A study on the course of the effect of stock
market development on economic growth in
the Eurozone.
Abstract
This research examines if the effect of stock market development on economic growth declines at a greater level of stock market development. The inducement of this study is the discussion whether stock market development leads to economic growth. Due to contradictory results, this discussion has not reached its conclusion. In this research, a decline in the effect of stock market development is examined as a possible cause of these inconsistent results. As a measurement for the stock market development in this research, an index is computed with the average of the market capitalization, value of total shares traded and the turnover ratio of the stock market. The relation is studied over a panel data set with annual observations from 1975-2015 for the eleven countries which joined the Eurozone in 1999. The findings of this research are that there is no proof of a decline in the marginal effect of stock market development. Some evidence is found for a positive effect of highly developed stock markets on economic growth. Due to the findings of this paper, it contributes to the discussion and implies the need for further research on this subject.
Wiecher Dalebout – 10679987 Date: 30-01-2017
Supervisor: M.A. Rola-Janicka Faculty of Economics and Business University of Amsterdam
1
Statement of Originality
This document is written by Student Wiecher Dalebout who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
2
Table of contents
1. Introduction ... 3
2. Theoretical Review... 5
2.1 Financial Development ... 5
2.2 Effect of Stock Market on Economic Growth ... 5
2.3 Stock Market Development ... 6
2.4 Empirical Literature Review ... 6
3. Research Methodology ... 8
3.1 Research Model ... 8
3.2 Data Collection ... 12
3.3 Research Hypothesis ... 12
4. Results and Discussion ... 13
4.1 Results Analysis ... 13
4.2 Discussion ... 15
5. Conclusion ... 16
Bibliography ... 18
3 1. Introduction
This thesis examines the relationship between stock market development and economic growth with the focus on the development of the relationship. Historically economic research studied the role of the financial system as an instrument for economic growth and proved the importance of banking and their ability to provide finance for listed companies, which leads to economic growth (Schumpeter, 1911). Since then much new research has been conducted, and the focus of the research shifted to the relation between stock market development and economic growth. The results of those studies are not consistent and therefore there exists discussion in the literature about this relationship.
The positive effect and importance of stock market development on economic growth are revealed by Demirgüç-Kunt & Maksimovic (1996). Furthermore, Levine and Zervos (1998) concluded that stock market liquidity and the banking sector both have a positive effect on economic growth. Levine, Loayza & Beck (2000) also provide evidence for the positive effect of financial intermediary development on economic growth. Levine and Zervos (1996) find
evidence that stock market development positively and robustly associates with long-run economic growth.
On the other hand, there is evidence for a negative relationship as well. Ram (1999) finds that contrary to the conclusion of the previous studies, no evidence support the relationship between financial development and economic growth. In his research, he finds a negligible or weakly negative relationship. Dawson (2003) tested the relationship, with panel data consisting of 13 Central and East European Countries (CEECs), and concludes that financial development has an insignificant effect on economic growth. Boubakari and Jin (2010) explore the relationship between stock market development and economic growth on 5 Euronext countries for 1995 to 2008 and find a positive effect for countries with a liquid and highly active market. However, they reject the causality for the small and less liquid stock markets.
Thus, the findings in the literature are that there are both positive and weakly negative relations between financial development, consisting of stock market development and banking
development, and economic growth. Since the relationship is important for the policy of countries, it is desired to conduct further research on this subject. One of the reasons for the conflicting results in the literature can be the difference between the levels of development of the
4
countries in the sample of the past studies. This difference could be the case if the degree of development of the stock markets plays a role in the relationship with economic growth. Another view on the tilting point of the relationship between financial development and economic growth is the dual effect of liquidity, defined as the ease and speed with which individuals can convert assets into cash. This stock market liquidity induces a faster steady economic growth on the saving rates and therefore the bilateral effect on economic growth (Levine, 1997).
Following the debate about this relationship, this thesis focusses on the stock market
development. The goal of this research is to determine whether the impact of an increase in the stock market development on economic growth has a downturn when the level of stock market development is higher. Therefore this thesis tries to find the answer to the question: "Does the benefits of stock market development on economic growth decline at a greater level of stock market development?”
It is expected that the marginal effect of stock market development decreases with a
well-developed stock market. The stock market allows listed companies easy access to capital through the issuance of equity, which leads to an improvement of the allocation of capital, which is an important channel of economic growth (Arestis, Demetriades, & Luintel, 2001). Another characteristic of the stock exchange is liquidity; the ease, and speed with which individuals can convert assets into cash. Levine (1997) states that an increase in liquidity results in higher investment returns and lower uncertainty, those higher returns ambiguously affect saving rates due to income and substitution effects. Therefore he concludes that an increase in stock market development can result in a rise or a fall of the saving rates. This bilateral effect shows that the development of stock markets can indeed set back economic growth.
