The Freehold Premium Has to Earn Its Keep. In Most of Singapore's Resale Market, It Hasn't.
Scotts Square sits on Scotts Road in Orchard. It is freehold. It is one of the most recognisable addresses in Singapore private property. Over the past 17 years, 82% of sellers walked away with less than they paid.
That number is not a data anomaly. Across five CCR freehold projects, the pattern holds: most sellers lost money, or barely broke even, over holds of 8 to 17 years. Meanwhile, the 99-year leasehold D'Leedon — in the same CCR region — posted a 96% win rate and 2.89% annualised return over a 10-year hold.
This article tests whether the freehold premium — which can cost an extra $400 to $500 per sqft in OCR, and more in CCR — has actually paid off in Singapore's resale market. The answer depends heavily on where you bought, what you paid, and when you bought it. The deed itself turns out to be the weakest variable of the three.
The callout before the investigation: At Regent Heights in Bukit Batok — a 99-year leasehold — 98% of sellers made money over the past decade, averaging around 4.2% a year. The freehold Hillbrooks a few streets away: also profitable, at 4.9% a year. The gap is smaller than the upfront PSF difference of nearly $470 per sqft suggests it should be.
0.7 percentage points a year: what the freehold label costs you in Bukit Batok
The cleanest near-comparable pair in the data comes from District 23 — Bukit Batok and Hillview — where freehold and 99-year leasehold condos of similar age and size have been transacting over the same approximate holding periods.
| Project | Tenure | Completed | Typical size (sqft) | Typical PSF | Avg annual return | Win rate | Transactions |
|---|---|---|---|---|---|---|---|
| Hillbrooks | Freehold | 1999 | 1,071 | $1,533 | 4.87%/yr | 100% | 16 |
| Hillview Park | Freehold | 1995 | 1,249 | $1,562 | 3.94%/yr | 78% | 9 |
| Natura@Hillview | Freehold | 2015 | 635 | $1,587 | 1.66%/yr | 100% | 19 |
| Hillington Green | 999-year | 2002 | 1,528 | $1,641 | 5.01%/yr | 100% | 17 |
| Regent Heights | 99-year | 1999 | 1,023 | $1,063 | 4.17%/yr | 98% | 48 |
| The Madeira | 99-year | 2003 | 1,249 | $1,212 | 4.41%/yr | 97% | 29 |
| Guilin View | 99-year | 2000 | 1,259 | $1,202 | 3.81%/yr | 100% | 28 |
| Maysprings | 99-year | 1998 | 1,292 | $1,161 | 4.30%/yr | 97% | 36 |
| Kingsford Hillview Peak | 99-year | 2017 | 775 | $1,469 | 1.08%/yr | 93% | 55 |
The gap between Hillbrooks (4.87%/yr) and Regent Heights (4.17%/yr) looks meaningful in a headline. Over 15 years, on a $1.64M freehold purchase versus a $1.09M leasehold purchase, it means you paid about $550,000 more upfront — and earned roughly 0.7 percentage points more per year in annualised return.
Whether that is a good trade depends on what you think the freehold title is worth beyond the investment return. But the data makes one thing plain: the gap is narrow, not dramatic. This is not the story of freehold obviously dominating.
The more disruptive row in that table is Hillington Green. A 999-year leasehold — functionally perpetual — completed in 2002, it delivered 5.01% a year, beating the pure freehold Hillbrooks and every 99-year peer in the same suburb. The lesson from the 1998–2003 vintage cohort is not about tenure. It is about entry price. Every project in that group — freehold, 999-year, and 99-year alike — posted strong returns because buyers in that era paid $600–900 per sqft. That is what drove the profits.
The 2015–2017 cohort tells a different story. Natura@Hillview (freehold, 2015) returned only 1.66% a year. Kingsford Hillview Peak (99-year, 2017) returned 1.08%. Both disappointed. The shared cause: higher entry prices, not different tenures.
The 99-year condos in Bedok returned 5% a year. The freehold neighbour returned less than 1%.
The Bedok D16 cluster sharpens the argument considerably.
| Project | Tenure | Completed | Typical size (sqft) | Typical PSF | Avg annual return | Win rate | Transactions |
|---|---|---|---|---|---|---|---|
| Waterfront Waves | 99-year | 2011 | 1,346 | $1,579 | 5.30%/yr | 96% | 25 |
| Waterfront Key | 99-year | 2012 | 1,281 | $1,725 | 5.12%/yr | 100% | 20 |
| Costa Del Sol | 99-year | 2003 | 1,346 | $1,873 | 4.29%/yr | 100% | 71 |
| Siglap V | Freehold | 2013 | 705 | $1,529 | 0.57%/yr | 72% | 18 |
Waterfront Key (Bedok Reservoir Road, roughly a 15-minute drive from Bedok MRT) and Waterfront Waves — both 99-year leasehold, both completed around 2011–2012 — averaged more than 5% annualised return. Nearly every seller made money. The freehold Siglap V, completed in 2013 in the same broad area, returned 0.57% a year. Nearly three in ten sellers lost money outright.
