The Land Value Tax: How to Increase Efficiency, Fairness, and Economic Growth with a Single Policy Change
Note: Since this post was originally written in 2021, Lars Doucet published Land is a Big Deal, which addresses many of these issues in far more detail. You can also check out his shorter series of essays here.
The US is currently experiencing a major housing deficit in large metro areas, resulting in skyrocketing real estate costs and millions of people who would otherwise prefer to live in high-demand places being unable to do so. As we’ve discussed elsewhere, the primary culprit here is overregulation of land use making lots of new development uneconomical or outright impossible. Reforming these rules is the most important step we can take towards making cities more prosperous and affordable, but it’s not the only available lever of change.
The land value tax (LVT) is a tax on the unimproved value of land. Most commonly associated with the 19th-century economist Henry George, who advocated for it as a means of reducing urban poverty during the industrial revolution, LVT is both an intuitively just method of taxation and an economically elegant approach to aligning individual incentives with socially optimal outcomes, including an increase in the amount of available urban real estate.
Economic Efficiency
Not all taxes are created equal. Different types of taxation vary considerably in terms of their effectiveness (how well they actually generate the desired revenue), their incidence (who bears the cost they impose), and their impact on economic activity. Replacing less optimal taxes (e.g. those on income, capital, or consumption) with more optimal ones (LVT or Pigouvian taxes) can improve incentives and outcomes without increasing the overall tax burden.
Effectiveness: Many types of taxation run into problems when it comes to collection. With a bit of legal sleight of hand, income and capital gains can be sequestered to polities with more favorable tax regimes, resulting in decreased revenues for the home country and deadweight loss from the effort expended in the process. Taxes on consumption (i.e. sales tax/VAT) don’t face this specific problem, as most consumption can’t be easily transferred across national borders, but nevertheless encourage the emergence of black markets--see the prevalence of illegal cigarette sales in response to high taxes in places like New York City. Taxing immovable (or nearly so) assets like land or real estate overall avoids both of these problems--the assets in question can’t be sent to tax havens, and they can’t be easily hidden from the authorities (though value assessment can be complicated--see below for more detail).
Incidence: Tax incidence refers to who bears the ultimate economic burden of taxation--whose welfare is actually decreased by the imposition of a tax. This is not necessarily the same as the party from whom the tax is collected, as price elasticity can shift the true burden to other participants in the market. In the graph below, example A shows a market with high elasticity of demand and low elasticity of supply; consumers can easily adapt to the imposition of the tax and so bear a lower portion of its cost than do producers. The reverse is true in example B.
Image: “Elasticity and Pricing”, OpenStaxCollege
Unlike most goods, the supply curve for land (not real estate overall, which also includes buildings and other improvements) is perfectly inelastic, as the owners of land generally can't create or destroy it. This means that the incidence graph for a LVT looks like this:
The incidence of the tax falls entirely on the suppliers, passing no cost on to consumers. Since landowners tend to be wealthier than land consumers, this makes the LVT a more effective tool for fighting inequality than most other taxes, which have non-zero elasticity of supply and therefore pass on some cost to consumers.
Incentives/Economic Impact
Perhaps the most important consideration to take into account when designing a tax system is the degree to which the imposition of taxes will distort the behavior of the agents within the system. Economic actors respond to changes in the expected value of different activities they can undertake, and taxes by definition lower the expected value of the activities to which they apply: income taxes lower the value of labor relative to leisure, sales taxes lower the value of consumption relative to investment, and property/capital gains taxes lower the value of their respective kinds of investment relative to consumption. In all of these cases, economic actors would be expected to respond to the imposition of new taxes by shifting their activity away from the newly taxed categories in ways that tend to be economically suboptimal by virtue of deadweight loss. To put it in somewhat plainer terms, most taxes discourage economically productive activity, thereby reducing overall output.
This is not true of LVT. As dictated by common sense, taxing the unimproved value of land doesn’t reduce the amount of land that exists, and as shown in the incidence graph above, this results in no deadweight loss, as supply and demand intersect at the same point as they would without any tax. This means that landowners still have a strong incentive to increase the value of their property (i.e. land + buildings) as much as possible, creating value for themselves and society in the process. To better understand the social value of LVT relative to standard property taxes, consider the following illustrative examples:
Example 1: Jane owns a single-family home on a quarter-acre parcel of land in a densely populated neighborhood of a major US city. Suppose the land itself is worth $600,000 and the home is worth $400,000, for a total of $1M. As mentioned at the beginning of this post, cities like Jane’s desperately need to build new housing, and Jane’s large, centrally located lot would be an ideal place to build more. She could, for instance, replace her single-family home with a ten-unit multi-family building, changing the property’s value to $600k worth of land and $3M worth of building for a total of $3.6M. The construction cost for this type of project might be $2M, so Jane still comes out ahead by $600k (new value - construction costs - original value). However, we’re missing a key factor here: property taxes. Let’s say that combined state and local property taxes in this case are 1.5%, so completing this construction would cause Jane’s annual property taxes to go from $15,000 to $54,000. This would make a considerable dent in any additional rent that Jane would expect to gain from the upgrades and could potentially dissuade her from taking on the project, to the detriment of anyone who might want to move into her neighborhood.
Now consider how this would look under LVT. Jane’s taxes would be based only on the $600,000 worth of land she owns, which would be unlikely to change significantly if she were to make the upgrades. Without an expected tax increase, Jane would be much more likely to invest in her property and her neighborhood, making both her and her city better off in the long run.
Example 2: Robert is a developer who owns a three-acre empty lot in an up-and-coming area of a major US city. Rents in the area are rising, causing the value of the lot to increase along with them. Robert could take the risk of developing the lot now, facing a similar tradeoff to Jane’s between higher revenues and property value on one hand vs. construction costs and higher property taxes on the other, or he could leave the lot vacant for a few years and then either sell it for a profit or develop it at that point. If he chooses the latter (which could be entirely rational from his perspective), many people will again lose the opportunity to move into this neighborhood.
Under a well designed LVT, Robert’s decision becomes much easier. A large portion of the increase in the value of the land itself will be captured by the tax, meaning that Robert will have to improve the property in order to make a significant return. In this scenario (and not in the property tax scenario), Robert’s interests are well aligned with those of the rest of the neighborhood; he’s more strongly incentivized to add value to the city.
"There's a sense in which all taxes are antagonistic to free enterprise … and yet we need taxes. We have to recognize that we must not hope for a Utopia that is unattainable. I would like to see a great deal less government activity than we have now, but I do not believe that we can have a situation in which we don't need government at all. We do need to provide for certain essential government functions — the national defense function, the police function, preserving law and order, maintaining a judiciary. So the question is, which are the least bad taxes? In my opinion the least bad tax is the property tax on the unimproved value of land, the Henry George argument of many, many years ago." - Milton Friedman
Moral Grounding
In addition to the economic arguments for it, LVT makes intuitive moral sense (to most people, at least). A key principle of capitalist systems is that people deserve credit for the value they add for others, typically either via contributing labor or capital, and that profits are a just compensation for these activities. Land ownership (again, as distinct from any work done to improve the land) doesn’t fit within these categories; the value of land is for the most part an emergent property of the land and its surroundings and not of its owner’s work. This is especially true in big, modern cities in which the typical landowner controls a tiny fraction of a percent of the total value of the nearby land, the value of which is primarily determined by network effects. The consequence of this is that when productivity gains are made in an area, a large portion of the profits associated with that increase is captured in the form of rents by landowners who had little to do with the creation of that value. LVT realigns the incentives here, allowing value creators to benefit more from the value they create for others and nudging landowners to invest more in their property.
Relevance to Modern Urban Problems
Cities have no shortage of problems to tackle, and no shortage of policy proposals for fixing them. Many of these proposals, however, create problems of their own, often in the form of net higher taxes and spending, regulations that constrain useful innovation, or regulatory capture. Below are a few examples of urban problems that can be at least partially addressed by LVT without these drawbacks:
Cost of Living: The prevailing government approach to addressing the high cost of living in cities is to spend tens of billions of dollars of taxpayer money each year on housing subsidies. While this does help some people afford housing in the short run, its long-run impact is to drive up the price of housing by increasing demand without addressing supply (not to mention increasing government budget deficits). On the other hand, LVT and other supply-side interventions drive down housing costs by increasing supply without requiring additional government spending.
Underutilization of Space: Large amounts of urban land are dedicated to relatively unproductive uses like parking or even vacant lots. While addressing the former will require zoning reform in many cases, LVT can play a role in addressing this issue more generally by incentivizing landowners to seek out more productive uses of the land as described in the Incentives/Economic Impact section.
Environmental Efficiency: LVT increases the incentives to build upward rather than outward (because this consumes less land) and for governments to invest in public transportation rather than automobile infrastructure (because of value capture), both of which (alongside the aforementioned abatement of parking lots) improve the ratio of economic activity to environmental costs.
Drawbacks of LVT
To be clear, there are some drawbacks to LVT. First, it’s difficult to assess the value of land in isolation from the rest of the property contained on it. Assessing the value of an entire property is relatively easy, as assessors can use its previous sale price or the values of comparable properties as reference points, but since land is rarely sold independently from the rest of its associated property, it can be difficult/politically contentious to agree on a standardized way of assessing its value for the purpose of determining taxes.
Second, if LVT were to be applied evenly across the country, it would be hugely disruptive to rural landowners with a high ratio of land value to cash flow without producing commensurate gains in productivity. On the other hand, applying dramatically different tax rates between adjacent areas (i.e. lower in rural areas and higher in urban ones) could encourage suboptimal allocation of resources, encouraging productive activity to move farther from population centers.
Perhaps he most important obstacle is a matter of political economy; without tax abatement elsewhere, the implementation of LVT would amount to a one-time taking from whoever owns land at the moment--future owners wouldn’t bear any cost since the sale price would decrease, so the introduction of LVT would be reasonably perceived as distributionally unfair. The obvious solution here is to couple LVT with a decrease in property taxes (or e.g. income taxes) of similar magnitude, but this introduces an additional layer of policy complexity that could slow implementation.
Since LVT is not yet common in the US outside of Pennsylvania, there’s a lot of work still to be done to determine the best ways to implement it. Should we eliminate taxes on non-land real estate entirely, or just tax it at a lower rate than land? At what rate should we tax land? Should any other taxes be decreased alongside the implementation of LVT? And how much should LVT policy vary across jurisdictions? Lots of trial and error will be required here; fortunately, the US has fifty states and thousands of cities and towns that can independently try to find the best policies and then broadly adopt the ones that work the best.
by Chris Lehman
