Reduce Bonds to Reduce Volatility

I like David Dredge’s metaphor of a portfolio like a racing car: you do not need to reduce cruise speed when you have brakes.

The 60/40 is a car moving always at the same speed, the one that would keep it on the track even on the nastiest turns (what about 2022?!? eheheheh). The car could go faster on straight portions of the route but this would mean to crash at the first turn because…remember? no brakes.

Can we find portfolio elements as brakes?

David’s preferred tool is the CBOE Eurekahedge Long Volatility Index, an index of active long-volatility managers. A paper from JP Morgan AM illustrates David’s idea well.

from here.

By employing hedge funds, the concept is to create value with alpha rather than beta. This sentence puts many readers on the defensive. Stay with me: neither I nor you have access to those managers anyway, whether we believe in alpha or not. Still, the paper has a useful point.

We want our diversifiers, the hedge funds in the paper, to have three characteristics:

• low correlation
• capital preservation in equity drawdowns
• alpha generation positive expected returns

The picture posted above illustrates two relevant conclusions in the third bar:

• by simply introducing assets that go up when stocks dive, the optimal portfolio can hold MORE stocks
• we do not care about the single diversifier return potential, we want the whole portfolio to work

In fact, the difference between the second and the third portfolios is that we are more than fine with stock returns: we do not want diversifiers to work when stocks are doing fine, we want them to pull when stocks are down.

Now a reality check.

We have to be careful going all-in with this optimisation process because parts of the above characteristics are not…set in stone. As I wrote in this post about my allocation to TAIL, CAOS and BTAL, there is a trade-off between those 3 axes: if you want to be sure about the capital preservation part, you have to give up on expected returns. We have higher certainty that things would not go as planned, as in the past, than a perfect continuation on that path.

For this reason, leaning towards diversification rather than going ‘full barbel’ with high risk and low correlation is a reasonable choice. Handle the insights with care. The main conclusion is that bonds are not as good as you thought and that using just stocks and bonds is not enough.

On page 24 of the paper, they discuss what here at TILS we already know: it is better to “walk on the capital market line than the efficient frontier”, as Delta Airlines pension fund manager once said. This means employing leverage on the highest Sharpe portfolio instead of taking more risk by going right on the efficient frontier.

Going back to Mr Dredge, we know that unfortunately there is no equivalent to long volatility hedge funds available to retail investors. But we can get creative:

from here.

This paper shows the benefits of using factors along standard stock and bond betas. The authors build a portfolio, called Lowvol+, which first screens for low vol and then for value (as net payout yield) and momentum. The formula is not complicated and can be run on a GoogleSheet but it requires you to run a 100 stocks portfolio out of a 1k stocks DB. Probably you have something better to do in your spare time…

That said, the concept can be used with off-the-shelf factor funds: each factor has <1 correlation to the stock benchmark and >1 correlation with bonds. By adding factors to your portfolio, you can reduce the bond sleeve and improve the ensemble Sharpe. It is implied leverage for investors that cannot digest the word ‘leverage’.

What I am reading now:

Follow me on Bluesky @nprotasoni.bsky.social