Forget about the 1%’ers, look at the 10%’ers, says AllianceBernstein. They wrote an interesting blog post a week or so ago about the widening gap between in the rich stocks and the value stocks.
In the aftermath of last year’s market party, the most expensive quintile of global stocks now trades at 8.1 times book value, or almost four times higher than the global stock market as a whole—the second-highest premium since 1971 (Display 1). Earnings multiples are similarly hefty. Meanwhile, the cheapest quintile sells at 0.9 times book value, a 57% discount to the market and around its historical average in both absolute and relative terms. In the past, differences of this magnitude between the valuations of cheap and expensive stocks have heralded outsized value outperformance. Eventually, either the controversies that caused the disparity resolved themselves or cheap stocks simply became too enticing to resist.
Author Chris Marx notes that Consumer Cyclicals & Staples, Medical and Technology sectors are among the highest P/BV valuations and Financials, Housing and Utilities are among the lowest. Neither of those two facts should be all that surprising as those sectors typically exhibit such a split due to accounting treatment of assets and intangibles as well as valuation preferences regarding growth. However, I think his point is the magnitude of the gap.
Using Bloodhound’s AlphaFactor tool, we are able to quickly analyze factors contributing to performance. Bloodhound’s simulation engine queries our proprietary database of U.S. and Canadian stocks (and U.S.-listed ADRs) back to 1987. The “as-reported” and unadjusted dataset enables the recreation of real-world peformance free of survivorship bias and look-ahead bais. As such, we chose to look at the return profile of holding the richest and cheapest stocks. Rather than focus on book value which can be skewed by accounting treatment of tangible and intangible assets, we looked at forward P/E valuations. We compared portfolios of the highest forward P/Es against a portfolio of the lowest forward P/Es, each qualified by being listed on the NYSE or NASDAQ and greater than $1bn in market cap. We set up rules to buy the 100 highest (lowest) stocks, and held them for at least 30 days, and rebalanced the value amongst the portfolio once a month.
Due to the assymetric returns of the equity market, both strategies have positive geometric returns; however, the low P/E strategy has substantially outperformed the high P/E strategy over the the last 28 years. In the last eleven years, the S&P 500 has finished in the black in all but one (2008), and the high P/E strategy underperformed in 50% of them, while the low P/E startegy only underperformed in one. Neither strategy did particularly well during the high correlation period of 2008. Both strategies massively underformed the S&P 500.
The correlation between the two strategies are stong (87% on a yearly basis, and 89% on a monthly basis), their return profiles are widely different. While high P/E is outperforming so far this year, low P/E outperformed for the 16 straight months ended December 2013.
As Bernstein points out, the widening gap in valuation may be tilting the field in favor of value stocks, but the field was already tilted.
January was a dificult month for sure. No effect, that is. February was a little different story. Equities recovered from January’s losses in fine style. The Nasdaq lead the pack among the major indicies, but by the end of the month the S&P 500 had set an all-time closing high. The small caps of the Russell 2000 also had a strong month, leaving the Dow industrials the only one of these four domestic indices down for the year despite its February gains.
The second month of cuts in the Fed’s bond purchases seemed to have little impact on the benchmark 10-year Treasury yield. Meanwhile, gold saw a rebound from its recent losses, gaining almost $100 an ounce and hitting its highest level so far this year before settling back a bit to end at roughly $1,320. The U.S. economy grew a bit more slowly in Q4 2013 than previously thought (2.4%). According to the Bureau of Economic Analysis, that put growth for all of 2013 at 1.9%. Congress agreed to avoid renewed conflict over an increase in the debt ceiling by passing legislation that resolves the issue until March 2015 – and President Walker and Vice President Underwood relaunched their efforts to speak to the American people February 14th.
As defined by Barron’s, “junk bonds continue to defy gravity and mathematics, soaring through the first two months of 2014 with a 2.76% return, even though the market entered the year yielding just 5.6%. That average yield is down to 5.2% now, and the average junk-bond price is up to 104.9 cents on the dollar, with an average spread over Treasuries of 3.81 percentage points, a fresh post-crisis low.”
