The Ground

The building ages while the address grows dear. A roof needs replacing, a staircase wears, the wiring belongs to an earlier century; yet the house can sell for far more than it did when all these things were new. Some of the difference may reward improvements, and some may reflect a change in the price of credit or the expectations of buyers. But the address carries a claim the builder never supplied: a place within reach of particular work, schools, streets, and other people. The city has grown around the property, and the owner holds a portion of the access it creates. This is how the oldest subject in the theory of rent returns inside a thoroughly modern asset. The buyer sees a house. The purchase joins a structure that can be made again to a position that cannot be copied, and the distinction between them determines how much of the price rewards building and how much pays for being there.

Housing brings the distinction between wealth and income into unusually close view. A property's sale price gathers up what buyers expect ownership to provide over time, including the value of living there or letting someone else do so. Its annual rental income is a flow; its price is the present valuation of an expected stream. A lower discount rate can raise that valuation without an equivalent increase in the rent charged this year. An improvement in nearby transport can raise the worth of the location without improving the house. A change in permitted use can alter what a developer expects to earn from the parcel. Each can enter the same sale price. The household that owns its home receives housing services even when no tenant sends a cheque, while the household that has borrowed against it also owes a lender. To read rising house values as a single measure of pure land rent would collapse these different claims. The ground matters precisely because it must be separated from the building, the financing, and the services of shelter, rather than made to stand for all of them.¹

The return of housing to the center of distribution is nevertheless more than a story told by estate agents. Rognlie's reconstruction of postwar income shares in the G7 found that the long-run increase in capital's net share came from housing. Outside housing, a later recovery largely reversed an earlier decline. The result concerned income after depreciation, over a sample ending in 2010, and its force lies in the composition it revealed: the account of capital could not be read as though every asset were another factory machine. Housing includes both reproducible structures and scarce locations, and its income may rise because access has become dearer as well as because more shelter has been supplied. This brings the volume back to its central question. When a growing claim on income accompanies the expansion of productive capacity, one history has to be explained. When it accompanies the rising cost of entry to an existing place, another history has to be explained. The aggregate word capital does not decide between them.²

A fixed site can accommodate a changing amount of shelter. The corner cannot be reproduced, but a second dwelling, another floor, or a larger building may let more households share access to it. Transport can bring additional sites within reach. These adjustments cost something: materials, skilled work, infrastructure, the inconvenience of construction, and sometimes the loss of amenities that neighbors value. Geography imposes its own limits. Yet permission can become a further scarce input, through a restriction on height, a minimum lot size, or a rule allowing only one dwelling where several could physically stand. Where demand rises against a supply that can expand, part of the response can take the form of new homes. Where expansion is obstructed, more of the pressure can enter prices. Glaeser and Gyourko's account of housing supply places this difference at the center of the divergence among cities. It also explains why a costly house is insufficient evidence by itself: the inquiry must establish which costs and constraints prevent an additional home, and what public benefit, if any, the restriction secures.³

The person who owns a scarce home has a different stake in that adjustment from the person trying to acquire one. More neighbors may bring trade, public services, and a livelier street; they may also bring congestion, construction, and competition for existing properties. An owner whose savings are concentrated in a house can reasonably fear a loss in its value, even when the wider city would benefit from admitting more people. Prospective residents face the reverse problem: they can gain from a building that does not yet exist, but they may have no vote where permission is decided. The political imbalance grows from these positions without requiring every homeowner to share the same motive. It also divides generations within a single family. A parent may welcome the appreciation that supports retirement while watching the same rise move a first home beyond a child's earnings. The wealth gain is real to the owner, yet its social meaning depends on what produced it. Making a neighborhood more valuable and making access to it more difficult can both enrich someone who already holds a title.⁴

Permission can change the amount built, though the effect has to be established in the place where the rules change. Auckland's 2016 reform allowed more intensive housing across much of its residential land. Ryan Greenaway-McGrevy and Peter Phillips compared permitted dwellings in upzoned and non-upzoned areas and found evidence of increased construction activity. Their central difficulty was that building might move from one part of the city to another, making a simple comparison overstate the addition to the whole. They therefore examined how their conclusion changed under alternative assumptions about that displacement. The evidence supports the possibility of enlarging housing supply by relaxing a binding restriction; the measured permits do not by themselves establish how far rents fell, how many homes were completed, or how every tenant fared. The distinction matters for the argument about land. A city need not make another acre of the same location to let more people live near it. It must make the added shelter feasible, and then make provision for the services and access that allow the additional households to use it.⁵

The remedies consequently address different parts of the house. Allowing additional homes loosens a constraint on the supply of shelter. Taxing the value of the underlying site, the proposal examined in The Single, addresses who receives the location's value. The first does not abolish the scarcity of a particularly desirable address; the second does not pour a foundation or issue a building permit. Both require attention to the distinction between land and improvements, and both meet owners who may have paid the capitalized value of an expected advantage long before a reform was proposed. The ground has become a dividing line because a title can confer access, security, and a claim on the growth of the surrounding city at once. Those outside it must buy or rent their way across that line. What portion of the charge reflects the place itself, what portion the work of building, and what portion a scarcity maintained by rules is the inquiry that the price alone leaves unfinished. The next plate follows the rule into another market, where the permission to work can itself become something scarce.