HDB lease decay is the one risk in Singapore housing that nobody can negotiate, refinance or wait out. Every flat is on a 99-year clock, and the clock only runs one way. But in seventeen years and 500-odd transactions I have never once been able to show a client a straight answer to the obvious question: how bad is it in my town, and when does it get worse? So my team and I went and built it. Over the past fortnight we profiled all 26 HDB towns from the government’s own block register, and this article publishes the result — plus the part no dashboard anywhere does. We put dates on it.

Here is the headline. Across Singapore, 4,129 of 10,657 residential HDB blocks — 38.7% — already have under 60 years of lease left. That is not a future problem. That is today. And the spread between towns is brutal: from 100% of Marine Parade to 0.7% of Pasir Ris.

HDB lease decay by town: all 26 towns, ranked

This is the table that did not exist before. One row per HDB town, built from the HDB property register — every residential block, its completion year, its town. The middle column is what matters: how many of that town’s blocks have already dropped below 60 years of remaining lease.

Why blocks and not flats? Because block counts come straight out of the register as exact integers I can verify. Unit counts require summing dwelling-unit fields across thousands of rows, and I would rather publish a number I can stand behind than a bigger number I can’t. A town with many tall blocks will have more flats behind each row than a town of walk-ups — bear that in mind when you read Marine Parade against Tampines.

Tengah is not a row. Tengah has 139 residential blocks in the register and no resale market whatsoever — the first MOP wave of roughly 3,435 units only reaches the market in 2028. Its oldest block dates to 2023, so no Tengah flat crosses the 60-year line until 2063. Including it as a “0%” row would flatter the national average with homes nobody can buy resale yet.

A few things in that table genuinely surprised me, and I have been doing this a long time.

Marine Parade is finished. Not declining — finished. Every single one of its 59 residential blocks is under 60 years, because the newest one was completed in 1976. There is no young stock to average against. When people ask me about Marine Parade I no longer talk about lease decay as a risk; I talk about it as a fact you price in on day one.

Bukit Merah has been living with this since 1977. Its oldest block dates to 1937 — the pre-war SIT stock. Bukit Merah crossed its first 60-year line nearly fifty years ago and the town is still one of the most liquid in Singapore. Hold that thought; it matters later.

Bukit Panjang crossed its very first 60-year line this year. 2026. That is not a coincidence of the calendar — it is the leading edge of a wave, and Bukit Panjang is about to demonstrate the single most violent move in this entire dataset.

Why 60 years is the number that actually matters

People throw “60 years” around as if it were a valuation rule. It isn’t. It is a financing rule, and it is not even really about 60 — it is about age 95.

The governing test, in HDB’s own words on its mode of financing page, is this: “The total amount of CPF savings that can be used for your flat purchase and/or paying monthly mortgage instalments will depend on the extent the remaining lease of the flat can cover the youngest buyer up to the age of 95.”

Unpack that and three rules fall out:

1. If the remaining lease covers the youngest buyer to 95 — full CPF use, up to the Valuation Limit (the lower of purchase price and valuation at time of purchase). HDB loan-to-value sits at the full 75%.
2. If it does not — CPF is pro-rated. HDB says the amount is “pro-rated based on the extent the remaining lease of the flat can cover the youngest buyer up to the age of 95,” and that the LTV limit “will be pro-rated from 75%.”
3. If the remaining lease is under 20 years — no CPF at all. CPF Board sets a Minimum Lease Period of 20 years “to ensure prudent use of CPF savings.” Below that, the flat is a cash purchase.

That is where “60 years” comes from. It is simply the lease a 35-year-old needs — and 35 is roughly the middle of the HDB resale buyer pool. Cross below 60 and you start losing the biggest, hungriest, most loan-dependent segment of your market: young couples upgrading out of a BTO or buying their first resale flat. Those are the buyers who bid hardest, because they are the ones who can borrow most.

Above 60 years, everyone can buy your flat. At 55 years, your 35-year-old buyer is pro-rated but your 45-year-old is not. At 45 years, you have lost everyone under 50. The pool narrows from the bottom up, and the bidding gets thinner as it goes.

