Using proprietary Idealista quarterly data (2019-2024) for Barcelona, Madrid and Málaga, the paper argues that mid-term rental (MTR) growth is driven mainly by regulatory evasion in Barcelona and by market dynamics in Madrid and Málaga, with transnational demand acting as an enabler rather than a driver.
Thesis (C1) MTR expansion responds to an interplay of regulatory incentives, market dynamics and transnational demand: regulatory avoidance is the primary driver in Barcelona, market dynamics and asset valorisation in Madrid and Málaga; transnational demand is an enabler rather than a driver. partial
Valuable first multi-city picture of Spain's MTR boom. The Barcelona descriptive evidence — MTR listings surging at each rent-control milestone while LTR listings collapse — makes regulatory avoidance a compelling explanation, and all reported figures reconcile. But the regression layer meant to adjudicate among hypotheses is not fit for purpose (24 trending observations, no regulation term, reverse causality in the rent-gap variable), so the city-specific attributions for Madrid and Málaga and the 'definitive' rejection of profit motives in Barcelona are not established.
Claims (10)
12331
Issues (6)
222
Numeric checks (8)
611
Summary
Three hypotheses are posed: H1 regulatory blind spots (seasonal 1-11 month contracts are exempt from rent control and tenure rules), H2 market navigation (flexibility/liquidity in appreciating markets), H3 transnational demand (digital nomads, expats, students). Idealista data (24 quarters per city) provide listings by type, median asking rents, leads and the origin/language of searches. In Barcelona MTR listings reached 75.45% of residential listings by end-2024 (from 4.76% in 2019), jumping when rent controls were applied; residential listings fell 76.73%. Madrid's MTR ratio reached ~18%, Málaga's 11.57% (listings ×9). The MTR-LTR asking-rent premium ('rent gap') ended at 44.32% (Madrid), 16.71% (Barcelona) and 10.02% (Málaga). Simple and multiple OLS regressions on quarterly levels find: Barcelona — negative rent-gap coefficient (interpreted as regulatory evasion, H1); Málaga — rent gap, residential listings and price/m2 significant (H2); Madrid — price/m2 and listings significant (partial H2); foreign searches not significant in multivariate models (H3 as contextual enabler). Neighbourhood-level foreign-origin and foreign-language searches correlate at R2 > 0.91 and concentrate where MTRs are dense. Policy: extend regulation to seasonal contracts.
Research questions
What drives the expansion of MTRs across Barcelona, Madrid and Málaga: regulatory blind spots, market navigation, or transnational demand?
How are MTRs and international demand distributed within cities?
Stated contribution
First multi-city empirical analysis of mid-term rentals in Europe, comparing three explanatory mechanisms.
How the argument is built
Introduction with H1-H3 → literature on platform rentierism, regulatory blind spots, liquidity → policy context of the three cities → data/methods (Idealista, rent-gap proxy, regressions) → supply, price and rent-gap trends → demand (leads, foreign searches/language) → multiple regression → conclusions and limitations.
Argument map
Claims (boxes) coloured by how well the paper supports them; concepts as ellipses. Arrows point from premise to what it supports.
Thick black border = thesis; blue border = conclusion/recommendation. Click a node to jump to it. Layout engine: graphviz.
MTR expansion responds to an interplay of regulatory incentives, market dynamics and transnational demand: regulatory avoidance is the primary driver in Barcelona, market dynamics and asset valorisation in Madrid and Málaga; transnational demand is an enabler rather than a driver.
Assessment: The Barcelona timeline (MTR ratio jumping at each regulatory step) is strong descriptive evidence for H1. The regression evidence used to discriminate hypotheses is weak: OLS on 24 trending quarterly levels without treatment of autocorrelation or non-stationarity, four correlated predictors, no regulation variable at all, and interpretation of single significant coefficients as decisive.
In Barcelona MTRs rose from 4.76% to 75.45% of residential listings (peaking at 89%) with jumps coinciding with rent-control measures, while residential listings fell 76.73%.
Assessment: Figures are mutually consistent (4.76% x 3.689 / 0.233 = 75.46%; total −61.03%). The timing with regulatory steps is clear. Caveat: listings measure advertised flow, not stock — MTRs re-list more often, and under longer leases/rent control fewer LTR vacancies reach the platform — so the 75% ratio overstates MTRs' share of the actual rental stock.
