Self-Funding Harberger Taxes
gwern.net
gwern.net
The idea is that you are taxed based on your own estimate of the value of a thing, with the estimate being public and binding as a sale price.
This sounds incredibly simple, with a natural mechanism of enforcement; if you undervalue a thing to pay lower taxes then you risk losing it at that price.
But I don't think this would work in reality. Pricing things is super hard. In a very liquid market it is difficult but doable, in an illiquid market (of one non-fungible asset) it is practically impossible. At what cadence do you update the estimate? How long is it binding for? If the value suddenly changes, how long a grace period do you have before you have to update the price?
You can try to patch this by saying that you have a right of first refusal -- that if someone offers to buy the asset at the price then you can accept their valuation as the new estimate of the price, and pay taxes going forward on the new value.
But then this defeats the point of the process because effectively the valuation is non-binding. So you have to put in all sorts of structures to prevent abuse -- maybe levy fines for past underpayment when the pricing is adjusted, or limit the number of re-pricings per time period.
And it incentivizes nuisance bidders, because owners will defend their property. So you have to put in restrictions like forcing escrow before the sale, or limiting bidding by a single entity.
Not to mention the difficulty in maintaining this registry of ownership of these assets, along with how to deal with more complicated deals about the rights involved -- film rights vs. tv rights vs. print rights, rights to individual characters, temporary licenses, etc., and all of that needs to be mechanized to allow a certain fixed list of the types of rights that can be granted, which right now is as extensive as a contract can contain -- if Sherlock Holmes were still in copyright, what if I just wanted to license Watson or make a film about Moriarty?
This is all just hopeless technocratic pie-in-the-sky dreaming.
If you own 80% of something worth $1B - you have $800M. If you can't sell $2M at your preferred price to pay your taxes, then it's not worth $1B, and you wouldn't have to pay that much tax anyway.
But the whole point is that you don't _want_ to sell it. And it's entirely possible to own something that is work more than 20x (given a 5% tax rate) than you currently have in accessible cash.
For example, assume you write a story and MegaCorp decides they want to make a movie of it. Right now, they need to pay you to license it. With the Harberger tax scheme, they can force you to sell it to them; because you don't currently have the money to pay taxes on what it's worth to them.
If someone just wants to create something they like, and then own what they created, a big corp or rich jerk could easily bankrupt them through the tax/force them to sell it off.
Some things don't need to exist in the context of a market and markets are often less efficient than folks seem to assume.
1. Private — untaxed, lasts for 14 years (or until the death of the last author), goes into Public Domain. No need for registration; and,
2. Public — any work (before it's expired) an be transferred to a Public work. At that point some sort of Harberger tax scheme would kick in.
Obviously, picking a number like "14" distorts the market — works that take off after 14 years are going to be undervalued, which sucks. OTOH, the point of copyright is that the opposite distortion — free-for-all — is considered bad.
Most people don't even correctly understand how progressive taxation works, so anything more complicated than that is dead on arrival.
Scenario: Your local town assessor hates $ethnic_minority and starts assessing everything in $minority_neighborhood at 5x its former value so that they get taxed to death.
Scenario: Your local town assessor hates rich people and starts assessing everything in $wealthy_neighborhood at 5x its former value so that they get taxed to death.
Scenario: Libertarian demagogue mayor decides to eliminate income and sale taxes, and this is his proposal to fund what few public services will remain (at a very low rate). It works, but having virtually no public services remaining has predictable effects and gets him thrown out of office. Rather than rebuild the old system (and start an interminable fight over the specifics in the legislature) the decision is made by the next administration to simply adjust the rate on the new one upwards by a thousand percent.
I'm not saying these are credible scenarios, but treat this like a novel where you've written the last chapter; What is the most plausible way that a polity got from here to there? It's a more circumstantial way of thinking - "How might this occur?" than planning hypotheticals, which work more like "How will I achieve this next year in our current universe using my own actions".