The money: two buyers, one flat, two leases

Abstractions don’t move anybody. Let me put real numbers on it. Take one 4-room flat, 90 sqm, transacting at $560,000 with valuation also at $560,000 — so the Valuation Limit is $560,000. Now run the same flat past two buyers at two different remaining leases.

Scenario 1 — the flat has 65 years of lease left

Buyer A, aged 35 needs 60 years. She has 65. Test passed. Full CPF up to the $560,000 VL, HDB LTV at the full 75% — a maximum loan of $420,000, leaving $140,000 to find, all of which can come from her CPF Ordinary Account.

Buyer B, aged 55 needs 40 years. He has 65. Test passed. Same 75% LTV, same $420,000 loan, same $140,000 upfront.

Gap between them: zero. Above the line, lease is invisible.

Scenario 2 — the same flat, now with 55 years left

Buyer A, aged 35 needs 60 years and has 55. Test failed. Her entitlement pro-rates by the extent the lease covers her to 95 — 55 ÷ 60, or 91.7%. Her CPF cap drops to roughly $513,000, and more painfully her LTV drops from 75% to about 68.8%. Maximum loan: roughly $385,000. She now has to find $175,000 upfront.

Buyer B, aged 55 needs 40 years and has 55. Test passed. Full 75% LTV, $420,000 loan, $140,000 upfront — completely unaffected.

Ten lease years did that. Not a market crash, not an interest-rate move — ten years on a clock. And notice what it does to a seller: at 65 years you are selling to everybody. At 55 years, every buyer under about 40 is bidding with one hand tied, which means they bid lower or they walk. That is how lease decay shows up in your sale price long before it shows up in any valuation table.

These are illustrative pro-rations to show the shape of the problem — your own numbers depend on your CPF balance, your age, your existing loans and the actual valuation. Run yours properly with the free calculators at listings.sg/tools before you make any decision, and sanity-check against CPF’s own housing usage calculator.

HDB lease decay by town, projected forward to 2036

Here is the part that makes this article worth your time, and the part I have not seen anywhere else. The under-60 share is not a static statistic. It has dates. Every block in the register carries its completion year, so I can tell you exactly when each town’s share moves — and by how much.

The arithmetic is simple and unforgiving. A block is under 60 years remaining on 1 January of year Y if it was completed in or before Y − 40. So the share on any future date is just a cumulative count of the town’s build history. Build history is fixed. The future of this table is already written.

I projected 12 towns properly. I picked them for spread — the two extremes, the biggest towns by block count, and the towns where the curve does something interesting.

And the column every seller actually wants — how long until this becomes your problem:

Cliffs and plateaus: why build gaps decide your exit

Look at that projection table again and you’ll notice the curves are not smooth. They lurch, then they stall. That is not noise — it is Singapore’s building programme showing through forty years later.

Before you read your own town off that table, one practical note: what a cliff costs you is not an abstraction, it is the extra cash your buyer suddenly has to find — and therefore the discount they will ask for. Model it on your own numbers with the free calculators at listings.sg/tools.

HDB never built at a constant rate. It built in bursts, and between bursts there were years where a town got almost nothing. Forty years on, those bursts become cliffs and those gaps become plateaus. If you own a flat, which side of a cliff you sit on is worth more to your exit price than almost any renovation you will ever do.

The cliff: Bukit Panjang, 2028

Bukit Panjang goes from 8.4% to 43.6% in a single step — the sharpest move in the country. The reason is blunt: 122 of its 346 blocks were completed in 1987 and 1988 alone. The town was essentially built in two years, so it ages in two years.

Then it stops dead. Zero blocks were completed in Bukit Panjang between 1991 and 1994. So after the 2028 cliff the share sits at 44.8% and does not move through 2030, 2032 or 2034 — a six-year plateau — before jumping again to 52.3% in 2036.

The other plateau: Woodlands

Woodlands shows the same signature. Nothing at all was completed there between 1989 and 1992, so its share sits pinned at 30.0% right through 2028, 2030 and 2032 — then more than doubles the pace, hitting 36.8% in 2034 and 58.2% in 2036 as the big mid-90s cohort rolls over. A Woodlands owner has six genuinely quiet years and then a fast one. That is a planning window, and most owners there have no idea they have it.

