LO 63.5: Compare value and momentum investment strategies, including their risk and return profiles.
The fact that small stocks tend to outperform big stocks, after adjusting for the firms beta, was discovered by Banz (1981 )3 and similarly by Reinganum (1981).4 Following the publication of this finding, the effect disappeared. In other words, if you examine the returns to an SMB strategy from 1965 to 2011, returns to the strategy peak in the early 1980s, with no evidence of a small stock premium in subsequent years. The two possible explanations for the disappearing size effect are as follows: Data mining. Fischer Black (1993)5 suggested data mining following the publication of
the Fama and French study. If a finding is discovered with in-sample data (i.e., in the data used in the original study) but is not substantiated in further studies using out-of-sample data, then data mining provides a possible explanation for the result. Investor actions. Upon the publication of the Banz and Reinganum studies, investors, acting rationally, bid up the prices of small-cap stocks until the SMB effect was removed. This is consistent with the efficient market hypothesis (EMF1) in which investors exploit anomalies until they can no longer profit from them. If this is true, then size should be removed as a risk factor in the Fama-French model.
Note that small stocks do tend to have higher returns (i.e., weak size effect), partially because they are less liquid than large-cap stocks. Also, the value and momentum effects, discussed next, are stronger for small stocks. Flowever, the ability to capture small-cap excess returns over the market (on a risk-adj usted basis) is no longer present.
Value Investing
Unlike the disappearing size premium, the value risk premium has provided investors with higher risk-adjusted returns for more than 50 years. Value strategies have suffered periods of loss, including the 1990s recession, the dot com bull market of the late 1990s, and the 20072009 financial crisis. The notion of value investing dates back to when Graham and Dodd (1934)6 published Security Analysis with a focus on finding stocks that had prices lower than their fundamental values.
There are generally two explanations for the value premium, one rational and the other behavioral.
Rational Theories o f the Value Premium
Value stocks move with each other and co-vary with growth stocks in the rational story about the reason a value premium exists. They perform well together and poorly together.
3. Rolf W. Banz, The Relationship Between Return and Market Value of Common Stocks,
Journal o f Financial Economics 9 (1981): 3-18.
4. Marc R. Reinganum, Misspecification of Capital Asset Pricing: Empirical Anomalies Based on
EarningsYields and Market Values, Journal of Financial Economics 9, no. 1 (1981): 19-46. 5. Fischer Black, Beta and Return, Journal ofPorfolio Management 20, no. 1 (1993): 8-18. 6. Benjamin Graham and David Dodd, Security Analysis (New York: McGraw-Hill, 1934).
Page 22
2018 Kaplan, Inc.
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
Value is risky and, as such, value stocks sometimes perform poorly. The value premium is compensation for these periods of poor performance, for losing money during bad times. Value did perform poorly during the bull market in the late 1990s. This means rational stories must define bad times and that value earns a premium on average, not all of the time. Also, not all value risk can be diversified away. The remaining value risk is captured in the value premium.
Labor income risk, investment growth, luxury consumption, long-run consumption risk, and housing risk are factors that have been used to explain the value premium. Value stock betas often increase during bad times defined by these risks, causing value stocks to be particularly risky. Macro-based and CAPM risk factors turn out to be the same factors that affect value firms.
Consider the difference between growth and value firms. Growth firms are more adaptable and can adjust when times change because the bulk of their capital is human capital. Value firms are more old school with capital in the form of fixed assets that cannot be redeployed when times change. Thus, value firms have high and asymmetric adjustment costs. This makes value stocks fundamentally more risky than growth stocks.
The average investor holds the market portfolio. Some investors choose a value tilt and others a growth tilt. The decision boils down to how well the investor can withstand bad times. Given the factors defined previously as bad for value (i.e., labor income risk, investment growth, etc.), the investor must ask himself, Are these times bad for me (versus bad in general)? If, for example, an investor can manage well during times of low investment growth, that is not a bad time for that investor relative to the average investor. The investor, who has a comparative advantage in holding value stocks, can bear value risk and, therefore, can earn the value premium.
Behavioral Theories o f the Value Premium
Behavioral theories of the value premium revolve around two basic ideas: (1) overextrapolation and overreaction and (2) loss aversion and mental accounting.
