Utility Patents Granted per Calendar Year, 1840-2020

By Jason Rantanen

I’m getting ready to teach my Fall 2021 Patent Law class, and that means doing an updated patent grant graph.  This year’s version shows U.S. utility patents granted per year from 1840-2020:

Data for 2021 isn’t included in the table, but as of July 31, 2021, the authority file contains 195,480 patents.  If the pace remains the same, that would predict about 335,000 patents granted this calendar year–around 10%  [correction: 6% – 356k minus 335k is 21k, not 31k] lower than 2020 (356,640 granted patents).  Looking at the spreadsheets on the Patents Dashboard, it looks like filings are a smidgen lower, but the real difference is the number of allowances: as of the end of June, 2020, the PTO had issued 278k allowances for UPR’s that fiscal year, while as of the end of June, 2021, it had issued 249k allowances for FY 2021.

Methodology:

I used the 2021-07-31 Patent Grant Authority file, parsed it by date, and tabulated the number of A1, B1 and B2 kind code records by year.  (For those who might be confused by the use of A1, the authority file uses A1 to indicate pre-2001 utility patent grants).  I included patents that are marked as Withdrawn.  This graph uses calendar years (Jan. 1 – Dec. 31) instead of fiscal years (Oct 1 to Sept. 30).  UPR stands for “utility, plant, reissue.”

93 thoughts on “Utility Patents Granted per Calendar Year, 1840-2020

  1. 11

    Good Lord! My theory that public choice theory is operating writ large in the patent system is proving true – by the numbers!

  2. 10

    Here are some past blogs with related statistics that might be of interest. I only went back a couple of years.

    Patent grants per year 2020: link to patentlyo.com

    Patent grants per year 2019:
    link to patentlyo.com

    Average number of inventors increasing: link to patentlyo.com

    link to patentlyo.com

    Women inventors increasing: link to patentlyo.com

    Allowance rate by entity size:
    link to patentlyo.com

    Estimated design patents by 2018:
    link to patentlyo.com

    Growth in citation of US patent applications:
    link to patentlyo.com

    Use of priority claims increasing:
    link to patentlyo.com

    The increase in foreign filings as a percentage of total US filings (nowadays more than half) is shown by comparing the WIPO chart on page 12 of this 2020 WIPO report

    link to wipo.int

    with a similar chart on page 30 of this 2017 WIPO report:

    link to wipo.int

  3. 9

    For those who’ve asked, I use this chart to show how many U.S. patents there are, and that there have been a lot of patents issued over the last twenty years. I think that having a sense of the scale of the patent system, and how that scale has changed over time, is important for students who are just starting to learn about it. I also draw the inference that with a lot of patents, there’s a lots of potential for complexity within the patent system. Getting into causality can come later, but at the start of the class we’re just figuring out the shape of the elephant.

    Also, I made a math error above – 335,000 would be 6% lower than 356,640, not 10%. I’ve corrected it in the post.

    1. 9.1

      Scale has changed? See Ron’s answer below for a proper way to view scale.

      Innovation feeds innovation. If you think that proper scale is linear, then perhaps you should not be teaching.

      1. 9.1.1

        Anon, re your comments like this “perhaps you should not be teaching.” Rather than continue to misuse this blog for personal professional insults, perhaps you should be doing tweets on Twitter or some other outlet for hostility therapy.

        1. 9.1.1.1

          Indeed. Paul, he posts with a clear conscience because, he tells himself, he does it only to provoke a response. The greater the insult, the more revealing the reply, he supposes.

          And, the more responses he provokes, the more he smiles, the more he sees himself contributing constructively to the thread.

          When I respond to an anon post, it is only when I think that i) what I’m writing is of interest to other readers and ii) also contributes something to the analysis of the subject under debate.

          Let’s both of us now see what reaction your post and mine elicits.

          1. 9.1.1.1.1

            When I respond to an anon post, it is only when I think that i) what I’m writing is of interest to other readers and ii) also contributes something to the analysis of the subject under debate.

            Good for you. Your contributions are consistently worthwhile and informative. I appreciate the way that you manage to extract value from even the most unpromising starting material.

            1. 9.1.1.1.1.1

              Your contributions are consistently worthwhile and informative.

