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

Transcription

1 The Performance, Pervasiveness and Determinants of Value Premium in Different US Exchanges George Athanassakos PhD, Director Ben Graham Centre for Value Investing Richard Ivey School of Business The University of Western Ontario

2 Motivation A large body of academic research has documented the existence of a value premium (Basu (1977), Chan, Hamao and Lakonishok (1991), Fama and French (1992, 1993, 1996), Lakonishok, Shleifer and Vishny (1994) and Chan and Lakonishok (2004)) As a result, most arguments in academia now revolve around the reasons of such superior performance of value stocks

3 Motivation Proponents of efficient markets, such as Fama and French (1992, 1993, 1996, 1998), argue that the value premium exists because value stocks bear more risk Others, such as La Porta, Lakonishok, Shleifer and Vishny (1997), and Chan and Lakonishok (2004), argue against market efficiency and rational pricing They advocate that systematic errors made by investors and agency problems faced by institutional investors prevent the value premium from disappearing

4 Motivation Doukas, Kim and Pantzalis (2002) reject the non-risk based explanation of the value premium, while Doukas, Kim and Pantzalis (2004) find support for a risk based explanation of the value premium Petkova and Zhang (2005) argue that value stocks are riskier than growth stocks, at least in the adverse states of the world Phalippou (2004) finds evidence consistent with behavioral explanations of the value premium and so is Lettau and Wachter (2007)

5 Motivation Consequently, the jury is then still out with regards to the drivers of the value premium

6 Objectives Using US data for each of AMEX, NASDAQ and NYSE for the period , the purpose of this paper is: To shed further light on the performance of value and growth stocks To shed light into the argument of whether the value premium is driven by risk or behavioral factors The paper will provide tests that prevent confounding inferences, utilizing a more comprehensive set of data and tests than previous studies and a research methodology that minimizes any potential data snooping problems

7 Contribution I document a consistently strong value premium over the sample period, which persists in both bull and bear markets, as well as in recessions and recoveries I show that the value premium is not driven by a few outliers, but it is pervasive as the overwhelming majority of stocks in the value portfolio have positive returns, and the majority of industries in the sample contain stocks that have positive value premiums The value premium, in general, remains positive and statistically significant over time

8 Contribution In terms of explaining the drivers of the value premium, and having looked at this question from many angles, the findings seem to support the notion that it what drives the value premium is (Most likely) errors in expectations Not risk, as argued by the market efficiency school of thought

9 Proxy Variables Employed Risk Proxies Recessions & Bear Markets Liquidity Standard deviation of returns Dispersion of analysts forecasts Behavioral Proxies % of Institutional holdings Firm size Analyst following Analyst Optimism

10 Data This study uses data from four data bases The first data base is the I/B/E/S data base from which data on analysts mean and median EPS forecasts, forecast dispersion, analyst optimism and number of analysts are obtained/calculated The second data base is the CRSP database from which prices per share, shares outstanding, trading volume and stock returns are obtained and used to estimate market cap, the measure of liquidity and the standard deviation of stock returns The third is CDA/Spectrum from which the number of institutions holding a stock, the number of shares of a given stock held by institutions, shares outstanding are obtained The final database is COMPUSTAT from which trailing EPS and BVPS and the industry in which a stock belongs are obtained

11 Data The timing of recessions/recoveries and bear/bull markets is obtained from The data is for each month in the 1985 through 2006 sample period After all screens and further adjustments for missing observations, the intersection of the four databases resulted in a total of 12,804, 313,779 and 344,712 cross sectional-time series observations for the final sample of 583, 4908 and 2977 AMEX, NASDAQ and NYSE firms, respectively, representing 10 industries as classified by the 1-digit Standard Industrial Classification (SIC) code

12 Methodology Starting in June 1985, firms are ranked based on P/E (trailing) and P/BV ratios from low to high and the ranked firms are divided into four groups of equal size Returns are then obtained for each month over the following year (starting in July 1985) for each stock within each quartile and equally weighted mean returns for each quartile are derived This process is repeated for each year of the sample Quartile-1 (Q1) is the low P/E (P/BV) ratio quartile or the value stocks, while Quartile-4 (Q4) is the high P/E (P/BV) ratio quartile or the growth stocks A time series of non-overlapping monthly returns are obtained for each quartile from July 1985 to June 2006, sub-periods, recessions/recoveries, and bear/bull markets

