Why PSF alone won't tell you new launch versus resale

A buyer walks out of a showflat in the Rest of Central Region quoting around $2,800 PSF on a 700 sqft two-bedder, then drives ten minutes to a 2014 project where similar units last transacted closer to $2,200 PSF. The premium looks like 25%. The buyer wants to know if the new launch is overpriced or if the resale is the better deal. Neither answer is in the PSF number alone. Total cost over a ten-year hold turns on payment timing, lease, defects, sinking fund, and how the next buyer values the unit at exit. PSF is the headline. The math sits underneath.

What the PSF number actually shows

PSF normalises price by saleable strata area. It doesn't adjust for the variables that move total cost.

Three distortions show up first. New launches in the post-2019 cohort run smaller absolute floor areas than equivalent units a decade earlier, so a higher PSF on a smaller unit can produce a similar or lower ticket price. Resale buyers pay the full price at completion. New launch buyers pay under the Progressive Payment Scheme, with stage payments tied to certified construction milestones, and the time value of those deferred payments isn't in the headline. Resale PSF reflects an asset with a known maintenance history and a measurable sinking fund balance, both visible from the Management Corporation Strata Title. New launch PSF reflects an asset with no operating history.

URA's Q1 2026 final release showed overall private PPI up 0.9% q-o-q, with non-landed CCR up 0.6%, RCR up 0.8%, and OCR up 2.2%. The flash estimate had the OCR lift at 1.3%, so the final revision was material, and any per-region read on new launch versus resale moves with it.

The eight dimensions that matter

Dimension	New launch	Resale
Price discovery	Developer-set; few transacted comparables nearby at launch	Caveats lodged on URA; the bid-ask is observable
Payment	Progressive Payment Scheme over construction; 5% option fee, 20% within 8 weeks	Full price at completion, typically 4 to 14 weeks from OTP
Physical inspection	Showflat and floor plan only	Direct inspection; defect history visible
Tenure	Fresh 99-year leasehold or freehold from completion	Remaining lease net of years elapsed
Defect liability	12-month Defects Liability Period from notice of vacant possession under the standard form S&P	None from the seller
Buyer pool at exit	Competes with later phases and adjacent older stock	Established competitive set, pricing benchmarked over years
Secondary liquidity	Limited transaction history; bank valuations may anchor to developer pricing	Deep caveat history supports valuation
Renovation	Developer-spec finishes; minimal at handover	A 10 to 15-year-old unit commonly needs kitchen, bathroom, flooring works

The Defects Liability Period sits inside the standard form Sale & Purchase Agreement issued by the Controller of Housing under the Housing Developers (Control and Licensing) Act. The exact terms applicable to any purchase live in the executed contract.

The PPS question buyers usually skip

The Progressive Payment Scheme isn't a discount. It's a cash-flow profile. A buyer pays roughly 20% by the eight-week mark, then 10% on foundation works, 10% on reinforced concrete framework, 5% on partition walls, 5% on roofing, 5% on door and window frames, 5% on car park, drains and roads, 5% on TOP, and the remaining 15% on the certificate of statutory completion. Total still 100%, but stretched across two to four years on a typical project.

Two consequences. First, less cash is locked up early, so the opportunity cost of capital favours the new launch over a resale of the same headline price. Second, mortgage interest only kicks in on each disbursed tranche, so total interest paid during construction is lower than on a resale fully drawn at completion. Q1 2026 fixed-rate packages sat in the 1.40 to 1.60% range per industry surveys, with 3M SORA around 1.07%. At those rates the carry savings are real but modest. Five years ago at 3.5% they were significant.

The cost on the other side: maintenance and property tax start at TOP, but a tenant rarely walks in on day one. Carry the void.

Stamp duty and ABSD don't care which one you bought

BSD applies the same way on both. On a $2 million purchase the BSD comes to roughly $64,600, calculated on IRAS brackets that step up to 5% on the slice from $1.5m to $3m and 6% above that. The duty is owed on the higher of price or market value, and it's payable within 14 days of contract.

ABSD is where buyer profile bites. A Singapore Citizen buying a first property pays nothing. A second property attracts 20%. A third or more, 30%. A Singapore PR pays 5% on the first, 15% on the second, 30% on the third. Foreigners pay 60% flat. None of that distinguishes new launch from resale, but it dominates total outlay for anyone past their first home and is worth pricing in before any PSF comparison.

