Failure Happens


Evolution, Extinction & Economics

By Paul Ormerod

Pantheon; 255pp; $24.95

( below)

Editor's Review

Three Stars
Star Rating

The Good A sometimes eloquent examination of why failure is so common.

The Bad There's little nuts-and-bolts advice in this big-idea book.

The Bottom Line Why Most Things Fail largely succeeds.

Paul Ormerod, an unorthodox free-market economist in the tradition of Friedrich Hayek, made a surprising discovery a few years ago when he compared the failure rate of businesses with the extinction rate of species. In both cases, instead of an even flow, there were long stretches with few extinctions, interrupted by huge spikes in the failure rate.

Ormerod, a former British economic forecaster, noticed that these patterns didn't fit the well-behaved bell curve that economists are so fond of. In bell-curve economics, equilibrium is the norm. But he found that mass extinctions of companies and species are not all that unusual. Instead, the patterns fit what scientists call a power-law distribution, which is commonly observed in networks. (The Web follows a power-law distribution: The vast majority of sites have almost no links to them from other sites; a thriving handful attract thousands.) Ormerod and two collaborators published their results in Physica A, a journal of statistical physics.

Ormerod was stunned first by the spiky patterns and second by the strong resemblance between species and companies. Why should the patterns be the same, after all? Corporate executives constantly plan how to cope with the changing business environment. Plants and animals mostly don't plan at all. Yet a company is about as vulnerable to sudden death as, say, some zero-IQ South American fungus.

Why Most Things Fail is Ormerod's explanation of why failure is so common and apparently unavoidable, and what we can do about that unfortunate fact. He says the reason that companies are almost as likely to be blindsided as dodo birds is that the world is simply too complex and nonlinear for anyone to predict more than a short while ahead. What you can't predict, you can't avoid. And the reason that mass extinctions aren't so rare is that species, and companies, exist in networks analogous to the World Wide Web. The complex webs of relationships in networks -- some competing, some cooperating -- can breed chaotic outcomes. Both successes and failures tend to cascade. Winners take all, or most, and losers disappear in great bunches.

What does this mean for the average overstressed CEO? Ormerod is under no obligation to rescue executives from the dangerous world he has identified in this and two earlier works, The Death of Economics and Butterfly Economics. But he does try to be useful, as befits his day job as co-founder of a small but brainy business-consulting firm in London called Volterra Consulting Ltd. For government, he counsels restraint because most interventions in the workings of the market will just make matters worse. For business, he offers two bits of advice. First, keep trying to predict the future because, "Even a tiny bit of genuine knowledge goes a very long way." Second, keep experimenting, because eventually something is bound to work. Here is how he ends the book: "Karl Marx famously wrote that the motto of capitalists was 'Accumulate, accumulate, that is the law of Moses and the Prophets!' As in many other respects, Marx was completely wrong. 'Innovate, innovate!' -- that is the guiding principle...."

Although Why Most Things Fail is no management handbook, Ormerod has his likes and dislikes. Coca-Cola Co. (KO ) impresses him because it had the humility to reintroduce the classic formulation of Coke less than three months after New Coke fell flat. He mocks Long-Term Capital Management, not just because the hedge fund failed in 1998 but because of why it failed. According to Ormerod, its egghead partners, including Nobel laureates Robert C. Merton and Myron Scholes, stuck to a misguided academic faith in order, linearity, and equilibrium. The real world had other plans.

It's fascinating to compare Ormerod's book with another volume on failure, Jared Diamond's 2005 Collapse: How Societies Choose to Fail or Succeed. Diamond argues, contrary to Ormerod, that long-term planning is essential. But on closer inspection, his success stories don't illustrate much truly forward thinking. Example: Diamond shows that Japan under the shoguns staved off a deforestation disaster by curbing tree-cutting and beginning silviculture. But the shoguns' actions were more reactive than proactive. As Ormerod might say, even animals adjust when a resource they've depended on becomes scarce.

Although Ormerod waxes eloquent about the "iron law of failure," Why Most Things Fail mostly succeeds. Please don't hold that against his thesis.

By Peter Coy

    Before it's here, it's on the Bloomberg Terminal.