• No results found

S TOCK M ARKETS T HE S YNCHRONICITY OF THE E URO ZONE

N/A
N/A
Protected

Academic year: 2021

Share "S TOCK M ARKETS T HE S YNCHRONICITY OF THE E URO ZONE"

Copied!
28
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

1

T

HE

S

YNCHRONICITY OF THE

E

URO ZONE

S

TOCK

M

ARKETS

F.A. Plescau (s2230011) July, 2013

Rijksuniversiteit Groningen Faculty of Economics and Business Master of Science: Business Administration

Specialization: Finance Supervisor: dr. J.O. Mierau

Keywords: Stock price synchronicity, institutional quality, financial crisis JEL-classification: G01, G15, G38

ABSTRACT

(2)

2

Table of Contents

I. Introduction ...3

II. Literature review ...4

Stock price synchronicity ...5

Institutional index ...6

III. Data collection and descriptive statistics ...8

Dependent variable ...8

Independent variables ... 10

IV. Methodology ... 14

Measurement of stock price synchronicity ... 14

Model specifications ... 15

V. Empirical Results ... 16

VI. Conclusions ... 21

References ... 23

(3)

3

I.

Introduction

A growing body of recent work investigates linkages between stock prices synchronicity and the real economy as well as the institutional environment of a country. Synchronicity of stock price movement refers to the level of deviation from a random walk of the stock prices, up or down, en masse.

We may ask ourselves why it is so important to observe if the markets are synchronous or not. One of the reasons is linked with the arbitrageurs’ work. They make money by predicting the move that we are referring to, as a response to swings in other prices, laws, consumer tastes, technology, or other factors.

Roll (1988) argues that stock synchronicity can push the value of a stock from its fundamental value and this over- or under-valuation of stocks can adversely affect the overall equity market. The mechanisms that underline the risk arbitrage and many other specific risks show the stock market as an information processor. Further, Campbell et al. (2001) argue that large investors are exposed to greater risk as stock synchronicity increases.

Another reason aims at the immediate implications of the stock price movement for corporate governance. Roll (1988) finds that when stock prices begin to fall, various corporate governance mechanisms come into effect. If the way that the stock price is moving brings an efficient capital allocation, Tobin (1984) names it a functionally efficient stock market. If the stock market is functionally inefficient, this causes serious problems, because the mechanisms of the corporate governance can misfire and this brings undesired microeconomic capital allocation measures. Furthermore, this reflects in a waste of economy’s capital and finally in a slower economy growth. We can easily see how the movement of the share prices has immediate implications in corporate governance. In concordance with this, a high synchronicity of the stock prices predicts signs of inefficient capital allocation while, in contrast, the independency of the stock market is attributed to a stock market that is functionally efficient.

(4)

4

in countries with below average corruption, stronger investor protection laws are associated with higher firm-specific variation.

Following this path of thinking, the purpose of this study is to analyse the co-movement behaviour of the Euro zone stock markets and to observe the relationship between various institutional governance mechanisms and the stock price synchronicity in the context of global financial crisis.

While the euro zone stock market has been well addressed in many ways there is very little empirical analysis on its stock price synchronicity behaviour. Also, despite the number of empirical researches concerning the stock market synchronicity, there is a lack of papers analysing the effects of the late financial crisis on the stock price co-movement. This problem is relevant mainly due to the fact that recent financial turbulences affected all aspects of economy. Therefore, the paper will contribute to this literature gap with new evidences which show that during unexpected financial turmoil a good perception of public institutions does not explain the stock market synchronicity adequately. The introduction of the crisis influence in our analysis renders the development of the stock markets insignificant in explaining stock price synchronicity.

Such a study can draw a warning concerning the necessity of developing stronger preventive measures for the protection of financial markets, and also a more thorough investigation of the cross-country connections which leads to the transmission of country-specific shocks.

We arrange this paper as follows: Section 2 reviews the existing literature on stock price synchronicity and institutional indices. Section 3 includes the data development and data description. Section 4 describes the research methodology. The presentation and analysis of the empirical results are discussed in Section 5. The concluding remarks are outlined in Section 6.

II.

Literature review

(5)

5 Stock price synchronicity

The synchronicity of stock price movement has received a lot of attention in the literature. Practically, synchronicity is the tendency for the stocks in a market to co-move. The stock market movement can be downward or upward, but only in the same direction, according to the trend of that market.

Previous literature has shown that the stock price co-move more in low GDP economies than in high GDP economies. Morck et al.(2000) found that there is a clearly negative correlation between per capita GDP and stock price synchronicity. This means that the stock prices in countries that are poorer present a more synchronous movement over time, while those in rich economies do not present this particularity. This finding is an important breakthrough in understanding the effects of the capital market infrastructure on the development of capital markets around the world.

Durnev et al. (2004a) also noted that stock returns in economies in transition, precisely the low GDP ones, present a synchronized movement. In addition, countries with a low GDP per capita were found to register smaller equity and debt markets, and also a high degree of corruption rates.

We have stated in the first part that the synchronicity of the markets is a signs of efficient or inefficient capital allocation. Indeed, a wide area of papers argues that the share price synchronicity influence the efficiency of capital allocation through its effect on stock price informativeness. For example, Durnev, Morck and Yeung (2004) show that companies which exhibit less synchronicity tend to use more external financing and allocate capital more efficiently. In the same vein, Wurgler (2000) finds that the stock price synchronicity is negatively correlated with efficiency of capital allocation across countries in domestically traded stock returns.

