Erdmann Housing Tracker · Housing & Cities
TIER 4 Wed, 27 May 2026 18:24:54 +0000
I recently wrote a short post laying out a thesis that where land values are elevated by scarcity or amenity value, property taxes are Georgist taxes (taxes specifically on land). ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- --- # Property Taxes vs. Land Value Taxes | | Kevin Erdmann --- | May 27 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- I recently wrote a short post laying out a thesis that where land values are elevated by scarcity or amenity value, property taxes _are_ Georgist taxes (taxes specifically on land). Upgrade to paid Greg Miller and Lars Doucet have written a thoughtful response to it at "Progress and Poverty", which merits a follow-up post. My basic idea was that the price of a structure is set at the cost of constructing it. Net rental income on structures reflect the market's required return on that investment. Property taxes that fall on structures raise the required rent, but don't effect the price of the structure. In the long run, where there isn't excess land value and property taxes fall on structures, I would expect expenditures on housing to settle at a similar percentage of incomes, but with marginally smaller structures. Rents on land reflect market conditions related to locational amenities or scarcity. Property taxes that fall on land don't change those conditions, so they lower the price of the land. Where any of the value of a property is associated value unrelated to the structure itself, higher property taxes are land value taxes, on the margin. In effect, the land rents are paid to the municipality instead of to the property owner. To be honest, I kept the post short and simple because there are a lot of margins to consider on how these taxes would affect development under conditions of binding zoning constraints vs. non-binding, high demand vs. low, changing tax rates vs. changing conditions under stable tax rates. And, I'm having trouble wrapping my head around all of those contexts and what property taxes vs. land value taxes mean in every permutation. But, I think the arithmetical point is sound, and was worth airing. So, I wrote it without going into complications. Miller and Doucet make a number of reasonable points, and you should read their post, if you're interested in the topic. Below are my caveats, excuses, and explanations. #### Housing is a bundle of relatively binary elasticities Before we get into their post, I should probably discuss one subtle issue. The classic simple supply and demand curve doesn't really apply well to housing, and this is frequently an important element when discussing housing markets. I frequently discuss the issue of the housing "bundle". Housing is a bundle of luxuries, necessities, and endowments. We easily trade away luxuries when they become more expensive. This includes extra square footage. We don't easily trade away necessities when they become more expensive, but the necessities for almost all Americans are such a tiny part of the bundle that they aren't really important. This isn't the 19th century. We don't have streets lined with 12-person households packed into 700 square foot tenements. In fact, one reason we have a housing shortage is that where families would hunker down in housing that is mostly necessities, we have made those housing forms illegal (dorms, single-room occupancy, large numbers of non-family occupants, etc.) We won't let families make marginal consumption decisions on housing necessities. Where housing is very expensive, families are generally consuming the least amount of luxuries they are comfortable with, sometimes together with some negative amenities like high crime, and paying a premium for endowments - idiosyncratic reasons that a location is meaningful and valuable to them, but not to others. In short, if swimming pools are unaffordable, families go without them and spend the comfortable amount on housing. If living close to mom and grandma and the local industry where you built 20 years of professional credibility costs more, you pay it until you can't. Square footage is closer to the swimming pool end of things than the grandma end of things. In a marketplace where property taxes fall entirely on structures, the average home might be 1,700 square feet instead of 2,000 square feet. I think the demand for those 300 square feet is highly elastic. Instead of paying each year for those 300 square feet, you pay the municipality for other services. Practically nothing else in the economics of that market will change. So, I think, on the margin, higher property taxes on structures actually have a similar effect as property taxes on land. Total expenditures won't change that much. In this case, the average structure will be smaller, but that is a marginal change that households don't change their nominal expenditures to counter. Families aren't going to spend more to get their homes back up to 2,000 square feet. I also think the effect on household formation will be weak. I don't think an average home size of 1,700 square feet vs. 2,000 square feet will cause young adults to live with their parents longer, for instance. In their case, it might be an 850 sq. foot apartment instead of a 1,000 sq. foot apartment. I don't think anyone is delaying household formation because apartments are 150 sq. feet too small for them. I think they are delaying household formation because some families are paying premiums to stay in homes in locations with endowment value, and where they are desperate for that, they are willing to put up with negative locational amenities that the young adult looking to form a household won't put up with, or they are paying more for endowments than the young adult is willing or able to pay for tradable locational amenities. One recent market response to this has been to build more small apartments where those are legal, where locational amenities are not negative, and where much of the property tax falls on the land value. Another aspect of these "bundle" issues is that the gross rental yield on the average structure is about 10% whereas the gross rental yield on land is more like 3%. So, if we think about property taxes as a consumption tax, it's actually a consumption tax that is highly weighted toward the land rent component of a property. So, even where property