Against Platform Monopolies: Zero-Sum Competition

by on July 23, 2006

On May 8, 1998, Paramount Pictures released a summer blockbuster in which a comet was discovered on a collision course with Earth. A team of astronauts is dispatched to destroy the comet with nuclear weapons before it hits Earth. After some setbacks, the astronauts do save the planet from total destruction, but they lose their lives in the effort.

On July 1, 1998, Touchstone Pictures released a summer blockbuster in which an asteroid was discovered on a collision course with Earth. A team of oil drillers is dispatched to destroy the asteroid with a nuclear weapon before it hits Earth. After some setbacks, the drillers do save the planet from destruction, but one of them loses his life in the effort.

The first film, Deep Impact cost $75 million to make and brought in $349 million worldwide. The second, Armageddon cost $140 million to make and brought in $553 million.

One of the assumptions behind the pro-platform rights argument I laid out last week was that increasing the rewards for platform creation lead firms to engage in socially-beneficial R&D spending. The hope is that increasing the returns to platform-creation will stimulate new R&D spending, which will in turn lead to innovations that expand the size of the economic pie. And, we hope, the pie should grow enough to offset the social costs of platform rights that I’ve laid out in previous posts.

But the asteroid/comet blockbuster example suggests one of the reasons that assumption might not hold true, at least at the margin. It seems unlikely that consumers were clamoring for two asteroid/comet disaster movies in the same summer–especially movies that had almost identical plots. If most consumers considered the two movies to be close substitutes–that is, if they would have been indifferent between seeing one or the other–the creation of the second movie was probably a net negativ from a social point of view. It would have been better if only one movie had been made and all the consumers had seen that one, saving the hundred million dollars or so it cost to make, distribute, and promote the second movie.

Now, suppose copyright policy had been changed to halve the returns on summer blockbusters (say, by reducing the term of copyrights to 3 months). It’s likely that only one asteroid disaster movie would have been made that summer. Consumers would have enjoyed the movie just as much (indeed, more probably would have enjoyed it, since they could have watched it for free after the copyright expired), but the studio that produced the movie would have captured less of the surplus than the studios captured in our world. But more importantly, the resources that previously went to make the second disaster movie would have been unspent, and would therefore be available for doing something more useful. The size of the pie increases, but the studios’ share would have shrunk.

Now, I hasten to emphasize that this is an example, not a serious policy proposal. It’s possible that the movies were not close substitutes–that consumers did reap substantial value from having two disaster movies to choose from. And even if having two disaster movies was wasteful in this case, there are probably lots of worthwhile movies that couldn’t be financed under a 3-month copyright term.

My point isn’t about movies in particular, but about the incentives created by too-large rewards for new products. When government policy raises the expected return for developing some product (be it a movie, an iPod, or an operating system) too much higher than the costs of developing that product, you induce what might be called rent-seeking competition–that is, the introduction of too many substantially similar products, with the goal of capturing some of the too-large reward from the incumbent.

Now, obviously this sort of behavior isn’t unique to intellectual property markets. Coca Cola and Pepsi engage in a similar kind of rent-seeking when they spend millions of dollars on soda ads. There can’t be very many consumers who haven’t heard of Coke and Pepsi–the primary function of the ads is simply to divert some of Pespi’s customers to drink Coke products, and vice versa. It’s likely that society would be better off without such ads, but I don’t think it would be a good idea for the government to restrict them.

So my point isn’t to say that the state should always police such behavior, but simply to point out how it undercuts the assumption that increasing the rewards to platform creation will necessarily increase innovation. Increased rewards will probably increase R&D spending, and that will inevitably lead to at least some increased innovation. But it will also lead to the creation of more platforms than is optimal, and to firms expending resources trying to get consumers to switch to their platform. These are wasteful activities that should be avoided if possible.

Hence, the ideal policy isn’t one that maximizes the rewards to product development. A reward that’s artificially made too big stimulates rent-seeking behavior by firms seeking to appropriate more of the reward for themselves. The goal is just to make the reward big enough so that socially beneficial innovations occur.

Of course, it’s just as bad (perhaps worse) for the reward to be too small, since that will lead to the product not being made at all. And since we can’t be sure how much reward is needed for any given innovation, one could argue that it’s better to err on the high side to make sure that all worthwhile inventions are properly incentivized. Next I’ll make the case that the pre-DMCA legal regime–sans platform monopoly rights–provided more than adequate rewards to high-tech innovators.

Comments on this entry are closed.

Previous post:

Next post: