Effect of Globalization on International Diversification

Transcription

1 NAA301 Bachelor thesis in Economics, 15 hp Tutor: Johan Lindén Mälardalens Högskola Effect of Globalization on International Diversification Tom Schöblom : Chukwuemeka Okoye: Yan Zhu:830212

2 Abstract Date: Level: Authors: Tudor: Bachelor thesis in economics, 15hp Tom Schöblom, Chukwuemeka Okoye, and Yan Zhu. Johan Lindén Aim: As Swedish investors seek global diversification, it is important to indentify the correlation between international stock market indices and the Swedish market. In this study, the authors will be looking at the effect of globalization and international diversification. It looks at how globalization, if any, has reduced or even removed the diversification effects of investing abroad. The investigation will be conducted with the OMXS 30 as portfolio and using the theorem of correlation. Methodology: A total of 25 stock indexes were taken from different economics located from different continents. The authors compare the OMXS30 Index with the other 24 stock market indexes. All index prices are collected as secondary data from Yahoo Finance, and were taken from 1 st of January, 1998 to December 31s t, 2009 on a weekly basis. Conclusion: The result of the investigation clearly shows that the correlation between the Swedish stock market and other countries markets in the world has increased a lot over the last twelve years. The increases have developed differently throughout the continents, but there has been an increase in all investigated markets. The average increase in correlation is The countries that are most correlated with Sweden are Australia, Austria, and the Netherlands, whilst the least correlated markets are Mexico and Dow Jones Index in the United States. No market has lower correlation with Sweden than Key words: Globalized stock market, OMXS30, Portfolio, Diversification, Correlation.

3 List of figures: Figure 1: Total correlation Figure 2: Correlation between South America Figure 3: Correlation between Asia/Pacific Figure 4: Correlation between Europe Figure 5: Correlation between U.S Figure 6: Correlation Sub-periods South America Figure 7: Difference between first and last period South America Figure 8: Correlation Sub-periods Asia/Pacific Figure 9: Difference between first and last period Asia/Pacific Figure 10: Correlation Sub-periods Europe Figure 11: Difference between first and last period Europe Figure 12: Correlation Sub-periods U.S Figure 13: Difference between first and last period U.S Figure 14: Correlation in the first and the last period... 25

4 Table of Content 1. Introduction The problem A short review of the literature The aim of the thesis The limitations Diversification only by investing abroad: No effect of exchange rate risk: The methodology The analysis Total Year to year basis Correlation between the Swedish market and markets in South America Correlation between the Swedish market and markets in Asia/Pacific Correlation between the Swedish market and markets in Europe Correlation between the Swedish market and markets in United States Sub-periods Correlation between Sweden and South American countries Correlation between Sweden and Asian/Pacific countries Correlation between Sweden and European countries Correlation between Sweden and markets in the United States The results Summary and Conclusions References: Appendix... 30

5 1. Introduction 1.1. The problem The recent economic crises had made people and investors wonder how so interconnected the world market is. The economic meltdown involved most developed and underdeveloped markets across the globe and these markets were in a free fall plunging their economics into recession. In recent years, stock investments abroad have been very popular but with the recent occurrence of Information technology which had been improved upon by rapid internet development 1, investment aboard has been made convenient for investors hence investments can now be made over the internet and across continents and countries. In investments there are two major types of risk, the systematic risk and the unsystematic risk. The unsystematic risk are internal risk, risk within the stocks. These risks can be diversified away by different means but mostly by increasing the number of stocks on a portfolio. Systematic risks are risk that arises due to movement in the market (economics). These risks cannot be diversified and investors a rewarded for bearing that risk. Most investors are concerned with return on their investment and only few look at the risk involve in that investment. These few investors tend to remove that risk (unsystematic risk) by diversification. Some investors tend to diversify their risk by investing in other world markets as well as their home country while some just invest in different stocks thereby forming a portfolio of stocks. In this study, the authors will be looking at the effect of globalization and international diversification. The authors will be using the various market indexes as a proxy to individual portfolio Each stock market can be considered as bundle of portfolios of stocks. 1 Information Transparency of Business-to-Business Electronic Markets: A Game-Theoretic Analysis,Kevin Zhu, Management Science, Vol. 50, No. 5 (May, 2004), pp ,Published by: INFORMS, Stable URL: 1

