December 3, 2014

Diversification or Deworsification?

Most of us learned long ago that diversification is a good thing. In fact, it is often called the closest thing to a “free lunch” in the world of investing. This is because, when used wisely, diversification can reduce portfolio volatility with little or no diminution in return. But the key is “when used wisely.”

Working with individual stocks, diversification reduces company-specific (idiosyncratic) risk that comes from earnings surprises or other bad news that can affect individual companies. A carefully selected (no strong biases) portfolio of 50 or more stocks will diversify away most idiosyncratic risk. The main benefits from more diversification are reduced benchmark tracking error and an increased ability by active managers to handle larger amounts of capital.

Investment managers often want to reduce tracking error for reasons of job security and to trade larger amounts of capital to receive more compensation. But from an investor’s point of view, larger portfolios are no better than smaller ones once you eliminate most idiosyncratic risk. Larger portfolios may, in fact, be worse than smaller ones in offering up profit opportunities. Active managers might better serve their clients’ interests by having more focused portfolios of their best holdings rather than diluting their portfolios with less attractive issues. Investors wanting broader-based portfolios can buy less costly index funds.

Over diversification is a problem especially for momentum investors because studies show that momentum profits are highest in the most concentrated momentum ranked cross-sections of the market. Top momentum deciles outperform top momentum quintiles, which outperform top momentum terciles. There is a significant reduction in expected return from a momentum portfolio as the number of stocks in the portfolio increases. Yet, as I point out in my blog post “Individual Stock Momentum – that Dog Won’t Hunt”, there are some momentum funds that own half their benchmark universe of individual stocks. Investors in those funds are paying for what amounts to an index fund plus a modest exposure to momentum.

Over diversification can also be a problem with asset class momentum. To better understand this, you need to consider how investors earn their profits. Investors are compensated for giving up use of their capital (which can earn them the risk-free rate) and for bearing risk (which can earn them a risk premium above the risk-free rate).

Companies receive and use invested capital for productive purposes when equity investors become beneficial owners of these companies. Stockholders share in the fortunes or misfortunes of such companies and are compensated with a high risk premium. In fact, among all investment opportunities, stocks (especially U.S. stocks) have offered the highest risk premium. Those who have trouble accepting the evidence that I present in my book, Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk, should read Stocks for the Long Run by Jeremy Siegel. He devotes most of his book to that subject.
Source: Jeremy Siegel, Stocks for the Long Run: The Definitive Guide to Financial Market Returns & Long Term Investment Strategies, McGraw-Hill Education, 2014

Bonds also provide a risk premium, but one that is substantially lower than stocks because bond investors have a senior claim on company assets and are guaranteed a return of capital when their bonds mature. It is uncertain what kind of risk premium, if any, investors in assets other than stocks and bonds receive. For example, investors in aggregate commodity futures (a zero-sum game, less transaction costs) once received plenty of risk premium from commercial interests looking to hedge their business risks by using those markets. But with the proliferation of speculative commodity trading, as well as a large number of institutions adding passive commodities to their portfolios, much of that risk premium has now vanished. One might sometimes still earn short-term speculative trend following profits from assets other than stocks and bonds, but the odds are much better having a proven risk premium behind you as a tailwind.

The main reasons investors continue to hold assets other than stocks and bonds is the mistaken belief that more is always better with diversification and that holding less correlated assets will lessen portfolio volatility and reduce bear market exposure.

But bonds can be just as risky as stocks, and stocks and bonds are not always negatively correlated. Here is a chart of the correlations over time between U.S. stocks and 10-year Treasury notes.

Many markets that were once non-correlated now move together under economic stress. Diversification can fall short when it is needed the most. With increased globalization, the world is now much more inter connected. Widespread diversification is no longer as useful as it once was in reducing downside risk exposure. What is useful for that purpose is trend-following momentum, which has shown the ability to both enhance returns and reduce downside exposure among different assets going back to the turn of the last century.[1] The effective downside protection offered by absolute momentum is even more reason why over diversification is unnecessary for momentum investors.

A better approach, as presented in my book, is to invest in the stock market when it is strong to capture the highest amount of risk premium.[2]  When stocks are weak, you switch to bonds which offer a more modest risk premium than stocks. Since the stock market is a leading economic indicator, a weak stock market can indicate a future economic slowdown, declining interest rates, and a healthy bond market. So stocks and bonds should complement each other at the most appropriate times. This is a more effective approach than having a permanent allocation to both. Diversification into asset classes with lower risk premiums dilutes long-run returns and leads to investment mediocrity.

