CCR, RCR and OCR in 2026: the prime-district rebound lasted exactly one quarter
Prime condo prices rose 1.8% in the second quarter and everyone called the bottom. Then the July index came out on 28 August and the Central Region gave back 1.8% in a single month. Here is what URA and NUS actually published, why the three regions keep swapping places, and which of these numbers should change what you do.
The short answer
Singapore's private residential price index rose 0.5% in the second quarter of 2026, after 0.9% in the first. That is the number that ran in every headline. It is also the number that hides what happened underneath it.
Strip the index apart and you get a different picture. Non-landed homes, which is what almost everyone reading this owns or is buying, fell 0.1% in Q2. The whole of the headline gain came from landed property, which jumped 2.5% after falling 0.4% the quarter before. Six things worth knowing:
- The three regions swapped places in one quarter. In Q1 the suburbs led at +2.2% and prime lagged at +0.6%. In Q2 prime led at +1.8%, the city fringe fell 1.2%, and the suburbs went flat at −0.1%.
- The prime rebound did not survive July. NUS released its July index on 28 August: the Central Region fell 1.8% in that one month, giving back the entire quarterly gain.
- Small units are cooling fastest. Homes of 506 sq ft and under fell 1.2% month on month in July, after rising in June. That is the investor end of the market changing its mind.
- New sales and resale are now two different markets. Developers sold 731 units in July, up from 156 in June. Resale volume went the other way, 1,009 down to 984. Buyers are showing up for launches and walking past the second-hand shelf.
- The CCR's problem is vacancy, not demand. 8.3% of completed prime homes sat empty at end-June, against 5.6% in the suburbs. Yet prime rents still rose 1.2% in the quarter. Those two facts sit together for a reason I explain in section 5.
- 33,963 unsold units are queued behind all this, plus roughly 60,600 completions coming. Supply is the story of 2027, and it is already visible in the 2026 numbers.
If you are buying, the useful conclusion is that "which region" has become a much worse question than "which price band". If you are selling, the useful conclusion is in section 7.
1. The headline number hid the story
URA published its second-quarter statistics on 24 July. Overall private residential prices: up 0.5%. Add that to Q1's 0.9% and the first half of 2026 comes to about 1.4%. Modest, orderly, nothing to frighten anyone at MAS. On that basis several commentators wrote that the market was cooling gently and moved on.
Here is the split that did not travel as far:
Private residential price index, quarter-on-quarter · Source: URA quarterly real estate statistics, Q1 2026 (released 25 Apr 2026) and Q2 2026 (released 24 Jul 2026)
| Segment | Q4 2025 | Q1 2026 | Q2 2026 |
|---|---|---|---|
| All private residential | +0.6% | +0.9% | +0.5% |
| Landed | +3.4% | −0.4% | +2.5% |
| Non-landed | −0.2% | +1.3% | −0.1% |
Swipe to see the full table →
Landed property is a thin market. A few hundred transactions a quarter, no two houses alike, and an index that lurches around as a result. It went +3.4%, then −0.4%, then +2.5% in three consecutive quarters. Nobody thinks Singapore's landed housing stock genuinely gained 3.4% of value, lost 0.4%, then gained 2.5% again in nine months. That is a small sample doing what small samples do.
Non-landed is the market with volume in it, and non-landed prices went backwards in Q2. Slightly, by 0.1%, but backwards. When you read "Singapore private home prices rose 0.5%", what you are mostly reading is a handful of bungalow sales.
This matters because the 0.5% figure is the one that gets quoted back at me in negotiations. A seller shows me a headline and asks why my pricing advice is more cautious than the news. The answer is that the news is quoting an index that includes a segment they do not own.
2. The three regions swapped places
URA splits non-landed housing into three regions. Core Central (CCR) is the prime districts, roughly Orchard, Marina Bay, Sentosa Cove and the traditional prime stretch. Rest of Central (RCR) is the city fringe. Outside Central (OCR) is everywhere else, which in practice means the suburbs where most Singaporeans actually buy.
For most of the last two years, the story has been the same one: the suburbs lead, prime lags, the gap narrows slowly. In Q2 2026 that reversed completely.
Non-landed private home prices, quarter-on-quarter change by region · Source: URA quarterly real estate statistics, Q1 and Q2 2026
Swipe to see all three regions →
Read the two bars for each region side by side and the swing is startling. OCR went from +2.2% to −0.1%, a 2.3-point turn. RCR went from +0.8% to −1.2%. CCR tripled its rate of gain, from +0.6% to +1.8%.
Before anyone declares a prime-property renaissance, it is worth being honest about what a regional index measures. URA's index is quality-adjusted, so it is not simply an average of what sold. But the regional sub-indices are still built from the transactions that occurred, and in a quarter when the CCR's activity is concentrated in a small number of well-received launches, the CCR index reflects those prices. Q2 was a light quarter for launches overall: developers sold 2,141 units across the whole market, against 2,013 in Q1.
