A Lot Of Crazy Topped Off With A Little Sanity

Some quick hits today.

Barron's had advice about how to invest a Roth IRA for people unlikely to ever need that money who might have the "goal of passing on as much they can to their heirs." The article solicited input from Harold Evensky, William Bernstein and Larry Swedroe. Most of the advice was to just buy very plain vanilla, market cap weighted funds which of course makes sense. Swedroe included some serious alts in his recommendation including Cliffwater funds. 

The article isn't really an advice piece, it is a thought exercise. Buying Vanguard Total Stock Market Index ETF (VTI) is of course valid but not really something that merits an article in Barron's. It's far more interesting as a thought exercise to do something crazy.

A 50% allocation to the Direxion 3X Long Tech ETF (TECL) is crazy but the 50% to cash tops it off with a little sanity. Sticking with Barron's objective of leaving as much as they can to an heir, this mix gives the opportunity for crazy growth while building up what at first might be an emergency fund and then later a huge stockpile of cash as it rebalances to offer real, financial utility for the heir.

Quantpedia had an interesting article noting that most portfolios are under diversified based on a portfolio of ten common ETFs that have the following correlations;

Quantpedia plugged those ten ETFs into an equal weight portfolio because equal weighting creates the appearance of being well diversified. I am paraphrasing so if you have a better take on that please leave a comment. They ran the ten ETFs through some sort of tool for risk attribution. The way to read the following is that a 10% allocation to BITO, a Bitcoin ETF, contributes 29% of the risk.

Here I model out the equal weight version above along with Finominal's risk weighting against a couple of benchmarks.

I don't actually think anyone would build a portfolio that equal weighted those ten ETFs but doing so for the backtest does support the notion of being under diversified. The portfolio appears to look different but there's no differentiation on the way down. On the way up it lags meaningfully as does the risk weighted version. Finding adequate differentiation to actually add diversification takes some work and if you've been reading this site for a while you probably have a good idea of how I like to approach solving that problem.

Playing around with quadrant inspired portfolios notwithstanding, I think equal weighting is a tough way to make a living. We've looked at a few examples in the past where it can work but I don't love it. 

Here's one we probably haven't looked at before.

The correlations are very low which creates the opportunity for diversification. Modeling those four funds out with equal weighting looks good;

It's simple work to look at different funds and how they correlate to each other as a starting point if equal weighting appeals to you. The correlation matrix above is from Portfoliovisualizer.  QDSIX is a fund of funds so it's not 25% into one strategy but that would be difficult for me. Ditto cat bonds and gold. Weighting the four funds for risk contribution yields an interesting result for anyone who want to plug that into Finominal. 

And One River took a look at work from Meketa about making a portfolio more robust with first responder defensives, second responders and diversifiers. If that seems familiar, we did the same thing almost a year and a half ago. One River titled their piece The Perfect Hedge. Great title. The paper has an element of why say in 100 words what you can say in 1000 words. I didn't take anything new or too definitive from the paper but there comments about managed futures prompted me to try a different take on an idea we've looked at before. 

Managed futures is a big part of the Meketa framework and we obviously spend a ton of time on it here too.

QMHIX is relatively volatile, ABYIX might be the least volatile managed futures fund out there and RYMFX aside from being one with a long track record is in between the two others in terms of volatility. Comparing the blue and red lines, they appear to be (almost) perfectly negatively correlated. They take very different paths to a similar result. The yellow blend blends them together for an interesting result. It has just about all of VBAIX' upside with a much smoother ride. Unfortunately, the time period is cherry picked to omit a dreadful run for managed futures in the 2010's.

Maxing out the backtest actually isn't that bad. It doesn't capture VBAIX' upside too well but it's a decent absolute return type of result, CPI plus four and a quarter. 

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.