The Rational Part of Momentum

Similar documents
The Rational Part of Momentum* Abstract

Stock price synchronicity and the role of analyst: Do analysts generate firm-specific vs. market-wide information?

Time-variation of CAPM betas across market volatility regimes for Book-to-market and Momentum portfolios

Economics of Behavioral Finance. Lecture 3

Decimalization and Illiquidity Premiums: An Extended Analysis

Premium Timing with Valuation Ratios

Lecture 5. Predictability. Traditional Views of Market Efficiency ( )

Exploiting Factor Autocorrelation to Improve Risk Adjusted Returns

Size and Book-to-Market Factors in Returns

Some Features of the Three- and Four- -factor Models for the Selected Portfolios of the Stocks Listed on the Warsaw Stock Exchange,

Volatility Appendix. B.1 Firm-Specific Uncertainty and Aggregate Volatility

Unpublished Appendices to Market Reactions to Tangible and Intangible Information. Market Reactions to Different Types of Information

Smart Beta #

Applied Macro Finance

An Online Appendix of Technical Trading: A Trend Factor

The New Issues Puzzle

The Performance, Pervasiveness and Determinants of Value Premium in Different US Exchanges

A Lottery Demand-Based Explanation of the Beta Anomaly. Online Appendix

Investment Performance of Common Stock in Relation to their Price-Earnings Ratios: BASU 1977 Extended Analysis

Fundamental, Technical, and Combined Information for Separating Winners from Losers

Online Appendix - Does Inventory Productivity Predict Future Stock Returns? A Retailing Industry Perspective

Is Information Risk Priced for NASDAQ-listed Stocks?

Internet Appendix to The Evolution of Financial Market Efficiency: Evidence from Earnings Announcements

Size, Beta, Average Stock Return Relationship, 19 th century Evidence

The Economic Consequences of (not) Issuing Preliminary Earnings Announcement

Statistical Understanding. of the Fama-French Factor model. Chua Yan Ru

Investor Sophistication and the Mispricing of Accruals

What is the Expected Return on a Stock?

Master Thesis Finance THE ATTRACTIVENESS OF AN INVESTMENT STRATEGY BASED ON SKEWNESS: SELLING LOTTERY TICKETS IN FINANCIAL MARKETS

Appendix. In this Appendix, we present the construction of variables, data source, and some empirical procedures.

Return Reversals, Idiosyncratic Risk and Expected Returns

ALL THINGS CONSIDERED, TAXES DRIVE THE JANUARY EFFECT. Abstract

Market Efficiency FINANCE 352 INVESTMENTS. Professor Alon Brav. Fuqua School of Business Duke University

Post-Earnings-Announcement Drift: The Role of Revenue Surprises and Earnings Persistence

Trading Costs of Asset Pricing Anomalies Appendix: Additional Empirical Results

Style Timing with Insiders

Betting against Beta or Demand for Lottery

Dispersion in Analysts Earnings Forecasts and Credit Rating

Herding and Feedback Trading by Institutional and Individual Investors

Are Idiosyncratic Skewness and Idiosyncratic Kurtosis Priced?

Heterogeneous Beliefs and Momentum Profits

Appendix Tables for: A Flow-Based Explanation for Return Predictability. Dong Lou London School of Economics

Internet Appendix for: Cyclical Dispersion in Expected Defaults

Do Investors Fully Understand the Implications of the Persistence of Revenue and Expense Surprises for Future Prices?

Is Default Risk Priced in Equity Returns?

Portfolio strategies based on stock

Internet Appendix for: Cyclical Dispersion in Expected Defaults

The cross section of expected stock returns

Discussion Paper No. DP 07/02

Internet Appendix to Is Information Risk Priced? Evidence from Abnormal Idiosyncratic Volatility

Value at Risk and Expected Stock Returns

Market Frictions, Price Delay, and the Cross-Section of Expected Returns

High Idiosyncratic Volatility and Low Returns. Andrew Ang Columbia University and NBER. Q Group October 2007, Scottsdale AZ

BOOK TO MARKET RATIO AND EXPECTED STOCK RETURN: AN EMPIRICAL STUDY ON THE COLOMBO STOCK MARKET

THE VALUE OF VALUE INVESTING. Stephen Horan, Ph.D., CFA, CIPM Managing Director, Credentialing CFA Institute

Value versus Growth. The sources of return differences**

Market Reactions to Tangible and Intangible Information Revisited

Liquidity and IPO performance in the last decade

The Value Premium and the January Effect

Further Test on Stock Liquidity Risk With a Relative Measure

Online Appendix for. Short-Run and Long-Run Consumption Risks, Dividend Processes, and Asset Returns

Ambrus Kecskés (Virginia Tech) Roni Michaely (Cornell and IDC) Kent Womack (Dartmouth)

Radu BURLACU Patrice FONTAINE 1 Sonia JIMENEZ-GARCÈS. C.E.R.A.G. UMR CNRS 5820 Université Pierre Mendès France

Time Dependency in Fama French Portfolios

Industry Concentration and Average Stock Returns

Do Investors Understand Really Dirty Surplus?