6 It should be noted that the recent financial crisis inspired the authors in doing this research. Research questions: Has globalization made the world market a single factor? Has globalization reduced the advantages of international diversification? Which countries is the best alternative for Swedish investors in terms of international diversification? How do the correlation between Swedish market represented here by (OMXS 30) and other chosen stock market in the world look like in recent years? Key Words: Globalized stock market, OMXS30, Portfolio, Diversification, Correlation. 2

7 1.2. A short review of the literature The authors have found some interesting previous study, which includes Stockholm stock exchange efficiency 2 - The deviation of OMXs 30 was statistically verified, and the Stockholm stock market is not efficient on a semi-strong level after the positive report issues, Does increased international influence cause higher stock market volatility 3 - the main finding in that paper is: the world markets influence the volatility of the Swedish stock market rather than just purely influenced by the flows of domestic news, Stock market volatility : evidence from Sweden 4 - This paper has found the Sweden stock market has becoming more volatile from 1979 to Stock market volatility in OECD countries: recent trend. Consequences for the real economy. And proposals for reform 5 - The stock exchange in the OECD countries is more interrelated in the last thirty years. And correlation index of return among those countries are on average higher and the volatility among those stock markets are more correlated, and Volatility driven changes in stock return correlation dynamics 6 - The paper found that high volatility could rise the correlation among the stock markets and the Nordic markets one-selves can influence the correlation with other markets through the local volatility, in other words, the Nordic markets are correlated with other world stock market but still have own difference. Those previous studies inspire the authors and at same time the authors realize that the research between OMXS30 and the mixed gathering of other main stock exchange markets of Europe, Asian-pacific area, North America, U.S. is still rare in English for those mature stock market and emerging stock market in updated recent years. So in this paper, the authors explore the influence of the chosen world stock market on the main Swedish index-omxs30 in Sweden stock market with other chosen stock markets which are either 2 stock exchange efficiency Gardhage Erik, Gyllefjord Fredik, Lolic Vladmir, 3 rd, June, John Hassler, Scandinavian Journal of Economics, 101:(1), TanjaHellberg & Dr Mohammand I qbal Tahir, draft in January Paul Kupiec. 6 cles/ html, Johan Knif, HANKEN and Seppo Pynnönen,2007 3

8 mature market or emerging market in the investment market. OMXS 30 which includes 30 most actively traded stocks on the Stockholm stock exchange market is abbreviation of OMXS Stockholm 30. The OMXS30 is a market weighted price index and the composition of the index is revised two times each year. And the 30 stocks in the index strengthen the liquidity of corresponding shares of stock. 4

9 1.3. The aim of the thesis The aim of this paper is to investigate the effect of globalization on international diversification. It looks at how globalization, if any, has reduced or even removed the diversification effects of investing abroad. The investigation will be conducted with the OMXS 30 as portfolio. As major Swedish investor seek global diversification, it is important to indentify the correlation between international stock market indices and the Swedish market. 5

