I am not in the habit of reading classic horror stories but this weekend I picked up John Kenneth Galbraith's 1955 book, The Great Crash: 1929. Unfortunately it is non-fiction. And even more unfortunately it is selling well in the university bookstore. Galbraith is gone but his book lives on. In a new Foreword written in the 1990s he noted that it has never gone out of print since its publication more than 50 years ago, mainly because every decade or two we have a new stock market crisis to renew interest. Since 1929 these crises have all been harbingers of recession, not depression. It isn't clear if this one will be different or another 1929 disaster. Reading the book, however, is frightening because the parallels are eerily similar, from the incessant incantations of convenient wisdom (the fundamentals are sound, the big players are stabilizing things, the market is poised for rebound after being oversold, etc., etc.) to the actual behavior of the market.
The free fall of Black Thursday (October 24, 1929) was presaged by a number of other sell-offs and in fact the market rebounded and finished off only a bit below its start. The panic was over by noon that day as the "organized support" of a cabal of Wall Street bankers moved in to support prices. They held steady on the Friday after Black Thursday and were off only a little on Saturday (a normal but shorter trading day at that time). Then on Monday the real disaster started to unfold. Free fall and no closing rally. On Tuesday, the worst day in Wall Street history (until now). Then a brief respite on Wednesday, with a gain. But for the next two and half years, the direction was generally and inexorably down as the country and the world went over the cliff into the abyss.
Will this be averted this time around? Possibly. Possibly not. Those at the helm don't seem to have a firm grasp of the situation, which is not surprising since people are still arguing over what caused the 1929 crash. Those with their hands on the wheel in the US seem more than usually clueless. They've wasted three weeks in to staunch the bleeding that many predicted wouldn't work. The obvious approach -- essentially to nationalize the banks -- was not taken for ideological reasons. Europe, led by the UK, has now taken the lead. If we are lucky, the US will be a tardy follower.
There's no doubt there are a lot of smart people involved in trying to figure this out. But the system is extremely complex. We have a hard time predicting the weather and the arguments over climate modeling are well known. But it turns out that scientists are in agreement that we have better models for climate than we have financial models:
With Wall Street's vaunted financial models looking shaky, could other models of complex systems -- say, the climate models that underpin our understanding of global warming -- have similar faults?
In two words, say scientists and financial engineers: not really. It turns out that it's much harder to model human sentiment, the basis of value, than particle interaction.
"It's the physics. The issue is that economic models aren't based on any underlying physically observed facts. They're based on people's feelings," said Gavin Schmidt, a climate modeler at Goddard Institute for Space Studies. "We're not having a climate crisis because there's a lack of confidence in water vapor." (Wired)
These are the models that the technicians use to price the derivatives and other complicated financial instruments that are part of this unholy global mess. But we are much farther ahead with climate models which have a sound scientific basis:
"Climate models are very complex but you more or less understand the basic physics or chemistry," said [Emanuel Derman, a physicist turned financial engineer, who teaches at Columbia University]. "[Finance papers] look like physics but a lot of the similarity is syntactic more than semantic."
For example, stock options are priced with the Black-Scholes model, which says that stock price movement can be seen to move like the random movements of particles suspended in a liquid, i.e. Brownian motion. But stock price models differ from particle models because they describe the aggregate actions of people.
"When you put out a weather forecast, the weather doesn't read your forecast and get affected by it," Derman said.
In other words, Derman argues, in a soon-to-be released essay, the primary difference between physical and financial models is that the accuracy of financial models could be fundamentally unknowable. No test can really validate how they works.
"The gap between a successful financial model and the correct value is nearly indefinable," he writes.
Functionally, the ability to generate returns determines how useful a financial model is.
"What then is the test of the [Black-Scholes] model?" asked Jeremy Bernstein in a prescient 2004 Commentary article. "Presumably, it is that if one uses it as a guide to buy these options and, as a result, goes broke, one will be inclined to re-examine the assumptions. Presumably." (Wired)
I guess we aren't presuming this any longer. Or shouldn't be. But who knows? The same people who are climate change deniers because they don't believe the science-based models are often the same people who were willing to put financial policy in the back seat while much worse models were running the autopilot.
Now we can have the best of both worlds: climate change and global depression. Lucky us.
- Log in to post comments
More like this
The key thread in denial of both systemic problems - AGW and the current financial melt-down - is an approach to situations that are based in ideaology not in analysis. Ideaology says that markets should be left alone because it is not in their interestes to melt down because then they can't make any money. Ideaology says that humans can't be causing global warming because the earth warms and cools. Neither allows any room for analysis of the actual facts on the ground.
