Risk-Reward of Exchange Traded Funds: A Study of Canadian ETF's by Carl Overennay Diploma of Business Administration, Selkirk College, 1990 Diploma of Technology, British Columbia Institute of Technology, 1997 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF BUSINESS ADMINSTRA TION THE UNIVERISTY OF NORTHERN BRTISH COLUMBIA April2010 ©Carl Overennay, 2010 UNNERSITY of NORTHERN BRITISH COLUMBIA LIBRARY Prince George, s.c. Abstract Exchange Traded funds are the fastest growing investment product in capital markets today. Total funds deployed in ETFs are approaching US$700 billion globally which is nearly 15 per cent of $4.5 trillion held in traditional equity mutual funds . This study examines the risk-reward of 74 Canadian ETFs across three major sponsors representing nearly $29 Billion in assets under management for the last 5 years (out of which, the last two years - 2007 and 2008- witnessed a general decline of the Canadian stock market). The study found the best performing ETFs were international funds especially emerging market funds which witnessed high growth rates in the last decade (especially BRIC countries). Currency-hedged funds also performed relatively better reflecting the role of exchange rates in impacting the returns of cross-border investments. Commodity ETFs generally had shown mixed results: the bear commodity ETFs (reflecting the macro-economic performance) generally did well as compared to bull commodity ETFs. In terms of risk-reward, the results are somewhat different. International and emerging market ETFs performed well in terms of positive and high alphas (excess returns) but also had displayed relatively high risk. In terms of risk-reward, the Canadian and US broad equity ETFs performed well (Treynor ratio of 0.11) while fixed income ETFs had the lowest Treynor ratio (-2.05). In terms of ranking, the currency-hedged ETFs performed relatively better than Canadian sector ETFs. The international and emerging market funds while displaying positive Treynor ratios (risk-reward) were the ETFs with relatively modest performance. Ill !TABLE OF CONTENTS Chapter 1 Chapter 2 Section 2.1 Section 2.2 Section 2.3 Section 2.4 Section 2.4 I Chapter 3 Approval Abstract Table of Contents List of Tables List of Charts Investment Terminology Acknowledgements Introduction The Evolution of ETFs 11 Ill lV v Vl Vll Xl 1 4 Definition of Exchange Traded Fund Passive Management Lower Management Expense Ratio Liquidity Tax Advantages Transparency ETF: The Creation Process ETF vs. Mutual Fund vs. Index Fund Growth of World Wide ETF Industry The Canadian Experiment 4 5 6 7 8 9 10 12 14 19 Section 3.1 Section 3.2 Section 3.3 Section 3.4 Section 3.5 Section 3.6 Section 3.7 Risks and the Reward Relationship Measures of Systematic Risk Risk Preferences Optimum Portfolios Capital Asset Pricing Model (CAPM) Fama-French Three Factor Model Risk-Reward of Stocks, Mutual, Index Funds and ETFs Literature Review: Risk - Reward 21 1 Chapter 4 Data Sources and Methodology 41 Section 4.1 Section 4.2 Database Methodological Framework 41 43 Chapter 5 Market Risk of Canadian ETFs Empirical Results 44 Fixed Income Canadian and US Equity International and Emerging Markets Commodities Gold and Precious Metals Alternative Investments Currency Hedged 44 46 51 52 55 57 59 Risk - Reward of ETFs in Canada Empirical Results 61 Section 5.1 Section 5.2 Section 5.3 Section 5.4 Section 5.5 Section 5.6 Section 5.7 Chapter 6 Section 6.1 Section 6.2 Section 6.3 The Analytical and Empirical Framework Excess Retums (Alpha) of Canadian ETFs Risk-Reward of Canadian ETFs - Overall Performance lV 22 22 25 27 31 37 39 61 63 66 Section 6.4 I Chapter 7 Conclusions 67 Conclusions of ETF Study jTABLES/CHARTS/FIGURES Figure 2.1 Creation and Redemption Simplified 11 Figure 2.2 Process of Creation ofETF Units 12 Figure 3.1 Risk Classifications 26 Figure 3.2 The Risk-Reward Pyramid 27 Figure 3.3 Markowitz Efficient Frontier Model 29 Figure 3.4 Security Market Line 33 Figure 3.5 Fund Investment Style and Size of Companies 38 Table 2.1 Net Issuance of ETF Shares 16 Table 3.1 Global Equity Risk Premium 35 Table 5.1 Fixed Income Risk: R2 46 Table 5.2.B Canadian and US Equity R2 48 Table 5.2.D Canadian and US Equity R2 50 Table 5.3 International and Emerging Markets R2 52 Table 5.4.8 Commodities R2 56 Table 5.4.D Commodities R2 55 Table 5.5 Gold and Precious Metals R2 58 Table 5.6.B Alternative Investments R2 58 Table 5.6.D Alternative Investments R2 59 Table 5.7 Currency R2 60 Table 6.1 Basic Assumptions underlying CAPM Model 62 Table 6.2 Summary of Expected Returns, Actual Returns and Excess Returns 63 v Table 6.3 Commodity ETF's Performance 65 Table 6.4 Treynor Ratio Rankings of Canadian ETFs 66 Chart 2.3 Progress of Exchange Traded Funds - 1998-2008 15 Chart 2.4 ETF New Equity Issue - Year End 2008 I7 Chart 2.5 Growth of ETF's 18 Chart 2.6 Growth Curve of ETFs Canada 20 Chart 5.I Fixed Income Risk: B 45 Chart 5.2.A Canadian and US Equity B 48 Chart 5.2.C Canadian and US Equity Risk B 49 Chart 5.3 International and Emerging Markets B 51 Chart 5.4.A Commodities: Positive B 56 Chart 5.4.C Commodities: Negative B 54 Chart 5.5. Gold and Precious Metals B 58 Chart 5.6.A Alternative Investments Positive B 57 Chart 5.6.C Alternative Investments Negative B 58 Chart 5.7 Currency B 60 Appendix Tables Appendix 4.I Fund Company Sponsor and Inception Date 72 Appendix 5.1 Fund Company Trading Symbol 75 Appendix 6.I Performance Analysis by Asset Class 77 Appendix 6.2 Treynor Ratio of Various Canadian ETFs 8I References & Internet Sources 85 1 VI Investment Terminology SOURCE: THE ETF BOOK, RICHARD FERRI AND DICTIONARY OF FINANCE AND INVESTMENT TERMS, GOODMAN. SEVENTH EDITION, 2006 JOHN DOWNES, JORDAN Active Management An investment management strategy where the portfolio manager actively makes investment decisions and initiates buying and selling of securities in an effort to maximize a return. Alpha A measure of performance in percentage above or below what would have been predicted by risk as suggested by its beta. Positive Alpha means a fund performed greater than its risk would suggest, while negative means it underperformed. Arbitrage Mechanism At any point in the trading day, authorized participants can come to the fund manager with a basket of the underlying securities given in the published holdings, which the fund will then exchange for a creation unit consisting of a set number of shares in the ETF. Authorized Participant (AP) An institutional investor that is authorized to buy and sell ETF creation units directly with a fund company Benchmark Index An index that correlates with a fund, used to measure a fund manager's performance. Beta A measure of the magnitude of a portfolio's past share-price fluctuations in relation to the ups and downs of the overall market. The market has a beta of 1.0. So if a portfolio with a beta of 1.2 would have seen its share price rise or fall by 12% when the overall market rose or fell by 10% Bid-Ask Spread The difference between what a buyer is willing to bid (pay) for a security and the seller's ask (offer) price. Vll Creation Unit A set of shares or securities that makes up one unit of the fund held by the trust that underlies an Exchange Traded Fund (ETF). One creation unit is the denomination of underlying assets that can be redeemed for a certain number of shares Capital Asset Pricing Model (CAPM) Models the relationships between expected risk and expected return. The model is grounded in the theory that investors demand higher returns for higher risks. It says that the return on an asset or a security is equal to the risk-free return - such as the return on a short-term treasury security- plus a risk premium. Efficient Market The theory, disputed by some experts, that stock prices reflect all market information that is known by all investors. Also states that investors cannot beat the market because it is impossible to determine future stock prices. Exchange Traded Fund (ETF) An ETF is an index fund that trades on the stock market. A common ETF is the Standard & Poor's Depositary Receipts (SPY), which tracks the S&P 500. A Canadian equivalent is XSP by iShares. Expense Ratio The percentage of a portfolio's average net assets used to pay its annual expenses. The expense ratio, which includes management fees, administrative fees and directly reduces return to investors. Indexing An investment strategy to match the average performance of a market or group of stocks. Usually this is accomplished by buying a small amount of each stock in a market. Index Providers Companies that construct and maintain stock and bond indexes. The main providers are S&P, Dow Jones, S&P/TSX, MSCI, Russell and Wilshire. Vlll lntraday Value Ongoing estimates of the underlying value of securities and cash that makeup ETF shares and are quoted every 15 seconds by the exchange listing the ETF. Also called the Intraday Indicative Value (IIV) Management Expense Ratio (MER) The amount an ETF or Mutual Fund pay to its investment advisor for the work of overseeing the fund's holdings. Mutual Fund A Closed-end fund that has a fixed number of shares, usually listed on a major stock exchange. An Open-end fund has the ability to issue or redeem the number of shares outstanding on a daily basis. Prices are quoted once per day, at the end of the day, at the net asset value of the fund (NA V) No Load is fund that charges no sales commission or load. Real Return The actual return received on an investment after factoring in inflation. For example, if the nominal investment return for a particular period was 8 percent and inflation was 3 percent, the real return would be 5 percent. Risk A chance that invested capital will drop in value. With reference to fluctuating market values of securities and portfolios, risk means exposure to uncertainty, which is measured by standard deviation Risk Averse Refers to the assumption that, given the same return and different risk alternatives, a rational investor will seek the security offering the least risk. Risk-Free Return Yield on a risk-free investment. For example, a 3-month Treasury bill is considered to be risk-free investment because of the unlikelihood of government default. IX Risk Premium In portfolio theory, the difference between the risk-free return and the total return from a risk investment. In the CAPM, the risk premium reflects market-related risk as measured by Beta. Risk Tolerance An investor's ability or willingness to endure declines in the prices of investments while waiting for them to increase in value. Sharpe Ratio A measure of risk-adjusted return. To calculate a Sharpe Ratio, an asset's excess return (its return in excess of the return generated by risk-free assets such as Treasury Bills) is divided by the asset's standard deviation. It should be compared to an appropriate benchmark. The higher the ratio, the safer the strategy. Single Index Model (SIM) A model of stock returns that decomposes influences on returns into a systematic factor, as measured by the return on the broad market index, and firm specific factors. Unit Investment Trust (UIT) Common type of ETF that requires exact duplication of an index and prohibits derivatives in operation. Examples include ETF's. Volatility The degree of fluctuation in the value of a security, mutual fund or index. Often expressed as a mathematical measure such as standard deviation or beta. The greater a fund's volatility, the wider, the fluctuations between high and low prices. X Acknowledgement The writing of this project 1s highly dependent on the support and encouragement the author receives from others. With this in mind, I have many people to personally thank and acknowledge. First of, I would like to acknowledge the support, guidance and assistance of Dr.Ajit Dayanandan. His wise counsel and relentless editing guided me through the process on topic selection, generating a thesis idea and completing the work is greatly appreciated. I would also like to acknowledge the UNBC School of Business for their support. To my colleagues and friends; Paul and Rick, I offer my sincere thanks as I could not have made through the program with-out your help and support. And last, a special mention for my wife, Angela. Your un-wavenng support from the day I enrolled in the UNBC Master of Business Administration on through my entire academic period of study is greatly appreciated. A project and time commitment of this magnitude cannot be done successfully without the full support of loved ones. Your continuing support during my academic pursuits is greatly recognized and appreciated. I am so very thankful to you for your active support and encouragement. As this chapter in my life comes to a close, it is now time for me to reflect back on my accomplishments and to begin a new career challenge in a new location. Also, it is now time for me to support my wife as she embarks on her academic journey. I just hope I can offer the same measure of support you have shown me. XI Chapter 1 Introduction to Exchange Traded Funds Exchange Traded funds (ETFs) are the fastest growing investment products in capital markets today (Gastineau, 2001, Poterba and Shoven, 2002). ETFs are shares which closely track the performance of an index, sector, or region 1• They offer the benefits of diversification and index tracking at a low cost. The ETFs industry began in 1993 and is the fastest growing (around 20 per cent per annum) asset class in the world 2 . Total funds invested in all ETFs is approaching US$700 billion globally, representing nearly 15 per cent of $4.5 trillion held in traditional equity mutual funds (Santoli, 2009). The industry is rapidly approaching one trillion dollars of global investment and is expected to grow at double digits into the near future (Ferri, 2008). With the financial crisis in the last two years, most mutual fund investors have experienced negative returns on their portfolios, while many ETFs have recorded positive returns; on average ETFs generated a total return of seventeen per cent in 2008 (Santoli, 2009).With management expenses of ETFs well below 1 per cent (compared to 2-5 per cent for mutual funds) and with real time quotes and liquidity, ETFs have become a popular short-term instrument amongst the investing public and with large institutional investors. The number of ETFs has expanded exponentially over the past fifteen years; it is estimated by then end of 2010, there will be more 1 For an overview on ETFs, see Gastineau (2001) and Poterba and Shaven (2002). Today there are 104 ETFs in Canada, 768 USA, over 1200 Worldwide; widely popular as a short-term investment tool. 2 1 than 1000 available on U.S. Exchanges - with assets totaling well over $1 Trillion. 3 . There are currently 104 Canadian sponsored ETFs across five major sponsors in Canada representing nearly $29 Billion in assets under management4 . In Canada, ETFs trading accounted for 14 per cent of the total trading on the TSX in June 2009, representing a year over year increase of 374 per cent 5 . In a survey of investment professionals conducted in March 2008, 67 per cent of respondents called ETFs the most innovative investment vehicle of the last two decades, and 60 per cent reported ETFs have fundamentally changed the way they construct investment portfolios 6 • The risk and return of financial assets like stocks and mutual funds are adequately researched (Fama and French, 2004, Perold, 2004). There are very few studies on the risk-reward relationship of ETFs except for Agrrawal and Clark (2007) which is centered on US-based ETFs. The present study contributes to the literature in this field by examining the risk-reward relationships of publicly listed ETFs in Canada, which has not been attempted earlier. The main aim of the present study is as follows: (i) (ii) (iii) To the estimate the risk of various ETFs. To estimate the return and excess return (alpha) ofETFs. To rank the Canadian ETFs in terms of risk-reward (popularly known as the Treynor Ratio) The study uses the popular analytical framework of asset pricing, viz., the capital asset pricing model (CAPM) of Sharpe (1964) and Littner (1965) 7 . The 3 Ferri, Richard, (2008) Advisors News I lndus t1y news I ADVISORS - ETFs continue to gather steam 5 TSX ETFS listings surpass l 00- Economy & markets- News Investment Executive 6 State Street Global Advisors and Knowledge @ Wharton "The Impact of Exchange Traded Products on the Financial Industry" June 10, 2008 7 For an overview of the CAPM model, see Fama and French (2004) and Perold (2004) . 