Performance of Mutual Funds and Exchange Traded Funds (ETFs)

10552 words (42 pages) Dissertation

16th Dec 2019 Dissertation Reference this

Tags: Finance

Disclaimer: This work has been submitted by a student. This is not an example of the work produced by our Dissertation Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of NursingAnswers.net.

Table of Contents

Abstract 2

Introduction 3

Literature Review 6

Data and Methodology 12

Findings and Discussions 15

Fund Performance 16

Expense Ratio 18

Dividend Taxes 21

Fund Listing and Fund Performance 27

Conclusion 28

Reference List 31

Abstract

In this paper, I investigate the performance of mutual funds and exchange traded funds (ETFs) listed in Europe for the period March 2008 until March 2018. The empirical result found in this research indicates that the median of the 12-month return difference between the benchmark index and the mutual funds and ETFs in my sample is 34 basis point. Thus, most mutual funds and ETFs underperform the benchmark index tracked. I also examine the relationship between fund expense ratio and fund performance and find that fund expense ratio is significant in explaining the fund return. However, the expense ratio has an extremely low explanatory power on the fund performance. Furthermore, I investigate the impact of dividend taxation on fund returns. Similar to expense ratio, dividend taxation is one of the contributor towards the fund underperformance. However, dividend taxation yield an even lower explanatory power than expense ratio. Lastly, I examine the effect of fund listing on the performance of funds. It is observed that when fund listings are jointly taken into account with expense ratio and dividend taxation, the explanatory power for fund listings is extremely minimal.

1. Introduction

Exchange traded fund (ETF) is an investment product which has grown tremendously since its first creation in the United States (US). According to Poterba and Shoven (2002, p1), even though the ETF is first created in the year 1993, it has been growing exponentially throughout the years, especially in the year 2000. By 2001, the ETF has gestated to an astounding amount of $79 billion in assets held, which made up 2.4 percent of the total assets in equity mutual fund. Even after its initial introductory phase, the ETFs continue to grow at an astonishing pace. According to the EY Global ETF Research 2017 Report, ETF which totaled to $417 billion in 2005, has expanded to $4.4 trillion by September 2017 with an impressive growth rate of 21% per annum. With the persistent annual growth, ETFs will yield increasing assets under management and would be a vital substitute for investors interested in mutual fund investments.

ETFs are of interest for investors mainly because the funds give investors the ability to buy and sell the investment vehicle throughout the day, which is very similar to trading a common stock on the stock exchange. This gives the investors flexibility in their investments. This feature is also the main principal that distinguishes ETFs from traditional mutual funds. Furthermore, another benefit of the ETF is that it gives the ability for investors to get an exposure to a specific, narrowed sector or market in an “inexpensive and efficient method” as stated by Gallagher and Segara (2005, p.53). Gallagher and Segara (2005) also suggests that ETFs are more tax efficient due to limited realized capital gains allowed. Moreover, ETF can also be traded in smaller quantity as compared to futures contracts and hence yield advantages over the mentioned derivative product. As a result, the ETF has grown immensely popular among investors.

Therefore, in this paper, I examine the performance of the Exchange Traded Fund (ETF) that are listed in the stock exchanges in the European market. The main objective of this research is to compare the performance of ETFs and other passive funds listed in the Europe market with regards to their tracked benchmark index using data that covers the period from March 2008 until March 2018. According to PWC’s 2nd Annual Global ETFs survey (2015), American ETFs is expected to have a cumulative annual growth rate of 23%, the European ETF market is expecting a 27% annual growth while the Asia market is expected to have 18% compounded annual growth over the next five years. Despite the evidence showing a robust growth of ETFs in European market, there are relatively less researches done to investigate the Europe ETFs in comparison to the literatures done on the US ETFs. Hence, this paper adds value by providing recent researches on the European ETF. The key question that this paper intends to answer is how are the performances of these ETFs in the Europe market?  How are the performance of the ETFs and other traditional mutual funds listed the Europe market in comparison to the performance of the tracked benchmark index? What are the causes for the ETFs to yield such return?

This essay is organized as follows. First, I document the total return for the mutual funds and ETFs in my sample and compare them against the tracked benchmark index. I observe that the funds listed in Europe yield returns ranging from an outperformance of 3 basis points to an underperformance of 54 basis points. The underperformance of the fund is slightly larger than the underperformance of funds listed in the US market with an amount of 28 basis point as found by Elton et al (2000). Elton et al (2004) later identified that the fund performance has a negative relationship with the fund’s expense ratio which is also similar to the results that I have found in my study. I also observe from my result that the fund underperforms the benchmark index by a slightly larger amount than the reported fund expense ratios. Hence, this indicate that fund underperformance cannot be attributed solely to expense ratios.

Besides that, I also note that there are variations in the performance of the funds that tracked different geographical indexes. As an example, mutual funds provided by Vanguard that tracked S&P500, MSCI Europe and MSCI Japan has different fund returns across the different region even though the fund expense ratios are within a similar range. Therefore, this indicates that there are evidences where the fund expenses are unable to explain the funds’ underperformance.

Third, I observe the impact of dividend taxation in explaining the variation of the fund underperformance in my sample. In the results that I have found, the explanatory power of dividend taxation is extremely low. I find that the R square for dividend taxation is only 0.5 basis point, which is one quarter of the explanatory power of fund expense ratios. Hence, this suggests that although the funds listed in Europe are subjected to different dividend taxation based on the different benchmark index’s regions, it cannot be attributed entirely in explaining the fund returns.

Fourth, I examine the impact of fund listings on fund returns. I identify that when fund listings are jointly taken into account alongside fund expense ratio and dividend taxation, the explanatory power of fund listings disappear entirely.

