Innovation in Access: Market Neutral ETFs

In my opinion, innovation in the ETF space has been almost entirely constrained to innovation in access.  By access, I mean the continued slicing-and-dicing of asset classes into specialized verticals and the efficient packaging of those verticals into tradable securities.  As participants in the ETF managed portfolio arena, we love this innovation: it’s like providing a chef with more and more ingredients to choose from. Four ingredients that we’ve had fun playing with lately are from our friends over at QuantShares.  The ETFs are:

  • BTAL: U.S. Market Neutral Anti-Beta
  • MOM: U.S. Market Neutral Momentum
  • SIZ: U.S. Market Neutral Size
  • CHEP: U.S. Market Neutral Value

We think these ETFs are particularly interesting because they provide access to pure premia harvesting strategies.  Even better, we can express this strategy through a long position — which we think is far more palatable to retail investors.

By removing market beta, these strategies gain a fair bit of independence from market returns.  But they also look pretty … well, weird.

MN Growth of $1BTAL, in particular, looks pretty unattractive from a standard “total return” perspective.  But when you think about what it is — long low-beta & short high-beta — you can almost think of it as a market hedge.  For example, in a market dislocation, when people seek to reduce market exposure, they may move from high-beta to low-beta securities.  In comparison to the other long only “hedging” alternatives like VXX (which since inception has lost 99.3% of its value) it is a pretty attractive play.


Market neutral has always lived in a strange place.  From an asset class analysis, it would be classified as an equity.  From a return analysis, it looks totally unique.  The problem is that the returns often look like picking up pennies in from of a steam-roller.  When you consider what these long/short portfolios are from a theoretical perspective, it makes sense: you are isolating a risk premium and collecting it.  When that risk is realized, however, you tend to get run over.  And it is why, in isolation, these strategies can look pretty unattractive:

L/S Performance DifferenceThe dispersion in performance between these long/short factors is astounding.  Consider the 2009 dispersion between Value and Momentum: 25% to -48%.  What client, in their right mind, is going to continue holding an ETF that is down 48% in 2009?

But what happens if we look at these premium harvesters as a group?  What happens if we look at their return profiles and ignore their composition?  We see something that is potentially very, very attractive:

LS Performance Correlation

MN Up CaptureMN Down CaptureWe see returns that are diversified to equities, fixed-income and to each other.  In cases like BTAL, we even see asymmetric return profiles versus the market; while it may be negatively correlated, historically it has lost less when the market goes up than it makes when the market goes down.  Sounds like a great hedge…


Most portfolios are comprised of “return enhancers” and “risk mitigators.”  Equities fall in the former category and fixed-income in the latter.  Because the majority of investors are risk-averse, we tend to place some of our capital into the risk mitigator category to protect from losses due to return enhancers.

So where do market-neutral strategies fall?  Purely based on the size of their long-term return characteristics, you’d probably have to lump them in with the risk-mitigators.  In particular, I think they can serve as an excellent fixed-income alternative for those who are fearful of duration.  But I also see them as return enhancers,

The issue with these strategies is that alone, they look like picking up pennies in front of a train.  How can something be a risk mitigator when it is down nearly 50% in 2009?  It is when packaged together that these strategies become particularly interesting.  Ignoring their correlations (which leaves some returns on the table, especially with the fairly stable inverse relationship between the momentum and value premia), a simple inverse-volatility weighted portfolio (rebalanced monthly based on trailing 63-day volatility) ends up with an annualized return of 4.3% with an annualized volatility of only 3.1%.  The max drawdown?  5%.

LS Portfolio

ETFs like these have me really excited for the years to come.  Innovation in access means more tools in my toolbox.  The greater the number of tools in the toolbox, the more ways we can look to preserve and grow client wealth.

As a side note, for those concerned about using ETFs with low AUM and liquidity, I recommend giving EGShares’ “Thinly-Traded ETFs” paper a read.


The Importance of Long/Short Portfolios

Corey Hoffstein's picture

I thought that David Cantor’s thoughtful response to our recent blog post “Innovation in Access: Market Neutral ETFs” deserved its own article.  The “Innovation in Access: Market Neutral ETFs” post discussed four long/short ETFs that were launched by our friends at QuantShares in 2011.  One of these ETFs, QuantShares US Market Neutral Anti-Beta (BTAL), implements a long position in low beta U.S. equities and a short position in high beta U.S. equities.  David wondered why BTAL* is down since 2003 even though low beta stocks have outperformed high beta stocks both on an absolute and a risk-adjusted basis over this same period.  David is absolutely correct in his assessment of BTAL’s performance relative to the long and short sides of the strategy in isolation.


What is going on here?  Shouldn’t a long/short portfolio in which the long side outperforms the short side make money?  The answer is yes if we are considering a period without any rebalances.  For example, if an investor bought the low beta stocks and sold the high beta stocks at the end of 2003 and didn’t touch the portfolio, the results would be as follows:


BTAL, which rebalances monthly, that the answer to the underlined question is no over periods with rebalances.  A simple example might provide some more clarity on this point.  Consider two stocks with the following returns:

Without rebalancing, a long/short portfolio that is long ABC and short XYZ would return 2% over the two-year period.  However, with annual rebalancing the portfolio would be down 1.6% [ (1+20%-0%) * (1+0%-18%) – 1 = 1.6%].

This result implies that investors with a specific view on two securities that they wish to express through a long/short portfolio should not rebalance during their investment horizon if they wish to profit if their view turns out to be correct.

This does not mean that there are not good reasons to rebalance a long/short portfolio.  The need for and frequency of rebalancing should be tailored to the portfolio’s objective.  For example, a market neutral long/short portfolio whose objective is to diversify core holdings should be rebalancing periodically since market fluctuations can erode the market neutral nature of the portfolio over time.  In a similar vein, retail long/short products should in many instances rebalance periodically so that investors entering the fund at different times get the same exposure.