Employment
01
Unemployment sits at 2.9%.
DOSM's monthly Labour Force Survey for August 2026, read live from the official open-data API. The rate is below its pre-pandemic decade average of 3.2% — who is working, who is looking, and who is outside the count altogether.
Unemployment · Aug 2026
2.9%
of the labour force was without work, looking and available
Participation
71.0%
of working-age people are in the labour force at all
Employed
16.87M
up 189.9 thousand people on a year earlier
Unemployed
511.6k
people — each one counted as actively seeking work
Official latestMonthly official releases, published in arrears — not a live count. The page re-reads the feed daily and always shows the latest surveyed month, never a projection.
DOSM — lfs_month & lfs_month_sa (monthly Labour Force Survey), via the OpenDOSM / data.gov.my open API · retrieved 2026-10-12 17:27 UTC
Official DOSM household-survey estimates, Jan 2010–Aug 2026, data as of 2026-08. Person counts in thousands; rates in per cent. Published monthly in arrears — the latest surveyed month, not a live count.
The long view
02
A decade of routine, then a spike.
Unemployment held near 3.2% through the 2010s, hit 5.3% in May 2020, and now sits at 2.9% — below the pre-pandemic decade's average. Every month of the survey since January 2010.
Unemployment rate
Seasonally adjusted
The dashed line is DOSM's seasonally adjusted companion series — the same survey with the regular within-year rhythm removed. The two lines sit close; where they part is where something other than the calendar moved the rate.
Participation
03
71.0% in, 29.0% out.
The participation rate — the share of working-age people either working or looking for work — stands at 71.0% for August 2026, against a 2010s average of 66.9%. 7.11 million working-age people remain outside the labour force.
At today's working-age population, a single point of participation is worth about 245 thousand people. The 4.1-point climb from the 2010s average therefore represents roughly 990,000more people in the labour force than that decade's rate would put there — before the population grew at all.
These are national headline rates. The monthly series this chapter reads carries no state, gender or age breakdown — a companion youth series exists on OpenDOSM but is not wired in yet.
Did you know
Only 20 of 200 months have been this low.
Unemployment at or below today's 2.9% appears in just 20 of the surveyed months since January 2010 — including each of the latest 1, in a row.
Computed from DOSM Labour Force Survey monthly series via OpenDOSM
The sharpest month moved the rate +1.2 points.
From 3.8% in March 2020 to 5.0% in April 2020 — the steepest single-month rise in the series. The series peak, 5.3% in May 2020, stands 2.1 points above the 2010s average.
Computed from DOSM Labour Force Survey monthly series via OpenDOSM
The labour force has grown 40.6% since 2010.
12.37 million people in January 2010, 17.39 million in August 2026. Employment grew alongside it, from 11.93 million to 16.87 million.
Computed from DOSM Labour Force Survey monthly series via OpenDOSM
Deep data
03.3
What “unemployed” actually means.
These figures come from a household survey, not a register. In labour force surveys of this kind, a person counts as employed if they did any work at all in the survey's reference week; as unemployed only if they had no work, were actively looking and were available to start. Everyone else — students, homemakers, retirees, people who have stopped looking — is outside the labour force, and outside the unemployment rate's denominator.
Series start
2010
monthly and unbroken since January 2010, on one survey definition
Cadence
Monthly
published in arrears; the page refetches daily
Latest month
Aug 2026
the most recent surveyed month in the open series
Data as of
2026-08
the vintage DOSM stamps on the feed itself
Headline ≠ underemployment
Any work in the reference week is enough to count as employed. The headline rate says nothing about hours, pay, or whether the job matches the worker — underemployment needs tables this series does not carry.
Rates versus people
The rate is lower than in January 2010 (2.9% against 3.5%) while unemployed persons are more numerous (511.6k against 436.7k) — a growing labour force moves both numbers at once. Read rates for stress, counts for scale.
Survey versus registrations
These are estimates from a household survey. Counts of people registered with placement services or claiming assistance measure different populations — neither is a subset of the other, and the two will not match.
DOSM — lfs_month & lfs_month_sa (monthly Labour Force Survey), via the OpenDOSM / data.gov.my open API · retrieved 2026-10-12 17:27 UTC
Official DOSM household-survey estimates, Jan 2010–Aug 2026, data as of 2026-08. Person counts in thousands; rates in per cent. Published monthly in arrears — the latest surveyed month, not a live count.
Work is where household income meets household bills — the labour force here is the denominator for wages, prices and output elsewhere in the library. Next shelves: Cost of Living and Economy.