My paper contributes to the research of ETFs by providing recent insights towards the relationship of fund performance listed in the Europe market with respect to expense ratios, dividend taxation and fund listings. There are not many researches on the ETF universe that involve recent data. Most researches such as the ones from Elton et al (2004), Mussavian and Hirsch (2002), Gallagher and Segara (2005), Poterba and Shoven (2002) used data prior to 2008. Thus, my research intends to fill in this gap by providing more up to date study regarding ETFs.

In Section 2 I describe the Literature Review and in Section 3 I outline the process of extracting the data required for this study and the methodology used in analyzing the data. In Section 4, I discuss the empirical results that I have found and in Section 5 I summarize the conclusion of this paper.

2. The Literature Review

Exchange Traded Fund (ETF) definitely has been a continuously unfolding issue in the academic literature. Since the first ETF has been created in 1993 in the United States (US), ETFs have been growing exponentially, up to 19.8 billion dollars by the end of 1999. Elton et al (2000) is the first literature to discuss ETFs after its creation. Referring to Elton et al (2000)’s literature, in 1998, Spider, an ETF that track the S&P500 has achieved the highest daily dollar volume for any share traded. This indicates the vast importance of ETFs. Meanwhile, the first ETF issued in Europe was in April 2000 which has been discussed by Mussavian and Hirsch (2002) in their literature. Both Elton et al (2000) and Mussavian and Hirsch (2002) noted that the ETF in the US and Europe are transparent, and that deviations of the price of ETFs from their Net Asset Value (NAV) are minute. This is due to the ability of the creation and deletion of ETFs. Gallagher and Segara (2006) concur with Elton et al (2000) as they found consistent results that indicates the ETFs instruments in Australia does not have a large variation between the ETFs price and NAV.

Hence, following the creation of ETF in the US and the issuance of ETF in Europe, other countries in the world have also issued ETFs to replicate indexes. However, despite the vast growth of ETF, Gallagher and Segara (2005), on the other hand, noted that ETFs in the Australian market are not as heavily adopted in comparison to the US and Europe market. Gallagher and Segara (2005) identified that the reason for investors to not embrace the Australian ETFs is due to the up-front fee that is required in order to enter the funds in Australia are absent in most funds in the US. Therefore, this creates a disadvantage towards fund managers in Australia to initiate the issuance of ETFs.

One area of study that is significant in observing the behavior of ETF is the study of the performance of ETF. In general, the average performance of the ETFs could be identified from Svetina and Wahal (2008) literature. This is due to the large samples of ETFs used and the inclusivity of the ETFs across different sectors and markets. The samples of ETFs used amount to 584 ETFs that include a wide range of sectors such as domestic equity, international equity, and fixed income ETFs. The literature shows that the ETFs underperform the benchmark index on average. Another finding that Syetina and Wahal (2008) generate from their study is that 83% of the ETFs focused on a specific and narrow market in order to give investors niche exposures and hence cannot be directly investable using index mutual funds. There are also studies that intend to investigate whether using trading strategies are effective in trying to gain excess return in investing in ETF. Tse (2015) examines the utilization of momentum trading strategies using the ETFs. However, the results found that the momentum trading strategies are statistically insignificant to the return on ETFs. Only few ETFs exhibit excess return, especially during the 2008 financial crisis. However, the return plummeted immediately post financial crisis. Overall, the momentum strategies using ETFs yield worse return than the simple buy and hold investment strategies. The limitation of this is that it does not include the results momentum strategies might exhibit in relation to other investment strategies using ETFs such as value strategy and other trading strategy.

There are also comparisons of studies found in the performance of ETFs across different countries and the reason for the ETF to yield such return. In the US, Elton et al (2000) noted in his result that Spider underperform the S&P500 Index and the main causes of the underperformance are the management fees and dividend reinvestment that results in loss of return. Similarly, as Blitz et al (2012) concluded in their result, the European ETFs also underperform their benchmark index. However, Blitz et al (2012) has found that the underperformance is not only due to the expense ratio, but it is also explained by dividend withholding taxes. Milonas and Rompotis (2014) narrow their research further to focus on the ETFs traded in the Swiss market. One of the most surprising findings that they find is that the tracking error that the ETF exhibit is substantially large. This is the same case as the global emerging markets ETF. The global emerging markets which include countries such as Russia, China, South Korea, India, South Africa and Brazil, according to Blitz and Huij (2012), the returns of ETFs in comparison to the returns of the benchmark index do not yield any significant meaning as the ETFs manifest high levels of tracking error. The situation is different in markets where the ETFs are in the initial growth phase. Prasanna (2012) show that the ETFs listed on Indian Stock Market has a staggering growth of 37% annually from 2006 until the year 2011. Unlike the researches done on ETFs in developed markets and the emerging markets discussed before, the Indian ETFS outperform the benchmark index, CNX NIFTY, by 3% per annum. Besides that, Gold ETFs yields excess return over the market by 13% after the financial crisis period. Moving on to South Africa, Strydom et al (2015) has identified that the South African ETFs have tracked the benchmark index more closely than the index funds. Hence, the author concludes that investors searching for exposure to the South African market could invest in ETFs rather than the mutual index fund for better replication of the index benchmark. However, the performance of ETFs is similar to the developed market ETFs, where the ETFs underperform the benchmark index, which is FTSE/JSE Top 40 Index in this case.

Apart from that, there are also studies that aims to identify the benefits of ETF. One of the advantage of ETF is that it is tax efficient. Poterba and Shoven (2002, p.8-9) conduct a study to show the pre-tax and post-tax returns on SPDR in comparison with the pre-tax and post-tax return on Vanguard Index 500. The authors find that pre-tax and post-tax return for SPDR and Vanguard Index 500 are very similar, albeit the return for the ETF is slightly lower than the mutual fund. This is due to the “tax efficiency” characteristic of the ETF. The main feature of ETF that leads to lower tax on the return is; “the ETF distribute securities with a basis below the market price to eliminate the potential capital gains tax liability ETF investors might face if these shares were sold”. However, the limitation of this paper is that it only uses SPDR that tracks the S&P 500 index in the paper’s analysis. The other ETFs have expense ratios which are significantly higher, which translate to lower return.