Networks of Common Asset Holdings : Aggregation and Measures of Vulnerability

Debt/Equity Ratio and Asset Pricing Analysis

Does Book-to-Market Equity Proxy for Distress Risk or Overreaction? John M. Griffin and Michael L. Lemmon *

The Effect of Kurtosis on the Cross-Section of Stock Returns

The Determinants of Informed Trading: Implications for Asset Pricing

Momentum, Business Cycle and Time-Varying Expected Returns. Tarun Chordia and Lakshmanan Shivakumar * FORTHCOMING, JOURNAL OF FINANCE

Keywords: Equity firms, capital structure, debt free firms, debt and stocks.

Do Stock Prices Fully Reflect Information in Accruals and Cash Flows About Future Earnings?

Online Appendix for Overpriced Winners

The Naive Extrapolation Hypothesis and the Rosy-Gloomy Forecasts

Deviations from Optimal Corporate Cash Holdings and the Valuation from a Shareholder s Perspective

Dissecting Anomalies. Eugene F. Fama and Kenneth R. French. Abstract

Internet Appendix for The Joint Cross Section of Stocks and Options *

EFFICIENT MARKETS HYPOTHESIS

This is a working draft. Please do not cite without permission from the author.

HOW TO GENERATE ABNORMAL RETURNS.

Changes in Analyst Coverage: Does the Stock Market Overreact?

Does the Fama and French Five- Factor Model Work Well in Japan?*

Fama-French in China: Size and Value Factors in Chinese Stock Returns

Note on Cost of Capital

A Sensitivity Analysis between Common Risk Factors and Exchange Traded Funds

Investment-Based Underperformance Following Seasoned Equity Offering. Evgeny Lyandres. Lu Zhang University of Rochester and NBER

Reevaluating the CCAPM

SIZE EFFECT ON STOCK RETURNS IN SRI LANKAN CAPITAL MARKET

Vas Ist Das. The Turn of the Year Effect: Is the January Effect Real and Still Present?

Appendix. A. Firm-Specific DeterminantsofPIN, PIN_G, and PIN_B

International Journal of Management Sciences and Business Research, 2013 ISSN ( ) Vol-2, Issue 12

University of California Berkeley

Liquidity Risk and Bank Stock Returns. June 16, 2017

LIQUIDITY, STOCK RETURNS AND INVESTMENTS

Momentum Life Cycle Hypothesis Revisited

Liquidity Variation and the Cross-Section of Stock Returns *

Asymmetric Information and the Impact on Interest Rates. Evidence from Forecast Data

Price, Earnings, and Revenue Momentum Strategies

Transcription:

The Rational Part of Momentum Jim Scott George Murillo Heilbrunn Center for Graham and Dodd Investing Columbia Business School Value Investing Research Consortium 1

Outline The Momentum Effect A Rationality Argument Our Measure of Fundamental Value and its Relation to Stock Returns Time Paths of Fundamental Value and Momentum An Empirical Hypothesis and Tests Value Investing Research Consortium 2

The Current Null Hypothesis: Market Efficiency E.g., Fama, JF,1970; JF, 1991. At any time prices fully reflect all available information. If stocks are ranked by their returns in one period, their returns before and after should equal normal returns (i.e., no Momentum Effect) Value Investing Research Consortium 3

6 month period returns Hypothetical Momentum Deciles if the Market were Efficient 80% 60% 40% 20% 0% -20% -40% -60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 Low Mom 2 3 4 5 6 7 8 9 High Mom Period 0 is the portfolio formation period (6 months) Period 1 is the holding period, skipping one month between formation and holding Periods -1, -2, 2 and 3 are six month periods pre and post formation of the firms Value Investing Research Consortium 4

Data Description U.S. stock returns from CRSP: 1985 2006 Exclude REITs, ADRs, LPs and Closed-end Funds IBES mean EPS estimates for the current fiscal year (FY1), next year (FY2) and long term growth rate (LTG) Value Investing Research Consortium 5

6 month period returns Momentum Deciles (1985 2006) 80% 60% 40% 20% 0% -20% -40% -60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 Low Mom 2 3 4 5 6 7 8 9 High Mom Value Investing Research Consortium 6

