Groningen, Netherlands – January 11th, 2021
MSc Finance - Master Thesis
Stock market and banking development as
economic growth determinants.
An empirical research on Eastern Europe
University of Groningen - Faculty of Economics and Business
Abstract:
The goal of this research is to study the long-term relationship, if any, between the stock market, banking development and economic growth in Eastern Europe between 1995 and 2019, using a panel data approach, and then to compare the findings with those obtained using the same methodology in the Western European sample. The findings show a clear and positive relationship between the size of the stock market and economic growth in both areas, while in the banking sector there has been a strong and positive relationship in the Eastern countries, while in the Western countries there has been a negative relationship.
Author: Stefan Ungurasu Student number: S3819221 Supervisor: prof. dr. B.W. Lensink
3
1.Introduction
The relationship between financial development and economic growth has always been under discussion, questioning which leads to the other one. There are two important items in this process, namely stock markets and bank development. In general, the state of the economy is expressed in the stock market. As the economy expands, production will increase, and most businesses should experience enhanced profitability. Also, the other way is valid, when it is predicted a recession, the share prices will drop. The reason behind this phenomenon is explained by the lowered profits and fewer dividends brought by the recession and even by the prospect of firms going bankrupt.
In some historical events, it has been shown that the stock market has an impact on the economy. An example worth mentioning is the Wall Street crash from 1929 to 1932. This rapid decline in stock markets has had a significant impact on business and consumer confidence. It also caused the banks to lose their money. This was undoubtedly a factor in the duration and severity of the Great Depression. There is a counterexample, during the Black Monday of 1987, when the world's stock markets crashed without pushing the economy into a financial recession like previous similar events. Among other scholars, King et al (1993), Levine et at (1998), Rajan et al(1998) and Capasso (2008) concluded that there is a positive correlation between development and economic growth .
Banks are important financial intermediaries in many countries and play a significant role in bridging savings and investment, and it was concluded by McKinnon (1973) and Shaw (1973) that the growth in the economy is inspired by financial development through savings and investments.
Most of the other studies were focused solely on the development of banking, or only on the development of the stock market in developed countries. The objective of this research is to bridge the gap in literature by analyzing the relationship of stock market and banking development with the economic growth in Eastern Europe, during 1995-2019. In addition, the same analysis will be carried out on a sample of Western European countries between 1995-2019, and the results of the two studies will be compared to check how the development level of the region affects the impact of stock market and banking development on the economic growth.
The selected countries have participated in the Warsaw Agreement, which was formed in 1955 and dissolved in 1991. The following study concerns Romania, Bulgaria, Hungary, the Czech Republic, Slovakia, and Poland. The Eastern part of Germany was not included in the study due to its fusion with the Western part of Germany, making the inclusion irrelevant.
4
regime, the stock exchanges in this area were reopened. The operations of the stock exchanges in Bulgaria, Poland, the Czech Republic, Romania, and Hungary were then revived from 1989 to 1997.
The purpose of this paper is to research the impact of stock market and bank development for economic growth in the Eastern Europe and contrast the results with the ones from Western Europe. In order to perform this research, six most important countries were selected from the east of Europe, namely Romania, Bulgaria, Hungary, the Czech Republic, Slovakia, and Poland. Similar for west of Europe were chosen Austria, Belgium, France, Netherlands, Ireland and Switzerland.
In the next section will be presented the existing literature which intends to answer the research question regarding the relevance of the stock market and bank development. Section 3 defines the methodology and Section 4 describes the statistics of the data. In Section 5 the results of this research will be presented, and the last part of the study will conclude.
2. Literature review
There have been many debates regarding the relationship of stock market and bank development with the economic growth. This subject was intensively studied by many scholars in different regions of the world, using different research approaches. The results varied with the countries and time frame under the study.
2.1 Banking development and economic growth
Miller (1988) stated that the relationship of financial markets with the economic growth is positive, too important, and obvious for further discussions. Lucas(1988) argued that this relationship is not necessarily positive and concluded that financial markets in economic development were severely exaggerated in academic discussions. Zingales (2015) provided evidence that the financial system could easily turn into rent-seeking activities during the disturbance period.
In his research, Goldsmith (1969) argued that financial development promotes economic growth. Although the sample size was generous, including a large number of countries over a period of 103 years, his work was criticized for not controlling a number of important factors and failed to provide any conclusions.
country-5
specific indicators, borrowing the Barro (1991) approach, which includes variables that may impact economic growth, such as initial wealth, secondary school enrolment, and population growth. In order to assess financial developments, the researchers included various proxies in their study, such as liquid liabilities of the financial system (percent of GDP), the ratio of bank credit to bank credit plus central bank domestic assets and private sector credit as a percentage of GDP. The conclusion was that the initial level of financial development was a good predictor of economic growth.
La Porta et al (2002) used a different approach for measuring the financial dept sizes. They examined the extent to which a certain feature of the financial system had an impact on the growth of the economy. It was concluded that a higher level of public ownership of the banking sector leads to slower growth, and the authors argued that banks' state ownership had a limited and generally negligible impact on potential investment and had a major negative impact on future growth in productivity.
Beck, Levine, and Loayza (2000) inferred that there is a positive statically significant effect on economic growth of financial development.
