I agree with many of the arguments here about the theoretical impacts of a land value tax, especially the section "an LVT implicitly taxes improvements to nearby land" which is often overlooked or glossed over in these discussions.
But my main argument is practical. I content that it is simply not possible to evaluate the "unimproved value" of a given parcel. Any discussion of a practical LVT has to start with the fact that it is an approximation to a theoretical ideal, and define exactly what the basis for "land value" estimation is, because it's really a tax on that process. While some of these may have overlap with the benefits and detriments of a theoretical LVT, they have to be looked at from first principles rather than by comparison with the LVT because the fundamental assumptions are often broken.
> I content that it is simply not possible to evaluate the "unimproved value" of a given parcel.
How familiar are you with existing property taxes?
It might surprise you that land value estimation is literally already happening at scale.
Also, you don't need to be 100% accurate with the estimation. Even a 50% lvt would be a huge improvement and would mean that you could literally be off by 100% which is extremely unlikely. How many houses do you see selling for twice the listing price?
Indeed. When I see these discussions, I am always struck by two things:
1) that many people don't seem to realize that a tax on land value is not novel, but is one of the oldest ways to fund government, with tons of experience behind it (past and present); and
2) that whatever the problems associated with figuring out land value, they pale in comparison to figuring out an individual's real income, which obviously didn't stop us from taxing that.
Right. So frustrating to hear people speak of LVT as some pie in the sky concept when it's the most original and basic form of tax, for centuries, worldwide.
It's equivalent to a modern day person talking about how cooking food over an open fire is never going to work.
The valuation issue is also a non-issue. You just have people self-assess the value, but any stated value is an option for the state to buy at that price. Also an ancient and well known mechanism.
But the difference is that assessing total property value can appeal to recent sales, so you have actual objective data to bound the realism of your assessment. What somebody would pay for the land without the improvement is stuck in imagination land.
> especially the section "an LVT implicitly taxes improvements to nearby land" which is often overlooked or glossed over in these discussions.
The claim, which I disagree with, is that spreading those taxes across nearby land incentivizes those property owners to sell their land to someone else who will improve it.
Which gets at another LVT problem that is glossed over in discussions: Everything assumes that selling properties and moving is cheap and easy. If grandma's forever home is surrounded by apartment complexes when she's 85 years old, her taxes would become unaffordable because she's paying her share of those apartment complex value taxes. She would just pick up and move, which we're supposed to assume is cheap and easy.
It can all be fixed by making the tax structure a combination of land value and structure value, which happens to be how existing property taxes are constructed in most places.
>The claim, which I disagree with, is that spreading those taxes across nearby land incentivizes those property owners to sell their land to someone else who will improve it.
The only claim here is that if you own land which makes you $1000 in income and pay $2000 in taxes for that underutilized land you'd probably prefer to sell up.
Which has to be the least controversial part of LVT.
>Which gets at another LVT problem that is glossed over in discussions: Everything assumes that selling properties and moving is cheap and easy. If grandma's forever home is surrounded by apartment complexes when she's 85 years old, her taxes would become unaffordable because she's paying her share of those apartment complex value taxes. She would just pick up and move, which we're supposed to assume is cheap and easy.
We're seeing the net result of your desired policy right now where retired boomers sit on 4 bedroom houses with 3 empty bedrooms while anything resembling this type of family home is unaffordable for actual families.
Personally I think I preferred it when retirees were given tax incentives to sell up and downgrade to a smaller property, because even though moving day is stressful, not easy and costs money, it's not worth sacrificing an entire society over trying to avoid it.
Retirees have been doing this for as long as humans have owned property. Then they die and the house moves on. There is good reason someone will want to live in a house that is larger than they need.
If there are not enough houses don't blame that on existing houses.
> Retirees have been doing this for as long as humans have owned property.
Absolutely not. Extended families lived under the same roof for most of human history. This is a Nuclear Family problem, which only emerged in the 20th century.
more than just that. the very concept of "retirees" is a modern phenomenon. for most of human history you'd have to keep working on the farm or whatever until you were no longer physically able
Multigenerational households were the norm until recently. The eldest son gradually took over the household and raised his family there, or something like that. Both because it would have been terrible waste to have an entire house for some old people, and because household chores were hard work before modern amenities.
