If you've been reading these posts for the last few months or maybe longer, you know how intrigued I am with the concept of ReturnStacking or is it is also known capital efficiency. My interest goes back long before the ReturnStacked ETFs existed and I believe long before the term capital efficiency was common, to Nassim Taleb writing about barbelling returns where most of the risk is allocated to just 10% of a portfolio with the rest in very conservative things like T-bills. Even before that, a story I've told many times, when I was at Fisher Investments in 2002 there were a couple of guys who talked about getting a return equal to the S&P 500 by shorting Nikkei Futures with just 2% of the portfolio and 98% in cash. I don't know if those previous co-workers were correct about that, but it doesn't matter for this blog post it is just an interesting concept.
I've also though, been very skeptical of using the ReturnStacked funds and I still am but you can maybe tell that I've been fumbling around with how to articulate why I am skeptical, at least skeptical of going heavy in them anyway.
When we talk about allocating a lot to simplicity and hedging with a little complexity, it is important for anyone needing normal stock market growth for their plan to work to have plain equity exposure. Whatever allocation to equities they think they need, most of it should be in very simple, very plain holdings. I am not saying index funds only, there are countless plain exposures ranging from index funds of course to individual stocks and all the narrow based ETFs in between.
Buying a semiconductor ETF or individual name, may or may not work out but they are both plain vanilla, simple exposures. It just boils down to being right or wrong about the choice, there's no complexity that could potentially malfunction leading to a blow up like with the short VIX products in 2018.
RSST is stocks and managed futures leveraged up in one fund, RSSY is stocks and carry leveraged up in one fund and SPY is plain S&P 500 exposure. If you wanted a plain vanilla S&P 500 index fund, SPY would be one of many choices. The chart is short because of how new RSSY is but do the two leveraged fund from ReturnStacked track close enough to the S&P 500 to be thought of as equity proxies? There are probably arguments on both sides and I am not saying what you should think. For you, are they close enough? I could envision a prolonged period where the equity exposure in the complex products gets offset by the more complex portion of those funds, the managed futures or carry portion of the respective funds. I think there is the possibility of kneecapping upcapture with heavy allocations to these funds. In 2016, SPY was up 12%, managed futures as measured by EBSIX was down 11% and an RSST replication using the two would have been flat which is what I mean by kneecapping upcapture.
The second chart looks at the ReturnStacked bond funds versus plain vanilla benchmark type funds, AGG and IEF which is 7-10 year treasuries. If you wanted traditional bond benchmark funds in your portfolio, I do not, but for anyone who does, are they close enough? Again that is for the end user but the first impression they are making is to be more volatile than the plain vanilla bond proxies.
There are ways to make portfolios capitally efficient that go more a long the barbelling risk and volatility into a narrower slice of the portfolio than a full 60% into an index fund as we've looked at several times before. Here is some modeling we did on August 19th.
Looking back, a 26% allocation to Blackstone (BX) with the rest in iShares Treasury Floating Rate ETF (TFLO) had a lower standard deviation than VBAIX with higher returns. Obviously betting on one stock like that won't make a lot of sense but it makes the point. BX was chosen as a proxy for private equity even though it is more of an operating company than a portfolio of private equity investments. Putting 15% each into NOC, XLY and IYW, all of which are client holdings, with the rest in TFLO Had a CAGR 170 bp higher than VBAIX and the standard deviation was 1/3 less than VBAIX. That one is a couple of steps closer to reality in terms of spreading risk more. Consumer Discretionary and Tech both tend to go up more than the S&P 500 on the way up and down more on the way down.
Regardless of how crazy or interesting you think the above two ideas are, BX, NOC, XLY and IYW are simple stock exposures. Going forward, they will either be good choices or bad choices but there's no leverage to break, there's no complexity that can malfunction, they'll just be good or bad choices (repeated for emphasis). Both portfolios would be forms of capital efficiency via barbelling as opposed to capital efficiency via leverage.
We stumbled into a way to incorporate leverage for anyone believing that is the answer while managing to avoid the multi-strategy complexity that I think still needs to prove itself in the ReturnStacked funds.
SPYB and SPYM are the new 2x S&P 500 ETFs from Tradr that we've looked a couple of times already. SPYB resets weekly, SPYM resets monthly and SSO is the long standing 2x fund that resets daily. Tradr should have a quarterly reset fund coming out on October 1st. We've looked at SSO many times and while yes, it clearly is a daily reset and could absolutely deviate when held for longer periods but it tracks fairly closely to 2x the S&P 500 for longer periods far more often than not. Is it close enough far more often than not? That is up to the end user but the Tradr funds are a work around to the daily risk.
The first portfolio is called Simpler Leverage and is built as follows.
QGMIX is a client and personal holding. The other portfolios are easily understood by how they are labeled.
The Tradr funds could be used instead of SSO once they prove themselves. Portfolio 2 is there to show how similar AGG and IEF can be. In 2022 they all performed as follows;
Despite all the work that went into thinking this through and the uptick in complexity with the leverage, the thing that mattered more than the capital efficiency was the simple decision to avoid fixed income duration. I'm not saying it was easy, but swapping out duration (AGG or IEF) in favor of floating rate involved simpler decisions that embedding leverage into the portfolio.
If you know me well, you might push back and ask why I own Standpoint Multi-Asset Fund (BLNDX), what the difference between it and the ReturnStacked suite? I don't view BLNDX as any kind of equity proxy. It is an uncorrelated return stream that is weighted in the portfolio as an alt, not a core equity exposure and it is marketed as an all-weather exposure. So sure, maybe RSST or RSSY can be held the same way but the funds are marketed as core exposures with alts added on top. Additionally, harsh comment coming, I think Standpoint is a lot better at this sort of blend than ReturnStacked is, at least so far. Maybe I will be proven wrong on that. I'm inclined to give the similar, and very new, fund from AQR with symbol QNZIX an initial benefit of the doubt believing AQR is also better at this than ReturnStacked and if I am proven wrong I will own it here.
Summing up, I think portfolios should be mostly allocated to simplicity, hedged with a little complexity.
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.
2 comments:
what do you think of DBMF and kmlm? They both seem to have a much larger track record. I feel like if you can hold DBMF and KMLM, it seems easy to see one would just modify it to include another exposure. I agree with your view on it seeming easier/cheaper to hold leveraged market exposure instead of always combining different betas, but my suspicion is line item risk is probably the reason of combining these different exposures (BLNDX CIO says the same).
At the same time I realize Corey's previous complicated products (ROMO) seemed to underperform expectations.
Hi Greg,
DBMF is a replication product, not sure about KMLM but both seem to be plenty valid for capturing managed futures. You're right about the priority given to line item risk. The ReturnStacked guys talk about that regularly and the Standpoint guys talked about that when they were spooling up in 2019 before the actual launch (i've told that story about how I know them many times).
When you look at a managed futures, single strategy fund what are you seeing versus a multi-strat fund like RSST versus plain equity exposure? No wrong answer, how do you see it?
However you view is probably your answer. Tough for me to see line item risk through the eyes of a do-it-yourselfer or another advisor but I ask clients regularly that if everything goes up together, what's likely to happen when the market goes down? They usually understand the point but I do reiterate it when asked in several different ways. Knowing ahead of time, for example, that BTAL is probably going to go down when stocks go up (except this year I guess) makes it much easier to digest. "Oh yeah, he said it would be down." Hope that helps.
Post a Comment