Overextrapolation and overreaction. Investors have a tendency to assume that past growth rates will continue in the future. This is called overextrapolation. For example, a technology company may have a period of tremendous growth as it developed new products that are in high demand. Many investors may assume that this company will continue this growth into the fixture. Investors often bid up the prices of growth stocks beyond their intrinsic values due to unwarranted optimism. Prices fall when the high expected growth doesnt materialize, leading to lower returns than those earned on value stocks.
Loss aversion and mental accounting. Investors dislike losses more than they like gains (i.e., loss aversion), and they tend to view investment gains and losses on a case-by-case basis rather than on a portfolio basis (known as mental accounting). Barberis and Huang (2001)7 use this notion to explain the value premium. They argue that the reason value stocks have high book-to-market values is that they have undergone a period of very poor
Nicholas Barberis and Ming Huang, Mental Accounting, Loss Aversion, and Individual Stock Return s ) Journal o f Finance 56, no. 4 (2001): 1247-92.
2018 Kaplan, Inc.
Page 23
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
performance. Loss-averse investors view the stock as risker and, therefore, require a higher rate of return.
Professors Note: The extrapolation/overreaction behavioral explanation o f the value premium is different from the rational one in that in the behavioral explanation, value stocks are not riskier, they are just cheap relative to growth stocks. Investors tend to underestimate the growth prospects o f value stocks and overestimate the growth prospects o f growth stocks. This bids up the prices o f growth stocks and bids down the prices o f value stocks, allowing value stocks to outperform on average. Investors must determine i f they tend to overextrapolate or not. Investors who act like other average, non-over or under-reacting investors should hold the market por folio. Investors who overextrapolate will lean toward growth stocks, and those who underreact will lean toward value stocks.
Why are there not enough value investors in the market to push up prices and remove the value premium, as described in the section on the small-cap effect? Maybe investors find value investing difficult, although it is easy to sort stocks on a book-to-market basis using internet screening tools. Perhaps investment horizons must be too long to engage in value investing. The book-to-market value effect described here requires at least a three month to six month horizon. It is possible that not enough institutions have a long enough investment horizon to adopt a value investing approach.
Value investing exists in all asset classes. Strategies include: Riding the yield curve in fixed income (i.e., capturing the duration premium). Roll return in commodities (i.e., an upward or downward sloping futures curve
determines the sign of the return).
Carry in foreign exchange (e.g., long positions in currencies with high interest rates and short positions in currencies with low interest rates). In this case, high yields are akin to low prices in equity value strategies.
Retail investors can implement value strategies via low-cost index products. Large, institutional investors can, at least theoretically, cheaply implement value strategies across markets.
Momentum Investing
In 1993, the same year Fama and French captured the size and value/growth effects, Jagadeesh and Titman8 identified a momentum effect. Momentum strategies (also called trend investing) consist of buying stocks that have gone up over a period (e.g., six months or so) and short stocks that have fallen over the same period (i.e., buy past winners and sell past losers). The momentum factor, WML, stands for winners minus losers. It is also sometimes denoted UMD for up minus down, buying stocks that have gone up in price and selling stocks that have gone down in price. A momentum premium is observed in fixed income (government and corporate bonds), international equities, commodities, real estate, and specific industries and sectors.
Narasimhan Jegadeesh and Sheridan Titman, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, Journal o f Finance 48, no. 1 (1993): 65-91.
Page 24
2018 Kaplan, Inc.
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
The returns to momentum investing exceed size and value investing premiums by a wide margin. Figure 2 illustrates the differences in returns across the three strategies. One dollar invested in the WML premium in January 1963 reached a high of more than $60 before following precipitously (below $30) during the 20072009 financial crisis. Correlation between the value premium and the momentum premium was only approximately 0.16 during this period. This means that value returns are not opposite momentum returns.
Figure 2: Returns for SMB, HML, and WML strategies
Year
Value and momentum strategies are, however, opposite each other in the following sense. Value investing is inherently stabilizing. It is a negative feedback strategy where stocks that have fallen in value eventually are priced low enough to become value investments, pushing prices back up. Momentum is inherently destabilizing. It is a positive feedback strategy where stocks that have been increasing in value are attractive to investors, so investors buy them, and prices increase even more. Momentum investing can lead to crashes (e.g., the more than 30% drop during the 20072009 financial crisis). Notice that value and growth returns did not fall in quite so dramatic a fashion. An investor following a momentum strategy should still rebalance his portfolio.
