It's not surprising that so many "rationalists" oppose Georgism. They fancy themselves intelligent creators but actually cannot prosper unless they're allowed to own something that makes them money without any real effort on their part.
Regarding part III, there is a critical subtlety in the question of "can you measure the unimproved value of land cleanly", and how that relates to the idea that the tax has no deadweight loss.
The normal sense of "unimproved value of land" is: if I took this particular parcel of land and knocked down the building, what would it be worth?
A subtle feature of this definition is that the value of land is sensitive to parcel size and is not linear. That is, two separate parcels A and B are worth *more*, added together, than a parcel C combining them would.
The reasons for this is the same as the reason that the LVT notionally raises a lot of money with no deadweight loss: because most of the value of any specific parcel is spillover - a positive externality - from all the improvements on the neighboring structures and really most of the structures in the region. If someone builds an apartment building in one parcel, the neighboring parcel becomes more valuable because you can put retail on it. A developer that owned *both parcels* could internalize this externality by building both structures - but if the parcels are taxed separately, then the developer is actually paying a tax on its structures! It pays a tax on the retail by way of the apartment-land and on the apartment by way of the retail-land.
It's important to note that this is not a hypothetical. In general this sort of thing - buying adjacent parcels and building synergistic structures on them - is fairly common. (There is a lot of it going on in Tysons Corner in Northern Virginia, near where I live.) Even more specifically, this is a norm for the development of suburban subdivisions - a single developer buys up the land for the entire neighborhood, builds a bunch of houses (and amenities like pools, sometimes), sets up an HOA and so on, and then sells the houses. This is profitable because the houses are worth more in a nice little neighborhood than when built one at a time as stragglers off of a main road.
So: if you take parcels as fixed and immutable, there is actually a deadweight loss from an LVT.
Is there a way to get around this? Well, you could allow people to combine parcels for the purpose of appraisal, and try to revalue the combined parcel without *any* of the buildings on it. But this creates a straightforward tax strategy in which ownership of ever-larger contiguous regions is combined so that more and more spillover value gets subtracted out and the tax base is eroded. (In the limit you value the Earth as a single parcel with no buildings at all, and ask "what is the value of a square kilometer of Manhattan if no buildings exist in the entire world?")
The existence of nonzero (and nontrivial) deadweight loss really changes things. The general rule with deadweight loss of taxes is that, for a given tax, it's proportional to the square of the tax rate. So for an LVT with an extremely high rate, this could be quite serious. But in particular this informs the comparison with an equal-revenue property tax.
The LVT falls on land and the property tax falls on land+structures. So the property tax has a larger base and a lower rate, but the structures portion of the base has a big coefficient for the deadweight loss while land just has what we have above + whatever you get from other problems in land valuation. The result is that, unless you actually know all of the relevant parameters, it's ambiguous which of the two taxes has a higher total deadweight loss.
Finally: it's overlooked, but a large, large portion of the tax on structures *also* has no deadweight loss: the tax on structures that already exist. With the exception of a few structures that might actually be torn down, you cannot disincentivize the construction of a building that already exists. So simply eliminating property taxes is a windfall to that particular factor, and - why pay out a windfall? As long as we are doing estimates and approximations, I personally favor reforming existing property taxes to have abatements for new improvements rather than trying to go full LVT.
OK but like, there's not enough houses in America being built to satisfy the demand. So the landlords can just charge whatever they want because there's not really anywhere for people to go? Maybe their property values decrease, so they don't want to sell their houses, but why should that have any effect whatsoever on rent prices? If you implemented this LVT tax, what prevents a landlord from passing it on to the tenant if every other landlord is also passing it on to the tenants, and no one is building new houses because building new houses is illegal? People have to live somewhere and moving is very hard.
I was really hoping this section would be convincing but I just can't see how this would concretely change the situation in America. I guess I just don't believe the Danish study because presumably it's not illegal to build houses there or something. Help me understand how this applies to America. Can you explain, concretely, how in a city where landlords are already gouging people on rents and their taxes go up, and it's illegal to build new houses, that they wouldn't just increase the rents further? What is supposed to happen to the people who are living in that city? What does the person who can never afford a house and is already spending 50% of their income in rent actually do that "gives them leverage"? They can't buy land or a house. Everyone is raising rents everywhere because of this new tax. I predict that in America, the tenants of that city would just eat it and spend 60% of their income on rent instead of 50%, and just give up on ever having kids. How, concretely, are you imaging it could be any different?
Lars, could Donald Hagman's 1965 book, "The Single Tax and Land Use Planning: Henry George Updated," actually be a journal article and not a book?
I couldn't locate a book by that title in several usual places, including WorldCat and AbeBooks. However, HeinOnline offers a preview of a paper in the UCLA Law Review by that author, with that title, in its 1964-65 edition.
While this evidence is intriguing, I think you'd still need to red team the new rules to really make a compelling argument that they can't be gamed.
Suppose, for example, that one of the metrics used in practice to assess the value of the property is what other properties sell for versus what they previously sold. Great, that's hard data, even if it requires some interpretation; wear and work and so on. But you could easily imagine a degenerate case where only a couple properties were sold, and a cabal of property owners could gather and drive the prices down on those few sales so their taxes are artificially depressed; they then split the difference with the seller.
I want to believe that it's hard to just raise the rents and soak people, but having watched friends try to buy houses in the last year ... sellers' markets are rough for buyers.
So, I was thinking about the observation that Georgist and Pigouvian taxes both have no economic deadweight loss. On the surface this makes perfect sense: force actors to price in negative externalities, and don't let them benefit from things they did nothing to earn or create.
But the Georgist definition of land isn't just a matter of uncreated wealth, because the value of one parcel of land (especially in cities) is primarily a matter of all the other improvements humans have made to surrounding land. On its face a LVT prevents me from benefitting from those positive externalities, which, fine, I didn't create them. But to me this suggests a problem in regards to defining how large each taxed unit should be. If I own a plot of land and build a house on it, then sure, I have no right to earn higher rent just because someone else builds a mall or a highway or a better school nearby. But what if I buy up many plots and build a bunch of things on them that mutually increase the value of one another? It seems like unless I can somehow arrange for all the land I own to get assessed as a single unit, I'm being prevented from benefitting from positive effects of improvements I myself create. An edge case for sure, but developers do, in fact, sometimes buy up large tracts to build mixed use communities for exactly this kind of reason today.
If you own land and rent it as a mobile home park, and suddenly you have to pay an additional $25k/yr, and so does every other mobile home park owner in the country, then of course prices will rise. When labor costs rose over the past year I raised my bids to compensate for it, just like every other contractor did - and they're not making more labor(ers) on cue either. I didn't shrug and tell my gf we were living off rice and beans so that I could have the privilege of dealing with the physical and emotional stress of contrstruction, nor would I react this way to have the privilege of dealing with the stress of owning a mobile home park. If prices didn't rise, then services would be cut - the septic wouldn't be pumped on time and would back up into the trailers (this actually happened at a park down the road from where I grew up until after many years of complaints it was finally condemned), the well pump would burn out and be replaced by a hand pump, and so on. I don't understand how anyone can possibly think otherwise. And if rents do rise or costs are cut, then the tub and flow analogy doesn't hold water either because the flow is essentially cranked up commensurate with the amount siphoned off.
What are the mechanisms preventing bad actors from artificially lowering the value assessment of the land they own for LVT calculation purposes?
Given that many landholders are already wealthy and powerful, wouldn't they find a way to game the system in their favour, through political influence or otherwise?
A lot sans improvements is perhaps easier to valuate comparing to the neighbouring lots, so if anybody in vicinity is selling, then it gives us a hint about the market price of lots in the area and helps us spot artificially lowered valuations?
I understand that it's not a new problem, because many current tax schemes may suffer from tax evasion, but if we base all taxation on just unimproved land value, doesn't it make it easier to evade?
Hmm. I still don't buy the "can't pass on the tax" argument, at least on the theory side.
To begin with, a change in the tax rate seems like a perfect Schelling point for landlords to coordinate rent increases. If all rents go up by the same amount, do renters really have much of a choice? Neither supply nor demand immediately change when the tax rate changes...
More generally, my gut instinct would be that shifts in the the way landlords pay taxes would result in changes in rents, such that the landlord's income remains constant (at a minimum). That seems to be what happens when property taxes go up, anyway. So if the overall tax income of an area shifts toward being based on an LVT, then the LVT will increase and other taxes will decrease, and more and more of the tax income will be funneled through rent and landlords. Ultimately, in the case of all taxes being replaced with an LVT, a renter would pay their entire tax burden as part of their rent, passed through their landlord to the government(s). I can't quite imagine that the total tax burden would change too much, which means the result would be that everything we currently pay in income tax withholding and sales tax, would instead be added onto whatever we currently pay for rent. (Or paid directly, to the extent that we actually own some of the property that we use.) (I have no idea what social security withholding would look like.) The upshot being that although we pay the same amount in taxes, superficially it looks like rents go way up.
Is this sort of thing what you're talking about when you mention landlords passing on the LVT to their tenants? Or am I confused and you mean something else?
Another thing: it might be good to distinguish two financial functions of landlords. One is the "speculative" function of simply owning property and hoping that values increase. The other is the ongoing revenue stream from rentals, which is partly offset by maintenance, taxes, and that sort of thing. As I see it, the LVT ignores the second function and aims directly at that first function. So it seems like there'd still be room in that second function for someone to own property, charge rent, use some of that rental income to pay a property management company to handle the day-to-day stuff, and still keep enough of the rental income to be able to live without having to do much work. Right? But from what I've read, I get the impression that there's some confusion between the ideas of eliminating land speculation, and of eliminating the "rentier class" entirely. Which of these do Georgism and the LVT aim for?
Am I correct that the annual rates analyzed are all on the order of the Danish rates, which are annual taxes set at 2% to 3% of land value? In that case, the annual rates in some areas increased by an average of 0.34% and others decreased by an average of 0.26%. I'd be hesitant to assume that full capitalization of relatively small changes in this annual tax rate imply full capitalization of far larger tax rates.
It's perhaps tangential to this blog, but understanding what happened in New Zealand seems quite important. It seems New Zealand achieved a full LVT in 1912, but it became progressively less relevant until it was abolished in 1991. So what went wrong, and how could we avoid repeating those mistakes? A quick Google showed me https://www.ethicaleconomics.org.uk/2017/02/a-history-of-land-value-taxation-in-new-zealand/.
> In fact, many of the strongest opponents of LVT seem opposed precisely because they agree that land is a big deal.
Hello.
> Well, I'm already charging as much as the market will bear. If I charge any more, my tenant will move out. [...] Price signals from the market are telling them to stop producing and do something else.
Yeah, no. The fact you don't respond to very basic economics like transaction costs or the idea that what the market will bear can change in response to taxation is such a fundamental flaw I'm tempted to stop. You don't even talk about demand elasticity! This theory section is really, really bad and ignores mainstream economics on the issue. The normal position is that while the supply of land is completely inelastic the demand is also highly inelastic. You can choose to stop buying cigarettes but you can't stop consuming land because you need to physically exist somewhere.
