The textbook policy rationale for the existence of a patent system is a quid-pro-quo -- a tradeoff in which an inventor is granted a time-limited property right over his or her invention in return for disclosing it to the public.  Such disclosure is expected to, over time, spur further innovation, and bolster research and development.  But empirical measurements of the impact of the patent system on innovation and the economy have been difficult to come by.  Patents are just one factor influencing ongoing scientific and engineering advances, and their impact is perhaps impossible to determine in an aggregate sense.

Heidi L. Williams of the National Bureau of Economic Research reviewed the available evidence of how patents on existing technologies might affect follow-on innovation and research investment.  Her conclusion is that more evidence is needed:  "[G]iven the limitations of the existing literature we still have essentially no credible empirical evidence on the seemingly simple question of whether stronger patent rights -- either longer patent terms or broader patent rights -- encourage research investments into developing new technologies."

Nonetheless, in a recent article, Vera Ranieri of the Electronic Frontier Foundation (EFF) has used this essentially neutral position to argue in favor of recent changes to U.S. patent law that arbitrarily and subjectively limit patentable subject matter.  In doing so, Ms. Ranieri falls back on a number tired tropes and misconceptions about the workings of U.S. patent law, and fails to consider the damage done by these changes to the certainty function of the patent system.

The Alice Decision

In 2014, the Supreme Court handed down a highly controversial decision in Alice Corp. v. CLS Bank Int'l.  In a nutshell, the Court ruled that patent claims should be analyzed to determine whether they are too "abstract," and if they are, these claims should be further analyzed to determine whether they incorporate "significantly more" than an abstraction.  In doing so, the Court went well beyond the plain language that Congress set forth in 35 U.S.C. § 101, the section of the patent statute that defines subject-matter eligibility.  Furthermore, the Court purposely declined to define what it meant by "abstract" and "significantly more," essentially leaving it up to the Federal Circuit, the district courts, and the USPTO to provide meaning for these ambiguous terms.

While not explicitly stating it, in Alice the Court effectively targeted software and business method inventions (and certain types of life science inventions, such as diagnostic methods) for an additional level of scrutiny.  In fact, the EFF's own amicus brief in Alice encouraged the Court to use the case to "reign in overbroad software patents."  Whether or not this argument ultimately swayed the Justices to rule as they did, the effect of Alice is clear -- over 15,000 patent claims have been invalidated under § 101 since this decision came down, the vast majority of these in the software and business method space.  Additionally, some software and most business method claims are now harder to obtain from the USPTO, with rejection rates of over 90% in some business method art units.

In many cases, courts have invalidated patents under § 101 without undertaking claim construction.  Let that sink in for a while -- courts are throwing out patents as allegedly being abstract without determining the actual scope of the claimed invention.

The Alice Decision Has Not Been, and Cannot Be, Applied Consistently or Objectively

The problem with Alice is not just that it limits the scope of patentable subject matter, but that it does so in an arbitrary and subjective fashion.  Courts and the USPTO have struggled mightily with making sense of the decision.  Judge Wu of the United States District Court for the Central District of California criticized Alice for setting forth an "I know it when I see it" test.[1]  Judge Pfaelzer, a colleague of Judge Wu, wrote that the Supreme Court's patent-eligibility cases "often confuse more than they clarify [and] appear to contradict each other on important issues."[2]  Former Chief Judge of the Federal Circuit Paul Michel stated that Alice "create[d] a standard that is too vague, too subjective, too unpredictable and impossible to administer in a coherent consistent way in the patent office or in the district courts or even in the federal circuit."

While the Federal Circuit has attempted to provide a consistent line of post-Alice case law, this has resulted in some very thin bacon slicing.  For instance, in 2014 a panel in Ultramercial v. Hulu found a complex and narrow web-based transaction to be unpatentably abstract, while another panel in DDR v. Hotels.com found a web server that combines the look and feel of two different web sites meeting the requirements of § 101.  Last year, in Synopsys v. Mentor Graphics, an invention automating circuit design procedures was found to be abstract because these procedures had been performed manually by humans.  But in McRO v. Bandai Namco Games America, an invention automating lip-synchronization in animated characters was found not abstract even though the automated procedure could be performed by humans.

While one could argue that there are at least tenuous distinctions between the claimed inventions that were found eligible and ineligible in these four cases, the recent Recognicorp v. Nintendo case clearly illustrates the Federal Circuit's inability to apply Alice consistently.  In the decision, the panel characterized a 154-word claim as nothing more than "encoding and decoding image data," despite the panels in Enfish v. Microsoft and McRo explicitly forbidding such claim over-simplification.

At the USPTO, despite the administrators' admirable efforts to provide clear examination guidelines with respect to Alice, it is notoriously well known amongst practitioners that different art units and different examiners will apply the Alice test in radically different ways.  Further, it is common for an applicant to receive an Alice rejection that is purely conclusory in nature -- the examiner essentially saying "Your claims are abstract, no patent for you!" without any rationale or reasoning to support the rejection.  While the latter type of rejection can be rebutted, it imposes additional and unnecessary costs on the applicant to go through extra cycles of Office actions or appeals.

