GUIDES & INSIGHTS · FRAMEWORK 05

The Yield Play. Computed on a number you never paid.

Buy for income rather than appreciation, select for what rents reliably rather than what shows well, and hold something that carries itself. It is a sound framework and there is nothing naive about it. The problem is the arithmetic everybody accepts without checking: the yield you are quoted is rent over the price, and the price is not what left your account.

Farhan Adenan · 30 Aug 2026 · 10 min read

Duty rates and lending limits are as published by IRAS and MAS, current as at 30 August 2026. Worked examples are POV arithmetic on those rates for a Singapore citizen buying a second residential property, and are gross of maintenance, sinking fund, property tax, agency fees and vacancy. Rent figures are illustrative and are not a forecast.

You buy for income rather than appreciation, and you select for what rents reliably rather than what shows well at a viewing. A smaller, duller position that pays beats a larger one that does not. That is the entire framework, and there is nothing naive about it — a property that covers its own outgoings is a position you can hold through a bad decade, which is the thing that actually determines outcomes in this market.

It suits you if you want the asset to carry itself rather than to be carried, if you would rather have a modest reliable number than a large uncertain one, and if you are prepared to buy something unglamorous. It does not suit you if the plan depends on the capital value doing the work, because then you are running a growth position and calling it a yield one.

The framework’s weakness is not the idea. It is the arithmetic everybody accepts without checking.

Yield is a ratio, and almost every yield you will be quoted uses the purchase price as its denominator. That is the right number for comparing buildings. It is the wrong number for deciding whether to commit your money, because the price is not what you committed.

If this is a second residential property, ABSD sits in that denominator whether you put it there or not, and so does Buyer’s Stamp Duty. For a Singapore citizen that is 20% plus roughly 3% — nearly a quarter added to the capital committed, none of it recoverable from rent, none of it appearing in the yield on the brochure. For a permanent resident on a second property it is 30%; for a foreign buyer, 60% on any purchase.

So a yield that looked adequate against the sticker price is materially thinner against the sum that left your account. 3.60% becomes 2.93%. That is not a rounding difference — it is roughly a fifth of the return, and it is the difference between a position that clears its financing and one that does not.

One line in the outgoings deserves separating out, because it is the one people forget entirely. Property tax in Singapore is charged at different rates depending on whether the property is owner-occupied, and a let property is not. The non-owner-occupier schedule is materially higher, it is charged on annual value rather than on rent, and it rises as annual value rises — so a property that is doing well on rent is also, by construction, being taxed more heavily. Check the current schedule against the property’s annual value before you model anything; it is published and it is specific to the unit.

And that is still gross. Maintenance, sinking fund contributions, property tax at the higher non-owner-occupier rates, agency fees on each new tenancy, insurance, and the months between tenants all come off it. We are not going to put a single number on those, because they vary enormously by building and by how the property is run — but a net figure is always meaningfully below the gross one, and any pitch that never mentions the distinction is not being straight with you. Our map of where yields are actually real works from measured rents rather than asking prices.

TDSR sizes your loan at a 4.0% medium-term floor rather than at the rate you are actually offered. Buyers routinely take comfort from this: the bank has checked I could pay 4%, so I am covered. That is a misreading. The floor tests whether the bank should lend you the money. It does nothing whatsoever to stop your actual rate rising to 4% or beyond.

Which matters more here than in any other framework on this list, because a yield play is a spread business. You are earning rent and paying interest, and your return is what is left. A growth position can absorb a rate rise by waiting. A yield position cannot — the spread compresses immediately, and if it inverts you are funding someone else’s housing out of your own income while calling it an investment.

So the useful test is the opposite of the bank’s. Do not ask whether you could service the loan at 4%. Ask whether the position still pays at 4%, with a month or two of vacancy in the year, after the outgoings. If the answer is no, the framework has not been implemented — something else has been, wearing its name.

Rates and limits as published by IRAS and MAS, current as at 30 August 2026. Dollar figures are arithmetic on those rates for a Singapore citizen buyer.

The first two are the framework’s real problem, and they compound. A 3.60% headline recomputed on committed capital and then net of outgoings can land somewhere a long way below the number that made the property look interesting. That is not an argument against yield investing. It is an argument for doing the division yourself before you fall in love with a listing.

1. Recompute every quoted yield on committed capital. Price plus ABSD plus Buyer’s Stamp Duty. It takes thirty seconds and it is the single most useful thing on this page.

2. Then take it net, with your own assumptions. Maintenance, sinking fund, property tax at the non-owner-occupier rates, agency fees amortised over the expected tenancy, and a realistic void allowance. Write the assumptions down so you can argue with them later.

3. Test the position at 4%, not at the rate you are quoted. If it only works at today’s rate, it does not work — it is waiting.

4. Underwrite a void. One month, minimum, every year. Two if the unit is unusual, large, or in a building with a lot of identical stock coming to market at once.

5. Select for the tenant, not for yourself. Proximity to employment, a station, and a school catchment lets more reliably than a view or a finish. The framework rewards dull and well-located over striking and awkward.

6. Read the tenant pool before the building. A yield is a claim about who will want to live there and for how long. Proximity to a large employer, a station and a school catchment produces longer tenancies and shorter voids; a striking unit in a thin submarket produces the opposite. The framework rewards the boring answer here more consistently than anywhere else in these six.

7. Check whether the yield is already priced in. A high advertised yield is sometimes a bargain and is more often the market telling you something about the lease, the building, or the tenant pool. Ask which it is before you assume the first.

If you want the income

If you want the asset to carry itself, do the division before the viewing

Recompute the quoted yield on price plus ABSD plus Buyer’s Stamp Duty, take it net of maintenance, sinking fund, property tax and a void month, and test the whole thing at a 4% rate rather than the one you are offered.

If it still clears, you have found something real and you should move on it — positions that genuinely pay are scarcer than the advertising suggests. If it does not clear, no amount of appreciation later makes a negative spread a good idea now.

If you want growth

If the plan quietly depends on the price rising, this is not a yield play

That is a growth position, and it should be sized, financed and timed as one. Calling it income does not make the cash flow work, and it removes the one discipline this framework actually gives you.

There is nothing wrong with buying for growth. But the Quantum Ladder is a more honest structure for it, and it does not ask you to defend a yield figure that was computed on a number you never paid.

MARKET & POLICIES · INSIGHT 005

Where the yields are actually real

The projects where rental income covers the position — measured, not advertised.

VALUE FINDS · ISSUE 003

The 4% club: 99 projects where rental yields are actually real

Where the income case survives contact with the entry price.

Dataset — Duty rates and lending limits as published by IRAS and MAS, current as at 30 August 2026. Rent figures are illustrative; dollar figures are POV arithmetic.