Is Bridgewater’s risk parity toast?

News from Dealbreaker that Ray Dalio of Bridgewater was waxing pompously about evolution got me thinking about how natural selection actually works.  Contra Dalio, there is no design or purpose or improvement. Rather, species are adapted to the current environment until it changes and then they are not.

Given Dalio’s much ballyhooed culture of radical transparency and honesty, I was surprised nobody on the staff down at Bridgewater had the guts to tell him how childish his take on Darwinism is.

And then I got to thinking if I could be a bit of a dick about this.  Maybe Ray Dalio is himself not really superior but is merely a mutant who happened to be selected by the environment of the past thirty years. That would be a fun thought, and so I am going to take some liberties here to see if I can make it work.

Consider the case of risk parity as applied in Bridgewaters’ all-weather fund. According to the firm’s own description of the strategy, the fundamental premise is that “betas”, by which they mean excess returns, tend to rise or accumulate over time. For them, that is one of the few iron laws of finance on which one can rely.

The trick, then, is to find a set of assets classes with “betas” that accumulate over time and are weakly or negatively correlated with one another in the short run. That would give you a portfolio with excess returns and low vol, the holy grail of finance.  If I understand the strategy correctly, they do this with equities, nominal bonds, real bonds, and commodities.  Perhaps there is some cash in there too. I am not sure.


I don’t know anything about commodities, but I do know that the environment of the past thirty years has been extremely friendly to a species of investor that would assume “betas” in equities, nominal bonds and TIPS would just accumulate over time. That premise is false. Some “betas” go to -100% as the asset class gets eliminated. But I am talking here about the effect over the past 30 years of having that mutant  premise.

Consider the charts above, noting that I show a hodge podge of time periods, based in part on data availability and in part on the points I am trying to raise.  You can use different time periods if you want. I am not trying to pull the wool over your eyes. And nor am I being remotely authoritative. I am pretty new to even thinking about this. And I fully admit to wanting to hoist Dalio on his own petard, given how pompous his Principles are.

Since 1986, when the all-weather strategy got going, even Shillerized earnings have grown at a rate of about 5 ½%.  In the current inflation environment, assuming as I do that the Fed hits its inflation target, that would work out to about 3 1/2% real.  Starting from here, that would be a rather aggressive expectation for earnings growth.

Meanwhile, the Shillerized earnings yield in in the S&P500 is a full standard deviation below its long-term average. So it seems almost inevitable that equity returns, real and nominal, are going to be far below their 30-year average over the coming years.  I do not believe that equities are overvalued, but I do believe they are fairly priced to deliver the crap returns consistent with slow but non-volatile growth.  One man’s view.

The same is more than fully so for the TIPS returns and the real component of conventional bonds.  Bond returns have shot the lights out in the past 30 years, in part because the real yield in 10s has dropped from over 3 ½% at the end of 1985 to less than ½% now.  TIPS yields have also come down.  We have a shorter run of data for TIPS yields than for my proxy of the real yields in conventional bonds. But as TIPS liquidity has improved, they have actually outperformed proxies of real yields in the conventional bond market.

Finally, inflation expectations implied by the nominal – TIPS yield comparison have fallen from far above the spread consistent with the Fed’s inflation objective to well below. Some of this decline would double count the point I made about TIPS liquidity. But the point is that the inflation breakeven now looks awfully low.  Moreover, a decline from here would probably be negative equities, challenging the correlation premise of risk parity, at least as I understand it to be applied at Bridgewater.

I have no idea how commodities will do or how much they have contributed to risk parity strategies in recent years.

But my point is simple. Just being long everything worked great in the environment of the past 30 years. And it seems an extremely dubious strategy looking forward. So be careful about your support for Darwin. He may be coming for you — or at least for your main strategy.  Maybe Dalio will adapt and come up with a different strategy. That would stick it to me!

There is another aspect of risk parity that people occasionally like to talk about. Apparently, the strategy gets more aggressive as vol declines. This can set up self-reinforcing dynamics in the markets if risk parity is sufficiently popular. I have no idea about that.  This post is just about likely average returns and biology 100.

This seems like something it would be good to learn more about, so I will try.  The post above is intentionally provocative, almost Business Insiderish in its click baitiness.