Efficent Market Hypotesis

& Empirical Evidence on Security Returns

Efficient Market Hypothesis (EMH)

Do security prices reflect information ? Why look at market efficiency
Implications for business and corporate finance Implications for investment

Random Walk and the EMH

Random Walk - stock prices are random
Actually submartingale
Expected price is positive over time Positive trend and random about the trend

Random Walk with Positive Trend

Security Prices

Time

Random Price Changes

Why are price changes random? Prices react to information Flow of information is random Therefore, price changes are random

EMH and Competition

Stock prices fully and accurately reflect publicly available information Once information becomes available, market participants analyze it Competition assures prices reflect information

So, What is market efficiency ?

Prices reflect all available information. Thus, fin. asset prices are fair prices. They are neither too high, nor too low. What is meant by all available information? Historical trading data Publicly available information All (private and public) information

Three Forms of Efficiency

Strong form of capital market efficiency.
Current prices reflect all information that can possibly be known to anyone.

Semi-strong form of cap mkt efficiency.

Current prices reflect all publicly available information.

Weak form of capital market efficiency.

Current prices reflect only the information contained in past prices.

Information set of market efficiency

Strong Form

(All information affecting the assets value)

Semi-Strong Form

(All publicly available information)

Weak Form

(Information contained in historical trading data)

Implication of Efficiency for Investors

Future market prices cannot be predicted based on available information
Random Walk

Investments in these markets have a zero NPV. The expected rate of return equals the required rate of return. The expected rate of return compensates the investor for the risk borne. Abnormally high returns are earned by pure chance.

Frictions in Capital Markets

Frictions in the capital markets prevent these markets from being perfectly efficient. Frictions include:
Transaction Costs: time, effort, and money required to make a transaction. Asymmetric taxes. Asymmetric information.

Information and Price Movements

In an efficient capital market, prices reflect all available information. When new information arrives, prices react instantaneously. Since new information is that which cannot be predicted, it would arrive at random points in time. Price movements are random (i.e. cannot be predicted).

Types of Stock Analysis

Technical Analysis - using prices and volume information to predict future prices
Weak form efficiency & technical analysis

Fundamental Analysis - using economic and accounting information to predict stock prices
Semi strong form efficiency & fundamental analysis

Reaction of Stock Price to New Information in Efficient and Inefficient Stock Markets Price
Overreaction and reversion

Delayed response (Underreaction) Efficient-market response to new information

30 20 10

+10 +20 +30

Days before (+) and after (-) announcement

Implications of Efficiency for Active or Passive Management

Active Management : Security analysis, Timing Passive Management: Buy and Hold, Index Funds

Market Efficiency and Portfolio Management

Even if the market is efficient a role exists for portfolio management: Appropriate risk level Tax considerations Other considerations

Empirical Tests of Market Efficiency

Event studies Assessing performance of professional managers Testing some trading rule

How Tests Are Structured? 1. Examine prices and returns over time

Returns Over Time

-t

0 Announcement Date

+t

How Tests Are Structured (contd)

2. Returns are adjusted to determine if they are abnormal Market Model approach a. Rt = at + btRmt + et (Expected Return) b. Excess Return = (Actual - Expected) et = Actual - (at + btRmt) c. Cumulate the excess returns over time:

-t

+t

Issues in Examining the Results

Magnitude Issue Selection Bias Issue Lucky Event Issue Possible Model Misspecification

What Does the Evidence Show?

Technical Analysis
Short horizon Long horizon

Fundamental Analysis Anomalies Exist

Anomalies
Small Firm Effect (January Effect) Neglected Firm Market to Book Ratios Reversals Post-Earnings Announcement Drift Market Crash of 1987

Mutual Fund and Professional Manager Performance

Some evidence of persistent positive and negative performance Potential measurement error for benchmark returns Style changes May be risk premiums Superstar phenomenon

Why? Perfect Capital Markets?

No barriers to entry. Perfect competition. each participant is sufficiently small and cannot affect prices by her/his actions. Financial assets are infinitely divisible. No transaction costs. All information is fully available to every participant, at no cost. No tax asymmetries. No restrictions on trading.