              The only thing consistent from Paul is IPR cheerleading and positions AGAINST holders of patents.

              That he may be ‘polite’ to you (and calls for OMFG politeness), is NOT a good thing.

              A bunch of Nancy’s need to toughen up a bit.

          2. 9.1.1.1.2

            because, he tells himself, he does it only to provoke a response

            Wrong and wrong.

            YOU are the one that claims to post inane items ‘to provoke responses.’

            My posts are NOT merely ‘to provoke,’ and IF you are only feeling provoked, then YOU are missing the point.

            Simmer down, get a grip of your emotions, and then try reading again.

        2. 9.1.1.2

          Jiminy – try not to have such thin skin, you whining Nancy.

          The point I provide is a valid one: if you are in a position to be teaching about innovation, you SHOULD have a better understanding of innovation.

          Then again, even when ‘polite’ (as Ron K was to you), you are intent to NOT understand the driving issues.

          This fully comports with the John Maynard Keynes call for words being a bit harsh as they SHOULD BE an assault on the unthinking.

  4. 8

    The surge from 2021 to 2020 is pretty impressive. Is there any value to this discussion in comparing one jurisdiction, the USA, with another, say, Germany, for the years 2011 to 2020? Suppose the respective year-on-year surge is very different.

    We can do this using the stats published by the EPO.

    In 2011, there were of the order of thirteen thousand European patents granted to Applicants resident in Germany and also about thirteen thousand to Applicants resident in the USA. The comparable figures for 2020 were DE: 20k and USA: 35k. Is that i) surprising or ii) only to be expected?

    Here a Link:

    link to epo.org

      1. 8.1.1

        What do I think? Pretty much what Night and Paul have suggested.

        For some reason, those deciding how much to spend on patent procurement have for the last ten years been demanding ever higher numbers of issued patents. That reason, I suppose, is that those deciding in the USA whether to invest in an enterprise make their decision on the basis of how many issued patents the enterprise has under its control.

        I recall that old catch phrase from the carpet salesman “Never mind the quality; feel the width!”

        1. 8.1.1.2

          Really?

          Way to be definitive…

          Pretty much what Night and Paul have suggested.

          Have you grokked that these two are in fundamental disagreement?

          Throw some more at the wall and hurry to pat yourself on the back.

          1. 8.1.1.2.1

            You see more disagreement between Night and Paul than I do. I don’t see any “fundamental” disagreement between them.

            I think that the pressure on the CEO’s of US-based corporations, these past 10 years, to swell the size of their patent portfolio, ie Paul’s point, has resulted in a serious increase in the number of patents issued by the USPTO off of continuation applications, Night’s point.

            That’s all.

    1. 8.2

      I am not sure that I would have expected the rate of DE grants to US residents to have grown faster than the rate of DE grants to DE residents. That surprising find will affect my baseline expectations about EP patenting by US applicants in future.

      1. 8.2.1

        Greg you are in a better position than me, to explain the “surprising find”. I have already offered my best guess to explain it, namely pressure from CEO’s in the USA, in the computer-implemented invention space, to build a bigger pile of patents than possessed by the competitor.

        But if that is the reason, perhaps the figures are revealing an inability of Europe to compete with the USA and China in the field of SEP’s?

        Who knows? I sure don’t.

  5. 7

    No amount of “normalization” of any kind, even of sea turtles,* is going to eliminate this charts demonstration of a huge and historically unique increase in issued U.S. patents between 1999 and 2020. The reason TBT. E.g., if there really was a huge increase in the number of patents issued as divisionals or continuations [in spite of RCE availability] AND also issuing rather than abandoning their parent applications, that number should should be readily determinable statistically rather than conjecturally.] I do however, think a factor might be the huge increase in private venture capital investments in software/internet related business ideas over this time period?
    *The percentage of foreign-origin patents has NOT significantly increased in that time period either.

        1. 7.1.1.1

          >Hoisted by your own petard. “The total number of U.S. inventions” is not what this graph is showing.

          Greg, you misinterpreted the graph using the context of patentlyo.

          I did not express my feeling but legitimate objections to the presentation of the data. You, the hoisted one, proved my point.

          1. 7.1.1.1.1

            But note that specifically, Greg’s own feelings have been hurt so often that he refuses to note certain posts.

            Look at the “company he keeps” (or prefers): examiners and IPR cheerleaders.

            Anything outside of his “Big Pharma” views — he discounts and dismisses.

            (and also note that there is a not-small enjoyment to be had with his cognitive dissonance between his Big Pharma views and his political Liberal Left leanings, with Biden’s tilt to the India/South Africa FULL IP (and critically more than just patent) waiver. You won’t catch him putting his money on those particular beliefs.

            Hypocrisy is his hallmark.

            1. 7.1.1.1.2.1

              anon >>First attempt to dismiss.
              Next attempt to denigrate.

              Maybe try actually addressing on the merits, Ben…

              Could you try that Ben? Oh wait you know you’d lose on the merits so you want to be disruptive.

              1. 7.1.1.1.2.1.1

                I’d be embarrassed to post something like that after ignoring a polite question (post 5.1.2.1.1.1.2), but I agree it’s quite in line with your brand.

                And “the merits”? As to whether Paul’s points are feelings while your points are legitimate objections?

                Grow. Up.

                1. Ben: Your attempt to be disruptive and just ignore the merits of my objections are obvious. And your attempt to deflect and call me childish when your brand is one of being disruptive and childish is ridiculous.

                  I am not going to restate all my points as they are within this post. And Ron before you made this comment certainly answered your question. And Greg misinterpreting the graph answered your question. And Lemley presenting the graph as evidence of a healthy patent system answers your question.

                  The fact is that my post is all substance and your posts all about being disruptive. Par for the course.

                2. Greg misinterpreting the graph answered your question.

                  Oh for pity’s sake. Go back and re-read 5.2 carefully. Then try to point to the exact line where I “misinterpreted” the chart. It cannot be done, because I did not misinterpret the chart. I did not want to make a big deal of your sloppy reading and monomania, so I was content to pass over your error in silence. I see now, however, that this is about to become a founding premise of a long run of fallacious arguments, so I suppose that I need to address the error now before it expands further.

    1. 7.2

      Paul: “No amount of ‘normalization’ of any kind, even of sea turtles,* is going to eliminate this charts demonstration of a huge and historically unique increase in issued U.S. patents between 1999 and 2020.” (My emphasis). Really? Have you actually looked at the numerical data on the chart? Perhaps it is Jason’s graphical presentation that misled you here. There is an unfortunate tendency for those who graphically present historical patent time-series to use linear scales that mask the appearance of any natural growth trends that are inherently exponential. Therefore, plotting the number of issued patents in a log scale would reveal the true growth rate by observing the slopes. In contrast, the slopes in linear scales are highly misleading as to growth rate. For example, a log scale would have revealed that in the same span of 21 years, the number of issued U.S. patents from 1849 to 1870 grew by a factor of 12.3 — from 988 to 12,157. This growth factor dwarfs the growth factor of about 2.4 from issuance of about 146,000 UPR in 1999 to 356,640 in 2020. Paul’s pronouncement of “historically unique increase” is clearly wrong.

      What about Paul’s conclusory declaration that no normalization of any kind can explain the observed growth in patenting in this 21-year span? Wrong again. Normalizing for commercial activity in the market by indicators of growth in such market activity is an appropriate way to determine whether the increased patenting activity is unique to patents, or more generally related to underlying economic drivers for patenting such as commercial market activity. For example, trademark applications are a good indicator of new commercial activity in brands and business growth. According to the PTO’ Annual Reports, the number of trademark applications grew from 295,165 in 1999 to 738,112 in 2020 — by a factor of 2.5. Therefore, the patenting growth factor may be fully explained by the growth in commercial activity.

      1. 7.2.1

        Thanks Ron. I’ve made these points before and yet patentlyo just continues to put up the same chart.

        I’ve said before normalizing based on GDP growth in technology areas would be a good way to normalize the data.

        I’ve also pointed out that in previous charts that show an increase in patent applications that there was actually a decrease in patent applications for inventions made in the USA.

        I’ve also pointed out that continuations and divisionals are far more popular now in my practice post AIA as companies want multiple patents on similar inventions as they assume IPRs will be used on patents that are valuable.