In order to be able to deal with heterogeneity, Loayza and Ranciere (2006) used a dynamic panel pooled mean group estimator. This technique was initially used by Pesaran and Smith (1995) and allows for short-term heterogeneous country effects, using fixed effects that control time-invariant unobservable characteristics, while constraining the long-term effect of the regressor to have the same value throughout the entire panel. The advantage of this approach is that financial development can have different impacts across countries. Loayza and Ranciere (2006) concluded that, in the long run, there is a positive, significant and robust relationship between banking development and growth, that in the short run is negative for a group of countries, and that high credit growth can lead to slower growth or even a financial crisis. Rosseau and Wachtel (2002) conducted a similar study, stating that the relationship under study fluctuates with inflation and financial deepening. They concluded that growth is no longer affected when annual inflation exceeds the 13 per cent threshold.
Dawson (2008) concluded that banking development, using historical evidence, has a positive impact on economic growth and that the positive impact of finance on growth dissipates above the threshold stage of financial development. . It has also been concluded that as GDP per capita grows, financial systems begin to change to non-banking funding. This finding is in line with Atje and Jovanovic (1993) who encouraged the stock market development for the economic growth.
6 2.2 Stock market development
By issuing new financial services to businesses, Levine (1991) found a positive connection between the financial stock market and economic growth. Filer et al. (1999) investigated the nexus of stock market growth and showed a strong casual link between the development of stock markets and economic activity. Spears (1991) concluded that in the early stages of development, financial intermediation triggered economic growth. . Higher investment and capital allocation and, indirectly, economic development is promoted by the financial stock market. Often investors avoid investing directly in businesses because, whenever they want, they cannot easily withdraw their capital. But they can buy and sell stocks easily, with more freedom, via the financial stock exchange. Schumpeter (1912) argued that technological progress is linked to the efficiency of financial intermediaries, which redirect investment funds to entrepreneurs with the best prospects for successful implementation of new products. A research was later carried out by Atje and Jovanovic (1993) in which they concluded that the degree of stock market development had a substantial and positive impact on economic growth. The researchers argued that in their study, which consists of 40 countries over an 8-year period, there is a strong association between stock market development and economic growth. In a similar manner, Beck et al (2000) and Beck and Levine (2004), studied the relationship between stock market development and economic growth. They concluded that stock market development has a major positive impact on economic growth.
The same conclusion was reached in a more recent study by Grbic (2020), who examined the relationship between the development of the stock market and economic growth in Serbia between 2002 and 2018 and concluded that policymakers in the Republic of Serbia should concentrate on a stock market promotion strategy to improve economic growth.
With respect to liquidity, Bencivenga et al. (1991) concluded that investing in liquid markets is easier, so people are encouraged to invest their money in long-term assets because they can quickly turn the stakes into money before maturity. New companies and enterprises also benefit from stock market liquidity, allowing them to access capital and finance. Paudel (2005) conducted a study in which the stock markets were identified as the main source of funding for firms. Bahadur and Neupane (2006) analysed stock market fluctuations and economic indices in their paper, concluding that stock markets could be a good predictor of growth. Reverse effect is available, financial crashes (1929 and 2008) were anticipated due to stock market downturns.
2.3 Stock market and banking development
7
that growth is not inked to the size of the stock market, volatility, and international integration. Numerous recent papers show that as capital markets rise, the contribution of banking growth to economic growth decreases. This is a consequence of a greater capacity to foster stock market creativity and productivity.
It can be inferred that the relationship between the stock market and development of banking with the economic growth was analysed intensive since 20th century. Although the opinions are divided among the researchers, most of them argued in favor of it, attempting to find the best logical model for investigating this matter. Considering everything stated before, this paper attempts to validate these approaches, with a particular emphasis on stock markets and banks, as one of the determinants of economic growth.
Taking into account the literature presented it can be concluded that stock market development had in most of the cases a positive impact on the economic growth for both developing and developed countries. In the case of banking development, there was a significant relationship for growth in the developing countries, while for the developed ones this relationship was not always depicted. Thus, for enhancing the economic growth, in the developing countries both stock market and banking sector should be developed to a certain extent, while in developed countries, the accent should be on stock market growth.
Counting on empirical evidence from the existing literature, I formulate the following hypotheses:
H0: Stock market development promotes economic growth in both regions.
H1: Banking development stimulates the economic growth in the eastern Europe, but not in western Europe.
3.Methodology
8
From all of the above, it can be concluded that the best method for performing this research is described by a panel data analysis using a two-way error factor model to allow both entity fixed effects and time fixed effects within the same model. According to Brooks (2014) there are certain benefits of this approach, a broader range of questions and complex topics can be discussed, since it includes both time-series and cross-section details at once and it is important to be able to analyse how variables or their interactions change over time. There are several advantages if cross-sectional data is combined with the time series one. First, the additional variance will be added to the model and, in this way, the multicollinearity problem that may arise when the time series is modelled on its own will be solved, and second the impact of omitted variables bias in regression results can be removed if the model has an appropriate structure.
Pooled OLS regression will be estimated. The assumption behind this model is that there is no heterogeneity, but panel data has to deal with different entities over multiple periods, therefore it is constrained to heterogeneity. This possible issue can be mitigated using two types of panel data models. The first one is Fixed Effects model, which control for omitted variable bias caused by the constant unobserved heterogeneity over time, and it assumes that the individual-specific effects are related to independent variables. The second one is Random Effects model, which controls the unobserved heterogeneity when is not correlated with the independent variable and it is constant over time. This model assumes that there is no correlation between the individual unobserved heterogeneity and independent variables. If the assumption of random effects holds, then it can be concluded that random effects estimator is more effective than the fixed effects model, otherwise random effect estimator is not reliable.