[1] See "A Century of Evidence on Trend Following Investing" by Hurst, Ooi, and Pedersen.
[2] Our Global Equities Momentum (GEM) model stays mostly in U.S. stocks. It switches to non-U.S. stocks when the odds shift in their favor according to relative strength momentum. For validation of this switching approach, see

November 10, 2014

Individual Stock Momentum - That Dog Won't Hunt?

Dead or dying academic ideas latched on to by unwary institutional investors litter the investment graveyard landscape. My new book, Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk, describes some of these, such as portfolio insurance, small cap premium, and portfolio diversification with passive commodities. Most of these occurred because of incomplete information or omitted variables. 

Shortly after Rolf Banz published a paper based on his University of Chicago PhD dissertation that identified a small cap premium from 1936 through 1975, Dimensional Fund Advisors (DFA) and others quickly tried to exploit this "small cap anomaly." They learned later that this apparent anomaly was perhaps driven by a mistake in how researchers treated missing data for delisted stocks, many of which were small caps. As noted in my book, studies since then have shown that the small cap premium by itself may no longer offer above market risk-adjusted returns.

Portfolio insurance was based on the elegant idea that you could synthesize protective options through a combination of stock buying and selling combined with short-term borrowing and lending. You could "insure" your portfolio using these synthetic options. However, what portfolio insurers failed to take into account was the short- term mean reverting nature of stocks. At that time, most academics thought stocks followed a random walk and had little or no auto-correlation, even though for many years stock exchange specialists made handsome mean reversion profits by trading against public order flow. Portfolio insurers quietly packed up their bags and disappeared not long after they first appeared.

In the mid-2000s, academics published studies showing that passive commodities could be a decent portfolio diversifier. Investors jumping on to that bandwagon failed to realize that very large inflows of speculative capital from institutional investors could eradicate the premium that had previously been flowing from hedgers to speculators. Front-running costs of over 3% annually from simultaneous rolling over index future contracts would also take its toll on speculative investor profits. After half a dozen years, there were new research papers showing that adding commodities to a stock/bond portfolio was no longer beneficial.

It may very well be that relative strength momentum applied to individual stocks is the latest academic concept that may not work so well in practice because of investors blindly following academics without seeing the bigger picture. Many academic studies of momentum ignore transaction costs, which can be significant.

A trading costs study by Frazzini et al. (2012) of AQR, covering 13 years of data in 19 developed markets, stated that "...the main anomalies to standard asset pricing models [including momentum] are robust, implementable, and sizeable." The authors were looking at long/short momentum using proprietary data over a relatively short period using broad range of  large cap momentum stocks. Lesmond et al.(2004) found that momentum stocks were relatively less liquid than other stocks. They concluded that "… the magnitude of the abnormal returns associated with these trading strategies [stock momentum] creates an illusion of profit opportunity when, in fact, none exists." Korajczyk and Sadka (2004) found the profit breakeven allocation  to be only $2-5 billion for long-only strategies using the top 10% of momentum stocks.

Transaction costs, however, are only half the story. Academic researchers validate cross-sectional relative strength momentum by looking at winners versus losers and segmenting the stock market into deciles, quintiles, quartiles, or terciles. According to Siganos (2007), beyond the first few extreme winners and losers, there is a continuous decline of momentum gains from larger momentum portfolios. Siganos found maximum momentum returns using a portfolio limited to the 40 top and bottom performing stocks.

Yet all publicly available stock momentum funds so far use more than 40 stocks, and some use ten times more than that! The iShares MSCI USA Momentum Factor ETF holds 125 stocks, and the PowerShares DWA Momentum Portfolio has 100 holdings. The AQR Momentum Fund, AQR Small Cap Momentum Fund, and AQR International Momentum Fund hold 479, 953, and 440 stocks, respectively. These represent nearly 50% of the underlying indexes from which these momentum funds draw their holdings. 

The annual expense ratios of AQR's momentum funds range from 0.50 to 0.90%. In contrast to this, you can invest in the Russell 1000 broad-based index from which AQR draws their momentum holdings for a cost of only 0.11% per year. When you add in another 0.7% per year that AQR estimates as transaction costs for their quarterly rebalanced large/midcap momentum index, it raises serious questions about how investors can really capture momentum profits from individual stocks.

Yet, as they say, the proof is in the pudding. In Chapter 6 of my new book, I show readers a simple way to find style-based alternatives to so-called smart beta funds. In many cases, style-based ETFs with lower expense ratios and lower transaction costs offer similar or better performance than their "smart beta" counterparts.  I thought it would be interesting to use the same technique to look at lower cost style-based alternatives to the largest and longest existing publicly available funds that use momentum applied to individual stocks. 

