Financial Openness, Financial Markets Development, and Economic. Growth: Evidence from Americas, Asia, and Europe

Transcription

1 Financial Openness, Financial Markets Development, and Economic Growth: Evidence from Americas, Asia, and Europe ABSTRACT Financial openness can provide additional financial resources for domestic financial markets and firms by providing access to international financial markets. As a result, with banks and capital markets reducing financial constraints, firms have been able to invest in valuecreating projects. These investments should then enhance economic growth. Using a panel sample of countries in Americas, Asia, and Europe over the period , we find empirical support for the negative effect of financial openness on banking sector development. We do not find empirical evidence for the effect of financial openness on stock market development. We find consistent evidence to support the notion that financial development enhances economic growth. JEL Classification: E4 Keywords: financial openness, financial development, economic growth 1 Introduction Financial openness could affect financial development in a positive or negative way (Baltagi et al., 2009; de la Torre et al., 2007; Hauner et al., 2013). Empirical evidence on the nexus between financial openness, financial development and economic growth remains inconclusive. Therefore, we empirical test the relationships between financial openness, 1

2 financial development and economic growth by using a sample of countries in Asia, Europe, and Americas over the period A country may benefit from financial openness if financial openness strengthens the development of domestic financial sector but limits the adverse effects of increased volatility of asset prices, capital flows, and macroeconomic fluctuations. In a similar vein, the association between financial development and economic growth appears to be either positive or negative (Herwartz and Walle, 2014; Law and Singh, 2014; Rajan and Zingales, 2003; Rioja and Valev, 2004). Financial development could boost economic growth by (1) increasing efficiency of fund mobility and resource allocation and (2) facilitating trade, diversification, and management of risk (Sahay et al., 2015). However, too much finance could adversely affect economic growth as resources may be excessively tapped into the financial sector and away from the real sector, and funds may be allocated disproportionately to non-productive uses. Recent studies (Law and Singh, 2014; Rioja and Valev, 2004) have documented the inverted U-shape relationship between finance and growth. Empirical evidence from prior studies suggests that the growth-enhancing effect of financial development diminishes at higher level of financial development. When financial sector becomes too large, it may compete with the rest of the economy for scarce resources and crowd out resources away from the real sector. In addition, deepening financial development could create economic and financial instability if, poorly regulated and supervised, financial institutions take excessive risk and high leverage. The contribution of financial development to growth seems to differ across regions, countries, and income levels (Barajas et al., 2013; Nili and Rastad, 2007). The nexus between financial openness, financial development and economic growth remains inconclusive. 2

3 This study examines the relationships between financial openness and financial development and the finance-growth nexus in Asia, Europe, and Americas. In particular, we attempt to answer the following research questions: (1) whether the financial openness has led to more development of financial markets in the country that adopts it; and (2) how the financial development affects economic growth of a country. The objectives of this paper are twofold. First, we attempt to empirically examine the relationship between financial openness and financial development. In particular, we test whether openness of a country in terms of finance could link to the development of domestic financial markets. Financial openness enables a country or a firm to access to a larger pool of capital and tap funds in international markets. However, financial openness might have a crowding-out effect on domestic financial markets as in some cases financial openness could lead to outflows of funds. Financial openness could also lessen the effectiveness of domestic monetary policy. Second, we empirically test the effect of finance on economic growth. Prior studies (e.g., Law and Singh, 2014; Rajan and Zingales, 2003) provide some evidence for the positive effect of financial development on economic growth. Deepening financial development could promote growth by increasing efficiency of fund mobility, more effectiveness of resource allocation, as well as facilitating trade, diversification and management of risk (Sahay et al., 2015). To address our research questions and objectives, we construct an initial panel sample of 119 developing and advanced countries in Asia, Europe and Americas over the period of We estimate a series of panel OLS regressions using four measures of financial development as in prior studies (Baltagi et al., 2009; Law and Singh, 2014; Rajan and Zingales, 2003): namely, FD1 (domestic credit provided by financial sector as a percentage of GDP), FD2 (domestic credit to private sector as a percentage of GDP), FD3 (market 3

