Hacker News Reader

Comment by startupdiscuss | Hacker News Reader

Some of the comments here by people who think that he got spectacularly lucky one year and might be slowly bleeding out might be missing a subtlety.

There is a perfectly rational strategy whereby you lose a little bit of money every year if you think that the catastrophic tail event is more likely than other people think and so the risk is mis-priced.

One finance professor described it as a "100 year flood that happens every 7 years."

Taleb believes that this is true of markets and he may very well have a similar strategy (consistently buy out of the money puts).

We won't know till a certain amount of time has passed if this strategy is genius or the opposite.

View on HN

That's true up to a point - but there are also ways to mitigate against those losses without completely giving up all the alpha you can get from betting on extreme tail events.

I'm sure there are some fund managers who do follow this strategy - but they also mitigate the risks associated with it, which means their extreme gains are not as extreme as Paulson's, but their losses are not as extreme either.

We hear about Paulson because he's an edge case. We don't hear about the funds that won big, but not quite as big as Paulson but who have had much better risk management practices.

This is the equivalent of saying you should never lose money. Not sure there is a fund that has a strategy that is a perfect oracle.

The only funds that "never lose money" are the ones us mere peons can't invest in.

Eg: quant-driven funds that try to buy/sell the edge then profit the minimum movement 24/7, if those are losing money then they cease to exist - and they haven't done that yet. The minimum investment in those funds is high 7 figures though, and that's out of at least my price range for risk capital(and non-risk capital for that matter).

There is only one such fund that has survived over a long period of time and that is RenTec. As far as I'm aware, there aren't many others.

And my understanding is that no one can invest in RenTec anymore either because they can't scale the strategy.

>We hear about Paulson because he's an edge case. We don't hear about the funds that won big, but not quite as big as Paulson but who have had much better risk management practices.

At least we know a bit about their average performance, which has been very poor over the past 5 years:

https://www.hedgefundresearch.com/family-indices/hfri

Agreed but I don't think that Paulson ever positioned these as tail risk funds. And his losses aren't small - losing 18-20% per year will have you out of money quick.

That is true.

However, he doesn't have to consciously think of himself as a tail risk fund for an investor to use him as an instrument.

The investor has to believe that risk is mispriced. Paulson is just a mechanism for expressing a short view.

Why not just put it in an index fund? Because Paulson has access to different kinds of shorts (CDS, convincing banks to allow him to short housing and so on).

I have no answer for 18-20%. That is, as you pointed out, unsustainable.

Hard to make bets via CDS these days given the central clearing (and thus collateral requirements) regulations in Dodd-Frank.

Sure, but tail risk funds are a specific "thing" - you're generally doing exactly what you described - selling out of the money puts on a constant basis in the event of a tail event, knowing you are going to lose that money 99% of the time. Betting on gold is certainly a bet like insurance, but it's not a tail risk. There's even an ETF for this now: TAIL

I think you mean buying OoM puts, not selling them.

Yup - thanks for noting the typo!

It's a lot easier to lose other peoples money when you get a bit cut of the upside.

You've gotta win pretty big to reverse out a 26% loss followed by a 14% loss. That's 36% over two years.

What not enough people consider is the null hypothesis: that he never had any skill in the first place and just got lucky. There's acres of research supporting this.

the hole is a bit steeper when you consider the sp500 was up 1.38% in 2015 and 11.74% in 2016.

so -36% vs +13.12% baseline (didnt bother to compound)

This kind of strategy is also useful for taxes as well, because losses carry over to following years but profits don't.

So, 4 years of losses and 1 year of profits could be better than 5 years of even (but smaller) gains.

> Taleb believes that this is true of markets and he may very well have a similar strategy (consistently buy out of the money puts).

im not sure that statement is an accurate description of Taleb's barbell strategy; for one thing it ignores one half of the barbell

There are other tail-risk strategies: One can have the majority of assets in treasuries but have small slices in diversified, focused portions of the market with higher expected returns (Small / Value is the typical tail-risk portfolio).

This is also known as a positive skew strategy. You gotta have some real justification to follow it.

Reply on news.ycombinator.com