So the more careful reading of Q2 is not "prime is back". It is "prime transacted at higher prices in a quarter where prime had the market's attention, and the suburbs took a breather after a very strong Q1". Which is exactly what the next month's data suggests.
3. Then July happened
On 28 August, the NUS Institute of Real Estate and Urban Studies released the July flash estimate of its Singapore Residential Price Index. The SRPI is a different animal from URA's index: it tracks completed non-landed private homes monthly, using a fixed basket, and it excludes new sales. Because of that it is faster and noisier than the quarterly index, and it tells you about the resale market specifically.
July's reading:
NUS SRPI flash estimates, month-on-month · Source: NUS Institute of Real Estate and Urban Studies, released 28 Aug 2026
Swipe to see all four measures →
The Central Region fell 1.8% in a single month. In Q2 the CCR gained 1.8% across three months. One month of July took it all back on the resale side.
Two other lines are worth staring at. The Non-Central Region rose 0.5% after falling 0.4% in June, so the suburbs quietly recovered while prime dropped. And small units, meaning strata areas of 47 sqm or 506 sq ft and under, fell 1.2%. Shoebox units are the purest expression of investor sentiment in this market. When they fall while family-sized suburban stock rises, the money that is leaving is rental-yield money, not homebuyer money.
Resale volume moved the same direction. 984 non-landed resale transactions in July, against 1,009 in June, excluding executive condos and collective sales. A 2.4% dip is not a collapse. But it is the fourth data point in a row pointing the same way, and four is enough to stop calling it noise.
One honest caveat, because I would rather you hear it from me than find it later. A single month of SRPI is a flash estimate and gets revised. The Central Region sub-index in particular is built on a smaller basket, so it moves more. I would not rebuild a strategy on one month. I would absolutely stop quoting the Q2 CCR number as though it settled anything.
−0.1%
Non-landed prices, Q2 2026
−1.8%
Central Region, July alone
8.3%
CCR vacancy at end-June
33,963
Unsold units in the pipeline
4. New sales and resale are now two markets
If you only read the price indices you would conclude that demand is fading. Then you look at July's developer sales and the conclusion falls apart.
Monthly volumes, June vs July 2026 · New sales: URA developer sales survey (excludes ECs). Resales: URA Realis non-landed resale transactions, excluding ECs and collective sales, cited by NUS Ireus on 28 Aug 2026
Swipe to see all four bars →
Developers moved 731 units in July, up from 156 in June. That is a 368.6% jump and the third-best month of 2026 so far. June was a two-year low, so some of this is a base effect and a school-holiday lull unwinding. Most of it is simpler than that: two projects launched.
- Lentor Gardens Residences sold 270 of its 499 units, a 54% take-up on launch weekend.
- Dunearn House sold 212 of 380, or 56%.
Between them, 482 units, which is 63.6% of everything developers sold in Singapore that month. Two addresses carried the market.
Now the number that actually tells you something. At Lentor Gardens Residences, 222 of the 270 units sold, 82.2%, went for under $2.5 million. Market-wide, 422 of the 731 new sales, 58.1%, were under $2.5 million. Buyers turned up in force for a specific price band, and the projects that hit that band sold out half their stock in days.
That is the shape of this market. Not "the CCR is hot" or "the OCR is cooling". A quantum ceiling that most buyers will not cross, and fierce competition for anything well-located that sits under it.
The buyer mix backs this up. From URA caveats as at 17 August, Singaporeans took 87.5% of July's new sales, 635 units. Permanent residents took another 10.9%, 79 units. Which leaves foreigners at roughly 1.6% of the whole month.
In the CCR specifically, non-PR foreigners bought seven units in July. Seven. Across Dunearn House, Aurea, Grange 1866, River Green, Skywaters Residences and UpperHouse at Orchard Boulevard. The 60% ABSD rate for foreign buyers has been in force since April 2023 and this is what it looks like three years in: the prime market now runs almost entirely on local money. Anyone still pricing a CCR unit on the assumption that an overseas buyer will appear is pricing on 2019 conditions.
5. Empty prime homes, rising prime rents
Two figures from the Q2 release look like they contradict each other.
Vacancy rate of completed private residential units, end of Q2 2026 · Source: URA quarterly real estate statistics, 24 Jul 2026
Swipe to see the full scale →
The CCR carries 8.3% vacancy. The OCR carries 5.6%. Roughly one in twelve completed prime homes had nobody in it at the end of June, against one in eighteen in the suburbs.
And yet:
Non-landed rental index, quarter-on-quarter change, Q2 2026 · Source: URA quarterly real estate statistics, 24 Jul 2026
Swipe to see the full scale →
Prime rents rose 1.2% while suburban rents fell 0.3%. The region with the most empty homes had the strongest rental growth.