13 Methodology I carry out uni(bi)variate analysis that looks at value and growth stock performance and carries out a first stage examination of the drivers of the value premium To further examine the drivers of the value premium, I also carry out regression analysis and robustness tests First, I regress subsequent returns of the value and growth stocks against a number of explanatory variables drawn from previous research and the paper s earlier findings Second, I examine the value premium using trailing P/E as of December of year (t-1) and forward P/E ratio Finally, I examine whether the risks specified by a formal asset pricing model, such as the Fama and French three factor model, explain the returns of the value and growth stocks

14 Empirical Results

15 Figure Mean Monthly Value Premia to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Year: June July AMEX NASDAQ NYSE

16 Figure 2 Mean Monthly Value Premia to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Subperiod and State of the World: June July AMEX NASDAQ NYSE

17 Figure Mean Monthly Value Premia to P/BV Ratio (June, Trailing) Based Value and Growth Strategies by Year: June July AMEX NASDAQ NYSE

18 Figure 4 Mean Monthly Value Premia to P/BV Ratio (June, Trailing) Based Value and Growth Strategies by Subperiod and State of the World: June July AMEX NASDAQ NYSE

19 Percentage of Positive and Negative Monthly AMEX Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies: July June 2006 P/E Ratio Sorted Quartiles Q1 (Value) Q4 (Growth) % Positive % Negative

20 Percentage of Positive and Negative Monthly NASDAQ Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies: July 1985-June 2006 P/E Ratio Sorted Quartiles Q1 (Value) Q4 (Growth) % Positive % Negative

21 Percentage of Positive and Negative Monthly NYSE Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies: July June 2006 P/E Ratio Sorted Quartiles Q1 (Value) Q4 (Growth) % Positive % Negative

22 Mean Monthly AMEX Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Industry: July 1985-June 2006 No of OBS Name Industry Q1 (Value) Q4 (Growth) Value Premium Q1 Q4 (P-Values) N/A Agriculture, Forestry and Fishing N/A 408 Mining Construction N/A 1350 Manufacturing Transporation & Public Utilities Wholesale Trade Retail Trade Finance, Insurance & Real Estate Services N/A Public Administration N/A

23 Mean Monthly NASDAQ Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Industry: July 1985-June 2006 No of OBS Name Industry Q1 (Value) Q4 (Growth) Value Premium Q1 Q4 (P-Values) 328 Agriculture, Forestry and Fishing Mining Construction Manufacturing Transporation & Public Utilities Wholesale Trade Retail Trade Finance, Insurance & Real Estate Services N/A Public Administration N/A

24 Mean Monthly NYSE Stock Returns to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Industry: July 1985-June 2006 No of OBS Name Industry Q1 (Value) Q4 (Growth) Value Premium Q1 Q4 (P-Values) 360 Agriculture, Forestry and Fishing Mining Construction Manufacturing Transporation & Public Utilities Wholesale Trade Retail Trade Finance, Insurance & Real Estate Services Public Administration

25 Mean Monthly AMEX Stock Returns, Sigma, Dispersion, # Analysts, Size, Optimism and Liquidity to P/E Ratio (June, Trailing) Based Value and Growth Strategies by % Inst Based categories: July 1985-June 2006 Q1 (Low % Inst.) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q4 (High % Inst.) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity

26 Mean Monthly NASDAQ Stock Returns, Sigma, Dispersion, # Analysts, Size, Optimism and Liquidity to P/E Ratio (June, Trailing) Based Value and Growth Strategies by % Inst Based categories: July 1985-June 2006 Q1 (Low % Inst.) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q4 (High % Inst.) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($M) Optimism Liquidity

27 Mean Monthly NYSE Stock Returns, Sigma, Dispersion, # Analysts, Size, Optimism and Liquidity to P/E Ratio (June, Trailing) Based Value and Growth Strategies by % Inst Based categories: July 1985-June 2006 Q1 (Low % Inst.) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q4 (High % Inst.) Returns Sigma Dispersion # Analysts Size ($000) Optimism Liquidity Q1 Q4 (P-Values) Returns Sigma Dispersion # Analysts Size ($M) Optimism Liquidity