Besides that, another advantage of ETF is that it provides a specific exposure to a certain market desired by the investors. This is clearly shown in the literature produced by Meric et al (2010) where the author used ETFs in order to determine the sector that experience the largest losses during the financial crisis. Meric et al (2010) provided an insight into the performance of the ETFs in bear market which is from the year 2007 until the year 2009. The authors conclude from the results that they have generated using the Sharpe and Treynor ratios that the highest losses are generated by ETFs from the financials and the home construction sector whereas the sectors that yield the least losses and therefore, the best performances are the ETFs from the healthcare and consumer staples sector. Hence, the feature of ETF that targets a narrow, niche market is also useful not only for investors but also for academic researches.

Moving on to another area of study, there are also various researches in determining the difference between ETFs and other traditional index mutual fund. In Kostoversky (2003)’s paper, the author show that the main difference between index mutual fund and ETFs are management fees, shareholder transaction fees and taxation efficiency. The author also concludes that there are a few qualitative difference between the two investment instruments. One advantage of ETF is that it is relatively simple to trade ETF as it can be bought or sold throughout the day, yielding a similar characteristic to common stock. Apart from that, the investors also have the ability to place stop-loss and limit orders on the ETFs which are vital in restricting the amount of loss an investor could incur. On the other hand, the advantage of index funds is that it is relatively easier than ETFs for investor to accomplish the entire process of investment. Index fund requires less procedures for an investment to be executed. The limitation of this paper, as stated by the author is that the models proposed are not a perfect representation of real-world scenarios.

Another literature, Agapova (2010) compares the index fund and ETFs and finds that the two investment instruments are substitutes for each other, albeit not perfect substitutes. On the contrary to Kostoversky’s (2003) paper, the evidence in Agapova (2010) paper’s result indicate that the different features of the index funds and ETFs are subjected to more than just price and cost. The coexistence of the funds naturally target different market segments resulting in the funds being focused on different market niches.

Apart from that, there are also a research in another area that study the difference between the performance of ETFs and closed-end country funds. Harper et al (2005) overall finding is that ETFs yield higher return than closed end funds due to the lower expense ratios. Furthermore, ETFs have higher Sharpe ratio than the closed end funds. Besides that, closed end funds exhibit negative alpha, which indicates that investing in a passively managed investment using ETF yield higher return than actively managed investment using closed end funds.

Taking into account all the literature reviewed above, it is noticeable that there are less researches on European ETFs despite the European market being the second largest market where ETFs are heavily and frequently traded. Furthermore, there are less researches that use recent data and hence this is the gap that I intend to fill. The finance research that I am planning to conduct is to examine the performance of the index mutual funds and ETFs that are listed in Europe by using up-to-date data. Although it is also observable that there are not many ETFs research in Asia market and it is interesting to conduct one due to its high growth and the potential for the ETFs to outperform the market, there are limitations in terms of data availability from Asian countries. Apart from that, the largest ETFs being traded are mainly in the US, followed by Europe, Asia Pacific and the rest of the world. Hence, the vast availability of data for Europe market will make this research possible. One of the incentive for conducting this research is also to investigate whether the results found by Blitz et al (2012) regarding the performance of the ETFs still holds true in the current financial market.

3. Data and Methodology

I focus my analysis on the traditional index funds and ETFs which are listed in Europe. According to EY Global ETF Survey 2017, ETF is predicted to gain asset growth of 15% per annum in the next five years. Besides that, EY believes that ETFs asset could grow to $7.6 trillion by the year 2020. In Europe, ETF has been growing up to 20.4 percent on average per annum as reported by Deloitte in the Growth of ETFs in Europe report. The European ETF’s asset under management has an average growth of 20.1 percent per annum for the past decade.

I specifically narrow the analysis of my study to index funds and ETFs that track the main market indexes which give exposures to investors towards diversified equity markets. The market indexes include S&P500 and MSCI USA for US related exposures, MSCI Europe for Europe exposure, MSCI Japan for Japan exposure, MSCI World and MSCI Emerging Markets for exposures to the respective regions. The primary source of fund-level data for this research are Thomson Financial Datastream and the Morningstar website. Besides that, there are also index related data that are gained from the MSCI website and the individual traditional index funds’ websites.

This study is first started with the creation of a comprehensive list consisted of all available traditional index fund and ETFs. The lists of index funds and ETFs are collected from Thomson Financial Datastream. The list is constructed to focus on the European market. Therefore, the list includes funds listed in the stock exchanges from Austria, Belgium, Bulgaria, Channel Islands, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Kazakhstan, Latvia, Lithuania, Luxembourg, Netherlands, Norway, Poland, Portugal, Russia, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine and the United Kingdom. In order to gather the traditional index funds, I filter the list to include only funds that consist of the word “index” and “idx”. Then, the list is further restricted to only include funds that are not enhanced indexing funds, institutional funds and insurance funds. Due to a large amount of data amounting up to more than 5,000 funds, only a sample of the whole list is taken into consideration. This is due to the limited availability of duration given for conducting the research and limited manpower to analyze the fund-level data on a case by case basis. Hence, only index funds that are issued from established providers are taken. The providers are Vanguard, State Street Global Advisors (SSgA), Pictet and HSBC. Next, I gather funds that only focus on tracking the major benchmark indexes such as S&P 500, MSCI USA, MSCI Europe, MSCI Japan, MSCI World and MSCI Emerging Market. The funds are then checked on an individual basis using the Morningstar website, the Financial Times and the website of the companies that issue these funds in order to find the fund’s objective and investment policy. This is to investigate the nature of the fund and the clients that the funds attract. Any enhanced funds and institutional funds are excluded from this list.