4 capitalization of listed domestic companies as a percentage of GDP), and FD4 (money supply (M2) as a percentage of GDP). Our key findings can be briefly summarized as follows. First, the findings suggest that financial openness adversely affects the levels of banking sector and credit markets development, positively affects the level of financial markets development in general, and has no effect on the level of stock market development. Second, we find that financial development enhances economic growth. This result is consistent with prior studies (Law and Singh, 2014; Rajan and Zingales, 2003). Countries with a higher degree of financial development seems to have higher economic growth than those with a lower degree of financial development. Third, we find that trade openness seems to strengthen financial development and economic growth. Financial openness is also good for economic growth. Last but not least, we find that countries with stock markets do not necessarily have higher economic growth than countries without stock markets and that the stock market development does not weaken the positive effect of banking sector development on economic growth. The rest of the paper is organized as follows. In Section 2, we provide a brief overview of the studies on financial openness, financial development and economic growth. We describe our data, variables, and research methodology in section 3. We report and discuss empirical results in Section 4. Section 5 concludes the paper. 2 Financial Openness, Financial Development, and Economic Growth A number of studies (e.g., Baltagi et al., 2009; de la Torre et al., 2007; Hauner et al., 2013) have examined the effects of financial openness on the development of financial sector from various perspectives. In theory, financial openness could benefit the country by 4

5 strengthening the development of domestic financial markets (Ranciere et al., 2006). Financial openness, however, may have the adverse effects by increasing volatility of asset prices and facilitating more pronounced macroeconomic fluctuations (e.g., Herwartz and Walle, 2014; Park, 2013; Ranciere et al., 2006; Umutlu et al., 2010; Yilmazkuday, 2011). Financial openness could induce more volatility of capital flows that trigger boom-bust cycles in economies, hereby increasing the risk of capital flow reversals, financial contagion and crises, especially in the countries with small financial sectors (Herwartz and Walle, 2014). Empirical evidence has documented that the effect of financial openness on the development of domestic financial markets can occur in both directions. Baltagi et al. (2009) find that openness in terms of trade or finance but not necessarily both could induce financial development. de la Torre et al. (2007) find that financial openness leads to the stock market development in the country that adopts it. On the contrary, Hauner et al. (2013) find little evidence on the effect of financial openness on domestic financial development in the sample of 91 countries over the period of Their results indicate that capital account liberalization can predict domestic financial liberalization only at short horizon and the effect is limited to securities markets development. The link between financial development and economic growth also seems to be bidirectional. On the one hand, financial development could boost economic growth through increasing efficiency of fund mobility, improving resource allocation to productive uses, promoting information sharing, facilitating trade, diversification, and management of risk, and monitoring investments and exerting corporate control (Sahay et al., 2015). A number of studies (e.g., Law and Singh, 2014; Rajan and Zingales, 2003) have found empirical evidence on the positive relationship between financial development and economic growth. On the other hand, too much finance could possibly have adverse effects on economic growth. When 5

6 the size of financial sector becomes too large relative to the real economy, resources could be excessively tapped into the financial sector and away from the real economy. In some cases, financial sector could provide too much consumption loans that are non-productive and tend to impede growth than business loans that tend to promote growth in the economy. Recent studies (e.g., Law and Singh, 2014; Rioja and Valev, 2004; Sahay et al., 2015) examine the effect of financial development on economic growth and find that the relationship between financial development and economic growth is non-linear. Financial development tends to have a positive effect on growth only up to a certain threshold. Beyond this point of financial development, the benefits of financial development start to decline. The level of financial development for which the positive effect on economic growth begins to decline varies. Berkes et al. (2012) find that financial development starts having a negative effect on output growth when credit to the private sector reaches 100% of GDP. Meanwhile, the study of Law and Singh (2014) indicates that the private sector credit threshold level is 88% of GDP, but good institutions could moderate the negative impact of financial development on growth. Using a new, broad measure of financial development that assesses depth, access and efficiency of financial markets and institutions, Sahay et al. (2015) find that the level of financial development for which the positive effects on growth begin to decline lies between 0.4 and 0.7 on the financial development index. The literature suggests that the growth-enhancing effect of financial development tends to diminish at higher level of financial development (e.g., Aghion et al., 1999; Cecchetti and Kharroubi, 2015; Rioja and Valev, 2004; Sahay et al., 2015) as the quality of finance deteriorates (e.g., the allocation of financial resources toward non-productive activities, the increasing composition of consumer credits relative to business credits, and the shifting of human capital to financial sector). When the financial sector becomes too large, it may compete with the rest of the economy for scarce resources and crowd out resources away 6

7 from the real sector. Financial development, if proceeds too fast, could also lead to economic and financial instability as enlarging the size of financial institutions encourage risk-taking and high leverage, especially when they are poorly regulated and supervised (Sahay et al., 2015). Overall, the results of prior studies are somewhat inconclusive. The relationship between financial openness and financial development and the finance effect on growth could be either positive or negative. Financial openness could strengthen (weaken) the development of domestic financial sector. In a similar vein, the development of domestic financial markets may positively (adversely) affect the country s economic growth. Based on the above discussion, we propose two testable hypotheses as follows: Hypothesis 1: Countries with a high degree of financial openness would exhibit a high degree of financial development. Hypothesis 2: Countries with a high degree of financial development would exhibit a high degree of economic growth. Hypothesis 2a: Countries with a high degree of banking sector development would exhibit a high degree of economic growth. Hypothesis 2b: Stock market development strengthens the influence of banking sector development on economic growth. More specifically, countries with stock markets would have a higher degree of economic growth than countries without stock markets. 3 Data and variables Our sample consists of 119 countries in Asia (34 countries), Europe (49 countries), and North and South America (36 countries). We retrieve annual macroeconomic data over the period from Datastream and World Databank of the World Bank. We primarily 7

8 use the estimation technique of ordinary least squares (OLS) regressions to assess the impact of financial openness on the development of domestic financial markets and to test the effect of financial development on economic growth. First, we estimate a series of panel OLS regressions of financial openness on domestic financial development and a set of control variables. As appropriate, we include country fixed-effects to control for omitted time-invariant country characteristics, period fixed-effects to control for any unobserved time-variant effects, or both. We rely on Hausman tests to indicate whether fixed-effects estimates are preferred to random-effects estimates. Consistent with prior studies such as that of Chang et al. (2008), we deal with the potential endogeneity problem by lagging all right-hand side variables by one period, which should address reversecausality concerns (e.g., causal effects running from financial development to independent variables). Accordingly, we estimate regressions of contemporaneous financial development on lagged financial openness and lagged control variables as follows: FD i,t 1 FO i,t 1 CON i,t 1 i t i,t, (1) where i and t index country i and time t, respectively. FD i,t is the indicator of financial development for country i at time t; FO i,t-1 denotes the indicator of financial openness for country i at time t-1; CON i,t-1 is a vector of country-level control variables for country i at time t-1; η i is a country-fixed effect; ν t is a period-fixed effect, and i,t is the zero-mean disturbance term. In line with prior studies (Baltagi et al., 2009; Herwartz and Walle, 2014; Law and Singh, 2014; Rajan and Zingales, 2003), we use four measures of financial development as follows: FD1 is measured as domestic credit provided by financial sector as a percentage of GDP, FD2 is measured as domestic credit to private sector as a percentage of GDP, FD3 is 8

