Forums  > Basics  > Tail Risk Strats  
Page 1 of 1
Display using:  


Total Posts: 78
Joined: May 2010
Posted: 2020-04-30 21:32
You may have come across this back and forth between Meng and Taleb, where the former, CIO of Calpers, had apparently divested from Universa at the worst time possible saying that tail risk hedging basically is not optimal for a large money manager. Taleb shot a 3 minute video refusing this. Now MSCI reports that "unless they could have been timed to go into effect just before a crisis, the protection they provided was more than offset by their cost."

Does having an exposure to Tail Risk strategies make sense for a large, long term oriented asset managers?



Total Posts: 1075
Joined: May 2004
Posted: 2020-05-01 00:37
There are tail risk strategies and tail risk strategies... Simply buying otm puts or VIX futures is horribly expensive long-term, which is why Universa has a more elaborate strategy of selling some options to fund the purchase of otm puts, plus a bunch of trading / liquidity provision to reduce the bleed. Still, the nature of the beast is to bleed money for years before the eventual big payday, and that is hard to stomach for most people.

The argument for Universa is that it will cut the drawdowns of an SPX portfolio and increase CAGR without using leverage. When SPX crashes, the Universa hedge pays off, allowing the investor to rebalance into SPX when it is cheap. More info here

Now I have to wear my flame-retardant suit for the Taleb-hating brigade Smiley

"Earth: some bacteria and basic life forms, no sign of intelligent life" (Message from a type III civilization probe sent to the solar system circa 2016)


Total Posts: 437
Joined: Jan 2015
Posted: 2020-05-06 02:12
I'll bite as someone who's generally skeptical of tail-risk strategies. In particular, here's what bedevils me about the tail risk thesis in general.

Let's say Universa does have some sort of secret sauce in terms of gaining long-vol exposure and reducing the bleed with sophisticated trading. Why couldn't they simply transmute that to a vanilla alpha product, by overlaying a standard short-vol hedge on top of their sophisticated tail-risk core?

The value proposition is they know how to cheaply buy vol that pay off similarly to OTM index puts in the case of a major drawdown. So isn't the next step to take that long-vol portfolio and short the equivalent in VX futures and collect the carry? Now instead of just avoiding the bleed, they'll be collecting the rich variance premium every month (while only paying the cheap premium on their core tail-risk position). This provides the high-Sharpe regular income of selling puts in good times, while avoiding blowups that normally accompany that strategy.

That seems like a much easier product to market than a tail-risk fund that pays off once a decade. At the very least there's no reason they couldn't sell *both* products. But the fact that they don't makes me suspicious that they're not doing anything much different than vanilla OTM put buying.

To caveat, I'm not a vol-guy and this is a very off-the-cuff analysis. So, I'm assuming I'm definitely wrong about something here.

Good questions outrank easy answers. -Paul Samuelson


Total Posts: 1271
Joined: Feb 2005
Posted: 2020-05-06 03:10

An example would be guys that supposedly have alpha being long value stocks, and then dabbles in long short using their edge in selecting value plays. Countless tried, and only a handful succeeded to say they are statistically significant which they probably still aren't. Just because you are good at buying doesn't always mean you are good at selling.

But I am with you actually, I am skeptical but for other reasons such as crossing too much bid offer, there aren't many blow up scenarios where you can stress test/see diversity, and using historicals in terms of their asset allocation weightings for someone like Calpers to buy into them. Not to mention the secret sauce of selling some vol to bleed less vs other tail risk peers. This type of alpha is very hard to measure in the grand scheme of things compared to hiring vol traders in house to do something similar.

Dilbert: Why does it seem as though I am the only honest guy on earth? Dogbert: Your type tends not to reproduce.


Total Posts: 103
Joined: Apr 2018
Posted: 2020-05-06 06:59
I'm under the impression that Taleb doesn't believe in the variance premium in the first place. In fact I seem to recall he didn't even believe in the equity premium. The argument probably goes something like: the risk premia are inferred from historical data, but tail risk events cannot be estimated from history. Traditional concepts like alpha, sharpe, correlations don't really apply here. A tail risk event either happens or it doesn't. It's meaningless to talk about the probability of the tail risk since it can't be measured.

IMO, the question of whether you should invest in a tail-risk fund is largely philosophical. You may sleep a little better. 9 out of 10 years you look like an idiot, and 1 year you get to say "I told you so". In the long long run, nobody really knows who will win.


Total Posts: 1143
Joined: Jun 2005
Posted: 2020-05-06 12:28
Just a thought:

imagine very simplified market situations:
at t=0, S=100,
at t=100, S=200

there are two ways to arrive there.
1 < t < 99, S = (200-100)*t/100+100 (linear) and
1 < t < 99, S = 100, then it jumps to 200 (jump)

Marginal distribution at t=100 is the same, but market reaction to them at t=100 will be very different.
Again, if we see these jumps many times then they smooth out again into the Gaussian.

So, Taleb most of time speaks about how unexpected the "unexpected" is, which links directly to the mismatch between calibrated/expected and realized. This will always be. And time scale is irrelevant here.
The main difficulty is 1) how to account for this un-un-expected 2) how to build the system which will not be (financially) destroyed when it crashes into the wall in the dark.


Total Posts: 78
Joined: May 2010
Posted: 2020-05-06 17:20
Relevant article on Bloomberg today:



Total Posts: 351
Joined: Feb 2014
Posted: 2020-05-07 21:02
from where Taleb-hating brigade coming from?

don't know the guy or read any of the books (yet), but he is clearly legit and universa did what actually supposed to do so he is crystal clear to me.

they (and I - [bias warning]) believe in amor fati as the correct approach, so it would be unlikely if they deceived themselves that they would have an advantage if they did not

First Commander of the USS Enterprise


Total Posts: 591
Joined: May 2006
Posted: 2020-05-07 21:14
The case is for hedging systemic tail risk. But it isn't great.

The basic idea is that it is all nice to be beating your benchmark, but when the benchmark crashes spectacularly, you will crash spectacularly as well. If you return -20%, the investors won't be impressed by your "but the index is down 30%".

So something that, say, floors the index loss at 10% makes you flat. Which is good.

Now, like @nerotulip says, just buying puts is suicide. You'll spend all your money long before any crash. You don't need a crash.

So then the next order is collars. You buy protection against downside beyond some limit, and you fund it by seling your upside beyond some other limit.

But that's not great either. Vol is skewed, so a zero cost collar is signifcantly asymmetric. There is a lot more downside than upside. You can only make it symmetric by paying money for the difference, which adds up to significant negative cash flows.

Which is all fine if all you are is a tail hedge. It's not your job to generate returns.

But if you are in the business of generating returns, it's no good. When you have paid for the hedge, there are little or no returns left.

"There is a SIX am?" -- Arthur
Previous Thread :: Next Thread 
Page 1 of 1