Introduction. РЭШ EFM 2005/6 2 Plan Motivation Motivation Specifics of financial data Specifics of financial data –Time and cross dimensions –Market microstructure. - презентация

1 Introduction

2 РЭШ EFM 2005/6 2 Plan Motivation Motivation Specifics of financial data Specifics of financial data –Time and cross dimensions –Market microstructure effects Modeling financial data Modeling financial data –Rational vs. behavioral models –Different types of empirical tests –Methodological issues

3 РЭШ EFM 2005/6 3 How to explain asset prices?

4 РЭШ EFM 2005/6 4 From prices to returns Returns Returns –Simple: R t = (P t -P t-1 )/P t-1 Easy to compute portfolio return Easy to compute portfolio return –Log: R t = ln(P t /P t-1 ) Easy to aggregate over time Easy to aggregate over time Doesn't violate limited liability Doesn't violate limited liability –Arythmetic vs geometric average

5 РЭШ EFM 2005/6 5 How to explain asset returns?

6 РЭШ EFM 2005/6 6 Stylized facts Non-normality Non-normality –Thick tails –Asymmetry Autocorrelation of returns Autocorrelation of returns –Ultra-short horizon -, short +, long – Cross-correlation of returns Cross-correlation of returns

7 РЭШ EFM 2005/6 7 Gazprom in

8 РЭШ EFM 2005/6 8 Stylized facts (2) Volatility Volatility –Clustering in time –Inverse relation with prices –Smaller when the market are closed –Higher in times of forecastable releases of info –Inverse relation with auto-correlation –Common factors for different assets –Too high relative to fundamentals Often explosive growth or crashes Often explosive growth or crashes

9 РЭШ EFM 2005/6 9 Stylized facts (3) Cross-sectional anomalies Cross-sectional anomalies –Price-related company characteristics –Calendar effects

10 РЭШ EFM 2005/6 10 Market microstructure effects How to measure returns? How to measure returns? –Average vs. close prices Low liquidity Low liquidity Impact of the bid-ask spread Impact of the bid-ask spread –Can you make profit out of asset mispricing taking into account transaction costs?

11 РЭШ EFM 2005/6 11 Rational vs. behavioral models Traditional analysis: explain asset prices by rational models Traditional analysis: explain asset prices by rational models Only if they fail, resort to irrational investor behavior Only if they fail, resort to irrational investor behavior –Behavioral finance models

12 РЭШ EFM 2005/6 12 The efficient market hypothesis Informational efficiency: asset prices accommodate all relevant information Informational efficiency: asset prices accommodate all relevant information –History of prices / all public variables / all private information Why is it important? Why is it important? –For corporate finance –For (financial) investments

13 РЭШ EFM 2005/6 13 Is this market efficient?

14 РЭШ EFM 2005/6 14 Efficient markets Price movements must be random! Price movements must be random! –Otherwise one can forecast future price and make arbitrage profit –Prices should immediately respond to new information Sufficient conditions: Sufficient conditions: –No transaction costs –No information costs –Homogeneous expectations

15 РЭШ EFM 2005/6 15 Basic model for prices: random walk with trend Asset price Time

16 РЭШ EFM 2005/6 16 What does it imply for the business? No place for active investment policy No place for active investment policy –There are no under- or overpriced assets, no opportunities for arbitrage –Most portfolio managers should be fired! –Still, role for diversification, choice of risk, and tax optimization No place for active corporate policy No place for active corporate policy –It does not matter which capital structure to choose

17 РЭШ EFM 2005/6 17 How realistic is it? The paradox of Grossman-Stiglitz (1980): The paradox of Grossman-Stiglitz (1980): –No one will make research on the market, if all info is already reflected in the prices –There must be some inefficiency in the equilibrium to provide incentives for information acquisition Operational efficiency: Operational efficiency: –One cannot make profit on the basis of information, accounting for info acquisition and trading costs

