Investments

CAPM performance, Value and

Momentum

Nicholas Hirschey
Nova SBE
 Overview
• Last time
– CAPM: same beliefs → market = tangency portfolio
– Greatly simplifies portfolio/discount rate problems
– Also useful for performance analysis

• Today
– How well does the CAPM explain returns in practice?
• Not very well
– Then what explains differences in returns across securities?
• Size—small stocks have higher returns than large stocks
• Value—value stocks have higher returns than growth stocks
• Momentum—stocks that did well in the past do well in the future

Lecture - CAPM Perf, Value & Momentum Page 2

 Testing the CAPM
• What should we expect if CAPM is true?
– E[r] are a linear function of 𝛽 and the market risk premium
– Only 𝛽 explains expected returns
• Caveat: the Roll critique
– What if you find that 𝛽 (and only 𝛽) explains returns?
• How do you know your market proxy is “the market”?
– What if 𝛽 doesn’t explain returns?
• You may just have a bad market proxy.

• Still, CAPM tests check efficiency of the market proxy

Lecture - CAPM Perf, Value & Momentum Page 3

 Fama and French (1992)
• Form 10 portfolios on the past 5-year’s Beta
• Look at Next year’s Return
(1963-1990)
• High beta stocks 18
16

Annual % Return
earn no higher 14

return than low- 12

10
beta stocks. 8
6
• Beta is DEAD! 4
2
In 1926-3/2012 period, Frazzini and
Pederson (2013 JFE) show low-beta 0
stocks earn 10.9% pm vs. 11.64% for 0.80 0.92 1.04 1.13 1.19 1.26 1.32 1.41 1.52 1.68
high-beta stocks. Portfolio Betas the next period

Lecture - CAPM Perf, Value & Momentum Page 4

 Why doesn’t beta work?
• Wrong theory
– Behavioral biases
– No relation between return and marginal risk
– Friction-less market assumption in CAPM invalid.
• Leverage? Non-tradable risks?
• Wrong test
– Tests don’t capture the correct market portfolio
– If you had the right portfolio, you’d see the risk priced

• Frictions (leverage, non-tradable risks) are likely an important part of the explanation.

• CAPM failure has motivated search for other risk factors

– In practice leads to wide-spread adoption of multi-factor models, whose basis is the APT (coming
soon).

Lecture - CAPM Perf, Value & Momentum Page 5

 So what did the CAPM teach us?
• We learned what is meant by a “priced” risk factor
– If a risk factor is “priced”, that means that investors require compensation for exposure
to the factor.
– Thus, variation in betas on a priced risk factor should explain expected returns.
• This is how Fama, French (1992) tested the CAPM

• The theory is appealing and may work in the future

– Do you really think market exposure is not risky?
• Seems likely that asset managers with high beta portfolios going into the 2008/2009 crisis think
beta is risky today.
– What happens if everyone realizes that high-beta securities do not have higher expected
returns?
• Despite empirical shortcomings … the CAPM is still widely used in practice.

Lecture - CAPM Perf, Value & Momentum Page 6

 The leverage explanation?
• We have assumed investors can leverage the tangency
portfolio.

• Black (1972 JB) argues that one explanation for a flatter-than-

CAPM relationship between Beta and expected return is
borrowing constraints.
– Investors that want high returns but have borrowing constraints will
hold high standard deviation portfolios with lower Sharpe ratios than
the tangency portfolio.
– With borrowing constraints, “model predicts only that the slope of the
line relating expected returns and beta is positive”

• Frazzini, Pedersen (2013 JFE) extend and test this idea in

detail.

Lecture - CAPM Perf, Value & Momentum Page 7

 Portfolio selection with constraints

Tangency Example from Frazzini,

Portfolio Constrained frontier Pedersen where some
investors are constrained
to hold a fraction of their
portfolio in the risk-free
asset.

D’ is a combination of the risk-

free asset and portfolio D.

Where constraint on investment in risk-free bond binds

Lecture - CAPM Perf, Value & Momentum Page 8

 “Betting on Beta”
• Title of Frazzini, Pedersen (2013 JFE) paper.
– Maybe leverage-constrained investors bid up the price of high-beta stocks,
lowering their expected returns.

• Black (1992 JPM)

– “… if the line is really flat, that implies dramatic investment opportunities for
those who use beta. A person who normally holds both stocks and bonds or
stocks and cash can shift to a portfolio of similar total risk but higher expected
return by emphasizing low-beta stocks.” pg. 75

Lecture - CAPM Perf, Value & Momentum Page 9

 Sharpe ratios of Beta-sorted Portfolios
From Frazzini, Pedersen (2013 JFE) Data: 1926-2012

Lecture - CAPM Perf, Value & Momentum Page 10

 Betting on Beta cont’d
• So unconstrained investors could leverage low-beta securities and earn
better risk-adjusted returns.

• Frazzini, Pedersen find the result (low-beta SR > high-beta SR) in a

variety of asset classes.
– US/Int. Equities, Gov/Corp. bonds, futures.