10 1.4. The limitations Diversification only by investing abroad: As stated above, our main objective is to find out if there is the effect of globalization on international diversification. Investors are looking at ways they can diversify their risk so as to make their investments riskless. No investor want to bear unsystematic risk but only systematic risk and form the theory of CAPM; investors should be rewarded for bearing systematic risk. This paper looks only at the effect of globalization on international diversification of unsystematic risk assuming that the systematic risk as measured by beta is constant. In deciding where to invest, the research will not be looking at the systematic risk (Beta) but only on how to obtain the best way to diversify the unsystematic risk so that the portfolio will be risk free. It should be noted that in any investment decision, Beta plays a very important role as it indicates the degree of responsiveness between stocks/portfolios to movements in market index. Investors would like to invest in stocks or portfolios that have high beta because according to CAPM, a higher beta yields a higher return (this theory had been proven not to be absolutely correct). Diversification of risks can be achieved by different means; one of the most common ways to diversify risks is to invest in many stocks that are less correlated to each other. The higher the number of less correlated stocks the less the risk of the portfolio. This research acknowledges this method of diversification but is merely sticking to diversification by investing in other countries. Thus, for the purpose of this study, the research is only limited to diversification of unsystematic risk by investing abroad as a way of deciding where to invest their money No effect of exchange rate risk: The expected return and risk (unsystematic risk) were computed from the Swedish investor s point of view. Since this research is based on Swedish investors and their investments abroad, all currencies should be in the Swedish kroner. The authors acknowledge the fact that there might be slight difference in the returns because for example, kroner have not fluctuated perfectly with the dollar if the US market is considered. 6

11 Thus the country of domicile affects the expected returns and risk (including the correlation coefficients) more so; investors can protect themselves partially against exchange rate risk by hedging 7, but this is beyond the scope of this research. Having that in mind, this research did not convert or put in place the effect of exchange rate risk. This is so because the data was taken from a long period of time and these times consist of different years. This means that converting all currencies would mean using the prevailing exchange rate at a given date for a given return. This would have made the research more cumbersome. Instead, the authors assumed there are no effects of exchange rate risk and returns were converted into percentages, that way no matter how the exchange rate is; the returns would show how much an investor would get in terms of percentage. 7 Modern portfolio theory and investment analysis(seventh edition) by Elton and Gruber 7

12 2. The methodology A total of 25 stock indexes were taken from different economics located from different continents. The authors compare the OMXS30 Index with the other 24 stock market index: MERV Index, AORD Index, ATX Index, BFX Index, BVSP Index, HIS Index, FCHI Index, GDAXI Index, BSESN Index, JKSE Index, N225 Index, KSII Index, KLSE Index, MXX Index, AEX Index, STI Index, OMXSPI Index, SSMI Index, TWII Index, FTSE Index, DJI Index, IXIC Index, NYA Index, GSPC Index. All indexes prices are collected as secondary data from Yahoo Finance, and were taken from 1 st of January, 1998 to December 31s t, 2009 on a weekly basis. The OMXS 30 which includes 30 most actively traded stocks on the Stockholm stock exchange market is abbreviation of OMXS Stockholm 30. It s a market weighted price index. The composition of the index is revised two times each year. The 30 stocks in the index strengthen the liquidity of corresponding shares of stock. It should be noted that all prices are adjusted for dividend. As stated earlier, the aim of this paper is to look at the effect of globalization on international diversification. Since no investors want to bear unsystematic risk, the theory of diversification has become an important subject in line with making sure that unsystematic risk can be diversified away. Part of the way to look at this issue is to examine the globalization effect on correlation. Over the years, the world and everything that happens in it has become a village, this include economic activities. The world of finance is no exception to this as investments and trade can be done with the snap of the fingers. This had been made possible with new developments in information and communication technology. Correlation is going to be a key tool in our investigation and the reason for this is because correlation looks or examines the relationship between two securities and their movements. It shows how securities tend to move in relation to other security. The idea in using this theory in this research is that it makes it easy to understand the movements associated between these markets. If these markets are highly correlated, then diversifying by investing abroad might yield little or no result because if the market in a country abroad goes down, then the markets in Sweden will also go down. The same is true if he markets goes up. 8