The Redundant Fixed Effects test will be performed and will help to choose between the Pooled OLS and the fixed effects model. The null hypothesis of this states that the fixed effects of the observed and unobserved fixed effects is null. Breusch-Pagan Lagrange multiplier test will be conducted to determine which approach is better between Pooled OLS and random effects model. The null hypothesis states that the error variances are equal. Hausman test will be used to choose the best method between fixed or random effects. The null hypothesis states that the most appropriate test to perform is random effect.
In order to test the relationship between the banking development and stock market with the growth of the economy, OLS regression (1), Fixed Effects (2) and Random Effects (3) will be used, with the following structure:
Yit=𝛼 + 𝛽1𝑋1𝑖𝑡+ 𝛽2𝑋2𝑖𝑡+𝛽3𝑋3𝑖𝑡+ 𝜀𝑖𝑡 (1) i = 1, 2, …, N (cross sections)
t = 1, 2, …, T (time series)
Yit = β1Xit + αi + uit (2) i = 1, 2, …, N (cross sections)
9
Yit = βXit + α + uit + εit (3) i = 1, 2, …, N (cross sections)
t = 1, 2, …, T (time series)
In the equations presented above, the dependent variable Y stands for the economic growth. In terms of independent variables there is stock market development, which is defined by more than one variable and banking development. The stock market development will be quantified using the size and liquidity. The banking sector development will be measured through domestic credit to private sector by banks.
The data is compiled annually for the period 1995-2019 and is collected from the World Bank using different databases (World Development Indicators, World Development Indicators Beta World Bank Open Data, Global Financial Development Database), Ameco, EconStats and the International Monetary Fund Database).
In order to determine the effect of the stock market and the advancement of banking on Eastern European's economic growth , it is important to gather various indicators relevant to the progress of both variables under research, as well as an acceptable proxy for economic growth. The general equation mentioned above can be adjusted more specific for this research:
GDPPCt= 𝛼+𝛽1MCAP𝑖𝑡+𝛽2TR𝑖𝑡 +𝛽3TVS𝑖𝑡+𝛽4BC𝑖𝑡+𝛽5FDI𝑖𝑡+𝛽6IR𝑖𝑡+𝜀𝑖𝑡, i = 1, 2, …, N (cross sections)
t = 1, 2, …, T (time series)
Where β1 to β6 represents coefficient of the parameters of estimation, i represents cross- section, i.e individual countries of data sample and t is the period in question.
The dependent variable is GDPPC, which stands for real Gross Domestic Product per capita growth and it analyses the economic growth. The same indicator was used in the studies performed by Levine and Zervos (1998), Beck and Levine (2001), Ogbeide et al (2018) and other scholars.
Based on work done by Levine and Zervos (1998) the stock market development will be measured through size and liquidity. The stock market size will be measured using the market capitalization over the GDP, in the model abbreviated MCAP, and it is calculated as the share price times the number of shares outstanding (including their several classes) for listed domestic companies.
10
is calculated as the total number of shares traded, both on domestic and foreign market, multiplied by their respective corresponding prices.
The banking development is measured through Domestic credit to private sector (% GDP) and it refers to financial resources provided to the private sector by financial corporations, such as through loans, purchases of nonequity securities, and trade credits and other accounts receivable, that establish a claim for repayment. In the model is denoted as BC. This indicator is in line with Levine and Zervos (1998) and Beck and Levine (2001) methodology, and it has been selected because it enables the identification of where capital is allocated by the financial system.
Last but not least, the model also includes two control variables. The first is related to financial integration, which is represented by Foreign Direct Investment over Gross domestic product and can be found as FDI in the model. The second control variable is the inflation rate (IR), and according to Eaterly and Sergo (1993) it is a valid way of measuring economic activity and macroeconomic instability.
The same method will be used for a set of six Western European countries, Austria, Belgium, France, Netherlands, Ireland and Switzerland, with the aim of comparing whether the impact of the selected indicators is higher in developing countries than in developed ones.
4. Descriptive Statistics
Table 1. Summary Statistics: Annual Averages 1995 – 2019 for Eastern Europe. Table
shows mean, median, maximum, minimum, standard deviation, kurtosis, skewness, and number of observations.
Mean Median Max Min Std. Dev. Kurtosis Skewness N
GDPPCG 0.035 0.041 0.111 -0.137 0.035 7.461 -1.406 149
Market Cap 0.148 0.133 0.493 0.000 0.110 3.267 0.844 149
Value Traded 0.061 0.032 0.341 0.000 0.073 5.636 1.633 149
Turnover ratio 0.411 0.332 2.155 0.002 0.423 7.107 1.880 149
Bank Credit 0.390 0.390 0.689 0.071 0.159 2.164 -0.200 149
Notes: The results are based on historical data of Romania, Bulgaria, Hungary, the Czech Republic, Slovakia and Poland;
GDPPC = real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Value Traded = value of the trades of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Bank Credit = bank credit to the private sector as a share of GDP.
11
Checking the tables and figures presented in this section, it can be inferred that there is a significant variation between countries in the indices of economic growth, stock market development and banking development. Focusing on statistics in Table 1, all variables have a positive value for Kurtosis, different than 3 , which leads to the conclusion that the distribution is Leptokurtic, with a peaked curve, and indicates that there are higher values than the sample mean. As regards the asymmetric level of the series, it can be seen that there is a positive skew in the case of Market Capitalization, Value Traded and Turnover Ratio.