Manufacturing showed signs of slowing in the United States, where durable goods orders were down for the third of the last four months thanks to a decline in transportation-related orders and the Institute for Supply Management’s gauge fell more than 5%. Meanwhile, Markit/HSBC’s survey of Chinese purchasing managers showed contraction there, though seasonal distortion may have played a role. Housing suffered from frigid weather throughout much of the country. Housing starts, building permits, and sales of existing homes all saw declines, though new-home sales were up slightly for the month and construction spending also rose. Inflation remained well within the Fed’s comfort level. The biggest monthly increase in the cost of electricity since March 2010 pushed up consumer prices by 0.1% for the month, putting the annual rate for the last 12 months at 1.6%. Meanwhile, the Bureau of Labor Statistics said the wholesale inflation rate was up 0.2% in January, but the annual rate was only 1.2% over the last year.
Among industries, we track six of the major categories through the largest 10-, 20- and 50-capitalized stocks. Even though each sector performed well, Healthcare continues to lead the pack. Depending upon the size of the portfolio, the Healthcare sector is up between 7-8% year-to-date.
Only five of the 50 largest capitalized healthcare stocks generated a negative return for February, and two of those would have rounded to flat. Cigna (CI) was the only real negative drain down 8.9%. Conversely, 40th ranked Forest Labs (FRX) and 41st Mylan Labs (MYL) ripped up 45% and 22% respectively. Actavis (ACT) agreed to acquire Forest in a cash-and-stock deal valued at $25 billion. Upon the heels of that merger, investors sought what generic drug maker might be next – and their answer was Mylan.
But as we noted above, the two generics were hardly alone in their February rally. Take Merck (MRK) which has been on the move since November, and really an upward trajectory since Summer 2011.
Among Financials, only Bank of America (BAC) lost ground among the top names, and has been trending that way all year.
Despite strength as a group, only Consumer Cyclicals showed some variability among the largest names. Wal-Mart (WMT; -0.7%), Toyota Motor (TM, -1.0%); Comcast Corp (CMCSA, -6.3%); Honda Motor (HMC, -3.0%); and Starbucks (SBUX, -0.9%) lost value in February.
The S&P 500 is made of 500 stocks – duh. But each stock is not equally important. Perkins Elmer (PKI) has shown a nice steady rise since May 2013 and is a component of the S&P.
If equally weighted, PKI would represent 0.2% of the index. However, the S&P 500 is a capitalization-weighted index – a type of market index whose individual components are weighted according to their market capitalization, so that larger components carry a larger percentage weighting. The value of a capitalization-weighted index can be computed by adding up the collective market capitalizations of its members and dividing it by the number of securities in the index. As such, PKI represents 0.03% of the S&P 500, or roughly 1/100 of the representation of Apple.
The top ten weights of the S&P are the ten largest capitalized companies, and represent almost 18% of the entire index.
Apple (AAPL) 2.8%
Exxon Mobile (XOM) 2.5%
Google (GOOG) 2.1%
Microsoft (MSFT) 1.7%
Johnson & Johnson (JNJ) 1.6%
General Electric (GE) 1.5%
Chevron (CVX) 1.3%
Wells & Fargo (WFC) 1.3%
J.P Morgan (JPM) 1.3%
Proctor & Gamble (PG) 1.3%
Being cap-weighted, the S&P 500 certainly provides a solid representation of the economy as a whole. On one hand, Apple is surely more representative than, say, Jabil Circuit (JBL). However, on another hand, a large proportion of Apple’s market cap is its ungodly cash balance. Apple grew its massive cash hoard to a new record-breaking level last quarter taking the total of cash and marketable securities to $158.8 billion, or roughly a third of its entire market cap. Therefore, is it properly positioned in the S&P 500?
These cap-weighted portfolios automatically rebalance as security prices fluctuate, and thus easy to maintain, only when new companies become large enough to merit inclusion in an index or when others disappear through merger, failure, or relative changes in capitalization. However, cap weighting may lead to suboptimal portfolio return characteristics. Mathematically, cap weighting gives additional weight to rising stocks and reduces weights in stocks that have traded lower. Therefore, by definition, cap-weighted equity index funds have the tendency to overweight overvalued securities and underweight undervalued ones. This mismatch leads to a natural performance drag in cap-weighted and other price-weighted portfolios.
Traditional indices characterized by capitalization weighting are based upon the theories of the Capital Asset Pricing Model (CAPM). This model recognizes that all investors in a given market are exposed to systematic market risk. For investors who had neither the time for alpha generation nor the inclination that active managers can persistently capture excess returns, the cap-weighted market portfolio was the only sensible passive portfolio.