28 Mean Monthly AMEX Stock Returns, Dispersion, # Analysts, Size, Optimism, Liquidity and %Inst to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Sigma Based categories: July June 2006 Q1 (Low Sigma) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q4 (High Sigma) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($M) Optimism Liquidity % Inst

29 Mean Monthly NASDAQ Stock Returns, Dispersion, # Analysts, Size, Optimism, Liquidity and %Inst to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Sigma Based categories: July June 2006 Q1 (Low Sigma) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q4 (High Sigma) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($M) Optimism Liquidity % Inst

30 Mean Monthly NYSE Stock Returns, Dispersion, # Analysts, Size, Optimism, Liquidity and %Inst to P/E Ratio (June, Trailing) Based Value and Growth Strategies by Sigma Based categories: July June 2006 Q1 (Low Sigma) Q1 (Value) Q4 (Growth) Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q4 (High Sigma) Returns Dispersion # Analysts Size ($000) Optimism Liquidity % Inst Q1 Q4 (P-Values) Returns Dispersion # Analysts Size ($M) Optimism Liquidity % Inst

31 Regression Estimates of Monthly AMEX Stock Returns Against P/E Ratios, # Analysts, Liquidity, Sigma and Dispersion (Fama-MacBeth Procedure): July 1985-June 2006 P/E Ratio Sorted Quartiles Independent Variables Total Sample Q1 (Value) Q4 (Growth) Intercept (p-value) (0.05) (0.02) (0.58) PERATIO (p-value) (0.42) (0.39) (0.32) #ANALYSTS (p-value) (0.58) (0.11) (0.14) LLIQUID (p-value) (0.90) (0.32) (0.93) LSIGMA (p-value) (0.03) (0.00) (0.01) LDISP (p-value) (0.00) (0.10) (0.25) Adjusted R (p-value) (0.00) (0.09) (0.04)

32 Regression Estimates of Monthly NASDAQ Stock Returns Against P/E Ratios, # Analysts, Liquidity, Sigma and Dispersion (Fama-MacBeth Procedure): July 1985-June 2006 P/E Ratio Sorted Quartiles Independent Variables Total Sample Q1 (Value) Q4 (Growth) Intercept (p-value) (0.42) (0.08) (0.06) PERATIO (p-value) (0.00) (0.05) (0.19) #ANALYSTS (p-value) (0.00) (0.17) (0.05) LLIQUID (p-value) (0.16) (0.06) (0.63) LSIGMA (p-value) (0.33) (0.30) (0.04) LDISP (p-value) (0.00) (0.00) (0.00) Adjusted R (p-value) (0.00) (0.00) (0.01)

33 Regression Estimates of Monthly NYSE Stock Returns Against P/E Ratios, # Analysts, Liquidity, Sigma and Dispersion (Fama-MacBeth Procedure): July 1985-June 2006 P/E Ratio Sorted Quartiles Independent Variables Total Sample Q1 (Value) Q4 (Growth) Intercept (p-value) (0.58) (0.21) (0.24) PERATIO (p-value) (0.81) (0.40) (0.21) #ANALYSTS (p-value) (0.00) (0.80) (0.04) LLIQUID (p-value) (0.03) (0.82) (0.38) LSIGMA (p-value) (0.50) (0.40) (0.19) LDISP (p-value) (0.00) (0.00) (0.03) Adjusted R (p-value) (0.04) (0.00) (0.05)

34 Time Series Regression Results of Size and Book-to- Market (B/M) Stock Portfolio Monthly AMEX Returns on the Three Fama-French Factors: July 1985-June 2006 Sub-sample Intercept RMF SMB HML R-Square a b c d Small Value (P-Value) (0.78) (0.05) (0.03) (0.00) Small Growth (P-Value) (0.42) (0.01) (0.00) (0.00) Large Value (P-Value) (0.48) (0.39) (0.41) (0.00) Large Growth (P-Value) (0.27) (0.52) (0.67) (0.20)

