FINANCIAL SYSTEMS, THE BRICS AND ECONOMIC PERFORMANCE Marcelo Bianconi Department of Economics Tufts University Joe A. Yoshino Department of Economics University of Sao Paulo EEA-NYC
What is this paper about? Does market versus bank-based financial systems impact macroeconomic performance?
What is this paper about? The Test Sample: Under a given classification of countries into market versus bank-based financial system, where do BRICS fit?
Bank-based vs Market Based Financial Systems Bank-based financial system: banks play a leading role in mobilizing savings, allocating capital, overseeing the investment decisions of corporate managers, and providing risk management vehicles.
Bank-based vs Market Based Financial Systems Bank-based financial system: banks play a leading role in mobilizing savings, allocating capital, overseeing the investment decisions of corporate managers, and providing risk management vehicles. Market-based financial system: securities markets share center stage with banks in terms of getting society s savings to firms, exerting corporate control, and easing risk management.
Bank-based vs Market Based Financial Systems MICHIEL J. BIJLSMA AND GIJSBERT T. J. ZWART THE CHANGING LANDSCAPE OF FINANCIAL MARKETS IN EUROPE, THE UNITED STATES AND JAPAN 2013, WP Brueguel Institute, Feb. Levine, Ross (2011) Regulating Financial Markets and Institutions to Promote Growth, in Proceedings of the Jackson Hole Economic Policy Symposium 2011 Demirguc-Kunt, A. and Ross Levine BANK-BASED AND MARKET-BASED FINANCIAL SYSTEMS: CROSS-COUNTRY COMPARISONS Development Research Group, The World Bank, and Finance Department, University of Minnesota WP, 2000 Rajan, R. and L. Zingales (2003) Banks and Markets: The Changing Character of European Finance, in The transformation of the European financial system, eds. V. Gaspar, P. Hartmann, O. Sleijpen, Frankfurt, ECB
Data Unbalanced Panel: 34 countries, yearly observations 1989-2013 (25 years)
Data Country aggregation Bijlsma and Zwart (2013) via PCA Market based EU: Netherlands, UK, Belgium, France, Finland and Sweden Bank based EU: Austria, Denmark, Germany, Greece, Italy, Portugal and Spain Eastern EU: Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, Slovenia Outliers: Ireland, Malta, Cyprus, Luxembourg BRICS: Brazil, Russia, India, China, South Africa US Traditional market-based Japan Traditional bank-based
Data: Equity funding Stock market capitalization BRICS, Bank-based, Mkt-based, Outliers, Eastern EU, US_Japan smc 0 1 2 3 smc 0 1 2 3 smc 0 1 2 3 smc 0 1 2 3 smc 0 1 2 3 smc 0 1 2 3
Data: Equity funding Growth of stock market capitalization BRICS, Bank-based, Mkt-based, Outliers, Eastern EU, US_Japan g_smc -2 0 2 4 6 g_smc -2 0 2 4 6 g_smc -2 0 2 4 6 g_smc -2 0 2 4 6 g_smc -2 0 2 4 6 g_smc -2 0 2 4 6
Data: Bank-intermediated credit Credit to Private Sector BRICS, Bank-based, Mkt-based, Outliers, Eastern EU, US_Japan crdps 0 1 2 3 crdps 0 1 2 3 crdps 0 1 2 3 crdps 0 1 2 3 crdps 0 1 2 3 crdps 0 1 2 3
Data: Bank-intermediated credit Growth of credit to Private Sector BRICS, Bank-based, Mkt-based, Outliers, Eastern EU, US_Japan g_crdps -2-1 0 1 2 g_crdps -2-1 0 1 2 g_crdps -2-1 0 1 2 g_crdps -2-1 0 1 2 g_crdps -2-1 0 1 2 g_crdps -2-1 0 1 2
Data: Macroeconomic Performance Growth of real GDP BRICS, Bank-based, Mkt-based, Outliers, Eastern EU, US_Japan grgdp -.3 -.2 -.1 0.1.2 grgdp -.3 -.2 -.1 0.1.2 grgdp -.3 -.2 -.1 0.1.2 grgdp -.3 -.2 -.1 0.1.2 grgdp -.3 -.2 -.1 0.1.2 grgdp -.3 -.2 -.1 0.1.2
Econometric Models and Empirical Results Panel regressions: 0 1 1 2 3 4 5 1 2 _ _ 3 _ 4 5 _ where i indexes the country, t indexes the year, interactions are between stock market capitalization and credit to the private sector (and growth rates) and the classifications; trend is a time trend.
