How To Tell If A Smart Beta Trade Is Crowded

Astor Investments’ Deepika Sharma discusses the firm’s approach to smart-beta strategies.

etf
|
Reviewed by: John Swolfs
,
Edited by: John Swolfs

With smart beta and factor strategies growing in popularity, getting in and out of them has become a real concern for today's investors. Are they getting too crowded? How long is too long to be in a strategy? Can I bet against these factors and make money?

Deepika Sharma, portfolio manager and managing director of investments at Astor Investment Management, has done the research and knows what it will take for investors to have success using smart-beta and factor-based ETFs.

Ahead of our Inside Smart Beta conference in New York City on June 8 and 9, Sharma explains why looking at the economic cycle is vital to smart-beta investing and why we still have a long way to go before we reach capacity in these innovative strategies.

Inside ETFs: Would you walk us through a little bit of Astor’s approach to asset management and your firm’s thoughts on smart-beta or factor investing? Sharma: Absolutely. At Astor, we believe in the predictive ability of economic fundamentals. We call it the Astor Economic Index, which we use for asset allocation.

The Astor Economic Index aggregates trends in economic fundamental data, looking at coincident and leading indicators to determine where the economy is now, and where it is headed. We put the biggest loading into the macro factor or the business cycle risk factor to determine our ratio of stocks and bonds, and how to direct asset allocation for the next six to nine months.

Inside ETFs: Does Astor distinguish between factors and smart beta?

Sharma: The way that most investors understand smart beta is that it’s a “new” way to invest in equity markets. We disagree, which is why we often use the term “factors” rather than “smart beta.”

When we say “factors,” we cover broad and persistent drivers of performance across asset classes that have been used by active managers for decades. These factors—value, carry, momentum, volatility, etc.—work in other asset classes like fixed income, currencies and commodities as well, and often overlap with smart-beta strategies.

The perception of what smart beta is will change as there is more information and more products that are looking at factors or smart-beta strategies outside of equity markets.  

Inside ETFs: When we talk about smart-beta factors, they’re typically long-term solutions. Astor employs a tactical approach to investing. How do the two approaches work together?

Sharma: We try to look at factors using the same principles we use for asset allocation, which is an economic regime and risk perspective.

We implement that perspective though the Astor Economic Index, and we would rather use the dynamic rather than a tactical trading approach; that is, we can change exposure when economic conditions warrant, not just for the sake of trading. And we can apply the same philosophy of not timing stocks to avoid timing factors.

And by “timing,” I mean making short-term shifts on a month-to-month basis. So you're not going from 100% value to 100% momentum strategy in a month or a quarter. What we’ve found is that people who try to do that, and those who try to chase performance and buy popular factors, usually end up getting burned by that kind of strategy.

We’ve come to the conclusion that, just as timing stocks doesn't significantly impact your bottom line if you're a long-term investor—asset allocation does—timing factors will not either, and will just add to your transaction costs.

So, it's better to tilt than time. We tilt by overweighting or underweighting a certain factor depending on whether we’re in a growing economy or we observe signs of weakness, whether or not investors have a high-risk appetite, etc.

 

Inside ETFs: That makes sense, and it sounds like there are times to make changes, but not just at the drop of a hat.

Sharma: Yes, there are times you’d rather be in a defensive- or in a factor-focused ETF, like quality or minimum volatility. This is why, just as we use our economic and risk indicators to determine when to get defensive in asset allocation, we use the same approach for factor rotation; that is, when to move from a factor that gives you return appreciation to one that offers downside protection.

Inside ETFs: I recently saw some research where Astor was talking a little bit about a short-term approach to factors versus a long-term approach to smart beta. Why can the diversification work in the short term, but it’s not ideal for the long-term investor when it comes to factor-based investing?

Sharma: The main reason is that timing or performance chasing hasn’t worked for investors in the short term when it comes to selecting which factor or smart-beta ETF is the best bet for the next month or quarter.

Let’s take a step back: When we talk about smart beta or factor selection, there are two ways to do that. One is by picking one side of value versus growth, or large-cap versus small-cap. We know that value has historically done better than growth. But there are periods when growth beat value by as much as 11.4% in one year.

Another way is that you look at the universe of smart-beta ETFs and ask, is this a good time to be in momentum? Or is it a good time to be in minimum volatility, or a good time to be in value?

In any of those cases, recent research has shown it’s extremely challenging to select and time which factor will outperform either next month or quarter. You can do this by picking the best-performing factor from last month, but how do you predict when the performance will reverse?

Or you can use a valuation-based measure such as P/E or P/Book, which we don’t like, because it further overweights value. You may even get good results on a backtest, but does your backtest accurately measure transaction costs and market impact?

