Sentosa Cove, in one number. Developers charged up to $3,252 per square foot. The market settled at $1,679.

Sentosa Cove is a waterfront enclave of about 2,500 homes on the eastern end of Sentosa, built on reclaimed land and sold mostly to foreign buyers between 2004 and 2011. It is the only place in Singapore where a foreigner can buy a landed home.

Almost everyone who bought at the launches has lost money.

We have spent five notes working out why. This is what they add up to, and it comes down to one number that was public at the time.

The one number that mattered

Since 2012, the year after the last new launch, the middle price for a resale apartment in Sentosa Cove has been $1,679 per square foot. That figure has barely moved in fourteen years. It is what the market has been willing to pay, consistently, for a long time.

Now look at what the developers charged when the buildings were new.

• Launch price
• What the resale market settled at

Until 2006 the developers were charging less than that level. From 2007 they were charging far more, and they never came back down.

A buyer in 2009 paid $3,252 per square foot for a home the market has since decided is worth about half that.

Read those launch figures as an upper bound rather than an exact measure. Before 25 May 2015 a new sale price was recorded as the price on the option, before any discount or rebate the developer might have given. Resale prices are not affected, because those are sales between individuals.

The full analysis.

Everyone who bought new after 2006 lost money

Not most of them. All of them.

Bought new in	Share who made money on resale
2006	67.4%
2007	0%
2008	0%
2009	0%
2010	0%
2011	0%

Five consecutive years, forty resales, no exceptions.

We looked for something the losing buyers did wrong, and did not find one. The size of the apartment does not explain it: within a single project, larger units did better in three of seven cases. How long they held does not explain it. Which building they chose does matter, but every building's buyers from those years still lost.

What explains it is the price on the day they signed, and that was set before they walked in.

Which building you chose mattered too

Outcomes differ enormously between projects, and the pattern is stable.

Project	Median resale outcome
The Berth by the Cove	+5.1%
The Oceanfront	-3.9%
The Azure	-9.5%
The Coast	-10.0%
Seascape	-30.1%
Marina Collection	-32.4%
Turquoise	-41.7%

Forty-six points between the best and the worst, in the same enclave, with buyers who bought in the same years and held for about the same time.

The three at the bottom are the three that launched at $2,600 per square foot and above. The four at the top launched below $1,700.

Marina Collection's figures carry a caveat the others do not. In 2022 the High Court found its developer had misled a lender about the prices of 38 units, having given undisclosed rebates of 25% to 34%. Where a buyer received one, the loss we show is larger than the loss they took. The detail is in the project note.

The money did not disappear, it changed hands

There were three ways to buy an apartment in Sentosa Cove.

New from the developer. Resale, from someone who already owned it. Or sub-sale, which means buying from someone who had bought off-plan and was selling before the building was finished.

That third route was not a sideshow. In 2007 and 2009 it was two thirds to three quarters of everything traded in the enclave. People were buying unbuilt apartments from each other.

For the people doing the flipping it worked. Of the flips we can follow from purchase to sale, 93% made money, at a median gain of 28% in under two years.

For the people they sold to, it did not. Those buyers paid the higher price, waited for the building to finish, and eventually sold it themselves. Only 20% of them made money.

One apartment at The Azure shows it in a single line: sold by the developer in 2005 for $1,632,000, resold three times within twenty-four months to reach $3,850,000, and then sold in 2021 for $2,500,000. The first three sellers took $2.2 million out between them. The fourth owner held for fourteen years and lost $1,350,000.

The full analysis.

Rent held up, and it is not enough

Sentosa Cove rents well. Size for size, an apartment there rents for more than one in prime central Singapore, from 2.5% more in the smallest sizes to 21% more for large waterfront homes.

And because prices fell while rents did not, it now yields more than prime central Singapore does, by roughly a quarter to a half depending on the size of the apartment. That sounds like a recommendation and is not. Yield is rent divided by price. Sentosa's is high because the price collapsed.

For someone who bought at a launch and rented the apartment out, sixteen years of rent covers something over half of what they lost on the price, once the costs of letting a home are taken out. An owner from 2007 who is still holding would need close to another decade of rent to break even, having already held nineteen years. Both of those depend on what letting actually costs, which the records do not show, so treat them as estimates rather than measurements.

The full analysis.

The buyers it was built for have gone

Sentosa Cove was designed and priced for foreign buyers.

In 2023 they were 20.9% of purchases there. By 2025 they were 5.8%.

The stamp duty on foreign buyers rose five times between 2011 and 2023, reaching 60%. That last increase removed about two thirds of remaining foreign buyers from every part of the Singapore market, not just Sentosa. The shock was national and even.

What was not even was the exposure. Sentosa lost 15.1 percentage points of its buyer base. Prime central Singapore lost 4.8. The country as a whole lost 2.3.

A market that loses five points of its buyers sells fewer homes. A market that loses fifteen has to drop its price. Prime central held its prices and sold more. Sentosa's prices fell 21% in a year.

What this series does not answer

• Why the buildings differ so much from each other. We can show that The Berth by the Cove and Turquoise are forty-six points apart and that the gap is stable across seven years of buyers. Transaction records do not say whether it is the marina berths, the sea frontage, the developer, the maintenance, or something else.
• Why the developers priced where they did, or why buyers paid it. The records carry prices, not reasoning.
• Whether any of this repeats elsewhere. These are launches from 2004 to 2011 in one enclave with a buyer base later narrowed by policy. Nothing here says a project launching at a high price today will do the same.
• Much about the landed homes. Roughly 392 landed homes sit inside Sentosa Cove, almost all detached. They are covered only in the landed note, on a population of 54 resales.
• What one particular home did. Everything here is a middle value across a group. Floor, facing, berth access and condition matter for a specific apartment in ways a median cannot see.
• What a declared price includes. Every price here is as declared to the authorities, and in one project a court has found that declared prices misstated what buyers actually paid.

The five notes

• Which building you bought. Outcomes from +5.1% to -41.7%, and the same ordering in every year.
• Landed against apartments. Landed homes lose less often and lose more money.
• What the launches cost against what the market pays. Every new buyer from 2007 to 2011 sold below what they paid.
• What separated the buyers who did well. The price they paid, and the flippers who took the gain.
• What the rent did while they held. A yield above prime central, and a loss it does not close.

Method

Figures are transacted prices from URA caveat data and rental contract records for the Sentosa Cove non-landed projects, excluding collective sales. The Residences at W Singapore is excluded throughout: eighty-three of its sales fall in a single year, 2024, which does not describe a market.

Every figure in this note comes from one of the five articles it summarises, and each is recomputable from the underlying records. Where a figure appears here and in an earlier note, it is the same figure rather than a new calculation. The single Azure apartment traced from 2005 to 2021 is an illustration drawn from the entry-price note, not evidence.

Prices are as declared to the authorities. New sale prices before 25 May 2015 record the price on the option, before any discount or rebate; URA has published net new-sale prices only since that date, and every launch here predates it.

Sentosa Cove is part of the Core Central Region under URA's definition, so prime central Singapore means the rest of that region.

Sale data to 4 August 2026. Rental data to June 2026.