The opposite shape: Choa Chu Kang

Choa Chu Kang does the reverse — no cliff, just a relentless escalator. 4.2% today, then 10.4%, 21.2%, 34.2%, 53.1%, 63.1%. It was built steadily across the late 80s and 90s, so it ages steadily. There is never a bad year in Choa Chu Kang and there is never a good one either. It just goes up, every single year, for a decade.

The town that barely moves: Bukit Batok

And then Bukit Batok, which is the most stable town in the dataset for the strangest reason: it already happened. 63.9% today, 66.7% in 2034 — a move of under three percentage points in eight years, because only 12 of its 435 blocks are due to cross in that window. Bukit Batok did its lease decay in the 2010s and early 2020s. The shock is behind it. (Our Bukit Batok town piece goes into the psf detail; it’s drafted but not yet live on the site.)

The Pasir Ris problem

If you read one section of this article, read this one.

Pasir Ris is, today, the safest town in Singapore on lease. Three blocks out of 441 are under 60 years — 0.7%. Lower than Choa Chu Kang, lower than Bukit Panjang, lower than anywhere except the three towns sitting at a clean zero. If you bought in Pasir Ris and someone told you lease decay was a concern, you would have laughed.

Now look at the eight-year path:

0.7% to 87.5% in eight years. A swing of 86.8 percentage points — 383 blocks — which is comfortably the largest movement of any town in Singapore. Pasir Ris goes from the safest town in the country to one of the most exposed, inside a single property cycle.

The cause is the same mechanism as Bukit Panjang, just bigger. Pasir Ris was built almost entirely between 1988 and 1994 — a compressed, concentrated town-building programme. A town built in six years ages in six years. There is no old stock to have already absorbed the hit, and no new stock to dilute it. The entire town crosses the line together.

Bishan runs the same play on a shorter fuse: 39.5% today, 57.6% in 2028, 84.0% by 2030. Bishan was built in a tight burst in the late 80s and it will cross almost entirely within four years. For a town that people think of as blue-chip and safe, that is a fast-moving number.

What the market actually pays for lease

Now for the uncomfortable part, and the reason I have not told a single client to dump a flat because of this table.

The market does not price lease the way the financing rules suggest it should. Our town-by-town analyses across this series kept turning up the same thing: inside a town, lease barely moves psf until you get to the very long leases — and location beats lease almost every time.

Bukit Batok’s flat-spot. Four-room psf by remaining-lease band ran $499 under 60 years, $505 at 60–69 (up just 1.2%), then $579 at 70–79 (+16%) and $684 at 90-plus. Read that again: the market paid a 1.2% premium for the difference between a 58-year lease and a 66-year one. Eight years of lease, priced at almost nothing.

Choa Chu Kang is starker. Four-room went $487 at 60–69 to $503 at 70–79 — 3.3% for a full decade of lease. And the five-room number actually went backwards, down 0.4% across those same ten lease years on 244 transactions, before jumping 19.0% once you crossed 80 years.

Central Area is the limit case. It is three-room stock almost exclusively, and there lease is worth nothing against address. Chinatown and Tanjong Pagar traded at $817 psf on an average 49.3 years remaining, while Little India and Jalan Besar traded at $718 psf on 56.3 years — 13.9% more psf for seven fewer years of lease. The sharpest single pair: Klang Lane, 74.3 years left, $721 psf; Tanjong Pagar Plaza, 49.3 years left, $863 psf. Twenty-five extra lease years, sold at a 16% discount.

Where does that leave the table above? Exactly here: the towns to watch are not the ones with high shares already — those flats have taken their haircut and are trading on location now. The towns to watch are the ones about to take it: Pasir Ris, Bishan, Bukit Panjang, Tampines, Hougang. That is where the repricing is still ahead of the owner, not behind him.

I should be straight with you about one thing: those psf figures come from our individual town analyses in this series, computed from the same HDB resale dataset. I did not re-derive them for this article. The 26-town table and the projections above, I did compute fresh, and I’d stake my licence on them.

Method, sensitivity, and what I could not compute

I would rather show you the seams than pretend there aren’t any.