Same goes for the Harberger tax on copyright, which is basically "only people and corporations who are already rich deserve to extract market value from their creations"
"I can make your life miserable and I have enough money to do so" is something that can happen in the real world and that the average person doesn't want to happen. Tthe Statute of Anne was created b/c previously printers could essentially publish any work that wasn't already in publication by another house... which is in some ways closer to a Harberger tax situation, but was widely hated by everyone who wasn't one of the publishers who monopolized access to the printing presses.
the stamp act, the townshend act, the boston tea party, etc...
might be "tax what you want less of" could be... citizens.
If anything, product dumping (which happened to improve the lives of tea consumers, but hurt smugglers) was the motivation, not taxation. In that case.
You pay a flat percent of most purchases as a sales tax. It varies by geography, but most people spend their day to day life in a single place (their home town).
You pay a flat percent of your home as a property tax. Most people who own property don't move, so while it varies regionally, you don't have to think about any but yours.
You pay a percent of your salary as income tax (it's technically progressive, but i don't think most Americans really think about this conceptually outside of tax filing). Your employer takes out this for you, and then once a year you fill out some forms that seem scary, but for the average American are actually somewhat straight forward. The government writes you a check back if you over paid.
Then everyone talks about all the loopholes and stuff that rich people and businesses get. Most Americans never touch 99% of any of that, and just think it's a way for the rich to get richer. Some people talk about charity tax deductions, but most people don't know how that works, but also wouldn't really benefit from this. Similarly there is the vague idea of business write-offs that no one really understands nor needs to know about.
I'm outside the "average" nowadays, but I've been paying someone to do my taxes for me for decades because I don't have the mental bandwidth to figure out all the things that could save me money on my taxes. All the way back to when I was earning well below the poverty line (though I paid someone less back then).
That's it, that's the hack that solves the problem of all the weird deductions - reducing it to word problems that you can answer - there's definitely a few dozen of them though.
The DeepMind personality cloner just asks questions. It is definitely not simple.
At some point there is irreducible complexity, but most tax questions are of this nature for all normal people - a huge portion of people could file a form 1040ez and be done with it.
Most people should probably take the standard deduction, and not try to itemize their taxes? Most deductions require you to actually spend a certain amount of money to save anything, and the standard deduction is like 30k.
There are millions of permutations of different deductions, and that makes taxes seem scary, and society talks a ton about all these different rules, but I’m pretty confident that the average W2 American stays well within the happy-path of “employer takes a percentage of your income, and you ack that in April”
One can go too far the other way, though. Sometimes problems are both too big to ignore and don't have known, simple, efficient, and effective solutions. It's also easy to ignore efficiency or effectiveness in a drive for simplicity, and end up either not solving the problem or solving it at too great a cost.
That doesn't sound so simple to me, everything has to be for sale all the time and\or you over pay on your taxes. Value your vehicle at market rate to pay the "right" taxes, anybody in the market for a vehicle can come buy yours at $1 over market rate if they want to. (It costs 2$ to drive to the closest one being sold at market rate, so it makes economic sense.) The only way to even partially defend yourself is to value your property way over market rate and pay taxes higher than the market value would suggest. Basically you have to value your stuff at market rate plus how much I would be annoyed to loose said item.
Piss off a billionaire? Guess who is never going to live indoors again.
And even if my car is worth $5k, not having to sell it and buy another one (even if I can find it) is worth a premium.
It's likely that something like Harberger taxes could work in certain situations, but not normal people ones.
It also doesn't make the tax onerous; a revenue-neutral switch from our current taxation scheme to harberger taxes would be revenue-neutral, whether it was based on market value or personal value.
Like many people I'm taxed at source with minimal options to change my tax bill. I pay the going rate, as per the rules. Putting the obligation on me to do a bunch of work and worry about correctness would be a huge source of frustration and unhappiness.
If you mess up an income tax, it's easy to get your house taken away and auctioned, with $0 of the proceeds going to you. If you mess up a harberger tax, you get all the proceeds from the sale at a price which you literally just said you'd be happy to sell for.