• Is the market inefficient? Or maybe just constrained?

Lecture - CAPM Perf, Value & Momentum Page 11

 If 𝛽 doesn’t work, does anything?
• A primary motivation for CAPM is identifying expected returns and
discount rates.

• While 𝛽 does not explain returns, certain stock characteristics are

consistently related to future returns:
– Small stocks earn higher returns
• Banz (1981); Fama, French (1992)
– Value stocks earn higher returns
• Basu (1977); Fama, French (1992); Lakonishok, Shleifer, Vishny (1994)
– Stocks with high past returns have high future returns
• Jegadeesh, Titman (1993); Jegadeesh, Titman (2001)

Lecture - CAPM Perf, Value & Momentum Page 12

 Value Investing
• Value investing is an idea developed
by practitioners.
– Ben Graham credited as the father of
value investing.
– Warren Buffet most famous pupil.

• Essential idea is like PV analysis

– Buy stocks with market prices lower than
discounted value of cash flows.
– Ben Graham, for example, would try to
buy stocks with prices < liquidation value.
– Many implementations involve buying
stocks with low P/B (or high B/M) ratios.
Source: amazon.com

Lecture - CAPM Perf, Value & Momentum Page 13

 More about Fama & French (1992)
• Fama & French published comprehensive study of relations between
accounting value of equity (book) and market value (B/M ratio) and
future returns
– They controlled for other phenomena such as P/E, size, financial leverage, &
beta
• Key result: high B/M (value) stocks easily outperformed low B/M
(growth) stocks
• Small Stocks earned more than large stocks
– AND beta, P/E, leverage became redundant
• What does this finding really mean?

Lecture - CAPM Perf, Value & Momentum Page 14

 Returns, sorts by ME and BE/ME FF ‘92
Small-cap Value

Large-cap Growth

Lecture - CAPM Perf, Value & Momentum Page 15

 Is Value Risk?
• Maybe value is a risk factor
– For example, distress risk
– Fama and French argue value stocks and small-cap stocks are riskier.
• DFA: “Many economists believe small cap and value stocks outperform because the
market rationally discounts their prices to reflect underlying risk. The lower prices give
investors greater upside as compensation for bearing this risk.”
(http://www.dfaus.com/philosophy/dimensions.html)

• Maybe value is mispricing

– People like “glamourous” stocks?
– Lakonishok, Shleifer, Vishy argue it is mispricing
• LSV: “The fundamental premise on which our investment philosophy is based is that
superior long-term results can be achieved by systematically exploiting the judgmental
biases and behavioral weaknesses that influence the decisions of many investors.”
(http://www.lsvasset.com/about/about.html)

Lecture - CAPM Perf, Value & Momentum Page 16

 The case for risk
• Fama & French in several subsequent studies have argued that high
B/M stocks are riskier, or “marginal” firms and that the CAPM is too
simple, “CAPM is dead”
• Firms with low market value relative to their BE are firms that the
market perceives to have poor growth opportunities.
– They are more sensitive to the business cycle
– Investors require a higher rate of return for these risky securities.

Lecture - CAPM Perf, Value & Momentum Page 17

 The case for glamour
• Lakonishok, Shleifer, Vishny (1994)

Lecture - CAPM Perf, Value & Momentum Page 18

 Are value strategies risky? (LSV ’94)
• Remember, risk is when 𝑐𝑜𝑣 𝑟𝑒𝑡𝑢𝑟𝑛𝑠, 𝑤𝑒𝑎𝑙𝑡ℎ > 0

R = Recession
D = Mkt < 0

LSV ’94 argues

that value does
not perform
poorly in bad
times, so it is
not risk.

Lecture - CAPM Perf, Value & Momentum Page 19

 Value in last ~10 years
In recent sample, value covaries
more positively with the market.
Does this make it more or less
risky?

Lecture - CAPM Perf, Value & Momentum Page 20

 Is value risk or mispricing?
• Strong debate on both sides.

• No clear answer.

• Recently, value looks more risky.

– Effect of quant funds following value more?

• Conclusion left to the reader.

– But if you try to trade value, remember it is mostly in very small-cap stocks.

Lecture - CAPM Perf, Value & Momentum Page 21

 “Drifts”
• Other things that explain returns
– Momentum
– Post-earnings Announcement Drift
– Delayed reaction to news
– Equity issuance

• All potentially related to an underreaction story

Lecture - CAPM Perf, Value & Momentum Page 22

 Momentum

Lecture - CAPM Perf, Value & Momentum Page 23

 Buying Winners & Selling Losers
• Jegadeesh & Titman (1993) studied a portfolio
investment strategy that exploits lagged
adjustment of stock prices to information

• Each month from 1964-89, 5400 NYSE stocks

sorted into past winners/loser deciles by their
recent 3-, 6-, 9- and 12-month performance &
winner-net-of-loser portfolios held for next 3- to
12-months

