A quick skim indicates that on the whole it’s pretty good, though it seems to overestimate the difficulty of assessing land value, and repeats the error of some previous studies which conflates owners of land occupied by low income people with the low income people themselves. (More likely, low income people are renters living on land owned by someone else, and when taxes on such land increase the owners can’t pass the cost on to their tenants.)

There is also mention of a study, new to me, that seems to document an anti-sprawl benefit from a land tax. The study unfortunately is secured by ssrn; I shall have to try to find it elsewhere.

This study was requested and funded by the Minnesota legislature.

Hat tip to lvtfan.

This discrepancy isn’t due to any inherent problem with the real estate tax, but may have something to do with the fragmentation of taxing units, particularly school districts.  Areas with relatively little taxable real estate need to collect a greater percentage of its value than do areas with a larger tax base, other things being equal.  But there’s no reason we couldn’t have an equalization system under which the strong-tax-base communities share revenue with the others, as has been done since 1971 in Minnesota.

It is said that lower-income neighborhoods have a greater share of their real estate value in improvements rather than land, in which case exemption of improvements from the tax would also tend to equalize the burden.