A new 3 bedroom outside central rents for more per square foot than a 3 bedroom 26 years old or older in the core central region, and has since 2020

A 3 bedroom in a building 5 years old or newer rented for a median S$4.92 per square foot outside central in 2025. A 3 bedroom in a building 26 years old or older rented for S$4.02 in the core central region. The newer suburban unit costs 22% more per square foot than the older prime unit.

In 2017 that comparison ran the other way: S$2.82 in prime against S$2.60 outside central. Nothing about the buildings changed. What changed is how much the market pays for newness.

What is in this note

• What age is worth: median rent per square foot by how old the building is, nationally, from 2017 to 2025.
• When it changed: the test the title rests on: whether a new 3 bedroom outside central (OCR) rents for more per square foot than an old 3 bedroom in the core central region (CCR), and in which years.
• It is not size and it is not region: two other explanations for the age pattern, tested against the data.
• The stock is ageing: how many leases fall in each age band, and how that has shifted.

What age is worth

Median rent per square foot by the building's age when the lease began, all regions.

• 0-5 years old
• 6-10 years old
• 11-15 years old
• 16-25 years old
• 26+ years old

Year	0-5	6-10	11-15	16-25	26+	Newest / oldest
2017	S$3.03	S$3.33	S$2.92	S$2.48	S$2.43	1.25
2018	S$3.07	S$3.28	S$2.96	S$2.44	S$2.41	1.27
2019	S$3.06	S$3.34	S$3.10	S$2.46	S$2.34	1.31
2020	S$3.08	S$3.34	S$3.05	S$2.48	S$2.19	1.41
2021	S$3.54	S$3.38	S$3.34	S$2.65	S$2.38	1.49
2022	S$4.81	S$3.92	S$4.09	S$3.19	S$2.83	1.70
2023	S$5.79	S$4.89	S$4.78	S$3.94	S$3.56	1.63
2024	S$5.68	S$4.69	S$4.68	S$4.04	S$3.58	1.59
2025	S$6.01	S$4.73	S$4.78	S$4.25	S$3.64	1.65

In 2025, rent falls with age at every step of the ladder inside each of the three regions. The earlier years are not quite so clean: 2017, 2018 and 2021 each carry one or two reversals of a few cents, 2019 has a tie in prime, and outside central in 2017 and 2018 the oldest band's median is just above the 16-25 band's.

The national column above pools three regions whose age profiles differ, which is why the middle of it inverts twice: in 2022 and 2025 the 11-15 band's median is above the 6-10 band's. The 2025 regional table below has no such reversal.

The top of the ladder is also untidy before 2021. From 2017 to 2020 the 6-10 band out-rented the newest band nationally, and only from 2021 does the newest band lead outright.

What changed is the size of the gap. In 2017 the newest band rented at 1.25 times the oldest. By 2025 it was 1.65 times. Most of that arrived in a single year: 1.49 in 2021 to 1.70 in 2022.

When it changed

The claim in the title is a specific test: does the newest band outside central beat the oldest band in prime?

Year	Oldest, core central	Newest, outside central	Which leads
2017	S$2.82	S$2.60	Core central
2018	S$2.86	S$2.62	Core central
2019	S$2.79	S$2.73	Core central
2020	S$2.58	S$2.74	Outside central
2021	S$2.73	S$3.17	Outside central
2022	S$3.26	S$4.09	Outside central
2023	S$4.02	S$5.04	Outside central
2024	S$4.04	S$4.89	Outside central
2025	S$4.02	S$4.92	Outside central

Location led in 2017, 2018 and 2019, the last of those by six cents. It has lost in every year since.

The 2020 crossing came from a fall in prime, not a rise outside central. In 2020 the prime figure fell, from S$2.79 to S$2.58, while the suburban figure barely moved. That fall came in the first COVID year.

Over the 5 years since, both rise and the gap widens, because the suburban figure rises faster, from S$2.74 to S$4.92. Neither line rises every single year; the direction over the span is not in doubt.

This is not a claim that location stopped mattering. Inside any one age band, prime still rents for more than the suburbs, in every year. The claim is narrower: the gap between a new building and an old building has grown wider than the gap between the best address and the ordinary address. Prime has been losing ground on price too, growing slowest of the three regions on every measure over the same 9 years.

It is not size and it is not region

Two other explanations could account for the age pattern, unit size and region, and the data rules out both.

The first is unit size. Newer 3 bedrooms are smaller, and a smaller unit rents for more per square foot. That is real: the median leased area rises with building age, from about 95 square metres in the newest band to 125 square metres in the oldest, though not at every step. But splitting 2025 by unit size does not overturn it, though only one of the four bands has enough leases on both ends to quote directly. A 3 bedroom of 90 to 110 square metres rented for S$5.87 per square foot in the newest band and S$3.89 per square foot in the oldest. Rent per square foot falls from the newest building band to the oldest in that size band, and in the same direction in the other three size bands.

