We had mentioned this in passing on previous posts, but never expounded upon it here on the blog. We thought it would be too boring and down in the weeds. But the emails just keep coming (we forgot our members tend to be as analytically-minded as we are), so here are the details…
Meta is a combination of 10 tactical asset allocation (TAA) strategies tracked on this site, and is designed to reduce the “specification risk” of choosing any one particular strategy.
Each month, Meta picks 10 strategies based on the approach discussed here. All of our tests of Meta are “walked forward”, so the 10 strategies selected could change each month, but those changes are not applied to the historical track record (because that would be egregious lookahead bias).
While Meta strongly outperformed the benchmark during the downturn in March, it underperformed the average strategy on our site (read more). Not by a lot mind you (-2.8% vs -2.1%), but more than we’d goal for, and that was disappointing to us.
The reason: Because very active TAA had been pretty meh recently compared to this go-go market (where anything less than long risk to the gills has been suboptimal), the 10 strategies selected began gravitating towards less active, less skittish models. Unfortunately, this crisis called for the exact opposite approach. More active strategies that were quicker to reduce risk outperformed, and Meta was holding too “slow” a portfolio.
We expected the issue to work itself out over time naturally, but the speed this crisis came on required a little more intrusive approach. So from 03/31 onward we set some reasonable limits on how many low turnover strategies that Meta can hold at any given time in order to enforce it’s “tactical” function.
Normally that would be an issue because history wouldn’t match the strategy as it looks today, but not in this case. This was a very recent phenomenon – within the last year – that Meta began heavily-weighting these low turnover strategies. It had never previously happened in the historical record.
We would never want to appear to be cherry-picking, so we’ve only applied this change on/after 03/31. Because this was such a recent issue, even had it been in place since inception, it would have only had a significant impact on Feb/March anyways. Further, in all future analysis of Meta on this site, no matter how many strategies we add, asset allocations from end-of-month Dec/2019 to Feb/2020 (which determined returns from Jan/2020 to Mar/2020) will be frozen as they were. We wear our wounds.
In summary:
- Small common sense change to enforce the “tactical” function of Meta.
- Would have had no significant impact on historical performance other than Feb/Mar 2020.
- Allocations from the first quarter of this year are locked in all future analyses.
Members: click to see the 10 strategies that Meta is trading now.
New here?
We invite you to become a member for about a $1 a day, or take our platform for a test drive with a free limited membership. Put the industry’s best tactical asset allocation strategies to the test, combine them into your own custom portfolio, and follow them in near real-time. Not a DIY investor? There’s also a managed solution. Learn more about what we do.