What determines who pays taxes is demand elasticity. If it's high then the producer eats it (because if they raise prices their customers go elsewhere) while if it's low then the consumer eats it because the producer can raise the price and the consumer has no choice but to pay it. The classic example of this is, and I swear I'm not making this up, rent.
> Even a dedicated cartel like OPEC can't enforce high oil prices by fiat. They do it by cutting off production and driving down the global supply of oil until people are forced to pay the price OPEC wants.
They could. They choose not to because it's cheaper to do it their way. We also do see landlords in aristocratic regimes (where landholders also hold political power) doing exactly that: coordinating to set higher rents. The idea "landlords can't coordinate to raise rents" is ridiculous. I mean, they can't now because it's a crime and the government investigates and prosecutes people for it. But "it's illegal" is different from "it's impossible."
> This phenomenon is known as Ricardo's Law of Rent.
Ricardo's Law of Rent is specifically concerned with land as a factor of production. Ricardo is explicitly excluding domestic land like homes. You are not. Yet you don't bring this up.
> All the leverage is on the side of the tenants, which forces the landlord to eat the tax.
If all the leverage in tenant-landlord relations is on the side of the renters then why don't we see tenants using that to renegotiate lower rents now? This is very opposite of most people's default intuition but you don't really justify it.
> One day, Denmark decided to redraw all its municipal boundaries. Regions that had been under one local government woke up the next day under a different one, immediately adopting a new set of local regulations and rules, including changed tax rates. This caused a large-scale, semi-random shuffling of Land Value Tax rates overnight.
This is extremely confounded. Firstly, Denmark has a huge amount of control over its housing market that would prevent a full free market response. Secondly, it's widely accepted that social services affect property rates. I haven't read the paper but you don't mention such compensation. Is it there? Further, even if it could be proved (and it probably could) that taxes decrease property value, this doesn't mean the rent goes down too. You assert this ("This just means that if you tax land, absent any other interventions, the price of land goes down. The rental income of the land available to the landlord goes down too, which means the landlord is eating the tax and can't pass it on to the tenant.") but I didn't see it in the paper. I'll read it more thoroughly later or you can point it out. But there's plenty of examples of high rent, low value areas. Rent is not a fixed portion of value by any means.
> land taxes do capitalize into land values, whereas the monthly rent level remains unaffected by the land tax.
In fact, one of your studies finds this is the exact effect it causes! (Though I haven't read the study, just your summary.)
> Wyatt cites another source (Pillai 1987) that claims that LVT hasn't worked in developing countries, but notes that the "LVT" imposed there was a flat tax based on land acreage rather than actual land market value.
How else would it be assessed unless you're taxing the value of improvements on the land too?
> That's according to a 1983 article originally published in Fortune Magazine. I don't know how Pennsylvania fares today, given it's been 38 years.
I do. Still bad. As are most of the states with higher property taxes. Those property taxes, by the way, are widely believed to drive up rents because landlords price them into rents. I don't know if it's true econometrically. But I know both tenants and landlords believe that and act as if it's true. Of course, Denmark (the example you give at the beginning) is rent controlled. They've been retreating from that policy but remember how I said that study was hugely confounded? Yeah.
> LVT proponents claim that an LVT can't be passed on to tenants, but Wyatt is saying that if you turn around and spend that LVT money on making your city better and more desirable, then the increased demand for land in your city might more than offset the negative capitalization of the tax into the sales price of land. That's a solid argument. Notice that Wyatt is here implicitly admitting to capitalization of land taxes into land prices; he's just also arguing that there are other effects in play. What Wyatt doesn't realize is that the natural policy conclusion here is...a 100% LVT that recaptures all the added gains to land value from public spending.
No, what you don't realize is that you're assuming public spending is the highest and best use of that money. This is another studied area: the whole crowding out theory, basically whether public or private spending gives better returns. If any tax, not just land taxes, give better returns than they cause in economic damage then they are net positives for society. So this is in no way a unique trait of land taxes. This is in fact the justification for property taxes: they are relatively less damaging and the money can be put to relatively good use. But as I understand it that's not Georgism, that's just mainstream economics. Georgism is the idea they are not damaging AT ALL if levied solely on land and so can be set high without bad economic effects. "Least bad" and "not bad" have a huge chasm between them.
This is true of the section that you can increase land prices by taxation. Yes, you can do that with anything that's positive sum. You can tax businesses and spend it on business incentives such that more businesses get created, for example. At least theoretically, the problem is often that the government can't actually manage to do that.
> Conclusion: Land Value Tax can't be passed on to tenants.
You've completely failed to support this. What you've shown is a bunch of mixed evidence with one example where rent control was in effect and a huge number of confounding social service changes occurred and apparently weren't controlled for.
> If land is truly chronically underassessed, than simply making land assessments more accurate across the board will give you a similar effect to raising the rate of LVT, without touching the nominal tax rate or changing any laws.
No, you couldn't because a lot of the adjustments are based on improvements. If you wanted a pure land tax you'd need to figure out some way to do adjustments and to ban reassessments for improvements. Otherwise the status quo, where building a pool leads to a reassessment and an increase in taxes, will continue. Which is not really all that Georgian.
I can potentially see the argument working for land rents. But presumably when people complain about rents, it's usually about their housing, not their land? And *here*, the supply isn't constant. The cost of housing *will* be driven up - either, the landlords can raise rents outright (which might well work, as everyone has to do it), or there will simply be less housing built until the rents are driven up by scarcity.
Also, is land actually constant in any way that matters? In mere total area, I guess it mostly is (the Dutch would disagree), but is that the important measure? Farmland isn't a constant, for instance - you can create new farmland, or plant forests on your old ones when it doesn't pay. There are remarkably few things that depend exclusively on total land area in a nation, as opposed to what it's for, and what it might be fit for at the margins when the situation changes.
The concept that landowners of different parcels who are using them for different purposes (one has a large garden & orchard, one has a shop, one has several rental units) are going to be each taxed (equally) enough to support the city & state & federal taxes necessary to support the wants of a larger renter class - even disregarding the "eat the rich" and "fuck the landlords" attitude thick in the comments to these posts - doesn't pass the smell test. All the words in the articles aren't really helping - I feel like something major is being ignored or glossed over.
I find two things unconvincing about this article:
1. The studies cited show LVT rate changes being capitalized into the price, which is supposedly theoretically consistent with no rent increase, but there’s no direct demonstration of no rent increase afaict. So the argument that rent wouldn’t increase under an extreme LVT seems still theoretical. (Also, as some other commenters mentioned, some of the studies were in jurisdictions with rent control. If raising rent is illegal, then of course it won’t fall on renters.
2. In practice, property tax rate changes _are_ passed onto renters. It seems a priori unlikely that taxing only the land and not structures would have a different outcome. It should take strong evidence to move off this prior, and the provided evidence is indirect at best,.
On net, this article reduced my credence in Georgism due to the above two factor.s
I don’t get how the tax doesn’t get passed on to tenants. Suppose in a 100% LVT world two landlords build identical buildings, one in the middle of nowhere and another one in a desirable location. Wouldn’t the tenants pay vastly different rents, in each case consisting of the part due to the building (the landlord pockets it) and the part due to the land (the landlord collects it and repays it as the tax)?
If I understand it correctly, not only does the 100% LVT get paid 100% by the tenants, Georgism doesn’t preclude rent hikes in gentrified areas either. The difference is whose pockets those hikes go into, which is no longer the landlords but the state.
I don't know if I'm pointing out the obvious or being dumb here, but wouldn't this kind of policy immediately result in maximum exploitation of all land (i.e all natural) resources? Surely we're already making enough of a mess of the Earth, between contamination, climate change and biodiversity depletion... I'm a bit skeptical of the wisdom of a global policy that would make it impossible virtually for anyone to hold land without exploiting it to the max.
Wyatt mentioned an issue that you touched on very briefly, but seems quite central to the overall claims.
You say "With a Land Value Tax, the owner has to pay that tax every month whether they have a tenant or not. They're already charging the highest amount the market will bear, and as we've already shown, they are unable to change the supply of land."
But the supply of [usable/used] land does change. In a small urban country this is harder to see, but in a large rural country this is actually quite obvious. Consider a brand new town/city (Chicago in 1833 for instance). At the time Chicago was formed, it was an alternative to other places where people may have lived, but we can see even more so that it became a better alternative over time, and the very small central core expanded over time into more and more land. "Chicago" was very small - 4,000 people in 1837 - and more land was added to "Chicago" over the years as the population expanded. I put the name in quotes because people in the 1840s would not have considered the full boundaries of modern day Chicago as part of the town back then - in other words the town expanded to include more land and the supply of land in "Chicago" increased! Right now there are rural areas with completely unused land. The land isn't even a national park or held by a mining company, but may just be completely unused land. Not to say it's not owned, but it's being put to no productive use at all. The LVT on the land would have to be $0, based on current use. Think of the very top of a high mountain, deep within a mountain range or a patch of desert far from any human habitation, if you can't envision anything less severe (though less severe does exist). Over time, humans may find reasons to move closer to this unused land, thereby moving it from "this is worthless land that nobody would ever pay rent to use" to "this land has some rent value" and therefore the supply of land is increased. As a more extreme but more obvious example, consider Mars. If humans were ever to colonize Mars, we would significantly increase the amount of land available.
To push this back towards your point, if the supply of land can change, then we are no longer in a rigid system where taxes can be raised without recourse. This may be approximately true in a developed urban area, so an LVT for Manhattan may be able to work approximately as you describe. Even then, there are alternatives available. Consider the real life example of a significant expansion of New York City residents moving to Connecticut. They are close enough to the city to interact with it as they need, but not deal with the problems of living in the city. In other words, "New York City" actually expanded its supply of land! And we know that happened in the past as well, with northern New Jersey acting as an overflow for NYC proper for many years. If the cost to live in a place is too high (as with your tobacco examples), then the supply of land can actually decrease as well. Consider Detroit, which has significantly contracted from its population peak, and land has gone fallow even within what used to be known as the city proper.
If the supply of land can change, which I think I've demonstrated it both can and does, then it no longer follows that a tax on land behaves differently than other taxes and that it cannot be passed on to tenants. In retrospect I think it's pretty obvious why this would happen. It's the same reason people want to live in cities in the first place, and why urban costs increase so much compared to rural areas. Being close to your job, or city amenities, or whatever reason people want to live in a city, means people will pay more for land within that area, and less for land elsewhere. You can raise taxes on living in a popular city and still see people paying more to live there. Local to me, the best school district in the county has housing prices significantly more than similar houses in worse school districts. Property taxes are also higher in the district with better schools. People still want to live in the more expensive area - in other words the prices are passed on to tenants who want to live in that particular place.
Again, I think LVTs make a lot more sense in dense urban areas, and approximately resemble the theory. But, considering how cities expand and even contract over long periods of time, it's simply not true that the Georgian theory can hold over a long period of time without causing distortions similar to other taxes.
