Sixty-one thousand units completing, 13,484 flats reaching minimum occupation period and a S$2 million floor only 1,016 sales cleared
By: Robert Sterling – SeaPRwire – Singapore’s private housing market rose 0.5 percent in the second quarter of 2026. That is the weakest quarterly gain in seven quarters, and it is the number that gets reported. It is also the least useful number in the dataset, because almost nobody buying or selling a home in Singapore this year is making a decision about the national index. They are making a decision about whether they can get out.
The exit is where the market has actually changed. Transactions tell the story better than prices do. In the second quarter, 5,420 private homes changed hands. Developers sold 2,141 new units, resales accounted for 2,634, and the remainder was a thin layer of subsales. Against that, the completion pipeline holds 61,000 private homes. Divide 61,000 by an annualised new-sale pace of 8,564 and you get 7.1 years. That arithmetic is crude — the pipeline stretches across several years and resale volume absorbs some of it — but the direction is not crude at all. Supply is arriving faster than the market is clearing, and it is arriving into a buyer pool that has been narrowed by stamp duties, loan limits and a rental market that has stopped helping.
So the useful question is not “will prices fall”. It is “when I need to sell, who is legally able to buy this from me, and how many identical units will be on the market that month”. Answer that and the price question answers itself.
Diagnose which squeeze you are in before you look at a listing
Four buyer types are active in this market and each one faces a different binding constraint. Getting the diagnosis wrong is the most expensive mistake available, because the wrong property in the right market still loses money.
| Profile | Binding constraint | What actually kills the deal |
| First-time resale buyer | Loan-to-value and total debt servicing ratio | A valuation gap at the S$2 million mark that raises the cash down payment |
| HDB upgrader | Minimum occupation period and the 15-month wait before buying private | Timing mismatch between selling the flat and the completion date |
| Investor seeking yield | Additional buyer’s stamp duty and rental index flat at 0.7% | Gross yield of 3.1% against a carry cost that assumes rent rises |
| Family office or entity buyer | 60% ABSD for foreigners and the housing trust route | Six years of carry before a resale that the next buyer can also finance |
The third row deserves a number. The private rental index rose 0.7 percent quarter on quarter in Q2 2026. The three-month compounded Singapore Overnight Rate Average stood at 1.05 percent. A gross rental yield of 3.1 percent against a 1.05 percent funding cost looks like a 2-point spread until you subtract property tax, maintenance, vacancy and the fact that the loan is amortising. The spread that survives that subtraction is thin, and it is entirely dependent on the rent line continuing to rise at a rate it did not rise at last quarter.
The pipeline is the number that sets your exit date
Sixty-one thousand units. That is the private residential completion pipeline, and it is the single largest determinant of what happens to a resale price in 2028.
Developers sold 2,141 new units in Q2, up from 2,013 in Q1. The take-up rate on launched units was 77.5 percent, which is healthy. The General Land Sales programme will release 9,320 units for the full year 2026, which is the government signalling that it intends to keep the tap open.
Now the concentration problem. Of those 2,141 new sales, 67.5 percent went to the top three projects. That is 1,445 units. The remaining 696 sales were spread across every other project on the island.
Read that again from the position of an owner. If you bought in one of the top three projects, you own something the market currently wants and you have pricing power. If you bought in any other project, your exit depends on a buyer choosing your building over 60,000 other units reaching completion in the same window, most of them newer, many of them sold by a developer with a showroom, a deferred payment scheme and a marketing budget larger than your entire transaction.
This is the mechanism by which a flat market produces very uneven outcomes. The index rises 0.5 percent and 67.5 percent of the volume goes to three developments. The average is real, and it describes almost nobody.
The HDB wave is supply and demand at the same time
Thirteen thousand four hundred and eighty-four HDB flats reach their five-year minimum occupation period in 2026. That number is usually discussed as a supply shock for the resale market, and it is — 6,268 flats resold in Q2 against a wave of newly eligible units is a lot of competition for a buyer’s attention.
But it is also the upgrade pipeline, and that is the more useful reading.
Every household whose flat clears the MOP and sells is a household with cash and a reason to buy private property, subject to the 15-month wait that applies to former HDB owners buying private housing. Those sellers are the natural buyers for the outer central region stock that developers are completing now. The resale price index slipped 0.3 percent quarter on quarter to 202.7, which means the upgraders are selling into a slightly softer market and therefore arriving at the private market with slightly less equity than they expected a year ago.
Then add the Build-To-Order tap. HDB launched 7,960 BTO units in October 2026. Every one of those is a household that will not be buying resale for at least eight years. The policy is deliberately suppressing a slice of resale demand in order to keep the resale price index flat, and it is working — a 0.3 percent quarterly decline after years of increases is a policy achievement, not a market failure.
For an upgrader the implication is specific: your flat will sell, but not at the number you saw in a headline eighteen months ago, and the gap between what you get and what you need for the private purchase has widened. Plan the financing on the actual recent transacted price in your block, not on the index.
The S$2 million line, and the 144 units that cleared it
Here is the most actionable number in the Singapore market right now.
In Q2 2026, 1,016 new units sold below S$2 million. That is 47.5 percent of the 2,141 total, and it is the segment where the loan limits, the stamp duty tiers and the buyer’s actual cash position all still fit together. Above S$2 million, the financing structure changes, the buyer pool narrows, and the exit takes longer.
Now the tighter number: 144 units sold below S$2,000 per square foot. Out of 5,420 total transactions across new, resale and sub-sale, 2.7 percent cleared under that threshold.
| Segment, Q2 2026 | Volume | Share of transactions |
| All private transactions | 5,420 | 100% |
| New sales below S$2 million | 1,016 | 18.7% |
| New sales below S$2,000 psf | 144 | 2.7% |
| Resale transactions | 2,634 | 48.6% |
| Top three projects’ share of new sales | 1,445 | 26.7% |
One hundred and forty-four units at under S$2,000 per square foot is not a market segment. It is a rounding error with a queue attached, and the queue is the reason the price is what it is. If your strategy depends on finding value below S$2,000 per square foot in the new launch market, your strategy depends on being faster than every other buyer with the same idea and the same spreadsheet.
The realistic version of that strategy is the resale market, where 2,634 transactions happened in a quarter and where the seller is a household rather than a pricing committee. That is where a 3 to 8 percent negotiation exists, and it is where an owner with a deadline — a job move, an estate, a divorce — will meet a price the index does not record.
Landed property is running a separate market
Everything above describes the non-landed segment. Landed housing is doing something else entirely, and treating the two as one market is how buyers misjudge both.
In the first half of 2026, 869 landed transactions were worth S$5.4 billion — an average of S$6.2 million per deal. Landed prices in the core central region rose 20.9 percent year on year. In the rest of central region they rose 6.4 percent. Outside central region, 9.1 percent.
At the top, ten good class bungalow transactions totalled S$304 million, an average of S$30.4 million, and a record land rate of S$4,550 per square foot was set. The prestige landed segment rose 19.3 percent year on year.
That is not a market responding to interest rates. It is a market responding to the number of good class bungalows that exist, which is fixed at roughly 2,500 island-wide and does not increase when demand does. A 20.9 percent annual gain in a supply-constrained segment during a quarter when the overall index rose 0.5 percent is the clearest evidence available that Singapore now contains at least two property markets with almost nothing in common except the same planning authority.
The trap for a buyer is assuming the landed market’s momentum transfers. It does not transfer downward. A S$6 million semi-detached house in the outside central region is competing with new launch condominiums at the same price and losing on facilities, on financing, and on the fact that the next buyer needs a larger down payment.
District-level divergence is where the mistakes happen
The regional numbers for Q2 2026 describe three markets moving in three directions. Core central region non-landed prices rose 1.8 percent. Rest of central region fell 1.8 percent. Outside central region slipped 0.1 percent.
Then the reconciliation problem. The Urban Redevelopment Authority’s flash estimates for the same quarter put core central region at 2.0 percent and landed at 2.6 percent, against final figures of 1.8 percent and 2.5 percent. Four districts, four disagreements between the preliminary and final series. That is normal statistical revision, and it is also a warning: anyone quoting a quarterly district number to justify a purchase is quoting a figure that will be revised.
What survives the revision is the ranking. Core central region outperformed, rest of central region underperformed, and the gap between them was 3.6 percentage points in a single quarter. That gap is the whole argument for buying location over buying timing, and it is also the reason a rest-of-central-region owner should not expect the index to rescue them.
Three exits, worked through
Abstractions are cheap. Here are three positions that exist in this market right now, and what the numbers say about each.
Position one: a 2022 launch buyer in a mid-tier project, 1,100 square feet, bought at S$2,100 per square foot. The unit cost S$2.31 million. Today, 144 units island-wide cleared below S$2,000 per square foot in a quarter, which tells you the comparable set is thin. The owner needs to sell in 2027, when a meaningful share of the 61,000-unit pipeline completes within two kilometres. The realistic exit is a resale at or slightly below entry, minus the seller’s stamp duty if the holding period is under three years, minus commission. The mistake to avoid is listing at the 2026 index level and waiting. In a completion window, the first 90 days on market are the best 90 days.
Position two: an HDB upgrader in Ang Mo Kio, five-room flat, MOP cleared, buying a rest-of-central-region condominium at S$1.9 million. The resale index is 202.7 and fell 0.3 percent last quarter, so the flat sells at roughly the last transacted price in the block, not the peak. The 15-month wait means the private purchase either happens before the sale with stamp duty paid up front and reclaimed later, or 15 months after, with rent in between. At S$4,000 a month, 15 months of rent is S$60,000, which is more than most upgraders budget for and is the single most common cause of the plan failing at the completion date.
Position three: a yield buyer holding a one-bedroom in the outside central region, purchased 2021, gross yield 3.1 percent. The rent line rose 0.7 percent last quarter. The funding cost is 1.05 percent on SORA, but the loan is amortising, so the cash yield after principal is closer to 1.5 percent. Property tax, maintenance and one month of vacancy a year take another 0.8 points. The position is roughly cash-neutral and its return depends entirely on capital appreciation in a segment where the index fell 0.1 percent in the quarter and 61,000 units are completing.
| Position | Entry | The number that decides it | Action |
| Launch buyer, 2022 vintage | S$2.31 m at S$2,100 psf | 144 units cleared below S$2,000 psf in a quarter | List in the first 90 days of the window; do not wait for the index |
| HDB upgrader | Flat at index 202.7, private at S$1.9 m | 15-month wait at S$4,000 rent = S$60,000 | Sequence the sale first; budget the rent as a transaction cost |
| Yield holder, OCR one-bedroom | 2021 entry, 3.1% gross | Cash yield after principal and costs near 1.5% | Model flat rent for 24 months; sell into strength, not into need |
The common thread is that in all three cases the decisive number is not a price. It is a duration — 90 days of listing, 15 months of waiting, 24 months of flat rent. Singapore’s market has become a market where time costs more than price, and the buyers who do well are the ones who priced the time.
The data centre constraint is a property trade in disguise
One more segment deserves attention because it connects to something most residential buyers never look at.
Singapore approved 200 megawatts of new data centre capacity on a provisional basis. Occupancy in existing facilities is 95 percent. Colocation vacancy is 4.9 percent. Both numbers describe a market that is effectively full.
The consequence is not a data centre investment opportunity for a private buyer — that market is institutional, and the entry ticket is measured in tens of millions. The consequence is geographic. Compute demand that cannot be served in Singapore moves to Johor, and the industrial and residential property around the Causeway follows the employment. Anyone buying in the north of the island is buying a derivative on a power allocation decision made in a boardroom they will never enter.
That is worth knowing before you pay a premium for a Woodlands or Sembawang unit on the theory that the data centre boom will lift the whole north. It will lift the industrial belt and the rental market around it. Whether it lifts a 99-year leasehold condominium three MRT stops away is a different question, and the honest answer is that nobody has the data yet.
A six-month checklist, in the order the decisions arrive
This is the part that converts all of the above into something you can do on a Monday morning.
- Fix your exit date before your entry price. Write down the month you expect to sell, not the year. If it is within 36 months, you are buying into the completion window of 61,000 units and you need a discount, not a view.
- Price the financing at the stressed rate, not the current one. SORA at 1.05 percent is the input to today’s instalment. Your loan will be stress-tested at a higher floor, and the number that matters is whether you can carry it for 24 months with zero rent.
- Check the top-three concentration in your target project’s district. If three developments account for most of the recent sales there, your unit is competing with a marketing machine. Buy the machine’s output at resale, not the imitator at launch.
- Count the units completing within 500 metres in the next 24 months. This is public information and almost nobody uses it. It is the single best predictor of your negotiation position when you sell.
- For HDB upgraders, sequence the sale before the purchase and budget for the 15-month wait. The gap between an HDB sale and a private purchase is where most upgraders lose more money than they save on the flat.
- For yield buyers, model the rent at a 5 percent vacancy factor and a flat index. The rental index rose 0.7 percent last quarter. A model that assumes 3 percent annual rent growth is a model of a market that no longer exists.
Price the exit before you price the unit
The Singapore market in 2026 rewards a specific discipline: knowing who your buyer is, what they are allowed to borrow, and how many alternatives they will have when you need them to choose you.
Every number in this article points the same way. Prices are flat, supply is arriving, transaction volume is concentrated in a handful of projects, and the segments that are still moving — landed, core central region, data-centre-adjacent industrial — are moving for supply reasons that will not repeat downward. A buyer who understands that can still do very well in this market. A buyer who reads a 0.5 percent quarterly index and assumes it describes their building cannot.
Start with the completion map, not the showroom.
Sources
- Urban Redevelopment Authority, Q2 2026 private residential price index and developer sales data, via EdgeProp and AsianPrime
- Housing and Development Board, Q2 2026 resale price index, resale transaction volume, MOP and Build-To-Order figures, via PropertyNet
- PropNex research via Singapore Business Review, first half 2026 landed and good class bungalow transaction data
- Fitch and CBRE estimates via Business Today, Singapore data centre capacity and occupancy, 2026
- Knight Frank, Singapore data centre colocation vacancy, first half 2026
- Global Property Guide, Singapore gross rental yield series, 2026
- Track B real estate dataset, Singapore cell, knowledge base compiled October 2026