The best causal studies of short-term rentals find that Airbnb’s growth raised rents and house prices by a few per cent on average, and by several times as much in the central, tourist-heavy neighbourhoods where listings are densest. They measure prices; the replacement of residents that the word gentrification describes lies outside what they observe. The restrictions evaluated so far reversed effects of the same modest size, while those in New York, Barcelona and Spain have not yet been evaluated in peer-reviewed work.
Between September 2023 and May 2026, New York, Barcelona, Spain and the European Union each adopted or began to enforce new rules on short-term rentals. New York City began enforcing Local Law 18 on 5 September 2023.1 Barcelona announced in June 2024 that it would stop renewing its 10,101 tourist-flat licences from November 2028.2 Spain now runs a single national registry for short-term rentals,3 and EU Regulation 2024/1028 on short-term rental data has applied since 20 May 2026.4 Each measure was defended in part by the claim that short-term rentals raise housing costs, a process the public debate often calls gentrification.
The causal evidence on short-term rentals and housing is narrower and more consistent than the public debate. The best studies find that Airbnb’s growth raised rents and house prices by a few per cent on average, and by several times as much in the central, tourist-heavy neighbourhoods where listings are densest. They measure prices, and the replacement of residents that the word gentrification describes lies outside what they observe; concentration is the better word for what they find. The restrictions evaluated so far reversed effects of the same modest size.
Three tests can falsify this reading on published evidence.
- Size. Peer-reviewed causal studies for the United States, Barcelona, Los Angeles, Berlin and Portugal must report average effects of a few per cent or less. If the best estimates are large on average, the thesis fails.
- Concentration. Studies that report effects by neighbourhood must find them several times larger where listings are densest. If effects are uniform, the thesis narrows to size alone.
- Reversal. Evaluations of restrictions must show effects of similar size in the opposite direction, with stated methods. Where no credible evaluation exists, the test stays open.
The first two tests pass, each with one qualification set out below. The third passes for Los Angeles and Berlin. It is open for New York, Barcelona and Spain, the three most recent restrictions reviewed here.
The question matters because of what governments have said. New York City’s enforcement office states that illegal short-term rentals “have been linked to rising rents and the removal of homes from the local housing market”.5 The economic literature on that link is about a decade old and now includes several studies with credible causal designs. This Dispatch, which opens Titanica’s Hospes Domain on travel and hospitality as an economy that reshapes places, sets out what those studies measured, where the effects fall, how much housing the listings occupy, and what the restrictions have shown so far.
What the word gentrification describes
The term has a precise origin. In 1964 the sociologist Ruth Glass wrote that “one by one, many of the working class quarters of London have been invaded by the middle classes”, and that once the process starts in a district “it goes on rapidly until all or most of the original working class occupiers are displaced and the whole social character of the district is changed”.6 In Glass’s usage, and in most of the urban-studies literature that followed, the defining feature is a change of population: who leaves, who arrives, and what kind of place results. Higher rents can be a cause or a symptom of that change. They are a different variable.
Urban geographers have connected short-term rentals to gentrification through rent-gap theory, which concerns the difference between what a property earns in its current use and what it could earn in another. David Wachsmuth and Alexander Weisler, in a 2018 article on three years of Airbnb activity in New York City, argued that the platform creates “a new form of rent gap in culturally desirable and internationally recognizable neighborhoods”, and framed their study around “the fear that short-term rentals are facilitating gentrification”.7 Their study set out to identify the neighbourhoods where that gap was opening and to estimate how much rental housing had been lost to the platform.
The econometric studies discussed below ask a narrower question. They estimate how rents and house prices respond to more Airbnb listings, and in two cases how the supply of long-term rental housing responds. None of them tracks which residents moved out and which moved in. A finding that rents rose by two per cent where listings grew is evidence about prices in a housing market. It is consistent with gentrification and also with many other outcomes, including higher costs for residents who stay. The word that describes what these studies find, across every design, is concentration: the listings cluster in a small number of places, and so do the measured effects.
Five designs, one direction
Estimating the effect of Airbnb on rents is difficult because listings appear where demand is already rising. A neighbourhood that attracts tourists may attract new residents and investment at the same time, so a simple correlation between listings and rents overstates the platform’s role, or occasionally understates it. The studies that carry most weight each use a source of variation in listings that is plausibly unrelated to other local trends.
Kyle Barron, Edward Kung and Davide Proserpio, in the study published in Marketing Science in 2021, used listings across the United States and an instrumental-variables strategy.8 Their instrument combines worldwide Google searches for “airbnb”, which measure the platform’s national growth over time, with each zipcode’s tourism intensity in 2010, measured by its number of food-service and accommodation establishments.9 The regressions cover monthly zipcode data for the 100 largest metropolitan areas from 2011 to 2016. At the zipcode with the median owner-occupancy rate, “a 1% increase in Airbnb listings leads to a 0.018% increase in rents and a 0.026% increase in house prices”.8 Applied to median annual listing growth, the authors translate this into an annual increase of $9 in monthly rent and $1,800 in house prices for the median zipcode, which they calculate as “about one-fifth of actual rent growth and about one-seventh of actual price growth”. They also find that total housing supply did not change: listings moved units from the long-term to the short-term rental market.