One qualification matters and should be stated plainly: Siglap V's typical unit is 705 sqft — mostly 1-bedrooms — while Waterfront Key and Waterfront Waves average around 1,300 sqft of family-format units with strong rental demand from the waterfront corridor. The comparison is not perfectly controlled. Product type and size contributed to the gap.
But the headline still stands: freehold status did not protect Siglap V sellers. A smaller unit in a weaker demand profile performed badly regardless of what the title deed said.
Scotts Square is freehold, in Orchard. 82% of sellers still lost money.
Here is the part of the data that inverts everything a buyer thinks they know.
| Project | Tenure | Completed | Typical size (sqft) | Typical PSF | Avg annual return | Win rate | Hold | Transactions |
|---|---|---|---|---|---|---|---|---|
| Scotts Square | Freehold | 2011 | 947 | $3,230 | −1.25%/yr | 18% | ~17 yrs | 17 |
| Espada | Freehold | 2013 | 560 | $2,290 | −0.27%/yr | 13% | ~15 yrs | 16 |
| 26 Newton | Freehold | 2016 | 560 | $2,217 | −0.74%/yr | 21% | ~8 yrs | 14 |
| Belle Vue Residences | Freehold | 2010 | 1,981 | $2,146 | +0.18%/yr | 50% | ~14 yrs | 20 |
| The Laurels | Freehold | 2013 | 883 | $2,793 | +0.47%/yr | 60% | ~9.6 yrs | 15 |
| Marina Bay Suites | 99-year | 2013 | 1,625 | $1,931 | −1.19%/yr | 10% | ~15 yrs | 10 |
| D'Leedon | 99-year | 2014 | 1,281 | $2,053 | +2.89%/yr | 96% | ~10 yrs | 68 |
Scotts Square, Espada, and 26 Newton are all freehold CCR condos. Together, they represent some of the most coveted addresses in the Singapore private market. Over holds of 8 to 17 years, they delivered negative returns for most of their sellers. Scotts Square at $3,230 per sqft and Espada at $2,290 per sqft were simply bought at prices the market never confirmed.
A fair objection applies here and must be named: Scotts Square, Espada, and 26 Newton were all bought at or near the 2007–2011 CCR price peak. Any CCR asset — freehold or leasehold — purchased at that entry point would have faced headwinds. Marina Bay Suites, a 99-year leasehold, has a 10% win rate and −1.19%/yr average return from the same era. CCR timing, not tenure, explains a meaningful portion of these losses.
But timing alone does not explain the gap with D'Leedon. A 99-year leasehold in the same CCR region — District 10, completed 2014, entry prices around $2,053 per sqft — posted a 96% win rate at 2.89% annualised return over roughly 10 years. The difference is not that D'Leedon buyers had better timing by a decade. The difference is that D'Leedon buyers paid less for a well-located leasehold in a recovering corridor, and the market rewarded that discipline. The freehold label on Scotts Square didn't add enough to the resale value to overcome the entry price.
The CCR 99-year story does not prove leasehold beats freehold universally. It proves that location and price discipline outranked tenure in this region during this period.
There are cases where freehold paid off — but they're not the ones most commonly pitched
The data does show freehold working. Just not broadly, and not cheaply.
| Project | Tenure | Completed | Typical size (sqft) | Typical PSF | Avg annual return | Win rate | Transactions |
|---|---|---|---|---|---|---|---|
| Leedon Residence | Freehold | 2015 | 2,131 | $2,790 | 3.03%/yr | 96% | 23 |
| The Trizon | Freehold | 2012 | 2,013 | $2,150 | 3.46%/yr | 95% | 21 |
Leedon Residence and The Trizon are both freehold, both in District 10, and both delivered strong results: 96% and 95% win rates, with annualised returns above 3%. These are large-format units — over 2,000 sqft — in established CCR enclaves. They represent freehold doing what freehold is supposed to do: large, illiquid, long-hold assets in locations with durable scarcity value and underlying estate demand.
But a typical purchase here is $5–6 million or more. And the performance, at 3–3.5% annualised, is below what the best-located 99-year leasehold RCR projects produced in the same period.
Parc Esta — a 99-year leasehold in RCR Geylang, completed 2022, next to Eunos MRT — returned 5.38% a year with a 100% win rate across 236 transactions. That is the largest and most reliable sample in this dataset. A leasehold city-fringe project outperformed freehold in Orchard. The reason is not tenure. It is that buyers entered at about $1.69M on average for a well-connected RCR project when the price still left room for resale gains.
The real variable was always the price you paid
The most consistent finding across every region and tenure in this dataset is this: projects completed between 1998 and 2003 — OCR, regardless of tenure — almost universally delivered 4–5.5% annualised returns. Projects completed between 2013 and 2017 — in all regions and tenures — compressed to 1–3% returns. The tenure composition of each cohort barely shifted the outcome.
| Era | Region/Tenure | Return range |
|---|---|---|
| 1998–2003 OCR (any tenure) | Freehold, 999-yr, 99-yr | 3.8–5.5%/yr |
| 2013–2017 OCR/CCR (any tenure) | Freehold and leasehold | 0.6–2.9%/yr |
| 2011–2014 CCR freehold (peak entry) | Freehold | −1.25% to +0.47%/yr |
| 2014 CCR 99-year (post-peak entry) | 99-year | +2.89%/yr |
The vintage effect is more powerful than the tenure effect. Every generation of buyers in the 2000s benefited from low entry prices — not from what their deed said. Every generation of buyers from 2011 to 2016 entered at elevated prices and found the market had less room to give.
The freehold premium, in that context, is a structural headwind. When you pay $470 more per sqft for a freehold label in Bukit Batok, that extra cost has to be recovered through superior resale appreciation before you earn your first dollar of return above a leasehold alternative. In OCR, the data says it mostly earns back — narrowly. In CCR at peak entry prices, it mostly has not.
The two-sided verdict
The freehold vs leasehold debate is genuinely conditional, not settled in one direction. Here is what the resale record shows clearly, and what it leaves open.
What the data supports: In OCR, freehold produced slightly better annualised returns than 99-year leasehold over long holds from low entry prices — but the gap was small enough (under 1 percentage point) that the higher purchase cost compressed income yield substantially. A 99-year Regent Heights buyer today faces a ~4.2% indicative gross yield at ~$1.09M. A freehold Hillbrooks buyer faces ~2.7% at ~$1.64M. Those are rough directional estimates, not confirmed yields — but the direction is consistent. In CCR, freehold has been a worse bet than well-located 99-year leasehold in most of the last 15 years, driven primarily by peak-era entry prices on small-format units in districts that did not re-rate.
What the data cannot settle: The freehold case for store-of-value — estate planning, en bloc optionality, the 60-year horizon — is not tested here. The resale record measures 8 to 17 year holds. A buyer genuinely intending to hold for 60 years, or who values the land title for succession purposes, is asking a different question than the one this data answers.
The honest summary: The freehold premium is real and persistent. Whether it pays off depends almost entirely on what you paid, where you bought, and whether the corridor re-rated after you entered. The deed is the last variable to check — not the first. Most buyers check it first. The data says that is the wrong order.
Full dataset reference: 15 projects, three regions
| Project | Region | Tenure | Completed | Size (sqft) | PSF | Annual return | Win rate | Transactions |
|---|---|---|---|---|---|---|---|---|
| Scotts Square | CCR D9 | Freehold | 2011 | 947 | $3,230 | −1.25%/yr | 18% | 17 [Freehold loss] |
| Espada | CCR D9 | Freehold | 2013 | 560 | $2,290 | −0.27%/yr | 13% | 16 [Freehold loss] |
| 26 Newton | CCR D11 | Freehold | 2016 | 560 | $2,217 | −0.74%/yr | 21% | 14 [Freehold loss] |
| D'Leedon | CCR D10 | 99-year | 2014 | 1,281 | $2,053 | +2.89%/yr | 96% | 68 [CCR leasehold contrast] |
| Leedon Residence | CCR D10 | Freehold | 2015 | 2,131 | $2,790 | +3.03%/yr | 96% | 23 [Freehold works here] |
| The Trizon | CCR D10 | Freehold | 2012 | 2,013 | $2,150 | +3.46%/yr | 95% | 21 |
| Parc Esta | RCR | 99-year | 2022 | 743 | $2,275 | +5.38%/yr | 100% | 236 [Largest sample] |
| The Interlace | RCR | 99-year | 2013 | 1,593 | $1,683 | +3.03%/yr | 95% | — |
| Hillbrooks | OCR D23 | Freehold | 1999 | 1,071 | $1,533 | +4.87%/yr | 100% | 16 |
| Hillington Green | OCR D23 | 999-year | 2002 | 1,528 | $1,641 | +5.01%/yr | 100% | 17 [999-yr beats freehold] |
| Regent Heights | OCR D23 | 99-year | 1999 | 1,023 | $1,063 | +4.17%/yr | 98% | 48 |
| Maysprings | OCR D23 | 99-year | 1998 | 1,292 | $1,161 | +4.30%/yr | 97% | 36 |
| Waterfront Waves | OCR D16 | 99-year | 2011 | 1,346 | $1,579 | +5.30%/yr | 96% | 25 [Best leasehold return] |
| Waterfront Key | OCR D16 | 99-year | 2012 | 1,281 | $1,725 | +5.12%/yr | 100% | 20 |
| Siglap V | OCR D15 | Freehold | 2013 | 705 | $1,529 | +0.57%/yr | 72% | 18 |
The table above shows 15 projects across CCR, RCR, and OCR. The pattern that runs through it is not freehold versus leasehold. It is entry price and location quality — almost every time. The wins and losses sort cleanly by what buyers paid in their era, not by what the title deed said when they signed.