Via memeorandum, a pretty fascinating piece in Wired magazine on the market meltdown:
A year ago, it was hardly unthinkable that a math wizard like David X. Li might someday earn a Nobel Prize. After all, financial economists—even Wall Street quants—have received the Nobel in economics before, and Li’s work on measuring risk has had more impact, more quickly, than previous Nobel Prize-winning contributions to the field. Today, though, as dazed bankers, politicians, regulators, and investors survey the wreckage of the biggest financial meltdown since the Great Depression, Li is probably thankful he still has a job in finance at all. Not that his achievement should be dismissed. He took a notoriously tough nut—determining correlation, or how seemingly disparate events are related—and cracked it wide open with a simple and elegant mathematical formula, one that would become ubiquitous in finance worldwide.
For five years, Li’s formula, known as a Gaussian copula function, looked like an unambiguously positive breakthrough, a piece of financial technology that allowed hugely complex risks to be modeled with more ease and accuracy than ever before. With his brilliant spark of mathematical legerdemain, Li made it possible for traders to sell vast quantities of new securities, expanding financial markets to unimaginable levels.
His method was adopted by everybody from bond investors and Wall Street banks to ratings agencies and regulators. And it became so deeply entrenched—and was making people so much money—that warnings about its limitations were largely ignored.
Then the model fell apart. Cracks started appearing early on, when financial markets began behaving in ways that users of Li’s formula hadn’t expected. The cracks became full-fledged canyons in 2008—when ruptures in the financial system’s foundation swallowed up trillions of dollars and put the survival of the global banking system in serious peril.
David X. Li, it’s safe to say, won’t be getting that Nobel anytime soon. One result of the collapse has been the end of financial economics as something to be celebrated rather than feared. And 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.
There have been a number of articles about this phenomenon, including this 2004 piece noting that a number of theoretical physicists were heading to Wall Street, and this Washington Post piece from 2007 about the same topic. I distinctly remember another long piece in the Washington post about this, but I can not find it right now.
At any rate, every time I think of the theoretical thinking behind all this, I am reminded of a story one of my old international relations profs told me almost two decades ago about Mike Tyson (I have no way of verifying the quote, as I have looked for it and can never find it. Might be my prof was bs-ing). As the story goes, in an interview with Tyson, a reporter told Tyson that the fighter he was about to face had a strategy for this, and a strategy for that, and a strategy to counter all of Tyson’s strengths, and how did Tyson plan to deal with that. According to my prof, Tyson responded: “They all have their strategies. Then I hit ’em.”
Andrew J. Lazarus
Hey, if we had more academic positions for Ph.D. mathematicians and physicists, we wouldn’t have had to do finance.
Signed, Ph.D. mathematician who works in finance.
And not the only one who reads BJ.
Jeffro
I just got my copy of WIRED yesterday and jumped to this very article. WIRED really is the last subscription I’ll ever cancel (or should I say, if I needed the money, I would cancel all of my subscriptions except for WIRED)
Having said that, the article plainly lays out how this financial disaster was caused by yet more magical thinking at the top, used to justify whatever the folks in power wanted to do anyway.
Sound familiar?
HT
My understanding of the quote is "Everyone has a plan until they get punched in the mouth."
James Gary
The "copula" formula? Am I the only one who sees this name as appropriate, given that the financial system seems to be fucked?
Keith
The Tyson quote is a variation on a famous Cus D’Amato (Tyson’s original trainer) line: "Everybody has a plan until they get hit"
Noonan
John, the quote you’re thinking of goes: "Everybody’s got plans until they get hit."
cervantes
An economist, a physicist, and a sociologist are marooned on a desert island. Then a case of canned spaghettios washes ashore. "That’s just great," groused the sociologist, "how are we gonna get these open?"
"Assume a can opener," said the economist.