Miquel-Àngel Garcia-López, Jordi Jofre-Monseny, Rodrigo Martínez-Mazza and Mariona Segú studied Barcelona, using data on rents and on both transaction and posted prices. They combined panel fixed-effects models, a shift-share instrumental-variables design in which tourist amenities predict where listings locate and Google searches predict when, event studies, and a case study of the area around the Sagrada Família.10 Their main result is that, for the average neighbourhood, “Airbnb activity has increased rents by 1.9%, transaction prices by 4.6% and posted prices by 3.7%”.
Hans Koster, Jos van Ommeren and Nicolas Volkhausen used regulation as the source of variation. Eighteen of the 88 cities in Los Angeles County adopted Home Sharing Ordinances that severely restricted short-term rentals. Comparing properties on either side of those cities’ borders in a panel regression-discontinuity design, they find that the ordinances “reduced listings by 50% and housing prices by 2%”, and difference-in-differences estimates show that rents fell “also by 2%”.11 The baseline estimates are −1.8 per cent for prices, with a standard error of 0.7 percentage points, and −2.3 per cent for rents, with a standard error of 0.9.
Tomaso Duso, Claus Michelsen, Maximilian Schaefer and Kevin Tran exploited two policy interventions in Berlin. In the published version of their study, they estimate that “each additional commercial Airbnb listing displaces 0.23 to 0.37 rental units” and “increases rent per square meter by 1.3 to 2.4 percent”.12 Sofia Franco and Carlos Santos studied Portugal, where a 2014 decree simplified the registration of short-term rentals. An earlier working-paper version describes the design as a difference-in-differences comparison of areas with high and low Airbnb presence before the reform.13 The published article finds that “on average a 1pp increase in a municipality Airbnb share results in a 3.7% increase in house prices”, and that in the civil parishes of Lisbon and Porto a one-point increase leads to a 3.2 per cent price increase.14
| Study | Place and period | Design | Estimated effect |
|---|---|---|---|
| Barron, Kung & Proserpio, Marketing Science (2021) | United States, zipcodes in the 100 largest metro areas, 2011–2016 | Causal: instrumental variables (Google searches × 2010 tourism intensity) | 1% more listings: rents +0.018%, prices +0.026% (median owner-occupancy zipcode); about $9 a month in rent and $1,800 in price per year |
| Garcia-López, Jofre-Monseny, Martínez-Mazza & Segú, J. Urban Economics (2020) | Barcelona, neighbourhoods | Causal: fixed effects, shift-share instrument, event studies | Average neighbourhood: rents +1.9%, transaction prices +4.6%, posted prices +3.7% |
| Koster, van Ommeren & Volkhausen, J. Urban Economics (2021) | Los Angeles County, 2014–2018 | Causal: panel regression discontinuity at the borders of 18 regulating cities; difference-in-differences for rents | Ordinances: listings −50%, house prices −2%, rents −2% |
| Duso, Michelsen, Schaefer & Tran, Reg. Sci. & Urban Economics (2024) | Berlin | Causal: two policy interventions as shocks to listings | Each additional commercial listing: 0.23 to 0.37 long-term rental units displaced; rent per m² +1.3% to +2.4% |
| Franco & Santos, Reg. Sci. & Urban Economics (2021) | Portugal; parishes of Lisbon and Porto; to 2016 | Causal: difference-in-differences around the 2014 registration reform | 1 point more Airbnb share: prices +3.7% (municipalities), +3.2% (Lisbon and Porto parishes) |
| Horn & Merante, J. Housing Economics (2017) | Boston | Correlational: authors report an association | 1 standard deviation more listings: asking rents +0.4% |
The units in the table differ, and the differences matter. Barron and his co-authors report an elasticity; Garcia-López and his co-authors report the cumulative effect of the listings that existed by the end of their period; Koster and his co-authors report the effect of removing about half the listings; Duso and his co-authors report the effect of one listing; Franco and Santos report the effect of a one-point rise in the share of housing listed. They cannot be averaged. They can be read for sign and order of magnitude, and on both they agree. Every causal estimate implies that more listings raise rents or prices. Where a study reports an average effect over its period, as in Barcelona and Los Angeles, it is about two to five per cent. A correlational study of Boston by Keren Horn and Mark Merante, which uses the word “associated” for its result, found that a one-standard-deviation increase in listings went with asking rents 0.4 per cent higher.15
The first test therefore passes on levels. The qualification concerns growth. Barron and his co-authors attribute about a fifth of the rent growth in their median zipcode to Airbnb. At the sample median, the working paper reports that listings grew 28 per cent a year.9 Multiplying that growth by the elasticities gives about 0.5 per cent a year for rents and 0.7 per cent for prices, computed by Titanica from the paper’s figures. An effect that is small in level can be a meaningful part of the change over a few years. Kung said at publication that concerns about the effect of Airbnb on the housing market “do not appear unfounded”.16
Where the effect falls
The Barcelona study reports its results by intensity of Airbnb activity. For neighbourhoods in the top decile of the distribution, the authors estimate that rents increased by 7 per cent, transaction prices by 17 per cent and posted prices by 14 per cent.10 Each figure is 3.7 to 3.8 times the average-neighbourhood estimate.
