The turn-of-the-month effect and liquidity : an empirical study on several European

The turn-of-the-month effect and liquidity:

An empirical study on several European

stock markets.

Abstract:

This thesis investigates whether evidence can be found for the turn-of-the-month effect and if higher liquidity is the cause of the effect. The effect will be examined on the following European indices; AEX (Netherlands), CAC (Franc), DAX (Germany), FTSE Athex (Greece), FTSE MIB (Italy), BET (Romania), WIG (Poland) and IBEX (Spain). A dummy variable is added for the 2008-2009 financial crisis. For this thesis a period from January 2004 to December 2014 is used. Results show the turn-of-the-month effect can be found for several markets but is inconsistent in its appearance. The effect is affected by the financial crisis. This thesis has not found any proof for liquidity being the cause of the turn-of-the-month effect.

Chris Jan Weijzen 10075151

February 2, 2015 Bachelor's thesis

Program: Economics and Business Administration Specialization: Finance and Organization

Thesis specialization: Finance Supervisor: Mark Dijkstra

2 This document is written by Student Chris Jan Weijzen 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.

3

Inhoudsopgave

3. Methodology and Data description ... 9

3.1. Methodology ... 9

3.2. Data description ... 10

4. Results ... 13

4.1. Results for the turn-of-the-month effect and liquidity ... 13

4.2. Results during the crisis ... 14

5. Discussion ... 16

6. Conclusion ... 16

4

1. Introduction

Ariel (1987) was the first that documented empirical regularity at the turn of the month, over the period 1963-1981. The calendar anomaly that was found explains the increase of stock returns on the turn of the month compared to the remaining days of the month and is called the turn-of-the-month effect. Ariel (1987), Lakonishok & Smidt (1988) and Zumpano (2009) find evidence of the turn-of-the-month effect for the US markets. Cadsby & Ratner (1992) and Agrawal & Tandon (1994) find international prove for the effect.

Ogden (1990) finds empirical evidence for the cause of the turn-of-the-month effect. He uses his liquidity hypothesis to prove the common payoff time for dividends, salaries and interest payments in the US at the turn-of-the-month result in higher liquidity. Investors reinvest the liquid profits causing an increase in stock market returns at the turn of the month. Liquidity growth is influenced by monetary policy; therefore the impact of monetary policy is also expected to affect liquid profits. When liquid profits are higher during the turn-of-the-month compared to the remaining days of the turn-of-the-month, monetary policy should have more affect during the turn-of-the month compared to the remaining days.

The evidence of the turn-of-the-month effect is mainly investigated in the US markets and less on the European stock markets. This thesis will try to answer if the turn-of-the-month effect is present in several European countries. Furthermore, this thesis will try to prove if higher liquidity during the turn-of-the-month is the cause of the turn-of-the-month effect. The effect will be investigated on the following indices; AEX (Netherlands), CAC (France), DAX (Germany), FTSE Athex (Greece), FTSE MIB (Italy), BET (Romania), WIG (Poland) and IBEX (Spain). Daily stock prices from January 2004 to December 2014 are collected using Datastream.

Chapter 2 reviews corresponding literature on the turn-of-the-month effect. In chapter 2.1 the theoretical framework describes the contradiction between the efficient market

hypothesis and the turn-of-the-month effect. In chapter 2.2 existing literature that provides evidence on the US and international markets for the turn-of-the month effect will be discussed together with other calendar anomalies. Chapter 3 describes the methodological approach and the data that is used in this study. Chapter 4 presents the empirical results of the research. Chapter 5 presents a discussion. In the last chapter a conclusion is given.

5

2. Literature review

2.1. Theoretical framework

When information about risk of a stock is being obtained by investors, the efficient market hypothesis suggests that this information should be absorbed by the price of the stock (Fama, 1970). This means that when relevant information becomes available this should directly be reflected by the price. High competition among investors eliminates the positive Net Present Value of an asset caused by the information. For investors it becomes impossible to

structurally outperform the market if markets are efficient. Consecutive stock price changes are independent and identically distributed random variables and therefore make a random walk and are unpredictable. The random walk of stock prices is linked to the efficient market hypothesis (Fama, 1965). If stock prices were predictable this would prove not all information available is reflected by the price which would mean the market is inefficient and would therefore be inconsistent with the efficient market hypothesis. The efficient market hypothesis is contradictory with the theory of calendar anomalies. Calendar anomalies show a pattern in stock returns at a particular time of the year. One of the calendar anomalies is the month effect. The month effect shows higher average returns at the turn-of-the-month compared to the remaining days of the turn-of-the-month.