13 The authors have chosen the quantitative methods in this research. All analysis and the final conclusions are based on the secondary data which was collected from Yahoo Finance between July 1997 and April 2010 Since we are considering the Swedish market and the rest of the world, the authors decided to use the major indices as a proxy to individual portfolios or stock. The market index indicates the overall performance of the securities involve in the market. So it s a good proxy to individual securities or portfolios. The performances of individual portfolios are related to the market index according to CAPM: if the market does well then a well constructed portfolio will also do well. This theory explains why the authors used the OMXS 30 as a proxy to individual portfolio. In calculating the correlation coefficients, the returns of the various markets were first calculated. What is a return? It is the relationship between money gains or lost to the amount of money invested in an investment. It can be seen as a ratio between money gain and the initial investment. The authors calculated the returns on each of the indexes including the Stockholm OMX. The formula used is giving by, R = P 1 P 0 P 0, where R is the rate of return, P 1 is equal to price at period 1 and P 0 price at period 0. Where period 1 and 0 indicates the prices at the preceded and previous week respectively, the prices were all closing prices. As explained earlier, correlation says a lot about the co-movement between securities and in calculating the correlation, the returns must be known. Having gotten the returns on each index, the authors went further to calculate the correlation between the Swedish market index (OMXS 30) and the various markets around the world. The formulation for the correlation is given by the formula: CORREL(x, y) = n i=1(x i x )(y i y) n i=1 x i x 2 n i=1 y i y 2 9

14 The formula above shows the correlation between x and y. In this thesis, the authors used excel to calculate the various correlations for different indexes in relation with OMXS 30. The formula in excel is giving by = correl(array1, array 2) where array 1 is the independent variable, X and array 2 is dependent variable, Y. this formula is to compare how the Swedish market index tends to move in response to the various market index or vice versa. The results were then graphed to see the relationship at a glance and a trend line was constructed to analyze the slop of the various points. All calculations were made in excel and shown in the appendixes located at the back page of this paper. 10

15 3. The analysis The aim of this thesis is to see if there is a globalization effect on diversification. Diversification is a way of reducing risk in any investment. Most investors tend diversify by investing in other countries other than theirs. The authors are looking at the effect of globalization on such investment. The idea is that when an investment is made outside the home country of an investor in this case Sweden, a market meltdown in Sweden will not affect his investment abroad or vice versa. Does it mean that if a Swedish investor invests in other countries it makes their investment safe if the Swedish market crashes? The one way to answer this question is to look at the correlation and covariance effect between Sweden and most of the world market. Correlation is simply defined as the relationship between two variables during a given time period. It shows the various movements between given data. A positive correlation means that the two variables are positively related. Simply put, a movement in one variable means the other variable will also move in the same direction as the first. A correlation of 1 means that both variables will move in the same direction and a correlation of -1 mean movement in opposite direction. Covariance looks at the way both variables co-vary with each other. In our thesis, we are considering the market indexes of individual countries in relation to the Swedish market index. We are taking the index because it tells us the general performance of the stock market of a giving county and so it severs as a good proxy to individual portfolio. Let look at the result from our data. Our data was divided into 1 total and 4sub groups which represents the various continents. 1 Americas which included countries from Argentina, Mexico and Brazil 2. Asia/ pacific which included countries from Australia, Hong Kong, India, Indonesia, Japan, Korea, Malaysia, Singapore and Taiwan. 3. Europe which includes countries from Austria, Belgium, France, Germany, Netherland, Switzerland and the United Kingdom 4. The United State of America 11

16 Correlation 3.1 Total First we look at the graph that show the correlation between the Swedish market index (OMXS 30) and all the other indexes as a total between 1998 and Total Correlation Total Correlation Indices Figure 1 The chart above shows the correlation between OMXS30 and the other indexes reviewed, it goes from lowest correlation to highest correlation. During the time period, it can be seen that all correlations between Sweden and the other countries are positive. This implies that all investigated countries markets moves in the same direction as the Swedish. In the case of diversification this is not a good result. The goal is to find a market that moves in the opposite direction as the Swedish market, i.e. have a negative correlation. It also displays that Indonesia and Malaysia have the lowest correlation with Sweden whilst Germany and France have the highest correlation with Sweden. 12

