Long/Short Diversification To The Rescue

Let's start with The Endowment Syndrome: Why Elite Funds Are Falling Behind. Before we really jump in, it's important to understand that endowment funds and foundation type accounts have infinite time horizons that we do not. That reality can change some of the calculus between endowments and individual investor accounts but there are things we can learn from their asset allocations all the same. 

The big idea is that endowments have drifted into having huge allocations to alternative assets which have generally lagged simpler building block exposures accessible to individual accounts through brokerage platforms like a 60/40 portfolio. An interesting comment in the article is that the allocations to alternatives are typically far bigger now than Jack Meyer from Harvard or David Swensen from Yale ever had. The two of them were pretty much pioneers with using alts but based on the article, Meyer and Swensen appear to have better understood the drawbacks and limitations.

The article blames behavioral factors for the huge allocations to alts including vanity. We've touched on that here once or twice, there is an ego stroke to having sophisticated strategies in a portfolio and I think we could sub in the word complex for sophisticated. A repeat theme here though for years has been that equities are the thing that goes up the most, most of the time. If you need normal growth for your plan financial to work then you need an equity-based portfolio maybe hedged with small exposures to alternatives, not an alternative-based portfolio hedged with a little equity exposure. Put differently, simplicity hedged with a little complexity. 

This brings us to a paper from Man Institute that appears to conclude that mixing "long/short quality" with managed futures can help offset the random results that managed futures have when volatility spikes. In the 2020 Pandemic Crash, managed futures did very well at one end of the scale but did poorly in the Great Dip of Early August 2024 as an example at the other end of the scale. 

They have a scatter chart that shows how truly random the performance of managed futures is during volatility spikes. The paper assumes that volatility spikes equate to negative stock market events. The paper distilled a pretty good explanation for why managed futures does well in a volatility spike or why it does poorly. They attribute the different to how it is positioned with longer dated treasuries, long or short. Bonds tend to rally when there is some sort of external event that adversely effects markets. So if bonds were in a negative trend resulting in a short position and then bonds rally as happened in August, managed futures do poorly. If bonds were already in a positive trend like they were in the 2020 Pandemic Crash then managed futures will do well. 

Looking at the chart, I can see why managed futures would have been short TLT, a proxy for treasuries, back in August of this year. TLT started to turn higher in the spring but using a 10 month signal, managed futures could very well have been short. You can see BTAL going up last August which is should do as more of a first responder type of alternative. Putting on my Karl Popper hat, seeing QLEIX not work in August, maybe the paper is wrong. 

The second chart leads into the 2020 Pandemic Crash. You can see why managed futures might have been long treasuries. BTAL went up of course. EBSIX didn't do that well so maybe it was something with that fund, another fund I use did well through this event. 

Where there can be divergences between different managed futures funds, a common suggestion is to split the allocation between multiple funds. We've talked about that once or twice. An easy type of differentiation is to split between a fund that implements the full strategy which might be trading 90-100 markets and a replicator which might trade 10-20 markets. A more difficult way to try to split up the exposure would be to try to differentiate how funds risk-weight differently. Larry Swedroe wrote an article that cites splitting a managed futures allocation between six different managers. 

Back to QLEIX which again struggled in 2020. For what it's worth, in the 2020 event, although QLEIX fell quite a bit, it was about 900 basis points better than the S&P 500. 

The long/short discussion seemed to drift between two different types, the kind offered by the AQR Long/Short Equity Fund (QLEIX) and the kind offered by client/personal holding AGFiQ US Market Neutral Anti-Beta ETF (BTAL). That's why I included both funds in the charts.  Below are a couple of ways to implement what Man appears to be talking about. 

Neither Portfolios 1 or 2 are stock market proxies but offer compelling long term results versus VBAIX. The standard deviation and betas of both are much lower than VBAIX and of course they both did better in 2022, each going up close to 10%. Look though from the start of the study until the end of 2021, both lagged considerably. If we stop the study at the end of 2021, Portfolio 1 compounded at 6.89% and Portfolio 2 compounded at 9.25% versus 11.09% for VBAIX. 

That really can be a long time to languish depending on someone's specific makeup but a portfolio that can get most of the broad market's return with much lower volatility is a pretty good way to go. 

I think this also helps create context for the drawbacks of backtesting. They can be useful for creating some understanding and setting expectations for things like volatility, how well the defensive holdings might do and with setting some performance expectations. If VBAIX is up 20% next year, you can expect Portfolios 1 and 2 to be up less but you have no way of knowing ahead of time whether less means it goes up 4% or 16%. If VBAIX goes down 20% next year, you can expect that Portfolios 1 and 2 would be down less but you have no way of knowing whether less means down 3% or down 16%. A lucky outcome against a 20% decline for VBAIX could be that 1 and 2 go up like they did in 2022 but there is no way to know ahead of time. 

The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.