When you're calculating a royalty base in a patent damage case, when do you apply the entire market value rule (EMVR) and when do you use the smallest saleable unit (SSU)? And if it's the latter, how do you even determine the SSU?
The problem, as with most patent damage issues, is that the law is continually evolving. As technology advances and machines become at the same time more complicated and more compact, the courts are racing to keep up with the constantly shifting landscape when determining royalties.
A brief review: The EMVR comes into play when it is determined that a patented feature drives the demand for a larger item. Determining exactly what that means has become more complicated as devices become more integrated and harder to break down, so a three-pronged test was established for the EMVR by the Federal Circuit in 2009 [Cornell Univ. v. Hewlett-Packard Co., 609 F. Supp. 2d 279, 268-87 (N.D.N.Y. 2009) a.k.a. Cornell II].
First, the infringing components must be the basis for customer demand of the entire machine. Second, the individual components must be sold together, constituting a functional unit or parts of a complete machine. Third, the individual components must make up a single, functioning unit.
When the EMVR is established in a case, royalties can be based on the value of the larger, more complex item. When the EMVR doesn't apply, the SSU comes into play, and calculating the royalties becomes much more complicated. As recently as June 2014, the Northern District of California issued a Daubert opinion in the case of MediaTek Inc. v. Freescale Semiconductor, Inc., where the plaintiff's expert apportioned the royalty base to the smallest saleable patent-practicing unit—in this case, a computer chip.
In this case, the court found that the chips were the smallest saleable unit, since it was impossible to separate the patented features of the chip from the chip as a whole. The expert determined that the royalties would depend on the balance of the infringing and non-infringing features. Determining the value of a feature should not change simply because it is or isn't easily divided into smaller parts.
In May 2014, the Middle District of Pennsylvania issued a Daubert opinion in the case of Kimberly-Clark Worldwide, Inc. v. First Quality Baby Products, LLC, on whether the plaintiff's damages expert failed to properly apply the EMVR. In this case, the court ruled that the requirements of the EMVR didn't apply because the reasonable royalty was calculated based on the SSU of the claimed product. Here, the expert's royalty calculations were based on the SSU because the materials at issue—disposable absorbent articles made up of several different components—could not be broken down into smaller parts. In the words of the court:
"Defendants have not identified, nor can the court envision, a manner in which the articles could be further subdivided into component parts that still practice the patents at issue."
In contrast, in March of 2013 the plaintiff's expert testimony in VirnetX Inc. v. Cisco Systems, Inc., was thrown out because he did not tie VirnetX's royalty damages to the SSU. While the plaintiff's expert attempted to apportion total revenues between the infringing and non-infringing elements, the Court found that because Cisco sells both software and components separately from its larger packages, those separate components should have been considered when calculating the royalty base. By failing to account for the SSU and not showing the EMVR had been satisfied, the expert's calculations were, in the court's words,
"a poor substitute for the type of analysis one should undertake when parsing an alleged infringer's profits for patented versus unpatented features."
If calculating future infringement damages were a primary concern at product inception, recent case law might suggest that a company selling complex products should, at the outset, divide the product into as many parts as possible, assign a cost to each part, and then put the whole thing back together again rather than build an integrated product with a single cost. This, of course, isn't a practical solution, so in the meantime attorneys and patent damages experts must explore ways to identify the SSU while we wait for practical guidance from either Congress or the higher courts to help navigate these murky waters.