(6)

6

companies, using as descriptive variable the trading volume of stocks. They conclude that when there are few large firms in a market, they tend to dominate the entire market movements and that the size of a company is important for the market-wide stock price changes.

The findings of Morck et al. (2000) had an important impact on further studies, but in literature exists some other different results, depending on the sample that was used or the new economic frameworks. For example, Hsin and Liao (2003) discovered that when the sample of the countries that are analyzed is restricted to a specific group, in their case, emerging markets, the hypothesis of Morck et al. (2000) does not hold anymore. Therefore, low income countries do not exhibit higher price synchronicity within the spectrum of emerging markets.

Also, Walti (2011) applied their analysis on fifteen developed economies and focussed their attention mainly on the relationship between the co-movement of the stock markets and the monetary integration. They found that monetary integration leads to higher stock market synchronicity both through the common monetary policy and through the elimination of exchange rate volatility. Trade and financial integration also contribute to a stronger synchronicity.

Previous literature presents us a strong nexus between stock price synchronicity and structural characteristics of an economy. Furthermore, the stock market co-movement is intensively studied because of its informative power.

Institutional index

(7)

7

Empirically, trying to find a way to capture the degree of private property rights protection for the countries that form their data base, Morck et al. (2000) used an index named Good Government. This index is composed of the sum of three indices and was constructed before by La Porta et al. (1998): (i) government corruption; (ii) the risk of expropriation of private property by the government; (iii) the risk of the government repudiating contracts. These indices can reach values from zero to thirty; lower values indicate a low protection of private property rights.

In a study of Griffin et al. (2007a) the findings of Morck et al. (2000) were replicated and they argued that the result obtained by the latter is not solid to other combination of indices that can explain good government. Furthermore, they stressed that La Porta et al. (1999) did not suggest a specific combination of the indices they developed as the most appropriate definition of good government.

In other studies, when the index of the good government is instilled into the model with other competing variables such as the anti-director rights index (Morck et al., 2000), market liquidity, short sales restrictions (Bris et al., 2007), or insider trading laws (Fernandes and Ferreira, 2009), the explanatory power becomes weaker.

As an answer to this controversy, we propose another indicator - the Worldwide Governance Indicators (WGI) from Kaufmann, Kraay and Mastruzzi (2009). WGI is one of the most comprehensive and popular compilations of cross-country data on institutional governance. This indicator follows six dimensions of the governance quality: (i) voice and accountability; (ii) political stability and absence of violence; (iii) government; (iv) regulatory quality; (v) rule of law; (vi) control of corruption. The definition of the terms, as presented in Kaufmann et al. (2009) is detailed in Appendix 1.

We will name it Institutional Quality Index and it will be the aggregate value of the previously mentioned dimensions. Faria and Mauro (2005) name this index ”the state of the art” among indicators of institutional quality because they are the summary compounded measure that contains the widest attainable set of such indicators.

(8)

cross-8

country stock price synchronicity. As a result, in our paper, we analyze whether this relation is supported when using a different sample of countries and a different period of time. In addition to this, we study whether the financial crisis had any impact on these relations. The result can help the policymakers have a clear perspective on developing an appropriate macro framework in order to ensure the macro-efficiency of the stock markets (Jung and Shiller, 2005).

III. Data collection and descriptive statistics

We begin by analysing weekly stock returns generated by Datastream service, in order to calculate annual synchronicities for representative stock markets of Europe. We calculate the fraction of stocks in a country that move in the same direction, filtered out those whose prices did not move, and charted the number that rose and fell. Our sample covers the 17 markets from Euro zone and the analysed period extends from January 1, 2005 to December 30, 2010. We consider the blue chip companies that formed the main index of each country in this period. We also take into account the firms that are no longer traded, but whose equity prices were covered by the Datastream information service. A company was included in our estimation only if more than 30 weeks were available for recently delisted stocks or newly listed ones, to ensure that we use apposite information for our synchronicity measure.

Dependent variable

Our dependent variable refers to the level of the stock price synchronicity in a country. The summary of data statistics for and also, for measure is presented in Table 1. As we can observe, the average synchronicity result for the whole sample is 72.9% from 2005 to 2010, with a minimum value of 68.6% and a maximum of 77%.

(9)

9

and Malta (68.6%) have the lowest level of synchronicity with the lower standard deviation (Italy (1.4%), Luxembourg (2%) and Malta (1.7%)).

Table 1

Description of main variables

Invariable statistics (Sample: 17 Euro zone countries) - This table reports the descriptive statistics used to analyze the data series. The average value is described through either the mean value or the median value. The maximum and minimum represent the higher synchronicity value and the lower synchronicity value that the market index has taken during the full sample period (from 2005 to 2010).