2.2. Evidence for the turn-of-the-month effect

Ariel (1987) discovered a monthly pattern in stock returns for the US from 1963 to 1981 and described it as the monthly effect. He shows the mean cumulative return on the last trading day of the month to the first nine trading days of the following month is 1.411% against -0.021% for the remaining days of the month. Ariel (1987) proves that the monthly effect is independent and not an explanation from other calendar anomalies.

One of the first researchers that found statistical evidence for a calendar effect in stock returns was Wachtel (1942). His findings showed significant higher stock market returns in the month of January compared to the other months of the year. This yearly pattern is known as the January effect. He did an examination on the Dow-Jones industrial Average from 1927 to 1941. Earlier work concluded that there was no need to investigate the years before 1926 as

6 it turned out to be insignificant.

Analysis on the behavior of stock prices on Mondays and Fridays were done by Cross (1973). He proved significant lower returns during his researched time frame, on the Mondays compared to the Fridays. The results that were found on the US stock market are known as; the weekend effect, Monday effect and or day-of-the-week effect (calendar-effects.net).

Lakonishok & Smidt (1988) find evidence for higher returns around the turn-of-the-month, using daily data from the Dow Jones Industrial Average index over the period of 1897 to 1987. Average returns are 0.4732% on the last day of the month and the first three days of the following month, whereas for an average four day period within the month the average return rate is 0.0612 %. Because Lakonishok & Smidt discovered that the monthly pattern of higher average returns was on the turn of the month the phenomenon is being described as the turn-of-the-month effect.

Cadsby & Ratner (1992) find evidence for the turn-of-the-month effect in Canada, the UK, Switzerland and West Germany during a period from 1980 to 1989. They use the same turn-of-the-month period for their research as Lakonishok & Smidt (1988), which are the last day of the month and the first three days of the next month.

Ziemba (1991) studies the turn-of-the-month effect on the Japanese NSA and TOPIX indices from 1949 to 1988. He takes different periods of the turn-of-the-month in

consideration and documents strong significant higher returns for the last five day of the month and the first two days of the following month. For these days the average return is approximately 0.10 % higher in comparison with the other days of the month. The last day of the month is 0.22 % higher. The returns on these days were about two thirds of the total monthly return.

Agrawal & Tandon (1994) find international evidence for the turn-of-the-month effect for eighteen countries. Belgium, France, Germany, UK, Denmark and Luxembourg show on average exceeding returns at the-turn-of-the month compared to the rest of the month for the European countries. The Netherlands and Italy did not show the effect. The study uses a timespan from 1971 to 1987.

Marquering et al. (2006) examine the turn-of-the-month effect before and after the anomaly was published. They use the published studies from; Ariel (1897), Lakonishok & Smidt (1988), Cadsby & Ratner (1992) and Kunkel et al. (2003). The effect was expected to disappear after the anomaly was reported. An explanation for this this expectation is the attention of the publication of the anomaly among investors. The effect will disappear as more

7 investors will trade based on the published calendar anomalies. The results of the study show that the turn-of-the-month effect was still present after the publications.

2.3. Possible causes

Ziemba (1991) suggests that the turn-of-the-month effect is caused by the timing of financial flows. Most salaries are paid on the twenty-fifth of the month which leads to buying pressure shortly after these days for individual investors. Investor will invest their liquid profits when the accumulative assets are sufficient to invest in order to reduce the transaction costs (Elton and Gruber, 1974). Ziemba (1991) proves the turn-of-the-month effect to appear on the last five days of the month to the first two days of the next month which corresponds to the day of salaries being paid. Ziemba (1991) shows the turn-of-the month period arrives earlier in Japan in comparison to the US (Sick and Ziemba, 1991). This explanation seems supported by the fact that salaries in Japan are paid earlier in Japan than other countries. Ziemba (1991) claims window dressing in the Japanese market on the last day of the month as a possible cause of the turn-of-the-month effect. Window dressing is a strategy used to

improve the appearance of the portfolio before presenting them to the stock holders. Poor performing stocks are sold while stocks with high returns are purchased (Zweig, 2012). Lakonishok et al. (1991) study window dressing by pension fund managers in the US. The managers sell bad performing stocks before they are evaluated by their sponsor and rebalance their portfolio after the evaluation at the end of the turn-of-the-month by buying high-risk stocks to maintain their strategy.

