The Impact of PCT Documents on the US Patent System, the Child Living Under the Stairs, or a Valued Family Member?

It sometimes seems like PCT applications, often referred to as WO documents, are a little like Rodney Dangerfield, they don’t seem to get much respect. They never grant, so that’s a knock against them, and they have special, National Stage, filing rules that can be confusing to inexperienced users. For instance, many analysis projects focus on patents from a particular country, but don’t consider PCT applications, which might designate that country. In almost all aspects the PCT applications are given the same deference as non-provisional applications generated within the country itself, and should be thought of as valuable family members. Even though this is the case I am sometimes met with a blank stare when I suggest that even if the analysis is country specific PCT applications designating that country should be included, and reconciled when performing a family reduction. To some, PCT applications are like the cousin living in the room under the stairs, they know they’re there, but they would just as soon ignore them if they could. In order to understand the impact that PCT applications have on the US patent system this post will examine filing patterns associated with them.

According to data in Thomson Innovation, as of September 10th, 2013, there are 2,906,588 WO applications in their system. These documents are differentiated by kind codes, a complete list of which, can be found here, but the first three are the most significant, and account for 97% of the total number of documents. Kind codes A1-A3 are defined as:

A1 – International application published with ISR

A2 – International application published without ISR

A3 – Later publication of ISR with revised front page

Based on the kind codes, it immediately becomes apparent that A2 and A3 documents are associated with one another, and in almost all cases when an application is published without an International Search Report (ISR), the A2, a corresponding copy of the ISR, the A3 document, will follow. When counting these documents this relationship needs to be factored in.

Looking at all PCT applications the following distribution among the top three kind codes is found:

A1 – 1,772,948 documents

A2 – 535,166 documents

A3 – 515,756 documents

This analysis suggests that ~70% of the WO documents have a kind code of A1, were published 18 months from their priority filing date, and included an ISR. Things begin to get more interesting when these numbers are studied using the lens of how many of these documents list the United States as the priority country.

A1 with US as priority country – 598,065

A2 with US as priority country – 358,229

A3 with US as priority country – 345,536

It is immediately apparent that while 70% of all PCT applications have an A1 kind code less than 34% of these list the US as the priority country. Alternatively, A2, or A3 documents with US as the priority country account for about 67% of the total number of documents of these types. Looking at documents with the US listed as priority country published over the last 30 years starts to suggest some interesting patterns with regards to the importance of WO documents to the US patenting system as seen in the line drawing below:

WO Documents with US as the Priority Country Published over the Past Thirty Years by Kind Code

WO Documents with US as the Priority Country Published over the Past Thirty Years by Kind Code

Up until 2001 the majority of the WO documents with the US listed as the priority country were published as A1 documents. This suggests that for inventions,  which an organization thought they would be filing internationally, a PCT application was initially filed. These documents could take advantage of the full 18 months from priority filing before publication, for a ISR to be generated. This trend changes dramatically staring in 2001, and culminates in 2007 when the majority of the published WO documents were listed as A2 documents instead of A1s. This could represent a change in filing strategies suggesting that US non-provisional applications were being filed first, and twelve months later a PCT application was filed. This would only leave six months from the priority filing date for an ISR to be generated, and would likely cause the increase in A2s versus A1s. After 2007, A1s have re-emerged as the kind code most frequently seen but it would be difficult to speculate whether this represents applicants reverting to pre-2001 filing behaviors, or if it can be accounted for by other variables introduced in the worldwide patenting ecosystem.

Most people working with patent information will recognize the 2001 time frame as the period when the United States began publishing pending patent applications 18 months after their priority filing date. This provision was covered in the American Inventor’s Protection Act (AIPA), which was signed into law on November 29th, 1999. The specific provision was Subtitle E – The “Domestic Publication of Foreign Filed Patent Applications Act of 1999” and details on this portion of the law are provided below from the USPTO:

This subtitle provides for publication of patent applications 18 months after filing unless the applicant requests otherwise upon filing and certifies that the invention has not and will not be the subject of an application filed in a foreign country. Provisional rights are available to patentees to obtain reasonable royalties if others make, use, sell, or import the invention during the period between publication and grant.

If the foreign-filed application is less extensive than that filed with the PTO, the applicant may submit and request publication of a redacted version by the PTO.