Momentum is often added to the Fama-French model as follows:
E(Ri) Rp + (3ijMKT x E(Rm Rf ) + Pi,SMB x E(SMB) + Pi5HML x E(HML)
+ Pi,WML x E(WML)
As mentioned, momentum can be riskier than value or size investing in that it is more prone to crashes. There have been 11 momentum crashes on record: seven during the 1930s Great Depression, three during the financial crisis starting in 2007, and one in 2001. During the 20072009 crisis, financial stocks were hit hard. Losers tend to keep losing, and they likely would have, but the government bailout put a floor on stock prices. Momentum
2018 Kaplan, Inc.
Page 25
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
investors were short these stocks. When the government bailed out financial firms and other firms that were hit hard, momentum investors experienced large losses as the market rebounded. During the Great Depression, policymakers also influenced asset prices, causing losses to momentum investors.
Momentum risk includes: Tendency toward crashes. Monetary policy and government risk (i.e., the government gets in the way of the natural
progression of asset prices).
Macro factors such as the business cycle, the state of the stock market, and liquidity risk. Behavioral explanations suggest that investor biases explain momentum. Investors overreact (a delayed overreaction) to good news about firms. This causes prices to drift upward. Alternatively, investors may underreact to good news, causing prices to increase less than they should have given the good news. As investors acquire more information, prices go up in the next period. Thus, behavioral explanations for the momentum premium fall into two, difficult-to-distinguish camps: 1. Overreaction to good news. In some cases overconfident, informed investors, like retail
investors and hedge fund managers, observe positive signals in stock performance. They attribute the performance to their own skill. The overconfidence leads to overreaction, pushing prices up above their fundamental values.
2. Underreaction to good news. In this case, news watchers ignore information in the
history of stock prices and other investors trade only on history (i.e., past price signals) and ignore fundamental information about the firm. In both cases, information is only partially incorporated into stock prices, causing an underreaction.
Whether there is momentum that results from overreaction or from underreaction, prices eventually revert to their fundamental values over the long run. An investor considering momentum investing must assess whether he leans toward overreaction or underreaction. Also, the investor must know that he can tolerate large losses during crash periods, historically concentrated around periods when policymakers (e.g., central banks) interrupt momentum, changing the course that asset prices would naturally take. In sum, assets are exposed to factor risks like value and momentum. Factor premiums compensate investors for losses during bad times.
Page 26
2018 Kaplan, Inc.
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
Ke y Co n c e pt s
LO 63.1 A value-growth investment strategy is long value stocks and short growth stocks. Value stocks have high book-to-market ratios, and growth stocks have low book-to-market ratios. Historically, value stocks have significantly outperformed growth stocks.
Risk premiums, including a value premium, exist to compensate investors for losses experienced during bad times. There are rational and behavioral explanations for why a value premium may exist. Value stocks are risky, thus the value premium compensates investors for losses during bad times (rational explanation). Investors undervalue the growth prospects of value stocks and overextrapolate past growth into future prospects, overvaluing growth stocks. Value stocks are underpriced relative to their fundamental values, and growth stocks are overvalued, leading to a value premium (behavioral explanation).
LO 63.2 Macroeconomic factors, like inflation and economic growth, affect all investors to varying degrees. Economic growth, inflation, and volatility are the three most important macro factors that affect asset prices. It is unanticipated changes to a risk factor that affect asset prices, not the level of the factor. In other words, it is not the level of inflation, but an unanticipated increase or decrease in inflation that causes stock and bond prices to rise or fall. Risky assets generally perform poorly during periods of low economic growth.
Other macroeconomic factors, like shocks to productivity, demographic risks, and sovereign risks, also affect asset returns.
Stocks and bonds generally perform poorly in periods of high inflation. Stock returns drop when volatility (measured by the VIX) increases.
LO 63.3 Volatility increases in periods of economic stress. There are two basic approaches to mitigating volatility risk:
Invest in less-volatile assets like bonds. One challenge to managing volatility is that asset prices, including less volatile assets, tend to perform poorly during periods of economic stress (e.g., 20072009).
Buy volatility protection in the derivatives market (e.g., buy out-of-the-money put
options). Sellers of volatility protection (i.e., those selling put options) collect volatility premiums.
