For as long as India has measured employment, the answers have arrived at the wrong scale. A national unemployment rate tells a job-seeker in Darbhanga nothing. A State figure for Gujarat averages Surat's textile mills with the dry districts of Kachchh. The person deciding whether to migrate, and the official deciding where to put a skilling centre, have both been working from numbers drawn one or two levels above the place they actually care about.
On 18 September 2026 the National Statistical Office released district-level key labour market indicators for the first time. Labour force participation, employment, unemployment and youth disengagement, district by district.
Why this was not possible before
The reason is a sampling one, and it is the most useful thing in this release.
A sample survey estimates a population from a subset. Its reliability depends on how many observations fall within the group you want a number for. India's Periodic Labour Force Survey (PLFS) was historically designed to produce reliable national and State estimates. Within any single district it might interview only a handful of households — enough to contribute to a State average, nowhere near enough to publish a district figure without an unacceptable margin of error.
From January 2025 the PLFS sample design was revamped. Districts are now treated as basic strata, separately for the rural and urban sectors, when selecting first-stage units, and the overall sample size was substantially enlarged. That redesign is what makes district publication statistically defensible, and it is why this release appears now rather than a decade ago.
The general principle is worth carrying beyond this story: the geography at which a statistic can be published is a property of the survey design, not of the data processing. You cannot obtain a credible district number by slicing a survey that was never built to support one.
The four indicators, defined precisely
The release uses the usual status approach and reports four indicators. Getting the denominators right is where marks are won.
Labour Force Participation Rate (LFPR) — the number of persons in the labour force, meaning those working or seeking or available for work, per 1,000 persons in the population.
Worker Population Ratio (WPR) — the number of employed persons per 1,000 persons in the population.
Unemployment Rate (UR) — the number of unemployed persons per 1,000 persons in the labour force. The denominator is the labour force, not the population. This single difference generates more wrong answers than any other definition in Indian economics.
NEET — the share of youth aged 15-29 who are not in employment, education or training. It captures a group that the other three indicators hide: a young person neither working, nor looking for work, nor studying does not appear as unemployed at all.
Usual status classifies a person by their activity over the preceding 365 days, using a majority-time criterion. It is the long reference period, and it produces lower unemployment estimates than current weekly status, which uses the preceding seven days. The same population can yield two different unemployment rates depending on which lens is used, which is why the reference period is always stated.
What the map shows
Participation and employment are remarkably uniform. About 98.8 per cent of districts record LFPR between 40 and 80 per cent, and about 98.3 per cent record WPR in the same band. Within that band the two diverge: a majority of districts (52.7 per cent) sit in the 60-80 per cent range for LFPR, while a majority (54.3 per cent) sit in the lower 40-60 per cent range for WPR. Among the top half of districts by population, WPR ranges from 37.3 to 76.7 per cent.
Unemployment is low almost everywhere. About 77.8 per cent of districts report UR between 0.5 and 5.5 per cent — 43.1 per cent between 0.5 and 3.0, and 34.7 per cent between 3.0 and 5.5. Only 14.9 per cent of districts exceed 5.5 per cent, and 7.2 per cent fall below 0.5 per cent.
Youth disengagement is the real problem. 45.9 per cent of districts record NEET between 20 and 30 per cent, 26.1 per cent between 10 and 20, and 23.7 per cent above 30 per cent. Only 4.2 per cent of districts are below 10 per cent. Roughly one district in four has more than three in ten of its young people neither working nor studying nor training.
Female participation clears 40 per cent in most places. About 57.8 per cent of districts record a female LFPR of 40 per cent or above — and the State-wise spread is enormous. Among the ten most populous districts of each State, the best performers include Surguja in Chhattisgarh (69.0 per cent) and Tinsukia in Assam (66.1 per cent), while Bihar's best is Siwan at 24.7 per cent. A gap of forty-four percentage points between the leading district of one State and the leading district of another is the single most striking number in the release.
Among the 100 most populous districts:
| Indicator | Highest | Lowest |
|---|---|---|
| LFPR (15+) | Surat 68.6%, Araria 68.0%, Salem 67.8% | Darbhanga |
| WPR (15+) | Surat 67.8%, Araria 67.4%, Banaskantha 66.6% | Darbhanga |
| UR (15+) | — | South Twenty-Four Parganas 0.9%, Araria 0.9%, Ahmedabad 1.1% |
| NEET (15-29) | — | Ernakulam 9.6%, Coimbatore 14.5%, Bengaluru Urban 15.0% |
Among State capital districts, Shimla leads on LFPR and WPR (70.0 and 65.1 per cent), followed by Thiruvananthapuram (64.4 and 63.2) and Raipur (61.5 and 59.4). The lowest unemployment among capitals: Thiruvananthapuram and Kolkata at 1.9 per cent, then Bengaluru Urban at 2.5 per cent. The lowest NEET: Srinagar 14.3, Bengaluru Urban 15.0, Guntur 15.1 per cent.