      2. 7.2.2

        RZ, that is changing the subject, from the huge 1999 – 2020 increase in total issued patent numbers per year to a comparison to a % increase in an ancient period starting with very much smaller total patent numbers in a peak growth period of the industrial revolution. The great growth in total 1999 – 2020 patent numbers is significant because of repeated claims on this blog that U.S. patent filings are greatly decreasing due to 101 rejections, IPRs, KSR, etc. over much of that 1999 – 2020 time period.
        The huge increase in trademark applications last year has been reported to be an anomaly due to many stay-at-home small business startups, not to any big R&D increase. If U.S. R&D spending doubled 1999 – 2020 [but did it?] that would be a logical explanation for what I said was “The reason TBT.”

        1. 7.2.2.1

          >>U.S. patent filings are greatly decreasing due to 101 rejections, IPRs, KSR, etc. over much of that 1999 – 2020 time period.

          Here is more proof that the data is being misinterpreted (based on the context) to mean that all these patents are issuing for US inventions. That is not what the data is showing.

          1. 7.2.2.1.1

            NW, first, I never even said “U.S. inventions.” Secondly, as I did note [as established in an earlier blog here] that the % of U.S. patents of foreign origin has not significantly changed over that time period. Thirdly, 101, Alice, Mayo, IPRs, KSR, etc. apply equally to such current U.S. patents of either foreign or U.S. origin.

        2. 7.2.2.2

          This is decidedly NOT “changing the subject.”

          It is directly on point.

          Further, your attempt to mislabel — and Greg’s faux “dignity” are exactly why data presentations like this NEEDS to be accompanied by caveats.

          Ron captured this perfectly, given that innovation breeds innovation — this should be portrayed on a time-log scale.

          You — and Greg — have chosen to clench tight your eyes to the valid points presented by Ron. This places both of you as part of the problem and attacks on a pro-innovation culture.

        3. 7.2.2.3

          Paul: “that is changing the subject, from the huge 1999 – 2020 increase in total issued patent numbers per year to a comparison to a % increase in an ancient period starting with very much smaller total patent numbers.”

          Your insistence on comparing absolute increases in the number of issued patents per year rather than percentage increases reflects fundamental lack of established quantitative analyses skills. In what you call the “ancient period starting with very much smaller total patent numbers,” there was also a “very much smaller total” number of people in the population from which patenting could emerge. In 1850, the U.S. population was merely 23 million. See link to census.gov. It was 331 million in 2020. Comparison of absolute increases in the number of issued patents is meaningless.

          I have not “changed the subject” by using percentage growth – rather, I have merely explained why your statement about the “huge and historically unique increase in issued U.S. patents between 1999 and 2020” was simply nonsense.

      3. 7.2.3

        [A] log scale would have revealed that in the same span of 21 years, the number of issued U.S. patents from 1849 to 1870 grew by a factor of 12.3 — from 988 to 12,157. This growth factor dwarfs the growth factor of about 2.4 from issuance of about 146,000 UPR in 1999 to 356,640 in 2020. Paul’s pronouncement of “historically unique increase” is clearly wrong.

        Agreed, “historically unique increase” was a silly and factually indefensible framing of the recent trend in patent counts. The other interesting take-away from your more careful analysis is that patent count trends seem (at least at first blush) remarkably unresponsive to changes in the legal incentive. From 1999 to 2020 the U.S. granted a longer patent term with a greater overall likelihood that an asserted patent might be found valid an infringed than was the case from 1849 to 1870, and yet the market still responded by seeking fewer patents (relative to either per capita or per $GDP) in the later 21 year window than the earlier 21 window. Evidently, demand for patents is less responsive to legal incentive structures than to other factors (presumably market factors, although I confess not to know enough about market conditions in the immediately pre- and post-Civil War era to speak to this point with any certainty).

        1. 7.2.3.1

          Wow Greg. You are just making things up at this point. And doing based on this graph that only gives us a count of grants without telling us how many are from patents application of inventions not made in the USA nor of the affect of continuations/divisionals.

          And yet we know, for example, that the number of Chinese applications has skyrocketed.

          Although it is nice for you to acknowledge some of your mistakes based on Ron taking more time than I did to explain the patentlyo obvious problems with the graph.

