Understanding the Enumeration of Women’s Work in India

Every ten years, India counts its people. But does it count all the work they do? When it comes to women’s economic contributions, the answer has consistently been: not quite. Despite having two major data systems – the Census of India and the National Sample Survey (NSS) – the country has struggled to accurately capture the full range of work that women perform. From unpaid care work at home to seasonal agricultural labour, much of what women do remains invisible in official statistics. This gap is not just a technical problem. It is deeply tied to how society defines and values “work” itself.

How India measures its workforce 🔗

India primarily relies on two instruments to estimate the size and composition of its economically active population: the decennial Census and the periodic rounds of the National Sample Survey (NSS), now conducted under the Periodic Labour Force Survey (PLFS) framework.

The Census, conducted by the Office of the Registrar General, asks every household a set of questions about the main activity of each member. It classifies individuals as workers (main or marginal) or non-workers. The NSS, managed by the National Statistical Office (NSO), uses a more detailed questionnaire with multiple reference periods – usual status (one year), current weekly status, and current daily status – to capture different dimensions of employment.

In theory, these tools should give us a comprehensive picture of India’s labour force. In practice, especially for women, they fall significantly short.

The persistent undercount of women’s work 🔗

One of the most striking features of Indian labour statistics is the consistently low female labour force participation rate (FLFPR). According to the Census 2011, only about 25.5% of women were classified as workers, compared to over 53% of men. While the PLFS has shown some improvement in recent rounds, the underlying issues with data collection remain.

Why do the numbers look so low? 🔗

The problem starts with how “work” is defined in these surveys. Both the Census and the NSS have historically used definitions rooted in the International Labour Organization’s (ILO) framework, which focuses on market-oriented economic activity. This framework was originally designed for industrialised economies where most production happens in formal markets. It does not translate well to a context where a huge volume of productive activity – especially by women – happens within the household, outside the formal market.

Consider a woman in rural India who spends her morning tending to livestock, her afternoon processing grain for the family’s consumption, and her evening fetching water and firewood. If a Census enumerator asks her, “What is your main occupation?”, she is likely to say “housework” – and the enumerator is likely to record her as a non-worker. The same activities, if performed for wages or sale, would be counted as economic work.

The role of reference periods 🔗

The NSS attempts to address some of these issues through multiple reference periods. The “usual status” approach captures work done over the past year, which is better suited to picking up seasonal work. The “current weekly status” and “current daily status” measures provide more granular data. Women’s participation rates tend to be higher under the usual status measure than the daily status measure, because many women work seasonally – during harvest periods, for instance – and may not be working on the specific day or week of the survey.

However, even the usual status measure misses a great deal. The Time Use Survey conducted in 2019 revealed that Indian women spend an average of nearly 300 minutes per day on unpaid domestic services and caregiving activities, compared to about 97 minutes for men. Almost none of this gets classified as “economic activity” in the Census or PLFS.

Feminist critiques of data collection methods 🔗

Feminist economists have long argued that the way labour force data is collected in India systematically excludes women’s contributions. These critiques operate on multiple levels – from the design of questionnaires to the training of enumerators to the conceptual categories used to classify work.

The problem with questionnaire design 🔗

Scholars like Indira Hirway have pointed out that standard labour force surveys use a “main activity” approach that forces respondents into a single category. For men who typically have one primary occupation, this works reasonably well. For women who juggle multiple activities – domestic chores, care work, subsistence agriculture, home-based production – reducing their day to a single “main activity” inevitably erases much of what they do.

Furthermore, many surveys use a filter question early on – something like “Did you work in the last week?” If a woman says “no” because she does not perceive her unpaid household production as “work,” the survey skips all subsequent questions about economic activity. She is classified as a non-worker without any further probing.

Enumerator bias and respondent perception 🔗

The data collection process itself introduces biases. Enumerators, often male, may not probe adequately when a woman describes her activities. There is a well-documented tendency for household respondents – frequently the male head of the household – to describe women’s status as “attending to domestic duties” rather than listing their productive activities. This is not always a deliberate misrepresentation. In many communities, patriarchal norms dictate that women’s place is at home, and any work they do is simply part of their domestic role, not “real” work.

Research published by the United Nations Entity for Gender Equality (UN Women) has consistently highlighted this global pattern. Across developing countries, the social perception of women’s roles heavily influences how their economic activities are reported – or not reported – in official surveys.