The trap in the table
Look again at Araria in Bihar. It records among the highest labour force participation and employment of India's hundred largest districts, and among the lowest unemployment rates — 0.9 per cent. On a naive reading it is one of the best labour markets in the country.
It is one of the poorest districts in India.
Both facts are true, and reconciling them is the most important analytical point in this entire release. The unemployment rate measures job search, not job quality. To be counted as unemployed you must be without work and seeking or available for it. That is a status only affordable to someone who can survive while searching — someone with savings, a remitting household, or a family that can carry them. A landless agricultural labourer with no cushion cannot be unemployed for a month. They take whatever work is available at whatever wage, and are recorded as employed.
Economists call this distress employment, and its statistical signature is exactly what Araria shows: high participation, high employment, near-zero unemployment, deep poverty.
The mirror image appears in Kerala, which has historically reported some of India's higher measured unemployment rates alongside its best human development outcomes. An educated young person in Ernakulam with a remitting relative in the Gulf can afford to reject unsuitable work and keep searching for a job matching their qualifications. That search registers as unemployment. Note that Ernakulam simultaneously records the lowest NEET among the hundred largest districts, at 9.6 per cent — its young people are overwhelmingly working or studying. High unemployment and low disengagement together describe a labour market where people are choosy because they can be.
This is why NEET is the more revealing indicator for a developing economy, and why its inclusion here matters. It does not ask whether someone is searching. It asks whether they are doing anything at all. Our CDS/OTA economy notes on unemployment and labour work through the full set of definitions and their traps.
What this changes
District estimates are not merely a finer version of State estimates. They change what is administratively possible.
Targeting. Skilling centres, employment exchanges and livelihood missions can be placed where measured need is, rather than distributed by population or by political weight.
Accountability. A State average conceals its own internal failures. Gujarat contains both Surat and Banaskantha; Bihar contains both Araria and Darbhanga. Publishing the district exposes the dispersion that the average was smoothing away.
Migration analysis. Origin and destination districts can now be compared directly, which is the natural unit for understanding internal migration — a person migrates from a district, not from a State.
The caution to state alongside: district-level survey estimates carry wider confidence intervals than State estimates even under the improved design, because the sample within each district is still far smaller. Small differences between districts should not be over-read, and a ranking is less robust than a band. Treating a first-ever statistical product as precise to the decimal point is the standard error of enthusiasm.
🔑 Revision block
The release. 18 September 2026 — National Statistical Office (NSO), Ministry of Statistics and Programme Implementation. District-level key labour market indicators, released for the first time.
Why now. The PLFS sample design was revamped from January 2025: districts treated as basic strata, separately for rural and urban sectors, with an enlarged sample. Publication geography is a property of survey design, not of processing.
The four indicators. LFPR — working or seeking/available for work, per 1,000 population. WPR — employed per 1,000 population. UR — unemployed per 1,000 in the labour force (denominator is the labour force). NEET — share of youth 15-29 not in employment, education or training.
Reference periods. Usual status = preceding 365 days, majority-time criterion. Current weekly status = preceding 7 days. Usual status yields lower unemployment estimates.
Distribution. LFPR 40-80% in about 98.8% of districts; WPR in about 98.3%. 52.7% of districts have LFPR in 60-80%; 54.3% have WPR in 40-60%. WPR of the top half of districts by population: 37.3-76.7%.
Unemployment. 77.8% of districts between 0.5 and 5.5% — 43.1% in 0.5-3.0, 34.7% in 3.0-5.5. Above 5.5%: 14.9%. Below 0.5%: 7.2%.
NEET. 45.9% of districts at 20-30%; 26.1% at 10-20%; 23.7% above 30%; only 4.2% below 10%. 76.2% of districts below 30%.
Female LFPR. 57.8% of districts at 40% or above. State leaders include Surguja, Chhattisgarh (69.0%) and Tinsukia, Assam (66.1%); Bihar's leader is Siwan (24.7%).
Top 100 populous districts. Highest LFPR: Surat 68.6, Araria 68.0, Salem 67.8. Highest WPR: Surat 67.8, Araria 67.4, Banaskantha 66.6. Lowest UR: South Twenty-Four Parganas 0.9, Araria 0.9, Ahmedabad 1.1. Lowest NEET: Ernakulam 9.6, Coimbatore 14.5, Bengaluru Urban 15.0. Lowest LFPR and WPR: Darbhanga, Bihar.