Thing is the Black-Scholes model was tested and failed back in 1998 when Long Term Capital Management went bust. One of the authors of the Black-Scholes model was one of the principals of LTCM. I guess they don't take back your Nobel Prize if your ideas turn out to be losers.
One of the problems with the model is exactly what you stated - the model assumes that the model itself will not affect future inputs in the model. That's only one of the problems with it though. Another is the fact that the model assumes the future will be the same as the past (past few years - not the whole big crazy past). There are actually more problems with it that just that, but I'll leave the rest to the finance professionals.
Revere, since you deal in models, you may be able to understand the Black-Scholes model much better than a regular no-model person like me. One of the questions I always had about it was the valuations are based on probability. But from what I know of statistics I would see any valuation based on a probability totally breaking down as soon as you found the probability incorrect. I'm not putting that very well, but, in other words, as soon as an unknown event occurs (a fat tail event or what they are calling a black swan event - some event that was not factored into your valuations), the valuations would go to zero rather than simply decrease in value. Much like after a dograce, the bets on losing dogs are worthless. I don't know if this has anything to do with the current financial crisis, but it's a question I've always had. Sorry, I shouldn't be posting musings about this on your site, but since you brought it up...!
I love your blog. I read it every day. (I'm a librarian.)
Thanks,
Steph
Paulson is doing little more than rearranging the deck chairs on the Titanic; shuffling the money from one set of financial intermediaries to another does not increase either liquidity or solvency. It merely delays the problem for a few brief moments.
We are now living in a nation where Federal social and economic programs are our government, where our currency status hinges on the stock market rather than the gold standard and where our current President has taken more liberties with the United States constitution than any other President in our country's history.
We've grown accustomed to, and quite comfortable with, large Federal government and the act of re-shaping laws for convenience.
And ... It is beyond me why there are not more conversations about Congressman Dr. Ron Paul here.
http://www.pr.com/article/1110
Ron Paul: Washington's True Maverick Talks Bailouts, the United States Constitution and Re-Making the US Dollar
By Allison Kugel, Senior Editor - October 15, 2008
Often called the most honest man in Washington, Ron Paul, the unsinkable Congressman from Texas's fourteenth district, has positioned himself as resident watchdog for the United States Constitution. Ron Paul holds a point of view that, although echoing the wording of our nation's founding fathers, rings downright alien to the United States government of today.
...................
Steph: You prove once again how well informed and smart a readership we have here. Yes, I'm a modeler, but not a financial modeler. I asked someone more familiar than I about these and he had some interesting comments. Essentially the B-S pricing model is used to predict future values of things otherwise very hard to price (like options) on the basis of market volatility. You are not modeling the expected value but the variance. These models use probability concepts (like Martingales) and partial differential equations (PDEs) and are very complicated (PDEs are ntorious for their intractability). There is an old adage of modeling (due to George Box) that "all models are wrong, but some models are useful." But even the useful ones can be wrong in some circumstances. B-S models, like most economic models, assume efficient markets with rational actors, but panics are not rational actor events. This is a little different than the LTCM episode where the problem was a twelve sigma event, not a model specification failure. But what do I know? It's not really my field. So good question. Long non-answer.
In fact, the financial execs in question used somebody else's money to leverage the buy, and got big fat bonuses, even when the financial institutions they worked went bankrupt. At the top, many of the people who made the bad decisions have not 'gone broke'. Nor have they lost any money. They profited immensely. When an investment bank goes belly up, the 'assumptions' of the execs are not 're-examined'. There has been no accountability.
"the B-S pricing model"
You said it all with that one little phrase.
BS is right.
Thank you, Revere! I now have some new terms to look up and new places to dig into (just my cup of tea!). Thank you very much!
Although Ron Paul can be relied upon to vote against most forms of corporate welfare, he's also been a big promoter of the deregulation woo that got us into this mess. Ron Paul has also been a big promoter of the historical delusion that metal-based currencies are not subject to hyper-inflation. There were several episodes of hyper-inflation during the 15th-17th centuries, and all of the nations involved used metal currencies. (There also many hyper-inflation episodes in the 18th-19th centuries, although most of the involved currencies were not directly metal - instead they had legislative links to metal.) The only good thing that can be said about Ron Paul with respect to the present financial crisis is that when the Glass-Steagall (which had long prevented the merging of deposit and investment banks, and prevented the grotesque over-leveraging, both of which played fundamental causative roles in this disaster) act was repealed, he voted 'present' instead of 'yes'. (The tiny number of 'no' votes all came from liberal democrats, but most democrats voted 'yes'.)
