11.0339 (1.000) −0.6123 (0.5403)

13.2905 (1.000) −0.4512 (0.6519)

−0.1554a (0.000) 13.929 (1.000) −1.4313 (0.1523)

This table reports the regression results based on the GMM difference estimator for the European panel of countries. Model 1 is based on the market capitalisation proxy for financial market development, model 2 is based on the stocks traded proxy for financial market development, while model 3 reports results based on domestic credit proxy for banking sector development. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. a Denotes statistical significance at the 1% level. ⁎ Denotes 10% level of significance.

76

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

Table 6 GMM difference estimator based results for Central and South American countries.

INF GFCF OPEN MCAPY

Model 1

Model 2

Model 3

−0.0438 (0.425) 0.000 (0.357) 0.2096a (0.000) −0.0102 (0.475)

−0.0412 (0.489) 0.000 (0.599) 0.2083a (0.000)

−0.0374 (0.405) 0.000** (0.014) 0.2236a (0.000)

STOCKSY

0.1219 (0.195)

Credit Sargan test AR(2)

10.260 (1.000) 0.1897 (0.8495)

9.7331 (1.000) 0.5762 (0.5645)

−0.2034a (0.000) 9.8946 (1.000) −0.1971 (0.8437)

This table reports the regression results based on the GMM difference estimator for the Central and South American panel of countries. Model 1 is based on the market capitalisation proxy for financial market development, model 2 is based on the stocks traded proxy for financial market development, while model 3 reports results based on domestic credit proxy for banking sector development. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. a Denotes statistical significance at the 1% level.

example, only for the Asian panel, do both stocks traded and market capitalisation have a statistically significant and positive effect on economic growth. In the European and the African panels, only market capitalisation has a statistically significant and positive effect on economic growth. In the Middle Eastern and Central and South American panels, both stocks traded and market capitalisation are statistically insignificant. These results, taken on the whole, seem to suggest that there is some, but not overwhelming, evidence of a positive relationship between stock market development and economic growth in developing countries.

Table 7 GMM difference estimator based results for African countries.

INF GFCF OPEN MCAPY

Model 1

Model 2

Model 3

−0.0661⁎ (0.054) 0.0000 (0.195) 0.0640⁎⁎⁎

−0.1188⁎⁎⁎ (0.000) 0.000 (0.792) 0.0535⁎⁎⁎

−0.0996⁎⁎⁎ (0.009) −0.000⁎⁎⁎ (0.001) 0.0631⁎⁎⁎

(0.000) 0.0096⁎⁎ (0.048)

STOCKSY

(0.000)

0.0182 (0.357)

Credit Sargan test AR(2)

2.7804 (1.000) −0.6535 (0.5134)

6.3463 (1.000) −0.1353 (0.8924)

(0.001)

−0.0749⁎⁎⁎ (0.001) 3.7853 (1.000) −1.3079 (0.1909

This table reports the regression results based on the GMM difference estimator for the African panel of countries. Model 1 is based on the market capitalisation proxy for financial market development, model 2 is based on the stocks traded proxy for financial market development, while model 3 reports results based on domestic credit proxy for banking sector development. The last two rows report the Sargan test and the AR(2) test results. The Sargan test examines the null hypothesis that the over-identifying restrictions are valid. The p-values are reported to test the null. The AR(2) test examines the null hypothesis of zero autocorrelation in the first-differenced errors. The p-values are reported to examine the null. ⁎ Denotes 10% level of significance. ⁎⁎ Denotes 5% level of significance. ⁎⁎⁎ Denotes statistical significance at the 1% level.

