Macro Paper Warehouse Forthcoming macro & monetary research
Published [American Economic Review] doi:10.1257/aer.20240513 Vol. 116, No. 3, pp. 1014-1051

Dynamics of the Long-Term Housing Yield: Evidence from Natural Experiments

Veronica Backer-Peral

Jonathon Hazell

Atif Mian

What this paper finds — and why it matters

Each month a fraction of UK property leases are extended by 90 years or more, creating thousands of natural experiments in which the same property’s rent and capital value are revealed simultaneously. This paper uses these lease extensions — and Massachusetts and Cambridge rent-control removals as a second identification strategy — to estimate the expected long-term housing yield (annual rent-to-price ratio) and decompose its dynamics into rent-growth expectations and discount-rate components. The central finding is that housing yield movements are dominated by discount-rate shocks: variation in required returns on housing explains the overwhelming majority of yield variance, while expected rent growth contributes less than 10 percent. Housing booms are therefore primarily driven by falling required returns, not by rational expectations of higher future rents. The yield responds to real long-term interest rates with a slope significantly below one, consistent with a non-pecuniary convenience yield on housing that is not fully displaced by interest rate changes.

Summary of a forthcoming paper, AI-assisted and human-reviewed. See the linked original for the authoritative claims and full conditions.


In depth

Q1. What do the natural experiments identify?

Lease extensions reveal the market’s valuation of the same physical dwelling at two points — just before and just after the 90-year extension — with the extension itself creating a clean variation in the remaining lease term (and hence in the present value of ownership) without changing the property’s rent-generating characteristics. This design separates the rent and price components of the yield at the property level, allowing identification of discount-rate and rent-growth contributions free of compositional differences across properties.

Q2. Why do discount rates dominate yield variation?

A present-value decomposition of the housing yield into expected rent growth and the discount rate assigns more than 90 percent of variance to the discount rate component, implying that periods of low housing yields (high prices relative to rent) reflect primarily that investors demand a lower return on housing — not that they expect rents to rise faster. This result mirrors Campbell-Shiller findings for equity markets but is especially striking for housing, where naive narratives often attribute booms to expected rent appreciation.

Q3. What does the convenience yield interpretation imply?

Housing yields respond less than one-for-one to real interest rate movements — a slope well below one in the yield-rate regression — implying that housing carries a non-pecuniary convenience yield (liquidity, collateral value, direct utility of ownership) that buffers the required return on housing against interest rate changes. When real rates rise, housing yields rise by less, so price-to-rent ratios decline by less than a frictionless model would predict.

Key concepts

housing yield : the annual rent-to-price ratio on residential property; the paper’s central object, decomposed into discount-rate and rent-growth components.

discount-rate channel : the dominant source of housing yield variation in this paper; movements in investors’ required return on housing, not expected rent growth, drive the observed yield dynamics.

convenience yield : the non-pecuniary value of housing ownership (liquidity, collateral, direct utility) that drives a wedge between the housing yield and the risk-free real interest rate; explains the less-than-one slope in the yield-rate relationship.

How this summary was made. Bibliographic fields are pulled from Crossref and OpenAlex and are not model-generated. The summary was drafted from the open-access manuscript , checked by a claim-grounding and calibration review pass, and approved before publishing. Found an error or a misrepresentation? Flag it here — corrections are welcome, especially from the authors.