The Thai state pays a citizen over 60 a basic pension of 600 baht a month, around 17 US dollars. It steps up with age, to 700 baht in your seventies, 800 in your eighties, 1,000 once you pass 90 [3]. Set that against the line the same government draws for poverty, 3,078 baht a month in 2024 [4], and the Old Age Allowance lands at roughly a fifth of the income below which the state already counts you as poor. It is not a pension anyone could live on. It was never built to be.
So who actually carries an old Thai? Run the national accounts by age, the exercise the UN Population Fund and the NESDC publish for Thailand as its National Transfer Accounts, and the answer is blunt. Not the state. In those accounts public transfers play only a small net role in old-age support next to private transfers and the drawing-down of assets; the gap between what older Thais consume and what they still earn is filled mostly by their own savings and by money from their children [5]. In Thailand the family is the pension system. This is the second of three briefs on the same demographic break. The first took the population baseline and what it does to housing, and it ends on the fact that turns this one dark: the family that is meant to carry the old is the exact thing the country’s demographics is dismantling. Live median age, old-age dependency, and the support ratio behind all of it sit on the Regional Demographics dashboard [1][2].
The pension that isn’t one
Thailand introduced the universal Old Age Allowance in 2009, and for a while it was exactly that, universal: any Thai over 60 not already drawing a civil-service or Social Security pension could register at their district office and collect the monthly amount. In August 2023 the government changed the rule. New claimants from the 12th of that month face a means test, payment only to those with no income or too little to live on, the threshold set by a national committee on older persons; people already registered were left untouched [3][6]. The timing says something. In Thai policy terms the country crossed into a “complete-aged” society in 2024, with people aged 60 and over passing a fifth of the population [1]. As more people cross into old age each year, the state’s adjustment to the allowance ran the other way, narrowing access for new claimants rather than lifting the floor.
Every attempt to raise that floor has died on the same wall, which is cost. In April 2025 a citizen-initiated bill for a 3,000-baht People’s Pension, an amount that would have cleared the poverty line, was rejected before it reached a vote; the Paetongtarn government put the price near 400 billion baht a year and said the country could not afford it [7]. A rival idea, a flat 1,000 baht for every senior regardless of age, has been floated, costed, and shelved more than once [8]. The arithmetic is real, and there is no painless number. But look at what the argument is actually about. It is whether to lift the weakest, sub-subsistence tier of old-age support from a fifth of the poverty line to something nearer it, and the answer so far has been no.
The fund that runs dry
Above the allowance sits the only real contributory pension most private-sector workers have, the old-age benefit of the Social Security Fund, paid for by a payroll contribution split between worker, employer, and state. On its own actuaries’ numbers it is heading for a wall. The count of Thais drawing the SSF old-age pension is set to climb from around 760,000 in 2022 to some 2.3 million by 2032, while the working-age base paying in contracts [9]. A 2020 review by the Fund and the International Labour Organization put the trajectory plainly: contributions cover annual payouts only to about 2034, and on current rules the reserve, headed toward 5 trillion baht, is drawn down to nothing by roughly 2054 [9].
The fixes are known, and politically heavy. The standard menu is to raise the contribution rate, lift the wage ceiling that caps what higher earners pay in, and push back the pension age, which for the SSF still sits at 55 [9]. One of those levers has finally moved. A ministerial regulation published in December 2025 raises the contribution wage ceiling in stages from January 2026, from 15,000 baht a month toward 23,000 by 2032, the first increase since the ceiling was set in the 1990s; the maximum monthly contribution rises from 750 baht to 875 [17]. It helps at the edges and comes nowhere near an actuarial gap measured in trillions of baht. The harder levers, a higher contribution rate and a later retirement age, have not moved, because each is a heavier charge on the working-age population at the exact moment that population is shrinking. That is the recurring shape of the problem: the reforms that would actually close the gap ask a contracting group of workers to pay more, and they get harder to legislate every year the group contracts.
The majority without a contributory pension
Here is the part that drops out of most discussion of Thai pensions: most Thai workers are not in the formal system at all. Depending on how it is counted, informal work covers somewhere between roughly half and two-thirds of Thai employment, more than 20 million people: street vendors, motorcycle-taxi riders, farmers, domestic workers, the self-employed. WIEGO puts the 2024 figure at 65 percent of total employment; the OECD’s 2025 survey reads it closer to half, the difference being mostly one of definition [10]. They have no employer to split a payroll contribution with, and the schemes meant to reach them are thin: Article 40, the voluntary track for informal workers, has drawn roughly 11 million members, and the National Savings Fund fewer than 3 million, both paying out in modest lump sums or small annuities rather than a pension you grow old on [11].
