The new patent frontier

Out of sheer laziness, I am usually reluctant to report on bills that are still in the legislative process.

After all, you never know whether your report will still be consistent with what comes out of the process in the end. You may call that the UPC syndrome.

But the reform of the French patent granting procedure currently in gestation represents such an upheaval of the world as we French patent attorneys know it that it is getting more and more difficult to ignore it on this blog.

So, today, I will update readers on the “Projet de loi relatif à la croissance et la transformation des entreprises“, also known as the PACTE bill. Don’t ask me what the A in the acronym stands for. It is probably just easier to pronounce than “PLRCTE“, that’s all.

This is a very complex and long (some would say catch-all) bill that spans labor law, tax law, commercial law, etc. And there are a few provisions on patent law, which are rather far-reaching.

The bill, updated as of today, can be found here. All provisions have now been approved by the Assemblée nationale (the lower chamber of our parliament), which prompted today’s post. Please bear in mind that these provisions can still be modified during the rest of the legislative process though, but by now it is more likely than not that they will indeed become law.

The first relevant part of the bill is article 40, which deals with utility certificates. That’s the local name for utility models. These are seldom used in this country and probably not very well known. The three current main features of the utility certificate are the following:

  • no search report is established for a utility certificate;
  • the maximum duration of the utility certificate is 6 years from the filing date; and
  • a patent application can be converted into a utility certificate application but not the other way around.

In order to make the utility certificate a more attractive option for innovators, the maximum duration of the certificate will be brought to 10 years instead of 6. Its duration will thus be on a par with that of utility models in foreign countries, notably Germany.

A second change is that it will now be possible to convert a utility certificate application into a patent application. The conditions for this conversion will be further specified by way of an executive order.

If we assume that flexibility is always appreciated by applicants, these changes will indeed make the utility certificate more palatable to them. While we are at looking at what the Germans do, we might as well have imported the “Abzweigung“, i.e. the possibility to file a utility model application as a split-off from a patent (including European) application. This can be quite a powerful tool for applicants. Maybe next time – together with a (much awaited) re-opening of the French national phase PCT route?

The second topic addressed in the bill is a much bigger prize, though.

Article 42 authorizes the government to create national opposition proceedings, so as to allow third parties to request the revocation or modification of a French patent, by way of an “ordonnance”. 

An “ordonnance” is a special kind of executive order which has the same effect as a law passed by the parliament, provided that it is later ratified by the parliament.

There are absolutely no details as to this future opposition in the bill, as the parliament has precisely surrendered its prerogative to set rules on this matter to the government, by way of this article 42. So we will have to wait until we see the “ordonnance”. 

For what it is worth, a survey was conducted by the French patent and trademark office (INPI) on this topic a few months ago. Among the questions which were addressed in the survey, were the following:

  • Should any person be allowed to file an opposition, like at the EPO, or should an opposition be reserved to persons and entities having standing? (And, boy, we know that the appraisal of standing in nullity suits can be tricky.)
  • Should the opposition period be set to 6, 9, or 12 months?
  • What should the grounds for opposition be?
  • Should it be possible to file an opposition against a granted utility certificate?
  • Should the decision on the opposition be open to appeal in front of the Paris Tribunal de grande instance (first instance court) or in front of the Paris Cour d’appel (appeal court)?
  • What should the effect of an opposition be on parallel litigation, in particular concerning a possible stay of proceedings?
  • Should oral proceedings be summoned?
  • Should an accused infringer be able to intervene in opposition proceedings?
  • Should the examiner who granted the patent be part of the panel of three examiners who will decide on the opposition?
  • Etc.

As you can see, things were (still are?) very open. Rumor has it that the INPI is contemplating a rather extensive alignment of the future French opposition proceedings on the EPO model.

As a side note, article 42 of the bill specifies that the future “ordonnance” should make sure to prevent abusive oppositions. I have no idea what they mean by this, but I do hope that there will be no standing required to oppose a patent. This would certainly lead to useless complexity and legal uncertainty.

