This is a linkpost for https://www.bhauth.com/blog/institutions/patents.html
The FDA used to have a long backlog of drug approval applications due to understaffing. This problem was eventually addressed by the Prescription Drug User Fee Act, which established filling fees for drug approvals and a deadline to review new drug applications. The filling fees were a way to make the cost of additional staffing politically palatable, and are much less than the costs of approval delays.
Most of the cost of patents usually comes from patent lawyers, anyway. Yes, the US patent office will probably have to raise its fees at some point, but it's also had trouble hiring enough examiners, and not just because of the salaries offered.
I think all of these problems are minor compared to the underlying difficulties that patents inherently bring.
There are industries and equilibria where patents are the least-bad workaround for the broken systems around everything. Especially in old-school process patents, where HOW to do something is the hard part. But the VAST majority of patents should go away. Most ideas and designs are the easy part nowadays, the hard part is operations and regulatory compliance.
If you want to argue against the very existence of the patent system, you should make your own post for that. This isn't the right place for that.
Since I have a whole pile of technology I designed (with my blog posts largely being an occasional byproduct of that) a few people have asked me why I haven't patented lots of things. The main answer is, it's usually just not worth it. I also like designing technology more than I like dealing with the patent office.
Would you patenting more things actually lead to you doing something to turn the patented inventions into products that you currently aren't doing?
If it wouldn't and it would just allow you to sell your patent to some patent troll who then uses the patent to sue people who independently invented the same thing, that might overall have a bad effect on innovation.
Would you patenting more things actually lead to you doing something to turn the patented inventions into products that you currently aren't doing?
That's not the right question, because I'm not the one with the power to turn them into products. There are people who have power - money, executive positions in companies, government positions, media influence, etc - that could be used to accomplish (part of) that. The question is whether it would lead to them doing something to turn the patented inventions into products.
I'm not sure what your model here happens to be. I would expect that at a minimum you would need to lobby people with power to turn patents into products for that to happen.
From the progress study perspective the key question is whether if you make it easier to get patents and sue people with those patents, you will get less or more or new products.
Both are plausible outcomes and the effects are hard to know. Just pointing to the fact that life would be easier for some people in the ecosystem if it would be easier to get patents for them and sue people, does not automatically mean that this would be good for progress.
It's not about making it "harder" or "easier to get patents". That's not what increasing the time that reviewers have to look at patents does.
There's also a distinction between reducing the procedural difficulty of applying for patents and reducing the "real" difficulty of making a legitimate novel invention. You don't seem to be making that distinction, but surely you agree it's relevant?
If you reduce the backlog of patents, it's "easier to get patents" and people are likely going to fill more patents.
I do agree that having a single reviewer spend more time to review a patent doesn't make it harder or easier to get a patent.
If you worry about effects such as different reviewers making different decisions about when to accept or reject a patent, it would however be better to let every patent be independently reviewed by two patent clerks. If both vote approve, the patent gets approved. If both vote reject it gets rejected. If they differ a new patent clerk gets to be tie brackets.
Then you give bonuses to the patent clerks that least often need a third patent clerk as a tiebreaker and fire those patent clerks that need the most tie brackets.
Going from the current system to one where every application gets two reviewers isn't easy but wouldn't be impossible either.
One way would be to allow a process, where an inventor that request a patent can pay $10,000 to get two reviewers and jump ahead of the one-reviewer patent queue. At the end of each calendar month when the two-reviewer process takes longer than a month to reply to a patent application the price goes up by 10%. If the queue is empty the price goes down by 10%.
The money then goes to the patent office to hire more patent clerks. Maybe +200% fee would be better than a blanket sum given that different patent claims have different complexities.
There's also a distinction between reducing the procedural difficulty of applying for patents and reducing the "real" difficulty of making a legitimate novel invention. You don't seem to be making that distinction, but surely you agree it's relevant?
If you are making a "legitimate novel invention" and do nothing to produce a product but sue all the people who make the same legitimate novel invention after you, I think that's still a drag to progress overall.
If you reduce the backlog of patents, it's "easier to get patents" and people are likely going to fill more patents.
Well, if you reduce the backlog by rejecting far more patents (especially those that aren't demonstrated as feasible), then maybe people won't file more.
Yes, it's possible to reduce the backlog that way but it wasn't the strategy proposed in the OP.
Also, there are big problems with the idea of patents in general.
If Alice and Bob each invent and patent something, and you need both ideas to be a useful product, then if Alice and Bob can't cooperate, nothing gets made. This becomes worse the more ideas are involved.
It's quite possible for a single person to patent something, and to not have the resources to make it (at least not at scale) themselves, but also not trust anyone else with the idea.
Patents (and copyright) ban a lot of productive innovation in the name of producing incentives to innovate.
Arguably the situation where innovators have incentive to keep their idea secret and profit off that is worse. But the incentives here are still bad.
How about
That has all been considered extensively before and this post isn't a good place to discuss it. Prizes have been found to be generally worse than patents and research funding.
The current patent process has some problems. Here are some of them.
patenting is slow
The US Patent Office tracks pendency of patents. Currently, on average, there's 20 months from filing to the first response, and over 25 months before issuance.