Wiley and Zumpano (2009) distinguish the level of impact that institutional compared to individual investment has on the turn-of-the-month effect. They use the daily stock returns from real estate investment trusts (REITs) from 1980 to 2004. REITs are interesting for pension fund managers and other financial institutions because of the reduced volatility, diversification and dividend payout (Wiley and Zumpano, 2009). Wiley and Zumpano (2009) suggest that the trusts are particularly useful for this investigation because there was a tax law adjustment in 1993 that loosened ownership requirements and allowed institutional investors to invest significantly more in real estate investment trusts. Before the adjustment,

institutional investors were restricted in the amount they could invest in REITs. The institutional investors held on average approximately 23% at the end of 1992 and 49% in 1995 of the total shares outstanding. To analyze the influence of institutional investment

8 Wiley and Zumpano (2009) examine the turn-of-the-month effect before and after the tax law adjustment. Result show the stock returns around the turn-of-the-month are influenced by the level of institutional investment. The payday at the last day of the month is given as a

plausible explanation. After the employees receive their paycheck, a part of them choose to add some of that money to their investment account. The profits provide institutional

investors to reinvest these contributions. Secondly, Wiley and Zumpano (2009) give window dressing as a possible cause of the turn-of-the-month effect.

Ogden (1990) considers the turn-of-the-month period as the common payoff time for dividends, salaries and interest payments in the US. Ritter (1988) proves that these liquid profits are often reinvested shortly after being received by institutional investors and therefore have a positive impact on the investments at the turn-of-the-month. When liquid profits are large, investors will invest more in the stock markets at the turn-of-the-month causing an increase in returns. Liquidity growth is influenced by monetary policy, therefore the impact of monetary policy is also expected to affect liquid profits. When liquid profits are higher during the turn-of-the-month compared to the remaining days of the month, monetary policy should have more affect during the turn-of-the month compared to the remaining days of the month (Ogden, 1990). Ogden (1990) his turn-of-the-month liquidity hypothesis is tested using CRSP value-weighted and equally-weighted daily stock index returns from 1969 to 1986. Results show that the fed fund rate is inversely related to the returns during the turn-of-the-month period and has no significant impact on the returns of the remaining part of the month1.

2.4. Literature conclusion

The turn-of-month effect has been studied by many financial economists and most of them found significant prove on the turn-of-the-month effect. Most of the evidence is found in the US. There are studies that proved international evidence for the turn-of-the-month effect but no study provides an explanation for the difference between the results found for countries. Window dressing and the increase of liquid profits at the turn-of-the-month are given as common causes of the effect. Ogden (1990) finds empirical evidence for an increase in liquidity at the turn-of-the-month to be the cause of the turn-of-the-month effect.

1 _{Ogden (1990) focuses his research on the same turn-of-the-month period as were Lakonishok & Smidt (1988)}

9

3. Methodology and Data description

3.1. Methodology

The estimation method used for the model is ordinary least squares (OLS). The following model is estimated:

Rt = β0+β1TOM+β2Euribor+β3TOM*Euribor+β4S&P500+β5CRISIS+β6CRISIS*TOM+Ɛt Where Rtis the daily index return (used indices are described in chapter 3.2) in period t. TOM

is a dummy variable that takes the value 1 for the turn-of-the-month trading day and 0 otherwise. TOM is expected to be positive and significant. For the TOM variable a turn-of-the-month period starting from the last trading day of the month to the first three trading days of the following month (-1,3) is used based on Lakonishok & Smidt (1988) and Cadsby & Ratner (1992). The Euribor is the 1-month Euribor rate and TOM*Euribor is the interaction term of TOM and Euribor. The variables are based on the examination from Ogden (1990). The coefficients from the Euribor variable are expected to be insignificant. The coefficient from the variable TOM*Euribor is expected to have a negative and significant effect during the turn-of-the-month period. S&P500 is the daily index return of the Standard and Poors 500 index. The variable is added to control for a possible market-effect. CRISIS is a dummy variable that takes the value 1 for trading days during the CRISIS and 0 otherwise. We date the start of the financial crisis at the 15th of September 2008, the day Lehman Brother files for bankruptcy. The end of the most turbulent face is dated at the end of March 2009. By the end of March, the G20 committed to a fiscal expansion of 5 trillion USD by the end of 2010 to repair the financial system (Arner, 2011). The variable is edit to capture the change in returns during the crisis. CRISIS*TOM is the interaction term of CRISIS and TOM. Ɛt is the error term. In this thesis a turn-of-the-month period starting from the last trading day of the month to the first three trading days of the next month (-1,3) based on Lakonishok & Smidt (1988) and Cadsby & Ratner (1992) is used. If the null-hypothesis is accepted, there is no turn of the month effect and the average daily returns are equal. Secondly there is no effect of the 1-month Euribor on the daily returns at the turn-of-the-1-month