This allows us to state that this distribution has a long right tail, with higher values than the sample mean. There is a negative skew in the case of explained variable and bank credit, which concludes exactly the opposite of the previous case. The values of the standard deviation indicate that there is a moderate variability across the dataset.
Last but not least, the panel data is strongly balanced without any missing observations. Bellow, Figures 1 to 4 present the average value of the indicators among the East European countries.
The differences between the countries selected in Eastern Europe can be seen in Figures 1 to 4. In terms of GDP per capita growth, Poland recorded the highest average growth in the region of close to 4.28 per cent, while the Czech Republic recorded the lowest average growth of 2.59 per cent between 1995 and 2019. For the independent variables, Poland outperforms in comparison with the sample in the case of Market capitalization, having an average value of 25.26%
Fig.1 Real GDP per capita growth Fig.2 Market Capitalization
Fig.3 Liquidity indicators Fig. 4 Bank Credit
0 0.02 0.04 0.06 Bulgaria Czechia Hungary Poland Romania Slovakia 0 0.1 0.2 0.3 Bulgaria Czechia Hungary Poland Romania Slovakia 0 0.2 0.4 0.6 0.8 Bulgaria Czechia Hungary Poland Romania Slovakia
Total value traded Turnover ratio 0 0.1 0.2 0.3 0.4 0.5
12
On the other hand, Slovakia has the lowest market capitalization of 4% for the given period. Maximal value for the liquidity indicators was recorded in Hungary, having turnover ratio of almost 63.87% and a Total value of stocks traded close to 14.22%. On the opposite side, Romania is the lowest point of the sample for this variable, having a turnover ratio of 13.42% and a total value traded of 0.82% in average.
The sample shows that there is a high turnover ratio, which concludes that the transaction costs in this market are low. According to Demirgiig-Kunt et al (1996), Levine (1991) and Bencivenga et al. (1995), it can be inferred that the market in selected countries is small and liquid due to the high turnover values associated with the small value traded. Regarding the Bank Credit, it can be noticed that there is not a big variation across the countries. Slovakia represents the highest point, with a value close to 46.18%, while Romania has the lowest one, nearly 23.49%.
In terms of control variables, Hungary outstands compared to the eastern region, having a Foreign Direct Investment in value of 9.3%, almost double than the sample average. Czech Republic had the best control of inflation over the last 25 years, with a value close to 3.4%. The least FDI level in the sample was in Poland, only 3.3%, and the worst management of inflation was in Bulgaria, with values over 50% in the last quarter of century.
Table 2. Correlations Matrix for Eastern Europe. This table presents the correlation
coefficients among the different variables in the model.
GDPPCG Market Cap Turnover Value traded Bank Credit
GDPPCG 1.000
Market Cap 0.124 1.000
Turnover -0.159 0.007 1.000
Value Traded -0.006 0.745 0.446 1.000
Bank credit -0.118 0.256 0.009 0.140 1.000
Notes: The results are based on historical data of Romania, Bulgaria, Hungary, the Czech Republic, Slovakia and Poland;
GDPPC = real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Value Traded = value of the trades of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Bank Credit = bank credit to the private sector as a share of GDP.
Source: STATA output based on data collected from World Development Indicators of World Bank
13
Table 3. Summary Statistics: Annual Averages 1995 – 2019 for Western Europe. Table
shows mean, median, maximum, minimum, standard deviation, kurtosis, skewness, and number of observations.
Mean Median Max Min Std. Dev. Kurtosis Skewness N
GDPPCG 0.018 0.016 0.240 -0.064 0.030 22.459 2.780 150
Market Cap 0.883 0.720 2.912 0.119 0.643 4.035 1.323 150
Value Traded 0.534 0.314 2.624 0.025 0.567 5.278 1.580 149
Turnover ratio 0.540 0.459 2.498 0.052 0.370 7.826 1.629 149
Bank Credit 0.998 0.951 1.746 0.370 0.320 2.678 0.583 142
Notes: The results are based on historical data of Austria, Belgium, France, Netherlands, Ireland and Switzerland; GDPPC =
real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Value Traded = value of the trades of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Bank Credit = bank credit to the private sector as a share of GDP.
Source: STATA output based on data collected from World Development Indicators of World Bank
In table 3 are presented the summary statistics for the west European countries. Similar to the Eastern sample, all variables have a positive value for Kurtosis, which is different than 3, leading to the conclusion that the distribution is Leptokurtic, with a peaked curve, and it indicates that there are higher values than the sample mean. The skewness is positive both for the dependent and independent variables, therefore it can be concluded that hat this distribution has a long right tail, with higher values than the sample mean.
14
Figures 5 to 8 present the average value of the indicators among the West European selected countries.
Fig. 5 Real GDP per capita growth Fig. 6 Market Capitalization
Fig. 7 Liquidity indicators Fig. 8 Bank Credit
From figure 5 it can be inferred that the average economic growth was lower in the west side of Europe, compared to the east. Developing countries, with poor economies and lower capital stocks, can grow faster compared to developed ones, according to the neoclassical growth model of Solow(1956).
The stock market indicators for size and liquidity are considerable bigger compared to the eastern ones, which exposes a more mature and developed market for the western countries. Taking into account the results from Market Capitalization, Total Value traded and Turnover ratio, it can be inferred in the case of Switzerland and Netherlands, that the stock market is bigger and less liquid.
From figure 8 can be mentioned that Western Europe has almost double Bank Credit that the average of Eastern side.
