Thu, 8 Sep 2016
Low-volatility strategies have a role to play, but investors shouldn't expect to get higher long-run returns by taking on less risk, says Vanguard's John Ameriks.
Alex Bryan: From Morningstar, I'm Alex Bryan. Low-volatility strategies have gained popularity over the last few years. Since first being introduced in 2011, low-volatility branded ETFs and mutual funds have garnered nearly $41 billion in assets. Here to discuss these trends is John Ameriks from Vanguard. He's the head of the quantitative equity group there.
John Ameriks: That's right, thanks Alex.
Bryan: John, thanks for joining me.
Ameriks: Happy to be here.
Bryan: Low-volatility strategies are basically equity-oriented strategies that are trying to give investors a smoother ride than the broader market. But what's the case for owning these strategies, rather than just adopting a more conservative asset allocation?
Ameriks: That's a great question. Probably I would start out just by saying, I guess we've called the category low-vol. But as you know there are different approaches within that; the way that Vanguard implements it is more of a minimum-volatility strategy. But the case for building a minimum-volatility product is simply that it allows an investor to stay 100% invested in the equity markets, while delivering a more controlled, more disciplined, way to get lower volatility into the portfolio.
Bryan: OK. So if I'm an investor and I want exposure to stocks, this might be a good way, if I'm more risk-averse, of cutting my downside risk?
Ameriks: You've got it. Again, to some degree, there really is nothing new under the sun, right? We've always had defensive strategies, or strategies that are focused on what people see as lower-risk type investments. What minimum volatility and low volatility do is apply the quantitative tools that quant managers have been using for a long time, to try to be very disciplined about how we think about lowering volatility. You do it in a rules-based way or using an optimization engine to really provide a more disciplined, hopefully more consistent approach, to delivering a portfolio that should have less of the ups and downs than the broader market does.
Bryan: Now historically low-volatility stocks have offered better risk-adjusted performance than the broad market, largely by conserving capital on the downside. Is it reasonable to expect that type of outperformance to continue going forward?
Ameriks: Well, I think I can be pretty definitive on this one. We don't think it's reasonable for investors to expect higher returns with lower risk. While I'm an active manager, and for someone from Vanguard maybe that's news to folks, that the QEG group of Vanguard runs our active strategies. With that, we believe there have to be some inefficiencies in the marketplace. There have to be some ways for us to add value. The issue is with minimum volatility or low volatility, there is a significant, we think a significant, volatility reduction is achievable. With that, there's not enough efficiency in the market for us not to be giving up at least a little bit of return. I would say our read of the anomalies literature on this, and min-vol and low-vol is in one way the granddaddy of the anomalies. No sooner had the CAPM been out there, then somebody found out, "Well, it doesn't seem to work in this particular way." There's strong evidence, there's out-of-sample evidence, that there is a risk-adjusted return premium in the space, historically. Going forward, open question about whether that's going to continue. And we're also very skeptical that the size of that anomaly, and the history, is going to be able to persist. Particularly in light of the way you led the introduction; $41 billion now in products that are specifically designed at looking at this.
Bryan: And one concern that a lot of people have is, given the increased popularity of these strategies and the attractive performance, particularly over the past few years, many of these stocks are trading at above-average valuations, higher valuations than they have historically. Does that portend worse performance going forward?
Ameriks: I'd say, I do agree with some of the sentiment out there that whenever you're investing you need to think about the valuation of what you're investing into. So that part's important; people should be conscious of this. After saying that, let's now go back and look at the data. And there are a couple things there that make me feel a little bit more comfortable. First of all, our read of the historical record is that these low-volatility strategies have tended to have a slight premium, even over the longer-term history. Secondly, when you look at the current situation, that volatility premium's actually not that high. I think in our particular case, we're at around 22, in terms of a P/E. Maybe a little higher than that, whereas the market itself is 21 something. While we pay attention to this thing, and it's something that investors should always keep their eye out for, in this particular case, we don't think that evidence is strong enough for people to be sounding the alarm bells.
Bryan: Now there's two major approaches that providers have gone out there to try to reduce volatility. Some funds out there simply target the least volatile stocks in the market, and then try to tilt toward the least-volatile members of that portfolio. Then there's others out there, including Vanguard, that use a more sophisticated optimization framework that tries to construct the least-volatile portfolio possible under a set of constraints. Could you talk about the relative advantages and disadvantages of that more optimized approach?
Ameriks: Sure I'm happy to do it, and I appreciate the compliment. I'm never sure when people say it's more sophisticated or less sophisticated; you can get pretty sophisticated with how you construct the weighting that's based on volatility, so I definitely don't want to denigrate that approach. I do want to differentiate it from what we do, which is to use an optimization, to use a risk model, to try to control volatility. I think for us, it goes back to what's the objective. We're trying to lower volatility in the portfolio. That's the thing that we care about the most. We'll use our commercial risk model to help us identify what kind of expected volatility and correlation do we see among stocks, and how can we put together a portfolio optimally, to deliver the lowest possible volatility, subject to a couple of other constraints, which you mentioned.
The big ones for us are constraints around particular industry exposure; that helps us to avoid concentration risk. The others are just around liquidity. We want to make sure that there's enough liquidity in the portfolio through time. It's a very different process, but it's all driven from what are you trying to achieve? If somebody's out there trying to achieve outperformance, has a particular theory on how you define the particular stocks that you want to overweight to try to achieve outperformance, I could see why a different methodology would make some sense. But in our case, where the sole focus is really on, can we provide lower relative volatility, we think the approach we've got is the right way to go.
Bryan: And do you feel that there are any misconceptions about low-volatility investing? Certainly a lot of people are new to this type of strategy. Do you think that they might be using it poorly?
Ameriks: Well, we worry about that a lot. And even from the start, when we launched the fund and we did the materials and put informational literature out on the website, we tried to talk about realistic expectations. We have no interest in disappointing clients in terms of putting strategies out there. I've mentioned one thing already: Expecting outperformance I think is a bridge too far over the long term. Risk and return we do think are related, so I would dial back those expectations. In light of the run that we've seen in the strategy recently, you worry about that. Are people just looking at the performance number and putting money there?
The other misconceptions are that there's sort of superiority in terms of one way or the other in delivering a lower-volatility, minimum-volatility strategy. I hope the discussion we've had here today says, "Look, you've got to be careful and think about what you want to do here." These are active strategies. They involve tilting a portfolio away from market capitalization to achieve a particular objective. You need to get clear on what's your objective. You're trying to outperform, you're trying to lower volatility, that could guide your choice among high-level construction methodology, and then you've got to peel back the onion a little bit more, just like you would for any other active strategy.
Bryan: Absolutely. Thank you, John, for sharing those insights.
Ameriks: It's great to be here. Thank you.
Bryan: For Morningstar, I'm Alex Bryan.