To find an answer to address the research question, this thesis is structured as follows. In the first section, the literature on the subject is reviewed by describing the theory of financial
development, the effect of the stock market on economic growth, stock market development and further explanation about the empirical evidence on this topic. Subsequently, in the research methodology, the model, data collection, and the hypothesis are described. In part four the results are analyzed and interpreted with the economic theory. Finally, the conclusion is drawn from the results, and how those findings relate to the literature, the limitation of the thesis and the
5 2. Theoretical Review
2.1 Financial Development
Financial development is described by Levine (1999) as the ability of the financial system to analyze firms and identify profitable ventures, exert corporate control, manage the risk, mobilize savings and the ease of transactions. Thus, the development of financial institutions and stock markets are comprehended in financial development. Levine and Zervos (1998) find that stock markets do not substitute for the banking sector. In their research measures of the bank sector development and stock market development both significantly influence economic growth. Thus, they conclude that banks and stock markets both provide different financial services with a positive effect on economic growth. Furthermore, Demirgüç-Kunt & Levine (1996) find a strong positive correlation between stock market development and financial intermediaries. In this paper, the literature on the effects of financial development and stock market development are both reviewed, though the main focus is on the effect of stock market development.
2.2 Effect of Stock Market on Economic Growth
Within financial development, theory suggests that the stock market provides several channels which enhance economic growth. Some of these channels are the mobilization of savings, risk diversification, improvement of the allocation of resources and information sharing.
A mobilization of savings occurs in the stock market since it provides extra channels within a country for individuals to invest their money (Levine & Zervos, 1998). This mobilization of savings results in a better fit with investors their liquidity and risk preferences compared to a country without a stock market. By diversifying their portfolio, investors can reach their preferred level of risk encouraging them to save more (Demirgüç-Kunt & Levine, 1996). Obstfeld (1992) links this diversification to a shift in high-return investments, which promotes economic growth.
Another influence of stock markets is the effect on the allocation of resources because of a shift in portfolios of investors. Firms have permanent access to the external capital through the stock market. Therefore they can invest in long-term projects with potentially high returns. While for the shareholder the investment remains liquid. This results in a shift of investors’ portfolios towards long-term investments and thereby improving the allocation of capital which results in
6
economic growth (Levine, 1997).
Additionally, stock markets have an important role in the gathering and sharing of information. Listed firms are obliged to share information about its financial activities in their annual report, which enriches the available information. Investors can acquire information on the listed companies to gain insights on their investments. Therefore the information is more symmetric. Asymmetric information and imperfections in capital markets have an adverse effect on firm's ability to raise funds and invest (Demirgüç-Kunt & Maksimovic, 1996). Thus, an increase in available information leads to more efficient monitoring of the firms, making it more profitable for investors to invest in those companies (Demirgüç-Kunt & Maksimovic, 1996).
2.3 Stock Market Development
Stock market development is a multidimensional concept which is difficult to measure (Garcia & Lui, 1999). They state that it is usually measured by the market size, liquidity, volatility,
concentration, integration with capital market and the regulations in the market. Demirgüç-Kunt and Maksimovic (1996) also state that the stock market development is multidimensional and choose to measure the development by creating an index based on the ratios of stock market capitalization to GDP, the total value of shares traded to GDP, and values traded to market capitalization. This last measurement has been used by Pagano (1993b) and Demirgüç-Kunt and Levine (1996). Therefore this study follows the latter approach to determine the stock market development. Sahay, Čihák, N'Diaye, & Barajas (2015) find evidence that financial development entails trade-offs and espouse the “too much finance” theory of Arcand, Berkes, and Pinzza (2012). That is, beyond a certain level of financial development there is a tilting point where the benefits start to decline, and the costs of economic and financial volatility start to increase. Thus the economic theory about the stock market shows that the effect of the stock market can both have a positive or negative effect on the economic growth. This reciprocal effect induces that an increase in stock market development can have the same bilateral effect on economic growth. In the following part, the discussion about this effect in the literature is reviewed.
2.4 Empirical Literature Review
The literature on the role of stock market development on economic growth gives contradictory results. There is a discussion in the literature about the existence and direction of the relationship
7
and the tilting point of the effect. Levine (1997) describes the dual effect of liquidity of the stock market, defined as the ease and speed with which individuals can convert assets into cash, on the saving rates and therefore the potential adverse effect of liquidity on the economic growth. As for Pagano (1993a), he states that the relationship between financial development and saving rate is ambiguous. Thus, it may also reduce the saving rate and slow down the economic growth. This ambiguous effect is also studied by Arcand et al. (2012). They find the so-called "too much finance" effect, where the effect of financial development has a negative impact on output growth. In this thesis, the development of the relationship is researched to control for the
marginal effect. This effect is backed up by the paper of Sahay et al. (2015). They find evidence for the tilting point where financial development leads to an increase in costs because of
economic and financial volatility. Another finding in their paper is that regulatory frameworks can mitigate this effect. Furthermore, the pace of the stock market development is relevant for the financial stability risks.
Caporale, Howells & Soliman (2004) research the effect of stock market development on
economic growth. Moreover, they examine the linkage between financial development and stock market development. Their findings are that economic growth is caused by stock market
development, where they measure stock market development by the market capitalization ratio and the value traded ratio. Without the stock market development, they find little evidence of financial development causing economic growth. However, with the stock market development taken into account, the casualty between financial development and economic growth is proven. This relationship is also researched by Levine and Zervos (1998), they do find that both stock market liquidity and banking development positively predict economic growth.
Demirgüç-Kunt & Maksimovic (1996) create an index for stock market development where the average of stock market capitalization, the total value of shares traded and the total value of shares traded to market capitalization is taken. This index is used in this research as stock market development index as well. In their research, they conclude that highly developed stock markets lead to a substitution of equity for debt financing. As a result, large firms become more levered, whereas small firms are not significantly affected by stock market development.
Levine et al. (2000) prove in their research that there exists a positive link between financial development and economic growth and show that this link is not due to potential bias from
8
omitted variables. They also explain the influence of legal rights of creditors, the efficiency of contract enforcement and accounting system standards on the cross-country differences. From this, they conclude that countries with laws that give high priority to creditors, legal systems that rigorously enforce contracts and with high accounting standards tend to have a better financial development and therefore a higher economic growth.
Boubakari & Jin (2010) find a positive relation between stock market development and economic growth in some Euronext countries. A Granger casualty test is used to prove this causal relationship. The relationship is rejected for countries with a small and less liquid stock market. These contrary results between small and large stock markets suggest an important role of liquidity within a stock market on economic growth.
The negative effect of the financial development on economic growth is examined by Ram (1999), this study does not support the view that financial development has a positive impact on economic growth. At first, the paper concludes that the data for 95 countries suggest that the relation between financial development and economic growth is negligible or weakly negative. Secondly, the average individual-country correlation is in contrast with the cross-country correlation between the same variables. These conclusions could indicate that the studies with the cross-country data might be spurious. Finally, the paper concludes that positive parameters estimates for the financial development are not reliable when statements are made for individual countries or sample groups as used in the research of Levine (1997). This negative effect is supported by Dawson (2003) since his study finds no evidence of a positive and significant relationship between the financial development and economic growth in CEECs.
3. Research Methodology 3.1 Research Model
In this section, the relationship between stock market development and economic growth is examined, and the foundation for the analysis of the time-series/cross-section (TSCS) data is set. Beck and Katz (1995) use Ordinary Least Squares (OLS) for the TSCS data. They state that OLS is only optimal if the assumptions hold that there is no serial correlation and no spatial
correlation within the data, thus the errors of the model are generated in an uncomplicated spherical matter. These assumptions are commonly violated in TSCS datasets because some
9
properties of models for TSCS data allow the errors to be temporally and spatially correlated and enable them to be heteroskedastic. Although the OLS regression seems to perform well in
practical research, it might not produce the optimal estimates of the TSCS model parameters (Beck & Katz, 1995). In this paper, a fixed effects model is used for the analysis of the TSCS data. This model is used to analyze the impact of the variables over time within the country and adjust for exogenous factors of each country separately, thus without correlation between those error terms. The fixed effects model is chosen, above the random-effects model, because it is likely that the data contains omitted country-specific characteristics that are correlated with other explanatory variables (Rousseau & Wachtel, 2000).
In this panel of 11 countries over a 40 year period, the following regression is used to analyze the TSCS data and see if the marginal effect of SMDI decreases from a certain point. Thus, the significance of the negative effect of 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑦𝑦𝑦𝑦−12 on 𝐺𝐺𝑆𝑆𝐺𝐺𝑦𝑦𝑦𝑦 is tested. Since it is expected that stock market development causes economic growth, the lagged data of stock market
development is used in the model to measure the long-term effect on economic growth. The lagged SMDI is used to prevent reverse causality; this can occur since GDP can also influence variables within the SMDI. This delay also prevents omitted variable bias if there is a variable that affects GDP and SMDI in the same year.
The first model is specified as:
𝑮𝑮𝑮𝑮𝑮𝑮𝒚𝒚𝒚𝒚= 𝜷𝜷𝟎𝟎+ 𝜷𝜷𝟏𝟏𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺𝒚𝒚𝒚𝒚−𝟏𝟏+ 𝜷𝜷𝟐𝟐𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺𝒚𝒚𝒚𝒚−𝟏𝟏𝟐𝟐 + 𝜷𝜷𝟑𝟑𝑮𝑮𝑮𝑮𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟒𝟒𝑿𝑿𝑺𝑺𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟓𝟓𝑺𝑺𝑰𝑰𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟔𝟔𝑰𝑰𝑮𝑮𝑺𝑺𝒚𝒚𝒚𝒚+ 𝑰𝑰𝑮𝑮𝒚𝒚+ 𝜺𝜺𝒊𝒊 (𝟏𝟏. 𝟎𝟎)
Where SMDI is computed as:
𝐒𝐒𝐒𝐒𝐒𝐒𝐒𝐒𝒚𝒚=𝟏𝟏𝒏𝒏 �(𝑺𝑺𝑴𝑴𝒚𝒚+ 𝑻𝑻𝑻𝑻𝑺𝑺𝑻𝑻𝒚𝒚+ 𝑻𝑻𝑻𝑻𝒚𝒚) (𝟏𝟏. 𝟏𝟏)
A second regression is used to test if the impact of 𝛥𝛥𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 on 𝐺𝐺𝑆𝑆𝐺𝐺 is lower at higher levels of 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆. Therefore the second model is specified as:
𝑮𝑮𝑮𝑮𝑮𝑮𝒚𝒚𝒚𝒚= 𝜶𝜶𝟎𝟎+ 𝜷𝜷𝟏𝟏(𝜟𝜟𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺)𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟐𝟐(𝑮𝑮𝑫𝑫𝑫𝑫𝑫𝑫𝒚𝒚)𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟑𝟑((𝑮𝑮𝑫𝑫𝑫𝑫𝑫𝑫𝒚𝒚)𝒚𝒚𝒚𝒚∗ (𝜟𝜟𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺)𝒚𝒚𝒚𝒚) + 𝜷𝜷𝟒𝟒𝑮𝑮𝑮𝑮𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟓𝟓𝑿𝑿𝑺𝑺𝒚𝒚𝒚𝒚+
𝜷𝜷𝟔𝟔𝑺𝑺𝑰𝑰𝒚𝒚𝒚𝒚+ 𝜷𝜷𝟕𝟕𝑰𝑰𝑮𝑮𝑺𝑺𝒚𝒚𝒚𝒚+ 𝑰𝑰𝑮𝑮𝒚𝒚+ 𝜺𝜺𝒊𝒊 (𝟐𝟐. 𝟎𝟎)
10
The dependent variable economic growth (GDP) is measured in real GDP per capita percentage annual change (Levine & Zervos, 1996). In this research, the annual percentage growth rate of GDP per capita is based on a constant local currency. Aggregates are based on a constant 2010 U.S. Dollar. The GDP per capita is calculated by dividing the gross domestic product by the midyear population.
The independent variable stock market development index (SMDI) is computed by the average of the following stock market indicators (equation 1.1) according to the theory of Demirgüç-Kunt and Levine (1996), Levine and Zervos (1996) and Levine (1997). The market size is determined by the market capitalization ratio (MC). Market capitalization ratio equals the value of share price times the number of shares outstanding of listed domestic companies as a percentage of the GPD. This indicator is chosen because it reflects the ability to mobilize capital and diversify risk (Demirgüç-Kunt & Levine, 1996). As a liquidity indicator of the stock market, the total value of shares traded ratio (TVST) is computed by dividing the total shares traded on the domestic stock market by GDP. The second indicator of stock market liquidity is the turnover ratio. The
turnover ratio (TR) is the value of domestic shares traded divided by their market capitalization. The value is annualized by multiplying the monthly average by twelve. This index is a complete index available from the literature and is preferred over other indexes with fewer variables; this is in agreement with the paper by Sahay et al. (2015) which stated that financial development is multi-faceted and should be measured by many indicators. This index is also used in the research of Boubakari & Jin (2010), they find evidence for a positive relation with economic growth in some countries and proof of the contrary for countries with a small and less liquid market. The variable 𝛥𝛥𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 represents the change in SMDI between t and t-1. The variable 𝑆𝑆𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 represents a dummy variable with value 0 if SMDI<Z and value 1 when SMDI>Z, where Z = mean of SMDI of the entire sample. This variable is used in the second regression to examine the effect of a large stock market development index compared to a small index. Since the value is set as the mean of SMDI of the entire sample, it is arbitrary, and this is taken account of in the interpretation of the results.
As control variables, government expenditure ratio (GE) is used to control against the macroeconomic stability, because theory suggests a negative relationship between
11
expected that government expenditure has a positive effect on economic growth (Devarajan, Swaroop, & Zou, 1996). This variable is computed by the annual percentage growth of general government final consumption expenditure as a yearly percentage of GDP.
Another control variable is the total trade ratio (XM) represents the openness to international trade. This variable is used because openness to international trade can influence economic growth (Edwards, 1993). This variable is computed by the sum of exports and imports of goods and services measured a percentage of GDP.
The following control variable inflation (IF), represents the inflation as measured by the consumer price index (CPI) and it reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services. It is measured by the CPI since the European Central Bank (ECB) uses this measurement for their inflation target (Svensson, 1999). This control variable is used to control against the influence of inflation on economic growth because economic theory states that inflation reduces investment and productivity growth and therefore has a negative effect on economic growth (Fischer, 1993).
The final control variable is the foreign direct investment (FDI). This variable controls for the effect of foreign capital inflows on economic growth. This is based on the theory of Borensztein, De Gregorio & Lee (1998), they state that FDI leads to higher productivity when the host
country has sufficient absorptive capability of advanced technologies. Alfaro, Chanda, Kalemli-Ozcan, & Sayek also find that well-developed stock markets gain significantly from FDI (2004). The fixed effect (FE) estimator is added to both regressions to adjust for the fixed effect. It is assumed that time-independent effects for each country can be correlated with the regressors. (Borenstein, Hedges, Higgins, & Rothstein, 2010).