The less wealthy the family, the more likely you'll see it.
So your aggressive taxes will hit those people - displacing additional generations, not just the land-owning-but-otherwise-fairly-poor retiree - before it will hit the stereotypical middle class boomer retiree.
Outside of CA's Prop-13 territory, the multi-generational shabby-old-home-owners pay less taxes currently than their richer neighbors who moved more recently and renovated or expanded. The land value of both is going up, but the improvement value is lower for the poorer family. So now you'll get rid of the improvement value and even it out for both, which will hit the poorer land owners the hardest.
But land is a scarce resource, at least desirable land.
Blaming existing houses is exactly what you should do because instead of them you could build higher density.
There are plenty of houses for sale at any time. If building densely is the goal then any one of them can be used to build. Speculators buy houses in hopes that in the future the house next door will sell and then they can combine the lots to a larger building - where this is allowed.
If you're concerned with "it's underutilized because the population density is low and other people need more housing" than it would be much easier and more effective to directly pursue building housing on low-population-density non-residential land. Direct construction driven by the government, vs a multi-step strategy of "make people miserable with tax payments until they sell, hope the people they sell too will be deep-pocketed developers who will build super-high-density stuff instead of just fancier homes for richer families, and hope all this happens quickly."
Cause a tax amount that goes up based on what people with more money than you do on other pieces of property simply gives more power to the wealthy. "Underutilized" as far as tax implications go then means "people with more money than you would like there to be something else there."
And, of course, this already happens with US property taxes in many jurisdictions. And people absolutely hate it.
"Tax incentives to downgrade to a smaller property" sounds great in theory for retirees sitting in huge properties, but I think is limited in practice. The people with the really big places are wealthy and politically influential, so you'll get Prop 13s, or you'll get the recent cuts to property tax in Texas. "Make housing more affordable by cutting the taxes!" And the people impacted by more aggressive taxes will less be the boomers in giant houses and more be the poorer retirees in multi-generational living situations, or ones in fairly small condos.
How would LVT be easier or quicker? In what states are you going to be able to campaign on this?
"Change commercial and industrial under-utilized areas to allow residential, then aggressively subsidize developers or build directly there" sounds much easier to me than "pass a property tax increase that people will fight tooth and nail." Your LVT is gonna need to come with zoning changes too, after all.
California has Prop 13. Texas lowered property taxes recently. Florida is considering property tax rebates. You're gonna have a hard time convincing people to increase property taxes when housing is expensive already. Big overlap between "property owners" and "highly motivated voters" already.
Incentivizing density in single-family areas is quite hard, I'm not convinced it would do much even if it passed! Even in CA where neighborhoods can now have ADUs, sales prices for 5bd big-as-box-as-possible single family units are generally higher than prices for 3bd or 4bd + an ADU properties with separate units. Because the people who can afford those would rather have a giant-ass house than be a small-time landlord on the side. So maybe you just displace the old retirees and a bunch of high-income DINKS move from their condos to those SFHs. Sales price goes down a bit to compensate for the now-higher tax burden; remains unreachable for many but reachable enough for enough that it doesn't spur massive change.
> We're seeing the net result of your desired policy right now where retired boomers sit on 4 bedroom houses with 3 empty bedrooms while anything resembling this type of family home is unaffordable for actual families.
Some of this is due to tax policy, at least in the US. If you own an oversized house that you’ve had for long enough, then most of the value is a capital gain. If you sell it, you pay taxes on all but $500k of that gain, even if you promptly buy a new, smaller house that costs almost as much. If, instead, you hold the house until you die, the tax is waived completely.
California has additional perverse incentives due to property taxes.
Some states are allowing you to “carry around” your assessed value, perhaps the $500k cap gains exclusion should be made similar - you can carry your basis similar to a 1039 exchange if you sell and rebuy in the same area soon enough.
The reusable $500k cap gains exclusion is a huge huge thing, bigger than some realize. (Though it is interesting that you can exclude gains from a house sale but not losses, but you if you own a rental you can claim losses but not exclude gains (though you can 1039).)
The reusable $500k exclusion is great in some markets. In the markets with serious housing cost problems, it’s nowhere near adequate. If you have a house that you’ve owned for 30 years in a market where the cheapest houses cost $3 million, and you want to downsize and move into a nice $2 million condo (sigh), and you live in a high-tax state like CA, you will owe long term capital gains tax on nearly $2.5 million, which comes out to around $750k. Real estate agent fees can easily eat up most of the remaining profit, and you probably end up with much higher HOA fees. And these expenses are basically a complete loss to your heirs. And, in CA, your property taxes increase from near zero to $20k/year.
So you probably don’t make the trade, and those spare bedrooms remain unavailable to anyone who would want them for the rest of your life.
Grandma's problem has been historically solved with a homestead exemption. Of course the value could rise above the exemption, but that just means it should be set high enough to ensure that the proceeds of selling will give Grandma a lot of options.
Well, if grandma's home is surrounded by apartment complexes already, then even without LVT some property developer would definitely make an offer to buy her house for a ton of money, and even throw in some relocation package, for example by gifting her an apartment in one of those neighborhood complexes. Happens in my country all the time - your neighborhood being converted from single family homes to apartment complexes usually means a windfall for you.
This is already part of property tax assessment - I get a separate price estimate for the land and the structure. I also own some undeveloped land and pay property taxes on that. All a LVT does is get rid of the structure part and raise the rate on the land part.
Saying that they assess the value tells me nothing. How, specifically, do they arrive at these assessments.
In my experience there is often an assessment process that is essentially just made up. And when properties do sell, the sale price is always a "surprise" relative to the assessed combined value of the property.
In a sense the question is "What in particular makes you confident that the estimate accurately reflects the price of the land" but in a deeper sense what does the concept of "price of the land" even mean in practical terms? How would you know that the answer is right even if you were omniscient? And given the practical divergence from whatever theoretical standpoint, does then this value serve the same objectives as a "true" LVT?
The assessments are roughly based on the sale price of similar lots and the approximate rebuild value of the structures; usually erring on the low side in my experience (though that is probably more about the direction of the housing market than inherent to the process). I agree there is a qualitative element that strikes me as icky compared to pure quant taxes like wages, but it's already happening so LVT doesn't materially change things.
> How, specifically, do they arrive at these assessments.
So is your argument that you don't understand something and so it must be wrong?
There's an extremely large amount of existing material that is used by property assessors available for you to look up to research how they do this. It is a well-established field.
Property assessments have been done across a huge number of countries for decades probably billions of times at this point. There are probably trillions of dollars of capital that flows according to these assessments.
No, my point is that two things both calling themselves a "land value tax" based on a putative "unimproved land value" can be wildly different in terms of incentives and results based on the specific mechanism used for computing that value, because there is a lack of any sort of objective reality underlying the assumptions that go in to computing that number.
At least for an entire value tax, we have the benchmark of "when a house is sold, how closely does the price hew to the imputed value for tax purposes" to determine if the tax is, in some sense, fair. Some municipalities do better than others by this benchmark, but many of them continue their totally broken methodologies (by the standard of predicting sale prices) for a variety of political and bureaucratic reasons.
For a specific attempted implementation of a land value tax, how do you go about measuring whether you are doing it well? You can sometimes get sparse data points when vacant lots go up for sale, but otherwise you're just benchmarking against other models.
>> I content that it is simply not possible to evaluate the "unimproved value" of a given parcel
There's a decent discussion on that topic here [1], as a starting point. Not saying it's absolutely conclusive, but gives some food for thought. I suppose where determining an accurate value might be most difficult is parcels that rarely turn over, so have little direct or nearby sales data.
It is an interesting discussion, but I don't see anything in there about how to assess the accuracy of any of the methods against some sort of objective truth. I would have to read the underlying papers I think to get at this, but I don't feel a strong need because I feel like the larger epistemic point is unaddressed in any of the summaries.
The closest I saw was one study that compared the model to a human generated data set, which is just kicking the can down the road. The article semi-concludes
> I think it's quite plausible but not a slam dunk. That said, if the objection is, "valuing land separately from improvements is fundamentally impossible, and we can never get better at it, so we shouldn't try," I think that's plainly ruled out.
I do not agree with this assessment -- you can create a bunch of models and show that the models have good intra-model agreement, but the fundamental point has not been touched.
Maybe but at some point, isn't your problem the same valuing a single dollar? We can compare the USD against other fiat currency, but they're all valued against the USD or some other bucket of fiat currencies. Or, you could try to tie the value of the dollar against a bucket of goods, but that's also just moving the goalposts around.
Now we're wandering a bit far afield, but estimates and measurements like this are fine if they are designed for an operational purpose. If I want to know the unimproved value of my land so that I can evaluate whether to buy a similar property and build a similar house on it, then it's fine for me to use whatever estimation protocol is going to help me make a decision. If my protocol is bad then my estimates will be bad then <shrug emoji> it's my problem.
Similarly trying to measure the value of the dollar -- what is the operational purpose of that measurement? This is a real problem in any sort of macroeconomic analysis, and Goodhart's law makes it far far far worse when trying to apply it for practical purposes. Mostly you have to accept that there is not going to be a quantitative metric that captures the underlying squishy concept so better not to think about the problem of, say, inflation, in purely quantitative terms.
No man is an island, unless you're Robinson Crusoe, who was a character in a story.
No parcel of land is an island in this sense, not even an island, because what makes an island is its being independent of a landmass or continental shelf. But in the sense of being moored, all land values are grounded in discrete monetary terms. Even if they are subjectively determined, the land value is an idea space that has limits on what concerns are common, and thus are typically already priced in, and concerns that are less common, and may have already been sold, such as mineral rights.
Unimproved land value is value relative to comparable[0] plots in that same market if possible, because it's meant to be as close to apples-to-apples, all-else-being-equal as it can be. The basis for comparison is whether it's improved or not, as you seem to be specifying, so the relation of land value to improvements seems to be the only independent variable that a potential or actual landowner or anyone else could even manipulate, even in a hypothetical where we aren't bound by real-world factors or limits.
Do you believe you can determine the true value of improved land, given how illiquid a market it is? Obviously, the last sale price provides some true information about the value, but it could literally be a decades old number. Do you believe states and municipalities shouldn't update property taxes for a parcel of real estate unless it's sold? I think we've seen from CA's experience with Prop 13 that this creates pretty distortionary incentives.
> Do you believe you can determine the true value of improved land
It is at least theoretically answerable. In the extreme, yes. We can simply force the sale of the land. Practically speaking, no, we cannot answer that question in a deeply illiquid market.
For the unimproved value I'm not certain that there is a consistent and useful theoretical definition that can be translated to practice. Even in the extreme the question of the unimproved value of the land becomes difficult. Were we to raze all improvements and force the sale would that give us an answer? Do we include the cost of razing? What counts an unimproved? Can we leave trees or grass?
To expand on this differently -- if we have a model that produces estimates of the "true value of improved land", then we can validate that model whenever property sells. It would be a slow process but if we have a model and the model parameters capture a significant portion of the variance, then we would expect the model to converge over time.
Nobody does this, of course; it is not usually politically expedient to do so for a number of reasons, not least of which is the predictability of taxes for a given parcel. But at least it is theoretically asymptotically achievable. Not so for unimproved land value because the value you are trying to estimate is not an actual quantity.
Tear downs, sites with existing structures which are demolished to make way for building something else, are sold all the time. The value of the land in those instances is pretty much exactly the purchase price minus the demolition costs.
Real estate is liquid enough. I cannot sell my house this afternoon, so it isn't fully liquid, but a real estate agent can give me a number to list my house at this afternoon and be within a few % of what I get in a few months in most cases so that is close enough to liquid.
>>I agree with many of the arguments here about the theoretical impacts of a land value tax, especially the section "an LVT implicitly taxes improvements to nearby land" which is often overlooked or glossed over in these discussions.
That's a feature not a bug.
The whole argument is that it's good to tax wealth that you didn't work for. It's surely more just than taxing labor or the improvement you have made yourself.
>I agree with many of the arguments here about the theoretical impacts of a land value tax, especially the section "an LVT implicitly taxes improvements to nearby land"
Which is more of a feature than a bug.
The alternative to "local land value improvements feed the local tax base" is that schoolteachers who make the local schools good make the local landlords more money.
The idea mooted in the article that developers would be unwilling to build 20 houses on a plot of land because having 10 houses would jack the LVT up for the other 10 is entirely backward. The value of those houses will be predicated almost entirely on infrastructure (roads, rail, schools, etc.) or services (shops) provided by the community you're paying taxes to.
>I content that it is simply not possible to evaluate the "unimproved value" of a given parcel.
Did you read the wikipedia page about LVT which describes how? Which part is impossible?
> The alternative to "local land value improvements feed the local tax base" is that schoolteachers who make the local schools good make the local landlords more money
Or indeed that a headteacher works long and hard to improve their school and all they get for it is a reduction in their real income because the plot of land their house sits on costs 20% more.
Or the schoolteachers get driven away by a horde of NIMBYs who really don't want to be forced to move because the schools are good...
And that headteacher's salary can increase too because the tax base went up.
No such luck for the poor schoolteacher whose rent went up. All she will contribute to is her landlord's vacation fund.
>Or the schoolteachers get driven away by a horde of NIMBYs
Really??? You think NIMBYs will protest a good school?
If you want NIMBYs go visit San Francisco. Theyve detaxed land there to such an absurd extent that the locals with $1.5 million mortgages will flock to town council meetings to try to declare a launderette historic to prevent it from being turned into apartments (because if any more apartments are built they will go underwater on their absurdly sized mortgage).
> Really??? You think NIMBYs will protest a good school?
I mean NIMBYs protest educational facilities and transit links all the time, even when on paper at least they reap benefits in terms of house price appreciation.
Imagine arguing with a straight face they wouldn't protest such improvements more actively when someone else's kids getting a good education pushes them below the poverty line or forces them to leave the area. If you align any improvements to an area with increases in costs, you get incentives to ensure an area doesn't improve.
> Did you read the wikipedia page about LVT which describes how? Which part is impossible?
The wikipedia article describes some processes, including assessments, regressions, and interpolation from fixed landmarks.
Those are all means of estimating something, which you can call the "unimproved land value" if you are so inclined, but what exactly is the thing that they are estimating? How do you know if they are accurate?
You can implement a framework based on any of those measures, but crucially as above they are not an LVT, they are a "proportionate tax on total value based on extrapolating previous sales minus human estimates of improvement value according to a rubric" for example, and have different advantages and disadvantages than an LVT even theoretically, so every time you make an argument that "LVTs have such-and-such a property" you have to expand the definition of LVT to be the specific case and verify whether that property makes sense in the context of that particular methodology. As a shorthand it becomes useless.
My point is not that there are attempts to have an LVT that are approximations of the ideal reality; my point is that this ideal simply does not exist in any sort of cogent way so you might as well tax based on how much God loves the property or how many potatoes you could grow on the land.
>You can implement a framework based on any of those measures, but crucially as above they are not an LVT, they are a "proportionate tax on total value based on extrapolating previous sales minus human estimates of improvement value according to a rubric"
>>>I content that it is simply not possible to evaluate the "unimproved value" of a given parcel.
I think there's a perfectly fine way; you estimate it the way we currently estimate properly values for tax purposes. If they owner doesn't like that value, you allow them to contest it, and we immediately accept any contested claim and value it as the owner desires, with two small caveats:
a) they pay tax on the claimed value, to ensure they don't over value and
b) they are required to sell to anyone at claimed value + 10%, to ensure they don't under value.
edit: two points to address some responses. First, it can be claimed land value + assessed/claimed improvement value + 10%, that's fine. Second, I'd only require they sell at that value if the owner contests the original appraisal. If they accept it, nobody can buy their stuff out from under them for any price.
But the point of LVT is that it doesn't include the value of the stuff on the land. A house can easily be worth more than 10% of the land it's on, my house is valued at about twice that of the land, or 20x what your plan would require me to accept for the house.
>they are required to sell to anyone at claimed value + 10%
I like the goal here, but I think a less intrusive way to achieve it would be to charge back taxes at the time of a future sale if the sales price is in excess of the valuation. To ensure the back taxes are paid, they would encumber the title of the new owner, so that in practice the buyer would require them to be paid at closing.
But my main argument is practical. I content that it is simply not possible to evaluate the "unimproved value" of a given parcel. Any discussion of a practical LVT has to start with the fact that it is an approximation to a theoretical ideal, and define exactly what the basis for "land value" estimation is, because it's really a tax on that process. While some of these may have overlap with the benefits and detriments of a theoretical LVT, they have to be looked at from first principles rather than by comparison with the LVT because the fundamental assumptions are often broken.