Some Imperfections
Asymmetric taxes These change the zero-sum nature of capital market transactions. Asymmetric information Information is not equally (and costlessly) available to all market participants. Transaction costs Generally less important an imperfection. Frictions in the capital markets prevent markets from being perfectly efficient

STRATEGY IN AN EFFICIENT MARKET

Diversify; select suitable asset allocation Dont try to time security price movement Keep tax consideration in mind Passive investing: Index Fund, Dollar cost averaging

Empirical Evidence on Security Returns

Overview of Investigation
Tests of the single factor CAPM or APT Model Tests of the Multifactor APT Model Results are difficult to interpret Studies on volatility of returns over time

Tests of the Single Factor Model

Tests of the expected return beta relationship First Pass Regression Estimate beta, average risk premiums, unsystematic risk Second Pass: Using estimates from the first pass to determine if model is supported by the data Most tests do not generally support the single factor model

Single Factor Test Results

Return %
Predicted Actual

Beta

Rolls Criticism
Only testable hypothesis is on the efficiency of the market portfolio Benchmark error

Measurement Error in Beta

Statistical property If beta is measured with error in the first stage Second stage results will be biased in the direction the tests have supported Test results could result from measurement error

Tests of the Multifactor Model

Chen, Roll and Ross 1986 Study. Factors: Growth rate in industrial production Changes in expected inflation Unexpected inflation Changes in risk premiums on bonds Unexpected changes in term premium on bonds

Study Structure & Results

Method: Two -stage regression with portfolios constructed by size based on market value of equity Findings: Significant factors: industrial production, risk premium on bonds and unanticipated inflation Market index returns were not statistically significant in the multifactor model

Anomalies Literature
Is the CAPM or APT Model Valid? Numerous studies show the approach is not valid Why do the studies show this result Other factors influence returns on securities Statistical problems prohibit a good test of the model

Fama and French Study

Size and book-to-market ratios explain returns on securities Beta is not a significant variable when other variables are included Study results show no support for the CAPM or APT

Researchers Responses to Fama and French

Utilize better econometric techniques Improve estimates of beta Reconsider the theoretical sources and implications of the Fama and French-type results Return to the single-index model, accounting for nontraded assets and cyclical behavior of betas

Jaganathan and Wang Study

Included factors for cyclical behavior of betas and human capital When these factors were included the results showed returns were a function of beta Size is not an important factor when cyclical behavior and human capital are included

Stochastic Volatility
Stock prices change primarily in reaction to information New information arrival is time varying Volatility is therefore not constant through time

Stock Volatility Studies and Techniques

Pagan and Schwert Study Study of 150 years of volatility on NYSE stocks Volatility is not constant through time Improved modeling techniques should improve results of tests of the risk-return relationship GARCH Models to incorporate time varying volatility

IS THE MARKET EFFICIENT ?

According to Eugene Fama According to Grossman-Stiglitz According to Fischer Black According to You ?

Efficient markets: intuition

Price

Realization

Expectation

Price change is unexpected

Time

Weak Form Efficiency

Random-walk model:
Pt -Pt-1 = Pt-1 * (Expected return) + Random error Expected value (Random error) = 0 Random error of period t unrelated to random component of any past period

Implication:
Expected value (Pt) = Pt-1 * (1 + Expected return) Technical analysis: useless

Empirical evidence: serial correlation

Correlation coefficient between current return and some past return Serial correlation = Cor (Rt, Rt-s)

S&P500 Daily returns

0.08 0.06 0.04

0.02

Return day t+1

0 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08

-0.02

-0.04

-0.06

-0.08 Return day t

Semi-strong Form Efficiency

Prices reflect all publicly available information Empirical evidence: Event studies Test whether the release of information influences returns and when this influence takes place. Abnormal return AR : ARt = Rt - Rmt Cumulative abnormal return: CARt = ARt0 + ARt0+1 + ARt0+2 +... + ARt0+1

Efficient Market Theory

Announcement Date
Cumulative Abnormal Return (%)
39 34 29 24 19 14 9 4 -1 -6 -11 -16

Days Relative to annoncement date

Example: How stock splits affect value

40 35 30 25 20 15 10 5 0
-29 0 30

Cumulative abnormal return %

Month relative to split

Source: Fama, Fisher, Jensen & Roll

Event Studies: Dividend Omissions

Cumulative abnormal returns (%)
Cumulative Abnormal Returns for Companies Announcing Dividend Omissions
1 0.146 0.108 -8 -6 -4 -0.72 0.032 0 -0.244 -2 -0.483 0 -1 -2 -3 -3.619 -4 -5 -6 -4.49 -4.563 -4.685 -4.747 -4.898 -5.015 -5.183 -5.411

Efficient market response to bad news

Days relative to announcement of dividend omission

S.H. Szewczyk, G.P. Tsetsekos, and Z. Santout Do Dividend Omissions Signal Future Earnings or Past Earnings? Journal of Investing (Spring 1997)

Strong-form Efficiency
How do professional portfolio managers perform? Jensen 1969: Mutual funds do not generate abnormal returns

Rfund - Rf = + (RM - Rf)

Insider trading

Insiders do seem to generate abnormal returns

(should cover their information acquisition activities)

What moves the market

Who knows? Lot of noise:
1985-1990: 120 days with | DJ| > 5%
28 cases (1/4) identified with specific event
(Siegel Stocks for the Long Run Irwin 1994 p 184)

Orange juice futures (Roll 1984)

90% of the day-to-day variability cannot explained by fundamentals

Pity financial journalists!

Trading Is Hazardous to Your Wealth

(Barber and Odean Journal of Finance April 2000)

Sample: trading activity of 78,000 households 1991-1997 Main conclusions:

1. Average household underperforms benchmark by 1.1% annually 2. Trading reduces net annualized mean returns
Infrequent traders: 18.5% Frequent traders: 11.4%

3. Households trade frequently (75% annual turnover) 4. Trading costs are high: for average round-trip trade 4% (Commissions 3%, bid-ask spread 1%)
PhD 01-1 |40

US Equity Mutual Funds 1982-1991

(Malkiel, Journal of Finance June 1995)

Average Annual Return Capital appreciation funds 16.32% Growth funds 15.81% Small company growth funds 13.46% Growth and income funds 15.97% Equity income funds 15.66% 17.52%

S&P 500 Index

Average deviation from benchmark (risk adjusted)

PhD 01-1 |41

-3.20%

 : Excess Return
Excess return = Average return - Risk adjusted Return Expected return expected return
Average return

Risk

Risk

PhD 01-1

|42

Jensen 1968 - Distribution of t values for 115 mutual funds 1955-1964

35 30 25 20 15 10 5 0 -5 -4 -3 -2 -1
PhD 01-1

Not significantly different from 0

4
|43

US Mutual Funds Consistency of Investment Result

Successive Period Performance

Initial Period Performance Top Half Goetzmann and Ibbotson (1976-1985) Top Half 62.0% Bottom Half 36.6% Malkiel, (1970s) Top Half 65.1% Bottom Half 35.5% Malkiel, (1980s) Top Half 51.7% Bottom Half 47.5%
Source: Bodie, Kane, Marcus Investments 4th ed. McGraw Hill 1999 (p.118)

Bottom Half
38.0% 63.4% 34.9% 64.5% 48.3% 52.5%

Decomposition of Mutual Fund Returns

(Wermers Journal of Finance August 2000)

Sample: 1,758 funds 1976-1994 Benchmark Gross return Expense ratio Transaction costs Non stock holdings Net Return 14.8% +1% 15.8% 0.8% 0.8% 0.4% 13.8%

Funds outperform benchmark

Stock picking +0.75% No timing ability Deviation from benchmark +0.55%

Not enough to cover costs

Empirical challenges
Explaining the cross section of returns Explaining changes in expected returns

Beta
18.00 16.00 Durb 14.00 NoDu 12.00 Telc
Average return

Chem Mone Oil MktPort Util Other

Shop Manu

10.00

8.00

6.00

NoDu Durb Oil Chem Manu Telc Util Shop Mone Other MktPort RF

4.00

RF

2.00

0.00 0.00

0.20

0.40

0.60

0.80 Beta

1.00

1.20

1.40

1.60

Fama French
Average return vs market beta for 25 FF stock portfolios 1926-2004
2.00 1.80 1.60
LowB/M

