Saturday, 25 May 2013

Looking Back On 2012 (Part 2)

Although the comments in response to my review of the examination appeals in 2012 (here) were mostly critical, the post was quite popular, which I interpret to mean that the topic is not without interest. 

So let us now have a look at opposition appeals.

The corresponding data are more complex and, therefore, more difficult to present than the outcome of examination appeals.

The following data are extracted from 757 decisions issued between January 10 and December 20, 2012. Needless to say, there may be 2012 decisions that have not yet been published. My extra-sensorial capacities being what they are, I have not been able to take them into account.

English was the most common language of the proceedings (62%; to be compared with 78% for examination appeals) but the share of German was not negligible (almost 29%) and French took a 9% share. That German is more often used than in examination proceedings might be a consequence of bifurcation of patent infringement proceedings in Germany.

In principle, there are three possible outcomes to oppositions. The opposition can be rejected, the patent can be maintained in amended form or even revoked. When we consider opposition appeals, we have to deal with a situation that is more complicated than for examination appeals because almost all examination appeals have their origin in a refusal of the application, whereas first instance opposition proceedings have three possible outcomes. Depending on who files the appeal, the range of action of the Board of appeal may be limited, because of the prohibition of reformatio in peius.

Let us first consider what the origin of the 757 opposition appeals was:

Most appeals resulted from a decision maintaining the opposed patent in amended form (43%). Revocation appeals were second (almost 36%). Almost 21% of the appeals had their origin in a rejection of the opposition. The single other case was related to a correction.

Before we go into details, let me present you the overall outcome for oppositions that went before the Boards of appeal. If we consider only the 669 cases where the Boards took a final decision on the outcome of the opposition (leaving aside remittals for further prosecution, lapses of the patent and decisions on more exotic cases), we obtain the following graph.

In opposition cases that went through first and second instance, the most frequent outcome in 2012 was the revocation of the patent (more than 52%). The patent survived without amendments in only 13% of the cases.

Please note, however, that this graph does not show the overall fate of oppositions because, obviously, my results do not take account of oppositions that were terminated after the first instance decision (i.e. where no appeal was filed).

If we add the cases where the Boards remitted to the first instance (but leave aside the more exotic cases), we obtain the following graph (750 cases):

I have undertaken a comparison of the Technical Boards of appeal on this basis. Please note that this is only a snapshot and that the numbers of decisions per Board (indicated in brackets) are too low for allowing to draw statistically significant conclusions.

(click to enlarge)

Although there were considerable differences between the Boards, the situation varied not too much as a function of the technical field to which the Boards belong (although mechanics have revoked less).

Let us now delve into more detail. For obvious reasons, I will limit myself to global observations; the numbers of decisions per Board are too small to allow for reasonable comparisons between Boards.

First, I have had a closer look at first instance rejections of the opposition. In such cases only the opponent is adversely affected but all the options are available to the Board. As can be seen from the graph, in the 158 cases on which a decision was taken in 2012 the Boards either confirmed the rejection (i.e. dismissed the appeal; 40%) or revoked the patent (41%); maintenance in amended form was relatively rare (14%).

A possible reason could be that many patents do not have strong fall-back positions: when the claims as granted fall, the patent falls.

When the OD had revoked the opposed patent (269 cases on which a decision was taken in 2012) the Boards confirmed the revocation in a majority of cases (55%).

The second most frequent outcome was a remittal to the first instance for further processing (23%), followed by maintenance in amended form (almost 17%). The rejection of the opposition after a first-instance revocation was relatively rare (4.5%).

Perhaps the most interesting cases are the ones where the OD maintained the patent in amended form. As the prohibition of reformatio in peius applies to some of these cases it is necessary to have an even closer look.

First of all, one may ask oneself whether it was the opponents or the proprietors (or both) that appealed against such decisions. As the following graph shows, it was more often the opponents that were unhappy with the outcome of the first instance proceedings. 

This might be an indication that the OD merit their reputation of being proprietor-friendly. Be that as it may, let us now have a look of what the outcome of such appeals was.

