Examining Exchange Traded Funds

CHAPTER I INTRODUCTION

In the past few decades many performance evaluation studies indicated that actively managed mutual funds, which seek to obtain excess returns than the market by actively forecasting returns on individual stocks, do not actually obtain statistically significant excess returns (Jensen, 1968; Grinblatt and Titman, 1989; Elton et al., 1993; Malkiel, 1995; Gruber, 1996; Carhart, 1997; Edelen, 1999). This was consistent with the ‘Efficient Market Hypothesis’ which suggests that due to the availability of all kinds of information, obtaining excess returns should be difficult in a competitive market. These researches suggested a superior investment strategy: the index fund. Instead of actively engaging in stock picking, an index fund passively replicates the risks and returns of an underlying market index by investing in the securities constituting the index in the same weightage as represented in the index.

Since the first index fund launched in early 1970s, investors all over the world have discovered that there are substantial benefits from utilizing index funds as an alternative to actively managed funds. Some such benefits of indexing includes lower management expense ratio (due to passive management of fund), lower turnover and its related expenses due to buy and hold nature of most index funds and tax efficiency when compared to actively managed funds.

Like any other mutual fund, index funds may be structured as either open-ended funds or close-ended funds. Open-ended funds, as the name suggests, are open for subscription and redemption for all the investors throughout the year. However, one of the limitations of such funds is that they are priced only once a day, after the close of business. Since all the trades in such funds during the business day are executed at the closing Net Asset Value (NAV), investors are unable to react expeditiously to dramatic changes in a market during the business day.

Close-ended funds on the other hand, are open for subscription only once, and can be redeemed only on the fixed date of redemption. Moreover, in order to provide liquidity to such funds, these are listed on stock exchanges (like corporate security) and traded throughout the business day on real time basis. Though the continuous trading of such funds overcome the pricing limitation of open-ended funds, but at the same time also raises the issue of deviation between the trading price and NAV of such funds. Since the overall corpus of close-ended funds remain constant, the daily demand and supply forces often leads to significant premiums or discounts on such funds in the secondary market which is popularly known as closed-end fund puzzle[1].

Recognizing both the appeals of open-ended index funds (continuous creation and redemption of fund units) as well as of close-ended index funds (continuous trading on exchange like a stock), an innovative financial product named Exchange Traded Fund (ETF) was designed in the early 1990s, which combines the beneficial features of both these types of funds.

The study entitled “Cross-Country Analysis of Exchange Traded Funds: A Study of Performance and Trading Characteristics” aims at examining this relatively new financial product available to investors, namely, Exchange Traded Funds (ETFs).

The purpose of this chapter is to present the rationale, objectives, testable hypothesis, methodology and the organization of the study. Towards this purpose, the chapter is divided in six sections. Section I is introductory in nature, which briefly describes the nature, origin and growth of Exchange Traded Funds. Section II points out the gap in the existing literature on ETFs and elaborates upon the rationale of the present study. The specification of the objectives of the study is provided in Section III. Section IV contains the testable hypotheses of the study. The research methodology adopted, data set and the time period covered under the study are described in Section V. Finally, Section VI provides the chapter-wise layout of the study.

I

1.1 EXCHANGE TRADED FUNDS

Exchange Traded Funds (ETFs) represent a security that tracks a stock index, a commodity or a basket of assets like an index fund, but trades like a stock on an exchange. They are hybrid investment instruments which combine the beneficial characteristics of a corporate stock and an index mutual fund. Like an index fund, they provide investors with the benefit of diversification through one investment product, improved tax efficiency relative to active portfolio management, lower expenses and transparency of portfolio. And like a stock, they can be transacted in small quantities, can be short sold or bought on margin and are priced and traded continuously on a stock exchange throughout the trading day.

All these features of ETF rely on its specific dual trading system, characterized by a primary market open to authorized participants (mainly institutional investors) for the ‘in-kind’ creation and redemption of ETF shares in lots directly from the fund, and a secondary market open to all investors, where ETF shares can be traded on real time basis, with no limitation on order size.

Since an ETF is negotiated on two markets, it has two prices: the Net Asset Value (NAV) of the shares on the basis of which creation and redemption takes place in the primary market and the price in the secondary market which depends on the supply and demand for ETF shares on the exchange. If buying or selling pressure is high, these two prices may deviate from one another. However, the possibility of ‘in-kind’ creation and redemption facilitates an arbitrage mechanism which ensures that such departures are not too large. For example, if ETF shares begin to trade at a price below the NAV (i.e. at discount), arbitragers may purchase ETF shares in secondary market and after accumulating enough shares to equal a creation unit, redeem the shares from the fund and thereby acquire the underlying securities in the index, which the arbitrager may then liquidate at a profit. A similar and reverse process may apply in case of ETF trading at a premium. An effective execution of this arbitrage mechanism would thus enable the ETFs to trade at prices equal to or very close to their NAVs, thereby eliminating the problem of significant premiums or discounts often associated with closed-end mutual funds.

