Emanuel Derman

Bad economic decisions have nonlinear effects

Emanuel Derman
Jan 17, 2012 20:13 UTC

I have been reading an interesting post by Mark Thoma on how the Fed can prevent the next financial crisis, inspired by the release of minutes of Fed meetings in 2006.

Professor Thoma makes the following good points:

The Fed’s errors can be placed into two broad categories, the failure to ask the right questions before the crisis, and the failure to act quickly and aggressively enough once the crisis began. The first problem had a lot to do with economists’ undue faith in their own models and abilities – the financial meltdown problem had been solved so no need to worry about that – while the second problem is at least partly due to the way in which the public interest is represented on the Fed.

I’d like to add a third error. The Fed’s blithe confidence in their models and forecasts influenced society’s behavior. Therefore, not only were their models wrong, but their unacknowledged wrongness arguably made things worse.

In the social arena, wrong models have psychological consequences that affect the things they are modeling. Nonlinearity is built into the system, and is especially large and significant when forecasts made using the models are widely disseminated from the top.

How is belief in the Fed’s current projections affecting the projections themselves?

Probably maybe

Emanuel Derman
Jan 9, 2012 15:37 UTC

When I say that the probability of throwing 3 heads in succession is 1/8, that refers to the fact that if I toss three coins a large number of times, I believe that the number of times I get three simultaneous heads will asymptotically approach a ratio of 1/8. Each individual throw is of course governed by well understood mechanical laws, but the sequence of uncontrolled tiny effects at the start of and during each coin toss effectively produces a pseudo-random result.

When I say the probability of Ron Paul getting elected is small, I don’t refer to an ensemble of identical events at all. Randomness doesn’t come into it either. I think that what I really mean is that I CANNOT EASILY IMAGINE A PLAUSIBLE DETERMINISTIC SERIES OF EVENTS THAT WILL CAUSE THAT TO HAPPEN.

These are two very different uses of the word probability. There should probably be a different word for the second use. All of this makes me increasingly suspicious of the use of probabilities in describing societal events.

To me you’re a wave, but to myself I’m sometimes a particle

Emanuel Derman
Jan 4, 2012 17:01 UTC

In quantum mechanics — i.e. in the real world as we understand it today — matter can have two kinds of formerly apparently contradictory qualities.

The same applies to people.

Matter propagates as a wave but materializes (for observers) as a particle.

Similarly, people can be both free and yet enslaved, depending not only on circumstance but on the observer too.


When I do something GOOD, I like to say I acted freely, and I experience it that way.
When I do something hurtful or BAD, I sometimes excuse myself by saying I couldn’t help it (meaning I experience the cause as compulsion, provocation, reaction, environment, upbringing, parents, circumstances …)

All I want for New Year’s …

Emanuel Derman
Dec 30, 2011 15:33 UTC

Lately it has become fashionable to disparage intuition in favor of careful statistical analysis.

Me, I’m still a fan of intuition. Intuition isn’t merely quick thinking. It’s an arrow to the heart of things. Sometimes, maybe often, it’s an arrow that misses, but when it doesn’t it’s the deepest form of perception. Once the arrow tells you where to look, slow thinking will help confirm or negate its mark. But without the arrow, there’s no target to think slowly about.

Dave Edwards recently wrote:

There are three ‘styles of knowing’:

1. Euclidean: Basic principles are clear and precise and consistent with one another; examples: Euclidean geometry, classical mechanics, von Neumann’s quantum mechanics.


Emanuel Derman
Dec 29, 2011 19:08 UTC

I have been visiting family in Hong Kong for ten days, a place I’ve been to many times before for business, but always so briefly that I never really paid much attention to anything other than work and getting a run in.

This time, in Hong Kong Island, Kowloon and Macau, the thing that has impressed itself on me through airports and downtowns is the bombardment of luxury advertising: Vertu conspicuous consumption cellphones, fancy hotels, apartment complexes, jewelry …. And the presence of luxury stores in profusion. Kowloon main street is a mile-long luxury mall, block after block jammed with super designer stores, multiple Tiffany’s, literally several Chow Tai Fooks per block, etc. These stores run at constant density for a long long distance and then fade rapidly into a few midscale stores for only a few blocks, and then suddenly you’re into Yau Ma Tei. The slope is very steep; one minute it’s luxury, then it’s hotels advertising two-hour rooms and some areas that look like Blade Runner’s LA.