The cost of that gap shows up at the other end of life. About two-thirds of Thais now over 60, on the order of 66 percent, draw no contributory pension whatsoever, because they never accumulated the years of formal-sector contributions an annuity requires [12]. For them the entire apparatus, the SSF, the civil-service scheme, comes to nothing. What is left is the 600-baht allowance, whatever they managed to save, and then the family.
So the family pays, and the family is shrinking
This is where the demographics of the first brief stop being abstract. The Thai system, by default far more than by design, rests on one assumption: that adult children carry their parents. For a long time they did, and many still do. The National Transfer Accounts put a number on it. Family transfers fund roughly a third of the gap between what older Thais consume and what they earn, with most of the remainder drawn from the old person’s own savings and assets, and the state a thin sliver of the rest [5]. Old-age support in Thailand is, in the accounting sense, privatised, run through the household ledger.
Now read that assumption against the population it depends on. In 2020 Thailand had about 5.4 working-age adults for every person over 65. By 2050 it will have about 1.9 [1]. The old-age dependency ratio doubles over the same window, from 22 to 50 per hundred working-age Thais [1][2]. The pool of children available to carry each parent is collapsing, and it is collapsing fastest exactly where the parents are most numerous.
It is not only that there are fewer children. They are also further away, and the ones still at home are often not the ones the model assumed. John Knodel’s long-running surveys of Thai ageing found co-residence between elderly parents and an adult child falling from about 77 percent in the mid-1980s to 59 percent two decades on, as the young left for Bangkok and the industrial estates [13]. The slide did not stop there: by 2021 about 12 percent of over-60s lived alone, some 1.6 million people, roughly double the 2002 share, on the National Statistical Office’s own survey [18]. What rose alongside it is the skipped-generation household, grandparents raising grandchildren while the working-age middle is away earning, now a recognised and growing category in Thai ageing research [18]. It quietly hands the old a second job, childcare, at the age the old-age system assumed they would be the ones receiving care.
And the carrying is not shared evenly. The unpaid work of looking after old parents in Thailand falls heavily on women, daughters and daughters-in-law, who put in on the order of two to two and a half hours a day of elder care, far more than men, and who routinely cut their paid hours or quit work to do it while men do not [14]. So the cost not carried through the public system does not vanish. It lands on the household, and inside the household it lands on women, as forgone wages, forgone pension contributions of their own, and a thinner retirement a generation later. The family-as-pension model is not only shrinking. It is quietly manufacturing the next cohort of under-pensioned old women.
The bill comes due either way
The state can keep the pension floor low. It cannot make the cost of ageing disappear; it can only choose where that cost sits, and on the public books it is already rising. The World Bank projects, on a no-reform path, Thailand’s combined public spending on pensions and health care, the civil-service pension, the Old Age Allowance, and the health system together, climbing from 6.2 percent of GDP in 2020 to 11.3 percent by 2060 [15]. Inside that total the pension piece does the heavy lifting. OECD and IMF projections have public pension spending more than quadrupling as a share of the economy, from 1.4 percent of GDP in 2017 to 5.6 percent by 2060, with public health spending rising from 3 to 5 percent over the same span [16]. Poonpolkul and co-authors, in the cleanest recent study of the Thai case, name the trap directly: the country is caught between the inadequacy of what it pays the old and the fiscal constraint on paying more, and their own modelling finds that a pension-income-tested top-up, paid in full to the majority with no other pension and tapered for those who already draw one, reaches the poverty line more affordably than lifting the floor uniformly for all [12].
This is where the dull reforms the first brief flagged stop being dull, because the alternative to reform is not the status quo. It is the family breaking under a load that grows every year. The honest list is short and graceless. Raise the Old Age Allowance toward something a person can actually live on, and aim the new money at the poorest old rather than spreading it to a fifth of the line for everyone. Fix the Social Security Fund now, while there is still a contributor base to fix it with, through some mix of higher contributions, a higher ceiling, and a later pension age. Build a genuine route into coverage for the informal majority instead of two voluntary schemes most never join. And the piece everyone skips: construct an actual long-term-care system and pay for the care work women now do for nothing, so that carrying a parent stops meaning impoverishing a daughter.
