92.9% of 2 bedroom condo owners in Singapore sold for more than they paid, a median gain of $230,000 over 8.3 years

92.9% of the 24,232 two-bedroom condominiums bought and later resold between January 2020 and August 2026 sold for more than the owner paid. 22,500 owners made money and 1,661 lost it. The median gain was $230,000, 21.7% on a median purchase price of $1,007,000, after a median hold of 8.3 years, which works out to about 2.8% a year in price alone. The median loss, among the 6.9% who lost, was $70,600.

These are price outcomes only, the price paid against the price received, before rent collected and before stamp duties, interest, commission and maintenance. Rent is covered in the rental yield notes; the full cost picture for one unit is a separate piece of work.

Where the losses sit is the useful part. In the CCR, 18.4% of two-bedroom resales lost money and the median loss was $135,500; in the OCR, 4.6% and $40,000. Buyers at the 2010 to 2014 peak lost money 12.4% of the time; buyers from 2019 to 2022 have lost 1.8% of the time and gained a median 20.7% in four years. The shoebox series found the same shape one size down, with worse numbers. 88.8% of shoebox owners sold above what they paid, for a median gain of $92,000, and 44.2% of CCR shoebox resales lost money.

Key takeaways

• 92.9% of two-bedrooms resold since 2020 sold above their purchase price; the median gain was $230,000, 21.7%, over 8.3 years. What actually happened
• The median gain works out to 2.8% a year in price alone, before rent and all costs; a unit bought as a resale returned 3.3% a year, one bought at new launch 2.6%. What it works out to per year
• Where you bought mattered most. 18.4% of CCR resales lost money, against 5.4% in the RCR and 4.6% in the OCR. Where the losses are
• When you bought mattered next, and the two are independent. 12.4% of 2010 to 2014 purchases lost money; 1.8% of 2019 to 2022 purchases have. Timing and region together

What actually happened

Every two-bedroom condominium unit that was bought and later resold, with the resale completing between January 2020 and August 2026, matched on the unit itself so the purchase and the resale are of the same home.

	All two-bedrooms
Units resold	24,232
Sold above purchase price	92.9%
Sold at a loss	6.9%
Median gain	$230,000 (21.7%)
Median hold	8.3 years
Middle half of gains	11.4% to 36.0%
Median loss, among those that lost	$70,600 (5.4%)
Median purchase price	$1,007,000
Median purchase size	850 sqft

Seventy-one units sold at exactly the purchase price. The middle half of gains ran from 11.4% to 36.0%; one owner in ten made less than 2.7%, and one in ten made more than 66.6%. Among the 1,661 who lost, the worst tenth lost more than 16.4%.

What it works out to per year

Annualised over each owner's own hold, the median gain is 2.8% a year in price alone. Whether the unit was bought new or as a resale moved that figure a little. A two-bedroom bought as a resale returned 3.3% a year and a median $272,610; one bought at new launch returned 2.6% a year and $209,200, largely because the resale buyer paid less to begin with. Sub-sale purchases, bought from another buyer before the building was finished, were the least likely to sell above the purchase price at 87.5%, returned the least per year at 2.1%, and were held longest.

Bought as	Units resold	Sold above	Median gain	A year	Median hold
New launch	14,039	92.9%	$209,200 (20.0%)	2.6%	8.4 years
Resale	8,612	93.8%	$272,610 (25.2%)	3.3%	7.1 years
Sub-sale	1,581	87.5%	$250,000 (23.5%)	2.1%	11.2 years

Where the losses are

Region	Units resold	Sold above	Sold at a loss	Median gain	A year	Median hold	Median loss
CCR	3,491	81.1%	18.4%	$257,000 (15.7%)	1.8%	10.1 years	$135,500 (7.3%)
RCR	8,517	94.4%	5.4%	$248,888 (21.2%)	3.0%	7.2 years	$60,200 (5.1%)
OCR	12,224	95.1%	4.6%	$213,950 (23.3%)	2.9%	8.4 years	$40,000 (4.1%)

Nearly one CCR two-bedroom resale in five lost money; one OCR resale in twenty-two did. The CCR is worse on every measure, more likely to lose, a larger loss when it did, a smaller percentage gain when it did not, and the longest hold to get there. Its median dollar gain is the largest of the three only because its purchase prices are. The same ordering held for shoebox units, where the CCR loss rate was 44.2% against 6.0% in the OCR; for two-bedrooms the two figures are 18.4% and 4.6%.

Is it timing

Partly. When an owner bought explains a large part of how they did, and the 2010 to 2014 cohort is the weak one.

Bought	Units resold	Sold above	Sold at a loss	Median gain	A year	Median hold	Median loss
Before 2010	3,568	96.5%	3.5%	$571,650 (79.0%)	3.5%	16.4 years	$270,000 (12.1%)
2010 to 2014	9,004	87.2%	12.4%	$170,000 (17.0%)	1.6%	10.3 years	$70,000 (5.7%)
2015 to 2018	6,402	94.9%	4.8%	$208,000 (20.0%)	3.2%	5.9 years	$50,000 (4.1%)
2019 to 2022	5,169	98.0%	1.8%	$235,000 (20.7%)	4.6%	4.1 years	$77,000 (4.5%)

The 2010 to 2014 cohort is 37% of all units resold, with the lowest share sold above purchase, the smallest gains and the slowest annual rate, because those owners bought at the last peak and sold into a market that took a decade to pass it. Annualised, the windows separate further. Owners who bought before 2010 made 3.5% a year over sixteen years; owners who bought in 2010 to 2014 made 1.6% a year over ten; owners who bought in 2019 to 2022 have made 4.6% a year over four, having bought before the 2021 to 2023 rise and sold into it. 89 units bought in 2023 or later have already resold; at a median two-year hold they are too few and too short to read, and are kept out of the table.

