In the July 31 issue of The New Yorker, James Surowiecki takes a closer look at the logic behind the recent knocks that Airbus has suffered in the business press. The story nicely illustrates a number of the analytic biases that we discuss so often here on Micromotives. The first is attribution error — commentators see Boeing’s recent successes as an outward manifestation of the skill with which Boeing is choosing and executing its business plan, rather than what could simply be fortuitous timing, or sheer dumb luck. Another issue is recency bias, or the popular fetishization of the new. How quickly analysts forget that from 2001-2005 it was Airbus, not Boeing, who had the lead in new plane orders. Third, we see the law of small numbers in action — Boeing’s seeming victory in the competition for dominance in the passenger mega-plane is a vanishing sample size on which to base any long term conclusions.

Here’s an excerpt of Surowiecki questioning the logic of those predicting Airbus’ imminent demise:

The problem with such prognostications is that they infer basic truths about a company’s prospects from its short-term performance. In fact, present success is often determined as much by context and chance as by fundamental viability. This is particularly true of the aerospace industry, because success is heavily dependent on a small number of big gambles. If you bet right, you look like a genius for a few years, even if the success of your bet was due to factors out of your control. The 787 may now look like Boeing’s salvation, but Boeing built it only after more ambitious plans—for a plane, known as the Sonic Cruiser, that would have been the fastest passenger jet in the air—fell through, partly because of the slowdown in air travel after September 11th. And had Boeing not been in such straits in 2003 it probably wouldn’t have risked the investment required for the 787.

People are generally bad at accepting the importance of context and chance. We fall prey to what the social psychologist Lee Ross called “the fundamental attribution error”—the tendency to ascribe success or failure to innate characteristics, even when context is overwhelmingly important. In one classic demonstration, people shown a person shooting a basketball in a gym with poor lighting and another person shooting a basketball in a gym with excellent lighting assume that the second person hit more shots because he was a better player. This problem is compounded by the tendency to extrapolate big conclusions from small samples, something that behavioral economists call “the law of small numbers.” In the decade or so that Airbus has been a serious competitor to Boeing, this is its first really bad patch, and its difficulties are due mainly to making one bad bet while Boeing made one good one. That’s a minuscule sample size on which to base any kind of conclusion. But this is exactly what we like to do: sports fans assume that a few excellent performances are proof of a player’s underlying ability, while investors assume that a mutual fund’s record over one year is a reliable indicator of the manager’s skill.

Because we underestimate how much variation can be caused simply by luck, we see patterns where none exist. It’s no wonder that management theory is dominated by fads: every few years, new companies succeed, and they are scrutinized for the underlying truths that they might reveal. But often there is no underlying truth; the companies just happened to be in the right place at the right time. In 1999, after all, it was hard to find a business book that didn’t hold up Enron as the embodiment of one important principle or other. Of course, some strategies and structures work better than others, but real meaning emerges only over the long term. Let’s give Airbus a few more years of floundering before we decide that it should be put out of its misery.

Read more: The Fatal-Flaw Myth

UPDATE: Over at Crooked Timber, Henry Farrell adds that studies have shown that The Economist has made the same type of errors in evaluating the relative success of different countries over time, as well:

This applies not only to judgements about the success of companies, but to judgements about the success of countries. A few years ago, the political scientist Peter Katzenstein went through a couple of decades worth of those special issues that the Economist runs on particular countries for his own amusement. He found that there wasn’t any long term consistency in judgement – a country cited as a model of how to create a thriving economy in one special issue might be cited as a prime example of political dysfunction the next time round, and back in the good books a few years later. This isn’t a problem that’s specific to the Economist; it’s a more general one of how the political wisdom on the sources of economic success is incredibly unstable. A couple of decades ago, the shelves were filled with books on Japan Inc., and nasty xenophobic bestsellers like Michael Crichton’s Rising Sun claiming that Japan was going to gobble up America unless it fought back. Before that, there was a lot of talk about Modell Deutschland as the way forward. Und so weiter. We don’t know very much at all about the root reasons why economies succeed or fail, for some of the reasons that Surowiecki cites. Countries too can happen to be in the right place at the right time, and may find their luck running out unexpectedly when conditions change.