What are the odds of losing money on a condo?
Most published resale figures describe winners only, because the losing sales are removed before the average is taken. We keep them. Every rate on this page counts all 149,329 sales.
Where did resale prices actually go?
The price per square foot the middle seller got each quarter since 2010, split by area. The second view shows how many of that quarter's sales made money. The two move together, which is the point: price and the odds of winning are the same story told twice.
Does holding longer improve your odds?
Yes, but two things happen at once. The chance of losing money falls, and past fifteen years it almost disappears. The return per year of ownership falls too. A long hold buys you safety, not speed.
Share of exits that lost money
Median annualised return, all exits
| Held for | Exits | Profitable | Lost money | Median ann. | Median gain | Median loss |
|---|
Sales inside the first year are mostly new launch units resold before the building was finished. Turning a few months of gain into a yearly rate produces very large numbers, so read that band as a description of those deals rather than as a return anyone could repeat.
Did the year you bought matter more than the project?
Group every sale by the year its owner bought, and the cycle shows up plainly. People who bought at the 2011 to 2013 peak lost money most often. People who bought in the quiet years after did best. The year you enter is not something you control, which is exactly why it deserves attention.
Which districts returned the most?
Every measure on one map. Switch between the return per year for the middle seller, the share of sales that made money, the rent-to-price yield, and the price per square foot. Each answers a different question, so check more than one before you judge a district.
| District | Segment | Projects | Exits | Profitable | Median ann. | Median gain | Median yield | Median exit psf |
|---|
District 24, covering Lim Chu Kang and Tengah, has very few completed condo sales and is not drawn on the map artwork. It appears in the table wherever data exists.
Does a high rental yield cost you growth?
Yield is a year of rent divided by the price. So it goes up whenever the price falls relative to the rent. That makes a high yield a useful warning sign about price growth, not a free bonus on top of it. This section shows how the two have moved together.
Median gross yield by segment and tenure
Yield against realised capital return, by project
| Project | District | Gross yield | Avg psf | Avg rent psf | Exits | Profitable | Median ann. |
|---|
The highest rent-to-price yields among projects with at least 20 paired sales and recorded rental data. These are gross figures, before maintenance, property tax, agent fees, empty months and income tax.
When a sale went wrong, how bad was it?
The losing sales are the part of this data nobody else publishes. So they are worth looking at directly, rather than as whatever is left over after the winners.
How big were the losses?
Where every exit landed
Which projects ranked best?
Ranked on whichever measure you pick. Sorting by how often a project made money gives a very different table from sorting by how much. Reliability and size of gain are not the same thing, and most buyers care about both.
| Project | District | Exits | Profitable | Median ann. | Yield | Avg psf |
|---|
How to read this page without fooling yourself
Data updated —
A price index tells you what the market did. A paired sale tells you what one owner did. They bought one unit on one day. They sold that same unit later. They walked away with a real number. This page is built entirely from the second kind of fact. There are 149,329 of them here, across 683 projects, from 1995 to 2026.
Four terms this page uses
- Paired sale — one unit bought once and sold later. Both prices are real and recorded.
- Profitable share — how many sales went out above the price the owner paid.
- Return per year — the gain spread evenly across the years the owner held. A 40% gain over ten years is about 3.4% a year.
- Gross yield — a year of rent divided by the price, before any costs at all.
What changed in this edition
Every earlier version of this page showed winning sales only. That was a real limit, and it is now fixed. The data now holds 18,952 losing sales next to 130,377 winning ones. Wherever this page used to say "what sellers made", it now separates three things: what winners made, what losers lost, and how often each happened. A sale at exactly the purchase price counts as a loss, because the owner still paid stamp duty going in, and agent and legal fees coming out.
Three ways to be wrong with this data
First, do not read a typical figure as a forecast. Every buyer here bought under stamp duty rules, loan limits and interest rates that have since changed. Someone buying in 2013 paid no extra stamp duty on a first home. Someone buying in 2026 faces a different table entirely.
Second, do not treat the 87.3% as your personal odds. It is how often it happened in the past, over one particular stretch of years. It is also incomplete in a specific way. An owner sitting on a paper loss can simply refuse to sell, and many did. So the real loss rate is almost certainly higher than 12.7%. The worst cases never became sales at all.
Third, do not add a yearly return to a rental yield. They are not the same currency. The return here leaves out every cost of buying and selling. No stamp duty, no legal fees, no agent fees, no loan interest, no renovation. The yield leaves out upkeep, property tax, empty months and tax on the rent. Both are honest as published. Neither is what reaches your bank account.