The following comparative charts using momentum funds that can be accurately matched up with stylistic index funds begin at each momentum fund's inception date using the lowest cost class of momentum shares.

The stylistic equivalent fund to the AQR Momentum Fund (AMOMX), with an annual expense ratio of 0.40%, is the Vanguard U.S. Large Cap Growth ETF (VUG), with an expense ratio of .09%.


The stylistic equivalent fund to the PowerShares DWA Momentum Portfolio (PDP), with an annual expense ratio of 0.65%, is the Vanguard U.S. Mid Cap Growth ETF (VOT), with an expense ratio of .09%

The stylistic equivalent fund to the AQR Small Cap Momentum Fund (ASMOX), with an annual expense ratio of 0.60%, is the Vanguard U.S. Small Cap Growth ETF (VBK), with an expense ratio of .09%.

The stylistic equivalent fund to the AQR International Momentum Fund (AIMOX), with an annual expense ratio of 0.55%, is the iShares MSCI EAFE Growth ETF (EFG), with an expense ratio of .40%.

The above charts give some evidence of why I am not a fan of using momentum with individual stocks. It should also be mentioned that relative strength stock momentum does little or nothing to reduce portfolio drawdown. To accomplish that, you would need something like absolute momentum or another trend-following filter (cross asset diversification may also help somewhat). In terms of both risk and return, momentum is more effective when it is used with asset classes or broad indexes having lower transaction costs, and when it incorporates trend-following absolute momentum, as described in my book.

October 31, 2014

Dual Momentum Investing Is Now Released

Today is the official release day of  Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk. I wrote the book to help as many people as possible earn attractive returns and minimize bear market drawdowns.
book by Gary Antonacci

 Here is an excerpt from a thoughtful review by Reading the Markets:

Antonacci's extensive research and his clear-headed thinking have led to a book that every investor should read. The academically oriented reader will be grateful for his occasional excursions into the weeds, his carefully laid-out data, and his lengthy bibliography. The practically oriented investor will find a road map for moving ahead and staying out of really big trouble ...This one's a keeper!

Scott's Investments wrote:

Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk is a must-read for individual investors and financial professionals…Antonacci has done the heavy lifting for his readers by thoroughly researching the history and data behind momentum investing. The result is a well-researched and overwhelming argument for momentum investing. Readers are rewarded with a simple, robust strategy that anyone can implement.

And from Alpha Architect...:

Antonacci demonstrates returns to Dual Momentum and the empirical evidence through extensive backtesting across multiple decades; the analysis includes various risk metrics (returns, standard deviations, Sharpe, drawdowns, etc.), and robustness studies, the interpretation of which he explains in detail, so that the reader can have an informed view of the data.

The evidence culminates in a simple but powerful applied momentum model: Antonacci’s Global Equities Momentum (GEM) strategy, which uses these dual momentum ideas to tactically allocate across and among domestic and international equity and bonds. And the results are nothing short of spectacular: superior returns, with low volatility.

The Kindle version of the book was released on October 9, so there are also a dozen reader reviews on Amazon, as well as endorsements from prominent industry professionals. Click here to read these, find out more about the book, or to order it now.

October 25, 2014

Value Investing Redux

Believe it or not, I was once a value investor. From an early age I was impressed with the long-run success of such value luminaries as John Neff, Bill Ruane, Walter Schloss, and Max Heine. Being of a contrary temperament, I also liked the idea of buying stocks that were out-of-favor and ignored by the multitudes. Supporting this approach, DeBondt and Thaler (1985) presented evidence that stocks overreact to bad news, and that poor performers over the past 3 to 5 years tend to outperform when moving forward. Value investing seemed to have representative bias going for it, whereby investors overreact to poor performance and project it far into the future while failing to account for long-run mean reversion.

In 1992, Fama and French presented their groundbreaking paper indicating that value and small cap stocks offered up a risk premium to investors. I, along with the rest of the civilized world, readily accepted this idea.

What got me away from value wasn't anything about value itself, but rather was my discovery, after reading a plethora of research papers on momentum, of just how strong momentum is compared with anything else. Looking at relative strength momentum applied to individual stocks showed a higher premium.

What also got me excited about momentum though was my own research showing that momentum can be applied not just across different investment opportunities, but to single investments themselves in the form of trend-following absolute momentum.  Absolute momentum not only enhances returns like relative momentum, but it can also greatly reduce drawdown. As far as I could tell, no other anomaly could do this.

