The patent opposition system recognises that the examination process is imperfect. Examiners have limited time, resources and technical expertise. Therefore, they may not always find the closest and most relevant prior art, or spot every technical and legal issue that might be identified by a motivated competitor to the patent applicant, equipped with a team of technical and legal experts. Furthermore, opposed applications are presumably those that are of greatest concern to competitors, enabling the system to weed out invalid claims that have the greatest potential to unfairly stifle competition.
In the years prior to 2016, the number of oppositions filed each year was fairly consistently between 100 and 120. In recent years, however, it has commonly been between 40 and 60. In other words, there are now only about half the number of patent oppositions being filed than was the case just a decade ago. So, if oppositions play an important role in the Australian patent system – and the policy rationale for having them asserts that they do – is it possible that they are now less effective than they once were? And, if so, then why?
In this article, I will present data on all patent oppositions filed between 2008 and 2023. I will demonstrate that the decline in patent oppositions appears to be associated with the commencement of the Raising the Bar (‘RtB’) IP law reforms in 2013. Among other things, the RtB reforms introduced more stringent standards of patentability, particularly in relation to inventive step and the level of disclosure required to support broader patent claims. The reforms also changed the standard of proof to be applied during examination and opposition proceedings. I will show that in the post-RtB era, opposition proceedings have more frequently progressed through to a final decision, and that opponents have had somewhat greater success in completely eliminating opposed applications. However, the overwhelming majority of opposed applications still result in granted patent rights, and in nearly half of all cases the opponent has been wholly unsuccessful and the patent has been granted with the originally accepted claims.
While the data cannot directly reveal the reasons for the significant reduction in opposition filings, I tentatively argue that the change in the standard of proof applied in patent oppositions may have had the unintended consequence of suppressing patent oppositions, and reducing the effectiveness of the opposition system.
Data Source
All of the analysis presented in this article is based on IP Australia’s ‘IP Refreshed Automated Product for Information and Data (IP RAPID)’ data set. IP RAPID is updated weekly, and the data I employed was up-to-date as at 15 April 2024. It therefore reflects the state of IP Australia’s records, and the status of all patents, applications, and ongoing oppositions, as at that date.
Pre- and Post-RtB Opposition Filings
The chart below shows the number of patent oppositions filed each year, between 2008 and 2023, broken down into those application that are subject to the former (‘pre-RtB’) laws and those subject to the current (‘post-RtB’) laws. There were, in total, 1418 oppositions filed during this period, of which 932 were pre-RtB and 486 were post-RtB. Of these, 1332 have been finalised (926 pre-RtB and 406 post-RtB). There is, therefore, plenty of data available on outcomes of both pre- and post-RtB oppositions.
In the years prior to the reforms, the rate of patent oppositions was reasonably consistent, at around 100 to 120 per year. For two years following commencement of the reforms, the pre-RtB laws continued to apply to most accepted patent applications, and oppositions filed against these applications continued in similar numbers. From 2014 onwards, an increasing number of applications were accepted subject to post-RtB laws, and the number of oppositions to such applications began to rise.
It is clear, however, that once the dust had settled, the number of oppositions filed each year against post-RtB applications was only around half the number previously filed against pre-RtB applications, and that this remains the case to the present day. This naturally prompts the classic Julius Sumner Miller question – why is it so? Let’s see if a more detailed look at the data can provide any insights.
Fewer Accepted Applications?
An Australia patent application can only be opposed during a three month period following publication of a notification of acceptance. In practice, oppositions are almost invariably filed on the final day of this period, to deny the patent applicant the opportunity to file a defensive divisional application after becoming aware of the opposition. Since only accepted applications can be opposed, all else being equal we might expect the number of oppositions to be roughly proportional to the number of acceptances in a given period. Could this explain the decline in oppositions following the RtB reforms? Are fewer applications being accepted?
In short, no. The chart below shows the annual numbers of ‘available’ opposition deadlines, where the date of such a deadline is calculated by adding three months to the corresponding date of publication of acceptance of an application.
