For more than a decade, public policy makers, competition agencies, courts, and government authorities around the world have been developing an increasingly detailed set of rules governing and defining fair, reasonable, and nondiscriminatory (FRAND) terms for the licensing of standard-essential patents (SEPs). It is time to take a step back, to think hard about where we are, and to recognize that some core propositions need to be reconsidered to better align FRAND analysis and policy with good public policy.

Unlike in a competitive marketplace, and unlike in the markets for any other component in an implementer’s product, in a market for patented technology, implementers are free to “take” the technology goods (the patented technologies) and incorporate them into their products, without having first agreed to a price—in fact, without having agreed to pay at all.

This anomaly—the inability of the patent holder to prevent an implementer from taking and using the technology without having agreed to pay—has significant ramifications for both the patent holder and the implementer regardless of whether the patents are SEPs.  For any patent, if and when the implementer is finally tracked down and asked to pay, the implementer will be free to increase the patent holder’s costs to enforce its patents by challenging both whether its products infringe the patents, and whether the patents are valid.

This is the issue of “holdout.” From a public policy perspective, holdout is troubling, because far from simulating the outcomes of a competitive market, it incentivizes conduct that is contrary to what would be obtained in a competitive market.

The ramifications of holdout are even more troubling and contrary to public policy when the patents at issue are SEPs.  An implementer engaged in holdout when the patents are not SEPs runs the risk of being found liable for willful infringement, and therefore subject to enhanced penalties.  By contrast, an implementer engaged in holdout when the patents are SEPs runs no such risk: the only “penalty” the holdout implementer will face is being required to pay FRAND royalties, the same royalties it would have paid had it not engaged in holdout and—even more to the point—the same royalties that its competing good faith implementers have been paying all along. The term “efficient infringement” is an apt description and explains the strong incentive to engage in such conduct.

Holdout thus creates exactly the effect that FRAND, as a public policy, was intended to prevent. In effect, it permits the holdout implementer to award itself a subsidy not available to the good-faith implementer, and the effect on the good-faith implementer is the same as if the holdout implementer were permitted to raise its rivals’ costs directly. Thus “efficient infringement” harms the good faith implementer both directly and indirectly.

There is an additional distortion that has developed in implementing FRAND: the ability of implementers to engage in holdout distorts the notion of what is a “reasonable” rate. The common view that the price to be paid by the implementer for the intellectual property (IP) is an externality, a “tax,” invariably does not recognize that, as with any other component, the cost of the IP should have been factored into the bill of materials, as part of the total cost of goods sold. Not only does this approach distort the determination of the appropriate price to be paid for the IP, but also calibrating that price to the implementer’s “margin” encourages courts to determine FRAND rates on a base that ignores the very IP costs at issue, rewarding implementers for ignoring their obligation to pay for the IP they use. And it can lead courts to conclude that implementers with the lowest margins should pay the least, with the result that the least efficient producer gets the best outcome, a reward for inefficiency that is certainly contrary to public policy.