These innovative financial products were first introduced on the U.S. and Canadian exchanges in the early 90s. Officially, the Standard and Poor’s Depository Receipts (SPDRs) is considered to be the first ETF which was created in 1993 in order to replicate the performance of S&P 500 Index. In the first several years, ETFs represented a small fraction of the assets under management in index funds. However the launching of an ETF named cubes (or QQQ) in 1999 which follows the return of NASDAQ 100 Index was accompanied by a spectacular growth in trading volume, making ETFs the most actively traded equity securities on the U.S. stock exchanges.

While ETFs were gaining popularity in the U.S, European stock exchanges started listing their first ETFs in 2000. Following the same trend as the one observed in the U.S., exchanges began by quoting broad-based national and regional equity index ETFs. They then quickly diversified the benchmarks to a variety of underlying indices. This included ETFs based on Euro zone or European indices, emerging country indices, style (socially responsible, growth, value, small caps, mid caps, etc.) or sectors indices. Besides these equity-based ETFs, sponsors launched fixed-income ETFs, ETFs based on precious metals and lastly, on commodities. Similar growth was experienced by ETFs in other global markets as well, including the Asia Pacific and Latin American markets which started listing these instruments in the early 2000s.

The popularity of these innovative structures around the globe can be gauged from the fact that at the end of March 2012, there were over 3,169 ETFs, with 6,795 listings, having assets of $1,536.7 billion, managed by 157 providers around the globe[2].

II

2. RATIONALE OF THE STUDY

A large number of performance evaluation studies have been undertaken for actively managed funds (Elton et al. 1993, Malkiel 1995, Gruber 1996, Elton et al 1996, Carhart 1997). However, despite the significant growth in the value of assets under ETFs across the globe, empirical research evaluating the performance of this passively managed fund is quite limited. Moreover, these studies have concentrated on the U.S markets, with only a few of them focusing on the European, Australian or Asian markets. Also, these researches done on various countries have at times used different methodologies for assessing the ETF performance and no empirical study has yet compared the ETF performance across countries using uniform performance criterion.

The limited evidence on the performance of ETFs internationally, and the absence of empirical research comparing such performance across countries provides the rationale and motivation for the present study. This study will contribute to the literature by providing analysis of the performance and trading characteristics of ETFs tracking popular market indices across the globe and providing a comparison between them based on uniform performance criterion.

The work will be of direct usefulness to the investors in these markets, as it would help in assessing the extent to which ETFs deliver on their promise of exactly tracking the index. From the perspective of market makers, this work will help in assessing the pricing efficiency of ETFs in these markets, which would possibly indicate the presence or absence of profitable arbitrage opportunities in these markets. Also, a comparison of ETFs across the globe can provide useful indications to the regulators and policy makers of these markets regarding the competitiveness of their products and the areas of strength and weaknesses so as to plan for any corrective policy actions in respect of their markets, if required.

III

3. OBJECTIVES OF THE STUDY

The first objective of the study is to empirically examine the performance of Exchange Traded Funds (ETFs) across the globe. To examine the performance of ETFs, the study analyses them in terms of their risks and returns, replication strategy, tracking ability and performance effectiveness.

The second objective of the study is to provide a comparison of performance of ETFs across various countries.

Finally, the third major area that this study addresses to is the examination and comparison of pricing efficiency of ETFs across the globe, i.e. whether significant premiums or discounts exist in these ETF markets, and whether they persist over a number of days, which could present profitable arbitrage opportunities to the market makers.

IV

4. RESEARCH HYPOTHESES

The study attempts to test the following broad hypotheses concerning the performance and pricing efficiency of Exchange Traded Funds (ETFs) across the globe. The specific null and alternate hypotheses are formulated in Chapter IV- Data and Research Methodology.

• ETFs are able to perfectly track the daily and weekly returns of their underlying indices and experience insignificant tracking errors.

• ETFs are effective in replicating the returns of their underlying indices, and do not experience significant net under-performance or out-performance over half yearly intervals.

• All ETFs across the globe experience similar tracking errors (if any), and are equally effective in their performance.

• ETF markets around the world are efficient in terms of pricing, i.e. no significant premium or discount (more than 1%) exists on any given day.

• The premium or discount in the ETF markets (if any), disappears within a day and does not persist for long.

V

5. RESEARCH METHODOLOGY

As mentioned earlier, the focus of the present study is to examine and compare the performance and trading characteristics of ETFs across the globe. This section provides a brief overview of the research methodology adopted to examine each of these issues and the data used in the study.

1. Performance of ETFs

Conventionally, the performance of mutual funds has been evaluated using three classical performance measures, namely, Sharpe ratio (1966), Treynor ratio (1965) and Jensen’s alpha (1968). However, since the underlying investment objective of passively managed ETF is different from those of actively managed mutual fund[3], these traditional performance valuation methods are not considered appropriate for analyzing the performance of ETFs.

