The Model Portfolio and the 2024 Teddy Bear

The first days of August have been engaging. I am unsure if many people were waiting so eagerly for some action for so long that they jumped at the first opportunity or if what happened was short but meaningful. For the right or wrong reasons, everyone had to comment. I made fun of many and yet, here I am: I would like to mention a few valuable insights related to the Model Portfolio.

A few weeks before the recent events unfolded, I read the following (from here):

“The Sharpe World standard practice of driving without brakes (no explicit risk mitigation) and targeting average lap-speeds (expected returns), leaves investors forever susceptible to the siren-song of bearishness. In that world, risk management is about driving ever more slowly in anticipation of unknown but oft-predicted future bad patches. Proper investment management is all about building effective explicit risk mitigation, well-functioning brakes, that allow you to safely put more capital to work in participating investments, aka drive faster.“

I am not sure the guys at Convex Strategies would look at the Model Portfolio and conclude it is a good mix to achieve the above-stated goal but…that’s what I’m trying to do. So, what about my brakes? Or, as someone else put it, responders?

TAIL

Long-vol strategies are first responders, like bonds: if stocks crash, they increase in value.

The main difference between the two is that long-vol is convex: big movements in equities generate even stronger gains for $TAIL; the best an investor can hope for bonds is to compensate for the loss in stocks.

This is what convex looks like.

The first type of protection that $TAIL offers me is against margin calls. On top of the leverage I get from the ETFs, I use a bit of additional margin from IBKR even if its cost is not optimal (but you know, YOLO). In this context, I lower the probability that IBKR would unwind unilaterally my portfolio.

This has nothing to do with the Model Portfolio, so back to it. To get the full benefit from long-vol strategy, we have to monetise them, ideally selling them at the top to buy stocks when they are cheap. Easier said than done. Luckily, even a dumb strategy like quarterly rebalancing works in this regard; since I am stuck in front of a screen for the best part of the day, I can try to do better than that (ahahahahah).

$TAIL broke above its 200MA on the 1st of August, ringing the first “hey, look at me” bell. On Friday, it started to go parabolic, so I refreshed the whole portfolio values but at that point any rebalancing activity was still really premature. Too late is better than too soon here because you do not want to remove the protection before the storm has gathered all its strength.

Then Monday happened, but when I tried to login into IBKR around 4.30pm CET…I couldn’t (I later discovered this was a shared issue with many US-based brokers). 4.30pm CET is early trading in the US but it is pretty fucking late for a family man in Switzerland, meaning it is August and I want to spend time with the kids outside. I can always try later, after I put the kids to bed, I thought…and then I forgot. So much for what every financial content creator out there labelled it as “the greatest catastrophe since COVID”: I forgot about it while I was in it.

On Tuesday, it was already too late. It was also when I discovered that $CAOS, another long-vol ETF I use, did this (crying face):

This is usually the moment when you think you can leave some take-profit orders on IBKR, before realising that their maintenance is too much of a burden. They do their job and if you are disciplined enough to do it, great; I am juggling too many balls ATM to add this as well… That Monday many ETFs like $GDE had a flash-crash, so there are plenty of opportunities for these “deep OTM” orders on the buy-side too. But again, maintenance (and sometimes capacity). These are once in at least a couple of years events.

DBMF

Trend following is a second responder.

If you look at the above graph, your first thought is probably “where is my diversification?”. A few trends started to reverse/whipsaw recently: short bonds in June, then the short YEN, then the long stocks.

The way trend models are designed determines how they behave in these circumstances: short-trend models react faster; models that do not include equities are less correlated to equities when bull markets die (duh).

This was sort of the perfect storm for $DBMF: the model trades only 10 markets, 3 of which are equity indexes, and it is very VERY slow to turn its positions. As it often happens, this performance arrived just after they published their rosiest monthly update.

$DBMF is built this way because it is one of the best designs to achieve long-term results. For example, if you exclude equities from your model, you cannot be short stocks when bear markets are in full display. There are trade-offs (what do you think you can buy with an 85bps management fee? :D) that, hopefully, will translate into long-term profits. $DBMF is a second responder because it pulls its weight in a portfolio when calmer and prolonged trends arise, like in 2022.

It is diversification again and again

Even in August 2024, some elements of the Model Portfolio had gains and other losses. Like in July 2024. Like in June 2024. It is the same story, again and again. Unfortunately, there will be months when everything will be red…but who do you think I am, Marco Casario Jim Simons? So far (as of 13/08/24) in Q3, the Model Portfolio lost 0.5% and it is up 11.7% for the year…imagine if I managed to monetise $TAIL ahahahah.

What I am reading now:

Follow me on Twitter @nprotasoni