Sunday, July 04, 2010

The complexity gurus and our margins of safety

Human societies have long relied on specialists to help in complex matters involving the natural and human-built worlds. Ptolemaic astronomers used what would seem to us moderns as a needlessly complex Earth-centered system to explain the heavens and predict celestial events. And yet, they were amazingly accurate. Complex societies of the past have employed specialists in war, statecraft, engineering, agriculture, shipbuilding, and a variety of other tasks that would be difficult to accomplish without in-depth knowledge.

And, yet these specialists typically lived not within societies that were managed along completely rational principles. Instead, the role of religion was far more prominent that it is today and tightly interwoven with the workings of the state. In recognition of the irrational and unknowable in human affairs, the gods were invoked to help steer events which humans were incapable of fully understanding or controlling.

Fast forward to today. The modern managers of the rational state rarely invoke the divine except for ceremonial purposes. They imagine that they can model the world of social and natural events accurately enough to understand and control it. But today these managers enlist the help of an entire class of theorists and researchers who do nothing but study the complex systems we rely on. And, it has become a matter of faith that this class of complexity gurus can amass enough understanding to overcome the vast uncertainties that drove ancient humans to seek help from the gods.

Recent events should give us pause. Daniel Yergin, the ever smiling guru of world oil supplies from Cambridge Energy Research Associates, is regarded as perhaps the most influential energy analyst on the planet. We cannot say what his private counsel to clients has been. But we can judge his public pronouncements about supplies and prices for the first decade of this century. That record is miserable. A site called "The Sad Record of Daniel Yergin and Cambridge Energy Research Associates" details his and his firm's prognostications.

We should not expect one person to be omniscient. But if he pretends to be omniscient, and we go along with it because we don't have expert knowledge, then we are only doing what most people are simply forced to do in our society: Defer to specialists because we have neither the time nor the inclination to master the subject at hand. But people in high places are making public policy and important corporate investment decisions based on such advice. And, that affects us no matter who we are.

Also in the last decade, Wall Street wizards told us that the complexity of the supposedly new financial instruments they cooked up were simply a response to a more complex world. Perhaps. But even these wizards failed to understand the risks involved. Alan Greenspan, former U. S. Federal Reserve Bank chairman, admitted as much in congressional testimony in 2008. Prior to that Greenspan and his fellow travelers in the financial community put their faith in ideology, not the gods. The so-called free market would sort everything out and keep everyone within bounds. Instead, a bloodbath ensued that still threatens the viability of the world economy.

Finally, the promoters of offshore drilling managed to convince themselves and others that no untoward major spill could come from the advanced technology now available. It was, of course, full speed ahead until the BP well blowout in the Gulf of Mexico. There were concerns prior to the blowout. But in the arcane world of offshore drilling, few outsiders were qualified to judge the risks of such drilling properly. As far as I know, there was no public pronouncement that a blowout on the scale of the BP one was likely before the accident.

What should be taken from this record of the complexity gurus? Certainly, they have technical knowledge which few can match. And, they keep track of developments in their respective areas much more thoroughly than nonspecialists. So, given that, why are their judgments frequently so mistaken?

First, no one can know the future, not even well-informed specialists. Second, specialist knowledge by its very nature tends to crowd out a broader and more nuanced view of the world. The perennial optimism of technophiles is alluring. But a good dose of history and literature would teach the specialist that in the affairs of men and women, things often don't go according to plan no matter how carefully constructed that plan has been. Third, and probably most important, is that the systems we depend on have become so complex that we simply cannot know all the ways that they can go wrong hampering our efforts to prevent accidents or outcomes with severe consequences.

Should we do away with complexity gurus? In a word, no. But our expectations of them need to be far more circumscribed. We must simply stop taking their word for it that the risks inherent in our modern systems are well-understood and under control. And, we need to widen greatly our margins of safety to account for what we do not or cannot know.

4 comments:

  1. Kurt, I agree that the word of specialists can't be taken as the only word. They are experts within their field and have an eye for what's possible on a small scale but aren't very good at estimating the ability to scale or the results of interaction with other factors outside of their area of expertise.

    It's not much of a surprise that they are overly optimistic. The gatherings and conferences they attend put forth a hyper-optimistic attitude at the creation of knowledge for its own sake. The only ones that seem to come out with a negative outlook for the future are philosophy and humanist conferences.

    But we seem to have the supposed solution in place already. People who set policies usually aren't specialists, although they do consult with specialists. It seems to work well in a lot of areas, but obviously not in the area of energy policy.

    Or, rather, it works well until it stops working well. People who are pessimistic about energy assume it will stop working well at some point, but in the present and in the minds of people who don't look much into the future, it works well. They can still gas up their car, buy their food cheaply, and live in the suburbs. Fuel price increases motivate incremental changes in behaviour.

    There is probably a lot of faith in the free market to take care of it and find the solution. I'm not sure that's a wholly bad idea because the outcomes are very uncertain and it threatens the stability of society when people are told they must behave in one way when it's not obviously apparent that they must.

    Specifically on the financial instruments issue, I have heard the Black Swan guy quite repetitively going on about how banks don't understand their risk and in particular criticizing the notion of "value at risk" measurements in approximating the risk position. He ignores the fact that Canadian banks (which did not suffer much in the recession) use the same measurement to estimate their risk, but regulators here have a more stringent framework around the interpretation of this information.

    ReplyDelete
  2. Anonymous2:28 PM

    Daniel Yergin was a co-author of a book called :--> Energy Future : Report of the Energy Project at the Harvard Business School.

    Curious how his views changed, huh ?

    ReplyDelete
  3. We should not expect one person to be omniscient.

    Now there's an understatement... What's the word for someone who's always wrong? "Antiscient"?

    But even these wizards failed to understand the risks involved.

    Here, I'm less convinced. A lot of those "wizards" were shorting their own products, even as they were talking them up to the marks. Sorry, "investors"...

    ReplyDelete
  4. scaredwitless3:44 PM

    Couldn't the same blistering post be made about the dire predictions of AGW doomsayers and the end of the world as we know it. I have an article in my "save and see if this scary stuff will come true" file that predicts dire floods in Bangladesh in 2007 and mass migration in the ensuing years. I suppose its coming.

    ReplyDelete