Japan’s Statistics Bureau reported 68.49 million employed people in the basic results of its August 2026 Labour Force Survey, released on 2 October. The estimate was 140,000 higher than in August a year earlier. The number of unemployed people was estimated at 1.80 million, down 20,000 on the same annual comparison. The seasonally adjusted unemployment rate stood at 2.5%, an increase of 0.1 percentage point from July. These are survey estimates and they refer to different comparison periods.

The Bureau mainly presents unadjusted values for the numbers employed and unemployed, while it highlights the seasonally adjusted unemployment rate because month-to-month movement matters more for that measure. Thus, 68.49 million and 1.80 million are estimates of people for the reference month, whereas the 2.5% rate is a series adjusted to remove seasonal patterns. A 0.1-point monthly increase does not mean the rate rose by 0.1 percent; it moved from 2.4% to 2.5%. A percentage point describes the difference between two percentages.

The annual count changes and the monthly unemployment-rate movement should also be kept separate. The employment estimate was 140,000 higher than in August 2025, and the unemployed estimate was 20,000 lower on that annual comparison. The increase in the unemployment rate compares the seasonally adjusted July and August 2026 figures. Together, these measures track different aspects of labour-market status, but they do not share one reference period and cannot be added together or substituted for one another.

The survey is national in scope. The Bureau says it asks about employment and unemployment among roughly 100,000 people aged 15 and over in about 40,000 households randomly selected across Japan. Its definitions conform to International Labour Organization standards and use recent actual activity. A person who worked for at least one hour for pay or profit may count as employed; people temporarily away from work can also qualify under the rules. This is not a measure of full-time or permanent jobs alone.

The Labour Force Survey does not by itself describe workers’ pay, the quality of their hours, vacancies, or job security. Nor is the employment count a direct tally of newly created jobs: it is a sample estimate expanded to represent the country. Someone who meets the survey’s employment definition may, for example, have worked only briefly. The figures therefore describe employment status under the survey’s rules; separate series are needed to assess job quality or other workplace conditions.

With the August release, the Bureau noted that the employed count had increased year on year for the first time in two months and that the unemployed count had declined year on year for the first time in 13 months. These comments provide context about the recent direction of the estimates; they are not forecasts of what will happen next. Seasonally adjusted series can be recalculated, so comparisons should use the same current data vintage. The Bureau’s English guidance explains why it emphasizes seasonal adjustment for rates and unadjusted estimates for population counts.

The main August 2026 findings are an estimated 68.49 million employed people, 1.80 million unemployed people, and a seasonally adjusted unemployment rate of 2.5%. Readers should keep the count, rate, and reference periods distinct. This monthly survey is a snapshot, not a census or a complete administrative record of every employment contract. It provides a core indicator for monitoring Japan’s labour market, while not representing the experience of every individual or sector on its own.