Size S: S1 smallest - S5 biggest B/M: BM1 lowest - BM5 highest

Average monthly returns Big 1.01 1.08 1.01 0.92 1.33 1.26 1.10 0.92 1.46 1.30 1.30 1.03 1.51 1.40 1.35 1.11 1.64 1.53 1.46 1.34 1.39 1.32 1.24 1.07

1.40
HighB/M

1.20

Small 0.91 1.29 1.50 1.69 1.83 1.45

0.99 1.18 1.32 1.41 1.56

BM5 BM4 S2 BM3S3 S4 BM2 S5 S1

Mean return

1.00 0.80 0.60 CorpB 0.40 0.20 0.00 0.00 GovB RF

Mkt

BM1

-0.20

0.20

0.40

0.60

0.80 Beta

1.00

1.20

1.40

1.60

1.80

Size and B/M

2.00 m 1.80 e a n 1.60 m 1.40 o n t 1.20 h l 1.00 y r e t u r n % 0.80 0.60 0.40 0.20 S3 0.00 1 Small 2 Size Value S2 3 Low B/M 4 Big 5 S4

Low B/M Series2 Series3 Series4 High B/M

High B/M

PhD 01-1

|49

Fama French
SIZE Av.Ret. St.Dev. Beta Small 1.45 9.47 1.35 1.39 7.59 1.17 1.32 7.03 1.15
B/M Av.Ret. St.Dev. Beta Low 0.99 7.38 1.17 1.18 6.91 1.12 1.32 6.75 1.10 1.41 7.10 1.13 High 1.56 8.85 1.32

1.24 6.70 1.13

Big 1.07 6.45 1.05

Based on monthly data 192607 200411

File: 25_Portfolios_5x5_monthly.xls

-60 20 40 60 80 0
1927 1930 1933 1936 1939 1942 1945 1948 1951 1954 1957

-40
1960 1963

-20

Fama French

Fama French Factors - Annual

RM SMB HML
1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002

Predictability: Interest Rates and Expected Inflation

Rmt R ft t
Sample period (Sample Size) 1831-2002 1831-1925 1926-1952 1953-1971 1972-2002 (2,053) (1,136) (324) (228) (357) -2.073 (-3.50) -3.958 (-4.58) 0.114 (0.03) -5.559 (-2.57) -1.140 (-1.08)

Schwert, W., Anomalies and Market Efficiency,WP October 2002 http://ssrn.com/abstract_id=338080

100.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

0.00

1926 1928 1930 1932 1934 1936 1938 1940 1942 1944 1946 1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002

Predictability: D/P
Price/dividend ratio

Predictability
Nobs 77 R(t+1)=a+b*R(t)+e(t+1) Mean StDev Slope Standerror Stock 0.1190 0.2050 0.03 0.1154 Tbill 0.0421 0.0350 0.92 0.0465 Excess 0.0769 0.2083 0.04 0.1155 Excess(t+x) = a + b (D/P)(t) + e Horizon 1 year 2 year 3 years 4 years 5 years

t 0.27 19.79 0.31

R 0.001 0.838 0.001

4.17 8.13 11.27 13.69 15.02

1.60 2.26 2.62 2.95 3.21

2.61 3.60 4.30 4.64 4.67

0.082 0.147 0.200 0.228 0.233

ER(+5)=a+b*(D/P)(t)+e
1.5

Excess Return +5

0.5

0 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

-0.5

-1 D/P

0.5

1.5

-1.5
19

-0.5

-1
19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 58 56 54 52 50 48 46 44 42 40 38 36 34 32 30 28 26

Econometrician wanted
Excess Return + 5 : Residuals

Reference: Investment, 2008. Bodie-Kent-Markus. Lecture Handout - Prof. Roy Sembel (2008) International Investment, Prof. Andr Farber, Solvay Business School Universit Libre de Bruxelles