| Measure | Average neighbourhood | Top decile of Airbnb activity | Ratio, top decile to average |
|---|---|---|---|
| Rents | +1.9% | +7% | 3.7 |
| Transaction prices | +4.6% | +17% | 3.7 |
| Posted prices | +3.7% | +14% | 3.8 |
| Licensed tourist flats by district, August 2024: l’Eixample 4,467 of 9,818 (45.5%); Nou Barris 40 (0.4%); eight of the city’s neighbourhoods had none | |||
The licence records show why a top decile can drive the result. Of the 9,818 tourist-flat licences valid in Barcelona in August 2024, 4,467, or 45.5 per cent, were in the Eixample district, and the Dreta de l’Eixample neighbourhood alone held 1,655.17 Nou Barris held 40. Eight neighbourhoods had none.
The same pattern appears in the other studies, measured in different ways. In the United States, the effect is larger where fewer residents own their homes, which the authors describe as consistent with non-owner-occupiers being more likely to move homes from the long-term to the short-term rental market.8 In the working-paper version, at the 25th percentile of owner-occupancy, 56 per cent, a 1 per cent increase in listings raises rents by 0.024 per cent and prices by 0.037 per cent; at the 75th percentile, 82 per cent, the figures are 0.014 and 0.019.9 In Los Angeles, Koster and his co-authors extrapolate from their estimates that Airbnb raised house prices by almost 15 per cent within 2.5 kilometres of Hollywood’s Walk of Fame, and by 5.8 per cent within 2.5 kilometres of beaches, compared with the roughly 2 per cent by which the ordinances lowered prices in the regulating cities.11 In Portugal, compared with a parish with few tourists, house prices in a highly touristic parish rose 24.3 per cent in 2015 and 32.3 per cent in the first quarter of 2016 relative to the period before the second quarter of 2014, according to the published abstract.14 The working-paper version reports, for areas with high Airbnb presence before the 2014 reform, difference-in-differences estimates of 34.0 per cent for house prices and 10.9 per cent for rents by 2016.13 These are the largest estimates in the literature reviewed here, and they apply to the places where listings were densest.
Berlin adds the second qualification. In the 2020 discussion-paper version of the Berlin study, the authors found “some effect heterogeneity across the city”, with “areas with lower Airbnb density” tending “to be affected more by additional Airbnb listings”.18 The effect of one more listing is therefore not larger where listings are dense. The total effect in a neighbourhood is the effect per listing multiplied by the number of listings, and the number is what varies most. On Titanica’s reading of these results, the concentration lies in the listings, and the measured effects follow the listings. The second test passes in that form: effects are several times larger where listings are densest, because that is where most of the listings are.
How much housing the listings occupy
The size of the average effect follows from the size of the stock. In the largest cities, short-term rentals occupy around one per cent of housing. New York City’s enforcement office counted more than 38,000 active listings on a single site at the beginning of 2023.19 The city’s 2023 Housing and Vacancy Survey, conducted between January and mid-June that year, counted 3,705,000 housing units.20 The ratio is about 1.0 per cent. That figure is an upper bound for entire homes, because listings include private rooms and some hosts list more than one unit, and it compares two different sources. By September 2026 the city counted more than 3,500 active approved hosts,5 about 0.09 per cent of the 2023 stock.