4 2 study employs the econometric methods like ordinary least square (OLS) method for empirical investigation and is based on the data of 74 Canadian ETFs for the last 5 years (2004 to 2009). The study is organized as follows : Chapter 1 provides an introduction to Exchange Traded Funds. Chapter 2 discusses the global evolution of ETFs and in particular, the evolution of ETFs in Canada. Chapter 3 provides a brief review of literature on the risk-reward relationship. Chapter 4 discusses the data sources and methodology. Chapter 5, we present the empirical results of risk as measured by beta. Chapter 6 we present the empirical results of excess returns as measured by alpha and ranks the risk of ETFs by the Treynor ratio. Chapter 7 summarizes the conclusions ofthe study. 3 Chapter 2 The Evolution of Exchange Traded Funds This chapter provides an overvtew of the evolution of Exchange Traded Funds (ETFs). The chapter is organized as follows: Section 2.1 defines and discusses the pros and cons of ETFs. Section 2.2 summarizes the creation process. Section 2.3 discusses the differences between ETFs, Mutual Funds and Index Funds. Section 2.4 discusses the world-wide growth of the industry and in particular, the growth of ETFs in Canada. Section 2.1: Definition of an Exchange Traded Fund ETFs are a passively managed basket of financial assets which includes stocks, bonds or currencies that are traded throughout the day on a stock exchange in the same manner an individual stock is traded. ETFs offer the conveniences of diversification found in a mutual fund and the flexibility to trade like an individual stock 8 . ETFs are priced continually on a market exchange and fluctuate in price according to market conditions of supply and demand. The objective of ETFs portfolio managers is to match the performance of an underlying index before fees and expenses. In addition, they possess the characteristics of diversification just like a benchmark index such as the S&P 500. According to Ferri (2008), ETFs are account structures, not investment strategies 9 . These structures are only limited to one's imagination; for example, there 8 9 Exchange Traded Fund, Street Authority.com. http ://www.streetauthority.com/terms/e/ETFs/.asp Ferri (2008) 4 are ETFs which move contrary to the market (called inverse ETFs) and thus mimic some the extreme (high risk) investment strategies witnessed in the market (like shorting). Investment companies' could design ETFs according to the type of fund, investment style and/or investment strategy. The investment return is subject only to the benchmark index it follows. Generally ETFs offer a passively managed investment strategy and usually have lower management fees, more liquidity, better tax advantages and complete transparency. These fundamental features will be discussed in greater detail for the remainder of this chapter. Passive Management Unlike actively managed funds which seek to generate a higher return then their benchmark, the main objective of all ETFs is to replicate the return of the benchmark index they were created to follow. For example, symbol XIU (iShares Canada) is the simplest case of passive management in Canada. It is designed to replicate a well-defined index of common stock, in this case the TSX 60. The portfolio manager buys each stock comprising the TSX 60 in exactly the same proportion it represents on the TSX 60 index. As such, the role of portfolio managers is to maintain the correct weighting rather then actively trying to create value by trading and selecting and de-selecting stocks. Passive management is a key strategy as ETFs offer the investor less reason to "shop around" for a variety of broad based funds. A few well-chosen ETFs can provide the average investor ample diversification. 5 Replication is the simplest technique used to construct passive ETFs. However, sometimes managers of ETFs have to make decisions in replicating the index. This is known as "tracking error" and can lead to increased transaction costs. In the case of the TSX 60, the manager has to conclude if it is in the best interest to replicate all sixty companies or exclude some of the smaller companies in order to reduce transaction costs. For index funds, cash kept in the fund for redemptions and dividends distribution are also key considerations for the manager ofETFs. Lower Management Expense Ratio (MER) Generally, ETFs have a lower MER than mutual funds and this is attributed to: (1) the passive management strategy adopted by ETFs portfolio managers which reduces costs associated with buying and trading securities in an attempt to beat the benchmark index and (2) lower marketing and accounting fees. A study by Deloitte (2007) 10 concluded the average ETFs had an MER of0.41 per cent compared to the average mutual fund, which had an MER of between 1-3 per cent. The lower fees results in more of the investor's money going towards the purchase of ETFs rather then administrative costs. Conversely the higher the fees, the lower the overall return the investor receives. A small change in a fee structure can result in a significant change in an investor's portfolio over the long term. The reality of compounding interest can impact portfolio values with small changes in MER fees. There are also challenges for ETFs to keep their costs down relative to mutual or index funds . Due to their popularity, both advertising spending and licensing index fees have been steadily increasing. These two variables apply cost 10 Rongala (2009) , p.9 6 pressures, which have resulted in slight MER increases. As the industry continues to grow, these two cost factors show signs of increasing fees rather then decreasing. This will pose a challenge to the industry as it attempts to negate the effects of fee increases rather than pass the increases along to the customer. All investors dread management fees but realize they are a necessary evil of the investment industry. In today's world, to maximize returns, investors are seeking out low fee investment products. ETFs offer an alternative to the investor trying to minimize the impact of expense fees. Liquidity Liquidity is an important consideration when selecting an investment product. As mentioned earlier, ETFs trade like individual stocks on an exchange. They are traded throughout the day in the secondary market. The average daily trading volume reflects the natural liquidity of ETF shares trading on the open market. ETF shares of total US Equity dollar volume increased from 24 percent in July 2007 to 34 percent in July 2008 11 • Eight of the ten most actively traded US Equities by dollar value were ETFs in July of 2008 12 • ETFs growth has been rapid. Liquidity is far from a concern as ETFs are now commonly traded on many exchanges worldwide. With Authorized Participant's (AP's) managing the creation and redemption process along with the secondary market trading and arbitrage pricing mechanisms on the stock exchanges, make 11 12 NYSE Area, Faetset & BGI, July 2008 NYSE Area, Faetset & BGI, July 2008 7 liquidity a non-issue. These reasons explain why liquidity is considered a major strength of the ETFs structure. Tax Advantages In a mutual fund, redemptions can have an adverse tax impact for investors. When a mutual fund manager sells securities to raise cash for fund redemptions, these sales generate capital gains. In an ETF structure, the shares in-kind redemption mechanism generally does not lead to any capital gains. Any capital gains tax on ETFs can be delayed until the ETFs are finally sold. Mutual funds on the other hand are quite different. Mutual funds generate unrealized tax gains for stocks, which have gone up in value. Once these stocks are sold, the fund calculates the tax liability to the owner in the proportion of how many fund units were held. ETFs are structured differently for tax purposes. Through a regulatory loophole, ETFs are created by trading an equivalent certificate (the ETFs for the many stocks that make up the basket) in what is called an in-kind trade. This exchange of essentially identical items does not trigger capital gains according to the IRS. Traditional mutual funds must go into the open market and exchange cash for stocks and vice versa, which trigger realization of gains. It's a subtle difference but which results in an advantage for ETFs investors 13 • In summary, the best way to look at the tax benefits of ETFs is to regard them as a stock trade. ETFs often have nontaxable trades of ETFs shares for underlying stock and vice versa, traditional mutual funds generally have sales events, which trigger tax consequences 14 • 13 14 Bloomberg, from May, 2002 issue of Financial Planning Magazine, Wiandt, (2002) http://www.indexfunds.com/articles/20010928_ETFstax_adv_ETFs_JW.htm 8 Transparency Stocks held within ETFs are published daily. By law, ETFs sponsors must disclose to the public the securities holdings and their weights in the index while active open-end mutual fund holdings are only disclosed periodically, usually several week or months later 15 • As shares trade throughout the day, this disclosure is necessary for the shares to be correctly priced. Mutual funds on the other hand, do not disclose their holdings on a daily basis. This allows investors to trade ETFs in the market at known prices. This simplifies the process for the investor. When an investor begins to build a portfolio, choosing the right asset allocation strategy or diversification strategy becomes easier when the investor knows exactly what index or what sector they are investing in. This in direct contrast to an actively managed mutual fund, where the fund manager may on a regular basis may decide to adjust the fund by selecting or de-selecting stocks. If the adjustments were made public or transparent immediately, the fund manager's strategy would be exposed and may even be exploited or copied by other competitors trying to match the return performance of the fund being actively managed. In summary, ETFs transparency includes no hidden fees or costs. This signals to the buyer you getting exactly what you pay for. Secondly, ETFs holdings are published daily for anyone to review, usually on the issuing fund company's website. The market prices are set in an open fair price based on bid and ask spreads. Lastly, owning ETFs indicates owning a proportion of the underlying stocks making up the ETFs. 15 Ferri (2008) 9 Section 2.2: ETFS - The Creation Process All ETFs are publically traded securities. Gastineau (2001 , 2002) provides extensive literature on how ETFs are created. According to Gastineau (200 l , 2002), each ETFs share is a claim on a trust that holds a specified pool of assets. ETFs shares are created when an authorized participant (AP) deposits publically traded shares with a trustee and in return receives ETFs shares in return known as creation units. These are usually created in units of 50,000 and then are broken up and distributed amongst individual investors (Ferri 2008). The secondary market is where ETFs are traded amongst individual investors. The price of any ETFs is nearly in-line within the stated index. The Net Asset Value (NA V) of the underlying basket of stocks primarily determines the price of ETFs. During the trading day, market forces of supply and demand can have a discount or premium effect on their price as compared to their NAV. If a discount or premium is identified, new ETFs shares can be created to meet the additional demand, thus bringing the price back closer the NA V. In addition, if an arbitrage opportunity is created, usually large institutional investors will step in and arbitrage the price of the ETFs closer to the NA V. It is the primary market which sets ETFs apart (Demaine, 2002) 16 • In this market, ETFs are different from any other investment vehicle. New units can be either created or redeemed through an ordinary exchange of constituent shares that make up an index, and this also ensures an organized supply of shares to the market. 16 Demaine (2002) 10 In Figure 2.1 and Figure 2.2, Gary Gastineau (2003) illustrates the creation/redemption process. 17 Creation and Redemption Simplified Creation Portfolio Securities ETF Shares Redemption Authorized Participants ETF Shares ETF Portfolio Securities Everything Priced Consistent \Nith Net Asset Value Each Day Source: Ferri, 2008, p 5 "Specific procedures allow index baskets, in the same shape as the funds existing assets, to be received by or delivered from the fund. This dissipates any imbalance in the supply or demand for the fund's units. Furthermore, since the creation/redemption process offers arbitrage potential (between cash values of the components and the cash values of the ETFs in the secondary market) the AP is motivated to address any premiums/discounts as part of their market-making activity" 18 . 17 18 Gastineau Gary L(2003), http://www.hofstra.edu/pdflbiz_ mlc_gastineau.pdf Demaine (2002), p. 354-366 11 Figure 2.2: Process of Creation of ETFS Units Buyers & Se,llers .. ..... ... . ETF Shares ...... .. Securit:ies Cash ETF Cr-ea.u o n Units Gash tl Capita l IVI.arkets C:reat:ion Basket: Se<:urit:Ji es So u rce : F ir st Tr ust: Excha nge- Traded FurKjs _ - A Tra d e d Fun ds" Guide o E xch a n ge - Section 2.3: ETFs vs. Mutual Fund vs. Index Fund As discussed earlier, ETFs are commonly traded on a stock exchange. They can be bought through a margin account 19 and be sold short 20 if the investor chooses. In contrast, Mutual funds can only be acquired directly through the fund company. Secondly, ETFs vary from mutual funds in terms of pricing: ETFs are actively traded on a stock exchange; therefore, they are continuously being priced every fifteen seconds with a bid and ask price spread21 through economic forces of supply and demand. In contrast, Mutual Funds can only be purchased or redeemed at the end of the trading day. Mutual funds pricing is pre-determined or forward priced as investors can place orders to buy or sell shares throughout the day and then will 19 Margin Account: Buying securities with borrowed money usually from a broker and using the securities purchased as collateral. By utilizing this leverage strategy, the investor can magnify the effects of returns. 20 Selling Short: Technique used by investors whereby they "borrow" securities from a brokerage firm and sell them believing the value of the securities is about to drop in. Then they wait to buy them back at a lower price and return to the broker. The sale at higher price less the purchase at the lower price is the profit or return. 21 Bid or Ask Price Spread: Bid price is the maximum price a buyer is willing to pay for a security and the Ask is the minimum a seller is willing to sell the security. The difference between the two is the price spread. 