These are not in conflict once you separate the two pools. Vacancy measures completed units with no tenant. The CCR's vacant stock is disproportionately large, expensive units in newly completed prime projects, plus units held empty by owners who would rather wait than accept a lower rent. Those take months to let, and some are never seriously marketed. Meanwhile the CCR units that are tenanted are being renewed at higher rates, because the expatriate demand that remains is concentrated on a smaller, better pool of homes.
Two practical consequences.
For landlords in the CCR: the 1.2% rent gain is real but it is being earned by the units that let quickly. If yours has sat empty for two months, you are not in the group that number describes. Reprice or reposition.
For anyone modelling a prime purchase as a rental investment: use the vacancy figure, not just the rent figure. An 8.3% vacancy rate means budgeting for genuine void periods. Run it through the mortgage affordability calculator with two months of vacancy a year built in and see whether the numbers still work.
6. The supply queue behind 2027
Everything above is about the last three months. This section is about why I think the next eighteen matter more.
At the end of Q2 2026 there were 15,810 unsold units in projects with planning approval, and another 18,153 unsold units in projects awaiting approval, the second figure including 4,745 units from the H2 2026 Government Land Sales Confirmed List. Total: 33,963.
Compare that to the end of Q1, when the same two buckets held 17,032 and 13,265, for a total of 30,297.
Look at what moved. The approved bucket shrank by 1,222 units as projects launched and sold. The pending bucket grew by 4,888. That is not a wave of new supply appearing out of nowhere. It is the same supply taking one step closer to the market, and the queue behind it getting longer at the same time.
Separately, URA expects around 60,600 units to be completed in the coming years. Completions are what turn into vacancy if demand does not keep pace, and we have just established that CCR vacancy is already at 8.3%.
None of this is a crash signal. Singapore does not do disorderly housing markets, and the cooling measures introduced since 2023 exist precisely to prevent one. But it does mean the phrase "no supply" that gets used in showflats is not supported by the numbers, and it means the buyer's negotiating position in 2027 is likely to be better than it is today.
URA's Q3 2026 flash estimate is due on the first working day of October. I will update this page the same day.
7. What I would actually do with this
Data is only useful if it changes a decision. Here is how I would use these numbers, depending on where you are standing.
If you are buying a home to live in
Stop shopping by region. The regional indices moved in three different directions in two consecutive quarters, which tells you the region label is no longer carrying much information about price direction. What is carrying information is the $2.5 million band, where 58.1% of July's new sales landed and where competition is genuinely tight.
Below that band, expect to compete. Above it, expect to be one of few buyers in the room, and negotiate accordingly. That is true in every region right now.
Also compare resale seriously. Resale volume is falling and the SRPI is soft, which is the definition of a market where sellers are more flexible than they were six months ago. My new launch versus resale guide lays out the full trade, and the resale condo page covers the process.
If you are selling a condo
Price against your own region's July direction, not against the national 0.5%. If you are in the RCR, your regional index fell 1.2% in Q2. If you are selling a completed unit in the Central Region, the July SRPI fell 1.8%. Bringing a Q2 headline to a September negotiation will cost you weeks.
If you are selling a small unit, move sooner rather than later. That sub-index is the one falling fastest.
And check your seller's stamp duty position before you list. The holding period was extended to four years for units bought from July 2025, with rates from 4% to 16%, and I still meet sellers who have the old three-year table in their heads.
If you are an HDB upgrader
This is the group with the best hand right now, and most of them do not know it. The HDB resale market has held up far better than private resale. High-floor flats with long leases are still setting town records, which I track in the million-dollar HDB report. So you are potentially selling into strength and buying into softness at the same time.
The thing to get right is sequencing and cash flow, not timing the index. The HDB-to-condo upgrade guide covers the sequencing, and the ABSD calculator will tell you what the bridge costs if the two sides do not line up.
If you are looking at prime as an investment
Underwrite it on 8.3% vacancy and on a buyer pool that is 98.4% local. Both of those are facts from the last release, not predictions. If the numbers still work on those assumptions, the entry price is more reasonable now than it has been in years. If they only work on the assumption that foreign buyers return, you are betting on an ABSD change that the government has given no signal of making.
8. What I am watching before 1 October
Four things, in the order I expect them to land.
- The August SRPI flash, due late September. One down month in the Central Region is noise. Two consecutive down months is a trend, and it would mean the Q2 prime rebound was a single-quarter transaction-mix effect rather than a recovery.
- URA's Q3 flash estimate on the first working day of October. This is the one that settles whether non-landed prices have actually turned negative or whether Q2's −0.1% was a pause.
- August and September developer sales. July's 731 units came almost entirely from two launches. If August drops back toward June's 156 without new launches to carry it, that confirms demand is launch-driven rather than broad.