| Place | Measure | Date | Share of housing units |
|---|---|---|---|
| New York City | Active listings on one site (38,000+) vs housing units (3,705,000) | Early 2023 | about 1.0% (computed) |
| New York City | Active approved hosts (3,500+) vs 2023 housing units | Sep 2026 | about 0.09% (computed) |
| Madrid | Tourist dwellings, INE (15,242) | May 2025 | 0.99% |
| Barcelona | Tourist dwellings, INE (9,579) | May 2025 | 1.18% |
| Valencia | Tourist dwellings, INE (6,553) | May 2025 | 1.58% |
| Sevilla | Tourist dwellings, INE (7,137) | May 2025 | 2.18% |
| Málaga | Tourist dwellings, INE (8,597) | May 2025 | 3.28% |
| Marbella | Tourist dwellings, INE (7,934) | May 2025 | 7.91% |
| Lisbon, Santa Maria Maior parish | Airbnb listings vs housing units | Q1 2014 · Q3 2016 | 13.0% · 22.6% |
| Yaiza (Lanzarote) | Tourist dwellings, INE; highest municipality in Spain | May 2025 | 24.4% |
Spain’s statistics office measures tourist dwellings with a consistent method across municipalities. For May, in figures reported in July 2025, it counted 15,242 in Madrid, 0.99 per cent of dwellings, and 9,579 in Barcelona, 1.18 per cent.21 The shares rise along the coast and on the islands: 3.28 per cent in Málaga, 7.91 per cent in Marbella and 24.4 per cent in Yaiza, on Lanzarote, the highest in the country. Within cities, shares can be far higher in historic centres. In the parish of Santa Maria Maior, which covers much of central Lisbon, Airbnb listings equalled about 13.0 per cent of housing units in the first quarter of 2014 and about 22.6 per cent in the third quarter of 2016, according to the tables of the working-paper version of the Portuguese study.13
The per-point estimates allow an illustrative comparison. Franco and Santos find 3.7 per cent higher prices for each point of housing share; Koster and his co-authors, using the ordinances as an instrument, find 3.0 to 5.1 per cent for a one-point increase in the listings rate, defined as listings per hundred housing units within 200 metres of a property.11 Applied to a citywide share of about one per cent, these coefficients imply price effects of about three to five per cent. This is illustrative arithmetic by Titanica, not a measured result for any city, and the coefficients cannot be extended in a straight line to shares of 20 per cent. In the Portuguese areas with the highest shares, the measured price increases were 24 to 34 per cent. The arithmetic shows why the averages are modest: most housing in most cities has few listings near it.
What the restrictions have shown
A restriction that removes listings offers the reverse experiment. If listings raised rents by a given amount, removing them should lower rents, or slow their growth, by a similar amount where the listings were. Two restrictions have been evaluated in peer-reviewed work. The restrictions in New York, Spain and Barcelona have not.
Berlin’s ban on the misuse of housing, the Zweckentfremdungsverbot, has applied since 1 May 2014. For the 6,300 holiday flats registered under it, a two-year transition period ended on 30 April 2016, after which they were to return to the rental market.22 Under the amended law, every advertisement had to show a registration number from 1 August 2018, a permit to let a second home generally required letting on no more than 90 days a year, and breaches of the new duties carried fines of up to €250,000.23 In the 2020 version of their study, Duso and his co-authors found that the policy changes reduced entire homes listed on Airbnb by eight to ten listings per square kilometre.18 In the published version, “the first intervention, which affected commercial listings, significantly impacted long-term rental markets”, while the second mainly affected non-commercial listings.12
In Los Angeles County, the Home Sharing Ordinances cut listings by about half and lowered prices and rents by about 2 per cent each, an effect of the same order as the increases estimated elsewhere.11 Portugal’s 2014 reform ran in the other direction, easing registration, and was followed by the price increases described above.13
| Place | Measure | Dates | Documented outcome | Evaluation of rents |
|---|---|---|---|---|
| Berlin | Zweckentfremdungsverbot; registration numbers in advertisements | 1 May 2014; transition ended 30 Apr 2016; 1 Aug 2018 | Entire-home listings down 8 to 10 per km² | Peer-reviewed: per commercial listing, 0.23 to 0.37 rental units and 1.3% to 2.4% rent per m² |
| Los Angeles County | Home Sharing Ordinances in 18 of 88 cities | Study data 2014–2018 | Listings −50% | Peer-reviewed: prices −2%, rents −2% |
| New York City | Local Law 18: registration; platforms barred from unregistered bookings | Enforced from 5 Sep 2023 | Short-term listings 21,900 (Jul 2023) to 3,700 (Jul 2024), company data; 3,500+ approved hosts (Sep 2026); 618 rent-regulated units protected | None peer-reviewed found; one independent analysis finds no detectable effect |
| Spain | National registration number for short-term rentals | Required from 1 Jul 2025 | About 65,000 listings removed by Airbnb; 54,728 more reported (Jul 2025) | None found |
| Barcelona | Non-renewal of tourist-flat licences | Announced Jun 2024; from Nov 2028; upheld Mar 2025 | Not yet in force | None possible yet |
| European Union | Regulation 2024/1028 on short-term rental data | Applies from 20 May 2026 | Data obligations | Not a restriction |
Among the restrictions in force reviewed here, New York’s is the most recent and the least evaluated. AirDNA, a company that sells short-term rental data, reported that active listings for stays of under 30 nights fell from 21,900 in July 2023 to 3,700 in July 2024, a decline of 83 per cent, while listings for stays of 30 nights or more rose 29 per cent and total Airbnb listings fell 24 per cent.24 Part of the stock therefore moved to monthly rentals. The city reported 3,000 active registrations in September 2025 and more than 3,500 active approved hosts in September 2026, and it counted 618 rent-regulated units protected through denied applications.195 It did not publish an estimate of the effect on rents. Airbnb said in a blog post a year after enforcement began that the law had “failed to deliver on the promise to combat the housing crisis”.25 The consultancy HR&A Advisors lists an impact assessment of the law among its projects, with Airbnb as client; its project page reports no figures.26 One independent analysis, published on a newsletter platform in July 2026 and not peer-reviewed, compared asking rents in high- and low-density areas within the same borough with an event-study difference-in-differences design and estimated savings of about $52 a year for a typical renter, with a confidence interval from −$71 to +$175, and concluded that “there was no notable effect size”.27 As of 10 November 2026, Titanica found no peer-reviewed evaluation of the law’s effect on rents.
