Universa Investments

[jeffsreng]

FYI:

 

https://www.youtube.com/watch?v=8vrtPxj7NTY

 

https://www.bloomberg.com/news/articles/2020-04-08/taleb-advised-universa-tail-risk-fund-returned-3-600-in-march

 

https://www.universa.net/riskmitigation.html

[Gregmal]

So the assumption is what? The Ackman trade? Didn’t really see anything in the article detailing any specific trade. Just the same type of index and hedge stuff that’s been floating around like spam on twitter lately. Actually after checking, it’s from the same dude who is responsible for a lot of those tweets.

 

This stuff is always cool but often it’s either inaccessible to most investors(the instruments to put these trades on) or its some wild awkward looking angle some guy sees that make him look genius when otherwise he s out to lunch. Kind of like John Paulson. Greatest trade ever and then lost most of his fund in an easy to see bull market over the next decade.

[thowed]

some wild awkward looking angle some guy sees that make him look genius when otherwise he s out to lunch. Kind of like John Paulson. Greatest trade ever and then lost most of his fund in an easy to see bull market over the next decade.

 

Ha ha - agreed!

 

The Bloomberg article sounded cool, but all we know is the results for 2020 - no sense of how much was lost before that.  If you're down 95%, then making 3,612% isn't going to be quite so special.

[SHDL]

I think they were up a lot in 2008 too, like 8x or something. The fund is marketed as something that you might put a small portion of your portfolio into for hedging/volatility reduction purposes and there are certainly people/institutions that will pay for that sort of thing.

 

I was always interested in what exactly they own, but I imagine that is a trade secret.

[Gregmal]

Yea they'll just ask for 3% of your capital. Probably EVERY year...

[opihiman2]

Big fan of these guys.  Learned some really interesting things from them, including the barbell strategy.  Artemis and Ruffer also shaped my current view on investing.  Basically, the more I see these boom / busts events with equities, I am moving away from it.

[scorpioncapital]

It's insurance with a twist. Namely the payout is very high when it hits. So you pay very low premiums each year and every decade or two when a big disaster hits you make 1000x. It's not a bad strategy to add to any portfolio. But I'm not sure how to do it as a private investor. Universa probably isn't publicly traded or take all clients. You could buy deep out of the money puts but you need to do it regularly. Not sure if there is a vehicle to do this. Most people can just buy an inverse ETF or something? But that wouldn't hit big time as it's not buying tail risk insurance.

Any thoughts on how to buy tail risk insurance would be appreciated!

[jeffsreng]

FYI:

 

Universa Investments Q1 2020 Letter

Universa_Investments_Q1_2020_Letter.pdf

[thowed]

Thanks for this.  Interesting to read.

 

I'd still like to know, for example, how the figures looked at the end of 2018 or 2019.

[scorpioncapital]

I wonder if his fund is scalable , i.e if it was much bigger if he could do these things. Maybe there is counterparty risk on a larger scale?

 

http://universa.net/UniversaResearch_SafeHavenPart3_Tenbaggers.pdf

 

This article seems to suggest that you must buy catastrophe insurance in the amount of 4x-8x roughly return in a crash event. No discussion of pricing however. It seems they use out of the money put options.

But another question is what are they buying puts on? Volatility? SP500? specific basket or individual stocks? Seems there is still a judgement call unless you just do the entire market. Guess this is where they make their admin fees, deciding. I suspect there is some theory about what they feel is highly likely to implode during a crash. SP500 can't possibly be the best choice but it is definitely a choice. In fact it may fall much less than say specific sectors. Maybe a good question to ask is: What has ALWAYS collapsed in a crash, any crash. So far real estate is prime candidate but that's only going back 20-30 years. Would be interesting to see sensitivity analysis to crashes somewhere.

 

This article explains the mechanics of what they do. I found it very useful as you can do the same thing (not  now though as premiums are sky high) - https://thefelderreport.com/2016/08/15/worried-about-a-stock-market-crash-heres-how-you-can-tail-hedge-your-portfolio/

 

 

[SHDL]

Classic pension fund move:

 

https://www.wsj.com/articles/calpers-unwound-hedges-just-before-marchs-epic-stock-selloff-11587211200?mod=searchresults&page=1&pos=1

[JimBowerman]

So I'm probably misreading this, but he mentions that the mean annual return was 76%....but the geometric return is likely very different?  Anyone have any estimates on what geometric mean return since inception has been?

 

I'd guess its much lower than 76%...

[Kaegi2011]

Just got around to reading this.  Super interesting to think about.  A few interesting tidbits that I picked up:

 

1 - 36x return in Feb, 41x return YTD, but against a 96.66%/3.33% SPX/tail hedge portfolio, March performance was 0.4% vs. SPX March performance of -12.4%.  This would imply that only .44% of the tail hedge portfolio was actually "at work" at the beginning of March vs. the 3.33% allocation (or 13.1% of the total 3.33% allocation, not exactly 1/12, but maybe due to the big drawdown in Feb carried over?). 

 

2 - In the public articles he talks about tail hedge being able to return 9-10x in a big drawdown event (defined as >-15%), and over backtesting period even assuming 0% return it does aid the portfolio in a fairly dramatic way. 

 

3 - While the articles do not explicitly state this, I have to assume the tail hedge costs are expected to be expensed on an annual basis. 

 

I quite like the elegance of the theory.  We know emotionally we are our own worst enemies, so to be able to stay calm and invested during times of turmoil is extremely valuable.  Furthermore, not only does it provide a hedge, it also allows one to be tax efficient (assuming crash levels are still above cost bases), and add to exposure during the most opportune time.

 

My questions related to the strategy are as follows:

 

A - What instruments are being used for this strategy?  It has to be a derivative of some kind.  Likely not a straight out of the money put but something far more exotic (binary options? barrier options? I looked up "exotic options" and saw a few I've never even heard of like Rainbow and Timer options - fascinating stuff...). 

 

B - Given the instruments, what is the execution strategy?  Based on #1 above it seems like the hedge costs are not being spent all at once, but rather sparingly throughout the full year.  So this implies some type of rolling exposure, but how? 

 

C - Lastly, but likely most pertinent to us, is there a similar replication strategy we can implement with instruments more accessible to the common investors (e.g., puts?)?

 

On the last question, based on *today's prices* it seems that one would have to go with a put spread strategy to get close to a 10-15x max return (and that's going long 15% OTM put and short 20% OTM put a month out).  Using instruments longer term (e.g., 6 months), it would produce closer to 5x max return.  Using a more "normalized" vol and an options calculator, the returns above can 2x.  But remember that's the max return, and also at expiry, so it's possible that MTM return is far less. 

 

As a thought experiment, I used the options calculator here - https://www.optionseducation.org/toolsoptionquotes/optionscalculator - to see what SPY puts for June 30 would have been priced at the beginning of the year (324 underlying, 275 strike, and 250 strike, 20% vol) - the 275 strike would have been $3.24, and the 250 strike would have been $0.70.  Ignoring bid/ask for a second, assuming the spread cost is as such (3.24 - 0.70 = 2.54), then max payoff was close to 10x (25/2.54).  However, if one held it until now, the return is close to 2.14x as the 275 strike is 10.09 mid and 4.66 mid, for a total value of $5.43.  Not interesting (but we're also looking at prices after a new bull market so...).

 

So what are some ways to make the simple instrument hedging strategy more interesting?

 

a - Shorten the duration of the puts.  Of course this comes with its own issues such as time decay, so instead of one month out, maybe do a rolling 2 month hedge (e.g., ride it from day 60 to day 30, then go buy new 60 day hedges again)?  I need to have historical options data to see how this would work out.

 

b - During big drawdowns, sell off the long put and keep the short put.  That way, one could realize the explosion in vol on the long side, and ride it back down on the short side.  Obviously if the timing is incorrect, and the market goes down a lot more, the hedging strategy starts to accelerate the decline of the entire book.  Kinda dangerous... 

 

I'm out of ideas at this point.  Any thoughts on this?  Does anyone have data access on historical options pricing?

 

[scorpioncapital]

Not sure I understand the short put portion of your strategy. Does that somehow work to boost the payout in a cataclysm? I looked at pricing on Vanguard sp500 etf (lower management fees) but the prices are still too elevated. I mean you want pennies on the dollar like maybe 10 or 20 cents but I think it's still very elevated, which suggests fear in the market has not yet dissipated. Also there is a real risk in future periods you have inflationary depression which means sp500 may go up but not fast enough relative to costs. But all those options would always be worthless for a very long time.

 

[Kaegi2011]

Not sure I understand the short put portion of your strategy. Does that somehow work to boost the payout in a cataclysm? I looked at pricing on Vanguard sp500 etf (lower management fees) but the prices are still too elevated. I mean you want pennies on the dollar like maybe 10 or 20 cents but I think it's still very elevated, which suggests fear in the market has not yet dissipated. Also there is a real risk in future periods you have inflationary depression which means sp500 may go up but not fast enough relative to costs. But all those options would always be worthless for a very long time.

 

Yes, vol is currently elevated due to fear in the market.  That's why the trade would have paid out 2x payout now if done at the beginning of the year when vol was lower (approximately).  With regard to your point about inflation - that's covered by the 97% of the risk portfolio, not the insurance side.  The insurance side is for drawdowns only. 

 

Your question on the short put option - I'm pairing it to lower the initial upfront cost so the payout is higher in between the two strikes, but lower in the extreme scenario (e.g., if you think market is going down 50% it's better to buy 45% OTM nakes puts than 20% OTM puts or a bear spread with 15% / 25% downside strikes).  Since his description is that a drawdown is 15% I'm trying to construct a payout where insurance "pays off" past 15% but does not require 50% for a 10x.  Hope I'm explaining my logic clearly. 

 

Obviously would love to hear other opinions on constructing this tail hedge. 

[Kaegi2011]

In case anyone is curious, Universa only manages institutional capital, with min account size of $150mm.  I assume that's per year, so a cool $5bn AUM should do it. 

[SHDL]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

 

[BG2008]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

Which high yield bond ETF went up 100x?  I'm curious

[SHDL]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

Which high yield bond ETF went up 100x?  I'm curious

 

It was either HYG or JNK, maybe both. The caveats are: (a) you had to get the strike and expiration just right to get the max return (duh), and (b) trading was pretty thin pre-crash so I am not sure how much you could actually buy at those historical prices.

 

Edit: Randomly pulling out one example ... a few HYG 5/15 70 puts traded < $0.05 in Feb and went > $5.00 in late Mar

[Kaegi2011]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

Which high yield bond ETF went up 100x?  I'm curious

 

It was either HYG or JNK, maybe both. The caveats are: (a) you had to get the strike and expiration just right to get the max return (duh), and (b) trading was pretty thin pre-crash so I am not sure how much you could actually buy at those historical prices.

 

Edit: Randomly pulling out one example ... a few HYG 5/15 70 puts traded < $0.05 in Feb and went > $5.00 in late Mar

 

So you have access to historical options data?  Could we PM and work together on some backtest? :)

[SHDL]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

Which high yield bond ETF went up 100x?  I'm curious

 

It was either HYG or JNK, maybe both. The caveats are: (a) you had to get the strike and expiration just right to get the max return (duh), and (b) trading was pretty thin pre-crash so I am not sure how much you could actually buy at those historical prices.

 

Edit: Randomly pulling out one example ... a few HYG 5/15 70 puts traded < $0.05 in Feb and went > $5.00 in late Mar

 

So you have access to historical options data?  Could we PM and work together on some backtest? :)

 

I don’t have any special data subscriptions, I just pulled the numbers above from my broker (IB). The option hasn’t expired yet so the numbers were easy to fetch. Thx for the offer though. :)

[bizaro86]

I would suspect there is some sort of volatility bets in there as well. Possibly way out of the money options on the VIX? They are european exercise, iirc, so you would need to own every expiry.

[Kaegi2011]

I have no insider info but looking at their numbers I think it’s quite likely that they have bets tied to the bond market as well. Like Eurodollar futures/options (to bet on falling interest rates) or credit default swaps (to bet on rising credit spreads). As an individual investor one could have done very well during the recent crash with put options on certain high yield bond ETFs as well — some went up like 100x or so IIRC.

 

Which high yield bond ETF went up 100x?  I'm curious

 

It was either HYG or JNK, maybe both. The caveats are: (a) you had to get the strike and expiration just right to get the max return (duh), and (b) trading was pretty thin pre-crash so I am not sure how much you could actually buy at those historical prices.

 

Edit: Randomly pulling out one example ... a few HYG 5/15 70 puts traded < $0.05 in Feb and went > $5.00 in late Mar

 

So you have access to historical options data?  Could we PM and work together on some backtest? :)

 

I don’t have any special data subscriptions, I just pulled the numbers above from my broker (IB). The option hasn’t expired yet so the numbers were easy to fetch. Thx for the offer though. :)

 

Got it. So you can see historical pricing on existing options?  Still think that’s helpful. I can’t even find that info on Merrill!

 

Out of curiosity what’s the max pricing on 275/250 spy puts for June (noted above)?

[SHDL]

Got it. So you can see historical pricing on existing options? 

 

Yes, you can view it on TWS (their trading software) or through an API.

 

 

Out of curiosity what’s the max pricing on 275/250 spy puts for June (noted above)?

 

It looks like it got very close to 20 in late March.

[scorpioncapital]

Mark's main thesis is you need massive leveraged upside in rare events. Doesn't even matter anything else. So anything that doesn't go up a whole lot isn't going to work. I was looking at say TTT, 3x leveraged short 20 year bond etf. But it just isn't going to work as even that is not enough leverage. Maybe the options on such an etf would be better.