Recent years have seen much of the CAPM criticized or even rejected on both theoretical and empirical grounds, yet a trillion-dollar industry has emerged on investing in or benchmarking to cap-weighted indexes. CAPM says that a “market portfolio” is mean–variance optimal. While CAPM is still taught in business schools as a valuable conceptual tool, the state of the art in return modeling is the multi-factor framework based on the Arbitrage Pricing Model (APT) which reflects the sensitivity of the underlying asset to economic factors.
Capitalization-based, but equal-weight portfofolios have different problems. An equal-weighted portfolio containing the Russell 1000 stocks gives as much weight to the 1000th largest company as to the largest company, and gives no weight whatsoever to the 1001st largest company.
Based on APT, financial theorists now believe that there are numerous sources of equity premia, some risk-based and some behavior-based. These factor-based premiums, which appear to be robust over time and economically significant, can be identified and are associated with value, growth, momentum, volatility, capitalization and other fundamental or technical factors.
A paper on Fundamental Indexation (Arnott, Hsu, Moore, 2005) in the Financial Analyst Journal quantified the difference between a straight up cap-weighted return versus a similar portfolio weighted and built with intuitive fundamental factors (BV, income, revenue, sales, dividends, and employment). They concluded that sourcing particular equity premia yielded both a benefit in retun as well as in some cases volatility and correlation to other asset classes. Such difference was persistent under many market environments.
Although our process is somewhat different, the concepts presented by Arnott, Hsu, and Moore are similar to the methods used by Bloodhound. Our system’s methodology allows a user to seek and identigy specific factors that lead to performance enhancement – whether it be for alpha generation or smarter beta. Our simulation engine allows users to quickly and precisely model equal-weighted portfolios based on customized fundamental and technical factors.
Equal-weighted portfolios eliminate the dominating effect that one stock can claim. Strategies can be built and validated with a finite number of holdings. While not weighted by fundamantal factors, Bloodhound’s simulation engine utilizes ranking methodologies to identify the most appropriate holdings.
Bloodhound’s point-in-time dataset contains fundamental and technical data since 1987 for over 40,000 U.S. and Canadian equities and ADRs. Every data element remains in original, ‘as reported’ form. This adherence to data integrity enables investors to generate historical simulations completely free of survivorship, selection and restatement biases.
Systematic, factor-based indexes have been found to deliver consistent, significant benefits relative to standard cap-weighted indexes. As such, a more efficient system exists than the existing mass-produced, financial products. Bloodhound enables the customization of low cost, liquid, transparent, easy-to-implement strategies that either reflect broad market exposure or identify unique sources of equity premia.
One of the articles we referenced in a blog post last Summer won the CFA Institute’s Graham & Dodd Award that recognizes outstanding articles published last year. “Active Share and Mutual Fund Performance,” by Antti Petajisto (July/August 2013) which analyzed using Active Share and tracking error in active fund management.
The author unequivically notes that active managers are not all equal: They differ in how active they are and what type of active management they practice. These distinctions allow us to distinguish different types of active managers, which turns out to matter a great deal for investment performance. The author sorted all-equity mutual funds into various categories of active management. The most active stock pickers outperformed their benchmark indices even after fees, whereas closet indexers underperformed. During the period of 1990-2009 one group in particular added value for investors: the most active stock pickers, who beat their benchmarks by 1.26% a year after fees and expenses; before fees, their stock picks beat the benchmarks by 2.61%, displaying a nontrivial amount of skill.
Note that “Factor Bets” are not factors in the sense that we speak of them with Bloodhound, but rather timing and tactical asset allocation.
Should a mutual fund investor pay for active fund management? Generally, the answer is no. A number of studies have all concluded that the average actively managed fund loses to a low-cost index fund, net of all fees and expenses. However, active managers are not all equal: They differ in how active they are and what type of active management they practice. These distinctions allow us to distinguish different types of active managers, which turns out to matter a great deal for investment performance.
I divided active managers into several categories on the basis of both Active Share, which measures mostly stock selection, and tracking error, which measures mostly exposure to systematic risk. Active stock pickers take large but diversified positions away from the index. Funds that focus on factor bets generate large volatility with respect to the index even with relatively small active positions. Concentrated funds combine very active stock selection with exposure to systematic risk. Closet indexers do not engage much in any type of active management. A large number of funds in the middle are moderately active without a clearly distinctive style.
