Labour, unemployment and the informal sector in India

Take the population aged 15 and over, N. A person is employed if they did any work for pay or profit (including unpaid work in a family enterprise) in the reference period, unemployed if they had no work but were available for it and looking, and outside the labour force otherwise: students, home-makers, the retired and those who have stopped looking. The labour force is employed plus unemployed (E + U) and the work force is only the employed. Three rates follow:

The unemployment rate divides by the labour force, not by the population, and this has two consequences that exams use. First, someone who gives up looking leaves the labour force, so the rate falls and so does the participation rate, with no job created: the discouraged worker effect. Second, a rise in participation can raise the unemployment rate when new entrants look and do not find work. The worker-population ratio avoids both problems, since it counts only people who work, which is why labour economists often prefer it for comparisons.

The calculator uses the headline all-India figures of the PLFS for July 2023 to June 2024, usual status, as shares of 1,000 persons aged 15 and over: 582 employed and 19 unemployed. That gives a participation rate of 60.1 per cent, a worker-population ratio of 58.2 per cent and an unemployment rate of 3.2 per cent. Try the other cases to see how the same population can give very different rates.

Whether a person is counted as employed depends on when you ask. The National Sample Survey Office, and since 2017 the Periodic Labour Force Survey of the Ministry of Statistics and Programme Implementation (the ministry that publishes the data on India’s informal sector, UGC NET 2024), use three reference periods (JAM 2025):

• Usual status (reference period one year): the activity on which a person spent the longer time in the previous 365 days is the principal status; a person whose principal status is unemployment or outside the labour force, but who worked for at least 30 days in some other activity, is counted as employed in a subsidiary status (JAM 2024). Taking principal and subsidiary status together gives the “usual principal and subsidiary status” employment rate. The older rule of “usual status employment” as a person working for at least six months of the year is the one the question from UOH 2018 names.
• Current weekly status (the previous 7 days): a person is employed if they worked for at least an hour on any day of the week.
• Current daily status (each day of the week): the activity on each day, so a person can be part-employed, part-unemployed over the week. It gives the largest estimate of unemployment, since a short spell of idleness is counted.

The Census counts main workers, who worked for six months or more in the year, and marginal workers, who worked for less than six months, and the work participation rate is the share of both in the population (UGC NET 2026). The three NSS measures give different unemployment rates for the same people, and the longer the reference period the lower the measured unemployment, since over a year most people find some work. So an unemployment rate always has to be quoted with its status.

In rich countries unemployment is mostly frictional (people between jobs), structural (skills do not match vacancies) or cyclical (too little demand). In a country like India the typical problem is underemployment: people work, but for too few hours or at too low a wage to live on, because open unemployment is a luxury the poor cannot afford. Two forms are named in exams.

• Disguised unemployment, the type found in Indian agriculture (CUET 2021): more workers are on the land than the output needs, so removing some leaves output unchanged. It is a case of zero marginal product, taken as the premise of the Lewis model in Lesson 3.
• Seasonal unemployment: work follows the farming year, with slack seasons when no work is available, which is the gap that public works programmes try to fill.

Open unemployment in India is also higher among the educated young than among the uneducated, because those who can afford to wait look for better jobs. The features that JAM 2026 asks for since 2000 include a shift of employment out of agriculture towards the non-agricultural sectors, a high employment elasticity in construction, and growth led by the service sector rather than by more labour-intensive production.

Keith Hart (1973) observed that many urban Ghanaians worked in small, unregistered activities outside the wage economy and called their income opportunities informal; the International Labour Office’s mission to Kenya (1972) used the term in a policy report. The informal sector became a standard category: small, easy to enter, family-owned, relying on local resources, with low skills, unregulated and competitive markets. Whether it is a trap for the poor or a source of dynamism is argued: the dualist view treats it as a residual for those the modern sector cannot absorb (as in Lessons 3 and 4); the structuralist view treats it as part of the same system, supplying cheap inputs to formal firms.

In India the official term is unorganised sector. The National Commission for Enterprises in the Unorganised Sector (NCEUS, 2007, chaired by Arjun Sengupta) defined the unorganised sector as all unincorporated private enterprises owned by individuals or households, run as proprietorships or partnerships, with fewer than ten workers. It added the idea of informal employment: unorganised workers include everyone in those enterprises, together with workers in formal firms who have no employment or social security benefits. The distinction matters because many jobs in registered firms are informal, on contract and without protection, which is part of what GATE 2025 means by a dual structure of the labour market with increasing gig and contractual employment, alongside high work-force participation in agriculture and without a predominantly formal sector. The Ministry of Statistics and Programme Implementation publishes the data on the informal sector.

The chart gives four World Bank series, built from the International Labour Organization’s modelled estimates so that countries can be compared on one definition. India’s participation rate for people aged 15 and over was 55.6 per cent in 2024, and for women 32.4 per cent, against 38.7 in Bangladesh and 68.9 in Vietnam (2024). The modelled unemployment rate was 4.17 per cent in 2024. Vulnerable employment, own-account and contributing family workers, who typically lack a formal contract and social protection, was 72.1 per cent of total employment.

These modelled figures differ from the PLFS ones in the calculator (participation of 60.1 per cent from the survey, against 55.6 per cent in this series), because the two differ in definition and method (the survey uses usual status and its own questions). Use the survey for India and the modelled series for comparisons between countries, and do not mix them. The most visible gap is for women: India’s female participation is among the lowest of these countries, a fact that is part of why the quality of work, not only its quantity, is the issue.

An employment guarantee promises unskilled manual work, at a set wage, to anyone who asks for it. It does two things: it gives a floor to rural incomes in the slack season and in bad years, and, because the wage is set low, it is self-targeting, since only those without a better option will take the work. The first scheme of this kind in India was the Maharashtra Employment Guarantee Scheme (CUET 2020), begun in the drought of the early 1970s and given statutory form in the Maharashtra Employment Guarantee Act of 1977. National schemes without a legal guarantee followed: the National Rural Employment Programme (1980) and the Rural Landless Employment Guarantee Programme (1983), with a Food for Work programme in 2001 (all dated in CUET 2022).

The National Rural Employment Guarantee Act, 2005 made the right law. It began on 2 February 2006 in 200 districts, covered 330 in 2007 and every rural district from 1 April 2008, and was renamed Mahatma Gandhi NREGA in October 2009 (the exams of 2012 to 2021 test its provisions). What the examiners keep asking:

• Who: every rural household whose adult members volunteer to do unskilled manual work, not every individual and not only households below the poverty line (UOH 2016, 2018, 2020).
• How much: up to 100 days of employment in a year, a minimum of 100 days for a household that asks for it (UOH 2012).
• The legal right: work was to be provided within 15 days of an application; if it was not, the applicant was entitled to an unemployment allowance, payable by the state (UOH 2017, 2021). A guarantee that has a penalty for failing to deliver is a legal right, unlike a scheme with a budget limit.

The Act was repealed with effect from 30 June 2026 and replaced by the Act known as VB-G RAM G. Questions that mention MGNREGA, set before that date, are about the Act as it stood, and the provisions above are those of the old Act; this page does not describe the new one, so read it from the ministry’s own text before you rely on it. The economics does not change: a guarantee gives a wage floor, smooths consumption and creates assets, at a fiscal cost, and its effect depends on how well the administration delivers work and pays on time.

Questions on the labour force, survey status, unemployment, the informal sector and employment programmes. Dated figures, such as the share of the unorganised sector in a given year, are not worked here because they change with each survey.