0 0.01 0.02 0.03 0.04 0.05 Austria Belgium France Ireland Netherlands Switzerland 0 0.5 1 1.5 2 2.5 Austria Belgium France Netherlands Switzerland Ireland 0 0.5 1 1.5 2 Austria Belgium France Ireland Netherlands Switzerland 0 0.5 1 1.5 Austria Belgium France Ireland Netherlands Switzerland
15
5. Results
The aim of this part of the research is to present the results obtained using the above-mentioned methodological part. More specifically, for each single independent variable, the coefficient values and their confidence levels will be shown and the impact on the research will be noted. It is important to reiterate that the aim of this study is to examine whether there is a relationship between economic growth, stock exchange and banking development in selected countries in Eastern Europe and, if there is one, to compare the magnitude of the relationship with that in developed countries in Western Europe.
Table 4 . Stock Markets, Banks, and Growth for Eastern Europe, 1995 – 2019. Pooled OLS, Fixed Effects and Random Effects results. Dependent variable: Real Gross Domestic
Product per capita growth. P-values smaller than 0.01, 0.05 and 0.10 are indicated by ***, ** and *.
(1) (2) (3)
VARIABLES Pooled OLS Fixed Effects Random Effects
Market Cap 0.095** 0.108** 0.095** (0.042) (0.051) (0.043) Value Traded -0.172** -0.142 -0.172** (0.071) (0.088) (0.074) Turnover 0.011 0.013 0.011 (0.008) (0.010) (0.009) Bank credit 0.153*** 0.126** 0.153*** (0.036) (0.059) (0.055) FDI 0.015 0.016 0.015 (0.022) (0.030) (0.029) Inflation -0.012*** -0.013*** -0.012*** (0.002) (0.004) (0.004) Constant 0.026*** 0.022*** 0.026*** (0.007) (0.007) (0.006) Observations 142 142 142 R-squared 0.265 0.262 Number of countries 6 6
Redundant Fixed Effects test p-value
0.616
Hausman p-value 0.863
Notes. The results are based on historical data of Romania, Bulgaria, Hungary, the Czech Republic, Slovakia and Poland;
GDPPC = real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Value Traded = value of the trades of domestic shares as a share of GDP; Bank Credit = bank credit to the private sector as share of GDP. Control variables included in the regression: Foreign Direct Investment (FDI) and Inflation. Standard errors are presented in parentheses.
16
Table 4 presents the Pooled OLS, Fixed Effects and Random Effects models for Eastern Europe sample. The significance level of the variable of interest differ across the tests. Market capitalization has a positive value, and it is statistically significant at 5 % level. The coefficient of Value traded is negative and it is statistically significant at 5% level only under the Pooled OLS and Random effects when it has an influence on the economic growth. Bank credit impacts positive the real GDP per capita growth in all the cases, being significant at 5% level under the fixed effects and at 1% under the other 2 tests.
The Redundant fixed effects test helped to choose between the Pooled OLS and Fixed effects method. According to Table 4, the most suitable test is Pooled OLS.
From a financial perspective, the results in Table 4 suggest that one standard deviation increase in market capitalization (0.11) would increase the dependent variable by 1.05 per cent annually (0.095*0.11) over the period under investigation. Taking into account the 25-year period, by the end of the sample period, the real Gross Domestic Product per capita would have been almost 30% higher (𝑒(25∗0.0105)) . The opposite effect is observed for the value traded which has a negative value, allowing us to state that the unit increase in the standard deviation (0.073) reduces real GDP per capita by 1.26% per year. Summing up the whole period, the left side variable would have been almost 27 per cent lower (𝑒(25∗(−0.0126))) .
Table 4 shows that the other liquidity indicator, the Turnover Ratio, does not significantly regress, discarding the relevant relationship between this indicator and real GDP per capita growth.
On the other hand, the development of banks and their influence on economic growth have been studied using variable bank credit, which is defined as private-sector domestic credit by banks as a percentage of GDP. This variable is significant in the regression at 1% and it can be concluded that if there is one unit increase in the standard deviation (0.159), real GDP per capita would increase by 2.43 per cent (0.159*0.153) in one year, leading to an increase of almost 84 per cent (𝑒(25∗0.0243)) by the end of the time period.
Assuming that there is no heterogeneity, the results of previous interpretations are based on the best method for assessing the relationship under analysis, which is the Pooled Ordinary Least Squares. Squares estimate. Taking into account the fact that the countries present in the sample had different indicators, the assumption was surprising.
17
Table 5 . Stock Markets, Banks, and Growth for Eastern Europe, 1995 – 2019. Robustness check for Fixed Effects. Dependent variable: Real Gross Domestic Product per capita growth.
P-values smaller than 0.01, 0.05 and 0.10 are indicated by ***, ** and *.
