- Additional factors creep into single-factor smart-beta ETFs
- Funds marrying active and passive traits grow in popularity
Turns out, one of the hottest parts of the $2.3 trillion ETF industry has a little-noticed quirk with the potential to blindside users.
Fund titans like BlackRock Inc. and Invesco Ltd. have amassed billions in assets by selling their clients on smart-beta ETFs. By using computer models instead of human judgment to target strategies like growth or low volatility, they aim to combine the best of both the active and passive worlds. The pitch has helped swell assets in the ETFs by almost $30 billion this year.
Yet for all their popularity, there’s an inconvenient fact that even the most sophisticated investors may be overlooking.
In mutual funds, when a value manager strays from his mandate and ends up owning a bunch of high-priced growth stocks, they call it style drift. While the algorithms that do the stock picking in smart beta are too clever for that, a related hazard exists. It’s when funds tuned to one strategy start to be influenced by another: a low-volatility portfolio that gets infused with momentum stocks, for example.
“You can use smart beta for implementing factor investing, but you have to be very careful with how you do it,” said David Blitz, Robeco Asset Management’s head of quantitative equity research, who published a study in April about the complexities of using smart-beta indexes in pure factor investing. “What you end up with is very different than what you had in mind.”
Unintended exposure caused annualized returns for smart-beta ETFs tied to dividend stocks to vary by as much as 80 percent over the past 10 years, according to a research paper by Northern Trust Corp. It’s a big enough concern that the Chicago-based firm developed a rating system for how closely funds track their goal.
Nobody’s saying the fund providers are to blame, or that they’re trying to dupe unwitting investors. Basically, the issue comes down to simple math and the difficulty of isolating single characteristics, or factors, in baskets of stocks that don’t go short. And how much anyone should care about this depends on whom you ask. Northern Trust markets ETFs designed to root out impurities, while BlackRock says people who use its products correctly won’t be harmed.
“It doesn’t alter the risk reduction characteristics at the portfolio level,” said Robert Nestor, BlackRock’s head of iShares U.S. equity and fixed income smart beta strategy. The firm’s minimum volatility ETF “is designed to deliver roughly market returns with 15 to 20 percent less risk, and that’s exactly what it’s done since inception.”
Low-vol ETFs have been a goldmine for Invesco’s PowerShares and BlackRock’s iShares franchises in 2016, amassing the bulk of the smart-beta inflows. As a buyer, you might think you’re getting an exchange-traded fund with shares that don’t jump around too much. And almost always, that’s true. But using Bloomberg’s portfolio analytics tool on their low-volatility ETFs also shows your fortunes can also align with stocks that have the highest price momentum or valuations, in addition to what you paid for.
It’s no small matter. As the investing public embraces passive investing, fund companies have used smart beta to offer a wider array of ETFs that focus on specific characteristics. Today, one in five U.S.-listed ETFs uses smart beta while more than $400 billion is tied to the strategies.
Smart-beta ETFs are prone to the same influences that affect any portfolio as circumstances in the market change. One reason is that many use relatively simple methods to rejigger their holdings -- pick the 200 least-volatile stocks in the S&P 500 and build a security out of it, for example.
Perhaps the most famous example occurred in May, when the $7.94 billion PowerShares S&P 500 Low Volatility Portfolio started posting bigger price swings than the tech-heavy Nasdaq Composite Index itself. Some analysts have speculated the sheer popularity of the fund has been behind the turbulence. In the first half of 2016, the ETF had more than $1 billion of inflows, putting it on track for its biggest ever year.
“I don’t think investors should be concerned by it, but they should be aware,” said Todd Rosenbluth, the director of ETF and mutual-fund research at S&P Global Inc., which compiles the index underpinning the PowerShares low-volatility fund. “If you’re holding on to them for a longer period during a full market cycle, which is how they’re intended, it shouldn’t matter as much.”
BlackRock’s $15.1 billion iShares U.S. MSCI Minimum Volatility Fund, whose underlying index does well on the Northern Trust purity scale, consists of stocks whose day-to-day fluctuations have been muted. But the underlying portfolio sometimes is just as apt to rise or fall due to another investment factor: momentum.
“The problem comes when that extraneous exposure comes to dominate performance,” Michael Hunstad, director of quantitative research at Northern Trust, said by phone. “If all of a sudden momentum went against you when the volatility factor itself wasn’t going against you, that’s an issue.”
BlackRock contends it’s a feature rather than a flaw.
Although investors frequently view it as such, the ETF isn’t strictly a collection of low volatility stocks, according to BlackRock’s Nestor. Instead, its intent is “risk mitigation,” which means the fund will have a diversity of styles, including momentum. In this case, the ETF includes stocks like Newmont Mining Corp., with a one-month historical volatility almost double that of the S&P 500.
“It’s hugely overweight low volatility and low beta, but certain other factors will come into it if it’s risk diversified,” Nestor said. “Momentum over the last year has been often the second most-weighted factor inside it, and that’s because momentum has been highly diversified to low volatility stocks.”
The extent of style drift also differs depending on how you measure it.
Single-factor funds often have variations in how they define value, momentum, growth, on and on. The iShares low volatility ETF takes the MSCI USA Index and rearranges the stocks based on the Barra risk model. The idea is that stocks with the lowest historical volatility get the heaviest weightings in the index. Invesco’s version is made up of the 100 stocks with the lowest volatility in the S&P 500. Though the models aren’t necessarily simple, without additional mathematical programs that can hedge out other risks, other factors creep in.
Concern about factor purity is related to a debate about smart-beta styles ignited in February by one of the industry’s pioneers, Research Affiliates LLC’s Rob Arnott. He sent tremors through investing circles with a paper warning that some strategies like low volatility or momentum are succeeding only because their valuations are inflating. One day, they’ll crash, he said.
“These are things investors don’t pay as much attention to,” said Jason Hsu, chief executive officer of Hong Kong-based Rayliant Global Advisors and Research Affiliates’s co-founder. “They focus on primary factors and don’t pay attention to the tertiary or secondary factors, which sometimes could come to dominate performance in extreme conditions.”