Market Efficiency

1
Market Efficiency

1. Performance of portfolio managers
2. Anomalies
3. Behavioral Finance as a challenge to the EMH

2
2. Performance of Portfolio Managers

• Implication of the semi-strong form EMH managers
  cannot consistently beat the market
• Information set of managers (supposedly) public
  information
• Collectively, U.S. evidence based on this type of
  tests support the semi-strong form EMH
• Issue of survivorship bias

3
Canadian Evidence

• Largest 76 Canadian equity funds from 1988 to
  1997, none beat the category average in all ten
  years
• Canadian Investment Review, Fall 2002, Does
  Aggressive Portfolio Management Work?
• Market timing test non-linear regression covered
  in class
• Stock selection ability test Jensens alpha

4
Performance of Portfolio Managers

• Conclusion no evidence of consistent
  market-timing or stock-picking abilities
• And past performance is not an indicator of
  future performance

5
3. Anomalies

• Exceptions that appear to be contrary to market
  efficiency
• Earnings announcements affect stock prices
• Adjustment occurs before announcement, but also
  significant amount after
• Contrary to efficient market hypothesis because
  the lag should not exist

6
Anomalies Examples

• Low M/B ratio stocks tend to outperform high M/B
  ratio stocks
• Low M/B portfolios typically have higher
  risk-adjusted returns (risk measured by ? or
  constant ?)
• Value investing
• Why is it an anomaly?
• M/B is public information!

7
Canadian Evidence(Deaves 2005)
8
Anomalies Examples

• Size effect
• Tendency for small firms to have higher
  risk-adjusted returns than large firms
• January effect
• Tendency for small firm stock returns to be
  higher in January
• Half of the size premium can be accounted for in
  January (known as Small-firm-in-January effect)

9
Anomalies Examples

• Time trend
• Evidence of short-term momentum (3-12 month
  horizon) in stock prices
• But evidence of long-term reversal (3-5 year
  horizon) in winner and loser portfolios

10
Explanations for Anomalies

• Risk Premiums or market inefficiencies?
• Data mining or anomalies?

11
The Value Premium

• Risk-based explanation
• Relax the assumption in the conventional CAPM
  that beta and the market risk premium are
  constant
• HML has higher beta when market risk premium is
  high. Translation value stocks do not do well in
  down markets, and hence are riskier to investors
  (Petkova and Zhang 2005)
• Value firms tend to have greater amounts of
  tangible assets, and hence less flexibility to
  adjust capacity during downturns (operating risk)

12
The Value Premium

• Behavioral finance explanation Investors tend to
  overreact
• Growth stocks are glamour stocks
• Price bidded up beyond fundamental value
• Correction in the long term
• Opposite is true for value stocks

13
4. Behavioral Finance

• Behavioral finance provides an alternative view
  of financial markets
• Challenges the EMH on both theoretical and
  empirical grounds
• Theory model investor behavior, using theories
  and observations from the psychology literature
• Empirical existence of anomalies (anomalous from
  the EMH perspective)

14
Three Theoretical Challenges

• Investors can be irrational
• Trade on irrelevant information (noise)
• Trade on sentiment
• Follow advice of financial gurus
• Fail to diversify
• Over-active trading

15
Theoretical Challenges

• Irrational investors trades are not random
• If random and uncorrelated, tend to cancel each
  other out, so that on average, stock price
  fundamental price
• Behavioral finance irrational investors trades
  are positively correlated, and hence move in the
  same direction
• Investor sentiment reflect common judgment errors
  made by a substantial number of investors
• Listen to the same rumours, and imitate neighbours

16
Theoretical Challenges

• There are limits to arbitrage
• If there is a significant number of irrational
  investors, arbitrage is risky
• If arbitrageurs are risk-averse, their activities
  will be limited (fundamental risk, implementation
  costs, model risk)
• Mispricing can exist, particularly in the short
  term