Arnold Kling  

Pushback on "Fair Use" for Patents

PRINT
Socialized Medicine... Goods, Bads, Marginal Utility,...

Michael Rosen and Thomas J. Van Gilder write,


The patent system has not developed such a mechanism ["fair use"]. This is so in part because patents tend to deal more with the physical implementation of ideas rather than their expression. Thus, to practice a patented method or to make a patented device is to exploit the central right granted by patents. Fair use of copyrighted material, on the other hand, does not implicate the central right granted by a copyright — the exclusive use and control of the copyrighted item — as the "fair user" has merely been given limited access to portions of the item.

But, after saying that patents differ from copyrights because the former involve physical implementation, they write,

What he [Kling] neglects is that IP revolves around rewarding people's intangible ideas...Furthermore, 'holders of patents with no products' fits the description of the many start-up companies, in biotechnology for example, that have been the engines of innovation in many industries.

So they say that patents are different because they involve tangible things and then say that the patent system is there to protect intangible ideas. I'm sorry, but you can't have it both ways.

I'm afraid the intellectual property is a difficult area. I do not find any dogmatic solution satisfactory. If you are strongly in favor of government protection of ideas that get filed with the government (patents, copyrights), then I think you wind up creating a situation where somebody who does 100 percent of the work to implement an idea (including independently coming up with the idea) gets ripped off by somebody who did 1 percent of the work--coming up with the idea and filing a patent.

On the other hand, if you strongly oppose any IP protection, then you allow someone who has spent years working on something in a lab to have his or her work stolen as soon as it bears fruit.

Edison said that invention is one part inspiration and 99 parts perspiration. I think it is important to protect the perspiration. I think that protecting the inspiration ends up leading to perverse results more often than not.


Comments and Sharing





COMMENTS (10 to date)
Bill Stepp writes:

Arnold writes:

If you are strongly in favor of government protection of ideas that get filed with the government (patents, copyrights), then I think you wind up creating a situation where somebody who does 100 percent of the work to implement an idea (including independently coming up with the idea) gets ripped off by somebody who did 1 percent of the work--coming up with the idea and filing a patent.

The problem is worse than this. The 100%/1% of the work in creating something doesn't stand scrutiny. RIM did 100% of the work that RIM did; NTP did 100% of the work that NTP did and none of the work that RIM did.
They had no contractual relationship; NTP, which has no product or service, wasn't a supplier to RIM, although the local Staples and pizza stores might have been.

Richard Epstein and two patent lawyers have a lame letter in the WSJ today, in which they claim that there is no legal reason to force a patent holder to make a product.
The problem is that a patent isn't property, contrary to the law & economicst establishment;
it's just a rent-seeking device, a license to steal, if you will.
"IP" is really intellectual monopoly. It prevents other people using their own property as they wish. In doing so, it impedes innovation and makes society poorer, as it unjustly enriches a small rentier class of patent and copyright holders and their lawyer-crime facilitators.
For that's exactly what "IP" is: theft.
Proudhon famously said that "property is theft"
What he should have said is that "intellectual property is theft."

Also, Van Gilder and Rosen are mistaken about biotech start ups. Some do have products, contrary to their presumption. They also have access to venture capital and IPOs.
Patents are far from their only source of revenues. Some are also acquired by larger firms that have products.
The "IP" system should be abolished. This would unleash a torrent of innovation and end the type of injustice that was perpetrated against RIM's shareholders.

Chris Rasch writes:

On the other hand, if you strongly oppose any IP protection, then you allow someone who has spent years working on something in a lab to have his or her work stolen as soon as it bears fruit.

Suppose I invented a duplicator gun with which I could duplicate physical goods as easily as I can now duplicate intellectual goods (movies, music, books). And suppose I used my gun to duplicate your Toyota Camry. Now we both own identical Toyotas. Have I stolen your Toyota Camry?

Stefan writes:

I don't really care if someone "steals" your idea that you took years to develop. Other people are not your slaves, and have no obligation, enforceable or otherwise, not to copy the design or makeup of something freely found or given to them. Roderick Long has the best libertarian view on the subject:

http://www.libertariannation.org/a/f31l1.html

Arnold Kling writes:

In response to Chris Rasch:

If goods are easily duplicated, then they become like pills. The cost of manufacture is relatively small, but the cost of discovery and testing is high. (If the cost of discovery and testing is also low, then the economic problem just about goes away.) If the cost of discovery and testing is high, then some mechanism needs to be in place to allow those who invest in R&D to earn a fair return. Patents are one such mechanism, but not necessarily the best. Alternatives included prizes, for example.

Bill Stepp writes:

Arnold writes, in response to Chris:

If goods are easily duplicated, then they become like pills. The cost of manufacture is relatively small, but the cost of discovery and testing is high. (If the cost of discovery and testing is also low, then the economic problem just about goes away.) If the cost of discovery and testing is high, then some mechanism needs to be in place to allow those who invest in R&D to earn a fair return. Patents are one such mechanism, but not necessarily the best. Alternatives included prizes, for example.

The MC of making a pill is low, but not zero, as Paul Romer seems to think. Otherwise generic drug companies would incur no costs. (Actually, I fibbed here a bit: they do incur R&D expenses, but not as big a percentage of their revenues as proprietary drug firms. But the point remains.)

Why must their shareholders be guaranteed a return? We have a mechanism to enable successful ones to earn back their cost of capital, and that's the market.
After all, drug and biotech firms are not unique in having large upfront capital requirements and "fixed" costs. So too do airlines and steel companies, and we don't have a "must allow to earn cost of capital" rule in place for them. Nor are they patent-protected.
Extending this point to the media business, startup book and music publishers don't have large capex needs and succeed quite often, so why must they be copyright-mollycoddled?