This model is based on the theory Demirgüç-Kunt and Levine (1996), Levine (1997) and Levine and Zervos (1996) to measure the stock market development accurately, the rest of the model is in line with the theory to control against other variables. Therefore the model is sound to test the effect of stock market development on economic growth within the available data. The
shortcomings of this model are that not all influences on economic growth can be determined due to the lack of data. With all data available the model would be enriched with more variables
12
to control for other factors influencing the economic growth. In reality, it is likely that the economic growth of a country is influenced by factors that are not possible to measure, one of those variables can be black market activity. This missing data may result in omitted variable bias.
3.2 Data Collection
Yearly data is collected from The World Bank (2017) for eleven developed countries within the Eurozone to create a cross-country and cross-time panel. The data set covers the years 1975 – 2015 from the countries that joined the Eurozone in 1999: Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, Netherlands, Portugal and Spain.
The combined data from the market capitalization of listed domestic companies as a percentage of GDP, turnover ratio of domestic shares and the total value of stock traded as a percentage of GDP is used to determine the stock market development index as computed by Demirgüç-Kunt and Levine (1996).
The SMDI is only calculated when the data of all three variables within the model (model 1.1) is available. Otherwise, the value of SMDI is left out in the dataset. There exists an error in the dataset regarding the market capitalization of Spain in 1977; this value is 0, which is not presumable to be an actual value since it is unlikely that there was no trading during an entire year on the Spanish stock market. Therefore, the value of the SMDI of Spain in 1977 is removed. This solution is preferred over the smoothing of the value by taking the average. With this
removal, the total observations decrease by one. Therefore the remainder of the total data set for the regressions consists of 11 countries with a total of 347 observations. The summary of the variables is displayed in Appendix B.
3.3 Research Hypothesis
As an explanation of the discussion described in the literature review, it is expected that the effect stock market development is positive on economic growth, but the marginal effect decreases when stock market development increases. Therefore, the expectation is that the 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆2 variable in regression 1.0 is significantly smaller than 0.
13
𝐻𝐻𝐷𝐷𝐻𝐻𝐻𝐻𝐻𝐻ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝐴𝐴: 𝐻𝐻0: 𝜷𝜷𝟐𝟐= 0, 𝑎𝑎𝑎𝑎𝑎𝑎𝑒𝑒𝑎𝑎𝑒𝑒𝐻𝐻 𝐻𝐻1: 𝜷𝜷𝟐𝟐< 0.
The second expectation is that the impact of 𝜟𝜟𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺 on 𝑮𝑮𝑮𝑮𝑮𝑮 is lower at higher levels of 𝑺𝑺𝑺𝑺𝑮𝑮𝑺𝑺, this is tested against the following hypothesis on the regression 2.0:
𝐻𝐻𝐷𝐷𝐻𝐻𝐻𝐻𝐻𝐻ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝐵𝐵: 𝐻𝐻0: 𝜷𝜷𝟑𝟑= 0, 𝑎𝑎𝑎𝑎𝑎𝑎𝑒𝑒𝑎𝑎𝑒𝑒𝐻𝐻 𝐻𝐻1: 𝜷𝜷𝟑𝟑< 0.
Both hypotheses are tested with a one-sided t-test with 1%, 5%, and 10% significance levels. The results of these test are described in the following section.
4. Results and Discussion 4.1 Results Analysis
The dataset is regressed with robust standard errors to adjust for assumed heteroskedasticity. A fixed effect regression is used to analyze the impact of the variables over time within the country. This option is used to adjust for exogenous factors of each country separately, thus without correlation between those error terms. Under the fixed effect model, the overall error variance is smaller, compared to the random effect model (Borenstein, Hedges, Higgins, & Rothstein, 2010).
The results of the first fixed effect regressions are presented in Table 1. According to these results, the coefficients of 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 and 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆2 are not significant. The coefficient of government expenditure is significant, which is expected since government expenditure is part of the GDP accounting identity. Inflation is also proven to have a significant coefficient; this variable is also included in the GDP accounting identity, and therefore this relation is expected. The openness to trade is only significant in the regression with all the available variables and has a positive effect on economic growth. An explanation for this result is the correlation between FDI and XM, as seen in Appendix A the correlation between those variables is 0.5151, so in the regression where FDI is included, XM is significant because the variables correlate with each other. This
significance of XM is expected according to the theory about the influence of international trade on economic growth.
Because of the insignificant results of SMDI and SMDI2, there is not enough evidence to reject the null-hypothesis of the A hypothesis. Thus, the results in Table 1 provide no evidence that the
14
economic growth is influenced by stock market development or a decrease in the effect of stock market development on economic growth.