The statistically insignificant relationship between financial markets and economic growth for the Middle Eastern countries can be explained as follows. First, in many of the Middle East countries financial sector reform has progressed at a relatively slow pace compared with the other parts of the world, such as Asia. As a result, Creane, Goyal, Mobarak, and Sab (2003) argued that the challenge for policymakers is to move away from financially repressive policies. Second, the Middle East stock markets have different features, particularly with respect to the participation of foreign investors compared with other emerging markets. For instance, the Saudi Arabian stock market only allowed foreign investors in the market in 1997. In Bahrain, their stock exchange stipulates that non-resident foreigners would be allowed to buy up to 24% of the shares of a company listed on the exchange that was previously closed to non-Gulf Cooperation Council investors (Yu & Hassan, 2008). Similarly in Kuwait foreign investors were allowed to trade in the Kuwait stock exchange only in 2000 through a Ministerial resolution passed in that year. Our second most important finding is the statistically significant and negative relationship between bank credit and economic growth. Except for the Middle Eastern panel, for the rest of the panels there is strong evidence against the hypothesis that bank credit has a positive effect on economic growth. This finding is inconsistent with recent empirical studies, in particular Beck and Levine (2004), who for a panel of 40 developed and developing countries found that bank credit has a statistically significant and positive effect on economic growth. One reason for this difference in the results could be due to the make-up of their panel (which includes both developed and developing countries). It may simply be the case that the bank credit-led growth evidence may be arising purely from the developed countries making up the panel. 5 We also notice that the results on the effect of the banking sector on economic growth for the Middle Eastern countries is statistically insignificant. This result is in sharp contrast to results from the rest of the panel. This result may be due to the following. First, policymakers in the Middle Eastern countries are concerned regarding the absorptive capacity of banks to recycle oil surplus funds. This has challenged policymakers to design policies for more efficient and stable banking systems (Turk-Ariss, 2009). Second, the banking systems in the Middle Eastern countries have traditionally been very highly concentrated. Data presented in Turk-Ariss (2009) reveal that the three-bank average concentration ratios (based on total assets) over the period 2000–2006 have ranged from 49.6% in the case of UAE to 68% in Kuwait, 79% in Oman, and 80% in Bahrain. Third, the banking systems in the Middle Eastern countries are undergoing a period of financial liberalisation, from banking institutions primarily family-owned or state owned to private ownership and foreign ownership. Fourth, the banking system in the Middle Eastern countries functions under the framework of “Islamic banking”, which is different from conventional banking in that banks are not allowed to offer a fixed rate of return on deposits and are not allowed to charge interest on loans. A unique feature of Islamic banking is their profit-and-loss sharing regulation: borrowers share profits and losses with banks, which in turn share profit and losses with depositors. Thus, Khan and Mirakhor (1989) and Iqbal (1997), among others, argued that Islamic banks are well-equipped to absorb external shocks. Moreover, the risk-sharing feature of Islamic banks allows them to lend on a longer-term basis to projects with higher risk-return (Chong & Liu, 2009; Mills & Presley, 1999). Fifith, Creane et al. (2003) found that in many of the Middle Eastern countries the banking sector is dominated by public sector banks, which are characterised by government intervention in credit allocation, losses and liquidity problems, and wide interest rate spreads. They recommend developing modern and financial skills in the Middle Eastern banking system. 5 In a recent study, Chang, Jia, and Wang (2010) found no empirical evidence that bank loans lead to regional economic growth in China.

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

Theoretically, as explained, the outcome of a negative relationship between banks and economic growth can be traced to the McKinnon– Shaw model, which contends that government restrictions on the banking system in the form of either interest rate ceilings, high reserve requirements, or directed credit programmes retard financial development. In a relative sense, developing countries with weak capital markets are at a greater risk of these types of banking restrictions. The upshot of these banking restrictions is a negative effect on economic growth. The negative effect emerges potentially from three sources. First, low interest rates discourage savings in favour of current consumption. Second, potential lenders may engage in relatively low-yielding direct investment instead of lending by depositing money in a bank. Third, bank borrowers able to obtain all the funds they want at low loan rates will choose relatively capital-intensive projects. Financial repression is, therefore, anti-growth. Generally, the failure to find overwhelming evidence in favour of stock market-led growth and banking sector-led growth in our developing country regional panel may also mean that in these countries the legal environment is not well developed to protect the rights of investors. To this end, La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997) and Levine (1999) argued that establishing a legal environment that credibly protects the rights of investors is paramount. Levine (1999), in particular, showed that in countries with legal and regulatory systems that give a high priority to creditors receiving the full present value of their claims on corporations have better functioning financial intermediaries than countries where the legal system provides much weaker support to creditors. Not only the legal environment, the macroeconomic environment, in particular macroeconomic policies need to be well-developed in order to facilitate the proper functioning of the financial/banking system. Policies related to the taxation of the financial/banking sector will have implications for its growth. The quality of macroeconomic management will determine the degree of uncertainty that characterises the domestic economic environment, having an influence on the magnitude of loan evaluation and monitoring costs. Uncertain or unstable macroeconomic policies, as is the case in many developing countries, increase the costs of financial intermediation. To this end, Montiel (2003: 18–19) noted: “… instability in the form of the ‘boom-bust’ cycles that too often has afflicted developing countries, may make it much more difficult for financial intermediaries to evaluate and monitor the activities of their borrowers, not only increasing the cost of doing business, but also lowering the average quality of loans and thus imperilling the health of the financial system itself”.