Reinforcing my view were the surprising results about value by Israel and Moskowitz (See my post "Momentum…the Practical Anomaly?"). Working with U.S. equities data back to 1926, the authors found that:

The value premium… is largely concentrated only among small stocks and is insignificant among the largest two quintiles of stocks (largest 40% of NYSE stocks). Our smallest size groupings of stocks contain mostly micro-cap stocks that may be difficult to trade and implement in a real-world portfolio.

Earlier research by Loghran and Hough (2006) also found the value premium to be illusionary.
Research by Das and Rao (2012) showed that large cap value only works in January.  On another front, Chen et al. (2011) proposed an asset pricing model in which investment and profitability are the main explanatory variables, rather than value and size. Fama and French (2014) then expanded their established three-factor model to include investment (expected future changes in book equity) and profitability (expected future net income relative to book assets). When doing so, they concluded that value was redundant.


So did all this put a nail in the coffin of value investing? Not necessarily. Israel and Moskowitz had looked at value using the popular book-to-market ratio. They found similar results using other value measures, such as dividend yield and long-term reversals that had data going back to at least the 1930s. However, these were only singular measures of value.

Dhatt et al. (2001) found that composite measures of value were superior to any individual metric. Moreover, in their book Quantitative Value: A Practitioner's Guide to Automating Intelligent Investment and Eliminating Behavioral Errors, Gray and Carlisle identified a valuation metric based on enterprise multiple, defined as total enterprise value (TEV), divided by earnings before interest, taxes, depreciation, and amortization (EBITDA). Here is a table of valuation metric comparisons by Gray using equal weight portfolios sorted into quintiles. Data is from 1971 through 2010 and excludes micro caps. But Loughran & Wellman (2010) found only a difference of only .02% per month between book-to-market (HML) and the enterprise multiple (EM).

comparative valuation metrics


Some researchers have found that adding a financial strength or quality metric may improve the risk-adjusted results of value portfolios. Paying attention to these extra factors can help overcome the problem of the "value trap," which happens when stocks remain depressed (and may go bankrupt) because their poor fundamentals warrant it. We should keep in mind though that there is always the potential of overfitting and selection bias when combining factors.

Using global data from 1988 through 2012 and U.S. data from 1963 through 2012, Kozlov and Petajisto (2013) found that going long stocks with high quality earnings (based on high return on equity, high cash flow, and low leverage) and short stocks with low quality earnings gave higher Sharpe ratios than a similar value only strategy. They also found earnings quality to be negatively correlated with value. The best Sharpe ratio came from combining high quality with value, since there were significant diversification benefits.

Using a different approach, Piotroski and So (2012) came up with a multi-factor scoring method (F-Score) to measure a firm's financial strength. This method was positively correlated with profitability and earnings growth. Piotroski and So found that strategies formed jointly on F-Score and value outperformed traditional value only strategies.

Novy-Marx (2012) found he could simplify the quality factor to just gross profitability, defined as revenues minus cost of goods sold, scaled by assets. Novy-Marx (2013) found on U.S. stock data from 1963 through 2012 that profitable firms generated significantly higher returns (0.31% per month) than unprofitable firms, despite having higher price-to-book ratios. Novy-Marx (2012) also found that joint strategies combining value with Piotroski/So's F-Score, Greenblatt's magic formula, or gross profitability outperformed traditional value, with profitability plus value being the strongest combination.

growth of value plus quality


Despite the above potential enhancements to value investing, given the enormous advantages of dual momentum investing, if I were ever to get the urge to do value investing, I would just lie down until the urge went away. But if I were actually going to add a value component to my portfolio, it would need to be calculated on a more robust basis than price-to-book, and it would need to be combined with profitability/quality to mitigate the potential value trap problem. Here is what I might look for in a value fund if I were going to invest in one:

1)    It should combine value with other criteria, such as quality or profitability screens.
2)    It should determine value based on several value metrics and/or a value metric that incorporates the enterprise multiple.
3)    It should not dilute returns by having too broad a portfolio. In Chapter 6 of my new book, Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk, I show that many so-called smart beta funds are like closet index funds with modest stylistic tilts. Many value funds are the same. Even though the value effect is more pronounced in the top 10-20% of value rated stocks, most funds dilute this effect by using the top third or top half of value stocks instead of the more profitable top 10-20%.
5)    The expense ratio should be reasonable. (This is true of all investments.)
6)    For taxable accounts, ETFs are preferred over mutual funds and hedge funds. Capital gains occur only when you sell your ETF holdings, whereas mutual funds have yearly taxable distributions of capital gains.