17 Correlation 3.2 Year to year basis What is more interesting is to see how the correlation has changed during the last 12 years which will be displayed below, in year to year graphs. The countries are divided into 4 groups according to location; South America, Asia/Pacific, Europe, and U.S Correlation between the Swedish market and markets in South America 0.80 Correlation between South America Argentina 0.70 Brazil 0.60 Mexico Year Figure 2 The figure above shows the correlation between the Swedish market index and the indexes from Argentina, Brazil and Mexico. From the data above, it can be seen that they are all positively 13

18 Correlation correlated to Swedish market. It means they have a co movement between the markets of Sweden and the markets in the Americas. Mexico has the highest correlation which occurred in 2001 and we noticed that during the later part of the years (2006 to 2009), the individual countries were more correlated to Sweden than ever before. As time progressed, globalization has taken a toll on the financial markets and has made these countries more and more correlated to the Swedish market Correlation between the Swedish market and markets in Asia/Pacific 0.90 Correlation between Asia/Pacific Australia Hong Kong India Indonesia Japan Korea Malaysia Singapore Taiwan Year Figure 3 14

19 Correlation The figure above shows the correlation between the Swedish market and the markets from Asia and the pacific. The countries are displayed on the right hand Conner of the graph. It also shows a positive correlation except for India in It had a negative correlation with the Swedish market. Consider the early stages of the time period under consideration, it would be seen that the correlation was not as high as it is now. When we consider the relationship between the individual markets in the region, we could also see that the effect of globalization has made the whole financial market be combined as one. This is evident from the later stages of the time period under consideration. A trend line was also constructed to see how the slop looks like and from the figure above we could see that it has a positive slop and its increasing. What does this mean, it means that the correlation between the countries in the Asia continent and the pacific is increasing and may continue over the years to come Correlation between the Swedish market and markets in Europe 0.80 Correlation between Europe Year Austria Belgium France Germany Netherlands Switzerland U.K Figure 4 The figure above shows the correlation between the Swedish market and the markets from Europe. The countries taken into consideration are listed on the right hand side of the graph. 15

20 Correlation This graph is not too different from the graphs we have in figure 1 and 2. Notice that the correlation between countries in Europe increased over the years. In 1998, countries in the Euro zone were not that correlated as it is in The relationship between these countries in terms of the correlation between them increased a great deal and most countries in 2009 crossed the 0.6 threshold unlike what it was in Put together, all of the countries are positively correlated with the Swedish market. The trend lines also show a positive slope. From 2003 and onward the trend seems to be that the correlation is becoming more stable than before. In 2000 the Swedish market were negatively correlated with both Belgium and Switzerland, and also had very low correlation with Austria and U.K Correlation between the Swedish market and markets in United States Correlation between U.S. Dow Jones Nasdaq New York Stock Exchange S&P Year Figure 5 The figure above shows the correlation between the Swedish market and the markets from the largest economy in the world, the United State of America. The authors looked at the 4 major 16

21 Correlation market indexes, namely the Dow Jones industrial average, NASDAQ, NYCE and the S&P 500. A look at the graph shows some interesting result. The correlation during 2001 and 2003 were the highest and from there it became lower until the year 2009 but is still positive. We also notice that the trend line for the nasdaq has a negative slope. The correlation between the Swedish market and the nasdaq is decreasing. Other trend lines have a positive slope. What is noticeable is that the correlation seems to have become more stable the last four years, with a small increase in correlation as a trend. From 98 to 05 there correlation were very volatile with large differences from year to year. It is very hard to see a trend or even some sort of stability in the correlation in these years. 3.3 Sub-periods Here we have divided the time into four sub-periods in order to get a clear overview of how the correlation has changed during the last 12 years Correlation between Sweden and South American countries 0.70 Correlation Sub-periods - South America Argentina Brazil Mexico Countries Figure 6 17

