Monday 21 May 2012

Some Statistics


If, like me, you were a little disappointed by the information regarding the Boards of appeal offered in the latest annual report of the EPO, you might be interested in some more statistics. In what follows, I present some statistical data regarding the 443 decisions published on the DG3 webpage during the first four months of 2012. These decisions were taken between October 28, 2010, and April 11, 2012.
1. Nature

Let us first have a look at the nature of these decisions:


60 per cent of the cases were opposition appeal cases, compared to 37 per cent of examination appeal cases. “PETREV” refers to petitions for review. Note that there were still two PCT cases left.

2. Boards involved

The following Table shows the numbers of decisions per Board of appeal (click to enlarge):



3. Language of proceedings

Unsurprisingly, the three official languages of the EPO were very unevenly represented:


English took the lion’s share (68 per cent) whereas French was very little used (6 per cent).

What is more surprising is that the situation depends on whether examination or opposition appeals are considered.

In examination appeal proceedings, English was even more preponderant (81 per cent, more than twenty times the share of French (4 per cent):


In opposition appeal proceedings, however, German was relatively more often the language of the proceedings:


English was used in 61 per cent, German in 32 per cent of the cases. Might it be that German language patents attract more oppositions? That would be weird.

Obviously, the discrepancy between examination and opposition appeals may have very complex reasons that are beyond the reach of so simple a study; it might also express the fact that English is gaining momentum – because opposition proceedings, generally speaking, deal with patents based on applications that were filed well before the applications dealt with in examination appeal proceedings. (In other words, when examining opposition appeal cases, one looks further back in the past than when studying examination appeal cases.)

4. Examination appeals

Considering the examination appeal proceedings, what was the outcome?


In almost 60 per cent of the cases, the appeal was dismissed, the first instance refusal becoming final. In a quarter of the cases, the Boards ordered the Examining Division to grant a patent. In 16 per cent of the cases, the case was remitted to the first instance for further prosecution. As far as the remittals are concerned, I have not been able to go any further and examine the subsequent fate of the applications, all the more as many of these cases may not have been settled yet.

NB: The “OTHER” cases are cases where the first instance granted interlocutory revision and the Board had to decide on reimbursement of the appeal fee etc.

5. Opposition appeals

Let us now consider the opposition appeal proceedings, which are more complex to analyse.

It turns out that the three possible reasons for an appeal against a decision of the Opposition Division (rejection of the opposition(s), maintenance of the opposed patent, revocation) all had about the same weight.


What was the overall outcome?

As the following graph shows, the most likely outcome of opposition appeal proceedings was the revocation of the patent (46 per cent), followed by maintenance of the patent in amended form (30 per cent). The opposition was rejected in only 14 per cent of the cases, and in another 8 per cent of the cases the case was remitted to the Opposition Division for further processing (again, I have not studied the subsequent fate of the patent). In two cases the proceedings were terminated by the lapse of the patent in all Contracting States.


Let us now have a closer look at these data.

5.1 Appeals subsequent to a rejection of the opposition

For appeals filed by the opponent(s) after their opposition(s) had been rejected (78 cases), the most likely outcome was … the revocation of the patent (44 per cent), closely followed by a confirmation of the rejection (42 per cent). The maintenance of the patent in amended form was quite unlikely an outcome (12 per cent).


I found this result quite surprising; I would have expected to find maintenance in amended form more often. A possible reason could be that many patents do not have strong fall-back positions: when the claims as granted fall, the patent falls.

5.2 Appeals against the maintenance of the opposed patent in amended form

First of all, I was interested to see who filed the appeal: was it predominantly the patent proprietor or the opponent, or did both parties file appeals?

In the 98 cases published in 2012, it was mostly the opponent(s) alone who filed the appeal (68 per cent of the cases), whereas appeals by the patent proprietor alone were pretty rare (5 per cent):


I understand this to mean that the proprietors mostly were happy with the outcome, which confirms my (admittedly, subjective) impression that the first instance is quite proprietor-friendly in opposition proceedings.

The most likely outcome was the maintenance of the patent in amended form (62 per cent, which encompasses both a dismissal of the appeal and amendments made before the Board of appeal), followed by the revocation of the patent (32 per cent). A rejection of the opposition at this stage was relatively rare (3 per cent).


This view is, however, oversimplified because it blurs the distinction between cases where the principle of prohibition of reformatio in peius applies and cases where it does not. Therefore, it is necessary to have an even closer look.

 5.2.1 Cases where the patent proprietor was the sole appellant

Obviously, the patent cannot be revoked in this situation. As a matter of fact, in all cases but one (where the case was remitted for further prosecution) the outcome of the appeal was the maintenance in amended form:


Note, however, that there were only five cases in total, which means that the results may not be significant, statistically speaking.

 5.2.2 Cases where the opponent was the sole appellant

In such cases (67 in total) the Board could not reject the opposition. It turns out that the patent was maintained in amended form in about two thirds of the cases (64 per cent) and revoked in the remaining cases:


5.2.2 Cases where the proprietor and the opponent filed appeals

In these cases (26 so far), the principle of prohibition of reformatio in peius does not find application and all options are open to the Boards. The outcome was as follows:


In 54 per cent of the cases, the patent was finally maintained in amended form; in about 27 percent of the cases it was revoked. The rejection of the opposition was less frequent but not nonexistent (12 per cent). “FURTHER” refers to remittals for further prosecution.

5.3 Appeals against the revocation of the opposed patent

When the patent had been revoked by the Opposition Division (91 cases), there was little hope even before the Board of appeal: in two thirds of the cases (65 per cent), the Board dismissed the appeal:


I find this quite logical because, according to what I have seen in my practice, Opposition Divisions are reluctant to revoke patents, in particular in the (frequent) case where the first examiner is the one who delivered the patent. If even the Opposition Division comes to the conclusion that the patent is to be revoked, the Board is likely to agree.

6. Outlook

In case DG3 will publish further decisions in 2012 - I have not given up hope yet - I shall try to pursue these statistics. Perhaps at the end of the year it will even be possible to compare the most prolific Boards and see whether there are significant differences.

NB: My reservoir is now empty. This blog will be back as soon as fresh decisions are available.

4 comments:

Rimbaud said...

Interesting figures.
Great job, Oliver.

Anonymous said...

Oliver, this is very impressive. However, I do not like the expression "the most likely outcome of opposition appeal proceedings was ....", because you do not analyse the cases on their merits, only the outcome. For this reason I would prefer "frequent" to "likely", which has a predictive ring to it.

Kind regards,

George (Brock-Nannestad)

MaxDrei said...

In early years, the EPO Annual report revealed the nationality of the Opponent: every year, consistently, 70% were from German-speaking parties. I understood this to be a consequence of "bifurcation" of patent infringement proceedings in Germany. Any thoughts on that?

or said...

MaxDrei’s explanation appears to be correct, as I just discovered here .