Skip to Content

Round-off Error? Hardly!

A brief disquisition as to why–when it comes to innovation–superficial economic analysis is worse than no analysis at all.
October 9, 2007

As pop economists go, Tim Harford is pretty good. His The Undercover Economist is a decent read. His Dear Economist column in the Weekend Financial Times can be quite clever–except when it’s not.

The source of irritation? The cloying glibness of Harford’s response to a simple but important question about voluntary contributions to the Internet. So he dismissively writes:

J. K. Rowling’s books attract thousands of reviews on Yet the overwhelming majority of her readers–more than 999 in every thousand–don’t bother to post a review. Frankly, if 0.1 per cent of people make unrewarded contributions to the internet, that’s just a rounding error away from nobody at all. Economists love efficiency, and it is not very efficient to produce an explanation of behaviour that hardly anyone engages in.

As Glenn Reynolds likes to say, you should read the whole thing. However, what we have here is bad analysis and worse math–the sort of bad analysis and quantitative peabrainedness that annihilates rather than promotes economic awareness and understanding. Let’s start with the math: treating .1 percent of people as a “rounding error” may make economic sense if they don’t do anything of substance or significance. But if they actually create and add value to your network and/or your business, you might be wiser to treat that magnificent tenth of a percent as an “outlier” to be nurtured rather than a statistical aberration to be ignored. Indeed, as Mr. Harford well knows, 10 percent of a thousand is larger than 90 percent of a hundred. If we have–and we do!–Net enterprises with tens of millions of users, then that measly-looking .1 percent can be misleadingly robust. One-tenth of a percent of a million is a thousand, and, of course, .1 percent of 10 million = 10,000. That’s quite a workforce. But why grant Tim his self-serving assumptions? I cheerfully argue that as a “culture of contribution” and creative volunteerism facilitates a vibrant innovation marketplace, that tenth of a percent might really be closer to seven-tenths of a percent or even 1 percent. Again, it’s not the volume of contributors that matters; it’s the value of their contributions. Isn’t that what economics is supposed to be about? Hence my dismay with and dismissal of our Undercover Economist’s “analysis.” The simple truth is that we can look at any industry–automobiles, semiconductors, telecoms–and find only a tiny fraction of “entrepreneurs” who are making real contributions. Now, I’m not for a moment suggesting that folks who volunteer their videos on Youtube or comments on Amazon or story preferences on Digg are entrepreneurs in the Schumpertarian or Hayekian sense. The (obvious) profit motive is absent. However, the community/audience/marketplace clearly sees some value in these contributions–as do the undeniably entrepreneurial hosts. The spirit–and seduction–of “increasing returns” suggests that literally tens of thousands of people will be making value-added contributions to these arenas and that, yes, noncontributing consumers will also extract some value from them. In other words, Harford’s “rounding error” assertion is a bit like saying, “You know those tiny little things that huge trees have called seeds? Well, they’re so small and lightweight, they don’t really matter. I mean, they’re not even one-tenth of one percent the size or weight of the tree….” Just as bad botanical understanding can wipe out forests and biodiversity, poor economic analysis can wipe out value and its creation. The truth is that we’re just beginning to grasp the underlying behavioral economics of “open source” and “cooperative” innovation economies. Dismissing what we don’t understand as “round-off error” makes poor punditry and worse economics.

Keep Reading

Most Popular

10 Breakthrough Technologies 2024

Every year, we look for promising technologies poised to have a real impact on the world. Here are the advances that we think matter most right now.

Scientists are finding signals of long covid in blood. They could lead to new treatments.

Faults in a certain part of the immune system might be at the root of some long covid cases, new research suggests.

AI for everything: 10 Breakthrough Technologies 2024

Generative AI tools like ChatGPT reached mass adoption in record time, and reset the course of an entire industry.

What’s next for AI in 2024

Our writers look at the four hot trends to watch out for this year

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.