Maintenance Fees 2015

By Dennis Crouch

A substantial percentage of the USPTO budget arrives in the form of maintenance fee payments. This is “easy money” for the USPTO because the Office has already done the work of examination. A patent’s term runs 20-years from the effective filing date, not counting provisional or foreign national priority filing and with the addition of any patent term adjustment. They typical result is a term of 17 years post-issuance, give or take. During that time, a patent applicant must continue to pay maintenance fees at regular intervals: due 3 ½, 7 ½, and 11 ½ years from issuance (with an additional six-month grace-period added on).

The chart below shows the percentage of patents whose maintenance fees have been paid at each of the three stages. You’ll see that the vast majority of applicants pay the first maintenance fee (currently set at $1,600 for large entities); More patents drop out as the second maintenance fee is due ($3,600), and fewer than half of the cases see the third maintenance fee paid ($7,400). I expect that a number of factors drive the apparent trends in maintenance fee payments, including the ever-rising cost, the perceived value of future patent right, patent term remaining, general economic outlook, and current access to funds, for example.

The current fees were put into place in March 2013. It is not surprising that the increased price resulted in the diminished number demanded as shown in the chart. Importantly, however, the drop in maintenance fee payments was slight in comparison to the size of fee increase (3rd Fee increased by 50%).

In addition to the revenue generated, a number of folks (often non-patent-holders) see maintenance fees as a costly-screen configured to weed-out relatively worthless patents so as to ease any freedom-to-operate search.

41 thoughts on “Maintenance Fees 2015

  1. interesting blog..

    http://www.tiongsonlaw.com
    immigration, accident, divorce, hurt in las vegas, litigating malpractice, injured in vegas, hurt by a doctor, las vegas personal injury attorney, negotiating clark county Nevada cases, traumatic injuries, hotel accident, decades of experience, casino negligence, nevada dui, medical malpractice, devastating accident, client rights, protecting victims

  2. Question: How much does it cost, in time spent and dollars billed, to determine whether a patent is useful or economically viable?

    Answer: Likely more than $7,400. (And even if it’s less than $7,400, if the determination is to keep the patent, the money spent making that determination is sunk.)

    Conclusion: If there is any possible reason to keep the patent, just pay the final maintenance fee. Then you get to keep the patent, and your money goes to your patent, not your patent attorneys. Yes, it’s like settling a case with a troll because it’s cheaper. But, in the end, it’s still cheaper.

    1. Question: How much does it cost, in time spent and dollars billed, to determine whether a patent is useful or economically viable?

      Answer: Likely more than $7,400.

      Depends on the patent.

      Great free advice is sometimes easy to come by. You just have to know where to look and what questions to ask.

    1. Joe, good observation. If a patent is not protecting a current or expected product or service, it should be freely licensed, sold or abandoned.

      But that makes every patent owner a troll, not so? Where do we draw lines in defining a troll? Is a troll anyone with a patent that does not have a product, or only those who assert invalid or not infringed patents, and use the high cost of litigation to make some money regardless of validity or infringement?

      1. I don’t think we need to go into a patent troll discussion directly.

        From my perspective at MIT and in my work with companies I see many companies looking at innovations to grow their businesses. They are embracing Open Innovation and Collaborative Innovation as never before. The Patent Maintenance Review Process described in my link is a systematic way to think about otherwise stranded assets. In one case I saw a business line that was going to be abandoned turned into a new collaborative joint venture. That business line had sunk costs (patent and R&D expenses) in the multi-hundred million dollar range. The company had mentally written off the assets until they took a hard look at what they had and who might value those assets.

        Innovations and patents get stranded for many reasons including timing, management and people issues. Not all monetization efforts need to be assertion based. There is that old expression “stocks are sold, they are not bought”. Let’s focus first on whether these assets can create positive value in the right hands – that is what my Best Practices article is about.

        I will leave it to others to deal with the Patent Troll issue – I don’t have a solution to that one yet.

      2. Because its a perversion Ned. It was in somebody’s interest to confuse the public that NPE’s have somehow less right to monetize their patents. True, many bad patents have been subject to acquisition and exploitation, but being an NPA is not a sine qua non of a Troll.

