To view a PDF of the report, click here.
As more investors adopt smart beta, scrutiny is intensifying—not only of factor and style investing, but also of the fine points of portfolio and index construction.
From a few brave early adopters after the 2008 financial crisis, use of the investment strategies now known as smart beta has grown dramatically. And they are becoming mainstream. Growth over the past year alone hit 7.1 percent as stable-value assets under management globally reached a record $429 billion (see “Volatility drives growth,” page 2). “AUM in funds tracking our indices has grown by about 50 percent year-to-date over all of last year,” says Eric Shirbini, global product specialist at Scientific Beta—a pace he expects will continue. “Our pipeline is just as strong as last year.”
Smart-beta vehicles consist of non-market-cap-weighted indexes, usually tilted toward recognized investment factors and styles such as low volatility, quality, value and momentum. An annual smart-beta survey by Market Strategies International reveals that nearly 75 percent of current users view smart-beta strategies as part of their core portfolio. The survey shows investors’ understanding of smart beta rising rapidly as well, 63 percent of institutional decision-makers indicating they are familiar with smart-beta exchange-traded funds (ETFs)—up nearly 10 percent from just two years ago.
“We certainly see more applications,” says John Feyerer, director of equity ETF product strategy at Powershares, the greatest percentage of investors using smart beta to manage volatility and enhance performance. “There is a clear upward trend in utilization of smart-beta or factor strategies.”
| 
|
| Ronen Israel, principal and portfolio manager, AQR | John Feyerer, director of equity ETF product strategy, Powershares |
| “As investors look at these alternative means of accessing these return sources, they are changing the way they build their portfolios.” | “We certainly see more applications. The greatest percentage of investors using smart beta to manage volatility and enhance performance. “There is a clear upward trend in utilization of smart-beta or factor strategies.” |
| Volatility drives growth Assets are pouring into smart-beta ETFs and ETPs. At the end of June, global assets invested in these vehicles reached a record $429 billion, with a five-year compound annual growth rate of 31.3 percent, according to equity research firm ETFGI. Those listed in the U.S. reached record levels, at $390.2 billion, while those listed in Europe hit a new peak of $26.7 billion. Deborah Fuhr, managing partner at ETFGI, attributes much of the growth to market uncertainty following the Brexit vote. Year to date, smart-beta products have seen global net inflows of $16.15 billion. Morningstar reported asset flows to U.S. smart-beta ETFs growing 11 percent in 2015 and 5 percent year-to-date at the end of June. Asset flow growth among European smart-beta ETFs has been even more rapid, hitting 34 percent in 2015 and 18 percent year-to-date. |
A look under the hood
The smart-beta field has been through an education phase over the past five years as more investors implement the strategies and become more familiar with smart beta’s characteristics and how it works. Product proliferation means investors must be more diligent about making sure they understand exactly what they are buying (see “Selection criteria for smart-beta ETFs,” below). “We’ve reached the point in the development of smart beta where people are really taking a closer look at some of these issues and seeing how construction techniques make a difference,” says Jennifer Bender, director of research for the Global Equity Beta Solutions group at SSGA.
Until recently, few investors gave much thought to how index construction techniques actually affect outcomes; today, more are taking a closer look under the hood. “The strategies can have similar names, but when you look closely, they have vastly different approaches,” says Feyerer. For example, there are two common ways to implement an index-based low-volatility strategy. One is simply to rank and weight stocks from a given universe based on their realized volatility. The other is optimized and constrained in a way that provides more direct exposure to the value factor. Should volatility spike in a given sector or region, the constrained approach may not be nimble enough when investors need it most.
“These stark differences can be misunderstood, but due diligence on behalf of investors has gotten increasingly more sophisticated over the past 12 to18 months,” Feyerer says.
Shirbini also sees more focus on methodology. Once simply a concept, with only backtests to prove its effectiveness, smart beta now has a stronger track record. Investors can analyze live performance, compare different providers and understand how construction methodology affects outcomes. Why, for example, do two value strategies produce different returns? “That puts the focus on the construction methodology,” Shirbini says.
Multifactor portfolio construction can become much more complex. There are two basic approaches. In the combination method, one first creates single-factor portfolios, then combines them; in the bottom-up or integrated approach, one creates an aggregate score for each stock across all the desired factors, then assigns weights. It’s an open question which is most effective.
Scientific Beta uses the combination method. “We build our single-factor portfolios first,” says Shirbini. The firm selects half the stocks from the broad universe that have the highest exposure to a particular factor, applying a multi-strategy weighting scheme to avoid concentrations and other unintended risks. The higher the concentration in certain sectors or companies, the more idiosyncratic risk will work its way in.
For example, many value indexes overweighted Volkswagen, which worked against them when the automaker was exposed as having used software in 11 million of its vehicles that cheated on emissions tests. “That is exactly what we want to avoid by diversifying across all of the value stocks,” says Shirbini. Scientific Beta then combines the size, value, low volatility, high profitability and momentum indexes that make up its multifactor index by allocating equally to each. “This provides a second level of diversification, because when one factor does poorly another one may compensate,” he says. It’s worked, as the multifactor index has outperformed the MSCI World Index by 3.3 percent per year.
SSGA, by contrast, uses the bottom-up approach. “We take each individual stock and assign it a score for each of the factors we’re trying to capture,” says Bender. “If you’re interested in capturing more than one factor, it makes sense to weight the stock based on how that stock ranks on, for example, value, momentum and quality at the same time.” Otherwise one could miss out on the interaction effects. In a comparison of single-factor portfolios blended in combination versus a bottom-up portfolio built by considering all the factors simultaneously, SSGA found the bottom-up approach results in a higher annualized rate of return with lower volatility. “There’s no way to capture these differences when you simply combine single-factor portfolios,” Bender says.
| 
| 
|
| Jennifer Bender, director of research for the Global Equity Beta Solutions group, SSGA | Jeremy Baskin, Global CEO and CIO, Rosenberg Equities | Eric Shirbini, global product specialist, Scientific Beta |
| “We’ve reached the point in the development of smart beta where people are really taking a closer look at some of these issues and seeing how construction techniques make a difference.” | “You can also manage overall exposures collectively with more efficiency and transparency than if you simply buy individual factor funds and shift the weights around. But you have to be wary of a few things.” | “AUM in funds tracking our indices has grown by about 50 percent year-to date over all of last year. Our pipeline is just as strong as last year.” |
Some say the bottom-up multifactor portfolio is the most efficient because the strategy avoids buying or selling securities at the same time. “You can also manage overall exposures collectively with more efficiency and transparency than if you simply buy individual factor funds and shift the weights around,” says Jeremy Baskin, Global CEO and CIO of Rosenberg Equities, a unit of AXA Investment Managers. “But you have to be wary of a few things.”
For example, there was a time in 2009 when the correlation between momentum and value was almost -0.9 and few stocks scored favorably on both factors. “If you weighted only those stocks that had the highest combined score, you would end up with very mediocre exposures on both dimensions,” Baskin says.
Keeping it simple
The hallmark of smart beta is to use rules-based methodologies to assign weights to portfolios—disposing with optimization. “But product proliferation has completely confused people about what the so-called best methodology is, and each provider has a different view,” says Bender. The idea has always been to keep it simple by overweighting the stock that ranks higher according to a given characteristic and underweighting the stock that ranks lower. “Once you get more complex, a lot of the benefits might just break down,” she says.
AQR takes the bottom-up, or integrated, approach a step further. “Integrated” means using multiple factors or styles simultaneously in a single portfolio by selecting the securities that look best across all the factors taken together. “It’s the most efficient way to capture those exposures and it leads to better portfolios over time,” says Israel.
But this is one point on the smart-beta spectrum. At one extreme is classic smart beta, which includes single-factor, long-only equity strategies like low-volatility, value or momentum; at the other is multifactor long-only equity. “At this point, you’ve added diversifying, long-term return sources, which can lead to more consistent returns,” says Israel.
The strategy can also go long and short. “Long-short can more fully capture the underlying themes because it is a purer representation of the desired exposures,” says Israel. It’s also uncorrelated to a traditional portfolio, because the primary risk is no longer the equity market but the relative performance of the styles and factors. Finally, the concept can be applied to other asset classes, including those that do not rely on a benchmark, like fixed income, currencies and commodities, resulting in a multi-style, multi-asset-class, long-short framework. “At this point you have a more efficient representation of these returns and one that is more diversifying,” says Israel.
The sheer number of smart beta products can make selection confusing, especially as the degree of selectivity and constituent weighting becomes more intricate in a multifactor strategy. “As you move toward the increasingly complex end of the multifactor spectrum, you start to move away from the beta in smart beta,” says Feyerer. These more complex methodologies tend to require more judgment and reliance on rules-based, quantitative strategies. “There is a place for those,” he says. Investors simply need to know the difference.