          1. 7.2.3.1.1

            was a silly and factually indefensible framing of the recent trend in patent counts

            You will note that he did not actually state that DIRECTLY to Paul.

            You will also note that MaxDrei seems oblivious about the very marked difference between your views and Paul’s.

            It’s as if acknowledging these things is itself something to be avoided (VERY much like a good simian trained to obey and to not apply critical thinking, or to you the Wizard of Oz phrase, these people have been trained to ‘pay no attention to that man behind the curtain’).

            And HEAVEN FORBID, if someone dare raise their response over a hush whisper or show ANY hint of disagreement, then THAT person should be ‘shunned.’

  6. 6

    The real outrage here is that both this graph, this post, and every prior comment under this post fails to highlight how endangered sea turtles are. Everyone who has distributed any opinion or information without referencing this issue (myself included) is clearly an unethical individual.

    1. 6.2

      Suppose that someone presented a graph of sea turtles and their increase was presented in a salutatory fashion, but as it turns out green sea turtles were actually becoming extinct but no mention was made of this. Is that misleading? Is that unethical if presented at a conference on Ocean pollution and fishing? What about if it put up constantly on a blog for people to discuss the Ocean and sea turtles?

    2. 6.3

      You may laugh, but the most favorite patent applications that I wrote (and became patents) were directed to protecting turtles from being hit by boats in the water. That, and my famous prosecution of the ‘turtle ear’ patent. Oh, how I long for the good old days in the 90s!

    3. 6.4

      Sorry, Ben. Friday the 13th was LAST week.

      (no turtles were harmed during the submission of this comment)

    4. 6.5

      First attempt to dismiss.
      Next attempt to denigrate.

      Maybe try actually addressing on the merits, Ben…

  7. 5

    This should be split out by where the invention was made. I believe many of these were not made in the USA and the rise merely indicates the greater use of continuations because of IPRs and the huge increase in filings from China.

    The data is almost meaningless as a measure of patent health in the USA without factoring out these two factors.

    1. 5.1

      I wonder too Jason what are the ethics of putting up charts to indicate a healthy patent system when one knows that there are significant factors that have changed over the last 10 years that may account for the data.

      I know that Lemley does this as proof that the weakening of the patent system hasn’t hurt innovation.

      1. 5.1.2

        There is nothing unethical about Prof. R.’s presentation of these data (unless he has falsified them, but I see no reason why we should suppose that he has). The data are not “misleading,” because data can only mislead if they lead you somewhere in the first place, and these data do not.

        If you think that there are some other data that would tell a story that you want to tell (clearly these frustrate you in that they do not tell the story that you want to tell), you are free to assemble and present those data. It is a shabby business, however, to intimate aspersions against Prof. R.’s ethics merely because he has not done that work for you.

        1. 5.1.2.1

          Greg, with all due respect, you are FOS. First, the only person I clearly made “asperations against [his] ethics” is Professor Lemley who has put up a version of this chart in the past and said that it indicates the patent system is as healthy as ever and innovation is not hurt by weakening patents. Second, given the context of how this chart has been used and its implicit meaning, one can say that without further information that it is unethical to present this chart.

          Full stop. In fact, Lemley’s use of the chart is clearly an ethical violation in regards to Stanford. Jason’s use of the chart in a celebratory fashion to indicate the health of the patent system is suspect at best.

          And, no kidding that data is data. The words used to present the data and where and how the data is presented can be misleading, though.

          1. 5.1.2.1.1

            First, the only person I clearly made “asperations against [his] ethics” is Professor Lemley… Second,… it is unethical to present this chart.

            Er, o.k., so you had not clearly slandered Prof. R.’s ethics at the time that I wrote, but now you have. How that makes me “FOS” is rather opaque, but no doubt I will puzzle it out eventually.

              1. 5.1.2.1.1.1.1

                Greg does not accept the fact that the raw data can be (and has been) misused.

                Note that every time this type of data is presented, the same call for caveats has been made, — and ignored.

                1. But Prof. R has not used the data to show anything (at least in this post). I don’t see any conclusions about the data, what the data means, or what the data says about the best choice of meat to put on a hoagie. The data is just facts. I understand that you would like Prof. R. to add a baseline level of analysis to preemptively address any conclusions that you think are misleading. Prof. R. doesn’t owe this to you and him not providing it does not make anything bad or unethical.

