market efficiency lectures

Professor Gordon Phillips 1

Market Efficiency - 1

CAPITAL MARKET EFFICIENCY

IN THIS LECTURE WE WILL DISCUSS:

Efficient Capital Markets: A. Price BehaviorB. The HypothesisC. Forms of Market EfficiencyD. Common Misconceptions about Efficient Capital Markets.

We will take a RANDOM WALK down Wall Street and discuss: Fundamental Analysis, Technical Analysis, Chartists, Momentum Players & Contrarians

We will then discuss the evidence on efficient capital markets.

1.0 CAPITAL MARKET EFFICIENCYEfficient Capital Market - A market in which current market prices fully reflect available information. That is, one in which costless trading rules do not consistently beat the market.

A. Price Behavior in an Efficient MarketStock price reaction in efficient and inefficient markets: NEARLY INSTANTANEOUS

B. The Efficient Market HypothesisEfficient Market Hypothesis (EMH) - asserts that modern U.S. stock markets are, as a practical matter, efficient.

The important implication of the EMH is securities represent zero NPV investments - meaning that they are expected to return just exactly their risk-adjusted rate.Why Efficient? Competition among investors and traders makes a market efficient.

CAPITAL MARKET EFFICIENCY

IN THIS LECTURE WE WILL DISCUSS:

Efficient Capital Markets: A. Price BehaviorB. The HypothesisC. Forms of Market EfficiencyD. Common Misconceptions about Efficient Capital Markets.

We will take a RANDOM WALK down Wall Street and discuss: Fundamental Analysis, Technical Analysis, Chartists, Momentum Players & Contrarians

We will then discuss the evidence on efficient capital markets.

1.0 CAPITAL MARKET EFFICIENCYEfficient Capital Market - A market in which current market prices fully reflect available information. That is, one in which costless trading rules do not consistently beat the market.

A. Price Behavior in an Efficient MarketStock price reaction in efficient and inefficient markets: NEARLY INSTANTANEOUS

B. The Efficient Market HypothesisEfficient Market Hypothesis (EMH) - asserts that modern U.S. stock markets are, as a practical matter, efficient.

The important implication of the EMH is securities represent zero NPV investments - meaning that they are expected to return just exactly their risk-adjusted rate.Why Efficient? Competition among investors and traders makes a market efficient.

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

Stock Price

-30 -20 -10 0 +10 +20 +30

Days before (-) and after (+) announcement

Efficient market response to bad news

Overreaction to bad news with reversion

Delayed response to bad news

C. The Forms of Market Efficiency

1. Weak Form Efficiency - A form of the theory that suggests you can't beat the market by knowing past prices.

2. Semi-strong Efficiency - Perhaps the most controversial form of the theory, it suggests you can't consistently beat the market using publicly available information. That is, you can't win knowing what everyone else knows - annual reports etc.

3. Strong Form Efficiency - The form of the theory that states no information of any kind can be used to beat the market. Evidence shows this form does not hold.

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

Stock Price

-30 -20 -10 0 +10 +20 +30

Days before (-) and after (+) announcement

Overreaction to bad news with reversion

Delayed response to bad news

C. The Forms of Market Efficiency

1. Weak Form Efficiency - A form of the theory that suggests you can't beat the market by knowing past prices.

2. Semi-strong Efficiency - Perhaps the most controversial form of the theory, it suggests you can't consistently beat the market using publicly available information. That is, you can't win knowing what everyone else knows - annual reports etc.

3. Strong Form Efficiency - The form of the theory that states no information of any kind can be used to beat the market. Evidence shows this form does not hold.

Capital market history and the EMH:

1. Prices respond very rapidly to new information.

2. Future prices changes are difficult to predict.

3. Mispriced stocks (those whose future price level can be predicted accurately) are difficult to identify and exploit.

D. Some Common Misconceptions about the EMH

- Market efficiency does not mean

it doesn't make a difference how you invest, since the risk/return trade-off still applies, but

rather that you can't expect to consistently "beat the market" on a risk-adjusted basis using costless trading strategies.

- Stock price fluctuations are evidence that the market is efficient since new information is constantly arriving - prices that don't change are evidence of inefficiency.

- The EMH doesn't say prices are random,

but that price changes in an efficient market are random andindependent. That is, they can't be predicted before they happen.

Capital market history and the EMH:

1. Prices respond very rapidly to new information.

2. Future prices changes are difficult to predict.

3. Mispriced stocks (those whose future price level can be predicted accurately) are difficult to identify and exploit.

D. Some Common Misconceptions about the EMH

- Market efficiency does not mean

it doesn't make a difference how you invest, since the risk/return trade-off still applies, but

rather that you can't expect to consistently "beat the market" on a risk-adjusted basis using costless trading strategies.

- Stock price fluctuations are evidence that the market is efficient since new information is constantly arriving - prices that don't change are evidence of inefficiency.

- The EMH doesn't say prices are random,

but that price changes in an efficient market are random andindependent. That is, they can't be predicted before they happen.

2.0 A RANDOM WALK DOWN WALL STREET(STRONG FORM OF MARKET EFFICIENCY DOES NOT HOLD)

A. FUNDAMENTAL ANALYSIS: Important Factors(OR FIRM FOUNDATION THEORY)

1. EXPECTED GROWTH RATE OF EARNINGS Longer growth rate ==> higher price Growth does not go on forever

2. EXPECTED DIVIDEND PAYOUT Higher price the more paid out. Stock splits: Why? Small shareholders

3. DEGREE OF RISK Higher price the less risky

4. LEVEL OF INTEREST RATES Higher price the lower the interest rates

CAVEATS / PROBLEMS:

1. EXPECTATIONS

2. PRECISE Calculations cannot be made from imprecise decisions-Don't Justify buy or sell decisions.

3. Information / Analysis may be incorrect

4. Market may not converge to "true value"

Premiums change over time: Growth, Risk etc.

2.0 A RANDOM WALK DOWN WALL STREET(STRONG FORM OF MARKET EFFICIENCY DOES NOT HOLD)

A. FUNDAMENTAL ANALYSIS: Important Factors(OR FIRM FOUNDATION THEORY)

1. EXPECTED GROWTH RATE OF EARNINGS Longer growth rate ==> higher price Growth does not go on forever

2. EXPECTED DIVIDEND PAYOUT Higher price the more paid out. Stock splits: Why? Small shareholders

3. DEGREE OF RISK Higher price the less risky

4. LEVEL OF INTEREST RATES Higher price the lower the interest rates

CAVEATS / PROBLEMS:

1. EXPECTATIONS

2. PRECISE Calculations cannot be made from imprecise decisions-Don't Justify buy or sell decisions.

3. Information / Analysis may be incorrect

4. Market may not converge to "true value"

Premiums change over time: Growth, Risk etc.

B. TECHNICAL ANALYSIS: MARKET IS INEFFICIENT!!!

CASTLES IN THE AIR: Tulip bulb craze in 1634 South sea trade in 1711 1929 stock price valuation: INVESTMENT POOLS Growth stocks in 1961 NIFTY 50 in 1972 Internet stocks in 2000-2001

==> Can we predict the start of stopping of the above castles? Stick to your fundamentals

Technical analysts do not adhere exclusively to the view that prices reflect fundamental values. Adopt the view that prices are driven by prevailing sentiments.

C. SOME OF THE MIS-INFORMATION

1. CHARTISTS: PREDICT PRICE FROM PAST PRICES.Special class of mechanical trading rules:/Filters /Head and Shoulders/flags, pennants, support and resistance

Levels, accumulation levels, corrections, waves, and breakthroughs.

SLOGANS"Hold the winners, sell the losers""Switch into 'Strong' stocks"Don't fight the tape"

COMPUTERS JAZZ THIS "ANALYSIS" UP.The technician relies on prices and, perhaps, volume to reveal

information.Presuming that prices react slowly over long periods of time to new

information or changes in investor sentiment.

