In Saturday's post we touched on all-weather portfolios and assumed a lot of overlap with the Permanent Portfolio (PP). All-weather is a term typically associated with Ray Dalio and PP is typically associated with Harry Browne. Both seek out diversification that hopefully results in a robust result regardless of prevailing market events.
All-weather and PP are in Portfoliovisualizer's dropdown choices for portfolios to study. Looking at a reasonably long time frame, All-weather and PP get left noticeably behind 60/40 but not dramatically so IMO. The standard deviation of both is quite a bit less than 60/40. For 10 years, 60/40 compounded at 7.81% versus 5.20% for Dalio and 5.29% for PP.
The arrows highlight years that would be been challenging to hold. Also you can see that both did fantastically well in 2008 but in 2022 Dalio did worse and PP was down a little less because they are both heavy in bonds which got pasted. If you compare both to 60/40 starting in 2009, stripping out the great year of 2008, 60/40 compounded at 9.54, All-weather at 6.08% and PP at 5.96%. The results, in nominal terms can work, I am not being critical, but going this route will create periods of misery. In 2013, both of them missed out on a very good year and then they've each taken turns completely missing out in other years. If my thesis about bonds' unreliably volatility and less diversification benefits from bonds turns out to be correct, then Dalio All-weather and PP might struggle more often in the future.
Just because bond-heavy might not offer as much protection in the future doesn't mean the high level objective of robust results regardless of prevailing market events or a portfolio in the direction of 75/50 can't be achieved. I think it can.
Apparently Charlie Munger thought that diversification could be had with just three stocks. A common type of portfolio in the Boglehead realm is built on three mutual funds; domestic equity, foreign equity and bonds. With that inspiration, here is a three fund portfolio that back tests with all-weatherish attributes.
Keep in mind that in the period available to study, managed futures did poorly. In the ten full years in the study, EBSIX was down in four of them and up less than 5% in two other years. The period isn't that long but a lot happened with the Pandemic Crash where our attempt at all-weather did much better and the worst year was only -0.88% versus -17.06. The CAGR is competitive and the standard deviation is much lower. Included in the results for 60/40 is significant outperformance in 2023.
VOO is plain vanilla equities. I'm not worried about a malfunction there as opposed to the occasional very large decline. Merger arbitrage could have some sort of problem I suppose but the under the symbol MERFX, which goes back to 1990, the worst year for the fund was 2002 when it dropped 5.67% but during that year it did go down 14% before recovering most of that decline before the year ended. Looking at the Societe Generale website, I can't find a period where equities and managed futures were both down a lot. Managed futures dropped mid teens a few times but not when equities were down a lot. There's a handy slide-tool that starts in 2000 that you can look at. That can only be taken as a proxy but it's better than nothing.
In 2002, the year that both equities and merger arbitrage did poorly, the SG Trend Indicator was up 28%. Trend and equities would have just about canceled each other out, Trend was up a little more than the S&P 500 was down so factoring in MERFX' decline, maybe the whole mix was down about 5% which looks like would have been the worst year but click through and play around with it and let me know if you see it differently.
I have to say that when I had the idea for this post, I didn't expect to find something that would backtest that well. The idea was to bundle plain vanilla equity, something that would trade with very little volatility like how I think people want their bonds to trade and managed futures. The managed futures sleeve was about capturing something that is mostly negatively correlated to equities but that can still go up.
In the real world, I would be more comfortable splitting the 40% MERIX sleeve into multiple, unrelated strategies that trade the same way as MERIX. They're out there and we've discussed them plenty here. I also would not consider 30% in managed futures. While I have unyielding faith that the strategy does work over the long term it is pretty clear to me that there are times where the holding does make you want to puke as Jason Buck said. Puking on a 5% weighting is nothing like puking on a 30% weighting.
I was going to point this post toward "we probably don't want all-weather even if we think we do." Portfolios I manage absolutely have all-weatherish attributes but maybe we do want all-weather. If you think you do, ok but I would encourage using more than three or four funds. I think you could get away with one broad based equity fund for that sleeve keeping in mind that at some point, small cap and foreign will again outperform for an extended time. Obviously I believe in splitting up the negative correlation sleeve and the low volatility sleeve into multiple exposures that take different risks.
I labeled the portfolio with the number 1, I'll try to work on others and we can compare them in future posts.
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