Paying Maintenance Fees

A patent's twenty year term is cut short when a patent holder fails to pay regular maintenance fees. In the US, maintenance fees are due 3.5, 7.5, and 11.5 years after issuance. The USPTO provides an additional 6-month grace period before the patent is considered abandoned. The current fee schedule requires a payment of $980 for the first maintenance fee; $2480 for the second fee; and $4,110 for the third fee. Qualified "small entities" receive a 50% reduction in fees. Small entities typically include universities, non-profit organization, and businesses with fewer than 500 employees.

Although the USPTO does not provide any ongoing services to patent holders, the maintenance fee revenue is used to fund ongoing patent office expenses (such as the examination of new patents). In fact, more than one-third of the USPTO's two-billion-dollar annual revenue comes from maintenance fee payments. For manufacturers and non-patent-holders, the maintenance fees are useful because they encourage patent holders to drop their patent rights.

Using USPTO records of maintenance fee payments, I looked at how fee payments have varied over time. The graphs below show the percentage of patents abandoned four years after issuance. These are patents that issued and then the patentee failed to pay the first maintenance fee due. In order to better visualize the trends, I used a 12-month moving average in the charts — thus, the data point for January 2011 represents abandonments for February 2010 through January 2010. The second chart divides the results between small and large entities.



Two initial thoughts on these results:

  • Over the past two years, the abandonment rate has been rising. This rise is more notable for patents held by large entities. For large entities, the abandonment rate is the highest in the past decade.
  • Even though they have much lower maintenance fee charges, small entities are much more likely to abandon their patent applications.

37 thoughts on “Paying Maintenance Fees

  1. 37

    When patents were 17 years from issue, I did a quick and dirty survey of members of ACPC. I didn’t try to account for the technology b/c I thought I already knew the basic answer to four questions. How much would your company’s R&D budget go up if terms were 21; 24 years? How much would it go down if terms were 13; 9 years? There were outliers (Pharma?) but I was very surprised when most respondents said no effect. See Perceptions of Chief Patent Counsel…, 32 Idea 277, 279 (1992). I expected to be asked to follow up or that someone else might, but. to my knowledge, no one has.

    Comparing payment of maintenance fees in various technologies might be similarly useful but it would not account for those whose R&D is supported by secrecy, margins associated with good will….

  2. 36

    Dennis and Ping- thank you for your responses.

    Malcolm- you really should know who you are talking to before you write something rude.

    It is not obvious that the tab “bibliographic data” means payment history. And I should have invented a bag for my head. I have cancer and lost all of my hair last week.

  3. 35

    Lolz sunshine (at you).

    I give my condensension to those who think too highly of themselves whilst you condescend those who bravely venture forth with what you so snobbily perceive as a “dumb” question.

    And yet, there are those who will never question or comment on your choices.

    In the words of one Bartholomew JoJo Simpson:

    Eat my shorts.

  4. 32

    Where do you find on the website whether a patent has had its maintenance fees paid or if it has been abandoned?
    thank you, independent inventor nina

  5. 31

    Lionell, perhaps they have, in the past. But, given BRI and the extreme manner it is being employed by the examining corps today, I would suggest that no reasonable patent owner would ever have his patent reexamined, for any reason.

    Anyone who has ever posted here on the topic of BRI has nothing by hyper criticism of it. Ever since In re Morris, or thereabouts, BRI has become a monster. Examiners go out of their way to misconstrue claim terms. You and I both know this to be true.

  6. 30

    small entities are far more likely to abandon there patents because of the 40 methods of having there patents stolen or declaired invalid and there inability to afford 15 year legal battles to get any money from them. This is why trolls ate so vital to maintaining an incentive to create for these entities.

  7. 29

    LH – “whitewash” is hardly a term to use when ya havta surrender the patent and give up the C&C.

    Sides which – the examiner fails at his job to turn up the reference in the first place and it is the patentee on the line – yet ya have no problem with that.

  8. 28

    Yes, because no patent holder has ever used the reexam process to whitewash an issued patent when a new reference is discovered.

  9. 27

    Ya everything good and fair in the system since 1953 was me. I should have been all done with non-practicing entity status by the time I was six months old in 1954 after selling my first 10 or twenty patents id have been rolling in reinvestment dough if we had had a perfected system like the one I am proposing to congress. I Got stuck in the steal the candy from the baby era.

  10. 25

    Dear DC, thanks for the interesting graphs and data, which of course bring to mind James Carville’s famous phrase. But note in your “Two initial thoughts on these results”, the second point says, “Even though… lower maintenance fee… small entities are much more likely to abandon their patent applications.

    Apps s/b “patents”.

  11. 24

    Here is a similar story

    The United States Patent and Trademark Office (USPTO) published an Official Gazette (OG) notice outlining special accommodations available to patent and trademark applicants and owners affected by the catastrophic events that took place in Japan on March 11, 2011.