Variables Mean Standard

Deviation Minimum Maximum Stock co-movement indices

Average fraction of stocks moving the same direction )

0.729 0.006 0.686 0.770

Logistic transformation of synchronicity

( ) -0.073 0.020 -0.228 0.074

GDP per capita, Euro 27121 3703 10538 75835

Logarithm of per capita GDP 4.379 0.053 4.020 4.879

Logarithm of Market capitalization 1.641 0.326 0.800 2.268 Structural variables

Annual growth in GDP per capita, % 1.014 0.315 -0.698 4.930

Variance in GDP growth 9.783 10.134 2.601 43.206

Inflation, % 2.314 0.196 1.556 4.728

Variance in domestic inflation rate 2.501 3.001 0.292 11.699 Institutional variables

Voice and accountability 1.243 0.022 0.847 1.702

Political stability and absence of

violence 0.828 0.042 -0.316 1.577 Government effectiveness 1.316 0.045 0.302 2.240 Regulatory quality 1.322 0.031 0.653 1.924 Rule of law 1.295 0.047 0.279 1.975 Control of corruption 1.250 0.065 -0.121 2.490 Institutional Index 0.484 0.155 0.225 0.746

(10)

10

Table A4 illustrates the descriptive statistics for the level of stock price synchronicity over these two sub-periods. The results presented in Table A4 show important differences between the levels of synchronicity in the two sub-periods. The highest increases in synchronicity were observed in Malta (13.55%), Germany (12.56%) and Portugal (10.72%). On the other hand, a decrease in synchronicity was encountered in Slovenia 3.03%), Italy 2.38%), Greece (-2.08%), and Luxembourg (-1.01%). The overall increase in synchronicity was of 5.20% in the second period, compared to the first one.

We can also observe in Table A4 that some countries, like Slovakia (217.39%), Slovenia (104.76%) and Cyprus (164.28%) registered a climb in the value of the standard deviation, which means that in the second sample the synchronicity measure results are spread out over a large range of values, while in Belgium 62.5%), Portugal 61.53%) and Germany (-47.82%) we observe a slump in value, which indicates that the new results are even closer to the mean value.

Independent variables

We will consider a first set of variables which is based on the economic characteristics of each country. These variables were used in a majority of studies that dealt with the synchronicity topic. Subsequent studies included those variables either to replicate earlier findings or to observe the explanatory force of additional ones (Jin and Myers, 2006; Griffin et al., 2007; Bris et al., 2007; Fernandes and Ferreia, 2008).

The set of variables used to highlight the link of the synchronicity with the cross-country economic characteristics are part of the proxies used by Morck et al. (2000) and refers to the logarithm of marketing capitalization (or market value) to observe the effect of market size. We use this measure as representative for stock market development and we consider it to be appropriate because it is less arbitrary than other individual measures and indexes (Levine and Zervos, 1996) of stock market development. The assumption underlying the use of this variable in our analysis is that the size of the share market is positively correlated with the possibility of diversifying risk and mobilizing capital.

(11)

11

synchronicity and per capita GDP. We notice that even if the simple correlation between those two is negative, as expected, our regression result shows, contrary to the ones of Morck et al. (2000), that GDP per capita is positively correlated with the stock price synchronicity measure at the 1% significance level. This can be surprising because we assumed that high GDP per capita economies renders lower level of synchronicity in the markets. We confined our analysis to the 17th countries from Euro zone, which means that we use data from developed economies. Our results show that GDP per capita is positive and statistically significant in explaining the cross-sectional differences in stock price synchronicity.

Therefore, the main reason for this discrepancy in results is that Morck et al. (2000) based their analysis on both emerging markets (mostly low-income economies) and developed markets (mostly high-income economies). By enclosing our sample to developed markets, the statement that high-income economies exhibit low stock market synchronicities does not hold anymore. This finding further implies that there are some other factors (other than per capita GDP) that explain the market synchronicities. We consider that in our case, market capitalization level is a better indicator for a country stock market development, because it represents the way a stock market works. More, this indicator is less biased by our choice of Euro zone stock markets sample.

Another variable is per capita gross domestic product (GDP) which stands for the development of an economy. Further, we use variance of per capita GDP growth to represent macroeconomic instability (higher synchronicity might result from higher macroeconomic instability). The last ones stand for structural variables.

If we include these variables in a multivariate regression and we notice that the market development becomes insignificant, we can say that market capitalization stands for these structural characteristics.

To calculate the variance in GDP growth, we will apply a rolling window method in order to obtain different values for the 2005-2010 period. The variable is measured as the variance of the previous three year GDP per capita growth. The same approach will be used to calculate the variance of inflation, as a control variable.

(12)

12

and inflation. The average GDP per capita is € 27,121, the mean annual GDP growth is 1.01% and the average inflation is 2.31%.

A second set of independent variables refers to the Institutional Index components. Morck et al (2000) analyze the effects of several country and firm level variables to identify the major reasons for the variations in stock price co-movements between countries. They find that the primary reason for the variations is the differences in the property rights protection arrangements of the countries they examined. Following Morck et al’s (2000) contention that investor protection is the primary driver of stock price synchronicity we use several proxies for investor protection, focusing on corporate governance development. Our proxies are Regulatory Quality, Government Efficiency, Rule of Law, Political Stability, Control of Corruption, and Voice and Accountability, which form The Worldwide Government Indicator.

The Worldwide Government Indicator (WGI) is a project developed by researchers from the World Bank - Kaufmann, Kraay and Mastruzzi (2009) - which reports individual and aggregate governance indicators for 212 territories and countries. The values of the indicators are obtained based on the view of a large number of survey respondents, citizens, or companies and the source of these views are policy institutions, survey institutes, international organizations, or non-governmental institutions. The World Bank uses a model which aggregates the wide sample of responses to six broad clusters. The name of the model is Unobserved Component Model (UCM) and has the particularity of approximating an indicator by aggregating the scores from previous responses, as soon as an indicator like Violence, for example, cannot be observed directly, but can be approximate.