        A Troll, correctly used by the Supreme Court, is a patent litigation abuser.

        Nothing more or less.

        What is litigation abuse? No single pattern, but there are clues:

        very old patents
        broad/abstract/non-technical or method patents
        very high/nutty settlement demands
        sues first, with no warning or attempt to license
        excessive discovery demands
        presenting unprepared/irrelevant witnesses
        acquired patents w/ checkered litigation or prosecution history
        vexatious motions or procedural excess
        changing legal positions on material issues

        Trolls know when they are trolling. They know when their patents are real, and mean something, and when they are junk, and mere rent-seeking pestilence.

        They just don’t care, and why should they?

        1. Martin, item one on your list threw me, “very old patents.” That is what we are talking about here, is it not? — whether to pay the steep maintenance fees for old patents? If the age of the patent were a data point on determining whether assertion of the patent outside the context of protecting a product was indicative of troll status for an award of attorneys fees, then, again, we would all be trolls.

          I don’t think age has anything at all to do with being a troll. However, it may indicate the presence of functional claiming, and asserting functional claims is a problem because such claims should be deem per se invalid under Halliburton.ed

        2. Patent litigation abusers are no different than any other type of court abusers.

          The answer is NOT in patent law.

          The answer is in civil procedure.

          (but be careful of getting the answer you are asking for)

          1. “Patent litigation abusers are no different than any other type of court abusers”

            Utterly false. Patent litigation is always complex litigation, so in both money and time, patent cases stand apart.

            Beyond that, patents are shrouded in mystery and obscurity and the implicit expectation that what is being fought about must be valuable, as untrue as that may be.

            Other litigation requires at least some relationship between the litigants- an accident, contract, or incident….whereas a patentee can reach out from nearly anywhere, from the ether, to demand what they may not deserve….

            Other litigation does not require experts to explain the obvious, and other litigation does not start all over again at the appellate level.

            1. You are letting your unfamiliarity of the territory color your perceptions.

              Read (and understand) Sun Tzu to begin to grasp just how wrong you are – complexity of subject matter (and the fact that patents are public) just wreck your arguments. As to your stumbling with civil procedure matters – again, I will refer you to your retained counsel. They already should be operating with your informed consent, and your parading about your being ill-informed does not reflect well on them.

    2. Yes, and as noted, the savings from the maintenance review can be a very great deal more than just the U.S. maintenance fee if equivalent foreign applications of that U.S. application had been filed. [Foreign patents can have high and increasing maintenance fees].
      Furthermore, long-pendency U.S. applications – patents issuing from continuations, divisionals or plural RCEs – may have little actual patent term left from the 20 years running their claimed original filing date by the time of even the second, 7.5, maintenance review time, and much greater likelihood of obsolescence.
      But one needs to check in those cases to see if long pending serial continuation and divisional patents were the result of deliberately keeping them pending a long time so that claims could be amended or added to cover actual products of others. [This tactic is enabled by the long history of poor PTO docket control over actual [total] application pendencies.] That is, check to see if that tactic actually worked. [E.g., I doubt if many Lemelson or Hyatt patents have maintenance fee abandonments?]

  3. Here is a list of the top 9 [from a 3/09 blog] of companies having the most U.S. patents that expired early due to not paying maintenance fees.

    1. NEC Corporation [2484 expired patents]
    2. International Business Machines Corporation [2408 expired patents]
    3. Canon Kabushiki Kaisha [1954 expired patents]
    4. Eastman Kodak Company [1821 expired patents]
    5. Sony Corporation [1566 expired patents]
    6. Hitachi, Ltd. [1387 expired patents]
    7. Matsushita Electric Industrial Co., Ltd. [1155 expired patents]
    8. Mitsubishi Denki Kabushiki Kaisha [1034 expired patents]
    9. U.S. Philips Corporation [980 expired patents]

  4. For comparison, about 2/25/09 Prof. Crouch posted on this website a study calculated from OG notices of abandonments. Here is some key text:
    “..of the 2.229 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.”* (He did not add-back a few cases that were revived due to alleged unintentional abandonment.)
    He also observed that such patent abandonment rates are higher for foreign owned patents.