Meanwhile, the details for ETFs are also collected from Thomson Financial Datastream. The list are filtered to include only European listed ETFs. Similar to the index funds data, only a sample of the whole list is taken into the paper’s analysis as the total data amount to 7,000 ETFs. The sample is taken by choosing ETFs from main ETF providers such as iShares, Lyxor and Street Tracks. The funds are then check on a case by case basis by using the Morningstar website, and the Financial Times in order to filter for funds that track only the major benchmark indexes. Apart from that, due to the characteristic of ETFs that are created with the purpose of gaining exposure to a specific industry or region, I exclude any ETFs that are too narrowly focused on a certain sector of the market indexes.

I obtain the data for the return of the funds in my sample by searching Thomson Financial Datastream for the Price of the funds and the benchmark index tracked. The gross total return for the funds are then calculated. The total return data are assembled from the period of March 2008 until March 2018 for a period of 10 years. The median for the 12 month return of the funds are then compared to the respective benchmark index’s return and the difference are then calculated. The expense ratio data for the traditional mutual funds are collected by searching the Total Expense Ratio data in Thomson Financial Datastream and the expense ratios for ETFs are obtained from the Morningstar Website by checking them on a case by case basis.

As mentioned above, the funds in my sample are selected from leading fund providers. The providers for the traditional mutual funds include Vanguard, State Street Global Advisors (SSgA), Pictet and HSBC. Meanwhile, the ETFs are from iShares and Lyxor. There are also funds that have multiple listings in the European stock market in which I have included in my sample. In total, a sample of 85 funds consisting of the most relevant traditional mutual funds and ETFs is created for this research. Since this paper aims to investigate the performance of funds listed in Europe for the most recent period, this paper only include the period from March 2008 to March 2018. The data for the previous period is already covered in Blitz et al (2012) research in which the authors investigate the performance of European fund for the period January 2003 to December 2008.

4. Findings & Discussion

In this empirical results section, I examine the explanatory power of fund expense ratios and dividend withholding taxes on the performance of funds listed in the European market.

 

 

 

 

 

 

4.1 Fund Performance

Table 1

Performances

This table exhibit the fund performance in my sample. The fund performance is reported from my measure of 12-month return difference between the funds and the benchmark index tracked. The data covered the period from March 2008 until March 2018.

Performance of Funds
  S&P500 OR MSCI USA MSCI EUROPE MSCI JAPAN OR TOPIX MSCI WORLD MSCI EM Median
ETFs
iShares -0.54% -0.31% -0.34% -0.36% -0.14% -0.34%
Lyxor -0.44% -0.33% -0.39% -0.27% -0.20% -0.33%
INDEX FUNDS
Vanguard -0.30% -0.12% -0.36% -0.02% -0.24% -0.34%
State Street -1.00% -1.00% -0.33%
PICTET 0.03% -0.32% 0.002% 0.02% 0.01%
HSBC -0.47% -0.47%
Median -0.44% -0.32% -0.35% -0.32% -0.20% -0.34%

In this section, first of all, I summarize the performance of the funds in my sample in the table above. I began by investigating the performance of ETFs and index mutual funds with respective to the performance of the tracked benchmark indexes. Table 1 summarizes the median of the 12-month return difference between the ETFs and mutual index funds against the tracked benchmark index. According to Table 1, almost all funds underperformed the benchmark indexes. Only mutual funds provided by PICTET have excess return over the tracked benchmark indexes. PICTET funds that tracked S&P500, MSCI Japan and MSCI Emerging Markets (EM) outperform the equity indexes by 3 basis points, 0.2 basis points and 2 basis points respectively, which is not significantly large. For the rest of the funds, the extent of the fund underperformance ranges from 2 basis points to 54 basis points per year, which is slightly different from the underperformance found in Blitz et al (2012) paper. In the paper, Blitz et al (2012) identified the underperformance of the European index funds and ETFs to be in between 50 basis point and 150 basis point per annum. Moreover, the median for the underperformance of the funds in my sample is 34 basis point. This value is much lower than the median identified by Blitz et al (2012) in their research where the underperformance of the funds listed in Europe has a median of 84 basis point. This indicates that the funds’ performance might have improved for the period 2008 to 2018 as compared to the funds listed in Europe for the year 2003 to 2008 as identified in Blitz et al (2012) study. However, the underperformance of European ETFs and index funds in my study are larger in comparison to the US oldest ETF, The Standard and Poor’s Depositary Receipts (Spider) that tracked the S&P 500 index, where the funds underperform the benchmark index by only 28 basis point as mentioned in Elton et al (2000)’s literature.

As provided in Table 1, all the funds do not outperform the benchmark indexes by a significantly large amount. It is identified that the best performing funds are mutual funds issued by PICTET which track the S&P 500 with an outperformance of 3 basis point. In my sample, only mutual funds provided by PICTET outperform the tracked benchmark index. Meanwhile, the worst performing funds are mutual funds provided by HSBC and ETFs from iShares that track the S&P500 and MSCI USA with an underperformance of 47 basis point and 54 basis point respectively. In general, I conclude that these results suggest that most mutual funds and ETFs in my sample yield returns that are lower than the tracked benchmark index.

I also note that there are return differences between funds that track different benchmark indexes. As shown in Table 1, the median for the underperformance of funds tracking S&P 500 and MSCI USA is 44 basis points, whereas the funds tracking MSCI Europe, MSCI Japan and MSCI EM have medians of 32, 35 and 32 basis points respectively. In contrast to the findings from Blitz et al (2012), the lowest median underperformance of funds in my sample is from funds that tracked MSCI EM, with a median of only 20 basis points, and not from funds that track the Japan benchmark indexes as found in the authors’ research. However, the fund tracking the MSCI EM has a high tracking error and thus the median does not yield any significant meaning. In conclusion, there is a different range of fund performances which varies across the benchmark indexes that the funds tracked.

