Tuesday, February 24, 2009

Don't Blame The Quants, Felix

It would be nice to think that this crisis was a huge math error. Some geek in a cubicle hit the wrong button on his HP12-C. Doh! Alas, it is not the case. Any error this pervasive is not from some obscure assumption, but an assumption that everyone was comfortable with. If an obscure technical assumption drove investors into various bonds, there should be a large group of investors who would have quants who disagreed, and they would have not been exposed. That is, quants may have no common sense, but on obscure technical results, they are smart and often disagree. Yet not one of the major investment banks worldwide escaped this craze. It's as if all the geeks thought the 5th digit of Pi was 6.

The only reason everyone made the mistake was that it was from an assumption that everyone seemed to think was reasonable, something quants did not have the authority to affect: that housing prices would not fall significantly. Look at Shiller's update Irrational Exuberance from 2005, and there's a chapter on the Housing Bubble, where he notes that recent housing price increases 'probably won't continue', hardly a clarion call to avoid housing. When everyone is doing something, and think it's right, the quants will perforce generate supporting documentation. Scientists in that way are like lawyers, advocates, not for a paid client, but rather, the ephemeral verities of the current zeitgeist. With $4B for global warming research, and much less for anti-global warming research, it is no surprise many scientists are documenting various evidence of global warming: people like the results, they aren't as skeptical, more likely to get published, get a grant, get on a government project, etc.

UBS used a 10-day Value-at-Risk on its residential mortgage backed portfolio to assess its risk, using data from the benign 2000-2005 period. This was an error. But UBS was going to invest in mortgages anyway, because presumably this collateral does not decline much if ever, so this was all standard rationalization of preconceived objectives.

Felix Salmon wrote a piece in Wired, where he fingers the copula function as the primary culprit.
And (David) Li's Gaussian copula formula will go down in history as instrumental in causing the unfathomable losses that brought the world financial system to its knees.
He then quotes some quants, like Darrell Duffie, who states that "corporate CDO world relied almost exclusively on this copula-based correlation model". Well, derivatives textbook authors might like to believe that their formulas underlie most financial activity, but I suspect the number of people using copulas in their sales pitch was low, no more than 5%, because most investors, big shots who actually make portfolio decisions, do not like pretentious mathematics. Perfunctory metrics may be ubiquitous, but they are still perfunctory. The decision makers are rich, powerful, kind of smart, do not feel embarrassed by their lack of knowledge in obscure technical trivia, and surely are not intimidated by it. Of that limited set, the copula itself, irrespective of the broader knowledge about mortgages and housing price trends, was probably the driver of 10% of those purchases, because even for investors who like formulas, most require a bigger picture. Thus, I generously estimate about 0.5% of all mortgages were bought because of David Li's Frankenstein formula.

Copulas, value at risk, correlations, credit scores, one might add standard deviations, means, and Microsoft Excel's 'solver'. These concepts can get tricky, but the basic assumption in the mortgage crisis, that housing prices would not fall significantly, was not tricky, and people with the authority to make decisions were not swayed by these technical issues, rather, they used them to validate their earlier beliefs.

21 comments:

Anonymous said...

bs on the standard error diatribe. no statistical metric, technique, test etc. is predictive in the markets. they're all descriptive. i dare you to prove me wrong. but of course you won't, cause that's not why you write this blog.

ccm said...

"So leave the geeks alone. They merely prove what the suits want them to prove."

Paul Volcker calls this the Nuremberg defense: http://network.nationalpost.com/np/blogs/fullcomment/archive/2009/02/17/paul-volcker-the-banking-world-needs-more-canadas.aspx

Eric Falkenstein said...

Anonymous: Ok, I'll sell you call option on PEP (Pepsi), and buy a call on HOV (hovnanian), at the same implied vol +/- 10 vols (but, different rebate). That is, the higher historical volatility for HOV relative to PEP, should be irrelevant to you, and I'm giving you a 10 vol vig. It can be 90 vs. 80, 200 vs 190, whatever.

Per the Nuremberg defense, those blamed for Nazi atrocities at Nuremberg were the decision makers, not scientists, even though they supported Nazi policies. What they did was wrong, but I think it would be wrong to blame the atrocities on the geeks. I would blame Rubin, Thain, and the rest. But David Li? No.

Anonymous said...

your example is badly chosen. why not PEP vs KO and you adjust the vig accordingly? that's where we can test the true value of historical vol. and we do it 100 times? i thought so

Eric Falkenstein said...

First, I find it odd that anonymous commenters seem so angry at the poster, in this case me. It highlights how anonymity brings out the worst in people (manners wise).

But in any case, my point is that historical vols are highly correlated with implied vols, and one can use historical vols to generate a profitable strategy if you are trading against someone who does not have mere 'historical' data. Past is not prologue, but it's all we got for empirically based forecasting.

Anonymous said...

not angry, sorry if it's coming that way. that's how i usually talk :). and i choose anon cause it's the easiest button there. and keeps the focus on the ideas vs ppl. i guess.

Anonymous said...

Relating to the scientists and the Nurenberg defense ... the life story of Fritz Haber is pretty interesting. Check it out on Wikipedia. Nobel Prize winner that developed a process that helped create gas for the WW I gas attacks. He directly supervised gas attacks and was honored by the German military. In protest, his wife committed suicide.
A Jewish scientist, people working at his institute in Germany developed the Cyclon B gas. This gas was used to "exterminate" Jews in the camps.
He eventually had to leave Germany because of the Nazis, although Hitler considered making an exception for him.

Alex R said...

Paul Wilmott doesn't think the quants are entirely blameless.

Eric Falkenstein said...

