stephen dubner – Techdirt (original) (raw)

Stories filed under: "stephen dubner"

Freakonomics Obsession With Patents Strikes Again: Says If More Women Got Patents The Economy Would Grow

from the logical-leap dept

While the Freakonomics guys have shown a willingness in the past to be skeptical about the traditional claims concerning copyright, they often seem to have a blindness to similar criticisms of patents. In fact, in the Freakonomics book sequel Superfreakonomics, there’s what’s basically a puff piece about massive patent troll Intellectual Ventures. Thus it’s disappointing, but not surprising, to hear that one of the Freakonomics duo, Stephen Dubner, is talking up some claims about how “closing the patent gap” between the number of patents that men get and women get (by basically convincing more women to patent stuff) could increase our GDP by a staggering 2.7%.

That’s a huge leap, and a rather astounding claim. And you would think that these guys, who are so focused on trying to sniff out interesting points from data, wouldn’t leap so blindly past a whole variety of questionable assumptions, including the big one: the assumption that there’s a causal relationship between more patents and economic growth. Similarly, Dubner, in his radio piece, goes so far as to claim that patents are a reasonable proxy for innovation — directly asserting that the fact that fewer women get patents means that women have “the most room for improvement in the innovation field.”

To put it simply, that’s ridiculous. Just because people aren’t getting patents, it doesn’t mean that they’re not being innovative. In fact, multiple studies have shown absolutely no link between patents and innovation. It’s a shame that Dubner would take the easy way out and leap to the false conclusion that patents are a reasonable proxy for innovation — and, even worse, that he would then encourage an increasing rate of patenting, when one of the biggest problems facing the innovation sector today is over-patenting.

To be fair, Dubner is basing his report on some new research into the “patent gap” between men and women, which makes these particular claims (including the GDP growth claim). There is some interesting data in the paper showing just how few patents are given to women, but there are a bunch of questionable assumptions and logic leaps as well. Others have been pointing out some of the problems with the report, including new data showing that, based on the trendline, women are already increasing the rate at which they get patents very quickly. The bigger problem, though, is just the facile (and simply unsupportable) claim that more patents automatically leads to greater GDP. As economist Alex Tabarrok notes sarcastically in response:

Right; and since only 10% of construction workers are women, closing the gender gap would result in many more houses

This is a first year stats student type error, assuming this kind of causal relationship where none exists. The authors of the paper should have known better, but Dubner and the Freakonomics crew, who give this kind of work a stamp of approval, should be even more careful. Disappointing.

Filed Under: freakonomics, gdp, stephen dubner, women