Momentum Deciles Returns (1985 2006) Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Low Mom 14.57% 10.08% -37.32% 3.23% 7.24% 10.14% 2 11.11% 8.90% -18.19% 5.09% 6.76% 8.28% 3 9.54% 8.56% -9.15% 6.26% 6.97% 8.01% 4 9.29% 8.95% -2.59% 7.03% 7.41% 7.72% 5 9.33% 8.43% 3.07% 7.17% 7.32% 7.47% 6 9.39% 8.63% 8.58% 7.45% 7.35% 7.63% 7 9.68% 8.88% 14.61% 7.86% 7.55% 7.26% 8 10.10% 9.57% 22.19% 8.36% 7.69% 6.98% 9 11.38% 10.85% 33.96% 8.87% 7.03% 7.02% High Mom 11.74% 14.80% 73.09% 11.33% 7.15% 5.63% Average 10.46% 9.69% 8.80% 7.26% 7.25% 7.56% High - Low -2.82% 4.73%** 110.40%** 8.10%** -0.09% -4.51%** Value Investing Research Consortium 7 `

Momentum Regressions with Individual Stocks Current 6 month return as a function of lagged returns Standard errors corrected using Newey-West estimates R i,t = α + β 1 *R i,t-1 + β 2 *R i,t-2 + β 3 *R i,t-3 + ε i,t Months α β 1 β 2 β 3 R 2 Avg # Obs 251 0.0653** 0.0688** 0.0140 1956 245 0.0704** 0.0053 0.0106 1764 239 0.0744** -0.0368** 0.0085 1615 245 0.0636** 0.0623** 0.0039 0.0233 1763 239 0.0738** 0.0063-0.0357** 0.0185 1605 * Significance to 95% ** Significance to 99% Value Investing Research Consortium 8

A Rational Viewpoint In a noisy rational expectations equilibrium, price changes may signal future value changes More generally, in a model of capital market equilibrium with heterogeneous expectations, informed investors expectations will only be partially revealed Value Investing Research Consortium 9

A Noisy Rational Equilibrium Market On the Impossibility of Informationally Efficient Markets Grossman and Stiglitz, AER, 1980 Information is costly Informed investors must have an incentive to collect it Prices only partially reflect informed investors expectations Since some investors become informed, returns today will, on average, predict future returns (As found Momentum) Efficient Capital Markets: II, Fama, JF, 1991, Since there are surely positive information and trading costs, the extreme version of the market efficiency hypothesis is surely false Value Investing Research Consortium 10

The role of Present Value Many models of capital market equilibrium imply that rational investors use present value formulas Many professional investors use present value formulas in their investment processes If the market is rational, then changes in present value of future cash flows should mirror returns Value Investing Research Consortium 11

Estimating the Change in Fundamental Value Two period Dividend Discount Model: v it = le it+1 /(1+r) + le it+2 /(1+r) 2 E it+1 = w FY1 + (1 w) FY2 E it+2 = w FY2 + (1-w) FY2 (1 + LTG), where w = number of months left in the fiscal year divided by 12 A weighted earnings estimate allows us to capture one year ahead EPS estimates each month We assume changes in near-term earnings expectations capture much of the firm-specific change in fundamental value Cross-sectionally, the change in a firm s fundamental value should be proportional to R v, where R v t = v it / v it-1 Value Investing Research Consortium 12

Sample Distribution by Size and Book-to-Market Avg # of Firms 1977 Low B/M 2 3 4 High B/M Total Size Small 6.1% 6.4% 6.1% 6.0% 5.9% 30.5% 2 6.5% 5.6% 5.0% 3.9% 2.5% 23.5% 3 5.7% 4.2% 3.7% 2.6% 1.5% 17.7% 4 4.7% 3.6% 3.0% 2.2% 1.2% 14.6% Big 5.4% 3.1% 2.3% 1.8% 1.1% 13.7% Total B/M 28.5% 22.8% 20.2% 16.4% 12.0% 100.0% Firms are classified using Fama French quintile breakpoints, which are based on NYSE stocks, however our sample includes NYSE, AMEX and NASDAQ stocks Value Investing Research Consortium 13

Fundamental Values and Returns R t =.064 +.214R vt R 2 =.107 (4.4) (16.9) n =1956 R t =.014 +.203R vt +.704R ind,t R 2 =.173 (1.86) (16.91) (37.5) n = 1956 At individual security level, 6 month returns are significantly correlated with concurrent 6 month fundamental returns Value Investing Research Consortium 14

6 month period returns 6 month change in Value Momentum Deciles (1985 2006) 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% -20% -20% -40% -40% -60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3-60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 Low Mom 2 3 4 5 6 7 8 9 High Mom Low Mom 2 3 4 5 6 7 8 9 High Mom Value Investing Research Consortium 15

Fundamental Value Changes Corresponding to Momentum Deciles (1985 2006) Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Low Mom 18.56% 14.94% -12.86% -5.24% 0.95% 8.42% 2 11.78% 10.91% -1.17% 0.71% 2.88% 7.54% 3 10.29% 8.30% 2.59% 3.01% 3.50% 6.06% 4 8.97% 8.33% 5.59% 4.32% 4.92% 7.02% 5 8.40% 7.58% 7.52% 6.22% 3.54% 3.99% 6 8.31% 8.38% 10.18% 6.71% 4.40% 4.27% 7 9.31% 8.76% 11.67% 7.72% 6.05% 3.81% 8 9.27% 9.66% 14.80% 9.75% 6.45% 3.77% 9 10.92% 12.30% 22.47% 12.90% 7.24% 3.59% High Mom 12.02% 17.52% 43.36% 20.21% 9.86% 3.93% Average 10.61% 10.54% 10.40% 6.62% 5.01% 5.15% High - Low -6.55%** 2.58% 56.22%** 25.46%** 8.91%** -4.49%** Value Investing Research Consortium 16

Average Number of Observations per Momentum Deciles Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Low Mom 146 170 196 196 179 145 2 159 176 196 196 188 166 3 165 179 196 196 191 172 4 166 180 196 196 192 174 5 168 181 196 196 192 177 6 169 181 196 196 193 179 7 168 180 196 196 193 179 8 165 179 196 196 193 178 9 160 176 196 196 193 178 High Mom 142 164 195 195 192 177 Value Investing Research Consortium 17

6 month period returns 6 month change in Value Change in Value (R v,t-6 ) Deciles (1985 2006) 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% -20% -20% -40% -40% -60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3-60% Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 Low Rv 2 3 4 5 6 7 8 9 High Rv Low Mom 2 3 4 5 6 7 8 9 High Mom Value Investing Research Consortium 18

Stock Returns for Change in Value (R v ) Deciles (1985 2006) Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Low R v,t-6 3.22% -8.16% -15.07% 5.42% 8.79% 9.49% 2 5.68% -1.10% -5.91% 5.94% 7.49% 8.05% 3 7.03% 2.93% -0.01% 6.34% 7.32% 7.32% 4 8.00% 5.82% 3.97% 6.79% 6.99% 7.35% 5 9.26% 8.23% 6.85% 7.04% 7.20% 7.28% 6 11.21% 10.71% 9.73% 7.72% 7.20% 7.35% 7 13.56% 13.59% 13.05% 7.63% 6.99% 7.21% 8 16.09% 18.21% 17.05% 7.66% 7.04% 7.28% 9 18.43% 23.70% 23.31% 8.88% 7.09% 7.14% High R v,t-6 13.96% 26.50% 35.15% 9.21% 6.35% 7.56% Average 10.46% 9.69% 8.80% 7.26% 7.25% 7.56% High Low 10.74%** 34.66%** 50.22%** 3.79%** -2.45%** -12.70%** Value Investing Research Consortium 19

Fundamental Value Changes Corresponding to Value Deciles (1985 2006) Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Low R v,t-6 10.21% 10.17% -43.65% 19.19% 13.45% 16.78% 2 6.79% 6.93% -15.30% -0.29% 4.11% 6.75% 3 5.69% 6.63% -5.29% 0.25% 1.36% 3.69% 4 6.00% 6.46% 0.32% 1.86% 1.97% 2.98% 5 7.10% 7.13% 4.15% 2.83% 2.46% 4.45% 6 8.44% 8.49% 7.48% 4.48% 3.23% 3.51% 7 10.59% 10.09% 11.20% 5.87% 4.23% 3.82% 8 13.08% 13.16% 16.48% 7.29% 4.53% 3.83% 9 18.82% 18.04% 26.64% 10.02% 5.67% 3.87% High R v,t-6 24.22% 21.35% 102.32% 14.73% 10.73% 4.08% Average 10.61% 10.54% 10.40% 6.62% 5.01% 5.15% High Low 14.01%** 11.18%** 145.96%** -4.46% -2.72% -12.70%** Value Investing Research Consortium 20

An Empirical Hypothesis Current prices partially reflect future changes in the market s estimate of fundamental value P t = l t V t b 0(1+Rv,t+1 ) b 1 (1+Rv,t+2 ) b 2 Taking logarithms and first differences yields a regression equation R t = a + b 0 r v,t + b 1 (r v,t+1 r v,t ) + b 2 (r v,t+2 r v,t+1 ) + u t Value Investing Research Consortium 21