taxes are taxes on structures, I think the economic consequences of property taxes aren't much different than the economic consequences of land value taxes. They reduce some quantity of consumption of structures that families easily go without in order to keep total nominal expenditures comfortable. I don't think they materially reduce the production of new units or household formation. Texas and New Jersey seem to confirm that intuition. High property taxes, at worst, have little effect on the number of homes permitted, under current supply-constrained conditions. Some of this is related to our current housing market context, where on some margins, both demand and supply are inelastic. When prices go up, families pay more and the higher rents don't trigger more supply. In the long history of housing markets in previous centuries, supply and demand were elastic. Demand for more housing followed rising incomes, and when it did, new homes were constructed. Under those conditions, land values were generally much lower, and property taxes weren't as much like land taxes. #### Exceptions to my point Miller & Doucet note what they consider to be an exception to my example in Cincinnati, where they highlight 3 lots. Two have triplexes on them and have a property tax bill of $3,600 and one is vacant and has a property tax bill of $1,500. Cincinnati is the sort of location that my explanation doesn't apply well to. In most of Cincinnati, there isn't much land value to tax. Cincinnati's residential property market is mostly structures, and most of its property taxes fall on those structures. The tax on structures isn't constraining aggregate construction activity in Cincinnati. Before the 2008 mortgage crackdown, there were plenty of new single-family homes to keep the region affordable, and, since the mortgage crackdown, zoning has prevented multi-family construction from filling the gap. The lot they highlight must have some locational value, because $1,500 is a decent sized property tax for a vacant lot in Cincinnati. That suggests that more than 1/3 of the value of the triplexes is still in the land. So, this particular example is unusual for Cincinnati, in that it is a pretty Georgist tax (mostly taxing land). Since this is an example of that, I'm not sure how it is an example of property taxes being a disincentive to building. The lot only has value because of the option value of potentially building something on it. That's why it has a $1,500 tax bill. The triplexes should be returning something near 10% ROI on the structures - probably something like 7% on the property as a whole (structures + land). If the lot has value, it will pay to develop it under the current tax regime. The owner is leaving potential income on the table by leaving it vacant. This does point to the context where I think strict land value taxes might help with development, but not because of the traditional "tax the value that comes from social investment and agglomeration value" reasoning of Georgist land taxes. And, if this lot has enough value to be taxed $1,500 while vacant, I don't think it's a good example of this context. There are cities where old parts of the city are hollowed out. Parts of Cincinnati are in this condition, but empty lots in those parts of the city have negligible property taxes, because they have no locational amenity value. Those lots wouldn't have a large tax bill even under a pure land tax. The problem in some cities like that is that the property taxes become disorderly, and the taxes on old, depreciating homes are too high. Owners sometimes let them depreciate to uselessness with outstanding tax bills and let the city have them. Speculators aren't holding on to those lots. They are considered worthless. There are a number of places where land value taxes have been implemented, to some success, in Pennsylvania. And, they appear to me to generally be in this type of context, where there are old neighborhoods that have transitioned back to vacancy, where new construction is happening on the city's periphery. Some literature on these programs finds positive results, and finds that, contrary to my intuition, land value taxes (or two-tier property taxes that tax land more) increased the density of homes rather than the size of them. However, it seems to me that the land value taxes in those contexts were increasing the tax on suburban neighborhoods. I am not intimately familiar with this literature, but my impression is that the land in the vacated neighborhoods would not have been taxed highly in either case. It could be that these land taxes were implemented within broader programs meant to revitalize those areas. Also, the reduction of taxes on structures would lead to a one-time shift. Whereas a given home might have previously required $1,000 in rent to justify construction, maybe it would now only require $800. The value of existing structures would increase and the value of land would decrease after the change from property taxes to land taxes. But, in this context, it seems like the revitalization building would happen on the least valuable land with the lowest taxes; not because valuable land was induced to development because high taxes reflected its potential value. Some literature on property taxes does find that higher property taxes lead to fewer homes, not just smaller homes, where supply conditions allow for reactions in quantity rather than price. This contradicts my intuition about demand for the "basket" of housing. There are a lot of complications here, such as inter-regional migration, that are hard to control for. Also, the existing literature was written when there wasn't pent up demand for more households, so the supply and demand conditions I discussed above weren't so binary. But I note it for the reader's consideration. I frequently take positions against the existing literature based on in depth considerations. Here, some existing literature seems to contradict my intuition, and admittedly, I have not written 200 pages of analysis to defend my position. In any case, Miller & Doucet's example seems to have Georgist enough taxes to induce development. The empty lot with $1,500 in annual taxes would surely provide better returns with something on it, even if that increases the tax to $3,600. #### How does a given tax rate affect development? They note, accurately, that under any given property tax, the tax on structures will lead to fewer structures. They write, "Multiply (me: higher property taxes) across every owner in the city making decisions like that, and you get a city that _builds less_ and _builds smaller_." I agree, and I said as much in my post. I think, in consideration of our current housing crisis, property taxes will mostly lead to building smaller. But, as noted above, I may not have the support of existing studies in my favor. Maybe pure land value taxes deserve more due than I am giving them. #### Property taxes are big. They note that an annual tax of 1% is a big deal if discount rates are low. This goes back to my claim that gross rents on structures tend to be around 10%, on average, and rents on land tend to be around 3%. This is a decent part of my reasoning. The first-order effect of a 1% property tax on a property is like a 9% consumption tax on the house and a 25% consumption tax on the land. Partly what led me to thinking about property taxes as a de facto land value tax was my thinking about these dual yields on property. Land definitely has a much lower yield than structures do. How much may be a matter of debate, but I don't think there can be much debate that land yields are substantially lower. So, property taxes are already a two-tier tax. Explicit two-tier tax regimes might be increasing the difference between the tax rates, but on a consumption basis, a straight property tax is already a two-tier tax. #### Land Hoarding One of the motivations for land value taxes is to prevent land hoarding - speculators buying up land and sitting on it for the price appreciation as a city grows up around it, rather than developing it. I apologize, if you're wanting more on this topic. I'm not going to develop this argument here. But, when I was involved some time ago in an in depth discussion of Henry George's "Progress and Poverty", I didn't find the hoarding issue compelling. Land hoarding doesn't keep cities from building enough to be affordable when mortgage and land use regulations aren't binding. To the extent that development patterns have been suboptimal, it is because of what landowners aren't allowed to do, not because of what they won't do. I don't think heterogeneous decisions about how intensively to develop each plot of land in a city over time create systemic problems that are particularly bothersome. In some respects, the effects of some undeveloped lots in high value locations are positive and allow for unforeseen opportunities for new developments within a changing city. #### Our Current Dumb World But, that's neither here nor there. Land taxes are good, and my main reason for defending property taxes is that they are a pretty good approximation of land value taxes in most contexts where there is demand for new development. I generally eschew political considerations to focus on the analytical. But, on this issue, part of my motivation for thinking through it comes from the weird political intuitions people seem to have about these taxes. Think about Proposition 13 in California, and other similar state laws that prevent property taxes from rising along with market values. Or, think of the current property tax revolts happening in Republican strongholds. The housing shortage was mostly hurting renters, because the initial effects of the 2008 mortgage crackdown reduced the number of funded buyers and pushed home prices too low. Families that could get mortgages had generationally low housing costs. And, when that was accompanied by low mortgage rates, it was especially the case. But, as the housing shortage has pushed up property values, those low mortgage payments are being countered by rising property taxes as market values become more inflated (and property taxes become, mathematically, more Georgist). Now homeowners are feeling the effect of the housing shortage, too. Many homeowners live in homes they wouldn't be able to afford if they had to repurchase them at today's prices with today's mortgage rates, and one way they feel that is through rising property taxes. Their reaction to that is sometimes, "Why are we paying property taxes at all?!" Or, old homeowners who live in $600,000 homes they purchased 30 years ago for $100,000, will say, "This extra value you're taxing me on doesn't have anything to do with anything _I_ did. I just bought a $100,000 home that I want to keep living in. It's not fair to tax me for things I had no control over." The tax revolts are _quite specifically_ revolts against the Georgist portion of the tax. Oddly, in this case, people are especially _offended_ at the idea of taxing wealth they didn't earn. The more Georgist the property tax becomes as families live in inflated markets, the more they revolt. So, I guess, part of my motivation here is to defend the traditional, single rate, property tax from its current antagonists. Pushing for or implementing land taxes might be a case of the perfect being the enemy of the good. In our dumb world, with wide use of the good tax, the good tax is under attack specifically because of the part of it that is perfect. On that topic, land is a much larger portion of the average property value today than it was a decade ago. Thus, property taxes are much more Georgist today than they were a decade ago. I wonder if there are ways to analyze the market through that lens to estimate the marginal effect of more or less Georgist taxes on various aspects of the housing market. In any case, I think we're all, more or less, on the same side here. They end their post with "Erdmann's not wrong to be a fan of property taxes in general. By all means, localities should be encouraged to drop their sales and income taxes and trade them for increased property taxes instead. However, there's an even better policy after you've done that: split the property tax into two taxes, and tax the land more, and buildings less." My response is that property taxes already do that. And that's why I think it's pretty good. But, unfortunately, it's the same reason why so many taxpaying laypeople disagree. Upgrade to paid You're currently a free subscriber to Erdmann Housing Tracker. For the full experience, upgrade your subscription. Upgrade to paid --- | | | Like --- | | Comment --- | | Restack --- (C) 2026 Kevin Erdmann 548 Market Street PMB 72296, San Francisco, CA 94104 Unsubscribe