catallaxy files

catallaxy in technical exile

long term capital management

with 33 comments

what can we learn from academics and their models?

you can make a model that comforts you with certainty in an uncertain world. look at this statement, its one of those necessarily false statemnts:

“in a strict sense, there wasn’t any risk — if the world had behaved as it did in the past” –merton miller, nobel laureate

no, you fool, in a strict sense, of course there was risk. the nerds at LTCM had obviously never heard of the great depression. any historian could have told them the risk had they cared to ask. but they were caught up in ivory towers divorced from the real world, meriwether took them out and put them in ivory towers with money. (plus the bankers were complete nutters for funding the whole operation)

something to keep in mind for the lovers of climate change models.

Written by Admin

November 21, 2006 at 11:37 am

Posted in Uncategorized

33 Responses

Subscribe to comments with RSS.

  1. This AGW scare reminds me a lot like trading. In fact the human behaviour element is almost identical.

    There’s the model guys. The technicians, the funadmentalists. The whole kit and kit and caboodle. Even the sceptics.

    What’s so funny is that the current climate reminds me of traders opinion on the US Dollar. There’s hardly anyone who thinks the US dollar can go up from here. Anyone who does is shouted down for being an idiot.


    November 21, 2006 at 11:14 pm

  2. “but they were caught up in ivory towers divorced from the real world, meriwether took them out and put them in ivory towers with money. (plus the bankers were complete nutters for funding the whole operation)”

    There was one director there, an old time Irish-American bond trader who kept telling them to be careful with their risk. he was shouted down and treated like a leper.
    Not formally educated…. just understood market dynamics.

    This is why I think a lot of what Bird says makes sense.


    November 21, 2006 at 11:18 pm

  3. rebel traders as well…but not the good kind like in jason’s post…


    November 21, 2006 at 10:40 pm

  4. c8to, apologies but the comment sequence in the “Democrats return to the pro-slavery cause” thread appears to have gone awry.


    November 21, 2006 at 10:42 pm

  5. are you talking about a guy at salomon or at long term…

    so many irish guys…

    then there was tully at merril lynch who was chummy with them…


    November 22, 2006 at 3:39 pm

  6. you’re right about global warming…

    the model guys are going to be shot down…but unlike long term unfortunately they aren’t going to lose their shirts…(or actually they will get bailed out by the government)

    the fundamental guys…those trying to understand why this all happens (me and bird) should win the day…the technical traders can go to hell…


    November 22, 2006 at 3:41 pm

  7. It is a puzzle to me quite what we can learn about academics and their model from the LTCM debacle, which had much less to do with a failure of academic models and more to do with a failure of corporate controls.

    Further, one might point out much better examples of extraordinary failures of academics and their models and anyone else with a vaguely acadamy based model. For example, to my knowledge every forecast ever made of analog mobile telephony demand was so far off base to be laughable looking back. The question is what is the alternative. Is the Iris-American bond trader always right? How do you know when he isn’t and the Dutch-Australian will be? Or put slightly more generally, what should you rely on?

    I’m not so sure academics and their models deserve so much disapprobation (or why do we visit doctors?). As the bumper sticker reads, if you think education is expensive, try ignorance.


    November 22, 2006 at 3:42 pm

  8. actually I have to agree with Kodjo here except on the regulation issue. LTCM came up with a great product. They just took leveraged themselves too much using it. Their risk taking was irrational. Nothing to do with regulation or bad models.

    Jason Soon

    November 22, 2006 at 3:50 pm

  9. their models we’re bad because they worked in the normal time but not during crashes…and they couldn’t predict crashes…

    any naive model would work as well as theirs did…i could do exactly the same in currency markets which are cyclical until there is a big change…big deal…i dont have a nobel prize…

    the lesson is just because everyone in a certain industry gets carried away with a bad model doesn’t mean you have to…

    larry summers said efficient market hypothesis would have to go down as one of the biggest errors in economic history.


    November 22, 2006 at 3:54 pm

  10. That’s rubbish. Why should markets be good aggregators of information everywhere except in the financial markets.

    The EMH is better than any of its alternatives. It’s the most logical model there is.

    Jason Soon

    November 22, 2006 at 4:01 pm

  11. Cato.
    It was Jim McEntee.
    Kept telling them to be careful, they were getting to be too big and the spreads they were trading too small. he undertood risk in a way the academics never did.

    There’s lot to be said about horse sense.

    You see, Bird may be wrong on certian things, i don’t know, but he sure as hell has a lot of horse sense.
    He bloody right about mdels too. If they can’t be back tested they’re not worth shit.


    November 22, 2006 at 4:03 pm

  12. The EMH is better than any of its alternatives. It’s the most logical model there is.

    …and it’s still wrong.