Timing and region together

The obvious objection to the regional finding is that CCR buyers bought at the 2010 to 2014 peak and everyone who did fared badly. Crossing the two answers it.

CCR

Bought	Units resold	Sold above	Sold at a loss	Median gain	A year
Before 2010	929	90.7%	9.2%	58.9%	2.9%
2010 to 2014	1,258	68.9%	30.7%	8.2%	0.8%
2015 to 2018	794	83.9%	15.5%	13.2%	2.0%
2019 to 2022	500	89.8%	9.4%	12.4%	2.9%

RCR

Bought	Units resold	Sold above	Sold at a loss	Median gain	A year
Before 2010	1,142	96.7%	3.3%	84.3%	3.7%
2010 to 2014	2,647	87.8%	11.8%	16.6%	1.5%
2015 to 2018	2,515	96.9%	2.9%	20.0%	3.3%
2019 to 2022	2,186	98.8%	1.1%	20.7%	4.6%

OCR

Bought	Units resold	Sold above	Sold at a loss	Median gain	A year
Before 2010	1,497	99.9%	0.1%	84.4%	3.6%
2010 to 2014	5,099	91.4%	8.2%	19.2%	1.8%
2015 to 2018	3,093	96.1%	3.6%	21.6%	3.3%
2019 to 2022	2,483	99.0%	1.0%	21.9%	4.9%

The CCR is the worst region in all four windows, and 2010 to 2014 is the worst window in all three regions, so neither effect is the other in disguise. A CCR buyer in the best window, before 2010 or 2019 to 2022, still lost money 9% of the time, more than an OCR buyer in the worst window, 2010 to 2014, at 8.2%. Region mattered more than timing, and the two are independent.

One cell carries most of the damage. CCR two-bedrooms bought in 2010 to 2014 are 5.2% of all units resold and 23.2% of all the losses; 30.7% of them lost money, and the cell's median outcome and its median loss were the same size, a gain of $135,000 against a loss of $135,888. Take that cell out and the CCR loss rate falls from 18.4% to 11.5%, still more than double the other regions.

No two-bedroom cell is a majority loss. The shoebox series' worst cell, the same region and window, was. 59.8% of CCR shoeboxes bought in 2010 to 2014 lost money, and the median owner lost 2.6%. The median two-bedroom owner in that cell made 8.2% over eleven years, barely above nothing after costs, but a different sign. The two-bedroom's worst cell, at 30.7%, sits roughly where the shoebox's 2019 to 2022 CCR cell did, at 26.4%.

What this means for judging a specific unit

Region is the single strongest signal in this data. Not a rule that CCR two-bedrooms lose, 81% of them made money, but the base rate is three to four times worse than elsewhere and has been in every purchase window. Purchase year is the second signal, and the two are independent; a CCR buyer in the best window still lost money more often than an OCR buyer in the worst one. Whether the unit was bought new or as a resale mattered less than either. What none of this tells you is the answer for one unit. The middle half of outcomes spans 25 points, and floor, facing, project and the price actually paid sit inside that range.

What this does not tell you

• Rent. These are price outcomes only. A two-bedroom's gross rental yield of 3.29% in 2025, over an eight-year hold, is a large part of total return and is not counted here.
• Costs. Stamp duties, interest, commission and maintenance are not deducted; on a $1 million purchase they run to six figures over eight years. A worked example for one unit is a separate note.
• Units that have not resold. Only completed resales are counted; owners still holding, including those who bought at the peak and have not sold, are not in the data.
• Who these owners were. The record shows the unit and the two prices, never the buyer.
• What happens next. These are completed resales; someone buying in 2026 faces a different market.

Method and data notes

Repeat sales from URA caveat data for private condominiums and apartments, each pair a purchase and a later resale of the same unit, with the resale completing between January 2020 and August 2026 and the purchase in any earlier year back to 1995. Bedroom count is resolved from the unit record; the class of unit records with more than 1,000 sqft per claimed bedroom is excluded (120 pairs), as are two units whose bedroom count was corrected on floor-plan review. Region is the URA market segment on the purchase caveat. Purchase type is the purchase's sale type, new sale, resale or sub-sale. Gain is the resale price less the purchase price, as a share of the purchase price; the annual figure compounds it over the owner's own hold and is not computed for the 19 holds under six months. Medians are of units, not of dollars; the median loss is the median among those that lost. The crossed tables use the same pairs, split by region and purchase window. Latest resale in the data 19 May 2026, from a scan of 554,608 caveats across 335,948 unit identities. The shoebox figures quoted are from the shoebox series on the same construction.