One couple, two properties

Most BTO owners reaching their Minimum Occupation Period are presented with two choices.

• Stay in it.
• Sell and buy bigger.

There's a 3rd option, and one that is rarely given much thought. Because by the time you run it through an advisory, you're shut down because of punitive taxes involved.

But there is still a way to restructure your household's property portfolio so it holds two properties instead of one. One you live in. One that earns income for you, from the day you collect the keys.

This is what that looked like for one family. Names changed, figures exactly as they were.

IMAGE 1 · REPLACE-IMG-WORKSESSION
Real client working session. Table, printed CPF statements, calculator, your handwriting on the sheet. Hands and paper are enough, faces optional. 1600×900.

The before.

Syaza is 35 and earns $10,000 a month. Fadly is 39, income $7,100.

Their first matrimonial home: a 4-room BTO in Tampines with just $43,000 left on the loan.

By any normal reading they were doing fine. One flat, nearly paid off, two incomes. The obvious next move was to sell and buy a bigger place.

For context, here's what their finances were like.

So what did they actually have to spend?

Not just the $338,651 cash. The CPF refunds, $374,200 between them, are their own money coming back and it can go straight into the next purchase. On top of that they already had $157,800 sitting in their Ordinary Accounts.

Cash plus CPF, they were working with $870,651.

IMAGE 2 · REPLACE-IMG-CPFSHEET
Close crop of a real CPF statement, redacted, beside a handwritten deployable-capital tally. Office desk, natural light. 1600×900.

What most people do next.

Take cash in hand, combine it with their total CPF OA, plan for a condo upgrade, move in.

One property. One mortgage. Two names on it.

What they did instead.

Fadly bought Sol Acres, an ex-Executive Condo. 2-bedder, 57 sqm (614 sqft), $885,000, in his name alone. Rented out at $3,300 a month.

Syaza bought Palette. 99 sqm (1,066 sqft) 3-bedder, $1,550,000, in her name alone. They live in it.

Two properties. $2,435,000 of real estate ($1,550,000 plus $885,000). Two separate loans, each in one name, without incurring any additional buyer's stamp duty.

IMAGE 3 · REPLACE-IMG-STRUCTUREDIAGRAM
Brand-coloured diagram, navy / gold / cream. One household box splitting into two title boxes: Fadly, 57 sqm, $885,000, tenanted. Syaza, 99 sqm, $1,550,000, owner-occupied. Arrow from the rent line to the shortfall line. No photography needed. 1600×1000.

Putting logic into cents.

They did not put in a cent of their own money beyond the sale proceeds.

Here is where every dollar went.

Fadly: Sol Acres, $885,000.

He had $370,000 of CPF, but only $285,000 of it was liquid. The other $85,000 was invested through CPFIS and he had no intention of touching it.

He used $201,150. That left him $83,850 of liquid OA, plus the $85,000 still invested. This is important and we'll refer back to it later on.

Syaza: Palette, $1,550,000.

She had $247,000 of CPF OA. She used all of it. Upon collecting the keys to their new home, she had emptied out her CPF savings.

This was not a mistake. It is the single most important number in this case study, and we will come back to it.

Why she put in more cash than the rules required.

Syaza is allowed to borrow 75% of $1,550,000, which is $1,162,500.

But her income wasn't strong enough to clear that borrowing amount. Thanks, TDSR.

Total Debt Servicing Ratio caps your total monthly loan repayments at 55% of income, and the bank assesses that at a 4% stress rate, not the interest rate you actually pay. On $10,000 a month over a 29-year tenure, that framework limits her borrowing at $1,131,749.

She was $30,751 short of getting the full loan required.

So they put in an extra $30,800 of cash and brought the loan down to $1,131,700. Her loan-to-value came in at 73% instead of 75%.

They were comfortable doing that because of how the rest of the structure worked, which is the next section.

What it takes to own two condos.

Mathing this out will decide if your 2-condo household adventure will make or break your bank.

When they bought, in 2024, fixed rates were just over 3%. Please never return to such highs.

Fadly pays his mortgage using his CPF OA in full. Zero cash top-ups involved.

His $1,428 monthly contribution does not cover the $3,182 instalment, so the balance draws down the $83,850 of liquid OA he kept. Remember we mentioned this earlier?

At that rate it lasts about four years. After the 2026 refinance dropped his mortgage to $2,670, the drain fell to $1,179 a month and the same balance now lasts closer to six.

And when it runs out, he liquidates the $85,000 sitting in CPFIS and gets roughly another six years. He will sell them when he needs them, and not before.

So the rent he receives, $3,300 from the tenant, is his cash to spend.

Syaza's CPF OA is empty. Every month, she earns $1,564 in her OA. Her mortgage is $4,934. She has to top up $3,370 in cash.

So Fadly's $3,300 rent income goes straight to covering Syaza's shortfall. $3,300 in, $3,370 out. She is $70 a month short, and that $70 comes out of the cash they kept.

That was the design. Not a happy coincidence. It was the reason Fadly took the smaller property and Syaza took the bigger one. To maximise their lopsided CPF versus income ratios.

By 2026, they refinanced. This was when we caught up, and it inspired me to write this article and prepare a full marketing campaign around their 2-property approach.

Rates had fallen to around 1.45% to 1.65%.

Fadly's rent of $3,300 now covers Syaza's $2,360, with $940 a month left over. The household went from $70 short to $940 clear.

IMAGE 4 · REPLACE-IMG-CASHFLOWLOOP
Two-state cashflow diagram, brand colours. Left, 2024: rent $3,300 in, shortfall $3,370 out, $70 short. Right, 2026: rent $3,300 in, shortfall $2,360 out, $940 clear. Identical layout both sides so the change reads instantly. 1600×900.

What we missed out (nearly).

In 2024, they never assumed interest rates would fall. They planned based on 3% as baseline and possible increments in future. They were ready to top up in cash to own 2 properties.

The initial plan was that at the end of the two-year loan lock-in, they would refinance Syaza's loan and stretch the tenure to bring the monthly payment down. Stretching from 29 years to 35 would have taken her instalment from $4,934 to around $4,419. That gives enough breathing room, at the cost of paying interest for longer.

Thankfully by the time the lock-in expired in 2026, rates had dropped far enough that a simple refinance did the job of reducing the mortgage on its own, without extending the tenure at all.

The lesson here is that they were prepared for the reality that interest rates remained elevated.

What made this work.

One: each of them could carry the loans they needed.

Their incomes were lopsided, as most households are. Wife $10,000 and husband $7,100. A 59/41 split.

What mattered is that the split roughly matched the properties. Syaza took the $1,550,000 home, which is 64% of the total value, and she earns 58% of the income. Fadly took the $885,000 unit. Each of them could qualify for their own side on their own payslip.

Let me explain. The same household earning $17,100.

• $14,000 and $3,100 could not have done this. The $3,100 earner would have qualified for a loan of roughly $360,000, which buys around $480,000 of property. There is very little worth holding at that price, and the whole structure collapses.
• $8,500 and $8,500 incomes. Perfectly even, yet still won't work. Neither of them can clear the loan for the $1,550,000 home. It would mean deploying too much cash and CPF to clear the purchase, leaving almost nothing for the second property.

Two: Fadly had deep CPF.

$370,000 in his Ordinary Account. That is what let him buy a property whose mortgage he could carry without touching his salary, freeing the entire rent to support the other side.

Take away either of those and this does not work.

It won't work if.

Let me repeat.

Fadly is the opposite: roughly six years of liquid OA behind his mortgage, and another six behind that in CPFIS.

One of them has twelve years of runway. The other has none. That is survivable in one household because his rent covers her shortfall. It would not be survivable if they had both bought like Syaza.

The one thing to note: that $85,000 is invested, so it will be worth whatever it is worth in six years. It is a reserve, not a guarantee. Which is exactly why the $73,001 in cash stayed in cash.

What they bought, and why it matters.

The restructure was only half the decision. The other half is what property number two should be.

Fadly bought 57 sqm (614 sqft), 2 bed 1 bath, at $885,000 and had a tenant paying $3,300 from the start. That rent is crucial.

Without a tenant from day one, there is nothing covering Syaza's shortfall, and the household is $2,360 a month down instead of $940 a month up.

To help you understand why I got them to buy an older resale condo instead of a brand new one, my team did another research recently.

We looked at what actually happened to 24 condos and ECs bought around a million dollars between 2018 and 2019 and sold in 2024. The median resale returned $436,171 after all expenses. The median new launch returned $173,011.

What we discovered was, it's not about buying old versus buying new. It was about how long you could collect rent versus waiting for construction of a property. On average, resale properties were tenanted for a median of 63 months. The new launches, just 13 months in roughly 5 years.

The full table, every project named, with the methodology and cost assumptions, is here: REPLACE-STUDY-TITLE ↗

IMAGE 5 · REPLACE-IMG-TENANCYCHART
Paired bars in brand colours. Months tenanted, 63 versus 13. Median net return, $436,171 versus $173,011. Label the sample as n=24, 2018 to 2019 purchases sold 2024, URA caveats. 1600×900.

Before you get excited.

This worked for Fadly and Syaza. It is not going to work for everybody.

What decides it is your CPF strength, your current loan balances, what your BTO is worth today, how your incomes are split, and your ages. Five variables, and they all affect the upgrading options you can explore.

Work with someone knowledgeable on this. That part takes a sitting and about an hour of arithmetic.

Heikal Shafrudin, PropNex Realty, CEA R016820G

Client names and identifying details changed with permission. All figures are from the actual transaction. CPF Ordinary Account inflows reflect the Ordinary Wage ceiling and age-band allocation rates in force in each year shown. Past transactions are not an indication of future performance. ABSD, BSD, SSD, TDSR and LTV rules are as at August 2026 and subject to change. Transaction data is drawn from URA caveats and HDB resale records. Supply counts and transaction data are factual records. Nothing here is a forecast of prices or market direction.