A declining rent gap alongside expanding MTRs in Barcelona (and a significant negative rent-gap coefficient) 'definitively rules out profit maximisation' and shows regulatory evasion is the dominant driver.
“definitively rules out profit maximisation as the primary driver”
Assessment: The negative association is open to other explanations: MTR supply growth itself lowers MTR asking prices, and shrinking LTR listings raise LTR asking prices (reverse causality/equilibrium), so the gap falls because MTRs expand. Moreover, under rent control the landlord's alternative is the capped legal rent, not the LTR asking price, so the asking-price gap does not measure the incentive. Evading controls is itself profit maximisation under a constraint; the dichotomy is false. 'Definitively' is not warranted from a single coefficient with N = 24.
In Málaga MTR growth follows market logic (H2): rent gap, residential stock and price/m2 are significant (adj. R2 0.761).
Assessment: With trending series (MTR listings ×9, prices +61%) a high R2 and significant price coefficient are expected under a common time trend; without detrending or differencing the association may be spurious. The simple rent-gap regression for Málaga was null (R2 0.005), so the multivariate significance depends on controls.
Transnational demand is not a significant driver of MTR supply in any city (multivariate), though simple regressions in Madrid and Málaga and spatial coincidence are 'broadly consistent' with H3.
Assessment: The conclusion is appropriately nuanced, but the introduction states H3 'does not emerge as a statistically significant driver ... in any of the three cities', which is inconsistent with the significant simple regressions reported. Non-significance with N = 24 and correlated predictors is weak evidence of absence.
In Madrid, structural market pressure (price/m2, residential stock) rather than the rent gap or international demand drives MTR growth (partial H2), despite a persistent ~40%+ premium that 'makes conversion economically rational'.
Assessment: Same time-series concerns. There is also an internal tension: the descriptive section presents Madrid's large premium as a clear economic incentive, while the model finds the rent gap insignificant.
International demand is spatially concentrated in central/tourist neighbourhoods where MTR density is highest; foreign-origin and foreign-language searches are almost perfectly correlated across neighbourhoods (R2 > 0.91).
Assessment: Clearly documented for the top/bottom neighbourhoods. The 'reinforcing dynamic' and 'displacing local residents' are interpretations not tested.
Leads per listing rose dramatically (Barcelona residential 9 → 71.4), signalling unprecedented intensification of competition for housing.
Assessment: Per-listing leads rose ~7.9x, but residential listings fell by 76.7%, so total leads rose only ~1.85x; most of the per-listing increase is a denominator effect. It does indicate harder access to advertised units.
Regulation should be extended to seasonal contracts and designed to anticipate new forms of avoidance; in less regulated contexts broader affordability policies are needed.
Assessment: Reasonable policy implication of the Barcelona descriptive evidence.
Quantitative. Comparative descriptive time-series analysis of platform listings for three cities plus small-N OLS regressions (city-level quarterly levels; neighbourhood cross-sections).
Scope: Three Spanish cities, 2019-2024.
Methods
Descriptive trends (cumulative change, shares)
Rent-gap proxy: % difference in median asking €/m2 (MTR vs LTR)
Simple OLS of MTR listings on rent gap and on foreign searches (N = 23 per city)
Neighbourhood OLS of foreign-language on foreign-origin searches
Multiple OLS with four predictors (N = 24 per city)
Transparency notes
Rent-gap proxy limitations are candidly explained. Not reported: unit-root/autocorrelation diagnostics, whether variables are levels or changes (text says 'growth', models use levels), standard errors, and raw series. Two references are left as '[to be confirmed]'. Data are proprietary.
Limitations acknowledged by the authors
Idealista does not capture all listings; asking prices differ from contract rents.
Rent-gap proxy ignores vacancy, turnover and costs.
Small N (24) — 'results should be interpreted with caution'.
Regulatory effects only partially captured in the time window.
No data on tenant profiles.
Data sources
ID
Source
Type
Description
Coverage
Loc.
D1
Idealista proprietary dataset
scraped platform data
Quarterly listings by type (residential, seasonal; STR excluded), median asking rent (€/month, €/m2), leads, share of searches from abroad and in foreign languages; city and Idealista zones.
