How the Census of India Measures Women’s Work

Every ten years, India undertakes the world’s largest peacetime administrative exercise: counting its more than 1.4 billion people. The Census of India, conducted by the Office of the Registrar General and Census Commissioner under the Ministry of Home Affairs, does much more than just tally heads. It tries to map who is working, where, and in what capacity, including the millions of women whose labour fuels households, farms, and small enterprises across the country. Yet, when it comes to measuring women’s work, the Census tells only part of the story, and the gaps are as instructive as the data itself.

The decennial Census and how it tracks work 🔗

The Indian Census is a decennial exercise, meaning it takes place every ten years. The tradition began with a non-synchronous count in 1872 under Viceroy Lord Mayo, with the first synchronous all-India census conducted in 1881 under Lord Ripon. Since 1951, it has been governed by the Census Act, 1948, and conducted in two phases: a Houselisting and Housing Census, followed by Population Enumeration. During the second phase, enumerators record each person’s age, marital status, education, migration history, fertility (for women), and crucially, their economic activity.

To capture work, the Census uses a simple but powerful classification system. A worker is defined as a person whose main activity is participation in any economically productive activity, whether physical or mental. This definition explicitly includes effective supervision and direction of work, and unpaid work on a farm or in a family enterprise. Workers are then split into two categories based on the duration of work in the year preceding the Census.

Main workers and marginal workers 🔗

The distinction between these two categories is central to understanding labour data in India. According to the Open Government Data Platform, those workers who had worked for the major part of the reference period (six months or more, that is, 183 days or more) are termed Main Workers. Those who had not worked for the major part of the reference period, that is, less than six months, are termed Marginal Workers. A person who did not work at all during the reference period is treated as a Non-worker.

This three-way split matters enormously for women. A large share of women in rural areas perform seasonal agricultural labour, work intermittently on family farms, or take up short stints of paid work. They are therefore far more likely to be classified as marginal workers than as main workers. According to analysis of Census data, the share of female workers in non-agricultural occupations rose particularly sharply during 1991-2001 among the main workers category, signalling a shift in the structure of female employment even within these official definitions.

Work Participation Rate: what 1991 and 2001 revealed 🔗

The Work Participation Rate (WPR) is the headline indicator that comes out of Census data. It is defined as the percentage of total workers (main plus marginal) to the total population. It is the cleanest way to compare workforce engagement across men and women, rural and urban areas, and over time.

The 1991 and 2001 censuses are particularly revealing for women’s economic participation. The official Census of India explainer notes that the overall WPR rose from 37.5% in 1991 to 39.3% in 2001. While the WPR for males marginally increased from 51.6% to 51.9% during this decade, for females it improved significantly from 22.7% to 25.7%.

This was widely cited as evidence that more women were entering the recorded workforce. But the increase came with a catch: it was driven largely by a rise in the share of female marginal workers rather than main workers. In other words, more women were being counted, but in short-duration, often poorly paid roles.

Rural and urban differences in women’s WPR 🔗

The rural-urban gap in female work participation is one of the most striking features of Indian labour data, and one that the Census has consistently captured. Rural women have always recorded higher WPRs than urban women, almost the inverse of what one might expect if industrialisation were the main driver of female employment.

A regional study based on the 2001 Census found that rural areas had a women’s WPR of 30.79% compared with only 11.8% in urban settings. By Census 2011, the pattern persisted: the female workforce participation rate stood at 25.51% against 53.26% for males, with rural female WPR at 30.02% compared with 53.03% for rural males. Urban female WPR continued to languish in the mid-teens.

Why is rural female WPR so much higher? Most rural women work in agriculture, either as cultivators on family land or as agricultural labourers, and these activities are at least partially captured by Census enumerators. Urban female WPR has hovered close to 15% over decades, reflecting a mix of social norms, the structure of urban job markets, and the difficulty Census methods have in identifying home-based and informal urban work.

Where the Census struggles: measuring unpaid and home-based work 🔗

Counting women’s work in India is fundamentally complicated by the fact that much of it is unpaid, informal, and folded into household routines. The Census’s definitions, however inclusive on paper, run into serious practical limits during enumeration.

The “housewife” problem 🔗

One of the most persistent biases in Census data is that women who do substantial productive work, beyond cooking and childcare, often end up classified as housewives or non-workers. As economist Bina Agarwal has pointed out, biases in definitions of work and in reporting by typically male enumerators and respondents lead to many hardworking women being labelled as housewives. Tasks like cattle care, fetching firewood and fodder, keeping poultry, and cultivating family fields are widely performed by rural women but go invisible because admitting that women work outside the home is also seen as lowering the family’s status.

Unpaid family work on farms and enterprises 🔗

Family enterprises, whether small shops, weaving units, or agricultural plots, are the backbone of livelihoods for hundreds of millions of Indians. They depend heavily on the unpaid labour of women. Recent research argues that standard labour force surveys underestimate women’s contribution to paid work because family enterprises rely on the free labour provided by women, and this work often goes unnoticed and unrecognised, leading to underreporting by both men and women. The same problem applies to the Census. Even when the rulebook treats unpaid family work as economic activity, household members and enumerators often do not see it that way.

Domestic work outside the production boundary 🔗

Cooking, cleaning, fetching water, caring for children and the elderly: these are activities that consume hours of women’s days but fall outside what statisticians call the “production boundary” of the System of National Accounts. India’s statistical system follows this convention. As one analysis explains, the Central Statistical Organization does not include the processing of primary goods and collection of free goods in India’s GDP, which means workers engaged in these activities, mostly women, are excluded from the estimated workforce of the National Sample Survey Organisation and the Population Census. They are typically clubbed under codes for “domestic duties and allied activities,” outside the workforce altogether.

Seasonal, intermittent, and short-duration work 🔗

The 183-day cut-off that separates main from marginal workers is itself a source of undercounting. A great deal of women’s paid work is short-duration, fluctuating with the agricultural cycle or with care responsibilities at home. Work done over shorter durations fails to get adequately captured in household surveys, leading to economically active women not getting classified as workers. The Census’s annual reference period was designed for stable, full-year employment; it works less well for the rhythms of rural women’s labour.

Why these gaps matter for policy 🔗

If the Census undercounts women’s work, the consequences are not merely academic. WPR figures feed into employment policy, social security design, and the framing of women’s economic role in development plans. When women’s labour is invisible in the data, it is also invisible in the design of welfare schemes, training programmes, and labour protections.

The first full-scale Time Use Survey conducted by the National Statistical Office in 2019 represented a major attempt to fix this blind spot. By asking respondents to log activities through the day, including unpaid domestic and care work, time-use data provides a better framework for understanding gender gaps in the world of work. The Time Use Survey complements rather than replaces the Census, but it shows what a more honest picture of women’s labour might look like.

The Census’s reach, scale, and decennial regularity remain unmatched. It provides the denominator for almost every social and economic statistic in the country, from fertility rates to electoral delimitation. But on women’s work, it is best read as a careful but conservative estimate, the floor of what women contribute to the economy, not the ceiling.

What do you think? 🔗

What do you think? If a woman spends her morning cooking for the family, her afternoon working in the family shop, and her evening tending to the family’s livestock, how should the Census classify her? And if our economic statistics counted unpaid care and domestic work, how might that change the way India designs its labour and welfare policies?

References

1. https://censusindia.gov.in/census.website/en/node/378
2. https://anantamias.com/first-census-in-india/
3. http://labourbureau.gov.in/WL 2K5-6 Chap 1.htm
4. https://www.data.gov.in/catalog/main-workers-marginal-workers-non-workers-and-those-marginal-workers-non-workers
5. https://mpra.ub.uni-muenchen.de/48003/1/MPRA_paper_48003.pdf
6. https://censusindia.gov.in/data_products/library/post_enumeration_link/eci6_page3.html
7. https://www.academia.edu/1388302/Education_and_womens_participation_in_Indian_economy_A_regional_analysis
8. https://www.mospi.gov.in/sites/default/files/reports_and_publication/statistical_publication/social_statistics/Chapter_4.pdf
9. https://www.business-standard.com/article/specials/urban-india-and-its-female-demographic-dividend-115073000229_1.html
10. https://www.businesstoday.in/opinion/interviews/story/undercounting-of-women-in-workforce-is-not-just-a-statistical-artefact-60202-2016-01-18
11. https://voxdev.org/topic/labour-markets/indian-women-are-not-dropping-out-paid-work-voluntarily
12. https://www.nature.com/articles/s41599-020-0488-2
13. https://iariw.org/wp-content/uploads/2021/08/Deb_paper.pdf