3 ms·

In Japan it's reasonably common for someone to own the land while someone else (e.g. a homeowner) owns the buildings on the land and has certain rights to live

In Japan it's reasonably common for someone to own the land while someone else (e.g. a homeowner) owns the buildings on the land and has certain rights to live there that the landlord can't ignore (and hence can't just turf them out suddenly). Flats in the UK are leasehold, which means you can buy (part of) a property and still not own the land. It's entirely possible to separate the two.

mlyle 4y ago
It's possible to separate the two, but not in a way that gives a meaningful true market value for land alone in the short term-- rather just values for land in long-term option value after other rights are considered. That is, that leaseholder you can't turf is enjoying a lot of the value of the land, and it may not be correlated to past payments or transactions. In turn, lots of the benefits of a LVT evaporate if the taxable value (of landowner's eventual rights) is so decorrelated from the present value of land uses. That's the problem here: valuing the land alone is hard. Separating rights, adding more owners and options and preferences --- doesn't make valuing the land alone easier.
- brigandish 4y ago
  In the example where a landlord has a tenant with a building that that tenant owns, the landlord knows how much they are charging the tenant, which is the primary signal of the value they can extract from that land in that situation. Hence, they can calculate a tax level they're comfortable with paying (and I posit that it would indicate the value of the land to the owner). The tenant knows how much value they are getting from being there, at least relatively, because they pay the rent. If you're paying twice as much rent for half the space as your mate, unless there's some reason like being closer to the centre of the city or something, you know you're getting a bad deal. Pricing can be difficult but it doesn't exist in a vacuum, there's a whole market to compare to.
  - mlyle 4y ago
    > In the example where a landlord has a tenant with a building that that tenant owns, the landlord knows how much they are charging the tenant, which is the primary signal of the value they can extract from that land in that situation. Hence, they can calculate a tax level they're comfortable with paying (and I posit that it would indicate the value of the land to the owner). Yes, but... As I keep trying to make clear: * Anyone improving the property requires some kind of protection of this investment, usually in the form of a durable ownership interest or a very long term lease, with a value chosen based on both parties' estimates of the utility of the land over that lease term. * In turn, the rents that the landlord receives from the property can be not-too-closely related to the present value of the land. They may be higher or lower. * In turn, the land value tax doesn't succeed in capturing the true market rent of the land when it is based upon bidding by landowners, whether we're dealing with separate owners for improvements or not. Imagine the case of someone who entered into a long term lease (60 years? 99 years?) in 1975 to build a house surrounded by orchards in San Jose. The real value of land has appreciated >60x over that term. A land value tax would capture a tiny fraction of that value if the landowner was bidding based upon the lease returns + the present value of the land being returned in 40 years. > Pricing can be difficult but it doesn't exist in a vacuum, there's a whole market to compare to. The level of friction is so high that you can't solve this one with bidding; you're going to have to rely upon comparable sales and assessed/appraised values.
    - brigandish 4y ago
      > Yes, but... As I keep trying to make clear: It takes two to tango. Perhaps you'd be better off keeping your frustrations to yourself and improving the clarity of your writing. > The real value of land has appreciated >60x over that term. What is the "real value"? It's what someone is willing to pay for it. If the returns on the land are not that much, as you contend, then it's not worth >60x what it was. If the property owner has some protection than means they can't be shifted off the land for 99 years, then that land is not worth as much as you contend because it can't be used for something else. If someone else thinks the land is undervalued then they can bid on it and make more money, but they'll have to pay the tax and find a way to extract that value. If the land really is that valuable then the landowner can say so and pay the tax on it now because they think they'll get the return in future. In short, you're assuming that the "real" value and the actual value are different and erroneously so. This assumption makes the reasoning self-contradictory - the value is determined by people's bids, not some Pythagorean value floating around somewhere waiting to be happened upon by some perfect system.

mlyle 4y ago

It's possible to separate the two, but not in a way that gives a meaningful true market value for land alone in the short term-- rather just values for land in long-term option value after other rights are considered. That is, that leaseholder you can't turf is enjoying a lot of the value of the land, and it may not be correlated to past payments or transactions. In turn, lots of the benefits of a LVT evaporate if the taxable value (of landowner's eventual rights) is so decorrelated from the present value of land uses. That's the problem here: valuing the land alone is hard. Separating rights, adding more owners and options and preferences --- doesn't make valuing the land alone easier.

brigandish 4y ago
In the example where a landlord has a tenant with a building that that tenant owns, the landlord knows how much they are charging the tenant, which is the primary signal of the value they can extract from that land in that situation. Hence, they can calculate a tax level they're comfortable with paying (and I posit that it would indicate the value of the land to the owner). The tenant knows how much value they are getting from being there, at least relatively, because they pay the rent. If you're paying twice as much rent for half the space as your mate, unless there's some reason like being closer to the centre of the city or something, you know you're getting a bad deal. Pricing can be difficult but it doesn't exist in a vacuum, there's a whole market to compare to.
- mlyle 4y ago
  > In the example where a landlord has a tenant with a building that that tenant owns, the landlord knows how much they are charging the tenant, which is the primary signal of the value they can extract from that land in that situation. Hence, they can calculate a tax level they're comfortable with paying (and I posit that it would indicate the value of the land to the owner). Yes, but... As I keep trying to make clear: * Anyone improving the property requires some kind of protection of this investment, usually in the form of a durable ownership interest or a very long term lease, with a value chosen based on both parties' estimates of the utility of the land over that lease term. * In turn, the rents that the landlord receives from the property can be not-too-closely related to the present value of the land. They may be higher or lower. * In turn, the land value tax doesn't succeed in capturing the true market rent of the land when it is based upon bidding by landowners, whether we're dealing with separate owners for improvements or not. Imagine the case of someone who entered into a long term lease (60 years? 99 years?) in 1975 to build a house surrounded by orchards in San Jose. The real value of land has appreciated >60x over that term. A land value tax would capture a tiny fraction of that value if the landowner was bidding based upon the lease returns + the present value of the land being returned in 40 years. > Pricing can be difficult but it doesn't exist in a vacuum, there's a whole market to compare to. The level of friction is so high that you can't solve this one with bidding; you're going to have to rely upon comparable sales and assessed/appraised values.
  - brigandish 4y ago
    > Yes, but... As I keep trying to make clear: It takes two to tango. Perhaps you'd be better off keeping your frustrations to yourself and improving the clarity of your writing. > The real value of land has appreciated >60x over that term. What is the "real value"? It's what someone is willing to pay for it. If the returns on the land are not that much, as you contend, then it's not worth >60x what it was. If the property owner has some protection than means they can't be shifted off the land for 99 years, then that land is not worth as much as you contend because it can't be used for something else. If someone else thinks the land is undervalued then they can bid on it and make more money, but they'll have to pay the tax and find a way to extract that value. If the land really is that valuable then the landowner can say so and pay the tax on it now because they think they'll get the return in future. In short, you're assuming that the "real" value and the actual value are different and erroneously so. This assumption makes the reasoning self-contradictory - the value is determined by people's bids, not some Pythagorean value floating around somewhere waiting to be happened upon by some perfect system.