2/ The first is this idea that traditionally constructed portfolios are based upon a series on concave payoffs.

What is a concave payoff? It's basically a payoff that collects small premiums but tends to lose big in the tail.
3/ Insurance is a very obvious example of a concave payoff. You collect a premium and then have to pay out a very large sum when the risk you are insuring materializes.
4/ One way to think of traditional asset classes is through this insurance framework.

The only rational reason to buy equities instead of U.S. Treasuries is if you expected to earn a premium for doing so.
5/ But where does that premium magically manifest from? Well, investors who do not want to bare the risk of equities will try to sell their exposure. Buyers will step in when they think the deal is fair. i.e. The market serves as a clearing house for risk.
6/ So while equity investors expect to earn the equity risk premium, that premium is payment for insuring other investors against large losses. Losses like 2008.
7/ This is true for most traditional asset classes. So when you build a strategic asset allocation, you are really building a portfolio that you expect to earn the risk-free rate plus a lot of systematic risk premiums.
8/ There are two potential problems, though. The first is that there are very few systematic risk premia to diversify across (that's one of the things that makes alternative risk premia so interesting).
9/ The second is that when risk manifests in one risk factor, it may be large enough to wipe out all the other premiums you are collecting.

(Though, risk parity folks may disagree).
10/ Imagine, for a moment, you sell fire, flood, and tornado insurance to homes around the country. Suddenly, there is a large wildfire, causing you to have to pay out significantly on your fire insurance. The premiums collected likely are not enough to offset this loss.
11/ Ignoring all the economics, that's basically one way to think about your portfolio in 2008. Huge economic shock causing large losses in equities that could not be offset by premiums collected from other systematic risk factors.
12/ So while a strategic asset allocation may be diversified from a systematic risk premia perspective, it's pretty much all-in on concave payoffs.
13/ And that's why trend-following strategies can be interesting: they exhibit convex payoffs, creating payoff diversification that can potentially offset negative skew and fat tails.

FIN.
You can follow @choffstein.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: