Asymmetrical Information - Megan McArdle

09.07.12

How much does America’s patent policy matter?

If IP matters so much, and ours is so bad, why do we have so many tech firms?

I was having a conversation last night with a very smart libertarian friend whose research focus is on economic growth.  During the course of our conversation, he said, “and of course, intellectual property policy is just a mess”.  Which naturally led us to the question: how much does IP policy matter?

There’s no question that American intellectual property policy could stand some serious reform.  Copyright terms are way, way too long, software process patents seem like a generally terrible idea, and there are a host of other systemic problems that make our intellectual property regime far from ideal, even if, like me, you tend to be something of an intellectual property hawk.

I frequently run in libertarian circles, and so of course, I know a lot of people who think that this is a Really Big Deal. You hear a lot of dire warnings about how this spells The End of the Technology Sector, how Big Media Are Shooting Themselves in the Foot, and so forth. But there is an uncomfortable fact glaring at us from the sidelines: America has the most robust technology and media industries in the world, along with (I’m told) some of the most restrictive and ineptly implemented IP rules. If the critics are right, and our IP rules are strangling innovation, why are those companies still here?

There are a few answers to that, I think. The first is that the rules have been steadily getting worse, so while we have a sort of legacy industry, any new growth is going to face stiff headwinds going forward. Yes, but… a lot of these really terrible copyright and patent abuses have been around for a decade or two at this point (I’m looking at you, Sonny Bono). So why does so much of the culture that the world consumes still come from America? Why have we launched Facebook and Google and so forth, and not somewhere else with better IP law? Why is so much research on new molecular entities done by and for the US market, and so little in India?

Another argument is that we shouldn’t measure this stuff just by how much is being sold. A lot of the reason to loosen IP restrictions is to encourage innovation in the free sector, the “gift economy”.

I find this somewhat plausible, but as far as I know, the US also plays a pretty big role in the free sector. You can start getting into per-capita measures, where for all I know maybe a lot of other countries perform better, but those kinds of measures are tricky. For one thing, sheer random chance is going to produce more variance in smaller countries than in larger ones, so even if IP has no effect at all, you’d expect to see some small countries beating us on almost any measure you choose. If everyone, or at least lots of countries large and small, outcontributes us on this score, that would tell you something, of course.

(To be sure, this is all hypothetical: I don’t even know how you would assemble data on which nation makes the most free stuff. Though my imagination is limited, so maybe someone has; if readers know of any, please point me to it.)

You also take the risk that you’re measuring things that aren’t really about national-level IP policy: how old the population is, how much free time they have, and so forth. I’d be willing to bet that Italy is a low generator of awesome free mashups because it’s an old country that is not on the cutting edge of technology adoption, and has a language that’s not spoken elsewhere. Meanwhile, it’s possible that, say, the Danish welfare state and broadband policy acts as a substantial subsidy to mashup culture. This could be good or bad from the point of view of the Swedes, but would not, either way, really tell us much about what our IP policy should look like.

Another way to respond is to point to all the other factors that make America an efficient provider of IP-intensive goods like drugs and movies. Take media. Hollywood grew up at a time when the US had the largest rich-world population (and therefore, the largest market), and also, was rich and urbanized enough to support mass distribution of films to almost everyone in that population. We also speak English, a language that is spoken in a lot of other wealthy countries that buy our movies. The result is that over a century of production has generated an extremely skilled industry which is hard for other countries to compete with. Ditto television. To some extent, these factors still matter: we’re the biggest rich country, and the richest big country.

Beyond that, we have the best post-secondary education system in the world, relatively free markets, a culture that prizes risk-taking, and laws that make it relatively easy to start a business. So we get a lot of people starting businesses.

There are quibbles one could make—we also used to be dominant in television manufacturing, so you need a theory that differentiates between television sets and television shows. But assume, arguendo, that this is true: our terrible IP policy exerts a drag on the economy, but in these sectors, it’s made up for by other factors.

The problem is that this seems like just another way of saying that intellectual property regimes just don’t matter that much. The US arguably has the strictest IP laws in the world, which in many of the arguments I’ve been in, seems to mean some of the worst IP laws in the world. So from that point of view, what the data seems to be telling us is that you can have the worst IP laws in the world, and still dominate many of the markets where IP is most prevalent.

Of course one can say that the US is about to lose its dominance because we’ve finally crossed some invisible line where our IP policy is just too bad to support innovation, or that our performance could be even better with better IP policy. But if IP policy can be this bad, and still leave us with world-dominance in these industries… industries that, moreover, offer some of the best jobs and growth potential in the country… then you kind of wonder how big the marginal improvement could possibly be.