I’m only sampling, not doing detailed statistical analysis, so forgive the exaggeration. But what strikes me is the rapid gradient from oligarchic wives’ appeal to very very survivalist stuff. In New York it feels to me as though the big ads are for H&M, Zara, Banana Republic … Uniqlo has bus-stop ads with Susan Sarandon touting cheap cashmere sweaters, aiming at old middle class people trying to buy cheaper nice stuff. Here, I don’t see ads aimed at middle class people. I see ads and stores aimed at people with lots of money to throw away, or people who aspire to that. It’s a little sad.

Entranced by causality

Emanuel Derman
Dec 28, 2011 15:35 UTC

Nassim Taleb highlighted economists’ and mankind’s ability to be fooled by randomness. Part of the reason economists and traders are likely to be fooled is because, underlying everything they do is the statistics of what is ultimately human behavior.

I’ve always had a guilty feeling about disliking statistics. When I came to graduate school in physics I almost didn’t take a course in statistical mechanics, a beautiful subject eventually taught to me in a class by T. D. Lee that turned out to be one of the best courses I ever took. I was put off by adjective ‘statistical’, which I misunderstood and therefore scorned. I believed in mechanics; I wanted explanations and I thought statistical mechanics would dodge them.

I was wrong. Statistical mechanics explains the properties of macroscopic matter from averaging over the microscopic properties of its constituents, and, vice versa, deduces the qualities of the microscopic constituents from macroscopic behavior.

Paraphrasing Hayek

Emanuel Derman
Dec 14, 2011 19:38 UTC

Self-schooled in finance and unschooled in economics, I was very glad to be sent a link to Hayek’s Nobel Prize acceptance lecture.  I’ve read around him, but not much in the original. One of these days …

As I read it, one of his many important points is the following:

Traditional quantitative science models rest on a basis one level down of either accurate theories (Maxwell’s equations, say, as I mention in my book) or reliable heuristics. In contrast,  quantitative economic or social science  models have no reliable base to rest on; neither theoretically accurate nor empirically reliable ones. And, at the bottom of the social science chain are  individuals, about whom very little that is quantitatively reliable is known. With individuals at the bottom of the tree, models of complex phenomena in society are going to be wrong. And therefore, social planning based on such models will inevitably go badly awry. Therefore, don’t theorize and act on the wrong theories of complex things you didn’t create and don’t understand.


Too Much Meta

Emanuel Derman
Dec 12, 2011 01:43 UTC

I have a friend who, when you tell him about something bad that happened to you (“I fell and scraped my knee”) doesn’t address the problem at hand, but goes one level higher (“Yes, well that’s what happens when you run in the presence of a gravitational field”). I call that going meta, and I’m not crazy about it.

I thought of this when I came across a post by Felix Salmon about consultants. When I worked at Goldman Sachs was when I first met professional management consultants, and what most impressed me (Not!) about the ones I met was that they were meta guys with meta-skills rather than actual skills. Few of them lasted long because they weren’t hands on; they were often mere human multiplexers who told you where to go for information rather than how to actually do something.

To go meta productively, you need skills and experience of a practical nature from which you can abstract. To be a boxing coach, you don’t have to be a great boxer, but you should have boxed. To be a programming consultant, you should have been a reasonably good programmer. To be a management consultant, you should have been a manager, made mistakes, recovered from them. I don’t know what management consultants learn in school but I suspect it’s formulaic, and that’s the problem.

Rainer Werner Fassbinder on contagion

Emanuel Derman
Dec 5, 2011 13:14 UTC

A couple of nights ago I watched “Lola,” one of the last movies made by the immensely prolific Rainer Werner Fassbinder, born in 1945 and dead at age 37. I’d seen a bunch of his movies in the Seventies, particularly liking “The Merchant of Four Seasons,” a sad sad movie of inevitability that reminded me a bit of Bresson’s “Au Hasard Balthasar.”