Andrew
There was an interview with a Chinese sovereign wealth fund manager in The Atlantic recently, and he said the same sort of thing: if you undervalue science, all of the really smart people will go into finance and come up with a whole bunch of crazy ass ways to make money, like CDOs.
amorphous
This is the only positive that can come out of this crash, that scientists will start to do more science, that our greatest minds won’t go directly to Wall Street to make more money for those who have too much already. If only we hadn’t wasted the last thirty years (or even just the absolute throw-away anti-science policies of the Bush years), think of where we could be. What astounds me is that the "fiscal conservatives" all get their panties in a bunch for science funding, which their oh so pwecious business buddies can then turn around and commercialize. Oh, that’s right, they still need the fundies. One group will sell their souls for an extra $10, the other would give up all money to save theirs. How are those people married politcally, again?
bago
Should have been a Bayesian formula. Consider the veracity of your sources.
Tim H.
That should have been the first warning. Simple and elegant formulae usually only apply in the ideal world.
DougJ
The trouble with all this stuff is that the models aren’t that good. If something you think is a 5 sigma event really turns out to be a 2 sigma event, then you’re screwed, no matter how clever your mathematical solution is.
I also think the key line is:
As a mathematician, I have very mixed feelings about all of this. On the one hand, it vindicates my decision to stay in academia doing pure research. On the other hand, I’m bummed out that my friends who I thought were making millions helping the world actually made millions helping to destroy it. On the whole, I find it depressing. Though completely predictable.
Walker
As a mathematician by training, let me say that these people were stupid. As an academic, you assume that random variables are independent from one another because it makes the computations easier and allows you to publish a paper. But when you start asserting that the world actually works that way and that people should bet huge sums of money on it, then you get systemic failure.
I love the related anecdote that Volker recently told on this matter:
DougJ
I’m not sure how much will change. Among the math/physics people, many of them ended up on Wall Street making millions because they failed to get a job in Urbana making 50 thousand. There are exceptions of course, but that was certainly the pattern.
zzyzx
@Andrew J. Lazarus: "Hey, if we had more academic positions for Ph.D. mathematicians and physicists, we wouldn’t have had to do finance. "
That’s one of the dirty secrets they don’t tell you in undergraduate school. It’s all "shortage of math PhD’s" there and it’s only after you get to grad school that they let you know about the glut.
I bailed a year after my MS degree because I didn’t want to spend another 3 years of my life working to get something that would probably make me less hirable than the MS… best. decision. ever.
Binkyboy
I don’t think it is time to completely throw out scientific finances. There are patterns and situations that can be modeled. The thing is, though, you have to have peer review, you have to have data and evidence to support the theories. You can’t just hand it to a CEO and say "this is a semi-perfect model and will make you millions" because all they hear is "millions" and they’re off.
If you can’t wait for a new idea to be tested, verified and tested some more, then you are only disrespecting science, you aren’t doing science.
Walker
Well considering all the hiring freezes in academia right now, we certainly aren’t pulling up the slack. My department actually had some faculty lines survive the cut, and it is buyers market for us right now. People we would normally lose to competitors in a good year.
DougJ
You’ve got that right.
Still, I’d rather these people go into something like biostats, where there still seem to be plenty of jobs.
amorphous
@DougJ:
Noted, however if we can create policies whereby a.) there is more and more easily accessible funding and b.) there is less financial incentive to go to Wall Street instead, then I think it is something we can change for the better. You’re absolutely correct that most scientists who want to do research can’t for whatever reason, and that they would take less money to do it. I know I would (I’m looking for jobs right now, and would gladly take less money to do some research… of course I’d rather not postdoc, but that’s a different story). It appears Andrew J. Lazarus would like to, also. We need more research universities, we can expand our national laboratory mission (which is nominally security), we can
more tightlyregulate Wall Street at all.Come to think of it, that’s probably what the "business Republicans" are afraid of (if I dare give them enough credit to think more than one step ahead). It’s all about the funding. We can find some cool stuff to do with the money if it’s there.
zzyzx
It’s sad that I’ve been bitching about people confusing the model for reality for years now, and the model I’ve been complaining about is nowhere near this sophisticated. The entire conservative economic theory revolves around an uber-simplified assumption about supply and demand where all transactions happen in a vacuum and nothing matters other than maximizing the profit for each transaction, because there never can be repercussions from someone feeling like they’re ripped off and angry mobs in the street could never affect people’s willingness to go to work…
…and then they have the nerve to mock others for "not understanding Economics 101," when the problem is that they’ve gone on to the junior and senior level classes and understand the shortcomings of the basic kindergarten model.