9 measured as market capitalization of listed domestic companies as a percentage of GDP, and FD4 is measured as money supply (M2) as a percentage of GDP 1. Consistent with the literature (e.g., Herwartz and Walle, 2014; Lane and Milesi-Ferretti, 2007; Rajan and Zingales, 2003), this study assesses financial openness (FO) in terms of the quantity of foreign assets and liabilities as measured by net foreign assets to GDP. 2 Differences in fundamental economic variables seem to be important determinants that explain why some countries have more economic growth than others (Park, 2013). Therefore, a set of control variables are included in regression (1), namely GDP growth ( GDP) as measured by the first difference in the natural logarithm of real GDP per capita, government spending (GOV) as measured by the final consumption expenditures of general government as a percentage of GDP, inflation (INF) as measured by the GDP deflator in annual percentage, interest rate (INT) as measured by the real interest rate in percentage, investment (INV) as measured by gross fixed capital formation as a percentage of GDP, and trade openness (TRADE) as measured by the total trade as a percentage of GDP. Standard errors that are adjusted for heteroskedasticity and serial correlation are clustered at the country level. Second, to test the effects of financial development on economic growth, we estimate regression (2) as follows: 1 Monetary aggregates (M2 or M3) were used as proxy for financial development in earlier studies such as de Gregorio and Guidotti (1995) and Law and Singh (2014). This measure of financial development reflects the overall development of a country s financial sector while private sector credit and stock market capitalization measure banking sector development and stock market development, respectively. 2 This study uses de facto financial openness measure as the measure of financial openness basing on legal restrictions that countries announced do not capture the degree of enforcement of controls on international capital movements, which can change over time even if the legal restrictions that the countries announced remain unchanged. 9

10 GDP i,t 1 FD i,t 1 CON i,t 1 i t i,t, (2) where GDP i,t denotes the first differential of the natural logarithm of real GDP per capita for country i at time t; FD i,t-1 denotes the indicator of financial development for country i at time t-1 as measured by four proxies (i.e., domestic credit provided by financial sector as a percentage of GDP (FD1), domestic credit to private sector as a percentage of GDP (FD2), market capitalization of listed domestic companies as a percentage of GDP (FD3) and M2 as of percentage of GDP (FD4); CON i,t-1 is a vector of country-level control variables for country i at time t-1; η i is a country-fixed effect; ν t is a period-fixed effect, and i,t is the zeromean disturbance term. The set of control variables in regression (2) is the same as before. Basing on the results of Hausman tests, we include country fixed-effects, period fixed-effects or both, when appropriate. 4 Empirical results and discussions 4.1 Descriptive statistics and univariate analysis Table 1 reports descriptive statistics for key variables in the full sample. To minimize the effects of outliers and/or recording errors, we winsorize all variables at the 5% and 95% quantiles. As can be seen in Table 1, during the sample period, GDP growth ( GDP) is about 2% per annum on average. Financial openness (FO) is about 16% on average. All financial development indicators (FD1, FD2, FD3, and FD4) have the mean value of around 51%-71% of GDP on average. Government spending (GOV) has the mean value of 15% of GDP while trade openness (TRADE) has the mean value of 74% of GDP. 10

11 (INSERT TABLE 1 ABOUT HERE) Table 2 presents correlation coefficients for key variables for the sample of 1,522 country-year observations during the period We find that GDP growth ( GDP) is positively correlated with financial openness but it is negatively associated with financial development (FD1). Investment (INV) is positively correlated with FD1 while trade openness (TRADE) is highly and positively correlated with financial openness (FO). (INSERT TABLE 2 ABOUT HERE) 4.2 The effects of financial openness on financial development In this section, we test whether financial openness affects financial development. As discussed earlier, we use four proxies to measure different dimensions of financial development. As Hausman tests suggest that fixed-effects models are preferred to randomeffects models, we estimate panel OLS regressions with country- and year-fixed effects. Table 3 reports panel OLS regression results of financial development. In Models 1-4, the dependent variable is FD1, FD2, FD3, and FD4, respectively. The results in Models 1 and 2 suggest that financial openness (FO) has a negative effect on FD1, which reflects the degree of banking sector development, and FD2, which reflects the relative size of credit markets. The results in Model 3 indicate that financial openness has no effect on FD3, which measures the level of stock market development. The results in Model 4 show that financial openness has a positive effect on FD4, which reflects the overall development of a country s financial sector. Overall, the results in Table 3 suggest that financial openness appears to 11