In Spain, Airbnb agreed in July 2025 to send the housing ministry monthly data on its listings, including registration numbers,3 and the Ministry of Consumer Affairs reported on 30 July 2025 that Airbnb had removed about 65,000 listings without a licence and that a further 54,728 lacked the registration number required since 1 July 2025.28 Barcelona’s measure is not yet in force. Spain’s Constitutional Court rejected an appeal by the People’s Party against the Catalan decree on which it rests, a ruling Catalan News reported on 14 March 2025.29 The city’s licence counts differ by source and date: 10,101 in the June 2024 announcement, 10,294 in the second quarter of 2024 according to city data reported by 3Cat, and 9,818 valid licences in August 2024.23017 The third test therefore passes where it can be run, in Los Angeles and Berlin, and is open in New York, Spain and Barcelona.
The case that short-term rentals are a rounding error
The strongest objection can be stated in three sentences. The housing shortage in these cities is a shortage of supply, and short-term rentals occupy about one per cent of the housing stock in New York, Madrid and Barcelona. The studies’ own average effects are a few per cent. Restricting short-term rentals therefore addresses a small part of the problem, chosen because it is visible, while the larger part, the shortfall of new building, continues.
The evidence supports much of this at the scale of a city. The shares in Exhibit 03 are close to one per cent, and Barron and his co-authors found that Airbnb did not change the total supply of housing.8 Listings moved existing units from the long-term to the short-term rental market. The New York data also show that part of the delisted stock moved to stays of 30 nights or more, which the registration requirement does not cover.24 Claims about large citywide gains depend on how many units return to ordinary renting. Barcelona’s mayor said the 2028 measure would return to the market housing “equivalent to a decade of new construction”;29 the figure is his, and no evaluation yet exists to test it.
The objection weakens at the scale of a neighbourhood. In a parish where one housing unit in five is listed, or in a top-decile Barcelona neighbourhood with rents 7 per cent and transaction prices 17 per cent higher, the estimated effect is several times the citywide figure. The median-zipcode estimate for the United States also attributes about a fifth of rent growth over its period to Airbnb.8 Both readings can be correct at once, because the citywide average and the central-district effect describe different populations of homes.
The experiments still running
The datasets that would allow the open tests to be run now exist. Spain’s registry and Airbnb’s monthly reporting to the housing ministry, New York’s registration records and, since May 2026, the data-sharing rules of the EU regulation give public authorities listing data tied to registration numbers, where earlier studies relied largely on scraped listings. New York’s law has been enforced for three years and two months. Barcelona’s licences lapse in two years.
The thesis implies where an evaluation should look. If the effect of short-term rentals follows the listings, the effect of removing them should also follow the listings: in the Eixample, which held 45 per cent of Barcelona’s licences in 2024, and in the New York neighbourhoods where listings were densest before 2023, more than in either city as a whole. A citywide rent index may show little change in Barcelona after 2028 even if rents in those districts change substantially. A design that compares districts by their share of listings before the restriction, as the Los Angeles and Berlin studies did, would measure the effect where the evidence places it. Concentration, the word the studies support, is also the hypothesis that the new data can test.