Table 1: Fixed effect regression on Economic Growth (GDP)
Number of variables: (1) (2) (3) (4) (5) (6) Constant 2.030924 2.046386 1.142548 1.730244 1.997859 -1.524682 (11.11)*** (6.29)*** (3.2)*** (1.34) (1.49)* (-0.74) SMDI -0.004942 -0.0056039 -0.0004052 0.0037756 -0.0043806 -0.0100769 (-1.22) (0.06) (-0.04) (0.38) (-0.32) (-0.64) SMDI2 0.00000417 -0.0000265 -0.000043 -0.0000111 0.00000325 (-0.48) (-0.42) (-0.7) (-0.15) (0.04) GE 0.3346153 0.3160642 0.3543455 0.4200793 (5.73)*** (3.92)*** (3.75)*** (3.69)*** XM -0.0068918 -0.0053458 0.0314488 (-0.56) (-0.4) (1.52)* INF -0.0898924 -0.0739697 (-2.85)*** (-2.17)** FDI -0.0050951 (-0.24) R-squared (overall) 0.012 0.0121 0.1046 0.0422 0.0776 0.0309 Observations 347 347 347 347 331 286 Groups 11 11 11 11 11 11
Note: constant included, t statistics in the parentheses. * Significant at 10%; ** significant at 5%; *** significant at 1%.
The output of the second regression is stated in Table 2. This regression is used to test the B hypothesis. According to the results, the “Dummy * Δ SMDI” variable is significant in most of the regressions and has a positive influence on economic growth. Furthermore, this regression also proves the significance of government expenditure, which is expected since government expenditure is included in the GDP accounting identity. Foreign direct investment is also proven to have a positive effect on the GDP, which is in line with the view of Borensztein et al. (1998). The openness to trade is significant in the regression with five and six variables, but not in the last regression. The difference between the significant in the two regressions is explained by the correlation of 0.5006 between XM and FDI (Appendix A). So the effect of XM in the last regression is covered by FDI.
15
Table 2: Dependent variable: Economic Growth (GDP)
Variables: (1) (2) (3) (4) (5) (6) (7) Constant 1.69156 1.684566 1.723019 0.7643175 2.351191 2.218655 -0.7651897 (66.05)*** (6.89)*** (6.83)*** (2.14)** (3.54)*** (2.77)*** (-0.57) Δ SMDI 0.3501713 0.3491347 -0.0209536 0.0991511 0.068732 0.0609559 0.0953909 (2.09)** (1.97)** (-0.13) (0.59) (0.38) (0.35) (0.61) Dummy 0.0140134 -0.0480949 0.0608408 0.4128111 0.4259477 0.6592763 (0.03) (-0.1) (0.15) (1.32)* (1.21) (1.68)** Dummy * Δ SMDI 0.4637785 0.3301118 0.3104862 0.3333866 0.3032045 (2.55)*** (1.46)* (1.57)* (1.57)* (1.24) GE 0.3835709 0.3402632 0.3541755 0.3976454 (4.45)*** (3.65)*** (3.74)*** (3.65)*** XM -0.0170523 -0.0157467 0.0106895 (-4.5)*** (-3.17)*** (0.94) INF -0.0178042 0.0214493 (-0.65) (0.63) FDI -0.0189143 (-3.21)*** R-squared (overall) 0.0136 0.0134 0.016 0.1574 0.0226 0.0272 0.1234 Observations 335 335 335 335 335 319 274 Groups 11 11 11 11 11 11 11
Note: constant included, t statistics in the parentheses. * Significant at 10%; ** significant at 5%; *** significant at 1%.
Thus, the findings of the "Dummy * Δ SMDI" variable is the contrary of the expected. Therefore the null hypothesis cannot be rejected. The positive effect indicates that a higher developed stock market promotes the economic growth more than a lower developed stock market. These
findings can be explained by the results of Levine et al. (2000).
4.2 Discussion
The influence of stock markets can affect the financial development as the economic growth simultaneously (Caporale, Howells, & Soliman, 2004). Therefore the regression can suffer from simultaneity bias. (Caporale, Howells, & Soliman, 2004) lagged data for stock market
development is used.
The Eurozone in this research is represented by the eleven countries which adopted the Euro as a currency in 1999. This sample is representative for the entire Eurozone, but a more accurate sample would be to take the data of the entire Eurozone at this moment. The sample may suffer from selection bias, due to the entry conditions of the Eurozone. Those conditions, known as the ‘Maastricht criteria’ and they influence the monetary policy of the adopting country. Eventually,
16
all the participating countries of the Eurozone have to adjust to the convergence criteria, and therefore this selection can still represent the entire Eurozone.
The data to compute the SMDI of Spain in 1977 is omitted due to the dubious value of zero for the market capitalization in that year. Considering the box plots in Appendix C, some values stand out in the data set, and this could indicate the possibility of outliers. These outliers could not be explained as invalid or wrong data. Therefore they are not excluded from the data set. These outliers may influence the findings of the research. Furthermore, the internal validity of the research potentially suffers from omitted variable bias; this occurs because there are many influences on the economic growth which are difficult of unable to measure, such as the effect of the black market on GDP.