• They found spread returns of 12% - 15% on

average over the entire sample

Lecture - CAPM Perf, Value & Momentum 24

 Momentum Investing
• What? Continuation in stock returns:
– Prior winners continue to outperform prior losers

• Why? Leading suspects investigated to date:

– Data Mining: special artifact of U.S. data
– Risk: momentum returns are reward for priced risk
– Behavioral Patterns: irrational agents induce price patterns

Lecture - CAPM Perf, Value & Momentum 25

 Data Mining?
• Has been widely
publicized since (1927-2011)

1991.
– (Jegadeesh and Titman
(2001))

• What about
Internationally?
Does momentum
persist in other
markets?
– Rouwenhorst (1998),
Griffin, Ji, and Martin
(2003, 2005), Fama and
French (2010)

Lecture - CAPM Perf, Value & Momentum Page 26

 Momentum explanations?
• Data mining? No.
• Risk
– Assumes stocks that have done well in the past are risky
• Underreaction
– Assumes stocks have good news, but some investors are slow to react.
• Overreaction
– Assumes stocks have good news, but some investors overreact, so there’s a
reversal later.

Lecture - CAPM Perf, Value & Momentum Page 27

 Further examination (JT 2001)

Risk

Underreaction

Overreaction
Lecture - CAPM Perf, Value & Momentum Page 28
 Further examination cont’d (JT 2001)

Lecture - CAPM Perf, Value & Momentum Page 29

 Underreaction? (JT 1993)
1
• Rank stocks into
winners/losers based on past 0.9

6-months stock return. 0.8

• Look at winner – loser (WML) 0.7

Return in bps
returns from days -2 to 0 0.6
around earnings 0.5
announcements made over
0.4
the next 6 months
0.3

0.2
• Positive WML returns mean
stocks that have done well in 0.1

the past have more 0

“positive” earnings surprises 1 2 3 4 5 6
than those that have done Months after formation
poorly.

Lecture - CAPM Perf, Value & Momentum Page 30

 Momentum in many asset classes
• US Equities
– Jegadeesh and Titman (1993, 2001).
• Developed Equities
– Rouwenhorst (1998)
• Emerging Equities
– Rouwenhorst (1999)
• Industries & Firm Specific (Equity)
– Moskowitz and Grinblatt (1999), Grundy and Martin (2001).
• Country Equity Indices
– Asness, Liew, and Stevens (1997)
• Currencies
– Okunev and White (2003)
• Commodities
– Erb and Harvey (2006)
• Futures
– Asness, Moskowitz, and Pedersen (2008), Moskowitz, Ooi, and Pedersen (2010)

Lecture - CAPM Perf, Value & Momentum Page 31

 Momentum and Value Everywhere?
• Asness, Moskowitz, Pedersen 2012
– We find Value and Momentum in equities
– What about other asset classes?
– What happens if we combine the two strategies?
• Their strategies
– Stock: US, UK, Japan, Europe,
– Non-stock: Country Index futures, country bonds,
currencies, commodities

Lecture - CAPM Perf, Value & Momentum Page 32

Lecture - CAPM Perf, Value & Momentum Page 33
Lecture - CAPM Perf, Value & Momentum Page 34
 Stock vs. Non-Stock
• Stock: US, UK,
Japan, Europe,

• Non-stock: Country
Index futures,
country bonds,
currencies,
commodities

Lecture - CAPM Perf, Value & Momentum Page 35

 Should you trade on this?
• Sharpe Ratios of all asset strategies
– Value: 0.63
– Momentum: 0.94
– Combo: 1.89

• Explanations? Risks?

Lecture - CAPM Perf, Value & Momentum Page 36

 Momentum Crashes
• Daniel (2010); Daniel, Jagannathan, Kim (2012)
– Momentum “blows up” during down markets

• Reasons?
– Quants all rush for the exits at the same time?
– The firms in the winner/loser portfolios are different during momentum
crashes?

Lecture - CAPM Perf, Value & Momentum Page 37

 Momentum during the Depression

Lecture - CAPM Perf, Value & Momentum Page 38

 Momentum during Financial Crisis

Lecture - CAPM Perf, Value & Momentum Page 39

 Can we just hedge this risk?

Lecture - CAPM Perf, Value & Momentum Page 40

 Momentum Summary
• Historically Momentum has done great
– (up to 12% per year)
• Most agree it does not relate to standard notions
of systematic risk
– But it is not riskless in the sense of being a sure thing.
• There are some behavioral explanations
– But no consensus on these either
• Industry knows not to ignore momentum, but
most secretly admit that they don’t fully
understand it either.

Lecture - CAPM Perf, Value & Momentum 41

 Summary
• Is beta risk?
– Historically, no.
– Going forward? Not clear.

• Several “anomalies” that seem to explain returns

– Small-cap stocks have higher returns
– Value stocks have higher returns
– High price-momentum stocks have higher returns

• Risk or mispricing?
– If it is risk, it will continue to exist.
– If it is mispricing, we expect it to go away.

Lecture - CAPM Perf, Value & Momentum Page 42