12 weaken the development of the banking sector (FD1) and the credit markets (FD2) but strengthen the amount of money supply (FD4) in the financial system. Measuring in terms of the overall development of financial market (FD4), our results are generally in line with prior studies (e.g., Baltagi et al. (2009) suggesting that financial openness induces financial development. Similar to the results of Baltagi et al. (2009), we also find that trade openness promotes financial development measuring in terms of the banking sector development (FD1) and the overall financial development (FD4). Contrary to the results of de la Torre et al. (2007) and Hauner, Prati and Bircan (2013), we find that financial openness has insignificant effect on stock market development. Financial openness appears to affect the development of financial markets differently. While financial openness promotes the overall financial development, it hampers the development of banking sector and credit markets. Financial openness could adversely affect banking sector and credit markets development by increased volatility of capital flows and asset prices, and more pronounced macroeconomic fluctuations (Herwartz and Walle, 2014; Umutlu et al., 2010; Yilmazkuday, 2011). Investment (INV) seems to have a strong and positive effect on credit markets development while interest rate is strongly and positively associated with the overall financial development. (INSERT TABLE 3 ABOUT HERE) 4.3 The effects of financial development on economic growth In this section, we test whether financial development affects economic growth. We measure economic growth using real GDP growth ( GDP), which is measured as the first difference in the natural logarithm in real GDP per capita. As Hausman tests suggest that 12

13 fixed-effects models are preferred to random-effects models, we estimate panel OLS regressions on growth with country- and year-fixed effects. Table 4 presents the results of panel OLS regressions of GDP growth. We separately add each proxy for financial development (i.e., FD1, FD2, FD3, and FD4) in Models 1 through 4, respectively. Overall, the results in Models 1 through 4 indicate that financial development has a positive effect on GDP growth. Countries with a higher degree of financial development would have higher economic growth than those with a lower degree of financial development. Our result is in line with prior studies (Law and Singh, 2014; Rajan and Zingales, 2003). Financial development seems to promote economic growth as suggested by the theories that higher levels of financial development increase efficiency of capital mobility, improve resource allocation, facilitate trade, diversification and management of risk, and promote information sharing and loan monitoring (Sahay et al., 2015). We also find that trade openness (TRADE) and financial openness (FO) have a positive effect on economic growth while inflation (INF) and real interest rate have a negative effect on economic growth. Interestingly, government spending (GOV) is not associated with economic growth. To test whether the presence of stock markets affects economic growth, we create a new variable, FD3M, which takes a value of FD3 or a value of zero when the value of FD3 is missing (which indicates no presence of the stock market). As the coefficient on FD3M in Model 5 is not statistically significant, this result appears to suggest that countries with stock markets do not necessarily have higher economic growth than countries without stock markets. To test whether the presence of stock markets weakens the influence of the banking sector on economic growth, we interact FD3M with FD1 in Model 6. The results show that the coefficient on FD1 remains positive and statistically significant while the coefficient on the interaction term is not statistically significant. Taken together, the results suggest that 13

14 stock market development does not weaken the positive effect of banking sector development on economic growth. Overall, the findings in Table 4 indicate that financial development enhances economic growth after controlling for government spending, inflation, real interest rate, investment, trade openness and financial openness. 5 Conclusion This study empirically examines the relationship between financial openness and financial development and tests the effect of finance on economic growth for a sample of 119 developing and advanced countries in Asia, Europe, and Americas over the period Financial openness appears to negatively affect financial development measuring in terms of the levels of banking sector development and credit markets development. Financial openness positively affects financial development, as measured by the amount of monetary aggregates in the financial system, but it has no effect on the level of stock market development. The effect of financial development on economic growth is positive. Countries with a higher degree of financial development seem to have higher economic growth than those with lower degrees of financial development. In addition, trade openness strengthens financial development and promotes economic growth. References Aghion, P., Banerjee, A., Piketty, T., Dualism and Macroeconomic Volatility. The Quarterly Journal of Economics 114, Baltagi, B.H., Demetriades, P.O., Law, S.H., Financial development and openness: Evidence from panel data. Journal of Development Economics 89,