5. Conclusion
This research analyses the marginal of stock market development on economic growth within the Eurozone. According to Demirgüç-Kunt & Maksimovic (1996), Levine and Zervos (1998) and Caporale, Howells & Soliman (2004) stock market development has a positive effect on
economic growth. While Ram (1999), Dawson (2003) and Boubakari and Jin (2010) found that there was no evidence of a link between stock market development and economic growth. One of the reasons for the differences in findings is the measurement of stock market development. In the literature about this relation, some researchers used market capitalization to determine the development of the stock market, while Demirgüç-Kunt & Maksimovic (1996) choose for an index. Demirgüç-Kunt and Levine (1996) and Boubakari & Jin (2010) used an index which consisted of the average of market capitalization, the total value of shares traded and turnover ratio as used in this research.
The relation was studied in this research over a panel data set with annual observations from 1975-2015 for the eleven countries which joined the Eurozone in 1999. The findings in this study do not prove that there is a relation between stock market development and economic growth in the Eurozone, this is in line with the research of Ram (1999) and Dawson (2003), they do not find empirical evidence that financial development promotes economic growth. Furthermore, there is no indication in the results that there is a decrease in the marginal effect of stock market development on economic growth. In contrast, the results indicate that the effect of stock market
17
development increases at higher levels of development. Thus, the "too much finance" effect of Arcand et al. (2012) is not proven to be similar in the stock market development of the Eurozone. Due to these findings, the implications for policymakers are that there is no prove for a “too much development“ effect and therefore the development of a stock market should not be constrained. On the other hand, there is also a lack of evidence of the positive effect of stock market development on economic growth. Thus, there is no need for policy makers to encourage stock market development with the goal to increase economic growth.
Although, there are limitations in this research. As described due to the lack of a complete data set, there can be an omitted variable bias. Furthermore, the sample of eleven countries within the Eurozone can suffer from selection bias, since the countries which first joined the Eurozone had to satisfy the entrance requirements sooner. Therefore they may conduct a different monetary policy.
Due to the different findings and the discussion in the literature, there is a need for further research on this subject. It is recommended to expand the sample to determine the effect in different countries and varying degrees of developed stock markets. Another recommendation is to add more variables to control for other influences. Lastly, it is recommended to compare the different models that are used to determine stock market development, since the measurement of stock market development remains abstract.
18 Bibliography
Alfaro, L., Chanda, A., Kalemli-Ozcan, S., & Sayek, S. (2004). FDI and economic growth: the role of local financial markets. Journal of International Economics, 64, 89-112.
Arcand, J. L., Berkes, E., & Panizza, U. (2012). Too much finance? IMF Working Paper No.12/161.
Arestis, P., Demetriades, P. O., & Luintel, K. B. (2001). Financial development and economic growth: the role of stock markets. Journal of Money, Credit and Banking, 33(1), 16-41.
Beck, N., & Katz, J. N. (1995). What to do (and not to do) with time-series cross-section data. The American
Political Science Review, 89(3), 634-647.
Borenstein, M., Hedges, L. V., Higgins, J. P., & Rothstein, H. R. (2010). A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, 1(2), 97-111.
Borensztein, E., De Gregorio, J., & Lee, J. W. (1998). How does foreign direct investment affect economic growth? Journal of International Economics, 45(1), 115-135.
Boubakari, A., & Jin, D. (2010). The role of stock market development in economic growth: Evidence from some Euronext countries. International Journal of Financial Research, 1(1), 14-20.
Caporale, G. M., Howells, A., & Soliman, A. M. (2004). Stock market development and economic growth: the causal linkage. Journal of Economic Development, 29(1), 33-50.
Dawson, P. J. (2003). Financial development and growth in economies in transition. Applied Economics
Letters, 10(13), 833-836.
Demirgüç-Kunt, A., & Levine, R. (1996). Stock market development and financial intermediaries: stylized facts.
The World Bank Economic Review, 10(2), 291-321.
Demirgüç-Kunt, A., & Maksimovic, V. (1996). Stock market development and financing choices of firms. The
World Bank Economic Review, 10(2), 341-369.
Devarajan, S., Swaroop, V., & Zou, H. F. (1996). The composition of public expenditure and economic growth.
Journal of Monetary Economics, 37(2), 313-344.
Easterly, W., & Rebelo, S. (1993). Fiscal policy and economic growth. Journal of Monetary Economics, 32(3), 417-458.
Edwards, S. (1993). Openness, trade liberalization, and growth in developing countries. Journal of Economic
Literature, 31(3), 1358-1393.
Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32(3), 485-512.
Garcia, V. F., & Lui, L. (1999). Macroeconomic determinants of stock market development. Journal of Applied
Economics, 2(1), 29-59.
Levine, R. (1997). Financial development and economic growth: views and agenda. Journal of Economic
19 Levine, R. (1999). Law, finance, and economic growth. Journal of Financial Intermediation, 8(1), 8-35.
Levine, R., & Zervos, S. (1996). Stock market development and long-run growth. The World Bank Economic
Review, 10(2), 323-339.