The second is region. Newer homes are concentrated outside central, where rents are lower, so a national figure could be picking up location. But rent per square foot still falls with building age inside each region separately, across all 15 age-by-region groups in the 2025 grid, with no reversal:

Age	CCR	RCR	OCR
0-5	S$7.43	S$6.28	S$4.92
6-10	S$6.34	S$5.38	S$4.34
11-15	S$5.41	S$4.78	S$4.04
16-25	S$4.98	S$4.42	S$3.65
26+	S$4.02	S$3.78	S$3.36

Both effects are real, and the age effect is larger than the region premium. In the core central region the newest band rents at 1.85 times the oldest; the widest region gap in any single band is 1.51 times.

The stock is ageing

The bands are not the same size, and they are moving.

Year	0-5 leases	26+ leases
2017	6,140	2,116
2018	6,120	2,323
2019	5,528	2,632
2020	4,286	3,052
2021	3,231	4,098
2022	2,075	4,251
2023	1,687	5,580
2024	1,798	5,839
2025	1,882	6,704

The newest band has shrunk by more than two thirds and the oldest has more than tripled, crossing in 2021. Fewer completions are reaching the rental market and more of the existing stock is passing 26 years.

That matters for reading the tables above. Each year is a valid picture of that year, but the buildings in a band change from year to year: a block that was 16-25 in 2017 is in the 26+ band by 2025. Following one age band across the years tracks a category, not the same buildings.

What this tells you

A 3 bedroom in a building 5 years old or newer rents for more per square foot than a 3 bedroom 26 years old or older, and the difference has been widening. In 2017 the newest band rented at 1.25 times the oldest. By 2025 it was 1.65 times, with most of that move happening in a single stretch from 1.49 in 2021 to 1.70 in 2022.

That gap has grown large enough to out-rank location. A 3 bedroom in a new building outside central rented for S$4.92 per square foot in 2025 against S$4.02 for a 3 bedroom 26 years old or older in prime. In 2017 it ran the other way, and location led until 2019, by 6 cents in that final year. Outside central has been ahead every year since 2020.

The flip was not a suburban surge. Prime rents in the oldest band fell from S$2.79 to S$2.58 between 2019 and 2020 while the figure outside central barely moved. Age overtook address because rents for old prime homes fell, not because rents for new homes outside central rose: they were S$2.73 in 2019 and S$2.74 in 2020.

Two other explanations for the gap do not hold. It is not unit size: newer 3 bedrooms are smaller, and smaller units rent for more per square foot, but holding size constant at 90 to 110 square metres the newest band still rents for S$5.87 per square foot against S$3.89 for the oldest. It is not region either: rent per square foot falls with building age inside every region, across all fifteen age-by-region groups in the 2025 grid, with no reversal.

In fact age separates rents more sharply than region does. In prime the newest band rents at 1.85 times the oldest. The widest gap between regions within any single age band is 1.51 times. Both premiums are real, and the premium for newness is the larger of the two premiums.

What none of this says is that the buildings themselves changed. A block that sat in the 16-to-25 band in 2017 is in the oldest band by 2025, and the bands have shifted underneath the figures: leases on homes 5 years old or newer have fallen by more than two thirds since 2017, from 6,140 to 1,882, while leases on homes 26 years old or older have more than tripled, from 2,116 to 6,704. Each year is a fair picture of that year. Following one age band across the years tracks a category, not the same buildings.

What this does not tell you

• Why newness commands a premium. This measures that it does, not what tenants are paying for. Facilities, layout, finishes and the absence of a renovation cycle are all candidates and none is measured here.
• Whether it holds. The premium rose from 1.25 times in 2017 to 1.65 times in 2025, and most of that came in 2021 and 2022, the same 2 years rents rose across the whole market. A premium that increased in a tight market may not survive a loose market.
• What a landlord nets. An older building's maintenance costs, sinking fund contributions and vacancy periods are not in these figures.
• Anything about a specific project. These are medians across dozens to hundreds of developments in each age band and region. The thinnest quoted here covers 35.

Method and data notes. Rental data to 1 July 2026.

Medians of URA rental contract data for private condominiums and apartments, 3 bedroom, by the date the lease began. Building age is the years between the project's completion and the lease commencement, banded 0-5, 6-10, 11-15, 16-25 and 26 or more. Rent per square foot uses the midpoint of the contract's recorded floor-area band, since URA publishes rental areas in bands rather than exact figures.

Each year is a separate cross-section. A building moves between bands as it ages, so a band in 2025 holds different buildings from the same band in 2017. Comparing one band across the years compares categories, not the same properties over time.

Every figure quoted rests on at least 390 leases. The regional split is a partition: the three regions sum exactly to the national total in every year.

Rental data to 1 July 2026.

30 October 2026. This note placed homes in market segments by postal district. URA places them by planning area, which moves about one in seven private sales between regions. The figures by region here use the earlier segments and differ from what the same records give now. How regions are set: How we measure prices and sales.