4.2 Expense Ratio

In this section I examine the extent in which the underperformance of European ETFs and mutual funds are caused by the fund’s total expense ratio. The total expense ratio is a specific fee charged to the investors of the funds. The expense ratio is consisted of the purchase fee, redemption fee, auditing fee and other related expenses. I summarized the reported fund expense ratios for my fund sample in Table 2 below.

Table 2

Expense Ratios

This table exhibit the expense ratios as reported by the funds in my sample.

Expense Ratio
S&P500 OR MSCI USA MSCI EUROPE MSCI JAPAN OR TOPIX MSCI WORLD MSCI EM Median
ETFs
iShares 0.33 0.32 0.51 0.44 0.65 0.44
Lyxor 0.20 0.25 0.45 0.30 0.55 0.30
 

INDEX FUNDS

Vanguard 0.18 0.20 0.23 0.30 0.27 0.23
State Street 0.07 0.55 0.31
PICTET 0.06 0.07 0.08 0.18 0.075
HSBC 0.06 0.06
MEDIAN 0.19 0.23 0.34 0.30 0.55 0.30

Table 2 exhibit the expense ratio charged annually on investors upon investing in the index funds and ETFs provided in the sample. As summarized in the table, for the traditional mutual funds, the lowest expense ratio is 6 basis point which is charged by the funds provided from PICTET and HSBC whereas the highest expense ratio charged is by the State Street fund family with an amount of 55 basis point. The mutual funds are funds that track the S&P 500 index and MSCI EM index respectively. On the other hand, the highest expense ratio for ETF are charged by iShares that tracks the MSCI Emerging Market while the lowest expense ratio is from Lyxor funds which track S&P500 and MSCI USA. The overall median for the results gained is 30 basis point which is only slightly lower than the underperformance of the funds that yield an overall median of 34 basis point. This indicates that there are a difference of 4 basis point. This implies that there is a slight gap in the funds underperformance that is not covered by the fund expenses. Based on the outcome reported in the tables alone, it can be concluded that the underperformance of the funds are almost entirely caused by the amount of the total expense ratio.

However, there is another observation that could be derived from the report of the expense ratios and fund performances shown in Table 1 and 2 above. Although Table 1 and 2 might exhibit a negative relationship between fund expense ratios and fund performances, the fund performance differences in Table 1 cannot always be explained by their expense ratio. This is because, as shown in Table 2, funds with a more expensive total expense ratio do not necessarily yield better return than funds that require less expense ratio. Table 2 presents that ETFs that track MSCI Japan and Topix provided by iShares has a high expense ratio cost with an amount of 51 basis point but the funds underperformed the market index significantly by 34 basis point. Meanwhile, funds provided by PICTET outperform the US benchmark indexes by 3 basis point but with a notably lower expense ratio which is 6 basis point.

Furthermore, as observed in Table 2, fund providers that charge similar expense ratios across funds that track different benchmark index yield a significantly different range of fund performance. To illustrate this point, PICTET funds charge a similar expense ratio for funds tracking the S&P500 or MSCI USA, the MSCI Europe and MSCI Japan with a range of 6 to 8 basis point, whereas the fund performance ranges from a return excess over the tracked benchmark indexes by 3 basis point to an underperformance of 32 basis point, with a large difference of 35 basis point. Similarly, Blitz et al (2012) has found in their research that there are significant differences in the fund performance even though the fund providers charge similar expense ratios for funds tracking different benchmark indexes. The author concluded that the underperformance of the funds cannot be explained solely by the expense ratios charged. Hence, I conclude that the expense ratio is not the sole reason for the underperformance of the funds in my sample as my research yield the same outcome.

However, I note that there are significant limitations in my research. The main limitation is that the fund sample used in this study is chosen at random among 5000 fund data before being filtered on a case by case basis in order to eliminate the funds that are not relevant to this study. There is a high possibility that funds that possess the right characteristic to be included in my sample have been excluded from the study and hence my findings might have been distorted.

4.3 Dividend Taxes

Here I examine the extent that the variation of the underperformance of ETF is explained by the variation of dividend taxes charged on the funds. Analysis are conducted on the taxation levied on dividends generated from the sample funds. There are certain dividend taxations that are influenced by the tax status of investors. However, I bear in mind that the after tax return received by investors are not relevant to this paper as the main objective of the research is to investigate the performance of mutual funds and exchange traded funds. Therefore, consistent with other researches done in examining the mutual funds and exchange traded funds, an assumption is made where the dividends received are fully reinvested in the funds.

Expense Ratio is an amount of cost incurred toward investors upon investing in mutual funds and ETFs. The costs include initial fees, management fees, administrative fees and other fees charged for investing in the funds. However, dividend taxation is not included in the calculation of expense ratios. Therefore, benchmark index such as MSCI has presented the returns on indexes by calculating the gross return on the indexes and the net return on indexes, in order to account for the effect of dividend taxation. The index gross return accounts for full dividend reinvestment in the funds, whereas the index net return is calculated using dividends that are assumed to be reinvested after being taxed using the worst withholding tax rate. Therefore, to account for the effect of dividend taxation on fund performance, the gross index return acts as an upper percentile for the total fund returns whereas the net index return acts as a lower percentile on the total fund returns. Table 3 below presents the list of withholding taxes rates obtained from the MSCI Index Calculation Methodology Report in January 2018 where it showcases the withholding tax rate used in MSCI’s calculation of the net index return. Table 3 also presents the dividend yield in regions relevant to my research. The dividend yields are obtained from Bloomberg and the range of period taken are from March 2008 to March 2018. Average dividend yield over the ten year period is then calculated for the respective region as shown below.