If value and quality actually do well together, it is natural to think of actively managed mutual funds when considering investment candidates based on such factors. This is because most active managers use fundamental analysis (which takes into account profitability in the form of financial strength), managerial acumen, competitiveness, quality of earnings, and other judgmental factors besides valuation.

But most active managers lack transparency. This gives them the appearance of possessing proprietary knowledge that may be worth paying a premium to access, but it also means it is often difficult to tell if they meet the first two criteria listed above. Active mangers also charge high fees. The Morningstar average large cap value fund annual expense ratio is 1.16%. Mutual funds also have a drag on their performance because of their reserves that are held for redemption. Also, as mentioned above, these funds are generally not tax efficient.

But there is one mutual fund that may be worth looking at. It is the AQR Core Equity Fund (QCELX).[1]  This fund meets our first two requirements above, since it uses three indicators of profitability along with five indicators of value. For good measure, it adds three momentum indicators. QCELX has an annual expense ratio of 58 basis points. It is not tax efficient, but AQR has plans for a more tax efficient version of the fund. QCELX currently holds 413 stocks, which is more than 40% of the stocks in its 1000 mid/large cap stock selection universe.[1] Investors are paying a high fee for the index part of this fund.

In the ETF realm, until this past week there was only one fund that qualified using the above criteria. It was the PowerShares Dynamic Large Cap Value ETF (PWV) based on the Intellidex methodology. PWV uses a ranking selection method with four indicators of value and four indicators of growth. They subtract their value from growth rankings then select the largest negative scores. This process gives PWV some profitability exposure, according to the research of Bridgeway Capital Management. Bridgeway found that a multi-indicator value approach (adding price/cash flow, price/earnings, and price/sales) provides greater exposure to gross profitability than a portfolio based only on price/book.

To help further with profitability exposure, after doing their basic screens PWV adds weightings for price momentum, earnings momentum, quality, and management action. PWV selects the top 20% of the largest 250 stocks from a potential universe of 2000. This gives them a focused portfolio of  50 stocks that PWV rebalances quarterly. (I would prefer that they derived their portfolio of 50 stocks by selecting the top 10% of the largest 500 stocks instead of the top 20% of the 250 largest stocks.) They allocate half their capital to the top 15 ranked stocks, and the other half of their capital is divided among the remaining 35 stocks. Their annual expense ratio is 58 basis points. PWV's performance since inception has been attractive compared to the iShares S&P 500 Value ETF.

Value plus quality versus value
  Past performance is no assurance of future success.

This week another ETF got on to the short list of qualified candidates. The new kid on the block is the Value Shares U.S. Quantitative Value ETF (QVAL) offered by Alpha Architect. QVAL incorporates quality in several ways. First, it has forensic accounting screens to avoid firms at risk for financial distress or manipulation. Then it filters for financial strength using a modified Piotroski/So F-Score. Finally, QVAL checks for sound business fundamentals through what it calls an "economic moat." This is a screen for firms having sustainable competitive advantages, a la Warren Buffett.

QVAL's management uses the enterprise multiple to determine value. They select  the top 10% of their large cap universe based on value, then drop the bottom half of these based on quality. The remaining 50 stock portfolio is equal weighted and rebalanced quarterly. QVAL's annual expense ratio of 79 basis points is high compared to PVW and QCELX, but QVAL is the most focused fund among the three, selecting only the top 5% of quality/value stocks in their mid/large cap universe, compared to PVW's 20% and QCELX's 40%. In looking at any of these enhanced value programs, you may want to consider an alternative such as the iShares MSCI USA Value Factor ETF (VLUE) that selects 20% of its benchmark universe and has an annual expense ratio of only 15 basis points. The VLUE simple strategy may also be less overfit to past data.


Value investing in general poses risks that all investors should be aware of. One of these is high tracking error relative to the market. Value can go through sustained periods of under performance, such as during the 1990s. From 1994 through 1999, value underperformed growth by over 10% per year! With focused portfolios of just 50 stocks, PVW and QVAL can have much higher tracking error than other value funds. Value investors need to have a long-term investment horizon and a high tolerance for prolonged periods of underperformance.

There is also the potential problem of overfitting the data which comes from data mining. The use of filters and multiple selection criteria by all the above mentioned funds increases their chance of disappointing ex-post returns going forward.

