When I was younger, and a more disagreeable libertarian, I was staunchly against almost all taxes. And not just all taxes in general. Each type of tax was a specific affront to human dignity in its own egregious way.
- Sales taxes represented government meddling in private contracts.
- Income taxes represented government stealing people’s time.
- Property tax represented that living on land was a privilege provided by the state landlord. Private property was a myth.
I won’t win the fight over whether the state governments should be spending money. But, given that we have to pay for services, I can definitely opine on the desirable and undesirable traits of one tax or another. Economists tend to like sales taxes because they encourage saving, investment, capital formation, and greater output. Maybe that’s a good idea. But it’s not clear to me that we should incentivize consumption tomorrow at the cost of consumption today. There is no singular right answer to that tradeoff.
I would love to have a per-adult lump sum tax in which everyone pays the same dollar figure no matter what. I would also love to receive a million dollars – and that ain’t going to happen either. In lieu of a lump sum tax on people, I think that the next best thing is a lump sum tax on land. Each acre in a county can pay the same tax bill. On the margin, firms would economize on land and tend toward density. That would bring lots of agglomeration and economies of scale. Jeremy wrote recently about land taxes, which have a lot of proponents. I share the concerns about estimating land value and I think that it’s a non-trivial challenge.
An nth best alternative is property taxes. It leaves the incentives to consume today vs tomorrow untouched. Property values tend to be *relatively* stable over time, which is good for state budgeting. And, except for big home improvements, people don’t change their behavior much because they must pay the tax ‘no matter what’. Property taxes are a few degrees away from lump-sum taxes, but they aren’t as distortionary as some other taxes (I’m looking you excise taxes, sales taxes, income taxes).
Most tax codes are full of exemptions, but property tax codes tend to be relatively straightforward. There may be exceptions for non-profits and a homestead exemption. And the various school districts, city boundaries, county-governed areas, and Mosquito Control Districts can be somewhat labyrinthian. But mostly, local people pay for local services and the tax bill shows up already calculated.
But all the while, I still have the sense that property taxes are like paying rent to the government. So which states have the highest property taxes? Well, that depends on how we want to cut the data. We could calculate property tax per person. After all, states provide services to people. We could also calculate property tax per privately owned acre.
How about both?
Below-left is a scatterplot of 2021 property taxes per person and per acre. First, I’m somewhat surprised at the positive correlation. Both axes include the same numerator, but big states don’t always have bigger populations. For example, North Dakota has 54 privately owned acres per person while, the much smaller New Jersey has less than half of a private acre per person. Also, note the outlier. Vermont is an exceptional state. They get almost 30% of their total state tax revenues from property taxes. So, it’s a lot of taxes both per capita and per acre. Below right is the same scatter plot in logs.
Personally, my main take-away is that there’s no escaping the costs of taxation. The slope coefficient on the right is not statistically different from unity. Taxing private property goes hand-in-hand with taxing people. That is, we can’t tax land while hoping to avoid the cost that people bear. This isn’t a dig at land-tax advocates. If anything, it’s supportive of their cause. If a land tax provides more desirable incentives for firms and home-owners, then we may as well do it because residents bear the tax bill either way. Therefore, it’s better if they bear the cost in a less distortionary way rather than a more distortionary way. This same conclusion is analogous to the corporate tax. Corporate taxes are borne by people – regardless of whether we collect them through corporations. If we bear the cost of taxation regardless of our method of collection, then we may as well adopt the more efficient tax methods.
*If you think that there is chicanery in the property tax being on both sides of the regression equation, then I don’t blame you. Separating it out produces the same result of log acres and log population having nearly identical effects on log property tax revenues.
Economist Writing Every Day 
Not going to debate libertarianism here, but… what about the effect on inequality of sales taxes and per-capita taxes (compared to property tax)?
LikeLike
It really depends on what’s replacing what and at what magnitudes. Let’s assume revenue neutrality. Also, is post-tax income what we care about? Or consumption? Or something else?
In a static sense and speaking of effective average income tax rates:
-A sales tax hurts most those who consume more of their income.
-A property tax harms most those with more valuable real property.
-A lump-sum tax harms most those who are less productive.
Obv, the types of people in each group don’t perfectly overlap. The wealthy are often old people with productivity (retirees), and the young often consume more than their real wage permits (they get into debt).
A bolstering argument in favor of a lump sum tax is that people respond by working more, producing more, and raising the overall output per person so that people can better afford their options. Dynamic effects are hard though.
LikeLike