10

3.2. Data description

This thesis uses daily closing values of eight different European indices and covers the period from 1 January 2004 to 31 December 2014. Recent returns are used because in this thesis we want to research the current markets. The indices included in this thesis are AEX

(Netherlands), CAC (Franc), DAX (Germany), FTSE Athex (Greece), FTSE MIB (Italy), BET (Romania), WIG (Poland) and IBEX (Spain). All of the indices are value-weighted. When a firm pays out dividend the index needs to be corrected for this. This thesis therefore uses the return index which assumes dividends are re-invested to purchase an additional unit of equity. Markets could differ in terms of liquidity at the turn-of-the-month. To see if there are different results at the turn-of-the-month this thesis does not use the overall European total market index like the STOXX index.

The days that the markets were closed are removed from the data. The daily return would have been 0% during these days which will lead to less accurate results. For the weekends this was already done.

The daily returns of the indices are not given by Datastream and have to be calculated using the following formula:

R_𝑡 = ln⁡( P𝑡

P_𝑡−1) ⋅ 100

Where Rt is the daily return in period t, Pt is the daily price of the index at time t and Pt-1 is the return price of the index at time t-1.

This thesis uses the 1-month Euribor rate to test if increased market liquidity is an important determinant of the turn-of-the-month effect in stock returns. The Euribor (Euro interbank offered rate) refers to the average interest rate at which a large board of European banks lend money to each other (Bodie, 2009).

For the S&P500 the value-weighted daily closing values of eight different European indices are used. The same period as the depended indices is used (1 January 2004 to 31 December 2014). All of the values are retrieved by Datastream.

11 Table 1 shows the descriptive statistics regarding the eight indices. Noticeable is the fact that the FTSE Athex (Greece) has a negative mean and a large standard deviation compared to the other indices.

Table 1: Daily returns for eight indices

Index Mean daily

return (%) Standard deviation (%) Minimum daily (%) Maximum daily (%) N AEX (Netherlands) .0218 1.3206 -9.5905 10.0261 2819 CAC (France) .0204 1.4019 -9.4715 10.5944 2819 DAX (Germany) .0323 1.3574 -7.4335 10.7975 2801 FTSE Athex (Greece) -.0536 2.1734 -13.844 16.3741 2772 FTSE MIB (Italy) .0285 1.5316 -8.5981 10.8769 2795 BET (Romania) .0417 1.6500 -13.117 10.5645 2860 WIG (Poland) .0328 1.2795 -8.2888 6.08375 2757

IBEX (Spain) .0288 1.4857 -9.4013 13.4831 2801

The results in table 2 show the daily mean returns for both the days during the turn-of-the-month (-1,3) as the remaining days of the turn-of-the-month (no turn-of-the-turn-of-the-month) for each index. The standard errors are showed in parentheses. The S&P 500 was added to correct for possible market-effects. The number of observations is different from one index to another because some markets appeared to be closed more often than others. We can see that for all the indices there are higher daily mean returns during the turn of the month period compared to the non-turn of the month days. The results show there are three indices with significant higher renon-turns at the turn-of-the-month compared to the remaining days of the month. The FTSE Athex (Greece) and WIG (Poland) shows the existence of a turn-of-the-month effect with a

significance level of 5%. The BET (Romania) has the strongest significance. It proves to be significant at a 1% level. The AEX (Netherlands), CAC (France), DAX (Germany), FTSE MIB (Italy) and IBEX (Spain) don’t show a significant turn-of-the-month effect. Only few studies have proven empirical results on the cause of the turn-of-the-month effect. This makes it difficult to establish evidence for the difference in results between the countries2. A possible

2 _{After finding international evidence for the turn-of-the-month effect, Agrawal and Tandon (1994) conclude}

there is a need to refine the existing explanations of the effect to prove why the turn-of-the-month effect is present in different countries.