                2. Meh, no – as I noted in my first comment, simple caveats would do (no need for any “baseline analysis” at all)

              2. 5.1.2.1.1.1.2

                Could you explain what “further information” the good professor provided that makes your statement ” without further information that it is unethical to present this chart” not apply to him?

      2. 5.1.3

        All patents are good and great before the Lord, NW. After all, article 1 section 8 doesn’t say: “To promote the Progress of Science and useful Arts in the United State, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries”

        1. 5.1.3.1

          You are most definitely off here, squirrel.

          Patent law was and is a Sovereign-Centric law.

          Sure, there are international treaties – but the basis for those are reciprocity rather than any sense of ‘One World Order.’

          The minute that any international treaty stops being in the benefit of this Sovereign, this Sovereign should break from that treaty.

    2. 5.2

      The total number of U.S. inventions made each year is a good and useful thing to know. The total number of U.S. patents issued each year is also a good and useful thing to know.

      Your complaint (“[t]he data [are] almost meaningless as a measure of patent health in the USA…”) appears to be that Prof. R. is answering a question that is not the question that you want to ask. Fair enough. If you think patent health in the U.S. is the more interesting/important question, no one is stopping you from doing the research and generating your own chart. I expect that Profs. C. & R. would even be willing to let you guest-post it once you had done so.

      1. 5.2.1

        Hoisted by your own petard. “The total number of U.S. inventions” is not what this graph is showing.

        1. 5.2.1.1

          Right. This chart shows the total number of U.S. patents, which is a good and useful thing to know. I agree that it would also be good and useful to know how many U.S. inventions are made each year, and that this chart does not show that info. What’cha gonna’ do? There are only so many hours in each day, and Prof. R. chose to make one chart and not the other.

          It is a bit peevish to complain that Prof. R. did not make your preferred chart. If you want to see a different chart, you are free to make it (and share it here or elsewhere, as you wish).

          1. 5.2.1.1.1

            No Greg it is not “peevish” nor a matter of my preference.

            It is about being ethical and not presenting charts that you know will be misunderstood (as you misunderstood it Greg).

            1. 5.2.1.1.1.1

              Some will choose to not understand even the concerns expressed.

              Some – like examiner Ben – would find disdain that concerns are even expressed.

            2. 5.2.1.1.1.2

              This is a silly position to take, NW, and you should feel silly for taking it.

              1. 5.2.1.1.1.2.1

                It is not silly at all, and the fact that you feel that way shows that you simply do not understand the larger picture here.

                Then again, your general views are not typically pro-patent anyway.

              2. 5.2.1.1.1.2.2

                Ordinary, silly? It is silly to object to data not being normalized on a blog about patents? It is silly to object to Mark Lemley presenting this data as an indication of the health of the innovation engine without normalizing the data?

                The only thing that is silly is wasting my time interacting with you.

                1. But he DOES reflect a certain win level of calculated — and purposeful — misinformation, so it IS worth noting this (even if he will not engage in an inte11ectually honest manner).

                2. anon, there is no doubt that the point is be disruptive and offer misinformation.

                  Writ large is that the issuing patents and patent applications are being used to say everything is fine. Look at Paul’s post above. And, yet, the numbers aren’t normalized and the number of these patents that are for US inventions is not given.

                3. Because dabbling in patent prosecution, you must also be a straw farmer.

                  This is a perfect example of your BSery that you just don’t like being called out on and will throw a fit when someone does. You want the data to be manipulated in a certain way to fit your agenda. Just admit it. You obviously think there is a narrow boundary around what is acceptable to discuss when talking about patents. You also obviously think that you are the final arbiter of where that boundary is. You throw a fit and call everyone a shill or unethical when they stray for the NW-proscribed talking points.

                  Feel free not to respond to me. I just want everyone else to know what your deal is.

  8. 4

    Is there a reason why ‘calendar year’ has been chosen (given that the Office operates on a non-calendar year)?

    Also, it may make a better shot of quarterly activity (and then trending the third quarter of each year….)

  9. 3

    At a minimum, you need to normalize the graph to U.S. population. Otherwise, the graph has very little meaning.