The frustration and uncertainty isn't nothing; but I think if you were on the harberger system your whole life, and someone asked you if you wanted to switch away from it to the current US tax system, you'd laugh in their face.
In practice the spiteful billionaire will monthly send you offers $10K over the higher of your estimate or fair market value. So you'd be moving every month and incurring all the moving costs continually, or just continually dealing with the legal dealings of such an offer.
The transaction closes and the billionaire just sells the house at a small loss. The situation is asymmetric because the billionaire never actually uses the house.
That is to say, the prices you'd have to set to prevent very wealthy bad actors from just buying up everything and holding everyone hostage to it are so high that regular people would never be able to sustain them. And once they have obtained all the Stuff, even if they can't change the rules to make it so they no longer have to pay, they only have to keep the Harberger prices high for a fairly short time for the rents they charge us to bleed us dry enough that we have no chance of buying any of it back.
I think you're saying that, under a harberger taxation scheme, if your asset's (let's say it's a house) fair market value is $500k, you wouldn't be able to afford the taxes if you declared its value at $750k to make sure you wouldn't have to move unless someone (e.g. Blackrock) paid you enough to deal with that hassle.
But there's no reason to expect the prices on an extra 50% above market value to be ruinous, unless the harberger tax is set to near the full rental price of the asset, e.g. harberger-enabled Georgism.
And if the harberger tax is set to the full rental price of the asset, then Blackrock cannot make tons of money by buying the asset and renting it back to you; they will have all the rental income taxed away. It would lose them money every day they held onto it, and commonly-proposed harberger amendments like requiring the buyer to pay on their initially declared value for at least a year would quickly turn their ledgers into Redrock.
> Pricing things is super hard
Not applicable to copyright, but perhaps something analogous would arise if there were a market for it: Real estate agents have to estimate the value of a home to the market all the time, and have some simple heuristics with a fairly reliable error rate. Use those, plus an extra percentage for whatever the additional value is to you of sentimental attachment, of not dealing with the hassle of moving, etc.
You could also buy "harberger insurance" which would cover the additional cost of rebuying your asset, in exchange for premiums that depended on how much the insurer thought you were underpricing your asset.
Copyright only has opinions and history, and for the vast majority of IP, the history is "nobody has ever paid for this".
But the real killer for this proposal is that it creates a new mechanism for rich people or corporations to take work away from others. If the market is not serving the needs of the people at large, the people at large will and should alter or abolish the market.
There's two ways you could mean this: 1. A rich guy really likes your IP, for personal reasons; he's willing to pay a bunch of money for the rights to it.
2. A rich guy making business decisions thinks he can profit at someone's expense from owning large swathes of IP; he swoops in and buys it all up, to the detriment of the public.
Case #1 could absolutely happen under the harberger IP scheme. If Elon Musk decides he likes your cartoon, he can buy it at your declared price, and say it's his, and everyone will say "yes, Elon, that is yours." But Elon Musk already does that every day; he's constantly cropping the watermark off some cartoon or other and posting it on X, and everyone says "yes, Elon, that is yours." The only difference under the harberger IP scheme is that the artist actually gets his declared value for the cartoon from Elon.
Case #2 is what the harberger IP scheme is designed to facilitate, but it's not taking work away from others--it's buying the products that others have created, at the price at which the owner is willing to sell. This is what professional artists want. There's nothing here that would allow rich people or corporations to profit in some way that normal people cannot.
In the case of a Harberger copyright tax, there's an obvious equilibrium - anyone who seriously values their copyright has a strong incentive to err on the side of caution and pay taxes based on a high valuation, while lots of copyrights with negligible value are voluntarily abandoned due to the costs of both the tax and the administrative burden. The difficult cases in the middle would primarily affect speculators with large portfolios of copyrights of questionable value, which is an entirely desirable outcome. We could still drive socially desirable outcomes with very low taxation rates, because a very large proportion of copyrighted works have essentially zero commercial value.
