Model Portfolio Enhancements

A year ago I introduced The Italian Leather Sofa model portfolio. The main goal was to put my money where my mouth is…even if, back in the day, it was only virtual money. In fact, the model portfolio is built using US-listed ETFs, and EU financial regulation prevented me from buying those ETFs as a retail investor. I judged it was better to show what I would do if I had access to the best instruments on the market that provide a (very) sub-par but tradeable version of what I had in mind.

If I piqued your interest, check the origin of the model portfolio here.

After acquiring the status of Swiss taxpayer a few months ago, I was able to implement the model portfolio in real life finally. The model uses only 4 ETFs, it is simple and for a good reason: I do not want the main message to be lost in obscure details.

Running the same strategy in an actual portfolio, I want to tackle the inevitable shortcomings that come with simplicity. In the end, this is my portfolio. I am the only one who needs to understand it and manage it. It might look like a Humpty Dumpty Portfolio to some readers but I am not here to sell anything to anyone: there is redundancy but not unnecessary complexity. It is the right level of complexity for me…and I definitely do not charge myself a higher management fee on the back of that complexity.

Without further ado, let’s go with the portfolio additions.

NTSI and NTSE

These are the “siblings” of NTSX, the WisdomTree return stacked (just to piss off Corey Hoffstein) 60/40 portfolio ETF. Whereas NTSX offers investor exposure to US large cap stocks, NTSI invests in developed markets large caps and NTSE in developing markets. For all three ETFs, the bond component is US Treasuries.

In the last 15 (?) years, diversifying away from US stocks, especially large ones, resulted in a drag on returns. But here at TILS, we believe in diversification covering our asses. So there you go, I try to build a sort of World stock index, so the original 100% NTSX sleeve of the model portfolio will look like 60% NTSX, 30% NTSI and 10% NTSE.

Other Capital Efficient ETFs

As (I believe) first on the market of return stacked ETFs, WisdomTree did not stop at the stock/bond combination but launched GDE, the Efficient Gold Plus Equity Strategy Fund. The ETF offers a 90% exposure to gold plus 90% exposure to US large stocks, so there is also a bit more leverage compared to NTSX&friends.

Corey Hoffstein finally decided to capitalize on his idea of capital efficiency and recently launched RSBT, the Return Stacked Bonds & Managed Futures ETF. The fund offers 100% exposure to broad US bonds and 100% to a managed futures strategy (even moar leverage, yay!!!).

Here I probably need to insert a reminder. The model portfolio targets the following weights: 60% stocks, 40% bonds, 20% trend/managed futures/CTAs (call it as you want, I KNOW they are NOT technically synonyms) 10% commodities, 4% tail risk. My “extended”, real portfolio maintains the same targets. The addition of the above funds make it even more capital efficient and further diversify each bucket of stocks, bonds and Trend.

I allocate 5% of the unlevered portfolio to each GDE and RSBT. This way, the 10% commodity bucket has equal contributions from GDE and COM, while the 20% Trend bucket is 5% RSBT, 7.5% DBMF and 7.5% CTA.

The Trend Bucket

While the diversification benefits provided by a trend-following strategy are well documented, there is no easy, beta-like, way to achieve it comparable to stocks and bonds. I am not even 100% sure what I should call it…

DBMF might be a good proxy but it still represents DBi stab at replicating the SocGen CTA Index. There are single manager and model risks involved in employing a single ETF for this bucket. For this reason, I decided to diversify investing in RSBT and CTA. CTA is Simplify Managed Futures Strategy ETF.

Holding an ensemble of strategies and funds will likely mean the portfolio won’t be the worst (if one of the strategies underperforms) but won’t be the best either.

In the past, I have also invested in ETFs that were subsequently closed by their sponsors: this way, I already have an alternative ready and do not have to rush to find one.

KRBN

I know, I did not even finish this post and already lied to you. Using GDE and RSBT I have the opportunity to further increase the portfolio leverage from 1.34x to 1.36x (obviously this increase depends on the weights assigned to the two ETFs. The more I allocate, the higher the leverage I get).

I decided to employ this additional space to invest in KRBN, the KraneShares Global Carbon Strategy ETF. From KraneShares’ website: KRBN is benchmarked to IHS Markit’s Global Carbon Index, which offers broad coverage of cap-and-trade carbon allowances by tracking the most traded carbon credit futures contracts.

This choice is not related to responsible investing (even if I have 2 kids and would love them to have the opportunity to snowboard in Switzerland consistently as I did) but to add another source of uncorrelated returns. It is a very unconventional choice but I had this ETF under my radar for quite a while and I find the proposition interesting. 2% is a moderate amount to gamble, we will see how it goes; a valid counterargument would be that 2% is a too small allocation anyway and it is not going to move the needle. I know.

TAIL

I would love to diversify in this bucket as well but TAIL is the only ETF I liked so far. Simplify has its own version, called CYA: to my “untrained” eye, it feels a more sensitive take compared to TAIL, in the sense that it bleeds more in a calm period but it is more reactive to market corrections. Here is a comparison from PortfolioVisualizer:

CYA might emerge as the better choice (and possibly the more capital efficient as well) but considering its very short history I am not ready to gamble on it yet.

There is not even ample consensus on the fact that Systematic Put Buying strategies, like TAIL and CYA, can offer the hedging qualities they advertise. The only sure thing is their negative EV. I recently listened to a podcast (cannot recall which one) where the guest suggested combining put buying with a volatility premium harvesting strategy (something like $SVXY?) to make the overall strategy “stink less”.

If you did not understand yet, I am disoriented. Hopefully, I will convince myself to add CYA before the next market crash but I doubt about it.

ACWV

Adding RSBT to the portfolio means I have to buy a 100% stock ETF to maintain the portfolio weightings I target (unless I fckd up the math, which is also highly possible). My choice was ACWV, the iShares MSCI Global Min Vol Factor ETF. It’s been years since I am a junkie of the Min Vol factor, I have a post about it in my draft for ages and never found the time to complete it. The main reason for its inclusion is that its defensive nature combines well with the leverage in the portfolio.

One of my main regret about the model portfolio is that I can only get market-cap-weighted exposure to stocks, no factors. This way, I can better express my beliefs; maybe one day I will add as well some value and momentum tilts.

Future Developments

I do not want to mess too much with the portfolio structure but I loooove to experiment. Unfortunately, a lot of the ETFs I am interested to use are relatively new, and it is hard to impossible to back-test how they would have performed in past crises and circumstances.

I am pretty positive new ETFs in the capital efficiency, tail hedging and trend-following realms will be offered to investors in the future. Also, new uncorrelated return strategies will be available. The portfolio will closely follow those developments.

If you made it so far and are not bored to death yet, I would suggest two great podcast episodes around the topics elaborated on this post:

• here the managers of DBMF and COM, both model portfolio components, stage the closest version of a catfight you will ever get on a finance-related podcast.
• here Value Stock Geek interviews the man behind PortfolioCharts, a huge proponent of the benefits of uncorrelated assets in a portfolio context

What I am reading now:

Follow me on Twitter @nprotasoni