Econometric Models and Empirical Results grgdp grgdp grgdp grgdp ---------------------------------------------------------------------------- L.grgdp 0.471*** 0.412*** 0.325*** 0.289*** smc 0.0108** 0.0115 0.0187*** 0.0617** crdps -0.0137** -0.0126* -0.0208*** -0.0603*** int_bric_smc -0.0482* int_bric_g~c 0.0153* int_bric_c~s 0.0761*** i~d_eu_crdps 0.0385*** int_mkt~_smc -0.0452* int_mkt~gsmc 0.0138* int_m~_crdps 0.0485*** int_out_cr~s 0.0291* FE Y Y Country FE Y Y Trend -0.000190-0.000390-0.000819** -0.000907** _cons 0.403 0.806 1.656** 1.871** ---------------------------------------------------------------------------- N 683 683 683 683 adj. R-sq 0.316 0.339 0.638 0.651 AIC -2810.1-2828.5-3261.9-3296.8 ---------------------------------------------------------------------------- * p<0.05, ** p<0.01, *** p<0.001 Note: All other variables included but not statistically significant. int_bric_smc = bric * smc int_bric_g~c = bric * gsmc int_bric_c~s = bric * crdps i~d_eu_crdps = bank_based_eu * crdps int_mkt~_smc = mkt_based * smc int_mkt~gsmc = mkt_based* gsmc int_m~_crdps = mkt_based * crdps int_out_cr~s = outliers * crdps
Econometric Models and Empirical Results grgdp grgdp grgdp grgdp ---------------------------------------------------------------------------- L.grgdp 0.471*** 0.412*** 0.325*** 0.289*** smc 0.0108** 0.0115 0.0187*** 0.0617** crdps -0.0137** -0.0126* -0.0208*** -0.0603*** int_bric_smc -0.0482* int_bric_g~c 0.0153* int_bric_c~s 0.0761*** i~d_eu_crdps 0.0385*** int_mkt~_smc -0.0452* int_mkt~gsmc 0.0138* int_m~_crdps 0.0485*** int_out_cr~s 0.0291* FE Y Y Country FE Y Y Trend -0.000190-0.000390-0.000819** -0.000907** _cons 0.403 0.806 1.656** 1.871** ---------------------------------------------------------------------------- N 683 683 683 683 adj. R-sq 0.316 0.339 0.638 0.651 AIC -2810.1-2828.5-3261.9-3296.8 ---------------------------------------------------------------------------- * p<0.05, ** p<0.01, *** p<0.001 Note: All other variables included but not statistically significant. int_bric_smc = bric * smc int_bric_g~c = bric * gsmc int_bric_c~s = bric * crdps i~d_eu_crdps = bank_based_eu * crdps int_mkt~_smc = mkt_based * smc int_mkt~gsmc = mkt_based* gsmc int_m~_crdps = mkt_based * crdps int_out_cr~s = outliers * crdps
Econometric Models and Empirical Results grgdp grgdp grgdp grgdp ---------------------------------------------------------------------------- L.grgdp 0.471*** 0.412*** 0.325*** 0.289*** smc 0.0108** 0.0115 0.0187*** 0.0617** crdps -0.0137** -0.0126* -0.0208*** -0.0603*** int_bric_smc -0.0482* int_bric_g~c 0.0153* int_bric_c~s 0.0761*** i~d_eu_crdps 0.0385*** int_mkt~_smc -0.0452* int_mkt~gsmc 0.0138* int_m~_crdps 0.0485*** int_out_cr~s 0.0291* FE Y Y Country FE Y Y Trend -0.000190-0.000390-0.000819** -0.000907** _cons 0.403 0.806 1.656** 1.871** ---------------------------------------------------------------------------- N 683 683 683 683 adj. R-sq 0.316 0.339 0.638 0.651 AIC -2810.1-2828.5-3261.9-3296.8 ---------------------------------------------------------------------------- * p<0.05, ** p<0.01, *** p<0.001 Note: All other variables included but not statistically significant. int_bric_smc = bric * smc int_bric_g~c = bric * gsmc int_bric_c~s = bric * crdps i~d_eu_crdps = bank_based_eu * crdps int_mkt~_smc = mkt_based * smc int_mkt~gsmc = mkt_based* gsmc int_m~_crdps = mkt_based * crdps int_out_cr~s = outliers * crdps
Econometric Models and Empirical Results Panel Vector Autoregressions (VAR) via GMM By classification, bivariate models, each variable and growth of real GDP, lag length one 0 1 1 2 where i indexes the country, t indexes the year, interactions are between stock market capitalization and credit to the private sector (and growth rates) and the classifications and trend is a time trend. Financial Development and Dynamic Investment Behavior: evidence from Panel VAR (Inessa Love with Lea Ziccino), The Quarterly Review of Economics and Finance, 46 (2006), 190-210.
