Post on 14-Apr-2018
transcript
7/30/2019 EMH Empirical
1/57
Efficent
Market
Hypotesis
&
Empirical Evidence
on Security Returns
7/30/2019 EMH Empirical
2/57
Do security prices reflect information ?
Why look at market efficiency
Implications for business and corporate finance
Implications for investment
Efficient Market Hypothesis (EMH)
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
7/30/2019 EMH Empirical
3/57
Random Walk with Positive Trend
Security
Prices
Time
7/30/2019 EMH Empirical
4/57
Why are price changes random?
Prices react to information
Flow of information is random
Therefore, price changes are random
Random Price Changes
Stock prices fully and accurately reflect publicly
available information Once information becomes available, market
participants analyze it
Competition assures prices reflect information
EMH and Competition
7/30/2019 EMH Empirical
5/57
So, What is market efficiency ?
Prices reflect all available information.
Thus, fin. asset prices are fairprices.
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
7/30/2019 EMH Empirical
6/57
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.
7/30/2019 EMH Empirical
7/57
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)
7/30/2019 EMH Empirical
8/57
Implication of Efficiency for Investors
Future market prices cannot be predicted based onavailable 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.
7/30/2019 EMH Empirical
9/57
Frictions in Capital Markets
Frictions in the capital markets prevent thesemarkets from being perfectly efficient.
Frictions include:
Transaction Costs: time, effort, and money
required to make a transaction.
Asymmetric taxes. Asymmetric information.
7/30/2019 EMH Empirical
10/57
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).
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
Types of Stock Analysis
f k
7/30/2019 EMH Empirical
11/57
Reaction of Stock Price to New
Information in Efficient and Inefficient
MarketsStock
Price
Days before (+) andafter (-) announcement
30
20
10 0 +10 +20 +30
Overreaction andreversion
Delayed response(Underreaction)
Efficient-marketresponse to new information
7/30/2019 EMH Empirical
12/57
Active Management : Security analysis, Timing Passive Management: Buy and Hold, Index Funds
Implications of Efficiency for Active
or Passive Management
Market Efficiency
and Portfolio Management
Even if the market is efficient a role exists for
portfolio management:
Appropriate risk level
Tax considerations
Other considerations
7/30/2019 EMH Empirical
13/57
Event studies
Assessing performance of professional managers
Testing some trading rule
Empirical Tests of Market Efficiency
How Tests Are Structured?
1. Examine prices and returns over time
7/30/2019 EMH Empirical
14/57
Returns Over Time
0 +t-t
Announcement Date
7/30/2019 EMH Empirical
15/57
2. Returns are adjusted to determine if they are abnormal
Market Model approacha. Rt = at + btRmt + et
(Expected Return)b. Excess Return =
(Actual - Expected)et = Actual - (at + btRmt)
How Tests Are Structured (contd)
c. Cumulate the excess returns over time:
0 +t-t
7/30/2019 EMH Empirical
16/57
Magnitude Issue
Selection Bias Issue
Lucky Event Issue
Possible Model Misspecification
Issues in Examining the Results
What Does the Evidence Show? Technical Analysis
Short horizon Long horizon
Fundamental Analysis
Anomalies Exist
7/30/2019 EMH Empirical
17/57
Small Firm Effect (January Effect)
Neglected Firm
Market to Book Ratios
Reversals
Post-Earnings Announcement Drift
Market Crash of 1987
Anomalies
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
7/30/2019 EMH Empirical
18/57
Why? Perfect Capital Markets?
No barriers to entry.
Perfect competition.
each participant is sufficiently small andcannot affect prices by her/his actions.
Financial assets are infinitely divisible. No transaction costs.
All information is fully available to everyparticipant, at no cost.
No tax asymmetries. No restrictions on trading.
7/30/2019 EMH Empirical
19/57
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
7/30/2019 EMH Empirical
20/57
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
7/30/2019 EMH Empirical
21/57
Empirical Evidenceon Security Returns
7/30/2019 EMH Empirical
22/57
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
7/30/2019 EMH Empirical
23/57
Single Factor Test Results
Return %
Beta
Predicted
Actual
7/30/2019 EMH Empirical
24/57
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 directionthe tests have supported
Test results could result from measurement error
7/30/2019 EMH Empirical
25/57
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
7/30/2019 EMH Empirical
26/57
Anomalies LiteratureIs 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 areincluded
Study results show no support for the CAPM or APT
Researchers Responses
7/30/2019 EMH Empirical
27/57
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
S h i V l ili
7/30/2019 EMH Empirical
28/57
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 Studiesand 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
7/30/2019 EMH Empirical
29/57
IS THE MARKET EFFICIENT ?
