1. Introduction

In patent litigation, the adequacy of proof of apportionment in reasonable royalty damage claims is often a challenging issue that is hotly contested by the parties. The Federal Circuit has recently focused on the use of comparable prior license agreements to suggest a royalty base and rate and to satisfy the requirement of apportionment. This article reviews some recent Federal Circuit decisions regarding use of comparable license agreements to establish a reasonable royalty and prove apportionment, and raises some questions that arise from such decisions.

2. Background Regarding Apportionment in Patent Litigation

Apportionment is a patent damages concept that refers to an evaluation of the extent to which patented technology contributes to the value of an accused infringing product or process. The concept was introduced over 130 years ago by the United States Supreme Court decision in Garretson v. Clark where the Court stated that the patentee must “give evidence tending to separate or apportion the defendant’s profits and the patentee’s damages between the patented feature and the unpatented features.”[1]

The Federal Circuit has followed suit and required apportionment in calculation of reasonable royalty damages. It has articulated several purposes for requiring apportionment. One purpose is to ensure that an award of damages for patent infringement is commensurate with the “claimed invention’s footprint in the marketplace.”[2] When the patented technology is an improvement over existing technology, then another purpose of apportionment is to ensure that an award of damages reflects the “incremental value that the patented invention adds to the end product.”[3] These purposes are apparent from the reasonable royalty provision in the Patent Act which requires that the “court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer….”[4]

In claims for reasonable royalty damages, apportionment typically comes into play as part of the hypothetical negotiation analysis set forth in Georgia-Pacific Corp. v. U.S. Plywood Corp.[5] In addressing the issue of apportionment in analysis of the Georgia-Pacific factors, the Federal Circuit has adopted a few rules for certain fact patterns. If the patented technology can be credited as embodying the entire market value of the accused infringing product, i.e., when the patented technology is shown to be the entire basis for customer demand, then no apportionment is required under the “entire market value rule” or EMVR.[6] On the other hand, when the accused product has elements of market value that are not attributable to the patented technology, then apportionment is required. One method of apportionment is to reduce the royalty base to the smallest saleable patent practicing unit (SSPPU) contained within the accused product.[7] However, even after apportioning the royalty base down to the SSPPU, further apportionment may be required where the SSPPU has non-patented elements of value.[8] Where there are sufficiently comparable licensing agreements from which a reasonable royalty base and rate can be derived, then further apportionment may not be required at all. This last scenario has been the focus of recent Federal Circuit decisions, as discussed further below.

3. Apportionment from Prior Comparable License Agreements

When reasonable royalty rates are calculated by reference to prior license agreements, the Federal Circuit has recognized the interesting and compelling reality that in the prior license agreements, the parties have usually inherently agreed upon the value of the patented technology in the agreed royalty base and rate. In other words, the parties to a license agreement have inherently considered and accounted for the value of the patented technology in agreeing on what royalty will be paid for the right to manufacture or sell the licensed product. Thus, where a prior license agreement is reached in circumstances that are sufficiently comparable to the hypothetical negotiation of an asserted patent, the Federal Circuit recognizes that apportionment is effectively “baked into” the prior license agreement, and the agreed royalty may be used or adopted without further apportionment. Examples of this are found in the Federal Circuit decisions in Commonwealth Sci. & Indus. Rsch. Org. v. Cisco Sys., Inc. (CSIRO)[9], Elbit Sys. Land & C4I Ltd. v. Hughes Network Sys., LLC[10], and Bio-Rad Labs., Inc. v. 10X Genomics Inc.[11]

An important foundational requirement for basing a reasonable royalty on a prior license agreement is demonstrating sufficient economic and technological comparability of the prior license agreement under circumstances that tend to confirm that the value of the patented technology was considered and “apportioned” in the agreed royalty.[12] Economic comparability may be evaluated by going through the Georgia-Pacific factors to compare and contrast the circumstances of the hypothetical negotiation and the prior agreement. Relevant circumstances may include the bargaining positions of and relationship between the parties, whether there was a settlement of litigation, the date of the prior agreement, the inclusion of other patents, technology, products or services in the prior agreement, the license structure (e.g., lump sum versus running royalty), the patentee’s licensing policies, the royalty base and rate, the geographic and/or temporal scope of the license, exclusivity versus non-exclusivity of the license, the importance of the licensed technology in driving sales and usage of the licensed products, the profitability of the licensed products, etc. Similarly, technological comparability of the prior agreement must be evaluated and differences accounted for.[13] Demonstrating economic and technological comparability, and accounting for economic and technological differences, of prior license agreements has been a focus of recent decisions of the Federal Circuit, as discussed below.