Madrid, Barcelona, Málaga; 31 Mar 2019 - 31 Dec 2024; 24 quarters per city.
pp. 4-5
D2
Legal and policy documents / literature
legal texts
RDL 7/2019, Law 11/2020 (Catalonia), Law 12/2023, Decree 3/2023, Start-up Law 28/2022, city strategies.
pp. 3-4
Evidence items (9)
ID
Kind
Description
Loc.
Source
Used by
E1
figure
Fig. 1 and text: MTR/residential listings ratio — Barcelona 4.76% (2019) → 8.59% (2020) → ~24% → 46% (2023-24) → 74.26% / 89.02% after controls → 75.45% (end-2024); MTR +268.88%, LTR −76.73%, total −61.02%. Madrid MTR +73.89%, LTR −58.58%, total −53.15%, MTR 17.97%. Málaga MTR 51 → 458 (+798%), LTR −22.93%, 11.57%.
Table 1: neighbourhoods with highest MTR concentration (e.g. Sant Pere–Santa Caterina 124.75% and El Raval 118.58% of residential listings; Chueca-Justicia 43.33%).
Asking €/m2 2019→2024: Madrid MTR 32.37 vs LTR 22.43 (2024); Barcelona 23.01/17.18 → 28.85/24.72; Málaga 12.59/10.47 → 18.55/16.86. Rent gap 2024: 44.32% / 16.71% / 10.02%; peaks 89.27% (Madrid Dec 2021), 243.74% (Málaga Mar 2022), 53.17% (Barcelona Dec 2021); Barcelona annual means 40.44, 40.32, 29.13, 22.47 (2021-24).
Foreign-origin searches (Dec 2024): Madrid 12.22%, Barcelona 20.71%, Málaga 20.25%; neighbourhood extremes (Sol 34.02% vs Pavones 3.39%); foreign-language searches Barcelona 41.38%, Málaga 31.76%, Madrid 19.88%.
Simple OLS MTR listings on foreign searches: Madrid p 0.002 R2 0.37; Málaga p 0.007 R2 0.29; Barcelona p 0.59 R2 0.01. Neighbourhood OLS foreign-language on foreign-origin: R2 0.948 / 0.933 / 0.919.
Regulatory chronology: 2019 lease extension; Catalan Law 11/2020 (annulled April 2022); Law 12/2023 and its application in Catalonia; Barcelona PEUAT and tourist-licence limits; seasonal contracts untouched.
Contracts for 1-11 months not for habitual residence, exempt from rent controls and tenure protections in Spain.
legal
K2
Regulatory blind spots
p. 3
Legal tools used by intermediaries to circumvent laws governing rent relations.
borrowed
Dimitrakou (2025)
K3
Rent gap (proxy)
pp. 4-5
% difference between median asking €/m2 of MTR and residential listings, interpreted as expected profitability.
operational
Authors explicitly acknowledge it is not a true rent gap. Under rent control, the relevant alternative for a landlord is the legally capped rent, not the LTR asking rent.
K4
Asset liquidity / market navigation
p. 3
Landlords' use of flexible tenure to retain control over timing of sale/re-letting in appreciating markets.
adapted
Gil (2024); Kemp (2020)
Issues (6)
I1majorcausal inferenceTime-series regressions on trending levels with N = 24pp. 4-5, 9, 13
OLS on quarterly levels of strongly trending series (listings, prices) without detrending, differencing, autocorrelation-robust errors or unit-root checks risks spurious significance; four correlated predictors (rent gap is a function of price/m2) with 24 observations; the text speaks of 'growth' while models use levels. The authors flag small N but not these issues.
Impact on conclusions: Regression results cannot discriminate among H1-H3 with confidence.
I2majorcausal inferenceH1 never directly testedpp. 9, 13
No regulation variable (dummies for Law 11/2020 in force, Law 12/2023, Catalan application in 2024) enters any model; H1 support is inferred from a negative rent-gap coefficient and descriptive timing. An interrupted time-series or difference-in-differences vs Madrid/Málaga would be the natural test.
Impact on conclusions: Barcelona's regulatory story rests on descriptive timing (which is actually quite persuasive) rather than the regression.
