Most national house-price indices mix two different things into a single line: how much actual buyer demand has shifted, and how the mix of properties getting sold has rotated month to month. Our hedonic factor model separates those out. The intercept (which we call Baseline Market) is the quality-adjusted, location-controlled price level — the apples-to-apples return after stripping compositional shifts. The chart below plots that Baseline Market for London, New York, Paris and Singapore, each rebased to 0% in January 2020 so the four are directly comparable.
| City | Window | Cumulative | Peak | Trough |
|---|---|---|---|---|
| Singapore | Jan 2020 → Sep 2025 | +54.7% | +57.4% | −7.9% |
| New York | Jan 2020 → Jan 2026 | +45.5% | +51.5% | 0.0% |
| London | Jan 2020 → Oct 2025 | +22.5% | +23.8% | −2.9% |
| Paris | Jan 2020 → Apr 2025 | −1.4% | +10.0% | −5.6% |
The Singapore HDB resale market is the standout. It actually drew down further than London or Paris during 2020 (trough of −7.9% by late 2020), then began the most sustained rally of any of the four cities. The Baseline Market crossed zero in mid-2021, +20% by mid-2022, +40% by late 2023, and is now sitting near its peak of +57%. Crucially, this is the quality-adjusted move — it controls for which flat sizes and which towns transacted each month, so it cannot be explained away by "more 5-room flats in mature estates sold this quarter".
New York's Baseline rose almost monotonically from 2020 through to a peak of +51.5% in 2023, then gave back about 6 percentage points over the rate-rise correction, before resuming its climb back toward +45%. The drawdown was shallow by US-cycle standards.
London's quality-adjusted price did its full COVID run-up by 2022 (peak +23.8%) and has essentially traded sideways since, finishing the period at +22.5%. That sideways move masks meaningful factor-level activity below the surface — Floor Area down, energy_score down, the house-vs-flat spread compressing — but at the index level, post-2022 London is a flat market.
Paris is the only one of the four with a negative cumulative return over the period. The Baseline rallied to +10.0% in 2022, then fell to a trough of −5.6% in 2024 and has hovered around −1% since. The Paris finding matters because the headline Notaires/INSEE index suggests far more volatility — once we hold composition fixed, much of that volatility turns out to be compositional rotation between arrondissements and unit sizes, not pure repricing of comparable units.
Index-level comparisons across cities are noisy: every country measures real estate slightly differently, and a "boom" in the headlines often turns out to be a shift in the kind of property selling, not a re-rating of the underlying stock. Holding hedonic characteristics fixed and looking at the intercept gives a cleaner cross-city signal. By that measure, the last five years did not deliver a synchronised global property bull market — Singapore and New York doubled the pace of London, and Paris went nowhere.
Interactive factor charts for every market, updated monthly.
Open the global comparison →Methodology: rolling 3-month cross-sectional OLS of log(price/sqm) on hedonic property characteristics (floor area, rooms, energy rating, construction era, type) with postcode/borough/arrondissement dummies as controls. See About for the full model spec.