What Every Patent Practitioner and Applicant Needs to Know About Divisional Applications in Australia

Since commencement of the Raising the Bar patent reforms in 2013 there has been a potentially fatal trap in the Australian rules relating to divisional applications.  I have always thought it inevitable that someone would eventually fall into this trap – and practically certain that when it did happen, the applicant would be from the United States.  A recent decision of the Australian Patent Office, in which a US-based applicant has been denied the opportunity to convert an application into a divisional of an earlier filing, in order to avoid having one of its own previous applications cited as invalidating prior art, has confirmed my prediction: Magnum Magnetics Corporation [2019] APO 3.  Of course, this may not be the first time this situation has arisen.  To the best of my knowledge, however, it is the first time it has resulted in an actual Patent Office decision highlighting the issue.

In this recent case, Magnum Magnetics Corporation (‘Magnum’) filed an independent patent application in Australia when (as it subsequently turned out) the application should really have been filed as a divisional of an earlier application.  So far, its efforts to correct this error have been unsuccessful.

As most readers will be aware, a divisional patent application is a type of patent application which is based on a previously filed application, commonly called the ‘parent’ application.  A divisional application inherits the parent’s filing date and, to the extent that it discloses and claims subject matter that was also present in the parent, those claims are entitled to the same priority date.  The primary purpose for which divisional applications were originally created was to enable further inventions that may have been disclosed in an initial application to be protected, since as a general principle a single patent may only claim a single invention.  However, over the years many other practical and strategic uses of divisional applications have been developed.  For example, it is permissible to update or add new subject matter in a divisional application, although any claims based on the added matter will not usually be entitled to the benefit of the parent’s priority date.

Historically, it has been very easy in Australia to convert between ‘regular’ and divisional applications, requiring only a straightforward amendment to the patent request.  Such an amendment could be made at any time during the lifetime of the application/patent.  However, the Raising the Bar reforms changed the rules, adding additional restrictions on when an application may legitimately be converted to a divisional.  These restrictions will rarely be an issue for applicants that are making appropriate and well-informed use of the Australian patent application system.  However, a failure to appreciate the limitations of the Australian system – and, to be honest, this is something I have seen on a number of occasions from US practitioners, since their system is quite different – can get an applicant into trouble.

Unfortunately, this appears to be what happened to Magnum, although I think that there may still be some hope of saving its application.

The Rules Applying to Australian Divisional Applications

There are two basic rules that a valid Australian divisional application must obey:

1. it must be filed no later than three months after publication of acceptance of the parent application or, if the parent is never accepted, before the parent application lapses (I will call this the divisional filing deadline from now on); and
2. it must include at least one claim that is supported by the disclosure in the parent application.

The first rule ensures that an applicant cannot wait an indefinite or unreasonable time before deciding to file a divisional application.  This rule was tightened up by the Raising the Bar reforms – previously, it was possible to file a more restricted form of divisional application after what is now the divisional filing deadline, but before a patent was actually granted.

The second rule is pretty much the definition of a divisional application – if nothing is being claimed on the basis of the parent application, then what you really have is a new, independent, patent applications, not a divisional!

But the Raising the Bar reforms went further than tightening the divisional filing deadline.  They also closed what IP Australia perceived to be a remaining ‘loophole’ in the rules, relating to the ability of an applicant to ‘convert’ between different types of application.  In particular, prior to the reforms it was possible to convert a ‘regular’ patent application to a divisional application at any time – even many years after grant (as was confirmed by the Federal Court in Weatherford Australia Pty Ltd v Screenex Pty Ltd [2009] FCA 331).  Since the reforms, however, there have been two restrictions placed on such conversions:

1. no patent may be converted after it has been granted; and
2. no application may be converted after the relevant divisional filing deadline for the proposed parent application has passed.

This second rule means that even if the later application was filed while the proposed parent was still pending – i.e. if it could have been validly-filed as a divisional application on its filing date – it nonetheless cannot be converted if, in the meantime, the proposed parent has lapsed, or been accepted, and the divisional filing deadline has passed.

