The Hedgie in Winter

Institutional investors want consistency more than they want returns. Career risk is a bigger concern than your retirement balance.

In turn, hedge fund traders are paid for returns, but fired for volatility. Firms like Point72 literally have a "down and out" clause in their contract: if you have a certain percent drawdown you will be fired. Internally, hedge funds backtest against measures such as IR (information ratio) and not against returns. IR penalizes for volatility and rewards consistency.

https://www.investopedia.com/terms/i/informationratio.asp

Seems like a slightly simplified study but I think it does highlight a pervasive problem. There's too much money chasing increasingly limited strategies. Things like start-arb and momentum used to be pretty new and relatively obscure 10 years ago but now your every-day retail trader has probably heard about them and your friendly neighbourhood data-service provider includes all sorts of factor indices as a package.

I used to be a trader at a bank, the entire field/discipline of trading is a gross distortion in terms of the work they do and what they are compensated. The increasing competition sort of makes it even more twisted, the amount of effort and technology being used to squeeze out returns and enrichen these fund managers doesn't really make sense to me.

I think this post (not very objective or scientific to be citing quora but I think it gives good insight) sort of shows the kind of brain drain finance has on society.

https://www.quora.com/Do-hedge-funds-use-the-generalized-met...

These are really smart people staring at mostly random price data all day and applying increasingly obscure techniques to eek out a signal just so they can squeeze some money out of the guy at the other end. Why????

For context, the title is a reference to the play, "The Lion in Winter"

> Nobody has ever marketed a fund or asset class with the tag line “we’ll likely tread water net, which is actually pretty impressive after we take out our massive fees.” But it’s not nearly the disaster that’s so often claimed.

So it's not theft or fraud, just merely a wealth transfer from the investors to the managers.

The article is advocating that the only appropriate counterfactual investment to compare to a hedge fund investment would be a long-only equity portfolio constructed to bear the same exposure to the overall market as the hedgefund portfolio (this exposure is called beta), and later extends this to a regression that controls not only for the level of beta but also the level of the same old dumb shit Fama-French factors (well, just SMB) that people keep head-in-the-sand believing are important.

The regression analyses are sort of just technobabble. They very much fall absolutely into the class of so-called “garbage can regressions” — assuming a linear link function is roughly physically realistic, then just dump shit in on the right hand side and assume, unilaterally and for no good reason, that summary statistics like t-stats & r-squared faithfully indicate confidence in the interpretability of the fitted model, without any attempt at accounting for non-linearity, interaction terms, or random effects (situations where the coefficient ought to vary by different subgroups due to stratification inherent to the data).

Even for comically simple cases, you can show that when these aspects are unaccounted for you can get effect sizes with the wrong sign that are simultaneously hugely statistically significant, see [0].

Active managers use this simplistic regression technobabble to slice and dice datasets, cherry pick time periods or asset groups or quantiles of exposures to any number of conveyor belt covariates from accounting data, news data, whatever (and, hilariously, all of them are using the same third-party data vendors and constructing the same models with the same factors -- so the modeling is surely not any form of differentiator or comparative advantage... cough marketing cough).

And, just like here, they work extremely hard on arguing disingenuously for what baseline they're supposed to be measured against.

But the money quote is:

> “Traditional active funds do much worse than hedge funds over the full period but, unlike hedge funds, they don’t deteriorate in the second half of the sample (they just continue to consistently underperform both the market and hedge funds but at their normal pace of underperformance).”

So, given this, why shouldn’t someone consider a broad market index as the correct counterfactual investment. If it’s obvious (and I think by now it is) that active asset managers & mutual funds just consistently underperform while also taking active management fees, why wouldn’t someone view their counterfactual option as a super low fee ETF that tracks a broad equity index?

What matters is what I could have invested in. Whether that is my idiot nephew’s idea for flying toaster drones or a sophisticated hedge fund product. And the more you tell me that a class of investments defined by characteristic exposures to e.g. the market index or SMB consistently underperforms, the more reason I have to believe there’s no skill there, and to question whether my counterfactual investment really needs to have a constraint on particular exposure to these characteristics— maybe it just means to the extent that hedge funds look like low-beta strategies, the hedge funds are stupid, and I shouldn’t have been on the market for the hedge fund’s advertised characteristics in the first place.

It’s actually kind of amazing that Cliff Asness can write a head-smacking quote like that right on the AQR website and probably be rewarded for it with a bunch of protect-my-ass board members from pension funds voting to give AQR more money for telling them that AQR itself is systematically unskilled at this.

What a world.

[0]: < http://www.columbia.edu/~gjw10/achen04.pdf >