18 РЭШ EFM 2005/6 18 What if the EMH is rejected? The joint hypothesis problem: we simultaneously test market efficiency and the model The joint hypothesis problem: we simultaneously test market efficiency and the model –Either the investors behave irrationally, –or the model is wrong Ex-ante expected profit within information and transaction costs Ex-ante expected profit within information and transaction costs

19 РЭШ EFM 2005/6 19 What if the EMH is rejected? (2) Empirical illusions Empirical illusions –Data mining –Survivor bias –Selection bias –Short-shot bias (rare events): it could be luck –Trading costs, esp. invisible market impact costs

20 РЭШ EFM 2005/6 20 What if the EMH is rejected? (3) Why not try more complicated rational models? Why not try more complicated rational models? –Multiple periods –Imperfect markets: liquidity, short-sale constraints –Uncertainty over other investors demand curves

21 РЭШ EFM 2005/6 21 Different forms of rationality Maximal: all investors are rational Maximal: all investors are rational Intermediate: asset prices are set as if all investors are rational Intermediate: asset prices are set as if all investors are rational Minimal: there are no abnormal profit opportunities, though Minimal: there are no abnormal profit opportunities, though –Sometimes a small group of irrational investors are able to determine asset prices E.g., acquiring firms tend to overpay E.g., acquiring firms tend to overpay

22 РЭШ EFM 2005/6 22 Examples of irrationality Reference points and loss aversion Reference points and loss aversion –Endowment effect / Status quo bias Overconfidence Overconfidence –Biased self-attribution / Disposition effect –Illusion of control Statistical errors Statistical errors –Gamblers fallacy / Misjudging very rare events –Extrapolation bias / Overreaction Miscellaneous errors in reasoning Miscellaneous errors in reasoning –Sunk costs / Cognitive dissonance

23 РЭШ EFM 2005/6 23 Information aggregation Example: locating the missing US submarine in 1968 Example: locating the missing US submarine in 1968 –5 months of extensive search efforts in the 20- mile circle brought no effect –The experts were asked for opinions, the average response gave the location, where the sub was ultimately found!

24 РЭШ EFM 2005/6 24 Information aggregation (2) Security markets Security markets –From the invisible hand principle of Adam Smith… –to Friedrich Hayek: the price system utilizes bits of knowledge not given to anyone in total –An investor must consider the vast amount of info already impounded in prices before making a bet based on his own info Recent improvement in market efficiency Recent improvement in market efficiency –Advances in technology and organized markets –Derivatives designed to trace different risks

25 РЭШ EFM 2005/6 25 Basis for minimal rationality Profitable trading strategies are self-destructible Profitable trading strategies are self-destructible Irrational investors self-destruct (become poor) Irrational investors self-destruct (become poor) –Excess trading volume, under-diversification, disposition effect –Even if all investors are irrational, in aggregate the market can be rational Overconfidence causes many investors to spend much money on research Overconfidence causes many investors to spend much money on research –Overly efficient markets? E.g., there are much more active mutual funds than passive ones E.g., there are much more active mutual funds than passive ones

26 РЭШ EFM 2005/6 26 How to test market efficiency? The joint hypothesis problem: The joint hypothesis problem: –We need the model for the expected prices (or returns) The rejection of the null implies that The rejection of the null implies that –Either investors behave irrationally, –or the model is wrong

27 РЭШ EFM 2005/6 27 Different types of tests Statistical significance: e.g., the market model R t =α+βR M,t +γX t-1 +ε t Statistical significance: e.g., the market model R t =α+βR M,t +γX t-1 +ε t –H 0 : γ = 0 Economic significance : Economic significance : –H 0 : (risk-adjusted) profit from the investment strategy = 0 –It is important to account for transaction costs –Should use only variables available to investors Fundamental efficiency: Fundamental efficiency: –H 0 : price = fundamental value –Otherwise the bubble partly explains the assets price

28 РЭШ EFM 2005/6 28 Empirical evidence Up to the end of 1970s: belief in the efficient markets and CAPM Up to the end of 1970s: belief in the efficient markets and CAPM –Past prices and other public variables do not predict future prices –Profits from technical analysis are close to 0 –The market quickly reacts to new info –Portfolio managers cannot beat the market

