Two rates, two rankings
In the second quarter of 2026, the state of Maranhão had an unemployment rate of 6.0%, against 7.1% in the state of Rio de Janeiro. The gap of 1.1 points is wider than the desk’s margin of 0.83 points, computed from the coefficients of variation IBGE prints, so the order holds.
On the composite rate of labour underutilisation the order reverses: 22.7% in Maranhão against 13.0% in Rio. Maranhão ranks 13th of the 27 federative units on unemployment and 4th on the composite rate; Rio ranks 8th and joint 15th.
Maranhão is one of seven: of the other 26 units, seven have lower unemployment than Rio and a higher composite rate, and none has the opposite. Amapá has the highest unemployment rate, 9.8%, and the 12th highest composite rate, 14.5%. Piauí has the highest composite rate, 26.6%, and Santa Catarina the lowest on both, 2.1% and 4.1%.
- UNEMPLOYED: aged 14 or over, without work in the reference week, having taken effective steps to find work in the previous 30 days and available to start that week, or having already found a job due to start within four months.
- UNDEREMPLOYED BY HOURS: aged 14 or over, usually working fewer than 40 hours across all jobs, wanting to work more hours, and available to do so within 30 days. These people have work.
- POTENTIAL LABOUR FORCE: aged 14 or over, neither employed nor unemployed, in one of two groups: people who looked for work but were not available in the reference week, and people who did not look but wanted a job and were available. Discouraged people are one part of the second group.
- EXTENDED LABOUR FORCE: the labour force, meaning the employed and the unemployed, plus the potential labour force.
- THE TWO RATES: the unemployment rate divides the unemployed by the labour force. The composite rate of labour underutilisation divides the underemployed by hours, the unemployed and the potential labour force, added together, by the extended labour force.
So the two rates have different tops and different bottoms: each answers its own question, and neither is the real version of the other.
Five definitions from the Instituto Brasileiro de Geografia e Estatística. A reading that adds the potential labour force to the unemployed and calls the sum unemployment says what the data do not.
What the unemployment rate leaves out
Maranhão’s underused labour came to 749,000 people, of whom 174,000 were unemployed, 175,000 underemployed by hours and 399,000 in the potential labour force. Rio’s came to 1,183,000: 631,000 unemployed, 330,000 underemployed by hours and 222,000 in the potential labour force.
So Maranhão has more than twice as many people in the potential labour force as unemployed, and Rio about one for every three. Rio has 2.6 times Maranhão’s working-age population and 3.6 times its unemployed, yet Maranhão has 1.8 times Rio’s potential labour force.
Set against the extended labour force, the potential labour force is 12.1 points of Maranhão’s 22.7 and 2.4 of Rio’s 13.0, the desk’s arithmetic; that difference of 9.7 points is in effect the whole gap between the two rates. The other two parts nearly cancel out: Rio is higher on the unemployed and Maranhão on the underemployed by hours.
Fewer people are in work in Maranhão as well: 49.2% of those aged 14 or over, against 56.9% in Rio and 66.6% in Santa Catarina. Outside the labour force are 47.7% of Maranhão’s working-age population and 38.8% of Rio’s, the desk’s arithmetic. Most of them are not in the potential labour force at all.
| Persons | Maranhão | Rio de Janeiro | Piauí | Santa Catarina | Brazil |
|---|---|---|---|---|---|
| Unemployed | 174,000 | 631,000 | 121,000 | 98,000 | 5,861,000 |
| Underemployed by hours | 175,000 | 330,000 | 148,000 | 48,000 | 4,075,000 |
| Potential labour force | 399,000 | 222,000 | 160,000 | 49,000 | 4,723,000 |
| All underused labour | 749,000 | 1,183,000 | 429,000 | 195,000 | 14,659,000 |
| Extended labour force | 3,303,000 | 9,115,000 | 1,611,000 | 4,697,000 | 113,641,000 |
Every state on both measures
In Brazil as a whole the unemployment rate was 5.4% and the composite rate 12.9%. IBGE reports the highest measures of underused labour in the Northeast and the lowest in the South.
