The OECD publishes a table about statutory pensions that measures something other than the amount they pay: the share of your salary they replace, and how that share moves as the salary grows.
For Thailand the answer is 41.5% for someone on the average wage, and 20.7% for someone on twice the average wage. The rate does not halve because the higher earner contributed less. It halves because their salary outgrew the ceiling their contributions stop at.
Two kinds of promise, wearing one name
The ILO divides ASEAN's statutory schemes in two. Indonesia, Malaysia and Singapore run provident funds, where you and your employer pay into an account with your name on it and you end up with a pot. Under the social insurance schemes of the Philippines, Thailand and Viet Nam, a formula decides what you get, whatever the arithmetic of your own contributions.
That distinction is worth knowing, and for the question at hand it is a red herring. Singapore and Malaysia both run provident funds and behave in opposite ways; so do Thailand and the Philippines with their social insurance. The design of the scheme turns out to be secondary to the rule that limits contributions.
One number decides it
Every scheme here except Malaysia's stops counting your wage at a ceiling. Compare that ceiling to the country's average wage and you get a ratio — the point, in multiples of an ordinary salary, at which your pension stops noticing your pay rises.
country ceiling / average wage replacement at 1x at 2x lost
Thailand 1.07x 41.5% 20.7% 50%
Singapore 1.22x 57.6% 33.0% 43%
Philippines 2.12x 72.4% 72.5% 0%
Indonesia 4.12x 53.5% 53.5% 0%
Viet Nam 6.27x 58.7% 58.7% 0%
Malaysia no ceiling 37.8% 36.4% 4%The rule holds without exception across all six: if the ceiling sits below twice the average wage, the replacement rate collapses for high earners. If it sits above, the rate stays flat.
The relationship is not a statistical correlation but a matter of direct calculation. Above the ceiling your contributions stop, so the benefit they buy stops with them, while the salary in the denominator keeps climbing. Thailand's ceiling sits at 1.07 times the average wage, which is to say barely above an ordinary salary. Singapore's sits at 1.22 times. Both stopped following their earners a long way below what a senior professional makes.
What this means for your own number
Calculating your own position against the contribution ceiling is more informative than a generic projection.
Take your country's contribution ceiling — THB 17,500 a month in Thailand, S$8,000 in Singapore, PHP 35,000 in the Philippines, IDR 11,086,300 in Indonesia. Divide your own monthly salary by it. If the answer is above 1, every baht, dollar or peso above that line is already invisible to your statutory pension, and has been for as long as you have earned it.
Nothing you do at work changes this. A promotion raises the salary your retirement will be measured against without raising the pension that has to meet it.
Both of the tight ones just moved, and both still bite
The two countries with the binding ceilings each raised theirs on 1 January 2026. Thailand went from THB 15,000 to THB 17,500, its first change since 1995. Singapore completed a four-year climb to S$8,000 a month.
Both increases were overdue, and neither changes the shape of the problem. A ceiling at 1.07 times the average wage was at roughly 0.92 times before the increase; Singapore's move took its own ratio from about 1.13 to 1.22. These are adjustments within the same regime rather than exits from it. The problem is structural, because a ceiling indexed to the average wage will sit near the average wage, and that is where it stops serving anyone who earns more.
Flat is not the same as adequate
Malaysia's EPF applies no wage ceiling at all. The Third Schedule of the 1991 Act tabulates contributions up to RM20,000 a month and applies the exact rate above that, so a director on RM50,000 contributes on all of it. Its replacement rate is accordingly almost flat.
It is also the lowest in the table, at about 38% wherever you sit on the income scale. Even a scheme that declines to cap you can fail to carry you. The ceiling ratio predicts whether your rate falls; it says nothing about where it starts.
The Philippines manages a high replacement rate that holds up across the income range. Malaysia holds up too, from a much lower starting point. Thailand is middling for someone on the average wage and becomes one of the least generous in the region for anyone above it.
And the finish lines are ten years apart
One more column in the same table rarely makes it into the comparison. The age at which these schemes expect to start paying you differs by a decade:
Malaysia and Thailand 55
Philippines 60
Viet Nam 62
Indonesia and Singapore 65A Malaysian retiring at 55 on a 38% replacement rate and a Singaporean retiring at 65 on 57.6% are in quite different circumstances. Comparing the percentages alone flatters Malaysia and understates Singapore, because the Malaysian's lower rate has to fund an additional ten years.
Where this comes from, and what will date
The replacement rates and average wages are the OECD's, from Pensions at a Glance Asia/Pacific 2024, Table 2.1: gross replacement rates from mandatory schemes for men on a full career from age 22, at half, one and two times mean earnings. The scheme classification is the ILO's. The ceilings are each country's own: the Royal Gazette regulation of 12 December 2025 for Thailand, the CPF Ordinary Wage ceiling for Singapore, the SSS maximum monthly salary credit, the BPJS Jaminan Pensiun ceiling from March 2026, and Viet Nam's twenty-times-base-salary cap. Read on 30 August 2026.
The ratios are ours, not the OECD's. Dividing each ceiling by each average wage, and the claim that the result predicts the table, is this article's arithmetic on the OECD's figures. The replacement rates are theirs; the explanation is ours, and you can check it in the table above.
A full career from age 22 is a modelling assumption, not a life. The OECD's figures assume unbroken contributions on the same relative earnings throughout. Career breaks, informal work and periods abroad all reduce these numbers, and in economies with large informal sectors that gap is the rule rather than the exception.
These are the mandatory schemes only. Voluntary top-ups, occupational plans, the Malaysian civil service pension and private saving are all outside the table. In several of these countries the statutory scheme was never meant to be the whole answer, which is a reasonable position and a different one from the scheme being adequate.
Men's figures, and the women's are mostly lower. Where the OECD reports a different rate for women it is generally below the men's, through earlier pension ages and sex-specific annuity rates. The single-number comparison above hides that.
Ceilings move, and two of these moved this year. Every ratio here is a snapshot taken at the end of August 2026. Thailand's is already scheduled to change again in 2029 and 2032. Recompute before relying on one.
