Some Criticisms of Georgism
How are we to assess it?
The land value tax is a tax on the unimproved value of the land. The idea behind it is that it is a perfectly efficient tax – it only taxes the value of the land which is due to the efforts of other people spilling over onto your property. Unlike a property tax, it does not disincentivize developing a property, but in fact encourages development. You either use the land, or you lose it. There is no incentive to sit on unused land, waiting for it to appreciate as the land around you is developed. In fact, in a stylized model of the economy, a single tax on the unimproved value of the land is sufficient to pay for all necessary public goods (Arnott and Stiglitz (1979)). I want to criticize it, though. I am optimistic about the possibility of a tax on the unimproved value of land being welfare enhancing, but it would have to be small.
First, how should you account for natural resources? Resources are not always obvious — they require cost and effort to find. We have several solutions, none great. We could just ignore resource value, though this leaves considerable rents to the landowner. We could deduct the cost of searching from the tax bill, although this is extremely vulnerable to misrepresentation of the actual cost of looking. It would be like Credit Mobilier, or the recent Italian subsidization of home renovations.
It is also not clear what the value of the unimproved land is at all. The ordinary method is to take the market value, and subtract off the value of the improvements. This is sensitive to how you depreciate assets. How are you supposed to know how much maintenance work has been done? If the buildings on the land are more valuable than you think, then you will go back to taxing improvements. I do think it is possible to construct a mechanism which accurately reveals valuations. Owners can state their own valuation of the property, and then anyone can buy at that price. Nobody who implements a tax on the unimproved value of the land does this, though. It’s very politically unpopular.
What I think the most important criticism is is that a single tax breaks down when spillovers are due to a single agent. Let’s assume that we tax the full value of the unimproved land. Suppose that someone owns two plots of land, and they build a stadium on one, and a hotel on the other. There are hotels which only make sense in the context of the stadium being built, but which would be counted as an increase in the unimproved value of the land. Unlike with natural resources, there is no easy solution – how are you to say which of the spillovers are due to your improvements on other parts, and which are due to other peoples? The reductio ad absurdum of this would be to divide up property into arbitrarily small plots, and then tax the value of each bit of land, never minding that a bathroom is only valuable with a bed room to go with it. We would not want to, in order to fund public goods, place a tax on public goods!
The Arnott and Stiglitz result mentioned earlier is predicated on there being a single public good, which the tax can then be used to fund. There are many types of public goods, though, and discovering what those public goods are is best done in a market. If the world were simple enough that a single good were all we required, private action would be entirely superfluous — we could simply centrally plan everything. Since the world is complicated, the single tax must fall, tarnished, to the realm of all other taxes. It is plausibly useful, and may be superior to other taxes. We need not expect us to make the world optimal.
1. Estimates of the revenue that could be raised by an LVT in the US come out to 30-50% of federal and state tax receipts at a low rate. That’s significant!
https://www.astralcodexten.com/p/does-georgism-work-is-land-really
As another estimate, US states use a ~1% property tax ($600B/yr), LVT proposals are about the size of a ~3% property tax ($1.8T/yr). And of course, this revenue can crowd out less efficient tax sources, improving growth.
2. Taxing natural resource extraction and taxing land possession are somewhat distinct issues in Georgism. It's rare that high value land in urban centers also has undiscovered oil. For natural resources, severance taxes are an established solution to taxing resource extraction but have some inefficiencies. Harberger taxes could work too.
3. Estimating the value of unimproved land sounds challenging but has a lot of solutions in practice:
https://www.astralcodexten.com/p/does-georgism-work-part-3-can-unimproved
To the above discussion I would add my auctions proposal:
https://splittinginfinity.substack.com/p/some-thoughts-on-using-auctions-for
Because Georgism aims to collect only 80% of the land value, there's a lot of wiggle room for inaccurate assessments.
4. For the hotel and stadium complementarities it’s helpful to walk through the example more fully. The owner starts by buying undeveloped or underdeveloped land at a certain price and then adds buildings to it. The price of the undeveloped land is a good estimate of the land value and any assessor should use that as the starting point for assessing the land, not the value after the buildings are built. Alternatively, using assessments on undeveloped areas nearby gives an estimate as well.
After the hotel and stadium are built, nearby land values rise, raising more taxes for local public goods or increasing the dividend that the hotel owner receives.
The end result is that raising local land values slightly reduces the owners incentive to build (since this will raise the value of the land they built on and thus raise taxes). But this distortion is probably lower than the property taxes we use today.
5. I think we agree that the Georgist proposals would probably be an improvement to tax systems even though they’re imperfect and fall short of a revolution. Still seems worth doing.
>What I think the most important criticism is is that a single tax breaks down when spillovers are due to a single agent....There are many types of public goods, though, and discovering what those public goods are is best done in a market.
I've been concerned about this for a while - LVT disincentivizes attempts by communities to produce positive rents they would normally capture. It creates a misalignment whereby a location's desirability, which is often improved intentionally by local agents working together to provision public goods (e.g. HOAs), is now nationalized by a taxing authority that lacks that rent-increasing incentive. High land rents are good! They're the closest price signal we have to everything that is otherwise impossible to price: air quality, community networks, how pleasant a view is, literally everything there is about living and working somewhere. We should want to maximize them. But I don't think this is a death knell for LVT, it just means we need to make slight adjustments and everything will become very well aligned.
1. Make the natural zone in which spillovers from investments occur (greater metropolitan areas ideally) the taxing authority.
2. Give the taxing authority an incentive tied to land rents - ideally they should be a single, for-profit firm that owns all land and has as much sovereign authority as can be secured without threatening exit rights. Reedy Creed Improvement District is one model of this.
3. Have a network of such entities with free movement between them such that a market is created for different bundles of public goods and different models of metro administration, as a more explicit version of tiebout sorting. I expect efficient entities will mirror this set-up internally for their satellite cities.
If done right this puts governments under market discipline to maximize land values, our closest proxy for "everything good about the place they're responsible for" while organically learning-by-doing the best methods for appraisal, HGT public good investment, and *the art of governing itself*.
Rather than seeing the problem you brought up as a challenge to Georgism, I think it leads us to something that might substantially improve upon the initial vision. You might even be able to keep much of the citizen's dividend 'social return of rents' elements of orthodox georgism if these firms are put under cash flow taxation.