To Professor William Landes and Judge Richard Posner, intellectual property consists in any potentially valuable human product...that has an existence separate from a unique physical embodiment. Nothing less is the subject of The Economic Structure of Intellectual Property Law, a new installment from Chicago's tireless bards of Law and Economics that employs empirical and non-empirical techniques to analyze and ultimately, simplify the fields of copyright, trademark, patent, and trade secret law. The book is an unofficial sequel to their influential 1987 volume, The Economic Structure of Tort Law, and like the earlier book, its general posture is positive (i.e. descriptive) rather than normative. Landes and Posner give economic meaning to the statutory rules and judge-made doctrines that have evolved over centuries, and, with the tools of economics, they challenge the common wisdom of today's policy makers. I recently asked Professor Landes a few questions about the book.

You and Judge Posner have been collaborating for nearly thirty years, and you wrote your first article about trademarks together in 1987. Why did you decide to put this book together now?

We had published several papers that used economics to understand the important doctrines in trademark, copyright, and trade secret law, and had also written papers that applied economics to less conventional areas of intellectual property such as appropriation art, parody and burlesque, moral rights, unpublished works, and the influence of antitrust law on intellectual property. In light of the rapidly growing scholarly and popular interest in intellectual property law and its growing importance in the practice of law, it seemed like the right time to update and revise these papers and put them together into a single volume. The book gave us the opportunity to tie together the many doctrines governing intellectual property law and show that they could be best explained as efforts to promote economic efficiency. Unlike our earlier torts book, the importance of economics in illuminating legal doctrines is now well accepted and no longer controversial.

One of the book's themes is that intellectual property law should be understood as more than just a tradeoff between incentives and access the incentives that a monopoly position offers to creators of intellectual goods and the access that the general public has to those goods once they are created. Why is this model too simplistic?

It is not that the incentive-access model is too simplistic but rather that access costs must be broadly defined to include the costs to producers of creating new intellectual property. The key insight of the economic approach is to show that too much protection of intellectual property is socially harmful by making it so difficult for producers to borrow and build on earlier works that the costs of creating new works far outstrips the potential incentive benefits from added protection. Paradoxically, the net effect of too much protection will be a reduction in the creation of new works of intellectual property.

You're careful to say that copyrighted or patented goods have public good characteristics, but they are not public goods per se. Why is that?

When we use the term public goods to describe aspects of intellectual property, we mean that the cost of allowing additional persons to use or consume a piece of intellectual property is negligible. Once a song has been created, the cost of allowing another person to perform or listen to the song is zero. That is not the case for private goods. In order to provide an automobile to another consumer, society must incur the cost of producing another automobile. The public goods aspect of intellectual property is what produces access costs some persons will not be willing to pay the price for a work even though the cost to society of providing the work to them is zero. This creates a social or deadweight loss. On the other hand, if intellectual property were given away for free in order to eliminate access costs, there would be no revenue to cover the costs of creating new works. The economic model tries to show that many doctrines of intellectual property law can be explained as rough ways to promote economic efficiency by balancing access costs against the positive incentive effects to persons creating, producing, and distributing intellectual property. Yet unlike pure public goods such as national defense, it is economically feasible to prevent free riders or persons who don't pay from using a work of intellectual property. Thus, the private production of intellectual property can provide creators and others with revenues to cover their costs.

What is rent seeking and why is it a problem?

I don't think this is a major problem in intellectual property. That doesn't mean it doesn't exist. Whenever there is government regulation and statutory activity, persons will spend resources to influence the outcome in their favor. This is primarily a redistributive rather than a productive activity in which some groups gain and others lose. A good example is the recent copyright term extension legislation that added twenty years to the copyright term including works that had already been created many years ago. From a social standpoint, the incentive benefits that may arise from longer copyright terms for works that already exist are, by definition, zero (or at best trivial) whereas access costs are positive. This is a good example of rent seeking behavior by Disney and other owners of particularly long lasting pieces of intellectual property.

Rent seeking also occurs in patent races where multiple parties expend resources to secure patent protection for a particular invention ahead of their competitors. The concern is that the aggregate investment in seeking this protection will surpass the level that is socially optimal. We tend to believe that patent races produce benefits of their own including alternative inventions and broader learning across an industry. But we are not indifferent to the costs.

And don't some scholars also see rent seeking problems in the area of trademarks?

Yes. For example, some think it socially wasteful for producers of brand name drugs to invest in their trademarks when chemically identical generics are available in the market. Yet the fact that two goods have the same chemical formula does not make them of equal quality. Trademarks enable consumers to economize on the real cost of products, which includes the time they spend searching for the quality they desire.

Let's talk about some of the empirical studies in the book. You find that depreciation rates on patents, trademarks, and musical copyrights are much lower (corresponding to a longer average life) than the depreciation rates on books and graphic works. How did you come to this conclusion and what explains it?

Posner and I have always tried to do some systematic empirical analysis in our law and economics papers, starting with a paper in the 1970s on legal precedent which tried to estimate the stock of legal capital and the depreciation or durability of different types of legal capital. The intellectual property book contains a number of empirical studies on a variety of topics. One looks at the effects of state moral rights laws on the earnings of artists and state funding of the arts. Another compares the statutory terms of intellectual property with their actual durability. We did this by analyzing data on renewals of copyrights (for works created before 1978), patents and trademarks. For example, the smaller the fraction of copyrights that were renewed after their first term, (adjusted for the cost of renewal), the higher the depreciation rate and the shorter the average economic life of a copyright. Roughly 80 percent of registered copyrights in the were not renewed in the 1910 to 2000 period suggesting that for the expected future economic benefits from a renewal term were not worth the cost of renewal. In turn, this implies that the average economic life of a copyright was about fourteen years, which is considerably shorter than the statutory term. Within copyrighted works, music had a lower depreciation rate and a longer life than literary (including both books and periodicals) and artistic works. This is understandable because music is more easily adaptable to changes in taste and context than other works. Depreciation of trademarks and copyrighted music are similar, in part because a trademark can be extended to new goods and services. Although the statutory term limit for a trademark that continues to be used in commerce, the average economic life of a trademark is around 15.5 years. A patent has a twenty year statutory term provided the patentee pays maintenance fees at 3.5, 7.5 and 11.5 years after the patent has been issued. Using patent renewal data, we estimated that a patent had an average economic life of about 16.6 years and that about 30 percent of patents lasted the full twenty year term.

Finally, you mention the humbling example of academic publishing. And you note that given the tiny readership of most academic works, their publication may be an artifact of copyright rather than a response to a social need. Pardon me for asking, but which is it in this case?

I'll get back to you in a few months.