Sunday, May 27, 2007

Economists are from Mediocristan

Mediocristan is a mythical land inhabited by economists (and other social scientists) who believe that the world's events fit neatly beneath a bell curve of outcomes. Economists live in this land not because they are mediocre--in fact, they are decidedly less than mediocre in their prognostications--but because they accept one very crucial idea. They believe that extreme outcomes such as market crashes and other major discontinuities in our economic life are so rare that we can all but ignore them.

Mediocristan is an invention of the mind of Nassim Nicholas Taleb, a hedge fund manager and self-styled philosopher of uncertainty. His most recent book, The Black Swan: The Impact of the Highly Improbable, provides an entertaining and thoroughly readable critique of how most of us have been taught to think about risk.

Taleb doesn't address oil depletion, global warming or other resource or environmental issues directly. But much of what he says turns out to be useful for those trying to assess risks in such areas.

Economists often like to insist that risk can be easily quantified and that future economic scenarios will be played out within the bounds dictated by models. But risk cannot be easily quantified except in cases where the possible outcomes are already known. An example would be a game involving dice. There are only six possibilities on each die so the range of outcomes (assuming perfect dice) is already known. As for economic models, they are usually backtested which means that they are tested using historical data. First, historical data, probabilistically speaking, represent only one possible outcome among many outcomes. Second, historical data are very unlikely to reflect the dynamics of world peak oil production or global warming since peak by definition happens only once and the effect of global warming on the economy is a wildcard for the future.

The admission that economists are from Mediocristan is sometimes made explicit. This piece lays out many of the arguments frequently used by economists when discussing world peak oil production. But rather than focus here on the counterarguments, let me call your attention to the following sentence in that piece: "Remember that barring any unforeseen tragedy or deliberate measure to limit the supply of oil, the supply will not drop suddenly, meaning that the price will not rise suddenly."

It is precisely this type of risk that peak oil theorists are pointing to, risks of an extreme outcome that could be very disruptive to modern society. The sentence quoted above explains in part why economists and those concerned about peak oil often talk past one another. Economists start by discarding extreme outcomes. Those concerned about the possible severe consequences resulting from peak oil focus on extreme outcomes.

So then, who's right? The rather unfortunate answer is that both are right if you stay within the confines of their assumptions. So you need to figure out which assumptions better reflect the world we live in. The economists are assuming that no unforeseen, disruptive event will occur that causes a dramatic and relatively sudden fall in oil supplies. (They are also assuming rather better knowledge than we actually have about oil reserves.) Those concerned about peak oil think we should look for clues foreshadowing such a disruptive shortfall and take out some insurance in the form of concerted action on conservation and the development of alternative fuels. In other words, most of those pounding the table about peak oil do not believe that the marketplace will solve the problem.

Using Taleb's words, the arrival of peak oil in the near future could be considered a "black swan." The black swan is a metaphor for a rare and unforeseen event that has an extreme impact. Of course, a fair number of people now believe peak oil is imminent. But the vast majority of the world's population knows nothing about peak oil or believes that it is something for the distant future. And, it is expectations among the populace at large that make the impact of such an event extreme (since very few people are, by definition, prepared for it).

Taleb has admitted to making a bet on peak oil, but not one that most people would expect. In keeping with his principle that we cannot really know the future, but that we can expect people to underestimate the number and impact of extreme events, he has bet on two extreme outcomes: very cheap oil and very expensive oil within the next few years. (He does this using the options market, and his record shows that he does not have to be right about any one bet very often to do well financially.)

By doing so, he at least demonstrates a principle that lies behind the thinking of peak oil believers, namely, that it's worth taking out insurance against seemingly unlikely events if their impact could be very severe. And, that places peak oil believers in the other mythical world which Taleb calls Extremistan. In that world unforeseen, extreme, but still rare events happen more often than anyone can calculate. Their consequences are so large that they alter considerably the previously assumed average results for such events.

In the world of easy-to-tally physical measurements such as height and weight, the chance of extreme outcomes that can alter the average is quite small, Taleb explains. We are very unlikely to find a human being who is 100 feet tall or who weighs 10,000 pounds. This is the proper place for using techniques from Mediocristan. But in the world of social and economic affairs where so many factors are unknown and the dynamics of the system are so unruly, exceptional and unexpected events are the ones we should be thinking about because they are the ones that change the course of history. (Oil depletion must be classed within this realm since economic trends, political decisions and technological change as well as geological constraints are all important factors in analyzing it.)

Taleb adds that, by definition, no one can forecast a "black swan" event. Its very nature is that it is unexpected. But he believes we can acknowledge the existence of such events and try to mitigate the damage of "bad" black swans while exposing ourselves to "good" black swans. In the case of oil depletion, a good black swan might be an invention that makes an electrically powered transportation system very easy to deploy and run.

Taleb tells us that financial models based on the bell curve would have predicted that a stock market drop similar to the 1987 crash would occur once every several billion lifetimes of the universe. And, it is from people relying on such models, who believe in the magic of the marketplace (under ideal bell-curve conditions with no "unforeseen tragedies"!), that we now get confident pronouncements that peak oil is nothing to worry about, even if it's here.

Somehow, I don't take any comfort in that.

2 comments:

Ken said...

Hi.
Thanks for this post I listened to a podcast ( http://www.econtalk.org/archives/2007/04/taleb_on_black.html )and was left asking for more and your post has helpd clarify alot.
I havent read the book yet but Nassim Taleb is one thinker I haved really come to respect.

Peter Cotton said...

The Extremistan/Mediocristan distinction was laid before us in The Black Swan and can only be interpreted as meta-guidance for applied mathematicians. Actually it was meta-nonsense for people who had never applied mathematics, never intended to apply mathematics, but wanted to think deep thoughts about other people who applied mathematics ... but let's set that aside and examine it on it's merits.

Taleb presented a perfectly clear instruction manual. The world has linear and non-linear phenomena. But mathematicians have to be careful, he argued, not to use the wrong kind of mathematics. You wouldn't want to use linear mathematics to model a non-linear problem best treated with fractals, power laws and so forth now, would you? And I suppose you wouldn't want to use fractals to model a linear system either - though Taleb is highly skeptical that any exist so that is a decidedly less important topic.

Take web page popularity. It follows a Zipf distribution, landing it squarely in the groovy world of power laws. So it would be a massive mistake to try to apply linear mathematics to it, right? To me more precise I'd say it would be a massive mistake, a violation of a universal principal no less, to apply Linear Algebra to it, much less a singular value decomposition which is about as close to the epicenter of "linear mathematics" (to humor that ridiculous phrase) as one could surely get. Strange then, that Larry Page sought fit to do so when inventing Page Rank, the algorithm distinguishing Google from Yahoo that powered the greatest commercial success in recent history.

Peter