29 Event study analysis How quickly does the market react to new information? How quickly does the market react to new information? 0+t-t Announcement Date

30 РЭШ EFM 2005/6 30 Reaction to the unexpected event

31 РЭШ EFM 2005/6 31 Performance evaluation Experiment: compare returns of two portfolios Experiment: compare returns of two portfolios –1: chosen by experts –2: chosen by the monkey throwing darts –No significant difference! Formally: e.g., based on CAPM Formally: e.g., based on CAPM R t - R F,t = α + β(R M,t -R F,t ) + ε t –The Jensens alpha was close to 0

32 РЭШ EFM 2005/6 32 Empirical evidence (2) After 1970s: inefficient markets? After 1970s: inefficient markets? –Price anomalies in 1980s and 1990s, unexplained by the CAPM Calendar effects: e.g., Monday (negative), January (positive) Calendar effects: e.g., Monday (negative), January (positive) Book/market, value/growth and size effects Book/market, value/growth and size effects Short-run momentum and long-run price reversal effects Short-run momentum and long-run price reversal effects –Mutual funds performance persistence

33 РЭШ EFM 2005/6 33 Possible explanations Heterogeneity of investors Heterogeneity of investors –Tax considerations Each investor chooses assets to minimize his own efficient tax rate (role of dividends) Each investor chooses assets to minimize his own efficient tax rate (role of dividends) Investors fix losses on losing stocks at the end of the year (January anomaly) Investors fix losses on losing stocks at the end of the year (January anomaly) –Liquidity considerations (sharp market movements)

34 РЭШ EFM 2005/6 34 Possible explanations (2) Transaction costs Transaction costs –Short-sales restrictions (cannot profit from overprices assets, e.g., losing mutual funds) –Bid-ask spread (reduces substantially profits, esp. from small stocks – size and January anomalies) –Market impact (invisible costs)

35 РЭШ EFM 2005/6 35 Possible explanations (3) Statistical illusions Statistical illusions –Data mining (5 out of 100 irrelevant variables will be significant at the 5% level) –Selection bias (e.g., survivor bias) Should not exclude stocks of small companies (funds) that disappeared Should not exclude stocks of small companies (funds) that disappeared –Short-shot bias (rare events): it could be luck

36 РЭШ EFM 2005/6 36 Possible explanations (4) Mechanical relationships Mechanical relationships –Financial leverage effects (lower stock prices increase beta and imply higher return) –All variables based on current prices (e.g., size and BE/ME ) are automatically related to future returns P t is negatively related to R t =P t+1 /P t P t is negatively related to R t =P t+1 /P t

37 РЭШ EFM 2005/6 37 Possible explanations (5) More complicated rational models More complicated rational models –Time-varying betas and risk premiums –Multiple factors –Imperfect markets: liquidity, short-sale constraints, etc.

38 РЭШ EFM 2005/6 38 Topics covered in this course Time series analysis of asset returns Time series analysis of asset returns –Predictability at different horizons –Event study analysis Speed of stock price adjustment in response to news announcements Speed of stock price adjustment in response to news announcements Cross-sectional analysis of asset returns Cross-sectional analysis of asset returns –CAPM and market efficiency –Return anomalies and multi-factor asset pricing models

39 РЭШ EFM 2005/6 39 Topics covered in this course (2) Performance evaluation of mutual funds Performance evaluation of mutual funds –Performance persistence, survivorship bias, dynamic strategies, gaming behavior, etc. Investor behavior Investor behavior –Overconfidence, herding, home bias, impact of demographic characteristics, etc.

40 РЭШ EFM 2005/6 40 Topics outside of this course Fixed income / derivatives Fixed income / derivatives Corporate finance Corporate finance Trading and market microstructure Trading and market microstructure –Ultra high-frequency analysis International finance International finance Financial intermediaries Financial intermediaries