VARIABLES 1 2 3 4 5 Market Cap 0.120** 0.125** 0.121** 0.111** 0.111** (0.056) (0.055) (0.052) (0.054) (0.053) Value Traded -0.127 -0.129 -0.124 -0.096 -0.095 (0.089) (0.089) (0.086) (0.088) (0.087) Turnover 0.014 0.014 0.013 0.002 0.002 (0.010) (0.010) (0.010) (0.009) (0.009) Bank credit 0.129** 0.134** 0.137** 0.241*** 0.241*** (0.062) (0.061) (0.059) (0.048) (0.047) FDI 0.002 0.002 0.002 0.002 (0.030) (0.030) (0.030) (0.030) Inflation -0.012*** -0.012*** -0.012*** (0.005) (0.005) (0.004) Exports 0.012 0.012 (0.052) (0.052) Unemployment -0.045 (0.113) Constant 0.010 0.005 0.010 0.010 0.010 (0.026) (0.023) (0.010) (0.010) (0.010) Observations 142 142 142 142 142 R-squared 0.360 0.359 0.359 0.317 0.317 Number of countries 6 6 6 6 6
Time effects Yes Yes Yes Yes Yes
Notes. The results are based on historical data of Romania, Bulgaria, Hungary, the Czech Republic, Slovakia, and Poland;
GDPPC = real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Value Traded = value of the trades of domestic shares as a share of GDP; Bank Credit = bank credit to the private sector as share of GDP. Control variables included in the regression: Foreign Direct Investment (FDI), Inflation, Exports and Unemployment. Standard errors are presented in parentheses
Source:Calculated by the author using data collected from World Development Indicators of World Bank
18
Bank Credit stays significant at 5% level, and it can be denoted a small increasing trend while removing the control variables, and it has a statistically positive effect on the economic growth. One percentage unit increase in Bank Credit, increases the GDP growth by 0.129 percentage points when all variables are included, and by 0.241 when the independent variables are included only.
There is not evidence that a change in Value Traded or in Turnover ratio is linked to differing rates of economic growth. Although the signs of the variables are negative, and positive respectively, there is not a statistically significant effect.
Table 6 . Stock Markets, Banks, and Growth for Western Europe, 1995 – 2019. Pooled OLS, Fixed Effects and Random Effects results. Dependent variable: Real Gross Domestic
Product per capita growth. P-values smaller than 0.01, 0.05 and 0.10 are indicated by ***, ** and *.
(1) (2) (3)
VARIABLES Pooled OLS Fixed Effects Random Effects
Market Cap 0.007 0.042*** 0.007 (0.005) (0.012) (0.008) Value Traded -0.002 -0.005 -0.002 (0.006) (0.011) (0.011) Turnover -0.015* 0.007 -0.015 (0.008) (0.013) (0.011) Bank credit -0.055 -0.062* -0.055 (0.057) (0.033) (0.035) FDI 0.066* 0.050*** 0.066*** (0.035) (0.018) (0.017) Inflation 0.350 0.170 0.350* (0.311) (0.203) (0.206) Constant 0.008 -0.028** 0.008 (0.009) (0.012) (0.008) Observations 130 130 130 R-squared 0.188 0.194 Number of countries 6 6
Redundant Fixed Effects test p-value
0.0006
Hausman p-value 0.0130
Notes. The results are based on historical data of Austria, Belgium, France, Netherlands, Ireland and Switzerland; GDPPC =
real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Value Traded = value of the trades of domestic shares as a share of GDP; Bank Credit = bank credit to the private sector as share of GDP. Control variables included in the regression: Foreign Direct Investment (FDI), Inflation, Exports and Unemployment. Standard errors are presented in parentheses
19
Table 6 presents the Pooled OLS, Fixed Effects and Random Effects models for Western Europe sample. The significance level of the variable of interest differ. Market Capitalization has a positive impact on growth and is significant at 1% level under the fixed effect model, meaning that one percentage increase in the size of stock market leads to growth of the economy of 0.042 percentage points. Turnover ratio’s coefficient is significant at 10% level for Pooled OLS, and a percentage increase in its value will slow the economic growth by 0.015 percentage points.
Value traded has a low negative coefficient and is not statistically significant in the economic growth process.
Bank credit has a negative value, and it has a significant influence when is regressed using the fixed effects model, and it can be interpreted as one unit increase in the bank credit leads to a decrease of the real GDP per capita growth by 0.062 percentage points.
Redundant fixed effects test has a p-value of 0.006, which leads to the conclusion that fixed effects are preferred over Pooled OLS. The p-value of Hausman test is 0.0130 which is statistically significant at 5% level, therefore fixed effects are preferred over random effects. LM test indicates that the most suitable test for this sample is fixed effects.
Analysing the output in terms of standard deviation it can be stated that if there is a one-off rise in the standard deviation of the size of the stock market (0,042) there will be an increase of 2.7% (0.042*0.643) per year in real GDP per capita and an increase of 83.7% (𝑒(25∗0.027)) by the end of the study period. Bank development (-0.062) has a negative effect on economic growth, with the result that a one unit rise in the standard deviation (0.320) would reduce real GDP per capita by 1.9% per year. Accumulating over 25 years, this means that real GDP per capita will decline by 37.8% (𝑒(25∗(−0.0198))).
Robustness checks for fixed effects are displayed in table 7, where the control variables where varied in different regressions. In order to add the time trend in this analysis, 5 years periods dummies were created. The results in the table below suggest that the control variables have an effect on some independent variables.
Market capitalization has a positive influence on the economic growth, and it is significant at 1% level, therefore one unit increase in the market capitalization leads to 0.046 units increase in the growth when all the control variables are added and to 0.054 increase when the independent variable are sole in the model.
20
Table 7 . Stock Markets, Banks, and Growth for Eastern Europe, 1995 – 2019. Robustness check for Fixed Effects. Dependent variable: Real Gross Domestic Product per capita growth.
P-values smaller than 0.01, 0.05 and 0.10 are indicated by ***, ** and *.