    November 22, 2006 at 4:04 pm

  13. “That’s rubbish. Why should markets be good aggregators of information everywhere except in the financial markets.”

    I actually think markets are horrible aggregators of information. Telstra is a great case in point.

    Markets are simply people making buying and selling decisions.


    November 22, 2006 at 4:07 pm

  14. Incidently I don’t think Merton Miller was involved in LTCM, Robert Merton was he’s merely commenting on the event.

    However, in the quote I believe he is using the term “risk” in a fairly technical way per Frank Knight where risk is uncertainties that you can asign probabilities, such as inside a model, as against uncertainty which are random events that you can’t quantify, in this case the chance that the model is just wrong or inappropriate.

    One of the great ironies of the LTCM thing is that several of the academics caught up in it (Scholes and Merton) had been strong defenders of the Efficient Markets Hypothesis. Black-Scholes equation for example holds that the market moves should be both random and normally distributed, but here they were trading which is essentially a pure gambling exercise if you believe in EHM. Which begs the question what did they think they were doing.

    Steve Edney

    November 22, 2006 at 4:13 pm

  15. Steve
    From what I understand their trading wasn’t inconsistent with their theory as it was based on the idea that certain pricing anomalies and anomalies in variances would eventually correct themselves.

    Jason Soon

    November 22, 2006 at 4:17 pm

  16. “….based on the idea that certain pricing anomalies and anomalies in variances would eventually correct themselves. ”

    Which what a trader would expect from academics. A good trader would tell these guys that you can go broke trying to prove your theory is correct.

    you saw this type of behaviour in interest rate swap markets when time and time again traders ended up getting crastrated believing that the swap spread could never tighten as much as it did.

    A nice guy sitting a good spit away from me once lost the firm $250 million believing that shit. LTCM was not a new thing. They just lost the biggest pile of cash.


    November 22, 2006 at 4:25 pm

  17. Jason,

    Well yes and no. It did hinge on a belief that they would converge to the efficient outcome, but to take advantage of this it relied on this being inefficient in the first place.

    As Keynes could have told them the market can stay irrational longer than you can stay solvent, particularly if you are leveraged 50-1.

    Steve Edney

    November 22, 2006 at 4:33 pm

  18. yeah im not sure whether miller was commenting from inside our outside LTCM…still a bonehead statement…

    its exactly that definition of risk that got them into trouble…

    the eggheads are walking around saying theres no risk…so the banks leap in (their own fault) and then it all comes crashing down….

    i can just see the nerds going “well there was no risk, there was just uncertainty”

    they needed someone to beat them over the head…


    November 22, 2006 at 7:17 pm

  19. There was jim McEntee trying to do that, but it’s hard to do when the other guy is a nobel prize winner and the firm is on a roll.

    I knew Jim and spoke to him about it. I believe him when he told me he warned them all the time but was shouted down.


    November 22, 2006 at 7:33 pm

  20. “Then there was James McEntee, who was Meriwether’s constant companion on the green. Close friends since Meriwether’s early years at Salomon, the duo spent that summer jetting back and forth between golf courses, from New York state, to California and then to Ireland.

    McEntee played the court jester to Meriwether’s prince. When Myron Scholes joined the pair that summer on one of his first ever golf trips, McEntee teased the bookish novice mercilessly, later commenting: “Myron read 100 books about it to figure out the physics of the swing.” But the prince himself still lacked a kingdom. ”

    That would be true.


    November 22, 2006 at 7:37 pm

  21. Thjis gets back to what Bird said a long time ago about scientists and science workers. You can read all the books, know all the theory and not have the imagination to be a real scientist.

    The two guys who won the nobel peace prize were given a very hard time for thinking bugs had a hand in creating ulcers.

    These days anyone in the climate community who doesn’t toe the party line is treated like filth, especially by non scientists who pose they are.

    This is a shocking indictment.

    It is also deadly. Economists have figure out there is one extra death for every $us 15 million additional add on costs to the economy. This works out to being 500,000 people in the first 20 years of trying to meet kyoto.