Walker
Part of that is the ever revolving "when everyone retires" story. The GI bill hires were supposed to retire. Then they didn’t. And then they were, but the Russians flooded U.S. math departments. Now people are retiring again, but faculty lines are being cut. It was always something — though the job market was pretty damn good in 2000 as I recall.
The success of a PhD program has to do with whether the program understands the "backup market". What do you do with your students who are not going to go to a flagship R1? In computer science departments, they typically go to YahoogleSoft. In mathematics departments, they train them as teachers at SLACs – to earn less than a high school teacher.
Walker
We are interviewing computational bio people this year. That’s one of those areas that I am referring to in which we have the pick of the litter. As in demand as it is, there are not a lot of job openings out there.
Ash Can
This issue mirrors my overall beef with economics as an academic discipline. I was a dedicated poli sci student throughout my college years (a discipline also worthy of derision, but it had its advantages), and as such, I felt the need to take economics courses as well, given the extent to which the two disciplines were intertwined in real life. I could never fathom what the overall point to economic study was, since it always seemed to me to be presented in terms of cut-and-dried mathematics, with various writers making value judgments on these formulas nonetheless. I mean, wtf? X+Y=Z is no good because it’s not Y+X=Z? Obviously, I’m oversimplifying, but that’s the basic idea. It wasn’t until, finally, a graduate-level professor with half a gift for communicating ideas lucidly explained to us students that the value of the study of economics was in identifying the various forces that exerted influence on real-life economic issues and structures that I finally appreciated the study of economics. Voila — in one simple sentence that professor turned my years of economic studies completely around, and I realized that I was finally looking at them from the proper perspective, rather than being forced by texts and professors alike to view each economic theory in a vacuum and determine its correctness or incorrectness separate from any kind of real-life circumstances.
This also explains why I look at Ayn Rand and her followers like they’ve just landed from Pluto. When I try to say to them that any given organization, whether social, political, or economic, is only as good as the people who make it up, rather than being moral or immoral on its own, I might as well be speaking Mandarin Chinese to them. They can’t seem to get it through their heads that none of the political/social/economic structures in our world would exist if people didn’t invent and populate them. (To exemplify: Communism is an immoral system? Fine, go tell that to a monastery.) Like with Li’s formula and the popular (among financial folks) fascination with it, there’s a cause-and-effect issue involved somewhere that’s messed up, and it’s enough to make the whole structure fail eventually.
Walker
Republicans love government research funding. Beltway Bandits live off of SBIRs.
amorphous
@zzyzx:
That, and that it almost necessarily implies time is not a variable and that everyone has equal access to information.
Martin
I do a fair bit of mathematical modeling in social settings, and I’ll tell you it’s fucking hard to get right for very long. Elegant formulas are pretty damn easy to come by, but people are particularly good at anticipating such elegant formulas and routing around them or taking advantage of them.
A model that works great for a few years will suddenly go to shit because the population sees patterns and changes their collective behavior (look at how internet users jump from service to service to route around things like spam and phishing). In fact, pretty much the worst thing you can do is set your behavior to an elegant formula, because it’s there wide open to be taken advantage of. That’s doubly true in a financial market when every side of a transaction is convinced they are getting the better deal and traders can take advantage of a situation no matter what direction it is going in, so long as they can predict the direction. And breaking such elegant formulas is trivial – just work off of variables not present in the formula and it’ll never see you coming.
