A key premise of the Australian Government’s Productivity Commission’s (“The Commission”) Draft Report is that an effective patent system requires patented inventions to be socially valuable. The report suggests that the present system fails to achieve this, as a “non-trivial number” of patented inventions do not benefit the Australian community. The report examines empirical evidence, including The Commission’s own analysis on this issue.
The report presents data on patent value to support the above contention, suggesting that a large number of Australian patents have low value. These low value patents purportedly block Australian entities from operating and/or deter follow on innovation.
The key proposal to fix this apparent issue is to again raise the standard of inventiveness necessary to achieve the grant of an Australian patent. As the vast majority of granted Australian patents are owned by foreign entities, effectively The Commission is suggesting that foreign rights should be curtailed in order to benefit local operators. Raising the inventiveness threshold would, of course, also affect patent applications made by Australian entities, further reducing the already small proportion of Australian patents granted to Australian based entities relative to foreign entities.
While we do not necessarily disagree that the test for inventive step be reviewed and discussed, the notion that raising that standard would have any significant influence on the “social value” of patents in Australia is fundamentally flawed. The issues with Australian innovation and its translation into successful commercial outcomes run much deeper than tinkering with the IP system.
After many years in a multinational corporate environment, this author believes that there is no clear correlation between patent value and inventiveness. An invention may be highly inventive yet a patent granted on it may be of little value. Market factors largely determine patent value and these may have little to do with the ingenuity of the inventor. For example, a new pharmaceutical may require a high level of inventive skill to discover, yet may be of low value if years later clinical trials uncover unacceptable side effects. Equally, a patent with low inventiveness could be of high market value, noting that the majority of innovations are based on incremental improvements to existing technologies and not on breakthrough events.
The Commission relies on a combination of proxy measures, such as forward citations, patent scope, years in force, etc., to generate a “patent value index”, however such an index can only be speculatively correlated to inventiveness. None of the proxy measures take into account commercial realities and market forces.
Further, most of the proxy measures for value are based on data concerning equivalent US patent applications (forward citations, generality index, radicalness index, non-patent literature citations and family size). Therefore, any proposed correlation drawn by The Commission between apparent value and inventiveness if anything, relates to the US test for inventive step, a test which is generally viewed as tougher than that in Australia. Even if the inventive step test in Australia was raised to the standard in the US, it would have no influence on the “patent value index” which has been relied upon by The Commission in so far as it relates to Australia.
In an analytical flaw, the “patent value index” credited is relevant for patents filed in the period 1995-2014. The “Raising the Bar” amendments to the Australian Patents Act, which came into force in 2013 modified the inventive step test in Australia for applications where examination was requested after April 2013. The number of patents that have been granted under the revised examination regime is small compared to the total number of granted patents currently in force in Australia. Further, few if any of these recently granted patents would fall within the 1995-2014 “patent value index” dataset reviewed by The Commission. The correlation between apparent patent value and inventive step has been made with patents examined using the old test for inventive step and therefore is hardly a proper basis for the conclusions drawn and the recommendations made.
Patents across every jurisdiction in the world have a range of values. Even if the inventive step test was identical in every jurisdiction and patents with identical claims were granted in every jurisdiction, the value of a particular patent would vary from jurisdiction to jurisdiction. This is simply because commercial and market factors account for the majority of patent value, not the degree of inventiveness.