Economics of Intellectual Property: Patents

I realize I’m coming fairly late to this topic, but it is one areas of economics which is in great flux in the last few years. A decade ago, economists mostly said, “Intellectual property is like other property and so property rights should be clearly defined and enforced.” That consensus is falling apart in the face of increasing litagation costs and the computer revolution. In 2009, the average cost of a patent lawsuit was $5.5 million dollars, and was growing rapidly.

There are two major areas in IP policy: copyright and patent. No one gets particularly excited about trademarks, because the laws are mostly reasonable.

The Law
The Constitution: Article I, Section 8, Clause 8.[1]
“To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;”

35 U.S.C. § 101 35 U.S.C. 101 Inventions patentable.
“Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefore, subject to the conditions and requirements of this title.”

The two major criteria for what is patentable is non-obviousness and novelty. While some more recent patents have violated these requirements, for the most part, inventors must satisfy both to get a patent. Given the diversity of the modern economy, patent examiners cannot be more than slightly familiar with any given industry. As such, they are likely to over approve patents that would be obvious to someone with more experience in the field. This effect is most pronounced in high tech fields.

Economic Justification for Patents
Since imitation is cheaper than innovation, without some incentive to innovate, no firm would invest in innovation, since their competitors would simply copy them. Without the innovation costs, the imitating firms could price lower than the innovating firms and thus outcompete them. The incentive that our society has chosen is to grant a monopoly for approximately 20 years to the holder of the patent.

What Type of Good is Intellectual Property?
As I have written before, the essence of a property right is the ability to get the government to enforce it. In the case of IP, that means getting the government to stop other people from using your ideas to produce goods. To claim to have a property right is to claim that a good is excludable. Ideas are non-rival, so IP must be either a public or club good. The marginal cost of disseminating ideas has always been low relative to coming up with them yourself, ever since the written language became widespread, but with the development of the internet, the cost of transmitting ideas has never been lower. Excludability is much easier when the patents are associated with some physical good. Software patents are very easy to avoid, so long as the violater is not engaged in public commecial distribution.

Neoclassical Analysis
The marginal benefit from adding a new product to the market is the sum of the consumer and producer surpluses. Even with monopoly pricing, it is more efficient to have an invention then not have it. The price alone tells us nothing about the benefit from an invention, since a higher price means less money for the consumer, but more for the producer. The loss from the higher prices comes from the reduced quantity of goods produced, times the difference between the willingness to pay and the cost to produce. Graphically, the loss is represented by the deadweight loss triangle.

The gain to a patent holding monopolist is the difference between their producer surplus as a monopolist and their producer surplus they would have gotten as a competitor. The gain to the consumer from the patent is equal to the entire consumer surplus of the product invented, assuming the product would have not existed if not for the patent. But is that a reasonable assumtion?

Most inventions are simultaneously discovered by several inventors. It could be that the patent system is encouraging many people to rush for an invention, but this fact seems to imply to me that inventions come when they come and perhaps there isn’t a lot we can do as a polity to alter that speed. Firms which invent things already have a first mover advantage. They already have engineers who understand the ideas, and their competitors will need years to reverse engineer the product. In the meantime, the first mover will have brand name recognition as the “real” version, as well as time to improve while others are merely imitating. In some industries, perhaps that’s not enough, as imitation is sufficient (such as pharmeceudicals), but in fast moving industries like personal electronics, imitation is just not good enough to make a profit. Firms need to be innovators in their own right.

Shoulders of Giants
In fields where patents are interlinking and products are very technologically advanced, patents slow down the rate of growth by stopping people from using several ideas together. Each separate part could be patented by a different company, and no one could make a functional device without the permissions of everyone else. In such a world, no production at all could take place until all of the patents expire 20 years later. In a world where obsolesence occurs in a matter of a few years, such a tremendous lag is totally unacceptable. Our society is increasingly dependent on combining ideas in new and innovative ways, and the patent system is designed to prevent such use.

Mexican Standoff

Alternatives
Shorter patents would limit the impacts of patents, both good and bad, and would be a good intermediate step. Industry specific lengths would be helpful in separating industries which are rapidly changing and/or have a lot of interconnected patents from industries where products are stand alone and don’t change much over time. Prizes are another option entirely. A large enough prize attracts efforts disproportionately to its cost, because both the winners and the losers pay for research, but the prize granter only needs to pay the winner of the contest. The problem with prizes is that they only work when people know the problem they want to solve beforehand. If something is invented that people did not know they wanted, a prize is useless. Ex post awards are a third option. If people were paid the cost of their R+D plus a percent when they invented something, you could theoretically get both the incentives to invent without the deadweight loss of the monopoly. The problem with this approach is time consistency. No one wants to pay someone for something they already have. Also the value of a widely dispersed idea is completely debatable and subjective.

Patents are a mess, but I don’t think there is a clear solution that works for all industries. I think software needs to either not be patentable or have a much shorter patent life. FDA problems notwithstanding, there needs to be some reward for coming up with new drugs, because they are so stinking expensive to invent relative to their marginal costs. Prizes would probably work pretty well for medical innovation. High tech fields in general probably don’t need patents, since imitation is so costly. In any event, this is a topic that needs to be seriously debated and deserves efforts at reform. Changing the rate of innovation is the most important factor for sustaining economic growth.

Further reading:
Tabarrock on Innovation
Against Intellectual Monopoly
Introductary material from Econlib.
Anything by Kevin Kelly.
The Economy of Ideas
Patents cost around half a trillion.