15 Barajas, A., Chami, R., Yousefi, R., The Finance and Growth Nexus Re-Examined; Do All Countries Benefit Equally? International Monetary Fund. Berkes, E.G., Panizza, U., Arcand, J.-L., Too Much Finance? International Monetary Fund. Cecchetti, S.G., Kharroubi, E., Why does financial sector growth crowd out real economic growth? Bank for International Settlements. Chang, E.J., Guerra, S.M., Lima, E.J.A., Tabak, B.M., The stability concentration relationship in the Brazilian banking system. Journal of International Financial Markets, Institutions and Money 18, de la Torre, A., Gozzi, J.C., Schmukler, S.L., Stock market development under globalization: Whither the gains from reforms? Journal of Banking & Finance 31, Hauner, D., Prati, A., Bircan, C., The interest group theory of financial development: Evidence from regulation. Journal of Banking & Finance 37, Herwartz, H., Walle, Y.M., Openness and the finance-growth nexus. Journal of Banking & Finance 48, Lane, P.R., Milesi-Ferretti, G.M., The external wealth of nations mark II: Revised and extended estimates of foreign assets and liabilities, Journal of International Economics 73, Law, S.H., Singh, N., Does too much finance harm economic growth? Journal of Banking & Finance 41, Nili, M., Rastad, M., Addressing the growth failure of the oil economies: The role of financial development. The Quarterly Review of Economics and Finance 46, Park, C.-Y., Asian Capital Market Integration: Theory and Evidence, ADB Economics Working Paper Series. No. 351, Asian Development Bank, Manila. 15

16 Rajan, R.G., Zingales, L., The great reversals: the politics of financial development in the twentieth century. Journal of Financial Economics 69, Ranciere, R., Tornell, A., Westermann, F., Decomposing the effects of financial liberalization: Crises vs. growth. Journal of Banking & Finance 30, Rioja, F., Valev, N., Does one size fit all?: a reexamination of the finance and growth relationship. Journal of Development Economics 74, Sahay, R., Cihak, M., N'Diaye, P.M., Barajas, A., Ayala Pena, D.B., Bi, R., Gao, Y., Kyobe, A.J., Nguyen, L., Saborowski, C., Svirydzenka, K., Yousefi, R., Rethinking Financial Deepening : Stability and Growth in Emerging Markets, IMF Discussion Note, SDN/15/08. Umutlu, M., Akdeniz, L., Altay-Salih, A., The degree of financial liberalization and aggregated stock-return volatility in emerging markets. Journal of Banking & Finance 34, Yilmazkuday, H., Thresholds in the Finance-Growth Nexus: A Cross-Country Analysis. The World Bank Economic Review. 16

17 Table 1: Summary statistics of key variables This table displays summary statistics for key variables in the full sample period. Financial openness (FO) is measured as the net foreign assets as a percentage of GDP. FD1 is measured as domestic credit provided by financial sector as a percentage of GDP. FD2 is measured as domestic credit to private sector as a percentage of GDP. FD3 is measured as market capitalization of listed domestic companies as a percentage of GDP. FD4 is measured as money supply (M2) as a percentage of GDP. GDP growth ( GDP) is measured as the first difference in the natural logarithm of real GDP per capita. Government spending (GOV) is measured as the final consumption expenditures of general government as a percentage of GDP. Inflation (INF) is measured as the GDP deflator in annual percentage. Interest rate (INT) is measured as the real interest rate in percentage. Investment (INV) is measured as gross fixed capital formation as a percentage to GDP. Trade openness (TRADE) is measured as the total trade as a percentage to GDP. Mean Median S.D. Minimum Maximu m Observation s GDP ,833 FO ,540 FD ,591 FD ,608 FD ,766 FD ,333 GOV ,517 INF ,832 INT ,809 INV ,538 TRADE ,685 17

18 Table 2: Correlation coefficients This table reports correlation coefficient for the sample of 1,522 country-year observations during the period Financial openness (FO) is measured as the net foreign assets as a percentage of GDP. FD1 is measured as domestic credit provided by financial sector as a percentage of GDP. FD2 is measured as domestic credit to private sector as a percentage of GDP. FD3 is measured as market capitalization of listed domestic companies as a percentage of GDP. FD4 is measured as money supply (M2) as a percentage of GDP. GDP growth ( GDP) is measured as the first difference in the natural logarithm of real GDP per capita. Government spending (GOV) is measured as the final consumption expenditures of general government as a percentage of GDP. Inflation (INF) is measured as the GDP deflator in annual percentage. Interest rate (INT) is measured as the real interest rate in percentage. Investment (INV) is measured as gross fixed capital formation as a percentage to GDP. Trade openness (TRADE) is measured as the total trade as a percentage to GDP. Symbols ***, **, and * denote statistical significance at the 1%, 5% and 10% levels, respectively GDP FO 0.113*** FD *** 0.045* GOV *** *** 0.168*** INF 0.074*** *** *** *** INT *** *** *** INV 0.261*** 0.204*** 0.062** *** *** TRADE 0.116*** 0.544*** *** *** 0.192***

