A patent owner is entitled to damages to compensate for infringement of patent rights in an amount not less than a reasonable royalty. According to venerable precedent, a prevailing patentee may seek the royalty that would result from a hypothetical negotiation between the parties, occurring when the infringement began. (In some circumstances, the patent holder will instead seek lost profits.)

For many years, patent owners have used the so-called “25% rule” as a starting point for determining just what level of royalties is “reasonable.” Under that rule, 25% of the pre-tax profit attributable to sales of products embodying the invention should go to the inventor.

Courts have long bowed to the rule’s widespread acceptance among the brotherhood of economic consultants, from whose ranks spring the army of expert witnesses who have touted the rule’s usefulness. Even infringers have rarely challenged the rule’s admissibility.

Now the Federal Circuit—the court that hears appeals from all patent cases—has emphatically announced that the 25% profit-split that the rule calls for is too arbitrary to suit the messy facts of particular cases. In Uniloc v. Microsoft, the court admits that it has “passively tolerated” the rule, but declares it will do so no longer.

For all the wrenching changes to patent law that the Supreme Court and Federal Circuit have wrought in recent years, the calculation of reasonable-royalty damages has been tucked into something of a time warp. Two approaches that were introduced decades ago have long held sway: the 25% rule and the Georgia-Pacificcase.

The 25% rule – a popular method for splitting profits from patented technology

A licensing consultant named Robert Goldscheider has employed the 25% rule for 50 years. Through constant advocacy of his approach in lectures and publications and in expert testimony by him and his acolytes in countless intellectual property lawsuits, Goldscheider has established broad acceptance of the 25% rule. The Licensing Executives Society has awarded him a gold medal for the prominence his rule has attained in patent valuation.

The rule springs from Goldscheider’s observation that many royalty negotiations arrive at a royalty rate that equals about one-quarter of the licensee’s anticipated pre-tax profits derived from the licensor’s technology.

The rule assumes that a licensee should retain a majority, say, 75%, of the profits of a patented product because he has shouldered the considerable risks of developing the product and bringing it to market. The patentee takes the remainder as a license fee.

Of course, royalties are usually assessed against revenues, not profits, so the 25% rule is applied to generate a rate that meets this goal.

For example, in industries with high profit margins, like software and other digital products, the parties may anticipate a 60% profit margin for the licensee. In that case, the 25% rule suggests that a starting point for negotiations be in the range of 15% of net revenues.

By the same token, retail businesses or restaurants that may expect, say, a 3% profit margin, may end up—under the 25% rule–with a royalty rate of less than 1% of revenues generated by the licensed product.

Proponents of the 25% rule allow that adjustments must be made in view of the relative importance, uniqueness or maturity of the technology being licensed, among other factors. Still, the rule establishes a range based on historical royalty rates and may inhibit parties from making unreasonable proposals.

The rule can set a reasonable running-royalty rate by estimating the licensee’s profits for the patented product, then dividing the total profit by the total cost of sales. This gives a profit rate, of which 25% is the running royalty rate. This rate is then applied going forward for patent licenses, or backwards for calculating damages in litigation, where the infringer has already exploited the patent.

The Georgia-Pacific case—40-year old opinion continues to shape patent damages

When used to establish a reasonable royalty in a patent trial, the 25% rule has routinely been twinned with the Georgia-Pacific v. U.S. Plywood case. That case produced a remarkably influential 1970 opinion by a federal judge in Manhattan who listed over a dozen factors that, ever since, have been continually cited to steer royalty rates higher or lower.

Some of the Georgia-Pacific factors are:

• royalties the patent owner receives from other licenses of the patent in suit;
• rates that the licensee pays for comparable patents;
• the scope of the license, in terms of exclusivity or territory;
• the commercial success of products made under the patent;
• whether selling the patented item may promote sales of other of the licensee’s products;
• the portion of profit that is due to the invention as opposed to non-patented features;
• the licensor’s status: is he a competitor of the licensee, or a non-practicing inventor?
• the remaining duration of the patent.

