It's been a while since we bagged on risk parity but Bloomberg gave us a good prompt to revisit the strategy. Apparently a few state pensions and similar pools of capital have been pulling money from risk parity funds managed by the likes of Bridgewater, Man and others.
The simplistic definition is that risk parity equal weights asset classes by their volatility. Where bonds, generically are less volatile than equities, a risk parity fund would have more exposure to bonds to get the volatility contribution of both to be equal. This often involves using leverage to get the bond allocation large enough for its volatility contribution to equal that of equities. There can be other things in there too like commodities and there can also be management of the weightings to account for changes the volatility profiles of the various asset classes. The RPAR Risk Parity ETF (RPAR) is indexed, the Fidelity Risk Parity Fund (FAPSX) is actively managed as two examples.
I always say the same thing about risk parity, it is intellectually appealing and I want it to work which I concede is silly, but it just doesn't, at least not in retail accessible funds and based on the Bloomberg article, maybe not in hedge funds either. Or maybe they are doing it wrong.
At times in the Bloomberg article they seemed to use risk parity and All-Weather interchangeably, attributing both to Ray Dalio. The following compares All-Weather. a home made risk parity replication that copies the target allocation, which is leveraged, of the Risk Parity ETF (RPAR) and plain vanilla 60/40.
RPAR has only been around since 2019 but the replication allows us to go back to 2008. It isn't necessarily a bad thing that risk parity lagged but it did so with a higher standard deviation.
The year by year for risk parity replication shows quite a few things. It was down a little less in 2008 and 2022. There were a couple great years, a couple good years and maybe a half a dozen years where it lagged by a lot.
Both All-Weather and risk parity don't keep up with simpler broad market proxies and risk parity doesn't lower the volatility, All-Weather does have a a lower standard deviation though. It is possible that they are both too clever by half. We've looked at countless ways to build around a normal weighting to very simple core exposures, equities, with small slices to complex strategies to try to reduce overall volatility in what I think is a similar way to what All-Weather and risk parity have in mind.
An update on something I mentioned in passing a few months ago, the CBOE S&P 500 Dispersion which has symbol DSPX. Google Finance recognizes the symbol and you can do some things like chart comparisons. Yahoo recognizes it but the charting doesn't work. You can also do some things on the CBOE site too.
When I wrote about it before I thought it captured some sort of put/call skew but that was incorrect. Basically it tracks when stocks are more likely to deviate away from the performance of the S&P 500 or less likely to deviate away from the performance S&P 500. By their work, the dispersion tends to increase going into earnings season and then come down some after earnings season. The process to derive the Dispersion Index is kind of similar as the process to derive VIX but the information is much different.
The plan is to create a futures contract based on DSPX in Q1 2025 which could then be a path to some sort of exchange traded product. CBOE, a client holding, among others things has about 19% of ETF listings.
On Google, it only goes back to last fall but I compared to a bunch of liquid alternative funds and it doesn't look like anything. I think it is uncorrelated to everything but we'll see how that proves out. If so then it becomes a way to harness volatility as an asset class and uncorrelated return stream which makes for good diversification when sized correctly.
Per the above backtest, DSPX went up during market declines including a massive spike during the 2020 Pandemic Crash. It briefly went to zero in 2018 and then came right back. They would need to address what the index going to zero would mean for a fund. If there was some number of shares and the index went to zero but shares still existed then when the index came back, the shares would have value again. Maybe it would be that simple? I listened to a webinar about it and the way I understood it, it can touch zero but it cannot stay at zero. Based on that understanding, if it touched zero after a fund launch, it should probably be bought hand over fist but I'd still have some learning to do to be comfortable with that idea.
DSPX is interesting on first and second glance and I am getting an early start learning about it.
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.
No comments:
Post a Comment