Buffett slams Black-Scholes and 'flat earth' economists

[DominiConnor]

This begs the question of what model Buffet's firm is using to hedge and price these options ?

An analogy:
My company is building a bridge.

The building regulators says I have to meet a safety standard that make certain assumptions about steel, vibration, wind, load etc.

I say that because I'm smarter than the regulators that I can build a bridge that will make my firm more profit if we use a model that says we have to spend less on steel and concrete.

Who do you believe in this case ?

Is it different to what Buffet is saying ?

 

[iHateVariance]

i dont think Buffet is hedging, prly got his forward at 4000 for SPX. he is the type of guy that would sell HY equity at 90c and use it go long WFC.

 

[daleholborow]

I would say your bridge analogy is in effect exactly what Buffett is saying. I.e. "I believe, and am willing to bet with a lot of money, that my ability to value this financial product is better (more accurate?) than the general consensus".

So, in Andy's article, he is complaining that he is forced to add some redundant reinforcing to a bridge based on regulatory requirements, when he is quite sure it would stand up of it's own accord without said extra reenforcement. Whether he bases this on a rigorous mathematical formula, or 50 years of gut feel, I couldn't say (but would hazard a guess).

The only pedantic element I am inclined to pick at further is that, to my mind, structural engineering (building bridges etc) is a largely observable and quantifiable "hard" science.... maths and a few thousand years of experience have given humanity a pretty good notion on how to build a good bridge given a set of requirements. (The guys in the budgeting department might argue for the discount price concrete, but at the end of the day, what is required to ensure whether or not it stands is fairly well established?). Quant finance, on the flip side, still seems subject to some emotional factors that aren't captured in a similarly strict fashion.

 

[Tsotne]

Actually it should be of no much concern whether someone condemns or not using a particular model. Even if he did not like BS, it doesn't at all mean that he mustn't have been taking advantage of using it. He really called derivatives the financial weapon of mass destruction but if we remember the effects of derivatives crisis then we also realize that he is right. The effects of derivatives on the leading economies is enormous and the mathematical complexity is taking matters from humans' hands and "with all of its sophistication can get ahead of human judgment." These are Alan Greenspan's words. I agree. It doesn't mean that we shouldn't be engaged in derivatives in any way. Instead,we must realize the importance and the possible effects of failure.

 

[Yike Lu]

Out of interest, do you disagree with either of the first two Buffett points you mention?

And it would seem to me that the aim of investing is to maximise your returns... if Buffett has the best returns over the longest times, wouldn't that suggest he's worth listening to?

If you are an INVESTOR (and very particularly a value investor) not a trader, then it makes sense to listen to him. Simply because the returns are mean zero doesn't imply that everybody trading so often is a loser... it's most likely a positively skewed distributions with the top hedge funds in the right tail.

Also, not everybody believes they will live forever...

As to "WMDs" let me point the following out: WMDs in the real world sense are dangerous. Put them in the hands of idiots and people die. But they are also deterrents. Should they be around? You can't uninvent the atom bomb....

 

[Tsotne]

Very interesting articles:

In 1988, the then-Fed chairman Alan Greenspan stated, "What many critics of equity derivatives fail to realize is that the markets for these instruments have become so large not because of slick sales campaigns, but because they are providing economic value to their users."

http://www.investopedia.com/articles/optioninvestor/08/derivative-risks.asp
http://www.fintools.com/docs/Warren Buffet on Derivatives.pdf

 

[bob]

I imagine Buffet is thinking in terms of what you might call an actuarial valuation of these positions: If you take the historical perspective, in all likelihood the chances that these puts wind up in the money are, he would say, overwhelmingly small and thus merit a small valuation. No doubt this is the reason he was willing to sell them for the price he got.

But you can't have it both ways. If your position is going to be marked and collateralized according to a particular method, you can't collect the premium on one basis and then demand that the thing be subject to different treatment once it's on your balance sheet. In fact, his own statement about the auditor's perspective illustrates one reason why a common standard, even if imperfect, is essential in zero-sum cases like this one: If the party long the option is able to mark it one way while the party short the option is able to mark it another, in aggregate the result creates money out of thin air. In the process, it hides risk.

Things of this sort do happen, all the time, in places and ways that might surprise you. Still, the fact that it is possible to vaporize risk in areas that are poorly regulated or complex does not therefore make it a good idea to institute such practices in what ultimately is a rather straightforward area.