Ultimately, the problem does not lie with the Federal Circuit or the USPTO.  The Alice test is inherently subjective, requiring one to look at a claim and imagine analogous "well-known, routine, and conventional" activities.  In doing so, examiners and judges often ignore key claim elements, resulting in a 10,000-foot view of the claim being analyzed, rather than the actual claimed invention.

Due to this subjectivity, reasonable minds can differ regarding whether a claim is abstract and whether it contains something more.  There is no way for thousands of examiners and hundreds of federal judges to reach the same conclusion on a consistent basis.

The Lack of Evidence Found in the Williams Paper Cuts Both Ways

For sake of argument, let's assume that Ms. Williams's analysis is correct, and that there is little or no evidence identified yet that establishes innovation is spurred on by a strong patent system.[3]  Ms. Ranieri interprets this result as an invitation to slam the door on software patents, arguing that there is no need to change the law to fix problems introduced by Alice.

But the same argument could be used to establish that there was no need for the Supreme Court to change the law in Alice either.  If the impact of patent system strength on innovation was unclear, the Court should not have instituted such a dramatic change.  Besides, Congress is better equipped than the Court to derive policy from such studies.  If anything, the dramatic growth of technology company profitability and market capitalization from the mid-1990s to the Alice decision in 2014 suggests that a strong patent system is not inconsistent with innovation and technological investment in the software market.

Ms. Ranieri also states that the Williams paper establishes that there is no evidence that the Alice decision has "done any harm to the innovation economy or innovation generally."  But the paper does not address Alice at all, and most of its cited research was conducted prior to the 2014 date of Alice.  Given the lengthy, multi-year time scales of patent lawsuits, patent prosecution, and research and development, the three years between then and now is way too short a time period to draw such broad conclusions.

Ms. Ranieri Mischaracterizes the Efforts to Overturn Alice

Throughout her article, Ms. Ranieri makes misleading statements about efforts to overturn Alice.  Particularly, she focuses on proposals by the Intellectual Property Owners Association and the American Intellectual Property Law Association to rewrite § 101.  A comparison of their efforts indicates that they would replace the Alice test with an analysis of whether the claimed invention as a whole (i) exists in nature independently of human activity, or (ii) solely in the human mind.  Such a test would certainly result in fewer patents being invalidated under § 101, but would also go a long way toward clarifying the law.

Unlike Ms. Ranieri's contentions, this would not "return us to a world where 'do it on a computer' ideas are eligible for a patent."  That world never existed.  If a claimed invention's sole difference from prior art is to perform a known series of steps on a computer, the invention is probably obvious and can be invalidated on those grounds.  In fact, the claims in Alice and many other so-called "overly broad" patents could be invalidated under the patent statute's existing novelty, non-obviousness, and written description provisions.  Ms. Ranieri's further statement that the proposed changes to § 101 would mean that "anything is patentable" aside from the aforementioned exceptions ignores the fact that eligible subject matter is just one requirement of many for a patent to be granted.

Furthermore, Ms. Ranieri asserts that "the recent reform proposals seem like little more than a bid by lawyers to create work for themselves."  If anything, the opposite will likely be the case.  The uncertainty surrounding Alice has been a boon for § 101 focused practice, opening more patent prosecution and litigation opportunities.  While some applicants are filing fewer patents as a result of Alice, many are filing more -- attempting to protect multiple narrow inventions rather than one or two broader ones.  There are more appeals in the USPTO, as well as more covered business method review proceedings as well.

When there is a gray area in the law, lawyers are needed more than ever to help their clients navigate the ambiguities.  If the Intellectual Property Owners Association and the American Intellectual Property Law Association succeed in their efforts, the patent law will be less ambiguous and there will be far fewer billable hours spent determining whether or not a claim is abstract.  Even though such a change might have a negative impact on their bottom lines, many patent attorneys would welcome the change because the patent system is broken as long as Alice stands.

Conclusion

Opposition to the Alice test is not based on it making patents easier to invalidate, but that it does so in an unworkable and intellectually dishonest fashion.  This has resulted in applicants with limited financial resources being unable to pursue patent protection for legitimate technical inventions.  If such a trend is widespread and continuing, small U.S. companies will be reluctant to innovate in certain market segments because of the risk of their U.S.-based or foreign competitors getting a free ride to copying or cloning a product or service.

This is the area that merits academic study -- has Alice actually harmed innovation?  As noted above, it may be a while before we can make that determination.  But if it is the case, that would be just one more reason of many for discarding the test.

[1] Eclipse IP v. McKinley Equipment (C.D. Cal. 2014).

[2] Cal. Inst. of Tech. v. Hughes Commc’ns Inc. (C.D. Cal. 2014).

[3] As far as we know, the paper has not yet been subject to thorough peer review, and therefore its conclusions should be viewed as preliminary at best.