PACTE, the final frontier?

Now, the third topic of interest in the bill is as big as the second one – if not bigger.

At present, the INPI has limited power to refuse a patent application. Basically, a French examiner can raise objections (such as lack of clarity or unity) before the issuance of the search report. Then the search report and a complete written opinion are drafted, either by the EPO or directly by the French examiner. Then the applicant files observations as a response – which can very well be very brief. And a refusal can only take place in exceptional circumstances, e.g. if it is “blatant” that the application relates to subject-matter which is not an invention (such as discoveries, aesthetic creations, etc.); or if there is a “blatant” lack of novelty.

There are further (more minor) grounds for refusal, such as lack of unity. But the major fact is that lack of inventive step is not a ground for refusal – although of course it is a ground for nullity of the issued patent in front of a court. And lack of novelty or lack of patentability are grounds for refusal only insofar as the objection is crystal clear and there is no possible defense.

Well, it seems like this fundamental peculiarity of French patent law is about to become a thing of the past.

Article 42bis of the bill will make it possible for the INPI to refuse an application for lack of novelty or non-patentability, period (i.e. the objection will no longer need to be blatant); and also for lack of inventive step.

A huge change indeed. Hailed by some, as it is believed to increase the value of French patents. Lambasted by others, as applicants (especially local ones) may be less incentivized to file national applications if the local granting procedure becomes similar to the European one.

To me, the success or failure of the above reforms will mostly depend on their implementation by the INPI.

Tremendous efforts will have to be undertaken to hire and properly train new examiners to perform these tasks. Let’s hope that this aspect has not been underestimated by the government and that sufficient funding will be available to support these efforts.

One good point is that the new, beefed up, granting procedure will only come into force two years from the entry into force of the PACTE law. This transition period will certainly be necessary in order for everyone to get ready for this new enterprise, in other words “to explore strange new worlds, […] to boldly go where no man has gone before“.

And by the way, who said the unitary patent and the UPC would be the kiss of death for national patent rights?

PS: for those eager to know everything about the debates that took place in the Assemblée in connection with the above provisions, here is the link to the full transcript. Don’t hold your breath though, it is fairly high level and there was no discussion on the nuts and bolts of the new system.

Suspect suspicions

What are the top three traits of the French system that foreign companies like best? 

Good transportation infrastructure? Maybe. L’art de vivre? Very probably. But first and foremost: the saisie-contrefaçon. OK, there may be some patent attorney bias here.

As readers of this blog are surely aware, the saisie-contrefaçon (infringement seizure) is a special tool in the hands of patent right holders, in the form of an ex parte order for an unannounced inspection at a third party’s premises in order to gather evidence of possible infringement.

Traditionally, the order for seizure has been viewed as straightforward to obtain.

Basically, you show up in front of a judge; you show your patent, the status of the renewal fees, an excerpt from the patent register; then you explain what kind of evidence you want to get and where. And you’re pretty much good to go.

But from time to time, a court decision takes a different approach and makes this traditional view entirely derail.

Such is the judgment recently issued by the Paris Cour d’appel in Arconic v. Constellium Issoire.

Arconic owns European patent No. EP 1392878 on a process to produce sheets of aluminum.

It may be useful to copy here claim 1 of the patent, as a reference:

A process for producing a sheet or plate product comprising:
(a) providing an aluminium alloy consisting of 0.5 to 1.8% Si, 0.5 to 1.5% Mg, up to 1.2% Cu; up to 1% Mn, up to 1% Zn, up to 0.4% Cr, up to 0.5% Ag, up to 0.3% Sc, up to 0.2% V, up to 0.2% Hf, up to 0.2% Zr, the balance being aluminium and incidental impurities;
(b) heating the alloy;
(c) hot rolling the alloy to reduce its thickness by at least 30%;
(d) thermally treating the alloy hot rolled in (c) at 543°C (1010°F) or more;
(e) further hot rolling the alloy to further reduce its thickness;
(f) solution heat treating the alloy at 543°C (1010°F) or higher;
(g) quenching the alloy.

