James Boyle, William Neal Reynolds Professor of Law at the Duke, has an interesting column, "A Natural Experiment," in the November 22 Financial Times on-line about property interests in databases. (You can find a copy of the article -- and lots of other interesting material -- on Professor Boyle's personal webpage here.) At issue is the very different treatment of IP rights in databases in the European Union and the US. In Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340 (1991) (the opinion is available here), the US Supreme Court held (as had many lower federal appellate courts before 1991) that "unoriginal" compilations of facts, such as a directory of telephone numbers, are not copyrightable. (Despite pressure from some interests, Congress has resisted pressure to create this IP interest by statute.) In contrast, the EU issued a directive in 1996 that established copyright protection for databases, even unoriginal compilations.
Professor Boyle says that these stark differences between the US and the EU present a wonderful natural experiment to answer this question, "Is there any discernible difference between the US and EU in the quality or quantity of databases?" And the answer he gives is "Not much."
That shouldn't be all that surprising. It is not clear that IP protection is necessary to maintain the incentive to create and maintain a database. Some US database creators contend that there are technological means of solving the nonappropriability problem (that is, the problem that one's customers for information may pass it on to other potential customers, thus depriving the compiler of sales revenues and thereby diminishing the incentive to compile). For example, some databases are available only to subscribers who have paid a fee and received a password. (This, of course, is not a foolproof solution.)
Ultimately, the question of whether IP rights or technological (or other) fixes are necessary to protect the incentive to create and maintain databases is an empirical question. And Professor Boyle says that there are three of these empirical questions that we can ask about the US and EU to answer those questions.
1. "Has the European database industry's rate of growth increased since 1996, while the US industry's growth has declined?"
No. "In the US, the database industry has grown more than 25-fold since 1979 [even though during most of that period, even before Feist, there was no IP protection for databases]." Professor Boyle points out that the two principal advocates for IP database protection in the US are the owners of the largest legal databases: Reed Elsevier, the owner of Lexis, and Thomson Publishing, the owner of Westlaw. But both of those publishers made their key purchases for the purposes of putting together US legal databases after Feist. They have cleary prospered despite IP protection. How? "In the old-fashioned way: competing on features, accuracy, tied services, making users pay for entry in the database, and so on." This makes you wonder why Reed and Thomson are advocating IP rights for databases.
And what about Europe? Professor Boyle cites a study in Science by Hugenholtz, Maurer, and Onsrud that finds that there was a sharp increase in the number of databases in Europe in the three years after the 1996 directive but that thereafter the rate of increase fell back to the levels prevailing before 1996.
2. "Are the principal beneficiaries of the database right in Europe producing databases that they would not have produced otherwise?"
Apparently not. The evidence, again from the Science magazine article, is that most of the new databases are those that probably would be created anyway -- telephone numbers, television schedules, concert times, community events, and the like. (There have been, Professor Boyle reports, some interesting recent cases before the European Court of Justice regarding whether cetain databases compiled by businesses in the normal course of their operation are entitled to IP protection under the directive. The central question is whether that information would be compiled in the normal course of business or required a "substantial [additional] investment." If so, then the database is entitled to protection.
3. "Is the right promoting innovation and competition rather than stifling it?" We need a touch of background here. Some contend that IP rights in databases might stifle compilations of information because it would be harder, under an IP rights regime, to get some information. This would be particularly true if IP protection would be inappropriately granted to compilations that involved no "substantial investment," that were, for instance, compiled as a normal byproduct of doing business.
There's no clear answer yet.
Professor Boyle's conclusion is that the empirical evidence persuades him that we do not need IP rights in databases: "If the European Database Directive were a drug, the government would be pulling it from the market until its efficacy and harmfulness could be reassessed."
Professor Boyle will return to this topic in a future column for the Financial Times. I'll cover it in a later post.
TSU