Permanent Portfolio 3.0

When it comes to alternative lazy portfolios, I have two main sources of inspiration: Nomadic Samuel and Roger Nusbaum.

Here Roger presents an alternative to the classic Permanent Portfolio (“PP”) that I found interesting. The concept behind the PP is simple and effective:

• 4 instruments
• equal allocations
• good protection no matter what happens in the world
• acceptable returns

This table from PortfolioCharts summarises well the issue with the OG PP, returns are lacklustre:

Roger designed a new version this way:

• momentum factor instead of stock-beta
• Natural Catastrophe bonds instead of Long Term Bonds
• CTAs instead of cash
• Gold as…gold 😉

Here are my comments regarding the instruments he chose:

• since the portfolio has “brakes” in other sleeves, you want to go full speed when stocks work. The momentum factor is one way of achieving this
• NatCat bonds provide 2 enhancements compared to plain-vanilla long bonds: they are REALLY uncorrelated and provide high yields without high duration (if you do not know what I am talking about, here is the full picture)
• I do not recall who suggested first to switch cash with CTAs/Managed Futures but, at least in some dark corners of asset management, this is a foregone conclusion. Higher REAL returns and convex protection when needed

Here are some numbers/graph to compare PP 3.0 with PP and the 60/40:

Not the broadest back test but enough to give you an idea.

The main issue with PP 3.0 is its high exposure to model risk: momentum, NatCat and MF results depend, to different degrees, on how the manager behind the fund implements them. This can be mitigated by employing different instruments in each bucket and each strategy has proven its value in both the academic and practitioners world. But it still represent a high psychological obstacle to overcome.

Plus, there is the ‘other’ known issue that no sane person has 25% of their portfolio invested in gold.

“The answer is always the same – stop trying to guess what is going to happen. Apply some of those ecological rationality heuristics that say convexity is the solution to uncertainty and intractability. As we like to put it, participate and protect. Construct investment portfolios with strong protection against unforeseen shocks, allowing active participation should asset inflation continue to be the policy-choice du jour.“

This morning I was reading the post that included the above paragraph and my mind came straight to what I was working on here. The core tenet of PP is indeed stopping to guess what is going to happen, participate and protect. While momentum and MF are convex instruments, and gold probably is?, NatCat is the opposite. Can its uncorrelated nature compensate enough for this?

I do not have in mind a good convex instrument for the low growth/low inflation quadrant. I’d probably do a mix of NatCat bonds and levered intermediate bonds (like $TYA, to get the full duration premium).

I would also add a bit of Bitcoin in the Gold bucket, to do 5% and 20% mix?, to enhance convexity there.

---

Ok, ok, ok.

I can already imagine the comments “hey but how can we get exposure to NatCat bonds?“. Here is a list, courtesy of the Rational Reminder forum, of UCITS funds:

from here

What I am reading now:

Follow me on Bluesky @nprotasoni.bsky.social