This title also provides a prior art effect for published patent applications; requires the GAO to conduct a three-year study of applicants who file only in the United States; and requires the Director to recover the cost of early publication by charging a separate publication fee after notice of allowance is given.

The subtitle’s provisions take effect one year after the date of enactment and apply to patent applications filed on or after that date.

In addition to providing insight on changes in filing strategies, the line drawing, as expected, also shows that trends associated with the A2, and  A3 documents mirror one another, with a predictable time lag associated with the additional time needed to produce the missing ISR, and publish the A3 document.

It is also interesting to look at the percentage of WO documents, with the US listed as the priority country, by kind code over the same 30 year period, as seen in the next line drawing:

Percentage of WO Documents with US as the Priority Country over the Past Thirty Years Along with Percentages of A1 and A2 Documents

Percentage of WO Documents with US as the Priority Country over the Past Thirty Years Along with Percentages of A1 and A2 Documents

Regardless of whether the A1 or the A2 is the primary kind code associated with WO documents, the overall percentage of documents with a US priority stays reasonably consistent over the years, until it drops beginning in 2010. The chart below looks at WO priority countries for publication years 2004, 2008 and 2012, and clearly shows that increased use of PCT applications from Japan, China and Korea accounts for the share being lost from the US as a percentage.

Top Priority Countries Represented in WO Documents for Selected Publication Years for the Past Ten Years

Top Priority Countries Represented in WO Documents for Selected Publication Years for the Past Ten Years

Looking at this using a slightly different approach, it can be seen that the use of PCT applications by US applicants follows an upward trend until 2007, when overall, the use of PCT applications drops off. In 2010 though, when the overall number of WO documents begins to increase again, the number of applications by US applicants stays at a reduced level. This trend can be seen in the line drawing below:

Total Number of PCT Applications Published per Year Compared to Those Filed by US Applicants

Total Number of PCT Applications Published per Year Compared to Those Filed by US Applicants

It is clear from this analysis that while the use of the PCT system by US applicants has flattened out recently, and as a percentage is not dominated by them as much as it used to be, WO documents still play a critical role in the filing strategy of US organizations. The transition within the US to a fast publishing country, from a slow publishing one, can be seen in this analysis, and this change in the law appears to have had a major impact on the use of PCT applications, at least initially. Without a doubt the use of PCT applications by applicants, as it applies to US filings, should be considered when doing patent analysis, and continues to be necessary when studying the inventive output from US applicants. In the next post, I will look at what impact PCT documents have on extended patent families, especially when the applicant comes from outside of the United States.

Comments 7

  1. Tony, fascinating as always, thanks for posting this. I agree with your line of reasoning about including WO documents into patent analysis targeted toward a specific country. You suggest only including WO documents that designate that specific country, which also makes sense. My side question is – don’t applicants generally designate the kitchen sink in their PCT applications? I thought that applicants could choose not to go national in those states later on, and that there was no downside to listing every authority. Although a lot of search systems offer a “designated states” search field, I always looked at it as somewhat useless, since everybody seems to designate every state. What are your thoughts?

    1. Hello Kristin,

      Thanks for the comment and question. You are absolutely right that an applicant can decide not to go national and they even have up to 30 months to make that decision. I am not certain what percentage of applicants select every authority but I agree that many applications pick a large number.

      In the US, before the initiation of the track system for examination I think some applicants may have started with a provisional in the US, filed a PCT as the first non-provisional application and then followed up in the US within the 30 month window. I am not certain whether this would save the applicant much money but it would keep the application from publishing in the US initially, which might help them stay under the radar.

      In any event, I treat WO applications as an option in the countries designated since the applicant still has an opportunity to file a National Stage application, and knowing that this option exists can be important in a number of ways.

      All of this is actually leading to some conclusions I plan to make in two future posts on this topic.


  2. Hi Tony
    What is most interesting to me about this piece (and some other recent pieces by others) is how amazing you guys in the US find these things. Out here in the rest of the world this is totally normal. We just swap between convention applications, PCT applications and national phase applications, as well as different kind codes, without a thought.
    I will always include PCT applications in freedom to operate or infringement searches as they may become a national phase application and potentially relevant. The client may also want to oppose them later.
    I too ignore the designated states field. Maybe at first applicants thought about which countries they might be interested in protecting, but now I think they just tick everything, just in case.
    Finally, the number of PCT applications dropping after 2007 is related to the global financial crisis, and is now recovering. The US not recovering so quickly is probably related to the general economic state of the country.

