Independent Singapore Property Research

Condo price trends, measured in real exits

Not an index — 32,414 matched buy-and-sell pairs across 199 Singapore condo projects, exits from January 2018 to April 2026. See where resale prices went, what profitable sellers actually made, and how entry year, holding period and district changed the outcome. Filter everything.

Loading

Where resale prices went

Median exit price per square foot by quarter, split by market segment. Each point is the median of real matched exits lodged in that quarter — the price sellers actually achieved, not an asking-price index.

32,414
Profitable exits, one chart

Holding period vs annualised return

Every dot is one unit bought and later sold at a gain. Left of three years is the sub-sale and seller's-stamp-duty zone; the dashed line is the CPF Ordinary Account rate — the return your downpayment gave up. Hover any dot.

Annualised return by holding period

Median annualised gain by years held, split by how the unit was bought. The 7–10 year band is the dead zone — dominated by units bought near the 2011–2013 peak.

The year you bought mattered most

Median annualised return by purchase year. Buying into the 2011–2013 run-up meant a decade near 2% a year; 2020–2021 entries exited fastest and best.

Returns by district

Median annualised return among profitable exits, ranked. The pattern is the opposite of the price table: the fringe and suburbs out-compounded prime over this window, because prime's higher entry prices left less room for percentage growth.

Districts with fewer than 100 filtered exits are shown muted — treat their medians as indicative only.

Project leaderboard

Every project with at least ten filtered exits, sortable by any column. Medians are computed from the exits matching your filters, so narrowing to a holding period or sold-year window re-ranks the table.

Project Dist Seg TOP Exits Med p.a. Med gross Med hold Med gain 2026 psf

Reading Singapore price trends through exits

Data updated —

Most price-trend charts describe the market. This one describes the people who left it. Every figure on this page is built from matched transaction pairs — the purchase caveat and the eventual sale caveat for the same physical unit — so a gain here is not an index movement or a valuation, it is money a real seller banked. Between January 2018 and April 2026, this dataset captures 32,414 such exits across 199 condominium projects, and the median one looks like this: bought, held about 7.9 years, and sold for roughly $280,000 more — a gross gain of about 25%, which annualises to about 3.6% a year before stamp duties, interest and agent fees.

What this dataset is — and is not

Be clear-eyed about one thing before drawing conclusions: this dataset records profitable exits only. Unprofitable sales are not included, so nothing on this page tells you the probability of making a gain — only the size and shape of the gains that were made. That selection effect matters most where you would expect losses to concentrate: prime districts bought at peak prices, and short holds forced by circumstance. Read every median here as "what profitable sellers made", never as "what every seller made". It is also why we lean on medians rather than averages throughout — a handful of multi-million-dollar exits would otherwise drag every figure upward.

The quiet story: the fringe out-ran prime

The quarterly psf chart at the top of this page holds the clearest trend of the window. Median exit psf in the Rest of Central Region — the city fringe — rose from about $1,500 in 2018 to about $2,050 by early 2026, a gain of roughly 37%. The Core Central Region managed about 23% over the same stretch, and the suburbs about 18%. The fringe did not just keep pace with prime; it closed most of the gap, compressing the RCR-to-CCR discount from around $230 psf to under $100. Whether that compression is a new equilibrium or a stretched one is the single most consequential question for anyone choosing between a fringe launch and a prime resale today.

Entry year beat almost everything else

Cut the exits by the year the unit was bought and the market's memory becomes visible. Units bought in the 2011–2013 run-up — the peak before the cooling-measure era — exited at median annualised returns of barely 2.1–2.3% a year, often after a decade of waiting. Units bought in 2020 and 2021, into the pandemic trough, exited at 5.9–6.2% a year. The same island, the same asset class, a three-times difference in compounding — determined mostly by the year of entry. This is also why the holding-period chart shows its counterintuitive dip: the 7–10 year band looks like the worst place to sell not because time hurts, but because in this exit window that band was full of 2011–2013 buyers. Holding period and entry cohort are two views of the same underlying truth: price cycles are long, and buying into a peak costs years, not months.

New sale or resale entry — the honest comparison

Across all exits, resale-entry units actually posted the higher median annualised return — about 4.3% a year against 3.5% for units bought at launch. That headline flatters resale unfairly: most launch-bought units exiting in this window were bought in the heavy-supply years around the last peak, while resale entries are spread more evenly across the cycle. Compare like with like — the same three-to-five-year holding band — and launch-bought units come out ahead. The practical reading is not "new beats resale" or the reverse; it is that the entry price relative to the cycle dominates the entry channel. A well-priced resale beats an overpriced launch, and vice versa.