None of this is dictated by the demographics. Thailand could keep the family as its pension system and let the weight fall where it falls. The point, the same one the first brief made about housing and labour, is that the demographics raise the cost of avoiding the choice, every year, as the support ratio drops.
I started with 600 baht against a 3,078-baht poverty line, because that gap is the whole argument in miniature. Thailand has built a society that is now old, one in five over 60 and heading for more than one in three by 2050 [1], on top of a pension system that assumes someone else, the family, does the carrying. For decades the assumption held, because Thai families were large and lived close. The demographics are removing both conditions at once, and on a clock the country can read precisely, because everyone who will be old in 2050 is already alive [1]. The cost of an ageing Thailand will not disappear because the state has kept it off its own books. It will keep moving onto households, and inside those households onto women, until the state either picks up more of the weight while it still has the workers to tax for it, or waits until there are two workers to a pensioner and a family net that has already torn. The third brief turns to the other half of the question, what an ageing, shrinking workforce does to growth and productivity, and whether automation and skills can keep Thailand rich enough to afford any of this. Demographics decided who gets old. It did not decide who pays. That part is still open, and it is the part worth fighting over.
References
[1] United Nations, Department of Economic and Social Affairs, World Population Prospects 2024. Primary source for the demographic figures unless otherwise noted: median age, working-age population, old-age dependency, the potential support ratio (5.4 working-age per person 65-plus in 2020 falling to about 1.9 by 2050), and the share of the population over 60. https://population.un.org/wpp/
[2] A1AYN, Regional Demographics dashboard. Live median age, old-age dependency, working-age share, and the support ratio for Thailand and 174 other countries. https://a1ayn.com/data/demographics/
[3] OECD, Pensions at a Glance Asia/Pacific 2024 (Thailand); US Social Security Administration, International Update, October 2023. The Old Age Allowance pays 600 baht a month at ages 60-69, 700 at 70-79, 800 at 80-89, and 1,000 at 90 and over; a means test applies to claimants registering from 12 August 2023, with existing recipients unaffected. https://www.oecd.org/en/publications/pensions-at-a-glance-asia-pacific-2024_d4146d12-en/full-report/thailand_eaeb7aea.html
[4] Office of the National Economic and Social Development Council (NESDC), 2024 report on poverty and inequality in Thailand (reported by The Nation); World Bank, Thailand Poverty and Equity Brief, October 2024. NESDC raised the national poverty line to 3,078 baht a month in 2024 (from 3,043 in 2023); 3.43 million Thais (4.89 percent) fell below it, of whom about 879,000 classed as “very poor” averaged 615 baht a month, close to the base Old Age Allowance. https://www.nationthailand.com/news/policy/40055537 ; https://documents.worldbank.org/en/publication/documents-reports/documentdetail/099011301032557829
[5] UNFPA Thailand and NESDC, National Transfer Accounts (Thailand), 2021; Ronald Lee and Andrew Mason, Population Aging and the Generational Economy (Edward Elgar / IDRC, 2011). The generational-economy accounting behind the finding that net public transfers to older Thais are small and that family transfers and the elderly’s own assets fund most of their old-age consumption deficit. https://thailand.unfpa.org/en/Thailand-NTA-2021
[6] Bangkok Post, “Parties slam new restriction on old-age allowance” (2023). Reporting on the 12 August 2023 shift from a universal allowance to a means-tested one for new claimants, and the political reaction. https://www.bangkokpost.com/thailand/politics/2629326/
[7] Thai PBS Verify, fact-check of the 3,000-baht “People’s Pension” proposal (2025). The citizen-initiated People’s Pension bill, costed at roughly 400 billion baht a year, was rejected; the government argued it was unaffordable. https://www.thaipbs.or.th/verify/en/content/8315
[8] The Nation, “Govt plans flat 1,000 baht allowance for all seniors.” Reporting on repeated proposals to replace the tiered allowance with a flat 1,000-baht monthly payment, and their fiscal costing. https://www.nationthailand.com/news/policy/40046541
[9] The Nation, “Thailand grapples with looming Social Security Fund crisis”; International Labour Organization, Thailand pension-reform analysis (2020). SSF old-age pensioners projected to rise from about 760,000 (2022) to 2.3 million (2032); contributions cover payouts only to roughly 2034 and the reserve is depleted by about 2054 without reform; the SSF pension age is 55. https://www.nationthailand.com/business/economy/40049591
