Matt Yglesias estimates that the value of all the land in the United States is about $14.488 trillion, and suggests we could replace all taxes on labor and investment with taxes on land:
This number is high enough that it tends to confirm that view that taxation of land and other natural resources, supplemented by pollution fees and things like congestion charges could replace all taxes on labor and investment and still fund an ample welfare state and public sector.
Whenever you bring this up, some people pop up to say land value is difficult for city governments to assess fairly and accurately, but I don’t buy that.
For starters, local governments in PA already do assess land values. We do it here in Philadelphia, and they do it in your town. If you take a closer look at your local government’s tax documents, the “property tax” millage rates are probably expressed as two values, one for land and one for improvements. It’s just that they set the same rate for both, and lump them together in public communications as the “property tax” rate. But it’s really two different taxes.
If people think these official land values aren’t sufficiently accurate, then what we need to do is try harder to get accurate values. The reassessment system most places have now, where we send a bunch of dudes to walk around and eyeball the value of everybody’s houses from outside, is far less accurate than using market prices. AVI was supposed to get Philly on track to start using market prices, but the manual assessors still badly screwed up the land values in the last assessment.
C’est la vie. There’s just no substitute for trying harder to get it right, and the best valuation system that exists for figuring out what people are willing to pay for stuff is the housing market.
Here’s what I want to know: why can’t we just use asking rents as a proxy for land, for tax purposes? Rents are basically equal to land values. In my neighborhood in South Philly we’ve been seeing “home” prices going up and staying up more than in other areas of the city:
But while we’ve definitely been seeing a lot of new infill home construction around, there aren’t *that* many new units, and asking rents for crappy oldster row homes that haven’t been improved in decades have been going up too. That’s all land. The asking price increases are driven by the desirability of the neighborhood, the school catchment district, and other amenities. You didn’t build that!
Shouldn’t it be possible to come up with an index where we keep track of the asking rents for different types of houses in each catchment zone or ward, average the rate of price increases or decreases together, and then use that number as a stand-in for land value changes in determining people’s tax bills?
That’s a bit crude, but I feel like the GIS wizards can definitely figure out a more sophisticated algorithm in areas where there aren’t that many vacant parcels left to assess directly.