The indicators cover a large area of governance and institutions definitions and for this reason has been widespread in the researches. The main benefit is that, like the International Country Risk Guide (ICGR) indicator, WGI indicators can be used either individually or as an aggregate value. For example, Rodrik et al. (2004) and Dollar and Kraay (2003) use in their research only the rule of law index, while Easterly and Levine (2003) construct a compact indicator composed from all six indexes.

(13)

13

In order to form the aggregate value of the Institutional Quality Index, we will estimate the normalized value. We divide their sum by the maximum value that they can take, multiplied by six - the total number of dimensions.

The descriptive statistics of the Institutional variables are shown in Table1. Each variable ranges from approximately -2.5 (weak) to 2.5 (strong), representing the governance performance. The lowest average value was obtained for Political stability and absence of violence, while the highest was for Regulatory quality. The deep drop in the Political stability and absence of violence score (-0.316) was generated by Spain in 2009, and by Greece in 2010 for Control of corruption score (-0.121). On the other hand, the most important climb in the governance score was obtained by Finland in 2006 (2.49) for Control of Corruption and for Government effectiveness in 2006 and 2010 (2.24).

The average aggregate Institutional Index value is 0.484, obtaining a minimum score of 0.225 and a maximum of 0.746.

Table 5 presents Pearson correlation among regression variables. Consistent with prior research, we obtain a negative correlation equal to 0.052 between stock price synchronicity and per capita income, which means that we encounter a reduced level of the stock market synchronicity in countries with developed economies. Furthermore, our preliminary analysis regarding the correlations among variables indicates that low-income economies tend to have smaller numbers of listed stocks and greater volatility in GDP growth and also in inflation rate.

The number of listed companies in a country has a negative correlation of 3.27% with synchronicity, which means that stocks co-move more strongly in more concentrated markets. Price synchronicity is positively correlated with both GDP growth variance (0.203) and Inflation variance (0.038). Overall, these correlations suggest that no structural variable is likely to explain the link between market capitalization and stock price synchronicity.

(14)

14

listed companies in each country, which reflects the fact that more stocks are traded on a market in a country with stronger institutions.

IV. Methodology

Our study wants to analyse the relation between synchronicity and a series of cross-country macro characteristics, which include structural and institutional variables.

We will follow a few steps: firstly, we assume that the attributes of stock market structure are important in determining the degree of stock market synchronicity within. In particular, a share market that is more developed, in terms of the level of market capitalization, is expected to experience lower levels of synchronicities. Therefore, we will estimate a panel model in order to see if certain economy characteristics, which we include in the model, influence or are related in any way to synchronicity.

Secondly, a vector formed by the dimensions of the Institutional Index will be included in the regression and will be analysed not only as a whole, but also individually, in order to observe which part has a greater influence on the stock price deviations from a random walk over time. Finally, we add crisis dummies, to see if there is any change regarding our results during the crisis period.

Measurement of stock price synchronicity

Morck et al. (2000) had a very important contribution in finding a measure for stock price synchronicity. They propose two measures: Classical synchronicity approach and R-square measure.

(15)

15

The fraction of stocks that move in the same direction in country j is:

) ∑ ( )

where denote the number of stocks in country j whose prices rise in period t, is the

number of stocks whose prices decline, and T is the number of periods used. To avoid any bias due to non-trading, we drop those stocks whose prices do not present any move.

This measure has a limitation of 1.0 for markets where the stock prices are perfectly synchronized and one of 0.5 where there is an equal number of rises and falls over the period, consistent with a market where prices do not present synchronized move. In order to enlarge our range of values, we use a standard econometric remedy and we convert into an adjusted measure of synchronicity using the logistic transformation, as presented in equation (2):

)

)

Now, in equation (2), changes from initial limits [0.5, 1] to R, the set of real numbers from negative to positive infinity.

Model specifications

In the first part of the analysis, we want to see how some economy characteristics, as proposed by Morck et al. (2000), might influence synchronicity of the share prices.

We create a vector that contains structural variables and we include it in the following regression model:

)

where is the logistically transformed price synchronicity variable, is the market

capitalization, and is a random error term. Our purpose is to see which characteristics that

form explain stock price synchronicity and render the natural logarithm of market capitalization insignificant.

(16)

16

at the same time. Modelling with panel data technique brings us some advantages. Firstly, we can talk about the length of the data set, which is larger than the two simple data sets -cross-sectional and normal time series. This determine the independent variables to vary over two dimensions and in this way panel data renders a more accurate result. Secondly, panel data comes against multicollinearity problems among variables, because it focuses more on the development of the variables and on how the connection between them evolves over time. Lastly, the identification problems, as presented by Verbeek (2000) are reduced using panel data technique.

Our panel is balanced, having an equal amount of time-series observations for each cross-sectional unit.

In the second part, we introduce the Institutional Index. Firstly, we take into consideration each of the six dimensions to see the separate impact and, in the end, we construct a single index formed by the aggregate value of the previous ones. We denote this index and include it in the following regression:

)

where is the logistically transformed price synchronicity variable, is the market

capitalization, is a vector that measures structural characteristics and is a random error

term.

If the institutional index appears to be significant, and including it makes the market capitalization insignificant, we can say that we have evidence that a low development of macro institutions explains the high cross-country variation in the degree of stock price deviations from a random walk over time.

V.

Empirical Results

In this section, we document the effect of different corporate governance mechanisms on stock price synchronicity. More specifically, we look at how different proxies of corporate governance mechanism (the six dimensions of Institutional Index) relate to stock price synchronicity.

