Government Accountability Office Study Confirms: Patent Office Encouraged Examiners To Approve Crappy Patents
This shouldn't be a surprise. All the way back in 2004, in Adam Jaffe's and Josh Lerner's excellent book about our dysfunctional patent system, Innovation and Its Discontents, one of the key problems they outlined with the system was the fact that there was strong incentives for patent examiners at the US Patent Office to approve shit patents. That's because they were rewarded for how "productive" they were in terms of how many patent applications they completed processing. Now, you might think that shouldn't encourage approvals -- except that there's no such thing as a true "final rejection" from the patent office (they have something called a final rejection, but it's not -- applicants can just make some changes and try again... forever). So rejecting a patent, inevitably, harms your productivity rates as an examiner. Approving a patent gets it off your plate and is considered "done." Rejecting it means having to spend many more hours on that same patent when the inventor comes back to get another chance.
After Jaffe and Lerner made that criticism clear, it seemed like the Patent Office started to take the issue to heart and they actually started changing some of how examiners were rated. And, for a few years, it seemed like things were heading in the right direction. But then, once David Kappos took over, he noticed that a lot of patent holders were complaining that it took too long to get patents approved. Apparently ignoring all of the evidence that pushing examiners to review patents quickly ends up in disaster, Kappos put back in place an incentive structure to encourage examiners to approve more patents. He kept focusing on the need to get through the backlog and speed up the application process, rather than recognizing what a disaster it would be. Of course, some of us predicted it and were mocked in the comments by patent lawyers who insisted we were crazy to suggest that the USPTO would lower its standards.
Of course, an academic study a few years ago found that was absolutely happening and now, to make the point even clearer, the Government Accountability Office, which tends to do really fantastic work, has written a report that agrees. It blames the Patent Office's focus on rapidly approving patents for the flood of low quality patents and the resulting patent trolling epidemic:
Permalink | Comments | Email This Story
After Jaffe and Lerner made that criticism clear, it seemed like the Patent Office started to take the issue to heart and they actually started changing some of how examiners were rated. And, for a few years, it seemed like things were heading in the right direction. But then, once David Kappos took over, he noticed that a lot of patent holders were complaining that it took too long to get patents approved. Apparently ignoring all of the evidence that pushing examiners to review patents quickly ends up in disaster, Kappos put back in place an incentive structure to encourage examiners to approve more patents. He kept focusing on the need to get through the backlog and speed up the application process, rather than recognizing what a disaster it would be. Of course, some of us predicted it and were mocked in the comments by patent lawyers who insisted we were crazy to suggest that the USPTO would lower its standards.
Of course, an academic study a few years ago found that was absolutely happening and now, to make the point even clearer, the Government Accountability Office, which tends to do really fantastic work, has written a report that agrees. It blames the Patent Office's focus on rapidly approving patents for the flood of low quality patents and the resulting patent trolling epidemic:
Examiners are rated largely on their production, auditors said, and they are given different times to complete reviews based on the experience of the examiner and the technical level of the field. For example, someone working on an artificial intelligence review gets an average of about 31 hours to complete it, while an application for exercise devices takes an average of 17 hours.The GAO says that even in the last few years (after Kappos left), when the USPTO said it was focusing more on quality, there's little evidence that's actually happening:
Timeliness and production produce bonuses: From fiscal 2009 to fiscal 2013, examiners who met these goals got an extra $6,000 a year, GAO said, citing the inspector general's office.
But they are not rated for quality work. Examiners work fast, creating "an environment where patents may be granted that do not fully meet patentability standards," GAO found.
But GAO noted that patent officials, to improve quality, still are focusing too much on the timeliness of reviews, customer service and "process or production goals" rather than quality. And they have not tied bonuses and performance reviews to quality, which needs definition. Is a good patent broad, or clearly defined? How does it prove that the invention is novel, useful, not obvious and clearly described?The amazing thing is that, in approving crappy patents, the Patent Office actually makes its own job harder, because more crappy patents will lead to more and more people applying for crappy patents, knowing they might get through. Want to crack down on the massive backlog of patent applications? STOP APPROVING BAD PATENTS. That will wake patent lawyers up to the fact that they shouldn't be submitting so many crappy patents, as it'll be a total waste of time and money.
Permalink | Comments | Email This Story