Deconstructing The Cockroach Portfolio (Again)

Last year we looked a couple of times at something called the Cockroach Portfolio created by Jason Buck. Buck got interviewed recently by blogger Nomadic Samuel and there were a couple of interesting points that we may have touched on earlier and I think there a couple of more thoughts we can add to our discussion of the idea. 

It seemed like it was inspired by Harry Browne's Permanent Portfolio which they talked about in the current interview. Browne equal weighted stocks, long bonds, cash and gold with the idea that no matter what, at least one of the four would be doing well. With long bonds doing so well for so long, that at least two of the four did well creating a great looking backtest. If my ideas about bonds going forward actually turn out to be correct, then the Permanent Portfolio will not be able to repeat its past performance. 

In that light, Buck exploring ways to modernize Browne makes sense. Obviously he's built a business on what he believes is a superior version but if you play around with this on something like portfolio visualizer and try to stick close to the exposures and weightings, it really isn't going to perform like the the typical 60/40 portfolio. It is intended to be more all-weather than that which means up less in equity bull markets and hopefully down a lot less during equity bear markets, maybe even up a little when stocks drop.

This is a simplified version of Cockroach equivalent of stocks/long bonds/gold and cash.

And this is the more nuanced version that slices up each of the four.

One important point to make is that I am quite certain that Buck doesn't build this portfolio using exchange traded products or mutual funds. There might be some funds in there but he uses managers for these slices. He monitors a huge universe of managers and allocates accordingly.

Building the stock sleeve is the least interesting part of this. Pick whatever broad-based fund you want, I'll just go with the Vanguard S&P 500 ETF (VOO). The income tranche is a little more challenging or maybe not. I do not want intermediate or long term bonds, I don't want that volatility. Going very simple, we could just use the SPDR 1-3 Month Treasury Bill ETF (BIL). Or we could go more complex with the AQR Multi-Asset Fund (AQRIX) which is actually a risk parity fund. I know I've crapped all over other risk parity funds this one might actually be a truer manifestation of the strategy that might actually work.

In 2022 AQRIX did a little better than most bond funds but it didn't completely avoid the carnage either. It has the potential to have big up years every now and then. That potential to go up a lot combined with what appears to be a low correlation to equities could help the long term performance of this portfolio and with only 25% in equities, it might need the help. If we used BIL, there's be no movement for better or for worse and we'd collect 5% for a while. I don't think T-bill rates going back to zero can happen but would we want to make a change if BIL's yield fell to 2%?

Volatility is the most interesting allocation to me. If you go with a long volatility strategy, like maybe a VIX product, that has a negative correlation to equities you'd pretty much offset any benefit from the 25% you put in equities. Client and personal holding AGFiQ US Market Neutral Anti Beta ETF (BTAL) is arguably a play on volatility, I think it counts anyway. Putting 25% in BTAL, 25% in VOO and 50% in BIL has a CAGR less than four and standard deviation of three. It looks like a horizontal line that tilts upward. 

As we've talked about a few times buying puts in a tail risk strategy is tough to make work in a fund for now and even if it weren't, 25% is way to much for me to want to put into tail risk. For this exercise then, instead of buying volatility, we'll sell it. There are some funds that do this and I imagine there will be more coming. I considered two funds, the WisdomTree Putwrite Fund (PUTW) and client/personal holding Princeton Premium Income Fund (PPFAX/PPFIX). PUTW sells puts that are much closer to the money than PPFAX which makes the former more vulnerable to market crashes. PUTW fell 20% during the Pandemic crash in the spring of 2020 versus about 10% for PPFAX. Backtesting the two against each other only goes back to 2017 but PUTW has twice the standard deviation of PPFAX with a slightly lower CAGR. That might be skewed by 2022 when PUTW was down 10% and PPFAX was up 1.44%.

And lastly, trend or managed futures. I've talked countless times about why 25% is way to much IMO to put into managed futures but for purposes of being true to trying to replicate Jason's fund simply, that's what we'll do. 

And the results.

Some things stand out in the results but I would note that none of the three was always the best. Portfolio 1 came out ahead but it lagged for years while bonds were still going up in price. 

With a nod to yesterday's post about models, there are likely skews galore in my model related to the short time available to study, the possibility that 2022 was an anomaly that favors alts and I would expect my Cockroach to lag pretty far behind a portfolio that had a "normal" allocation to equities. The defensive nature of my version of the Cockroach might mean it would catch up when equities go down a lot but who knows?

The influence here of adding volatility and trend as asset classes can add a lot of value, I did it a while ago, but 25% allocations just doesn't really make sense, it's too much.

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.