According to Eugene Fama
According to Grossman-Stiglitz
According to Fischer Black
According to You ?
7/30/2019 EMH Empirical
30/57
7/30/2019 EMH Empirical
31/57
Efficient markets: intuition
Expectation
Time
Price
Realization
Price change is
unexpected
7/30/2019 EMH Empirical
32/57
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 anypast 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)
7/30/2019 EMH Empirical
33/57
S&P500 Daily returns
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
0.08
-0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08
Return day t
Returndayt+1
7/30/2019 EMH Empirical
34/57
Semi-strong Form Efficiency
Prices reflect all publicly available information
Empirical evidence: Event studies
Test whether the release of information influencesreturns and when this influence takes place.
Abnormal return AR : ARt = Rt - Rmt
Cumulative abnormal return:
CARt = ARt0 + ARt0+1 + ARt0+2 +... + ARt0+1
7/30/2019 EMH Empirical
35/57
Efficient Market Theory
-16
-11
-6
-1
4
9
14
19
24
29
34
39
Days Relative to annoncement date
Cum
ulativeAbnorm
alReturn
(%)
Announcement Date
7/30/2019 EMH Empirical
36/57
Example: How stock splits
affect value
0
5
10
15
20
25
30
35
40
Month relative to split
Cumulativeabnormal
return %
-29 0 30
Source: Fama, Fisher, Jensen & Roll
E t St di Di id d
7/30/2019 EMH Empirical
37/57
Event Studies: Dividend
OmissionsCumulative Abnormal Returns for Companies Announcing
Dividend Omissions
0.146 0.108
-0.72
0.032-0.244
-0.483
-3.619
-5.015-5.411
-5.183-4.898-4.563-4.747-4.685-4.49
-6
-5
-4
-3
-2
-1
0
1
-8 -6 -4 -2 0 2 4 6 8
Days relative to announcement of dividend omission
Cumulative
abnormalreturn
(%)
Efficient market
response to bad news
S.H. Szewczyk, G.P. Tsetsekos, and Z. Santout Do Dividend Omissions Signal Future Earnings or Past Earnings?Journal of Investing
(Spring 1997)
7/30/2019 EMH Empirical
38/57
Strong-form Efficiency
How do professional portfolio managers perform?
Jensen 1969: Mutual funds do not generate abnormalreturns
Rfund - Rf= + (RM - Rf)
Insider trading
Insiders do seem to generate abnormal returns
(should cover their information acquisition activities)
7/30/2019 EMH Empirical
39/57
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!
7/30/2019 EMH Empirical
40/57
PhD 01-1 |40
Trading Is Hazardous to Your Wealth(Barber and OdeanJournal of Finance April 2000)
Sample: trading activity of 78,000 households1991-1997
Main conclusions:1. Average household underperforms benchmark by
1.1% annually2. 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 trade4%
(Commissions 3%, bid-ask spread 1%)
7/30/2019 EMH Empirical
41/57
PhD 01-1 |41
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%
S&P 500 Index 17.52%
Average deviation from benchmark -3.20%
(risk adjusted)
7/30/2019 EMH Empirical
42/57
PhD 01-1 |42
: Excess Return
Excess return = Average return - Risk adjusted
expected return
Risk
Return Expected return
Average
return
Risk
Jensen 1968 Distribution of t values for
7/30/2019 EMH Empirical
43/57
PhD 01-1 |43
Jensen 1968 - Distribution of t values for 115 mutual funds 1955-1964
0
5
10
15
20
25
30
35
-5 -4 -3 -2 -1 0 1 2 3 4
Not significantly
different from 0
7/30/2019 EMH Empirical
44/57
US Mutual Funds
Consistency of Investment Result
Successive Period Performance
Initial Period Performance Top Half Bottom Half
Goetzmann and Ibbotson (1976-1985)
Top Half 62.0% 38.0%
Bottom Half 36.6% 63.4%
Malkiel, (1970s)
Top Half 65.1% 34.9%
Bottom Half 35.5% 64.5%
Malkiel, (1980s)
Top Half 51.7% 48.3%
Bottom Half 47.5% 52.5%
Source: Bodie, Kane, Marcus Investments 4th ed. McGraw Hill 1999 (p.118)
D iti f M t l F d
7/30/2019 EMH Empirical
45/57
Decomposition of Mutual Fund