Value investing with focused portfolios, such as those of PWV and QVAL, is also subject to high volatility and high drawdowns. Investors should be prepared for these as well.[2] Unfortunately, investors often lose sight of this. They panic and act counter to their best interests when confronted with severe drawdowns once bear markets arrive. In a survey of its members since 1988, AAII found that the highest weight to cash and the lowest weight to equities was in March 2009, right at the bottom of the worst bear market since the 1930s.

There is a way, though, that one may mitigate some of this harmful behavior. It is by using absolute momentum. (You didn't think I would write a long post like this without mentioning momentum, did you?) My research paper on absolute momentum and my new book show how to use trend following absolute momentum to reduce the expected drawdown of most any investment opportunity. The chart below of absolute momentum applied to the S&P 500 illustrates this.One could also incorporate value into a dual momentum-based model, such as the one featured in my new book.
Downside reduction through absolute momentum
Unfortunately, value does not respond as well as relative momentum portfolios or the market itself to trend following filters. But by adding an absolute momentum filter to value-based holdings we should get some drawdown protection.

[1] AQR has this same kind of broad participation in their momentum fund, where they hold 47% of the 1000 stocks in their selectable universe. Why do they not target factor profits more? Perhaps they expect liquidity issues. This may also explain why their portfolios are, for the most part, capitalization weighted instead of value or equal weighted.

[2] We found in examining the Ken French data of value, operating profits, and the joint sort of value and operating profits from 1964 through 2014, that the combination of value and operating profitability had a higher volatility and worse drawdown than value or operating profits themselves.

This material is for informational purposes only. It is not a recommendation to buy or an endorsement of any securities. Investments are subject to risk including loss of principal. You should do your own research before investing. Please see our Disclaimer page for additional information.

October 9, 2014

Giving Investors a Chance

Researchers estimate that the worldwide cost of investment management is approximately $3 trillion per year. Some of this expense is unavoidable, such as the costs associated with custodial fees and for the periodic re-balancing of portfolios.

However, most of this high expense is in the form of compensation paid to the managers of actively managed mutual funds, hedge funds, and other managed programs. What do investors have to show for this large transfer of wealth from themselves to active money managers? The answer, unfortunately, is "not much." My book, Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk, reveals an abundance of research that confirms the general lack of value-added from active investment management.

Lack of performance is due not just from the higher fees and transaction costs associated with active investment management. Institutional investors use disadvantageous periods for the evaluation and selection of their investments. Goyal and Wahal (2008) show that investment managers and their consultants tend to select investments based on performance over the prior three or so years. Yet momentum research papers show 3 to 5 years to be a relatively poor period for performance evaluation. Equity performance tends to be mean reverting over that time frame. A one-year relative evaluation period gives better results. Vanguard issued a research note last July also documenting poor future performance based on a past 3- year evaluation period.

Making matters worse, yearly reports from Dalbar Inc., a market research firm, indicate poor timing decisions on the part of the investing public. For example, the average US equity investor achieved an annualized return of 5.0% over the past 20 years ending in 2013, which is 4.2% less than the 9.2% average annualized return of the S&P 500.

Investors are emotionally influenced by market volatility when getting into and out of the markets. Equities have provided investors with the highest risk premium, but they also have been subject to high volatility and extreme drawdown, since few investors have been aware of the risk-reducing benefit of absolute momentum. When investors instead try to dampen volatility through diversification with fixed income or alternative investments with lower inherent risk premiums, they also dampen their long-run expected return.

Some try to boost returns by looking for more of an edge from their equity investments. Historically, investors have used value and small cap portfolio tilts in their attempts to achieve higher risk-adjusted returns.

Yet the latest research shows that small size and high value portfolios may not always provide the higher risk-adjusted returns that investors have been seeking (See our post "Momentum...the  Practical Anomaly?") Momentum, on the other hand, does provide a proven edge, especially when dealing with indexes rather than individual stocks and when using both absolute and relative momentum together (dual momentum). Unfortunately though, the most popular momentum-based programs use only relative strength momentum, and they apply it to individual stocks, which necessitate higher transaction costs.

I know of only one public program that uses dual momentum applied to asset classes. However, their fees are high, and their portfolio choices leave much to be desired (My book also goes into considerable detail about asset selection, especially with respect to momentum-based portfolios.)

In the future, when the advantages of dual momentum become better known, there may be other dual momentum investment opportunities. However, they may still have fees that are too high, portfolios that are less than ideal, or models based on too little data. The biggest mistakes I see others make are using models that overfit the data and drawing conclusions based on limited amounts (typically around 15 years) of data. Ex post returns are not the same as ex ante returns.