5. Concluding remarks The goal of this paper was to revisit the financial developmenteconomic growth nexus for developing countries divided into various regional panels, namely, Asian, European, African, South American, and Middle Eastern panels. Our empirical analysis is based on data for the period 1995 to 2011 and uses the differenced and system-GMM estimators. We use two proxies for financial sector development, namely stocks traded and market capitalisation, and we use domestic credit provided by the banking sector as a proxy for banking sector development. Our main findings can be summarised as follows. In the case of the full panel of 65 countries, we find that financial sector development has a statistically significant and positive effect on economic growth, while banking sector development has a statistically significant and negative effect on economic growth. Similar results are found for the Asian panel. In the European and African panels, only market capitalisation has a statistically significant and positive effect on economic growth. In the Middle Eastern panel, financial sector development is found to be statistically insignificant. Finally, except for the Middle Eastern panel, banking sector development is found to have a statistically significant and negative effect on economic growth.

77

Our results for the full panel of 65 countries on the relationship between financial sector development while consistent with the bulk of the literature that uses panel data, are relatively weak when we consider regional panels, to the extent that we do not find any significant evidence of the financial sector-led growth hypothesis for the Middle Eastern countries.

References Amable, B., & Chatelain, J. -B. (2001). Can financial infrastructures foster economic development? Journal of Development Economics, 64, 481–498. Anwar, S., & Sun, S. (2011). Financial development, foreign investment and economic growth in Malaysia. Journal of Asian Economics, 22, 335–342. Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58, 277–297. Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error component models. Journal of Econometrics, 68, 29–51. Bagehot, W. (1873). Lombard street (1962 ed.). Homewood, IL: Richard D. Irwin. Baltagi, B. H., Demetriades, P. O., & Law, S. H. (2009). Financial development and openness: Evidence from panel data. Journal of Development Economics, 89, 285–296. Batra, R. (1992). The fallacy of free trade. Review of International Economics, 1, 19–31. Batra, R., & Slottje, D. J. (1993). Trade policy and poverty in the United States: Theory and evidence, 1947–1990. Review of International Economics, 1, 189–208. Beck, T., & Levine, R. (2004). Stock markets, banks, and growth: Panel evidence. Journal of Banking and Finance, 28, 423–442. Bekaert, G., Harvey, C. R., & Lundblad, C. (2001). Emerging equity markets and economic development. Journal of Development Economics, 66, 465–504. Bencivenga, V. R., & Smith, B. D. (1991). Financial intermediation and endogenous growth. Review of Economic Studies, 58, 195–209. Bittencourt, M. (2012). Financial development and economic growth in Latin America: Is Schumpeter right? Journal of Policy Modeling, 34, 341–355. Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87, 115–143. Bordo, M. D., & Rousseau, P. L. (2012). Historical evidence on the finance-trade-growth nexus. Journal of Banking and Finance, 36, 1236–1243. Campos, N. F., Karanasos, M. G., & Tan, B. (2012). Two to tangle: Financial development, political instability and economic growth in Argentina. Journal of Banking and Finance, 36, 290–304. Chang, P. C., Jia, C., & Wang, Z. (2010). Bank fund reallocation and economic growth: Evidence from China. Journal of Banking and Finance, 34, 2753–2766. Chong, B. S., & Liu, M. -H. (2009). Islamic banking: Interest free or interest-based? Pacific-Basin Finance Journal, 17, 125–144. Cole, R. A., Moshirian, F., & Wu, Q. (2008). Bank stock returns and economic growth. Journal of Banking and Finance, 32, 995–1007. Cooley, T. F., & Hansen, G. D. (1989). The inflation tax in a real business cycle model. American Economic Review, 79, 733–748. Creane, A., Goyal, R., Mobarak, A. M., & Sab, R. (2003). Financial development in the Middle East and North Africa. Working Paper WP/04/201.. Washington, DC: International Monetary Fund. De Gregorio, J. (1993). Inflation, taxation and long-run growth. Journal of Monetary Economics, 31, 271–298. De Long, J., Shleifer, A., Summers, L., & Waldmann, R. (1989). The size and incidence of the losses from noise trading. Journal of Finance, 44, 681–696. Demetriades, P. O., & Rousseau, P. L. (2011). Government, openness and finance: Past and present. The Manchester School, 79, 98–115. Devereux, M., & Smith, G. (1994). International risk sharing and economic growth. International Economic Review, 35, 535–550. Federer, P. (1993). The impact of uncertainty on aggregate investment spending. Journal of Money, Credit, and Banking, 25, 30–45. Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32, 485–512. Fry, M. J. (1997). In favour of financial liberalisation. The Economic Journal, 107, 754–770. Grossman, G. M., & Helpman, E. (1992). Innovation and growth: Technological competition in the global economy. Boston: MIT Press. Hasan, I., Wachtel, P., & Zhou, M. (2009). Institutional development, financial deepening, and economic growth: Evidence from China. Journal of Banking and Finance, 33, 157–170. Iqbal, Z. (1997). Islamic financial systems. Finance and Development, 43, 42–45. Kaminsky, G. L., & Reinhart, C. M. (1999). The twin crises: The causes of banking and balance-of-payments problems. American Economic Review, 89, 473–500. Keynes, J. M. (1936). The general theory of employment, interest, and money. London: Macmillan. Khan, M. S., & Mirakhor, A. (1989). Islamic banking: Experiences in the Islamic Republic of Iran and Pakistan. IMF Working Paper No. WP/89/12.. Washington DC: International Monetary Fund. King, R. G., & Levine, R. (1993a). Finance and growth: Schumpeter might be right. Quarterly Journal of Economics, 108, 717–737. King, R. G., & Levine, R. (1993b). Finance, entrepreneurship, and growth. Journal of Monetary Economics, 32, 513–542. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1997). Legal determinants of external finance. Journal of Finance, 52, 1131–1150.