[10] WIEGO, “Employment in Thailand, 2019-2024: Crisis, Recovery and Informality”; OECD Economic Surveys: Thailand 2025 (Tackling informality). Informal employment was about 65 percent of total Thai employment in 2024. https://www.wiego.org/research-library-publications/employment-thailand-2019-2024-crisis-recovery-informality/
[11] International Labour Organization, Thailand social-protection diagnostic. Coverage of the voluntary schemes for informal workers, Article 40 and the National Savings Fund, and the gap left for workers outside formal social security. https://www.ilo.org/media/376171/download
[12] Phitawat Poonpolkul, Ponpoje Porapakkarm and Nada Wasi, “Aging, inadequacy, and fiscal constraint: The case of Thailand,” International Studies of Economics 19(1), 35-67 (2024). Finds that about 66 percent of Thais aged 60 and over are ineligible for a Social Security annuity for want of contribution years, and that a pension-tested basic pension (the term is the authors’, meaning a top-up tapered against existing pension income, distinct from the income means test on the OAA) reaches the poverty line more affordably than a uniform increase. https://onlinelibrary.wiley.com/doi/10.1002/ise3.66
[13] John Knodel et al., The Situation of Thailand’s Older Population (Population Studies Center, University of Michigan; HelpAge International surveys). Co-residence of elderly parents with an adult child fell from roughly 77 percent in the mid-1980s to about 59 percent by the late 2000s, alongside the rise of skipped-generation households. https://www.psc.isr.umich.edu/pubs/pdf/rr15-847.pdf
[14] University of Manchester, “Social gender norms deepen elderly care burdens for Thai women” (2025), reporting research published in World Development. Drawing on a national time-use survey of more than 70,000 Thai adults: women provide about two to two and a half hours a day of unpaid elder care, far more than men, and are the ones who cut paid work to do it. https://www.manchester.ac.uk/about/news/social-gender-norms-deepen-elderly-care-burdens-for-thai-women/
[15] World Bank, “The Macroeconomic and Fiscal Impact of Aging in Thailand” (2021). Combined public spending on the civil-service pension, the Old Age Allowance, and health care is projected to rise from 6.2 percent of GDP in 2020 to 11.3 percent by 2060. https://www.worldbank.org/en/country/thailand/publication/the-macroeconomic-and-fiscal-impact-of-aging-in-thailand
[16] OECD, Financing Social Protection through General Tax Revenues, Social Security Contributions and Formalisation in Thailand (with IMF projections). Public pension spending projected to rise from 1.4 percent of GDP in 2017 to 2.7 percent by 2035 and 5.6 percent by 2060; public health spending from 3 percent to 4 and then 5 percent over the same years. https://www.oecd.org/en/publications/financing-social-protection-through-general-tax-revenues-social-security-contributions-and-formalisation-in-thailand_b5cc1a43-en.html
[17] DLA Piper, “Thailand’s new social security wage ceilings, effective January 2026”; Baker McKenzie InsightPlus; Royal Gazette ministerial regulation, 12 December 2025. The monthly wage base for Social Security Fund contributions rises in stages, from 15,000 baht to 17,500 (2026-2028), 20,000 (2029-2031) and 23,000 (from 2032), the first increase since the ceiling was set in the 1990s; the maximum monthly contribution rises from 750 to 875 baht. The contribution rate (5 percent) and the SSF pension age (55) are unchanged. https://www.dlapiper.com/en/insights/publications/2025/12/thailands-new-social-security-wage-ceilings-has-officially-published
[18] Thailand National Statistical Office, Survey of Older Persons; research on Thai older-adult living arrangements (e.g., Tangchonlatip et al.; PMC, 2022). The share of people over 60 living alone roughly doubled, from 6.3 percent in 2002 to about 12 percent (some 1.6 million) by 2021; skipped-generation households, grandparents raising grandchildren while working-age adults migrate for work, are a recognised and growing arrangement. https://pmc.ncbi.nlm.nih.gov/articles/PMC9655045/
Explore the data: Thailand’s median age, old-age dependency, and support-ratio trajectory, alongside every other ASEAN country, are live on the Regional Demographics dashboard.
Demographic figures are UN World Population Prospects 2024, medium variant, with estimates through 2024 and later years projected. Fiscal projections are World Bank, OECD and IMF baselines and assume no reform; they are scenarios, not forecasts. Current as of June 2026.