Reasonable Royalties and ‘Comparable Licenses’: Three Recent Court Rulings

Economists Ink: A Brief Analysis of Policy and Litigation



Thomas R. Varner, an EI Vice President, has extensive experience in complex commercial litigation involving intellectual property, breach of contract, antitrust, and class action matters. He has also conducted research in high-tech, biotech, telecommunications, and manufacturing industries.

Three recent court rulings in patent infringement suits address the methodologies used by damage experts to analyze reasonable royalties. In each of these cases, courts stressed the importance of properly considering “comparable licenses” in light of the patents-in-suit and the licensed products. Issues raised in these cases can be considered using the “Technology License Dataset” (TLD), a new dataset that comprises over 4,500 technology licenses from the high-tech and biotech fields that have been filed with the U.S. Securities and Exchange Commission (SEC).

The first case, Lucent Technologies v. Gateway, involved a “date-picker” tool that enables a user of Microsoft Outlook, among other programs, to enter dates using a monthly calendar grid. The jury in Lucent awarded $358 million for reasonable royalties; however, the Federal Circuit vacated the award.

The Federal Circuit identified a number of problems with the plaintiff’s damage theory. Although the jury awarded a lump-sum dollar amount, the Federal Circuit reasoned that the amount was based on Lucent’s expert’s estimate of damages using a running royalty rate expressed as a percentage of sales. The Federal Circuit noted that the jury in Lucent had almost no testimony with which to make an economic comparison between lump-sum agreements and running royalty agreements. “For a jury to use a running-royalty agreement as a basis to award lump-sum damages, however, some basis for comparison must exist in the evidence presented to the jury.”

Although one can mathematically convert a lump-sum royalty payment to a running royalty amount if the total sales in the royalty base are known, one should first consider the economic basis for such a conversion. An economic analysis of which form of royalty is preferred could include a number of factors, such as the risk aversion characteristics of the parties, the costs to observe covered sales, the existence of other licensees, the price elasticity of the licensed products, information asymmetries between the parties, and whether the expected use of the technology is known at the time of the negotiation. Analysis of the TLD supports the view that a running royalty rate is not automatically preferred for many types of agreements. The TLD indicates that over 40 percent of all software patent licenses submitted to the SEC specify financial consideration in terms of either a lump-sum amount or a fee per unit sold rather than as a percentage of sales.

The Federal Circuit in Lucent also objected to the plaintiff’s reliance on the “entire market value rule” in determining the royalty base. That rule allows an expert to use the entire market value of a product as the royalty base, even though a product may have features that are not covered by the patents-in-suit, as long as the patented features are the basis for customer demand for the product. In Lucent the Federal Circuit found that the product in question, Microsoft Outlook, clearly had many features other than the date-picker, and that those features accounted for most of the consumer demand for the product.

The second case, Cornell University v. Hewlett-Packard, involved a small sub-component of a computer processor. The processor would be used as part of a CPU brick, which in turn would be part of a larger computer server. The jury awarded the plaintiff $184 million in reasonable royalty damages based on a royalty rate of 0.8 percent applied to a royalty base of $23 billion, the sales of the defendant’s CPU brick products. The court rejected Cornell’s use of Hewlett-Packard’s CPU brick sales as the appropriate royalty base and reduced damages to $53 million. The court stated that a more appropriate royalty base would be “the smallest salable infringing unit with close relation to the claimed invention–namely the processor itself.”

The Cornell ruling highlights the importance of analyzing royalty rates in relation to the royalty base. For example, data in the TLD indicate that the median royalty rate for bare patent licenses covering software technology is 4.0 percent if the technology is used in software, but less than 1.5 percent if the product is used in computer hardware.

The third case, ResQNet.com v. Lansa, involved software technology that downloads screen information from a remote mainframe computer onto a local PC. The plaintiff’s damage expert estimated reasonable royalty damages using a royalty rate of 12.5 percent based, in part, on seven “comparable licenses.” Five of these seven licenses were for re-branding or re-bundling of software products and software code. The Federal Circuit rejected the expert’s reliance on these five licenses and stated, “[T]he trial court should not rely on unrelated licenses to increase the reasonable royalty rate above rates more clearly linked to the economic demand for the claimed technology.”

Royalty rates do vary considerably depending on the type and scope of agreement, as data in the TLD show. Among technology agreements filed with the SEC, the median royalty rate for software re-bundling or re-marketing licenses is approximately 14 percent, whereas the median royalty rate for bare patent software licenses is approximately 3 percent, almost a five-fold difference in royalty rates. The remaining two licenses considered by the plaintiff’s expert inResQNet.com arose out of litigation over the patents-in-suit. The Federal Circuit indicated a willingness to consider such agreements as part of a reasonable royalty analysis. In determining the relevance of settlement agreements on reasonable royalties, an expert should consider a number of factors including the stage of litigation in which a settlement occurred, the magnitude of fixed payments, the release from other ongoing litigation between the parties, the outcome of related legal proceedings, and the presence of other non-patent related agreements between the parties.

These three patent infringement rulings highlight the careful scrutiny courts have recently given to “comparable licenses,” which are often an important resource for a damage expert in determining reasonable royalties. The issues discussed above – use of licenses with different forms of payment, consideration of the royalty base in conjunction with the royalty rate, and recognition of differences in royalty rates based on the type of license agreements – are all important matters to address in a damages analysis. This heightened scrutiny by the courts should lead to more careful analysis by experts and improve the economic basis for patent infringement damages analysis.