[hundt]

Yes, I understand this point but the ratio between rents and property values varies wildly. So yes you can compute an average ratio and say it is "equivalent" to a tax on imputed rent if you adjust by that ratio, just as you could average out the number of bathrooms in each house and say it is "equivalent" to a bathroom tax. But it's not a very compelling argument!

[harryh]

ratio between rents and property values varies wildly

Does it vary all that much within a specific property tax jurisdiction?

[hundt]

Sure! Here are some things that could cause the rent/value ratio to be significantly different for a single property:

- There is no usable building on your property.

- There are plans to build a highway through your neighborhood in five years.

- Amazon has announced plans to build an office in your neighborhood two years from now.

- There are harvestable (with further development) natural resources on your property.

Many of these have impacts on hypothetical future consumption, but hopefully you agree that a tax on anticipated future consumption indefinitely into the future is not a consumption tax.

I don't have data as to precisely how strong the correlation is, but I think it's pretty clear that it is not 1, and furthermore if we really wanted to tax imputed rent directly we could presumably do so (we'd have to guess what it is but the same challenge applies to valuations of properties that haven't sold recently). So I just don't think "consumption tax" is a good way to look at a tax that clearly captures, and is designed to capture, other things.

[harryh]

I don't have data as to precisely how strong the correlation is

I'm actually pretty curious now, and might be go looking for data. It's surely not precisely 1, but I bet it's not all that far off. Especially if you limit the sample to residences.

Your list of outliers is a good one (I particularly like the natural resources one), but I think they're probably mostly exceptions to the rule.

In the end though sure, property taxes aren't precisely consumption taxes. But they clearly aren't wealth taxes either. They're something in between and, I think, much closer to consumption that wealth.

I've really enjoyed thinking about the points you brought up. Thank you for doing that.

[hundt]

Thanks, I enjoyed the exchange as well. I am also curious about the correlation coefficient but could not find any details in the amount of time I had to put into it. I did get the feeling that there is substantial research in the area of what drives housing prices, though.