    “Our thoughts and prayers are with the people of Japan who continue to suffer from the effects of the earthquake and resulting tsunami,” said Under Secretary of Commerce for Intellectual Property and Director of the USPTO David Kappos. “The USPTO is offering assistance in the form of flexibility on deadlines to the full extent allowable under our laws to Japanese applicants.”

  12. 22

    Just to make it clear, Paul. No patent owners are not clamoring for reexaminations. For generations and generations, they have had reissues to deal with unexpected prior art, if any such comes along and they really need to narrow their claims.

    The problem for patent owners is not expense. It is — hold your breath — “broadest reasonable interpretation” –. That right there is why reexaminations are torture for patent owners.

    The clamoring for reexamination and oppositions and the like are coming from only one source, and it is not patent owners or anyone who needs the patent system for patent systems purposes. It comes, by and large, from people who do not want a patent system and who are doing their level best to weaken it, remove its teeth, or make it the exclusive club for the well heeled and the established.

  13. 21

    imho the patent expiration date (assuming all maintenance fees are paid) should be calculated and printed on the cover page of each patent

  14. 20

    Daniel, thats a good theoretical point.
    But first, the whole point of the Byzantinely complex patent Term Adjustment systen we have had since 1999 [imposed largely by a certain CA Congressman] was to guarantee patentees at least 17 years [or more] of patent term after issuance even though running 20 years from the earliest claimed U.S. filing date. So, theoretically anyway, patents getting less than 17 more years from issuance are only those patents where applicant delays were substracted from PTA’s.
    But, more importantly, it’s already dangerous enough for everyone to keep track of maintenance fee due dates to avoid accidental lapsed patents. It would be a lot worse if they all varied individually depending on calculations [which might not even be correct] of how much term was left at issuuance for each.

  15. 19

    Ned, I was talking about the PTO recovering its [unavoidable] expenses for having to handle patent reexaminations, not whether you, or some patent owners, like reexaminations or not.
    [P.S. did you realize there will be FIVE reexamination systems per S.23 just passed by the Senate by a huge vote margin? A lot of other people must really like reexaminations, at least in preference to lay jury trials.]

    Even for individual patentees, if it were not for maintenance fees, the U.S. PTO would not be able to have low application filing fees, one of the lowest of all examination fees, and the 50% deductions for small entities, all of which encourage more application filings and are a major benefit to individuals, small companies and universities being able to do so.

  16. 18

    Dennis –

    Setting maintenance fee payments on the basis of the issue date made sense when 17-year patent term was also calculated on the basis of issue date. Although the term is now 20 years from the earliest filing date (subject to PTA for PTO delays and PTE for FDA delays – don’t you love acronyms?), the maintenance fee schedule was never adjusted accordingly.

    So while the state of the economy undoubtedly has been a big contributor to the uptick in abandonments due to non-payment of maintenence fees at 3.5 years after grant, it would be helpful to examine how much patent term was left when the non-payment occurred. If examination was taking longer than it had in earlier times – and it seems that under John Dudas, that was the case – then by the time the 3.5 year fee came due, many patents might not have been worth as much at that point in time as if they had issued more quickly. Put differently, if there are only 8 years of life left for the patent versus 14 years left when a maintenance fee comes due, you might be less inclined to pay the maintenence fee. This is particularly true if, for example, the technology is such that it stands a chance of being adopted in the next 14 years but not in the next 8 years.

    On that note, breaking the analysis down by technology area might also be helpful. If a drug candidate is still in early clinical trials when a maintenance fee comes due, the maintenance fee will probably be paid, especially since there’s the possibility of eventually obtaining a PTE; but if the candidate already failed clinical trials, the patent won’t be renewed. Conversely, in the high tech area, if a particularly technology is just being adopted across the industry, then it makes sense to pay the maintenance fee, but if the grant of the patent took so long that one knows the technology has been passed by for something else, there’s little incentive pay the fee.

  17. 16

    Maintenence payments were created by myself although my intention was for only the profitmaking ones to be assesed. Although 99% of issued patents are trash there is also a refusal to deal issue on the 1% good resulting in theft by the financiers after abandoment. investment advisors consider them these valuable undeveloped patents intangible and advise in cliants to not invest and invest in previous ip products with questionable titles.

  18. 15

    Oh and by the way, the longer you delay going to issue, the more time you have, to sculpt your EPO claims to catch the infringer, all the while relying on Art 67 EPC to demand from the infringer compensation for acts committed before issue that would have infringed had they been committed after issue. Why rush to issue? It’s a no brainer, isn’t it?