- New York City Mayor’s Office of Special Enforcement, “Registration Law” (official): enforcement of Local Law 18 from 5 September 2023; registration of hosts; platforms barred from processing transactions for unregistered short-term rentals; stays of 30 days or more exempt. https://www.nyc.gov/site/specialenforcement/registration-law/registration.page (accessed 10 November 2026) ↩︎
- Catalan News, report on the city’s plan to eliminate tourist apartments by November 2028, 21 June 2024 (press report): 10,101 licences; non-renewal from November 2028; statements by Mayor Jaume Collboni. https://www.catalannews.com/politics/item/barcelona-eliminate-tourist-apartments-november-2028-21-june-2024 (accessed 10 November 2026) ↩︎
- Government of Spain, La Moncloa, Ministry of Housing and Urban Agenda, “Vivienda acuerda con Airbnb la retirada de los anuncios de pisos turísticos sin código de registro”, 15 July 2025 (official): single short-term rental registry; monthly data from Airbnb from August 2025; removal within 48 hours of revoked numbers. https://www.lamoncloa.gob.es/serviciosdeprensa/notasprensa/vivienda-agenda-urbana/Paginas/2025/150725-airbnb-retirada-pisos-sin-registro.aspx (accessed 10 November 2026) ↩︎
- European Commission, Transition Pathways, “EU Regulation on data collection and sharing relating to short-term accommodation rental services” (official): Regulation (EU) 2024/1028 adopted 11 April 2024; to be implemented by 20 May 2026. https://transition-pathways.europa.eu/legislation/eu-regulation-data-collection-and-sharing-short-term-accommodation-rentals (accessed 10 November 2026) ↩︎
- New York City Mayor’s Office of Criminal Justice and Office of Special Enforcement, “New Report Reveals Number of New Yorkers Registered to Host Legal Short-Term Rentals Surpassed 3,500 for First Time”, 1 September 2026 (official): over 3,500 active approved hosts; 618 rent-regulated units protected; estimate of as many as 18,000 units used as illegal short-term rentals by 2018; vacancy rate 1.4%; rentals of less than 30 days limited to two guests staying with permanent occupants. https://criminaljustice.cityofnewyork.us/press-release/new-report-reveals-number-of-new-yorkers-registered-to-host-legal-short-term-rentals-surpassed-3500-for-first-time/ (accessed 10 November 2026) ↩︎
- The Bartlett, University College London, “Ruth Glass and coining ‘gentrification'” (institutional commentary, undated), quoting Ruth Glass (ed.), London: Aspects of Change, 1964. https://www.ucl.ac.uk/bartlett/ideas/100-years-bartlett/ruth-glass-and-coining-gentrification (accessed 10 November 2026) ↩︎
- David Wachsmuth and Alexander Weisler, “Airbnb and the rent gap: Gentrification through the sharing economy”, Environment and Planning A, 50(6), 2018, pp. 1147–1170, doi:10.1177/0308518X18778038 (peer-reviewed). https://upgo.lab.mcgill.ca/publication/airbnb-and-the-rent-gap/ (accessed 10 November 2026) ↩︎
- Kyle Barron, Edward Kung and Davide Proserpio, “The Effect of Home-Sharing on House Prices and Rents: Evidence from Airbnb”, Marketing Science, 2021, doi:10.1287/mksc.2020.1227 (peer-reviewed; published abstract). https://pubsonline.informs.org/doi/fpi/10.1287/mksc.2020.1227 (accessed 10 November 2026) ↩︎
- Kyle Barron, Edward Kung and Davide Proserpio, “The Effect of Home-Sharing on House Prices and Rents: Evidence from Airbnb”, working-paper version dated 5 September 2019, Wharton School seminar paper (working paper; figures by owner-occupancy rate and median listing growth of 28% a year). https://marketing.wharton.upenn.edu/wp-content/uploads/2019/08/09.05.2019-Proserpio-Davide-Paper.pdf (accessed 10 November 2026) ↩︎
- Miquel-Àngel Garcia-López, Jordi Jofre-Monseny, Rodrigo Martínez-Mazza and Mariona Segú, “Do short-term rental platforms affect housing markets? Evidence from Airbnb in Barcelona”, Journal of Urban Economics, 119, article 103278, 2020, doi:10.1016/j.jue.2020.103278 (peer-reviewed; open-access record, CC BY). https://diposit.ub.edu/dspace/handle/2445/175642 ; https://discovery.ucl.ac.uk/id/eprint/10172148/ (accessed 10 November 2026) ↩︎
- Hans R. A. Koster, Jos van Ommeren and Nicolas Volkhausen, “Short-term rentals and the housing market: Quasi-experimental evidence from Airbnb in Los Angeles”, Journal of Urban Economics, 124, article 103356, 2021, doi:10.1016/j.jue.2021.103356 (peer-reviewed; copy of the published article hosted by The Orcasonian). https://theorcasonian.com/wp-content/uploads/2021/06/1-s2.0-S0094119021000383-main.pdf (accessed 10 November 2026) ↩︎