22 Correlation The South American countries show a very distinguished difference in their development through the four periods. Argentina has very low correlation in the first three periods, the lowest in the region, around 0.20, but increases drastically in the fourth period by 0.4, to the highest in the region (0.59). Brazil has a similar development ; starts at 0.26 in the first period, moves up to 0.31, then goes down by 0.03, but then takes a big leap up to 0.52 in the last period. Mexico has a different development than the other two South American countries. Mexico has a steady increase in the correlation throughout all four periods. Interestingly it has the highest correlation in the first period (0.27), increasing to 0.34 in the second, has a slight increase in the third, 0.02, and ends up with the lowest correlation (0.45) in the last period in the region. Difference between first and last period - South America Argentina Brazil Mexico Countries Figure 7 The figure above shows that the country with the lowest correlation in period one has most increase in correlation, and that the country with the second lowest correlation in period one had the second most increase in correlation. Interesting is that Both Argentina and Brazil have gone from very low correlations in the first three periods to increase tremendously in the last period and end up with higher correlation than Mexico who has had a steady but slow increase over the periods. 18

23 Correlation Correlation between Sweden and Asian/Pacific countries Correlation Sub-periods - Asia/Pacific Countries Figure 8 As shown above the correlation between the Swedish market and the markets in Asia/Pacific shows an apparent trend of increasing correlation. Australia has a total correlation of 0.5 but has 0.69 in the last period; it has the highest correlation of the countries from this region and has increased with 0.29 from the first to the last period. Hong Kong s correlation has gone from 0.29 in the first period to 0.56 in the last. The correlation increased from period one to period two and from period three to period four but holds steady at 0.43 in the two middle periods. India has a low 0.18 correlation in the first period, increases to 0.27 in the second and further increases to 0.5 in the third, which stays steady through the fourth period. Indonesia has very low correlation in the first two periods, 0.13 and 0.15, but increases very much in the third to 0.5, and dips, as only country in this region, with 0.02 in the fourth period. Japan has a very steady increase in correlation throughout all four periods; going from 0.34 in the first to 0.56 in the last period, with the biggest difference in increase being between the third and the fourth period. Korea has the most volatile change through the periods going up from 0.31 to 0.47 between the first and the second, but then going down to 0.4 in the third, just to reach a high of 0.53 in the last period. Malaysia has, like Japan, a steady increase in all four periods, with the difference that Malaysia s increases are larger each period. The correlation goes from 19

24 Correlation 0.13 in the first, to 0.23 in the second, 0.30 in the third, and then a big increase to 0.56 in the fourth. Singapore has a similar development as India it has increases between the first, 0.27, and the second period, 0.41, as well as between the second and third, 0.56, but flattens out after the third with an increase of 0.01 in the fourth period. Taiwan s increase is unique for the region with a big leap between the first, 0.20, and second period, Then it has a steady small increase to 0.46 in the third and 0.48 in the fourth period Difference between first and last period - Asia/Pacific Australia Hong kong India Indonesia Japan Korea Malaysia Singapore Taiwan Countries Figure 9 One noticeable thing in the figure above is that it is the three countries that have the least correlation in the first period, India (0.18), Indonesia (0.13), and Malaysia (0.13), that have the biggest difference between the last and first period, suggesting that they have caught up with the rest of the region during this period. 20

25 Correlation Correlation between Sweden and European countries Correlation Sub-periods - Europe Austria Belgium France Germany Netherlands Switzerland U.K. Countries Figure 10 If we look at the total correlation (figure 1) the European countries are among the ones with the highest correlation, but the development in correlation among this countries has not been equal. Austria is one of three countries in this region that has a lower correlation between two periods. It starts with 0.33 but goes down in the second period to the lowest recorded correlation in a European country, In the third period Austria and Sweden s correlation goes up by 0.40 to 0.59, and continue to climb in the fourth period to end up at the highest correlation of the European countries studied, Belgium has the lowest correlation in period one, 0.23, increases to 0.35 in the second period and then as Austria increases a lot in the third to 0.61, then a slight movement up in the fourth period, France shows the most stable correlation throughout the four periods with small increases in all periods. Beginning at 0.46 in period one to 0.50 in the second, 0.59 in the third, and ending up at 0.62 in the fourth period. Germany has the smallest movement in correlation of all the countries examined. The biggest difference between two successive periods is 0.06 and the total difference between the first and the last period is Germany has the highest correlation of all European countries in the first period but in the fourth period it ends up as having the fifth highest correlation out 21

