Why Low-Volatility Investing Works

Raman Subramanian: Thank you, Alex.

Bryan: Let me start with a very basic question that I've always wondered about. If investors are concerned about the volatility of their investments, why wouldn't they just adopt a more conservative asset allocation by owning more bonds and fewer stocks rather than buying into a low-volatility stock strategy?

Subramanian: That's a great question to start our discussion. If we look at asset allocation, as you know, it's a very important process for investors. Asset allocation means how do you actually allocate between equities, bonds, fixed income, and other asset classes. That drives the performance of the overall portfolio in the long run.

Now, in that perspective, each asset class has its role. So, for example, equities are there to provide you long-term potential in terms of equity risk premia and then you have the cash which provide you more stability with a lower risk. But if you are an investor and you have been given a mandate to provide a growth in the long term in the portfolio, then you don't want to change the mix of the portfolio. You don't want to increase or decrease based upon what you have done on the asset-liability study.

So, in that context, changing the profile of the portfolio by increasing cash that means that you may end up with a lower performance in the long run which can lead to underfunded status. Now, that's why the importance of a minimum volatility or low-volatility strategy come in, because in the long run low-vol strategies have provided returns which are similar to the market but with a lower risk. So, investors who want to maintain the total asset allocation profile that was determined using the asset-liability studies will probably gravitate toward using more min vol-like strategy as a part of their equity allocation.

Bryan: Why has that been that low-volatility strategies have offered better risk-adjusted performance than the market? Is there any expectation that this will continue into the future?

Subramanian: There are various research studies which have focused on what causes the anomaly or the premia which we have seen in the min-vol effect. Predominantly, people talk about more of a behavioral reason but there are also constraints which have been put in by the institutional investors on how the money is managed. So, what we have seen, as long as those constraints are managed, as long as the behavior of investors is not rational, these kinds of strategies will tend to provide the premia in the long run.

Bryan: So, there may be an element of mispricing in the market.

Subramanian: Exactly.

Bryan: I want to ask, there's two main approaches to low-volatility investing out there. Either you could target stocks that have exhibited low volatility and just simply do that. And that's the approach that some PowerShares ETFs have taken. You could also use a more complex optimization approach that tries to look not only at individual stock volatilities but also how stocks interact with each other in the portfolio to affect the overall volatility. Could you explain why MSCI has elected to take the latter approach using this optimization framework?

Subramanian: So, if you look at, as you pointed out, there are two different approaches; one is simple heuristic ranking-based approach, the second is more sophisticated optimization-based approach. Now, to understand why one approach is better than the second approach, you have to start with how do you define the total volatility of an index or a portfolio.

Now, when you talk about stock level volatility, you are just focusing on the stock level return variation which has happened. But in terms of portfolio, the stock level volatility is important but also the interaction, the correlation among the stock is also important. And when you create a ranking-based approach index, the heuristic index, it ignores the correlation effect, and the correlation effect becomes more important when the cross-sectional variation is very high among the stocks. So, that's why I think what we see is that the second approach is more sophisticated. 

The second aspect of that is that when you do a simple ranking-based approach, it can create unintended exposure to various styles, countries, and sectors. That can be also avoided when you go for the more sophisticated approach.

Bryan: And that's an important point, right, because if you just rank stocks based on their volatility, you can get pretty large overweightings in certain sectors, right? Like, utilities tend to have much lower volatility than the rest of the market, but you can end up with a very big exposure to that sector that you may not even intend.

Subramanian: Exactly.

Bryan: Is that an area where the optimization approach can rein in some of those bets?

Subramanian: Exactly. So, basically, optimization approach put this constraint on styles, sectors, and countries, because even some countries are less volatile compared to the other markets because maybe the trading doesn't take place too much out there. So, an optimization approach put that constraint making sure that the index become more replicable and investable for investors. At the end of the day, you want to avoid unintended bets which may deviate from the early purpose of going into min vol, which is capturing the lower volatility effect in the market.

Bryan: I wanted to ask about Research Affiliates. That firm has put out some research indicating that low-volatility strategies, a lot of their historic outperformance has come from expanding valuations and that these strategies are currently expensive and may not be priced to offer attractive performance going forward. Should investors be concerned about the valuations of these strategies at this moment? What are your views on that?

Subramanian: If you look at any equity market or at any asset class, the valuation can change over time. It can rise and fall over time. But more importantly, you have to understand whether valuation may change the profile of the index or not. Now, in case of minimum volatility, what we have seen is that the valuation has not changed the profile. The profile here means that they still have provided lower risk compared to the broader market. So, that is the most critical thing for a low-volatility investor.

The second thing, our research has shown that irrespective of the fact whether the investors have come into the min vol strategy when the valuations were richer or valuations were lower, they have not underperformed relative to the broader market in the next subsequent five years. And this we have back-tested over the last multidecades.

So, from that perspective, I think if you are a low-vol investor, it's better to understand that you are there getting into the strategy for a specific reason and the reason is to lower the overall risk of your program. And in that context, min vol strategies have provided a lower risk compared to the broader market on a three-year rolling basis for multidecades, irrespective of the fact whether the valuations of the min vol strategies were higher or lower than the market.

Bryan: Just to make sure I understand--so, you agree that valuations are currently high but you don't believe that there is anything that investors should do differently if they believe that the risk reduction is kind of what they are targeting; these strategies will likely continue to provide that?

Subramanian: Yeah. In fact, on a trailing basis, valuations are a little bit cheaper compared to the broader market. But on a forward basis, yes, valuations are a little bit on the richer side.

Bryan: Interesting perspective. Thank you for sharing your insights with us.

Subramanian: Thank you, Alex.

Bryan: For Morningstar, I'm Alex Bryan.