The performance of an ETF is analyzed in terms of whether it has been able to achieve its investment objective of tracking or replicating the returns of its underlying benchmark index. In order to compare the returns of ETFs with those of their underlying indices, the present study uses the methodology adopted by Frino and Gallagher (2002), Gallagher and Segara (2004) and Rompotis (2006c). In particular, it uses the following performance measures, each of which analyses a different aspect of ETF performance.

1. Risk and Return Analysis: As a first approximation of ETF performance, the study compares the average daily and weekly risks and returns of ETFs with those of their underlying indices over half-yearly intervals. Similar risk and return characteristics of ETF and its underlying index indicates a satisfactory performance of the ETF.

2. Regression Analysis: The study uses the well known market model by regressing the returns of the ETF portfolio on the returns of the underlying market index. The beta estimate of this regression is used as an indicator of the portfolio replication strategy adopted by the ETF manager. A beta equal to unity reflects a full replication strategy, which implies that the fund invests in all the components of the underlying index in the same weightage as represented by the index.

3. Tracking Error Estimation: An important criterion for assessing a passive fund manager’s performance is through an estimation of ‘tracking error’, which quantifies the difference in returns of the fund and its underlying index. Following the methodology adopted by Frino and Gallagher (2002), the study uses three different measures of tracking error. These are:
   • the average of the absolute difference in returns between the fund and the index,
   • the standard deviation of the return difference between the fund and the index, and
   • the standard error of the regression of fund returns on underlying index returns.

4. Performance Effectiveness: The study then measures the performance effectiveness (or bias), i.e. the net under-performance or out-performance of ETFs over a period of time, using the following two measures:
   • the average of the arithmetic difference in returns between the fund and the index, and
   • the alpha coefficient of regression of fund returns on underlying index returns.

2. Cross-Country Comparison of Performance of ETFs

The study attempts to make a cross-country comparison and provide relative ranking of countries in terms of their ETF performance using three of the performance measures, namely, replication strategy, tracking ability and performance effectiveness.

3. Trading Characteristics of ETFs

In order to analyze the trading characteristics of ETFs, the study firstly tests their pricing efficiency, i.e. whether they experience significant premiums or discounts during any given day, and how much time does it take for such premium/discount (if any) to disappear. Towards this purpose the study quantifies the daily deviations between ETF’s trading price and NAV (Net Asset Value) in monetary as well as percentage terms, and reports their frequency distribution. The persistence of such deviations is then examined using a regression model. The study also provides a trading profile of ETFs to analyze other trading characteristics of ETFs, such as their average trading turnover.

4. Data and Period of Study

ETFs were first introduced in the U.S markets in 1993, however they picked up pace only after 1999. Since then ETFs have continued to grow and have proliferated across global financial markets both in terms of their number and assets under management. By the end of first quarter 2012, there were over 3,169 ETFs, with assets worth $1,536.7 billion, managed by 157 providers around the globe. (ETF Landscape, Q1 2012)

Since the present study aims at analyzing the ETFs across the globe, and the study of the whole universe of ETFs is not viable, there is a need to select a representative sample for the purpose of our study. Our study therefore analyses all the domestic ETFs that track the popular indices of five dominant countries namely U.S., U.K., Japan, Australia and India. These indices are the broad-based equity indices which represent the performance of the respective markets. In all we study 11 popular indices and 17 ETFs that track such indices.

The time period under study extends from January 1993- March 2012. Each selected ETF has been analyzed over a time period beginning from the first full financial year of its trading till the end of the period under study. The study uses daily as well as weekly data in respect of closing prices and NAVs of ETFs as well as the closing values of the underlying indices.

VI

6. ORGANISATION OF THE STUDY

The study is organized into eight chapters including the present one. Chapter II provides a conceptual framework of ETFs by describing its nature, trading mechanism, benefits and costs as well as its history, growth and development. Chapter III reviews major research studies concerning the performance and trading characteristics of ETFs and provides a framework for the present study. The data used and the details about the methodology adopted in this study are provided in chapter IV. Chapter V, VI and VII are devoted to the empirical findings of the study whereby Chapter V discusses the performance of ETFs, Chapter VI provides a cross-country comparison of ETF performance and Chapter VII discusses the trading characteristics of ETFs under study. Finally, chapter VIII presents the summary, conclusions as well as the implications of the study.

---

[1] For a review on the closed-end fund puzzle, see Dimson and Minio-Kozerski (1999)

[2] ETF Landscape Industry Highlights Q1 2012 from Blackrock

[3] An actively managed mutual fund aims at beating the market by providing superior return than the market indices, whereas, a passively managed index fund or ETF aims at replicating the returns of the underlying market index by joining the market, instead of attempting to beat it.

Cite this page