“Dynamic Bond” Variations of 10 Tactical Asset Allocation Strategies

Background (we’ve covered all of this before, but just to bring readers up to date):

• We track 70+ Tactical Asset Allocation strategies sourced from research papers, books, etc.
  A subset of those strategies have a significant flaw: when shifting to a defensive risk posture, they blindly dump the portfolio into bonds (mostly US Treasuries) regardless of how bonds are currently performing.
• This flaw was exposed in 2022 when risk assets and bonds fell simultaneously to a degree not seen in nearly 100 years (read more). As one would expect, strategies with this flaw performed poorly because they were trapped in bonds all the way down.
• Since launching this platform, we’ve tried to capture strategies’ exposure to this risk via our Exposure to Rising Rates report, in which we account for whether the strategy has this “dumping to bonds” flaw.
  But we want to do more. These flawed strategies may otherwise be good strategies, and we want members to be able to still find value in them.

So, how do we proceed? We don’t remove strategies from the platform – that would introduce survivorship bias. And we don’t modify existing strategies – that would introduce hindsight bias. We track strategies as closely as possible to the author’s original design, within some basic standards.

Our solution:

We’ve identified 10 strategies with this flaw of blindly dumping the portfolio into bonds. Those 10 strategies are as follows:

We’ve created a second variation of each of these strategies. On the list of All Strategies, you’ll find the author’s original, as well as a second variation that ends in the words [Dynamic Bond].

This Dynamic Bond variation follows all the same rules as the original, with one exception. Whatever momentum rule the strategy uses to determine whether to hold risk assets, we apply that same rule to defensive assets. If the strategy shifts to defense, but defensive assets fail the momentum rule, the strategy will instead allocate to cash.

Notice what we did not do. We did not create a universal bond rule for all 10 strategies (because that would be imposing our own beliefs, and we would lose diversification between strategies). And we did not try to find an “optimal” bond rule for each individual strategy (because that would likely lead to overfitting). We simply applied the author’s same thinking for risk assets to defensive assets.

Further thoughts:

• All of these Dynamic Bond variations score significantly better on the Exposure to Rising Rates report than the original. That makes sense given that, well, they have less exposure to rising interest rates.
• All of these Dynamic Bond variations exhibit strong correlation to the original. That also makes sense as these are still fundamentally the same strategies, with the exception that they will sometimes move to cash rather than longer duration bonds.
• We selected these 10 strategies based purely on the rules of the strategy. We looked for all strategies that used an “absolute” momentum rule for risk assets (ex. requiring momentum to be positive), but did not use an absolute momentum rule for defensive assets.
  We did not consider how they performed in 2022. Most of the Dynamic Bond variations would have performed much better in 2022, but in a couple of cases, they were still terrible. There’s nothing we can do about that; a bad strategy is still going to be a bad strategy.
  Setting aside 2022, some of these Dynamic Bond variations performed worse over the rest of the backtest. That’s unsurprising. There’s a reason why this flaw existed; most of the time, blindly assuming bonds would be a reliable defensive asset has been a winning bet.
• There are strategies with high exposure to rising rates that didn’t get a Dynamic Bond variation; it just wasn’t possible without fundamentally changing the strategy. Examples:
  • If the strategy only considers “relative momentum”, not “absolute momentum”. For example, a strategy that picks assets with the highest momentum, but doesn’t require that momentum to be positive. We can’t add a rule that bond momentum be positive without fundamentally changing the strategy.
    Example: Lewis Glenn’s Paired Switching
  • If the strategy trades based on something other than momentum. We can’t add a rule that momentum be positive to a non-momentum strategy without fundamentally changing the strategy.
    Example: Risk Premium Value

We think we hit the sweet spot of providing a fix to an obvious flaw in these affected strategies, without (a) going too far against of our site ethos of tracking strategies as close to the author’s original intent as possible, and without (b) introducing unnecessary survivorship or lookahead bias.

Check out these Dynamic Bond variations now in the members area.

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 membership. Put the industry’s best Tactical Asset Allocation strategies to the test, combine them into your own custom portfolio, and follow them in real-time. Learn more about what we do.