Why national governments should buy patents

Summary: By purchasing patents and selectively placing them in the public domain the government can both encourage innovation through post invention rewards, and remove barriers to competition and future innovation posed by restrictions on "core" technologies.

In my last post, I promised ideas on how to improve the patent system. I cannot say they will “fix” it, but I think they would help. I could start by suggesting that software patents themselves simply be abandoned, but this has been debated quite a bit and so far there has been no progress. My first idea is a bit more unorthodox, but I believe is better suited to making everyone happy, and in general encouraging society’s goals for patents.

The short form is, governments should invest in patents. It is not an entirely new idea (another example), but it has never been an institutional practice, which is what I would suggest. Also, while I think one goal should be to place patents within the public domain, I think there is room for true investment, where governments collect royalties upon patents. For several reasons this is necessary for the success of any large-scale program.

One reason is the international equation. If governments are not encouraged to invest in patents, but only to buy them out in the name of the public domain, there will be no reason for multiple governments to compete in patent acquisition. From the perspective of the inventors this is not a major problem, there is already plenty of competition in the private sector, but how would US/British/French/etc. taxpayers react if the discovered other governments of the world were not pulling their fair share. Encouraging investment simplifies this problem by encouraging the acquisition of patents directly, and leaving the concerns of release to the international public domain to the diplomats, legislatures and heads of states who love to argue about such details. In the end, we would hope for an international treaty establishing such an initiative.

Another reason is the concerns of favoritism that would arise from the government paying for an intangible like patents. I am sure we can all imagine the possibility of certain congressional representatives arranging “graft” patent purchases. But if most purchased patents are expected to perform monetarily for some period of time before being released into the public domain, then some measure of accountability is introduced.

The final reason is that a program that licensed some of the patents might have costs in the 9-figure range, rather than the 10-figure range. This makes the program more politically palatable.

Advantages

There are still some significant advantages to having the government own certain patents, even when it is collecting royalties on them, prior to releasing them to the public domain. One advantage is the government is far more likely to make certain concessions, such as giving educational and public works institutions royalty wavers.

As well, organizations may find it less difficult to negotiate cross-patent licensing agreements. The open-source community might find this to be a very attractive means to reduce their patent liability concerns. It is important however to insure such deals are not based upon favoritism. Open source organizations should be contributing to the overall patent body, not just consuming. This would mean licenses like BSD that contribute to the public domain would have no problem, but more restrictive tools like the GPL, that are based upon collective pools, would need to find ways to share those pools outside their closed domains.

Another advantage is that government issued licenses are more likely to have standard terms, rather than require case-by-case negotiations. This reduces risks based upon the unknown licensing costs when standard terms are not available, and saves on costs in general by not incurring legal and other costs associated with negotiations.

In addition to these short-term advantages, accelerated conversion to public domain would occur. I would suggest 3 years as a general rule, and much quicker for the most important inventions. Release would be based upon factors like the importance of a patent to future innovation. Profit patents would be those more focused on production, rather than bases for future innovation.

But which patents?

A difficult decision is which patents to purchase. An obvious starting point is patents that are necessary to future innovation and research. A good example of this is patents for research techniques in the biomedical industry. Because these techniques are useful to the design and testing of products, their broad availability is invaluable to future innovation.

It is tempting to suggest acquiring patents with the most value to society, but unless broad availability has a value greater than the acquisition cost, I would disagree. The point of this program would be to remove obstacles to future innovation, not to reduce manufacturing costs through government subsidy.

All in all, I believe a program like this could be a win for everyone involved. The inventors would have another possible buyer. The users of patents would have less difficulty obtaining rights, either through simpler terms, or through public domain release. Taxpayers would have some costs involved in supporting the program, but those costs should be returned through lower product development costs and faster innovation, producing better cheaper products.

I have one more big idea, somewhat complimentary to this one, which I will write about soon. I would like to hear what others think of this one however. I will not pretend to have spent years thinking about this, but ideas are like a ship. Most holes you can patch if you have not already dropped it in the water. I am sure I have some holes still, but this feels like a good idea once finished.