CAPM AND QUALITY STOCKS

Transcription

1 NVESTMENT NSGHTS March 2012 CAPM AND QUALTY STOCKS THE PROBLEM WTH BETA s beta an effective predictive measure of systematic risk? n this paper, Perkins nvestment Management evaluates how beta and the capital asset pricing model have largely failed to accurately forecast stock performance relative to market movements over the past 20 years. Our research has found that these metrics have generally overestimated the return potential of low-quality stocks and underestimated their downside exposure. They also have discounted the up-capture of high-quality stocks. As a result, highquality stocks have benefited from stronger performance in rising markets than predicted by beta, with less risk than low-quality stocks. This has generally delivered higher, more consistent returns over full market cycles, providing greater compounding efficiency. THE LMTATONS OF BETA Beta is one of the most frequently utilised systematic risk metrics and remains a core component of the capital asset pricing model (CAPM), a stock valuation methodology first developed in the 1960s. Beta measures the volatility of a stock in comparison to the market, and can be thought of as the tendency of a stock s returns to respond to swings in the market. t is calculated by running a linear regression of a stock s return against the overall market or a benchmark index. This measure of relative volatility is then used in CAPM to forecast expected performance based on anticipated market returns. Stocks with larger betas are assumed to exhibit higher levels of volatility in reaction to market movements. For example, a stock with a 1.0 beta is expected to move exactly with the market, and therefore would result in a 1% respective gain or loss for each 1% rise or fall in the market. A stock with a 2.0 beta is expected to move at twice the rate of the market, and so would rise or fall 2% with every corresponding 1% rise or fall in the market (see Exhibit 1). Conversely, a lessrisky stock with a 0.5 beta would be expected to exhibit lower volatility, at half the rate of the market, rising or falling 0.5% with respect to a 1% market change. Stock Return (%) -10 EXHBT 1: MEASURE OF SYSTEMATC RSK 10 Beta = 2.0 Beta = Beta = Source: Perkins Market Return (%) ntroduced by Janus Capital nternational Limited

2 nvestors employing beta as a research tool, however, should be aware of its somewhat problematic limitations. A growing body of research has demonstrated that stock price movements with respect to the market are rarely linear. We recently released research (High Quality nvesting: Compound the Benefits, September 2011) demonstrating the benefits of investing in high-quality stocks. n this paper, we expand on those findings by examining how stock quality affects actual price movements in relation to beta and the CAPM, grouped by Perkins Quality methodology. n our analysis, we found that beta significantly underestimates both the up-capture of high-quality stocks and downside risk of low-quality stocks. Accordingly, investors in highquality stocks have benefited from greater up-capture than predicted by beta, with less downside exposure than low-quality stocks. Hence, high-quality securities have offered more consistent returns with less risk and greater compounding efficiency. Methodology Time Period: 20 years (January 1, 1992 December 31, 2011). Data Source: FactSet and Fama/French. Fama/French The value-weighted return of all stocks on the NYSE, AMEX and NASDAQ exchanges. Broad-Market ndex: Universe: At the end of each month, a universe was created using the 1,000 largest U.S. stocks, based on market capitalisation, herein referred to as the Market Universe. The term market does not and should not imply the returns of any particular index or benchmark. Quality Scores: At the end of each quarter each stock in the Market Universe was ranked using an average quality score based on six equally weighted quality factors (shown in the table below) and placed into five portfolios of approximately 200 stocks. The fundamental ranks were lagged one quarter. For example, first quarter financials published in the second quarter were used to rank stocks in the third quarter. Quintile Definitions: Returns: The top quintile portfolio (quintile 5) is defined as high-quality stocks, while the bottom quintile portfolio (quintile 1) is defined as low-quality stocks. Performance was calculated monthly based on market cap weighted portfolios for each of the five quintiles, and the Market Universe. Fees and expenses were not included, and dividends were reinvested. Up-Capture: Down-Capture: Beta: Expected CAPM Return: Risk-Free Rate: The ratio of eac h quality quintile portfolio s return to the Market Universe s annualised return in months in which the overall market (as determined by the Fama/French Broad-Market ndex) was up. The ratio of each quality quintile portfolio s return to the Market Universe s annualised return in months in which the overall market (as determined by the Fama/French Broad-Market ndex) was down. Calculated by running a linear regression of each quality quintile portfolio s monthly return against the monthly returns of the Fama/French Broad-Market ndex. Calculated for each of the highest and lowest quality quintiles as follows: Beta of quality quintile *(Market Universe average monthly return - Risk-Free Rate) + Risk-Free Rate. Calculated as an average of the one-month U.S. Treasury bill rate, as reported by bbotson Associates, monthly for the 20-year Time Period. Quality Factors ndustry Rank of Leverage (Debt/Equity) ndustry Rank of nterest Coverage (EBTDA 1 /nterest Expense) Predictability of EBTDA Over Trailing 12 Quarters Percent of Quarters EBTDA ncreased Over Trailing 12 Quarters ROA 2 ndustry Rank (Net ncome/assets) Percent of Quarters ROA in Top Third of ndustry Group Over Trailing 12 Quarters (Net ncome/assets) Category Balance Sheet Strength Balance Sheet Strength Earnings Stability Earnings Stability Operating Efficiency Operating Efficiency 1 Earnings Before nterest, Taxes, Depreciation, and Amortization. 2 Return on Assets. 2 CAPM and Quality Stocks: The Problem with Beta NVESTMENT NSGHTS

