Law in Contemporary Society

Uncertainty

-- By PeterPark - 17 Apr 2010

According to one Felix Salmon, bad math destroyed our economy. That math looks something like this:

This is called a Gaussian-Copula formula. Simply put, it posits that the probability of two events occurring simultaneously can be determined by probabilities of each event occurring independently and a correlation factor which expresses how closely the two events are related. Easy.

This is probably indicative of my old-school writing style but I think there should be some type of thesis statement that pops out from all of this exposition.

Modeling Uncertainty on Wall Street.

If a Caucasian male in his twenties earning $300,000/yr defaults on his mortgage on a million dollar condo in NYC, how does that affect the chance of an Asian female in her fifties who makes $50,000/yr defaulting on her modest house in Chicago? If someone who earns six times as much defaults on his mortgage, then surely, probability of the woman’s default be up. But wait – maybe white guys making six figures in their twenties tend to be inexperienced and irresponsible whereas middle-aged Asian women are wise and conservative with their money, so perhaps there is no correlation. But then again, maybe the first default occurred because of an emerging trend where Americans are sick of condos - maybe condo prices plummeted while house prices soared, in which case the probability of the woman’s defaulting actually went down.

These lines of reasoning and empirical conjectures - sufficient perhaps for judicial decisions - will not comfort an investor who just bought a bond backed by thousands of mortgage obligations. His bet is that even as some of these mortgages inevitably go into default, the remaining “good” mortgages will be enough to pay him off when he cashes out. If default on one house is strongly correlated to default on the next house, which is strongly correlated to default on the next house, and so on, then there is a chance that the investor will be wiped out as a consequence of the first default. So he needs to know just how interrelated these events are. Less is better, of course.

Psychology of Uncertainty.

Uncertainty is all around us. Although we count on a lot of things in our lives, in reality, we don’t even know for sure that the sun will rise tomorrow morning. We don’t even know if the world outside our minds truly exists. And yet, ours is a history of tackling uncertainties. Lightening was first caused because one god or another was angry. Now it’s because negatively charged particles dissipate by connecting to the positively charged particles on the ground. The world was once flat, with hell below and heaven on top. Now the world is large beyond our comprehension, but at least we can explain its birth, life, and death. We don’t know what a big chunk of it looks like, but at least we know the collective effect that chunk has on us.

Psychologists consistently connect uncertainty to such negative feelings as dread and anxiety. Dread and anxiety elicit fear, which sets off a variety of pathological behaviors. On the one end of the spectrum are people who wear an armor of dogma. They insist on permanence where there is none. To them, the world operates under absolute and unyielding rules. We may not know all the rules, but they are there, maintained by a supernatural being or the nature itself. This is a classic defense mechanism, whereby irrationality trumps having to confront uncertainty and the resulting unpleasantries. By stifling uncertainties, they also stifle open-mindedness and the potential for creativity. On the other end of the spectrum are people who succumb to a solipsistic stupor. Somewhere in between, there are people who dismiss uncertainty for practical purposes while acknowledging it at a fundamental level. Scientific instrumentalists are a good example. So what if one cannot explain how penicillin works? The point is that it works. Also in the murky middle are the statisticians, economists, every manner of gamblers, and of course, psychologists, all of whom seek to tame uncertainty with operating rules derived from empirical data.

Outlining humans’ aversion to uncertainty and their attempts to control it is certainly a necessary part of your argument, but I feel this is a little too wordy and doesn’t do enough to set up your point that these attempts to tame uncertainty (irrationally, or maybe even rationally) lead people into traps.

Back to Wall Street

Modernity gave birth to a new and dangerous kind of instrumentalists. Having gained an unconscionable degree of asymmetrical advantage, these people formed organizations whose sole purpose was to exploit the anxiety of everyone else. Under a countenance of professionalism and expertise, they sold compelling but not legally binding promises to everyone else (See for example, Stocks for the Long Run by Jeremy Siegel). And yet, as individuals, these Harvard-trained math Ph.D.’s were just as vulnerable to the anxiety of uncertainty as the rest of us. It is not surprising, then, that bad math destroyed our economy.

Enter David Li, a math genius turned Wall Street “quant.” Li came up with an elegant solution to the correlation problem above. Instead of thoroughly investigating the qualities of each event and drawing meaningful connections - a prohibitively expensive task - he figured out a shortcut. He tracked credit default swaps (CDS) associated with each event and compared them. If they moved in the same direction, there was positive correlation. CDS is an insurance policy which one investor buys from another, just in case the investment goes bust – it would be a good proxy for the risk of investment, if the primary investors could correctly price the risk. Thus, Li’s model was a fatally circular one – e.g. this derivative investment is not risky because higher order investors said it wasn’t. How did those guys know? They used Li’s model, of course. The model suggested, among other things, that one default was not correlated to another. Therefore, mortgage-backed securities were bullet-proof. As it turned out, the only available CDS data was from the boom years and was not at all applicable during a downturn. In a bad economy, as one might suspect, one default could easily signal an onslaught of others - and triple-A, first tranche, risk-free bonds could become worthless. And they did. Thus began the apocalypse.

The primary lesson here is to be wary of anyone who claims to have tamed uncertainty. Chances are, it's a con and you are the mark.

Was this a con though? I don’t know enough about Li’s intentions, but could this result have simply been accidental (i.e. belief in the ability to tame uncertainty created stupidity on both ends)? Secondly, are you precluding the possibility that people can ever tame uncertainty? If you are, does that make anyone who isn't in a solipsistic stupor a con man?

These are just some questions and initial points that I had when first reading your paper. A lot of these may be the product of my specific editorial preferences or inability to grasp the complexity of what you are saying, but I do feel you have an argument here that is somewhat obfuscated. If there are points that you believe I am missing or misinterpreting please let me know (actually, your input would be helpful). Overall, I think it’s a really interesting topic and hopefully I can give you some editorial feedback and do some reworking that does justice to your ideas. FYI, it might be awhile before I post an edited version.

Uncertainty

According to one Felix Salmon, bad math destroyed our economy. That math looks something like this:

This is called a Gaussian-Copula formula. It posits that the probability of two events occurring simultaneously can be determined by the probabilities of each event occurring independently and a correlation factor which expresses how closely the two events are related. While its creator, David Li, has avoided the blame for the effects of this bad math, the obvious weakness of his equation and the nature of the environment he marketed it to indicates that Li’s formula may have been more of a con than mere errant calculation.

Modeling Uncertainty on Wall Street.

If a Caucasian male in his twenties earning $300,000/yr defaults on his mortgage on a million dollar condo in NYC, how does that affect the chance of an Asian female in her fifties who makes $50,000/yr defaulting on her modest house in Chicago? If someone who earns six times as much defaults on his mortgage, then surely, the probability of the woman defaulting on her mortgage will go up. But wait – maybe white guys making six figures in their twenties tend to be inexperienced and irresponsible whereas middle-aged Asian women are wise and conservative with their money, so perhaps there is no correlation. But then again, maybe the first default occurred because of an emerging trend where Americans are sick of condos - maybe condo prices plummeted while house prices soared, in which case the probability of the woman’s defaulting actually went down. These lines of reasoning and empirical conjectures - sufficient perhaps for judicial decisions - will not comfort an investor who just bought a bond backed by thousands of mortgage obligations. His bet is that even as some of these mortgages inevitably go into default, the remaining “good” mortgages will be enough to pay him off when he cashes out. If default on one house is strongly correlated to default on the next house, which is strongly correlated to default on the next house, and so on, then there is a chance that the investor will be wiped out as a consequence of the first default. So he needs to know just how interrelated these events are so as to understand the risk of his investment.

Psychology of Uncertainty.

Psychologists consistently connect uncertainty to such negative feelings as dread and anxiety. Dread and anxiety elicit fear, which sets off a variety of pathological behaviors. Often this pathological behavior takes the shape of an obsessive need to tame the uncertainties that produce the emotional uneasiness leading to such fearfulness. As Jerome Frank noted, history bears record to different methods that people have used to convince themselves that they can conquer the unknown. In the past, people relied on magic in attempting to tame uncertainty, and now, science has become the tool by which people can seemingly comprehend the once incomprehensible. While there are certainly individuals who have been able to accept uncertainty and disavow any system that attempts to control for it, most people (even those who acknowledge the prevalence of uncertainty in an infinitely complex world) are still attracted to models that purport to vanquish uncertainty. Statisticians, economists, every manner of gamblers, and of course, psychologists, all seek to tame uncertainty with operating rules derived from empirical data. Some of these individuals likely retain enough self-awareness to know when they cannot control uncertainty, but there are many out there who do not, and these people are the perfect marks.

Back to Wall Street

Modernity gave birth to a new and dangerous kind of con. The large and complex global marketplace created an environment perfect for those purporting to have solved its complexities to feed off the anxiety of everyone else. Under a guise of professionalism and expertise, these individuals sold compelling but not legally binding promises to everyone else (See, for example, Stocks for the Long Run by Jeremy Siegel). And yet, as individuals, these Harvard-trained math Ph.D.’s were just as vulnerable to the anxiety of uncertainty as the rest of us. It is not surprising, then, that bad math destroyed our economy. Enter David Li, a math genius turned Wall Street “quant.” Li came up with an elegant solution to the correlation problem above. Instead of thoroughly investigating the probabilities of each event and drawing meaningful connections - a prohibitively expensive task - he figured out a shortcut. He tracked credit default swaps (CDS) associated with each event and compared them. If they moved in the same direction, there was positive correlation. CDS is an insurance policy which one investor buys from another, just in case the investment goes bust – theoretically making it a good proxy for the risk of investment, if the primary investors could correctly price the risk. But Li’s model was overly simplistic and fatally circular – e.g. this derivative investment is not risky because higher order investors said it wasn’t. How did those guys know? They used Li’s model, of course. The model suggested, among other things, that one default was not correlated to another. Therefore, mortgage-backed securities were bullet-proof. As it turned out, the only available CDS data was from the boom years and was not at all applicable during a downturn. In a bad economy, as one might suspect, one default could easily signal an onslaught of others - and triple-A, first tranche, risk-free bonds could become worthless. And they did. Thus began the apocalypse.

Financial experts have been quick to blame investors that relied on Li’s formula, rather than Li himself, for the recent financial crisis. However, numerous mathematicians had expressed skepticism of any formula that claimed to reduce the uncertainties of the market to a few variables. Where does that leave our assessment of Mr. Li? Was this really just bad math, or was Li perpetrating the biggest con on Wall Street? While we may never know whether Mr. Li was consciously preying on the anxieties of investors, he obviously made the most powerful traders in the world look like marks.

This is my take on what you were trying to argue. I was unsure whether you were arguing that Li was a con man or whether Wall Street investors perpetrated a con on themselves (if that's possible). I tried balance the structural similarities to a con that you rightly notice against the fact that many scholars have not blamed Li for the damage his equation caused. Hopefully, this is what you were going for.

Navigation

Webs Webs

Attachments Attachments

r4 - 13 Jan 2012 - 23:34:44 - IanSullivan
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM