Internet business method patents

The ease with which software and business models can be replicated in cyberspace, along with the culture that maintains "if it's online it is open to all," has pushed many companies to create or expand patent portfolios to capture and protect all aspects of their online business operations. In addition

to protection, many firms and individuals have acquired such patents for profitable licensing opportunities, regardless of whether they intend to employ these business operations. These patented operations include not only the software and Internet technology needed to facilitate transactions, but also the Internet business methods-that is, the new ways of conducting a business or an aspect of a business that are enabled by the Internet infrastructure. (Amazon.com, for example, has patented both its 1-Click method of checking out and its website "affiliates" program. Under the affiliates program, an affiliate site that signs up with Amazon receives a fee every time a book is bought from Amazon as a result of a click-through from that site.)

Traditionally, most observers considered patents on business methods to be outside the scope of the Patent Act. By the mid1990s, however, the United States Patent and Trademark Office (PTO) made clear in its examination guidelines for patent applicants that the office would issue such patents.1 Moreover, in a 1998 patent infringement lawsuit over a software-- enabled business method, the Federal Circuit Court of Appeals-the main court for deciding patent matters-endorsed this policy. In the 1998 case, State Street Bank & Trust Co. v. Signature Financial Group, Inc., the court held that a financial service provider's patent on,"business method" software that operated a hub-and-spoke investment portfolio system was valid and enforceable against a competitor who was using the software-enabled business method. The court stated that the long-held business method exception to patentability was "ill-conceived" and should not be used for holding an invention unpatentable.

The rise of the World Wide Web, the change in PTO policy, and the endorsement of the Federal Circuit prompted a flood of Internet business method patent applications. Only 170 applications for business method patents were filed in fiscal year 1995. By 1999, the number of business method patent applications received and granted by the PTO had increased to 2,700 and 583, respectively; in 2000, to 7,900 and 899 in 2000. The number of applications was expected to reach 12,000 for 2001.

The Controversy: Inventiveness and Distribution

This increase in the number of Internet business method patents applied for and received concerns many critics of the patent system, who argue that granting Internet business method patents is bad for innovation, making it more difficult for the diffusion of ideas and entrepreneurial activity often associated with the Internet. Critics also contend that Internet business method patents are too easily granted by the PTO and are "weaker" than other patents because of inadequate reference to prior art in the patent applications. Prior art refers to related inventions that can be found in the body of previously patented inventions ("patent prior art"-or inventions publicly known through, for example, journals, public press, and technical reports ("nonpatent prior art"that the patent office or court analyzes before granting or denying a patent to a comparable invention. Nonpatent prior art in the area of business methods may include programmed ecommerce website applications, literature and websites suggesting specific sales techniques for the Web, and possibly the entire field of business practices (non-- software implemented).2 Some critics argue that larger business organizations will become Internet patent mills, able to squeeze out small entrepreneurs with new property rights over Internet business activities.3 Another concern is that U. S. firms may be gaining an unfair advantage in patenting in this area inasmuch as Japan and Europe have been slower to adopt a pro-patent stance to business methods.

Congress responded to the prior art criticisms with several proposed and enacted reforms, most of which focused on reforming the processes through which the PTO reviews business method patents, especially prior art.4 Congress also passed the First Inventor Defense Act of 1999, which allows firms to be excused from violating a business method patent if that firm could prove that it used the invention more than one year before the filing of the business method patent. This provision was important to ensure that companies that had relied on trade secret protection for their business methods would not be undermined by the expansion in patent protection.

In response to the criticism and the scrutiny of Congress, the PTO also initiated several reforms. These included, for example, an outreach to the business industry for their input in establishing within the PTO a non-patent prior art database of "business data processing" methods; the mandatory searching of U.S. and foreign patent documents and nonpatent literature; an enhanced examination process for business method patents, entailing a broader, mandatory search by examiners for prior art related to these applications; a new, second-level of review to ensure compliance with the mandatory search requirements; and the rare action of retracting a recently issued business method patent over subsequent concerns about its novelty and non-obviousness. These reforms have affected the rate at which business method patents have been granted. In a recent internal study, the PTO found that it had granted just 36 percent of the business method patents it reviewed in the quarter ending March 2001, down from 56 percent in the quarter ending March 2000. Overall, the PTO last year granted 182,223, or 72 percent, of the overall patent applications it reviewed.5

What the Data Show

To evaluate some of the concerns regarding Internet business methods, we examined a full set of 1,093 Internet business method patents issued from January 1, 1990 to January 1, 2000 (I-business method set), most of which were issued during the last three years of the decade.6 We compared these with a random set of 1,000 patents issued by the PTO across all fields of invention between mid- 1996 and mid- 1998 (general patent set) to see how Internet business method patents stacked up in terms of prior art and ownership, both in terms of entity size and foreign v. U.S. ownership.7 Most of the patents from the I-- business method set were issued after 1996, so the time periods of the two data sets are compatible for comparison.