26 Correlation of seven countries in the region. Germany is also one of the three countries that display a lower correlation from one period to another. The Netherlands has gone from the second lowest correlation (0.28), in the region, in period one to the second highest correlation in Europe in period four (0.66). The increase is continues through the four periods, but largest from the first to the third, flattening out in the fourth period. Switzerland is similar to Netherlands in its correlation development, going from 0.33 in the first period, to 0.43 in the second, and taking a big leap up to 0.67 in the third period, but instead of stagnate like the Netherlands correlation did, Switzerland and Sweden s correlation goes down in the fourth period by 0.1, which is unique in this investigation. United Kingdom on the other hand follows the pattern of the Netherland all the way, showing a steady climb through period one to three which flattens out in the last period. The difference is that U.K has a higher correlation in period one (0.35), but ends up in the fourth period with a lower correlation (0.58) than the Netherlands Difference between first and last period - Europe Austria Belgium France Germany Netherlands Switzerland U.K. Countries Figure 11 In period one the difference between the country with the highest correlation with Sweden and the country with the lowest correlation with Sweden in Europe was 0.32, in period four the difference is This follows the trend seen in Asia/Pacific that the countries with the lowest correlation 12 years ago have had the most increase and thereby flattened out the difference within the region. 22

27 Correlation Correlation between Sweden and markets in the United States 0.60 Correlation Sub-periods - U.S Dow Jones Nasdaq New York SE S&P 500 Indices Figure 12 The markets in the United States have as much difference within the country as the other regions has between the countries. The markets also have a different development than most of the other regions. Dow Jones and the Standard and Poor 500 have a similar development. They have their lowest correlation with Sweden in the first period, then goes up to their highest correlation in the second period, after that both decreases in the third period, just to almost climb up to the same correlation in the last period. The difference is that Standard and Poor lies on a slighter higher correlation than Dow Jones through all four periods. New York Stock Exchange don t follow the same trend, it do not have the same dip between the second (0.45) and third period (0.45) and reaches its highest correlation in the last period. New York SE is the only market in the U.S. that does not have its highest correlation in the second period. Nasdaq is the market with the lowest difference between the first period (0.46) and the last period (0.50), it also has the highest correlation of all periods in the region (0.57) in the second period. Nasdaq is the only market observed in this investigation that shows no trend. 23

28 Correlation 0.25 Difference between first and last period - U.S Dow Jones Nasdaq New York SE S&P 500 Indices Figure 13 The markets in the U.S. have not made large increases in correlation with Sweden over the period, but three of the four have increased with more than 0.1, and the pattern that the countries correlation evens out throughout the years sustain, in period one the difference between the highest and lowest correlation was 0.18 and in the fourth period it is

29 Argentina Brazil Mexico Australia Hong kong India Indonesia Japan Korea Malaysia Singapore Taiwan Austria Belgium France Germany Netherlands Switzerland U.K. Dow Jones Nasdaq New York SE S&P 500 Correlation 4. The results Correlation in the first and the last period Countries Figure 14 The result of the investigation clearly shows that the correlation between the Swedish stock market and other countries markets in the world has increased a lot over the last twelve years. The increases have developed differently throughout the continents, but there has been an increase in all investigated markets. The average increase in correlation is The countries that are most correlated with Sweden are Australia, Austria, and the Netherlands, whilst the least correlated markets are Mexico and Dow Jones in the United States. No market has lower correlation with Sweden than When investing abroad the purpose to lower risk against return there are four things to take into account; 1) The correlation with the abroad market 2) The Return on the foreign market 3) The Risk of the foreign market 4) The Exchange rate risk 25

30 This investigation shows that the advantage of investing abroad in the purpose to lower risk has severely decreased during the last twelve years. 26

