Zahratu Shabrina, Elsa Arcaute, Michael Batty (2021), Urban Studies 59(1), 197–221 (2022 issue; online 2021-01-16) — DOI-10.1177/0042098020970865
Key takeaways
- Using 2018 Airbnb data, the authors estimate that more than 2% of London properties, and up to 7% in some areas, are likely “misused” as short-term lets. Such misuse clusters where dwelling-type diversity is low (flats) and private renting is high. At borough level, a 100% increase in misuse density is associated with up to an 8% rise in rent per bedroom per week, about £90 a year.
Seeded from the abstract and a skim, before the reading in NOTE-071. What follows is what the work says about itself.
The article builds proxies for Airbnb’s impact on Greater London housing by relating likely Airbnb misuse to local housing attributes. Misuse is defined as whole-property listings that break local regulation and are offered by hosts with multiple listings. The authors examine dwelling type, tenure (owned or rented) and the spatial pattern of rent changes. They find more than 2% of London properties, and up to 7% locally, likely misused. Misuse correlates negatively with dwelling-type diversity and positively with flats in areas with high private renting. A doubling of misuse density is associated with up to 8% higher rent per bedroom per week, and they discuss the policy implications.
Standing in the record
Filed from the survey of 2026-09-25 of work the anthology set aside as out of scope (tier C): 450 seconds of active reading over 2 sessions in the papers-feed tracker. It was filed Deferred, unread. NOTE-071 is the close reading of 2026-09-25, and it placed the work: Active — worth reading as a descriptive map of possibly non-compliant Airbnb supply in London and for its misuse definition. Its rent result is a 33-observation cross-sectional association whose “8% per doubling” translation does not follow from the reported model, and it should not be cited as an effect size.
When seeded, before the reading in NOTE-071, the entry gave a priority for a deeper reading: low — 450 s over 2 sessions and “file later” triage. It is a domain-specific empirical study whose headline figures are captured here.
What the seeding asked a deeper reading to check (NOTE-071 is that reading):
- A spatial case study of platform effects on housing, using entropy-based urban-complexity measures from the Batty school. Its relevance to the anthology is mostly as social-science background.
- The rent result is associational, with borough-level aggregation (London has 33 borough-level units; I did not check the regression’s actual n). Check the model’s sample size and specification before using the 8% figure.
- The “misuse” operationalization (days available, multi-listing) is the key assumption, so check its thresholds.
Access when seeded: Crossref (online 2021-01-16) and the published PDF deposited in UCL Discovery (eprint 10120512, 25 pp.), which I read: abstract, keywords, introduction opening, data list, entropy and regression method passages, and discussion/conclusion. SAGE’s PDF returned 403. The journal records it as received March 2019 and accepted September 2020. There is a related earlier preprint by the same authors, arXiv:1903.11205 “Airbnb’s disruption of the housing structure in London” (v1 27 Mar 2019). Its abstract reports different figures (more than 1.4% of supply, up to 20% in some neighbourhoods), so I treat it as a precursor, not the same work.