A Blog by Jonathan Low


Jan 2, 2018

The Reason So Many Bogus Patents Are Approved

Failures not of intellectual or technological capability, but of organization and process. JL

Timothy Lee reports in ars technica:

Why don't more low-quality patents get rejected? The United States Patent and Trademark Office (USPTO) is funded by fees—and the agency gets more fees if it approves an application. Unlimited opportunities to refile rejected applications means sometimes granting a patent is the only way to get rid of a persistent applicant. Patent examiners are given less time to review patent applications as they gain seniority, leading to less thorough reviews.
If you've read our coverage of the Electronic Frontier Foundation's "Stupid Patent of the Month" series, you know America has a patent quality problem. People apply for patents on ideas that are obvious, vague, or were invented years earlier. Too often, applications get approved and low-quality patents fall into the hands of patent trolls, creating headaches for real innovators.
Why don't more low-quality patents get rejected? A recent paper published by the Brookings Institution offers fascinating insights into this question. Written by legal scholars Michael Frakes and Melissa Wasserman, the paper identifies three ways the patent process encourages approval of low-quality patents:
  • The United States Patent and Trademark Office (USPTO) is funded by fees—and the agency gets more fees if it approves an application.
  • Unlimited opportunities to refile rejected applications means sometimes granting a patent is the only way to get rid of a persistent applicant.
  • Patent examiners are given less time to review patent applications as they gain seniority, leading to less thorough reviews.
None of these observations is entirely new. For example, we have covered the problems created by unlimited re-applications in the past. But what sets Frakes and Wasserman's work apart is that they have convincing empirical evidence for all three theories.
They have data showing that these features of the patent system systematically bias it
in the direction of granting more patents. Which means that if we reformed the patent process in the ways they advocate, we'd likely wind up with fewer bogus patents floating around.

The fee system creates perverse incentives to approve patents

Three revenue sources pay for about 85 percent of the USPTO's patent operating budget, according to the researchers. There's an application fee paid when an application is filed. There's an issuance fee paid when a patent is approved. And there are renewal fees paid during a patent's term. If fee revenues fall, the patent office can face a budget shortfall.
The issuance fee creates an obvious potential conflict of interest for patent office administrators. When money is tight, the USPTO has an incentive to grant more patents to get more issuance fees. And in a 2013 paper summarized in their new report, Frakes and Wasserman found empirical evidence that this is exactly what happens.
The researchers used freedom-of-information requests to gather data about patent office decisions on millions of patent applications filed between 1983 and 2010. This large data set allowed them to compare the patent office's behavior before and after 1991—the year Congress first made the patent office dependent on fees to fund its patent operations.
Obviously, it would be too simplistic to merely compare acceptance rates before and after the change, since the patent office might have been allowing more patents for reasons unrelated to the new fee system. Instead, the researchers did something more sophisticated: they compared changes for patent categories that generate a lot of fee revenue to categories that are less lucrative for the patent office.
One category they considered is patents held by large companies. Patents filed by large entities are lucrative for the patent office because large companies pay the full fee while smaller companies get discounts on patent fees. The other category the researchers studied is patents from technology categories—like semiconductors and communications equipment—where patents get renewed (generating renewal fees) more often than average.
"Our findings suggest the PTO is preferentially granting patents on technologies with high renewal rates and patents filed by large entities," the researchers wrote in 2013. The researchers found that this effect was strongest in years when the budget office was facing the prospect of a budget shortfall. When money was tight, the patent office started becoming relatively more generous to bigger companies that paid higher fees—relative to other companies and individual investors who paid less. When the budget picture improved, this bias diminished.
To be clear, this doesn't prove that the patent office ever had a deliberate policy of favoring larger entities or technologies with high renewal rates. It's more likely that subtle—perhaps even subconscious—pressures explain these shifts. But it does suggest that insulating the patent office from these financial pressures could lead to a more rigorous patent examination process.
Frakes and Wasserman recommend eliminating the issuance fee and boosting the application fee by a corresponding amount. That way, the patent office would get the same revenue whether it approved a patent or rejected it.
Of course, renewal fees create similar problems, but the authors don't favor eliminating those because they encourage patent holders to let low-quality patents expire. Instead, Frakes and Wasserman argue that all renewal fees should be put into a special fund that would be used to rebate some of the fees of small and medium patent applicants.
This would allow the patent office to charge everyone the same fees—eliminating the incentive to preferentially grant patents to larger entities. However, smaller companies and individuals would get some of their fee revenue back, helping to ensure that smaller players are not priced out of the patent system. In this world, revenue shortfalls would lead to smaller entities paying modestly higher fees than they otherwise would—a better outcome than making the patent office financially dependent on large companies.