3 UP-CAPTURE ANALYSS CAPM implies that investors are compensated for risk. Therefore, low-quality stocks with leveraged balance sheets and less-predictable business models should compensate investors with greater market returns in the form of larger up-capture ratios. Unfortunately, over the past 20 years, beta has overestimated the up-capture for low-quality stocks, while underestimating the up-capture for high-quality stocks. We found the average up-capture for stocks in the lowest quintile during months with a positive general market performance to be 1.13, well below the 1.17 expected from the group s beta (see Exhibit 2). n comparison, high-quality stocks captured 0.91 of market performance in up months, considerably higher than what would be expected based on the segment s 0.85 beta. Consequently, it appears that, on average, low-quality, high-beta stocks have failed to compensate investors adequately during up markets, given their additional risk. n contrast, the upside cost for high-quality stocks has been substantially less than expected, with much greater performance delivered in rising markets than what these securities beta would suggest. DOWN-CAPTURE ANALYSS nvestors generally expect high-quality stocks from companies with less debt and more stable businesses to retain greater value in falling markets than low-quality stocks. To help quantify the downside risk of a portfolio, investors have traditionally looked to beta. However, beta has significantly underestimated the down-capture of low-quality stocks. This might lead low-quality stock investors to believe that their portfolios contain less risk than they actually do. For example, the average down-capture for high-quality stocks during months experiencing a general market decline has been 0.86, well in line with the segment s 0.85 beta (see Exhibit 3). The average down-capture for lower-quality stocks has been greater than what might be predicted by each group s beta, a trend most pronounced in the lowestquality quintile where average down-capture in negative months has been 1.22 compared to the segment s 1.17 beta. Coupled with the up-capture findings presented earlier, it appears that over several market cycles investors in low-quality stocks have been constrained on two fronts: greater-than-expected average losses during down markets and less-than-expected average price appreciation during up markets. Up-Capture/Beta EXHBT 2: UP-CAPTURE VS BETA Beta significantly underestimates up-capture of high-quality stocks Up-Capture Beta Down-Capture/Beta EXHBT 3: DOWN-CAPTURE VS BETA Beta significantly underestimates down-capture of low-quality stocks Down-Capture Beta (high) (low) (high) (low) Quality Quintile Quality Quintile Up-capture and beta were calculated for each of the five quality quintiles using 20 years of monthly returns (January 1992 December 2011). See Methodology for more information on the calculations and how the quality portfolios were constructed. Source: FactSet Down-capture and beta were calculated for each of the five quality quintiles using 20 years of monthly returns (January 1992 December 2011). See Methodology for more information on the calculations and how the quality portfolios were constructed. Source: FactSet NVESTMENT NSGHTS CAPM and Quality Stocks: The Problem with Beta 3

4 MPLCATONS FOR NVESTORS The CAPM calculates an expected return for a portfolio by multiplying its beta with the market return in excess of the risk-free rate and then adding the risk-free rate to this total. Expected Market Return = Beta * (Market Risk Free) + Risk Free Over the past 20 years, the market cap weighted portfolio of the Market Universe has returned, on average, 0.77% per month. During the period, this portfolio has risen in 149 months (62% of the time), with an average 3.41% gain, and has declined in 91 months (38% of the time), with an average -3.56% loss. Based on these returns, our evaluation found that: Low-quality stocks have disappointed on both the upside and downside, relative to their CAPM forecasts: Applying the low-quality stocks 1.17 beta to CAPM, investors would expect that low-quality stocks should have returned 3.94%, on average, in rising months, compared to the actual 3.80%, and -4.20%, on average, in down months, rather than the reali ed return of -4.55%. Furthermore, this beta suggests that lowquality stocks should have returned 0.85% per month, on average, for the entire period, but the actual monthly return has been much lower at 0.63% (see Exhibit 4). High-quality stocks have outperformed on the upside relative to their CAPM forecasts: Using the high-quality stocks 0.85 beta, investors would have expected high-quality stocks to have gained 2.93%, on average, in rising months, underestimating the actual 3.16% return, and overstating an expected -2.98% loss, on average, in down months, compared to the actual -2.97% return. This beta also suggests that high-quality stocks should have returned 0.69% per month for the entire period, but the actual 0.84% monthly return has been notably higher (see Exhibit 5). This average monthly outperformance of high-quality stocks has offered considerable return advantages over longer time horizons. Given their stronger downside protection and solid upside capture over the past 20 years, high-quality stocks have outperformed low-quality stocks by 3.6% per year with 26% less risk with the difference in risk measured by standard deviation (see Exhibit 6). EXHBT 4: LOW-QUALTY STOCK PERFORMANCE Relative to CAPM forecasts, low-quality stocks have disappointed in both up and down markets % of Months Market Universe Return Beta Expected CAPM Return Actual Return Difference Up Market 62% Down Market 38% Total 100% (January 1992 December 2011) See Methodology for more information on how Returns, Beta and Expected CAPM Return were calculated. Source: FactSet and Fama/French. EXHBT 5: HGH-QUALTY STOCK PERFORMANCE High-quality stocks have outperformed in up markets relative to CAPM forecasts % of Months Market Universe Return Beta Expected CAPM Return Actual Return Difference Up Market 62% Down Market 38% Total 100% (January 1992 December 2011) See Methodology for more information on how Returns, Beta and Expected CAPM Return were calculated. Source: FactSet and Fama/French. 4 CAPM and Quality Stocks: The Problem with Beta NVESTMENT NSGHTS