Wilmott was quoted in the article. he's one of the derivatives textbook authors who thinks the mathematics in his books are used to make decisions all the time. After all, he asks people, and they say, 'sure, we sophisticated analytics'. What else are you going to say? It's mostly about being polite, but also signaling. But I have seen a lot of important portfolio decisions made, and have yet to see an obscure math formula as crucial. It is useful to know that stuff, but the 'total derivative' of copulas on mortgage demand is near zero.

tc said...

The bigshots don't know what the formula means, but surely the fact that a formula exists makes people more willing go out on a limb - makes them less "ambiguity averse", to use econspeak. What was the volume of options traded before and after everyone heard of Black-Scholes?

Plamen said...

I provide the analysis for the decisions in a hedge fund of the more unsophisticated kind, and I stand behind Eric. In our case, hundreds of millions of dollars get allocated every single day based on analysis I would be ashamed to discuss with intelligent people - but that's what the boss wants, so that's what he gets. Most of the time I am reduced to submitting results with no statistical significance, and I have tried to argue against it, but I like my job more.

wrightak said...

Felix is correct in saying that your argument is weakened by inserting your own views on climate change. I agree otherwise.

"The decision makers are rich, powerful, kind of smart, do not feel embarrassed by their lack of knowledge in obscure technical trivia, and surely are not intimidated by it."

My experience is that this isn't true. People in high places tend to be proud, smart and always averse to displaying a lack of knowledge. If there's one thing I've learnt, it's that people pretend to know more than they do. Any discussion on quantitative subjects is usually met with neutral responses that do as little as possible to betray a lack of familiarity with the subject. "Sorry, I don't understand, can you explain?" is never heard in finance.

Anonymous said...

Following along Plamens' comments, I work at 'quantitative' hedge fund but the decision maker is neither a quant nor a Trader! Little understanding of what or how any thing works and only responds to $ signs, i.e. Model made money? Add. Lost money, cut it.

ngogerty said...

The label quant is applied a little crudely across time frames and applications. The high frequency trader is perhaps less dangerous than a group of banks all using similar co-variance asset allocation models for structured finance investments. Random quant views right or wrong, should be appreciated. Unified group-think or application of models is what is worrying. Portfolio insurance was an example. Moody's etc. ratings of securitized junk leading to "Free money" is another example. Collective myopia (my term) is more dangerous than individual delusion.

Nick said...

addendum to top comment, beware basel 2 and those who game Basel 2, namely everyone. SIV 2.0 will be a Basel 2 defect used institutionally in the next few years. Systemic approaches to risk management propogate systemic failure.

Anonymous said...

What drives you quants to work in this stuff vs. saving the world? That's what many poeple want to know. I guess money is a patr of it, but i think its more about feeling empowered, after a life of geekiness. I guess nice clothes, expensive restaurants are exciting. I also think that talking, and yes having power over these bankers is a high...self relefction poeple!! We dont just do our jobs. We need to understand our jobs, how they affect our companies, our world...

Daniel Beunza said...

Here's a different view.

The post and subsequent comments (wrightalk and plamen) make it seem as if the quants (and their equations) and the bankers were on radically different sides on this issue. I disagree.

My reading of Felix's comment is a different one, and based on my fieldwork research at a failed hedge fund. According to my informant --a frustrated modeler of CDOs, the problem was that non-quant managers were able to get away with bad decision-making for too long, because house prices were indeed going up for years and years.

But the REASON why prices were going up is not unrelated. It was caused by the influx of capital into the real estate market... crated, in turn, by the ease with which David Li's formula allowed banks to price CDOs.

The problem, in short, is not the formula nor the quants nor their bosses... on their own, but the vicious cycle that rose among them.

wrightak said...

"The problem, in short, is not the formula nor the quants nor their bosses... on their own, but the vicious cycle that rose among them."

I completely agree that the problem cannot be attributable to any group. Perhaps I should try and be clearer.

My small point was that in my experience of working on a trading floor structuring CDOs, people do not ask questions when they don't understand, there is an aversion to sharing knowledge and everyone tries to sell themselves as very knowledgable. There's good reason for this: if you're a trader making millions, you don't want to tell other people what your thoughts are. No one wants to decrease their perceived value within the company by sounding stupid asking basic questions. This goes for everyone: bosses, traders, salesmen, quants, whatever. Quants won't ask about economic issues, bosses won't ask what a copula is, salesmen won't ask about anything (and won't understand anything). It's an culture of individualism and highly competitive.

daniel beunza said...

Wrightack -- You're on to something. Lack of communication and a culture of not sharing certainly is a problem on Wall Street.

One of my papers on the organization of trading rooms (my best paper so far) explains how the manager of the trading room used all sorts of mechanisms... from compensation design to architectural layout to recruitment policy... to ensure that communication remained open at all times.

For the curious:

http://www.sociology.columbia.edu/pdf-files/stark3.pdf

Anonymous said...

to random anon: we don't try to save the world b/c we know it cannot be saved. and the human race doesn't deserve to survive anyway.

to eric: so i was right. there are no statistical tests predictive in the markets, that can be argued for with a straight face.

Anonymous said...

What drives you quants to work in this stuff vs. saving the world?

Capitalism saved humanity from a persistent animal-like existence. Instead of desperate scavenging from food, shelter and clothing, billions now can dote on their kids, read blogs for enjoyment, expand their minds and consciousness because of material wellbeing, intellectual distribution channels created from capitalism. While every European in the 17th century lived through multiple famines and plagues that killed off chunks of people, I can swing by Whole Foods for Tikki Masala, beer and In Touch.

I am a proud, dedicated soldier in that revolution.