The Thing They Stopped Making

Land is not capital, and treating it as though it were is responsible for a good deal of confused argument about housing, wealth and taxation. Capital is produced: if its return rises, more of it gets made. Land is not produced. Its supply is what it is, whatever anybody pays for it.

David Ricardo drew the conclusion in 1817 in a sentence that still does most of the work: corn is not high because a rent is paid, but a rent is paid because corn is high. The causal direction is the entire point. The price of grain is set at the margin, by the least productive land it is worth cultivating at all, and that marginal land, by definition, earns no rent. Better land produces more for the same effort, and the difference is the rent its owner can extract. Landlords are not adding a cost to the loaf. They are capturing a surplus that the loaf's price created.

From this follows the property that makes land unique in tax policy. Passing a tax on to somebody else requires reducing supply so that the price rises. Producers of cars, or beer, or labour can all do this to some degree. Nobody can produce less land. The owner is already charging what the market will bear, and a tax on the land's value therefore comes out of the owner's residual and stays there. It is the one substantial tax with essentially no deadweight loss, a conclusion that economists across every political division have agreed on for two centuries.

Henry George built a movement on it. Progress and Poverty, published in 1879, sold in the millions and made him one of the most widely read writers of the century. His argument was that the value of land is created by the community around it (by roads, schools, neighbours, the presence of other people's work) and that taxing it in place of almost everything else would raise revenue without discouraging any productive activity. The economic case has never been seriously refuted. It has also been implemented at scale essentially nowhere, because land is the principal asset of a large, wealthy and politically effective group who would bear the entire burden.

The contemporary relevance is not historical. Almost all of the rise in housing costs across rich countries over the past forty years is land, not buildings. Real construction costs per square metre have moved relatively little; the ground underneath has multiplied, and in the most constrained cities it is now the large majority of a home's price. This matters because it means productivity gains in construction cannot solve the problem. The expensive part is not being built.

The clearest illustration of the value-capture argument is infrastructure. A new railway station raises nearby land values as soon as the route is credible, before a single passenger travels, because buyers are paying for future access. That gain goes to whoever happens to own the ground. It was funded by taxpayers or fare payers generally. Schemes to recoup some of it exist in a handful of places, and the fact that they are contentious rather than automatic tells you most of what you need to know about where the political weight sits.