    1. Hello Frazer,

      Thank you for the comments.

      I have to confess though that your initial contention on amazement confused me somewhat. Perhaps you could explain which idea, presented here, was one we seemed amazed at?

      Personally, I was surprised at how steep the drop off was between the publication of A1 and A2 documents, but only because it suggests a change in strategy for applicants, which I don’t think has been discussed previously. I am not certain this observation makes any difference when doing prior art searching but it’s interesting from a corporate strategy perspective.

      You may be correct with your observation regarding global finances around 2007 but I don’t think I quite buy the same explanation for why US applicants are using the PCT system less frequently recently. Since 2010 the number of filings by US applicants in the United States has continued to rise at 5-10% a year (USPTO Statistics can be found here). So US applicants are filing more patents but not necessarily doing so internationally.

      Best regards,

  3. Hi Tony

    Thanks for the response.

    I guess my amazement comes from the recent attention some topics have had considering the length of time the US has been a party to the PCT. I started out at IPONZ and now work in Australia, and the whole kind code aspect of documents is drummed into you right from the start. It becomes second nature to know which kind codes are applicable in different situations. For instance, I am no longer surprised when an A1 document that appears relevant for infringement purposes, is not so relevant when I look at the B2 document.

    It seems as though your (the US) focus is on US publications, and foreign publications are an afterthought, whereas I think we (the rest of the world) have a whole world view.

    We are also quite happy with the concept of an A1 document having the same number as a B2 document, which is something I think you forget, or at least it leads to confusion as to what version of a document you are looking at, because you don’t need to have that additional focus on the kind code. For instance you may say “20050123456” or “7654321”, whereas I would say “US2005123456A1” or “US7654321B2” or “AU2010215456A1” or “AU2010215456B2”.

    Going back to the specifics of your post, I’m pretty sure the applicants don’t get a say as to whether an A1 or A2/A3 is published. It’s all a matter of time. If the ISR is available when the application is published after 18 months then it’s an A1, otherwise it’s an A2 with the A3 following as soon as possible.

    As such it can’t be a strategic decision by the applicant. The strategy is whether you file a PCT or not, and judging by the stats you provided, US applicants aren’t choosing PCT at the same rate as before 2007.

    You are probably more aware of the answer to this than I am, but most US based PCT applications will have the ISR conducted by the USPTO as the local ISA. Given the dramatic increase in WO A2 publications from 2000/2001, it seems that searching in the US slowed to a point where they couldn’t meet the 18 month publication deadline for an A1 publication. Other than the US deciding to publish applications from 2001, was there some fall or stall in examiner numbers, or some great policy upheaval that would explain it? As there is a righting of the trend since 2007, has the decreased number of total applications requiring an ISR allowed more to be completed.


  4. Hello Frazer,

    I’m afraid you’re not giving practitioners in the US much credit for understanding kind codes and being US centric. Most professionals I know in this country understand the differences and have no issues with the numbering systems or kind codes. Many of them also work for global organizations and search worldwide routinely.

    In the case of my posts, my readership is made up of people at all experience levels, and as such I tend to explain things that are inherently obvious to professional searchers so there is no confusion for the novices and for analysts just starting with patent analytics.

    The A1/A2 item is, of course, time dependent and the applicant has no control over this, but that was not my point. I certainly wasn’t implying that the applicants were doing anything intentional to get an A2 vs. an A1, it was just a glaring reversal around the time of a significant law change in the US.

    Having said that, you may be right that a backlog with regards to the ISA might have caused the sudden shift, but normally those sort of trends appear slowly and grow over time, as opposed to the dramatic change seen in the chart. I will have a look to see if the ISA is an exportable field in any of the major databases so this can be explored. The USPTO did start outsourcing ISA responsibilities in the late 2000’s and this could have had an impact on the increase in A1s.

    The point of all of this was discussed in the latest post that was published this weekend where I went back to the topic of family reduction methods. I felt like I needed some additional data on WO documents and their relationship to US filing patterns before I could re-address the issue of generating accurate patent statistics.

    Best regards,

Leave a Reply

Your email address will not be published. Required fields are marked *