If you look at the curves of the different markets, they track each other very precisely.
http://markets.on.nytimes.com/research/markets/overview/overview.asp
The DOW, NASDAC and S&P 500 curves are virtually superimposible (except for the magnitude). That is the minute ups and downs are virtually identical.
The 3 different exchanges all have different stocks on them. The virtually identical tracking is occurring in the values of stocks that are completely different. Does the real abstract value of those different stocks change in real time so that the aggregated measurements just happen to track each other so precisely? No, what the aggregated measurements are tracking is the perception of value.
It isnt an invisible hand that is causing those stock values to track each other so precisely, it is computer generated trading based on something. The price fluctuations due to the computer generated trading dominate fluctuations due to anything else, which is why the tracking on the 3 exchanges is so precise. The models those computers are using to trade on is the dominant source of fluctuation in the markets.
Daedalus2u,
I've noticed the similar tracking in the indexes and in individual stocks (in unrelated industries) for some time, but I hadn't been able to figure out what was causing the correlation.
One other question I've been puzzling over is VOLUME. Why did the volume of trading (and also the valuations) jump higher (and stay higher) starting in the 80s and then again higher (and stayed higher) in the late 90s. The jumps don't seem to correlate to money supply changes only. My guess? The 80s jump in volume was due to the use of computers (though why or how I don't have figured out) and the late 90s jumps is because the ISDA (the derivatives group) came out with standardized forms for creating derivative products. My interest in this was piqued because no one appears to have written on this phenomenon at all that I could discover. Thanks for you insights!
For a visual on what I'm talking about just look at this chart (all charts are similar!):
http://bigcharts.marketwatch.com/quickchart/quickchart.asp?symb=DJIA&si…
or
http://tinyurl.com/6e3u4s
Thanks again for your insights - and thank you Revere for allowing this digression!
Something I don't understand about the present situation is why the US dollar is maintaining or even increasing its value. A commentator on the CBC said it was because in times of crisis the dollar is seen as being safe. Could someone explain to me why the dollar is seen as safe when US debts, both government and personal, are increasing rapidly?
It is my understanding that the trend is to buy and sell stocks solely on economic models that do not include very much about the actual company being bought and sold. That is how the quants buy and sell, solely based on mathematical models. The economic fundamentals of a company don't change much in a few minutes, a few hours or even in a day. What changes are people's feelings.
I see short term buying and selling more as gambling, playing a game such as poker (which is a game of skill because it is against other people) or horse racing (which is also a skill because it is against other people). That kind of game is a zero sum. For every winner there is a loser.
Some amount of that is ok because it provides liquidity. Too much of it makes the market unstable and subject to manipulation, being gamed by players to extract value from other players.
What Warren Buffet does is invest, not gamble.
Question for y'all to ponder:
It's often said that Wall Street runs on greed and fear.
How far down the phylogenetic scale do you have to go to find organisms whose behavior is similarly motivated? "Eat when you can, run when you must, and that's all folks."
Yes, that's right. Human society is dependent upon an economy that is dominated by motives that are purely reptilian.
Humans have subordinated themselves to reptile values and reptile behaviors.
And speaking of climate change, in light of our subordination to reptile values, is it really any wonder that we're killing the planet?
I'll have more to say about that in a forthcoming essay in another forum.
g336, you are correct, it is mostly reptilian values. Most reptiles don't take care of their young the way that all mammals do. Maternal bonding is the archetypal social interaction, one that all mammals exhibit, even those mammals that are non-social.
But some reptiles do care for their young, alligators care for baby alligators until they become a certain size, then they become fair game and are pursued as prey.
Another theory on investing might posit that a significant amount of luck is involved. Hard work is certainly important as is diligence in making investments but luck can be very important. Who knows where the next microsoft, google, drug breakthrough (viagra was actually discovered as a side-effect of high blood pressure medication)will be.
When things are going well successful people like to talk about how smart they are. When things go badly they often point to things beyond their control.
If you think the US market ,or pick any other market, will rebound and want to have a stake in it be careful. Don't bet the rent money. Perhaps make a number of small bets on various start-up companies.
It is likely very hard if not impossible to predict exactly which companies are going to do well (now more than other with looming derivative uncertainies yet to unfold) but placing some bets on the general economy may be worthwhile..
"""Before I left England for China in 1936 a friend told me that there exists a Chinese curse "May you live in interesting times". If so, our generation has certainly witnessed that curse's fulfilment.
Hughe Knatchbull-Hugessen, Diplomat in Peace and War, 1949"""
Predictions are much easier to make in retrospect... It's amazing how many people appear to have predicted the current financial crisis..
No one wants to be on an airplane with a blind pilot..
http://kempton.wordpress.com/2008/10/12/global-financial-crisis-in-the-…