35 Time Series Regression Results of Size and Book-to- Market (B/M) Stock Portfolio Monthly NASDAQ Returns on the Three Fama-French Factors: July 1985-June 2006 Sub-sample Intercept RMF SMB HML R-Square a b c d Small Value (P-Value) (0.63) (0.00) (0.00) (0.00) Small Growth (P-Value) (0.38) (0.00) (0.00) (0.00) Large Value (P-Value) (0.22) (0.25) (0.08) (0.00) Large Growth (P-Value) (0.43) (0.39) (0.48) (0.15)

36 Time Series Regression Results of Size and Book-to- Market (B/M) Stock Portfolio Monthly NYSE Returns on the Three Fama-French Factors: July 1985-June 2006 Sub-sample Intercept RMF SMB HML R-Square a b c d Small Value (P-Value) (0.51) (0.00) (0.00) (0.00) Small Growth (P-Value) (0.30) (0.00) (0.00) (0.00) Large Value (P-Value) (0.12) (0.36) (0.19) (0.00) Large Growth (P-Value) (0.20) (0.43) (0.30) (0.30)

37 Conclusions I document a consistently strong value premium over the sample period, which persists in both bull and bear markets, as well as in recessions and recoveries I show that the value premium is not driven by a few outliers, but it is pervasive as the overwhelming majority of stocks in the value portfolio have positive returns, and the majority of industries in the sample contain stocks that have positive value premiums The value premium, in general, remains positive and statistically significant over time

38 Conclusions In terms of explaining the drivers of the value premium, and having looked at this question from many angles, the findings seem to support the notion that what drives the value premium is (Most likely) errors in expectations Not risk, as argued by the market efficiency school of thought.

39 Motivation of Extensions The problem with the academic classification of stocks into value and growth is that such stock selection approach is only part of what value investors do! This is the first step value investors follow in stock selection, which helps them identify possibly undervalued stocks.

40 Motivation of Extensions Once stocks are screened out, value investors then proceed to find stocks that are truly undervalued by valuing individually each stock and arriving at their investment decision. And it is this step in particular that previous academic research has not examined.

41 Objectives of Extensions Using Canadian data for the period May 1999 to April 2007, this paper has two objectives: To confirm that a value premium exists in my sample of stocks using a search process that consists of cross-sorting stocks by both P/E and P/BV ratios. To examine whether the second step of stock selection that value investors follow adds any value. That is: Do value investors add any value?

42 Contribution of Extensions There is a strong and pervasive value premium in Canada over the sample period that persists in a bull and bear market and during a recession/recovery. Value investors do add value, in the sense that their process of selecting truly undervalued stocks, via in-depth security valuation, produces significantly positive excess returns over and above a naive approach of simply selecting low P/E and P/BV ratio stocks.

43 Data Data are from COMPUSTAT from which trailing price to earnings (P/E) and price to book value (P/BV) ratios and market cap and total stock returns are derived. For the trailing P/E and P/BV ratios, the price (P) is as of the end of April of year (t) and E and BV are, respectively, the December (t-1) basic annual earnings per share and book value per share for companies with fiscal year end December (t-1), as reported in COMPUSTAT.

44 Data Companies are from the Toronto Stock Exchange (TSX). Companies are not in the financial services or income trust sectors. Negative P/E and P/BV ratios, as well as P/E ratios in excess of 150 and P/BV in excess of 20 are excluded. Firms had to have both P/E and P/BV ratios within the aforementioned boundaries to be included in the sample. To be included in the sample a stock had to have a price over $1 and fiscal year end December.

45 Data The data, which are adjusted for stock splits and stock dividends, are for each year in the 1999 to 2007 period. After all aforementioned screenings, there are 1301 cross sectional-time series (firm-year) observations left belonging to a cumulative number of 377 companies over the sample period. Since the sample only includes firms with fiscal year end December of year (t-1), all firms have released their annual reports needed for the valuations and information for EPS and BVPS by April of year (t).

46 Range of P/E P/BV ratios per year for the low P/E low P/BV basket (Q1) and the high P/E high P/BV basket (Q4) Q1 (Value) Q4 (Growth) Year P/E P/BV P/E P/BV 1999 Max Min Max Min Max Min Max Min Max Min Max Min Max Min Max Min

47 Number of observations for each basket per year Q1 (Value) Q4 (Growth) Year Number of observations Number of observations Total 79 85

48 Methodology To determine the truly undervalued stocks, all naively chosen stocks from Q1 were individually valued. The annual reports of the companies in question were obtained from Sedar.com.