There was a period of increased acceptance rates in the years following commencement of the RtB reforms, during which IP Australia increased patent examination resources in order to clear the backlog of examination requests filed shortly prior to the commencement date. These were, it should be noted, predominantly pre-RtB cases, because the applicant behaviour in question had been prompted by a desire to have applications – and the resulting patents – subject to the more permissive former legal standards of validity. Since then, however, acceptance rates have returned to much the same level as immediately prior to the reforms. So this does not provide an explanation for the drop in opposition rates in the post-RtB period.
To emphasise the point, the chart below is a scatter plot of annual ‘available’ opposition deadlines versus the number of oppositions actually filed during the same period. The two appear to be almost completely uncorrelated (if they were proportional, the points would be clustered around a line extending from the lower left to the upper right of the chart). Whatever factors drive decisions to oppose Australian patent applications, they seem to bear little relationship to the volume of applications accepted.
What Are the Outcomes of Patent Oppositions?
There are, broadly speaking, six ways that an opposition can end.
- The opponent can with withdraw the opposition. This could be, for example, because the opponent determines that the opposition has limited prospects of success, or is otherwise no longer worth pursuing. Quite commonly, however, withdrawal of an opposition is the result of a settlement agreement reached between the patent applicant and the opponent. This may or may not include an agreement by the applicant to amend the application (i.e. to narrow its patent claims) prior to grant of the patent.
- The applicant can withdraw the application. This is essentially an active surrender by the applicant to the opponent.
- The application can lapse. Typically this will be due to the applicant failing to pay a required fee – either the acceptance fee, or an annual renewal fee, both of which remain due in the course of an ongoing opposition – and is essentially a ‘passive’ surrender, in which the applicant simply gives up and/or disappears. This (as we shall see) is rare.
- The opposition may proceed to a hearing and decision, in which the opponent is wholly successful and the application is refused.
- The opposition may proceed to a hearing and decision, in which the opponent is partially successful in that at least some claims in the application are found invalid, but the applicant is afforded an opportunity to amend the application to retain only valid claims, which then proceed to grant.
- The opposition may proceed to a hearing and decision, in which the applicant is wholly successful in defending, and the opposition is dismissed.
The chart below shows a breakdown of these six outcomes, along with those recent oppositions that remain in progress, for all oppositions filed between 2008 and 2023.
It is immediately apparent that oppositions were more commonly withdrawn in the years prior to the RtB reforms. Other than this, however, it is hard to spot any obvious trends in pre-RtB versus post-RtB opposition outcomes.
To provide additional insight into this, the following chart summarises the percentage of all completed oppositions that resulted in each of the six outcomes, broken down by pre-RtB and post-RtB cases.
Firstly, this chart bears out the initial impression that withdrawals of opposition were relatively more common – 48.3% versus 36.7% – pre-RtB as compared with post-RtB. Furthermore, along with fewer oppositions being withdrawn, opponents have more often achieved complete victory over patent applicants: post-RtB nearly a quarter (23.4%) of all oppositions have ended with the application being withdrawn, abandoned, or refused following a hearing, compared with just 15.7% pre-RtB. On the other hand, just over a quarter (26.8%) of all post-RtB oppositions have ended in dismissal – i.e. a complete failure by the opponent – compared with 22.4% of all pre-RtB oppositions.
The RtB reforms appear to have had little impact on the likelihood of an opposition resulting in partial success for the opponent, i.e. grant of a patent following some narrowing amendment of the claims. Where an opposition proceeds to a hearing and decision, there has been a negligible decline in the proportion of cases in which the claims are amended as a result. Furthermore – as shown in the chart below – where oppositions are withdrawn, there has been only a slight increase – from 38.5% to 43.6% – in the proportion of such cases in which the applicant has amended the claims prior to grant.