    1. 3.1

      Why? The total number of patents issued each year is a relevant thing to know, quite regardless of normalization to population. Incidentally, it is not as if U.S. patents are awarded only to U.S. people, so why should one normalize to U.S. population?

      1. 3.1.1

        I agree that “US only” is not the best normalization factor to consider, but you are definitely in the weeds to think that no normalization is necessary.

        1. 3.1.1.1

          Anon, it’s tough to know how to normalize this kind of data because Greg was right on that point. Statisticians could play with this data for days and never agree on the “best” way to analyze it. My thought is that normalizing the number of patents to world population, starting only when the PCT was fully implemented, might be interesting. Another possibility would be to normalize it to the combined population of the five (maybe 10 or 20) countries that receive US patents.

          1. 3.1.1.1.1

            Greg was only partially right — as noted; the view of “raw data” is as misleading as the suggested “US population only.”

            I will also disagree with the notion that normalizing is difficult.

            Sure, there may be no single “best,” but that is a false conundrum.

            One normalizes in view of what one is attempting to use the raw data for.

            But also, mere presentation of the raw data certainly needs caveats (see my first comment on this thread). This is because there has been so much (and for so long) anti-patent propaganda that the sheer size of the raw data may be misconstrued and insinuated that there are just too many patents, or that attacks on patents “really aren’t that bad.” Both of these propaganda points are of course false.

          2. 3.1.1.1.2

            How or whether one normalizes these data depends on what question one hopes to answer. If you want to know whether the rate of innovation is changing. On the other hand, if you were the CFO for the USPTO and you wanted to know how much you could hope to take in this year from maintenance fees on patents granted four years ago (and how much you could hope to take in next year from three-year-ago grants, etc.), then there is no need to normalize the data—total grants per year is enough info.

            Seeing as Prof. R. did not specify a particular question that he meant to answer, it is not meaningful to say that he should have presented these data in this way or in some other way.

            1. 3.1.1.1.2.1

              “How or whether one normalizes these data depends on what question one hopes to answer.”

              Ir’s downright embarrassing that this needs to be said in this community.

              1. 3.1.1.1.2.1.1

                It is downright embarrassing that the data is provided without saying so.

                Or do you want to continue to pretend that this type of data is not maltreated in the screed against patents?

              2. 3.1.1.1.2.1.2

                I think, Ben, that we are seeing here a live-action demonstration of the maxim that “you cannot shame the shameless.” Some folks will resist even the best efforts to convince them not to make fools of themselves.

                1. The above really more properly belongs as a response to 7.1.1.1.2.1.1, but that end of the thread has turned into sort-of a bad neighborhood—too much inanity up that way. I do not want to go back into that valley of folderol unless and until the sense:nonsense ratio improves.

        1. 3.2.1.1

          GDP falls to the same weakness that YOU pointed out with population – it’s a measure related to a single country, and the GDP of other nations certainly has an effect (no doubt reciprocal) on those in other nations filing in the US Sovereign.

    2. 3.3

      Once one starts normalizing….because we “normalize” to add enough context to the data to either ascertain the health of the patent system, or the economy, or both. Not that the latter matters so long as patents are OK.

      Maybe include a measure of normalized GDP for patent per unit of economic output.

      And surely it would be interesting to know the relative numbers of compositions, manufactures, machines & processes.

      So we could take a look at the ratios when America was great, and discount the inherently greater ease with which information inventions can be varied.

      What would be incredibly useful to know, but virtually impossible to capture, is how often inventors use patent disclosures by other inventors to seed yet newer inventions, because that would be a great signal of a healthy patent system.

  10. 2

    I wonder how the chart would look if you could factor out grants of continuation/divisional applications.

    1. 2.1

      This was my question, too. Especially since patents in certain art areas no longer take 3-7 years to obtain, keeping a patent family alive requires an increasing investment in cons and divs.

      1. 2.1.2

        “patents in certain art areas no longer take 3-7 years to obtain”

        Nor to reject.

        Nor to enter the innovation-killing jaws of the PTAB Death Squad.

        Nor to thereafter be rubber-stamp kill-confirmed by the “We’re a Death Squad, too!” CAFC.

  11. 1

    Do you also share a running list of caveats?

    Like, for example, the usual comments when you post graphs like this?

Comments are closed.