VARIABLES 1 2 3 4 5 Market Cap 0.046*** 0.045*** 0.049*** 0.052*** 0.054*** (0.012) (0.012) (0.013) (0.013) (0.014) Value Traded -0.016* -0.013 -0.017* -0.019* -0.017 (0.009) (0.010) (0.010) (0.010) (0.011) Turnover 0.013 0.018 0.024* 0.030** 0.030** (0.013) (0.014) (0.014) (0.014) (0.015) Bank credit -0.096** -0.034 -0.084** -0.072** -0.118*** (0.038) (0.037) (0.035) (0.034) (0.035) FDI 0.069*** 0.065*** 0.074*** 0.072*** (0.015) (0.016) (0.016) (0.016) Inflation 0.054 0.412** 0.338* (0.208) (0.196) (0.203) Exports 0.145*** 0.142*** (0.042) (0.044) Unemployment -0.604*** (0.159) Constant -0.039 -0.092*** -0.023* -0.021 -0.019 (0.027) (0.025) (0.013) (0.013) (0.014) Observations 126 126 126 126 126 R-squared 0.508 0.443 0.390 0.375 0.268 Number of countries 6 6 6 6 6
Time Effects Yes Yes Yes Yes Yes
Notes. The results are based on historical data of Austria, Belgium, France, Netherlands, Ireland and Switzerland; GDPPC =
real GDP per capita growth; Market Cap = value of domestic shares as a share of GDP; Turnover = value of the trades of domestic shares as a share of market capitalization; Value Traded = value of the trades of domestic shares as a share of GDP; Bank Credit = bank credit to the private sector as share of GDP. Control variables included in the regression: Foreign Direct Investment (FDI), Inflation, Exports and Unemployment. Standard errors are presented in parentheses
Source:Calculated by the author using data collected from World Development Indicators of World Bank
When the additional control variables are included, the turnover ratio does not impact the real GDP per capita growth. It can be seen an increase in the significance level while these
variables are dropped, therefore one percentage increase in the turnover ratio leads to an increase of the growth in value of 0.03 percentage points when one or none control variables are included.
21
From our results it can be concluded that stock market development has a positive impact on economic growth both in eastern and western Europe. This finding is in line with the one of Adjasi and Biekpe (2006) and confirms the first hypothesis of this study.
The second hypothesis which states that banking development plays a significant role in the economic growth is confirmed for the east European sample, and this result is consistent with the one of Levine and Zervos (1998), Beck and Levine (2001), Guru and Yadav (2019), and it is confirmed also for the west European countries. Atje and Jovanovic (1993) and
Mosquera (2019) did not find positive effects for banking development on the economic growth.
This result is supported by the find of Allen and Gale (2000) who emphasized that banks inefficient monopoly is reduced by the competitiveness of stock markets which encourages innovative, growth-enhancing practices as opposed to banks' overly conservative approach.
6. Conclusion
The empirical relationship between the stock market, banking development and long-term economic growth of Eastern Europe in the last quarter of the century was studied in this paper, using a sample of six countries. The outcomes were subsequently contrasted with the one obtained for Western Europe. The growth of the stock market was quantified by measurements of size and liquidity, and the development of banks was measured using private sector credit. Real gross domestic per capita growth has been used for long-term economic growth.
One of the findings of this paper is that the first determinant of the analysis, namely stock market development, is positive and significantly related, through size, to economic growth, namely the market capitalization expressed as a percentage of GDP. This outcome is in line with the Adjasi and Biekpe (2006) findings and does not support the Levine and Zervos (1998) outcome, which concludes that size is not related to economic development. The disparity in the studies may be a reason for this and that both studies deal with different kinds of countries, timeframes, and methodologies. The premise behind this measurement is that the size of the market correlates positively with the capacity to gather capital and diversify risk across the economy.
A negative relationship between the value traded and growth has been noticed when it comes to liquidity indicators. This result is supported by Ake et al (2010), Osamwony and Kasimu (2013), who concluded that there is a negative relationship in countries where the stock market is small and less liquid. No turnover ratio relationship was reported, a finding that was overwhelming because the sample had high turnover ratio levels, which meant that the transaction costs were low. A similar result was reached by Prats (2016).
22
since they have not been able to put an important role in banking development for the economic growth.
Comparing the results between the analysed regions, it can be inferred that the economic growth is slower in developed countries, although are better performing. Developing countries have the potential to rise faster than developed countries, since falling returns are not as high as those in capital-rich countries. Furthermore, poorer countries can replicate the production processes, technologies and institutions of developed countries.
Similar to Eastern Europe, from a stock market viewpoint, the impact of market capitalization on economic growth on the Western side is important and optimistic. This means that there is a much greater demand in the west. From a banking viewpoint, it can be seen that relative to eastern Europe, there is a reverse impact. Since bank credit has a positive effect on the eastern sample and a negative effect on the western sample, this is somehow an anticipated outcome. Allen and Gale's (2000) research emphasizes that equity markets mitigate the inefficient monopoly power of banks and stresses that the competitive nature of markets encourages creative, growth-enhancing activities as opposed to inefficient monopoly power.
Taking all of the above into account, it can be concluded that the stock market has a positive effect on economic development in the two regions examined. The economy is less developed in Eastern Europe, so banking development has a major and positive effect on growth, while the impact of banks on growth is negative in the West, where the economy is more developed. It can be argued that people are less interested in these financial institutions due to the growth and increased participation in the stock markets and that the monopoly power of the banks is attenuated.
23
Bibliography
Adjasi, K., & Biekpe, N. (2006). Stock Market Development and Economic Growth: The Case of Selected African Countries. African Development Bank, 144-161.
Ake, B., & Ognaligui, R. (2010). Financial Stock Market and Economic Growth in Developing Countries: The Case of Douala Stock Exchange in Cameroon. International Journal of Business and Management.