“Lola” is executed with great panache, and loosely relevant to economic affairs today. It’s set in a small town in the boom of 1950s West Germany, and is about contagion, the moral kind. There is a business man who does property deals that involve minor amounts of corruption and bribery for the sake of profit. There is a regulator who wants to get things done and is willing to bend the rules for the sake of the greater good, the trickle down, but is otherwise morally kind of upright to the point of extreme primness and naivete in his personal life.

The businessman, a bit like Tony Soprano, runs a whorehouse/sex club/nightclub and enjoys everything that goes with it. The vector of contagion is Lola, his mistress, who, savvy, ambitious and maltreated, temporarily seeks respectability. Eventually, the movie seems to say, you can’t confine corruption.

On self-destruction

Emanuel Derman
Dec 1, 2011 19:53 UTC
I am thoroughly tired of Europe, America, Obama, the Republicans, banks central and peripheral, and everyone’s indignation about the above. And of my own indignation too. It’s affecting my quality of life. Part of this is owing to having worked much too hard and too obsessively over the past year, always trying to do one more thing to get it out of my in-tray. I got to the point where when I woke up in the middle of the night my mind began to race with thoughts of things that “needed” to be done and it became preferable to tackle them rather than try to go back to sleep, even though I knew it was a bad idea. And then, the other part is that the world news really is consistently disturbing and unfair. With this in mind, I have long admired/been jealous of people whom observe the Sabbath, whatever religion. My father often used to claim that the greatest historical contribution to humanity by the Jewish people was the idea of a day of rest, which appears right at the very start of Genesis. When you think about it, having a periodic day of rest isn’t obvious at all. I can’t say my father actually observed it most of the time, very rarely actually, but I see his point. It must be wonderful to rigorously observe one day as a retreat from the world of business, sport, competition, and effort, with intent and habit. I know a few people who do that (actually only one at this point, and he, I suspect, spends most days that way, so maybe it doesn’t count). One day, if I live long enough, I hope to try that myself. It’s quite clear that resting is very antithetical to life today. When I left New York for England in 1975, it was hard to do much of anything on a Sunday. Driving around the city was was a breeze. Shops were closed. And London too was a ghost town after 2pm on Saturdays. When I came back to NYC in 1977 Sunday shopping had begun, and department stores were open for a few short hours. Now Sunday traffic can be the worst of the week, in New York or in London. Not to mention the restless internet. In a way, the world has become aperiodic. You can do anything any time, denying the rule of the sun and the moon and the earth’s rotation. You can even have your periods aperiodically. One big part of this aperiodicity is an attempt to deny time and its working its weary way on you. But to be fair, whereas animals go into heat periodically, adult humans can mate any time, so maybe this tendency to aperiodicity was built into us by evolution. All science and discovery can be regarded as an attempt to short-circuit time by either figuring out the end of the story before Nature gets you there, or trying to prevent Nature from nudging you away from the beginning or the middle. In normal times, when one is very exhausted, either happily or sadly, sleep is a blessing to be embraced, and one yields gratefully to unconsciousness. Re true self-destruction: I have never been tempted. But, occasionally, maybe two or three times, I’ve been in situations where facing reality has been too unpleasant. Neither books nor movies nor people could distract me from what I didn’t want to face. Only sleep brought relief, in the form of a loss of identity and consciousness, a kind of temporary annihilation. Then you sleep, and if you can’t then you drink or take a sleeping pill until you can. That kind of urge for unconsciousness has always been  very occasional and temporary thing for me, Thank God, but by vast analytic continuation I can imagine some extreme form of that urge for unconsciousness and need to avoid oneself. That, I can imagine, would make permanent loss of consciousness a compulsive attraction. It wouldn’t be a plea for help, just such a strong desire for not being that one has to give in. It’s a kind of self-loathing, I would think. I wonder if there isn’t some more final version of this latter feeling, or even the former mere exhaustion, which would constitute the preconditions for a good and timely death.