Correlations between stock market development and economic growth

Transcription

1 Correlations between stock market development and economic growth - the case of Romania - Author: Năstase Ioana-Alexandra Supervisor: PhD Professor Laura Obreja-Brasoveanu Introduction The relationship between financial development and economic growth has been subject to intense debate in the literature of development and growth. Capital markets, particularly stock markets, are always of interest considering that stock markets attract attention as the paramount symbol of capitalism. During the last two decades, the growing importance of capital markets has opened a new path of research between financial development and economic growth which focuses on the effects of the capital market development. Although, capital markets are still poorly developed in most countries, capital market financing started to be more preferred to the banking financing. This is mainly due to the globalization which diminished the differences between bank dominated and capital dominated systems. Hence, there is a considerable debate in respect of the relative merits of the bank-dominated financial sectors and those that give equal weight or more to capital market. Within this context, the present paper aims to investigate the following questions: i) is it possible to find a statistical significant relationship between stock market development and economic growth? and ii) what is the direction of the between stock market and economic growth?. The answers to these two questions will be provided in light of the evolution of stock market and economic growth in Romania, during period This paper is structured as follows. After this introduction, the next section reviews the main theoretical approaches regarding the correlations between economic growth and financial development. Chapter three presents a review of the empirical studies aimed to analyze the relationship between stock market development and economic growth. Section four presents an overview of the Romanian capital market evolution during the analyzed period. The next section explains the econometric methodology used and presents the data employed in the study. Section six presents the empirical results of the VAR models and regressions performed aimed to test the correlations between economic growth and Romanian capital market. The last chapter is represented by conclusions and reconciliation with other studies.

2 Literature review The significant role of financial development received considerable attention in the recent financial literature. Most of these theoretical studies focus on financial development which involves an analysis of the impact of both the banking sector and capital markets. Most of the models aimed to explain the relationship between economic growth and stock market are based on the endogenous growth model in which economic output is an endogenous variable influenced by physical capital, human capital and level of technology. This approach can be found in the papers of Greenwood and Jovanovic (1990), Pagano (1993), Bolbol et al. (2005), FitzeGerald (2006), Baier et al. (2004). As a distinctive approach, the paper of Graff (1999) analyzes the relationship between financial development and economic growth through four possibilities: Financial development and economic growth are not causally related; Financial development follows economic growth; Financial development is a determinant of economic growth; Financial development is an impediment to economic growth. Thus, no assumption is made as regards the direction of the between economic growth and financial development. In order to assess which variable is endogenous and which is exogenous, the author constructed two models. The first model assumes economic output as an endogenous variable which is determined by financial development. The second model inputs economic growth as an independent variable and financial development as an endogenous variable. Empirical review The major part of the previous empirical studies treated economic growth endogenously and analyzed whether stock market development has any impact on output. There has been found empirical evidence that stock market does have a significant impact on economic growth. Moreover, there is a positive effect of the stock market on the economic growth. In most cases, the studies were performed on a pool of countries, over o significant period of time e.g. Arestis et al. (2001), Levine and Zervos (1998), Naceur and Ghazouani (2007). Nevertheless, there are empirical studies which employed time series for testing this relationship in the case of a sole country e.g. Bolbol et al.(2005), Hondroyiannis (2004), Obreja-Brasoveanu et a.l (2008). A summary of the previous empirical studies on this theme is presented below: 2 P age

3 Authors Antonios (2010) Arestis, Demetriades, Luintel (2001) Augustine, Pius (2010) Bolbol, Fatheldin, Omran (2005) Deb, Mukhrjee (2008) Duca (2007) Ergungor (2003) Correlations between economic growth and stock market empirical studies Econometrical methods - VEC model - pooled crosssection /time series regression - multiple regression (OLS) - time-series regression - Toda Yamammotto - pooled crosssection/time series regression (OLS) Variables used Economic Growth Stock market Results real GDP per capita GDP growth rate - stock market index - ln market capitalization/gdp ratio - market volatility (eight quarter moving standard deviation of the end-of quarter change of stock market prices) - market capitalization/gdp ratio - total value of shares traded/gdp ratio - turnover ratio - market capitalization/gdp ratio - turnover ratio - real market capitalization/gdp ratio - real value traded/gdp ratio - stock market volatility positive effect * Non-concludent effect* Positive effect* positive effect* Positive effect* nominal GDP - stock market indices Positive effect* - real per capita GDP growth rate; - per capita capital stock growth rate; - productivity growth rate - ln(value of domestic equities traded on domestic exchanges/value of deposit money bank credits to the private sector) Non-concludent effect* Filer, Hanousek, Campos (2000) Hondroyiannis, Lolos, Papapetrou (2004) Hsing (2011) Humpe, MacMillan(2001) Leahy, Schich, Wehinger, Pelgrin (2001) Levine, Zervos (1998) Liu, Hsu (2006) - time series regression - Johansen cointegration - pooled crosssection/time series regression - pooled crosssection/time series regression - pooled crosssection/time series regression (GMM) real GDP - market capitalization/gdp ratio - domestic shares listed growth rate - market capitalization - industrial stock capitalization Non-concludent effect* Positive effect* real GDP - Czech stock market index Positive effect** real industrial production real per capita GDP growth - stock market indices - market capitalization - market capitalization - domestic turnover - value traded - standard deviation of monthly returns - market capitalization to GDP ratio - turnover ratio - stock return positive effect** positive effect* positive effect* positive effect* Naceur, Ghazouani (2007) Nieuwerburgh, Buelens, Cuyvers (2005) Obreja-Brasoveanu, Dragota, Catarama, Semenescu (2008) - pooled crosssection/time series regression (GMM) - Johansen cointegration - VAR model - linear regression real per capita GDP growth rate log difference in real per capita GDP * stock market development promotes economic growth ** economic growth promotes stock market development - market capitalization/gdp ratio - value traded/gdp ratio - turnover ratio - market capitalization - total number of listed shares - cumulative number of IPO - real market capitalization - real trade volume - liquidity proxy No effect* positive effect* positive effect** 3 P age

4 Overview of the Romanian capital market As a general concept, the capital market includes the stock market and the bond market. In Romania, the capital market is mainly represented by the stock market which is the most developed segment. Only during the last years the bond market started to develop, but the rhythm of development is very slow. Therefore, I focused on the stock market in my empirical study as it representative for the Romanian capital market. The Romanian capital market has its roots back in the 1880s, when the first stock exchange Bucharest Stock Exchange ( BSE ) was opened. However, until 1995, the stock exchange functioned only for limited periods of time. After year 1995, BSE started its development. As key years in the recent history of the stock exchange, 2007 is an important year since BSE operated for the first time as a market operator in the EU integrated financial market which resulted in an increased performance and high liquidity of the stock exchange. The next year, 2008, the stock exchange experienced a downgrading slope of the stock prices as a consequence of the evolution of the international financial markets on the background of a severe financial and economic crisis. The evolution of the main indicators of the stock exchange shows that during BSE experienced an upward trend in respect of performance. Starting with 2008, the stock market incurred a decrease both in size (i.e. market capitalization) and in liquidity (i.e. turnover). However, the stock market recovered quickly. Also, it can be easily noticed that the development of the stock market is more pronounced starting with year 2005 due to the complete liberalization of the Romanian financial market. As for the evolution of the most important indexes (BET, BET-C and BET-FI), they had an ascending trend from the year 2000 and until the beginning of The maximum value achieved by these indices was registered in 2007 and one year later they reached the values incurred in After 2008, all three indices are on an upward trend, although BET and BET-FI register a higher performance than BET. As regards the evolution of the market capitalization to GDP ratio had an ascending trend from 2000 until 2007, from 1.3% to 20.6%. However, in 2008, market capitalization ratio reached a poor level mostly due to the decrease of the market capitalization and not to GDP which recorded an increase in that year. During , both market capitalization and gross domestic product had a steady growth. From 2008, market capitalization experienced fluctuations which were not similar to the ones of GDP. GDP experienced a decrease compared to the previous year only in 2008, which was not the case for market capitalization. Thus, the effects of the crisis were firstly reflected on the stock market and later on economic growth. This could be explained by the high degree of integration of the Romanian capital market with the international financial markets. At a first glance, one might say that GDP is influenced by the capital market. However, further detailed analysis should be performed in order to accurately assess the direction of this relationship, if any. According to the classification of the World Bank based on 2010 gross national income per capita, Romania is part of the upper middle income group ($3,976 - $12,275) 4 P age

5 together with other four EU members - Bulgaria, Latvia and Lithuania. Between , the Romanian capital market quantified by the market capitalization to GDP ratio slightly underperformed the average of its group. Surprisingly, during , the value of this ratio was above the average. This evolution indicates an increased level of development of the Romanian stock market. Compared to each of its peers, Romania underperformed Lithuania and overperformed Latvia and for some years, also Bulgaria. The same findings are true when the countries are compared based on stock traded as percentage of GDP. Data and methodology I tested for the existence of a relationship between the development of the stock market and economic performance for the case of Romania, over the period Dataset used This study uses quarterly data ranging between 2000Q-2011Q4 (48 observations) provided by the National Institute for Statistics and Economic Studies and the Bucharest Stock Exchange. The variables used were: Market capitalization real terms ( MCR ) Trading volume real terms ( TVR ) BET-C index real terms ( BETCR ) Gross domestic product real terms ( GDPR ) The first three variables were selected as proxies of the Romanian stock market. The decision of selecting these variables was made based on the previous empirical evidence and based on the fact that each of them quantify in a different way the stock market. Market capitalization is a size variable, trading volume is a liquidity variable and BET-C index is a proxy of the overall performance of the Romanian stock market. Economic growth was represented by the gross domestic product. All data used was expressed in real terms with the basis in the first quarter of 2000 in order to eliminate the influence of inflation. The data used entered in log because time series, generally, have, overall trends of exponential growth. Methodology For the purpose of testing the relationship between economic growth and stock market development, I employed the VAR methodology, followed by linear regressions. I chose to use this econometric model based on the assumption that it is not known the direction of the relationship between economic growth and stock market. The reduced form of the VAR model can be expressed as: where Y t is a j vector of endogenous variables, X t is a k vector of exogenous variables, and are matrixes of coefficients to be estimated, is a j vector of error terms and m is the number of lags. 5 P age

6 The lags were determined based on the Lag Length Criteria which computes various criteria to select the number of the lag order such as: Likelihood ratio (LR), Akaike information criterion (AIC), Schwartz information criterion (SIC), Hannan-Quinn information criterion (HQ). There were estimated three VAR models for each of the pairs: market capitalization and economic growth, trade volume and economic growth, BET-C index and economic growth. Each of the VAR models is further tested by applying a number of diagnostic tests: AR Roots Table / AR Roots Graphs for testing the stability of the VAR model Portmanteau Test and LM test for finding evidence of autocorrelation; Jarque-Bera Test for testing the normality of the residuals; White Heteroskedasticity Test for testing the heteroskedasticity of residuals. Based on the results of the VAR model, several linear regressions regressions were estimated using the Least Square Method. All the econometric models were performed using Eviews 5.0 Enterprise Edition. Empirical results Descriptive statistics GDPR, MCR and TVR variables have a negatively skewed distribution (long-tail to the left), whereas BETCR has a positively skewed distribution (long-tail to the right). However, the values of the skewness coefficient are closed to 0 indicating that the variables selected are almost normally distributed. Kurtosis coefficient is below 3 for all variables indicating that the time series are platykurtic (i.e. fat-tails distribution). The series are peaked less than normal. The values of the Jarque-Bera test for GDPR, TVR and BETCR are in the normal range which show that follow a distribution closed to the normal distribution. Based on the correlation matrix, the variables are not strongly correlated. The correlation matrix has only positive values which indicate a positive association between variables. Thus, there is a positive but not strong correlation between variables. Seasonality The statistical series have been tested in order to identify seasonal factors using two tools provided by E-views: Seasonal stack graphs and Seasonal split line. I have found that all the series, except TVR, were affected by seasonality. The most affected series by seasonality was GDPR. The series were seasonally adjusted by using the Tramo/Seats method. Stationarity tests 6 P age