12 receive the price when the Net Asset Value (NAY) is calculated at day's end. At the end of each trading day, the NA V of each Mutual fund is calculated and buy and sell transactions occur based on the underlying NAY. Consequently, the price at which investors buy and sell shares may not always equal the NAY of the stocks in the ETFs basket. As well, two investors selling the same ETFs shares at different times throughout the day may receive different prices for the same shares, both of which may differ from the ETFs NAY. ETFs have an arbitrage mechanism 22 , which controls any price discrepancy and keeps the market price close to its NA V. The arbitrage is permitted by a select few of intuitional investors called Authorize Participants (AP). When a pricing discrepancy occurs, AP's are allowed to buy or redeem ETFs shares. The arbitrage trade then allows AP's to exchange individual stocks for ETFs shares or the reverse and the shares can be turned in for individual stocks. An Index Fund is another form of a basket of financial assets. Index funds track an underlying index and are also considered a passive investment. Tracking is usually achieved by statistically sampling an equal proportion of the financial assets, which comprise the underlying index. In the process, an index funds achieve a diversification of portfolios as embedded in the underlying index. As a result of the passive investment strategy, index funds usually charge lower management fees. John Bogle of the Vanguard Group created the first index fund in 1975. The original name of the fund, First Index Investment Trust was created to replicate the 22 Arbitrage Mechanism : At any point in the trading day, authorized participants can come to the fund with a basket of the underlying securities given in the published holdings, which the fund will exchange for a creation unit consisting of a set number of shares in the ETFS 13 S&P 500 index. Later renamed the Vanguard 500 Index Fund, "it started with comparatively meager assets of $11 million but crossed the $100 billion milestone in November 1999; this astonishing increase was funded by the market's increasing willingness to invest in such a product" 23 • Index funds are priced during normal business days and are similar to the pricing of ETFs. Index funds tend to have slightly lower expense ratios than ETFs. Normally index funds do not generate a return greater than the returns of the underlying index it follows . When tracking errors and fees are a factored in, an index fund will normally generate a return slightly lower then the return of the benchmark index. The main fundamental difference is an index fund is priced at the end of the business day based on its Net Asset Value (NA V) and ETFs are continually priced through-out the trading day. Section 2.4: Growth of the World-Wide ETFS industry ETFs are a rapidly growing investment product (Gastineau 2001, 2002). The first ETFs were created in 1993 and new ETFs have been entering the market at an exponential rate. In 2007, assets under management (AuM) rose worldwide by 41 percent to USD 796.6 bn and in Europe by twenty seven percent to EUR 89.2 bn (USD 128.4 bn) (Dieckerman 2008) (Chart 2.3). The biggest single market is the USA with AuM of USD 580.7 bn (73 percent market share), followed by Europe with AuM ofUSD 128.4 bn (16 percent market share).24 23 https: I!persona I.vanguard. com/ us/whatweo tTer/mu noa Ifundi nves tin!! 1overview? Link= Deck B (accessed on December 23, 2009) 24 Source Morgan Stanely (2008) , p.2, PriceWaterhouseCoopers.pdf 14 Chart 2.3: Progress of Exchange Traded Funds- 1998-2008 Total Assets & Number of ETF's 700 600 500 m 400 en :::J 300 ~ 200 100 0 800 Ill 600 f= w .... 400 L.. 0 Cll 200 ~ ::I 98 99 00 01 02 03 04 05 06 07 08 0 z Year Source: Investment Company Institute (ICI), p.41, http://www.icifactbook.org/ Today, attracting the most new net cash inflows are ETFs focusing in the specialty sectors of growth, fixed income and commodities indices. In 2008 there were 92 Canadian ETFs with market capitalization of over $1 billion, compared with 72 at the end of 2006 and just 10 in 2001 (Anderson, 2008). Interest in ETFs among all types of investors has continued to grow at a brisk pace (see chart 2.1 ). Furthermore, since the breakthrough of ETFs in 1993 in the USA, the global market for ETFs has experienced average annual rates of growth in AuM in the high double digits 25 • ETFs accounted for 57.6% of Net inflows into equity funds in the USA (Morgan Stanley, 2008) 26 . The demand for ETFs has accelerated as institutional investors have found ETFs to be a convenient vehicle for participating in, or hedging against, broad movements in the stock market. Retail investors and their financial advisers have 25 26 Morgan Stanley, (2008), p.3 Morgan Stanley, (2008), p.3 15 become increasingly aware of these investment vehicles. In contrast to a buy and hold strategy, investments in ETFs provide an opportunity to make short term gains (profits). According to the Investment Company Institute, an estimated 2 percent of households, or 2.3 million, owned ETFs in 2008. Households that owned mutual funds, an estimated 4 percent also owned ETFs. Assets in ETFs accounted for 5 percent of total net assets managed by investment companies at year-end 2008. Net issuance of ETFs shares continued to rise in 2008, reaching a record $177 billion. (ICI, 2009). Table 2.1: Net Issuance of Exchange Traded Funds Millions of dollars, 1999-2008 Total Investment Objective 1999 $11,929 Broadbased $10,221 2000 42,508 40,591 1,033 884 2001 31,012 26,911 2,735 1,366 2002 45,302 35,477 2,304 3,792 3,729 2003 15,810 5,737 3,587 5, 764 721 2004 56,365 29,084 6,514 15,645 3,778 $1,353 2005 56,729 16,941 6,719 23,455 6,756 2,859 2006 73,995 21,589 9,780 28,432 5,729 8,475 2007 150,617 61 , 152 18, 122 48,842 13,440 9,062 2008 177,220 88,105 30,296 25,243 23,010 10,567 Sector Bond & Hybrid Global $1,596 $112 Commodities Source: ICI Fact-book (2009) p.49, http://www. icifactbook.org/ Even when the worldwide economy slowed rapidly in 2008, net inflows into ETFs have still grown considerably over the past decade with the net issuance of ETFS shares accounting for much of this increase (ICI 2009). From year-end 1998 through 2008, the number of ETFs issued was $661 billion in net new shares, and 16 n investor demand for broad-based domestic equity ETFs accounted for nearly 50 percent. New equity ETFs issued totaled $336 billion in net new shares during this 10-year period, and their assets were $266 billion at year-end 2008. Within the broad-based domestic equity category, ETFs that track large cap domestic equity indexes, such as the S&P 500, managed $185 billion or 35 percent of all assets invested in ETFs. Chart 2, illustrates in $Billions of dollars through 2008 the total net assets concentrated in large cap domestic stocks. Chart 2.4: ETFs New Equity Issue- Year-end 2008 Total Net Asset's of ETFs Concentrated in Large Cap domestic Stocks I!? 185 200 r-- ~ 0 150 114 c ....0 "'c ~ iii - 100 58 50 25 c.. ~ ~ D D 0 v 28 10 ~~ ~ 29 D ~ D 36 n n 57 ~ ~ Source: ICI Fact-book 2009, p.46, http://www.icifactbook.org/ As investor demand increased for ETFs, fund sponsors continue to introduce new products onto the market. Over the period of 2000 to 2008, there were 758 ETFs created with the majority being offered in the last three years. Until 2008, few ETFs had been liquidated. In 2008, market pressures, started affecting ETFs that track identical indexes as they competed against each other to gain market share. 17 ETFs, which were tied to or specialized in niche market indexes failed to generate enough interest from investors. As a result 50 ETFs were liquidated during 2008; the number ofETFs increased, on net, by nearly 100 to 728 at year-end 2008. Chart 2.5: Growth of ETFs Number of ETFs 2000-2008 ~ 700 600 Ill u... 500 ~ w DCreated 0... 400 • Liquidated DTotal at year-end Q) ..0 § 300 z 200 - 2000 2001 2002 2003 2004 n 2005 2006 2007 2008 Year Source: ICI Fact-book (2009) , p.47, http://www.icifactbook.org/ 18 The Canadian Experiment The birth of ETFs in Canada began m 1989 with the launch of "Index Participation Shares" (IPU) on the Toronto Stock Exchange (TSE) in 1990. The IPU's were similar in structure to modern day ETFs as they represented a basket of stocks, which replicated a benchmark index. Known as "TIPS" 27 these units were created to track the performance of the Canadian market index known as the TSE 35 (the top 35 Canadian companies based on market cap). The success of TIPS 35 led to a broader version, which became know as the TIPS 100. They also traded on the stock exchange at a designated IPU to index ratio. In 1999 a new IPU began trading on the Canadian Stock exchange. The iUnits S&P/TSE 60 symbol XIU replicated the benchmark of the top sixty Canadian companies on the TSE. This new IPU replaced the TSE 35 and TSE 100 respectively. In April 1999, Mid-cap and Small-cap IPU's were added to the exchange. The i60s were almost identical to the TIPS, with a few minor differences. The TIPS 35 and TIPS 100 were both passively managed indices. The S&P/TSX 60 is an actively managed index. An S&P selection committee manages the inclusion of companies in the index using fundamental valuation criteria. The key criteria are size (assets and market capitalization), liquidity and sector leadership. 27 Kirzner, Eric, "Get ready for the i60 's", Canadian Investment Rev iew 19 Chart 2.6: Growth Curve of ETFs in Canada 7 ,- - - - - - - - - r - - - - - - - - - - - --,- 8000 -Volume 6 --+-Shares Traded 7000 6000 5 5000 4 4000 3 3000 2 2000 1000 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 0 Source: Exchange Traded Funds- ETFs Market Statistics I TMXmonev Growth of ETFs has been dramatic over the past two decades. In the early 1990s, only one Canadian ETF was available to advisors and their clients. The chart above dramatizes the trading volume in recent years as investors are increasingly utilizing ETFs as investment products. 20 Chapter 3 Literature Review: Risk - Reward The analytical framework of Risk - Reward is widely popular in academic and empirical discourse (Campbell, 1996). The Capital Asset Pricing Model (CAPM) by William Sharpe (1964) and John Litner (1965) is the most widely used analytical framework in calculating the cost of equity and to evaluate investment portfolios (Bos and Newbold, 1984; Fabozzi and Francis, 1978). This chapter provides an overview of the analytical framework for analyzing risk-reward relationships. The chapter is organized as follows: Section 3.1 lays out the key ideas of the risk-reward relationship. Section 3.2 provides indicators of the most commonly used measures of risk-reward evaluation. Section 3.3 discusses the risk preferences of investors. Section 3.4 reviews the concept of optimum portfolio outlined by Markowitz (1965). Section 3.5 provides the analytical framework of the Capital Asset Pricing Model (CAPM) 28 the predominant theory of risk-return. Section 3.6 extends the CAPM theory with the Fama-French Three Factor Model. Section 3.7 critically evaluates the empirical literature on risk-reward in mutual fund, index funds and ETF s. 28 For literature on CAPM, see Perold (2004), Fama-French (2004). 21 Section 3.1 Risks and the Reward Relationship In modern finance the notion of risk is central to security analysis and portfolio selection 29 (Sharpe 1972). The literature refers to various types of risk such as interest rate risk, inflation risk, business risk, financial risk, liquidity, exchange rate risk and country risk etc. These risks are broadly classified as: Systematic (market risk) and Non Systematic (business specific risk) (Sharpe 1965). Systematic risk is attributable to broad macro-factors such as national income, employment, inflation and exchange rates. Non-Systematic risk factors effect individual securities or specific industries. The non-systematic component of risk can be managed through a well-diversified portfolio. It is systematic risk which requires pricing. The most widely used indicators of systematic risks are (1) Standard Deviation (2) Beta (3) R-square (4) Alpha (5) Sharpe Ratio and the (6) Treynor Ratio. This literature provides a brief description of these concepts in the following paragraphs. Section 3.2 -Measures of Systematic Risk Standard Deviation is the most widely used measure of variability. The higher the variability in financial assets like stocks, the higher the risk associated with investing in these assets. It is the measure of deviation of each observation from the arithmetic mean of the observations. Standard deviation is a measure of total risk and not market risk. In order to understand market risk, we have to examine the concept of Beta which replaced standard deviation in the literature. 29 Sharpe (1972),p. 2 22 Beta (B) measures the co-movements of returns of a financial asset against the overall stock market index (normally the S&P/TSX, the S&P 500 or the Dow Jones Index). Beta is a measure of systematic risk or market risk which cannot be diversified. Beta, along with Alpha (a) and (R 2) are the central pillars of modern portfolio theory - both at the theoretical and practical level. R2 is an alternative statistical measure of risk. It is derived from performing ordinary least square (OLS) regression analysis on an individual asset against an underlying benchmark index. The R 2 value measures in a percentage form the degree of market risk and non-market risk. The R 2 maximum value is 1.0. Therefore, the closer the value is to 1.0, the more risk or more movement in the asset due to overall market economic forces such as higher inflation, exchange rate fluctuations or higher unemployment. These factors affect everyone in the market. Conversely, the remaining R 2 value indicates more business specific risk, such as labor strikes or events like a commodity shortage, which affect only a specific industry. The latter risk can be diversified and therefore will not be rewarded with a higher expected return. Alpha, is a measure of excess return (over a risk-free asset) (Ri-Rj) where Ri is the actual return of the underlying financial asset and Rf is the risk free asset (normally the yield on treasury bills of a certain maturity period). Then simply, alpha is the measure of excess return above what was predicted by the CAPM model. Investors and portfolio managers are always seeking positive alpha investments. William Sharpe (1966) whose contributions have already been discussed with the CAPM also developed a useful tool designed known as the Sharpe Ratio, to 23 measure the excess return a stock generated above the risk free rate of return from a ninety-one day Treasury bill divided by the standard deviation or risk volatility of the stock. The ratio enables the investor to assess how well the stock performed given the additional risk assumed. The formula is defined as follows: Sharpe Ratio = (Investment Return -Risk Free Rate)/ Standard Deviation A positive Sharpe ratio would indicate a financial investment product performed better on a risk-adjusted basis than the risk free asset (T-Bill). Logically then we can conclude, the higher the ratio, the better the overall performance relative to the risk taken. On a cautionary note, the ratio is a reflection of past performance and does not suggest the future performance will continue. If an investment had a positive Sharpe ratio during a particular timeline, this would not necessarily indicate future performance would continue. Therefore, an investor should carefully consider all aspects before purchasing any investment product including ETFs. A Sharpe ratio within the range of 1.0 - 2.0 would be considered good performance. Jack Treynor (1965) recognized two components of risk; risk from general fluctuations and risk from unique fluctuations in the securities in the portfolio. His research primarily focused on risk adjusted performance of systematic risk. He went on to formulate his work into a work ratio still used today in the investment community. The ratio is utilized extensively in this study. The Treynor ratio is a risk to reward ratio which considers only systematic risk and not total risk. Treynor (1965) developed this ratio to standardize returns relative to beta. Similar to the Sharpe Ratio (Sharpe 1966), which calculates the risk 24 premium relative to standard deviation, the Treynor ratio calculates the risk premium relative to beta as the per unit of risk. The numerator in the formula measures the risk premium and the denominator is the measure of risk as measured by beta and expressed as the risk premium return per unit of risk: T1 R 1. - RFR (3.1) Risk adverse investors would prefer to maximize this value. The resulting steepness of the slope line indicates the better the risk return trade-off. Therefore, a higher Ti value indicates better overall performance. This ratio is primarily used to evaluate individual assets as compared to the Sharpe ratio, which is more appropriate for well-diversified asset portfolios. The research argues all assets should have the same Treynor ratio, the same risk to reward risk ratio where risk refers to systematic risk. Then to the extent they do not, there is evidence that at least some portfolios have earned excess returns (Jordan 2006). Section 3.3 Risk Preferences Investors are broadly classified into these categories: a) risk - indifferent, b) risk - averse, c) risk - seeker (Holt and Laury 2002). For the risk indifferent investor, the required return does not change with an increase in risk. For the risk- averse investor, the required return increases with an increase in risk like (Hedge funds and ETFs which offer leverage or double leverage financial assets). For the risk-seeking investor, the required return decreases for an increase in risk. Most of the investors would come under a category of risk-indifferent (investors of higher maturity) and 25 risk-seeking investors (investors who are relatively young). Figure 3.1 displays the risk appetite of investors in an illustrative way. Figure 3.1 Risk Classifications Source: http://www. investopedia.com/articles/bas ics/03105 0203. asp?vie11•ed= / 30 Depending on investors risk appetite, every investor would appear somewhere on the risk-reward line. The general principle then being the larger the risk assumed, the higher the required return on investment. By minimizing the risk taken, the investor should expect to be compensated with a lower return. A question can then posed; are there any investment products which do not have any risk associated with them? The simple answer is "yes". Using the definition of risk in this manner, a Canadian government bond can be considered a risk-free asset because it is highly unlikely the Canadian government would default at the time of maturity. If all investors chose to minimize the risk, the result would have everyone holding riskfree assets such as government bonds, checking or savings accounts or short-term money market funds. 