01 · The questionThe tax that reshapes decisions
Not the largest cost in a purchase, but the one that changes the plan.
ABSD does not stop the market. It changes who is in it.
Additional Buyer’s Stamp Duty rarely stops a purchase outright. What it does is rearrange the sequence: sell first or buy first, decouple or do not, buy in one name or two. Those decisions are worth six figures, and most are made on half-remembered rates.
Below: the rates as they stand, what they cost in dollars, and what the transaction record actually shows happened each time they moved.
ABSD changes sequencing more than it changes appetite. Sequencing is where the money is.
02 · The claimThe rates, in full
Four buyer types, three property counts.
Entities — companies and trusts — pay 65%, with an additional non-remittable charge for housing developers. ABSD is assessed on the higher of purchase price or market value, and is payable within 14 days of exercising the Option to Purchase.
These rates have held since 27 April 2023. Married couples with at least one Singapore Citizen may claim remission on a second property if the first is sold within six months of completion — a rule that is generous in principle and unforgiving on the date.
The citizen second-property rate is 20%. That single number drives most upgrade sequencing in Singapore.
03 · MethodHow we measured the effects
The rates are policy. The volume effects are ours.
- Rates as reported by multiple secondary sources, July 2026; verify against IRAS before acting.
- Volume and price effects: URA caveat records, 2021-07 to 2026-07, private non-landed, Executive Condominiums excluded.
- Event window: mean monthly volume in the six months before an announcement against the six months after.
- Caveats are lodged with a lag, so the month of an announcement is noisy. We compare windows, never single months.
- Minimum 40 transactions per reported group.
The obvious limitation: rates change alongside interest rates, supply cycles and sentiment. What follows is what the record shows, not proof of what caused it.
04 · Core evidenceWhat ABSD costs in dollars
The rate table converted into the number that actually matters.
On a S$2,000,000 purchase: a citizen buying a first home pays nothing. The same citizen buying a second pays S$400,000. A permanent resident buying a second pays S$600,000. A foreigner pays S$1,200,000 on the first.
This is on top of Buyer’s Stamp Duty, which everyone pays and which reaches 6% at the top marginal band on residential property.
Second-property ABSD on a S$2,000,000 purchase is S$400,000 for a citizen. Decoupling and sequencing decisions start there.
05 · The mechanismWhat happened to volume
Four policy events, and the transaction record around each.
After the December 2021 package, mean monthly volume fell from 2,438 to 1,562 — 35.9%. That is the textbook response, and it is the one everybody remembers.
The rate rises of 2022 coincided with a further 35.8% fall. Then the pattern breaks.
Can you work out my ABSD exposure and the cleanest sequence? →The first two tightenings cut volume by about a third each. The next two did not.
06 · Cohort proofThe 60% rule and the foreign buyer
What the April 2023 move actually did.
In April 2023 foreign-buyer ABSD went to 60%. In the six months that followed, mean monthly volume rose 47.1% — from 1,274 to 1,875.
That is not a paradox. Foreign buyers were already a small share of transactions; the binding constraint on the domestic market in that window was supply and rates, not stamp duty. Removing foreign competition from prime segments did not remove domestic demand from suburban ones.
What the 60% rate did do is close a channel almost completely. The policy achieved its aim precisely — it just did not achieve the aim most commentary assigned to it.
Read ABSD changes as a composition tool, not a volume brake. Ask who is excluded, then ask who is left bidding.
The 60% foreign rate coincided with volume rising 47.1%. It changed the buyer mix, not the volume.
07 · The counter-caseWhere ABSD did not work as expected
The honest correction: our own event study is weak evidence.
The fourth event in our window — the 2025 Seller’s Stamp Duty extension to four years — was followed by a 82.6% rise in volume. Taken at face value, that would mean tightening stimulates the market, which is nonsense.
The honest reading is that a six-month before-and-after window cannot separate policy from everything else moving at the same time: launch pipelines, interest rate expectations, and the caveat-lodgement lag itself. Two of our four events point one way and two point the other.
So we publish the number that undercuts the neat story alongside the number that supports it. What survives is narrow and useful: ABSD reliably changes who transacts. Its effect on how much transacts is not established by this data.
Two of four events cut volume; two coincided with rises. Anyone claiming a clean ABSD-to-volume rule is over-reading.
08 · For youWhat it means for you
Four situations, four numbers.
If you are upgrading
Sell first and you pay 0%. Buy first and you pay 20% up front, recoverable only if you sell within six months of completion and meet the remission conditions. On a S$2,000,000 purchase that is S$400,000 at risk on a deadline.
If you are buying a second property to keep
Budget the 20%% as a permanent cost and rerun your yield on the all-in price. A 3.5%% gross yield on the price becomes roughly 2.9%% on the price plus ABSD.
If you are a permanent resident
Your first property carries 5%. On S$1,700,000 that is S$85,000 — a real cost most first-time PR buyers under-budget.
If you are considering decoupling
The arithmetic can work, but it involves stamp duty on the transfer, legal costs and refinancing. Model it fully before treating it as a saving.
What is your ABSD exposure, exactly?
Tell us your situation. We will send your ABSD and BSD in dollars at your intended price, plus the cleanest sequence for your case and the dates that govern any remission.
- Your ABSD and BSD in dollars, at your price
- Sell-first versus buy-first compared on total cost
- The remission conditions and deadlines that apply to you
One WhatsApp message back — usually same day.
We’ll WhatsApp it shortly — usually within the day.
09 · ReferenceThe full rate table
All four buyer types, all three counts.
10 · MethodSources, limits and corrections
Data: URA caveat records, 2021-07 to 2026-07, private non-landed, Executive Condominiums excluded. Thresholds: minimum 40 transactions per reported group. Policy source: ABSD rates from published secondary sources, July 2026, pending verification against IRAS — confirm before acting. Event study: six-month means before and after each announcement; correlation only. Reproducibility: every figure is produced by a script against the source dataset; corrections are made in place and dated. Not investment advice. Analysis of public records, published to be argued with.
Twenty per cent of your next purchase is decided by sequence.
Two minutes in the Scenario Planner compares sell-first against buy-first on total cost, including ABSD.
Map my move →