Returns(Wermers Journal of Finance August 2000)
Sample: 1,758 funds 1976-1994
Benchmark 14.8%
+1%
Gross return 15.8%
Expense ratio 0.8%
Transaction costs 0.8%
Non stock holdings 0.4%
Net Return 13.8%
Stock picking +0.75%
No timing ability
Deviation from benchmark +0.55%
Funds outperformbenchmark
Not enough to cover
costs
E i i l h ll
7/30/2019 EMH Empirical
46/57
Empirical challenges
Explaining the cross section of returns
Explaining changes in expected returns
Beta
7/30/2019 EMH Empirical
47/57
Beta
NoDu
Durb
Oil
Chem
Manu
TelcUtil
ShopMone
Other
MktPort
RF
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60
Beta
Averagereturn
NoDu
Durb
Oil
ChemManu
Telc
Util
Shop
Mone
Other
MktPort
RF
Fama French
7/30/2019 EMH Empirical
48/57
Average return vs market beta for 25 FF stock portfolios 1926-2004
Mkt
RF
S1S2
S3
S4
S5
GovB
CorpB
BM1
BM2
BM3
BM4
BM5
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
-0.20 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80
Beta
Meanretu
rn
Size S: S1 smallest - S5 biggest
B/M: BM1 lowest - BM5 highest
Average monthly returns
Small Big
LowB/M 0.91 1.01 1.08 1.01 0.92 0.99
1.29 1.33 1.26 1.10 0.92 1.18
1.50 1.46 1.30 1.30 1.03 1.32
1.69 1.51 1.40 1.35 1.11 1.41
HighB/M 1.83 1.64 1.53 1.46 1.34 1.56
1.45 1.39 1.32 1.24 1.07
Fama French
7/30/2019 EMH Empirical
49/57
PhD 01-1 |49
Size and B/M
12
34
5
Low B/M
S2
S3
S4High B/M
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
m
e
a
n
m
o
nt
h
l
y
r
e
t
u
r
n
%
Size
Value
Low B/M
Series2
Series3
Series4
High B/M
Small
Big
Fama French
7/30/2019 EMH Empirical
50/57
Small Big
Av.Ret. 1.45 1.39 1.32 1.24 1.07
St.Dev. 9.47 7.59 7.03 6.70 6.45
Beta 1.35 1.17 1.15 1.13 1.05
SIZE
Low High
Av.Ret. 0.99 1.18 1.32 1.41 1.56
St.Dev. 7.38 6.91 6.75 7.10 8.85Beta 1.17 1.12 1.10 1.13 1.32
B/M
Based on monthly data 192607 200411
File: 25_Portfolios_5x5_monthly.xls
Fama French
Fama French
7/30/2019 EMH Empirical
51/57
Fama FrenchFama French Factors - Annual
-60
-40
-20
0
20
40
60
80
1927
1930
1933
1936
1939
1942
1945
1948
1951
1954
1957
1960
1963
1966
1969
1972
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
RM SMB HML
Predictability: Interest Rates and Expected Inflation
7/30/2019 EMH Empirical
52/57
Predictability: Interest Rates and Expected Inflation
Sample period (Sample Size)
1831-2002 (2,053) -2.073
(-3.50)
1831-1925 (1,136) -3.958
(-4.58)
1926-1952 (324) 0.114
(0.03)
1953-1971 (228) -5.559
(-2.57)
1972-2002 (357) -1.140(-1.08)
Schwert, W., Anomalies and Market Efficiency,WP October 2002
http://ssrn.com/abstract_id=338080
tftmt RR
Predictability: D/P
7/30/2019 EMH Empirical
53/57
Predictability: D/P
Price/dividend ratio
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
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
7/30/2019 EMH Empirical
54/57
Predictability
Nobs 77
R(t+1)=a+b*R(t)+e(t+1)
Mean StDev Slope Standerro t R
Stock 0.1190 0.2050 0.03 0.1154 0.27 0.001
Tbill 0.0421 0.0350 0.92 0.0465 19.79 0.838
Excess 0.0769 0.2083 0.04 0.1155 0.31 0.001
Excess(t+x) = a + b (D/P)(t) + e
Horizon
1 year 4.17 1.60 2.61 0.0822 year 8.13 2.26 3.60 0.147
3 years 11.27 2.62 4.30 0.200
4 years 13.69 2.95 4.64 0.228
5 years 15.02 3.21 4.67 0.233
7/30/2019 EMH Empirical
55/57
ER(+5)=a+b*(D/P)(t)+e
-1
-0.5
0
0.5
1
1.5
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
D/P
ExcessReturn+
5
Econometrician wanted
7/30/2019 EMH Empirical
56/57
Econometrician wantedExcess Return + 5 : Residuals
-1.5
-1
-0.5
0
0.5
1
1.5
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
7/30/2019 EMH Empirical
57/57
Reference:
Investment, 2008. Bodie-Kent-Markus.
Lecture Handout - Prof. Roy Sembel (2008)
International Investment, Prof. Andr Farber,Solvay Business School Universit Libre de Bruxelles