I said I would still do updates when things got interesting and the market has certainly been interesting lately.
The prior week was a good one. But there were a couple days where everything went up simultaneously. I said in a recent post, this makes me nervous and can sometimes lead to problems. Those problems showed up this week.
Today, if we round gold up a touch, all three of our assets fell a percent or more. The last time that happened was March of last year, when it happened 4 times. Stocks and bonds had been behaving this way all week, and today gold decided to join them. So needless to say, the portfolio has tightened up quickly and warrants a mid week adjustment.
This is an interesting change. The portfolio didn’t really lighten up much on stocks. Essentially, because correlations are so high, the much lower expected return on bonds doesn’t provide the same kind of boost to overall portfolio returns. Except for today, gold has been somewhat more ambivalent to bond and stocks moves, so it gets a slightly higher role in the portfolio. And because of the high correlations and the high volatility in each asset, cash comes back in to control risk levels. On May 12th the portfolio rebalanced to:
47% SPY , 22% TLT , 22% GLD , 9% Cash
See the Portfolio page for charts and tables.
Now who knows if this will continue? I certainly hope it doesn’t. But since correlations and volatility have a tendency to cluster, it certainly might.
Automatic Upates
Some have noticed that the portfolio has been updating automatically on Thursdays. The last time I posted this update was on a Thursday, and I just decided to leave it there, updating every Thursday. In theory, the day you rebalance doesn’t matter. But in practice I think it’s optimal to be in the best possible portfolio going into the weekend, as there is probably more likely to be a major surprise over the weekend. So I’m going to leave it to rebalance on Thursdays, although unless tomorrow is a mess, I’ll just it run through to next Thursday.
I also haven’t made the mid-week rebalance automatic yet. I’ll get around to it soon enough.
*All investing strategies come with the risk of loss, including this one, which is evident by the many losing weeks this year. This portfolio may not be appropriate for your investment goals and requirements, and it is not investment advice.
Before I go into the critique, let me thank you for the blog. It gave me much food for thought.
As for the critique, you criticize academics a lot when they choose convenience over accuracy, yet you yourself use a backtest without fees, spreads and taxes taken into account. This is a way more egregious error than an academic using a stock index as a proxy for the US stock market.
A system that rebalances as manically as this would have huge performance lag during live trading. Going 100% cash and back every other week during the early 90s sends shudders down my spine.
https://i2.wp.com/breakingthemarket.com/wp-content/uploads/2019/04/chart-of-portfolio-1.png?ssl=1
I’ve certainly never hidden from the fees and spreads topic. The same post you pulled that chart from has a table of returns showing the estimated trading costs as a percentage of the portfolio for the year. Those costs included both an estimated bid-ask spread and an estimated commission, and traded about every two days. Granted I used todays typical values, and not the values form 30+ years ago, but this makes more sense to me as I’m most worried about implementing this going forward.
I also discussed fees in this post https://breakingthemarket.com/the-great-age-of-rebalancing-begins/ . Trading costs for weekly rebalancing are not high, and the values I posted live before last July included trading costs(its says this in the footnote on the portfolio page about the italic lines) and come from the positions which I posted on here each week through one of the craziest markets we’ve ever seen–with very little performance lag. Fees and spreads have a very very low impact on the performance of the strategy today, and should into the future.
Now for taxes. I agree that could be an issue for some. I will ask however, do most academic studies report returns after taxes?
More importantly I didn’t create this strategy originally for use in taxable accounts as most of my wealth (and I would strongly suspect them majority of US citizens with wealth investing in the stock market) is in tax advantaged accounts. So it wasn’t really “convenient” for me to ignore taxes, it was simply a reflection of my, and many other’s, investing reality.
As Geometric Balancing has gotten more popular though this will be a headwind to its growth. The biggest problem with the taxes issue is come from the impact being unclear today. It’s never really been quantified. I firmly believe that with he correct accounting structure the taxes will be very reasonable, and I’m working very hard to try and truly calculate the tax implications of Geometric Balancing and how it best to approach it.