1,910,195 short of hours, 877,716 out of work
In August 2026, Rwanda's labour force survey counted 1,910,195 people who were employed and wanted, and were available for, more working hours. In the same round it counted 877,716 unemployed people: 2.18 underemployed for every unemployed person.
The two groups do not overlap. The unemployed had no work, were seeking it and were available for it. The time-related underemployed had work, but less of it than they wanted and were available to do.
The survey covers people aged 16 and over, and its third-quarter round is collected in August. The figures come from the quarterly tables the National Institute of Statistics of Rwanda published with its report for that round.
- UNEMPLOYED MEANS ALL THREE AT ONCE: without employment, actively seeking it and available for it.
- TIME-RELATED UNDEREMPLOYED MEANS EMPLOYED. They are people in employment whose working time is insufficient in relation to other work they are willing and available to do.
- THE POTENTIAL LABOUR FORCE IS OUTSIDE THE LABOUR FORCE. It is people not in employment who were seeking work but not available, or available but not seeking.
- PRODUCING FOOD FOR ONE'S OWN HOUSEHOLD IS NOT EMPLOYMENT HERE. The survey counts as employed those who worked for pay or profit, and leaves out subsistence foodstuff production.
So the underemployed and the unemployed are two separate groups in the same survey, and the first is counted among the employed.
These are the definitions NISR's report prints for the survey, restated here; the questionnaire itself was not read.
Both counts rose in a year, the underemployed by more
Between August 2025 and August 2026 the count of time-related underemployed went from 1,730,608 to 1,910,195, which is 179,587 more. The count of unemployed went from 754,312 to 877,716, which is 123,404 more.
The count of employed people went the other way, from 4,888,921 to 4,752,916, which is 136,005 fewer. So more people wanted more hours inside a smaller employed population.
The potential labour force, people outside it who wanted work but were not both seeking and available, went from 1,650,400 to 1,807,065, which is 156,665 more.
| Group, aged 16 and over | August 2025 | August 2026 | Change (derived here) |
|---|---|---|---|
| Time-related underemployed (in work) | 1,730,608 | 1,910,195 | +179,587 |
| Unemployed | 754,312 | 877,716 | +123,404 |
| Potential labour force | 1,650,400 | 1,807,065 | +156,665 |
| Employed, all | 4,888,921 | 4,752,916 | -136,005 |
40.2% of the employed, 15.6% of the labour force
The time-related underemployment rate was 40.2% in August 2026, against 35.4% a year earlier, and the unemployment rate was 15.6%, against 13.4%.
The two rates do not share a denominator. The underemployment rate divides by the employed, the unemployment rate by the labour force, which is the employed plus the unemployed. So 40.2 and 15.6 are not shares of the same people, and this edition compares the counts instead.
The survey's composite measure, which adds the unemployed, the time-related underemployed and the potential labour force, was 61.8% of the extended labour force, against 56.7% a year earlier. The 40.2% is above each of the 29 earlier quarters for which the tables publish that rate.
| Rate | August 2025 | August 2026 | Change in points (derived here) |
|---|---|---|---|
| Time-related underemployment, % of the employed | 35.4% | 40.2% | +4.8 |
| Unemployment, % of the labour force | 13.4% | 15.6% | +2.2 |
| Composite underutilization, % of the extended labour force | 56.7% | 61.8% | +5.1 |
More underemployed than unemployed in 29 of 30 quarters
The tables publish both counts for 30 quarters, from the first quarter of 2019 to the third quarter of 2026; the second quarter of 2020 is published without the underemployment figure. In 29 of those 30 quarters the time-related underemployed outnumbered the unemployed.
The one exception is the fourth quarter of 2021, with 1,124,897 underemployed against 1,135,637 unemployed.
In the third-quarter rounds the ratio was between 1.25 and 1.53 from 2019 to 2023, then 1.98 in 2024, 2.29 in 2025 and 2.18 in 2026. NISR enlarged the sample and updated the design in February 2024, so the years on either side are printed and no trend is drawn across that change.
| Round | Underemployed | Unemployed | Underemployed per unemployed (derived) |
|---|---|---|---|
| Third quarter of 2019 | 852,890 | 601,022 | 1.42 |
| Third quarter of 2020 | 1,016,825 | 697,210 | 1.46 |
| Third quarter of 2021 | 958,744 | 764,480 | 1.25 |
| Third quarter of 2022 | 1,251,338 | 819,936 | 1.53 |
| Third quarter of 2023 | 1,186,283 | 874,876 | 1.36 |
| Third quarter of 2024 | 1,611,245 | 815,267 | 1.98 |
| Third quarter of 2025 | 1,730,608 | 754,312 | 2.29 |
| Third quarter of 2026 | 1,910,195 | 877,716 | 2.18 |
What these figures do not say
They do not say that 1,910,195 Rwandans are unemployed. The time-related underemployed are employed, and none of them is in the 15.6% unemployment rate.
They do not say that the unemployed are idle. The survey does not count producing food for one's own household as employment, and 43.3% of the unemployed in August 2026 were engaged in subsistence agriculture.
They do not say how many more hours the underemployed wanted, or that they wanted full-time jobs. This edition read no figure on the hours they sought.
They do not say that the ratio is rising. It was 2.29 in August 2025 and 2.18 in August 2026, and the sample design changed in February 2024.
They do not say why. This edition read no figure that compares reasons, so it names no cause: not the harvest calendar, not prices, not policy. And it read nothing by region, sex, age, occupation or industry for these groups, no other economy and no forecast; no figure here comes from our own platform.



