Factor-based approach gains momentum

Trends in academic research are steering portfolio management to a more factor-based approach, challenging concepts of risk and return, and bringing more transparency and expectations of better performance. Investment approaches considered generators of alpha, or outperformance, now have been seen by academics as generators of beta, or market exposures and drivers of market returns.

“The biggest thing that academics have contributed to the practice of investment that's going on right now is thinking about the world more from a factor perspective,” said Tobias J. Moskowitz, Fama Family professor of finance, University of Chicago, Booth School of Business and research consultant at AQR Capital Management LLC, Greenwich, Conn.

Among those newer factors are profitability, liquidity and the concept of carry. They represent an expansion of factor-based drivers of market returns, along with more established factors of value, momentum and low volatility.

“Academics have been thinking about these sorts of factors as building blocks.” Mr. Moskowitz said.

Matthew E. Stroud, New York-based head of investment strategy-Americas at Towers Watson Investment Services Inc., said executives there have become “big proponents” of factor-based investing, influencing their thinking these days.

One appeal is that many factors can be applied across different asset classes, Mr. Stroud said.

Dimensional Fund Advisors LP has incorporated the profitability factor across all its equity strategies over the past year or so, said Savina Rizova, Santa Monica, Calif.-based vice president, research.

“The profitability research is probably the best example” of a recent integration of academic research into Dimensional portfolios, said Jed Fogdall, vice president and co-head of portfolio management.

“The basic motivation for incorporating profitability ... comes from academia” and the valuation equation, which links expected profitability to expected stock returns, Ms. Rizova said.

The idea was inspired by a 2012 paper by Robert Novy-Marx — associate professor of finance, Simon School of Business, University of Rochester, Rochester, N.Y., and papers in 2006 and 2008 co-authored by Eugene F. Fama — 2013 Nobel Memorial Prize in Economic Sciences co-winner and Robert R. McCormick distinguished service professor of finance, University of Chicago Booth School of Business — and Kenneth R. French — Roth Family distinguished professor of finance, Tuck School of Business, Dartmouth College, Hanover, N.H.

Profitability “is a different, separate dimension of expected stock returns” from the size or value premium, “which means that it gives you additional information about differences in expected returns and helps you target those differences ... in a better, more informed way in your portfolios,” Ms. Rizova said.

“But more than that, it enables us combine different sources of expected stock returns when building portfolios. By integrating different premiums, different dimensions of expected stock returns, we can actually improve the consistency of expected outperformance, the consistency of the reliability of achieving those higher expected returns in equity (portfolios).

“For us, profitability is the fourth dimension of expected stock returns,” after the market, size and value premiums, Ms. Rizova said.

Roger G. Ibbotson, professor emeritus in the practice of finance, Yale University School of Management, and chairman and chief investment officer of Zebra Capital Management LLC, Milford, Conn., has been examining research on the concept of liquidity and has begun applying it to portfolio management.

“What is it (investors) like and don't like?” Mr. Ibbotson asked. They are, respectively, liquidity and risk, he said.

The riskier the investment, the higher the premium over lower risk assets investors expect, Mr. Ibbotson said. Investors, in turn, willingly pay more for liquidity.

As research has uncovered, “low liquidity is a source of high returns, but it's not more risky” than highly liquid investments, Mr. Ibbotson said.

The inspiration for singling out liquidity as an investment factor and applying it to portfolios goes back to scholarly work he has done, including award-winning papers in 1984 and in 2014.

Zebra Capital has created both a long-only portfolio and a long/short hedge fund managed to the liquidity factor, Mr. Ibbotson said.

One factor called carry is related to dividend yields in equities and yields in fixed income. Academic research has found a tendency for higher-yielding securities to outperform lower-yielding securities, a concept generally called carry, Mr. Moskowitz said.

“A carry strategy, would overweight high-yielding securities,” such as high-dividend-yielding stocks, high-coupon-paying bonds, currencies that have high interest rates, or a high yield associated with commodities, Mr. Moskowitz said.

“From a currency point of view, (carry) is playing interest rate-differentials across countries and allocations toward the high-yielding ones,” Mr. Stroud said. “There is more to it than that, but that is the basic idea.”

The outperformance of the carry factor was shown in a 2013 paper “Carry,” co-authored by Mr. Moskowitz; Ralph S.J. Koijen, professor of finance, London Business School; Lasse Heje Pedersen, John A. Paulson professor of finance and alternative investments, Stern School of Business, New York University, also professor of finance, Copenhagen Business School, Frederiksberg, Denmark, and principal, AQR Capital Management; and Evert B. Vrugt, affiliated with Vrije Universiteit Amsterdam.

“Our global carry factor across markets delivers strong average returns and, while it is exposed to recession, liquidity and volatility risks, its performance presents a challenge to asset pricing models,” they wrote in their paper.

In general, academics' work in factor-based approaches shows concepts active managers have been using as part of their skill to produce alpha are really market exposures or market betas.

But alpha of active performance often includes a lot of market exposures, that is market beta. Alpha, a return beyond a benchmarked return and risk, is often misused, AQR's Mr. Moskowitz said.

“Most academics would say that is not really active stock picking or not really securities selection. Those are just other forms of beta.”

“(A)lpha is often cloaked inside a broader portfolio that contains simple market exposures (or betas),” according to a paper earlier this year by co-authors from AQR: Clifford S. Asness, managing and founding principal and chief investment officer; Antti Ilmanen, principal; Ronen Israel, principal; and Mr. Moskowitz. “Since a single fee is charged for the portfolio, investors willing to pay high fees for true alpha end up paying exorbitant fees for traditional market beta. ... For example, hedge fund investors have often paid too much and accepted unfriendly terms for strategies that may contain some alpha but are clearly mixed in with a lot of market beta.”

Towers Watson's Mr. Stroud said, “We are big proponents of ... cost-efficient replication of some basic market exposures,” or betas. They are ideas that are embedded in actively managed portfolios that have “become well understood over time and commoditized over time.”

“Why go through, say, for example an active management vehicle or even a hedge-fund structure paying two and 20 — or a flat fee of 2% of assets under management and 20% of investment profits — for an underlying exposure that is really just a basic market exposure that you pretty much are going to put on and probably keep it there steady for a long time,” Mr. Stroud said.

Andrew Ang's work on factors inspired consultants at Towers Watson, Mr. Stroud said. Mr. Ang is the Ann F. Kaplan professor of business at the Columbia University Business School, New York.

In his paper, “Factor Investing,” published in 2013, Mr. Ang noted, “Many of these factors could be collected more cheaply by passive management.”

“We've worked with an external manger to structure something pretty cost effectively” using factors to improve expected return, said Mr. Stroud, who declined to name the manager.

Mr. Stroud said in his discussions with active mangers, they would present these factor concepts in their strategies: “And we say, "gee, is this really manager skill that we should be paying a full actively managed price for? Or this really a factor that they are allocating to that we can do elsewhere for kind of an index fund fee?'”

Some of these factors “are observable, but there's not necessarily an investment product that exists today,” Mr. Stroud said. “So our process is to engage with the investment management community ... to build a product that can be run cost efficiently.”

“Do you have to go through a hideously expensive actively managed structure” to get such exposure? Mr. Stroud asked. “The answer is often no and (the factors) can be accessed cheaply if you go about it the right way.”

“There is some role for active management, but it is very limited,” Mr. Stroud said. “We are not talking about an area where there is wide discretion for an active manager.”

Mr. Moskowitz said, “One of the things I do for AQR is preach” the concept of market exposures, “and say there is nothing fancy going on here,” no alpha. “It's just exposure to these other betas.”

In applying market exposures to portfolio management, “they are a little more expensive than your typical plain-vanilla market index fund, but they are a lot cheaper than” using them in an actively managed portfolio “or something we think as pure alpha. This is just different forms of beta.”

“That in my mind has been the biggest contribution” from academia, Mr. Moskowitz said. “It's somewhere in between simple passive indexing and certainly very far from traditional active management.”

The factors are an extension of the Fama-French multifactor model, Mr. Moskowitz said.

“People call these (factors) smart betas,” Mr. Moskowitz said. “ “If it were me, I wouldn't call them smart beta or alternative betas. I would just call them additional betas.”

Academics “don't believe there is a single factor world,” Mr. Moskowitz said. “We think there is a multiple factor world, and we are starting to get some consensus of what those factors are.”

The factors “all work independently” in predicting returns, Mr. Moskowitz said. “They also aren't very related to each other,” with correlations ranging from close to zero to negative. AQR started offering the factors in what it calls a style premium mutual fund for institutional investors last November after it began with a hedge fund version about a year ago, he said.

Use of the factors adds value over a traditional market index fund, Mr. Moskowitz said. “I was part of that research” that led to the application of factors into AQR's style funds, Mr. Moskowitz said. Style “is another name for (factors), which also sounds a little sexier.”

Among new factors, Mr. Ibbotson said liquidity is as strong as any of the other factors.

Momentum, a shorter-term phenomenon, “is the tendency of a stock that has performed well over the past (three) to 12 months (compared with other stocks) to continue to outperform ... over the next (three) to 12 months,” Dimensional's Ms. Rizova said.

Liquidity is as strong as value” in predicting stock returns and “stronger empirically than size or momentum,” Mr. Ibbotson said.

“In the bond market you have the same type of thing,” Mr. Ibbotson said. “You've got the ... the liquidity premium in the bond market. That premium “actually has a big influence on what the returns and risk of (a) portfolio are going to be.” n