Arnold Kling  

Costs and Benefits of the Patent System

PRINT
The Power of Mistaken Identity... Financial Turmoil...

Patent Failure is a new book by James Bessen and Michael J. Meurer. Here is a lecture by Bessen. The book's introduction emphasizes the importance of what the authors call "the notice function."


An efficient property system notifies non-owners of property boundaries. For example, land rights have a well-developed and efficient system to notify third parties of boundaries. Because of this, only rarely does someone invest millions of dollars constructing a building that encroaches on someone else’s land without permission. Far more typically, would-be investors “clear” the necessary rights before investing. They locate markers, check land deeds, conduct surveys, and so forth, in order to determine the adjacent boundaries. They then either negotiate rights to the needed land or design the building to avoid encroachment.

While surveyors can plainly map the words in a deed to a physical boundary, it is much harder to map the words in a patent to technologies...Not only are the words that lawyers use sometimes vague, but the rules for interpreting the words are also sometimes unpredictable...There is thus no reliable way of determining patent boundaries short of litigation...
It is possible, however, for patent owners to hide the claim language that defines patent boundaries from public view for many years, a practice that is becoming increasingly frequent...
...tangible property rights are linked closely to possession, hence the well-known expression, “possession is nine points of the law.” Patent law also requires possession of an invention, but often this requirement is not rigorously enforced. Courts sometimes grant patent owners rights to technology that is new, different, and distant from anything they actually made or possessed...
Investments in land or structures rarely involve many parcels of land, and property law discourages fragmentation of land rights. In contrast, investments in new technology often need to be checked against many patents— even thousands
I have only just started the book, and I may not be sufficiently invested in the topic to go through it entirely. But the main thesis seems sound.

My intuition is that drug patents are ok, but software patents are bad. The authors' perspective supports this intuition. It is easier for drug patents to be based on clear boundaries and to satisfy the "notice function."

I am trying to think about intellectual property in terms of the Coase theorem. Suppose I have come up with an invention, and I don't want you to use it. Either you could pay me to use it, or I could pay you not to use it. Does it matter which? From a utilitarian perspective, do we care whether the law puts its thumb on the scale of the inventor or the infringer?

What happens when boundaries are unclear, so that the "notice function" operates poorly or not at all, is that the bargaining costs go way up. As the authors show, litigation costs are high relative to profits from innovation.

If there are reforms that could force innovators to give better "notice" with greater clarity on the boundaries of their patents, average welfare would improve. However, the welfare of patent attorneys would not improve, which leads me to suspect that patent reform will not happen. The term "rent-seeking" gets little usage in the book, but the concept comes up in their conclusion, on p. 257:


Clearer notice might undermine patent lawyers' abilities to manipulate patents...the boom in litigation provides a boom in business for lawyers and the growth in the patent bar has outstripped the growth of R&D...the pharmaceutical industry already has clear notice, so they might be loath to introduce changes...in the near term these groups will most likely oppose improved notice.


Comments and Sharing





COMMENTS (9 to date)
Caliban Darklock writes:

I wrote up a proposal for how to fix the patent system two years ago, but nobody cared. I was about to paraphrase it here, but then figured it would be more effective just to link to it.

I'm of the opinion that a patent - by design - secures a return from an invention, but has been improperly exploited in practice to prevent competition. I don't believe the inventor of a patentable discovery has the right to forbid others to use it, only to secure a fair and equitable royalty from its use.

Chris Milroy writes:

I suspect this is a situation where Coase doesn't produce socially efficient results. Specifically, because developing patentable inventions is a costly process, we expect that patent-holders will make smaller profits than people who would take the patent and profit from it directly.

If we further assume that both the inventor and the infringer are equally able to exploit the invention, then the inventor would be unable to pay the infringer enough to dissuade him from infringing (and therefore will make small or negative profits, certainly smaller than the infringer's). In essence, a lack of patents would turn the market into one with open entry, where anybody could take the patent and drive down the price to below the level that the inventor would need to make the invention worthwhile.

This analysis suggests that the total royalty amount (sum of all royalty payments) needed to incentivize invention would be any amount even slightly higher than the cost of inventing.

So it does matter on which side the law falls. Property rights would need to be allocated to the person who came up with the property. Coase still applies in the sense that the same efficient situation will result in this transaction, but since we are looking at a series of interactions, nobody will innovate if the market is blown wide open after somebody invents.

Matt writes:

Software patents were just plain rigged because you can change the medium from one set of molecules to electrons organized on a screen and repatent the obvious.

If I can "invent" a cardboard looking folder file by making it out of free electrons on my TV screen, then I should be able to "invent" the same file folder by constructing it from plywood. But, in either case, it is an incarnation of something thats been around for a thousand years.

TGGP writes:

Kevin Carson has been discussing the difference between "artificial" property like IP from "natural" property at http://www.theartofthepossible.net/2008/03/13/libertarianism-and-liberalism-what-went-wrong/
However, as a Tucker-Proudhon mutualist he also views Lockean property in land to be illegitimate, preferring a right to usufruct.

Vit writes:

You cannot patent a movie plot but you can patent a software architecture. Additionally a lot of software patents are granted for trivial or obvious "inventions".

Josh writes:

I always thought they should tie patents to a single end-product. Everyone entering a particular business space should know their competitors in that space, which should greatly reduce the number of patents to search.

So in software, you couldn't patent a search algorithm, but you could patent a database that uses a new search algorithm. Anyone building a database would need to find patents that their competitors use in their own databases. But they wouldn't need to find generic search algorithms used somewhere else. Nor would a search algorithm patent in a database apply to e.g. an operating system, etc.

This would also undermine the "trolls" who simply create patents with no product, looking for someone to sue. Without an end product, there is no (enforceable) patent.

Michael Martin writes:

I've been thinking and blogging about Coase and the patent system for the past few months. The Bessen and Meurer thesis is flawed for several reasons.

Publicly-traded companies are not the source of innovation in our economy:

http://www.patenthawk.com/blog/2008/03/when_failure_success.html#more

But universities and smaller companies are. Bessen and Meurer themselves say that these entities have profited from patents.

http://www.patenthawk.com/blog/2008/03/did_bayhdole_end_corporate_rd.html#more

Also, real property is not the only legal analogy to patents that ought to be considered in economic debates over the costs and benefits of the patent system. Patents also bear some of the characteristics of corporations in terms of how they structure the investment of human capital into a new venture.

http://www.patenthawk.com/blog/2008/03/patents_property_and_corporati.html#more

Increased scrutiny of the patent system is certainly merited because of the many benefits some reforms would have for our economy. But I feel that Bessen and Meurer were too sensationalist in their accusations during a period in which patent reform is being considered in Congress. Moreover, they seem to be rather disconnected from the realities of how new technology is financed and developed.

Michael Martin writes:

I forgot the post that is most germane to your question about what Coase would think.

http://www.patenthawk.com/blog/2007/11/coase_the_internet_and_a_marke_1.html

As Bessen and Meurer note, transactions costs of procurement are high relative to gains from exclusive rights in most industries (even they note chemicals and pharmaceuticals as an exception). But transactions costs are declining very quickly right now because of new communication technology, most prominently the Internet. Advocates of patent reform should be wary of converting a system of property rights into a system of torts and contracts at the precise moment in history when the system of property rights becomes cost effective for the entire market. Moreover, the characteristic that chemical and pharmaceutical companies share, and that software companies lack, is technology that canNOT be protected by trade secrets. It is no surprise that companies that can maintain a durable competitive advantage through trade secret protection would view most patents as a tax on successful businesses. This criticism, however, is best addressed by tailoring of eligible subject matter requirement or by the creation of stronger independent inventor defenses rather than by abolishment of the entire system.

Lex Spoon writes:

Arnold,

It's a good point, Arnold, but there is a more fundamental problem. The amount of property being protected per patent is tiny. I only know of 2-4 patents which cannot be re-developed with less than one man-year of work. (I'm thinking RSA and MPEG, and I think I had another one or two). Most require far less. It is as if people were getting 20-year monopolies on random patches of one-square-foot real estate.


Lex Spoon

Comments for this entry have been closed
Return to top