Who can actually afford a flat in Singapore?

A median HDB resale flat cost $628,000 in 2025. For the middle tenth of households that is 4.1 years of gross income; for the top tenth 1.5, and for the bottom tenth 19.5. The same flat, a 13-fold spread — which is why a single national affordability figure answers almost nobody’s question.

One flat, ten answers

Every row below is the same property — the median HDB resale flat of 2025, at $628,000. Only the buyer changes. Income is the average within each decile of resident employed households, and it is employment income including employer CPF contributions, which is the same concept as the median figure this site publishes beside the price index.

Average Monthly Household Employment Income (Including Employer CPF Contributions) Among Resident Employed Households by Deciles, 2000 - 2025. SingStat table 17880, last updated 09/02/2026. Prices are HDB resale medians for calendar 2025, the latest complete year with both figures published.

Every rung against every decile

The same arithmetic across the whole of the housing ladder: years of gross income for each kind of home, for each decile. Cells are years; the darker the cell, the more years. Everything here is a median of resales over one twelve-month window, on one income table.

Columns are income deciles, 1st (lowest) to 10th (highest). Rows are the ladder rungs, cheapest first. A figure of 100 or more is rounded to whole years, because the difference between 143 and 144 years of income is not a distinction worth drawing.

The private market is the top decile’s HDB

The pattern in that table worth saying out loud: Condominium costs the top tenth 4.3 years of gross income, and HDB 4-room costs the middle tenth 4.2. Those are the same burden. So private property is not a different order of expense from public housing — it is the same order of expense, for a household earning about 2.7 times as much.

Which recasts the 70 per cent wall between the top of HDB and the bottom of the private market that the ladder measures. In price it is a step. In years of income it is not a step at all for a household at the top, and it is most of a decade for one in the middle.

Has it got worse?

For the middle decile, a median resale flat cost 3.7 years of gross income in 2017 and 4.1 years in 2025. Price went from $410,000 to $628,000; that decile’s average income went from $9,348 to $12,635 a month. Both rose, and the ratio is the question of which rose faster — prices did, by enough to move the ratio against buyers.

One caveat SingStat states itself, and it bites hardest here. The deciles are not a panel. A household moves between them from year to year — a spell of unemployment drops it, resumed work lifts it back — so comparing the 5th decile in 2017 with the 5th decile in 2025 is not comparing the same households. It is comparing the middle of the distribution at two dates, which is a real thing to measure but not the same as asking what happened to anyone.

What this is not

It is not a mortgage. Nobody buys a flat out of gross income. The employer CPF contribution included in these figures cannot be spent on anything except housing, medical and retirement accounts; income tax and living costs come out of what remains; and the real constraint on a purchase is the monthly repayment against a loan limit, not a multiple of annual income. Years-of-income is useful for comparing one decile or one year with another on a consistent basis, and for very little else. For what a purchase actually demands month to month, use the cost calculator.

The bottom decile’s 19.5 years is the least meaningful figure on this page, and the reason is in SingStat’s own note on the table. In 2025, of households in the lowest decile, 10.0 per cent lived in private property, 19.3 per cent employed a domestic worker, 15.3 per cent owned a car, and 38.2 per cent had a household reference person aged 65 or over. That decile is substantially households with assets and little current employment income — retirees, mostly — not households a pay rise away from a flat. Dividing a flat price by their employment income produces a number, not a finding.

And it is not a statement about any individual household. A decile average is an average: half the households in the middle decile earn less than it, and household size varies across deciles, so the same income supports different numbers of people.

Nor is the price side a single number in practice. Everything above uses the national median resale flat, and what that costs depends on where it is: the busiest towns by resales are Tampines, Woodlands, Sengkang, Punggol, Yishun and Jurong West, each with its own medians by flat type. Dividing a town median by a decile income is the same arithmetic as above, with a smaller and more relevant numerator.

Why the income table matters more than the arithmetic

SingStat publishes eighteen household-income series that could all be called “income by decile”, and they disagree. Three choices move the answer: the concept (employment income, or market income including rent and dividends), the universe (all resident households, or only those with someone in work), and the statistic (a median, or an average).

This page uses table 17880 throughout, because it is the only decile series that matches both the concept and the universe of the median figure already published beside the price index — so the figures here and the price-to-income ratio on the growth page are two views of one quantity rather than two different quantities. Had it used the obvious alternative instead, 2025 household income would read $16,159 rather than $12,027 — 34 per cent higher, moving the ratio from 4.35 years to 3.24 — because that table counts market income across all resident households including those with nobody in work.

One difference remains and cannot be removed: these are averages within each decile, while the published national figure is a median. That is checkable rather than assumed. The median household sits inside the 5th decile, so the 5th decile’s average must land near the published median — and it does, within 1.6 to 6.5 per cent across the nine overlapping years. The pipeline fails the build if it ever drifts past ten per cent, which is the case where publishing the two together would be wrong. Full method.

Questions people ask

How many years of income does an HDB flat cost in Singapore?

For the middle tenth of resident employed households it is 4.1 years of gross income: a median resale flat at $628,000 in 2025 against an average $12,635 a month. But a single national figure hides most of the question. The top tenth of households needs 1.5 years for the same flat and the bottom tenth needs 19.5 — a 13-fold spread for an identical property. Income here is employment income including employer CPF contributions, among resident employed households, averaged within each decile.

Is housing in Singapore getting less affordable?

On this measure, yes, slightly. For the middle decile a median resale flat cost 3.7 years of gross income in 2017 and 4.1 years in 2025, as the price went from $410,000 to $628,000 and that decile's average income from $9,348 to $12,635 a month. Both rose; the ratio is the question of which rose faster. Note that this is the ratio for one decile — the answer differs by decile, and the deciles are not the same households from year to year.

Which income decile can afford which type of home in Singapore?

The table on this page crosses every rung of the housing ladder with every income decile. The pattern worth knowing: the private market costs the top decile roughly what public housing costs the middle. Condominium takes the top tenth 4.3 years of gross income, and HDB 4-room takes the middle tenth 4.2. So "private property" is not a different order of expense from an HDB flat — it is the same order of expense, for a household with roughly three times the income.

Why do affordability figures for Singapore differ so much?

Mostly because there are eighteen household-income tables and they measure different things. The three choices that move the answer are the income concept (employment income, or market income including rent and dividends), the household universe (all resident households, or only those with someone in work), and the statistic (a median, or an average). Swapping one table for another moves 2025's household income from $12,027 to $16,159 — 34 per cent — and the price-to-income ratio from 4.35 years to 3.24. This page uses Average Monthly Household Employment Income (Including Employer CPF Contributions) Among Resident Employed Households by Deciles throughout, SingStat table 17880, and nothing on it is divided by a figure from a different table.

Does the lowest income decile really need 19 years of income for a flat?

The arithmetic says 19.5 years, and it is the least meaningful figure on this page. SingStat notes in the table itself that the lowest decile is not simply the poorest households: in 2025, 10.0 per cent of them lived in private property, 19.3 per cent employed a domestic worker, 15.3 per cent owned a car, and 38.2 per cent had a household reference person aged 65 or over. It is substantially households with assets and little current employment income, for whom years-of-income is the wrong measure — not households one pay rise from a flat.

Is years of income the right way to measure affordability?

It is a comparison, not a mortgage. Nobody buys a flat out of gross income: the employer CPF share included in these figures cannot be spent on anything but housing, medical and retirement accounts, income tax and living costs come out of the rest, and the actual constraint on a purchase is the monthly repayment against a loan limit. What the ratio is good for is comparing one year with another and one decile with another on a consistent basis. For what a purchase actually demands month to month, use the cost calculator.

Citing this. Income is SINGAPORE DEPARTMENT OF STATISTICS, table 17880, last updated 09/02/2026 and read Oct 2026. Prices are HDB resale transactions to Oct 2026 and URA caveats, recomputed as both refresh — so quote the date you read them. Permanent link: https://www.dailystudysg.com/affordability.