I3moderatelogicOver-interpretation of the negative rent-gap coefficientpp. 9, 13
Reverse causality (MTR supply growth lowers MTR prices; LTR listing scarcity raises LTR asking prices) and mis-measured incentive (the capped legal rent, not asking LTR rent, is the landlord's alternative under control) can produce the negative sign. Evasion of controls is a form of profit maximisation, so 'rules out profit maximisation' is a false dichotomy.
Impact on conclusions: The H1 conclusion may be right, but not for the reason given.
I4moderatemeasurementListings are flows, not stockspp. 5-6, 9
MTRs turn over (and are advertised) more often than LTRs; longer LTR leases and rent control reduce LTR vacancies reaching the platform; landlords may fill controlled units off-platform. Listing shares and leads-per-listing therefore exaggerate shifts in the underlying stock and in competition.
Impact on conclusions: Direction credible; magnitudes overstated.
Barcelona's rent gap fell 36.5 points from its Dec-2021 peak, or 17.2 points since 2019 — described as 'a compression of 17 percentage points in three years'. Madrid's 17.97% is the MTR/residential ratio, not 'of total listings' (≈15.2% of total). Simple-regression β (−0.94) and multiple-regression coefficient (−140.8) appear to be on different scales without explanation.
I6minorinternal inconsistencyIntroduction vs results on H3; unfinished referencespp. 2, 10, 14-15
The introduction says transnational demand is not significant 'in any of the three cities', but simple regressions are significant in Madrid and Málaga (only multivariate models are null). Two references are printed with '[to be confirmed]' placeholders.
Impact on conclusions: Editorial; the conclusion itself is nuanced.
Scope internal = numbers printed in the document; cross paper = other corpus documents; external / replication = outside sources or re-analysis of original data (web mode only).
ID
Status
Scope
What was checked
Computation
Result
Loc.
Claims
V1
pass
internal
Barcelona MTR ratio and total listings change.
0.0476 x 3.6888 / 0.2327; (0.0476 x 3.6888 + 0.2327)/1.0476 − 1
Evaluation frame: Comparative quantitative case study with explicit hypotheses. Standard: validity of proxies, appropriateness of time-series inference, whether tests actually discriminate among hypotheses, and proportionality of causal language to a small-N observational design.
Dimension
Score
Rationale
conceptual clarity
4
Clear hypotheses and candid discussion of the rent-gap proxy.
logical coherence
3
Descriptive story coherent; 'evasion vs profit' dichotomy and Madrid premium/insignificance tension.
Interrupted time-series / diff-in-diff with Madrid and Málaga as comparisons for Catalan regulatory events.
Model first differences or include trends; HAC standard errors.
Use stock estimates or listing durations to correct flow bias.
Compare MTR rents with the legal capped rent rather than LTR asking rent.
Links to the rest of the corpus
X1contradictionmoderateDid pandemic STRs return to the residential market?
Gil (2023) states that during the pandemic 'STRs did not return to the residential market (Gil et al., 2023)' and uses this to argue for housing-policy (not tourism-policy) responses. Gil, Vidal & Yrigoy (2026) cite the same source for STR units being 'temporarily redirected to residential or mid-term segments'. Sequera et al. (2022) document central Madrid long-term listings rising from 1,300 to 4,600 and back to 1,900, and Gil (2021) reported hosts 'directly ending the business and renting in the residential market'. The evidence supports a temporary switch, largely under seasonal contracts; the 2023 paper's unqualified statement understates it.
Implication: The policy argument in Gil (2023) should be qualified: switching occurred but was temporary and often outside long-term residential contracts.
X12corroborationpositiveLater data temper earlier demand-side claim
Sequera et al. (2022) asserted that STR continuity rested on a demand shift to digital nomads, students and expats. Gil, Vidal & Yrigoy (2026), with Idealista search data, find foreign demand is not a significant driver in multivariate models (an 'enabler'), while confirming the seasonal-contract mechanism.
Implication: Shows the programme updating claims with better data.
Gil (2023): Liquidity/flexibility as a motive for landlords to choose less regulated rental forms. fragile
Provenance
Analyst: Claude (Anthropic), LLM-generated audit · 2026-11-08 · access: document only · text: text/gil_vidal_2026_midterm_rental.txt · schema v1.1
No external sources were consulted (offline run). Figures are assumed to be correctly reported; claims[].external, citation checks and external/replication checks were not performed.
Two-column layout; some table text interleaved in extraction.