To give investors a better chance to earn decent risk-adjusted returns, my new book fully discloses my simple Global Equity Momentum (GEM) model (see the Performance page of my website) and shows how to easily use it. GEM has performed well over 40 years of past data under different market conditions using the same approach validated in numerous academic research papers. It has also avoided most bear market equity erosion. Any investor can easily utilize GEM to benefit from dual momentum while using a sensible, minimal expense portfolio.

September 11, 2014

Value and Momentum Revisited

Most academic research on momentum deals with individual stocks. Most applications of momentum are also oriented toward individual stocks. The three largest momentum programs (AQR momentum mutual funds, PowerShares DWA Momentum ETFs, and iShares MSCI USA Momentum Factor ETF) all use individual stock momentum. The only public program using momentum applied to asset classes was the ALPS Goldman Sachs Momentum Builder, which recently went out of business due to lack of interest.

Yet momentum applied to individual stocks is not the ideal way to use momentum. High transaction costs can negate much of the benefit of momentum investing, and most stock momentum programs dilute the momentum effect by selecting hundreds of stocks instead of just the ones showing highest relative strength. Momentum applied to indexes or sectors, rather than individual stocks, can capture high momentum profits with much lower transaction costs.

Here is a table from my new book Dual Momentum Investing: An Innovative Approach to Higher Returns with Lower Risk. This table shows the performance of the AQR Momentum Index composed of the top one-third of the 1000 highest capitalization U.S. stocks based on 12-month relative strength momentum with a one-month lag. AQR weights their index positions based on market capitalization and adjusts the positions quarterly. For comparison, we show the performance of the Russell 1000 index and from applying absolute momentum to the Russell 1000 by moving into aggregate bonds whenever 12-month absolute momentum is negative.

Table 9.2 AQR Momentum, Russell 1000, and Russell 1000 w/Absolute Momentum 1980-2013

AQR Momentum Index [1]
Russell 1000 Index
Russell 1000 w/Abs Momentum
Annual Return
Annual Std Dev
Annual Sharpe
Max Drawdown

These figures do not account for the 0.7% per year in transaction costs for the AQR Momentum Index, would have put it at a disadvantage to even the Russell 1000 index on a risk-adjusted basis.[2]

Table 9.3 shows the AQR Momentum Index, the Russell 1000 Value Index, and a 50/50 combination of value and momentum, which was advocated in the Asness et al. (2013) paper "Value and Momentum Everywhere." This combination is supposed to be desirable due to the negative correlation between value and momentum. But the Asness et al study used long/short momentum and long/short value. Hardly anyone actually invests that way. Long only momentum and value are highly correlated. . 

We see that value combined with momentum (rebalanced monthly) does give a higher Sharpe ratio than either value or momentum alone. But there is little or no advantage with worst drawdown, and the results still pale in comparison to simple absolute momentum used with the Russell 1000 Index [3].

Table 9.3 AQR Momentum, Russell 1000 Value, 50/50 AQR Momentum with Value 1980-2013

AQR Mom Index
Russell 1000 Value Index
50/50 AQR Mom with Value
Russell 1000 w/Abs Mom
 Annual Return
 Annual Std Dev
 Annual Sharpe
 Max Drawdown
As a further check on the possible worthiness of combining value with momentum, I used the Global Equity Momentum (GEM) model described and tracked on the Performance page of our website. Full disclosure of GEM and instructions on how to use it are in my book. Using relative momentum, GEM switches between the S&P 500 and the MSCI EAFE when absolute stock momentum is positive. When absolute momentum turns negative, GEM moves into aggregate bonds. 

The table below shows GEM results from January 1974 through August 2014, as well as the results from adding the MSCI USA Value (large and mid-cap) index to GEM as a switching option and rebalanced monthly. We see that the inclusion of value into the momentum model adds nothing to the performance of GEM.

GEM GEM w/Value
Annual Return 17.43 17.24
Annual Std Dev 12.64 12.52
Annual Sharpe 0.86 0.86
Max Drawdown -22.72 -22.94

The above are hypothetical results, are NOT an indicator of future results, and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees. One cannot invest directly in an index. Please see our Disclaimer page for more information.