Mike writes:

Rosen and Van Guilder should haved used more precise terminology. Copywright is intended to protect creative expression like books, movies, sculpture, etc. Patents are intended to protect useful things whether that thing is a physical device or a process that has some useful result. When they say that "patents tend to deal more with the physical implementation of ideas rather than their expression," it is this division that I understnd them to be making. In more technical terms, to be protected by a patent, the subject matter must have some "utility." The more fair criticism, therefore, is sloppy terminology rather than inconsistency.

Arnold Kling writes:

Bill writes,

After all, drug and biotech firms are not unique in having large upfront capital requirements and "fixed" costs. So too do airlines and steel companies, and we don't have a "must allow to earn cost of capital" rule in place for them. Nor are they patent-protected.

But the capital investments that airlines and steel companies make must be matched by any would-be competitors. A generic drug company does not need to spend years on testing drugs--they just manufacture someone else's formula.

Thanks for the response! How high must the cost of discovery and testing be before granting monopoly protection? To use the Toyota example again, which do you think would be a wealthier society:

a. A society in which anyone was legally allowed to duplicate Toyotas at will?
b. A society in which Toyota could prosecute and jail anyone who duplicated their cars without permission?

Bill Stepp writes:

The cost of a new drug is high ($800 million per Tufts' data, which is often cited in the press).
This is inflated by the cost of regulation, litigation, taxes, and the patent system itself. Patents are a double-edged sword, as Microsoft has recently learned. The patent system that protects your patent prevents you from exploiting someone else's patented invention.

As David K. Levine, coauthor of _Against Intellectual Monopoly_ (forthcoming) points out, a drug firm has to disclose a formula as part of the FDA approval process, and must make its
clinical trial results available to would-be competitors.

They point out that the man-in-the-street view
that, absent patents, generic competitors would always enter just as the original innovator released a new drug, thus competing away the monopoly rents necessary to invest in new drugs,
is far from certain.

They have a paper (their work, including their book can be read at their homepages) citing work by J. O. Lanjouw on the Indian patent-free pharma market, which shows that generic firms wait at least a year to see how a new drug performs (this makes sense; why invest in a loser or one with obvious potential litigation problems that show up quickly, but which escaped the regulatory apparatus for whatever reason?)

B&L also point out that there is a substantial first mover advantage in the drug market; they also earn profis even after drugs go off patent.

The regulatory process takes at least 3-4 years and seems to be getting more onerous.
Absent drug patents, innovators end up having only have five-, not ten-year monopolies.
They cite evidence that generics don't cause the price of the original drug to fall much, and the proprietary innovator keeps about half the market share.
Patents also impede innovation, and this cost must be accounted for. B&L think the social cost is greater than the social benefit; I agree.

Btw, I e-mailed Malcolm Gladwell about the developing backlash against the IP regime, citing the work of B&L; Tom Palmer of Cato; Stephan Kinsella ("Against Intellectual Property," J. Libertarian Studies [Spring 2001]), and his homepage;
Roderick Long at Auburn, who maintains an anti-IP resource link at the Molinari Institute's webpage; and Sheldon Richman of The Freeman--see his Chicago Sun Times op-ed yesterday, which he posted to the Liberty and Power blog.
MG replied this morning that he is on the case and reading the literature I mentioned.

Rebentisch writes:

A patent system restricts competition. So there is a competition-patenting trade-off, right. Patents cover inventions, not ideas. Unfortunately the US broadened the term invention, e.g. you don't have a technical requirement like in Europe, industrial application is weakened to "useful". So in fact patented inventions today also mean "ideas". Scope in US is very broad and you hardly find sufficient empirical justifications for market interference in many of the fields covered by patent law today. In fact Machlups negative results still hold. Add the judicial governance problem, patent law development is not driven by an economic rationale but by intra-institutional mechanisms, often even the legislator has no say and the system governs itself.

In areas where "ideas" are ubiquitious like software and the real market bottleneck is implementation, patents are esp. harmful. A fundamental premise of the patent system is that the object which gets protection is "rare" in laissez-faire and significantly more of it is produced when you apply the incentive system. Now, the utility of more "ideas" in an ubiquitious "ideas" scenario is low. The incentive system only makes sense in fields where you are short of something.

I recently put it like this:

It is important to understand the patent system as an incentive system. A mouse trap as an incentive system does provide cheese to the mouse only to catch it. A perverted mouse trap would feed the mice with cheese because everybody knows "mouse traps trap mice", although the trapping mechanism of that mouse trap is empirically verified broken. A trapper will be told by the mice that they want to enjoy their right for cheese...

It is apparent that a "user friendly" approach which is very common to the patent offices around the world will fail the economic objectives of the system and will lead to a quality problem/patent inflation.

The existence of oppositions procedures does not change anything, as the patent system is not about balancing the rights of inventors with those actors who file oppositions or challenge the patents but - as an incentive system - has to serve its primary objective (e.g. generate more innovation, "trap the mice") when we are on the level of patent system design.

Regarding the intangible paradoxon:

Patents like all rights are intangible. Ideas and inventions are intangible. But it is probably a difference whether the subject matter of the invention is intangible or not.

--
Copyright is fundamentally different, narrow claims, natural law foundation, a creative act constitutes the right. Don't it lump it together with patent law. In patent law there are no rights prior to the system. If you do not agree on the latter you get into troubles to define the scope by an economic rationale.

Comments for this entry have been closed
Return to top