We’ve blogged in the past about the ILPTO’s strange, ultra vires policy announcement in early 2010 to forbid the filing of second-generation divisional applications (2GDs) after the grandparent case has been published for opposition purposes. That announcement, which was based on an absolutely untenable reading of the statute, and which contradicted over 40 years of practice under the statute, came not in the context of a decision in an opposition (where one might have expected to see such a statutory challenge to the grant of a patent raised) but in a circular issued sua sponte by the then-commissioner.
In a decision (in Hebrew) issued on January 12, 2016 in an ex parte proceeding in IL 234696, the present Commissioner has taken a step back from that view, and may be on his way to returning to the status quo ante in which it was possible to file a chain of divisional applications. Unfortunately, the Commissioner’s thinking in this matter is muddled, as will be explained below, and the decision does not contain a clear repudiation of the ILPTO’s anti-2GD stance. We’ll believe that the anti-divisional policy has been killed off when we see an announcement from the ILPTO to that effect.
Background: Statutory Basis for Divisionals in Israel and the 2010 Circular
Section 24(a) of the Israel patent statute says, “As long as the application has not been accepted [i.e. published for opposition purposes – DJF], the applicant is entitled to demand that it be divided into several applications.” Section 24(b) says, “If the application includes more than one invention, then the Registrar may, as long as he has not accepted the application, direct the applicant to divide his application.”
From 1967 when the patent statute was enacted until the spring of 2010, 24(a) was understood by all as referring to any given application: any application could be divided, provided it had not yet been published for opposition purposes. Thus a first divisional could be divided from an original application, which, while the first divisional was pending, could publish for opposition and then issue as a patent. Thereafter, during the pendency of the first divisional but after the publication of the original case, a second divisional could be filed from the first divisional, notwithstanding the publication of the original application in the interim. Both the first and second divisionals would have an effective filing date of the original case and would enjoy any priority claims made therein. The process could be repeated ad infinitum, subject to the 20-year term, which for all the applications in the series would run from the filing date of the original application.
In February 2010, in one of many circulars issued during his tenure, the then-commissioner announced that henceforth, the ILPTO would interpret 24(a) as referring only to the original application. (He subsequently pushed off implementation of the circular by several months.) Thus, once the original application published for opposition purposes, no 2nd- or higher-generation divisionals could be filed from any already-filed, still-pending divisional applications. The rationale given in the circular for this new “policy” was “the need to conclude examination proceedings in a reasonable time and because of considerations of legal certainty for the general public”. In other words, the ability to file divisionals ad infinitum kept “the public” uninformed about the acts that might be enjoined should a grandchild patent issue many years after the first patent.
As we noted then, the problem with this view is that statutory interpretation and administrative polices are different creatures. Administrative policies, like how many months’ extension the Commissioner will grant before a show of cause is required to obtain additional extensions, can only be implemented prospectively, and there may be a delay between an administrative policy being announced and its implementation. Statutory interpretation, by contrast, has immediate effect, both prospectively and retrospectively. If the statute precludes the filing of 2GDs, then (a) all patents granted in such cases until 2010 time were void, and (b) the ILPTO needed to stop accepting such applications immediately, not several months after the publication of the circular, as the ILPTO stated would be the case.
Challenge to the ILPTO
Since the new “policy” hinged on statutory interpretation, but without retrospective effect, it was only a matter of time until an applicant with a sufficiently valuable patent application and sufficiently deep pockets would challenge the ILPTO. This occurred in late 2014, when Genentech sought to file a 2GD, IL 234696, related to the antibody that is the active ingredient in its blockbuster drug Herceptin.
Aware of the 2010 circular, Genentech asked the ILPTO nevertheless accept the 2GD as a divisional. Genentech raised several grounds for its request, among them that the circular was based on an untenable reading of 24(a), and as a result was ultra vires; that the policy announced in the circular failed to achieve its stated goal, viz. more certainty for “the public”, and therefore should be withdrawn; and that proper statutory interpretation applies retroactively, thus invalidating many already-granted patents if the interpretation proffered by the ILPTO was correct. Genentech did, however, ask that if the Commissioner disagreed about the proper reading of 24(a), and about the scope of his powers (i.e. if the Commissioner believed that the 2010 circular was not invalid), that he nevertheless grant an exception in this particular case and allow the filing of the 2GD. In January 2015 Genentech pled its case in an ex parte hearing before the Commissioner.
On January 12, 2016 (yes, a full year after the oral hearing), the Commissioner delivered his decision: Genentech’s application 234696 would be accorded the status of a divisional application.
While the decision is being heralded by the firm that represented Genentech as a reversal of policy and a green light for the filing of 2GDs, in this corner we view the decision as moving in that direction, but not quite getting there. Nowhere does the decision explicitly say that the statutory interpretation expressed in the 2010 circular is wrong, nor does the decision say that that circular is rescinded. The decision does say that it’s time to reconsider the 2010 circular, in light of changed circumstances. But the actual holding in the decision is that in this particular case, the 2GD shall be treated as a proper divisional. Hopefully a rescinding of the 2010 circular is in the works, but until that happens, it’s unclear where the ILPTO stands on the issue of 2GDs generally.