One final point: There are many who have negative things to say about Black-Scholes, but the critiques I've seen generally hinge on the fact that they understate the risk of large moves--in equities, this is virtually always a downside risk, which gives rise to the much-talked-about vol skew. Buffet is using these arguments as camouflage, but is actually not in agreement with them. He is claiming that Black-Scholes valuations overstate the risk of the positions, and so to my way of thinking is really complaining that the standard in options pricing is insufficiently reckless.

 

[DominiConnor]

My problem with BS is that it is so hugely dependent on unobservable values, often that makes any assumption about lognormality look pretty trivial.

 

[bigbadwolf]

My problem with BS is that it is so hugely dependent on unobservable values, often that makes any assumption about lognormality look pretty trivial.

That's why it was turned on its head by traders, who used actual option prices to estimate the unobservable "volatility." More discussion about this in MacKenzie's book, An Engine, Not a Camera.

 

[Afamiii]

Overall much of what Buffet says and what has been written about him from an investing point of view is misleading. Most of Warren Buffet's outperformance has come from stock arbitrage, which could hardly be described as a strategy supporting an idea holding period of forever.

Up untill the late 80s his portfolio turnover was high and indeed even in his letter he separated his portfolio into 'permanent holdings' which were to hold 'forever' and everything else presumably was for trading.

By the 90s, his holdings were so large that it became impractical for him to sell in that in many companies he is the market. So all of a sudden the best holding period is forever.

In his letters and in the popular books about him Mary Buffet, Hagestram, Loenstein, et.al the role of growth is significantly down played (after all he is a value investor.) But again if you look at his 'permanent portfolio' the alpha here has indeed been as much about the growth over the holding period as the low price paid.

In fairness perhaps he has stated that value and growth are joined at the hip. And by the late 90s was saying that price paid is margin of safety, but can't make up for buying a bad business.

 

[Tsotne]

Assumption of lognormality vanishes the anti-analytical way to go through, for example simulations. If not the assumption that asset prices are lognormally distributed and follows the geometric brownian motion, we'd be left with artificial tools making the option price even more unobservable since the only way to price an option would be some numerical methods which are pretty difficult to observe in terms of effects of variables involved. Nor I agree with the idea that it depends on completely unobservable inputs. The major breakthrough of Black-Sholes was to state an analytical way to price an option in such a way that variables involved were observable. One difficulty Fisher Black and Myroon Sholes came across while developing the model was the level of risk to be measured.

So the main idea of revolutionary theory was to change the financial world from non-observable models to more observable ones. As for assumptions, better assumptions haven't yet been stated which could take into consideration the sped of calculation and practicality (especially for traders) and still being able to approximate an option value as good as BS.

 

[DominiConnor]

Also, the most common criticism of BS is that it fails to handle tail events all that well, ie is not well suited to risk management, typically underpricing risk.

So Buffet is saying he wants to take more risk than that ?

 

[Tsotne]

Also, the most common criticism of BS is that it fails to handle tail events all that well, ie is not well suited to risk management, typically underpricing risk.

So Buffet is saying he wants to take more risk than that ?

If this were his reasoning then he'd probably be saying he wants to better 'estimate' (rather than take) risk. I think he is saying from economic viewes contradicting mathematical steps. BS requires to break some rules in order to remain simple and fast and observable and practical. Options were mentioned as the weapons of mass destruction -and this opinion cannot be even analyzed from financial or mathematical point of view. He is speaking from pure economic theories...

 

[Mr Doe]

I think he is saying from economic viewes contradicting mathematical steps.

I seriously doubth he knows anything about mathematics of BS. By the way, have you ever looked at his derivatives holdings? Seems like a "spit in the face", to the whole weapons of mass destruction story.

 

[Tsotne]

I seriously doubth he knows anything about mathematics of BS. By the way, have you ever looked at his derivatives holdings? Seems like a "spit in the face", to the whole weapons of mass destruction story.

I haven't looked but I read the complete article about his speech on BS. That's why I doubt he is speaking from mathematical point of view. Criticizm is just very general (as economists love). Saying something about widespread layoffs caused from loss on options contracts and how great impact and danger options can pose on the economic stability. That's their story. Not a single word about approximations, assumptions, stochastic explanators of underlying movements, etc. He said one - two word about the incorrectness of BS not enough to criticize from financial side though.