That's a long time. Here's a paper on this issue. It notes:
A patent application backlog means it takes longer before work is published and other people can potentially build on it. It also means a longer period of companies having uncertainty about whether a new product would be patented.
examiners are inconsistent
Statistical analysis indicates that whether or not your patent is approved depends greatly on which examiner you get. This article notes:
Perhaps applicants and examiners should both be allowed to get a second opinion from another examiner on certain claims. But of course, this would require more examiner time in total. This situation might also indicate some problems with the incentive structure examiners have.
patents take effort
A lot of smart people who work on developing new technology spend an enormous amount of effort dealing with the patent system that could be spent on research instead. Even if the monetary costs are small in comparison to the total economy, they're applied in perhaps the worst possible places.
There are many arcane rules about the format of documents for the patent office. Even professional patent lawyers get things wrong about the formatting and wording, and that's their main job. LLMs do quite poorly with that, too. Even I've made mistakes on a patent application.
The US patent office does do most things electronically now. Its website is at least technically functional. Considering that it's a US government agency, I suppose it deserves some praise for that. However, I'd argue that if correctly submitting documents is a major problem and even professionals sometimes get it wrong, that's a sign that the format is badly designed and/or the software used is inadequate. For example, their website could theoretically remind people when required forms in a submission are missing.
Currently, the US patent office is trying to migrate from pdf to docx files. Maybe that's an improvement over using Adobe Acrobat to fill pdf forms, but personally, I think it should accept:
I used to say Powerpoint was bad and maybe companies should ban it, and business-type people explained why that was really dumb, and then Amazon did that and it ultimately worked well for them. The problem Amazon had to solve was that most managers just wouldn't read and understand documents and long emails, so when banning Powerpoint, they had to make everyone silently read memos at the start of meetings, and they lost a lot of managers who couldn't understand things they read. At least the US patent office people have the ability to read long documents, I guess.
international patents are hard
If you get a patent in the US or EU, that's not valid in other countries. Rather, the PCT gives you up to 30 months from your initial application to apply for patents in other countries, and most countries require their own application. Even within the EU, a "unitary patent" still isn't valid in every EU country.
If you read the earlier sections, you might notice that 30 months is not much longer than the average pendancy before approval in the US. For many US patents, the application process in many other countries would have to start before the applicant knows if the patent gets approved in the US or not.
Sure, sovereign countries shouldn't have to automatically respect every patent that one country was willing to approve, but the current process is a lot of effort. The amount of purely bureaucratic work generated probably costs tens of billions of dollars a year. Maybe some treaty could be negotiated such that any patent approved by any 3 of (the USA, the EU, Canada, Australia, China, Japan, South Korea) and not applied for and rejected by any of those countries would automatically be respected by other countries, but it wouldn't be easy.
courts
lawsuits are expensive
What a patent does is give you the right to sue people for infringing it. Doing that requires lawyers, and lawyers are expensive.
lawsuits are slow
Conflicts over patents get resolved with lawsuits, and those lawsuits take a long time to resolve. Much of the time it takes for lawsuits to finish comes from courts having a backlog of cases.
This problem doesn't just apply to patents, of course, but patent cases always go to federal judges, and the US has a lot more state judges than federal ones. The US no longer has enough federal judges for its population. The average federal criminal trial now takes over 2 years to resolve. That's bad.
judges are appointed by politicians
If you want to get good judges, you have to elect good politicians.
Voting is individually irrational, but I'm smarter and better-informed than average, so I sometimes vote as a sort of public service, or maybe just an attempt to understand and participate in society. Whenever I've voted in an election, whether a primary or general election, I've briefly looked into the main candidates online, for every race I voted in, before voting. I've spent a lot of time searching the internet, and I've gotten rather good at it.
And...in significant elections with 2 main candidates, overall, the candidate I voted for has won less than 50% of the time. You might suspect that this is because I have some sort of uncommon ideology, but there's been a wide variety of reasons for that difference in voting between me and the median voter. Most of the time it wasn't because of a value difference, but because of something that most voters didn't know about or understand. From my perspective, I suppose this makes current US elections somewhat worse than a random number generator.
I think people are generally able to accomplish things they try to do better than random chance, or at least equally well. The only explanation I've found is that media has negative information value about politics. I said "significant elections" above - in elections for something "nobody cares about" like a treasurer or surveyor position, the candidate I voted for has been much more likely to win. The difference seems to be the amount of media coverage.
most judges don't understand technology
Patent cases tend to be very technical. Federal judges are specialized in law, not the particular area of science/engineering that a patent case involves. Yes, some judges make a good effort towards learning about whatever their current case involves, but mostly, judges just trust whatever paid expert witnesses say, and those expert witnesses will sometimes just lie. I think having some amount of scientific judicial specialization would make sense.
Again, this obviously isn't only a problem for patent cases, but patent cases are more technical than most court cases.
conclusion
Sometimes I see people talk about "progress studies", saying that even a small increase in overall technological progress is hugely valuable, so society should try harder to make that happen. Usually their recommendation is that the world should do whatever that person wanted to do anyway. What I don't see them do is talk to the people who actually research or implement new technology and listen to what they say their problems are. If they did, they'd hear that the current patent system is a big problem, and not just in simplistic ways like "patent terms are too long" or "patent terms are too short".
Since I have a whole pile of technology I designed (with my blog posts largely being an occasional byproduct of that) a few people have asked me why I haven't patented lots of things. The main answer is, it's usually just not worth it. I also like designing technology more than I like dealing with the patent office.
The US patent office should have more examiners. Its backlog is large and its examiners should spend more time on patents. Doubling the number of US patent examiners might cost an extra $1 billion a year, but that would probably be worth it.
The US should have more federal judges. Politically, that's difficult to accomplish, but it's still true.