In hundreds of infringement trials, damages experts have cited the 25% rule as a baseline, then used the Georgia-Pacific factors to adjust the profit-split upward or downward.

For example, if the patented technology is untested; if the patent owner is willing to give only a non-exclusive license; if the infringer does not compete directly with the patentee; and if the infringer pays a relatively low royalty rate for comparable patents, the expert may suggest that the patentee be awarded only a 15% royalty against the infringer’s revenues.

Conversely, if the technology has already commanded high royalty rates; if the patented feature is essential to the infringer’s products and will make the infringer’s other products attractive to consumers; and if the patent has many years to run, the expert may urge a 40% royalty for the patent owner.

Appellate court scowls at 25% rule, smiles upon Georgia-Pacific factors

In this month’s Uniloc decision, the Federal Circuit confirmed Microsoft’s liability for infringing a patent on anti-piracy software that Microsoft infringed via the copy protection keys used in sales of Office and Windows. But the court reversed the $388 million award in Uniloc’s favor because the 25% rule was its foundation.

The court concluded that the rule is a “fundamentally flawed tool” for establishing a royalty benchmark in a hypothetical negotiations because it lacks any foundation in the facts of the case.

Importantly, the rule fails the test that the Supreme Court established in 1999, in the Daubert and Kumho Tire cases, for expert testimony: Experts must have a firm scientific or technical grounding for their opinions. Those cases mandate skepticism toward the application of a general theory to the facts of a particular case, and require federal judges to act as “gatekeepers” against faulty methodologies.

The Federal Circuit explains that, in the last year, it has tightened its standards for admitting damages-related testimony. In recent cases, it has instructed, for example, that you cannot use an existing license that relates to one technology to help set the royalty rate for a license of an utterly different technology. Otherwise, the patentee would inflate the rate with “conveniently selected licenses without an economic or other link to the technology in question.”

The court found the 25% rule to be even less reliable than unrelated licenses that involve at least the same general industry and one or more of the same negotiating parties. By contrast, the 25% rule says nothing about a particular hypothetical negotiation or about a reasonable royalty for a particular technology.

The court was not assuaged by the argument that the rule merely establishes a baseline and that industry- and technology-specific factors are used to adjust it. “Beginning from a fundamentally flawed premise . . . results in a fundamentally flawed conclusion.”

In Uniloc, the court yielded to the drumbeat of criticism that has sounded for years over the 25% rule: that it does not account for factors—like the availability of noninfringing alternatives—that would affect a real-world negotiation; that it is too generous to the owner of a narrow patent and too stingy for broad patents; that it fails to recognize the different risks assumed by licensor and licensee in particular cases.

At the same time, the court approved the continued use of the Georgia-Pacific factors “to frame the reasonable royalty inquiry” as long as they can be tied to the facts that are distinctly relevant to each case.

Entire-market-value rule cannot be misused to taint a jury’s damages award

The court also rejected the use by Uniloc’s expert of the entire market value rule, which he used as a check on the total damages. That rule allows a patentee to collect damages based on the entire market value of the accused product, but only where the patented feature creates the basis for customer demand or substantially creates the value of the component parts.

Here, however, Uniloc conceded there was no evidence that its patented software-registration component created the basis for customer demand for Microsoft Office or Windows. While the expert used the entire market value rule as a mere check on the reliability of his damages calculations, it enabled him to disclose to the jury that Microsoft reaped revenues of over $19 billion on Office and Windows.

The lofty figure tainted the jury’s deliberation and may have produced an inflated award. “This case provides a good example of the danger of admitting consideration of the entire market value of the accused where the patented component does not create the basis for customer demand.”

* * *

The Federal Circuit is continuing its trend of requiring a logical link between an expert’s damages calculation and the particular facts of the case. Patent owners must now take pains to ensure that their damages analysis considers the facts surrounding the hypothetical licensing negotiation. Using abstract principles to calculate damages, or even to establish a reality check on the outcome, will be met with heightened skepticism.

Although Mr. Goldscheider deserves his gold medal for the durability of his contribution to licensing, his 25% rule has abruptly reached the end of its run in patent-damages calculations.