On October 4, 2017, Arconic requested and obtained an order for performing a seizure in the premises of Constellium Issoire, which belongs to the Constellium group. Arconic and Constellium are direct competitors as they both manufacture aluminum plate products, notably for Airbus’ A380 aircraft.

The seizure took place on October 10, 2017. On October 26, Constellium filed a motion to obtain the cancellation of the order for seizure. The judge in charge of urgency proceedings granted Constellium’s motion and issued an order on December 22, 2017 canceling the order for seizure.

Arconic appealed, but its appeal has now been dismissed by the Paris Cour d’appel.

So what was the problem with the initial order for seizure, and why did it deserve to be canceled, according to the appeal judges?

The focus of the discussion was Arconic’s statement in its initial request that they had reasons to suspect that Constellium infringed the patent at stake.

Arconic provided evidence for this suspicion, and the court performed a thorough review of said evidence.

The evidence comprised in particular a scientific article from three Constellium employees, a specification from Airbus, a standard and a publication on aluminum alloys, and an excerpt from Constellium’s website.

Based on this evidence, the court concluded that:

  • Arconic’s aluminum alloy is designated as 6013 while Constellium’s is designated as 6156.
  • Both alloys, based on Airbus’ specification, are made by lamination and a so-called T4 thermal treatment.

But the court then stated that this was insufficient to infer that most features of claim 1 of the patent at stake were implemented by Constellium.

In particular, the fact that the 6156 alloy is made by lamination and thermal treatment is perfectly conventional. Claim 1 of EP’878, on the other hand, requires a precise sequence of process steps, including the application of a certain temperature, a defined reduction of thickness of the alloy, etc. Thus, the court said, Arconic in fact did not know which process is used by Constellium.

Admittedly, the composition of the alloy itself, which is also recited in the form of elemental weight ranges in claim 1 of the patent, does cover the 6156 alloy. But the claimed ranges are sufficiently general to also cover further competitors’ products of the same category (the 6061 and 6082 alloys). Besides, it is clear when reading the patent that the invention is about a manufacturing process, not about a new alloy, added the court.

The tin man – which alloy is he made of?

Another reason for Arconic’s infringement suspicion was that, according to them, the mechanical properties of the plates made by Constellium could only be obtained with the patented manufacturing process.

The court accepted that Constellium’s 6156 plates seem to have very similar properties to Arconic’s 6013 plates, since both products are used by Airbus for the A380 aircraft.

But, the court continued, this does not imply that Constellium’s plates are necessarily made with the patented process, since different processes can of course lead to the same product.

As a further remark, the court noted that Constellium itself owns a later patent (EP 1809779) for a process of making 6056 or 6156 aluminum alloy products, and asserts that their industrial process is based on this patent.

As a conclusion, the court held:

Thus, the reasons set out in support of the request [for seizure] turn out to be wrong. More particularly, it cannot be derived therefrom any “suspicion” of infringement, nor “good reasons to assume” that the alloy plates supplied by Constellium would infringe Arconic’s patent. 

The final part of the judgment contains a more legal discussion. Arconic argued that its right to a seizure

only depends on the demonstration that the right at stake exists, without any further requirement of evidence concerning infringement itself.

As I said in the beginning, at the risk of oversimplifying, this is indeed probably the general opinion among the French patent profession.

Arconic in particular pointed to the so-called “enforcement directive” No. 2004/48/EC. Article 7 of the directive begins like this:

Member States shall ensure that, even before the commencement of proceedings on the merits of the case, the competent judicial authorities may, on application by a party who has presented reasonably available evidence to support his claims that his intellectual property right has been infringed or is about to be infringed, order prompt and effective provisional measures to preserve relevant evidence in respect of the alleged infringement, subject to the protection of confidential information.