State capitals. Highest LFPR/WPR: Shimla (70.0 / 65.1), Thiruvananthapuram (64.4 / 63.2), Raipur (61.5 / 59.4). Lowest UR: Thiruvananthapuram and Kolkata 1.9, Bengaluru Urban 2.5. Lowest NEET: Srinagar 14.3, Bengaluru Urban 15.0, Guntur 15.1.
The analytical point. Low UR ≠ good jobs. Unemployment measures search, which only those who can afford to search undertake. High LFPR + near-zero UR + deep poverty = distress employment (Araria). Higher UR + low NEET + high human development = a labour market where people can be choosy (Ernakulam, Kerala).
The caution. District estimates carry wider confidence intervals than State estimates; read bands, not rankings.
🎯 Practice MCQs
Q1. The Unemployment Rate is calculated as the number of unemployed persons per 1,000 persons in the: (a) Total population (b) Labour force (c) Working age population (d) Employed population
→ (b) — the single most commonly missed definition in labour statistics.
Q2. NEET, as used in labour force statistics, refers to youth who are: (a) Not employed but enrolled in training (b) Not in employment, education or training (c) Newly entering employment or training (d) Employed in the non-formal economy
→ (b) — for the age group 15-29.
Q3. Under the usual status approach, a person's activity is classified with reference to the preceding: (a) 7 days (b) 30 days (c) 180 days (d) 365 days
→ (d) — current weekly status uses 7 days.
Q4. District-level labour market estimates became publishable mainly because: (a) The population census was updated (b) The PLFS sample design was revamped with districts as basic strata (c) Administrative employment records replaced survey data (d) The definition of unemployment was changed
→ (b) — from January 2025.
Q5. Among India's 100 most populous districts, the highest LFPR was recorded by: (a) Ahmedabad (b) Surat (c) Salem (d) Coimbatore
→ (b) — 68.6 per cent.
Q6. A district reporting very high labour force participation together with a near-zero unemployment rate and widespread poverty most likely indicates: (a) A highly productive local economy (b) Distress employment (c) Large-scale in-migration of skilled workers (d) A statistical error
→ (b)
Q7. The Periodic Labour Force Survey is conducted by the: (a) Reserve Bank of India (b) National Statistical Office (c) NITI Aayog (d) Ministry of Labour and Employment
→ (b) — under MoSPI.
Q8. Among State capital districts, the highest LFPR was recorded by: (a) Thiruvananthapuram (b) Raipur (c) Shimla (d) Kolkata
→ (c) — 70.0 per cent.
Q9. The Worker Population Ratio differs from the Labour Force Participation Rate in that WPR excludes: (a) Women workers (b) Persons who are seeking or available for work but not employed (c) Self-employed persons (d) Rural workers
→ (b) — LFPR counts the employed plus the job-seekers; WPR counts only the employed.
Q10. Consider the following statements: 1. Usual status generally yields a higher unemployment estimate than current weekly status. 2. About one-fourth of India's districts recorded a NEET rate above 30 per cent. Which is/are correct? (a) 1 only (b) 2 only (c) Both 1 and 2 (d) Neither 1 nor 2
→ (b) — the longer reference period yields a lower estimate; 23.7 per cent of districts exceeded 30 per cent NEET.
📋 How this gets asked (PYQ pattern)
Labour statistics questions in CDS and OTA papers divide sharply into two kinds, and candidates who prepare only for the first kind lose the second.
The definition question is almost guaranteed. It asks what LFPR, WPR, UR or NEET measures, or which agency conducts the PLFS, or what distinguishes usual status from current weekly status. These are pure recall, and the denominator of the unemployment rate is the most frequently tested single fact in the set. Learn the three denominators — population, population, labour force — and most of this category resolves itself.
The interpretation question is where the harder marks sit, and it is increasingly common. It presents a scenario — a State with low unemployment and low incomes, or one with high unemployment and high literacy — and asks what it indicates. The examiner is testing whether the candidate understands that unemployment is a measure of search behaviour, not of welfare. Kerala and Bihar are the standing illustrations, and they now have district-level counterparts in Ernakulam and Araria.
Data-point questions — which district led on LFPR, what share of districts exceeded a threshold — are less likely in a written paper but perfectly plausible in an interview, where the ability to cite a specific figure marks a candidate who reads rather than crams.
The efficient note is short: four definitions with their denominators, two reference periods, one sampling insight, and the distress-employment paradox with its two examples.
Preparing for CDS or OTA? Economic statistics reward the candidate who knows the denominator — most wrong answers in this area come from a correct concept applied to the wrong base. Build the base with our CDS/OTA economy notes, follow the daily CDS/OTA current affairs, and prepare with our faculty in the upcoming Cavalier courses in Delhi.
✍️ Written by Hitendra Deswal — Economy & international relations faculty at The Cavalier. Reviewed by the Cavalier Faculty Desk.