How the districts are grouped
URA draws its three rings by planning area, not by postal district. So a few districts sit across a line. District 5 covers both Queenstown and Clementi. District 21 covers both Bukit Timah and Ulu Pandan. We give each district one ring, so a district reads the same here as on the new launch tracker. We put those two in the city fringe and the suburbs. Where the split would change a conclusion, use the district table above instead of the region figure.
Schools, and the other 1km
School distance is the other thing families shortlist on, and it touches everything above. A project within 1km of a school in high demand draws a different buyer. It also fetches a different resale price. That work lives on its own pages rather than being repeated here. The primary schools map plots every MOE school with its ballot demand and the 1km and 2km rings. Our study of school distance and launch pricing tests whether the premium is real. One caution. Those distances are straight-line guides. They are not the way MOE measures.
The short version
What to take away
- 87.3% of 149,329 paired sales made money. 12.7% did not.
- The middle seller made 3.2% a year over 8.7 years, a gain of $274,000.
- When a sale lost money, the middle loss was $115,000 after a 7.4-year hold.
- Holding past fifteen years cut the loss rate to 1.3%, but the yearly return fell to 3.0%.
- The year the owner bought mattered more than which project they picked.
- A high rental yield usually signals weak price growth, not a bonus on top of it.
- The real loss rate is higher than 12.7%, because owners sitting on losses often just do not sell.
So what should you do with this?
Two things. Work out how long you can realistically hold, then read the odds for that band rather than the island-wide figure. Then look up your own shortlist on the resale map. It plots all 683 projects with their own records, buyer profiles and full sale histories. It also carries a ranking of the strongest performers and our resale top picks. For what the other side of the market costs today, the new launch tracker carries developer pricing and the land costs behind it.
Update log
- 2026-08-20 — Rewritten in plain English for readers who are not property professionals. Section headings are now the questions readers ask, a glossary of four terms was added, and a short-version summary with next steps sits before the questions. Fixed a build bug that left unfilled template placeholders (%-style) inside the page's structured data, where search engines and AI assistants were reading a literal placeholder instead of the date and the dataset size. The generator that writes this page needs the same fix.
- 2026-08-14 — Moved here from /price-trends/, which is now the project map. Dataset rebuilt: 683 projects (was 199) and 149,329 matched exits (was 32,414), now including 18,952 unprofitable exits for the first time. Every "what sellers made" figure re-cut as winners, losers and a profitable share. New sections added on rental yield and the anatomy of a loss; the district view gained yield and profitable-share measures.
- 2026-07-30 — Previous edition: 32,414 profitable matched exits across 199 projects, exits Jan 2018 to Apr 2026.
Common questions
What share of Singapore condo resales made a profit?
Across 149,329 paired sales covering 683 projects, from 1995 to 2026, 87.3% sold above the price the owner paid and 12.7% did not. A sale at exactly the purchase price counts as a loss, because the owner still paid stamp duty, legal and agent fees.
What is the median return on a Singapore condo resale?
Counting winners and losers together, the middle seller made 3.2% a year, held for 8.7 years, and gained $274,000 before costs. Looking only at the sales that made money, the middle one returned 3.6% a year. Where money was lost, the middle loss was $115,000.
How much do Singapore condo sellers lose when they lose?
The middle losing sale lost $115,000 before costs, after holding for 7.4 years. Most losses were modest: 46% of them were under $100,000. But a few were very large. The biggest single loss on record here was $11.70 million.
Does holding a Singapore condo for longer improve the odds?
Sharply, but only at the far end. Between one and ten years, roughly one sale in seven lost money, and the worst stretch was the seven to ten year band. Past fifteen years the loss rate falls to 1.3%. The catch is that a long hold lowers the return per year. Owners holding fifteen years or more made 3.0% a year.
What is the average rental yield for a Singapore condo?
The middle project yields 3.6%, across 609 projects with rental data. Most sit between 3.1% and 4.1%. By area: 3.4% in the prime core, 3.8% on the city fringe and 3.6% in the suburbs. These are before maintenance, property tax, agent fees, empty months and income tax, so what an owner keeps is lower.
Which market segment performed best?
It depends what you mean by best. On return per year, the city fringe and prime districts led. On the odds of making money at all, the city fringe led. The prime core produced both the biggest typical gain and the biggest typical loss, simply because the prices are larger at both ends.
Why does this dataset now include losses?
Because a figure worked out from winning sales only tells you how big the wins were. It tells you nothing about how often they happened. Earlier versions of this dashboard carried winning sales only. It now carries every paired sale, so every share on this page is worked out on the full set.
Does PropertyInsider.sg sell property or represent developers?
No. We are an independent research publication. We do not market projects, take developer fees for coverage, or act as agents. Our editorial policy sets out how we work.
