Improved opposition statistics in the EPO annual reports

Cees Mulder (NL)

Since the entry into force of the EPC 2000 in December 2007, approximately 1,345,000 patent applications were received by the EPO (Direct European applications and international applications entering the regional phase). In the same period, the EPO granted almost 600,000 patents (see Figure 1) which is approximately 44% of the number of patent applications filed at the EPO.

Figure 1: Number of annually filed European patent applications and granted European patents

On average, about 4% of the granted European patents is opposed. Figure 2 shows the number of notices of opposition filed annually in the period 2008-2016. In this period, on average about 2,900 oppositions are started every year. Some of these oppositions are inadmissible. Others are withdrawn because, for instance, the proprietor and the opponent have come to a license agreement.

Figure 2: Number of opposed European patents

In opposition proceedings, the Opposition Divisions can take three decisions: the patent is revoked, the opposition is rejected or the patent is maintained in amended form (Article 101(2) and (3) EPC). Hence, the average number of decisions taken in opposition proceedings should be around or less than 3,000 per year. Of course, this number is influenced by the throughput time of opposition proceedings. In the period from 2008-2014, the statistical reports of the EPO showed that on average 2,170 decisions were taken in opposition proceedings per year. This statistical information can be found in the separate annual reports, but the EPO also publishes so-called five-year reviews. The Annual Report of 2015 showed a sudden increase in the number of decisions in opposition proceedings. The number of decisions taken by Opposition Divisions in 2014 was 3,157. In addition, the number of decisions for the previous years in the corresponding five-year report was also adapted retroactively. For each of the preceding years (2011-2014) around 1,000 decisions were taken in addition to the ones reported in the earlier Annual Reports. This sudden increase in output may have been due to an one-year error. However, the statistics of 2016 (and the corresponding figures in the five-year report) showed the same higher number of decisions. Figure 3 shows the number of decisions taken annually by the Opposition Divisions in proceedings in the period from 2008 to 2016. The number of decisions in opposition proceedings in the older Annual Reports (2008-2014; 'old' statistics) are depicted in blue. The number of decisions in the more recent Annual Reports (2011-2016; 'new' statistics) are given in red. As can be seen in Figure 3, there is an overlap in the statistical results. For the years 2011-2014, both the 'old' and the 'new' statistics have been reported.

Figure 3: Annual number of decisions in opposition proceedings

It is interesting to observe that the decisions in the years where both statistics are available show different trends. For instance, there is an increase in the number decisions from 2012 to 2013 in the 'old' statistics, whereas the 'new' statistics show a decrease in the number of decisions in the same period.

Surprisingly, more decisions are taken in opposition proceedings under the 'new' statistical regime as compared to the 'old' regime. In particular, the number of decisions taken in the period 2011-2016 is every year higher than the number of oppositions filed in the same year. In particular, 17,819 oppositions were filed in the period 2011-2016, whereas in the same period 20,659 decisions were taken by the Opposition Divisions, i.e. almost 14% more decisions than cases started.

A higher number of decisions may happen in a certain year if there are extra efforts in dealing with opposition cases, for instance because of a backlog.

No explanation about the retroactive change of the statistical method was given in the Annual Report where this happened first (2016 Annual Report).

When asked, EPO Customers Services in its reply letter gave no specific explanation why the number of decisions is suddenly and retroactively higher. In particular, the letter stated:

"In 2015 we changed the parameters selected for representing the outcome of our internal procedures with the aim to offer a better insight into the EPO performance as it affects users. This change accounts for the apparent discrepancies you noticed."

No information was given on what 'parameters' were altered.

After the issuance of the 2017 Annual Report, another query was started to find out the change in the opposition statistics. The EPO now responded with a good explanation of the new statistics.

The opposition statistics reported in the annual reports up to the year 2014 displayed the number of opposition decisions that had taken legal effect in the respective reporting year. This means that e.g. in the 2015 Annual Report, all opposition decisions taking legal effect in 2014, were summed up.

As around 45% of the opposition cases are appealed with the consequence that the decision of the opposition division is suspended (Article 106 EPC). This means that most of these appealed opposition cases are only finally decided and thus formally take legal effect in later years, even though they were decided upon by the opposition division in previous reporting years. This manner of reporting, albeit being legally and statistically sound, does not give a full picture of the work done by the opposition divisions. In view of this, the EPO decided to change the statistics. As from reporting year 2015, the number of decisions taken by the opposition divisions of the EPO during the reporting year are presented, irrespective of the date on which they take legal effect. For consistency reasons, this new method of reporting the opposition statistics was included in the five-year overview of the statistics section on the EPO website and covered (retroactively) also the earlier years.

The above explains the systematically higher level of opposition decisions reported as from 2015 compared to the preceding years.

In addition, the increasing number of decisions taken in 2015 and 2016 are the result of the successful implementation by the EPO of its Early Certainty program. Even though the Early Certainty from Opposition scheme was only officially implemented in 2016 (Official Journal of the EPO (2016) A42, A43), initial progress in opposition proceedings was already achieved in the framework of the Early Certainty from Search program which prioritised not only search but also opposition actions.

The author would like to express his gratitude to Heli Pihlajamaa for stimulating discussions during the preparation of the article.