49 Methodology For each stock in Q1, two valuations were carried out. First, the net replacement value of each company s assets (called Net Asset Value - NAV) was estimated. Second, a Free cash Flow (FCF) based valuation for each company was produced (called Earnings Power Value - EPV), by normalizing FCFs and discounting them to infinity using a perpetuity formula.

50 Methodology Where exactly the company s intrinsic value lies depends on strategic analysis and the probabilities of possible outcomes. Most of the time the Intrinsic Value lies between NAV and EPV. The Entry Price was set at 2/3 of the Intrinsic Value, i.e., Intrinsic Value less 1/3 Margin of Safety. If a stock's current price was below the Entry Price, a decision was made to purchase the stock.

51 Final number of stocks per year in the invested sophisticated portfolio (Q1S) Year # of Stocks in Sophisticated Portfolio

52 Mean and Median Annual (%) Value Premia to P/E P/BV Ratio Based Value (Q1) and Growth (Q4) Strategies by Year: May 1999 April Mean Median

53 Mean Annual (%) Returns to P/E P/BV Ratio Based Value (Q1) and Growth (Q4) Strategies by Year: May 1999 April 2007 Mean Return Value Premium Year Q1 Q4 Q1-Q % 10.9% -5.2% % 4.8% -3.3% % 9.7% 35.7% % -4.2% -0.4% % 29.7% 63.1% % 33.4% -0.9% % 53.2% 31.6% % 5.6% 12.2% Overall average 34.5% 17.9% 16.6% Variance 14.0% 3.6% Risk-free rate 3.6% 3.6% SHARPE ratio

54 Median Annual (%) Returns to P/E P/BV Ratio Based Value (Q1) and Growth (Q4) Strategies by Year: May 1999 April 2007 Median Return Value Premium Year Q1 Q4 Q1-Q % -4.8% 6.2% % -17.7% 18.3% % 8.4% 11.6% % -8.8% 10.7% % 26.0% 63.4% % 16.2% 12.4% % 34.4% 7.7% % 2.5% 20.4% Overall average 25.9% 7.0% 18.8% Variance 8.8% 3.2% Risk-free rate 3.6% 3.6% SHARPE ratio

55 Mean and Median Annual (%) Returns to P/E P/BV Ratio Based Naïve Value (Q1) and Sophisticated Value (Q1S) Strategies by Year: May 1999 April Mean Median

56 Mean Annual (%) Returns to P/E P/BV Ratio Based Naïve Value (Q1) and Sophisticated Value (Q1S) Strategies by Year: May 1999 April 2007 Mean Return Value Investor Premium Year Q1S Q1 Q1S - Q % 5.7% 0.0% % 1.5% 12.4% % 45.4% 26.4% % -4.6% 32.2% % 92.8% 7.6% % 32.5% -7.8% % 84.8% 27.9% % 17.8% 7.2% Overall average 47.7% 34.5% 13.2% Variance 17.1% 14.0% Risk-free rate 3.6% 3.6% SHARPE ratio

57 Median Annual (%) Returns to P/E P/BV Ratio Based Naïve Value (Q1) and Sophisticated Value (Q1S) Strategies by Year: May 1999 April 2007 Median Return Value Investor Premium Year Q1S Q1 Q1S - Q % 1.4% 0.0% % 0.6% 9.1% % 20.1% 26.2% % 1.8% 23.8% % 89.4% -55.4% % 28.6% -3.9% % 42.1% 73.4% % 22.9% 2.1% Overall average 35.3% 25.9% 9.4% Variance 12.4% 8.8% Risk-free rate 3.6% 3.6% SHARPE ratio

58 Conclusions of Extensions A strong and pervasive value premium existed in Canada over the sample period that persisted in a bull and bear market and during a recession/recovery. Value stocks beat growth stocks even when using a very mechanical screening of the search process. Value investors do add value, in the sense that their process of selecting truly undervalued stocks, via in-depth security valuation of the possibly undervalued stocks, produced significantly positive excess returns over and above the naive approach of simply selecting low P/E - P/BV ratio stocks. Value investors proceeding to their second step of the stock selection process do add value.