30 Accessed December 12,2009 26 In Figure 3.2, investment assets are separated by into a three-part pyramid based on risk characteristics. The bottom of the pyramid classifies low risk assets and the top of the pyramid, high-risk assets. Therefore, a properly diversified portfolio would have many combinations of assets through-out the pyramid. The ultimate goal then is to diversify the portfolio to obtain maximum returns while taking measures to identify the risk - averse point where an investor is most comfortable while minimizing as much ofthe risk as possible (Markowitz 1959). Figure 3.2: Risk-Reward Pyramid High R Summit • aowmmenr IJonds I Debt • Money A41rkBt I Bailie AccouJts • CDs, Hotss, Sills, Banlulrs Accept. ·Cash IJIId Cash ~ Source:http:IIH·\\ ·11 ". investopedia. com/art ic/es/basics/03105 0203. asp ?Fie wed = 131 31 Accessed December 12, 2009 27 Section 3.4 Optimum Portfolios The pioneering attempt to study risk- reward in a framework of an optimum portfolio is the attempt made by Markowitz (1959). Markowitz's theory starts with an assumption, all investors are risk averse as measured by standard deviation. He goes on to argue we should look at the risk of an entire portfolio not just the individual securities making up the portfolio. Furthermore, the addition of an individual security should not be evaluated as to its individual risk but standard deviation effect does asset have on the entire portfolio. Does the addition of the asset make the entire portfolio more or less riskier? Markowitz then utilized the concept of "correlation" to determine how the individual securities moved in relation to each other within an existing portfolio. He further suggested, the lower the correlation, the more diversified the portfolio would be. The important concept Markowitz suggested; a properly diversified portfolio could be constructed which could yield a high return for the minimum amount of risk taken. From this body of work, became the common notion of "Do not put all your m eggs in one basket". Modem Portfolio Theory or (MPT) changed the way investors considered the investments they chose plus the portfolio they created. Chart 3.3, illustrates the key concept developed by Markowitz and is named the Efficient Frontier Model. What Markowitz illustrated was standard deviation and the correlation of assets can be used to plot the relative return of any asset. The chart shows the difficulty an investor faces when choosing financial assets. For example, 28 historically small cap stocks provide the highest return, but usually with the highest risk and they would appear in the upper right quadrant. Investors, who prefer a low risk strategy may choose T -Bills with lower returns, would appear in the bottom left quadrant, while other investors may choose riskier investment somewhere in between. There is no one investment that is best for all investors. In Figure 3.3, the curved line, better known, as the optimum market portfolio would represent the collection of diversified assets, which would optimally provide the maximum return and minimize the risk for any investor depending on risk appetite. Figure 3.3 Markowitz Efficient Frontier Model A portfolio above this curve i s impossible "" i Hf11:h R i sk/H i gh Return Opthnal portfolios should lie on thi s curve (know as the "Efficient Frontier'") Medium Risk/Medium Return Portrolio"s bel.o"W the curve are not e "("t-icient, because for- the same risk one could achfe-ve a (:-- Low Risk I Low Return gJ"eater r-eturn . R1'sk % (Standar d De-vi at.1on) Copyrlgtot 2>003 • lnve!IO'topocd la "com Source: http://www.investopedia.com/terms/ele(ficientfrontier.asp 32 Since its inception, academic scholars have been trying to disprove this theory. Murphy (1977) in separate studies concluded the risk-reward relationship was proved weaker than expected. He further concluded there was no stable relationship between risk and return and the higher risk did not necessarily translate 32 Accessed December 13, 2009 29 into a higher return. Another study by (Haugen and Heins 1975) concluded in their empirical research, there was no support for the conventional hypothesis that higher levels of systematic risk led to higher returns. The Single Index model (SIM) developed by William Sharpe (1963) describes the advantages of using a simplified model of the relationships among securities for practical applications of MPT. Sharpe developed this model and determined the movements are due to a single common influence. Hence, the measure of this relationship can be found by relating the stock return to a benchmark index. The formula of the single Index model is (3.2) Where: Ri = return on stock i ai = component of stock return is independent of the market's performance Rm = the rate of return on the market index Bi = constant that measures the expected change in r given a change in m The basic assumption underlying the SIM is through Ordinary Least Squares, the returns of a security can be linearly regressed against a benchmark index. Generally a broad market index like the S&P 500 is used as the benchmark. The SIM separates the stock's return into two components. The first, known as Alpha (a) and the second, market related risk as (Beta * Return of the Market) symbolized as fJ(Rm). Alpha represents the individual stock's return based on factors specific to its own company or industry. For example, a newly discovered gold deposit or a fire 30 destroying a key facility would affect a firm's ability to generate positive excess returns. This risk factor can be properly diversified away. The market related part is a macro-level event affecting the entire market as a whole. For example, a change in interest rates or an increase in inflation would affect the entire economy. These events are not specific to one company or industry. The market risk factor cannot be diversified away. Therefore, the SIM is an individual related risk plus market related risk added together. The key assumption about the SJM is stocks co-vary together only because of their common relationship to the market index. In other words, only the market influences the performance of stocks. As a result of Sharpe's work, this model is viewed as simplifying the calculation of portfolio variance. The accuracy of the model depends on the accuracy of the key assumption of covariance. Lastly, the objective of the SIM is to simplify the calculations necessary for an investor to determine their optimum portfolio. 3.5 Capital Asset Pricing Model (CAPM) Building on the work of Markowitz, William Sharpe (1964) and John Litner (1965) created a new economic model known as the Capital Asset Pricing Model (CAPM). This model integrates the statistical behavior pattern of securities with the risk aversion of typical investors. It is considered the centerpiece of modem investing and sheds light on how assets are priced and what risk factors are considered to correctly price an asset. This work garnered Sharpe the Nobel Prize in Economics in 1990. Today CAPM is used to determine the cost of equity for firms in their capital budgeting plans. As well, public utilities use this model for rate setting and in the 31 context of this study, finance and investment professionals in pricing stocks and building investment portfolios. The premise under CAPM is investors will not invest in a security unless the expected return is equal to the required return. CAPM then goes on to explain how required returns are determined. There are two types of risk an investor considers prior to investing. The first being, an investor reqmres compensation for having their money invested for a period of time. The second being, an investor requires additional compensation for the possibility the investment generating the expected return. This is a measure of the movement or sensitivity of stock's return against a benchmark index. Statistically speaking, it a measure of covariance of how the stock moves in relation to movements in the market. The calculation derives the Beta (Sharpe 1964). A stock with a beta of less than one is less variable than a stock with a beta greater than one. This suggests the lower the beta, the less risk associated with stock and the lower the firm specific risk premium would be required. Conversely, the higher the beta, the more firm specific risk is associated with the stock and requires a larger risk premium in order to be adequately compensated for the additional risk. The marker risk premium is a function of a market return minus a risk free rate. This is also known as the equity risk premium. Multiplying the beta by the market risk premium generates the equity risk premium. Adding this value to the value of the risk free rate derives the total risk premium an investor would require in order to be properly compensated for investing in the stock (Sharpe 1964). 32 To compensate for the risk, the investment is required to achieve a risk adjusted return achieved by the overall market. A risk free rate plus a market premium are the basis of the CAPM. The firm's cost of equity is the sum of the return of a risk free asset plus the market risk premium as measured by beta for assuming firm specific risk. The risk free rate compensates the investor for the use of their money and the risk premium compensates the investor for the uncertainty of the investment not achieving its expected return. According to CAPM, this would be the expected return an investor would require in order to properly be compensated for the additional risk incurred by investing in a stock and adding to their portfolio. Together, the total risk is then determined by the amount of variance between the returns. CAPM invokes beta as the degree of measure of a stock' s risk compared the overall market risk. The CAPM implies there is a linear relationship between risk and return by the formula: (3.3) Ri = expected return on the investment Rf = expected risk-free of return ~ volatility of investment relative to an index (beta) E(Rm) = expected rate of return of the market 33 Figure 3.4 - Security Market Line Security Market Line {SML) 20% ~ i ""' E ·;;; Cl" &!! 16% 12% 8% 4% ~ ~ ~ ~ - - Required Return Risk -Free Rate 0% 0 .0 0 .2 OA 0 .6 0 .8 1.0 1.2 1.4 1.6 1.8 2 .0 Relative Risk as Measured by Beta Source: Hedges (2009), PPT Slide 11, Corporate Finance, Comm 725, UNBC, 2009 The vertical intercept represents the investment in risk free asset such as short-term government bond. The SML concludes in an equilibrium situation, for every asset sitting on the line, the rate of return is equal to the required rate of return. It defines the required return on basis of risk. The SML represents equilibrium stability. If stability is upset, forces are created to push it back into equilibrium. For example, if the expected return on an investment becomes less then the required return, this would signal to rational investors to sell the investment because it no longer meets their criteria for an expected return. The "sell-off' would lead to excess supply and limited demand forces. This would result in a drop in price. This drop in price would entice new investors, because of an increase in expected return. The most important aspect of the SML is forces are continually created to push towards an equilibrium state. Determining prices of individual securities is accomplished primarily through the Gordon Model. The Gordon Model is not discussed as it is considered beyond the scope of this study. 34 The CAPM is just a theory. In order to apply the CAPM model a number of assumptions are required. First, investors choose investments based on their interpretation of risk and reward. Second, there is also an assumption investors can borrow or lend money at a risk free rate. Third, investors can buy or sell any amount of stocks but the total value of the stocks does not change. This eliminates issues surrounding liquidity risk. Fourth, transaction costs, short selling or taxes are not considered. Finally, investors have access to the same information. Table 3.1: Global Equity Risk Premium Equity market risk premium , 1900-2001 Re lative to long-term govern ment bonds Mea n Canada Stand ard deviation 5 .7 France 17.9 6 .7 21 .7 9.6 28 .5 Japan 10 .0 33.2 United Kingdom 5.5 16.7 United States 6.7 20 .0 Germany 1 1 Excludes 1922- 23. Sources: Dimson et al (2002). The historical returns have provided an insight to the equity risk premium (See Table 3.1). The empirical research done by Dimson (2002) calculated the equity market risk premium for sixteen countries over a 102-year period from 1900-2001 , and showed risk-premiums relative to risk-free assets are substantial. In its simplicity the model works extremely well and provides consistently good results. In addition, the CAPM outlines the differences between diverse and non-diverseable risk. For these reasons, it has become one of the most popular 35 models used in finance today; as mentioned earlier, companies use the CAPM to calculate their cost of equity for capital budgeting purposes. Public utilities use CAPM to establish rates they will charge and various financial professional utilize CAPM to determining asset prices. Critics of the model have made testing the CAPM controversial. For example, Fama and French (1992) found evidence that stock market returns were predictable and demonstrated the predictability in a cross section of stock returns. Fama & French (1992) showed value stocks with high book to market value ratios, outperformed growth stocks with low book-to-market ratios on a risk-adjusted basis. Jegadeesh and Titman (1993) presented analysis in which a strategy of buying past winners and selling past losers generated a positive return. This type momentum strategy has generated positive returns for nearly a decade when it was re-tested by the same authors in 2001. These studies argue the CAPM does not sufficiently explain the cross sections of stock returns and the sources of the deviations. CAPM is a static model by which the expected returns are assumed to be constant. Merton (1973) and Campbell (1993) proved the returns of assets by measuring its covariance with variables that forecast the entire market. As well, Guo (2003) provided additional insight into market risk premiums, whereby investors also seek liquidity premiums on stocks to due to limited stock participation. Although some of these criticisms are true and have been proven empirically, the simplicity and ease of using the CAPM is still considered worthwhile. The mainstream view on the failure of the CAPM is assuming investors care only about 36 the mean variance of portfolio returns and ignore other important dimensions of risk The argument that a single factor is not sufficient in the stock market to account for the variation in returns will go on indefinitely. The solution may lie in the development of more sophisticated asset pricing models such as the Fama-French Model (1992). 