7 In order to test for the stationarity of the series, I used the following tests: Augmented Dickey-Fuller (ADF test), Philips-Perron (PP test), Kwiatowski, Philips, Schmidt and Shin s test (KPSS), also called unit-root tests. The combined results from all unit-root tests performed show that all series are nonstationary in their levels, but stationary in their first differences. Thus, the series are integrated of order one I(1). The series were adjusted for non-stationarity by considering first level differences. Granger Causality Test In order to establish the relationship between the variables used, Granger Causality tests were performed. More specific, they are performed to see the potential direction of the relationship between economic output (GDP) and each of the variables which describes the stock market (i.e. BETCR, MCR, TVR). The following conclusions could be made based on the results of the Granger Causality tests: there is a relationship between real BET-C and real GDP, relationship from BET-C towards GDP. However, this is not true for a long-term basis; the relationship between GDP and stock market capitalization is not conclusive. Further analysis should be performed; no causal relationship was found between real trade volume and GDP. Granger tests performed revealed relevant information only in respect of the relationship between GDP and BET-C index. VAR models The relationship between Romanian stock market development and economic growth was tested through three VAR models, one for each of the pairs: market capitalization and GDP, trading volume and GDP, BET-C index and GDP. Each of the VAR models proved to be stable and all the residual tests were passed. The empirical analysis of the correlations between capital market and economic growth led to the following estimated regressions: Estimated regression R 2 R1: dlog(gdpr) t = dlog(gdpr) t dlog(mcr) t 46.4% R2: dlog(mcr) t = dlog(gdpr) t 21.7% R3: dlog(gdpr) t = dlog(gdpr) t dlog(tvr) t 53.6% R4: dlog(tvr) t = dlog(gdpr) dlog(tvr) t % R5: dlog(gdpr) t = dlog(gdpr) t dlog(betcr) t % 7 P age

8 R6: dlog(betcr) t = dlog(betcr) t % The main findings could be summarized as follows: economic growth is positively correlated with stock market development, measured by market capitalization. The values of R 2 are not very high which implies the variables are not strongly correlated. Also, there is a feed-back effect between economic growth and market capitalization with more from market capitalization to economic growth. economic growth is positively correlated with trading volume with feed-back effect There is more influence from the stock market, quantified by the growth of trading volume, toward economic growth. The findings are similar with the first relationship analyzed. there is no bi-directional relationship between economic and stock market development quantified by BET-C index. Economic growth is determined by the growth of BET-C index with one lag. These variables are positively but not very strong correlated. However, the small magnitude of the coefficient indicates that the correlation between these two variables is limited. Conclusions This paper empirically examines the relationship between stock market development and economic growth for Romania, using quarterly data for the period The study is an attempt to answer two questions: whether at all any relationship exists between the stock market development and economic growth and what would be the direction of the relationship (i.e. does economic growth promote the development of stock market or viceversa?). To test this hypothesis, I employed the methodology of unrestricted VAR and OLS regression. The three important indicators of stock market development included in this study are: real market capitalization, real trading volume and real BET-C index. Real GDP is used as a proxy for economic growth. The main findings of this paper could be summarized as follows: firstly, there is a feed-back effect between economic growth and market capitalization with more influence from market capitalization to economic growth. Secondly, there is also a bi-directional relationship between the growth of the trading volume and the growth of gross domestic product with more influence from trading volume to economic growth. Thirdly, BET-C index with one lag determines economic growth. However, the relationship is not valid for the viceversa case. 8 P age

9 The methodology employed indicates there is a positive correlation between economic growth and stock market development, with feed-back effect, but the strongest influence comes from stock market to economic growth. This finding suggests that economic growth follows stock market development. However, the estimated coefficients are small in magnitude which suggests that the correlation between stock market and economy is limited. The limited contribution of stock market to growth should not be surprising considering its minor role played in Romania. The results of this empirical study are consistent with the third possibility of the causal relationship between financial development and economic growth stated by Graff (1999). Furthermore, the results are in line with most of the findings of the previous empirical studies: Antonios (2010), Augustine and Pius (2010), Bolbol et al. (2005), Hodroyiannis (2004), Levine and Zervos (1998), Liu and Hsu (2006) and Niewerbrgh (2005). Based on the results obtained in the case of Romania, future steps should be taken in order to gain from the positive correlation between economic growth and stock market. In this respect, the stock market should further develop at higher rates of growth. This could be obtained by increasing the level of investments in stocks and also the number of investors. There should be granted incentives to investors and also there should be a focus on small investors. In order to increase the investments in the stock market, transaction costs should be lower and transparency and regulation of the stock market should be increased. The number of domestic companies listed on the stock exchange should increase and also foreign companies should be stimulated to be listed on Bucharest Stock Exchange since as at this moment there is only a couple of foreign companies listed. The main idea is that stock market may become an important determinant of economic growth, perhaps as important as fiscal and monetary tools provided that it would develop and would increase the trust of the investors. As a further research, it should be included in the study a variable as a proxy of the stock market volatility. Moreover, it would be useful to quantify the impact of the other sectors of the capital market. I believe that in the next years the bond market would significantly develop and would be more and more preferred by high risk-adverse investors in the detriment of the stock market. Also, it should be taken into account the impact of the derivatives market. Nevertheless, several years or perhaps decades would be needed for the derivatives market to develop and increase its importance in the structure of the Romanian capital market. References: Anghelache, A. Piata de capital in context european, Economica Publishing, 2009 Arestis, P., Demetriades, P. and Luintel, K.B. Financial development and economic growth: the role of stock markets, Journal of Money, Credit, Bank, 33, 2001, pp Barro, Robert J. Economic growth in a cross section of countries, Quarterly Journal of Economics, Vol. 106, No.2, 1991, pp P age

10 Beck, T., Levine, R. and Loayza, N. Finance and sources of growth, Journal of Financial Economics, 58, pp , 2000 Bencivenga, V. and Smith, B. Financial Intermediation and endogenous growth, Review of Economic Studies, 58, 1991, pp Bolbol, A.A. Fatheldin, A. and Omran M.M. Financial development, structure and economic growth: the case of Egypt, , Research in International Business and Finance, 19, 2005, pp Graff M.A. Financial Development and economic growth a new empirical analysis, Dresden Discussion Papers in Economics No.5, 1999 Hondroyiannis, G., Lolos, S. and Papapetrou, E. Financial markets and economic growth in Greece, , Journal of International Financial Markets, Institutions and Money, 15, 2004, pp Humpe, A. and Macmillan, P. Can Macroeconomic Variables Explain Long Term Stock Market Movements? A Comparison of the US and Japan, Center for Research into Industry, Enterprise, Finance and the Firm, Working Paper 11, 2005 King, R.G. and Levine, R. Finance and Growth: Schumpeter might be right, the Quarterly Journal of Economics, vol. 108 (3), 1993, pp Levine, R. and Zervos, S. Stock markets, banks and economic growth, American Economic Review, 88, pp Liu, W.C. and Hsu, C.M. The role of financial development in economic growth: the experiences of Taiwan, Korea and Japan, Journal of Asian Economics, 17, 2006, pp Obreja -Brasoveanu, L., Dragota, V., Catarama, D. and Semenescu, A. Correlations between capital market development and economic growth: The case of Romania, Journal of Applied Quantitative Methods, Vol. 3, No.1, Spring 2008, pp Obreja, L, Dragota, M., Dragota, V., Semenescu, A. and Stancu, I. Piata de capital ASE Publishing, 2010 Stancu I. Finante, Economica Publishing, 2007 Stancu D. and Stancu I. Finante corporative cu Excel, Economica Publishing, P age