4. Recent Federal Circuit Decisions

In three relatively recent decisions, the Federal Circuit reached interesting and seemingly contrary results on whether prior license agreements are sufficiently comparable to be the basis for calculating a reasonable royalty in a hypothetical negotiation of a patent in suit.

Vectura Limited v. Glaxosmithkline LLC

On November 19, 2020, the Federal Circuit decided the case of Vectura Limited v. Glaxosmithkline LLC.[14] In that case, the damages expert proffered a 3% reasonable royalty rate based on a 3% royalty rate in a prior license to over 400 patents without calculating any apportionment. The Federal Circuit stated that the fact that over 400 patents and patent applications, one application being for the patent in suit, were included in the prior license “does not fatally undermine” the damage expert’s “theory of comparability.”[15] The court relied on the testimony of Vectura’s damages expert that the prior license and the hypothetical negotiation cover “roughly very similar technologies” and noted that even GSK’s damages expert testified that the prior license was “a very close comparable.” The court also noted that the accused infringing mixtures would have been covered by the prior license. With the prior license agreement being deemed comparable, the Federal Circuit held that apportionment is already “built-in” to the royalty rate and base such that no further apportionment is required and the 3% reasonable royalty rate may be applied to all of the accused infringing sales.[16]

MLC v. Micron

In an August 26, 2021 decision in MLC v. Micron,[17] the Federal Circuit affirmed exclusion of a damage expert’s opinion on a reasonable royalty rate for several reasons, including the expert’s failure to account for apportionment of the technology in the accused product to reach a proposed reasonable royalty rate. The expert had opined that prior license agreements were comparable to the hypothetical negotiation of a reasonable royalty for the patent-in-suit so that apportionment would already be accounted for by using a royalty rate that the expert derived from the prior license agreements. In affirming the district court’s exclusion of the expert testimony, the Federal Circuit first held that the expert’s derivation of the royalty rate from the prior license agreements was inadmissible because it was “not sufficiently tethered to the evidence presented.”[18] The Federal Circuit criticized the expert’s opinion for not analyzing the lump sum payments and associated sales volumes from the prior licenses to derive an effective royalty rate associated with the prior license agreements. Instead, according to the court, the expert improperly relied on a most-favored-customer royalty rate stated in the agreements as the basis for deriving a royalty rate from the agreements.

On the issue of apportionment, the Federal Circuit in MLC held, “we reject the view that the Hynix and Toshiba agreements are comparable licenses.”[19] The court distinguished its earlier decisions in CSIRO, Elbit, and Bio-Rad, supra, wherein the court permitted derivation of royalty rates from prior license agreements without further apportionment because the prior license agreements were “comparable.” The court distinguished CSIRO because it involved the same patent and accused technology. The court added that “unlike the agreement in CSIRO, the Hynix agreement is not a license for the same single patent. To the contrary, the Hynix agreement granted a license to a portfolio of forty-one U.S. and international patents and patent applications, and only one of those forty-one patents is at issue in the hypothetical negotiation.”[20] The court distinguished Elbit because, although the technology was “slightly different” in that case, the expert in that case “appropriately accounted for the differences between the technology at issue in the settlement agreement and the accused technology.”[21] The court distinguished Bio-Rad because the expert “did account for the differences between the accused technology and the licensed technology.”[22] The court further pointed out that the expert should have further apportioned the SSPPU because it contained other technological features and therefore the claimed invention was “not commensurate in scope with the SSPPU.”[23]

Omega Patents, LLC v. Calamp Corp.