3.6. Fama- French Three-Factor Model The CAPM model is essentially based on two factors: risk free rate and risk premium. The empirical investigation based on CAPM has generally resulted in low explanatory power. Thus there was a need to incorporate other factors to improve the explanatory power in empirical investigation. The Fama-French Three Factor Model (1992) incorporated variables to reflect size and value to augment the CAPM model. They have done extensive research in this area and found factors describing "size" and "value" to be the most significant factors, outside of market risk, for explaining the realized returns of publicly traded stocks. To represent these risks, they constructed two factors: SMB to address size risk and HML to address value risk. The model goes on to prove a positive SMB indicates a small cap stock outperformed a large cap and vice versa. The SMB factor stands for "Small Minus Big", and measures the additional return average investors have historically received by investing in stocks of companies with a small market capitalization. This additional return is often referred to as the "size premium." The HML Factor; HML, which stands for "High Minus Low", measures the "value premium" provided to investors for investing in 37 companies with high book-to-market values. A positive HML indicates a value stock outperformed a growth stock and vice versa. Their research has proven over the time period of 1926 to 2002, there is a premium for value stocks over growth stocks and this premium has averaged approximately 5.1 percent annually (French 2003). Also the two additional factors the researchers have tested, often yielding an R2 value of approximately 0.95 (French 2003). Furthermore, the model has proven the fact positive alphas observed in a CAPM regression are more a function of exposure to HML or SMB factors, rather than actual portfolio manager's performance (French 1992). From their research, investors learned to weigh their portfolios in such a way that they have generated greater or lesser exposure to each of the specific risk factors, and therefore can adjust their asset allocation portfolio to different levels of expected returns. This has led to the mutual fund industry developing matrices to target different strategies and exposure to risk. For example, these "style boxes" are a common feature on the Morningstar® website. Figure 3.5: Fund Investment Style and Size of Companies S ze ~ s Source: 33 ~ Date accessed: December 14, 2009 38 ~ 118018 33 .a l l By combining the original market risk factor from the CAPM and incorporating it with the newly developed factors by Fama-French, this model calculates the expected return of an investment asset as a result of its relationship to market, size and value risk. The formula for the Fama-French Three Factor Model: Section 3. 7: Risk-Reward of Stocks, Mutual, Index Funds and ETFs The risk-reward of portfolios like mutual funds, index funds and ETFs are generally supposed to differ from individual assets in the portfolio. Given the fact these portfolios diversify the risk of individual asset classes, the risk of these portfolios are expected to be lower than individual assets which comprise these portfolios. Filbeck and Tompkins (2004) studied the value of management tenure in creating excess returns by including a measure of risk-adjusted returns as a variable. Moskowitz (2000) examined various types of balanced mutual funds. The study concluded lower net returns were primarily caused by transaction costs. Rompotis (2009) analyzed data from active and passive ETFs listed in the United States. His research concluded active ETFs underperformed both the passive ETFs as well as market indexes. The research utilized the Sharpe ratio as well as regression analysis to suggest active fund managers showed limited ability to select ETFs and choosing the correct time to enter the market. In addition, his research looked at tracking error 39 and determined tracking error estimates were greater for active ETFs rather then passive or index funds. However, there are no similar studies in the Canadian context in this area. The present study is an attempt to fill this gap. 40 Chapter 4 Data Sources and Methodology This chapter discusses the data source and methodology used in the riskreward empirical investigation of ETFs. Section 4.1 discusses the data base used for the study. Section 4.2 discusses the methodology. Appendix 4.1 at the end of the concluding chapter displays the name and symbol of the ETFs studied and the date of inception. Section 4.1 Data base The study is based on data of 74 Canadian ETFs for the last 5 years (2004 to 2009). The data period used is from September 1, 2004 to August 30, 2009. The data was downloaded from the DataStream database; provided by Thomson Reuters. The data relates to ETFs prices on a monthly basis and corresponding benchmark index such as S&P 500 which is a broad based market index. The returns on individual ETFs were generated from corresponding price data. In the case of ETFs which were created recently, the data points could be considerably lower and these are also indicated in Appendix 4.1. For the purposes of this study, the top three Canadian investment firms offering ETFs products were chosen. Horizon Bet Pro funds (HBP) has been actively traded since 2007 with over $100 million in assets under management in Canada 34 . HBP offer investors the opportunity to invest in ETFs, which use leverage techniques 34 Horizon Beta Pro (HBP) Accessed on December 12, 2009 http: //www. tmxmoney.co m/en/sector profi les/exchange traded funds/funds/funds.htm l 41 to magnify the performance results of the assets. Each ETFs attempts to generate returns, which are 200 percent (Bull) and an inverse -200 percent (Bear) performance of its underlying index. Thirty ETFs from this company were analyzed to gain an understanding of leverage effects their performance. The second largest ETFs investment company in Canada is Claymore Funds with assets in excess of $13.3 Billion as of September 2009 35 . Claymore offers a variety of ETFs ranging from Canadian, US, and Global core equity markets, sector strategies, dividend and income based strategies and fixed income based strategies. Twenty ETFs from all investment categories were analyzed in this study. Finally, iShares Canada is the leading ETFs provider with assets under management exceeding $22 Billion as September 2009 36 . In Canada, iS hares was founded in 1999. Thirty ETFs from this fund company were analyzed. They were divided into three sections. Section one, analyzes fourteen ETFs tracking various indices on the Canadian equity benchmark index. A further six ETFs studied the fixed income market of government and corporate bonds. Lastly, the remaining ETFs cover a broad range of alternative indexes such as international and emerging markets and as well as investment style classes like growth and value portfolios. Other ETFs providers include Bank of Montreal and the Invesco group. They are excluded from the empirical investigation as these products are relatively new and simply do not have enough trading activity to warrant a complete a thorough analysis of their performance. 35 Claymore Funds Website accessed on December 12,2009 http ://www.claymoreinvestments.ca/about/about-claymore/ 36 iShares Website accessed on December 12, 2009 http://www. tmxmoney .com/en/sector profiles/exchange traded funds/funds/funds. html 42 Section 4.2 Methodological Framework We adopt the methodological framework of CAPM for risk-reward analysis. The CAPM framework is used to derive the risk ofETFs as well as to calculate alpha (a) - the excess return of each ETFs as discussed in chapter 2. The individual risk of each ETFs was calculated using the single index model, viz, Ri=a+ p Rm, (4.1) Where Ri= is the return of each ETFs (generated from ETFs prices) and Rm is the market index (S&P 500). The coefficient associated with beta ~ is taken as the measure ofETFs risk. The excess return (a) is generated from the fundamental CAPM equation as given below: Ri = Rf + p [E(Rm)- Rf] (4.2) Where Ri = return of individual ETFs; Rf= risk-free rate (one year Canadian treasury bill rate); beta ~ is the estimate of market risk (from equation 4.1) and E(Rm) is the expected market return (estimated as the average return in the market index for the last 5 years). The excess return (a) is estimated by transforming equation 4.2 as follows: Ri -Rf = p [E(Rm)- Rf] (4.3) 43 Chapter 5 Market Risk of Canadian ETFs - Empirical Results Based on the analytical framework outlined in chapter three and the database discussed in the previous chapter, we present the results of market risk for Canadian ETFs. We have chosen to present the estimates of market risk of various ETFs in Canada based on the risk characteristics of ETFs rather than the originator of these ETFs. This chapter is organized as follows: Section 5.1, we present the estimation of market risk Beta (B) and R square (R2 ) for Fixed Income ETFs. Section 5.2, we present the results of Canadian ETFs. Section 5.3, we present similar results for international (global) ETFs sold in Canada. Section 5.4 presents the results of commodity ETFs. Section 5.5 we present the results of gold and precious metals. Section 5.6 we present the results of alternative investments. Section 5.7 we present the results of currency hedged ETFs. The appendix table 5.1 following the concluding chapter presents ETFs symbols with the name description and the benchmark index the ETFs is designed to replicate. Section 5.1 Fixed Income ETFs Fixed income assets are supposed to be lowest risk ETFs among the menu of financial assets. As the term "fixed" implies, these ETFs provides investors with fixed cash flows over a specified period of time. This removes the majority of risk for an investor. There does remain a small risk where default can still occur. It is however unlikely the government of Canada would default. With that in mind we 44 would expect to find government backed ETFs to offer the lowest risk. With corporate bonds there is slight increase in the possibility of default and such bond ratings apply risk premiums to these financial assets increasing the return slightly for investors. In Chart 5.1 , nine fixed income ETFs were ranked according to their market risk beta. As is evident, all the fixed income ETFs had negative betas. The government bond CLF offers the least amount of financial risk. These ETFs also employs a laddered strategy 1-5 year maturity period. The short-term nature of these ETFs also insulates it from risk as it can renew at different interest rates in a relatively short period. Therefore we would also expect to see long-term bonds with a larger Beta implying there is more risk over a longer period of time. The data supports, the US government 30-year bond HTD is the riskiest fixed income asset in this study. Reason being, the extent of bond maturity period extends 30 years and the ETFs employing a double leverage strategy to maximize the return. Chart 5.1 Risk (B) -0.20 -0.25 Exchange Traded Fund 45 Table 5.1 Risk (R2 ) R2 Date of Inception Fixed Income Fund Manager Premium Money Market Claymore ETF CMR 21.83% Feb-08 CON Government Bond Index iShares XGB 16.66% Nov-06 CON Bond Index iShares XBB 8.58% Nov-00 CON Long Term Bond Index iShares XLB 5.06% Nov-06 CON Real Return Bond iShares XRB 2.95% Oec-05 US 30 Year Bond Bear Plus Horizon HTD 2.79% Jun-08 CON Corporate Bond Index iShares XCB 2.00% Nov-06 CON OEX Short Term Bond Index iShares XSB 1.15% Nov-00 1-5 Year Laddered Government Claymore CLF 0.05% Jan-08 In Table 5.1, we can see market risk in percentage terms is very low amongst the fixed income assets. By incorporating them into a portfolio the market risk is minimized. The most significant risk factor affecting fixed income ETFs is interest rate risk. Therefore, depending on the maturity period we would expect the risk appetite of investors to change when there is significant changes in interest rates. Section 5.2 Canadian and US Equity ETFs Canadian Equity and US Equity are the most popular amongst investors. There is a large amount of variety and choices for an investor with any risk appetite to consider. In this category we would expect to find broad based ETFs, which track large segments of the economy to offer the least amount of risk. In contrast, ETFs employing leverage strategies or tracking single commodity indices would generally 46 higher in risk. For reference a straight-line beta of 1.0 indicates in the chart replicates the beta of the underlying S&P 500 Benchmark index. In Chart 5.2A, we identify the five ETFs from Horizon Beta Pro funds with the highest risk. All of these ETFs employ either a double leverage or a double inverse strategy. General principles of finance confirm when leverage or double leverage is employed the risk and return is magnified. HFU (financials) and HEU (energy) track specific sectors in the stock market. In recent years we have seen significant return variation in both of these sectors. The global economic downturn in the financial services sector and the rapid rise and fall of energy prices has had an effect on these assets in terms of risk. Further analysis in the next chapter will test the hypothesis, the riskier the asset the greater the return. On the opposite spectrum, XTR is an income trust ETFs. By its structure, it distributes a fixed income each month to the investor. Also, by distributing cash payments directly to the investor, this ETFs has tax advantages over other ETFs, again minimizing its risk. As expected XSP, is designed to replicate the return of the S&P 500 index and the beta is consistent with the hypothesis. Preferred shares are considered hybrids (neither a fixed income asset nor an equity asset). They do distribute monthly income to the shareholder and have a claim above equity investors in the event of default. The data illustrates this implies a reduced level of risk. CPD the preferred shares ETFs supports a lower risk level. There are ten ETFs which offer close approximation to the market index performance. This supports the notion, beating the benchmark index is very hard to do. 47 Chart 5.2A: Risk (B) c: 2.50 2.00 Q) ·a; CJ !E 1.50 q> 0 (.) 111 1.00 Q) Ill 0.50 0.00 --,-- ~~~ ~~~~~~~~~~~~~~ Exchange Traded Funds Table 5.28: Risk (R2 ) Canadian & US Equity Fund Manager ETF R2 Date of Inception S&P 500 Bull Plus ETFS Horizon HSU 99.4% Jun-08 CON S&P!TSX Hedged to CON$ iShares XSP 91.7% May-01 RAFI Core $US Hedged CON Claymore CLU 90.8% Sep-06 NASDAQ 100 Bull Plus ETFS Horizon HQU 84.7% Jun-08 RAFI Core Canada Claymore CRQ 82.5% Feb-06 S&P!TSX Capped Financials Bull Plus Horizon HFU 75.8% Jun-07 S&P!TSX 60 Bull Plus ETFS Horizon HXU 73.6% Jan-07 S&P!TSX60 iShares XIU 69.7% Sep-99 S&P!TSX Capped Financials Index iShares XFN 69.6% Mar-01 S&P!TSX Capped Composite Index iShares XIC 68.7% Feb-01 S&P!TSX Completion Index iShares XMD 65.0% Mar-01 S&P!TSX Small Cap Index iShares xes 63.5% May-07 48 S&P!rSX REIT Index iShares XRE 57.6% Oct-02 S&Prrsx Capped Energy Bull Plus Horizon HEU 45.8% Jun-07 Preferred Share Claymore CPO 42.3% Apr-07 S&Prrsx Capped Energy Index iShares XEG 33.2% Mar-01 S&P!rSX Capped Information Technology iShares XIT 32.7% Mar-01 S&P!rSX Capped Materials Fund iShares XMA 16.5% Dec-05 S&P!rSX Income Trust Fund iShares XTR 2.0% Dec-05 From Table 5.2B we can surmise equity ETFs are largely susceptible to market forces as evidenced by the higher R2 values. As this study utilized the S&P 500 as the underlying benchmark, both HSU and XSP, which are designed to track this index, demonstrated the highest correlation to S&P 500. XTR showed little correlation to the index consistent with it lower B. This suggests investors value the steady stream of interest payments and consider the value of those cash payments to be less risky than investing in other specific sector ETFs. Chart 5.2C: Risk (B) 0.00 'EQ) -0.50 w HXD HED HQD ·c:; ~ Q) 0 -1.00 u ....., Q) Ill -1.50 - - - -2.00 Exchange Traded Fund 49 HSD Chart 5.2.C illustrates six equity ETFs which generated negative B. The fund XSU replicates the Russell 2000 US index and is hedged in Canadian dollars. Three of the ETFs almost achieved their intended strategy of doubling the index. Exchange rate fluctuations have had an effect on the perceived level risk for this ETFs negative beta. The remaining five ETFs employed a double leverage inverse strategy. These ETFs are designed to replicate two times the inverse performance of their underlying indices. As indices fall in value, the ETFs would employ leverage to "short" to magnify the returns of this strategy. Also, these ETFs are relatively new to the trading market. As the market has recently rebounded, these ETFs have performed poorly and with the effects of leverage, have resulted in higher levels of risk. Table 5.2D: Risk (R2 ) Canadian & US Equity Fund Manager ETF R2 Inception Date S&P 500 Bear Plus Horizon HSD 95.2% Jun-08 NASDAQ 100 Bear Plus Horizon HQD 80.2% Jun-08 S&P/TSX 60 Bear Plus Horizon HXD 73.7% Jan-07 S&P/TSX Capped Financials Bear Plus Horizon HFD 70.9% Jun-07 S&P/TSX Capped Energy Bear Plus Horizon HED 26.1% Jun-07 CON Russell 2000 Hedged to CON $ iShares xsu 2.4% May-07 In this table we further illustrate the close relationship the inverse double leverage demonstrated with the benchmark index. The movements in these assets have strong correlation to the movements in the underlying index as measured by high R2 values. However, the inverse strategy has not been effective as further 50 evidence supports market timing is not a viable strategy. For double leverage ETFs to attain double returns of the benchmark index, would require higher R2 values to better approximate the movements of the benchmark index. Section 5.3 International and Emerging Markets International investing adds new options for domestic portfolios. Currency and stock markets usually move in different directions. Investing in global, emerging markets and country specific index offers investors stock and currency diversification. Overall, investing in these ETFs diversifies a portfolio and reduces overall risk. The Chart 5.3A illustrates the beta co-efficient for six ETFs in this asset category. We would expect little correlation with the S&P 500 index. Chart 5.3 Risk (B) 2.50 2.00 ....s:::: Q) ·u ;-- 1.50 - 1.00 :e