(17)

17

regression to test if there is any relation between the institutional governance indices and stock price synchronicity.

Anyway, the aggregate value – Institutional Quality Index, presents a higher variability, so we have to use fixed or random effects. In order to see which is worth being used, we are concerned to determine whether the random effect model passes the Hausman test in the case of the random effects being uncorrelated with the explanatory variables. The result obtained for the p-value is less than 1%, which implies that the random effects model is not appropriate and that the fixed effects model is preferred.

Furthermore, we conduct the likelihood ratio test to see if fixed effects are more appropriate indeed. The results show that in all three different redundant fixed effects tests which were employed, the p-values associated with the test statistics are zero, which means that fixed effects can be used. From the tests, we can also observe that only the cross-sectional fixed effects model parameters are qualitatively different from those of the initial pooled regression, which means that it is not the period fixed effects that make the difference (Brooks, 2008). The fixed effects in our model capture all variance specific to individual countries. Country fixed effects control all variables that are specific to a country and are not introduced in the regression - unmeasured variables that do not present any change in the study period but are different across countries.

Table 2 presents the results of the cross-sectional analysis, based on the pooled regression. For the first equation, which includes structural variables and also for the last one, which includes the aggregate value of the Institutional index, we use fixed-effect panel regression. The results show a statistically significant negative correlation between the dependent variable, stock market synchronicity, and the size of the market capitalization. Previous literature finds similar evidence, which means that a stock market with a high level of market capitalization presents a low level of stocks co-movement. Further we will notice if this relation of significance can be changed and if we can conclude that the market capitalization is responsible for the structural effects.

(18)

18

In equation 3.0 we have used only the structural variables embedded in vector . We observe that elements that form do not render market capitalization insignificant in this regression. This means that the size of the market is not representative for these structural variables, taken either separately or all together.

Table 2 Panel regressions

Panel A. Panel regression

Dependent variable is logistic transformation of stock price synchronicity measure . Equation 3.0 includes the logarithm of per capita GDP and structural variables. The following equations (from 3.1 to 3.6) include each component from the Institutional Quality index, separately; last, we include in equation 3.7 the aggregate value of the Institutional Quality Index. Data are for 2005 through 2010. Numbers in parentheses are coefficient standard errors.

Regression 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7

Logarithm of market capitalization -0.19(0.05) ***

-0.13*** (0.04) -0.15*** (0.04) -0.15*** 0.04 -0.14*** (0.04) -0.13*** (0.04) -0.15*** (0.04) -0.14*** (0.09)

Logarithm of per capita GDP 1.03*** (0.33) 0.18** (0.08) 0.19*** (0.07) (0.08) 0.10 0.18** (0.08) 0.17** (0.08) 0.10 (0.08) 1.14*** (0.08) Variance in per capita GDP growth 0.51 (0.07) 0.60 (0.07) 0.64 (0. 07) (0.04) 0.43 0.71 (0.05) 0.68 (0.04) 0.44 (0.06) 0.64 (0.04)

voice and accountability (0.08) -0.08 political stability and absence of

violence -0. 08*** (0.02) government (0.03) 0.02 regulatory quality -0.06 (0.04) rule of law -0.03 (0.03) control of corruption 0.01 (0.02)

Institutional Quality Index -1.30***

(0.44)

Country fixed effects Yes No No No No No No Yes

Regression R2 0.5670 0.5213 0.5279 0.5344 0.5244 0.5878 0.5059 0.6079

*** p= <0.01, ** p= <0.05, * p= <0.10

(19)

19

The general equation (3) is analysed for each value of the six components of Institutional Quality Index, and finally, takes the aggregate value of it.

Results in Panel A from Table 2 show that only one out of six dimensions of the Quality Index –the political stability and absence of violence-, and also the aggregate index, are negatively statistically significant. This means that the perception of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means has a very important impact on the stock price co-movement. The negative relation shows that a low score of this index, which means worrisome expectation of government destabilization, is significantly related to a high level of synchronicity.

Finally, the aggregate value of the Institutional Quality index is statistically significant and negatively correlated with the stock price synchronicity at 1% level, which means that a country where the macro institutions are reliable, presents a decline in the market stock price swings.

It is important to notice that even if it is clear that a high quality of the macro institutions is negatively correlated with the stock price synchronicity, this does not determine the market capitalization to become insignificant. Overall, these results suggest that neither structural nor institutional variables are sufficiently powerful to explain the relation between stock price synchronicity and market capitalization. So, after conducting the analysis, we noticed that, even if both sets of variables were significant, they did not underline our findings that stock prices in well developed stock markets are less synchronous than in low-developed markets. This means that factors beyond our explanations underline the negative relation between market capitalization and stock price synchronicity.

Anyway, we have to mention that we cannot categorically reject the structural explanation hypothesis using regressions like the one defined before. Additional variables can always be included and some combinations of specific variables may explain the stock price co-movement, rendering market capitalization insignificant.

(20)

20

To find out if our results are due to crisis associated with synchronicity changes, we repeat our regression and include a crisis dummy, one for each period of time spanned from 2008 to 2010 and zero otherwise.

The results in Panel A of Table 2 show that the institutional quality index and its components remain negatively correlated with the synchronicity of the stock markets, after controlling for structural variables, but this result does not determine the development of the markets to become insignificant, as we assumed.

Table 2 Panel regressions

Panel B. Panel regression with crisis dummy variables

Dependent variable is logistic transformation of stock price synchronicity measure . Each equation includes structural variables. The following equations (from 4.1 to 4.6) include each component from the Institutional Quality index, separately; last, we include in equation 4.7 the aggregate value of the Institutional Quality Index. The global financial crisis dummy is one for period 2007-2010, and zero otherwise. Data are for 2005 through 2010. Numbers in parentheses are coefficient standard errors.

Regression 4.1 4.2 4.3 4.4 4.5 4.6 4 .7

Logarithm of market capitalization -0.05 (0.04) -0.07 (0.04) -0.05 (0.04) -0.05 (0.04) -0.04 (0.04) -0.06 (0.04) 0.34 (0.37) Logarithm of per capita GDP 0.02 (0.09) 0.05 (0.07) -0.04 (0.09) 0.02 (0.08) 0.04 (0.07) -0.03 (0.08) -0.16 (0.66) Variance in per capita GDP growth -0.52 (0.07) -0.36 (0.08) 0.74 (0.08) -0.47 (0.07) -0.41 (0.03) -0.70 (0.07) 0.08 (0.07)

voice and accountability -0.04

(0.07)

political stability and absence of violence -0.05(0.02) **

Government -0.03 (0.03) regulatory quality -0.02 (0.04) rule of law (0.03) -0.04 control of corruption 0.01 (0.02)

Institutional Quality Index -1.24(0.41) ***

Crisis 0.12(0.03) *** 0.10*** (0.03) 0.12*** (0.02) 0.11*** (0.03) 0.12*** (0.02) 0.12*** (0.02) 0.13*** (0.03)

Country fixed effects No No No No No No Yes

Regression R2 0.5716 0.5299 0.5434 0.5330 0.5913 0.5117 0.6664 *** p= <0.01, ** p= <0.05, * p= <0.10

(21)

21

price synchronicity, at 1% level in all cases. In addition, the market capitalization became insignificant in all regressions.

Our strategy was to observe which development measures can render market capitalization insignificant in multivariate regression. The results show that structural variables and the components of the institutional quality measure do not have the power to explain the relation between stock price co-movemet and market capitalization. When we conduct our analysis in the context of the recent financial crisis, we observe that the crisis has an important impact on our results. The stock prices do move more during the global financial crisis period. This effect seems to explain the correlation between stock market co-movement and market development, as the variable does not remain highly significant.

In summary, our result is consistent with the previous findings, namely that a better quality of the institutions in a country is negatively correlated with the level of stock market synchronicity, but more important than that is the impact of the crisis, which increases in the same way with the synchronicity and which renders the development of the market insignificant.

VI. Conclusions

In our analysis, we followed three important steps: firstly, we calculated synchronicities of the Euro zone stock markets from 2005 to 2010, for all trading weeks, on an annual basis. This implies that we took into consideration specific types of countries – developed economies. Despite the fact that previous papers (Morck et al. (2000)) have found that poor economies present a more synchronous movement than developed ones, our results show an average fraction of 72.9% of stocks moving in the same direction, which is very high. Indeed, it seems that the change of sample and the different period of time had an impact on our results. While per capita GDP presents a weaker negative correlation, the market capitalization shows a stronger negative correlation, so we decided to use it as representative for market development, instead of capita GDP.

(22)

22

were not able to explain the mentioned connection. Afterwards, we introduced an Institutional Quality index, used as six separate variables and also as an aggregate vector, to establish if a good development of macro institutions was linked in any way with stock price synchronicity. We proved that a good perception of the political stability and the absence of violence are associated with lower co-movement of the cross country stock markets, but this does not affect significantly the relation between synchronicity and stock market development. Although some of these variables contribute to the synchronicity of the stock markets, a large residual effect remains.

(23)

23

References

Ashbaugh-Skaife, Hollis, Gassen, Joachim, and LaFond, Ryan, 2006, Does stock price synchronicity represent firm-specific information? The international evidence, Working Paper, University of Wisconsin.

Bissessur, Sanjay, and Allan, Hodgson, 2012, Stock Market Synchronicity- an Alternative Approach to Assessing the Information Impact of Australian IFRS, Accounting and Finance 52, 187-212.

Bris, Arturo, William, Goetzmann, and Ning, Zhu, 2007, Efficiency and the bear: short sales and markets around the world, Journal of Finance, 62(3), 1029-1079.

Brooks, Chris, 2008, Introductory Econometrics for Finance, Second edition, Cambridge University Press.

Brunnermeier, Markus K., 2009, Deciphering the Liquidity and Credit Crunch 2007-08,

Journal of Economic Perspectives 23(1), 77—100.

Chan, K. and Hameed, A., 2006, Stock Price Synchronicity and Analyst Coverage in Emerging Markets, Journal of Financial Economics 80, 115-147.

De Nicolo, Gianni, Laeven, Luc, and Ueda, Kenichi, 2006, Corporate Governance Quality: Trends and Real Effects, IMF Working Paper 06/293.

De Nicolo, Gianni, 2010, Financial Integration, Globalization, Growth and Systemic Real Risk, International Monetary Fund, Research Department and CESifo.

Dollar, David, Kraay, Aart, 2003, Institutions, trade and growth. Journal of Monetary

Economy 50(1), 133–162.