B. TECHNICAL ANALYSIS: MARKET IS INEFFICIENT!!!

CASTLES IN THE AIR: Tulip bulb craze in 1634 South sea trade in 1711 1929 stock price valuation: INVESTMENT POOLS Growth stocks in 1961 NIFTY 50 in 1972 Internet stocks in 2000-2001

==> Can we predict the start of stopping of the above castles? Stick to your fundamentals

Technical analysts do not adhere exclusively to the view that prices reflect fundamental values. Adopt the view that prices are driven by prevailing sentiments.

C. SOME OF THE MIS-INFORMATION

1. CHARTISTS: PREDICT PRICE FROM PAST PRICES.Special class of mechanical trading rules:/Filters /Head and Shoulders/flags, pennants, support and resistance

Levels, accumulation levels, corrections, waves, and breakthroughs.

SLOGANS"Hold the winners, sell the losers""Switch into 'Strong' stocks"Don't fight the tape"

COMPUTERS JAZZ THIS "ANALYSIS" UP.The technician relies on prices and, perhaps, volume to reveal

information.Presuming that prices react slowly over long periods of time to new

information or changes in investor sentiment.

2. MOMENTUM PLAYERS: ==> SELL THE LOSERS, KEEP THE WINNERS -Past trends up or down will continue as the information is slowly

absorbed by the market, or as the wave of optimism or pessimism spreads through the market.

This suggests that the series of returns should show positive autocorrelation. Price increases should tend to be followed byprice increases, and price decreases should tend to be followed by price decreases.

EVIDENCE: Shows that this is not the case in the long run. Some evidence that within the year momentum strategies produce profits Jegadesh and Titman (Journal of Finance, 1993) (the losers are where it holds perhaps reluctance to recognize losses, overconfidence..

3. CONTRARIANSSome technical analysts argue investors overreact to good and bad

news.Advice is to buy on bad news and sell on good news.This implies that returns should exhibit negative serial dependence

because price reversals are more likely than continuances of price changes.

We do not observe this statistically.

2. MOMENTUM PLAYERS: ==> SELL THE LOSERS, KEEP THE WINNERS -Past trends up or down will continue as the information is slowly

absorbed by the market, or as the wave of optimism or pessimism spreads through the market.

This suggests that the series of returns should show positive autocorrelation. Price increases should tend to be followed byprice increases, and price decreases should tend to be followed by price decreases.

EVIDENCE: Shows that this is not the case in the long run. Some evidence that within the year momentum strategies produce profits Jegadesh and Titman (Journal of Finance, 1993) (the losers are where it holds perhaps reluctance to recognize losses, overconfidence..

3. CONTRARIANSSome technical analysts argue investors overreact to good and bad

news.Advice is to buy on bad news and sell on good news.This implies that returns should exhibit negative serial dependence

because price reversals are more likely than continuances of price changes.

We do not observe this statistically.

3.0 EVIDENCE ON THE EMH

Which forms, if any, are supported by examination of the data?

Remember the two adjustments that need to be made:

1. Risk adjustmentConcomitant problem of incorrect risk adjustment.

2. Transaction costsNeed to subtract costs from dollar returns.

A. Evidence on EMH: Weak form

Technical Analysisvs. THE RANDOM WALK: Market is Semi-strong efficient

RANDOM WALK: Price changes should be uncorrelated and prices should look like a random walk.

EVIDENCE: PLOTS/CORRELATIONS

NO evidence in support of Technical Analysis

3.0 EVIDENCE ON THE EMH

Which forms, if any, are supported by examination of the data?

Remember the two adjustments that need to be made:

1. Risk adjustmentConcomitant problem of incorrect risk adjustment.

2. Transaction costsNeed to subtract costs from dollar returns.

A. Evidence on EMH: Weak form

Technical Analysisvs. THE RANDOM WALK: Market is Semi-strong efficient

RANDOM WALK: Price changes should be uncorrelated and prices should look like a random walk.

EVIDENCE: PLOTS/CORRELATIONS

NO evidence in support of Technical Analysis

B. Semi-Strong form, evidence

CAN WE MAKE RISK ADJUSTED PROFITS FROM PUBLIC INFO?