2018 Kaplan, Inc.
Page 27
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
LO 63.4 The Fama-French model explains asset returns based on three dynamic factors. The factors are: The traditional CAPM market risk factor. A factor that captures the size effect (SMB or small cap minus big cap), historically,
small-cap stocks outperform large-cap stocks. The strategy is long small-cap stocks and short large-cap stocks.
A factor that captures the value/growth effect (F1ML or high book-to-market value
minus low book-to-market value). Value stocks tend to outperform growth stocks. The value-growth strategy is long value stocks and short growth stocks.
LO 63.3 A value strategy is long value stocks and short growth stocks. A momentum strategy is long winners (i.e., stocks that have gone up in value over the last six months or so) and short losers (i.e., stocks that have gone down in value over the last six months or so). A momentum strategy has vastly outperformed both value-growth and size strategies since 1963. Flowever, momentum strategies are subject to crashes. Rational and behavioral explanations can be used to describe both value and momentum risk premiums.
Page 28
2018 Kaplan, Inc.
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
Co n c e pt Ch e c k e r s
1.
2.
3.
4.
3.
A low book-to-market value ratio is indicative of a: A. value stock. B. growth stock. C. small-cap stock. D. large-cap stock.
Which of the following asset classes has approximately the same returns in high economic growth periods and low economic growth periods? A. Small-cap stocks. B. Large-cap stocks. C. Government bonds. D. High-yield bonds.
Which of the following investment options provides a means of mitigating volatility risk? A. Buying put options. B. Selling put options. C. Buying equities. D. Buying call options.
Which of the following is not a factor in the Fama-French three-factor model? A. The capital asset pricing model market risk factor. B. The small capitalization minus big capitalization risk factor. C. The winners minus losers risk factor. D. The high book-to-market value minus low book-to-market value risk factor.
Which of the following investment strategies stabilizes asset prices? A. A value investment strategy. B. A momentum investment strategy. C. A size investment strategy. D. Value, momentum, and size strategies all stabilize asset prices.
2018 Kaplan, Inc.
Page 29
Topic 63 Cross Reference to GARP Assigned Reading – Ang, Chapter 7
Co n c e pt Ch e c k e r An s w e r s
1. B A companys book value per share is equal to total assets minus total liabilities all divided by shares outstanding. It indicates, on a per-share basis, what a company would be worth if it liquidated its assets and paid off its liabilities. Value stocks have high book-to-market ratios while growth stocks have low book-to-market ratios.
2. D During periods of recession, government and investment-grade bonds outperform equities and high-yield bonds. During expansion periods, equities outperform bonds. High-yield bond returns appear indifferent to changes in economic growth, yielding 7.4% in recessions and 7.7% in expansions.
3. A There are two basic approaches to mitigating volatility risk. They are investing in less volatile assets like bonds (instead of stocks) or buying volatility protection in the derivatives market, such as buying out-of-the-money put options.
4. C The Fama-French model includes the following three risk factors: The traditional capital asset pricing model market risk factor.
A factor that captures the size effect (SMB). A factor that captures the value/growth effect (HML).
The winners minus losers (WML) momentum factor was discovered by Jagadeesh and Titman.
5. A Value and momentum are opposite each other in that value investing is inherently stabilizing.
It is a negative feedback strategy where stocks that have fallen in value eventually are priced low enough to become value investments, pushing prices back up. Momentum is inherently destabilizing. It is a positive feedback strategy where stocks that have been increasing in value are attractive to investors, so investors buy them, and prices increase even more.
Page 30
2018 Kaplan, Inc.
The following is a review of the Risk Management and Investment Management principles designed to address the learning objectives set forth by GARP. This topic is also covered in:
A l ph a (a n d t h e Lo w -Ri s k A n o m a l y )
Topic 64
Ex a m Fo c u s
Investors are interested in generating alpha, which is the return earned in excess of a benchmark. It was traditionally thought that higher risk produced higher returns. However, in practice, strategies focused on lower volatility have actually been found to produce higher returns than higher-volatility investments. For the exam, be able to explain the impact of benchmark section on alpha. Also, understand how to apply factor regression to construct a benchmark with multiple factors, and how to measure alpha against that benchmark. Finally, be familiar with the potential explanations for return anomalies with regard to low risk.
Lo w -Ris k An o m a l y