0 (.) :--1.00 ~ Q) Ill - 0.50 0.00 HOU HMU CLO - CMW Exchange Traded Fund cww n cow HNU The two ETFs generating the highest betas were HOU, which IS benchmarked against the NYMEX Crude Oil index. The other, HMU, IS benchmarked against S&P/TSX base metal index. Both of these ETFs are intended to achieve 2 times the beta as they utilize a leveraging strategy. The lack of diversification verifies these two ETFs did generate higher betas. It is not uncommon for significant short-term price swings on these commodities. The amount of 53 available inventory and currency fluctuations can affect all ETFs in the commodity sector to have significant variability. HNU, did not achieve its double return objective due high inventories from mild weather and excess drilling activity. Table 5.48: Risk (R2 ) Commodities Fund Manager Oil Sands R2 Inception Date Claymore ETF CLO 45.1% Oct-06 Global Water (S&P 500) Claymore cww 41.9% Jun-07 S&P/TSX Global Base Metals Bull Plus Horizon HMU 37.8% Jun-08 Global Agriculture Claymore cow 34.2% Dec-07 Global Mining Claymore CMW 32.5% Jun-07 NYMEX Crude Oil Bull Plus Horizon HOU 28.8% Jan-08 NYMEX Natural Gas Bull Plus Horizon HNU 7.2E-12 Jan-08 Compared to other investing sectors, The R2 values in commodities were found to be lower. This would indicate supply and demand forces specific to the commodity are the main drivers of performance. However on a macro level, currency fluctuations and inflation pressures can also have an effect in the movement of this asset class. Chart 5.4C: Negative (B) cQ): "(3 0.00 -0.50 HND I n.&Q I HMO -1.00 !E -1.50 ~ 0 -2.00 ., -2.50 0 Qj Ill -3.00 -3.50 Exchange Traded Fund 54 HOD From this chart, HOD is the only ETFs to achieve its intended inverse double leverage beta. ETFs which use inverse double leverage of future contracts to seek positive returns on the falling oil prices. Three of the ETFs in this asset class use double inverse leverage futures contract to generate positive returns on falling commodity prices. Table 5.40: Risk (R2 ) Commodities Fund Manager ETF R2 Inception Date NYMEX Crude Oil Bear Plus Horizon HOD 42.1% Jan-08 GAS Commodity Claymore GAS 18.3% Feb-08 S&PITSX Global Base Metals Bear Plus Horizon HMD 16.4% Jun-08 NYMEX Natural Gas Bear Plus Horizon HND 4.7E-04 Jan-08 From this table, we can see business specific forces driving the performance and less overall market forces having an effect. This would suggest these instruments are utilized in market timing strategies. This provides additional support suggesting market-timing strategies are not very effective. For inverse ETFs, it is very critical to be on the "right side" in terms of timing as the percentage change can vary drastically over short periods of time. From this we can conclude, double inverse leverage ETFs are not an effective "Buy & Hold" strategy. These ETFs may be utilized on a short-term basis as the returns can be magnified greatly due to the leverage employed. Section 5.5 Gold & Precious Metals Gold and Precious metals are defensive investments. Meaning investors seek these commodities during global economic downturns. Gold has a finite supply. 55 n The performance of these ETFs would be driven more by the price of gold or the price of gold mining stocks than the underlying index. As such we would expect to see lower beta and lower R 2 values. Regardless of whether the price of gold fluctuates wildly, the beta relative to the underlying index would be low. Also, during periods of rising prices, gold and other precious metals are considered as hedges against inflation. Investors can access these markets without physically taking possession of the commodity. Chart 5.5 Gold & Precious Metals: (B) 0.80 "E 0.60 (I) r--- "(J r--- !E 0.40 'I' 0 u 0.20 ~ Ill 0.00 -0.20 HGD HGU n XGD ,---, HBU IGT ~ Exchange Traded Fund From the chart we can see the betas are lower as predicted. The two ETFs which employed double leverage and inverse double leverage strategies generated the highest Betas. None of the double leverage ETFs achieved the double beta in relation to the S&P 500 benchmark index. Table 5.5 Risk (R2 ) ETF R2 S&P!TSX Global Gold Bear Plus Horizon HGD 3.6% Jun-07 S&P!TSX Global Gold Bull Plus Horizon HGU 1.5% Jun-07 CON S&P!TSX Global Gold Index iShares XGD 1.0% Mar-01 COMEX Gold Bullion Bear Plus iShares HBD 0.6% Jan-08 COMEX Gold Bullion Bull Plus Horizon HBU 0.1% Jan-08 Comex Gold Trust iShares IGT 0.1% Dec-05 Gold & Precious Metals Fund Manager 56 Inception Date This sector has produced the lowest R2 values in this study. As predicted the performance of these ETFs are more affected by the price of gold or the performance on individual mining stocks then movements in the benchmark index. We do not find many gold companies in the S& P 500. Section 5 .6 Alternative Investments As the ETFs industry has evolved, various investment styles have been created. These styles include dividend income, growth, and value on through to specific industries like the North American Agribusiness HAU. In this section we can see nine ETFs with betas ranging from 1.49 HAU for a specific industry to 0.14 XCV, which is benchmarked to the Dow Jones Value index. The lower beta represents a significant difference between the S&P 500 and Dow Jones Value index. As well, R2 value indicated a high S&P 500 related correlation with HAU to basically no correlation with the two iShares ETFs designed to track a different index. Chart 5.6A: Alternative Investments: Positive (B) 1.60 ..... ~ '(j i 1.40 1.20 1.00 - b 0.80 (.) 0.60 ca ~ 0.40 o.2o D D XDV XCV 0. 00 +----'L-...J'---r_L_---'-- , -'---' - r- ' ----'-, --'---...___,.-'---'---,--_L.._...___,_._--L----,---'---'-------, HAU CYH CDZ CGR CBN CIF Exchange Traded Fund 57 CBD I From the table we can the movements with the two iShares ETFs were virtually not affected by movements with the benchmark index. These ETFs were designed to track a separate index. Chart 5.6C: Alternative Investments: Negative (B) 0.00 ....s::: -0.20 Q) 0 it: Q) I 0 (.) XEN -0.40 -0.60 -0.80 -1.00 ....Ill -1.20 Q) aJ -1.40 - -1 .60 I Exchange Traded Fund 58 Table 5.60: Risk (R2 ) Alternative Investment Strategy Fund Manager ETF R2 Inception Date S&P Agribusiness N.A. Bear Plus Horizon HAD 28.0% Mar-08 Conservative Core Portfolio Builder iShares XCR 10.6% Nov-08 Growth Core Portfolio Builder iShares XGR 8.1% Nov-08 Global Completion Portfolio Builder iShares XGC 5.9% Nov-08 CON Dow Jones Growth Index iShares XCG 0.0% Nov-06 CON Jantzi Social Index iShares XEN 0.0% May-07 Specialty funds like these, which invest in other assets not listed in the S&P 500 benchmark, all have lower betas because their performance is tied more to other indexes or other factors. The low R2 values support the hypothesis that other factors unrelated to movements in the benchmark index are the drivers of performance for this asset class. Section 5. 7 Currency Hedged ETFS In the past, institutional investors primarily invested in the currency markets. Today, with ETFs, all types of investors have access to these markets. In this section we consider two currency ETFs hedges to the US dollar. Also, these ETFs employ a double leverage or an inverse double leverage strategy. These ETFs are created to generate returns of an underlying bond or currency. This represents an underlying benchmark for investors creating exposure to underlying bond or currency markets. "The Indices are calculated based on daily returns of a Daily Contract Settlement Price published by the Chicago Board of Trade ("CBOT") in the 59 case of the US Long Bond Index contracts, and the Chicago Mercantile Exchange in the case of the C$ Currency contract, for the Designated Relevant Contracts" 37 . Chart 5. 7 Currency: (B) 1.50 c 1.00 ·o :::'P 0.50 Cl) 0 (.) CIS Qj Ill '" j -0.50 HOD HDU -1.00 Exchange Traded Fund The two ETFs did not double the S&P 500 index performance. Table 5. 7 Risk (R2 ) Currency Hedged & Non-Hedged Fund Manager US Dollar US Bear Plus Horizon ETF HOD US Dollar US Bull Plus Horizon HDU Inception Date Jun-08 37.1% Jun-08 The lower R2 values and betas illustrate these ETFs do not approximate the movements of the S&P 500. The primary movements are perhaps more dependent on currency exchange rates. 37 Horizon Beta Pro Currency Fact Sheet http://www. hbpETFs .com/ indexPerformance.asp (Accessed on December 14, 2009) 60 Chapter 6 Risk - Reward of ETFs in Canada - Empirical Results This chapter presents an empirical estimation of the risk-reward relationship of 74 Canadian ETFs based on the price data for the last five year period (September 2004 to August 2009). This chapter is divided into four sections. Section 6.1 reviews the analytical framework and assumptions underlying the empirical investigation. Section 6.2 presents the estimates of excess return (alpha) of Canadian ETFs asset classes. Section 6.3 presents the results of Canadian ETFs in terms of risk-reward and Section 6.4 presents the conclusions emanating from empirical investigation. Appendix 6.1 and 6.2 presents the performance of each of the individual ETFs in terms of excess returns and risk-reward respectively Section 6.1: The Analytical & Empirical Framework As discussed in chapter three, the analytical framework for estimation of excess returns (alpha) is the Capital Asset Pricing Model. The estimation of excess returns (alpha) requires two crucial inputs - the Risk-Free rate (Rf) and expected return of the market [E(Rm)]. We have chosen the three-month Treasury bill rate as a representative of Risk-Free Rate (Rf) in Canada (available from Bank of Canada website under the rates and statistics section) as this maturity segment was found to be the most active and liquid segment of the short-term treasury bill market. The 3month Treasury bill rate in Canada is approximately 1.5 percent. The Expected Return of Market [E(Rm)] was calculated from the monthly S&P/TSX intervals 61 during the period of study from beginning of September 2004 to the end of August, 2009. The return was rounded-off and this comes to 8 percent for the 5 year period and consequently the estimated risk-premium in the Canadian context comes 6.5 percent (Table 6.1). It may be recalled from chapter three; Canada's historical risk premium from 1900 to 2001 was calculated to be 5.7 percent. This provides support for the risk premium value used in this study. The empirical estimation of excess return is based on CAPM and is based on equation 4.2 in Chapter 4 [ Ri - Rf= ~ (E(Rm- Rf)]. Table 6.1: Basic Assumptions underlying CAPM Model CAPM Assumptions Risk Free Rate (Rf) 1.5% Return Market [E(Rm)] 8.0% Risk Premium [Rf-E(Rm)] 6.5% For evaluating ETFs portfolios in terms of risk-reward, the analytical framework in this area considers the Sharpe Ratio and Treynor Ratio. The most significant different between these indicators as discussed in chapter four is the Sharpe ratio measures the reward to total risk, while in Treynor ratio uses systematic risk instead of total risk. 62 Table 6.2: Summary of Expected Returns, Actual Returns & Excess Returns Total Return Expected Return (CAPM) Alpha International & Emerging Markets 9.84% 6.63% 3.21% Canadian & US Broad Equity 5. 79% 4.34% 2.43% Currency $US Hedged 7.8% 5.49% 2.82% Fixed Income 1.76% 1.37% 0.40% Alternative investment Strategies Gold & Precious Metals 0.87% 3.82% -2.95% 5.22% 4.33% 1.18% Canadian Sector Funds -2.62% 5.38% -8.00% Commodities -17.53% 2.85% -20.38% ETF Asset Class Section 6.2: Excess Returns (Alpha) Empirical Results of Canadian ETFs- In this section, we present the estimated excess returns (alpha) of Canadian ETFs by broad asset classes as well individual Canadian ETFs for the period September 2004 to August 2009. Table 6.2 provides estimates of excess returns of ETFs by broad asset class and these show international and emerging market ETFs have generated the highest excess returns (alpha) during the period of study, followed by Canadian and US equity ETFs, currency hedged funds and fixed income funds. These results are in tune with emerging market growth story especially that of BRIC countries (Brazil, Russia, India and China) which witnessed rapid economic growth in the last decade. Among the international and emerging market ETFs, the one based on Emerging markets (MSCI Emerging Market Bull Plus - symbol HJU) 63 generated the highest return of nearly 143 per cent during the period of investigation (Appendix table 6.1 ). Other funds which have generated substantial excess returns are the global real estate and global portfolio funds (symbols CGR and XGC respectively - see Appendix table 6.1 ). Among the Canadian and US broad equity funds, the one based on NASDAQ 100 Bull (symbol HQU) and S&P 500 Bull (symbol HSU) have generated good returns; in the range of 26 to 76 per cent (Appendix table 6.1 ). The currency hedged funds also generated substantial positive returns during the period of investigation bringing into focus the importance of exchange rate volatility and its hedge as crucial parameters in generating positive excess returns (See Appendix table 6.1 ). The fixed income funds reflecting the lower risk has generated relatively lower returns (less than 1 per cent). Among the worst-performing Canadian ETFs, were the commodity fund and Canadian Sector funds and alternative investment strategy funds (Table 6.2). The performance of commodity ETFs represents an interesting picture. Commodity funds, which were based on a bear market (like crude oil and natural gas); have performed relatively well reflecting the downward slide in the market over the last two years (Table 6.3 and Appendix table 6.1 ). The bull funds have generally performed poorly during the last 5 years. The bear ETFs performed well during a period of worldwide declining commodity prices. In particular the ETFs funds under the symbol HND, the natural gas bear ETFs benefited by consistent and prolonged pricing declines in the natural gas futures market due to excess supply and high inventories of that commodity. 64 HOU, the crude oil bear ETFs also benefited by falling crude oil prices during the period of this study as crude oil prices fell over 50 percent from their highs of $14 7 dollars barrel in July 2008 (NYMEX). Table 6.3 Commodity ETFs Commodities Fund Symbol Date Total Return CAPM Return Alpha NYMEX Natural Gas Bear Plus Horizon HND Jan-08 156.91% 0.83% 156.08% NYMEX Crude Oil Bear Plus Horizon HOD Jan-08 55.65% -17.59% 73.24% S&P Agribusiness N.A. Bear Plus Horizon HAD Mar-08 53.70% -8.40% 62.11% Global Mining Claymore CMW Jun-07 0.03% 7.33% -7.30% Oil Sands Claymore CLO Oct-06 -1 .31% 9.61% -10.92% Global Agriculture Claymore cow Dec-07 -7.92% 5.36% -13.28% Global Water (S&P 500) Claymore cww Jun-07 - 11.48% 5.74% -17.22% S&P Agribusiness N.A. Bull Plus Horizon HAU Mar-08 -25.27% 11.20% -36.47% S&P!TSX Global Base Metals Bull Plus Horizon HMU Jun-08 -28.23% 13.30% -41.53% S&P!TSX Global Base Metals Bear Plus NYMEX Crude Oil Bull Plus Horizon HMO Jun-08 -53.09% -6.11% -46.98% Horizon HOU Jan-08 -81 .00% 15.92% -96.92% GAS Commodity Claymore GAS Feb-08 -126.48% -1.59% -124.89% NYMEX Natural Gas Bull Plus Horizon HNU Jan-08 -159.37% 1.50% -160.87% In terms of performance, five out the seven worst-performing commodity ETFs employed double leverage or double leverage inverse strategies. The bottom two performers were benchmarked against natural gas. These results indicate high- 65 risk investment funds especially commodity funds do not perform over the medium term (over a 5 year period). They could however represent a good choice for shortterm investments and carry substantial risks over a longer investment horizon. Section 6.3: Risk-Reward Canadian ETFs Overall Performance The empirical results presented in chapter 6.2 identified international and emerging market ETFs as the asset class which generated the highest excess returns. These asset classes also possess higher market risk. When one evaluates these funds in terms of risk-reward, the rankings are quite different (Table 6.3). The Canadian and U.S Equity ETFs had the highest Treynor ratio's (Table 6.3). In this category, NASDAQ Bull ETFs (Symbol HQU) had the highest Treynor ratio. Table 6.4 Treynor Ratio Rankings of Canadian ETFS by Asset Class Asset Class Total Return Alpha (Excess Returns) Treynor Ratio Canadian & US Broad Equity 5.79% 2.43% 0.11 Currency $US Hedged -21 .2% -22.97% 0.09 Canadian Sector Funds -2.62% -8.00% 0.06 International/Emerging Markets 9.84% 3.21% 0.01 Alternative Investments 1.05% -2.80% ·0.02 Gold & Precious Metals -8.22% -11 .10% -1.01 Commodities -17.53% -20.38% -1.13 Fixed Income 1.64% 0.58% -2.05 66 According to the Treynor ratio, thirty-two or 43 percent of the ETFs generated a positive Treynor Ratio. XEN generated the highest ratio of 4.1 0, while CEW generated the lowest value. Two ETFs in the study CDZ & XIT generated a zero Treynor Ratio. According to the data, forty one or 55 percent of the ETFs studied generated negative Treynor ratios. This indicates, investors were not properly rewarded for the risk taken to invest in these particular ETFs. Most of the negative returns did not drift from zero. Two ETFs, HND & HNU both of which track the NYMEX Natural Gas Market generated the least return for the risk taken. Section 6.4 - Conclusions The empirical results presented m the preceding paragraphs provides interesting contrast about the performance (excess returns) of Canadian ETFs over a 5 year period. During the 5-year period of study, the terminal years (2007 to 2009) witnessed collapse of the stock market due to the financial crisis and the initial few years (2004 to 2007) represent a bull market. In that sense, the present study presents a balanced assessment of risk-reward of Canadian ETFs without being unduly influenced by only one phase of the business cycle. The ETFs performing well over 5 year period were international funds especially emerging market funds which witnessed high growth rates in the last decade (especially BRIC countries). The currency-hedged funds also performed relatively well reflecting the role of exchange rates in impacting the returns of cross-border investments. Commodity ETFs have generally shown mixed results: the bear commodity ETFs (reflecting the macroeconomic performance) have generally done well as compared to bull commodity 67 ETFs. The performance of commodity ETFs brings into focus the importance of aligning the macro-economic indicators into investment portfolio decisions. In terms of risk-reward, the results are somewhat different. The international and emerging market ETFs performed well in terms of positive and high alphas (excess returns) also had relatively high risks. In terms of risk-reward, the Canadian and US broad equity ETFs performed well (Treynor ratio of 0.11) while the fixed income ETFs had the lowest Treynor ratio ( -2.05). In terms of ranking, the currencyhedged ETFs funds performed relatively better than Canadian sector ETFs. The international and emerging market funds while displaying positive Treynor ratio's (risk-reward) demonstrated relatively modest performance. 68 Chapter 7 Conclusion Exchange Traded funds are the fastest growing investment product in capital markets today. ETFs closely track the performance of an index, sector, or region and offer the benefits of diversification and index tracking at a low cost. The total dollars deployed in ETFs is approaching US$700 billion globally which is nearly 15 per cent of $4.5 trillion held in traditional equity mutual funds. This study examines the riskreward of 74 Canadian sponsored ETFs across the three major sponsors representing nearly $29 Billion in assets under management. These account for nearly 14 per cent of trading in Toronto Stock Exchange. In modem finance the notion of risk is central to security analysis and portfolio selection. These risks are broadly classified as : Systematic (market risk) and Non Systematic (business specific risk). Systematic risk is attributable to broad macro-factors such as national income, employment, inflation and exchange rates. Non-Systematic risk factors effect individual securities or specific industries. The non-systematic component of risk can be managed through a well-diversified portfolio. It is the systematic risk which requires pricing. The most widely used indicators of systematic risks are (1) Standard Deviation (2) Beta (3) R -square (4) Alpha (5) Sharpe Ratio and the (6) Treynor Ratio. This study is based on the monthly data of 74 Canadian ETFs for the last 5 years (September 2004 to August 2009). The data relates to ETFs prices on a monthly basis and corresponding benchmark index such as the S&P 500 which is a 69 broad based market index. The returns on individual ETFs were generated from corresponding price data. For the purposes of this study, the top three Canadian investment firms offering ETFs products were chosen. We adopt the methodological framework of CAPM for risk-reward analysis. The CAPM framework is used to derive the risk of each ETF and to calculate alpha (a) - the excess return of each ETFs. The market risk of each ETF was calculated using the single index model (SIM). The excess return (a) is generated from the fundamental CAPM equation (Ri -Rf = ~ [E(Rm) - Rf]. The empirical results presented in the preceding paragraphs provides interesting contrast about the performance (excess returns) of Canadian ETFs over a 5 year period. During the 5-year period of study, the terminal years (2007 to 2009) witnessed collapse of the stock market due to the financial crisis and the initial few years (2004 to 2007) represent a bull market. In that sense, the present study presents a balanced assessment of risk-reward of Canadian ETFs without being unduly influenced by only one phase of the business cycle. The best performing ETFs over the 5 year period were international funds especially emerging market funds which witnessed high growth rates in the last decade (especially BRIC countries). The currency-hedged funds also performed relatively better reflecting the role of exchange rates in impacting the returns of cross-border investments. Commodity ETFs have generally shown mixed results: the bear commodity ETFs (reflecting the macro-economic performance) have generally done well as compared to bull commodity ETFs. The performance of commodity ETFs brings into focus the 70 importance of aligning the macro-economic indicators into investment portfolio decisions. In terms of risk-reward, the results are somewhat different. The international and emerging market ETFs performed well in terms of positive and high alphas (excess returns) but also had relatively high risks. In terms of risk-reward, the Canadian and US broad equity ETFs funds performed well (Treynor ratio of 0.11) while the fixed income ETFs had the lowest Treynor ratio (-2.05). In terms of ranking, the currency-hedged ETFs performed relatively better than Canadian sector ETFs. The international and emerging market funds displaying positive Treynor ratios (risk-reward) were the ETFs with relatively modest performance. The concept of risk-reward is one aspect of the evaluation of ETFs. It would be interesting to examine its performance (risk-reward) in terms of other competing funds like mutual funds, index funds. Similarly, the performance of these funds (ETFs, mutual and index funds) in terms of their cost - management expense ratio's (MER) is another area worth examining. These issues require further extension of the study. 71 Appendix 4.1: Fund Sponsor and Inception Date Fixed Income Fund Manager Symbol Inception Date 1-5 Year Laddered Government Claymore CLF Jan-08 CON Bond Index /Shares XBB Nov-00 CON Corporate Bond Index /Shares XCB Nov-06 CON Government Bond Index /Shares XGB Nov-06 CON DEX Short Term Bond Index /Shares XSB Nov-00 CON Long Term Bond Index /Shares XLB Nov-06 CON Real Return Bond /Shares XRB Dec-05 Premium Money Market Claymore CMR Feb-08 US 30 Year Bond Bear Plus Horizon HTD Jun-08 Canadian & US Broad Equity Fund Manager Symbol Inception Date NASDAQ 100 Bull Plus ETFS Horizon HQU Jun-08 S&P 500 Bull Plus ETFS Horizon HSU May-01 S&P/TSX60 /Shares XIV Sep-99 S&P/TSX Completion Index /Shares XMD Mar-01 RAFt Core Canada Feb-06 Claymore CRQ CON S&P!TSX Hedged to CON $ /Shares XSP May-01 S&P 500 Bear Plus Horizon HSD Jun-08 NASDAQ 100 Bear Plus Horizon HQD Jun-08 RAFt Core $US Hedged CON Claymore CLU Sep-06 S&P!TSX 60 Bull Plus ETFS Horizon HXU Jan-07 CON Russell 2000 Hedged to CON$ /Shares xsu S&P/TSX 60 Bear Plus Horizon HXD S&P!TSX Small Cap Index /Shares xes Canadian Sector Funds Fund Manager Symbol Inception Date S&P!TSX Capped Materials Fund /Shares XMA Dec-05 May-07 Jan-07 May-07 S&P/TSX Capped Energy Index /Shares XEG Mar-01 S&P!TSX Capped Composite Index /Shares XIC Feb-01 S&P!TSX Capped Financials Index /Shares XFN Mar-01 S&P!TSX REIT Index /Shares XRE Oct-02 S&P!TSX Income Trust Fund /Shares XTR Dec-05 S&P/TSX Capped Information Technology Fund /Shares XIT Mar-01 S&P/TSX Capped Financials Bear Plus Horizon HFD Jun-07 S&P!TSX Capped Financials Bull Plus ETFS Horizon HFU Jun-07 S&P/TSX Capped Energy Bull Plus ETFS Horizon HEU Jun-07 S&P!TSX Capped Energy Bear Plus Horizon HED Jun-07 72 Fund Manager Symbol Inception Date MSCI Emerging Markets Bull Plus Horizon HJU Ju/-08 Global Real Estate Claymore CGR Aug-08 Global Completion Portfolio Builder /Shares XGC Nov-08 BRIG Hedged CON Claymore CBQ Sep-06 Global Infrastructure Claymore C/F Aug-08 CON MSC/ EAFE Hedged to CON $ /Shares XIN Sep-01 RAFt Core International Claymore CIE Feb-07 International & Emerging Markets RAFt Core Japan Hedged CON Claymore CJP Feb-07 MSC/ Emerging Markets Bear Plus Horizon HJD Ju/-08 Currency Hedged Fund Manager Symbol Inception Date US Dollar US Bear Plus ETFS Horizon HOD Jun-08 US Dollar US Bull Plus ETFS Horizon HDU Jun-08 Commodities Fund Manager Symbol Inception Date NYMEX Natural Gas Bear Plus Horizon HND Jan-08 NYMEX Crude Oil Bear Plus Horizon HOD Jan-08 S&P Agribusiness N.A. Bear Plus Horizon HAD Mar-08 Global Mining Claymore CMW Jun-07 CLO Oil Sands Claymore Global Agriculture Claymore Global Water (S&P 500) Claymore cow cww Oct-06 S&P Agribusiness N.A. Bull Plus Horizon HAU Mar-08 S&P!TSX Global Base Metals Bull Plus Horizon HMU Jun-08 Dec-07 Jun-07 S&P!TSX Global Base Metals Bear Plus Horizon HMO Jun-08 NYMEX Crude Oil Bull Plus Horizon HOU Jan-08 GAS Commodity Claymore GAS Feb-08 NYMEX Natural Gas Bull Plus Horizon HNU Jan-08 Gold & Precious Metals Fund Manager Symbol Inception Date Comex Gold Trust /Shares IGT Jun-07 S&P!TSX Global Gold Bull Plus Horizon HGU Jun-07 CON S&P!TSX Global Gold Index /Shares XGD Mar-01 COMEX Gold Bullion Bull Plus Horizon HBU Jan-08 COMEX Gold Bullion Bear Plus Horizon HBD Jan-08 S&P!TSX Global Gold Bear Plus Horizon HGD Dec-05 Fund Manager Symbol Inception Date Growth Core Portfolio Builder /Shares XGR Nov-08 Preferred Share Claymore CPO Apr-07 Alternative Investment Strategy 73 Conservative Core Portfolio Builder /Shares XCR Nov-08 CON Dow Jones Growth Index /Shares XCG Nov-06 CON Dow Jones Dividend Index /Shares XDV Dec-05 CON Dow Jones Value Index /Shares XCV Nov-06 Canadian Dividend Claymore COl Sep-06 Income Balanced Claymore CBD Jun-07 CON Jantzi Social Index /Shares XEN May-07 Growth Balanced Claymore CBN Jun-07 Monthly Advantaged Hedged Claymore CYH Jan-08 74 Appendix 5.1: ETF Symbols Benchmark Index Symbol S&PfTSX 60 Bull Plus HXU S&PfTSX Capped Financials Bull Plus HFU S&PfTSX Capped Energy Bull Plus HEU S&PfTSX Global Gold Bull Plus HGU S&PfTSX Global Base Metals Bull Plus HMU COM EX Gold Bullion Bull Plus HBU NYMEX Crude Oil Bull Plus HOU NYMEX Natural Gas Bull Plus HNU S&P Agribusiness N.A. Bull Plus HAU S&P 500 Bull Plus HSU NASDAQ 100 Bull Plus HQU US Dollar US Bull Plus HDU MSCI Emerging Markets Bull Plus HJU Benchmark Index Symbol S&PfTSX 60 Bear Plus HXD S&PfTSX Capped Financials Bear Plus HFD S&PfTSX Capped Energy Bear Plus HED S&PfTSX Global Gold Bear Plus HGD S&PfTSX Global Base Metal Bear Plus HMO COM EX Gold Bullion Bear Plus HBD NYMEX Crude Oil Bear Plus HOD NYMEX Natural Gas Bear Plus HND S&P Agribusiness N.A. Bear Plus HAD S&P 500 Bear Plus HSD NASDAQ 100 Bear Plus HOD US Dollar US Bear Plus HOD MSCI Emerging Markets Bear Plus HJD Benchmark index Symbol RAFI Core Canada CRQ RAFI Core $US Hedged CON CLU RAFI Core International CIE RAFI Core Japan Hedged CON CJP Global Real Estate CGR Global Infrastructure CIF Global Mining CMW Global Agriculture cow cww Global Water (S&P 500) 75 Oil Sands CLO Equal Weight Balance CEW BRIG Hedged CON CBQ Canadian Dividend CDZ Monthly Advantaged Hedged CYH Preferred Share CPD 1-5 Year Laddered Government CLF Premium Money Market CMR Income Balanced CBD Growth Balanced CBN GAS Commodity GAS Canadian Equity Benchmark Index s&Prrsx 60 Symbol XIU S&Prrsx Capped Composite Index XIC S&Prrsx Completion Index XMD S&Prrsx Small Cap Index xes S&Prrsx Capped Energy Index XEG S&Prrsx Capped Financials Index XFN S&Prrsx Capped Information Technology Fund XIT S&PrfSX REIT Index XRE S&Prrsx Capped Materials Fund XMA S&Prrsx Income Trust Fund XTR Canadian Dow Jones Dividend Index XDV Canadian Dow Jones Growth Index XCG Canadian Dow Jones Value Index XCV Canadian Jantzi Social Index XEN Fixed Income Benchmark Index Symbol Canadian DEX Short Term Bond Index XSB Canadian Bond Index XBB Canadian Real Return Bond XRB Canadian Long Term Bond Index XLB Canadian Corporate Bond Index XCB Canadian Government Bond Index XGB Alternative Benchmark Index Symbol Canadian S&Prrsx Global Gold Index XGD Canadian S&Prrsx Hedged to CDN $ XSP Canadian Russell 2000 Hedged to CON $ xsu Canadian MSCI EAFE Hedged to CON $ XIN Conservative Core Portfolio Builder XCR Growth Core Portfolio Builder XGR Global Completion Portfolio Builder XGC Comex Gold Trust IGT 76 Appendix 6.1: Performance Analysis: Asset Class Category Expected Return (CAPM) Alpha 5.16% 1.48% 3.68% Total Return Symbol A. Fixed Income ETFs 1-5 Year Laddered Government CLF CON Bond Index XBB 5.05% 1.09% 3.96% CON Corporate Bond Index XCB 4.76% 1.32% 3.44% CON Government Bond Index XGB 4.64% 0.92% 3.72% CON DEX Short Term Bond Index XSB 4.45% 1.40% 3.05% CON Long Term Bond Index XLB 3.75% 0.99% 2.77% Preferred Share CPO 2.83% 4.06% -1.22% CON Real Return Bond XRB 2.42% 0.86% 1.56% Premium Money Market CMR 1.73% 1.46% 0.27% US 30 Year Bond Bear Plus HTD -17.20% 0.07% -17.27% 1.76% 1.37% 0.40% 87.27% 11.27% 75.99% Average B. Canadian & US Broad Equity ETFs NASDAQ 100 Bull Plus ETFS HQU S&P 500 Bull Plus ETFS HSU 39.78% 14.04% 25.73% S&P!TSX60 XIV 9.94% 7.11 % 2.83% S&P!TSX Completion Index XMD 5.83% 7.33% -1 .51% RAFI Core Canada CRQ 4.20% 7.32% -3.13% CON S&P!TSX Hedged to CON$ XSP -1 .83% 8.03% -9.85% S&P 500 Bear Plus HSD -3.34% - 10.32% 6.98% NASDAQ 100 Bear Plus HQD -7.56% -10.26% 2.70% 77 RAFI Core $US Hedged CON CLU -7.82% 8.90% -16.72% S&P!TSX 60 Bull Plus ETFS HXU -9.85% 12.82% -22.68% xsu -12.73% 0.15% -12.88% S&P!TSX 60 Bear Plus HXD -12.73% -8.09% 8.09% S&P!TSX Small Cap Index xes -15.95% 8.10% -24.05% 5. 