The Policy Economists Agree About

Rent control is the standard example of a policy economists agree about. Surveys find levels of consensus comparable to those on free trade. The consensus is also narrower than the slogan that carries it, and the gap between the two is where the useful argument lives.

The textbook case concerns what is called first-generation control: a cap attached permanently to the unit, regardless of who occupies it or when they arrived. Rent is held below the market level, the gap grows over time, and two things follow. Tenants who hold such a tenancy gain a great deal. Owners, facing a return that falls further behind alternatives every year, respond by converting units to other uses, redeveloping, selling to owner-occupiers or simply maintaining them less. The stock available to rent shrinks.

The best evidence is the San Francisco study by Diamond, McQuade and Qian, which used a quasi-experiment created by a 1994 ballot measure and followed individual tenants and buildings. Both halves of the theory happened at once. Covered tenants were substantially more likely to remain in the city, which was the policy's stated aim and a real benefit. Landlords reduced rental supply by around fifteen per cent through conversion and redevelopment, and that reduction pushed citywide rents up. The policy transferred to incumbents and charged the cost to everyone who arrived later: a group that is younger, poorer and more mobile than the beneficiaries.

There is a second effect that gets less attention and matters. A controlled tenant loses the accumulated discount by moving, so people stay in homes long after those homes stop suiting them: a single occupant in a family flat, a household commuting across a city rather than surrendering a below-market tenancy. The stock ends up allocated by length of tenure, which correlates only loosely with need.

None of this is the end of the argument, because first-generation control is not what most countries now do. Second-generation systems limit increases during a tenancy, allow market rents when a unit is re-let, and exempt new construction entirely. This targets what tenants actually complain about (unpredictable increases forcing an unplanned move) while leaving intact the return on building a new unit. Most European regulation looks like this. It does considerably less damage, and it delivers considerably smaller transfers, which is the trade being made rather than a free lunch.

The honest defence concedes the mechanism rather than denying it. Denying the supply response requires ignoring good evidence and it is not necessary to the case. Security of tenure has real value: it keeps children in schools, keeps adults near work and family, and removes a category of financial shock that falls hardest on people with no buffer against it. The question is whether that value exceeds the units not built or converted away. Stated that way, the disagreement becomes one about design and magnitude, which is answerable.

Stated the other way, as a slogan about price ceilings on one side and landlords on the other, it is not answerable, and it has not been answered in eighty years of arguing.