The other rule that's changed and stayed underweighted in buyer conversations: SSD was reset on 4 July 2025. Holding period extended from 3 to 4 years, and rates went up 4 percentage points per tier. Sell within a year of purchase and the duty is now 16%, not 12%. Sell in year three and it's 8%, not 4%. Sell in year four and it's 4%, not zero. A new launch buyer who plans a quick exit on early TOP appreciation needs to remember the SSD clock now runs four years from the date of acquisition, not from TOP.

When the new launch premium has a real case

Four situations where the premium isn't just a developer markup.

Precinct creation. Where a Government Land Sales site is the first or second project in a redevelopment precinct, the launch is pricing the precinct's future. Bayshore Road's first GLS site went to SingHaiyi Garnet at $1,388 PSF PPR in March 2025, and the Bayshore Drive mixed-use mega-site is now in tender. The first launches there will price the precinct as it will be, not as it is. The buyer carries the timing risk.

MRT proximity not yet capitalised in resale. Where a station is announced or under construction but not operational, resale prices in the catchment may not have fully repriced. The new launch may price the station as if it's running on TOP date. JRL slipped again in March 2026: JRL1 now mid-2028, JRL2 in 2028, JRL3 in 2029. A new launch in the Tengah catchment priced for end-2027 opening is pricing a service that's now six months further out.

Tenure reset. A new 99-year leasehold gives the full lease. A resale of a 15-year-old project gives 84 years remaining. Lease decay isn't linear: bank LTV starts contracting around 60 years remaining, and below 30 years CPF and bank financing fall away.

Lower strata maintenance age. Major capital works on lifts, façade, waterproofing, and M&E typically arise from year 10 onwards. A new launch defers them. A resale buyer can face a sinking fund top-up or special levy resolution within the holding period.

Risks the new launch carries and resale doesn't

TOP timing slips. The S&P specifies an estimated date, longstop dates trigger liquidated damages, but those payments don't eliminate the buyer's holding cost.

No physical inspection. The buyer transacts on a showflat. Discrepancies between marketing and the delivered unit are addressable only through the Defects Liability Period.

Developer execution varies. Quality of finishes, layout efficiency in built form, and common-area execution differ. There's no published execution-quality index.

Void period between TOP and first tenancy. If the unit is for letting, the gap is a carrying cost paid in full by the buyer.

What resale actually offers

Immediate occupancy or rental, no void. Price negotiability between two private parties rather than a developer with a launch grid. Visible strata condition: the MCST's audited accounts and minutes are obtainable, and the sinking fund balance, recent special levies, and pending capital works resolutions can be inspected before transacting. Established amenities, schools, transport, and retail that already exist rather than rendered.

There's also a quieter point on bank financing. A new launch project in the period right after TOP has limited resale caveats, so bank valuations may anchor to developer launch prices, which can sit above what an arm's-length buyer would pay. The valuation gap is a funding risk for the next buyer rather than the current one, but it shapes the next buyer's willingness to pay, and therefore the current buyer's exit price. A resale unit in an established project has a deep caveat history, and bank valuations track the recent transaction band more tightly.

Where the framework breaks down

It doesn't apply where the comparison is freehold resale against a 99-year leasehold new launch. That collapses into the freehold versus leasehold question, which is a different one. It also doesn't hold in segments with very thin transaction volume, where caveat-based price discovery thins out and the resale advantage narrows. And it doesn't apply where the buyer is underwriting an en-bloc outcome on the resale side, which is a redevelopment-probability question, not a hold-and-rent one.

Bottom line

The choice doesn't resolve on PSF. Payment timing, tenure, sinking fund position, defect cover, and exit liquidity typically dominate the headline differential over a ten-year hold. A 25% PSF premium can shrink to single digits once the PPS carry, the SSD profile, the renovation gap, and the void differential are reconciled, or it can widen if the new launch has weak resale comparables at exit.

What the framework doesn't do: pick the winner. That depends on the entry price relative to the project's eventual transacted band, the trajectory of URA PPI for the relevant region, and outcomes that aren't knowable in advance. It also doesn't say the new launch premium in any particular project is fair or excessive. That requires looking at the GLS land cost the developer paid, the construction cost base, and the precinct's pipeline supply, project by project.

The buyer's job before signing isn't to defend the PSF print. It's to populate the total-cost line items with current numbers and a stated holding period, then see which side actually comes out ahead.