Source. Every figure in the 26-town table and the projection table comes from the HDB property register published on data.gov.sg — the same register HDB maintains of every block, its street, its town and its completion year. I filtered to residential blocks only. Third-party portals were used as a sanity check, never as a source.

The lease convention, and why it matters. The register carries year_completed, not lease_commence_date. Those are not the same thing — a lease typically commences slightly before the block is completed, which means true remaining lease is usually a touch shorter than my estimate. So I’ve used a zero-year offset throughout, which makes every figure here a conservative floor. The real numbers are the same or worse, never better.

I tested the sensitivity, and you should know how big it is:

The pattern is clean and it is itself a finding: towns built in a burst are hypersensitive to the convention; towns built gradually are robust to it. Bukit Panjang’s published 8.4% would be 43.6% on a two-year offset, because its entire 1987–88 cohort sits right on the boundary. Bukit Batok barely twitches. When you read any lease-decay statistic anywhere — mine included — ask which kind of town it is describing.

What I could not compute, and why. Publishing the gaps honestly:

• Fourteen towns have no forward projection. I projected 12 and stopped. The remaining 14 are mostly towns already above 65% (Marine Parade, Central Area, Ang Mo Kio, Clementi, Jurong East, Bedok, Kallang/Whampoa, Geylang, Bukit Merah, Queenstown), where the projection is close to “it already happened,” plus the three zero-percent towns whose first crossings I have given exactly (2037, 2037, 2040) and Bukit Timah, whose 27-block sample is too small to project meaningfully. I would rather publish 12 done properly than 26 done loosely.
• Flat counts, not block counts. Everything here is blocks. A block-weighted share is not a flat-weighted share, and in towns with mixed block heights the two will differ.
• Tengah is excluded entirely — 139 blocks, no resale market until the 2028 MOP wave.
• The psf-by-lease-band figures were not re-derived in this run, as noted above.

One null result worth recording: I found no town where the under-60 share declines. It can’t — blocks only get older, and new BTO supply enters at 99 years and dilutes only slowly. Every town in Singapore is monotonically increasing on this measure. The only question is the slope.

Who should worry now — and who genuinely shouldn’t

I get asked “should I be worried” roughly once a day. The honest answer is that most people shouldn’t, and a specific minority really should.

That second-last row is the trade almost nobody makes deliberately, and it is a good one. If you are 58 and downsizing, a 55-year lease is not a defect — it is a discount funded by buyers who are not allowed to compete with you. Work out what that’s worth with the calculators at listings.sg/tools.

If you want the wider context around these numbers, three pieces pair well with this one. My HDB resale prices 2026 analysis explains why the seven-year appreciation run ended this year and what the BTO supply wave is doing to resale pricing — the demand-side backdrop to everything above. Lease decay in condos runs the same question on the private side, where the mechanics are similar but the buyer pool and the en-bloc option are completely different. And Hougang HDB resale 2026 is worth a look precisely because Hougang crosses 50% in 2028 while simultaneously setting million-dollar records — the clearest live proof that location and lease are separate markets.

Town-level deep dives on Queenstown, Tampines, Marine Parade, Clementi, Bukit Batok and Jurong West are drafted and will publish over the coming weeks, along with a piece on why an HDB flat sits unsold — which, nine times out of ten, has nothing to do with its lease.

FAQ

Disclaimer. This article is general information and market commentary only. It is not financial, legal or tax advice, and it is not an offer or recommendation to buy or sell any property. All lease, block and completion-year figures are derived from the HDB property register published on data.gov.sg as at 16 September 2026 and use completion year as a proxy for lease commencement; actual remaining lease for any specific flat must be confirmed against HDB’s own records and the title. CPF and loan illustrations are simplified and indicative — your actual CPF usage, loan quantum and cash outlay depend on your age, CPF balances, existing loans, the flat’s valuation and prevailing rules at the time of purchase. Verify all financing figures with HDB, CPF Board and your bank before committing.

Gary Lim · CEA Salesperson Registration No. R009877B · ERA Realty Network Pte Ltd · CEA Licence No. L3002382K. Regulated by the Council for Estate Agencies (CEA). Past transactions and market trends are not indicative of future results.