- Tomaso Duso, Claus Michelsen, Maximilian Schaefer and Kevin D. Tran, “Airbnb and rental markets: Evidence from Berlin”, Regional Science and Urban Economics, 106, article 104007, May 2024, doi:10.1016/j.regsciurbeco.2024.104007 (peer-reviewed). https://research-information.bris.ac.uk/en/publications/airbnb-and-rental-markets-evidence-from-berlin/ (accessed 10 November 2026) ↩︎
- Sofia F. Franco, Carlos Daniel Santos and Rafael Longo, “The Impact of Airbnb on Residential Property Values and Rents: Evidence from Portugal”, Nova SBE Working Paper 630, 8 May 2019 (working paper): difference-in-differences around Decreto-Lei 128/2014; Airbnb listings as a share of housing units in Santa Maria Maior. https://core.ac.uk/download/232115221.pdf (accessed 10 November 2026) ↩︎
- Sofia F. Franco and Carlos D. Santos, “The impact of Airbnb on residential property values and rents: Evidence from Portugal”, Regional Science and Urban Economics, 88, article 103667, 2021, doi:10.1016/j.regsciurbeco.2021.103667 (peer-reviewed). https://run.unl.pt/entities/publication/62ed34d4-0cfe-48c7-bbd1-de7300ea0c6e/full (accessed 10 November 2026) ↩︎
- Keren Horn and Mark Merante, “Is home sharing driving up rents? Evidence from Airbnb in Boston”, Journal of Housing Economics, 38, 2017, pp. 14–24, doi:10.1016/j.jhe.2017.08.002 (peer-reviewed). https://ideas.repec.org/a/eee/jhouse/v38y2017icp14-24.html (accessed 10 November 2026) ↩︎
- INFORMS, “What Impact Does Airbnb Have on Local Housing Prices and Rents?”, news release, 2 February 2021 (official release of the publisher), with quotations from Edward Kung and Davide Proserpio. https://www.informs.org/News-Room/INFORMS-Releases/News-Releases/What-Impact-Does-Airbnb-Have-on-Local-Housing-Prices-and-Rents (accessed 10 November 2026) ↩︎
- Roger Torres i Liñán, betevé, report on tourist-flat licences by neighbourhood, 29 August 2024 (press report of Barcelona City Council licence records): 9,818 valid licences; l’Eixample 4,467; Dreta de l’Eixample 1,655. https://beteve.cat/politica/nomes-vuit-barris-barcelona-no-tenen-cap-pis-turistic-turo-peira-ultim-a-caure/ (accessed 10 November 2026) ↩︎
- Tomaso Duso, Claus Michelsen, Maximilian Schäfer and Kevin Ducbao Tran, “Airbnb and Rents: Evidence from Berlin”, DIW Discussion Paper 1890, 2020 (working paper): policy changes reduced entire-home listings by eight to ten per square kilometre; areas with lower Airbnb density more affected by additional listings. https://www.diw.de/de/diw_01.c.796644.de/publikationen/diskussionspapiere/2020_1890/airbnb_and_rents__evidence_from_berlin.html (accessed 10 November 2026) ↩︎
- New York City Mayor’s Office of Criminal Justice and Office of Special Enforcement, “LL18 Report Sheds Light on Eliminated Illegal Rentals in NYC”, 3 September 2025 (official): over 38,000 active listings at the beginning of 2023 on a single site; 3,000 active registrations; estimated 60,000 illegal listings in 2018. https://criminaljustice.cityofnewyork.us/press-release/ll18-report-sheds-light-on-eliminated-illegal-rentals-in-nyc/ (accessed 10 November 2026) ↩︎
- New York City Department of Housing Preservation and Development, “2023 New York City Housing and Vacancy Survey: Selected Initial Findings”, 2024 (official): 3,705,000 housing units; 2,357,000 occupied and vacant available rental units; net rental vacancy rate 1.41%; survey January to mid-June 2023. https://www.nyc.gov/assets/hpd/downloads/pdfs/about/2023-nychvs-selected-initial-findings.pdf (accessed 10 November 2026) ↩︎
- El Debate, report on the Instituto Nacional de Estadística’s experimental measurement of tourist dwellings, 19 July 2025 (press report of official statistics, May data, published July 2025): Madrid 15,242 (0.99% of dwellings); Barcelona 9,579 (1.18%); Málaga 8,597 (3.28%); Marbella 7,934 (7.91%); Sevilla 7,137 (2.18%); Valencia 6,553 (1.58%); Yaiza 24.4%. https://www.eldebate.com/economia/20250719/pisos-turisticos-espana-menos-15-concentrados-madrid-barcelona-costa_318543.html (accessed 10 November 2026) ↩︎
- Jens Sethmann, “Kampf gegen Zweckentfremdung: Jetzt wird’s ernst!”, MieterMagazin 5/2016, Berliner Mieterverein, 4 May 2016 (press report of a tenants’ association): ban in force since 1 May 2014; transition period for 6,300 registered holiday flats ended 30 April 2016. https://www.berliner-mieterverein.de/magazin/online/mm0516/051614.htm (accessed 10 November 2026) ↩︎
- Johannes Hofele, “Ferienwohnungen: die aktuellen Regelungen in Berlin”, NWB Experten-Blog, 18 July 2018 (commentary): registration number required in advertisements from 1 August 2018 (§ 5(6) ZwVbG); 90-day limit for second homes; fines up to €250,000. https://www.nwb-experten-blog.de/?p=7755 (accessed 10 November 2026) ↩︎