Creating a streamlined registration, valuation and purchasing platform is entirely within the wit of man.
It works in abstract economic theory, but encounters serious problems when you introduce pragmatic complications like cash-flow, malicious actors, non-rational actors, asymmetric value, transaction costs, lumpy probability distributions, and other things. Fixes to those various problems can be bolted on, but the end result loses much of the theoretical advantages which made Harberger taxes appealing in the first place.
So in a parallel universe where the same work is written by a 20 year old and an 80 year old, the 20-year-old artist can sell the perpetual rights for more than the 80-year-old because the new rights holder will get 60 years of protection in the former case and only 10 years in the latter case.
I don't think I agree with the GP on the conclusion, but this part anyway seems logically sound to me.
The publisher might know that the work of a 75 year old might actually have some back catalog value, so might be willing to pay for it. For a 20 year old, they don't know that and aren't going to pay for it, even if the copyright they're buying likely has a much longer time frame.
If you used a 30 or 40 year old as your young example, I might buy the argument. Taylor Swift's back catalog as a 30 year old is worth hundreds of millions. Taylor Swift's back catalog as a 20 year old likely had an expected value close to zero, and there are very few 20 year olds as successful as Taylor Swift was at that age.
Meanwhile, virtually all of the copyright-driven industries are portfolio-based hit-seeking businesses. The comp for creators prices in some low (1-10%?) probability that a title will be a breakout success. Those returns would in fact be impacted by a drastic reduction in copyright terms.
Either way though: none of this matters. The point is that there isn't a logical argument for tying compensation to the lifespan of the author. Authors reap rewards today for returns on their IP that occur posthumously.
Copyright expiration during an author's lifetime directly impacts their incentive to make derivative works, because everyone else can do it too. In a world where Lustbader is churning out The Bourne Pontification, would Ludlum's own sequel be nearly as commercially successful?
If the author is dead, no derivative works can be made, so we may no longer want this protection.
Does this argument sway me personally? No. Are there workarounds like trademarks? Yes. Does (life + 20) make more sense than max(life,20)? No.
But you're being needlessly black and white here.
Less perhaps, but drastically less? You'd need to justify that. Even big hits generally make most of their money in their first few years.
And creating the same value is also a claim you'd need to justify. Maybe society legitimately prefers works from younger authors; maybe a work from a live author really is more valuable than a dead one. Maybe there are unique things that a living author can do with an in-copyright work they wrote in the past, that increase the value of that work, and we want to incentivise those things.
The "market" then decides how much its worth, and I decide how much I want to enforce it (either by assigning rights to people, or removing them entirely by applying a liberal licence.)
However Harberger taxes require me to _pay_ something first. This means that in practice only rich people have copyright, and aggregators like meta, google and ticktok will take all the profit allowing no legal recompense from me.
Which is almost like what we have now.
The issue with copyright is less about how copyright is assigned, its more about how fucking long it lasts. Life + 25 years is probably enough for almost all cases.
But the "its really bad that an artwork cant be used" sentiment is wrong. I made it, I want to control whos using it. The point is that the owner has the right to choose. Do they want thier artistic vision to be fucked up by someone whos only out for money? Do they want someone who they vehemently oppose to use thier stuff for thier own game? (think politician that you don't like)
So no, Harberger taxes will make the worst parts of copyright default, and limit access to the small people who actually really need it's protection.
How do you define "enough" and what would be a case where it wouldn't be enough?
but to make my thoughts a bit more clear, "just" life means that as soon as an "author" dies, all their works are immediately public domain, as you know. This would mean that estate owners would have no control over their (usually parent's) legacy. This would allow unscrupulous people (ie people with money) to muck up a legacy by assuming moral rights instantly at death. If a project is unsuitable, tough shit, your parent's legacy aint yours, get used to that Sherlock homes cypto currency.(extreme example)
25 years is enough for a legacy to be created, and a style to be continued, but not long enough for a work to be extinguished or strangled for all the money it's worth.
