You're still muddling two distinct concepts together. Let's start with a simpler case where it is easier for people to reason about.
There is a single lot in the urban core. There are two proposed businesses vying for the lot. The first is an empty parking lot, the second is a two story parking garage. Under a property tax, the second group would be taxed for the improvements of building the above ground garage they would be taxed on economic activity - that is a bad disincentive for society. Under an LVT they would not be taxed for improving the property, there would be 0 marginal tax on economic improvement of the land. Now where it may not have been profitable previously to add the garage it now is. That is what you want. You have a market for this property, and it is now beneficial to be sold to the buyer with plans for the best economic use of the property.
A second example. Imaging a single city block split into lots. The city changes from a property tax to a LVT. The amount of money they collect from this block stays the same - but to change they will reduce the taxes on improved lots and raise the taxes on unimproved lots. Now those who improve their lots are no longer subsidising the free riders who are sitting on unimproved lots. Again you are supporting beneficial economic improvements and their activty instead of penalizing it via a property tax.
The case you are thinking of, you are using the phrase "marginal" but you are not applying marginal reasoning. Marginal implies a small change with all else being equal. You cannot apply marginal analysis to land as land is not substitutible - there is not "margin" between being in a dense urban center and in the exurbs.
There can be more buyers for a single plot of land. Enough buyers that they can be though of as substitutible. There are not more sellers for a single plot of land - you can't create more land, or produce more land with those characteristics of that dense urban center. Marginal analysis from the perspective of the buyer fails here and that's the mistake.
> no matter where they locate the property tax changes less than a land value tax, meaning the decision on where to locate is based on efficiency rather than taxes.
Again this is thinking that second order effects are first order. The first order effect here is the price of the property. The price of the property in the urban center will at least 2x-3x the price of that same property in the exurbs. It will dwarf any differnece in method of taxes. Even if LVT were 3x the property tax most property taxes are roughly 1% of the purchase price per year. Price not tax is the dominant factor and the reason why that comparison you made isn't valid. Within a price band yes tax will impact decision making, across price bands price obviously dominates. And tax dominating within a priceband is a good thing because we have now changed the policy to no longer tax economic improvements to land - so the tax policy is actually better. It's improving behavior at the margin.
Then why do realtors do showings for multiple locations? It’s because the locations are substitutable, with buyers balancing price and amenities. Land is substitutable.
> confusing second and first order
Prices are a second order effect, not a first order effect (unless there’s price fixing). LVT makes land with amenities relatively less attractive, which lowers demand. Lower demand then moves the price, but the quantity demanded will still be lower since the demand curve has shifted down. This causes an exodus from urban centers to the periphery.
>property taxes disincentivize development
True. Assuming constant revenue, a move from property to land value tax would create a relative migration from city cores to a highly developed exurbia. If the LVT was high enough, we’d get single plot high rises with multistory garages only accessible by freeway surrounded by untouched nature reserves.
> Aassuming constant revenue, a move from property to land value tax would create a relative migration from city cores to a highly developed exurbia
At this point it appears you're just ignoring basic economics price sensitivity. Lowering overall taxes on a lot will not cause a migration from that lot.
I mean you can just keep repeating that phrase, but it doesn't then make it true. If you continue your argument that eliminating a tax on economic activity in an area will reduce the amount of activity in that area then there likely is little left for us to discuss.
There is a single lot in the urban core. There are two proposed businesses vying for the lot. The first is an empty parking lot, the second is a two story parking garage. Under a property tax, the second group would be taxed for the improvements of building the above ground garage they would be taxed on economic activity - that is a bad disincentive for society. Under an LVT they would not be taxed for improving the property, there would be 0 marginal tax on economic improvement of the land. Now where it may not have been profitable previously to add the garage it now is. That is what you want. You have a market for this property, and it is now beneficial to be sold to the buyer with plans for the best economic use of the property.
A second example. Imaging a single city block split into lots. The city changes from a property tax to a LVT. The amount of money they collect from this block stays the same - but to change they will reduce the taxes on improved lots and raise the taxes on unimproved lots. Now those who improve their lots are no longer subsidising the free riders who are sitting on unimproved lots. Again you are supporting beneficial economic improvements and their activty instead of penalizing it via a property tax.
The case you are thinking of, you are using the phrase "marginal" but you are not applying marginal reasoning. Marginal implies a small change with all else being equal. You cannot apply marginal analysis to land as land is not substitutible - there is not "margin" between being in a dense urban center and in the exurbs.
There can be more buyers for a single plot of land. Enough buyers that they can be though of as substitutible. There are not more sellers for a single plot of land - you can't create more land, or produce more land with those characteristics of that dense urban center. Marginal analysis from the perspective of the buyer fails here and that's the mistake.
> no matter where they locate the property tax changes less than a land value tax, meaning the decision on where to locate is based on efficiency rather than taxes.
Again this is thinking that second order effects are first order. The first order effect here is the price of the property. The price of the property in the urban center will at least 2x-3x the price of that same property in the exurbs. It will dwarf any differnece in method of taxes. Even if LVT were 3x the property tax most property taxes are roughly 1% of the purchase price per year. Price not tax is the dominant factor and the reason why that comparison you made isn't valid. Within a price band yes tax will impact decision making, across price bands price obviously dominates. And tax dominating within a priceband is a good thing because we have now changed the policy to no longer tax economic improvements to land - so the tax policy is actually better. It's improving behavior at the margin.