5 EXHBT 6: CUMULATVE 20-YEAR PERFORMANCE High-quality stocks have outperformed low-quality stocks with less volatility 20% 15% 10% 5% Market Universe High-Quality Stocks Low-Quality Stocks n Annualised Return n Standard Deviation (January 1992 December 2011) Source: FactSet and Fama/French CONCLUSON n this paper, we have demonstrated how CAPM overstates the return potential of low-quality stocks, while discounting the performance of high-quality stocks. This discrepancy is the result of beta s underestimation of the up-capture in high-quality stocks, coupled with its up-capture overestimation and down-capture underestimation for low-quality stocks. n fact, high-quality stocks have achieved, on average, higher returns with less risk than low-quality stocks, offering investors greater compounding efficiency by benefiting from greater up-capture than predicted by beta and by avoiding the considerable downside demonstrated by low-quality stocks. Throughout our 31-year history, Perkins nvestment Management has created a powerful investment process that is designed to produce a pattern of returns similar to that of high-quality stocks. Our research reaffirms our conviction in the efficiency of our investment philosophy and process to identify high-quality stocks that we believe are undervalued. NVESTMENT NSGHTS CAPM and Quality Stocks: The Problem with Beta 5

6 This page intentionally left blank 6 CAPM and Quality Stocks: The Problem with Beta NVESTMENT NSGHTS

7 This page intentionally left blank NVESTMENT NSGHTS CAPM and Quality Stocks: The Problem with Beta 7

8 ssued in Europe by Janus Capital nternational Limited, authorised and regulated by the Financial Services Authority. ssued in: (a) Taiwan R.O.C by Janus Capital nternational Limited, authorised and regulated by the Financial Services Authority of the United Kingdom; (b) Hong Kong and Australia by Janus Capital Asia Limited (ARBN ), which is incorporated in Hong Kong, is exempt from the requirement to hold an Australian financial services licence and is licensed and regulated by the Securities and Futures Commission of Hong Kong under Hong Kong laws which differ from Australian laws. n Australia, for wholesale client use only; n Taiwan R.O.C and the PRC, only available to select targeted institutional investors. This document does not constitute investment advice or an offer to sell, buy or a recommendation for securities, other than pursuant to an agreement in compliance with applicable laws, rules and regulations. Janus Capital Group and its subsidiaries are not responsible for any unlawful distribution of this document to any third parties, in whole or in part, or for information reconstructed from this presentation and do not guarantee that the information supplied is accurate, complete, or timely, or make any warranties with regards to the results obtained from its use. As with all investments, there are inherent risks that each individual should address. The distribution of this document or the information contained in it may be restricted by law and may not be used in any jurisdiction or any circumstances in which its use would be unlawful. Should the intermediary wish to pass on this document or the information contained in it to any third party, it is the responsibility of the intermediary to investigate the extent to which this is permissible under relevant law, and to comply with all such law. Past performance is not a guarantee of future results. There is no assurance that the investment process will consistently lead to successful investing. The opinions are those of the authors are subject to change at any time due to changes in market or economic conditions. The comments should not be construed as a recommendation of individual holdings or market sectors, but as an illustration of broader themes. Statements in the brief that reflect projections or expectations of future financial or economic performance of a strategy, or of markets in general, and statements of any Janus strategies plans and objectives for future operations are forward-looking statements. Actual results or events may differ materially from those projected, estimated, assumed or anticipated in any such forward-looking statement. mportant factors that could result in such differences, in addition to the other factors noted with forward-looking statements, include general economic conditions such as inflation, recession and interest rates Janus Capital Management will act as sub-adviser to Janus Capital nternational. For nstitutional use and wholesale client Only. Not for public viewing or distribution. KB-0412(2)0413 EAPM nst