It is this last rule that caught out the applicant, Magnum, in its recent case.  Having missed the opportunity to file its application as a divisional in the first place, and then having missed the opportunity to convert it, Magnum required a 22 month extension to the conversion deadline.  In order to obtain this extension, Magnum would need to establish that it had previously failed to claim divisional status (at filing, or by conversion) as a result of some excusable error or omission that occurred prior to the deadline passing.  In the circumstances, it was unable to establish this to the satisfaction of the Patent Office Hearing Officer, and its extension request was denied.

A (US) Recipe for Divisional Disaster!

While there are no doubt many circumstances in which an applicant could be caught out by the restrictions relating to divisional applications, there is one particular pattern that I have seen on a few occasions and which essentially seems to have trapped Magnum in this case.  The strategy, which seems to be popular with some US patent practitioners, goes something like this:

1. a first US application is filed – I shall call this application US-1;
2. within 12 months of the first filing, an international application is filed under the Patent Cooperation Treaty (PCT) claiming priority from US-1 – I shall call this application PCT-1;
3. subsequently, a second US application is filed, claiming some relationship to US-1 – I shall call this application US-2, and note that it is typically a ‘continuation-in-part’ (CIP) of US-1, which incorporates at least some of the contents of US-1 along with some updates and/or additional subject matter;
4. within 12 months of this further US filing, a second PCT application is filed claiming priority from US-2 – I shall call this application PCT-2; and
5. within the relevant time periods, both PCT-1 and PCT-2 enter the national phase in Australia, resulting in two corresponding (but, at this point, independent) Australian applications – which I shall call AU-1 and AU-2.

Australian patent practitioners (and, I hope, most other practitioners from all around the world) should immediately identify two problems in this scenario.  Firstly, the entitlement of PCT-2 and AU-2 to claim priority from US-2 is doubtful, since Article 4 of the Paris Convention permits an international priority claim only from the very first application filed in respect of an invention.  To the extent that PCT-2 and AU-2 incorporate and claim subject matter that was initially disclosed in US-1, the priority claim is therefore invalid.  Secondly, there is not, in any case, anything that legally links either PCT-2 or AU-2 back to PCT-1, AU-1, and/or US-1.  As far as the Australian filing is concerned, there is no priority link back to US-1.

So, the ultimate problem is that AU-2 has no valid priority linkage back to US-1.  As a result, depending on the exact timeline, one or more of US-1, PCT-1, and AU-1 may be prior art to AU-2.  Accordingly, insofar as AU-2 claims subject matter that was originally disclosed in US-1, or that is insufficiently different to establish an inventive step over US-1, those claims will be found invalid.  In a simplified form, this appears to be essentially what happened to the unfortunate applicant Magnum in the case at hand.

Why Does Nobody Notice This Problem?

You might think that these problems would be picked up early, for example upon filing of AU-1 or AU-2.  Often, however, they are not.  If the Australian patent attorney responsible for the two national filings was aware of the full history of the applications set out above then they might well spot the issue, and find some way to fix it while there is still plenty of time to do so.  But usually this is not the case.  All the Australian attorney typically receives is instructions from a US counterpart to file an Australian application based upon an existing PCT application, without any other context.  Sometimes, AU-1 and AU-2 might be filed more than a year apart, and may even be filed by different Australian attorney firms.  So if the instructing US attorney is not aware that there is a problem, it is very unlikely that the Australian attorney is going to pick it up at this stage.

Of course, you might be wondering why the US attorney in this scenario is unaware of the problem.  I can only speculate about this, and do not wish to cast aspersions on my overseas professional colleagues.  It is my understanding, however, that the issue does not arise in the same way within the confines of the US system.  The Paris Convention sets the rules only between nations, it does not impose the same restrictions on the internal operation of national patent systems.  And the US system has a much more flexible notion of ‘priority’, through application types such as ‘continuations’ and ‘continuations-in-part’ that do not have exact equivalents in most other jurisdictions.

The (Sometimes) Fix – Conversion to Divisional

In many cases, the problem described above has a relatively easy fix – simply make AU-2 a divisional of AU-1.  In this way, AU-2 obtains the benefit of the filing and priority dates of AU-1 to the extent that it discloses subject matter that can be traced all the way back to US-1.  Such subject matter can then be claimed without any of the earlier applications being available as prior art.

At least, the problem used to have a relatively easy fix.  However, since the Raising the Bar reforms placed limitations on the conversion of applications, there will be many cases – such as that of Magnum – in which the relevant deadlines may have passed by the time the problem is discovered.  If so, then unless an extension of time can be obtained, conversion to a divisional application is no longer possible.  Although the sequence of applications in Magnum’s case is more complex than my example, the result is the same.  Having failed to secure the required extension, Magnum is left with an application with claims that are invalid over its own earlier applications in the sequence.

Conclusion – Solutions in the Post-Reform World

Prevention is better than cure.  In an ideal world, foreign attorneys would all be aware of the various limitations on claiming the benefit of earlier-filed applications in Australia (and, frankly, pretty much everywhere in the world outside of the US), and would not engage in international filing strategies such as the example I have outlined above.  Just to be clear, the PCT system cannot be used directly to achieve the effect of a divisional application, or of a US continuation or CIP application.  This must be done at the national (or regional) level, and requires a full awareness of the deadlines that apply in each jurisdiction.  Also, in an ideal world, Australian attorneys would be fully informed of the history and context of each application when being instructed to file in Australia.

However, since we do not live in an ideal world, it is useful to consider what can be done to fix problems, such as that faced by Magnum, when they arise.

Firstly, with additional diligence, and perhaps a little luck, the problem in Magnum’s case might have been picked up earlier.  There was an extended period, between 22 March 2013 and 4 May 2016, when applications equivalent to ‘AU-1’ and ‘AU-2’ in my example were co-pending in Australia, and the ‘AU-2’ application could have been validly converted to a divisional.  The equivalent of ‘US-1’ had been cited in the international search report, and in multiple Australian examination reports, and so the applicant and their US attorneys – who would have been familiar with the relationship between the applications – had ample opportunity to identify the problem and seek appropriate advice from their Australian attorneys.  Again, however, a lack of familiarity with the differences in the rules between different countries is likely to have been a factor here.

Secondly, although Magnum was unsuccessful on this occasion, I believe that in many cases an extension of time to convert an application to a divisional would be available.  An issue in Magnum’s case was that they had changed their US attorneys (one wonders why…) and Magnum lacked any evidence from their former attorneys about what they had intended to achieve through the international filing strategy that they adopted.  Without this evidence, it was impossible for the Australian Hearing Officer to identify a relevant error or omission that had occurred in order to trigger the extension of time provisions in section 223 of the Australian Patents Act 1990.  Sometimes, an agent will need to fall on their sword and admit that they did not properly understand the rules, and they made a mistake that resulted in everyone’s genuine intentions being thwarted.  The Australian extension provisions are generous enough to forgive such ‘mistaken beliefs’.

Thirdly, there is another type of patent application available in Australia that might sometimes be able to save the day.  A patent of addition can be used to protect an invention that is novel, but does not involve an inventive step, over an invention disclosed in an earlier application by the same applicant, and that would therefore be unpatentable via a normal application.  Historically, this has been a somewhat esoteric circumstance, and patents of addition are rarely used in Australia.  For example, in the decade from 2008-2017, nearly 269,000 standard patent applications were filed in Australia, of which just under 38,000 (14%) were divisional applications, but only 214 (0.08%) were for patents of addition!  However, there are no restrictions on converting an application for a regular patent to an application for a patent of addition prior to grant.  So in some cases – where the applicant does not want to claim exactly the same invention that was disclosed in the earlier application – conversion to a patent of addition may be an available option where conversion to a divisional is not.

There is, therefore, some hope still alive that Magnum may be able to save its Australian application.  I, for one, will be keeping an eye on it to see what happens next.

Tags: Australia, Divisional applications, Extensions of time, International treaties, Paris Convention, PCT, Priority claim