12 cause of the existence of the turn-of-the-month effect in FTSE Athex (Greece), WIG (Poland) and the BET (Romania) could be the fact that more investors use window dressing to cover up bad results as explained in chapter 2.3. The results in chapter 4.1 will discuss if the turn-of-the-month effect is caused by higher liquid profits during the turn-of-turn-of-the-month.

Table 2: Daily mean returns during the (no) turn-of-the-month

Index Turn-of-the-month (%) No Turn-of-the-month (%) R2 N AEX (Netherlands) ,0850 -,0144 0.371 2819 (,0526) (,0220) CAC (France) ,0595 -,0116 ,367 2819 (,0560) (,0235) DAX (Germany) ,0473 ,0022 ,396 2801 (,0533) (,0222)

FTSE Athex (Greece) ,2588** -,1051** ,076 2772

(,1052) (,0443)

FTSE MIB (Italy) ,0651 -,0281 ,314 2795

(,0642) (,0268) BET (Romania) ,2473*** ,0823 ,054 2855 (,0805) (,0490) WIG (Poland) ,10254** -,0063 ,158 2757 (,0601) (,0249) IBEX (Spain) ,0751 -,0031 ,291 2801 (,0631) (,0264) *Significant at 10% level **Significant at 5% level ***Significant at 1% level

13

4. Results

4.1. Results for the turn-of-the-month effect and liquidity

Table 3 shows the results of the OLS regression. The standard errors are showed in parentheses. The S&P 500 was added to correct for possible market-effects. The AEX (Netherlands), CAC (France), DAX (Germany), FTSE MIB (Italy), BET (Romania), WIG (Poland) and IBEX (Spain) indices show no significant turn-of-the-month effect. The interaction term between the turn-of-the-month and the Euribor does not show any

significance. The FTSE Athex has a significant turn-of-the-month effect with a level of 10%. This could be due to the fact the added Euribor with the interaction term of Euribor and TOM have a small explanatory impact on the returns of the FTSE Athex so that the existence of the turn-of-the-month effect is still present. For the FTSE Athex there are negative coefficients for the Euribor variable and the interaction term of the Euribor and the turn-of-the-month as expected although they do not seem to be significant. The results of the variables are

inconsistent with the results Ogden (1990), proved for all of the indices. The interaction term of the turn-of-the-month dummy variable and the 1-month Euribor is expected to be negative and significant. The 1-month Euribor variable is expected to be negative but with a smaller coefficient compared to the interaction term. The intended prove for the turn-of-the-month effect as a cause of higher liquidity is not found for the given markets. The results of the BET (Romania) and WIG (Poland) show that the turn-of-the-month effect disappears after

introducing the 1-month Euribor and the interaction term of the turn-of-the-month dummy variable and the 1-month Euribor.

14

Table 3: OLS-estimation of turn-of-the-month on index returns

*Significant at 0.10 level **Significant at 0.05 level ***Significant at 0.01 level

4.2. Results during the crisis

Table 4 shows the results of the OLS regression for the crisis dummy variable with the interaction term of the crisis and the turn-of-the-month. All of the indices show higher daily mean returns on the turn-of-the-month days compared to the remaining days of the month. The AEX (Netherlands), CAC (France), DAX (Germany), BET (Romania) and IBEX (Spain) prove to have a significant turn-of-the-month effect during the crisis. For the AEX this effect has the highest significance at a level of 1%. The CAC, DAX and BET are significant at a level of 5%. The IBEX proves to have a significant turn-of-the-month effect at a level of 10%. For the FTSE MIB (Italy), FTSE Athex (Greece) and WIG (Poland) indices there is no

evidence for a month effect during the crisis. The results indicate the turn-of-the-month effect becomes significant for some indices when the economy is affected by a crisis. The AEX, CAC, DAX, BET and IBEX have a significant turn-of-the-month effect during the crisis. All of the indices show the difference between the daily mean returns at the turn-of-the-month compared to the remaining days of the turn-of-the-month become larger in times of the crisis. Furthermore, the effect disappears for the FTSE Athex and WIG indices. The sample of the

Index TOM Euribor TOM*Euribor R2 N

AEX (Netherlands) -,0207 -,0253* ,0610 ,372 2819 (,0804) (,0147) (,0350) CAC (France) -,0185 -,0095 ,0449 ,368 2819 (,0857) (,0157) ,0373) DAX (Germany) -,0488 -,0094 ,0551 ,397 2801 (,0817) (,0151 (,0355)