79% 4.34% 2.43% -14.5% -3.94% 23.97% -27.9% 7.51 % 21.97% -21.2% 1.78% 22.97% CON Russell 2000 Hedged to CON$ Average C. Currency Hedged ETFs US Dollar US Bear Plus ETFS US Dollar US Bull Plus ETFS HDD HDU Average D. International & Emerging Markets ETFs MSC/ Emerging Markets Bull Plus HJU 158.91% 16.25% 142.66% Global Real Estate CGR 27.02% 7.38% 19.64 % Global Completion Portfolio Builder XGC 21 .97% 0.78% 21.19% BRIG Hedged CON CBQ 18.05% 11.66% 6.38% Global Infrastructure C/F 6.45% 6.21 % -0.24% CON MSC/ EAFE Hedged to CON$ XIN 0.72% 7.69% -6.97% RAFI Core International CIE -7.45% 6.66% 14.11% RAFI Core Japan Hedged CON CJP -23. 18% 7.80% -30.98% MSC/ Emerging Markets Bear Plus HJD -113.89% -4.73% -109.16% 9.84% 6.63% 3.21% Average E. Commodity ETFs NYMEX Natural Gas Bear Plus HND 156.91 % 0.83% 156.08% NYMEX Crude Oil Bear Plus HOD 55.65% -17.59% 73.24% S&P Agribusiness N.A. Bear Plus HAD 53.70% -8.40% 62.11 % 78 Global Mining CMW 0.03% Oil Sands CLO -1.31 % 9.61% -10.92% Global Agriculture cow -7.92% 5.36% -13.28% Global Water (S&P 500) cww -11.48% 5.74% -17.22% S&P Agribusiness N.A. Bull Plus HAU -25.27% 11.20% -36.47% S&P!TSX Global Base Metals Bull Plus HMU -28.23% 13.30% -41.53% S&P!TSX Global Base Metals Bear Plus HMO -53.09% -6.11 % -46.98% NYMEX Crude Oil Bull Plus HOU -81 .00% 15.92% -96.92% GAS Commodity GAS -126.48% -1.59% -124.89% NYMEX Natural Gas Bull Plus HNU -159.37% 1.50% -160.87% -17.53% 2.85% -20.38% Average 7.33% -7.30% F. Gold and Precious Metals ETFs Comex Gold Trust IGT 16.89% 1.33% 15.55% S&P!TSX Global Gold Bull Plus HGU 16.62% 4.76% 11.86% CON S&P!TSX Global Gold Index XGD 14.28% 2.94% 11.33% COMEX Gold Bullion Bull Plus HBU -1 .61 % 1.92% -3.53% COMEX Gold Bullion Bear Plus HBD -11.41% 0.52% -11.93% S&P!TSX Global Gold Bear Plus HGD -84.09% 5.78% -89.87% -8.22% 2.88% -11.10% Average G. Alternative Investment Strategy ETFs Growth Core Portfolio Builder XGR 20.85% 0.87% 19.98% Conservative Core Portfolio Builder XCR 13.40% 1.05% 12.35% CON Dow Jones Growth Index XCG 2.29% 1.35% 0.94% CON Dow Jones Dividend Index XDV 2.04% 2.47% -0.43% CON Dow Jones Value Index XCV 1.29% 2.38% -1 .10% 79 Canadian Dividend coz 1.14% 7.41% -6.27% Income Balanced CBD -1.32% 4.83% -6.14% CON Jantzi Social Index XEN -5.63% 1.39% -7.02% Growth Balanced CBN -12.08% 6.36% -18.44% Monthly Advantaged Hedged CYH -13.26% 10.15% -23.40% 0.87% 3.82% -2.95% Average 80 Appendix 6.2: Treynor Ratio of Various Canadian ETFs Category Fund Manager Symbol Date Total Return Treynor Ratio Fixed Income 1-5 Year Laddered Government CON Bond Index Claymore CLF Jan-08 5.16% -14.50 /Shares XBB Nov-00 5.05% -0.56 Corporate CON Bond Index CON Government Bond Index CON DEX Short Term Bond Index CON Long Term Bond Index CON Real Return Bond Premium Money Market US 30 Year Bond Bear Plus /Shares XCB Nov-06 4.76% -1.17 /Shares XGB Nov-06 4.64% -0.35 /Shares XSB Nov-00 4.45% -2.00 /Shares XLB Nov-06 3.75% -0.29 /Shares XRB Dec-05 2.42% -0.09 Claymore CMR Feb-08 1.73% -0.36 Horizon HTD Jun-08 -17.20% 0.85 1.64% -2.05 Average Canadian & US Broad Equity NASDAQ 100 Bull PlusETFS Horizon HQU Jun-08 87.27% 0.57 S&P 500 Bull Plus ETFS S&PITSX60 Horizon HSU May-01 39.78% 0.20 /Shares XIU Sep-99 9.94% 0.10 S&PITSX Completion Index RAF/Core Canada CDNS&PITSX Hedged to CON $ S&P 500 Bear Plus NASDAQ 100 Bear Plus RAFI Core $US Hedged CON S&PITSX 60 Bull PlusETFS /Shares XMD Mar-01 5.83% 0.05 Claymore CRQ Feb-06 4.20% 0.03 CON Russell 2000 Hedged to CON $ S&PITSX 60 Bear Plus S&PITSX Small Cap index /Shares XSP May-01 -1.83% -0.10 Horizon HSD Jun-08 -3.34% 0.03 Horizon HQD Jun-08 -7.56% 0.05 Claymore CLU Sep-06 -7.82% -0.08 Horizon HXU Jan-07 -9.85% -0.07 /Shares xsu May-07 -12.73% 0.69 Horizon HXD Jan-07 -12.73% 0.06 /Shares xes May-07 -15.95% -0.17 5.79% 0.10 Average 81 Canadian Sector Funds S&P/TSX Capped Materials Fund S&P/TSX Capped Energy Index S&P/TSX Capped Composite Index S&P/TSX Capped Financials Index S&PITSX REIT Index S&PITSX Income Trust Fund S&P/TSX Capped Information Technology Fund S&P/TSX Capped Financials Bear Plus S&P/TSX Capped Financials Bull PlusETFS S&P/TSX Capped Energy Bull Plus ETFS S&P/TSX Capped Energy Bear Plus /Shares XMA Dec-05 17.17% 0.22 /Shares XEG Mar-01 13.23% 0.12 /Shares XIC Feb-01 8.98% 0.09 /Shares XFN Mar-01 8.66% 0.07 /Shares XRE Oct-02 7.28% 0.06 /Shares XTR Dec-05 1.18% -0.02 /Shares XIT Mar-01 1.10% 0.00 Horizon HFD Jun-07 -15.29% 0.09 Horizon HFU Jun-07 -15.77% -0.08 Horizon HEU Jun-07 -24.63% -0.13 Horizon HED Jun-07 -30.74% 0.25 -2.62% 0.06 Average A. International & Emerging Market ETFs MSC/ Emerging Markets Bull Plus Global Real Estate Global Completion Portfolio Builder BRICHedged CON Global Infrastructure CON MSCI EAFE Hedged to CON $ RAF/ Core International RAFI Core Japan Hedged CON MSCI Emerging Markets Bear Plus Horizon -/JU Ju/-08 158.91 % 0.69 Claymore "JGR Aug-08 27.02% 0.28 /Shares (GC Nov-08 21 .97% -1.85 Claymore "JBQ Sep-06 18.05% 0.11 Claymore CIF Aug-08 6.45% 0.07 /Shares XIN Sep-01 0.72% -0.01 Claymore CIE Feb-07 -7.45% -0.11 Claymore CJP Feb-07 -23.18% -0.25 Horizon HJD Jul-08 -113.89% 1.20 9.84% 0.01 Average Currency Hedged ETFs. US Dollar US Bear PlusETFS US Dollar US Bull Plus ETFS Horizon "-/DO Jun-08 -14.5% -0.17 Horizon -IOU Jun-08 -27.9% 0.35 -21.2% 0.09 Average 82 Commodities ETFs Horizon HND Jan-08 156.91% -15.09 Horizon HOD Jan-08 55.65% -0.18 S&P Agribusiness N.A. Bear Plus Horizon HAD Mar-08 53.70% -0.34 Global Mining Claymore CMW Jun-07 0.03% -0.02 Oil Sands Claymore CLO Oct-06 -1 .31% -0.02 Global Agriculture Claymore -7.92% -0.16 Claymore cow cww Dec-07 Global Water (S&P 500) S&P Agribusiness N.A. Bull Plus Jun-07 -11 .48% -0.20 Horizon HAU Mar-08 -25.27% -0.18 S&P/TSX Global Base Metals Bull Plus Horizon HMU Jun-08 -28.23% -0.16 S&P/TSX Global Base Metals Bear Plus NYMEX Crude Oil Bull Plus GAS Commodity Horizon HMO Jun-08 -53.09% 0.47 Horizon HOU Jan-08 -81.00% -0.37 2.69 NYMEX Natural Gas Bear Plus NYMEX Crude Oil Bear Plus NYMEX Natural Gas Bull Plus Claymore GAS Feb-08 -126.48% Horizon HNU Jan-08 -159.37% Average 174062 .78 -17.53% -1.13 Gold & Precious Metals ETFs Comex Gold Trust /Shares IGT Jun-07 16.89% -6.01 S&P/TSX Global Gold Bull Plus CDNS&P!TSX Global Gold Index COMEXGold Bullion Bull Plus COMEXGo/d Bullion Bear Plus S&P!TSX Global Gold Bear Plus Horizon HGU Jun-07 16.62% 0.30 /Shares XGD Mar-0 1 14.28% 0.58 Horizon HBU Jan-08 -1.61% -0.48 Horizon HBD Jan-08 -11.41% 0.86 Horizon HGD Dec-05 -84.09% -1.30 -8.22% -1.01 Average Alternative Investment Strategy Growth Core Portfolio Builder /Shares XGR Nov-08 20.85% -2.00 Preferred Share Claymore CPO Apr-07 2.83% 0.03 Conservative Core Portfolio Builder /Shares XCR Nov-08 13.40% -1 .71 CON Dow Jones Growth Index /Shares XCG Nov-06 2.29% -0.33 CON Dow Jones Dividend Index /Shares XDV Dec-05 2.04% 0.04 CON Dow Jones Value Index /Shares XCV Nov-06 1.29% -0.02 83 Canadian Dividend Claymore CDZ Sep-06 1.14% 0.00 Income Balanced Claymore CBD Jun-07 -1.32% -0.06 CON Jantzi Social Index Growth Balanced /Shares XEN May-07 -5.63% 4.10 Claymore CBN Jun-07 -12.08% -0.18 Claymore CYH Jan-08 -13.26% -0.11 1.05% -0.02 Monthly Advantaged Hedged Average 84 References Agrawal Pankaj, & Clark John, "ETF Betas: A Study of their Estimation Sensitivity to Varying Time Intervals", Institutional Investors Guide ETF and Indexing, Fall2007, pp 96-103. Anderson, Tom (2008) "Real Assets Inflation Protection Solutions with Exchange-Traded Products", Institutional Investors Journal, 18(1 ), pp.l424. Bloomberg, "Tax Advantages of ETFs ", Financial Planning Magazine May, 2002 Bos,T, & Newbold, P, "An Empirical Investigation of the Possibility of Stochastic Systematic Risk in the Market Model", Journal of Business, 57(1), 1984, pp 35-41 Campbell, John Y, (1993) "Inter-temporal Asset Pricing with-out Consumption Data". American Economic Review, 83(3), pp, 487-512. Cleary, Sean Wand Jones, Charles, P Investment Analysis and Management, John Wiley & Sons Canada, 2009 Damodaran, Aswath "Corporate Finance Theory and Practice", John Wiley & Sons. (200 1) De Giorgi, (2008) "Second Order Stochastic Dominance, Reward-Risk Portfolio Selection and the CAPM", Journal of Financial and Quantitative Analysis, 43(2), pp 525-546. Dieckerman Raimar , Deutsche Bank Research, Financial Market Special, EU Monitor 55, 2008, p.3 Demaine John (2002) Exchange Traded Funds for the Sophisticated Investor, Henry Stewart publications Dimson, E,P, Marsh, R & Staunton M (2002) Triumph of the Optimists: IOJ years of Global Investment Returns, Princeton University Press. Fabozzi Frank & Cark, Jack Francis "Beta as a Random Coefficient", Journal of Financial & Quantitative Analysis, 13(1) 1978, pp 101-116 Fama Eugene and French, Kenneth "Business Conditions and Expected Returns on Stocks and Bonds", Journal of Financial Economics, 25(1), 1989, pp.23-49. 85 Fama Eugene and French, Kenneth "Cross Section of Expected Stock Returns", Journal of Finance, 47(2), 1992, pp.427-65. Fama Eugene and French, Kenneth "The Capital Asset Pricing Model: Theory & Evidence", Journal of Economic Perspectives, 18(3), 2004, pp 25-46 Ferri, Richard The ETFs Book- All you need to know about Exchange Traded Funds, New Jersey, John Wiley & Sons, 2008, pp 3-57. Filbeck, Tompkins, "The Option to Issue, Performance of Mutual Funds", Journal of investing, 13(2), 2004, pp 72-80 Frino A, Gallagher, D "Tracking S&P 500 Index Funds", Journal of Portfolio Management, 28(1) 2001, pp 44-55 Gastineau Gary L, "Exchange Traded Funds: An Introduction", Journal of Portfolio Management, 27(3), 2001, pp. 88-96 Gastineau Gary L., "Will Exchange Traded Funds Replace Conventional Mutual funds". The Merrill Lynch Center for the study of International Financial Services and Markets, Hofstra University, Frank J. Zarb School of Business, May 5, 2003 Graham, John R, and Campbell Harvey. "The Theory and Practice of Corporate Finance: Evidence from the Field". Journal of Financial Economics, 60(2), (May/June 2001), pp 187-243 Haugen and Heins, "Risk and the Rate of Return on Financial Assets: Some Old Wine in New bottles", Journal of Financial and Quantitative Analysis, 10(5) December 1975, pp 775-84 Harding, David. "A critique of the Sharpe Ratio" Winton Capital Management, London England Hedges, Peggy, UNBC Finance Commerce 620, Power Point Presentation chapter 11, slide 11 class presentations, spring 2009 Holt, Charles A. and Laury, Susan K. "Risk Aversion and incentive Effects", The American Economic Review 92(5), pp 1644-1655 Gou, Hui. "On the Real Time Forecasting Ability of the Consumption Wealth Ratio", Journal of Financial and Quantitative Analysis, 42(1) 2003b Jegadeesh, Narashimham & Titman, Sheridan. "Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency" Journal of Finance, 48(1), March 1993, pp 65-91. 86 Jordan, Miller, Yuce, "Fundamentals of Investments", Management, Macgraw Hill Ryerson, 2006, pp 426-427 Valuation and King, Michael R. "The Cost of Equity for Global Banks: a CAPM Perspective" from 1990 to 2009. BIS Quarterly review September 2009 Linter, John "The Valuation of Risk Assets and the Selection of Risky Investments in Stock Portfolios and Capital Budgets", Review of Economics and Statistics, 47(1), 1965 pp. 13-37. Markowitz, Harry "Portfolio Selection: Efficient Diversification oflnvestments", John Wiley & Sons, New York, Chapman & Hall Limited, London 1959. Merton, Robert. "An intertemporal Capital asset Pricing Model" Econometrica, 41(5) September 1973, pp.876-87 Murphy, J, Michael, "Efficient Markets, Index Funds, Illusion and Reality", Journal ofPortfolio Management 41(1) Fall1977, pp 5-20 Moskowitz, Tobias, "Mutual Fund Performance: An Empirical Composition into Stock Picking, Transaction Costs and Expenses", Journal of Finance, 55(4), 2000, pp 1675-1703 NYSE Area, Factset & BGI, July 2008 Perold, AF," The Capital Asset Pricing Model", Journal of Economic Perspective 18(3), 2004, pp 3-24 Poterba & Shoven, Exchange Traded Funds: "A New Investment Option for Taxable Investors". The American Economic Review, 92(2), May 2002, pp 422-427 Rompotis, Gerasimos, "Active vs. Passive Management", new evidence from Exchange Traded Funds, University of Athens, Greece, February 2009 Rongala, Sunil, "Exchange-Traded Funds Challenging the Dominance of Mutual Funds?" Deliotte Research. Deloitte Support Services India Private Limited, 2009 Santolli, Michael (2009) 'Growing to the Sky', Barron's, October 19: 1-4. Sharpe, William (1963a) "Risk, Market Sensitivity and Diversification", Financial Analysts Journal, 17(4), pp 74-78. _ _ (1963b) "A Simplified Model for Portfolio Analysis", Management Science, 9(2), pp.277-293. 87 _ _ (1964) "Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk", Journal of Finance, 19(3), pp. 425-42. Scholz, Hendrick & Wilkens, Marco. "The Sharpe Ratio ' s market Climate Bias Theoretical and Empirical Evidence from US Equity Mutual Funds", Ingosladt School of Management, Catholic university of EichsteattInglostadt, Germany, 2005 pp. 1-3 State Street Global Advisors and Knowledge @ Wharton "The Impact of Exchange Traded Products on the Financial Industry" June 10, 2008 Treynor, JL, "How to rate Management of Investment Funds", Harvard Business Review, 43, 1965, pp-63-75 Wermers, Russell "Mutual Fund Performance: An Empirical Decomposition into Stock Picking, Transaction Costs and Expenses, Journal of Finance, 57(2), 2000, pp Wiandt, Jim "How ETFs Manage a Tax-Efficiency Edge over Traditional Mutual Funds", Exchange Traded Funds - An Insider's Guide to Buying the Market", John Wiley & Sons, 2002 Womack, John "Understanding Risk & Return, The CAPM, The Fame-French Three Factor Model", Tuck School of Business at Dartmouth, Trustees of Dartmouth College, 2003 , pp 4-14 Yansen, Ali "Simplifying the Portfolio Optimization via Single Index model", Industrial Engineering Honors Program, McCormick School of Engineering Northwestern University, June 2008 Zopoundis, Doumpos, Pardalos "Exchange Traded Funds: History, trading and Research" Handbook of Financial Engineering, Springer Science & Media Center, 2008, pp 67-70 88 Internet Sources 1. Arbitrage Mechanism http://news. morni ngstar. com/arti c Ienet/ arti c lc.aspx? id=2 83 302 (Accessed December 23 , 2009) 2. Claymore Funds Website (Assets under Management): http://www.claymoreinvestments.ca/about/about-claymore/ (Accessed on December 12, 2009) 3. Claymore Funds Website: http://www.claymoreinvestments.ca/about/about-claymore/ (Accessed on December 12, 2009) 4. Exchange Traded Fund, Street Authority.com. http ://www.streetauthority.com/terms/e/ETFs/.asp (Accessed October 12, 2009) 5. ETFS Advisor Handbook Http://www .advisor.ca/advisors/news/industrynews/articlc. j sp?content=20091 030 152400 10684 (Accessed September 19, 2009) 6. ETFs at a Glance - October 2009 and YTD October 31 2009 Exchange Traded Funds - ETFs Market Statistics I TMXmoney (Accessed November 14, 2009) 7. Horizon Beta Pro Funds (Assets under Management): http ://www .tmxmoney.com/en/sector profiles/exchange traded funds/funds/fund s.html (Accessed on December 12,2009) 8. Horizon Beta Pro Currency Fact Sheet http://www.hbpETFs .com/ indexPerfonnance.asp (Accessed on Dec14, 2009) 9. iShares Canada (Assets under Management): http://www.tmxmoney.com/en/sector profiles/exchange traded funds/funds/fund s.html (Accessed on December 12, 2009) 10. Investment Company Institute Fact-book 2009 Section 3: Exchange-Traded Funds (Accessed vanous times September December 2009) 11. Kirzner, Eric, "Get ready for the i60's" Canadian Investment Review (Accessed November 14, 2009) 12. Markowitz Efficient Frontier Model Efficient Frontier http://www.investopedia.com/terms/e/efficientfrontier.asp (Accessed December 13, 2009) 89 13. Morningstar Website http ://news.momingstar.com/ articlenetlarticle.aspx?id= I I 8018 (Date accessed December 14, 2009) 14. Risk Reward Pyramid Source: http://www.investopedia.com/articles/basics/03/050203 .asp?viewed= 1 (Accessed December 12, 2009) 15. TSX ETFS listings surpass 100 Http://www.investmentexecutive.com/client/en/News/DetailNews.asp?Id=5001 1 &cat= 148&IdSection= l48&PageMem=&nbNews (Accessed November 7, 2009) 16. Vanguard Group Website https ://personal.vanguard.com/us/whatweoffer/mutualfundinvesting/overview?Lin k=DeckB (accessed on December 23, 2009) 90