31 5. Summary and Conclusions. The authors found out some interesting results and have drawn out their conclusions from them. The effect of globalization has nearly made the world market a single factor from the view of a Swedish investor. Looking at the result from our data analysis, we would discover that correlation between the Swedish market represented here by the OMXS30 and other world markets under study had increased. This means that the markets in these countries tend to move together depending on the directions. The advantages of international diversification that was before being enjoyed by Swedish investors, has now be reduced due to the effect of globalization. As stated earlier, investors tend to diversify their risk by investing in stocks or portfolio that has a low correlation with each other. Due to the fact that correlation has increased over the years, these advantages are beginning to wear away because world markets are more interlinked now than ever before. Having said that, there are some countries that have some interesting results. We could see that the countries of Mexico and the United State (not all markets in the States) are less correlated with the Swedish market. This means the advantages of international diversification in these countries are still up and running and investors can take the opportunity to diversify in these countries all other factors also being considered. In general, the correlation between the Swedish market and other countries under study has increased over the years; this is due to the effect of globalization. The economics of the world are becoming interlinked, what affect one economy affects the other and what affects one financial market also affects the other. Globalization has brought together the world s financial markets and it is gradually eroding the advantages of international diversification. 27

32 6. References: Books and Online Resource: Chang, L., Regional Integration of Stock Markets in Asia Pacific Countries[Online] Available at: [Accessed 6 April 2010] Elton, E., Gruber, M., Brown, S., Goetzmann,W.,2007.Modern Portfolio Theory and Investment Analysis.7 th ed. New York: John Wiley& Sons Erik,G., Fredik,G., Vladmir,L., Stockholm stock exchange efficiency [Online] Available at: [Accessed 7 April 2010] Hanken, J.,Seppo P., 2007.Volatility driven changes in stock return correlation dynamics [Online] Available at: lltextarticle/articles/ html [Accessed 7 April 2010] John, H., Does Increased International Influence Cause Higher Stock Market Volatility? 101:(1),1-9, Scandinavian Journal of Economics [Online] Available at: [Accessed 11 April 2010] 28

33 Kevin Zh., 2004.Information Transparency of Business-to-Business Electronic Markets: A Game-Theoretic Analysis. Management Science, Vol. 50, No. 5 pp , INFORMS [Online] Available at: [Accessed 8 April 2010]. Ming, L., Andrew, C., Index Futures Trading and Spot Market Volatility: Evidence from the Swedish Market [Online] Available at: [Accessed 8 April 2010] OMX Stockholm 30 Index [Online] Available at: [Accessed 19 March 2010] Tanja, H., Iqbal Tahir, M., dra Stock Market Volatility Evidence from Sweden[Online] Available at: EVIDENCE-FROM-SWEDEN/ [Accessed 7 April 2010] Woo-Sung, J.,Okyu, K. Jae-Suk, Y., Hie-Tae, M.,2006. Effects of Globalization in the Korean Financial Market.Journal of the Korean Physical Society, Vol. 48, pp. S135-S138 29

34 Appendix The list of abbreviation: MERV - Argentina Merval Index AORD - Australia All Ordinary Index ATX - Austrian Traded Index BFX - Belgium 20 Index BVSP Brazilian Bovespa Stock Index HIS - The Hang Seng Index of Hongkong FCHI - Summary for CAC 40 Index of France GDAXI - German DAX Index BSESN Summary for BSE SENSEX Index of India JKSE - Jakarta Composite Index of Indonesia N225 - Nikkei 225 of Japan KSII - KOSPI Composite Index of Korea KLSE - Kuala Lumpur Stock Exchange Index of Malaysia MXX - Mexican Stock Exchange BMV: BOLSA AEX Netherland AEX Index STI Singapore Straits Times Index OMXS30 - OMX Stockholm 30 of Sweden SSMI - Swiss Market Index TWII - TSEC Taiwan Weighted Index FTSE - The FTSE UK Index Series of UK DJI - Dow Jones Industrial Average of U.S IXIC - NASDAQ Composite of U.S NYA - NYSE COMPOSITE INDEX of U.S GSPC S&P 500 INDEX of U.S 30