Despite its outperformance and nontraditional approaches to indexing, Dave Nadig warns investors that smart beta comes with its own set of risks and problems.
[This article previously appeared in the April issue of ETF Report.]
This issue of the ETF Report is full of good ideas on how you can use new approaches to indexing to make money, reduce portfolio risk and access new patterns of investment returns. But that's not what this article is about. Instead, I'm here to tell you what can go wrong.
With every new invention comes new dangers. When the steam locomotive started spreading across the United States, people were killed regularly crossing train tracks, even though they could see the oncoming train. Why? It took an entire generation for people to learn how to judge the intercept vector of something moving so fast and get out of the way in time!
It's not all that different with so-called smart-beta ETFs. They can be powerful tools, but you've got to slow down and take a look to avoid getting run over.
Problem 1: False Alpha
Perhaps the biggest problem with alternatives to traditional market-cap indexes is that they can lure you into believing you've found a better mousetrap, when in fact all you've done is found more risk. Consider the myriad ways people have found to improve on the S&P 500. At the time of this writing (1/22/14), the S&P 500 SPDR was up 26.08% for the trailing 12 months. Figure 1 shows some of the other ways you could have skinned those same 500 stocks with ETFs.
This dizzying list of funds has a one-year spread of returns of more than 20%, and employs everything from complex options strategies (PBP | D-52) to VIX futures overlays (VIXH | D-67) to simple equal weighting with regular rebalancing (EQL | D-83). The danger with a group of funds like this is looking at the fund on top (SPHB | A-38) and just assuming that since it's fishing from the same pool, it's simply a better mousetrap and should be bought.
Luckily, SPHB has the decency to put its strategy clearly in the name—it's only buying the 100 highest-beta stocks in the S&P 500. The resulting portfolio has an overall beta of 1.44, which can be interpreted as "when the S&P is up 1%, I should be up 1.44%. When it's down 1%, I should be down 1.44%." In other words, it's effectively acting—from a risk perspective—like leverage in your portfolio.
Worse, however, is the way in which that extra risk manifests. A quick trip to etf.com/sphb would reveal that SPHB has an upside beta of just 1.29, and a downside beta of 1.56. That means when the S&P 500 is up 1%, you only expect to be up 1.29%, but when it's down, it's down even more!
The conclusion here is simple: Smart beta, nearly by definition, takes on different risks than plain-old boring market-cap indexes. Anytime you consider a smart-beta product, you have to look hard at how the "smart" part is affecting the "beta" part.
Problem 2: Crowding
As long as investors go into their investments with open eyes, there's nothing wrong with having a different take on the market. If you want to equal-weight the S&P because you think the effect of larger companies is too great in the index and want a midcap skew, that's fine, as long as you know you're going to eat some returns in rebalancing costs and miss out on the next big Apple run-up to do it.
But sometimes returns look too good to be true, even unexplainably good. Such was the case with the "low volatility" craze that swept the ETF world for much of 2012 and 2013. For much of that period, low-volatility ETFs were the single fastest-growing group of ETFs, with the first entrant, the PowerShares S&P 500 Low Vol ETF (SPLV | A-47), pulling in nearly $5 billion in assets in its first two years on the market.
The flows were chasing what seemed to be an odd anomaly. Academics would tell us that the riskier an asset is, the more return should be demanded by investors. Yet for a while, it seemed as though lower-risk stocks in the S&P 500 were consistently outperforming higher-risk stocks, and not just in down markets, either. This "low-vol anomaly" spawned dozens of academic papers, articles in the Journal of Indexes and eventually, ETFs. At the peak, more than $12 billion in new assets chased these low-vol strategies.
What happened next was somewhat predictable (see Figure 2).
As more and more money started chasing low-vol stocks, they stopped being the outperformers they had been. Indeed, in 2013, a year that featured a correction both in May and in August, the "low vol" stocks significantly underperformed as the market went down, and ended the year trailing the S&P 500 by nearly 10%.
It's of course impossible to prove empirically exactly why the "low vol anomaly" evaporated, but my inclination is to chalk up another victory for efficient markets. When something is too good to be true, people start piling in. Once people pile in, whatever inefficiency was being exploited has a nasty tendency to evaporate.
Problem 3: Tracking
One of the glorious things about market-cap-weighted indexes is that they're comparatively easy to run. Once you own all the stocks you want in their market weights, you generally don't need to touch the portfolio except to reinvest dividends, handle corporate actions and deal with index changes. There's still a science to doing that well, but it's relatively straightforward.
Smart-beta strategies by definition are doing something different. They're holding securities in different weights, adjusting their portfolios to match their "smart" strategies, and often rebalancing regularly. All of that means that ETFs tracking smart-beta indexes often have a much more difficult time than their vanilla brethren.
Consider the iShares S&P 500 ETF (IVV | A-97). We won't bother with a chart, because it's literally impossible to tease out the difference in performance between IVV and the actual S&P 500 Index. By the numbers, it trails the index by precisely the amount of its expense ratio in any given year, with a variation of generally less than a basis point.
The iShares MSCI USA Minimum Volatility ETF (USMV | A-54) fares much worse. While it charges an expense ratio of 15 bps, on an average year, it trails its index by 22 bps. In a bad year, it trails by nearly 40 bps (see Figure 3).
Even something as seemingly simple as equal weighting can add significant variability to your returns. Guggenheim's S&P 500 Equal Weight ETF (RSP | A-72), has a median annual tracking difference of 57 basis points—more than its 40 bp expense ratio would suggest—and in its worst years, trails by almost 70 bps.
Problem 4: Trading
Finally, many smart-beta funds share something in common with all niche ETFs—they often don't trade particularly well. The reason this is particularly concerning for smart-beta products is that often the alternative products being considered—vanilla ETFs—trade very well. Put another way, if you're looking to invest in corn futures, you don't have many options, and will have to deal with some tricky trading to get at the corn ETFs. But global equities? There are dozens of funds trading at fair value and penny spreads, all day long.
Consider one of the fastest-growing funds of 2013, the Vident International Equity ETF (VIDI | F-39). Since its recent launch in November of last year, VIDI has pulled in more than $550 million in assets, making it one of the largest international equity ETFs on the market. Yet despite that initial size, it's struggled to find good on-screen volume. Consequently, the fund has traded at significantly different prices than fair-value might suggest (see Figure 4).
These kinds of premiums and discounts driven by flows aren't uncommon in smart-beta products, and for good reason. Because these strategies follow nonstandard indexes, they're more difficult for market makers and authorized participants to hedge against. That means they're more likely to allow the price of an ETF to drift from "fair" before stepping in to create or redeem shares.
None of these issues—trading, tracking, crowding and false alpha—is unavoidable. In fact, for the most part, they're very easy to spot using free tools like the analytics at ETF.com. As with everything in ETFs, you need to look closely before you leap. It's just doubly true when you're hunting for a better mousetrap. Because sometimes it's not the mouse who gets trapped.