Focusing on closet indexing, I started by looking at examples of different types of funds and then examined two famous funds in detail. I also investigated general trends in closet indexing over time and the reasons behind them. I then turned to fund performance, testing the performance of each category of funds through December 2009. I separately explored fund performance in the financial crisis of January 2008–December 2009 to see whether historical patterns held up during this highly unusual period. Finally, I tried to identify when market conditions are generally most favorable to active stock pickers.
I found that closet indexing has been increasing in popularity since 2007, currently accounting for about one-third of all mutual fund assets. Over time, the average level of active management is low when volatility is high, particularly in the cross-section of stocks, and also when recent market returns have been low, which also explains the previous peak in closet indexing in 1999–2002.
The average actively managed fund has had weak performance, losing to its benchmark by –0.41%. The performance of closet indexers is predictably poor. They largely just match their benchmark index returns before fees, and so after fees, they lag behind their benchmarks by approximately the amount of their fees. Funds that focus on factor bets have also lost money for their investors. However, one group has added value for investors: the most active stock pickers, who have beaten their benchmarks by 1.26% a year after fees and expenses. Before fees, their stock picks have even beaten the benchmarks by 2.61%, displaying a nontrivial amount of skill. High Active Share is most strongly related to future returns among small-cap funds, but its predictive power within large-cap funds is also both economically and statistically significant.
The financial crisis hit active funds severely in 2008, leading to broad underperformance in 2008 and a strong recovery in 2009. The general patterns were similar to historical averages. The active stock pickers beat their indices over the crisis period by about 1%, whereas the closet indexers continued to underperform.
Cross-sectional dispersion in stock returns positively predicts benchmark-adjusted returns on the most active stock pickers, suggesting that stock-level dispersion can be used to identify market conditions favorable to stock pickers. Other related measures, such as the average correlation with the market index, do not predict returns equally well.
A prime example in the analysis is the history of Fidelity’s Magellan Fund. To this day, Magellan is still famous for its spectacular record under GARP manager Peter Lynch from 1977 to 1990. In his last 10 years as fund manager, Lynch beat the S&P 500 by a stunning 150% leading the path to becoming the largest mutual fund in the U.S. The fund’s subsequent performance, however, was been mixed. During Robert Stansky’s tenure as fund manager from 1996 to 2005, performance was weak and the formerly active fund was suspected of being a closet indexer. Utilizing the Active Share analysis, Magellan did indeed start out as a very active fund under Peter Lynch, with an Active Share over 90%. Yet its Active Share declined as the fund grew. After Stansky took over in June 1996, however, Active Share plunged more than 30 percentage points to 40% in just two years, and it
then kept going down until stabilizing at 33%–35% for the rest of his tenure. This remarkable shift in the fund’s policy represents a conscious decision to become a closet indexer. Not surprisingly, performance suffered during the closet indexing period. The fund lagged behind the S&P 500 by about 1% a year for 10 years.
One of the continuous knocks against active management is consistent underperformance. However, true active management needs to be distinguished from closet index funds that claim active management for a reasonable comparison.
Sunday may be the Academy Awards, or should I say Oscars, but today Institutional Investor’s Alpha magazine released its annual survey on hedge funds where investors were asked to evaluate firms on a variety of factors, including alpha generation, alignment of interests, infrastructure and liquidity terms.
Silver Point Capital, managed by Edward Mule, whose max draw down was 35.5% between June 2007 and December 2008 took the top ranking. The fund was up 16% in 2013 and 20% in 2012. Started on January 1, 2002, the now $5bn distressed shop has an annual return of 13.11% and a Sharpe Ratio of 1.6x. Silver Point took the award for Best Alpha Generation, Best Alignment of Interests, and Best Transparency.
Activist Dan Loeb’s Third Point took second place with awards for Best Liquidity of Terms and Best Independent Oversight, followed by:
3. Adage Capital
4. Elliott Management
6. Perry Capital
7. Davidson Kempner
Baupost, Seth Klarman’s Boston-based firm that we have written on before in the Bloodhound Exchange, won the award for Best Risk Management.
Always the muckracker, I also like to see who fared the worst. All four of the following firms received a grade of “F.”
BlueCrest Capital (out of the UK)
Bain Capital (sorry Mitt)