It’s also naive to assume the same conditions that were true in the last decade or so will continue to hold today or in the future. That's why it’s a good idea to start with a diversified approach.

This approach is not ideal in the long term, because a static diversified factor portfolio can sometimes suffer quick and massive drawdowns. The best example is momentum. During adverse downturns, you can get a negative drawdown in a short period. As an example, during the three-month period of March through May 2009, you lost years’ worth of gains from a long/short momentum strategy.

So during those periods of drawdowns, when these factors fall out of favor, your portfolio can really get beat up. You want to avoid those periods where you’re choking your portfolio and wiping away the long-term gains. You need to be nimble, but not to the point of chasing performance.

What our research has shown is that performance among factors varies based on dramatic changes in economic and market conditions. You want to avoid the momentum factor when the economy is weak and market volatility is high. That’s when you’re better off in the minimum-volatility ETFs, which provide a buffer in bear markets. Economic and market trends that really matter don’t change month-to-month, but when they do, you have to be ready to identify those trends and shift your factor exposure quickly.

Inside ETFs: Let’s shift focus to your upcoming session at our Inside Smart Beta conference. One of the hot topics you hear about right now is the crowding of factor trades: Does Astor believe that factor or smart-beta trades are getting crowded? If so, how can investors tell if a trade is getting crowded?

Sharma: That’s something we hear a lot about from our clients. There's a lot of research around that too. I would distinguish between crowding versus a different concern that’s lack of capacity in these ETFs.

The way I think of crowding is, when a smart-beta ETF becomes very popular, there's a lot of asset flow going into that ETF, which can be short-term, and going to dissipate, leading to a temporary drawdown.

And this kind of short-term movement or crowding shouldn’t be relevant for anybody who’s a long-term investor. That kind of crowding is not going to erode your smart-beta alpha or factor premium.

 

The caveat is that this may not hold true if you see that the crowding into the smart-beta ETFs being replicated on the active side too. 

If you’re seeing flows into smart-beta ETFs only because investors are replacing their factor exposure with ETFs, but active managers are not also crowding into those factors, then you're OK. Active managers are twice the size of passive in U.S. equity assets. So factor crowding driven by active managers is a greater cause for concern.

A great metric to measure whether there's crowding in the factors is not looking at smart-beta ETF flows, but looking at the aggregate factor ownership of active managers: Are they placing collective bets into a single factor or strategy that can move the market?

The second part is—and I think that question comes up, but is less important right now—is the question of capacity, where assets under management of smart-beta ETFs gets so high that the market impact and the transaction costs can nullify the smart-beta premium. I don’t think we’re there yet in terms of AUM. But it’s a question we'll have to come back to in the next three to five years.

Momentum might be one of the first to see capacity becoming a problem, being a high-turnover strategy. Lower-turnover strategies like value and quality have much more capacity to sustain their risk premium.

Inside ETFs: Is there a place for a contrarian factor investor, or a contrarian factor strategy, in the marketplace? Or do the factors just not work that way?

Sharma: That’s a great question. We done extensive research into various factor strategies, including contrarian and mean-reversion. There’s something there, as it makes sense to buy a factor when it’s beaten down, trusting the rationale of long-term alpha in that factor. So, any short-term degradation will be followed by a recovery.

But you also have to understand the drivers behind the factor in question. Look at a factor like minimum volatility. Minimum volatility exists because of a behavioral anomaly; that is, the behavior of long-only active equity managers trying to target high returns.

These managers want to beat their benchmarks but are constrained by how much they can leverage. So they’re attracted to a high-return stock that also happens to be a high-risk stock.

This leaves low-volatility securities on the table, resulting in the minimum-volatility premium. But if more active managers start moving into low-volatility stocks or min-vol ETFs, this is going to erode the factor, because its source, or driver, is getting taken away.

So you have to understand what’s driving the factor to outperform, whether it’s behavioral or market structure, or whether it’s low-interest rates and muted equity market volatility and what’s happening to those drivers, to know to what extent a contrarian or mean-reversion approach will work. 

Inside ETFs: It sounds like, to some extent, it’d be hard to go against a factor that’s producing returns.

Sharma: It’s difficult, but sometimes you have to. So, when you typically look at going contrarian, say as an equity investor, or you're using valuation for individual stocks, you're not just going to say, “The valuation is too high. Let me get out of this sector, or company.” You’ll also look at where the high valuation is coming from. Is it coming from multiple expansion? Or is it coming from earnings growth?

It’s the same analysis for factors too. If the right valuation can be justified—because it’s coming from the back of economic growth and earnings growth—that’s a good reason to continue to be in that factor. But if it can’t be justified, there may be some dispersion you can take advantage of.