Returns and Future Value Changes R i,t = α + β 0 *R v,t + β 1 *(R v,t+1 - R v,t ) + β 2 *(R v,t+2 - R v,t+1 ) + β 3 *(R v,t+3 - R v,t+2 ) + u t Months α β 0 β 1 β 2 β 3 R 2 Avg # Obs 250 0.0704 0.3433 0.1434 0.0383 0.1234 1900 4.89 18.94 13.46 6.27 244 0.0740 0.3456 0.1291 0.0109-0.0183 0.1329 1720 5.09 16.90 9.79 1.08-2.45 Value Investing Research Consortium 22

Observed Returns, Sorting on Past Returns and Concurrent R v Low R v 2 3 4 5 6 7 8 9 High R v Avg. Low Mom -15.8% -3.4% 3.2% 8.4% 9.9% 14.3% 15.9% 21.0% 24.5% 31.9% 3.2% 2-17.4% -6.2% 1.3% 5.8% 9.7% 12.9% 16.3% 18.4% 23.8% 30.7% 5.1% 3-17.8% -5.9% 0.2% 4.9% 8.3% 11.5% 14.2% 17.9% 23.1% 32.1% 6.3% 4-18.0% -6.8% -0.5% 4.6% 7.7% 10.7% 14.6% 17.7% 22.0% 30.3% 7.0% 5-18.4% -7.5% -1.5% 3.0% 7.2% 10.0% 12.7% 17.6% 21.7% 30.5% 7.2% 6-18.5% -8.4% -1.9% 2.7% 6.0% 9.3% 11.8% 16.0% 21.0% 32.0% 7.5% 7-20.5% -10.5% -2.4% 2.0% 5.0% 8.2% 11.6% 16.4% 21.8% 31.4% 7.9% 8-19.2% -11.3% -4.0% 1.3% 4.0% 7.3% 11.0% 15.4% 22.2% 30.5% 8.4% 9-23.6% -12.7% -4.9% -1.0% 2.0% 5.1% 9.0% 13.6% 20.5% 32.5% 8.9% High Mom -25.2% -16.4% -7.4% -4.1% -1.9% 0.7% 6.6% 11.0% 18.7% 35.6% 11.3% Avg. -18.0% -7.6% -1.2% 3.2% 6.1% 8.9% 11.9% 15.8% 21.3% 32.5% 7.3% Value Investing Research Consortium 23

Summary Our measure of change in fundamental value is significantly correlated with stock returns Momentum Effect is consistent with rational behavior, either: A Noisy Rational Expectations Equilibrium, or more generally, A properly specified heterogeneous expectations equilibrium On average, stock prices reflect not only current fundamental value estimates, but informed investors expectations of fundamental value as much as 12 months into the future As a result, the Momentum returns are highly correlated with concurrent fundamental return and appear to predict changes in fundamental return over the next year. Value Investing Research Consortium 24

Thank you Value Investing Research Consortium 25

Correlations for Returns and Change in Value Change in Value Return Period -2 Period -1 Period 0 Period 1 Period 2 Period 3 t-18 to t-12 t-12 to t-6 t-6 to t t+1 to t+7 t+7 to t+13 t+13 to t+19 Period -2 0.131** 0.007** -0.008** -0.001-0.009** -0.007** Period -1 0.067** 0.133** 0.004** -0.010** 0.001-0.008** Period 0 0.026** 0.064** 0.131** 0.002-0.011** 0.001 Period 1-0.026** 0.019** 0.047** 0.135** 0.003-0.005** Period 2-0.004** -0.018** 0.012** 0.044** 0.089** 0.003 Period 3-0.010** -0.003-0.016** 0.016** 0.048** 0.090** Value Investing Research Consortium 26

Average Number of Observations Low R v 2 3 4 5 6 7 8 9 High R v Avg. Low Mom 62 32 19 12 9 9 9 10 12 21 196 2 34 31 25 19 16 14 13 14 14 16 196 3 22 26 26 24 21 18 16 15 14 15 196 4 15 22 25 25 24 21 19 16 14 13 196 5 12 18 23 26 26 24 21 18 15 12 196 6 11 16 20 25 26 26 23 20 17 13 196 7 10 14 18 22 25 26 25 23 19 14 196 8 9 13 16 19 22 25 27 25 23 18 196 9 10 12 14 14 17 21 25 29 30 25 196 High Mom 11 12 10 10 10 13 18 26 39 47 195 Avg. 196 196 196 196 196 196 196 196 196 195 1958 Value Investing Research Consortium 27