Tenants pay a smaller shoebox premium in newer buildings

Tenants pay a smaller premium for a shoebox in a newer building. In condominiums completed between 2009 and 2012, a small unit rents for 42.8% more per square foot than a family-sized one. In buildings completed since 2021, that falls to 24.1%.

We found no such pattern in sale prices by completion year: the price premium showed no gradient with building age at all, except in the newest stock. Rent behaves differently, and where the two agree is the interesting part.

Key takeaways

• The shoebox rent premium falls with newness: 42.8% in 2009–2012 stock, 24.1% in 2021–2024 stock. The four groups
• The reason is on the family side. Shoebox rents rise modestly with newness, $6.06 to $7.02 psf. Family rents rise much faster, $4.25 to $5.65. What is driving it
• The newest group is the weakest cell on both halves of the return (lowest rent premium and the widest price discount) from two separately collected datasets. Rent and price together
• Building age still does not explain the earlier regional or tenure findings. There is almost no old shoebox stock to vary. What this does not change

Four groups, rented today

Median monthly rent per square foot on leases signed between 2023 and 2025, grouped by the year each project was completed. Shoebox is URA's 40–50 sqm rental band, which ends at the same 538 sqft line as its shoebox definition. Family-sized is the 90–130 sqm bands.

Completed	Shoebox	Family	Rent premium	Projects (S / F)
2009–2012	$6.06	$4.25	+42.8%	98 / 248
2013–2016	$5.99	$4.16	+44.0%	217 / 230
2017–2020	$6.40	$4.82	+32.9%	72 / 80
2021–2024	$7.02	$5.65	+24.1%	79 / 99

The last column shows how many separate projects each median is drawn from (shoebox first, family second).

The premium is flat across the two older groups, then falls in each of the two newer ones. From 2013–2016 stock to 2021–2024 stock it drops by nearly twenty points.

We start at 2009 for the same reason our price article did: shoebox units are a recent format, and there is almost no older stock to measure. Buildings completed before 2009 contribute only 42 projects with enough leases to qualify, and they are heavily concentrated in the central regions, so they carry a locational signal as much as an age one. They are excluded rather than reported as a fifth row.

The family side is doing the work

Read the two dollar columns rather than the premium.

Shoebox rents rise with newness, but gently: $6.06, $5.99, $6.40, $7.02. About 16% from the oldest group here to the newest.

Family rents rise steeply: $4.25, $4.16, $4.82, $5.65. About 33% across the same span, roughly double the shoebox increase.

So the premium is not compressing because tenants have stopped valuing small units. Shoebox rents in new buildings are the highest in the table. It is compressing because a new family-sized unit commands proportionally much more than an older one, while a new shoebox commands only a little more than an older one.

Put simply: newness is worth more in a family unit than it is in a shoebox. A tenant paying for a recent building is paying for space, finishes and facilities that a 500 sqft unit has less room to deliver.

Rent and price together

Our price article found no age gradient at all (the premium ran −3.6%, +3.6%, −1.6% across the three older groups, in no order) and then dropped to −16.4% in the newest. Setting the two side by side:

Completed	Rent premium	Price premium
2009–2012	+42.8%	−3.6%
2013–2016	+44.0%	+3.6%
2017–2020	+32.9%	−1.6%
2021–2024	+24.1%	−16.4%

The two columns behave differently. Rent declines steadily; price wanders and then falls off a cliff. But they agree about the newest group, and they agree emphatically: 2021–2024 stock is the weakest cell for shoeboxes on both measures.

That agreement is worth something because the two datasets are collected separately. Rental contracts and sale caveats are filed by different parties for different purposes. When both say the same thing about the same group, it is less likely to be an artefact of either.

What they say is that in the newest buildings, the gap between a small unit and a family-sized one is wider than it has been in any earlier vintage: in what tenants will pay, and in what buyers will pay.

What this does not change

Our earlier articles found that building age does not explain why prime-district or freehold shoeboxes have underperformed. That still holds, and this article does not disturb it.

The reason is structural. Shoebox units in Singapore were built almost entirely in one fifteen-year window. There is no meaningful spread of old and new shoebox stock for age to explain anything with. What this article measures is not age as a cause: it is what the market pays for newness within a format that is uniformly recent.

What this does not tell you

• A trend over time. This is a snapshot of leases signed in 2023–2025, grouped by building age. It is not a series showing how any group's premium has moved, because newer groups have no earlier history to compare against.
• Buildings before 2009. Too few shoebox projects qualify, and those that do are concentrated in the central regions.
• Yield. Rent premium is not yield. Yield needs a price as well, and we cover it market-wide and by tenure.
• Net income. These are gross rents, before maintenance, property tax, agent commission, vacancy and income tax.
• Why tenants pay it. Rental contracts record the rent, not the reasoning. Anything about who rents small units, and why, would be inference.

Method

URA non-landed rental contracts, leases commencing 2023 to 2025. Cohorts are defined by project completion year.

URA reports rental floor area in 10 sqm bands rather than exact areas, so rent per square foot uses the midpoint of each band. Shoebox uses the 40–50 sqm band, which converts to 431–538 sqft and shares its upper boundary with URA's 50 sqm shoebox threshold. Family-sized uses the 90–130 sqm bands, 969–1,399 sqft. Because both sides use midpoints, the ratio between them is more reliable than either level on its own.

A project qualifies for a cell when it has at least three leases in that band across the window. Medians are taken across projects rather than pooled across contracts, which would let the changing mix of what was leased masquerade as a change in rent.

This is a cross-section rather than a fixed panel. The paired-window method used elsewhere in this series compares each project against its own earlier self, which cannot reach cohorts completed after 2020: a building finished in 2023 has no 2017 rental history.

Rental data to February 2026.