The Federal Circuit's fractured en banc decision in CLS Bank v. Alice probably pleased only those patent litigators who might benefit from the resulting uncertainty. The case could have provided clear guidance on the "abstract ideas" exception to patent eligibility (and thus the patentability of software), but the court instead issued 7 opinions in 135 pages, with nothing beyond the judgment having the weight of precedent. While there is much to be disappointed in here, I want to highlight a statement on page 12 of Judge Newman's opinion: "No substitute has been devised for the incentive of profit opportunity through market exclusivity."
No substitute? Patents are hardly the only tool for incentivizing innovation—most obviously, the U.S. government provides $130-140 billion of direct R&D support each year, and there are many proposals to replace (or supplement) patents with prizes. And what about tax incentives for R&D? As Daniel Hemel and I explain in Beyond the Patents-Prizes Debate (forthcoming in the Texas Law Review), the U.S. already spends over $10 billion each year on R&D tax incentives. We argue that tax incentives share patents' advantage of relying on private actors (not government officials) to decide which innovations to pursue. And because tax incentives can provide funding earlier in the R&D process, they can be preferable to patents for risky projects requiring significant upfront capital, for which the incentive of patents is dulled due to the delayed and speculative nature of the reward.
As support for her statement, Judge Newman writes that "Illustration is seen in the Orphan Drug Act . . . which provides patent-like exclusivity and is reported to have provided treatment for many previously untreated diseases." But the Orphan Drug Act only serves to illustrate the diversity of the innovation policy toolkit: as we explain in Beyond the Patents-Prizes Debate, the Orphan Drug Act took three approaches to increasing innovation incentives—in addition to allowing a seven-year period of patent-like exclusivity, it increased government grants for orphan drug development, and it allowed pharmaceutical companies to claim a tax credit for 50% of their clinical testing expenses for orphan drugs. Analyses of the Orphan Drug Act's success have not attempted to disentangle the effects of these different provisions. (And in any case, the general success of patents for pharmaceuticals reveals little about optimal innovation policy for software.)
While Judge Newman would prefer a more expansive understanding of patent eligibility, suppose Judge Moore is correct that the CLS Bank decision will lead to "the death of hundreds of thousands of patents, including all business method, financial system, and software patents." And suppose (controversially) there are many welfare-enhancing software projects (i.e., those for which the expected value to society is greater than the expected cost) that would not be pursued absent government intervention (because the expected private benefit from mechanisms such as first-mover advantage is less than the expected cost). Couldn't these projects be incentivized through larger research tax credits for software development, perhaps with an amendment to make it more useful to startups? Indeed, tax credits may even be preferable to patents in the software context due to lower administrative and transaction costs—this latest failure to provide clear guidance only further illustrates the general dysfunction of patents in this field.
Perhaps Judge Newman would respond that even if tax credits, grants, or prizes could effectively substitute for software patents, shifting incentives toward alternative policies can only be done by Congress, not the courts. But even if judges can't bring about alternatives to the patent system, they should not imply that such alternatives don't exist.