Singapore home prices fell over two years in about one window in three since 2000

Official price indices say waiting for a fall has usually been a losing bet, but not always. Since 2000, about one in three two-year periods ended with lower prices, for private homes and HDB resale flats alike 12. Large falls were rare, and nobody could see the bottom while it was happening.

How often have Singapore home prices actually fallen?

The cleanest way to test "wait for a fall" is to take every quarter in the official record, look two years ahead, and count how often the price index was lower at the end. Two indices make this possible. URA's private residential property price index runs from the first quarter of 1975 1. HDB's resale price index runs from the first quarter of 1990 2. Both are set to 100 in the first quarter of 2009 12.

URA compiles its index from caveats lodged at the option stage, stamp duty data and developers' sales figures 1. HDB calculates its index from resale transactions registered across towns, flat types and models 6. Each measures how the market as a whole moved, not what one home is worth.

Here is the count, using every two-year window in each index up to the second quarter of 2026.

Source: URA private residential property price index (All Residential) 1 and HDB resale price index 2, data.gov.sg; Sightlines analysis. A window compares one quarter with the same quarter two years later.

So in most two-year stretches, a buyer who waited paid more at the end. The median private window rose 6.9% and the median HDB window rose 8.9% since 2000 12. But "most" is not "always". Roughly a third of the time, waiting two years would have meant a lower index at the end 12.

A one-year wait shows much the same picture. Since 2000, the private index was lower after four quarters in 32.4% of windows, with a median change of 3.0%; the HDB index was lower in 34.3%, with a median change of 3.2% 12. The largest one-year falls were 24.9% for private homes and 9.1% for HDB resale flats 12.

The falls were usually small. Since 2000, only 6 of the 98 two-year private windows fell by more than 10%, and only 2 of the 98 HDB windows did 12. Among the private windows that ended lower, the median fall was 6.2%; among the HDB ones, 2.4% 12.

How big were the big falls, and how long did they take to recover?

The averages hide a few severe episodes. Taking every fall of 10% or more from a peak in the private index, the record shows five 1.

Source: URA private residential property price index, All Residential 1; Sightlines analysis. The 1983 peak of 37.5 was reached in both the third and fourth quarters of 1983.

The HDB resale index has had three falls of 10% or more since 1990 2.

Source: HDB resale price index 2; Sightlines analysis.

Two patterns matter for anyone waiting. Falls of this size came roughly once a decade, not every few years 12. And they did not follow a script. The 2008 to 2009 fall in private prices took only four quarters and was erased within two years of the peak 1. The 2013 to 2019 slide in HDB prices was gentle, at 12.4%, but stretched over 24 quarters 2.

What does waiting cost a household that rents?

A household that waits does not stand still. If it rents, it pays rent that buys no equity. A simple test is to ask how far prices would have to fall to cover the rent paid while waiting. The worked example below uses official HDB data.

Take a family renting a four-room HDB flat while it decides whether to buy one. Over the 12 months to September 2026, the median monthly rent approved by HDB for a four-room flat was $3,300 3. The median resale price of a four-room flat registered in July to September 2026 was $628,000 4.

• Two years of rent at the median: 24 x $3,300 = $79,200 3.
• That rent as a share of the median four-room price: $79,200 / $628,000 = 12.6% 34.

In other words, for waiting two years to pay off on rent alone, four-room prices would need to fall by about 12.6% over those two years 34. Since 2000, the HDB resale index has fallen by more than 12.6% over two years in 2 of 98 windows, both starting in the first half of 2000 2.

The same test works for other flat sizes.

Source: HDB renting out of flats 3 and HDB resale flat prices 4, data.gov.sg; Sightlines analysis.

Since 2000, no two-year window in the HDB index has fallen by more than 15% 2. Two windows, both starting in early 2000, fell by more than 11.3% 2.