Du, Julan, He, Hao, Chenggang Xu, 2007, What determines the Synchronicity of the Markets in the Shanghai and Shenzhen Stock Exchange?, Occasional Paper 18.

Durnev, Art, Rendall, Morck, Bernard, Yeung and P. Zarowin, 2003, Does Greater Firm-Specific Return Variation mean more or less Informed Stock Pricing? Journal of Accounting

Research, 41, 797-836.

(24)

24

Easterly, William, and Levine Ross, 2003, Tropics, germs and crops: how endowments influence economic development, Journal of Monetary Economy 50(1), 3–39.

Faria, Andre, Mauro, Paolo, 2005, Institutions and the external capital structure of countries, IMF Working Paper 04/236, Washington.

Fernandes, Nuno, and Miguel, A. Ferreira, 2009, Insider trading laws and stock price informativeness, Review of Financial Studies, 22(5), 1845-1887.

Griffin, John M., Kelly, Patrick J., and Nardari, Federico, 2007a, Measuring short-term international stock market efficiency, SSRN Working Paper.

Gul, Ferdinand, Jeong-Bon, Kim, and Annie, Qiu, 2010, Ownership concentration, foreign shareholding, audit quality, and stock price synchronicity: Evidence from China, Journal of

Financial Economics 95, 425–442.

Hsin, Chin-Wen, and Yuehtzu Liao, 2003, Stock Price Synchronicities in Emerging Markets, Yuan Ze University, Taiwan.

Jin, Li, and Myers, Stewart, 2006, around the World: New Theory and Tests, Journal of

Financial Economics 79, 257-292.

Johnston, Joseph, 2009, Accruals Quality and Price Synchronicity, Faculty of the Louisiana State University.

Jung, Jeeman and Robert, Shiller, 2005, Samuelson’s Dictum and the Stock Market,

Economic Inquiry 43(2), 221-228.

Kaufmann, Daniel, Aart Kraay and Massimo Mastruzzi, 2009, Governance Matters VIII: Aggregate and Individual Governance Indicators for 1996-2008, World Bank Policy Research Working Paper No. 4978, Washington, D.C.

Khattak, Adnan, Stephen M., Courtenay, and Asheq, Rahman, 2010, Capital Market Developments and Stock Price Synchronicity, School of Accountancy, Massey University, New Zealand.

(25)

25

Khandaker, Sarod, and Richard, Heaney, 2009, Do Emerging Markets Have Higher Stock Synchronicity? The International Evidence, Journal of Business and Policy Research 1, 79-98.

Khandaker, Sarod, 2011, R Square Measure of Stock Price Synchronicity, International

Review of Business Research Papers 1, 165-175.

La Porta, Rafael, Lopez-de-Silanes, Florencio, Shleifer, Andrei, and Vishny, Robert, 1998, Law and Finance. Journal of Political Economy 106, 1113-1155.

Levine, Ross and Sara, Zervos. 1996. Stock Market Development and Long-Run Growth. The

World Bank Economic Review, 2(10).

Li, Kan, Randall, Morck, Fan, Yang, and Bernard Yeung, 2003, Time Varying Synchronicity in Individual Stock Returns: A Cross-Country Comparison, University of Alberta Business School, Canada.

Mierau, Jochen and Mark, Mink, 2009, Measuring Stock Market Contagion with an Application to the Sub-prime Crisis, DNB Working paper 217.

Morck, Randall, Yeung, Bernard, and Yu, Wayne, 2000, The Information content of stock market: why do emerging markets have synchronous price movements?, Journal of Financial

Economics 58, 215-260.

Piotroski, Joseph, and Darren Roulstone, 2004, The Influence of Analysts, Institutional Investors, and Insiders on the Incorporation of Market, Industry, and Firm-Specific Information into Stock Prices, The Accounting Review 79, 1119-1151.

Rodrik, Dani, Arvind, Subramanian and Francesco, Trebbi, 2004, Institutions rule: the primacy of institutions over geography and integration in economic development, Journal of

Economic Growth 9(2), 131–165.

Roll, Richard, 1988, , Journal of Finance, 43(2), 541-566.

Tobin, James, 1984, On the Efficiency of the Financial System. Lloyd’s Banking Review 153, 1-15.

Walti, Sebastien, 2011, Stock market synchronization and monetary integration, Journal of

(26)

26

Appendix

Table A1 Definition of Variables

Synchronicity refers to the tendency for the stocks in a share market to move in the same direction. The stock market co-movement can be upwards or downward depending upon the trend in the market

Per capita GDP is gross domestic product divided by midyear population. GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products

Market capitalization also known as market value, is the share price times the number of shares outstanding. Listed domestic companies are the domestically incorporated companies listed on the country's stock exchanges at the end of the year.

GDP per capita growth annual percentage growth rate of GDP per capita based on constant local currency. GDP per capita is gross domestic product divided by midyear population.

Inflation as measured by the consumer price index, reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. The Laspeyres formula is generally used.

Voice and accountability

captures perceptions of the extent to which a country's citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media.

Political stability and absence of violence

measures the perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means, including domestic violence and terrorism

Government effectiveness

captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies.

Regulatory quality captures perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote

private sector development.

Rule of law

captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence.

Control of corruption captures perceptions of the extent to which public power is exercised

(27)

27 Table A3

Descriptive Statistics – synchronicity measure on full sample

This table reports the descriptive statistics used to analyze the data series. The average value is described through either the mean value or the median value. The maximum and minimum represent the higher synchronicity value and the lower synchronicity value that the market index has taken during the full sample period (from 2005 to 2010). Skewness and Kurtosis are the coefficients which describe the distribution of the data series, while Jarque-Bera represents the test applied for normality.