The directive thus refers to the presentation of “reasonably available evidence” as a prerequisite for obtaining e.g. an order for seizure. But, Arconic remarked, this notion was not transposed in the relevant provision of the Code de la propriété intellectuelle, namely article L. 615-5. Since article 2 of the directive makes it possible for Member States to provide means which are more favorable to right holders, this must be interpreted to the effect that French law does specifically not require “reasonably available evidence” as a precondition for an order for seizure.

The court did not frontally disagree, but replied as follows.

[…] The property right conferred by a patent is neither absolute nor discretionary, and is still subject, in the case of a request for seizure, to the appraisal of its merits by the judge […], notably when, as is the case here, the reasons set forth in its support turn out to be wrong. 

Besides, article 3 of the [enforcement directive] requires that proceedings which are necessary to ensure that IP rights are respected must be fair and proportionate. In the present case, as there is no tangible suspicion of infringement, the court can only remark that the requested seizure measure would, as a main consequence, allow Arconic to know about Constellium’s aluminum alloy product manufacturing process. As these two companies are direct competitors regarding this products in front of companies in the aeronautic business, such as Airbus, Constellium can rightly fear that manufacturing secrets could detrimentally be disclosed, or even that its credibility with common customers could be harmed. Lastly, the court is surprised that it is only on October 4, 2017 that Arconic decided to initiate infringement seizure proceedings, although this direct competition with respect to Airbus dates back to 2004. 

Matthieu Dhenne, whose firm represented the patent proprietor in this case, wrote to me that there are a number of reasons why he believes the decision was wrongly decided.

First, he says the proportionality principle does not apply to an infringement seizure which remains an exceptional measure and is disproportionate by nature. He states that the travaux préparatoires concerning the transposition of the directive show that the proportionality principle was voluntarily not introduced into our statute. As a result, the court’s line of reasoning compromises legal certainty and jeopardizes the entire balance of the system. He also insists that a seizure is intended for gathering evidence, not appraising infringement. Matthieu adds that the court’s suspicion of the patentee was unwarranted and that it cannot be possibly demanded from the patentee that they should search for patents owned by the defendant. Whether any other, especially later, patent is indeed implemented or not by the defendant is irrelevant to the possible infringement of the main patent at stake, he argues.

I agree that there are a number of statements in the two above paragraphs quoted from the judgment which are controversial.

It is true that a seizure can result in business secrets being disclosed to a competitor. But the usual way to prevent this is for the seized party to request that the seized information and documents be sealed, and that an expert be appointed by the court to sort out the evidence, so that only information directly relevant to the demonstration of infringement be communicated to the right holder.

As for the fear that the defendant’s reputation be harmed, it seems irrelevant here, as the seizure was performed in Constellium’s premises, not at a third party’s.

The fact that Arconic waited a lot before initiating legal proceedings should probably not come into play either. Indeed, a patentee’s right to sue for infringement is never time-barred.

That said, I am not sure I am fully comfortable with the notion that a patentee’s right to a seizure should be automatic.

It does make sense for the court to carefully look at the specifics of each case before deciding whether this right can be exerted in the circumstances at hand or not.

In this particular case, it seems to me that what the court really did not like was that the “suspicions” of infringement mentioned in the original request for seizure, were not actual or legitimate suspicions, based on the evidence offered.

Would the outcome have been favorable if the plaintiff had played it differently, for example by (1) not mentioning suspicions of infringement but explaining that a seizure is the only possible way for them to determine whether there is any infringement; and (2) immediately offering that the seized evidence be sealed by default so as to be later remitted to a court-appointed expert?

I guess we will never know and can only have suspicions in this respect.


CASE REFERENCE: Cour d’appel de Paris, pôle 5 chambre 1, September 11, 2018, Arconic Inc. v. SAS Constellium Issoire, RG 18/01099.