Being a plaintiff’s damages expert in the intellectual property arena may be one of the hardest jobs in valuation these days. That was one of the messages we got from last week’s AICPA Forensic & Valuation Services Conference 2013 in Las Vegas.

For a while now, professionals have wrestled with the various legal principles applicable to lost profits and reasonable royalty, but a few guideposts provided orientation, including the notion that unless the patented feature drives demand, the expert would use the smallest salable unit featuring the invention to calculate the royalty base. Increasingly, courts have chiseled away at that bit of certainty, as a recent Daubert ruling from the Northern District of California illustrates.

Stunted apportionment: Both parties were active in the electronic design automation industry. The plaintiff claimed one of the defendant’s verification products contained its patented design-level parallelism (DLP), among numerous other features, and sued for infringement. Its expert calculated reasonable royalty damages resulting from a hypothetical negotiation using the Georgia-Pacific factors as a framework. For the royalty base, he used the sales data from the defendant’s entire verification product, reasoning that this was the smallest salable unit that included the patented component. He made no further apportionment to account for the product’s many nonpatented parts.

The defendant challenged the calculation under Rule 702 and Daubert. The court agreed that it was inadmissible. At the outset, it acknowledged the difficulty in calculating a royalty base in a product where one component has patent protection, while other components do not. “Courts—including, to be fair, this one—have struggled with whether and how to evaluate an apportionment of the royalty base in this scenario.” The court looked to precedent for guidance.

Under the Federal Circuit’s Lucent decision, absent evidence that the infringing feature drives demand, the patent holder must apportion the royalty base to isolate the value of a patent-related feature. But, said the court, Lucent did not deal with the critical issue “of exactly how one is to apportion the base.” Under Chief Judge Rader’s holding in Cornell, a patent holder must use the smallest salable unit as the starting point for the royalty base. This principle, said the court here, did not imply that apportionment necessarily stopped with the smallest salable unit; rather, Cornell expressly rejects the use of the entire market value where the patented feature is not driving consumer demand for the product that contains it. Put differently, there is no logical basis to drop the apportionment requirement in a case “where the alleged smallest salable unit plainly is not closely tied to the patented feature,” the court in the case at bar concluded.

Here, the protected component, DLP, was an optional feature in the defendant’s product, and it was just one component, the court stated with emphasis. Instead of pinpointing DLP’s value relative to the product’s other components, the plaintiff’s expert made a blanket—and wrong—assumption that his analysis ended with his determination that the defendant’s entire product was the smallest salable unit. His royalty base was fundamentally flawed, and the opinion was inadmissible on this basis alone. The court also discredited the expert’s royalty rate analysis but—perhaps in sympathy with the Herculean task damages experts face—allowed him to submit a revised report.