zzyzx
@amorphous: and it also implies that we all start off in similar enough places that there are no power advantages that one party can use to take advantage of the other one…
The Grand Panjandrum
You also had the break up of the old Soviet Union at about the same time these events began to unfold in the late 80’s and early 90’s. Suddenly, you have young US educated PhD’s competing with mathematicians from the former Soviet bloc countries for an entry level position. These people had dozens of published papers and were willing to take jobs paying 40 grand a year. A lot of people took jobs that paid, at minimum, twice that because it really was the only option given the influx of more seasoned people from Eastern Europe. Given the opportunity to take a job that will begin paying 80-100 grand a year or keep swinging away and hoping to get that 40 thousand a year gig? It should not be to difficult to figure out why so many ended up in finance.
Walker
The Soviets were an excuse in the 90s, not the 00s. Everyone who was going to come over had already done so by 2000. Which is why I remember the job market being so good that year.
There were quants in the 90s, but the modern bunch are a decidedly different breed. The mortgage wizardry really took off because people were looking for a safe investment after the tech collapse — and saw real estate.
NonyNony
@DougJ:
Yes. I watch PhD’s nearly kill each other for jobs that make 60-70K a year, with the backup plan of "going into industry" and making far more than that if they can’t snag one of those handful of academic jobs.
It’s why I laugh in the face of people who say that if we cap CEO pay then we won’t get the best CEOs in the job. Ha! You’ll still get plenty of people who want to be CEOs – and you’ll probably weed out the ones who don’t actually want to do the damn job but just like to have the money.
amorphous
@zzyzx: I think we can both agree that there’s only one way to solve this problem: Tax cuts.
bago
That’s why you need to use pre-execution behavioral agnostic models. Set up a neural network that has no bias built into it, and then run it for a few million generations and see what happens. As long as the initial condition variables are all scalable it can work.
Kirk Spencer
The map is not the terrain. The file is not the man. The model is not reality. It amazes me how often people forget these things.
lutton
OMG that tyson quote is great because my brother-in-law recently used it to describe becoming a father. He reference Tyson’s statement like this: "Everybody has a plan until they get hit in the face."
Walker
That’s because those are the good jobs. Many SLACs start at 45k for an a tenure track assistant professor.
zzyzx
@amorphous: I’m still waiting for someone to seriously suggest that a 0% tax rate will produce infinite revenue.
bootlegger
@DougJ: My brother is a statistician (with two math degrees) in public health and they are doing some really cool work. The best part is that their models aren’t deployed foolishly to make someone millions of dollars, but rather they lead to a modest policy trial that will grow if it works.
Crockpot
I know it’ll never happen and I’m sure people will tell me it’s a bad idea, but I’ve always just wanted a simple law that states; "the most any executive of any publicly traded corporation can be paid (including stock options/bonuses and such) is exactly 100 times as much as the lowest paid employee of said corporation."
THAT would be a rising tide that raises all boats.
bootlegger
@NonyNony: That is 100% fucking-a right-on!! How much "incentive" does any one person need to do their best work? And if there is too much "incentive", then who are you attracting? Those dumbasses assume the can opener (I’m stealing that joke by the way) and end up with can full of shit.
bootlegger
@Crockpot: Germany has something like this though I’m not sure of the particulars. But it has obviously killed their economy amiright?
Barry
Walker:
"As a mathematician by training, let me say that these people were stupid. As an academic, you assume that random variables are independent from one another because it makes the computations easier and allows you to publish a paper."
Actually, no. If you can’t work with correlated variables, you’re in for a world of hurt.
" But when you start asserting that the world actually works that way and that people should bet huge sums of money on it, then you get systemic failure."
What I don’t believe is that the higher-ups were fooled; if you were 55 years old and a VP on Wall St in 2000, you started work back in the early 1970’s. In your first ten years in finance you saw:
The first and second Oil Shocks, with massive commodity effects,
Inflation going up to 13% (e.g., your 10% bond was now a money-loser),
Massive third-world sovereign debt defaults,
The US losing a war,
The worst recession since the Great Depression.