> Georgists assert that landlords cannot pass Land Value Tax (LVT) on to their tenants
Right off that bat, I don't get why I care about this at all. Lots of taxes are "passed on," whether efficient or inefficient. LVT seems efficient to me, so I don't care much if it's passed on or not.
The price of land going down with a LVT seems quite intuitive (why buy land if you can barely make a profit from its rents). However, one of my initial main concerns with a true 85%+ LVT was that the price of land would drastically plummet. This drop seems like it would overwhelm any improvements in the surrounding area made by governments flush with cash.
Governments typically try to avoid harshly punishing a specific area of investment without suitable compensation. Would the solution be to very slowly increase to a very high rate (like over the course of 50 years) and compensate current owners proportional to their current land? Or am I missing something entirely?
> It's just an illustration meant to make a rhetorical point, but now I'm curious to see a real-world version of this superimposed over, say, Houston or Philadelphia or New York City, and based on actual data.
The big problem with this argument is that it is very, very misleading. The tax will get passed on to the tenants, but that doesn't mean their rent will increase necessarily, as the value of the land will drop to the point that the rental income from charging rent plus the value of the tax is equal to the discounted income stream land owners are willing to accept.
The tax gets passed on to the tenants because so long as they want to rent land, they are going to rent the land that pays the tax, and "what the market will bear" goes both ways. Land owners rent at as high a price as the market will bear, and renters rent at as low a price as the market will bear. BOTH matter. Just as you can claim that land owners are already charging as much as possible, it is equally true that renters are paying as little as possible.
Now, if you put an X% tax on all land rental, the price of land will drop as owners sell it to other owners because the cash flow is not high enough to justify owning it. No one is going to rent land at a loss, not for long, so any rental fees are going to include the tax. If the land owners needs to profit 100$ a month to make the price of the land worthwhile instead of buying a stock or something, the tenants are going to have to pay 100/(1-X) every month. (In the case where X%= 85%, tenants have to pay $666.66 a month for the landlord to get 100$.)
Where the new rental equilibrium will be, whether higher monthly rents or lower, no one can know ahead of time. That number is going to depend on how many other investment options land owners have, their relative returns, the opportunities and abilities to change the structures on the land (changing apartments to single homes to commerce, etc.) and all that. No one can claim to know what the new prices will look like without being a liar.
Would an LVT be better than all other forms of taxation? Possibly, although assessment is a real bear.
Is it a magical tax whereby the cost isn't born partly by the sellers and partially by the buyers? Of course not! There is a fixed amount of land to own, and everyone has to live somewhere on it (barring owning a house boat) so people can't change their behavior to avoid the tax much, outside of moving out of high assessment areas. The plus side to that is that it is less distortionary, since people are still going to own and rent land. It does NOT however mean that the renters are not going to be paying those taxes; tax incidence relies on relative elasticity, and land owners have a lot more flexibility in investment vehicles than renters have in whether or not to live on land, so land owners are not going to be the one paying the bulk of the tax.
I have a few questions which probably exposes my lack of understanding.
First, Part 1 showed that all the land being used is really valuable and that a consistent LVT could pay for a lot of really great things. If a severe LVT were imposed though, I think that a lot of people would just stop owning land? Which is obviously the whole point to reduce speculation, but then if nobody wants the land because of the tax burden, then who would pay the LVT to fund the Citizens' Dividend and other nice stuff?
Second, if everyone is selling off their land because they have to pay LVT on it, that seems like it would crash the price of land around it, reducing the value of the land to begin with. So how can we know that the land values calculated in Part 1 would remain accurate (and thus be able to fund all the nice things we want) after a huge LVT is implemented?
Third, wouldn't Georgism applied at scale to a country like the USA result in a huge redistribution of wealth away from urban areas and towards rural areas? Are we ok with that? I think people would be upset about that.
I'd be careful about talking about the current NZ economy because of the distortionary effect of Chinese capital controls. Buying real estate abroad is one of the easier ways to get money out of the country. The distances and size disparities involved mean that this has a big effect on NZ's housing market.
Really enjoying the series so far, thanks for posting these.
A lot of the arguments presented against an LVT strike me as isolated demands for rigor.
Tax laws get changed all the time and it seems entirely reasonable to slowly phase in an LVT. Sharper increases for less efficient taxes have occurred without catastrophic consequences.
The tax bill from an LVT wouldn't be much different from property taxes in the US today. Assuming land value is roughly 70% of property value, we target an 85% tax on land rent, and the market capitalization rate is 5%, an LVT would cost roughly the same as a 3% property tax (70% * 85% * 5%). This is higher than typical rates in the U.S. (~1%) but doesn't seem like it will have terrible effects, especially since it is more efficient than taxing the property directly. Sanity check: median property tax rates were 1.1% in the U.S. and generated $323 billion, something on the order of 3% property tax rates should raise $874 billion, consistent with the low end results in part I (because I'm using the lower market capitalization rate).
The key here is to consider the alternative to an LVT. Many forms of taxation are inefficient, arguments against a LVT have to demonstrate that the LVT is significantly worse than a different tax scheme.
More important than policy details, Georgism raises an interesting question. We can roughly divide everything of economic value into the downstream products of Labor and Land (wealth is labor+land, money is receipts on wealth, capital is invested wealth, etc.). How do we distribute these basic goods?
Distributing labor is pretty straightforward, everyone has a right to use as much or as little of their own labor as they want, in any way they choose (though we might redistribute this somewhat via income taxation).
But how do we distribute land? Land is this weird thing which surrounds us, is crucial to producing economic value, and which nobody has a natural claim to. Should we give it to those who manage it best? Share it equally? Redistribute it to the poor? What taxation scheme would achieve these goals?
I think these broader questions are more important than policy quibbles, especially considering the fact that "Land" applies to far more than just patches of dirt (energy, natural resources, air rights, broadband spectrum, etc.)
I'm very pro-LVT, but one thing has always bothered me: the claim that land scarcity is unresponsive to price signals (ie, supply curve is a vertical line). While true on its face, improvements can make a huge difference to economy of scale, by dividing land value over a larger number of people.
Take a large plot of land currently used for a single-family home; if the value of that land skyrockets (more people willing to pay a premium to live in that area), the market can respond to the new price signal by building multi-family housing, even a skyscraper that houses a thousand people.
The counterpoint would be: there is still a land-value underneath, whether calculated for one resident or a thousand residents. (A single resident who chose not to build a skyscraper would still have to pay LVT based on the opportunity cost, in the same way as an idle land speculator.) All the same: how we do reconcile the claim that land is unresponsive to the price signals of changing demand, when improvements can act as a force multiplier, providing land's location value to a vastly larger quantity of people?
Really been enjoying these posts, great research and writing!
The tenants won't be paying the maximum the property is worth to them, and landlords will make more of an effort to get back what they're losing in tax.
Related - people/businesses work super hard to break even and make a small profit, they don't work as hard to make a bit more profit.
The whole concept of 'pass the costs to renters' is a red herring. Rental prices are a function of supply and demand almost entirely divorced from the cost of inputs.
The cost of inputs for the 'rents' in the Georgian scenario are going to be the LVT and the cost of capital to build/maintain improvements. LVT is obviously a new cost and the cost of capital is also going to increase (interest rates will be higher as land is no longer a relevant piece of collateral).
Given that the cost to supply 'rents' is going to increase dramatically I don't see how the supply of 'rents' doesn't also decline dramatically. How this effects actual rental costs will depend on how demand shifts in response.
My question is, why do Georgists seem to be so prevalent in right-libertarian circles? I know a lot of Georgists and I've met literally all of them in online libertarian capitalist spaces. Conversely, they seem to be almost non-existent in actual progressive and leftist spaces, despite the fact that their goals and plans should ostensibly appeal to people of a left-wing persuasion.
My guess is that it's because Georgism offers a solution for some of the problems with totally unregulated free-market capitalism without throwing out capitalism and markets altogether like the radical leftists want, and without all the taxes and regulations and market distortions inherent to center-left social democratic liberalism. Alternatively, it might be simply because Georgism and libertarianism are both heterodox ideologies that get attacked from the right, left, and center alike, which makes Georgists sympathize with libertarians and vice-versa.
Some numbers. For example, suppose a plot of land can be put to one use only; to bring it from the current state to a state fully suited for the purpose costs $X; this will then yield a rent, of which $Y/yr can be attributed to the land itself and $Z/yr to the improvement. Assuming 100% LVT, how to determine the price of this plot?
I'm sure someone else has already mentioned it, but there is a pretty easy (albeit indirect) sanity check that can be done using American states. In general, states with annual/biannual assessments on present value have both lower housing prices and more stables housing asset values than those with Prop. 13 like distortions. The main exceptions appear to be states like New Jersey that are adjacent to high land value urban centers.
I don't see how tax/cost isn't somewhat passed on to the tenant. If the cost of owning land in cities goes up, then all landlords will have to raise rents. Won't this policy drive people out of the city and into rural US? where land is cheap...
If you actually look at the studies, a number suggest that, in response to the increased property taxes, people increased density of construction - rather than prices going up on a per-unit basis, instead they increased the number of units.
Thus, they lowered unit quality to increase the number of units while selling them for the same price as they did previously - which is exactly what you'd expect to happen.
The flaw with all of these theories is that they ignore the fact that the cost of housing is relative to the price of it - if you increase the cost of housing, you might increase the cost of housing, or you might cause a decrease in the quality of housing while the price remains the same.
In an area where the price of housing is maximized, we would instead expect that the quality of housing would decrease, resulting in less space per person.
Thus you might see no increase in one measured quality (say, price per unit) but see an increase in another (price per square foot).
Moreover, making existing owners eat costs doesn't actually do anything against future owners - if the property is permanently worth less, then new owners will buy it for less money, but the new capitalization will be corrected for the lowered cost, so at best you are just applying a one-time penalty to present property owners, as future ones can compensate for the higher taxes.
> With a Land Value Tax, the owner has to pay that tax every month whether they have a tenant or not. They're already charging the highest amount the market will bear, and as we've already shown, they are unable to change the supply of land. All the leverage is on the side of the tenants, which forces the landlord to eat the tax. The price to buy the land goes down, the price for a tenant to rent it goes down, but the total amount of income the land itself produces ("land rent") stays the same. A portion of it is just being collected by the taxing agency.
I've missed something here. We've stipulated that there is a market rent for the land which maximizes the revenue the land can bring in. Thus, if the land gets taxed, the tax cannot be passed on to renters, because raising the rent on the land would lower total revenue from the land, which is counterproductive for the landlord. This is a perfectly coherent model.
And I agree that it goes on to predict that after the institution of such a tax, the price of land would fall. The land is worth less, so its price will go down.
But how is the price for a tenant to rent it going to go down? If that happened, total pre-tax revenue from the land would fall, and the value of the land would drop even further. Our whole argument above was premised on the idea that the price the tenant pays to rent the land stays exactly the same come hell or high water. The price of _rent_ falling could only happen if the marginal rate of land value tax exceeded 100% at some point.
Eh, Hong Kong has the government control virtually the entire land supply, and charge megacorps incredible prices for very small lots - and yet the megacorps have proven fully able to pass the spiraling cost of land onto homeowners.
I spent so long thinking about this that I've completely missed the discussion. But I think it's worth inserting my comment late just in case anyone else is having the same feeling I am.
I can't work out what this second is arguing at all.
It says, "I'm a landlord...a Land Value Tax is imposed on me...why shouldn't I be able to pass on the tax to the tenant?"
But in fact, the tax should be "passed on" to the tenant (in the case of an ideal, correctly assessed 100% LVT). It should be passed on perfectly to the tenant. The landlord should be able to profit from their improvements to the land. They spend $1M building an office block on it, and now they collect $200,000 more in rent than they could for the bare land, so they profit after 5 years. But the value of the land itself is paid by the tenant to the landlord (as part of the rent); and is paid by the landlord to the state (as the LVT). There should be perfect, 100% pass-through.
So far as I can see, this entire post is arguing the very opposite to what the outcome is.
I *think* that this is a case of bad wording, and what author means is something like this: the use of an LVT does not create large and continuous opportunities for arbitrage which will allow landowners to capture increasing amounts of wealth on an ongoing basis. But I'm not sure, and the arguments below haven't made it clear to me, either.
If anyone else is as confused as me, please let me know; or tell me what I've misunderstood.
Georgism is fatally flawed, and while through painful analysis regarding whether it may "work" or not, the core issue to reflect upon is whether it's moral and thus even worth discussion. In my view it clearly fails for what should be obvious reasons. That this matter has not even been approached here leaves me cold.
Here's two objections - a thought experiment, and a breakdown which demonstrates what seems like an even more fundamental flaw in your thought about landlords vs tenants paying for things.
So let's posit a small country. Let's call it Georgia (heh). It has a capital city, Georgetown, where most of the economic activity is, and a hinterland with farming, resource extraction, tourism, but not much else.
In Georgetown, on average, units rent for $1000/m. This covers the costs to landlords - mortgage, maintenance, insurance, property tax, risk of sitting empty, etc. along with a certain profit, let's say $500/month. The tenants are engaged in economically productive activity that allows them to pay this rent, and their other expenses. Vacancy rates are low because Georgetown is where the jobs are, unless you're a miner/farmer/hotelier.
We replace the property tax with an LVT that increases landlords in Georgetown's costs $100/m. They're now making $400/m. It seems that in your mind, this is where the story stops - landlord costs go up, landlord profit goes down. But why would they accept this?
Now, as units come available on the market, landlords start listing them at $1100/m. They want to maintain the profit level they had before. Tenants who need a place to live shop around, but it seems like pretty much every unit is more expensive than they were before (every landlord has the same idea, because all of their costs went up - no coordination required!). Tenants try to bargain, but the landlords won't budge - they know there are lots of tenants, and that moving out of Georgetown into the hinterland is either economic suicide or such a huge hit to quality of life due to the brutal commute that tenants are extremely averse to doing so. The landlords are willing to eat a month or two without rent in order to "lock in" a higher market rate to offset their increased costs.
Tenants who need a place to live start paying the new $1100/m, because they basically don't have a choice. They need a place to live. They can't move to the hinterland, or they wouldn't be able to work (or they'd have an awful commute that's worth paying $100/m to avoid). There was a previous market equilibrium at $1000/m, but you can't expect that market equilibrium to persist after changing the cost of the inputs.
Who is paying the new tax? The tenants...
---
Let's look at this in an even more fundamental way. Let's break down some hypothetical costs to the landlord pre-LVT on a unit grossing $1000/m:
- mortgage - $300
- property tax - $50
- maintenance - $100
- insurance and other - $50
- profit - $500
Who is paying the mortgage? The tenant. Who is paying the property tax? The tenant. The maintenance? The tenant. Insurance and other? The tenant. The landlord's profit? The tenant. What is the landlord paying for? Nothing.
Now let's consider your proposed scenario where the LVT increases costs to the landlord, but for some reason the landlord can't just increase the rent:
- mortgage - $300
- LVT - $150
- maintenance - $100
- insurance and other - $50
- profit - $400
Who's paying the mortgage? Tenant. Who's paying the maintenance? Tenant. Who's paying the insurance? The tenant. Who's paying for the profit? The tenant. And key - who's paying the LVT? The tenant!
The tenant actually pays for *every single thing* to do with the unit. That's what being a tenant *is* - paying for everything to do with a property, but only enjoying the use of it, and not the ownership. The economic activity of the tenant is the only thing actually generating value in either scenario, and the only thing actually paying for anything.
At best, the LVT could decrease the *margin* the landlord gets from the property, but it can't shift the cost onto the landlord, because in a fundamental way, landlords don't pay any of the costs of the units they own - the tenants do. Landlords don't generate economic activity, they just leverage their capital to exploit the economic activity of tenants.
So, in a way, LVT is like an inverse Pigovian tax: rather than paying the full cost of negative externalities you create, you pay the full value of *positive* externalities you *receive*?
I find fault with your assumption that speculation is bad. Several threads in the comments try to address this by making the claim that speculation creates no value.
A standard counter argument to this is that it creates liquidity, which is a form of value. Furthermore, it allows for price formation which is what allows (currently) to solve the coordination problem of deciding which land to improve - a value.
I think that the core false assumption of most arguments that distinguish speculation from investment is the assumption that there is infinite supply of labor and capital available for improving land. Clearly this is false.
Have you seen the movie Up? An old man is being hounded from his home by developers eager to make money. He just wants to live out his days in the house where he has fond memories of his late wife. How would LVT deal with this? Would he be forced to sell because of revaluation and tax hikes?
Your work is outstanding. I agree land is a big deal. I agree that land taxes will not be fully passed on to renters (though I disagree that none of it will be passed on - supply is not actually fixed in a country like the US with wide-open spaces). I agree that unimproved land value can be reasonably accurately assessed. I am in the process of working through the articles, but there is one glaring question that keeps distracting me from comprehending your work. Maybe this has already been addressed; I haven't read through all of the comments.
It is this: the value of the land to an investor is ultimately a function of the income it will return to that investor. If the government taxes 100% of it, then it is worth nothing (actually less than nothing, because it takes work to fill out tax returns). The classic formula for value = net income / interest rate minus growth (for a perpetuity), with net income being net of taxes and other carrying costs. If you live on it, then it is assessed by the rental cost of something identical to your residence. Land as pure speculation (no income) is a function of resale value, which is a function of rental income to future owners. There is some ambiguity over whether land is also a store of value, like bitcoin or gold, apart from other functions. But that is a debate for another day.
This doesn't necessarily mean that the price of land will go to zero, because there will be elevated demand for land in good locations. But the price will almost certainly be significantly reduced, because there will be no investor demand, except multi-unit operators that earn a spread between rental income and maintenance cost. Resident owners, most of whom are secretly investors (hence the popularity of NIMBYism), will work to avoid higher costs, though telecommuting or relocating.
This also appears to be a back-door nationalization of land - the government doesn't own formal title but claims all financial value. This opens the door to corruption.
"The price to buy the land goes down, the price for a tenant to rent it goes down," - why would rent go down?
In some comments here you seem to contradict this and say that the rents would stay the same - also in the preceding paragraph you seem to suggest that the rent relies only on productivity
"Without a Land Value Tax, the owner of that land can charge rent up to the difference between their land's productivity and the best freely available alternative, establishing the "margin of productivity." This means that as productivity rises, so does the rent. This phenomenon is known as Ricardo's Law of Rent."
Have you ever thought about making multiplayer games to test economic mechanisms like this one? I think it could be crucial. I'm not sure I'd be able to go ahead with Propinquity Cities, for instance, without running test games.
I have skimmed the comments but haven't found this topic addressed head-on: What protection can be put in place for low-income building owners in rapidly gentrifying areas? I understand the long phase-in approach, but they only solves part of the problem. While I'm pretty pro-georgism, I'd like to be able to address this question better. The specific example I'm thinking about is inherited properties in high-cost areas that may be the primary income stream AND place of living for some low-income families. While I agree that these rents are not economically efficient, what we're basically telling folks is that they either need to have the wherewithal to improve their land enough to justify the LVT, or eventually move. The effective result of this is that areas that start off with high LVTs edge out low income people. Not that it isn't happening now, but if you're going to change the status quo, it's sort of on the new system to address the issue
I enjoyed your book review and this series of articles. Thanks for all your work.
I am confused as to why there's no mention of elasticities in this article. If you want to talk about who "really" pays a tax, I think that elasticities is widely considered the most useful and correct foundation for that discussion. In fact, there exists a technical term for who is burdened by a tax: tax incidence. The Wikipedia article on tax incidence seems fine and you might find it useful; it's almost mostly about elasticities.
I think your arguments could be much stronger if you grapple with elasticities. Instead of making tenuous bathtub analogies, you could draw upon general economic principles that are universally recognized as solid. Your opponents would basically have to disagree with the basics of supply and demand, and even non-economists tend to accept supply and demand basics.
I already agree with you about LVT tax incidence, and even I was very doubtful of your bathtub arguments.
Saying that a LVT would almost entirely burden land suppliers doesn't require any specialized arguments that people only see in Georgist advocacy. You have super-basic-and-mainstream economics on your side, and talking about elasticities could show that to your audience.
Is this the book by Donald Hagman which you could not find? I found this by google searching "The Single Tax and Land Use Planning: Henry George Updated" and it is published by the UCLA law review.
Makes sense. But would be a lot easier to heavily tax inheritance. And that would fix the most egregious part of land wealth - that most of it is inherited. Simpler, politically easy, no drag... really wonder why the focus is not on that really.
So rents may not fall, but property prices (including land) would fall significantly? Because the present value of the land becomes zero, so the property is just valued based on the building?
> New Zealand [snip] over the course of the next century that figure dropped all the way to 0.5% in 1965 and 0.3% in 1970
The above completely ignores that land is taxed by the city, the quote is looking at national taxes only.
Example: Christchurch has rates $5,758.16 on $850k property value (land value $540,000 + value of improvements $310,000) and national GST(≈VAT) is $640 of those rates. Median national taxation is $10000. So:
A: Total LVT is $3658 (≈36% of all taxes),
B: national LVT is $406 (≈ 4% of all taxes).
City rates are based on valuation, and valuation is based on market rates, and land value is the free variable because building valuations don't vary across the city. Is that not an LVT?
Thank you for the article.
Magnificent stuff.
What is "the liquidity effect" talked about in Bourassa (1987)?
It's not surprising that so many "rationalists" oppose Georgism. They fancy themselves intelligent creators but actually cannot prosper unless they're allowed to own something that makes them money without any real effort on their part.
Regarding part III, there is a critical subtlety in the question of "can you measure the unimproved value of land cleanly", and how that relates to the idea that the tax has no deadweight loss.
