Lenz Blog

Purpose of the Patent System

Axel H. Horns writes in his recent FAQ on "computer-implemented inventions":

Q5: What is the most important purpose of the patent system today?

A5: Patents are instruments to secure fair market values for non-obvious technical problem solutions ("inventions"). This is also valid regarding such inventions for which the realisation requires the usage of a programmed computer or computer network:

He then goes on to dismiss our side of the debate as a bunch of communists, which is one reason why I wrote a few lines about the undisputable facts regarding the personal financial interest he and other patent attorneys might have in this debate.

(Update: Horns kindly clarifies that he did not mean communism when he wrote "lobbying against modern global capitalism". I agree with his point that there is no benefit to be found for either side in the debate in using labels like "communism", and my original point is that that is also true for the label "anti-capitalism", or "anti-patent" for that matter.)

If I dismissed everything Horns writes as motivated by greed, that would be unfair in exactly the same way as his dismissing our side as "lobbying against modern global capitalism". I call everybody's attention to this unfairness but refuse to reply in kind. Again, I do not assume that everything a patent attorney writes is worthless just because he might have a financial interest in the debate.

And his question is actually quite important and interesting. I am going to try to have a swing at it myself and then I will address his answer.

There are several purposes associated with the costs in patent attorneys' fees, patent offices and restrictions of everybody's liberty that come with having a patent system.

One of them is the idea that the patent system helps disclosing technology. Without patent protection, people would keep their technology secret. Society pays the costs mentioned above to get the disclosure of technology in return.

For software patents, that does not seem to be an interesting deal for society. As far as I know, people who actually do development work do not research prior art at the patent office. The recent paper in support of software patents by Ronald J. Mann also says that no one the author spoke to researched patent prior art.

If anything, software patents are generally regarded as the single most effective threat against open source software development. If that is true and open source projects will be taken down or crippled all around the place, society will actually get less open information on technology than without a patent system. The open source development method seems to be far more effective for the goal of disclosing technology than the patent system.

Any gain for society in disclosing of software technology is dubious at best, and probably far outweighed by the potential damage to a process of disclosure that everyone knows is really working.

Another purpose of the patent system is to reward development. Without such reward, there might be a lack of incentives to invest the necessary ressources to develop great new technology.

However, with software there already is copyright protection. That is quite enough to assure the ability to make a profit. If our side is wrong and software can be protected by both copyright and patents, then that would be the only such case. Books, movies, music are protected only by copyright. You can't get a patent on the idea "young sorcerer's adventures at magician's high school". So why do we need a patent on the idea of "adding a third dimension to menus by grouping them as tabbed palettes one behind the other"?

Therefore the above purposes of the patent system do not seem to give much of a good reason for society to bear the costs in freedom and money associated with them.

Now for a short discussion of Horns' answer. He says that the purpose is to "secure fair market values".

I think it is exactly the other way around. Any patent disables the market mechanism. If some company has a patent on some drug with a large market, generic makers can't compete as long as the patent is valid. That, in turn enables the patent holder to charge more than the price that would result if competition was not restricted by the patent. The difference between the price such charged and the price resulting from a working market mechanism is the value of the patent.

Then, again, maybe I didn't understand the answer correctly. Maybe Horns wants to say that the market price without patents would not be "fair".

That leads to another criticism. What exactly is a "fair" reward for the patent holder? And, if there is any way to describe the dollar amount of a "fair" reward, how exactly does the patent system guarantee that the patent holder receives that "fair" amount and neither less nor more than would be "fair"?

The answer to these questions doesn't seem to be easy. I am agnostic on the point. While I think it might be possible that the patent system as a whole does work to assure "fair" rewards, I don't think it is easy to prove or disprove this with hard economic data.

Anyway, even accepting that assuring a "fair" reward for the development of technology is the purpose of the system, in the case of software that is already assured by copyright protection.

Posted by Karl-Friedrich Lenz at April 14, 2004 07:56 PM | TrackBack