- Scott Sage, AirDNA, blog post on New York City’s short-term rental crackdown, 22 August 2024, updated 25 September 2024 (company data): active short-term listings 21,900 in July 2023 and 3,700 in July 2024; 30-night-plus listings up 29%; total listings down 24%. https://www.airdna.co/blog/nycs-short-term-rental-crackdown (accessed 10 November 2026) ↩︎
- The Real Deal, report on Airbnb one year after Local Law 18, 3 September 2024 (press report citing an Airbnb blog post as reported by Bloomberg). https://therealdeal.com/new-york/2024/09/03/one-year-in-airbnb-pleads-with-nycto-ease-regulations (accessed 10 November 2026) ↩︎
- HR&A Advisors, “Impact Assessment of NYC Local Law 18 on Airbnb Short-Term Rentals”, project page listing Airbnb as client (company-commissioned study; no figures on the page). https://www.hraadvisors.com/portfolio/economic-impacts-of-airbnb/ (accessed 10 November 2026) ↩︎
- The Data Stream, “What did Banning Airbnbs in NYC Accomplish?”, 1 July 2026 (commentary; independent analysis, not peer-reviewed): event-study difference-in-differences on Zillow and StreetEasy asking rents; estimate about $52 a year, confidence interval −$71 to +$175. https://datastream.substack.com/p/what-did-banning-airbnbs-in-nyc-accomplish (accessed 10 November 2026) ↩︎
- Government of Spain, La Moncloa, Ministry of Social Rights, Consumer Affairs and Agenda 2030, “Consumo logra la retirada de 65.000 anuncios ilegales en Airbnb y denuncia otros 55.000 sin número de registro”, 30 July 2025 (official): about 65,000 listings removed; 54,728 further listings without the registration number required since 1 July 2025. https://www.lamoncloa.gob.es/serviciosdeprensa/notasprensa/derechos-sociales-consumo-agenda-2030/paginas/2025/300725-retirada-anuncios-airbnb.aspx (accessed 10 November 2026) ↩︎
- Catalan News, “Barcelona mayor welcomes Constitutional Court ruling on tourist apartment restrictions”, 14 March 2025 (press report): Constitutional Court rejects People’s Party appeal against the Catalan decree; 262 municipalities; Collboni: housing “equivalent to a decade of new construction”. https://www.catalannews.com/society-science/item/barcelona-mayor-welcomes-constitutional-court-ruling-on-tourist-apartment-restrictions (accessed 10 November 2026) ↩︎
- 3Cat (ACN), report on the tourist-flat sector’s compensation claim, 16 September 2024 (press report): 10,294 licensed tourist flats in the second quarter of 2024 by city data; Apartur claim of €1,000 million. https://www.3cat.cat/3catinfo/el-sector-de-pisos-turistics-de-barcelona-reclama-1000-milions-per-leliminacio-de-llicencies/noticia/3311339/ (accessed 10 November 2026) ↩︎
- Context
- Restrictions on short-term rentals in New York (Local Law 18, enforced from 5 September 2023), Barcelona (non-renewal of tourist-flat licences from November 2028), Spain (national registration number from 1 July 2025) and the European Union (Regulation 2024/1028, applying from 20 May 2026), set against the peer-reviewed causal literature on Airbnb, rents and house prices.
- Causes
- The reallocation of existing homes from long-term to short-term rental; the clustering of listings in central and tourist-heavy districts and in areas with low owner-occupancy; the growth of the platform after 2012; the regulatory responses of city, national and EU authorities.
- Actors
- Airbnb and other booking platforms; hosts and absentee owners; New York City's Office of Special Enforcement; Barcelona City Council and the Generalitat de Catalunya; Spain's Ministries of Housing and of Consumer Affairs and its Constitutional Court; the Berlin Senate; cities of Los Angeles County; the European Union; the economists and geographers whose studies are reviewed.
- Perspectives
- Distinguishes causal from correlational studies and reports effects in their own units; separates rents and prices from the population change that defines gentrification; labels company data, company-commissioned work and non-peer-reviewed analysis; records the strongest counter-argument; makes no judgement of governments, platforms or hosts.
- Relationships
- Opens the Hospes Domain. Links to Vitruvio (San Siro Was Sold Five Days Before Its Heritage Deadline, https://titani.ca/dispatch/san-siro-was-sold-five-days-before-its-heritage-deadline/) and Diaspora (Syrians Are Going Home, Just Not From Europe, https://titani.ca/dispatch/syrians-are-going-home-just-not-from-europe/).