The honest caveats, before relying on this:

1. HDB says its rental data are owner-declared and indicative only, and that it does not verify them 3.
2. The example counts only rent. An owner also pays loan interest, property tax and maintenance, and a renter's savings can earn interest while they wait. Counting those would lower the fall needed to make waiting pay, so 12.6% is an upper bound for this family, not an exact hurdle 34.
3. An index tracks the market as a whole. A particular flat, town or project can move differently.

Where are the two indices now?

The two markets are not moving together in 2026. URA's private index stood at 219.4 in the second quarter of 2026, the highest value in the series since 1975 1. URA's flash estimate for the third quarter put it up a further 1.4% 5.

HDB's resale index has eased for three quarters. It peaked at 203.7 in the third quarter of 2025, then slipped to 202.8 in the second quarter of 2026 and to a flash estimate of 202.4 in the third quarter 26. That is 0.6% below the peak, a small move next to the 12.4% to 28.1% falls in the table above 26.

A gentle drift is not the start of a large fall, and it is not proof that one will not come. The 2013 to 2019 HDB slide also began with a small step: the index slipped from 149.4 to 148.1 in the third quarter of 2013, a dip of 0.9%, before falling 12.4% in all over six years 2. This time, the index rose in each of the first three quarters of 2025 and was flat in the fourth 6. The data show what happened next in past episodes, not what will happen this time.

Can anyone time the bottom?

The episodes above suggest not. The trough is only visible after prices have started to rise again, and the time from peak to trough has ranged from 4 to 24 quarters 12.

The 2008 to 2009 private fall is a clear example. A buyer who waited from the second quarter of 2008, when the index was at 126.9, would have seen prices fall 24.9% within a year 1. A buyer who waited two years instead found the index at 131.7 in the second quarter of 2010, 3.8% above where it started 1. The fall was real, but the window to act on it was short.

The 2013 to 2017 private fall shows the opposite pattern. Prices slid for 15 quarters by a total of 11.6%, and did not regain their 2013 peak until the fourth quarter of 2020 1. A buyer waiting for a "proper" crash in that period would have waited years for a fall that never became large.

Who this does not apply to

Buyers of new HDB flats. HDB's resale index is built from resale transactions 6. It says nothing directly about the price of a new flat bought from HDB.

Households selling one home to buy another in the same market. For an HDB upgrader selling a flat to buy another flat, a fall lowers the sale price as well as the purchase price. Waiting changes the gap between the two far less than it changes either price.

Buyers with a fixed date. A family that must move by a school year or a lease end cannot wait for a fall that may take years. For them, the question is what to buy, not when.

Buyers who cannot yet afford the purchase. If waiting is about saving a larger down payment, the cost of waiting is the price drift against the extra savings, and the arithmetic differs from the renter's example above.

What would change the answer?

The decision rule the data support is simple enough to reuse. Work out what waiting costs you each year, whether in rent or in other ways, as a share of the price. Then compare that share with how often prices have fallen by that much over the same period. If the fall you need has rarely happened, waiting is a bet on an unusual event.

Three things could change that answer for a particular household.

• Higher borrowing costs. URA said in October 2026 that the macroeconomic outlook is highly uncertain, including for interest rates 5. Higher rates could slow price growth, though the data in this guide cannot say whether, or by how much.
• A longer horizon. Over five years, the record has the same pattern: private prices were lower at the end in 29.0% of five-year windows since 1975, and HDB prices in 27.0% since 1990 12. Over a long stay in one home, the timing of the purchase tends to matter less than the price paid and the home chosen.
• A cheaper way to wait. A household that lives with family rent-free while it saves faces a much smaller cost of waiting than the renter in the example, and the case for patience is stronger.

The counter-case should be stated plainly. Waiting is sometimes right. In about one two-year window in three since 2000, prices were lower at the end 12, and in a few episodes the fall was large. Doing nothing is a valid choice when the home is not needed yet. What the record does not support is the belief that a large fall is due, or that it will be visible in time to act.

This is general information, not financial advice. Check the current rules with IRAS, HDB, CPF Board or your bank before you commit.