The Risk of using Volatility as a Risk Proxy

These days, chaps are busy thinking about Xmas gifts…and I am thinking about bonds.

This post is highly unstructured, as it is a sort of stream of consciousness. I have collected many thoughts about bonds this year from reading papers, looking at portfolios and listening to podcasts. Bonds can provide a risk premium that is decently diversified but their role as risk-dampeners in a portfolio is exaggerated at best.

In short: you should have some, but not a lot.

Let’s dive into some images I have found online.

The blue line represents what you think you earn, the grey line is the (sad) reality. Ok, maybe this is not the best way to visualize what I have in mind. If I buy a bond today, its current yield represents my return GOING FORWARD. Today’s inflation represents what happened IN THE PRIOR YEAR. I think this graph shows well how quickly things can go wrong when dealing with bonds: when inflation starts to run (often, unannounced), investors lose money fast. So much for that volatility dampener, innit?

Let’s look it this way:

In the graph on the LHS, the yellow line looks quite enticing. However, when you look at the RHS, the same asset didn’t go anywhere for 80 years – 80 years during which cash performed better by any metric you consider.

Sure the last 45 years (minus 2022) have been great and are the best example on why you would want to hold some bonds…but probably not too many.

This one is nice:

from here. Long Treasury is 20-year bond.

Remember the story that over long horizons, stock performance narrows to an almost constant rate? For bonds, returns volatility doesn’t meaningfully tighten: look what happened if you invested in 1989 or 1993 – realised real returns more than halved. The range of possible outcomes is wider than for equities.

Here is a different way to depict the same concept:

Past 20 years, volatility is comparable to equities but the risk of going under doesn’t decline.

Maybe I am too influenced by this guy 🙂

The biggest bond crash ever?

Below are the long bond historical drawdowns for the past 100 years.

Nominal biggest ever…real getting close… pic.twitter.com/4Lmu1Zxn8D

— Meb Faber (@MebFaber) October 18, 2023

If you go past vol and proxy risk in terms of max drawdown, bonds are not that safe. All the analysis shown so far has a huge survivorship bias as well: Treasuries have been the safest (sigh) choice, imagine including German or Italian bonds (or Russian and Argentinian, or Microstrategy convertibl….ah no, sorry).

The funniest thing about these posts is that I write them thinking “if you are 60, you should probably not hold 60% of your savings in bonds“. While the most common conclusion is a confirmation bias for lads that hold a 50% SPY / 50% QQQ ‘diversified’ portfolio and think “see, I am right going 100% equities”.

What I am reading now:

Follow me on Bluesky @nprotasoni.bsky.social