78

P.K. Narayan, S. Narayan / International Review of Financial Analysis 29 (2013) 70–78

Laing, D., Li, V. E., & Wang, P. (1998). Inflation and economic activity in a multiple matching model of money. Working Paper 1998-018A.. St. Louis, MO: Federal Reserve Bank of St. Louis. Levine, R. (1998). The legal environment, banks, and long-run economic growth. Journal of Money, Credit, and Banking, 30, 596–613. Levine, R. (1999). Law, finance, and economic growth. Journal of Financial Intermediation, 8, 8–35. Levine, R. (2005). Finance and growth: Theory and evidence. In P. Aghion, & S. Durlauf (Eds.), Handbook of economic growth (pp. 865–934). : North-Holland. Levine, R., Loayza, N., & Beck, T. (2000). Financial intermediation and growth: Causality and causes. Journal of Monetary Economics, 46, 31–77. Levine, R., & Renelt, D. (1992). A sensitivity analysis of cross-country growth regressions. American Economic Review, 82, 942–963. Levine, R., & Zervos, S. (1998). Stock markets, banks and economic growth. American Economic Review, 88, 537–558. Loayza, N. V., & Ranciere, R. (2006). Financial development, financial fragility and growth. Journal of Money, Credit, and Banking, 38, 1051–1076. Lucas, R. E., Jr. (2009). Trade and the diffusion of the industrial revolution. American Economic Journal: Macroeconomics, 1, 1–25. Mills, P. S., & Presley, J. R. (1999). Islamic finance: Theory and practice. London: Macmillan. Montiel, P. J. (2003). Development of financial markets and macroeconomic policy. Journal of African Economies, 12(Supplement, 2), ii12–ii52 (AERC). Mundell, R. (1965). Growth, stability and inflationary finance. Journal of Political Economy, 73, 97–109. Narayan, P. K., Mishra, S., & Narayan, S. (2011). Do market capitalisation and stocks traded converge? New global evidence. Journal of Banking and Finance, 35, 2771–2781.

Nieuwerburgh, S. V., Buelens, F., & Cuyvers, L. (2006). Stock market development and economic growth in Belgium. Explorations in Economic History, 43, 13–38. Obstfeld, M. (1994). Risk taking, global diversification, and growth. American Economic Review, 1310–1329. Pindyck, R., & Solimano, A. (1993). Economic instability and aggregate investment. NBER Macroeconomic Annuals, 8, 259–303. Schumpeter, J. (1911). Theorie der Wirtschaftlichen Entwicklung Leipzig. Germany: Duncker and Humblot. Schumpeter, J. A. (1934). The theory of economic development. Cambridge: Harvard University Press. Shen, C. -H., & Lee, C. -C. (2006). Same financial development yet different economic growth — Why? Journal of Money, Credit, and Banking, 38, 1907–1944. Singh, A. (1997). Financial liberalisation, stockmarkets and economic development. The Economic Journal, 107, 771–782. Stockman, A. C. (1981). Anticipated inflation and the capital stock in a cash-in-advance economy. Journal of Monetary Economics, 8, 387–393. Tobin, J. (1965). Money and economic growth. Econometrica, 33, 671–684. Turk-Ariss, R. (2009). Competitive behaviour in Middle East and North Africa banking systems. The Quarterly Review of Economics and Finance, 49, 693–710. Wacziarg, R., & Welch, K. H. (2008). Trade liberalisation and economic growth: New evidence. World Bank Economic Review, 22, 187–231. Yu, J. -S., & Hassan, M. K. (2008). Global and regional integration of the Middle East and North African stock markets. The Quarterly Review of Economics and Finance, 48, 482–504. Zhang, J., Wang, L., & Wang, S. (2012). Financial development and economic growth: Recent evidence from China. Journal of Comparative Economics, 40, 393–412.