Applicants have unlimited opportunities to refile patents

When you submit a patent application, it goes to a patent examiner, who studies the patent and then does research to decide if it meets legal requirements like novelty, non-obviousness, and adequate disclosure. If the examiner believes the patent meets the legal requirements, she approves the application and it becomes a patent. If she believes it falls short, she notifies the applicant that the patent has been rejected.
The applicant then has the option to try again. He can modify the application to try to address the examiner's objections and submit it again. Re-applying does generally involve paying the same fees as the original application. But because these fees don't come close to covering the USPTO's costs for considering the application, the agency winds up losing money every time a patent applicant requests another look.
This gives the patent office a financial incentive to simply throw up its hands and grant patents to particularly persistent applicants.Once again, the researchers find evidence of this effect. They looked at how the patent office handled repeat patent applications, comparing technology categories where re-applications were common to those where they were uncommon.
Our findings suggested that when the Patent Office begins to face mounting backlogs, it appears to act on its incentive to grant patents at higher rates for technologies that are associated with higher rates of repeat application," the pair write, drawing on research they published in 2015.
The researchers point to a simple solution: only allow applicants to refile an application once. If an application is rejected once, the applicant would be able to modify the application and submit it again. But if it was rejected a second time, that rejection would be considered final.
Under their proposal, applicants would still have the option to appeal a rejection to the courts—so bad decisions by the patent office could still be challenged. But if the courts agreed with the patent office, that would be it.

Senior examiners don’t get enough time to examine patents

Patent examiners are part of the federal civil service, and they gain seniority over time, rising from an entry-level rank of GS-7 to a senior rank of GS-14 over the course.
of their careers, getting raises along the way. The agency assumes that examiners will get more efficient as they gain experience, so more senior examiners are expected to process more patents per hour than entry-level ones.
You can think of examiners as doing research to try to find a basis for rejecting the patent. If the examiner can't find a good reason to reject the application, then it gets accepted. The longer they have to work on each patent, the more likely they are to reject it.
Frakes and Wasserman obtained a data set that showed the individual patent decisions of a large number of patent examiners. They found that within a single rank, examiners' acceptance rates tended to go down over time. This makes sense because examiners were getting better at their job, which made it easier for them to find grounds for rejection quickly.
When examiners got promoted, they were given less time to work on each patent. As a result their acceptance rate temporarily rose. This, too, makes good sense—since less time to look for problems with the patent makes it less likely that they'll find issues that could be grounds for rejections.Ideally, you'd want these two factors to roughly balance out over time—with senior examiners getting less time per patent but still enough time that they're able to reject
patents at about the same rate as their more junior colleagues.
But Frakes and Wasserman find that despite their greater skill and experience, more senior examiners approve patents at a higher rate. That suggests the patent office is systematically giving more senior examiners too little time to do their job. Rushed for time, senior examiners are forced to cut their research short and approve patents that they would have rejected if they had done more thorough research.
The solution, they argue, is to give senior examiners more time for each patent. Not as much time as their junior colleagues—patent examiners really do get better at their jobs over time. But enough that applicants' odds of getting a patent are about the same whether it's reviewed by an experienced examiner or a green one.
A broader question to ask here is whether the patent system might be skimping on patent examinations across the board. We know that senior examiners approve patents that their more junior colleagues would have rejected based on more thorough research. But maybe even junior examiners are approving a significant number of patents that they would have rejected if they had time to do more thorough research.The reason patents don't get more scrutiny, of course, is that it would cost more money. The USPTO has a limited budget that only allows it to hire enough examiners to look at a typical patent for around 19 hours. Increasing this number would either.
mean raising patent fees—potentially pricing ordinary inventors out of the patent system altogether—or else using general tax revenue to fund the patent system rather than insisting that the system be user-funded.
While a user-funded patent office sounds good in theory, the reality is that the social costs of bad patents are high—and they're borne by the general public, not patent holders. One study a few years ago estimated the annual cost of defending against patent lawsuits at $29 billion—an order of magnitude more than the Trump administration's $3.5 billion budget request for the patent office. So insisting that the patent system be self-funded might actually be penny-wise and pound foolish.
We would love to know what officials at the USPTO think of Frakes and Wasserman's research, but unfortunately a USPTO spokesman said in an emailed statement that the agency had no comment.


Lan Thu said...

Thanks for sharing, nice post! Post really provice useful information!

Máy đưa võng tự động là quà tặng tuyệt vời giúp bé yêu ngủ ngon giấc, là phương pháp ru con thời hiện đại của các ông bố bà mẹ bận rộn. Hệ thống website máy đưa võng TS của An Thái Sơn gồm: Mayduavong.biz, Mayduavongtudong.org, Mayduavongts.comTreem.org giúp bé ngủ ngon.

Post a Comment