When compared against the general patent set, the I-business method patents fared well in terms of the amount of prior art cited in the patent.8 Contrary to critics' arguments, the average (mean) number of total (patent and non-patent) prior art references in a patent was, at 24.9, larger for I-business method patents than for general patents (15.16). The same was the case for the non-patent prior art reference averages: the I-business method set average, at 10, was larger than that for the general patent set (2.37). This is especially relevant to the debate over whether I-- business method patents are being properly reviewed by the PTO inasmuch as most of the prior art for a business method would be expected to come from non-patent resources (e.g., business and academic literature, websites, and software).

We also compared the PTO entity status of the general and Internet patent owners, including whether a patent was owned by an individual, small business (500 or fewer employees) or a large business (more than 500 employees). We found that a larger share of I-business method patents were owned by small business (19.4 percent) when compared to small business ownership of general patents (10.7 percent). Large entities owned a smaller share of I-- business method patents (63.13 percent) when compared to large entity ownership of general patents (70.7 percent). These results were statistically significant. The percentage of individual owners of each patent remained roughly the same. The results suggest that large companies do not receive patents on Internet business methods any more than for patents in general.

With respect to international competitiveness, there is evidence to support the concern that U.S. companies are being awarded a disproportionate share of Internet-related patents. We found that inventors in Europe (Internet 2.3 percent, General 17.3 percent), Japan (Internet 5 percent, General 21.4 percent), and other foreign countries (Internet 0.5 percent, General 5.9 percent) obtain Internet business method patents at a much lower rate than U.S. inventors, compared to the patents foreign owners receive more generally.

Conclusion

Companies operating over the Internet have aggressively pursued business method patents in recent years. While there has been much criticism about the PTO's generosity in granting such patents, recent reforms may have curtailed much of the perceived abuse, or at least mollified many of the PTO's critics regarding business method patents. Evidence from our recent study suggests that critics of the PTO's handling of business method patents, at least those that are Internet-related, may have overstated the case. We found a surprising amount of prior art in I-business method patents. Furthermore, we did not observe a significant imbalance between firm size and the issuance of I-business method patents, although the evidence was very crude in its measurement. We did find that U.S. firms were far ahead of their European and Japanese counterparts in attaining these patents, compared to the division of patents more generally among these players.

IMAGE TABLE 13

Internet Business Method Patents in the United States, 1990-1999

FOOTNOTE

Notes

FOOTNOTE

1. See U.S. Patent& Trademark Office, Examination Guidelines for Computer-related Inventions (Feb. 1996).

2. American Intellectual Property Law Association, White Paper on Patenting Business Methods (November 27,2000).

3. William Gurley, "The Trouble With Internet Patents: Creating Rents for the Few to the Detriment of the Many," Fortune, July 19, 1999. (See <http://www.fortune.com/ indexw.jhtml?channel=artcol. jhtml&doc-id=43712>)

4. These include, for example, the publication of any patent application claiming a business method eighteen months after filing; the establishment of an Administrative Opposition Panel to allow any person, within nine months of the issue date of a business method patent, to challenge the validity of the patent; and the requirement that a person challenging the validity of a business method patent demonstrate invalidity only by a preponderance of evidence (rather than the clear and convincing standard applicable to all other types of patent).

5. William L. Bulkeley, "U.S. Cuts Issuance of Business Patents for Work Methods," Wall Street Journal, March 22, 2001 (reporting results from Esther Kepplinger, deputy commissioner for patent operations).

6. John R. Allison and Emerson H. Tiller, Statistical Analysis of Internet Business Method Patents, Report to the National Academy of Sciences, STEP Board (2001).

7. This random set was generated in a previous study on patent litigation by John Allison and Mark Lemley. John R. Allison & Mark A. Lemley, Who's Patenting What? An Empirical Exploration of Patent Prosecution, 53 Vanderbilt L. Rev. 2099 (2000).

8. A full description of the statistical tests performed can be found in John R. Allison and Emerson H. Tiller, Statistical Analysis of Internet Business Method Patents, Report to the National Academy of Sciences, STEP Board (2001).