Table 3

Withholding taxes and Dividend Yields.

This table exhibit the withholding tax rates used by MSCI in the calculation of fund net total return. The table also presents the average dividend yield for the benchmark indices tracked by funds in the period March 2008 until March 2018.

Withholding Taxes Dividend Yield
United States 30% 2.10%
Japan 15.315% 2.06%
Europe
United Kingdom 0% 4.14%
France 30% 3.66%
Germany 26.375% 3.10%
Emerging Markets
Korea 22% 1.38%
Taiwan 21% 3.91%
Brazil 0% 3.63%
Russia 15% 3.80%
India 0% 1.37%
China 0% 2.61%

Based on Table 3 alone, it can be seen that the withholding tax might have a significant effect on the return of mutual funds. To illustrate the point, taking into account the US withholding tax rate which is 30% and the dividend yield of MSCI US from March 2008 to March 2018 which is 2.10%, the potential fund underperformance that results from the dividend taxation amounts up to 63 basis point per annum. However, according to Table 1, the total fund underperformance for funds that track MSCI US is only 44 basis point. Hence, the inconsistencies of the results found might be due to the limitation of this paper which is to select only a sample of funds to be included in the research instead of using the whole fund population. Besides that, across other countries, the withholding tax rates varied from 0% to 26%, which demonstrate that different regions have different dividend taxation. In Emerging Market such as India and China, the withholding tax rate is 0% while the withholding tax rate levied on dividends in France is 30%. Consequently, the performance of mutual funds and ETFs might vary across different geographical exposures due to the different taxation rate on dividends which vary substantially across nations.

The loss of income of individual mutual funds and ETFs due to taxation on dividends are not reported by the funds themselves and hence an estimation is needed in order to account for the impact of the withholding taxes. Therefore, the difference between the total gross index return and the total net index return are used in order to proxy for the impact of withholding taxes, which is consistent with the method used by Blitz et al (2012) in their literature. In accordance to our hypothesis, taxation on dividends varies significantly across different regions. The return difference per annum between the gross index return and net index return is 63 basis points for the MSCI USA Index, 63 basis points for MSCI Europe Index, 22 basis points for MSCI Japan, 60 basis points for MSCI World Index and 34 basis points for the MSCI Emerging Markets Index.

In order to test for the explanatory power of the expense ratio and dividend taxation on the performance of the mutual funds and ETFs, I use a cross-sectional regression on the sample of funds. The funds are regressed on a number of variables which are the funds respective expense ratio, dividend taxes and also fund listings.  The results are exhibited in Table 4.

In regression 1, I test the explanatory power that the expense ratio yield on the performance of the funds. The null hypothesis is that the Beta of Expense Ratio is equal to zero and the alternative hypothesis is that Beta of Expense Ratio is not equal to zero. I note that the regression has a slope coefficient of -1404.61, with a t-statistic of -10.8, which indicates that the null hypothesis is rejected. Hence, expense ratio is highly significant. However, it is found that the adjusted R-squared is only 0.017 basis point which demonstrates that the variation of the performance of funds are only 1.7% explained by the variation of the expense ratios. In contrast to the findings from Blitz et al (2012), the result that I have found in my regression result does not indicate that the fund underperformance are caused considerably by expense ratios. I note that there are limitations in my research. The sample of funds used are selected from a vast fund population that are found from Thomson Datasream. Not all funds that track the benchmark index are included in the sample. This is due to the large amount of data with more than 7000 funds from ETFs and 5000 fund data for the traditional mutual funds. Hence, funds are chosen at random and are then check on a case by case basis to filter for funds that track the market indices.

In regression 2, I regress fund performances against dividend taxation as measured in the table 3 above. The null hypothesis is that the Beta of Dividend Taxes is equal to zero and the alternative hypothesis is that Beta of Dividend Taxes is not equal to zero. The result exhibits that the regression has a slope coefficient of 8.01 with a t-statistic of 5.88, which indicates that it is highly significant. However, the R-square is only 0.005 basis point, which indicates that the variation of fund underperformances are only 0.5% attributed to dividend taxes. This shows that the fund returns could not be entirely explained by dividend taxes.

In regression 3, I regress a multiple regression to test the explanatory power of both the expense ratio and dividend taxes on the performance of the funds. Table 4 indicates that the regression has a slope of -1726.92, which is lesser than the slope for Regression 1. The adjusted R-squared does not improve greatly, only amounts to 0.018 basis point, which indicates that the explanatory power of both the expense ratio and dividend taxes are low in explaining the fund underperformance. Deriving from the regression results, I conclude that expense ratios and dividend taxation are not the main reason for the fund underperformance in my sample.

Table 4

Regression Results

This table exhibit the results of Regression 1 until Regression 6 as in order to explain the explanatory power of expense ratios and dividend taxes on fund performances. In Regression 1, I test the explanatory power of expense ratio on fund performance in isolation, in regression 2 I regress the fund performance on dividend taxation in isolation. Regression 3 is where I regress both expense ratio and dividend taxes on fund performance and regression 4,5 and 6 are based on regression 1,2 and 3 respectively, augmented with dummy variables to include the impact of fund listings in France, Ireland and Germany.

