We’ve been following along with MA Cap ever since they lobbed a grenade at managed futures mutual funds with their piece: “Managed Futures Mutual Funds Don’t Work Here is Why” last December, and were excited to read their most recent piece tackling the concept of absolute returns and how most people are getting the concept wrong.
They start out with a nice little history of how investors have sought out absolute return over the past few decades:
“… constructing [an absolute return] portfolio requires asset classes that exhibit return and volatility characteristics that are not correlated to other assets in the portfolio. The earliest attempts were made to construct such portfolios with a mix of long-only assets such as domestic stocks and bonds. Then global securities were added to the mix as well, including developed markets, such as in Europe, emerging markets, and even frontier markets of Africa or South East Asia. Later, alternative strategies were also added to the mix using hedge funds, managed futures, real estate and other esoteric classes such as art and wine.”
You don’t have to be a market historian to know that none of those approaches have really worked all that well, with the correlations of supposedly non correlated assets moving to one during market stress periods such as 2000 and 2008. But it’s not just correlations – Monty Agarwal of MA Capital see’s these stress periods, or what he coins “Shock Markets” as changing environments where asset classes deviate greatly from their normal return, alpha, volatility, and correlation parameters.
Now, there’s nothing all that new in that thinking – but what is a bit different is how MA Capital looks at the diversification needed for an absolute return portfolio as a result of that thinking. You see, they look at it not as creating a portfolio which is good for all periods (the usual absolute return pitch), but rather creating a portfolio which changes and adds diversification based on the period it is in.
The first step is analyzing these different volatility/correlation periods over time and how asset classes react to them. They define three such periods – Trend, Shock, and Noise market environments, and look not just at the basic stats (return and volatility) in these periods, but also do a more sophisticated look at the volatility of these periods.
They believe a market’s volatility in and of itself is not telling enough as to how an asset class (especially alternatives like hedge funds and managed futures) will perform in those periods, and as such look at two additional layers of volatility: trend volatility (what some people call directional volatility) and volatility of volatility (regardless of the volatility level, is it expanding or contracting).
Managed futures know all too well that just having volatility isn’t enough. For example, a market going up 5% and down -5% every other day would have high volatility but no trend volatility, while a market going up 2% each day would have high trend volatility, but little instantaneous volatility. Here’s how that looks for the ‘shock market’:
Disclaimer: Past performance is not necessarily indicative of future results.)
We can see that managed futures would like this type of ‘shock market’, with higher volatility, higher volatility on volatility, and high trend volatility with an extended down market. The shocker (pun intended) is that managed futures don’t do well in a trend period, although that may be because of the way MA Capital is defining the trend period (not including bear market trends, which are put in shock category). Here’s their performance analysis of stocks, bonds, and managed futures in each of the market periods:
S&P: S&P 500 Yahoo Finance
Bonds: 30 Yr bonds/Yahoo Finance
BTOP: Barclay Hedge Index
Hedge Funds: HFR
(Disclaimer: Past performance is not necessarily indicative of future results.)
What to do about it?
So if just having an absolute return product doesn’t work because of its underperformance in trend and noise periods, and if splitting part of your stock investments to alternatives doesn’t work because the alternatives aren’t large enough during the shock period when the stocks suffer – what is one to do?
MA Capital just happens to have a solution for this in the paper – proposing that the better way to approach diversification and absolute returns is to dynamically shift assets between asset classes depending on the market volatility environment, moving into alternatives when in a shock market, and back out into stocks when in a trend environment.
Now, they are the first to admit that there are limitations to this, mainly the difficulty and necessity to accurately predict the upcoming volatility period on a consistent basis. But, they believe their Alpha Absolute program and its internal ‘Predator Pattern Recognition Model’ is up to the task on that front. And it can be yours in a separately managed account for the bargain price of $15 million…
Whether you have $15 million lying around to invest or not – the paper is still worth a read and you can find it here.