On September 14, 2021, the Federal Circuit again addressed apportionment in its decision in Omega Patents, LLC v. Calamp Corp.[24] The court held that a new trial on damages was warranted because a jury award of $5.00-per-unit did not reflect apportionment and the plaintiff failed to show that the patented improvement drove demand for the entire accused product. The court reiterated its rule that “‘when a sufficiently comparable license is used as the basis for determining the appropriate royalty, further apportionment may not be required’… For built-in apportionment to apply the license must be ‘sufficiently comparable’ in that ‘principles of apportionment were effectively baked into’ the purportedly comparable license.”[25] The plaintiff’s president had testified to the company policy of licensing its technology for a $5.00 per unit royalty no matter how many patents in the company portfolio are licensed. The Federal Circuit held this policy “does not speak sufficiently to ‘built-in apportionment’” between the patented improvement and the conventional features of the accused product.[26] The court further held that the plaintiff had failed to show that the eighteen license agreements presented to the jury “attributed a $5.00-per-unit royalty to the value of the ‘278 patent.”[27] The court rejected the suggestion of the plaintiff’s damage expert that the most-favored-nation clauses in some of the license agreements accounts for apportionment. The court further stated that in testimony discussing two license agreements that licensed the patent-in-suit along with multiple other patents, the plaintiff’s damage expert “merely identified” (emphasis in original) differences in the license agreements and a hypothetical negotiation “over a single-patent license to the ‘278 patent” and that “[w]hat’s utterly lacking is evidence that Omega met its obligation to ‘account for such distinguishing facts’ in invoking the licenses to value the ‘278 patent.”[28] The court stated that the testimony from the plaintiff’s president and damage expert regarding the $5.00 per unit royalty was “generic testimony” that “does not ‘account[] for the technological and economic differences between th[e] licenses’ and a hypothetical negotiation over a single, specific patent.”[29]

In a dissenting opinion in Omega, Judge Hughes stated his opinion that there was sufficient evidence to support the jury’s award of a $5.00 per unit royalty. Judge Hughes stated that the objections to the methodology of the damage expert and the comparability of the prior licenses “should have been made via Daubert motion and objections to the admission of the license at trial.”[30] Since the defendant had made no such objections, then in Judge Hughes’ view “the testimony and licenses are evidence capable of supporting the damages award.”[31] Judge Hughes further disagreed with the majority’s view that the prior licenses did not sufficiently account for the incremental value of the ‘287 patent and stated his opinion that once the testimony from the plaintiff’s president and expert “were before the jury, the ‘degree of comparability’ was appropriately left for the jury to decide.”[32] Judge Hughes further commented that the “majority’s insistence on further apportionment unnecessarily forces a patent owner to make a specific business decision about how to license its patented technology …long before a patent infringement suit may even be contemplated” and “forecloses the idea that ‘there may be more than one reliable method for estimating a reasonable royalty.”’[33]

The above-discussed Federal Circuit decisions demonstrate that analysis of apportionment in the context of using prior license agreements to derive a reasonably royalty is very fact-specific and can produce seemingly inconsistent results. The dissenting opinion in Omega decision suggests the possibility that there is some internal disagreement within the court on the subject. However, the court has consistently emphasized that “accounting” for technological and economic differences in prior agreements is important, and a failure to do so results in exclusion of evidence of a proposed royalty rate.[34]

5. Conclusions and Questions Raised by the Recent Court Federal Circuit Decisions.

The above-discussed cases show that demonstrating comparability of prior license agreements and proper apportionment in a reasonably royalty analysis is a heavily fact-dependent analysis. As seen from the above-referenced Federal Circuit decisions, it is difficult to draw many conclusions or general rules on how to demonstrate comparability and proper apportionment. Instead, the decisions raise many questions, including:

1. When is proof of comparability sufficient to establish admissibility of prior license agreements?
2. When license agreements are deemed comparable and admissible, what additional evidence or analysis, if any, must be proffered to show “built-in” apportionment?

What kinds of proof adequately demonstrate apportionment? What kinds of proof are inadequate to demonstrate