Allen, F., & Gale, D. (2000). Comparing Financial Systems. Cambridge, MA: MIT Press,.
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, 277-297. Bahadur, S., & Neupane, S. (2006). Stock Market and Economic Development: a Causality Test.
36The Journal of Nepalese Business Studies .
Barro, R. J. (1991). Economic growth in a cross section of countries. Quarterly Journal of, 407-443. Beck, T., Levine, R., & Levine, R. (2000a). Finance and the Sources of Growth. Journal of Financial
Economics, 261-300.
Beck, T., Levine, R., & Levine, R. (2000b). Financial intermediation and growth: Causality and Causes. Journal of Monetary Economics, 31-77.
Bencivenga, V. R., & Smith, B. D. (1991). Financial Intermediation and Economic Growth. The Review of Economic Studies.
Biplab Kumar Guru, I. S. (2019). Financial development and economic growth: panel evidence from BRICS. Journal of Economics, Finance and Administrative Science.
CAPASSO, S. (2008, February). ENDOGENOUS INFORMATION FRICTIONS, STOCK MARKET DEVELOPMENT AND ECONOMIC GROWTH.
Carp, L. (2012). Can Stock Market Development Boost Economic Growth? Empirical Evidence from Emerging Markets in Central and Eastern Europe. Procedia Economics and Finance, pp. 438-444.
Dawson, P. (2008). Financial development and economic growth in developing countries. Progress in Development Studies, 325-331.
Filer, R. K., Hanouse, J., & Campos, N. F. (1999). Do Stock Markets Promote Economic Growth? ERGE-EI Working Papers, The Center for Economic Research and Graduate Education - Economics Institute, Prague.
Grbic, M. (2020). Stock market development and economic growth: the case of the Republic of Serbia.
Joseph, S. (1912). Theorie der wirtschaftlichen entwicklung. Dunker & Humblot.
Jr., R. E. (1998). On the mechanics of economic development. . Journal of Monetary Economics, 3-42. King, R. G. (1993a). Finance and Growth: Schumpeter Might Be Right. Quarterly Journal of
Economics, pp. 717-38.
24
Levine, R. (1991, September). Stock Markets, Growth, and Tax Policy. Journal of Finance,, pp. 1445-65.
Loayza, N., & Ranciere, R. (2006). Financial Development, Financial Fragility, and Growth. Journal of Money, Credit and Banking, 1051-1076.
McKinnon, R. (1973). Money and Capital in Economic Development. The Brooking Institute. Miller, M. H. (1998). Financial markets and economic growth. Journal of Applied Corporate, 8-15. Muhammad Shahbaz, N. A. (2008). Stock Market Development and Economic Growth: Ardl Causality
in Pakistan. EuroJournals Publishing.
Nahla Samargandi, J. F. (2015). Is the Relationship Between Financial Development and Economic Growth Monotonic? Evidence from a Sample of Middle-Income Countries. World
Development, 66-81.
Osamwonyi, I. O., & Kasimu, A. (2013). Stock Market and Economic Growth in Ghana, Kenya and Nigeria. International Journal of Financial Research.
Paudel, R. (2005). Financial System and Economic Development. Nepal Rastra Bank in Fifty Years, Part –II,.
Pettinger, T. (2018, December 9). Relationship between stock market and economy.
Philip Arestis, P. O. (2001). Financial Development and Economic Growth: The Role of Stock Markets. Journal of Money, Credit and Banking, pp. 14-41.
Porta, R. L., Lopez‐De‐Silanes, F., & Shleifer, A. (2002). Government ownership of banks. The Journal of Finance, 265-301.
Prats, M. (2016). Stock Market and Economic Growth in Eastern Europe. Financial System and Economic Growth.
R, A., & Jovanovic B. (1993). Stock markets and development. European Economic Review, 632-640. Shaw, E. S. (1973). Financial deepening and economic development. Oxford University Press. Spears, A. (1991). Financial development and economic growth — causality tests. Atlantic Economic
Journal volume .
SULIMAN, M. S. (2009, December). INSTITUTIONS, BANKING DEVELOPMENT, AND ECONOMIC GROWTH.
Sunday, O., & Babatunde, A. (2018). Stock market development and economic growth of Brazil, Russia, India, China and South African (BRICS) Nations: An empirical research. Accounting, 83-92.
ZINGALES, L. (2015). Presidential Address: Does Finance Benefit Society? The Journal of Finance, 1325-1363.
25
Appendix.
Eastern European Sample
1. Redundant Fixed Effects:
The F-test has a P-value of 0.6163, which is statistically different than zero, therefore we fail to reject the null hypothesis and it can be concluded that Pooled OLS is more suitable than Fixed Effects. 2. Hausman test
26
3. Breusch and Pagan Lagrangian multiplier test for Random Effects:
The P-value of LM test is 1 and the null hypothesis cannot be discharged, therefore it can be concluded that fixed effects are preferred .
4. Time effects test
The t test is significant at 1% level, therefore the dummies add value to the model
Western European Sample
27
As the P-value is 0.0006 the null hypothesis can be rejected and therefore Fixed Effects is more suitable than POLS
6. Hausman Test
Since the null hypothesis can be rejected, because of a P-value of 0.0130, Fixed Effects model should be preferred instead of Random Effects.
7. Breusch and Pagan Lagrangian multiplier test for Random Effects:
The P-value of LM test is 1 and the null hypothesis cannot be discharged, therefore it can be concluded that Fixed effects method is preferred for this model.
8. Time effects test