Country Market

Index Mean Median Maximum Minimum

Std.

Dev. Skewness Kurtosis

Jarque-Bera Austria ATX 0.731 0.728 0.764 0.701 0.023 0.186 1.893 0.341 Belgium BEL20 0.729 0.731 0.743 0.710 0.013 -0.342 1.582 0.620 Cyprus CSE GENERA L 0.725 0.717 0.779 0.694 0.031 0.889 2.612 0.828 Estonia OMXT 0.721 0.718 0.763 0.693 0.023 0.774 2.892 0.603 Finland OMXH25 0.750 0.753 0.795 0.685 0.038 -0.669 2.618 0.484 France CAC40 0.756 0.759 0.792 0.704 0.034 -0.416 1.940 0.454 Germany DAX 0.741 0.742 0.776 0.702 0.026 -0.192 2.008 0.283 Greece ATS 0.706 0.708 0.743 0.651 0.036 -0.361 1.850 0.461 Ireland ISEQ20 0.747 0.746 0.795 0.690 0.039 -0.163 1.723 0.435

Italy FTSE MIB 0.696 0.700 0.716 0.676 0.014 -0.175 2.133 0.219

Luxembourg LUXX 0.695 0.693 0.722 0.674 0.020 0.166 1.368 0.693 Malta MSE 0.686 0.694 0.701 0.663 0.017 -0.613 1.522 0.922 Netherlands AEX25 0.746 0.741 0.790 0.708 0.035 0.161 1.331 0.722 Portugal PSI20 0.727 0.742 0.781 0.656 0.049 -0.481 1.721 0.640 Slovenia SBI20 0.729 0.714 0.786 0.705 0.032 1.058 2.586 1.163 Slovakia SAX 0.770 0.753 0.863 0.734 0.049 1.317 3.307 1.759 Spain IBEX35 0.745 0.746 0.776 0.712 0.027 -0.044 1.295 0.728 Average 0.729 Table A4

Descriptive Statistics: Synchronicity measure for sub-periods

This table reports the descriptive statistics used to analyze the synchronicity measure data series. The average value is described through either the mean value or the median value.

2005-2007 2008-2010

Country Market Index Mean Median Std. Dev. Mean Median Std. Dev.

Austria ATX 0.714 0.717 0.012 0.749 0.750 0.016 Belgium BEL20 0.718 0.717 0.008 0.740 0.741 0.003 Cyprus CSE GENERAL 0.709 0.714 0.014 0.741 0.740 0.037 Estonia OMXT 0.710 0.709 0.017 0.777 0.775 0.017 Finland OMXH25 0.723 0.738 0.033 0.733 0.719 0.026 France CAC40 0.729 0.734 0.024 0.782 0.787 0.013 Germany DAX 0.677 0.683 0.023 0.762 0.757 0.012 Greece ATS 0.720 0.727 0.016 0.705 0.700 0.009 Ireland ISEQ20 0.688 0.685 0.013 0.734 0.740 0.013

Italy FTSE MIB 0.714 0.724 0.020 0.697 0.704 0.019

(28)

28 Table A5

Simple correlation coefficients of structural and institutional variables with each other (including dependent variable)

Variables a b c d e f g h i j k l m n o Stock co-movement indices a. Logistic transformation of synchronicity ( ) 1 b. Logarithm of per capita GDP -0.052 1 c. Logarithm of market capitalization -0.327 0.590 1 Structural variables d. Logarithm of geographical size 0.317 0.159 0.243 1 e. Logarithm of population 0.389 0.135 0.033 0.738 1 f. Variance in GDP growth 0.203 -0.200 -0.412 -0.133 -0.202 1 g. Variance in domestic inflation rate 0.038 -0.216 -0.427 -0.158 -0.191 0.691 1 Institutional variables h. Voice and accountability -0.153 0.724 0.551 0.050 0.037 -0.123 -0.122 1

i. Political stability and

Referenties

GERELATEERDE DOCUMENTEN

The emergence in the seventeenth century of a familiar yet easy and graceful prose among well-educated English writers is ascribed to a num- ber of causes: the need for

3 september 2018 Ethics and the value(s) of Artificial Intelligence Martijn van Otterlo.. happen if AI achieves general human-level

Yet why should an oil-rich country like the UAE pursue a civilian nuclear power program, especially at a time when the future of nuclear-power around the world is uncertain..

The goal of this study is to find a gap between theory and practice regarding Dutch SME valuation methods and a trend in the difference between the initially calculated value and

Zuwe Zorg Regio de Ronde Venen e.o. Zuwe Zorg Preventie &amp; Welzijn Zuwe Zorg Preventie &amp; Welzijn

Hij moet (1) een gebrek aan voldoende middelen hebben voor een woning, (2) niet ingeschreven zijn in de bevolkingsregisters en (3) een vraag stellen voor ondersteuning aan

Ook projecten die 'niet in betekenende mate' (nibm) van invloed zijn op de luchtkwaliteit hoeven niet meer te worden getoetst aan de grenswaarden voor luchtkwaliteit..

Als je bij het herschrijven van een tekst 15% minder woorden met drie of meer lettergrepen gebruikt, wordt S kleiner?. Maar die verkleining van