Blog Post

Patent Office staff engaged in fraud and rushed exams, report says

Patent Office examiners, whose job is to decide when to award a patent, regularly billed for hours they did not work and also rushed their reviews in order to make quarterly performance targets.

Those are some of the conclusions set out in a damning internal report shared with the Washington Post. The findings come as another black eye for an agency already under fire for allowing too many low-quality patents.

Many of the report’s criticisms turn on the Patent Office’s award-winning work-from-home program, which was designed to make it easier for the office to recruit and keep examiners. As the Post reports, employees often padded their hours:

Some of the 8,300 patent examiners, about half of whom work from home full time, repeatedly lied about the hours they were putting in, and many were receiving bonuses for work they didn’t do […]

[The report] cited several detailed cases of time and attendance abuse. In one, an examiner missed 304 hours of work in a year but was paid for the time. Despite warnings, this examiner kept cheating and was caught twice but not fired. Another examiner claimed to have worked 266 hours for which there was no evidence she was on the job, and she received $12,533 in pay.

One version of the report also flags a culture of “end-loading” in which examiners “can go from unacceptable performance to award levels in one bi-week by doing 500% to more than 1000% of their production goal.”

From a policy perspective, this report of “end-loading” is especially serious because the practice is likely responsible for the issuance of sub-par patents that can then be exploited by patent trolls and others.

The Patent Office is struggling with a long-term challenge of how to attract examiners, who typically have specialized technical or scientific knowledge, and who can often command a higher salary in the private sector. Currently, the Patent Office employs around 8,000 examiners around the country.

As the Post reports, the version of the report that was sent to an inspector general at the Commerce Department was watered down, and didn’t contain many of the more serious allegations. (The Post article describes both versions of the report).

13 Responses to “Patent Office staff engaged in fraud and rushed exams, report says”

  1. Corrupt patent system always denies my stuff. That it requires $15000 in fees and lawyer costs to actually get a single patent or trademark approved is a HUGE motivator for somebody like me to just shelf all of my innovative and creative ideas, and go produce something not worth stealing; dog toys for example.

  2. patentlyo

    1) The US has the best patent system in the world, google it.
    2) the patent office, unlike a lot of other government work, is a pay for performance organization, it means that employees get paid for what they do, not how for long it takes them to do it. If you want employees to get paid for how long it takes them to process a patent application, you’ll end up with less able people.
    3) The report does not even attempt quantify the amount of work with no compensation that employees put in to have satisfactory performance. I bet that amount work freely given to uncle sam easily dwarfs 10 to 1 the number of hours that a few talented individuals are not giving uncle sam.

  3. >>Patent Office examiners, whose job is to decide when to award a patent, regularly billed for hours they did not work and also rushed their reviews in order to make quarterly performance targets.

    You need to add to this sentence “…according to the report”; otherwise, this sentence lacks merit/source.

    • quiviran

      They seem to be getting about $100K per year. How much more would you suggest? This is a work from home job, which means it’s more like a $125K job with commuting, dressing and actually showing up.

      The problem is one of perverse incentives. The objective of the agency should not be to approve great numbers of patent applications, it should be to use the applications to determine if the proposed invention qualifies for patent protection. A rejected application has just as much value to the public (in whose name these public servants work) as a granted application, depending on their relative merits. They get paid for approved applications and are essentially unsupervised, meaning they are accessible to outside influences as they go about the publics work. Not a strong organizational design, IMHO. Perhaps there needs to be a review process which requires the examiners to present an oral defense of their decisions to a panel of their peers. At least they’d be forced out of their jambes and slippers once in a while.

      What is disheartening is the complete failure of organization after organization within the Government, starting with the ones we have sent there via the ballot box. Maybe we need a flush handle the we can, as a nation of citizens, just pull. I’m talking a just vote No handle, and everybody is out. It would be messy, but this situation is just more BS.

  4. jonsmirl

    Limit the PTO to granting 5,000 patents a month instead of the current 25,000/mth. That’s the historical level before things started going crazy in the 1990’s. Make them figure out the 5,000 best from the pile they have available.

    The number of patents issued is growing at 10% a year compounded. We are not increasing the number of real inventions at 10% a year compounded. What is growing at 10% compounded is the patent litigation industry and government employees in the patent office.

    If the pattern continues ten years from now they will be issuing 65,000 a month. Of course there will be millions of ‘infringements’ by that time since everything on the planet will have patents on it.

    • Dilip Andrade

      Is that 5000 number supposed to take the place of their novelty and obviousness searches? If you’re number 5001 do you get held over to the next month? Why assume that all months have an equal number of “good” ideas?

      We already have problems with them doing patentability evaluations, now you want to get them to somehow figure out which are “the best”? Is a new chemical better than a redesigned piston? How about a new photonic switch?

      How on earth does this solve anything?