There are arguments for less, and more. But you also need to factor in that new works that contain old copyright, get a whole new time applied. for example casino royal the original will be public domain in ten years, but the daniel craig one wont.
Almost all countries are now party to the Convention, including the USA.
I mean, it wasn't entirely worthless, I did get to have a great conversation about it.
PM: Hey toast, we should try this cool technique for X from this company, they do Y.
toast: oh, we tried that, I have a patent for it, and it didn't really work. Oh and we couldn't do it at all on one platform and we just got rid of the permission we would need to do it on the other platform.
I mean, clearly the other company got it to work better than I did, so maybe it's not entirely worthless, but afaik, they didn't license my patent either (I suspect I would have heard about it, maybe not though; and my employer claims their patent portfolio is for defense only and afaik haven't yet taken any enforcement actions)
Mostly I'm just being pedantic, a lot of patents are assigned and the inventor doesn't make a choice about it.
This reduces the risk of a work being mis-valued, since it's probably already had time to mature in the market.
That said, it does feel like this would give even more creative control to holders of capital, which is already a problem in the United States.
This article shows that the paperclip optimization risk isn't limited to AI. We're running on that slope just as well.
Although I like the intuitive simplicity and game theoretic aspects of the arrangement, I've never seen it get through a partnership agreement or shareholder agreement round of redlines. Often lawyers feel nervous about it, and they'd prefer to litigate anyway; a space they are comfortable assessing and handicapping (and billing for). Often one party feels less sophisticated than the other, and thinks there might be a trap somewhere in there. Most sense, I think, that this sort of arrangement benefits the wealthier party. I think Harberger taxes suffer from the same sort of problem: the most money wins the asset.
You mention, rightly, the absolute trash heap that Brian Herbert has made of Dune. You speculate, wrongly, I think, that this arrangement would have cut him out and protected the IP. Au contraire -- I think Brian has done very well financially from this IP ownership -- sadly, enshittification pays -- and I don't see any reason to think that better stewardship would have resulted in more money to go buy out the IP than Brian's made.
Particularly in a world of art and human creative output, I'm negative on forcing IP rights to the wealthiest owners -- I'd prefer a system that leaves it in the hands of people who might have a variety of interests that could dominate, even if that gives us some tragedies in exchange.
Anyway, a system of transparent IP with buyout values (or reference values) attached is not, to my mind, uniformly better than what we have -- the reference values will be wrong, they will not capture non-monetary value, and they don't guarantee a ready buyer. It's hard to see what we'd gain; I agree that we wouldn't lose a lot, but there's a lot of social and legal work to implement between here and there.
Sounds a bit like a Shotgun Clause [0] when splitting assets from some kind of partnership.
Also, would licenses terminate when copyright gets bought out?
Better to just have a fee structure that starts off as the price of a carton of eggs and doubles every year that you extend it.
Not so good for anyone that cares about the artistic value - they are not economic specialists so they struggle to value a work and they perhaps don't wish to extract maximum $ from the work.
Bill Watterson was strongly opposed to merchandising calvin and hobbes, claiming that displaying images on commercially sold products would devalue the characters and their personalities. Bill's stance against commercialism is the main reason why the series never made it to TV or movies.
The US Constitution proposes them as a way to “promote progress in the useful arts” and specifically calls out their time-limited nature. I think that “authors life + 95 years” arguably stretches “limited time” beyond reasonableness, but I digress.
There is a real cost, both to the society the law serves, and in the operation of law enforcement and judicial infrastructure needed to support copyright and patent enforcement.
Even if you 100% believe in copyright and patent, it seems like a reasonable thing for the direct benefactors of copyrights and patents to substantially bear the cost of enforcement. A one-time filing fee (for patents) and nothing (for copyright) seems low.
There is also a societal cost in suppression of innovation by allowing these to last long periods of time, so it seems reasonable that the cost structure should be (1) related to the value of the work to the creator/inventor, and (2) incentivize shorter times, especially if the values involved are low.
I don’t fully grok Harberger taxes but it seems to me that they try to satisfy both my proposed criteria, but are flawed in that the value is stated by the inventor/author without proof. Maybe I’m misinterpreting but I see other comments to that effect.
My proposal is to make the fee structure impute the value, by increasing over time. For example, we could require a fee of $1000 for issuance of a copyright or patent (separate from the application fee). The term would be for one year. The patent/copyright holder can extend for as many years as they want, but the cost goes up 10x each year. If it’s a valuable asset, you’ll keep paying until the cost of extension exceeds your perceived value. Now the value is established by the most informed party in the market.
This is my proposal. You can of course tinker with lengths and amounts and increase factors.
In sum the idea is that for valuable works (a Marvel Avengers movie or the patent on the RSA algorithm or a cancer drug) the copyright/patent holder will keep paying for many years until it’s not profitable anymore. For works that have little market value, they would fall into the public domain quickly and be able to help improve other works or inventions, and inhibit rent seeking. It practically eliminates the incentive for patent trolling, or for suing an artist decades later for having a beat similar to a song 40 years ago, or for establishing huge defensive patent portfolios. The economics aren’t there anymore.
And then beautiful to see how the internet responded.
Dwarkesh Patel
How do you sustain yourself while writing full time?
Gwern
Patreon and savings. I have a Patreon which does around $900-$1000/month, and then I cover the rest with my savings. I got lucky with having some early Bitcoins and made enough to write for a long time, but not forever. So I try to spend as little as possible to make it last.
I should probably advertise the Patreon more, but I'm too proud to shill it harder.
It's also awkward trying to come up with some good rewards which don't entail a paywall. Patreon and Substack work well for a lot of people like Scott Alexander, because they like writing regular newsletter-style updates but I don't like to. I just let it run and hope it works.
Dwarkesh Patel
Wait if you’re doing $900-1000/month and you’re sustaining yourself on that, that must mean you’re sustaining yourself on less than $12,000 a year. What is your lifestyle like at $12K?
Gwern
I live in the middle of nowhere. I don't travel much, or eat out, or have health insurance, or anything like that. I cook my own food. I use a free gym. There was this time when the floor of my bedroom began collapsing. It was so old that the humidity had decayed the wood. We just got a bunch of scrap wood and a joist and propped it up. If it lets in some bugs, oh well! I live like a grad student, but with better ramen. I don't mind it much since I spend all my time reading anyway.
Was Dune mismanaged? We got a lot of movies, TV shows and a famous video game from it!
Was Borges estate mismanaged? Wiki talks about two examples that are not dramatic.
What is the idea?
We also got Brian Herbert's terrible Butlerian Jihad books
I think fundamentally this article seems to be asking "There are these guys right, and they have created this fantastic stuff and it's protected by copyright, but they're not choosing the most economically productive way to use their stuff, so we should find a way to take it off them". And I'm not sure that premise is correct. If you want to create economic value from a copyrighted material please... go create some copyrighted material with economic value.
Deliberately giving things to the public domain, software licensing, and other issues arise in the context of hostility and malice and bad people existing. Stuff like this would only work if everyone participated in good faith.
By the time we figure out any sort of sensible copyright rules, AI will have made a total mockery of our systems, anyway.
The fact that BigCo. might be able to exploit an IP for some purpose -- perhaps making commercials out of popular songs -- in a way that would lead to revenue does not intrinsically suggest that it would be a good thing. Why shouldn't the artist retain control? Why shouldn't his heirs (who have already paid inheritance taxes) retain control?
Should an individual unable to pay taxes on his own organs lose them to someone who will pay the market value? I think most of us would think that grotesque. And it's grotesque because the viewpoint behind it -- that the highest bidder should automatically win, that it is intrinsically the "highest" and best purpose -- is grotesque.
That is generally how property taxes are supposed to function.
It's better than taxing the output of hard work - work, unlike hoarding, is generally something we want to encourage.
To what hoarding are you referring?