Accounting changes:Should we buy firm that announces change of accounting policy or

based on their choice of choice of LIFO versus FIFOIf prices (inflation) are rising, LIFO accounting leads to higher cash

flows because of lower taxes. But it produces lower net income.

Are investors fooled by accounting changes? Not this one.

Stock splitsShould you buy the stock of firm that has just split?NO: Plot of stock returns before and after show that could only make

excess returns by being able to predict beforehand what stocks will split.

Stock repurchasesShould you buy stock of firm that announces share repurchases by

tender offers? NO - Again need to be able to have information beforehand.

Dividend changes (see next slide for omissions)Should you buy stock of firm that makes Announcements of changes

in firm's dividend policy, I.e. buy on dividend increase or sell if decrease/cut dividend.

NO evidence of violation of semi-strong form.

B. Semi-Strong form, evidence

CAN WE MAKE RISK ADJUSTED PROFITS FROM PUBLIC INFO?

Accounting changes:Should we buy firm that announces change of accounting policy or

based on their choice of choice of LIFO versus FIFOIf prices (inflation) are rising, LIFO accounting leads to higher cash

flows because of lower taxes. But it produces lower net income.

Are investors fooled by accounting changes? Not this one.

Stock splitsShould you buy the stock of firm that has just split?NO: Plot of stock returns before and after show that could only make

excess returns by being able to predict beforehand what stocks will split.

Stock repurchasesShould you buy stock of firm that announces share repurchases by

tender offers? NO - Again need to be able to have information beforehand.

Dividend changes (see next slide for omissions)Should you buy stock of firm that makes Announcements of changes

in firm's dividend policy, I.e. buy on dividend increase or sell if decrease/cut dividend.

NO evidence of violation of semi-strong form.

Peter BossaertsI inserted the evidence - see next page

Efficient Market Hypothesis And Behavioral FinanceIs A Compromise In Sight?

6

What FFJR found is that, on average, stock prices around the date of the split behaved as shown in Figure 5.

Figure 5. Averaged stock price performance around the split date

According to FFJR findings, the market begins to anticipate a stock split more than two years before it actually happens and figures out the consequences of the split the day it is announced.

The event study techniques were further refined by other researchers. Some of the research designs are quite clever. A bizarre example appeared in a 1985 article in the Journal of Accounting and Economics by Johnson, Magee, Nagarajan, and Newman. The title of the article, An Analysis of the Stock Price Reaction to Sudden Executive Deaths, is self-explaining. The authors found that unexpected CEO deaths are associated with stock price decreases. However, in cases when the CEO was the company founder, the stock market tends to react by a price increase, begging the inference that the ability to create a business is different from the ability to run one.

The efficacy of professional investors is another enduring question. Can they, on average, provide better investment performance? The research here was focused primarily on mutual funds. Regrettably, most professional money managers are not able to provide superior returns.

By 1975, the preponderance of evidence argued that markets were efficient. Statistical studies showed that technical analysis did not add value (consistent with the weak form of market efficiency). Event studies found that the market quickly reacts to new information (consistent with the semi-strong form of market efficiency). And studies of professional investors performance made a strong case for the strong form market efficiency.

Tests of Market Efficiency after 1975

As more and more researchers tested the efficient market hypothesis, some rather controversial evidence began to appear.

Event Studies: Dividend Omissions

Cumulative 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

Cumu

lativ

e abn

orma

l retu

rns

(%)

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

Value Line reclassifications

Should you buy stock of firm after Value line Announcement of reclassification of firms into higher or lower Value Line group.

Highest rank is 1 (buy recommendation); lowest rank is 5 (sell recommendation).

==> NO - Again need to be able to have information beforehand.

Earnings announcementsShould we buy stock of firm who suprises the market with stronger

than expected earnings? or sell with weaker than expected?Look at return versus amount of unexpected earnings change.

Classified into portfolios on basis on magnitude of earnings "surprise."

YES - It appears that excess returns are possible here even many daysafter announcement.