The following table summarises overall outcomes of opposition before and after commencement of the RtB reforms.
| Outcome | Pre-RtB | Post-RtB |
|---|---|---|
| Opponent wholly successful (application withdrawn, abandoned or refused) | 15.7% | 23.4% |
| Applicant wholly successful (opposition dismissed or withdrawn without amendment) | 52.1% | 47.5% |
| Application proceeds to grant after amendment | 32.2% | 29.1% |
Is Opposition More or Less Effective Following RtB Reforms?
The following observations may be made about the apparent effect of the RtB reforms on patent oppositions:
- considerably fewer oppositions are being filed in the post-RtB era when compared with the years immediately preceding the reforms;
- post-RtB oppositions are less likely to be withdrawn, and more likely to result in total success for the opponent, than was the case prior to the reforms;
- but post-RtB oppositions have only marginally more frequently resulted in patents being eliminated or narrowed in scope (52.5% of cases) than pre-RtB oppositions (47.9% of cases).
Combining the first and third of these points, the number of accepted patent applications that are eliminated or narrowed in scope each year, as a result of opposition proceedings, has actually fallen substantially in the post-RtB era.
This raises a question as to whether the pre-grant opposition system in Australia remains as effective following the RtB reforms if – as seems to be the case – it is substantively impacting a smaller proportion of all accepted patent applications.
Explanation of Observed Effects – Two Hypotheses
Assuming that there is a causal relationship between the RtB reforms and a reduction in use of the patent opposition process, I can identify two aspects of the reforms that might influence the observed behaviour: higher substantive patentability requirements; and a change in the standard of proof in proceedings within the Patent Office.
Substantive Patentability Requirements
The RtB reforms introduced higher standards for inventive step and disclosure. We know that the RtB reforms have had no effect on the proportion of examined patent applications that go on to be accepted (and I have today checked more recent data and confirmed that acceptance rates have continued to remain steady at around 75% of all applications that enter examination). But we know that the higher standards provide examiners with more potent tools to challenge the validity and scope of patent claims, and it is therefore possible – indeed logically likely – that at least some patent applications are being accepted with narrower claims under the post-RtB regime than would have been the case pre-RtB.
If that is the case, then competitors may have less reason to oppose accepted applications, because narrower claims may be less likely to restrict their activities. So, perhaps this could explain the reduction in the number of oppositions filed.
Furthermore, the same enhanced standards also provide opponents with more potent tools to challenge accepted patent claims. This could explain why opponents are less inclined to settle opposition proceedings with applicants, or to withdraw oppositions once commenced. It might also explain why the reduction in withdrawal rate has been accompanied by an increase in the rate at which opponents are wholly successful in eliminating apposed applications.
On the other hand, there has been very little change in the rate at which applicants are wholly successful in defending patent oppositions in Australia. It remains the case that nearly half of all opposed applications end up proceeding to grant with no further amendment. Given the considerable costs that can be associated with patent oppositions, this is not a great outcome for many opponents.
Changed Standard of Proof
Historically, for the Patent Office to refuse an application, it needed to establish that it was ‘practically certain’ that the patent, if granted, would be invalid. The logic behind applying this high standard of proof was that the consequence of erroneously refusing a valid patent (i.e. the irrevocable loss of the applicant’s rights) was objectively worse than that of erroneously granting an invalid patent (given that there are further mechanisms, post-grant, for challenging the validity of a patent in court, and before the Patent Office).
This standard had previously been relaxed somewhat, in relation to grounds of novelty and inventive step, but the RtB reforms uniformly replaced the ‘practically certain’ standard with the ordinary civil standard of the ‘balance of probabilities’. So now the Patent Office can refuse an application, in examination and in opposition, if the decision-maker determines that it is more likely than not that the granted patent would be invalid. Notably, this is the same standard that is applied by the courts in post-grant proceedings.
As with the enhanced substantive patentability requirements discussed above, this change in the standard of proof could result in a narrowing of claims in examination, and a higher success rate in opposition. But it could have had a less desirable consequence, as well.
When IP Australia was consulting on the proposals for the RtB reforms, it received a number of submissions raising concerns that ‘consideration of the same issue against the same standard of proof may raise an “issue estoppel”, thereby preventing an unsuccessful opponent from re-arguing the same issue in court revocation proceedings’ (see IP Australia’s Consultation Paper (November 2009) Toward a Stronger and More Efficient IP Rights System [PDF, 494.2kB], page 17, paragraph [75]). These concerns were dismissed, with IP Australia stating that it was ‘not convinced that this would be a significant policy problem’ (paragraph [76]).
Maybe IP Australia was wrong about that! The question of whether issue estoppel may arise as a result of patent opposition proceedings remains open, and I suspect that many patent attorneys and IP lawyers would be advising their clients of this potential risk should they proceed with an unsuccessful opposition. The consequence of such advice might be that a prospective opponent may elect to ‘keep their powder dry’, and elect to avoid the costs and risk of opposition in favour of challenging the validity of the patent only if and when it becomes an issue post-grant, e.g. if they are actually threatened with infringement proceedings.
If this is what is happening, then the change in the standard of proof might have had the adverse effect of discouraging the filing of around half of all prospective oppositions of which, on the basis of both pre- and post-RtB outcomes, about half would have resulted in refusal or amendment of the accepted claims. That would be completely counter to the policy rationale for having a pre-grant opposition system in the first place!
(I note, in passing, that in Orikan Group Pty Ltd v Vehicle Monitoring Systems Pty Limited [2023] FCA 1031, a judge of the Federal Court of Australia recently considered a complementary question, namely that of so-called Anshun estoppel. This kind of estoppel arises not when a party seeks to raise an issue in a subsequent proceeding that had already been raised and determined in an earlier proceeding, but rather when a party seeks to raise a new issue in the subsequent proceeding that could and should have been raised in the earlier proceeding. The court found that Anshun estoppel does not arise as a result of patent opposition proceedings. But this tells us nothing about what the court might have decided had issue estoppel been raised instead.)
Conclusion – Is There a Problem, and Can We Fix It?
My first hypothesis to explain the post-RtB reduction in patent oppositions is essentially a positive one. It supposes that the reforms have been effective in limiting the scope of accepted patent claims – or at least of those that are of sufficient commercial significance to warrant the scrutiny of a competitor – and that there is therefore less need for opposition proceedings.
Frankly, however, I find that very difficult to believe. The reasons why patent examiners fail to identify all grounds of invalidity has never been because the substantive legal requirements for patentability are not stringent enough. The issue is that examiners have far more limited resources at their disposal than a motivated opponent, and less access to the information and evidence available to competitors operating in the same industry as the patent applicant. Oppositions do not merely re-agitate grounds that were raised and overcome during examination. Opponents introduce previously undiscovered prior art, expert evidence, and additional grounds and arguments for invalidity. This was true prior to the RtB reforms, and it remains true today. An enhanced standard of inventive step, and a reduced burden of proof, are of no utility to an examiner if they are not, in fact, able to identify the most relevant prior art with the resources (i.e. time and information) available to them. This is in no way a criticism of patent examiners. It is how the system works, and it is why we have a pre-grant opposition system in the first place.
By the same token, there is no reason to think that the rate of acceptance of potentially invalid and overly-broad patent claims is any lower post-RtB than it was pre-RtB.
And if that is correct, then the most plausible explanation for fewer patent oppositions being filed is that many prospective opponents are choosing not to file patent oppositions. Of course, I don’t really know the reason for this, and it might be totally unrelated to the RtB reforms… but that would seem to be a pretty huge coincidence based on the data in the first chart above.
To my mind, that makes my second hypothesis the more likely of the two – although, of course, not certain. But if true, then this is unfortunate. Parties with the motivation and information necessary to weed out invalid claims prior to the grant of a patent should not be discouraged from doing so. And it seems to me that we should be able to fix this. Surely we could legislate to immunise opposition proceedings against raising issue estoppel? Or maybe someone will tell me why that would be impractical?
0 comments:
Post a Comment