The composite rate spreads wider than unemployment: from 4.1% to 26.6%, against 2.1% to 9.8%. The table lists all 27 units, so any state can be read on both.
| State | Unemployment | Composite |
|---|---|---|
| Piauí | 8.3% | 26.6% |
| Bahia | 9.1% | 24.0% |
| Alagoas | 7.9% | 23.4% |
| Maranhão | 6.0% | 22.7% |
| Pernambuco | 8.3% | 21.0% |
| Sergipe | 7.9% | 20.9% |
| Pará | 6.2% | 19.5% |
| Acre | 6.6% | 18.3% |
| Ceará | 6.6% | 17.2% |
| Paraíba | 5.4% | 17.1% |
| Rio Grande do Norte | 5.6% | 16.3% |
| Amapá | 9.8% | 14.5% |
| Amazonas | 7.5% | 13.8% |
| Distrito Federal | 6.5% | 13.7% |
| Roraima | 5.0% | 13.0% |
| Rio de Janeiro | 7.1% | 13.0% |
| Tocantins | 4.1% | 11.2% |
| Minas Gerais | 3.8% | 10.5% |
| São Paulo | 5.4% | 10.3% |
| Rio Grande do Sul | 4.2% | 8.3% |
| Goiás | 4.0% | 8.0% |
| Paraná | 3.1% | 7.9% |
| Espírito Santo | 2.3% | 6.2% |
| Mato Grosso do Sul | 2.7% | 6.2% |
| Mato Grosso | 2.2% | 6.1% |
| Rondônia | 2.6% | 6.0% |
| Santa Catarina | 2.1% | 4.1% |
| Brazil | 5.4% | 12.9% |
How long the gap has lasted
Unemployment was lower in Maranhão than in Rio in each of the last six quarters, and the composite rate higher in each. The closest the unemployment rates came was the first quarter of 2026, 6.9% against 7.3%; the composite rates were never closer than 8.7 points.
It was not so in 2014: in the second quarter of that year Maranhão’s unemployment rate, 7.3%, was above Rio’s, 6.5%, and the composite rates were 22.0% and 8.7%. Since then Maranhão’s unemployment rate has fallen by 1.3 points, more than the desk’s margin of 0.81, and Rio’s has risen by 0.6, inside its margin of 0.78.
Rio’s composite rate rose by 4.3 points, beyond a margin of 1.19, while Maranhão’s rose by 0.7, inside a margin of 1.71. Across Brazil the two rates fell from 6.9% to 5.4% and from 14.9% to 12.9%.
The national figures have since moved on: in the three months to September 2026 the unemployment rate was 5.2%, a tenth of a point above the lowest reading since the series began in 2012, 5.1%, and the composite rate was 13.0%. The state figures for that quarter were not yet published when this was read.
| Quarter | Maranhão, unemployment | Rio, unemployment | Maranhão, composite | Rio, composite |
|---|---|---|---|---|
| Q2 2026 | 6.0% | 7.1% | 22.7% | 13.0% |
| Q1 2026 | 6.9% | 7.3% | 25.0% | 13.2% |
| Q4 2025 | 5.6% | 6.9% | 22.8% | 13.0% |
| Q3 2025 | 6.1% | 7.5% | 22.7% | 14.0% |
| Q2 2025 | 6.6% | 8.1% | 24.3% | 14.2% |
| Q1 2025 | 8.1% | 9.3% | 25.8% | 15.4% |
| Q2 2014 | 7.3% | 6.5% | 22.0% | 8.7% |
What these figures do not say
They do not say that the potential labour force is unemployed, or that it is hidden unemployment: IBGE counts these people outside the labour force, and they did not meet the test of looking and being available at once. They do not say that anyone in it is idle or does not want work.
They do not say why the gap exists: the survey does not attribute it to benefits, farming, informal work, drought, education or migration, and this piece does not either. They measure neither wages nor poverty, and they do not say that living or working in one state is better than in the other.
They do not say that a state’s rate is a record, and nothing here is a forecast. The figures for Rio de Janeiro are for the state, not the city, and nothing compares Brazil with another country.