Event Studies: Dividend Omissions

Cumulative 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

Cumu

lativ

e abn

orma

l retu

rns

(%)

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

Value Line reclassifications

Should you buy stock of firm after Value line Announcement of reclassification of firms into higher or lower Value Line group.

Highest rank is 1 (buy recommendation); lowest rank is 5 (sell recommendation).

==> NO - Again need to be able to have information beforehand.

Earnings announcementsShould we buy stock of firm who suprises the market with stronger

than expected earnings? or sell with weaker than expected?Look at return versus amount of unexpected earnings change.

Classified into portfolios on basis on magnitude of earnings "surprise."

YES - It appears that excess returns are possible here even many daysafter announcement.

C. Strong form, evidence

Mutual funds (SEE graph next slide)Do mutual funds earn excess returns because fund managers are

either better at picking stocks than most people? or because fund managers have access to private information?

After adjusting for risk and transaction costs, answer seems to be that fund managers are NOT better at picking stocks on average.

Insider tradingDo insiders earn excess returns?

Yes, if returns are calculated from time of purchase or sale rather than from time of announcement of purchase or sale.

The Record of Mutual Funds

Annual Return Performance of Different Types of U.S. Mutual Funds Relative to a Broad-Based Market Index

(1963-1998)

-60.00%

-50.00%

-40.00%

-30.00%

-20.00%

-10.00%

0.00%All funds Small-

companygrowthfunds

Other-aggressive

growthfunds

Growthfunds

Incomefunds

Growth andincomefunds

Maximumcapitalgainsfunds

Sectorfunds

Ann

ual R

etur

n Pe

rfor

man

ce

Taken from Lubos Pastor and Robert F. Stambaugh, Evaluating and Investing in Equity Mutual Funds, unpublished paper, Graduate School of Business, University of Chicago (March 2000).

C. Strong form, evidence

Mutual funds (SEE graph next slide)Do mutual funds earn excess returns because fund managers are

either better at picking stocks than most people? or because fund managers have access to private information?

After adjusting for risk and transaction costs, answer seems to be that fund managers are NOT better at picking stocks on average.

Insider tradingDo insiders earn excess returns?

Yes, if returns are calculated from time of purchase or sale rather than from time of announcement of purchase or sale.

The Record of Mutual Funds

Annual Return Performance of Different Types of U.S. Mutual Funds Relative to a Broad-Based Market Index

(1963-1998)

-60.00%

-50.00%

-40.00%

-30.00%

-20.00%

-10.00%

0.00%All funds Small-

companygrowthfunds

Other-aggressive

growthfunds

Growthfunds

Incomefunds

Growth andincomefunds

Maximumcapitalgainsfunds

Sectorfunds

Ann

ual R

etur

n Pe

rfor

man

ce

Taken from Lubos Pastor and Robert F. Stambaugh, Evaluating and Investing in Equity Mutual Funds, unpublished paper, Graduate School of Business, University of Chicago (March 2000).

CONCLUSIONS

Evidence supports weak and semi-strong forms of market efficiency but not strong form.Implication: to make excess profits

Need information no one else hasor

Ability to process available information much better

There are anomalies:Little evidence that they produce large dollar profits consistently after

adjustments for risk and transaction costs. (Short-run momentum is one possible exception depends on transaction costs.)

ImplicationsDIVERSIFY, GET THE INFORMATION FIRST, Avoid transaction costs, MINIMIZE TAXES

CONCLUSIONS

Evidence supports weak and semi-strong forms of market efficiency but not strong form.Implication: to make excess profits

Need information no one else hasor

Ability to process available information much better

There are anomalies:Little evidence that they produce large dollar profits consistently after

adjustments for risk and transaction costs. (Short-run momentum is one possible exception depends on transaction costs.)

ImplicationsDIVERSIFY, GET THE INFORMATION FIRST, Avoid transaction costs, MINIMIZE TAXES

market efficiency lectures

Documents

market efficient

capital market efficiencyin

efficient capital markets

current market prices

forms of market efficiencyd

forms of market efficiency1

stock markets

price behaviorb