Levine, R., & Zervos, S. (1998). Stock markets, banks, and economic growth. The American Economic Review,
88(3), 537-558.
Levine, R., Loayza, N., & Beck, T. (2000). Financial intermediation and growth: Causality and causes. Journal of
Monetary Economics, 46(1), 31-77.
Obstfeld, M. (1992). Risk-taking, global diversification, and growth. NBER Working Paper No. 4093. Pagano, M. (1993a). Financial markets and growth: an overview. European Economic Review, 37(2-3),
613-622.
Pagano, M. (1993b). The flotation of companies on the stock market: A coordination failure model. European
Economic Review, 37(5), 1101-1125.
Ram, R. (1999). Financial development and economic growth: Additional evidence. The Journal of
Development Studies, 35(4), 164-174.
Rousseau, P. L., & Wachtel, P. (2000). Equity markets and growth: Cross-country evidence on timing and outcomes, 1980–1995. Journal of Banking & Finance, 24(12), 1933-1957.
Sahay, R., Čihák, M., N'Diaye, P., & Barajas, A. (2015). Rethinking financial deepening: Stability and growth in emerging markets. Revista de Economía Institucional, 17(33), 73-107.
Schumpeter, J. A. (1911). The theory of economic development. Cambridge, MA: Harvard University Press. Svensson, L. E. (1999). Inflation targeting as a monetary policy rule. Journal of Monetary Economics, 43(3),
607–654.
20 Appendix
Appendix A – Correlations
Correlations between the variables in model 1.0
GDP SMDI SMDI2 GE XM INF FDI
GDP 1 SMDI -0.093 1 SMDI2 -0.0733 0.919 1 GE 0.2801 -0.0093 0.0485 1 XM 0.0553 0.0108 -0.0479 0.0109 1 INF 0.0295 -0.3452 -0.1985 0.3726 -0.2524 1 FDI 0.0304 0.0664 0.0226 0.06 0.5151 -0.1119 1
Correlations between the variables (non-lagged SMDI)
GDP SMDIT SMDIT 2 GE XM INF FDI GDP 1 SMDIT 0.0692 1 SMDIT 2 0.1024 0.9195 1 GE 0.371 -0.0358 0.0242 1 XM 0.0016 0.0135 -0.041 0.0215 1 INF 0.0589 -0.3667 -0.2051 0.3782 -0.2514 1 FDI -0.0913 0.0878 0.048 0.0663 0.5003 -0.1063 1
Correlations between the variables in model 2.0
GDP
Δ
SMDI Dummy
Dummy*
ΔSMDI GE XM INF FDI
GDP 1 Δ SMDI 0.1521 1 DUMMY -0.0015 0.0726 1 Dummy* ΔSMDI 0.1592 0.896 0.1373 1 GE 0.3616 -0.0234 -0.0864 -0.0103 1 XM -0.0081 -0.0561 0.1616 -0.0281 0.0204 1 INF 0.0819 -0.0425 -0.4149 -0.0834 0.3818 -0.2453 1 FDI -0.0906 -0.0366 0.1569 -0.0217 0.0697 0.5006 -0.1058 1
21
Appendix B: Variable Summary
Variable Summary Obs. Mean Std. Dev. Min Max
Country 440 6 3.165877 1 11 Year 440 20.5 11.55654 1 40 GDP 440 1.964213 2.756384 -8.70669 25.63721 SMDIt-1 347 45.08282 37.58242 2.356035 234.6392 SMDI2t-1 347 3440.829 6243.06 5.550902 55055.57 GE 440 2.415417 2.410263 -4.98018 11.79841 XM 440 93.56187 63.83242 27.76174 391.4974 INF 424 4.512064 4.975543 -4.47994 28.78333 FDI 388 5.309347 16.97642 -58.9777 255.4233 Δ SMDI 335 0.149786 0.815456 -0.91979 11.61223 DUMMY 340 0.508824 0.500659 0 1 DUMMY* Δ SMDI 340 0.106127 0.728888 -0.63873 11.61223 SMDI (non-lagged) 341 45.68178 37.62956 2.356035 234.6392 SMDI2 (non-lagged) 341 3498.656 6282.461 5.550902 55055.57
22
Appendix C – Boxplots per variable, per country
-1 0 0 10 20 30 G DP Aust ria Bel gium Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in 0 50 1 00 1 50 2 00 2 50 SMD I Aust ria Bel gium Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in
23 -5 0 5 10 15 GE Aust ria Bel gium Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in 0 1 00 2 00 3 00 4 00 XM Aust ria Bel giu m Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in
24 -1 0 0 10 20 30 INF Aust ria Bel gium Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in -1 00 0 1 00 2 00 3 00 F DI Aust ria Bel giu m Finl and Franc e Ger man y Irela nd Italy Luxem bour g Neth erla nds Por tug al Spa in
