North Dakota voters recently considered taking advantage of their state's newfound oil wealth to eliminate local property taxes, but decided to stick with the status quo. Given the actual choice set on the table that may have been the right decision. But I wish more jurisdictions would think harder about property taxes, which I think are overwhelmingly a barbaric relic of a time when governments faced large logistical constraints in their ability to assess and collect taxes. A property tax is essentially a tax on buildings, and since buildings are the most expensive most important and longest-lasting kind of capital goods we have it winds up being about the most anti-growth form of taxation you can imagine. What's more, it's a tax that has only a hazy relationship to ability to pay which invariably inspires jurisdictions to lard their property tax rules down with all kinds of exemptions and complications.
What you want to do is tax land. I met Rick Rybeck yesterday who sent me this interesting paper (PDF) of his on one special case of the issue:
Transportation investments often increase nearby land values. This can choke off development, pushing new growth to cheaper sites remote from these investments. This “leapfrog” development creates a demand for infrastructure extension that starts the process over again. Transportation infrastructure, intended to facilitate development, thus chases it away. Resulting sprawl strains the transportation, fiscal, and environmental systems upon which communities rely. Several jurisdictions around the country utilize a value-capture technique embedded in their property tax to help finance infrastructure and motivate affordable compact development. They reduce the tax rate on assessed building values and increase the tax rate on assessed land values. The resulting compact development should facilitate better transportation and accommodate economic growth with reduced fiscal and environmental costs. This technique’s ability to foster affordable compact development might help bridge the gap between those who advocate growth boundaries and those who fear the impact of growth boundaries on affordable housing.
This is an excellent point, but I really do think the issue is quite general. If you read this skeptical take on land taxes (PDF) from Jeffrey P. Cohen and Cletus C. Coughlin (yes, that's a realy person's name) I think you'll see that the case for them is really quite strong. Their first objection is that doing assessments of land value ex structures accurately is difficult. That's true, but doing the theoretically correct tax with some estimation error is much closer to optimal than doing the wrong tax accurately. Their third objection is the ridiculous one that a land tax will redistribute resources away from wealthy landowners, which they correctly note "might be viewed by some as a virtue rather than a fault."
Last they claim that "if a pure land tax were to captue all current and future rent from the landowners, the market value of the land would become zero" leading to widespread abandonment of land. If that were in fact to occur then I think we'd have to cut the tax. But I also think it's wrong. For one thing, land would have consumption value to its owners. But more broadly, I think this is a bit like saying that since the average investor can't beat average stock market returns nobody will ever buy stock. Or that since the expected return on launching a small business is negative there will be no entrepreneurship. The whole economy is propelled forward by a Lake Wobegon effect where, almost by definition, investors and entrepreneurs all think they're above average. High land taxes will make land cheap, but entrepreneurs will want to get their hands on that cheap land in order to undertake their entrepreneurial schemes.