< Back to Blog Page

A Study in Correlation and Causation - How Often Do WO Documents, with US as the Priority Country, End up Not Publishing in the US?

A Study in Correlation and Causation - How Often Do WO Documents, with US as the Priority Country, End up Not Publishing in the US?

Back in September of 2013 I wrote a post examining the frequency of multiple, non-equivalent WO documents in the same extended family. During that study I stumbled on the fact that in 2006 a little less than 10% of the WO documents, with the US as the priority country, did not have a corresponding publication, either granted patent or pre-grant application, in the United States. At the time I speculated on this occurrence by writing the following:

Of the 24,219 cases in 2006, 7.5% or 1828 of them do not have a US publication in their extended family. This is worth pausing on for a moment since it is the first time [that I was aware of where] this situation has been examined. In these circumstances the US is the priority country, potentially arising from a provisional filing, but a US publication is never produced. There are a number of reasons why this could occur, but the one that seems most likely was alluded to in the previous post where it was suggested that some organizations might have used the PCT system as a means to delay publication in the US while the commercial implications of the invention were better understood, and the return on investment of filing a non-provisional application in the US were calculated. It would be reasonably easy to calculate this percentage over a longer time period to see if this rate changes over the years. In this case, it would be important to include the A2 documents, along with the A1s to get a complete picture. Stay tuned for a potential future post on this item.

There have been a few studies recently that looked at situations were pre-grant applications do not publish in the United States. I refer to these since they seemed relevant to the situation I was looking at, even though the WO system studied is a little more complicated with several additional variables involved.

In December, Dennis Crouch over at Patently-O looked at the percentage of US patents whose applications were not published. Dennis found the following for the US documents:

We have moved to a somewhat steady state of 7%, with the bulk of these being US originated applications. That figure may further drop because the $300 publication fee has now been eliminated.

There was also a discussion on the non-publication of US patent applications started by Dominic DeMarco, which took place on the PIUG wiki, but in this case the question focused on whether certain companies were more likely to intentionally request that their applications not be published before a patent was granted. Looking at both of these cases the conclusion seems pretty clear, when taken in bulk the percentage of US patents that grant without a US pre-grant application is steady, but when you look more closely there are instances where the percentage changes fairly dramatically, and perhaps intentionally.

My original post only looked at WO documents from 2006 but in this study, as promised, I have looked at nearly 700,000 WO documents from 1980 to 2008 with the US listed as the priority country. What follows are the results with regards to how often these documents did not publish in the US even though they originated there.

In order to preform this study I needed to get my hands on a very large number of WO documents, and their extended family members. I was also interested in having multiple sources so I could confirm my results with more than one collection. I would like to thank the good people at Relecura, PatSeer and Thomson Innovation for supplying the data that was used to conduct this research.

In the graph below the blue line represents the rise in the number of WO documents published where the US is listed as the priority country. As reported previously the use of the PCT system by US applicants enjoyed a steady rise until just recently. It was speculated that the leveling off observed in the early 2000’s might be accounted for by the introduction of pre-grant publications in the US. After a few years the pace of PCT filings accelerated again.

Plotted on the same graph is the percentage of A1 and A2 WO documents that did not have a US document listed in their extended family. In the early 1980’s there were a relatively small number of PCT documents filed, presumably due to applicants becoming accustomed to using the system, so the percentages fluctuate widely, but by the mid-1980’s a steady state of approximately 15% was reached, even though the number of filings is climbing. This percentage remained fairly consistent until the early 2000’s when the percentage dropped quickly to approximately 7.5% and has remained there, more or less, ever since. The study was stopped in 2008 to ensure that enough time had passed for a US document to publish and become associated with the extended family.

For almost 15 years, from the mid-1980’s to the early 2000’s, the percentage of WO applications that originated in the US, but never resulted in a US publication was a steady 15%. Suddenly, after the advent of US pre-grant publications, the percentage dropped almost immediately. This number correlates nicely with the number that Professor Crouch discovered for the number of US grants that publish without a pre-grant application so perhaps there is some cause associated with this. Perhaps we can begin to speculate on what happened when the US decided to start publishing pre-grant applications that caused the percentage of WO documents with the US listed as the priority country, but no US publications to drop as it did and begin to mirror what was happening with US grants and their pattern for the publication of US pre-grant applications.

Unfortunately, this data looks almost too clean, especially considering the complex nature of this question, and the number of variables involved. While a trend has certainly been established we may not have enough data available to be able to identify a simple cause for this effect. Clearly though, the introduction of US pre-grant applications changed the dynamics involved in this system.

Knowing this we can remove one variable from the system, the influence of US pre-grant applications on the publication of a US document in the WO extended family for a PCT document originating in the US. The line chart below shows the original line for all US publications that was seen in the previous illustration, and plots it against the percentage of PCT documents, originating in the US without specifically, a US grant in the extended family.

The line drawing clearly shows that if I had started by plotting percentage of US grants only we would see a trend, but it would tell a very different story about the percentage of WO documents, originating in the US that contained a US publication in their extended family. Keeping in mind that before the advent of US pre-grant applications the only time that a US document published was when it granted. This upward trend, over time seems to make a lot more sense, since it is known that US patent pendency times, and allowance rates have fluctuated significantly in the last ten years. It would have been somewhat surprising to see a continued steady-state for US grants in these extended families under the circumstances.

Plotted below is a line drawing of the number of non-US grants appearing as US publications in the extended families of these PCT documents:

The advent, and growth in the number of US pre-grant applications showing up in the extended families is increasing very quickly, and is having a bigger and bigger influence on whether a US publication will show up in the PCT extended family even while the trend in the percentage of US grants is shrinking as time goes by. Since it only takes one US publication, regardless of whether it is a grant, or pre-grant application to cause the PCT document to register as having a US publication in the extended family the rise in US pre-grant applications is certainly playing a role in the overall trend.

It will be interesting to see if the overall publication trend of US documents in the extended families of PCT applications will continue even though these are a number of variables involved in how this plays out. It will also be interesting to see if this trend continues to align itself with the percentage of US grants without US pre-grant applications. There appears to be a correlation here, but the prevalence of US grants publishing without a US publication can’t be directly causing this trend since one of the requirements for the non-publication of US applications is that no international application is filed. The 7% of granted patents that issue without a pre-grant publication will not be part of the PCT collection since international filing is prohibited. It is easy to think that since the numbers are similar there might be a close relationship between what is occurring in the two systems, but this can not be the case.

When dealing with complex systems, especially when working with patent data it is critical to understand the variables involved and how they influence one another when conducting patent analytics studies. In this particular example a trend in US publications appearing in the extended families of PCT applications originating in the US is clearly seen, but understanding what may have caused this effect will take a little more digging and understanding the variables involved in the system, especially as time progresses.

To receive our newsletter on machine learning and artificial intelligence in the field of Intellectual Property from our sister website, ML4Patents.com enter your email below:

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
patinformatics logo
linkedin logotwitter logo

©Copyright 2021 Patinformatics