How Patent Races Impact Innovation

By Neil Thompson
July 6, 2020

“This first broad-based view of patent racing in the real world affirms that it is prevalent and important for firm innovation.”

When it comes to patents, timing matters. If two inventors are working on the same invention, it is the one who races to the patent office first that gets the brass ring. Perhaps the most famous example of a “patent race” is when Alexander Graham Bell and Elisha Gray both filed a patent for the telephone on the same day. Bell won the patent, started a successful company, and is now synonymous with the telephone, while few people remember Gray.

For years, economists have used patent races as the quintessential example of how firms innovate in the presence of competition. But these discussions are usually only grounded in theory. To gain more insight on how this plays out in the real world, Jeffrey Kuhn and I created the first way to identify patent races—and we used it to analyze the effect of patent racing on innovation.

Winning the Race

The key to our method is that the patent office keeps applications secret for a while after they are filed. This means that two companies may end up filing patents for the same invention, not realizing that the other also filed. After some digital sleuthing, we were able to pick up the paper trail from these cases to see how often this type of neck-in-neck innovation happens and what the benefits are from being the winner.

It turns out that these races have a significant impact on innovation both in direction and magnitude. Protected by their patent, winners do 14% more follow-up innovation. In contrast, losers are almost three times as likely to abandon their invention. The ones that do keep going have to “invent around,” finding alternate technical paths to avoid the patent coverage of the winner.

Given the importance of winning a patent race, it is not surprising that firms in hotly contested technology areas—where patent races are frequent—do more R&D and much more patenting. We saw 53% more patenting among these firms. They also seem to patent in smaller “steps,” rather than waiting until bigger milestones are achieved.

Who’s Running?

We also found that patent racing behavior is surprisingly frequent, with 10-11% of all patents in races. In some technology areas, patent racing is even more frequent. For example, 16% of the patents in computing are in races and 13% in communications. In comparison, only 5% of patents in biotechnology are in races.

There are many possible reasons for the frequency of patent racing in those areas. Patents in computing and communications are usually considered weaker, whereas those in biotechnology are stronger. Another explanation is that firms in areas like biotechnology might actively avoid patent racing by buying up competitors or exiting research areas in order to reduce the risk of paying for expensive medical trials and ending up without patent coverage.

What it Means

This first broad-based view of patent racing in the real world affirms that it is prevalent and important for firm innovation. Winning a race protects the victor, allowing them to push their research agenda forward. At the same time, it deflects the loser, leading them to do less follow-on work and having to “invent around” the winner.

Image Source: Deposit Photos
Author: Elymas
Image ID: 39977065 

The Author

Neil Thompson

Neil Thompson is Innovation Scholar at the MIT Sloan School of Management and the MIT Computer Science and Artificial Intelligence Laboratory. He is a coauthor of “Does winning a patent race lead to more follow-on innovation?”.

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Discuss this

There are currently 14 Comments comments. Join the discussion.

  1. anonymous July 6, 2020 9:30 pm

    A jet ski and an aircraft carrier at port decide to race to a buoy two miles out to sea. The race begins and before the aircraft carrier has moved more than 25 feet from the pier, the nimble jet ski is 500 yards out. In less than 10 minutes, the jet ski has won the race and is tied off at the buoy, proud of his victory. The race has been won.

    Some time later, the aircraft carrier is seen bearing down on the buoy, the jet ski and the jetskier. A half mile out, the carrier looks huge. A quarter mile out, even bigger. 500 yards out and there is panic by the jetskier who thought the race was over. The jetskier knows the aircraft carrier is not slowing down, won’t turn, and cares not if it sinks the jetski and leaves the jetskier stranded at sea, about to drown. It turns out survival wasn’t about the race, but the momentum one’s pirate ship has when it gets to the buoy.

    We all know what happens next, when an Efficient Infringer like Apple or Amazon captains the aircraft carrier-sized pirate ship. Start-ups and individual inventors who won the race are not protected. Courts won’t enjoin the collision, even knowing the jetski/start-up will be sunk. We’ll just pay for the cost of a jetski, they say, ignoring that they’ve put the jetskier in the water, struggling for his life.

    I don’t know what “real world” you envision where winning the race is enough to protect the victor. In the real world, Efficient Infringers don’t bother inventing around a patent owner as though he’s the “winner”. Willful infringement is the business model for Apple, Amazon, and many others. They don’t care about competing fairly, period. They sail the largest pirate ship and take what they please. Perhaps it is time to step away from the “regression discontinuity approach” and talk to some real inventors.

  2. MaxDrei July 7, 2020 4:32 am

    Dear anonymous, I’m not sure your analogy is apt. The paper is about races within the 18 month Patent Office “black hole period” between filing and A-publication, in which none of the “racers” even knows whether there is a race at all, never mind anything about the characteristics of any other party they might be racing with. So how can they agree with each other to race?

    I suppose what you mean is that society agreed to such races when it switched to First to File. But the Paper points out that even in the days of First to Invent, it was de Facto a race to the PTO because even then the “spoils” (enforceable exclusive rights) nearly always went to the party swiftest to file at the PTO. BTW, the Paper finds that more than 10% of all patent filings are participants in a race (that will become evident only after the A-publications emerge). I realise that interferences were rare but clashes between racers during post A-publication PTO examination are (or ought to be) extremely common. It was ever thus, at the EPO, for example, where by now the law on deciding what is the “same invention” is compendious, well-established and very sophisticated. Premature and speculative first filings often end in tears. just like Goldilocks, you have to get the moment of first filing just right (not to early but not too late either) as Regeneron has learned, just last month, when the Supreme Court of the UK revoked its patent for insufficiency of disclosure.

    And now to the notion of efficient infringement. In my youth, clients (here in Europe) sometimes asked whether it was worth filing behind the Iron Curtain, where the government did what it pleased and the courts did what they were told. The answer though applies universally: The only thing you can do to prevent being ripped off is to get a patent. If you don’t file, you cannot get a patent and can have no complaint about being ripped off. If you do get a patent, you might still get ripped off, but at least you can protest about it. It is not unknown for Little David to win in court against Giant Goliath now, is it? Against the Goliaths of the corporate world, all that David needs is a fair system for the administration of justice, and enough resources to petition that system and get help from it. That should not be beyond the ability of a decent society to organise for David, should it?

    BTW, if it is as you assert, that Little David on his jet-ski routinely files at the PTO earlier than Goliath on board his aircraft carrier, be happy that the turbo-charged FtF law in the USA has a characteristic, unique in all the world, that delivers a mind-boggling advantage to the Party that files earliest. I speak of that first filing being available under 35 USC 103 to wipe out all later filed patent applications, even when they are not even for the same invention. Be thankful for that, and instead campaign for reform of civil litigation in the USA.

  3. Anon July 7, 2020 10:37 am

    MaxDrei,

    You ask seemingly simple questions vis a vis Efficient Infringement, but I have to wonder how very deaf you have been (for quite some time) on the plain fact that Efficient Infringement is a very real problem.

    Given the existence and extent of Efficient Infringement, how would your own answers to your own questions help anyone?

  4. MaxDrei July 7, 2020 12:21 pm

    I have already suggested how the unattractive aspects of “efficient infringement” can be curbed, namely by imaginative Rules of Civil Procedure that enable the courts to deliver the over-riding objective of doing justice, in particular between parties, whether tiger sharks and minnows, engaged asymmetrically in civil litigation. See England and Germany for examples, very different from each other but both addressing the same issue in accordance with their sovereign judgement. Where there is a will there is a way. Is there a will, in the USA? If yes, get on and reform your own Rules of Civil Procedure in the courts. If England can do it, so can you.

    Meanwhile, reverting to Comment #1 in this thread, there is, presumably, inside “the Homeland”, a strong and concerted, nationalistic, patriotic “America First” wish that every American “aircraft carrier” shall prevail against every adversary, by fair means or foul, whether that adversary is a jet ski or an aircraft carrier and regardless what flag it happens to be flying from its masthead. It’s a matter of national survival isn’t it?

  5. Anon July 7, 2020 4:55 pm

    MaxDrei,

    Perhaps you have some particulars of these “civil procedures” (and not just glad handling)…

  6. Anon July 7, 2020 10:17 pm

    To the point of the article here, there is a small — but not insubstantial — blindspot.

    We have a (more or less) default advice for entities seeking protection only within the US Sovereign to file non-disclosure requests as a matter of first recourse.

    I wonder how the authors take such a thing into account (if at all).

  7. MaxDrei July 8, 2020 2:32 am

    The Article (going by the summary here) is replete with references to the activity of “inventing around” that to which the first filer has laid claim. Is “inventing around” something of benefit, something which happens the sooner the better, something therefore to be encouraged, or to the contrary something that ought to be handicapped?

    I ask because putting a block on the emergence of the A-publication slows down and handicaps the activities of “inventing around”. The patent system is supposed to promote, not hinder, the speed of progress in the useful arts.

  8. Anon July 8, 2020 8:37 am

    MaxDrei,

    Your question at 7 has been addressed previously and hinges on what is the actual Quid Pro Quo offered by each Sovereign.

    In the US, the Quid Pro Quo is NOT ‘publication for mere chance at a patent.’

    Instead, our Sovereign choice is (and always has been, subject to more recent accommodations for multiple Sovereign attempts at protection) that the Quid Pro Quo is publication in exchange for the actual grant of a patent.

    One should not gloss over this difference (as you personally so often do).

    Such is not respectful of the differences in Sovereign choices.

  9. Benny July 8, 2020 4:26 pm

    Max (at 7), “inventing around” generally involves an inferior solution to the problem at hand than the original invention, thus does little to “advance the arts”. It merely gives the inventor a chance to grab a slice of the market, albiet with an inferior product.
    Anon – that is not how I understand the Quid pro quo: I understand it to be a patent in exchange for publication IF, and only if, the applicant is the first to disclose the invention, and the invention is not obvious. This is often not the case.

  10. MaxDrei July 8, 2020 4:43 pm

    Well Benny I’m glad we agree, that i) inventing around does deliver a different solution to a real technical problem and ii) in that way it does both enrich and advance the art. As to which of the in consequence available solutions is the “inferior” one, often only time will tell. What often happens is that the plurality of solutions stimulates further inventive activity ad the emergence of an even better solution. In the field of batteries for electric vehicles, for example, that’s what I expect will happen.

    And as to the Quid pro Quo you are again right: exclusive rights but ONLY in return for giving the public something it did not have already, namely, all the old stuff and all the stuff that was obvious to the skilled person. If that’s all you’ve got, your A publication gives the public nothing it didn’t already have, so when it publishes you’ve lost nothing that was yours to give away.

  11. David Boundy July 9, 2020 11:30 am

    Dear MaxDrei (and Prof Thompson, if you’re here) —

    You’re entirely missing the point. Patents are important when they’re important, and irrelevant when they’re irrielevant. Patents are irrelevant when the sizes of the companies are highly disparate. To the big tech companies (Google, Microsoft, Intel, Cisco, Amazon, and to a lesser degree Apple), whose market protection comes from network externaliities and similar competitive forces, patents don’t matter. Indeed, to the extent patents ever enter the attentional horizon, these companies are always on the other side of the “v.”–patents are a nuisance.

    anonymous @ 1 painted the picture accurately, although didn’t quite tie up the analogy tightly enough so that all could see it. The jetski won the patent race. No question who won that race. The problem is that when patents are irrelevant, they’re irrelevant, and winning the patent race is irrelevant. If an aircraft carrier is intent on invading the space that should, legally, be protected by the jetskier’s patent, the aircraft carrier gets the space. The jetskier drowns and gets chewed up in the propellers. Patents are no match for market dominance.

    Between competitors of similar size, yes, I am sure patents and patent races matter. But not in the race between a new-market-entrant vs $100billion market incumbent.

    20 years ago, our legal system was an effective counterbalance to size. Stacker v Microsoft was one example, But over the 2010s, Google/Intel/Microsoft etc so weakened the patent system that territory is won and held by being an aircraft carrier. In a big-guy-little-guy competition, patents and the patent race are irrelevant.

    David

  12. MaxDrei July 9, 2020 5:52 pm

    Mr Boundy, I see your point, that when

    “….market protection comes from network externaliities and similar competitive forces, patents don’t matter.”

    You say that:

    “……over the 2010s, Google/Intel/Microsoft etc so weakened the patent system that territory is won and held by being an aircraft carrier”

    but I wonder whether you are right, that the patent system has been weakened. Perhaps the reason patents don’t matter any more is the rise in importance of those “network externalities”?

    I say that because I think that, at least in many fields other than IT, patents do still matter.

  13. Anon July 9, 2020 5:54 pm

    Mr. Boundy,

    You describe the results of the Efficient Infringer efforts.

    Some, like MaxDrei, can not (read that as chooses not) see this, and ever wonders why people are upset.

  14. Anon July 9, 2020 5:59 pm

    Benny (and time some extent, MaxDrie),

    You miss the point.

    Of course, an application still must earn allowance (and meet all the legal requirements).

    But the point is (still, with the election of the non-publication option) that the applicant has ENTIRE control as to IF publication occurs.

    I know if cases that even after allowance, a client has decided to abandon a matter, as their calculations showed that the final claims were less valuable than maintaining the item as a trade secret.

Post a Comment

Respectfully add to the discussion.

Name *
Email *
Website