- The CCR supply signals. On 1 September, Singapore Land Group confirmed it had received written permission under URA's Strategic Development Initiative to partially redevelop Marina Square, adding a 49-storey tower with 204 large-format homes alongside a new hotel, serviced apartments and Grade A offices, for completion in 2031. Separately, Gilstead Court in Newton-Novena launched a fourth collective sale attempt at $198 million, with redevelopment potential for around 98 units. Both are years from completion. Both are the pipeline that the 2029–2031 CCR market will be priced against.
I update this page when each of those lands. If you want the update as it happens rather than when you next visit, message me and I will send it to you directly.
Common questions
CCR, RCR and OCR prices: your questions
They are URA's three market segments for private housing. Core Central Region (CCR) covers the prime districts including Orchard, Marina Bay, Bukit Timah's prime stretch and Sentosa Cove. Rest of Central Region (RCR) is the city fringe just outside it. Outside Central Region (OCR) is everywhere else, which is where most suburban condos sit. URA publishes a separate price index for each one every quarter.
Not overall, but non-landed prices did. The all-property index rose 0.9% in Q1 2026 and 0.5% in Q2, for about 1.4% across the first half. Within that, non-landed prices fell 0.1% in Q2 while landed rose 2.5%. The NUS SRPI then recorded a 0.4% month-on-month fall for non-landed homes in July 2026.
The Core Central Region, with non-landed prices up 1.8% quarter on quarter. The Rest of Central Region fell 1.2% and the Outside Central Region fell 0.1%. That is a complete reversal of Q1 2026, when the OCR led at +2.2% and the CCR managed only +0.6%.
One quarter said yes and the next month said no. The CCR gained 1.8% across Q2 2026, then the Central Region sub-index of the NUS SRPI fell 1.8% in July alone. Prime rents did rise 1.2% in Q2, but prime vacancy is 8.3%, the highest of the three regions. I would want to see two more months of data before calling it a recovery.
731 units excluding executive condos, up from 156 in June, a 368.6% month-on-month rise and the third-best month of 2026. Two launches produced 63.6% of it: Lentor Gardens Residences sold 270 of 499 units and Dunearn House sold 212 of 380. A further 27 EC units were sold in the month.
Barely, at the new-sale end. Singaporeans took 87.5% of July 2026's new private home sales and permanent residents another 10.9%, leaving non-PR foreigners at roughly 1.6%. In the CCR specifically, foreigners bought seven units in the entire month. The 60% ABSD rate for foreign buyers, in force since April 2023, is doing what it was designed to do.
33,963 unsold units at the end of Q2 2026: 15,810 in projects with planning approval and 18,153 in projects awaiting approval, the latter including 4,745 units from the H2 2026 GLS Confirmed List. That is up from 30,297 at the end of Q1. URA also expects around 60,600 units to be completed in the coming years.
For a home you will live in, the conditions are more favourable than they were a year ago: non-landed prices are flat to slightly down, resale volume is falling, and 33,963 unsold units are queued behind the market. The competitive pressure is concentrated under $2.5 million, where 58.1% of July's new sales landed. For an investment purchase, underwrite it on the vacancy rate for your region rather than on headline price growth. I am happy to run your specific numbers with you.
Sources and how to verify these figures
- URA — Release of 2nd Quarter 2026 real estate statistics, published 24 July 2026. Source for the Q2 price index changes (+0.5% overall, +2.5% landed, −0.1% non-landed, CCR +1.8%, RCR −1.2%, OCR −0.1%), the rental index changes, the vacancy rates, the 2,141 developer sales and 3,813 resales, and the unsold and completion pipeline figures.
- URA — Release of 1st Quarter 2026 real estate statistics, published 25 April 2026. Source for the Q1 comparatives (+0.9% overall, −0.4% landed, +1.3% non-landed, CCR +0.6%, RCR +0.8%, OCR +2.2%) and the Q1 unsold inventory of 17,032 plus 13,265 units.
- NUS Institute of Real Estate and Urban Studies — Singapore Residential Price Index (SRPI). July 2026 flash estimates released 28 August 2026: overall −0.4%, Central Region −1.8%, Non-Central Region +0.5%, small units −1.2%, with 984 non-landed resale transactions in July against 1,009 in June.
- URA — developers' sales of private residential units, the monthly survey behind July's 731 new sales and 27 EC sales. Project-level take-up and the $2.5 million price-band split are from agency analysis of that release and of URA caveats lodged as at 17 August 2026.
- The Marina Square Strategic Development Initiative details are from Singapore Land Group's announcement of 1 September 2026; the Gilstead Court collective sale is the fourth attempt, launched at $198 million.
- Related reading on this site: the million-dollar HDB report for the public housing side of the same market, and new launch versus resale for how to weigh the two channels.