There are too many patents filed for too many obvious inventions, especially in the computer arena. The patent office is understaffed and overloaded, and its employees believe that their task is not to reject patents to aid the social good but to process patents on behalf of their “customers” (the filers) as quickly as possible. Once granted, patents become competitive weapons. When an interested party inquires about using or buying the patent, the asking price for the patent jumps manifold—after all, it’s not an auction but a bilateral negotiation, and the fact of inquiring reveals an unknown high valuation on the potential buyer’s side. Worse, inquiring exposes oneself in front of a jury to claims of deliberate infringement. Thus, it is often better for a company simply to ignore patents, plow ahead, and litigate later on.

Amid this mess, Congress is wondering how to deal with patents.

Here is a simple solution to the problem, borrowed liberally from stake horse racing: every patent filing should be required to come with an estimate of the “expected invention value” to it. Upon payment of the filed value from anyone to the inventor, the patent extinguishes. The filing fee would be 1% of this value, payable to the US Treasury, reducing the taxes we all have to pay.

As an inventor, would you put an expected value of $1 million onto your frivolous patent if it cost $10,000 to file? Probably not. As a potential user, would you end up in a lawsuit over a patent that you can “buy out” for $1 million if the legal fees are $2 million and the risk to you is $10 million? Probably not.

Of course, this solution is not a panacea.

An ordinary inventor in her garage, say of the cure to cancer, may be limited by her ability to pay the filing fee. However, there are almost no such inventors nowadays. Quick: name a few. Who’s the last such inventor that truly became rich off a patent that was invented in a garage by someone who couldn’t afford a filing fee?

As it stands, the legal preparation fees to file patents are already large enough to make patent filing worthwhile more for corporations with existing legal intellectual property rights departments and rarely for individuals. But even if you are the rare private inventor, the return on the patent, even if it were to bought into the public domain by someone and hence become socially useful to all, would still yield 100 times your filing fee. You would think very carefully about the right expected value to file.

A variant of this scheme would be “payment-for-duration.” Each fee would cover 10 years worth of protection. If not “purchased into the public domain” by someone, the inventor could choose to extend it, e.g., for the same fee, for another 10 years. And again. Good. Fewer taxes for the rest of us to pay.

Another problem is that an upper value eliminates the upper right tail of the payoff distribution. What if you invent something that lateron turns out to be far more valuable than you thought? All true. But the intent of patent protection is not to provide a lottery for inventors. Instead, it is to incentivice inventions. How many private inventors will be dissuaded from inventing (not from filing!) because they are limited to a potential gain of a few million dollars? In addition, the ability to put a value on patent may even aid inventors, who currently have no simple market for their patents (short of patent trolls)—and my proposed system here of course allows only “buy for freeing” not “buy for litigation.”

Good idea?