And that’s just what I can recall, and just from your first 10 years. These guys have seen reality smash enough things to understand that a mathematical formula is just that.
IMHO, it came down to the higher-ups gambling with their shareholders’ money, taking massive bonuses when they won, and sticking the shareholders with the bills when they lost.
"I love the related anecdote that Volker recently told on this matter:
(snipped)
A year or so ago, my daughter had seen something in the paper, some disparaging remarks I had made about financial engineering. She sent it to my grandson, who normally didn’t communicate with me very much. He sent me an email, “Grandpa, don’t blame it on us! We were just following the orders we were getting from our bosses.” The only thing I could do was send him back an email, “I will not accept the Nuremberg excuse.” "
Considering that Volcker was deliberately responsible for the worst post-war recession in sixty years, perhaps he should be first to the gallows.
Uli Kunkel
Whatever happened to Harry Kat?
TenguPhule
Considering you’re an idiot, you should go ahead first.
It was a recession or stagflation from hell.
Volcker took the lesser of two evils.
TenguPhule
Fixed.
TenguPhule
I always thought it was because there is no reliable variable that can ever model human stupidity.
Waingro
In other words:Political Economy, which has been neglected in favor of "Econ 101 rulz!". The people who idolize Adam Smith don’t realize that he was more of a political economist than a simplistic Supply & Demand tool.
Punchy
If by this you’re referring to Urbana, IL…..uh….wow. You couldn’t pay me $100K to work in Urbana. Unless you also supplied me with a bulletproof vest and a Lo-jack.
over_educated
You kind of had me up untl there… Volker for his many faults, laid the groundwork for the Reagan ERA expansion and has an excellent understanding of monetary policy. He was sounding the alarms on the current craziness years ago.
Brachiator
Here’s the fun thing: The current Village Idiot Wisdom is that throngs of “irresponsible people” brought down the economy because they bought houses that they could not afford. But the model originally pushed by the banker-fools was that bad choices by individual homebuyers did not matter because millions of transactions – risky and prudent — “smoothed out” the impact of the bad decisions. In effect all the conservative pundits yelling about “moral hazard” ignore the inconvenient truth that Wall Street whiz kids foolishly believed that they had developed a magic elixir which could make risk, failure and retribution disappear forever.
This model was stoopid on so many levels that it is amazing that anyone ever believed any of it for a second.
Person of Choler
Barry, the lessons of past busts did not penetrate financial board rooms. Companies had staffed up with financial genius types who had been trained on various quack mathematical risk evaluation techniques (e.g. Black-Scholes, Risk-Metrics….) Everybody wanted to listen to these stellar intellects and ignore experience of the messy, nonmathematical world.
Read "The Black Swan" (2007) by Nicholas Nassim Taleb for an entertaining send up of some of this numerological fakery.
A quote from Taleb’s book, "…the government-sponsored institution Fanny Mae, when I look at their risks, seems to be sitting on a barrel of dynamite, vulnerable to the slightest hiccup."
Walker
What TenguPhule said. This is why Volker, was vilified at the time, but is now largely recognized as a hero on the international stage.
Wile E. Quixote
@Waingro
Here’s an Adam Smith quote that you don’t often hear from the Randroids and others who worship him, but who never actually read An Inquiry into the Nature and Causes of the Wealth of Nations
That Adam Smith, what a cynical bastard he was.
chriskillian
I wouldn’t so much say there’s a glut of mathematicians, but the ratio of US-born vs. foreign-born mathematicians in this country is rather unfavorable. There are a good number of jobs available with the three-letter-agencies for mathematicians, but the pay isn’t much better than in academia (~50K to start with the NSA).
I took a position with a defense contractor because they pay more (and I have kids to feed), but again there’s the issue of where the mathematician in question was born. I plan to go back to complete my Ph.D. when my kids are a bit older, and I don’t expect it’ll cause me to be any less hirable/employable.
In other words, I don’t see any inherent reason for mathematicians to land in finance (except, perhaps, greed). Even the (only) modestly greedy can do as I did and join the military-industrial complex. =) There is original and interesting work being done by mathematicians in other places than universities and Wall Street.
Rick Taylor
Wasn’t there a huge firm about a decade or so bag that figured out a formula for making money with no risk, and become immensely rich until it nearly brought down the financial system? Damn, my mind’s gone blank, but I remember the story. You’d think we’d learn.
Chuck Butcher
Jumpin’ Jayzus on a f**kin’ pogo stick, "I’ve got a system," is supposed to be the moment you put your wallet back in your pocket. Maybe making a living by following plans and having to deal with their shortcomings with actual sticks and nails has forced me to look at things from something like a concrete viewpoint. I understand that stocks are a representation of something concrete, that a house is something to goddam live in, that $1 of value doesn’t magically turn into $10 of value because you write paper that says it does.
I could tell you that you should buy my ’62 Chevy II for $100K and I could point to a split window Stingray and, and, and… You and I would make a pretty good deal at $14K where I’d get something out of my investment in research, time, money and you’d get a car that, if you didn’t pile miles on it, would at sale cover you. This is true because it is not daily transportation and gas costs have nothing really to do with it but selling it as such a thing is a flat lie and you’d be screwed if you bought it as such. The fun factor of a massively over-powered very nearly sports car braking and handling 48 year old car doesn’t make it something it isn’t – it will burn 92 octane premium at 15 miles to the gallon at the very best and if you bang it up parts will cost an absolute fortune.
This wall will not meet this roof line is real, no matter what that drawing shows, so something in the real house won’t be what the drawing shows. Stocks have a basis in fact – the physical plant, inventory, sales, and a certain amount of goodwill which does not trump the rest. A 14K Dow was idiocy in operation, probably nearly 2x the actual value of the companies so how does that work? How does $1 of house mortgage become more than $1 + interest? It doesn’t.
Corner Stone
@Rick Taylor
If you’re not being snarky then I think you’re referring to LTCM.
Of course, even if it is snark then I still think you’re referencing Long Term Capital Mgmt.
Bunch of genius quants tried to slice the pie so thin that even with a million counter parties they tried to eliminate risk. Which, anyone with a brain should know by now is impossible.
They did succeed in setting the gold standard in leverage – until now of course.
Corner Stone
@Walker
I think the correct phrase here is "fucking greedy".
Ken
you can use the word "apochryphal" to describe your Mike Tyson story.
Tim F.
I remember talking to some of the theoretical physicists and computational biologists who were on their way to Wall Street and wondering what the hell they were going to do there.
For christ’s sake, we already learned this lesson from Enron. We don’t need smart people managing money. We need sensible people. They are not the same damned thing.
TenguPhule
Contracts and the time value of money.
TenguPhule
Some people just never learn.
terry chay
These physicists were stupid. They were engaging in magical thinking. End of story.
Andrew J. Lazarus
I botched it so bad I ended up in a finance industry job that pays about the same as academia—we are a very small shop and we don’t run money (that may change if anyone has any left).
The job market of the early 90s was indeed very bad because of an influx of Soviet-bloc mathematicians, and, although they were in even worse shape than US-born students, a non-trivial contingent of Chinese grad students and postdocs who decided to stay after Tienanmen. The market of the 80s, OTOH, was pretty decent. Most of the people I know who finished in the mid-80s got good positions.
I don’t mind the ideas of financial engineering, in their place, but there’s nothing like an 8-digit bonus to make you forget GIGO.
I went to high school with one of Volcker’s children, and she was not the sharpest knife in the drawer. No, I don’t know what that explains.
Chris Andersen
When I heard about how all these mathematicians had gotten jobs in Wall Street and how their modeling was used to justify a lot of the questionable practices that got us into this mess, all I could ask myself was, "Didn’t these guys ever study Chaos Theory?"
Nutella
I guess we can leave Li’s Gaussian copula in the dustheap of history, right next to the Laffer curve. Too bad we didn’t do that before all that money was thrown away attempting to follow them.