The normal sense of "unimproved value of land" is: if I took this particular parcel of land and knocked down the building, what would it be worth?
A subtle feature of this definition is that the value of land is sensitive to parcel size and is not linear. That is, two separate parcels A and B are worth *more*, added together, than a parcel C combining them would.
The reasons for this is the same as the reason that the LVT notionally raises a lot of money with no deadweight loss: because most of the value of any specific parcel is spillover - a positive externality - from all the improvements on the neighboring structures and really most of the structures in the region. If someone builds an apartment building in one parcel, the neighboring parcel becomes more valuable because you can put retail on it. A developer that owned *both parcels* could internalize this externality by building both structures - but if the parcels are taxed separately, then the developer is actually paying a tax on its structures! It pays a tax on the retail by way of the apartment-land and on the apartment by way of the retail-land.
It's important to note that this is not a hypothetical. In general this sort of thing - buying adjacent parcels and building synergistic structures on them - is fairly common. (There is a lot of it going on in Tysons Corner in Northern Virginia, near where I live.) Even more specifically, this is a norm for the development of suburban subdivisions - a single developer buys up the land for the entire neighborhood, builds a bunch of houses (and amenities like pools, sometimes), sets up an HOA and so on, and then sells the houses. This is profitable because the houses are worth more in a nice little neighborhood than when built one at a time as stragglers off of a main road.
So: if you take parcels as fixed and immutable, there is actually a deadweight loss from an LVT.
Is there a way to get around this? Well, you could allow people to combine parcels for the purpose of appraisal, and try to revalue the combined parcel without *any* of the buildings on it. But this creates a straightforward tax strategy in which ownership of ever-larger contiguous regions is combined so that more and more spillover value gets subtracted out and the tax base is eroded. (In the limit you value the Earth as a single parcel with no buildings at all, and ask "what is the value of a square kilometer of Manhattan if no buildings exist in the entire world?")
The existence of nonzero (and nontrivial) deadweight loss really changes things. The general rule with deadweight loss of taxes is that, for a given tax, it's proportional to the square of the tax rate. So for an LVT with an extremely high rate, this could be quite serious. But in particular this informs the comparison with an equal-revenue property tax.
The LVT falls on land and the property tax falls on land+structures. So the property tax has a larger base and a lower rate, but the structures portion of the base has a big coefficient for the deadweight loss while land just has what we have above + whatever you get from other problems in land valuation. The result is that, unless you actually know all of the relevant parameters, it's ambiguous which of the two taxes has a higher total deadweight loss.
Finally: it's overlooked, but a large, large portion of the tax on structures *also* has no deadweight loss: the tax on structures that already exist. With the exception of a few structures that might actually be torn down, you cannot disincentivize the construction of a building that already exists. So simply eliminating property taxes is a windfall to that particular factor, and - why pay out a windfall? As long as we are doing estimates and approximations, I personally favor reforming existing property taxes to have abatements for new improvements rather than trying to go full LVT.
OK but like, there's not enough houses in America being built to satisfy the demand. So the landlords can just charge whatever they want because there's not really anywhere for people to go? Maybe their property values decrease, so they don't want to sell their houses, but why should that have any effect whatsoever on rent prices? If you implemented this LVT tax, what prevents a landlord from passing it on to the tenant if every other landlord is also passing it on to the tenants, and no one is building new houses because building new houses is illegal? People have to live somewhere and moving is very hard.
I was really hoping this section would be convincing but I just can't see how this would concretely change the situation in America. I guess I just don't believe the Danish study because presumably it's not illegal to build houses there or something. Help me understand how this applies to America. Can you explain, concretely, how in a city where landlords are already gouging people on rents and their taxes go up, and it's illegal to build new houses, that they wouldn't just increase the rents further? What is supposed to happen to the people who are living in that city? What does the person who can never afford a house and is already spending 50% of their income in rent actually do that "gives them leverage"? They can't buy land or a house. Everyone is raising rents everywhere because of this new tax. I predict that in America, the tenants of that city would just eat it and spend 60% of their income on rent instead of 50%, and just give up on ever having kids. How, concretely, are you imaging it could be any different?
Lars, could Donald Hagman's 1965 book, "The Single Tax and Land Use Planning: Henry George Updated," actually be a journal article and not a book?
I couldn't locate a book by that title in several usual places, including WorldCat and AbeBooks. However, HeinOnline offers a preview of a paper in the UCLA Law Review by that author, with that title, in its 1964-65 edition.
https://heinonline.org/HOL/LandingPage?handle=hein.journals/uclalr12&div=49&id=&page=
While this evidence is intriguing, I think you'd still need to red team the new rules to really make a compelling argument that they can't be gamed.
Suppose, for example, that one of the metrics used in practice to assess the value of the property is what other properties sell for versus what they previously sold. Great, that's hard data, even if it requires some interpretation; wear and work and so on. But you could easily imagine a degenerate case where only a couple properties were sold, and a cabal of property owners could gather and drive the prices down on those few sales so their taxes are artificially depressed; they then split the difference with the seller.
I want to believe that it's hard to just raise the rents and soak people, but having watched friends try to buy houses in the last year ... sellers' markets are rough for buyers.
And Goodhart's Law always looms.
So, I was thinking about the observation that Georgist and Pigouvian taxes both have no economic deadweight loss. On the surface this makes perfect sense: force actors to price in negative externalities, and don't let them benefit from things they did nothing to earn or create.
But the Georgist definition of land isn't just a matter of uncreated wealth, because the value of one parcel of land (especially in cities) is primarily a matter of all the other improvements humans have made to surrounding land. On its face a LVT prevents me from benefitting from those positive externalities, which, fine, I didn't create them. But to me this suggests a problem in regards to defining how large each taxed unit should be. If I own a plot of land and build a house on it, then sure, I have no right to earn higher rent just because someone else builds a mall or a highway or a better school nearby. But what if I buy up many plots and build a bunch of things on them that mutually increase the value of one another? It seems like unless I can somehow arrange for all the land I own to get assessed as a single unit, I'm being prevented from benefitting from positive effects of improvements I myself create. An edge case for sure, but developers do, in fact, sometimes buy up large tracts to build mixed use communities for exactly this kind of reason today.
If you own land and rent it as a mobile home park, and suddenly you have to pay an additional $25k/yr, and so does every other mobile home park owner in the country, then of course prices will rise. When labor costs rose over the past year I raised my bids to compensate for it, just like every other contractor did - and they're not making more labor(ers) on cue either. I didn't shrug and tell my gf we were living off rice and beans so that I could have the privilege of dealing with the physical and emotional stress of contrstruction, nor would I react this way to have the privilege of dealing with the stress of owning a mobile home park. If prices didn't rise, then services would be cut - the septic wouldn't be pumped on time and would back up into the trailers (this actually happened at a park down the road from where I grew up until after many years of complaints it was finally condemned), the well pump would burn out and be replaced by a hand pump, and so on. I don't understand how anyone can possibly think otherwise. And if rents do rise or costs are cut, then the tub and flow analogy doesn't hold water either because the flow is essentially cranked up commensurate with the amount siphoned off.
What are the mechanisms preventing bad actors from artificially lowering the value assessment of the land they own for LVT calculation purposes?
Given that many landholders are already wealthy and powerful, wouldn't they find a way to game the system in their favour, through political influence or otherwise?
A lot sans improvements is perhaps easier to valuate comparing to the neighbouring lots, so if anybody in vicinity is selling, then it gives us a hint about the market price of lots in the area and helps us spot artificially lowered valuations?
I understand that it's not a new problem, because many current tax schemes may suffer from tax evasion, but if we base all taxation on just unimproved land value, doesn't it make it easier to evade?
Hmm. I still don't buy the "can't pass on the tax" argument, at least on the theory side.
To begin with, a change in the tax rate seems like a perfect Schelling point for landlords to coordinate rent increases. If all rents go up by the same amount, do renters really have much of a choice? Neither supply nor demand immediately change when the tax rate changes...
More generally, my gut instinct would be that shifts in the the way landlords pay taxes would result in changes in rents, such that the landlord's income remains constant (at a minimum). That seems to be what happens when property taxes go up, anyway. So if the overall tax income of an area shifts toward being based on an LVT, then the LVT will increase and other taxes will decrease, and more and more of the tax income will be funneled through rent and landlords. Ultimately, in the case of all taxes being replaced with an LVT, a renter would pay their entire tax burden as part of their rent, passed through their landlord to the government(s). I can't quite imagine that the total tax burden would change too much, which means the result would be that everything we currently pay in income tax withholding and sales tax, would instead be added onto whatever we currently pay for rent. (Or paid directly, to the extent that we actually own some of the property that we use.) (I have no idea what social security withholding would look like.) The upshot being that although we pay the same amount in taxes, superficially it looks like rents go way up.
Is this sort of thing what you're talking about when you mention landlords passing on the LVT to their tenants? Or am I confused and you mean something else?
Another thing: it might be good to distinguish two financial functions of landlords. One is the "speculative" function of simply owning property and hoping that values increase. The other is the ongoing revenue stream from rentals, which is partly offset by maintenance, taxes, and that sort of thing. As I see it, the LVT ignores the second function and aims directly at that first function. So it seems like there'd still be room in that second function for someone to own property, charge rent, use some of that rental income to pay a property management company to handle the day-to-day stuff, and still keep enough of the rental income to be able to live without having to do much work. Right? But from what I've read, I get the impression that there's some confusion between the ideas of eliminating land speculation, and of eliminating the "rentier class" entirely. Which of these do Georgism and the LVT aim for?
Am I correct that the annual rates analyzed are all on the order of the Danish rates, which are annual taxes set at 2% to 3% of land value? In that case, the annual rates in some areas increased by an average of 0.34% and others decreased by an average of 0.26%. I'd be hesitant to assume that full capitalization of relatively small changes in this annual tax rate imply full capitalization of far larger tax rates.
It's perhaps tangential to this blog, but understanding what happened in New Zealand seems quite important. It seems New Zealand achieved a full LVT in 1912, but it became progressively less relevant until it was abolished in 1991. So what went wrong, and how could we avoid repeating those mistakes? A quick Google showed me https://www.ethicaleconomics.org.uk/2017/02/a-history-of-land-value-taxation-in-new-zealand/.
> In fact, many of the strongest opponents of LVT seem opposed precisely because they agree that land is a big deal.
Hello.
> Well, I'm already charging as much as the market will bear. If I charge any more, my tenant will move out. [...] Price signals from the market are telling them to stop producing and do something else.
Yeah, no. The fact you don't respond to very basic economics like transaction costs or the idea that what the market will bear can change in response to taxation is such a fundamental flaw I'm tempted to stop. You don't even talk about demand elasticity! This theory section is really, really bad and ignores mainstream economics on the issue. The normal position is that while the supply of land is completely inelastic the demand is also highly inelastic. You can choose to stop buying cigarettes but you can't stop consuming land because you need to physically exist somewhere.
What determines who pays taxes is demand elasticity. If it's high then the producer eats it (because if they raise prices their customers go elsewhere) while if it's low then the consumer eats it because the producer can raise the price and the consumer has no choice but to pay it. The classic example of this is, and I swear I'm not making this up, rent.
> Even a dedicated cartel like OPEC can't enforce high oil prices by fiat. They do it by cutting off production and driving down the global supply of oil until people are forced to pay the price OPEC wants.
They could. They choose not to because it's cheaper to do it their way. We also do see landlords in aristocratic regimes (where landholders also hold political power) doing exactly that: coordinating to set higher rents. The idea "landlords can't coordinate to raise rents" is ridiculous. I mean, they can't now because it's a crime and the government investigates and prosecutes people for it. But "it's illegal" is different from "it's impossible."
> This phenomenon is known as Ricardo's Law of Rent.
Ricardo's Law of Rent is specifically concerned with land as a factor of production. Ricardo is explicitly excluding domestic land like homes. You are not. Yet you don't bring this up.
> All the leverage is on the side of the tenants, which forces the landlord to eat the tax.
If all the leverage in tenant-landlord relations is on the side of the renters then why don't we see tenants using that to renegotiate lower rents now? This is very opposite of most people's default intuition but you don't really justify it.
> One day, Denmark decided to redraw all its municipal boundaries. Regions that had been under one local government woke up the next day under a different one, immediately adopting a new set of local regulations and rules, including changed tax rates. This caused a large-scale, semi-random shuffling of Land Value Tax rates overnight.
This is extremely confounded. Firstly, Denmark has a huge amount of control over its housing market that would prevent a full free market response. Secondly, it's widely accepted that social services affect property rates. I haven't read the paper but you don't mention such compensation. Is it there? Further, even if it could be proved (and it probably could) that taxes decrease property value, this doesn't mean the rent goes down too. You assert this ("This just means that if you tax land, absent any other interventions, the price of land goes down. The rental income of the land available to the landlord goes down too, which means the landlord is eating the tax and can't pass it on to the tenant.") but I didn't see it in the paper. I'll read it more thoroughly later or you can point it out. But there's plenty of examples of high rent, low value areas. Rent is not a fixed portion of value by any means.
> land taxes do capitalize into land values, whereas the monthly rent level remains unaffected by the land tax.
In fact, one of your studies finds this is the exact effect it causes! (Though I haven't read the study, just your summary.)
> Wyatt cites another source (Pillai 1987) that claims that LVT hasn't worked in developing countries, but notes that the "LVT" imposed there was a flat tax based on land acreage rather than actual land market value.
How else would it be assessed unless you're taxing the value of improvements on the land too?
> That's according to a 1983 article originally published in Fortune Magazine. I don't know how Pennsylvania fares today, given it's been 38 years.
I do. Still bad. As are most of the states with higher property taxes. Those property taxes, by the way, are widely believed to drive up rents because landlords price them into rents. I don't know if it's true econometrically. But I know both tenants and landlords believe that and act as if it's true. Of course, Denmark (the example you give at the beginning) is rent controlled. They've been retreating from that policy but remember how I said that study was hugely confounded? Yeah.
> LVT proponents claim that an LVT can't be passed on to tenants, but Wyatt is saying that if you turn around and spend that LVT money on making your city better and more desirable, then the increased demand for land in your city might more than offset the negative capitalization of the tax into the sales price of land. That's a solid argument. Notice that Wyatt is here implicitly admitting to capitalization of land taxes into land prices; he's just also arguing that there are other effects in play. What Wyatt doesn't realize is that the natural policy conclusion here is...a 100% LVT that recaptures all the added gains to land value from public spending.
No, what you don't realize is that you're assuming public spending is the highest and best use of that money. This is another studied area: the whole crowding out theory, basically whether public or private spending gives better returns. If any tax, not just land taxes, give better returns than they cause in economic damage then they are net positives for society. So this is in no way a unique trait of land taxes. This is in fact the justification for property taxes: they are relatively less damaging and the money can be put to relatively good use. But as I understand it that's not Georgism, that's just mainstream economics. Georgism is the idea they are not damaging AT ALL if levied solely on land and so can be set high without bad economic effects. "Least bad" and "not bad" have a huge chasm between them.
This is true of the section that you can increase land prices by taxation. Yes, you can do that with anything that's positive sum. You can tax businesses and spend it on business incentives such that more businesses get created, for example. At least theoretically, the problem is often that the government can't actually manage to do that.
> Conclusion: Land Value Tax can't be passed on to tenants.
You've completely failed to support this. What you've shown is a bunch of mixed evidence with one example where rent control was in effect and a huge number of confounding social service changes occurred and apparently weren't controlled for.
> If land is truly chronically underassessed, than simply making land assessments more accurate across the board will give you a similar effect to raising the rate of LVT, without touching the nominal tax rate or changing any laws.
No, you couldn't because a lot of the adjustments are based on improvements. If you wanted a pure land tax you'd need to figure out some way to do adjustments and to ban reassessments for improvements. Otherwise the status quo, where building a pool leads to a reassessment and an increase in taxes, will continue. Which is not really all that Georgian.
I can potentially see the argument working for land rents. But presumably when people complain about rents, it's usually about their housing, not their land? And *here*, the supply isn't constant. The cost of housing *will* be driven up - either, the landlords can raise rents outright (which might well work, as everyone has to do it), or there will simply be less housing built until the rents are driven up by scarcity.
Also, is land actually constant in any way that matters? In mere total area, I guess it mostly is (the Dutch would disagree), but is that the important measure? Farmland isn't a constant, for instance - you can create new farmland, or plant forests on your old ones when it doesn't pay. There are remarkably few things that depend exclusively on total land area in a nation, as opposed to what it's for, and what it might be fit for at the margins when the situation changes.
The concept that landowners of different parcels who are using them for different purposes (one has a large garden & orchard, one has a shop, one has several rental units) are going to be each taxed (equally) enough to support the city & state & federal taxes necessary to support the wants of a larger renter class - even disregarding the "eat the rich" and "fuck the landlords" attitude thick in the comments to these posts - doesn't pass the smell test. All the words in the articles aren't really helping - I feel like something major is being ignored or glossed over.
I find two things unconvincing about this article:
1. The studies cited show LVT rate changes being capitalized into the price, which is supposedly theoretically consistent with no rent increase, but there’s no direct demonstration of no rent increase afaict. So the argument that rent wouldn’t increase under an extreme LVT seems still theoretical. (Also, as some other commenters mentioned, some of the studies were in jurisdictions with rent control. If raising rent is illegal, then of course it won’t fall on renters.
2. In practice, property tax rate changes _are_ passed onto renters. It seems a priori unlikely that taxing only the land and not structures would have a different outcome. It should take strong evidence to move off this prior, and the provided evidence is indirect at best,.
On net, this article reduced my credence in Georgism due to the above two factor.s
> Note the "per mille" – 20.6 per mille = 2.6 per cent, etc.
*2.06 per cent
I don’t get how the tax doesn’t get passed on to tenants. Suppose in a 100% LVT world two landlords build identical buildings, one in the middle of nowhere and another one in a desirable location. Wouldn’t the tenants pay vastly different rents, in each case consisting of the part due to the building (the landlord pockets it) and the part due to the land (the landlord collects it and repays it as the tax)?
If I understand it correctly, not only does the 100% LVT get paid 100% by the tenants, Georgism doesn’t preclude rent hikes in gentrified areas either. The difference is whose pockets those hikes go into, which is no longer the landlords but the state.
I don't know if I'm pointing out the obvious or being dumb here, but wouldn't this kind of policy immediately result in maximum exploitation of all land (i.e all natural) resources? Surely we're already making enough of a mess of the Earth, between contamination, climate change and biodiversity depletion... I'm a bit skeptical of the wisdom of a global policy that would make it impossible virtually for anyone to hold land without exploiting it to the max.
Wyatt mentioned an issue that you touched on very briefly, but seems quite central to the overall claims.
You say "With a Land Value Tax, the owner has to pay that tax every month whether they have a tenant or not. They're already charging the highest amount the market will bear, and as we've already shown, they are unable to change the supply of land."
But the supply of [usable/used] land does change. In a small urban country this is harder to see, but in a large rural country this is actually quite obvious. Consider a brand new town/city (Chicago in 1833 for instance). At the time Chicago was formed, it was an alternative to other places where people may have lived, but we can see even more so that it became a better alternative over time, and the very small central core expanded over time into more and more land. "Chicago" was very small - 4,000 people in 1837 - and more land was added to "Chicago" over the years as the population expanded. I put the name in quotes because people in the 1840s would not have considered the full boundaries of modern day Chicago as part of the town back then - in other words the town expanded to include more land and the supply of land in "Chicago" increased! Right now there are rural areas with completely unused land. The land isn't even a national park or held by a mining company, but may just be completely unused land. Not to say it's not owned, but it's being put to no productive use at all. The LVT on the land would have to be $0, based on current use. Think of the very top of a high mountain, deep within a mountain range or a patch of desert far from any human habitation, if you can't envision anything less severe (though less severe does exist). Over time, humans may find reasons to move closer to this unused land, thereby moving it from "this is worthless land that nobody would ever pay rent to use" to "this land has some rent value" and therefore the supply of land is increased. As a more extreme but more obvious example, consider Mars. If humans were ever to colonize Mars, we would significantly increase the amount of land available.
To push this back towards your point, if the supply of land can change, then we are no longer in a rigid system where taxes can be raised without recourse. This may be approximately true in a developed urban area, so an LVT for Manhattan may be able to work approximately as you describe. Even then, there are alternatives available. Consider the real life example of a significant expansion of New York City residents moving to Connecticut. They are close enough to the city to interact with it as they need, but not deal with the problems of living in the city. In other words, "New York City" actually expanded its supply of land! And we know that happened in the past as well, with northern New Jersey acting as an overflow for NYC proper for many years. If the cost to live in a place is too high (as with your tobacco examples), then the supply of land can actually decrease as well. Consider Detroit, which has significantly contracted from its population peak, and land has gone fallow even within what used to be known as the city proper.
If the supply of land can change, which I think I've demonstrated it both can and does, then it no longer follows that a tax on land behaves differently than other taxes and that it cannot be passed on to tenants. In retrospect I think it's pretty obvious why this would happen. It's the same reason people want to live in cities in the first place, and why urban costs increase so much compared to rural areas. Being close to your job, or city amenities, or whatever reason people want to live in a city, means people will pay more for land within that area, and less for land elsewhere. You can raise taxes on living in a popular city and still see people paying more to live there. Local to me, the best school district in the county has housing prices significantly more than similar houses in worse school districts. Property taxes are also higher in the district with better schools. People still want to live in the more expensive area - in other words the prices are passed on to tenants who want to live in that particular place.
Again, I think LVTs make a lot more sense in dense urban areas, and approximately resemble the theory. But, considering how cities expand and even contract over long periods of time, it's simply not true that the Georgian theory can hold over a long period of time without causing distortions similar to other taxes.
> Georgists assert that landlords cannot pass Land Value Tax (LVT) on to their tenants
Right off that bat, I don't get why I care about this at all. Lots of taxes are "passed on," whether efficient or inefficient. LVT seems efficient to me, so I don't care much if it's passed on or not.
The price of land going down with a LVT seems quite intuitive (why buy land if you can barely make a profit from its rents). However, one of my initial main concerns with a true 85%+ LVT was that the price of land would drastically plummet. This drop seems like it would overwhelm any improvements in the surrounding area made by governments flush with cash.
Governments typically try to avoid harshly punishing a specific area of investment without suitable compensation. Would the solution be to very slowly increase to a very high rate (like over the course of 50 years) and compensate current owners proportional to their current land? Or am I missing something entirely?
> It's just an illustration meant to make a rhetorical point, but now I'm curious to see a real-world version of this superimposed over, say, Houston or Philadelphia or New York City, and based on actual data.
Check out https://www.officialdata.org/ca-property-tax/#37.83470860338346,-122.26070885080847,17 ; many San Francisco numbers are nuts (and hugely variable, as you would expect from Prop 13)
The big problem with this argument is that it is very, very misleading. The tax will get passed on to the tenants, but that doesn't mean their rent will increase necessarily, as the value of the land will drop to the point that the rental income from charging rent plus the value of the tax is equal to the discounted income stream land owners are willing to accept.
The tax gets passed on to the tenants because so long as they want to rent land, they are going to rent the land that pays the tax, and "what the market will bear" goes both ways. Land owners rent at as high a price as the market will bear, and renters rent at as low a price as the market will bear. BOTH matter. Just as you can claim that land owners are already charging as much as possible, it is equally true that renters are paying as little as possible.
Now, if you put an X% tax on all land rental, the price of land will drop as owners sell it to other owners because the cash flow is not high enough to justify owning it. No one is going to rent land at a loss, not for long, so any rental fees are going to include the tax. If the land owners needs to profit 100$ a month to make the price of the land worthwhile instead of buying a stock or something, the tenants are going to have to pay 100/(1-X) every month. (In the case where X%= 85%, tenants have to pay $666.66 a month for the landlord to get 100$.)
Where the new rental equilibrium will be, whether higher monthly rents or lower, no one can know ahead of time. That number is going to depend on how many other investment options land owners have, their relative returns, the opportunities and abilities to change the structures on the land (changing apartments to single homes to commerce, etc.) and all that. No one can claim to know what the new prices will look like without being a liar.
Would an LVT be better than all other forms of taxation? Possibly, although assessment is a real bear.
Is it a magical tax whereby the cost isn't born partly by the sellers and partially by the buyers? Of course not! There is a fixed amount of land to own, and everyone has to live somewhere on it (barring owning a house boat) so people can't change their behavior to avoid the tax much, outside of moving out of high assessment areas. The plus side to that is that it is less distortionary, since people are still going to own and rent land. It does NOT however mean that the renters are not going to be paying those taxes; tax incidence relies on relative elasticity, and land owners have a lot more flexibility in investment vehicles than renters have in whether or not to live on land, so land owners are not going to be the one paying the bulk of the tax.
I have a few questions which probably exposes my lack of understanding.
First, Part 1 showed that all the land being used is really valuable and that a consistent LVT could pay for a lot of really great things. If a severe LVT were imposed though, I think that a lot of people would just stop owning land? Which is obviously the whole point to reduce speculation, but then if nobody wants the land because of the tax burden, then who would pay the LVT to fund the Citizens' Dividend and other nice stuff?
Second, if everyone is selling off their land because they have to pay LVT on it, that seems like it would crash the price of land around it, reducing the value of the land to begin with. So how can we know that the land values calculated in Part 1 would remain accurate (and thus be able to fund all the nice things we want) after a huge LVT is implemented?
Third, wouldn't Georgism applied at scale to a country like the USA result in a huge redistribution of wealth away from urban areas and towards rural areas? Are we ok with that? I think people would be upset about that.
I'd be careful about talking about the current NZ economy because of the distortionary effect of Chinese capital controls. Buying real estate abroad is one of the easier ways to get money out of the country. The distances and size disparities involved mean that this has a big effect on NZ's housing market.
Really enjoying the series so far, thanks for posting these.
A lot of the arguments presented against an LVT strike me as isolated demands for rigor.
Tax laws get changed all the time and it seems entirely reasonable to slowly phase in an LVT. Sharper increases for less efficient taxes have occurred without catastrophic consequences.
The tax bill from an LVT wouldn't be much different from property taxes in the US today. Assuming land value is roughly 70% of property value, we target an 85% tax on land rent, and the market capitalization rate is 5%, an LVT would cost roughly the same as a 3% property tax (70% * 85% * 5%). This is higher than typical rates in the U.S. (~1%) but doesn't seem like it will have terrible effects, especially since it is more efficient than taxing the property directly. Sanity check: median property tax rates were 1.1% in the U.S. and generated $323 billion, something on the order of 3% property tax rates should raise $874 billion, consistent with the low end results in part I (because I'm using the lower market capitalization rate).
The key here is to consider the alternative to an LVT. Many forms of taxation are inefficient, arguments against a LVT have to demonstrate that the LVT is significantly worse than a different tax scheme.
More important than policy details, Georgism raises an interesting question. We can roughly divide everything of economic value into the downstream products of Labor and Land (wealth is labor+land, money is receipts on wealth, capital is invested wealth, etc.). How do we distribute these basic goods?
Distributing labor is pretty straightforward, everyone has a right to use as much or as little of their own labor as they want, in any way they choose (though we might redistribute this somewhat via income taxation).
But how do we distribute land? Land is this weird thing which surrounds us, is crucial to producing economic value, and which nobody has a natural claim to. Should we give it to those who manage it best? Share it equally? Redistribute it to the poor? What taxation scheme would achieve these goals?
I think these broader questions are more important than policy quibbles, especially considering the fact that "Land" applies to far more than just patches of dirt (energy, natural resources, air rights, broadband spectrum, etc.)
I'm very pro-LVT, but one thing has always bothered me: the claim that land scarcity is unresponsive to price signals (ie, supply curve is a vertical line). While true on its face, improvements can make a huge difference to economy of scale, by dividing land value over a larger number of people.
Take a large plot of land currently used for a single-family home; if the value of that land skyrockets (more people willing to pay a premium to live in that area), the market can respond to the new price signal by building multi-family housing, even a skyscraper that houses a thousand people.
The counterpoint would be: there is still a land-value underneath, whether calculated for one resident or a thousand residents. (A single resident who chose not to build a skyscraper would still have to pay LVT based on the opportunity cost, in the same way as an idle land speculator.) All the same: how we do reconcile the claim that land is unresponsive to the price signals of changing demand, when improvements can act as a force multiplier, providing land's location value to a vastly larger quantity of people?
Really been enjoying these posts, great research and writing!
In the real world and the short-medium term..
The tenants won't be paying the maximum the property is worth to them, and landlords will make more of an effort to get back what they're losing in tax.
Related - people/businesses work super hard to break even and make a small profit, they don't work as hard to make a bit more profit.
(All for Georgism, though)
The whole concept of 'pass the costs to renters' is a red herring. Rental prices are a function of supply and demand almost entirely divorced from the cost of inputs.
The cost of inputs for the 'rents' in the Georgian scenario are going to be the LVT and the cost of capital to build/maintain improvements. LVT is obviously a new cost and the cost of capital is also going to increase (interest rates will be higher as land is no longer a relevant piece of collateral).
Given that the cost to supply 'rents' is going to increase dramatically I don't see how the supply of 'rents' doesn't also decline dramatically. How this effects actual rental costs will depend on how demand shifts in response.
My question is, why do Georgists seem to be so prevalent in right-libertarian circles? I know a lot of Georgists and I've met literally all of them in online libertarian capitalist spaces. Conversely, they seem to be almost non-existent in actual progressive and leftist spaces, despite the fact that their goals and plans should ostensibly appeal to people of a left-wing persuasion.
My guess is that it's because Georgism offers a solution for some of the problems with totally unregulated free-market capitalism without throwing out capitalism and markets altogether like the radical leftists want, and without all the taxes and regulations and market distortions inherent to center-left social democratic liberalism. Alternatively, it might be simply because Georgism and libertarianism are both heterodox ideologies that get attacked from the right, left, and center alike, which makes Georgists sympathize with libertarians and vice-versa.
Article Part IV Request:
Some numbers. For example, suppose a plot of land can be put to one use only; to bring it from the current state to a state fully suited for the purpose costs $X; this will then yield a rent, of which $Y/yr can be attributed to the land itself and $Z/yr to the improvement. Assuming 100% LVT, how to determine the price of this plot?
I'm sure someone else has already mentioned it, but there is a pretty easy (albeit indirect) sanity check that can be done using American states. In general, states with annual/biannual assessments on present value have both lower housing prices and more stables housing asset values than those with Prop. 13 like distortions. The main exceptions appear to be states like New Jersey that are adjacent to high land value urban centers.
I don't see how tax/cost isn't somewhat passed on to the tenant. If the cost of owning land in cities goes up, then all landlords will have to raise rents. Won't this policy drive people out of the city and into rural US? where land is cheap...
You missed the forest for the trees here.
If you actually look at the studies, a number suggest that, in response to the increased property taxes, people increased density of construction - rather than prices going up on a per-unit basis, instead they increased the number of units.
Thus, they lowered unit quality to increase the number of units while selling them for the same price as they did previously - which is exactly what you'd expect to happen.
The flaw with all of these theories is that they ignore the fact that the cost of housing is relative to the price of it - if you increase the cost of housing, you might increase the cost of housing, or you might cause a decrease in the quality of housing while the price remains the same.
In an area where the price of housing is maximized, we would instead expect that the quality of housing would decrease, resulting in less space per person.
Thus you might see no increase in one measured quality (say, price per unit) but see an increase in another (price per square foot).
Moreover, making existing owners eat costs doesn't actually do anything against future owners - if the property is permanently worth less, then new owners will buy it for less money, but the new capitalization will be corrected for the lowered cost, so at best you are just applying a one-time penalty to present property owners, as future ones can compensate for the higher taxes.
> With a Land Value Tax, the owner has to pay that tax every month whether they have a tenant or not. They're already charging the highest amount the market will bear, and as we've already shown, they are unable to change the supply of land. All the leverage is on the side of the tenants, which forces the landlord to eat the tax. The price to buy the land goes down, the price for a tenant to rent it goes down, but the total amount of income the land itself produces ("land rent") stays the same. A portion of it is just being collected by the taxing agency.
I've missed something here. We've stipulated that there is a market rent for the land which maximizes the revenue the land can bring in. Thus, if the land gets taxed, the tax cannot be passed on to renters, because raising the rent on the land would lower total revenue from the land, which is counterproductive for the landlord. This is a perfectly coherent model.
And I agree that it goes on to predict that after the institution of such a tax, the price of land would fall. The land is worth less, so its price will go down.
But how is the price for a tenant to rent it going to go down? If that happened, total pre-tax revenue from the land would fall, and the value of the land would drop even further. Our whole argument above was premised on the idea that the price the tenant pays to rent the land stays exactly the same come hell or high water. The price of _rent_ falling could only happen if the marginal rate of land value tax exceeded 100% at some point.
Eh, Hong Kong has the government control virtually the entire land supply, and charge megacorps incredible prices for very small lots - and yet the megacorps have proven fully able to pass the spiraling cost of land onto homeowners.
I spent so long thinking about this that I've completely missed the discussion. But I think it's worth inserting my comment late just in case anyone else is having the same feeling I am.
I can't work out what this second is arguing at all.
It says, "I'm a landlord...a Land Value Tax is imposed on me...why shouldn't I be able to pass on the tax to the tenant?"
But in fact, the tax should be "passed on" to the tenant (in the case of an ideal, correctly assessed 100% LVT). It should be passed on perfectly to the tenant. The landlord should be able to profit from their improvements to the land. They spend $1M building an office block on it, and now they collect $200,000 more in rent than they could for the bare land, so they profit after 5 years. But the value of the land itself is paid by the tenant to the landlord (as part of the rent); and is paid by the landlord to the state (as the LVT). There should be perfect, 100% pass-through.
So far as I can see, this entire post is arguing the very opposite to what the outcome is.
I *think* that this is a case of bad wording, and what author means is something like this: the use of an LVT does not create large and continuous opportunities for arbitrage which will allow landowners to capture increasing amounts of wealth on an ongoing basis. But I'm not sure, and the arguments below haven't made it clear to me, either.
If anyone else is as confused as me, please let me know; or tell me what I've misunderstood.
Thank you!
Georgism is fatally flawed, and while through painful analysis regarding whether it may "work" or not, the core issue to reflect upon is whether it's moral and thus even worth discussion. In my view it clearly fails for what should be obvious reasons. That this matter has not even been approached here leaves me cold.
Here's two objections - a thought experiment, and a breakdown which demonstrates what seems like an even more fundamental flaw in your thought about landlords vs tenants paying for things.
So let's posit a small country. Let's call it Georgia (heh). It has a capital city, Georgetown, where most of the economic activity is, and a hinterland with farming, resource extraction, tourism, but not much else.
In Georgetown, on average, units rent for $1000/m. This covers the costs to landlords - mortgage, maintenance, insurance, property tax, risk of sitting empty, etc. along with a certain profit, let's say $500/month. The tenants are engaged in economically productive activity that allows them to pay this rent, and their other expenses. Vacancy rates are low because Georgetown is where the jobs are, unless you're a miner/farmer/hotelier.
We replace the property tax with an LVT that increases landlords in Georgetown's costs $100/m. They're now making $400/m. It seems that in your mind, this is where the story stops - landlord costs go up, landlord profit goes down. But why would they accept this?
Now, as units come available on the market, landlords start listing them at $1100/m. They want to maintain the profit level they had before. Tenants who need a place to live shop around, but it seems like pretty much every unit is more expensive than they were before (every landlord has the same idea, because all of their costs went up - no coordination required!). Tenants try to bargain, but the landlords won't budge - they know there are lots of tenants, and that moving out of Georgetown into the hinterland is either economic suicide or such a huge hit to quality of life due to the brutal commute that tenants are extremely averse to doing so. The landlords are willing to eat a month or two without rent in order to "lock in" a higher market rate to offset their increased costs.
Tenants who need a place to live start paying the new $1100/m, because they basically don't have a choice. They need a place to live. They can't move to the hinterland, or they wouldn't be able to work (or they'd have an awful commute that's worth paying $100/m to avoid). There was a previous market equilibrium at $1000/m, but you can't expect that market equilibrium to persist after changing the cost of the inputs.
Who is paying the new tax? The tenants...
---
Let's look at this in an even more fundamental way. Let's break down some hypothetical costs to the landlord pre-LVT on a unit grossing $1000/m:
- mortgage - $300
- property tax - $50
- maintenance - $100
- insurance and other - $50
- profit - $500
Who is paying the mortgage? The tenant. Who is paying the property tax? The tenant. The maintenance? The tenant. Insurance and other? The tenant. The landlord's profit? The tenant. What is the landlord paying for? Nothing.
Now let's consider your proposed scenario where the LVT increases costs to the landlord, but for some reason the landlord can't just increase the rent:
- mortgage - $300
- LVT - $150
- maintenance - $100
- insurance and other - $50
- profit - $400
Who's paying the mortgage? Tenant. Who's paying the maintenance? Tenant. Who's paying the insurance? The tenant. Who's paying for the profit? The tenant. And key - who's paying the LVT? The tenant!
The tenant actually pays for *every single thing* to do with the unit. That's what being a tenant *is* - paying for everything to do with a property, but only enjoying the use of it, and not the ownership. The economic activity of the tenant is the only thing actually generating value in either scenario, and the only thing actually paying for anything.
At best, the LVT could decrease the *margin* the landlord gets from the property, but it can't shift the cost onto the landlord, because in a fundamental way, landlords don't pay any of the costs of the units they own - the tenants do. Landlords don't generate economic activity, they just leverage their capital to exploit the economic activity of tenants.
So, in a way, LVT is like an inverse Pigovian tax: rather than paying the full cost of negative externalities you create, you pay the full value of *positive* externalities you *receive*?
Thanks for the interesting articles.
I find fault with your assumption that speculation is bad. Several threads in the comments try to address this by making the claim that speculation creates no value.
A standard counter argument to this is that it creates liquidity, which is a form of value. Furthermore, it allows for price formation which is what allows (currently) to solve the coordination problem of deciding which land to improve - a value.
I think that the core false assumption of most arguments that distinguish speculation from investment is the assumption that there is infinite supply of labor and capital available for improving land. Clearly this is false.
Have you seen the movie Up? An old man is being hounded from his home by developers eager to make money. He just wants to live out his days in the house where he has fond memories of his late wife. How would LVT deal with this? Would he be forced to sell because of revaluation and tax hikes?
Your work is outstanding. I agree land is a big deal. I agree that land taxes will not be fully passed on to renters (though I disagree that none of it will be passed on - supply is not actually fixed in a country like the US with wide-open spaces). I agree that unimproved land value can be reasonably accurately assessed. I am in the process of working through the articles, but there is one glaring question that keeps distracting me from comprehending your work. Maybe this has already been addressed; I haven't read through all of the comments.
It is this: the value of the land to an investor is ultimately a function of the income it will return to that investor. If the government taxes 100% of it, then it is worth nothing (actually less than nothing, because it takes work to fill out tax returns). The classic formula for value = net income / interest rate minus growth (for a perpetuity), with net income being net of taxes and other carrying costs. If you live on it, then it is assessed by the rental cost of something identical to your residence. Land as pure speculation (no income) is a function of resale value, which is a function of rental income to future owners. There is some ambiguity over whether land is also a store of value, like bitcoin or gold, apart from other functions. But that is a debate for another day.
This doesn't necessarily mean that the price of land will go to zero, because there will be elevated demand for land in good locations. But the price will almost certainly be significantly reduced, because there will be no investor demand, except multi-unit operators that earn a spread between rental income and maintenance cost. Resident owners, most of whom are secretly investors (hence the popularity of NIMBYism), will work to avoid higher costs, though telecommuting or relocating.
This also appears to be a back-door nationalization of land - the government doesn't own formal title but claims all financial value. This opens the door to corruption.
A key benefit is that NIMBYism will evaporate.
"The price to buy the land goes down, the price for a tenant to rent it goes down," - why would rent go down?
In some comments here you seem to contradict this and say that the rents would stay the same - also in the preceding paragraph you seem to suggest that the rent relies only on productivity
"Without a Land Value Tax, the owner of that land can charge rent up to the difference between their land's productivity and the best freely available alternative, establishing the "margin of productivity." This means that as productivity rises, so does the rent. This phenomenon is known as Ricardo's Law of Rent."
Have you ever thought about making multiplayer games to test economic mechanisms like this one? I think it could be crucial. I'm not sure I'd be able to go ahead with Propinquity Cities, for instance, without running test games.
I have skimmed the comments but haven't found this topic addressed head-on: What protection can be put in place for low-income building owners in rapidly gentrifying areas? I understand the long phase-in approach, but they only solves part of the problem. While I'm pretty pro-georgism, I'd like to be able to address this question better. The specific example I'm thinking about is inherited properties in high-cost areas that may be the primary income stream AND place of living for some low-income families. While I agree that these rents are not economically efficient, what we're basically telling folks is that they either need to have the wherewithal to improve their land enough to justify the LVT, or eventually move. The effective result of this is that areas that start off with high LVTs edge out low income people. Not that it isn't happening now, but if you're going to change the status quo, it's sort of on the new system to address the issue
I enjoyed your book review and this series of articles. Thanks for all your work.
I am confused as to why there's no mention of elasticities in this article. If you want to talk about who "really" pays a tax, I think that elasticities is widely considered the most useful and correct foundation for that discussion. In fact, there exists a technical term for who is burdened by a tax: tax incidence. The Wikipedia article on tax incidence seems fine and you might find it useful; it's almost mostly about elasticities.
I think your arguments could be much stronger if you grapple with elasticities. Instead of making tenuous bathtub analogies, you could draw upon general economic principles that are universally recognized as solid. Your opponents would basically have to disagree with the basics of supply and demand, and even non-economists tend to accept supply and demand basics.
I already agree with you about LVT tax incidence, and even I was very doubtful of your bathtub arguments.
Saying that a LVT would almost entirely burden land suppliers doesn't require any specialized arguments that people only see in Georgist advocacy. You have super-basic-and-mainstream economics on your side, and talking about elasticities could show that to your audience.
Is this the book by Donald Hagman which you could not find? I found this by google searching "The Single Tax and Land Use Planning: Henry George Updated" and it is published by the UCLA law review.
https://heinonline.org/HOL/LandingPage?handle=hein.journals/uclalr12&div=49&id=&page=
Makes sense. But would be a lot easier to heavily tax inheritance. And that would fix the most egregious part of land wealth - that most of it is inherited. Simpler, politically easy, no drag... really wonder why the focus is not on that really.
So rents may not fall, but property prices (including land) would fall significantly? Because the present value of the land becomes zero, so the property is just valued based on the building?
> New Zealand [snip] over the course of the next century that figure dropped all the way to 0.5% in 1965 and 0.3% in 1970
The above completely ignores that land is taxed by the city, the quote is looking at national taxes only.
Example: Christchurch has rates $5,758.16 on $850k property value (land value $540,000 + value of improvements $310,000) and national GST(≈VAT) is $640 of those rates. Median national taxation is $10000. So:
A: Total LVT is $3658 (≈36% of all taxes),
B: national LVT is $406 (≈ 4% of all taxes).
City rates are based on valuation, and valuation is based on market rates, and land value is the free variable because building valuations don't vary across the city. Is that not an LVT?