Econometric Models and Empirical Results: BRICS g_smc g_smc:g_smc grgdp:g_smc -.2 0.2.4.6 -.1 0.1.2.3 -.005 0.005.01.015 g_smc:grgdp 0.005.01.015.02.025 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_smc grgdp 1.116 1 0.291 ALL 1.116 1 0.291 grgdp g_smc 28.011 1 0.000 ALL 28.011 1 0.000
Econometric Models and Empirical Results: BRICS g_crdps g_crdps:g_crdps grgdp:g_crdps 0.05.1.15 -.1 -.05 0.05 -.01 -.005 0.005 g_crdps:grgdp 0.01.02.03 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_crdps grgdp 0.145 1 0.704 ALL 0.145 1 0.704 grgdp g_crdps 8.975 1 0.003 ALL 8.975 1 0.003
Econometric Models and Empirical Results: Bank-based - smc smc:smc grgdp:smc -.1 -.05 0.05.1.15 -.06-.04-.02 0.02.04 -.04 -.02 0.02.04.06 smc:grgdp -.02 0.02.04 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 smc grgdp 1.212 1 0.271 ALL 1.212 1 0.271 grgdp smc 4.335 1 0.037 ALL 4.335 1 0.037
Econometric Models and Empirical Results: Bank-based g_smc 0.1.2.3.4 g_smc:g_smc -.04 -.02 0.02.04 grgdp:g_smc g_smc:grgdp grgdp:grgdp 0.005.01.015 0.005.01.015.02 impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_smc grgdp 0.094 1 0.759 ALL 0.094 1 0.759 grgdp g_smc 22.112 1 0.000 ALL 22.112 1 0.000
Econometric Models and Empirical Results: Bank-based crdps 0.05.1.15 crdps:crdps -.08 -.06 -.04 -.02 0 grgdp:crdps -.01 -.005 0.005.01 crdps:grgdp -.01 0.01.02.03 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 crdps grgdp 4.816 1 0.028 ALL 4.816 1 0.028 grgdp crdps 11.004 1 0.001 ALL 11.004 1 0.001
Econometric Models and Empirical Results: Bank-based g_crdps 0.05.1.15 g_crdps:g_crdps 0.005.01.015.02.025 grgdp:g_crdps -.002 0.002.004.006.008 g_crdps:grgdp 0.005.01.015.02.025 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_crdps grgdp 6.987 1 0.008 ALL 6.987 1 0.008 grgdp g_crdps 0.311 1 0.577 ALL 0.311 1 0.577
Econometric Models and Empirical Results: Market-based - smc 0.1.2.3 smc:smc -.04 -.02 0.02.04 grgdp:smc 0.01.02.03 smc:grgdp -.01 0.01.02.03 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 smc grgdp 0.014 1 0.905 ALL 0.014 1 0.905 grgdp smc 6.184 1 0.013 ALL 6.184 1 0.013
Econometric Models and Empirical Results: Market-based g_smc -.1 0.1.2.3.4 g_smc:g_smc -.03 -.02 -.01 0.01 grgdp:g_smc 0.005.01.015.02 g_smc:grgdp 0.005.01.015 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_smc grgdp 2.388 1 0.122 ALL 2.388 1 0.122 grgdp g_smc 182.811 1 0.000 ALL 182.811 1 0.000
Econometric Models and Empirical Results: Market-based crdps 0.02.04.06.08.1 crdps:crdps -.005 0.005.01.015 grgdp:crdps crdps:grgdp grgdp:grgdp -.008 -.006 -.004 -.002 0 0.005.01.015.02 impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 crdps grgdp 1.450 1 0.228 ALL 1.450 1 0.228 grgdp crdps 14.320 1 0.000 ALL 14.320 1 0.000