Furthermore, as I pointed out in a blog post last year called "Momentum…the Practical Anomaly?", Israel and Moskowitz of AQR show in their 2013 paper that value based on book-to-price only offers a long-term premium when applied to very small stocks, such as microcaps. These are unusable by larger investors. How one can mix individual stock momentum (which may offer nothing special after transaction costs) with value (which may also not be all that it was once thought to be) and create something extraordinary seems challenging. This is especially in true in light of an earlier paper by Daniel and Titman (1999) showing that value strategies are strongest among low momentum rather than high momentum stocks, and momentum strategies are strongest among growth rather than value stocks.

Even so, researchers are nothing if not persistent and imaginative. When they found that Markowitz mean variance optimization (MVO) gave inconsistent results, researchers tried constraining the inputs, incorporating prior information to shrink the estimates, and even ignoring returns altogether to try to create portfolios that were more robust. In the end, they found that because of estimation error, equal weight portfolios were generally superior to MVO portfolios. The same overreach is true with the Capital Asset Pricing Model (CAPM). This started out as a single factor model that expanded to 3 and then 4 factors. Factor fishing has now come up with more than 80 possible data-mined factors, yet the factor pricing model may still not model the real world well. 

So it didn't surprise me to see recent a paper by Fisher, Shaw, and Titman (2014) called "Combining Value and Momentum" that tries hard to find other ways to use value and momentum together. (Yes, this is the same Titman who co-authored the paper that showed momentum working better with growth rather than value stocks and who co-authored the seminal momentum papers of the 1990s with Jegadeesh.)

What is perhaps most interesting are the various findings the authors came up in the course of their research. As the saying goes, the devil is in the details. Here are some of those details.

The authors separate stocks into 2 size categories, large cap corresponding to the Russell 1000 index, and small cap corresponding to all other stocks in the CRSP database from 1975 through 2013. They base momentum on prior 12-month performance skipping the last month. About value and momentum separately, the authors find:
1) Value, as measured by the price-to-book ratio, is beneficial only with small stocks and not with large stocks. This is the same conclusion reached by Israel and Moskowitz who used data back to 1926, and who also found it to be true of other valuation measures that had data back to at least the 1930's.

2) Despite high momentum portfolio Sharpe ratios before transaction costs, the high transaction costs associated with momentum portfolios negates much of the difference in Sharpe ratios between large momentum and large value portfolios. 

 3) Since small stocks have even higher transaction costs than large stocks, the authors incorporated higher transaction costs to conclude that none of the small momentum portfolio Sharpe ratios are higher than the Sharpe ratios of the small market portfolios.

In other words, based on high transaction costs, individual stock momentum may not be good with either small or large stocks [4]. So all we are left with that provides above market risk-adjusted returns are small value stocks that most investors (and particularly institutional ones) will find too expensive and difficult to trade. 
The authors then look for ways to salvage momentum by combining it with value in two different ways. The first is to rank firms by momentum and value, and then to compute an average rank. One signal can outweigh the other this way, and momentum still has high transaction costs with this approach. 

The authors' second approach is to use momentum as a filter for value-based portfolios. They buy stocks only when value and momentum are both favorable, and they sell stocks only when both factors are unfavorable. Momentum does not trigger any trades, but instead influences the portfolios by delaying or avoiding trades. Data mining for the highest ex-post Sharpe ratios with this second approach, the authors find much greater exposure to the value factor. The optimal small cap portfolios, for example, have value allocations of 79% or more. The role of momentum with this approach is very small. 

The authors' first approach gives higher Sharpe ratios when trading costs are low, and the second approach gives higher Sharpe ratios when trading costs are high. Of course, we do not know if these Sharpe ratios will continue out-of-sample into the future.

We can avoid the issues of high trading costs and less certain Sharpe ratios if we instead use momentum with indexes rather than with individual stocks. In our 2012 post called "Value and Momentum…Not Here" we asked if there should be just value and momentum everywhere. I didn't think so then, and I see even less reason to believe so now.

[1] http:///
[2] The AQR mutual fund using this index (AMOMX) has an annual expense ratio of 0.40%, while the Russell 1000 ETF (IWB) has an expense ratio of 0.15%.
[3] Results of momentum combined with value are better if a quality factor, such as profitability, is added, and if momentum, value, and profitability are applied to the same portfolio.See "Quality Investing" by Novy-Marx.
[4] A study last year by Frazzini, Israel, and Moskowitz looked at large institutional trades across 19 developed markets from 1998-2013. They found the trading costs of momentum to be low, despite a higher turnover than from other factors. A study by Lesmond, Schill, and Zhou (2004) called "The Illusionary Nature of Momentum Profits" showed that transaction costs reduced momentum strategy returns to close to zero. Fisher et al. uses transaction cost estimates that are between these two.