  19. 14

    Paul, if you don’t like it that the EPO finances its operation with annual maintenance fees, here are some more thoughts for you to play with:

    1. European FTO is easier, if duds drop earlier off the EPO’s pendency list.

    2. Pendency of worthwhile apps goes down, when duds drop earlier off the inventory

    3. You want speed to issue? You can have it. No fee, no reason, just ask for PACE.

    4. But is that what you want? after the EPO issues, you don’t pay EPO annuities any more. Instead you pay per EPC country. And there are already 40+ of them.

  20. 12

    Re the above question, yes, U.S. maintenance fees regularly increase with COLA’s. E.g., in 2003 the 11.5 yr fee was $3100. In 2004 the 3.5 yr fee went up to $910. Dennis notes above that the current maintenance fees are $980, $2480 and $4,110.
    So the 3.5 yr fee increases are not significant, but the 11.5 yr fee has increased in 8 years by $1,010.
    [Still a bargan compared to some of the cumulative ANNUAL fees, even for pending applications, charged by the EPO [even for its own prosecution delays], and some countries maintenance fees for older patents.]

  21. 11

    The cost of examining a patent is higher than the fees charged during prosecution. The costs are kept low up front to encourage people to file and therefore supposedly “promote the progress of science and useful arts”. The maintenance fees are meant to shift the burden of the costs of examining the applications to those that have successful inventions (i.e. those that continue to pay the fees). In theory this encourages more disclosure of new technology than would otherwise be obtained if people had to pay the true cost of what it takes to examine the applications.

  22. 10

    I assume that the maintenance fee schedule was steady during the observed period.

    It would be interesting to see…

    A) The first chart replotted distinguishing the mechanical, chem/bio, and computer/electronic arts.

    B) Co-plotting reassignments post issuance date to see if there are spikes near the maintenance fee dates to an entity willing to assume the fees.

  23. 8

    Does it count when the statute mandates the collection of the fee? 35 USC 41(b).

  24. 7

    I thought Federal Agencies were allowed to charge for cost recovery only.

    When you pay a maintenance fee you are not paying the USPTO to do something. You are paying them to *not* do something. You are paying them to not terminate your patent rights.

    And this Blog still doesn’t work with Firefox. (No Post Button.)

  25. 6

    BTW, the above study is just for the first (3.5 year) and cheapest maintenance fee, which many companies just pay automatically because the cost of any maintenance review at that date is not worth the savings.
    But note that on or about 2/25/09 Dennis had posted on this blog another study [which he calculated from OG notices of abandonments]. I had saved this key text:
    “..of the 2229 utility patents that issued on December 31, 1996, the first fee was paid for 84% of the patents, the first two fees were paid for 63% of the patents, and all three fees were paid for 49% of the patents.”
    That means that more than half of all U.S. [utility] patents are abandoned for non-payment by the end of the grace period following the 11.5 year maintenance fee payment. [These are desirably coordinated with abandonment of foriegn equivalents for even greater savings.]

  26. 5

    Just to quibble a bit with “Although the USPTO does not provide any ongoing services to patent holders.” [For its maintenance fees.]
    The PTO has to handle lots of patent reexaminations, which I suspect costs them more than their fees. The PTO also has to handle interferences, most of which involve at least one issued patent, for which they get no other fees, and expensive PTO APJ work.
    More importantly, the maintenance fee non-payments clear a lot of deadwood out of the system, eliminating their infringement search analysis, assisting new product clearances, etc.
    Ironically, many of the economists who complain about too many patents are working with very erronious statistics by including all these patents that are NOT in force, but rather have become donated public teachings for public use.

  27. 3

    The monthly journal “Managing Intellectual Property” has a “What’s Everybody Talking About Currently” column. In this month’s issue it is all about “How to Prune your Patent Portfolio”. Enough said?

  28. 2

    A few thoughts:

    1. Most still choose to pay the first maintenance fee. The numbers who choose not to pay are going up, but not by all that much. Yet.

    2. When money is tight, decisions to spend money come under stricter scrutiny.

    3. Experienced volume users of the patent system are under no illusions about the product they are buying from the PTO. Occasional users often do labour under misapprehensions, at least until the due date for paying that first renewal fee.

  29. 1

    I really wonder what’s the reason for the fluctuations?

    Two thoughts come to mind…
    1) the state of the economy – when the economy isn’t doing as well, people have less money they want to spend, and therefore, are more likely to drop an issued patent or are more likely to be losing money on it in the first place.

    2) the quality of issued patents – when the patent office issues them more freely, the actually issued ones, four years later, are less likely to be found to be useful, so they’re abandoned. Conversely, when less are issued, the ones the people got are so hard fought for and more likely to be the ones people wanted to spend on, that they are more likely to maintain them.

    Perhaps if we overlay this graph with a GDP graph, shifted by 4 years, and an allowance rate graph, shifted by 4 years, we could try and locate such trends?

Comments are closed.