Districts: percentage growth lives where quantums are small

The district ranking inverts the usual prestige table. Districts 5, 28, 21 and 20 — Clementi and the west coast, Seletar, the Clementi–Bukit Timah corridor, Bishan–Ang Mo Kio — all cleared 4% a year at the median, while Districts 9, 2 and 7 sat between roughly 1% and 2.3%. Prime property preserved larger absolute gains per exit, but as a percentage of capital deployed, the suburbs and fringe compounded faster over this particular window. For an investor thinking in returns on equity rather than trophy addresses, that distinction is the whole game — and it is exactly the kind of pattern that only matched exit data, rather than asking-price indices, can show.

How to use this page with the rest of our research

This dashboard is the historical half of a pair. The forward-looking half is our new launch and GLS pipeline tracker, where land costs foreshadow the launch prices of 2027 and 2028, and the new launch price tables, which show what today's developers are charging by bedroom type. The discipline we would suggest: use this page to understand what entry timing, holding period and location have historically done to returns, then judge any launch you are considering against that base rate — not against the show-flat brochure. If a project's pricing implies you need fringe-at-2021 style appreciation to break even after stamp duties, the exit data tells you how often that has actually happened.

All figures are compiled from URA caveat data and public project records, are indicative, and change as new caveats are lodged. This page is research, not advice — verify current figures before making any decision.

Frequently asked questions

What is a matched transaction pair?

A matched pair links the purchase caveat and the subsequent sale caveat for the same unit, so profit, gross return, annualised return and holding period are computed from two real transactions of one physical unit — not from index movements or averages.

Does this dataset include losses?

No. It records profitable exits only — 32,414 of them. It describes the size and shape of realised gains, not the probability of making one. Read the medians as "what profitable sellers made", never as "what every seller made".

What was the median realised gain?

About $280,000 — roughly a 25% gross gain — over a median holding period of about 7.9 years. That annualises to about 3.6% a year before stamp duties, interest and transaction costs.

Which region rose fastest since 2018?

The city fringe (RCR): median exit psf rose from about $1,500 in 2018 to about $2,050 by early 2026, roughly +37%, versus +23% for CCR and +18% for OCR in this dataset.

Did buying at launch beat buying resale?

Not unconditionally. In the same three-to-five-year holding band, launch-bought units led; across all exits, resale entries posted the higher median annualised return, mostly because many launch-bought exits were purchased near the 2011–2013 peak. Entry price relative to the cycle matters more than the entry channel.

What is the best holding period?

In this dataset, three-to-five-year holds produced the strongest median annualised returns outside of short-dated sub-sales, while 7–10-year holds were weakest — a band dominated by 2011–2013 buyers. Use the holding-period and entry-cohort charts together; they are two views of the same cycle.

Why use annualised rather than gross return?

Gross return ignores time: a 25% gain over four years compounds at about 5.7% a year, the same gain over fourteen years at about 1.6%. Annualised return is the only fair way to compare exits with different holding periods, so it is the primary measure on this page.

Which districts produced the highest annualised gains?

Districts 5, 28, 21 and 20 led with medians above 4% a year. Prime districts 9, 2 and 7 sat near the bottom at roughly 1–2.3% — higher entry prices left less room for percentage growth over this window.

What does the CPF line on the scatter mean?

The dashed line marks 2.5% a year — the CPF Ordinary Account interest rate. Any exit below it earned less per year than the buyer's CPF savings would have earned untouched, before even counting transaction costs. It is the minimum honest hurdle for calling a property gain a good outcome.

Where does this data come from?

Transaction caveats lodged with URA, matched into buy-and-sell pairs per unit, covering exits from January 2018 to April 2026 across 199 projects. Project attributes are compiled from URA and public project records. The refresh date is shown in the strip at the top of the page.

Does PropertyInsider.sg sell property or represent developers?

No. PropertyInsider.sg is an independent research publication. We do not market projects, take developer fees for coverage, or provide agent services. Our editorial policy sets out how we work.

Talk it through with an advisor

The research on this page tells you what the data says. If you want to work through what it means for your own situation — budget, ABSD position, timing an HDB sale, or comparing your buy-or-sell decision against other options — you can request a one-to-one consultation.

  • No obligation, and no pressure to transact — the first conversation is about your goals, not a product.
  • Personalised affordability and stamp-duty scenarios based on your actual numbers.
  • Launch and tender alerts for the specific projects you shortlist.

Disclosure: advisory consultations are provided by Jamus Lee (CEA Reg. No. R065771E, ERA Realty Network Pte Ltd, Licence No. L3002382K), the publisher of PropertyInsider.sg, via JamusProperty.com. This is a separate service from our editorial research and has no influence over what we publish — see our editorial policy. Submitting this form shares your details with the advisory practice; see our privacy policy.

Required — 8-digit Singapore mobile, starting with 8 or 9.

Get updates on Telegram

Launch alerts, tender results and price analysis the moment we publish them — now on Telegram instead of email. Free, no spam, leave anytime.

Join @PropertyInsidersSG →