19 Table 3: OLS regressions of financial development. This table presents the results of panel OLS regressions of the effect of financial openness on financial development for a sample of 88 countries during the period Financial openness (FO) is measured as the net foreign assets as a percentage of GDP. Financial development is measured as FD1, FD2, FD3, and FD4. FD1 is measured as domestic credit provided by financial sector as a percentage of GDP. FD2 is measured as domestic credit to private sector as a percentage of GDP. FD3 is measured as market capitalization of listed domestic companies as a percentage of GDP. FD4 is measured as money supply (M2) as a percentage of GDP. Please see other variable definitions in Table 1. Country- and year-fixed effects are included in all models. Standard errors, which are reported in parentheses, are robust to heteroskedasticity and serial correlation and are clustered at the country level. Symbols ***, **, and * denote statistical significance at the 1%, 5% and 10% levels, respectively. (1) (2) (3) (4) Variable FD1 FD2 FD3 FD4 Constant *** *** * * (13.006) (11.035) (21.516) (11.393) GDP *** *** *** *** (19.266) (15.882) (37.573) (15.192) GOV * (0.525) (0.451) (0.822) (0.481) INF ** (0.168) (0.133) (0.320) (0.134) INT *** (0.206) (0.160) (0.352) (0.154) INV *** 0.754* (0.246) (0.203) (0.389) (0.209) TRADE 0.160** *** (0.070) (0.060) (0.104) (0.059) FO *** *** *** (0.097) (0.080) (0.125) (0.083) Country-fixed effects Yes Yes Yes Yes Year-fixed effects Yes Yes Yes Yes R Adjusted R F-statistic *** *** *** *** Countries included Observations 1,452 1, ,304 19

20 Table 4: OLS regressions of economic growth. This table presents the results of panel OLS regressions of the effect of financial development on economic growth for a sample of 88 countries during the period Economic growth is measured as GDP growth ( GDP), which is computed as the first difference in the natural logarithm of real GDP per capita. Financial openness (FO) is measured as the net foreign assets as a percentage of GDP. Financial development is measured as FD1, FD2, FD3, and FD4. FD1 is measured as domestic credit provided by financial sector as a percentage of GDP. FD2 is measured as domestic credit to private sector as a percentage of GDP. FD3 is measured as market capitalization of listed domestic companies as a percentage of GDP. FD3M is FD3 or takes a value of zero when the value of FD3 is missing. FD4 is measured as money supply (M2) as a percentage of GDP. Please see other variable definitions in Table 1. Country- and year-fixed effects are included in all models. Standard errors, which are reported in parentheses, are robust to heteroskedasticity and serial correlation and are clustered at the country level. Symbols ***, **, and * denote statistical significance at the 1%, 5% and 10% levels, respectively. Variable (1) (2) (3) (4) (5) (6) Constant 0.038** 0.039*** 0.053** ** (0.015) (0.015) (0.021) (0.016) (0.015) (0.015) GOVGDP * (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) INFLATION *** *** * *** *** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) INT ** *** ** ** *** *** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) INVGDP *** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) TRADEOPEN 0.000*** 0.000*** *** 0.000** 0.000*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) FAGDP 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) FD *** 0.000*** (0.000) (0.000) FD *** (0.000) FD *** 20

21 (0.000) FD *** (0.000) FD3M (0.000) (0.000) FD1 x FD3M (0.000) Country-fixed effects Yes Yes Yes Yes Yes Yes Year-fixed effects Yes Yes Yes Yes Yes Yes R Adjusted R F-statistic 9.790*** 9.739*** *** 9.022*** 9.505*** 9.703*** Countries included Observations 1,471 1, ,314 1,486 1,471 21