Far more telling is the reasoning, vel non, of the decision. The decision starts out, correctly, saying that this is a matter of statutory interpretation - viz. when S.24(a) refers to “the application” to be divided, does it mean the first application filed in a chain, or does it mean any pending application? But then, rather than simply state “the correct interpretation of the statute is that “the application” refers to any application”, which was what Genentech asked the Commissioner to do, the decision veers in the direction of administrative law, treating statutory interpretation as if this is a matter for PTO administrative policy. The Commissioner thus spends pages and pages justifying why he can now, in accordance with Israeli administrative law, consider deviating from the 2010 circular. This discussion includes statistics on numbers of divisional filings pre- and post-2010 and their percentages among the overall numbers of applications; how the matter of 2GDs is treated in the US, EPO, UK, AU and JP; and public policy concerns about lack of proper notice to “the public” if chains of divisionals are filed. Such public policy concerns are, of course, not the province of the Commissioner but of the legislature.
After this long-winded discourse, the Commissioner states that perhaps the time has come to reevaluate the 2010 circular, since it doesn’t seem to have achieved its purpose (the number of divisionals filed annually since 2010 is up versus the four-year period preceding the circular), and the ILPTO has significantly reduced examination times as well. The Commissioner also acknowledges that in some cases, such as the present one, the early issuance of patents with narrow claims and the later pendency of other applications divided out from the earlier applications may “serve the aims of the statute and the desired legal certainty for the public at large”. He also wonders if implementing a progressive fee scale for divisional applications might not be a bad thing. But his ruling is that, “In view of the foregoing, I find no reason to apply the instructions of Circular MN 81 to the application before me, and I see no reason to prevent further examination of the ‘696 application…. In conclusion, the ‘696 application shall be examined in accordance with the date on which it was filed and in the regular examination queue.”
While reaching the correct conclusion – the Genetech ‘696 application should be treated as a divisional of a divisional – there are a number of problems with the Commissioner’s approach in the Genentech decision. Chief among these is the decision’s implicit assumption that administrative law is controlling in this case. That’s plainly incorrect: there's only one correct interpretation of the statute, and statutory interpretation is NOT a matter of PTO administrative policy. Consequently, the Commissioner doesn’t need to engage in hand-waving or go looking for reasons to rescind the 2010 circular, because that circular is not the sort of pronouncement that’s due any deference by subsequent commissioners, or for which changed circumstances need to be considered if the circular is to be revised or trashed. If the view reflected in the circular is correct (and by implication, the practice of the ILPTO for the 43 years before the circular was wrong), then ALL 2GDs are invalid and they always have been, and the ILPTO must continue to refuse to accept them, including in Genentech’s case. If the view in the circular is wrong (which it is), then the Commissioner needed to say so in the Genentech decision, and to say that the 2010 circular is void, because the statute grants applicants the right to file divisionals from ANY pending application, period.
Remarkably, Genentech made these very arguments in its pleadings, but the Commissioner did not deal with them head-on in the decision. It appears the Commissioner was reluctant to admit that his predecessor had exceeded his powers, and that the present Commissioner was complicit in this by failing to rescind the circular years ago (a reluctance reinforced by the year between the oral arguments in the case and the issuance of the decision). One wonders how the Commissioner would have decided this case had Genentech not pled an alternative ground for allowing its application to proceed, viz. that the Commissioner should exercise his discretion and not apply the 2010 circular in this case.
The Commissioner’s accounting for public policy considerations, particularly the “need” of some amorphous “public” for “legal certainty”, viz. to know whether or not that “public” is entitled practice a particular invention or not, is also in error, although entirely typical of Israel Patent Commissioners going back for many years. To the extent that the needs of an amorphous “public” are to be accounted for, the statute itself does that: it says that applicants can file divisonals of divisonals. End of discussion. If a particular individual (say, a multinational pharmaceutical company based in Israel) feels harmed by a particular allowed patent application, the statute gives that individual the power to file a pre-grant opposition to the grant of the patent (which is what happened to applications nos. 136201 and 214084, the grandparent and parent, respectively, of the ‘696 application). But it’s not the place of the Commissioner to re-write the statute in accordance with his perceptions of the need to protect some unspecified “public”.
An additional point, less germane to the analysis of the decision per se but nevertheless instructive, is that the statistical analysis provided in the decision posits that the rise in divisionals since 2010 is due to the issuance of the circular. While that circular certainly played a role, the rise in divisional filings is likely due in part to additional factors, beyond the MN 81 circular. In particular, the adoption of regulations in December 2009 that impose a fee of 500 shekels for each additional claim above 50 probably influenced the increase in divisional filings, since the filing fee for a new application was set at around 2000 shekels. Additionally, not long before the issuance of the 2010 circular, the ILPTO issued a different circular in which it announced that an examiner’s finding of lack of unity of invention under section 8 of the statute constituted an order under 24(b) to divide the application. Per that earlier circular, the filing of a divisional after the allotted time respond to the disunity requires the payment of extension fees. This policy likely also contributed to the rise in the filing of divisional applications. Why neither of these additional factors was mentioned in the Genentech decision is both enigmatic and disconcerting, as it suggests a poor grasp of patent prosecution by the head of the ILPTO.
Ultimately, one gets the feeling from the Genetech decision that the Commissioner intends to revise or rescind the 2010 circular, but on the grounds that circumstances have changed and the policy expressed in the circular no longer needs to be implemented, not because the circular expressed an incorrect view of the statute ab initio. We eagerly await the issuance of such a pronouncement.