1 2 3 4 5 6
coef t-stat coef t-stat coef t-stat coef t-stat coef t-stat coef t-stat
Intercept 843.79 14.52 -135.94 -1.92 1212.99 7.60 839.52 12.08 -140.52 -1.74 1208.85 7.36
Expense Ratio -1404.61 -10.84 -1726.92 -9.41 -1404.63 -10.84 -1726.95 -9.41
Dividend Taxation 8.01 5.88 -4.75 -2.48 8.01 5.88 -4.75 -2.48
IR 1.41 0.02 1.89 0.03 1.30 0.02
GE 15.49 0.25 15.58 0.25 15.42 0.25
FR 0.21 0.03 0.82 0.01 -0.02 -0.02
Adj Rsq 0.017 0.005  0.018 0.016 0.005 0.017

4.4 Listing and Fund Performance

Here, I examine the impact of fund listing on fund performance. I construct regressions using dummy variables to include the fund listings from Ireland (IR), France (FR) and Germany (GE). Examining the expense ratios of the funds listed in the stock exchanges in the different regions, I initially expect that funds listed in Germany to have the highest fund underperformance due to the high average fund expense ratio of 71 basis points. In comparison, expense ratios for funds listed in Ireland and France are 39 and 40 basis points respectively. Furthermore, I initially expect that the fund underperformances can also be different due to the different dividend taxation across regions. This is because certain regions allow funds to benefit from international dividend tax treaties and hence investors are able to expect better fund performances. However, I do not have any expectations regarding the relationship of fund listings with dividend taxation as funds do not disclose any dividend tax treaties activities.

In regression 4, I investigate the impact of the expense ratios on fund performances in different fund listings using the dummy variables IR, GE and FR. The main objective is to examine indirectly the relationship of fund listings and fund performance through the average expense ratios which are different across regions. The results of the regression indicate that the fund listings are insignificant with the t-statistics of 0.02, 0.25 and 0.003 for funds listed in IR, GE and FR respectively. Moreover, the adjusted R-square suggests that the fund performance is only explained 1.6% by the variations of the average expense ratio in different fund listings. This does not vary significantly from the regression done on expense ratio and fund performance in Regression specification 1. Hence, this imply that when fund listing and expense ratio are both taken into consideration, the explanatory power of different listings do not yield any impact. In regression specification 5, the results exhibit that the dividend taxations from various fund listings yield low explanatory power when regressed against fund performance. Therefore, fund listed in different geographical region has no relation with dividend taxes in terms of impact towards fund performance. To test for the explanatory of both average expense ratio and dividend taxation on fund performance with different listings, in Regression 6, I conduct a multiple regression with both variables and regress them on fund returns using dummy variables IR, GE and FR. The adjusted R-square suggests that the explanatory power of expense ratio and dividend taxes do not have a substantial change when fund listings are taken into consideration.

5. Conclusion

This paper has examined the return performance of the traditional mutual funds and exchange traded funds listed in Europe market in comparison with the performance of the benchmark index tracked. From this research, I note a substantial performance difference between the mutual funds and ETFs across different geographical market indexes tracked. However, in contrast to other studies, I found that the underperformance of the funds in my sample is not as large. The European funds underperform the benchmark indexes only by 34 basis point, which is close to the underperformance of funds listed in the US market with a value of 28 basis point as identified in Elton et al (2000)’s research.

Besides that, from this research, I find that the expense ratios do not have a substantial impact on the underperformance of the funds in my sample, which is not consistent with other studies. First of all, I observe that the mutual funds and ETFs underperform the benchmark index with amounts different from the reported fund expense ratios. Furthermore, I observe that there are considerable differences in fund performances across different benchmark indexes tracked albeit having similar fund expense ratios. Hence, this shows that the fund underperformance is not explained entirely by fund expenses.

Moreover, I also discover that the fund underperformance are not entirely explained when dividend taxation is taken into account. The results suggest that dividend taxation does not have a strong explanatory power on fund performances. The explanatory power of dividend taxes on the variation of fund performance is only one quarter of the explanatory power of fund expenses. Even so, the fund expense ratio is outstandingly low. The expense ratios only contribute 1.7 percent in explaining the performance of the fund in my sample whereas the dividend taxes only contribute 0.5 percent of explanatory power to the fund performance. However, I acknowledge the limitations of my research. The sample of funds that I have constructed might have excluded some funds which are relevant to this research. Thus, this might have given an impact towards the findings of my study.

Lastly, my results also indicate that fund listings do not have any impact towards the variation of fund performances. This empirical finding is consistent with the results found in Blitz et al (2012) literature. In addition, when taking into account the fund listings with expense ratio and dividend taxation, the explanatory power of the fund listing seems to disappear entirely.

My study has a few implications which might be important towards the research of mutual funds and ETFs listed in Europe. The results that I have found in my study contributed to the existing ETF literature by providing recent evidence on the relationship between fund performance, expenses and dividend taxes. I observe that there is a negative relationship between fund expense ratios and fund performance which is consistent with other studies. Besides that, the total expense ratio in my sample is lower, with a median of only 30 basis point as compared to the median found by Blitz et al (2012) in their research which is 59 basis points. This might be due to the increased demand for a more cost efficient ETFs. According to Deloitte’s Growth of ETF in Europe (2017) report, the recent years have seen changes in the ETF universe that is a material repricing of ETFs in order to create ETFs with the lowest expense ratio possible. Apart from that, in terms of the relationship between fund performance and dividend taxation, it is firstly vital to note that the ETFs listed in Europe are subjected to more taxation on dividend when compared to other regions. According to the dividend taxation reported, the European ETFs has the highest dividend taxation with an amount of 63 basis point, similar to the US. The dividend taxation is higher than for funds that tracked MSCI Japan, MSCI World and MSCI EM. Hence, it is expected that the fund underperformance would be attributed substantially to dividend taxation. However, the findings in my result showed otherwise. Hence, it might be beneficial for further researches to investigate the impact of dividend taxation and time variation on mutual fund returns in order to examine further whether dividend taxation has a substantial explanatory power on fund performance across different period.

Finally, it is noteworthy that this paper is an effort to provide research on the ETFs in the European markets. The ETF is a market that is constantly evolving to accommodate the demand from various investors. There are many ETFs created with different features in order to enhance the adaptability of the funds. In future work, it might be beneficial that the European ETF is explored further to determine the issues associated with the funds. Besides that, an empirical study on the ETFs in emerging markets and other regions would be extremely beneficial in providing comparisons towards the existing ETFs in the ETF universe.

6. Reference List

1. Blitz, D., Huij, J. and Swinkels, L., 2012. The Performance of European Index Funds and Exchange‐Traded Funds. European Financial Management18(4), pp.649-662. https://doi.org/10.1111/j.1468-036X.2010.00550.x

2. Gallagher, D.R. and Segara, R., 2005. The performance and trading characteristics of exchange-traded funds. Journal of Investment Strategy1(1), pp.47-58. https://www.fsadvice.com.au/media/library/CPD/PDFs/Journals/Investment%20Strategy/Volume%201/Number%202/JIS-v1n02-05_gallagher.pdf

3. Elton, E.J., Gruber, M.J. and Busse, J.A., 2004. Are investors rational? Choices among index funds. the Journal of Finance59(1), pp.261-288. https://doi.org/10.1111/j.1540-6261.2004.00633.x

4. Elton, E.J., Gruber, M.J., Comer, G. and Li, K., 2011. Spiders: Where are the bugs?. In Investments And Portfolio Performance (pp. 235-254). https://doi.org/10.1142/9789814335409_0012

5. Mussavian, M. and Hirsch, J., 2002. European exchange-traded funds: An overview. Journal of Alternative Investments5(2), pp.63-77. https://doi.org/10.2469/dig.v33.n2.1278

6. Svetina, M. and Wahal, S., 2008. Exchange traded funds: Performance and competition. https://poseidon01.ssrn.com/delivery.php?ID=687101067008100018025025000077086090096055058047023043031089013109086104029088016121061053000000033097052119077126027023097064009094010045014114127012127118065025095087041001100031020121090121086085109122091025119117001027111101071093015096102089127&EXT=pdf

7. Tse, Y., 2015. Momentum strategies with stock index exchange-traded funds. The North American Journal of Economics and Finance33, pp.134-148. https://doi.org/10.1016/j.najef.2015.04.003

8.Poterba, J.M. and Shoven, J.B., 2002. Exchange-traded funds: A new investment option for taxable investors. American Economic Review92(2), pp.422-427. http://0-www.jstor.org.serlib0.essex.ac.uk/stable/pdf/3083444.pdf?refreqid=excelsior%3A1cd201915d705921bf1bdf5c4c9598da

9. Milonas, N.T. and Rompotis, G.G., 2006. Investigating European ETFs: The case of the Swiss exchange traded funds. In Conference of HFAA in Thessaloniki, Greece. https://www.researchgate.net/profile/Nikolaos_Milonas/publication/228350796_Investigating_European_ETFs_The_Case_of_the_Swiss_Exchange_Traded_Funds/links/004635314d090ef655000000/Investigating-European-ETFs-The-Case-of-the-Swiss-Exchange-Traded-Funds.pdf

10. Blitz, D. and Huij, J., 2012. Evaluating the performance of global emerging markets equity exchange-traded funds. Emerging markets review13(2), pp.149-158. https://doi.org/10.1016/j.ememar.2012.01.004

11. Prasanna, K., 2012. Performance of exchange-traded funds in india. International Journal of Business and Management7(23), p.122. http://dx.doi.org/10.5539/ijbm.v7n23p122

12. Strydom, B., Charteris, A. and McCullough, K., 2015. The relative tracking ability of South African exchange traded funds and index funds. Investment Analysts Journal44(2), pp.117-133. https://doi.org/10.1080/10293523.2014.994454

13. Meric, I., Dunne, K., McCall, C.W. and Meric, G., 2010. Performance of exchange-traded sector index funds in the October 9, 2007-March 9, 2009 bear market. Journal of Finance and Accountancy3, p.1. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.610.852&rep=rep1&type=pdf

14. Kostovetsky, L., 2003. Index mutual funds and exchange-traded funds. Journal of Portfolio Management29(4), pp.80-92. http://www.simon.rochester.edu/faculty_research_pdf/leonard.kostovetsky/intellcont/ETF%20PAPER-1.pdf

15. Agapova, A., 2011. Conventional mutual index funds versus exchange-traded funds. Journal of Financial Markets14(2), pp.323-343. https://doi.org/10.1016/j.finmar.2010.10.005

16. Harper, J.T., Madura, J. and Schnusenberg, O., 2006. Performance comparison between exchange-traded funds and closed-end country funds. Journal of International Financial Markets, Institutions and Money16(2), pp.104-122. https://doi.org/10.1016/j.intfin.2004.12.006

17. Ernst & Young. 2017. “Global ETF Research 2017: Reshaping Around the Investor.” Accessed March 30, 2018. http://www.ey.com/gl/en/industries/financial-services/asset-management/ey-global-etf-survey-2017

18. PWC. 2016. “ETFs: A Roadmap to Growth”. Accessed March 30, 2018. https://www.pwc.com/gx/en/asset-management/publications/pdfs/etfs-a-roadmap-to-growth.pdf

19. MSCI. 2018. “MSCI Index Calculation Methodology: Index Calculation Methodology for the MSCI Equity Indexes”. Accessed March 30, 2018. https://www.msci.com/eqb/methodology/meth_docs/MSCI_January2018_IndexCalcMethodology.pdf

20. Deloitte. 2017. “The Growth of ETFs in Europe”. Accessed March 30, 2018.  https://www2.deloitte.com/content/dam/Deloitte/lu/Documents/financial-services/performancemagazine/articles/lu_growth-etf-in-europe-012017.pdf

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this dissertation and no longer wish to have your work published on the UKDiss.com website then please: