How to sell Georgism to the middle class
We need to tax land. But the wealth of the middle class is mostly in land. How do we untie this knot?
I’m very glad to see more people taking an interest in Georgism! For those who don’t know, Georgism is the economic philosophy based on the ideas of Henry George, the 19th century American economist who wrote Progress and Poverty. His basic idea was that as humankind generates more wealth, landowners siphon off that wealth through increased land values and rents, so that lots of people stay poor. His solution was something we now call the “land value tax”, or LVT, which would tax the value of land itself without taxing the useful stuff that people create on the land (buildings, farms, etc.).
Lars Doucet has a great (and long) multi-part write-up of the LVT idea over at Scott Alexander’s blog, and I highly recommend it. But the basic idea isn’t hard to grasp. The tax has two big things going for it: Fairness, and economic efficiency.
First, fairness. Humans didn’t create the land; we just found it and declared it was ours. Companies, buildings, and other repositories of wealth are created by human ingenuity and effort, but land was there without us and will be there when we’re gone. Why should some humans reap a huge financial reward from that land while others get none? It seems unfair. Now if you build a building or a farm on the land or if you work to mine out the minerals under the land, then sure, reap the rewards. But just squatting on a piece of territory that they didn’t create doesn’t seem like the kind of thing that should determine who’s rich and who’s poor.
Second, efficiency. In general, when you tax something, people produce less of it. Tax labor earnings, and people work less. Tax savings, and people save less. And so on. Maybe it’s only a little bit less (as with income tax), but there is some effect. Not with land, though. No matter how much you tax it, land never goes away, so you can theoretically raise a lot of money with an LVT.
In fact, LVT is even more important for productive efficiency. A lot of land’s value — and more and more as time goes on — comes from its location. Land in urban areas is really valuable because of clustering effects — if all the tech companies decide they want to be in San Francisco, then land values in San Francisco shoot up. So do rents. That siphons away wealth from the productive tech workers who moved to the city, which discourages people from moving there, which reduces economic growth. Some authors, like Hsieh & Moretti (2019), find a pretty large growth reduction from limitations on housing in industrial cluster cities like NYC and SF. LVT provides a partial fix for this, by siphoning away money from lucky urban landowners and — presumably — returning it to the productive residents of the city, in the form of cash payouts or local public goods like education and infrastructure. In fact, a number of famous economic theory papers, most notably Arnott & Stiglits (1979), find that local public goods should be entirely financed by land value taxes — a result that’s usually referred to as the Henry George Theorem.
There are some practical problems regarding the implementation of LVT. The two thorniest questions revolve around when the tax is collected, and how the value of land can be separated from the value of improvements on the land (e.g. buildings). The answer to the first is basically to just have normal property tax with exemptions for improvements. This is how they did it in Pennsylvania in the 90s, and it worked pretty well. The second question, about assessments, is something Lars Doucet and other Georgists are working on dealing with.
So that’s all good. And it’s good to see that people in both the tech and politics world are taking a renewed interest in the idea of taxing land value. But there’s another problem here, which will likely prove less easy to surmount than the technical issues: Political economy. The fact is, most of the wealth of the American middle class is tied up in houses (and the land under the houses), and taxing away a significant portion of that wealth is going to make a lot of Americans very angry.
Why Georgism will be a hard sell to the American middle class
I’ve been thinking about this problem for a while, but today there was a video by Bloomberg CityLab (my favorite source for news and info about urban issues) that really brought home the scale of the challenge. It basically showed a bunch of people in New York City talking about how much difficulty they’re having getting the cash to pay property taxes. Part of this is because of NYC’s weird assessment system; that part can be fixed.
But part of it is because land values have shot up, meaning homes are getting assessed for more. That demonstrates a perennial problem with property tax (and hence with LVT) — homes are illiquid but cities need cash every year. Owner-occupiers do derive financial benefit from their property — houses really are capital! — but those benefits aren’t in the form of cash. So when the IRS comes calling every year, homeowners have to either borrow or find some other source of liquidity to pay up, so that cities that run on property tax revenue can run.
There are various proposals to fix this problem, but they all have drawbacks. For example, property tax or LVT can be levied retroactively at the time of sale, like capital gains taxes. but this means that people can delay the tax for a long time by never selling (just as they do with stocks). And it means that revenue will come in unevenly from year to year, which presents a problem for cities that rely on that revenue to pay their annual bills for things like schools. Cities, like homeowners themselves, pay a relatively high cost for borrowing.
The problem of liquidity means that property taxes can only go so high before property tax payers revolt and implement policies like California’s notorious Proposition 13.
But in fact there’s an even deeper problem here, one which seems to defy any purely technocratic fix. Namely, the wealth of the American middle class is mostly in their houses. Edward Wolff wrote a good paper in 2014 documenting this. Here’s a graph I made of his data for a Bloomberg column I wrote a few years back:
Now, this doesn’t mean that most housing wealth is owned by the middle class (to conclude that from this graph would be to commit the base rate fallacy). Housing wealth is still skewed to the rich, it’s just much less skewed than stock and bond wealth. The middle class in America isn’t very wealthy on a per capita basis, but what wealth they do have tends to be tied up in their houses.
Why this is true is an interesting tale, involving both history and economics. The buildout of the suburbs, and various policies to encourage homeownership, both made sure that regular Americans put a lot of money in their houses. Houses are usually bought with mortgages, and mortgage payments act as a sort of forced saving that nudges people to save more of what they earn, while crowding out other forms of investment at the same time. And as Jorda et al. (2017) found, the rate of return on housing has been truly excellent in most developed countries over a very long time, providing stock-like high returns with bond-like low risk. (The downside is the lack of liquidity.) So housing was just a good investment option for people who also wanted the freedom that being an owner-occupier provided. Plus, houses are an asset whose value is probably easier for middle-class people to understand than stocks.
But regardless of the reasons, the deed is done. And that means that any policy that reduces homeowner wealth is a very tough sell, because it means clobbering the politically powerful middle class. And property taxes and LVTs certainly reduce home values. We’re in a suboptimal political-economic equilibrium, where, as we Texans like to say, “you can’t get there from here.”
If we could somehow extract middle-class wealth from houses and put it into stocks, that might do the trick. The simplest scheme for doing that is to create social wealth funds at the city level, and have LVT contributions go into those funds, basically forcing the middle class to transfer their wealth from houses to stocks. I expect to see some Georgists suggesting schemes like this, and they’re certainly worth considering.
But an alternative idea is to think a little bit more broadly about the idea of Georgism.
A broader definition of Georgism
A lot of people equate Georgism with the idea of the land value tax. And indeed, this was George’s most important policy idea — in fact, he thought it was so important that it could replace all other taxes. But later generations of economists who studied Henry George and his ideas thought more broadly about ways to productively redistribute the wealth from landownership.
One of these Georgist economists was Wolf Ladejinsky, who advised postwar Japan and Taiwan to redistribute agricultural land from landlords to tenant farmers. In his book How Asia Works, Joe Studwell sees this agricultural land reform as key not just to creating equality, but to improving efficiency and growth — former tenants spent more effort on their land once they owned it, and former indolent landlords were motivated to go become productive entrepreneurs in export manufacturing.
Of course, the U.S. is not a country with a large population of tenant farmers, but Ladejinsky’s success shows that there are ways to think about Georgism without hyperfocusing on land value taxes. What Georgist alternatives might work for urban land in modern tech cluster cities?
One idea I’ve been thinking a lot about is the Singaporean housing system. In this system, the government builds houses and sells them cheaply to first-time homebuyers (actually Singapore technically sells 99-year leases, but these function similarly to homeownership, and have drawbacks that probably make them inefficient compared to simple, classic homeownership). The government then uses its power of construction to build more housing as needed in order to control the rate at which home values appreciate. Slow and steady home value appreciation basically gives Singaporeans a pension with low risk and moderate returns, while the government ensures that the city has plenty of housing.
A modified version of this system might work for the U.S. The government owns a lot of land in and around urban areas, on which it could simply build housing and sell it cheaply to first-time homebuyers (and to low-income people). In fact, the Biden administration has proposed to do exactly this, though the plan may or may not ever become reality. If the government can win in court, it could even use eminent domain to buy existing properties from big landlords and redevelop them into denser housing, then sell the houses at below-market prices to young people and poor people.
If you use government construction to control the rate of housing creation, you can moderate house price appreciation. That will reduce the wealth of the middle class somewhat. But you can titrate it so their wealth doesn’t actually fall, but simply rises more slowly. And in exchange for slower appreciation, homeowners can enjoy lower risk — if government always adjusts its construction rates to make sure house prices grow at a slow and steady rate, homes go from being assets with good returns but lots of risk into one with OK returns and low risk. The millions of middle-class Americans who got wiped out in the crash of 2007-8 would probably rather put their wealth in something with less risk, even if the price was less return. If they want something with higher risk and higher return, they can put money in the stock market — and in fact, with this system in place, they probably would do so.
In other words, government housing construction is a way of slowly weaning the American middle class off of its longstanding addiction to housing wealth. It took us a century or more to get into this bad equilibrium, and we aren’t going to get out of it in a day.
And government housing construction is just one idea for how to extend Georgist thinking beyond the single, simple idea of a Land Value Tax. There are probably others, and I encourage Georgists to think about what these might be. Georgism, ultimately, is about finding ways to redistribute land wealth while leaving the wealth of productive enterprise alone (or even increasing it). As they attempt to sell land-redistributionist ideas to an American middle class for whom housing wealth is the only lifeline, Georgists should think very hard about how to implement systems that let the middle class come out ahead in the final equilibrium.
One thing that could be really interesting with your modified Singapore idea is that the entry of firms like Blackrock into the single family rental market creates the perfect opportunity for eminent domain. A state being able to make a cash offer to one or two resident owners and seizing the rest of the block from a corporation to build affordable housing could be an incredibly effective way of turning certain single family areas into more dense housing. The the opponents would lose most of the normal left-NIMBY arguments if the options are new affordable housing managed by the state or literally Blackrock.
It seems too obvious. Build government housing where there's not enough homes to match demand. Keep the revenue to extract tax from land value without much harm to current homeowners. Use this process as a tool to prevent overly fast land value appreciation in high demand areas. Bonus: Lower the risk of home ownership.
Perfect just like Henry George's LVT. Only the problem turns from illiquidity to NIMBYs.