macrocephalopod

so a big part is narrowing down how you implement risk-contributions. the most common mistake is levering up TRS on AGG vs levering up discrete exposures to duration, linkers, corp-credit. good implementations are more granular so you can effectively tweak exposures to specifics

this makes a lot of sense - will experiment around. thank you for sharing!

its my pleasure. risk-parity breaks down to being agnostic about risk factors and weighing them equally (tautology is useful sometimes) so the more granular you get the better you can express it. thats all there is to it really. at least in my opinion.

A few thoughts I have about risk parity (disclaimer, I don’t manage a risk parity strat except for a very simple one in my PA so I am likely missing some things).

The hardest bit to get right is inflation exposure. It’s tempting to ignore this since you haven’t needed it for the last 30 years and stocks + levered bonds have done really well. But look at the performance of stocks + levered bonds from 1970-1985 to see how bad it could get.

You get inflation exposure two ways — inflation linked assets (TIPS, gilt linkers, inflation swaps) and commodities. TIPS/linkers should in theory do we in a risk off environment (diversifying stocks) but in practice it’s trickier than that since risk off environments can be

deflationary, and also TIPS/linkers have an illiquidity premium and tend to drop in price when credit is tight (eg 2008). Underestimating how much TIPS would sell off in a crisis is why Bridgewater All Weather did so badly in 2008. I would be tempted to use inflation swaps to get

the required inflation exposure instead. Commodities also give inflation exposure but they are not supposed to be investment assets, and a simple long-only investment in commodities is not expected to perform well over the long term. Some commodities (eg natural gas) even have a