FTSE Athex (Greece) ,3028* -,0009 -,0252 ,076 2772 (,1614) (,0295) (,0700)

FTSE MIB (Italy) -,0462 -,0160 ,0639 ,314 2795

(,0984) (,0179) (,0428) BET (Romania) ,1217 -,0626*** ,0728 ,057 2855 (,1226) (,0226) (,0535) WIG (Poland) ,0579 -,0303* ,0391 ,159 2757 (,0919) (,0166) (,0400) IBEX (Spain) -,0688 -,0106 ,0830 ,307 2801 (,0963) (,0176) (,0419)

15 crisis period has fewer observations compared to the number of observation that was used for the sample without the crisis dummy variable. This could have an effect on the significance.

Table 4: OLS-estimation of the turn-of-the-month during the crisis

*Significant at 0.10 level **Significant at 0.05 level ***Significant at 0.01 level

Index CRISIS CRISIS*TOM R2 N crisis

AEX (Netherlands) -,4009*** ,9525*** ,377 128 (,1054) (,2614) CAC (France) -,1903* ,6136** ,369 128 (,1126) (,2793) DAX (Germany) -,1744 ,5486** ,398 127 (,1067) (,2690)

FTSE Athex (Greece) -,3533* ,2192 ,077 124 (,2116) (,5340)

FTSE MIB (Italy) -,2247* ,1427 ,315 126

(,1289) (,3246) BET (Romania) -,6109*** ,9745** ,062 128 (,1615) (,4019) WIG (Poland) -,2402** ,2163 ,160 126 (,1193) (,3000) IBEX (Spain) -,1721 ,8250* ,294 127 (,1267) (,3193)

16

5. Discussion

A few suggestions can be made regarding this research. Many causes of the turn-of-the month effect have tried to explain the effect, but most of these theories have not been empirically tested. This makes it hard to use them for international evidence. Alternative causes of the effect are needed to successfully explain why the turn-of-the-month effect is present in a certain market. This would be useful to help explore the markets for which a

turn-of-the-month effect is present. This

thesis investigates the period from January 2004 until December 2014. Future research could be improved if different research periods are used to find the turn-of-the-month effect. A different turn of the month period could also improve future studies. Furthermore, a broader selection of markets (countries) can be used. At last, the characteristics of the markets can be studied more deeply to see if they support the turn-of-month effect.

6. Conclusion

This thesis investigates the turn-of-the-month effect in several European countries and tries to prove if liquidity is the cause of the effect. The turn-of-the-month effect is a calendar anomaly for which the daily returns are higher at the turn-of-the-month compared to the remaining days of the month. Ariel (1987) discovered the turn-of-the-month effect in stock returns for the US from 1963 to 1981. Lakonishok & Smidt (1988) find evidence for higher returns around the turn-of-the-month, using daily data from the Dow Jones Industrial Average index over the period of 1897 to 1987. Cadsby & Ratner (1992) find evidence for the turn-of-the-month effect in Canada, the UK, Switzerland and West Germany during a period from 1980 to 1989. Ziemba (1991) proves the existence of the turn-of-the-month effect on the Japanese NSA and TOPIX indices from 1949 to 1988. Agrawal and Tandon (1994) find international evidence for the turn-of-the-month effect for; Belgium, France, Germany, UK, Denmark and Luxembourg.

Ogden (1990) considers the turn-of-the-month period as the common payoff time for dividends, salaries and interest payments in the US. The standardization of payments provides higher liquidity at the turn-of-the-month. Institutional investors reinvest the liquid profits causing an increase in stock market returns. Liquidity growth is influenced by monetary

17 policy; therefore the impact of monetary policy is also expected to affect liquid profits. When liquid profits are higher during the turn-of-the-month compared to the remaining days of the month, monetary policy should also have more affect during the turn-of-the month compared to the remaining days.