The social undervaluation of women’s work 🔗

The statistical invisibility of women’s work is not simply a measurement error that better survey techniques can fix. It reflects deeper social attitudes about what kinds of labour have economic value.

Patriarchal norms and the definition of “productive” work 🔗

In India’s patriarchal social structure, the domestic sphere is often seen as women’s natural domain, and the activities performed within it are not considered “productive” in an economic sense. Cooking, cleaning, childcare, elder care, water and fuel collection – all of these are essential for the functioning of any household and, by extension, the broader economy. Yet they carry no market price and generate no income, so they are not classified as economic activity under standard definitions.

This creates a circular logic: women’s work is invisible in statistics because it is undervalued in society, and it remains undervalued partly because it is invisible in the data that informs policy. The UNDP Human Development Report has repeatedly noted that failing to account for unpaid care work leads to policies that do not address women’s actual needs and contributions.

The GDP blind spot 🔗

India’s System of National Accounts (SNA), which follows international guidelines, includes some forms of household production – such as owner-occupied housing and subsistence agriculture – within the GDP boundary. However, it excludes unpaid domestic and care services. Various estimates suggest that if unpaid household work were valued at market rates, it could add anywhere from 15% to over 40% of GDP, depending on the methodology used.

This is not merely an academic exercise. When a significant share of productive work is excluded from national accounting, the economy appears smaller than it actually is, women appear less productive than they actually are, and resource allocation decisions are made on incomplete information.

Steps toward better enumeration 🔗

Recognising these shortcomings, several improvements have been attempted over the years. The NSS has refined its questionnaire design in successive rounds, adding activity codes to capture a wider range of tasks that women perform. The introduction of the Time Use Survey in 2019 was a significant milestone, providing the first nationally representative data on how men and women allocate their time across different activities throughout the day.

The Time Use Survey as a corrective 🔗

The Time Use Survey (TUS) asks respondents to account for every activity they perform in a 24-hour period, rather than asking them to identify a single “main” activity. This approach captures the multiplicity of tasks that characterise women’s daily lives. The 2019 TUS revealed that 81.2% of women aged 15-59 participated in unpaid domestic services on a daily basis, compared to just 26.1% of men. These numbers make the gendered division of labour starkly visible in a way that standard employment surveys simply do not.

However, the TUS has its own limitations. It has been conducted only once so far, and it has not been integrated into the regular statistical system. Until time-use data is collected periodically alongside labour force surveys, it will remain a supplementary tool rather than a corrective to mainstream employment statistics.

International best practices 🔗

The ILO’s 19th International Conference of Labour Statisticians (ICLS) in 2013 adopted a new resolution that broadened the definition of “work” to include own-use production of goods and services. Under this framework, activities like fetching water, collecting firewood, and producing goods for household consumption are all classified as work, even if they are not employment. India has begun to align its survey methodology with these revised standards, but full implementation is still a work in progress.

Why accurate enumeration matters for policy 🔗

Getting the numbers right is not an end in itself. Accurate data on women’s work is essential for designing effective policies in areas ranging from employment generation to social protection to infrastructure development.

If policymakers believe that women in a particular district are “not working,” they are unlikely to invest in childcare facilities, skill development programmes, or safe transport – all of which are critical for enabling women’s participation in the labour market. On the other hand, if data shows that women are already performing 8-10 hours of productive work daily (albeit unpaid), the policy response shifts from “how to get women to work” to “how to recognise, redistribute, and support the work they already do.”

This reframing – from labour force “participation” to the full spectrum of work women perform – is at the heart of the feminist critique. It challenges us to move beyond counting heads in factories and offices, and instead build a statistical system that reflects economic reality in all its complexity.

What do you think? If India’s national surveys fully accounted for unpaid domestic and care work, how might that change the way we design welfare programmes and economic policies? And in your own experience, how much of the productive work in your household goes unrecognised in any official count?

References

1. https://mospi.gov.in/periodic-labour-force-survey-plfs
2. https://www.ilo.org/global/statistics-and-databases/standards-and-guidelines/resolutions-adopted-by-international-conferences-of-labour-statisticians/lang–en/index.htm
3. https://plfrorg.in/wp-content/uploads/2021/07/Time-Use-Survey-2019.pdf
4. https://www.epw.in/journal/2012/24/perspectives/accounting-womens-work.html
5. https://www.unwomen.org/en/digital-library/publications/2019/10/counting-the-unpaid-work-of-women-and-girls
6. https://hdr.undp.org/content/human-development-report-2023-24
7. https://mospi.gov.in/time-use-survey