In cases where both the proprietor and the opponent file appeals, there is no prohibition of reformatio in peius, and the Board has all the options. It turns out that in 2012 the Board revoked the patent in almost half of the 103 cases (47%); further amendments was the second most frequent outcome (32%) whereas a dismissal of the appeal (leaving the patent as amended by the OD) and rejection were less frequently observed (10% and 7%, respectively). In 4 cases the Board remitted the case to the OD for further processing.

That revocation was so frequent in this situation came as a surprise to me.

In cases where only the patent proprietor appeals, the prohibition of reformatio in peius hinders the Boards from revoking the patent. There were only 29 such cases in 2012. The outcome was as follows:

It turns out that in a majority of cases (almost 60%) the appeal was dismissed; rejection and further amendments each occurred in almost 20% of the cases.

Finally, in cases where only the opponent appeals, the prohibition of reformatio in peius hinders the Boards from rejecting the opposition. There were 194 such cases in 2012. The following graph shows the overall result:

In almost half of the cases (46%), the opponent obtained the revocation of the patent; in 22% of the cases the patent proprietor could save the patent by means of further amendments. The appeal was dismissed in about 28% of the cases. All in all, the opponents improved their situation in about two thirds of the cases.

NB: I have tried to be very careful, but as such a survey involves big amounts of data, I cannot guarantee that the above figures are free of error. If you spot a mistake, please let me know.


Roufousse T. Fairfly said...


Where does your data set come from?

Did you assemble it manually during countless hours of reading decisions (in that case: WOW! and that would be in addition to translating them for an almost daily posting)?

Or are these data available in a predigested form from the EPO?

Does your set include "stillborn" cases? (e.g.: appeal fee not paid in time, grounds not filed, appeal withdrawn)

I would be curious about the distribution of the duration of proceedings, i.e., how many months/years elapse between the expiry of the Art. 108 deadline and/or the "Verk√ľndung" of the decision during an OP, and the publication of the decision. Did you look into that?

Anony said...

I think it was said that these numbers concern actual decisions, meaning that no "stillborn" cases are included. From what I see, the boards are now mainly dealing with 2010/2011 cases, so appeals filed two to three years ago. Which means one to two years after the written arguments have been filed. If the parties do not enter into an endless written "conversation", which seems to happen more and more often, from what I see. And depending on the board, obviously.
There may be cases that get priority, e.g. if a patent was revoked or refused for a formality, e.g. Art.123(2). Then it is likely to be sent back for further proceedings or finally refused/revoked. Or if OP are canceled, the boards will try to fill the gap with another case, which may be a recent one.
Also cases may be very much delayed by actions of the parties, such as accusations of bias, not being impartial, requesting a change of composition of the board. Or by requesting postponement of the OP. Or by consuming valuable board time in some way for rather useless things. Considering that in some boards summons for OP are sent half a year in advance, postponing the OP may mean half a year delay.
So many, many factors influence the time it takes from the filing of the appeal to the issue of the written decision. There is an average, I suppose, but that may well not apply to the case at hand.
BTW, fairfly, all the cases are public, with all the details given. So why not look into that kind of thing yourself, when it interests you? Then you may restrict yurself to the field of your particular interest, obtaining more relevant information.

the dude said...

Thanks for the work done Oliver!which obviously teaches that except for some information it's hard to conclude.
that's why i am axiously waiting for the results of the complementary study that "grumpy"(comment 1) will soon conduct as recommended in comment 2.

oliver said...

@ Roufousse

Sorry, I was in a place without Internet this weekend (a very commendable practice). My data were collected by myself on the DG3 website. There are some stillborn cases; the Boards do publish decisions on cases where no statement of appeal was filed in time, and at least once I have seen a decision where the appeal was deemed not to have been filed (I guess it was a decision refusing restitutio). I doubt that the Boards issue decisions when the appeal is withdrawn.

So far I have not looked at overall durations, but I agree that the average duration is of interest. I will see what I can squeeze out of my set of data.

I do not completely agree with the dude that 'except for some information it is hard to conclude'. I think that the above data contain some very interesting information. They are also highly consistent with what I had found in an earlier survey, performed on a much smaller set of data. I have not always drawn all possible conclusions because I think my readers should do some of the thinking themselves. He who has ears to hear, let him hear, as my boss once said.