Econometric Models and Empirical Results: Market-based g_crdps g_crdps:g_crdps grgdp:g_crdps -.05 0.05.1.15 -.01 0.01.02.03 -.006-.004-.002 0.002 g_crdps:grgdp 0.005.01.015.02.025 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_crdps grgdp 0.351 1 0.553 ALL 0.351 1 0.553 grgdp g_crdps 0.004 1 0.952 ALL 0.004 1 0.952
Econometric Models and Empirical Results: Outliers g_smc -.2 0.2.4.6 g_smc:g_smc -.04 -.02 0.02.04.06 grgdp:g_smc 0.005.01.015.02.025 g_smc:grgdp 0.01.02.03.04 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_smc grgdp 0.013 1 0.908 ALL 0.013 1 0.908 grgdp g_smc 11.116 1 0.001 ALL 11.116 1 0.001
Econometric Models and Empirical Results: Outliers crdps 0.05.1.15 crdps:crdps -.15 -.1 -.05 0.05 grgdp:crdps -.03 -.02 -.01 0.01.02 crdps:grgdp -.04 -.02 0.02.04 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 crdps grgdp 0.470 1 0.493 ALL 0.470 1 0.493 grgdp crdps 83.374 1 0.000 ALL 83.374 1 0.000
Econometric Models and Empirical Results: Eastern Europe - smc smc:smc grgdp:smc 0.02.04.06.08 -.005 0.005.01 smc:grgdp grgdp:grgdp 0.01.02.03.04 0.01.02.03.04.05 impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 smc grgdp 0.402 1 0.526 ALL 0.402 1 0.526 grgdp smc 36.091 1 0.000 ALL 36.091 1 0.000
Econometric Models and Empirical Results: Eastern Europe g_smc -.2 0.2.4.6.8 g_smc:g_smc 0.05.1.15 grgdp:g_smc -.01 0.01.02.03 g_smc:grgdp 0.01.02.03.04 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_smc grgdp 6.223 1 0.013 ALL 6.223 1 0.013 grgdp g_smc 2.853 1 0.091 ALL 2.853 1 0.091
Econometric Models and Empirical Results: Eastern Europe crdps crdps:crdps grgdp:crdps 0.1.2.3 0.1.2.3 -.01 -.005 0.005 crdps:grgdp -.01 0.01.02.03.04 grgdp:grgdp impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 crdps grgdp 0.416 1 0.519 ALL 0.416 1 0.519 grgdp crdps 1.388 1 0.239 ALL 1.388 1 0.239
Econometric Models and Empirical Results: Eastern Europe g_crdps g_crdps:g_crdps grgdp:g_crdps -.05 0.05.1.15.2 0.05.1 g_crdps:grgdp grgdp:grgdp -.004-.002 0.002.004.006 0.01.02.03.04 impulse variable: response variable Ha: Excluded variable Granger-causes Equation variable Equation \ Excluded chi2 df Prob > chi2 g_crdps grgdp 6.300 1 0.012 ALL 6.300 1 0.012 grgdp g_crdps 0.823 1 0.364 ALL 0.823 1 0.364
In summary: o o o The Question: Does market versus bank-based financial systems impact macroeconomic performance? Yes, bank intermediation (credit/gdp) impacts significantly, regardless of the financial system more is needed to identify the effects
In summary: o o o The Test Sample: Under a given classification of countries into market versus bank-based financial system, where do BRICS fit? Both, stock market cap and credit/gdp impact BRICS macro performance But, growth of stock market cap impact growth more significantly in the BRICS from PVAR, and give a closer fit in panel regressions
Thank You! Comments and suggestions welcome: Marcelo.bianconi@tufts.edu pyoshino@usp.br