Momentum investing has long been a thorn in the side of conventional market theories. That doesn't dim its power as a strategic investment tool, but it can still be an awkward beast.

The momentum approach-investing in stocks that have had high returns in the recent past and dropping those with sluggish returns in the same period to win outsize returns-is difficult to unify with the capital asset pricing model or the efficient market hypothesis. For the moment, it's the financial equivalent of trying to put a round peg in a square hole. But some financial economists are trying anyway, and they may even succeed as they continue to peel away the mystery of asset pricing in the years ahead. But no one will find the process elegant.

In the meantime, there's a list of theories arising to explain why momentum investing works, ranging from traditional risk-premium-based explanations to behavioral economics that study irrational investors.

The concept of momentum investing is compelling not just because investors are hungry for diversification and new strategies but also for it's durability in the real world. Relatively few other strategies survive the transition from paper to real-world portfolios the way momentum investing does.

In the textbooks, minting profits looks easy because the standard asset pricing theory suffers from so-called return anomalies-sources of excess returns above and beyond what's implied by the academic models. But exploiting these anomalies in actual portfolios is hard. Trading costs, taxes and other frictions take a toll. And many profitable return patterns that look solid in the financial laboratory have an annoying habit of disappearing when the crowd comes rushing in.

Is momentum investing different? It appears to be. Academics and money managers tend to agree that it is a resilient source of return that stands up to the usual lines of attack, such as criticism that it's simply a byproduct of data mining or that it's vulnerable to arbitrage. It doesn't hurt that the basic idea is as old as investing itself and so it's stood the test of time.

Since it was formally revived in the academic literature for the first time in the early 1990s, there's been a wide-ranging debate about why momentum investing exists and what it means for modern portfolio theory. Yet now there's a growing acceptance of it as a separate and distinct driver of return premiums. As if to herald this broader acceptance and the strategy's coming of age, the first publicly traded index funds that formally target the strategy were launched last year by AQR Capital Management.

Is it time to consider (or reconsider) momentum as part of a diversified portfolio strategy? The answer would seem to be yes, if the expanding menu of product choices linked to it is any indication.

But the fact that money is chasing the strategy isn't a persuasive argument in and of itself. Momentum investing doesn't offer easy profits or sidestep risk. Indeed, investors seeking out such an approach must be comfortable with who's running the strategy and understand the methodology. In short, the details matter.

A Formal Grasp Of The Obvious
The concept of momentum, generally, is unpretentious. Isaac Newton offered a useful working definition of it in Principia Mathematica more than three centuries ago: A body in motion tends to stay in motion.

As an investment concept, the idea has been around for about as long as organized securities markets have been operating. Momentum-based trading advice dominates the famous 1923 book Reminiscences of a Stock Operator, which says the trend is your friend. And the bull market of the 1960s brought the first wave of momentum-influenced mutual funds, inspired by the trading successes of money manager Gerald Tsai. John Brooks famously chronicled this era in his best seller, The Go-Go Years.

In the modern era, academic research on momentum begins with a 1993 paper in the Journal of Finance by Narasimhan Jegadeesh and Sheridan Titman, who showed how buying stocks that have performed well in the past and selling stocks that have poorly performed in the past can win an investor significant returns over three-to-12-month periods (in their now famous report titled, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency.)

Professors Eugene Fama and Ken French cited the momentum factor as an "embarrassment" for their own popular three-factor asset pricing model, which identifies small and value stocks, along with the overall market, as the primary risk factors driving equity returns. Fama and French couldn't explain the success of momentum investing, even if they did acknowledge its existence.

One researcher, Mark Carhart, a finance professor and former managing director of the quantitative strategies group at Goldman Sachs, took the hint and simply added momentum to Fama and French's three-factor model, coming up with a four-factor approach that he outlined in a 1997 study. Carhart decided that a richer framework for deciphering stock returns would require researchers to look to momentum and place it in the context of value and small-cap effects, along with the broad market beta.

"Momentum is ubiquitous across all major asset classes," says professor Craig Pirrong at the University of Houston, summarizing the conclusion in one of his own research efforts.

A similar finding echoes throughout the analysis of Mebane Faber, a portfolio manager at Cambria Investment Management. His work demonstrates that momentum investing's close cousin-trend following-has proved its worth as a risk management tool in connection with tactical asset allocation. Faber's research showed that using the buy and sell signals generated with simple ten-month moving averages in a portfolio of five asset classes dramatically lowered overall volatility without sacrificing return (in this case, the classes are represented by indexes for domestic and foreign stocks, U.S. Treasurys, REITs and commodities). He explains in his book The Ivy Portfolio that managing a mix of asset classes using only moving averages can help an investor achieve a high Sharpe ratio (in other words, relatively high performance for the amount of volatility) when measured over several decades and compared with a buy-and-hold portfolio of the same assets.

Faber calls momentum one of the timeless sources of benchmark-beating return. Why does this particular alpha endure? The answer is bound up with investors' behavior-the byproduct of greed and fear, he says.

Taking a page from Carhart and the four-factor model, researchers have thus continued to combine momentum-based strategies with other factors. For instance, a forthcoming study in the Journal of Portfolio Management reviews the benefits of combining momentum and value strategies in a global tactical asset allocation portfolio. "Momentum and value strategies applied to [such a portfolio] across 12 asset classes deliver statistically and economically significant abnormal returns," write David Blitz and Pim Van Vliet in the report, Global Tactical Cross-Asset Allocation: Applying Value and Momentum Across Asset Classes.

Meanwhile, Shafiq Ebrahim, a researcher with Aronson+Johnson+Ortiz (AJO), a value-oriented quantitative money management firm in Philadelphia, explains,  "Academic and practitioner research has shown pervasive support for the profitability of momentum strategies across several markets/asset classes."

No wonder that momentum is routinely cited as a central force in managed futures strategies, which cast a wide net across asset classes, including equities, bonds, commodities and currencies. "Momentum is the dominant component of performance for the managed futures industry," says Pat Welton, CEO of Welton Investment, a managed futures shop in Carmel, Calif., with $500 million under management.

If the momentum factor is pervasive across asset classes, surely it accounts for some portion of what's generally thought of as manager "skill." But it's been difficult to compare active management strategies to relevant benchmarks and make the kind of apples-to-apples evaluations for momentum investing that an investor would for small-cap and large-cap managers against their respective indexes. Analyzing returns through a momentum prism has only just begun in the wider world of investing.

Momentum considerations are one of several "pillars" in AJO's multi-factor valuation model, says Ebrahim. "We use it to distinguish attractive stocks from unattractive ones."

A number of studies published over the years suggest that momentum also shows promise in terms of asset allocation. That has inspired some financial advisors to consider one of a growing number of publicly traded managed futures funds in client portfolios for enhancing conventional asset allocation strategies. Russell Wild, a financial planner in Allentown, Pa., has recently been analyzing the Elements S&P Commodity Trends Indicator exchange-traded note (LSC) as a potential investment. For the moment, he's "dabbling" with it in his personal portfolio. The fundamental attraction is its mandate to hold long and short positions in a wide variety of futures contracts. "It's momentum both ways," says Wild, author of Exchange Traded Funds for Dummies.

A 2009 research report by MSCI Barra looks at the historical record for targeting momentum and other "alternative betas" for enhancing results in a traditional asset allocation of stocks and bonds (The report is called Portfolio of Risk Premia: A New Approach to Diversification, by Remy Briand, et al.). The basic conclusion: Adding momentum and other non-traditional approaches to a traditional stock/bond mix can substantially lower overall portfolio volatility with a minimal reduction in return.

That's old news, as far as the research literature goes. Ditto for the institutional world's use of alternative betas for augmenting conventional asset allocation strategies. What's new is the average investor's ability to formally allocate portfolio assets to momentum investments in something approaching the beta's pure form, courtesy of the new trio of AQR mutual funds.

Momentum Index Funds
"We thought the time had come," says Ronen Israel, a principal at AQR, about the timing of the firm's 2009 launch of index products. The new mutual funds come in three momentum-tracking forms: large-/mid-cap U.S. stocks, small-cap U.S. stocks and foreign developed-market equities. The underlying indices are proprietary momentum benchmarks designed by AQR.

The 12-year AQR has a long relationship with momentum research. In particular, Cliff Asness (a former quant at Goldman Sachs and one of AQR's founding principals) has penned several studies on the subject over the years, including his 1994 Ph.D. dissertation, which reviewed momentum as one of several sources of equity return.

Among the more compelling arguments for embracing a formal allocation to momentum investments, explains Israel, is the low historical correlation with other specialty betas. Notably, momentum tends to be negatively correlated with the value effect (stocks trading below some fundamental valuation yardstick, such as the book value or a particular price-earnings ratio), he says. On the other hand, momentum tends to be positively correlated with growth. But momentum is more than a proxy for growth stocks, Israel emphasizes. "You can argue it's capturing the best parts of growth."

As evidence, he reports that AQR's large-cap equity momentum index has outperformed large-cap growth stocks (based on the Russell 1000 Growth Index) as well as the broad market (the Russell 3000) over the past three decades. For the 30 years through December 2009, the AQR Momentum Index posted a 13.2% annualized total return, the firm reports. That's comfortably above both the Russell 1000 Growth's 10.1% annual rise and the broad large-cap market's 11.1%, as per the Russell 1000. . And while AQR Momentum's correlation with the Russell 1000 Growth index is positive, it's only mildly so, at 0.43. The implication is that the momentum beta is a complement to growth, or perhaps a replacement.

The Devil In The Details
Momentum's enduring qualities may be reassuring, according to academics and practitioners, but it comes in a variety of flavors. "While momentum may seem like an excellent candidate for alternative beta," advises Ebrahim, "there are variations in how it is defined that could complicate replication." AJO defines momentum in several ways, he explains-as a trend in earnings revision, a trend in prices and a trend in trading volume, among other things.

Ultimately, such differences influence the end result. All momentum strategies, in other words, aren't created equal, as a new study reminds us. A recent research paper co-authored by Hong-Yi Chen analyzes three momentum trading rules for equities according to price, earnings and revenue, reviewing how the three vary in terms of persistence and magnitude to generate an excess return through time (the study is called "Price, Earnings, and Revenue Momentum Strategies," available at SSRN.com)

AQR defines momentum for its three index funds as the top one-third of stocks in the target universe (large-cap U.S. stocks, for instance) ranked by performance over the past 12 months. The benchmarks are reconstituted every three months to keep the indices populated with recent winners.
Momentum, after all, is somewhat ephemeral as risk premiums go. Although the general effect endures, it requires a fair amount of maintenance through trading for an investor to capture the beta as a long-term proposition and thus beat plain vanilla index strategies based instead on market cap.

"The implementation of a momentum strategy involves significant costs because such portfolios tend to contain stocks that are relatively more expensive to trade," Ebrahim warns. "It is unlikely that [a momentum] strategy would be profitable in the absence of careful management of transaction costs." That doesn't deter AJO, AQR and other shops, but it does keep them humble and focused on costs. Or at least it should.

All things being equal, the stakes are higher for portfolio managers using momentum relative to more familiar risk premiums. Why? Successful momentum investing requires a deft hand. That inspires the question: Is the extra work worth the effort?

The historical studies say yes. But analyzing the past is one thing; showing real world results is something else. Momentum investing has clearly been a strong source of return for a number of active managers. It also casts a long shadow across the encouraging track record of managed futures as an industry. With the advent of AQR's index funds, a new era dawns for using momentum as a stand-alone strategy and, arguably, in a passive way. But it doesn't come cheap, at least by indexing standards. The new AQR funds carry net expense ratios of 49 to 65 basis points. That's quite reasonable for active management, though investors will look twice at the price in a marketplace where some ETFs offer broad equity betas for less than 10 basis points.

As for experimenting with new risk factors in public funds, we've been here before. In the early 1980s, for instance, there was hope for the newly recognized power of the small-cap premium. In the following decade, small-cap value earned an academic blessing. In both cases, new products arrived soon after to exploit these academically sanctioned ideas. As a general proposition, both have worked out well, albeit with varying results.

Will a similar fate bless momentum with the mass audience? Stay tuned.

James Picerno is editor of The Beta Investment Report  and author of Dynamic Asset Allocation (Bloomberg Press).