    *To look at the above percentages the other way around, by 12 years from issue 51% of patents were no longer in force due to non-payment of the 11.5 year maintenance fee, and 37% were already no longer in force after only 8 years from issue for not paying the 7.5 year maintenance fee.

    1. Paul,

      The meaning of the data is a little bit more tricky than first meets the eye.

      Since the date listed of each “check” is consistent, the “family” of applications for that date data point is not the same: any graph effects that appear to happen “at the same time” – the dip at 2010 and subsequent rise and fall – must be viewed as EXTERNAL to the patents themselves, since the family of patents that make up a data point do not align vertically, but align on a successive three and a half year slope.

      In other words, the data points on the “Paid Third Fee” group at the end of the timeline correspond to the data points on the “Paid Second Fee” group three and half years towards the center of the graph.

      The real lesson here is that external factors appear to have higher effects than anything else. Since there appears to be very little “delayed effect” event from an angled “rippling across time” and “instant” perturbations are more easily seen, speaking about any “quality” of actual patent families is NOT supported by this picturing of the data.

  5. Here is something some patent maintenance reviewers have not considered:
    Even though 11.5 maintenance is dropped, could you still sue for up to six years of back damages if you later found infringements that had occurred before the maintenance abandonment? I don’t know of any case law on this point. But
    35 USC 271(a) allows suit for back damages on an expired patent, and when you abandon, you’re abandoning future rights, not necessarily past rights (which you had maintained by prior maintenance payments)?

  6. There are several dubious reasons, some already noted below, for huge differences between patent owners in paying U.S. maintenance fees, even in clear cut cases of technical or marketing obsolescence by the 11.5 year of issuance. For example, wanting to be able to claim a larger number of “patents in force” to look more “high tech” to the stock market. Differences ranging from a professional cost saving maintenance review team at IBM [yes, a good maintenance review system costs some money, but it saves far more – enough for a NEW patent application each!] to other companies that never abandon anything in the U.S.
    Much worse is to keep on paying the much higher and rapidly increasing annual maintenance fees of their foreign equivalents for patents with claims to something no one, even themselves, has never used or would have any reason to use. Not even taking advantage of the foreign fee lowering “license of right” option in the final years of a patent no one has yet shown any interest in.
    Surprisingly few corporate accountants ever seem to ever intelligently question patent-related expenses in general, such as by bringing in independent patent-knowledgeable auditing experts to help overcome internal resistances to making decisions or changes and identifying patents that might actually be licenseable.

    1. PM bringing in independent patent-knowledgeable auditing experts to help overcome internal resistances to making decisions or changes and identifying patents that might actually be licenseable.

      I share the sentiment but question the underlying assumption (as I’m sure you do, Paul!) that “internal resistances to making decisions” may be surmounted by logic and reason in every instance. In some cases, worthless patents are maintained simply because the CEO’s name (or someone else’s name) is on them or because the CEO decided to file the applications in the first place (against the patent attorney’s advice) and sees abandonment as capitulation. The sunk cost fallacy and inane accounting methods (i.e., five patents is better than three, even if two of the patents are plainly worth nothing) also play a huge role, as you note.

      1. Yes, I did not even get into the strange effects of things like depreciation terms for some patent costs.

  7. If they want to maximize revenue, they should reverse the fees and charge more when the largest number of patents have the largest earning potential. That could also incentivize faster monetization or allowance of a lapse.

    1. Charging high maintenance fees a short time after a patent issues would be especially unfair to small inventors and start-up companies. Furthermore it would often be too soon to see if potential competitors or others might use the claimed invention. But by 11.5 years of issuance [at least 13.5 years from their filing date, and often more] non-use and obsolescence can be very much easier to identify.
      If you look at the ages of patents being sued on you will find that by far the larger number are not newly issued patents. That is further demonstrated by the non-existent number so far of brand new [9 months or less] patents that have been challenged in a declared AIA PGR proceeding, and the low % of patents in their very early years of life being challenged in IPRs and CBMs.

  8. If the data could be segregated by art field, you might have a more interesting tale to tell.

    From the rise in second and third payment rates, it appears that patented inventions are holding their value (or presumed value) longer. This may be opposite the gen eral notion that patent term for software innovation need not be as long since the rate of innovation makes older patents obsolete quicker. It appears that what people do is not matching what some pundits say.

  9. Dennis,
    Why is this illustrated as a scatter plot? More particularly, what does one dot on the plot represent? The percentage for maintenance fees paid in one calendar month?

    1. Sorry for the lack of description. Each dot represents one week’s worth of patents. As you may know, the USPTO issues patents once each week (on Tuesday mornings). Maintenance fee payment due dates are tied to the issue date and thus it makes sense to continue to group on the weekly basis.

  10. For freedom-to-operate, one always checks to see if patents have lapsed, but it doesn’t happen that often for patents that might be a problem. On the other hand if, as is commonly stated in my experience, only ten percent of patents have real economic value, the high rate of maintenance fee payments does not make sense. Perhaps it takes a long time to assess the value of the patent, but this cuts against it being valuable. I think it is a combination of inertia and hope springs eternal. It would be interesting if anyone has done a survey to find out why maintenance fees continue to be paid at these rates.

    1. Just a thought – lawyers are highly risk averse. Especially patent lawyers (in-house and at firms). We’d much rather keep paying too much in maintenance fees than see the rare case where a patent’s economic value is realized AFTER we recommended fees not be paid.

      1. Seems to me that, if I’m the one that has to answer the Big Question: “Pay renewal fee Y/N?”, I’m going to answer it Y except in cases where I know for sure that it’s safe to decide N.

        What if I decide N and then it turns out that the patent was important after all? Who gets the blame for the financial loss to the company?

        But what if I decide Y? Who can criticise me then?

        1. “But what if I decide Y? Who can criticise me then?”

          Uh, let’s see, the finance people, the accountants, the CEO. Just to name a few.

          1. If I’m the one calling the shots and the finance people tell me we are paying too many maintenance fees, I’d be happy to let them make the call and take the associated blame when the lapsed patent is later infringed by a competitor.

            1. That’s a cute answer, but not really that helpful. If you’re the one calling the shots, you’re supposed to know which 12 year maintenance fees are worth paying, and which ones aren’t. The answer isn’t, “They’re all worth paying!” and it’s not, “None of them are worth paying!” It’s somewhere in between. And it’s your job to know which ones are which. If you think you’re going to just point your finger at somebody else when it comes to assign blame, you’re kidding yourself.

              You: “I knew this one was worth paying, but the finance guys wouldn’t let me pay it. So I just let it lapse. It’s their fault.”

              The CEO: “If you knew it was worth paying, why didn’t you go up the ladder to make sure it got paid?”

              You: “Uh. Well. Uh. Gee. I didn’t know I could do that.”

              The CEO: “Here’s a box. Clean out your desk.”

              1. I had assumed it was a budget thing. Thus: here’s your wad. Do the best you can with it.

                Then all you have to decide is which cases don’t make the cut.

                Otherwise, if it is a case of “too many cases”, that reminds me of the line in the film “Amadeus” about the life of W A Mozart. At the first performance of his opera, the Emperor was not convinced. Mozart: Tell me, your Majesty, what exactly troubles you? Reply: Too many notes. Says Mozart: OK, tell me please, your Majesty, which ones of my notes are the superfluous ones. Which ones should I strike out? Needless to say, no answer from his Majesty!

                1. Even if it is just a budget thing and you do have to decide which ones make the cut, you still cannot “point the finger” as AAA JJ aptly point out.

                  His response is still very much on point.

                2. “Then all you have to decide is which cases don’t make the cut.”

                  Right. That’s all you have to do. Just don’t mess it up.

    2. I agree with athought. The relative cost of the 12th year fee is a drop in the bucket compared to potential infringement damages.

    3. Hindsight is your answer. The idea that someone can “pick the winners” from a portfolio of thousands of patents is similar to people believing they can “beat the market”. People stop paying maintenance fees when their patents are clearly dead in the water. Until then, the fees are probably worth paying.

Comments are closed.