This thesis investigates if there is evidence for a turn-of the month effect and if liquid profits are the cause of this effect. The daily 1-month Euribor rates are used to investigate if higher liquidity is the cause of the turn-of-the-month effect. A dummy variable for the 2008-2009 financial crisis is used to investigate if there are different results for the crisis period. The S&P 500 is added to correct for a possible market-effect. This thesis investigates the period from January 2004 until December 2014. Evidence is found for the turn-of-the-month effect for the FTSE Athex (Greece), BET (Romania) and WIG (Poland) indices. For the AEX (Netherlands), CAC (France), DAX (Germany), FTSE MIB (Italy) and IBEX (Spain) no significant proof for the turn-of-the-month anomaly is found. No sufficient evidence is found to prove higher liquidity during the turn-of-the-month to be the cause of the turn-of-the-month effect. For this test the anomaly is not present in any market except for the FTSE Athex, but no significant effect of liquidity can be found. The fact that the turn-of-the-month effect is significant could be due to the fact the added Euribor with the interaction term of Euribor and TOM have a small explanatory impact on the returns of the FTSE Athex so that the existence of the turn-of-the-month effect is still present. During the crisis the AEX, CAC, DAX, BET and IBEX have a significant turn-of-the-month effect during the crisis. The small sample size could have an effect on the significance during this period.

The main conclusion of this thesis is that the turn-of-the-month effect can be found on several European markets but is inconsistent in its appearance. The turn-of-the-month effect is affected by the financial crisis. This thesis has not found any prove for liquidity being the cause of the turn-of-the-month effect.

18

7. References

Agrawal, A. & Tandon, K. 1994, Anomalies or illusions? Evidence from stock markets in eighteen countries, Journal of International Money and Finance, 13 (1), 83-106.

Ariel, R., 1987, A monthly effect in stock returns, Journal of Financial Economics, 18, 161-174.

Arner, D. & Buckley, R., 2011, The Global Financial System and Regulatory Failure (International Banking and Finance Law), Wolters Kluwer, 137-138.

Cadsby, C.B. & Ratner, M., 1992, Turn-of-the-month and pre-holiday effects on stock returns: some international evidence, Journal of Banking and Finance, 16 (3), 487–510.

Cross, F., 1973, The behavior of stock prices on Friday and Monday, Financial Analysts Journal, 67-69.

Calendar-effects.net, http://calendar-effects.behaviouralfinance.net/weekend-effect/, viewed on January 5, 2015.

Dzhabarov, C. & Ziemba W., 2010, Journal of portfolio management, 36 (3), 93-104.

Bodie, Z., Kane, A., & Marcus, A.J., 2009, Investments, New York: McGraw-Hill, 7th, 30-31. ec.europa.eu, http://ec.europa.eu/economy_finance/crisis/2009-03_en.htm, viewed on 28 January, 2015.

Fama E.F., 1965, The behavior of stock market prices, The journal of Business, 38 (1), 34-105.

Fama, E.F., 1970, Efficient Capital Markets: a Review of Theory and Empirical Work, Journal of Finance, 25 (2), 386-417.

Fama, E.F., 1998, Market efficiency, long-term returns, and behavioral finance, Journal of Financial Economics, 49, 283-306.

19 French, K., 1980, Stock return and the weekend effect, Journal of Financial Economics, 8, 55-70.

Gibbons, M. & Hess, P., 1981, Day of the week effects and asset returns, Journal of Business, 54 (4), 579-596.

Hensel, C., Sick, G., & Ziemba, W., 1994, The Turn of the Month Effect in the S&P 500, 1926-1992. Review of Futures Markets, 3 (3), 827-856.

Lakonishok, J. & Smidt, S., 1988, Are seasonal anomalies real? A ninety-year Perspective, The review of financial studies, 1 (4), 403-425.

Marquering, W., Nisser, J., & Valla, T. 2006, Disappearing anomalies: a dynamic analysis of the persistence of anomalies, Applied Financial Economics, 16, 291-302.

Ogden, J., 1990, Turn-of-the month evaluations of liquid profits and stock returns: A common explanation for the monthly and January effects, The Journal of Finance, 16 (4), 1259-1272.

Stoxx.com, http://www.stoxx.com/indices/index_information.html?symbol=SX5E, viewed on January 10, 2015.

Wachtel, S.B., 1942, Certain observations on seasonal movements in stock prices, The Journal of Business of the University of Chicago, 15 (2), 184-193.

Wiley, J. and Zumpano, L. 2009, Institutional investment and the turn-of-the-month effect: Evidence from REITs, The Journal of Real Estate Finance and Economics, 39 (2), 180-201.

Ziemba, W.T., 1991, Japanese security market regularities: monthly, turn-of-the-month and year, holiday and golden week effects, Japan and the World Econnomic, 3, 119–146.

Zweig, J. 2012, Fund managers lift results with timely trading sprees, The Wall Street Journal, retrieved from: