ACIP Says ‘Abolish Innovation Patents’ Based on IP Australia Report

On 25 May 2015, IP Australia published a report entitled The Economic Impact of Innovation Patents.  The innovation patent is Australia’s second-tier patent right, characterised by a shorter term (eight years), a lower threshold of innovation (‘innovative step’ rather than ‘inventive step’), and lower cost (largely because substantive examination of an innovation patent is optional, unless enforcement is required) than a regular ‘standard’ patent.

The new report is the result of analysis conducted under the auspices of IP Australia’s Office of the Chief Economist, as part of its ongoing studies of the Economics of IP, and concludes that ‘the innovation patent is not fulfilling its policy goal of providing an incentive for Australian SMEs to innovate’.

And while the IP Australia report does not quite go so far as to explicitly recommend abolition of the innovation patent system (merely pointing out that some other countries have abolished their second tier patent rights on the basis of similar considerations), the Australian Government’s Advisory Council on Intellectual Property (ACIP) has been very quick to come out against the innovation patent in response, stating that it has failed to achieve its policy objectives, and that the Government should consider its abolition.

There is, in fact, a great deal of interesting and thought-provoking analysis and information in the IP Australia report.  Whether the facts actually support all of the inferences and conclusions set out in the report is another matter.  It is an even greater leap to conclude that the innovation patent system should be abolished.

In my opinion, the facts laid out in the report need to be considered from the perspective of a range of different stakeholders in Australia’s patent system.  ACIP’s new recommendation appears to have been reached on purely economic grounds, taking the many inferences and conclusory statements in the report at face value.  I believe that the facts support more than one possible set of conclusions, and that abolishing the innovation patent system rather than considering some sensible reforms could well be a case of throwing out the baby with the bathwater.

ACIP’s About-Face

Last June ACIP released its own report, following a lengthy review (over more than three years) of the Australian innovation patent system.  Despite the long duration of the review, and receipt of a range of inputs – including its own commissioned research by Verve Economics on the ‘Economic Value of the Innovation Patent’ – ACIP’s report was largely inconclusive.  In particular, ACIP made no recommendation supporting either the retention or abolition of the innovation patent system. 

However, immediately following the release of IP Australia’s new report, ACIP issued a one-page statement, suddenly taking a firm and conclusive stand on the innovation patent system (the copy of the statement available from the ACIP web site is in MS Word format only, however I have made a PDF version available here [229 kB]).  ACIP now states (the bolding is mine):

ACIP has given [the findings in the IP Australia report] careful consideration. In light of the information made available by the IPGOD [Intellectual Property Government Open Data] dataset and the analysis presented in this research paper, ACIP is now able to make an assessment of the innovation patent system’s effectiveness in stimulating innovation among SMEs. ACIP considers it likely that the innovation patent is not achieving this objective and the Government should therefore consider abolishing the system.

Kneel Before the IPGOD!

No, the IPGOD is not some sort of deity of intellectual property.  It is an acronym for Intellectual Property Government Open Data, and refers to a dataset that has been compiled and made available to the public, free of charge.  In IP Australia’s own words:

The IPGOD includes over 100 years of data on all IP rights administered by IP Australia, comprising patents, trade marks, designs and plant breeder’s rights. The data is highly detailed, including information on each aspect of the application process from application through to granting of IP rights. The IP data has been further augmented by a unique set of identifiers which allow IP rights to be linked to individual firms and firm-level business information.

It is the last point that makes the IPGOD so special – for the first time, the IPGOD makes it possible to link the IP data to other government data sources, such as the Australian Business Register, the Tax Office and data held by the Bureau of Statistics.  This has enabled the authors of The Economic Impact of Innovation Patents to obtain new insights into relationships and correlations between acquisition of IP rights and business outcomes, such as company survival rates and claims for benefits under R&D tax incentive programs.

Just the Facts

As I have already indicated, there is a great deal of interesting analysis in the report.  Some of the factual outcomes of the analysis are listed below.

1. ‘… firms that apply for patents spend more on R&D than firms that file no patent applications. These results are statistically significant at the 5% level for standard patent applicants in the mining, manufacturing, professional, scientific & technical services, education & training, healthcare & social assistance and arts & recreation services.’ (Page 10)
2. ‘…firms filing innovation patents in the manufacturing industry claimed to spend an average of $2.584m more on R&D than firms that filed no innovation patents.’  However, no evidence is found of a specific correlation between R&D expenditure and filing of innovation patents in other sectors. (Page 11)
3. There is a positive correlation between survival of companies and their filing of innovation patents.  There is, however, an even greater positive correlation between survival and the filing of standard patent applications, which increases even further for standard patents that have actually been granted. (Page 13)
4. The total volume of innovation patent filings is low.  Since its introduction, the proportion of innovation patents filed by Australian applicants has fallen (from 88% in 2001 to 68% in 2013 – page 9), although this remains much higher than the 10% of standard patent applications that are filed by Australian companies and individuals.  Interestingly, however, the number of innovation patents filed each year by Australian SMEs (as opposed to individual Australian inventors) appears to have grown steadily since the innovation patent was introduced. (Page 15)
5. The great majority (over 80%) of innovation patents are never examined and certified, and therefore do not result in enforceable rights. (Page 17)
6. A significant majority (74%) of SMEs and individual inventors who have applied for an innovation patent since the system was created have never (or at least not yet) filed any further applications. (Pages 19-20)
7. On the other hand, 15% of all SMEs and individual inventors (1385 applicants in total) who used the patent system during the study period started out with an innovation patent filing and went on to file one or more further applications for either innovation or standard patents. (Page 20)
8. Most innovation patents are not maintained for their full eight year term, and many expire due to non-payment of the first annuity, which is due only two years after filing. (Page 22)
9. Innovation patents are more likely to be maintained for longer if one or more of the following applies: the patent is owned by a large company; the patent is certified; the applicant used an agent (i.e. a patent attorney) to prepare and file the application; the patent was filed as a divisional of a standard patent application; the applicant is more experienced with the patent system. (Pages 23-24)

Dubious Assumptions, Conclusions and Inferences?

The statistical analysis described in the report appears generally sound.  However, as is so often the case (hence ‘lies, damned lies and statistics’) it is selective, and is used to draw inferences and conclusions that may be less defensible.

Take, for example, the positive correlation between innovation patent filings and R&D expenditure in the manufacturing sector.  The report repeatedly uses the word ‘only’ in discussing this: ‘…only the manufacturing sector…’; ‘…in only one industry…’.  This is damning with faint praise.  Why is this ‘only’?  I might just as readily conclude that firms conducting higher levels of R&D (around $2.5 million more expenditure each per annum) in the manufacturing sector are finding the innovation patent system to be meaningfully supportive of that effort.  Why would I then care that other sectors are seeing less benefit?  Is that not their choice to make?

As another example, on page 16 the report states that (my bolding):

SMEs and private inventors file more innovation patents, and more second-tier patents applications were filed after the innovation patent was introduced, but any net-positive effect of the innovation patent system is most likely to revolve around the enforceable rights.

It then goes on to conclude that the low level of certification of innovation patents is a negative factor, particularly due to the uncertainty around the validity of the rights that is imposed upon third parties (i.e. competitors).  Yet the report also notes that non-certification is correlated with early expiry, i.e. the period of uncertainty tends to be limited.

I see no justification for the assumption that enforceability is a requisite condition for any net positive effect, or for the assumption that uncertified innovation patents are imposing a significant cost on third parties.

An alternative interpretation (equally without solid justification, other than my own experience, and anecdotal evidence) could run along the following lines.  Perhaps SMEs and individual inventors file innovation patents at an early stage in commercialisation, so as to keep open the option of protection should the idea pan out.  In such cases, it is more likely that the applicant will try to minimise costs, so perhaps they decide not to engage an agent (i.e. a patent attorney), encouraged by the general belief that the innovation patent is a cheap and easy way into the patent system.  But cash-strapped applicants are also the least likely to invest the necessary resources for successful commercialisation.

Such applications are basically ‘noise’ in the system.  They are of little value, and they are likely to go nowhere.  They are the flip side of the data showing that innovation patents are most likely to be maintained if they are filed by larger (i.e. better resourced) applicants, who use agents, and who subsequently request examination and certification of those patents.

Which brings me to what is, in my view, the most problematic aspect of the report – the main economic analysis.

Net Economic Impact?

The economic analysis spans pages 24 to 31 of the report.  For present purposes I am going to grossly simplify, however I would encourage interested readers to go to the source.

Essentially, the report concludes that the private benefits of the innovation patent system are difficult to quantify precisely, due to a lack of suitable data, but are probably in the low tens of millions of dollars each year.  These benefits must, however, be offset by the costs to third parties – i.e. the additional costs to consumers, and the opportunity costs to competitors that are restricted from acting due to the existence of IP rights.

The report further concludes that the regulatory costs of the system lie somewhere between $5.8 million and $17.4 million per annum.

On balance, therefore, it is difficult to determine whether the innovation patent system provides a net benefit in aggregate terms, and the report is equivocal on this point.

Most damningly, however, the report points out that while the greater part of the private benefit accrues to a minority of relatively larger entities, the majority of the regulatory cost is born by smaller companies and individual inventors who derive little or no private benefit. 

It appears to be this conclusion that has significantly influenced ACIP to come out with its statement recommending that abolition of the innovation patent system should be considered.

An Alternative Interpretation

There is, it seems to me, a flaw in the economic impact analysis.  The report assumes that the regulatory cost of the innovation patent system is evenly distributed across all applications.  It is stated, for example, on page 29 that ‘since large firms file
5.55% of all domestic innovation patent applications they absorb 5.55% of all administrative costs tied to domestic applicants.’

By the same token, however – as noted above – around 15% of all SMEs and individual applicants have seen sufficient value out of their use of the innovation patent system that they have gone on to file further innovation and/or standard patent applications.

In all, therefore, some 20% of all innovation patents may be obtained by small and large domestic applicants who derive some value from their acquisition.

The problem is therefore not that innovation patents provide insufficient value to many domestic applicants.  The problem is that far too many innovation patents are obtained by small companies and individual inventors who derive no value from them whatsoever.  Many of these applicants are preparing and filing applications without the benefit of professional advice or assistance.  Others are perhaps being advised inappropriately that they would obtain some value from an innovation patent.  Still others may have legitimate reasons for filing, but have subsequently determined that the commercial opportunities for their innovations are more limited than they had hoped.

Whatever the case may be, it appears that most of the value in the innovation patent system is extracted from the ‘top’ 20-30% of applications/patents, while the cost of the system is assumed to be uniformly distributed across all applicants.

So if, for example, the ‘bottom’ 50% of applications were never filed, the regulatory cost of the system would be halved, while the total benefit would be substantially unchanged.  That would present a very different picture of the value of the innovation patent system.

An alternative conclusion based on the analysis in the IP Australia report is therefore that the innovation patent has been too successful in attracting individuals and small companies into the patent system.  While for a minority of applicants – particularly, perhaps, in the manufacturing sector – it has achieved its objective of stimulating lower-level innovation, far too many innovation patent applications have been filed simply because the system is ‘there’.

Conclusion – Not Quite Time for Abolition?

In my view, abolishing the innovation patent system would be an extreme, and premature, action.  There has been no attempt at reform or refinement since its inception.  The analysis in IP Australia’s report suggests that the system is working as intended for some users.  Surely it would be worthwhile to look at possible reforms directed to encouraging the effective use of innovation patents, while discouraging use that results in little or no net benefit?

Or perhaps we are all now just slaves to the economists?  ACIP has certainly fallen quickly into line!  I, for one, am not quite ready to join them.

Tags: Australia, Economics, Innovation patent, Law reform, Public policy