    November 22, 2006 at 7:58 pm

  22. you’re right jc…i got sworn at and abused for two hours the other day by some uni nutbag (like a pv engineer or something) for discussing climate change…

    and i was being very civil and reasonable when i should have been tearing him a new asshole…

    facts i proved him wrong on which he never retracted, just kept ignoring:

    1) recent paper(s) have shown antarctica is cooling
    2) its been warmer in the past when life was on earth, which he later change to mamallian life which i again showed it had been warmer then
    3) he claimed CO2 was 500 ppm which i corrected to around 320 ppm at which point he said, well its going to be 500

    he got so emotional and upset because he was so infected with left wing lunatic thinking. in the end he kept coming back to attacking me for not believing in global warming, despite the fact that i claimed from the beginning that

    1) its getting warmer
    2) CO2 levels have risen from 280ppm pre industrial to about 320ppm now (these are off the top of my head but reasonably close i think)

    what a waste of two hours…the discussion finally ended when he told me to fuck off for putting forth the statement “that free markets would be beneficial to africa”

    (as an aside he thought markets we a joke and the solution was to put some smart people in power and let them run the show)


    November 22, 2006 at 8:23 pm

  23. LTCM is an interesting study in risk management. In the long run, they were right. I remember seeing a study (sorry, can’t dig it up now) that shows that their position would have come good in the long run. That is why they were Long Term Capital Management. The problem, in the short run, was liquidity to cover their position. The Russian default was an outside 3 standard deviation event (i.e. their liquidity VaR models did not cover it) and, as noted above, they were leveraged to the hilt, so they folded.
    The people doing the return models were right, those modelling liquidity were also right – it was simply that they did not do their scenario analysis corretly or stress test their portfolio.
    LTCM was designed to be high risk for high(er) returns, so I did not lose any sleep over it.
    The modelling people are often right – within the parameters of their models. I just would not bet my house on them.

    Andrew Reynolds

    November 22, 2006 at 8:31 pm

  24. Andrew VAR is crap. It’s going to cause a few road deaths if people live by it.

    I recall that because yen went from 145 to 111 in 72 hours back in 98 we were limited risk taking everytime we had a yen position for about 2 years. Great when the horse has bolted.


    November 22, 2006 at 8:39 pm

  25. But these financial models always assume there is an open liquid market. We learnt the hard way through portfolio insurance in the 80’s that this is nonsense. The boys at LTCM built that assumption in their models. They lost a lot of cash as a result of not finding bids in Italian BTPS (Government bonds). Funny that.


    November 22, 2006 at 8:42 pm

  26. CATO

    It’s really scary how few people actually show any faith in the market process. That’s why i am a little down about the future. Very few of us actually think we are better off with unfettered markets.

    That’s why this recent election in the US could be such a disaster. Not even the GOP believes in it any more.

    As for the MSM… the only strongly pro-market editorial department is at the Wall Street journal. Everything else is red or pink.


    November 22, 2006 at 8:46 pm

  27. Andrew,

    The problem is when we say “outside three standard deviations”, this doesn’t really mean anything unless we are talking about a world of normal distributions, which are very poor models for these type of events. 3 SD events crop up a lot more often in financial markets that modelling to a normal will indicate.

    The 87 stockmarket crash comes out as being unlikely to have occurred ever, even if the market had been trading for the entire history of the universe if you use normals.

    In short they should have been aware of this – perhaps they were, but it would have shut down the amount of leverage they were getting which inevitably would have reduced their returns.

    Steve Edney

    November 22, 2006 at 9:47 pm

  28. thats like saying your number will eventually come up on roulette…who cares…if you cant stay liquid until then and take your profit…

    im optimistic jc…markets are so amazing that they will get around people acting foolishly…the history of the world shows great progress for much of the human race…

    hopefully competition and evolution will continue to steer us in a positive direction…(with more power comes greater internal risks to human civilisation, but also hopefully the ability to spread and progress civilisation faster)


    November 22, 2006 at 9:52 pm

  29. Yea Great points Cato As always.


    November 22, 2006 at 9:58 pm

  30. the world needs us JC – the cautious optimists…

    we will try to avoid making the big mistakes, and warn against stupid planning policies, but we must be optimistic, spread truth and give hope to others…


    November 22, 2006 at 10:00 pm

  31. shit you guys are starting to sound as self-important as bird 😉

    Jason Soon

    November 22, 2006 at 10:01 pm

  32. its ok jason…i’ll drag you with us even though you are wrong about global warming…

    in fact your completely backwards on the two issues of the day…you’re worried about global warming but not nuclear proliferation…

    and i’m worried about nuclear proliferation and not the planet…

    you’re scared of nature when you should be worrying about the other humans…


    November 22, 2006 at 10:21 pm

  33. c8to says:

    “1) recent paper(s) have shown antarctica is cooling”

    Parts of antarctica are cooling. So what? Who lives there?

    “2) its been warmer in the past when life was on earth, which he later change to mamallian life which i again showed it had been warmer then”

    Yeah, global warming isn’t going to eliminate all life from the planet. This isn’t so much lowering the bar as sinking it to the Earth’s core.

    “3) he claimed CO2 was 500 ppm which i corrected to around 320 ppm at which point he said, well its going to be 500”

    It was around 320 in 1964. It’s gone up since then but not to 500. You were slightly less wrong. Congratulations.


    November 22, 2006 at 11:50 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: