Berkshire Profit on Goldman Sachs Passes $2 Billion

[Martinghoul]

Quote from asiaprop:

you noticed they are talking about options of 10-15 years of maturity?
In that context I dont agree with your criticism about the level of implied he sold those options at.

I would actually agree with that...

If you're in the insurance biz, like Buffett, have enough cash to handle the mark-to-mkt fluctuations, which he obviously does, and are going to run the puts to expiry, what does vol, either implied or delivered, matter?

In fact, I think what he has done with this trade makes sense in terms of overall style. As to the whole 'derivatives =WMD' I think it was a disingenuous comment on his part in the first place.

 

[makloda]

Quote from atticus:

Buffett also discussed these positions at the company's annual meeting on May 2. He has been managing these positions in the open market, and has shortened up his S&P 500 put options from 18 years to 10 years while lowering the strike price from 1,514 to 994 because he thinks the latter options are an even better deal.

I wasn't aware of that. I just remembered those older SEC filings from 2006. What you quoted is much more recent.

I didn't know that doubling down on a failed bet is now called "managing a position" LOL

 

[ElCubano]

Quote from ByLoSellHi:

Who's going to deny Buffet's trade would have been wiped out had he not gotten the government to shove 180 billion down AIGs broken throat?

It's all horseshit.

that is an example of a true EDGE

 

[atticus]

Quote from asiaprop:

you noticed they are talking about options of 10-15 years of maturity?
In that context I dont agree with your criticism about the level of implied he sold those options at.

Secondly, you seem to ignore that he is massively long cash and if he held 6-12 months downside puts in addition that would actually make him long gamma and most importantly, stay long gamma even if the market moved quite a bit. I can't say to what extend he is long the nearer downside puts and he is obviously massively long delta but he is an investor not a market maker.

However after everything is said, the trade is by far not as stupid as you make it out to be. (obviously the timing could have been way better thats why I said it was not one of his better trades).

WTF are you babbling about? Long gamma?! Right, he took a $10B loss on his long gammas... yeah, that's the ticket.

The term structure of vol had nowhere to go but up. You may have had a point had Buffett sold vol in the mid-30s, but he didn't. 10-year vols were in the 13s with front-months at 11.

He rolled the 1514 trade to 994 because it was a better deal!

Stupid? It's mind-blowing stupid. The counterparty-trade was phenomenal.

 

[atticus]

Quote from Martinghoul:

I would actually agree with that...

If you're in the insurance biz, like Buffett, have enough cash to handle the mark-to-mkt fluctuations, which he obviously does, and are going to run the puts to expiry, what does vol, either implied or delivered, matter?

In fact, I think what he has done with this trade makes sense in terms of overall style. As to the whole 'derivatives =WMD' I think it was a disingenuous comment on his part in the first place.

All trades can work if maturity is infinite, right? They're path-independent, so it's a nit to worry about a MTM loss of $10B. The difference between right and wrong is $10B.

Why vol and PD matters -- the counterparty can take the huge mark-up on vol and price and lock the trade at something approaching that $10B gain. Buy the binary put at 1514 and sell >notional in a binary put struck at 1000. Or simply cross the trade at tremendous vol and price edge with another counterparty. Buy spot. The fact that Buffett has a functionally-unlimited bankroll doesn't make it a sound decision.

 

[asiaprop]

again, you completely ignore all his other positions in this context. You assume this was his only position at that time which is nonesense. Your comments on this trade hint at the possibility you understand options, but not much in the context of a large buy side investor PLUS you know NOTHING about his exposure in shorter dated options that he may have held against this basket.
You assume his 2008 annual letter to investors reveal everything he was sitting on and I find that reveals a bit of your naivity in this particular context.

Quote from atticus:

WTF are you babbling about? Long gamma?! Right, he took a $10B loss on his long gammas... yeah, that's the ticket.

The term structure of vol had nowhere to go but up. You may have had a point had Buffett sold vol in the mid-30s, but he didn't. 10-year vols were in the 13s with front-months at 11.

He rolled the 1514 trade to 994 because it was a better deal!

Stupid? It's mind-blowing stupid. The counterparty-trade was phenomenal.

 

[Martinghoul]

Quote from atticus:

All trades can work if maturity is infinite, right? They're path-independent, so it's a nit to worry about a MTM loss of $10B. The difference between right and wrong is $10B.

I don't understand your point. I do not know how to define an expiry trade if maturity is infinite. I also don't understand what path-dependency has to do with anything. All I am saying that if I have enough cash to reserve against a potential max loss of $10B, MTM loss (as long as it's less than $10B) actually doesn't matter. That's exactly why insurance companies are regulated differently than, say, banks and why they don't have to mark their assets to mkt.

Why vol and PD matters -- the counterparty can take the huge mark-up on vol and price and lock the trade at something approaching that $10B gain. Buy the binary put at 1514 and sell a binary put struck at 1000, for example.

That doesn't make sense... Of course, the counterparty can hedge any which way they like any time during the lifetime of the trade, but who cares. All that matters is what happens at expiry, when someone makes money, someone loses.

I think of it this way... Do I have to care about vol and path dependency if I had an infinite amount of cash/liquid assets during the lifetime of the trade? Obviously not...

 

[atticus]

Quote from asiaprop:

again, you completely ignore all his other positions in this context. You assume this was his only position at that time which is nonesense. Your comments on this trade hint at the possibility you understand options, but not much in the context of a large buy side investor PLUS you know NOTHING about his exposure in shorter dated options that he may have held against this basket.
You assume his 2008 annual letter to investors reveal everything he was sitting on and I find that reveals a bit of your naivity in this particular context.

DUDE, he stated the loss in the letter. You think he's going to neglect the rather salient point that he's got a nifty offset in down&out gamma? Right, he left out the netted $10B gain on hedge, sure.

The pro-Buffett responses on this thread are a black-hole of stupidity. What is naivity? I assume you mean naivete. Who knows, I may be that, but for now I will assume that Buffett was forthcoming with the netted-position mark to market. IMO, he makes an argument for going un-hedged due to the Euro-expiration.

 

[atticus]

Quote from Martinghoul:

I think of it this way... Do I have to care about vol and path dependency if I had an infinite amount of cash/liquid assets during the lifetime of the trade? Obviously not...

Right, but you don't. Whether Buffett can ride it out isn't the point. If it was, then selling Euro-convention vol at any price could be considered valid. After-all, vol @ expiration is zero.

 

[Martinghoul]

Quote from atticus:

Right, but you don't. Whether Buffett can ride it out isn't the point. If it was, then selling Euro-convention vol at any price could be considered valid. After-all, vol @ expiration is zero.

What is the point then?

You're absolutely right about selling vol, but it's only sensible if you don't have to mark-to-market. This is exactly what insurance companies do, they sell vol to us. Obviously, it's the nature of their biz that they also can't structurally avoid 'delta', so they have to be right about the underlying distribution and the price they sell their policies (puts/calls) at.