The Issue

For some time now, the slow and declining rate of women’s workforce participation has been a major concern and the subject of much discussion among social scientists and policymakers. The low and declining women’s work participation rates until recently were a stark comment on both the status of women in India and labour market conditions in the economy, which clearly did not generate enough productive employment to meet the needs of the existing labour force, leave aside the tens of millions of young people entering the labour force every year.

This is why the recent labour force surveys of the National Sample Survey Office (NSSO), which have suggested a very significant increase in women’s employment rates after a prolonged period of decline and/or stagnation, have been so welcome, especially in official circles. The Ministry of Labour and Employment has characterised this process as “A Rising Tide of Female Participation”, which reflects the “Unstoppable Rise of Women in India’s Workforce”. It has highlighted the “remarkable increase in the Female Labor Force Participation Rate (LFPR) and Work Participation Rate (WPR) for women, coupled with a drop in the female unemployment rate. This shift speaks not only to economic progress but also to India’s drive toward creating an equitable, vibrant labour market where women play a leading role in shaping the nation’s future”.

Male work participation rates have risen slightly from 71.2% in 2017-18 to 76.1% in 2023-24. But those for women increased much more dramatically in the relatively short time span of six years.

Indeed, ever since the new quarterly Periodic Labour Force Surveys were introduced by the NSSO, women’s work participation rates have been increasing continuously—and went up quite sharply in 2023-24—to the point that they have nearly doubled since 2017-18. In turn, this has been the dominant cause of the overall increase in male and female employment combined that is recorded in the same period.

Figure 1 shows overall work participation rates for the country as a whole. Male work participation rates have risen slightly from 71.2% in 2017-18 to 76.1% in 2023-24. But those for women increased much more dramatically in the relatively short time span of six years. The rate was at only 22% in 2017-18, and went up by more than 80%, or more than 26 percentage points, to reach 40.3% in 2023-24. (All the figures refer to shares of population of 15 years and above.)

Figure 1: Work Participation Rates in India, 2017-18 to 2023-24

Figure 2 shows that this increase was evident in both rural and urban areas, with the increase in women’s work participation particularly sharp in rural India. The last two years of this period showed the biggest increases.

Figure 2: Female Work Participation Rates in India, 2017-18 to 2023-24

Clearly, if taken at face value, this is a very welcome development. Could this increasing work participation in recent years, especially among women, suggest that the period of jobless growth is finally receding? Is it possible that the benefits of gross domestic product (GDP) growth are finally translating into higher employment, and affording greater work opportunities to women in particular?

…Because of the ways in which the NSSO defines and classifies “work”, it is necessary to interpret such data trends much more carefully.

Is the Ministry of Labour and Employment right in claiming that “As more women join India’s workforce, the country stands on the brink of a transformative economic shift that embodies the spirit of inclusivity and opportunity. As women step into roles across sectors—be it technology, healthcare, education, or manufacturing—they bring fresh ideas, skills, and dedication that fuel economic growth and enhance workplace dynamics”? The official note further argues that “a notable decline in the female unemployment rate signals an increase in job availability, with women capitalising on these opportunities, reflecting both enhanced employment access and growing job satisfaction among women”.

If so, this is clearly something to celebrate. However, because of the ways in which the NSSO defines and classifies “work”, it is necessary to interpret such data trends much more carefully. In particular, we must unpack them to discover which specific types of work have increased to decide whether this is indeed a sign of labour market dynamism and improved conditions for women’s employment, as has been suggested by official spokespersons.

Concepts and Definitions

It is first necessary to examine how the NSSO has addressed the complex question of defining, identifying, and measuring work and employment in India over the past few decades, and in the recent past.

Most standard dictionaries define work as “any activity involving mental or physical effort done in order to achieve a result”. Economic activities are typically defined in a more restrictive way, as actions that involve the production, distribution, and consumption of goods and services at all levels within a society, which, of course, begs the further question of what constitutes “goods and services”. Pared down to its essence, it can be noted that any activity that can potentially be delegated is economic activity, which leaves only personal consumption and leisure as non-economic activities.

This requires a broadening of the common perception of what such activity is. For example, the activities associated with motherhood are typically seen as “non-economic”. Yet, breastfeeding can be outsourced through the hiring of a wet nurse, which then makes it an economic activity, with the wet nurse engaged in paid work. An even more extreme but recently proliferating example is that of surrogate motherhood, in which a woman is paid to be impregnated, carry a child in her womb and go through child birth, making all these explicitly paid economic activities, which, in turn, also contribute to national income to the extent of the remuneration received.

Yet a woman who does this for her “own” child rather than someone else’s, and without any monetary reward, is classified as “not in the labour force” in most if not all national statistical systems—and, indeed, the very notion of “maternity leave” from paid work suggests that the mother is in effect on some sort of holiday, rather than actively engaged in the work of producing a child.

…Breastfeeding can be outsourced through the hiring of a wet nurse, which then makes it an economic activity, with the wet nurse engaged in paid work.

Some of this contradiction was resolved by the 19th International Conference of Labour Statisticians (ICLS) in Geneva in 2013. It distinguished between work and employment and expanded the concept of work—“Work comprises any activity performed by persons of any sex and age to produce goods or to provide services for use by others or for own use” (ICLS 2013; p. 2). This definition was subsequently adopted by the International Labour Conference (ILC) in 2014.

The inclusion of the last phrase “for use by others or for own use” provides the crucial difference, as it includes the production of goods and services performed in the home for other household members and for personal use. So work is now defined irrespective of its formal or informal character or the legality of the activity. It only excludes activities that do not involve producing goods or services (for example, begging and stealing), self-care (for example, personal grooming and hygiene), and activities that cannot be performed by another person on one’s own behalf (for example, sleeping, learning, and activities for recreation).

The significance of this definition is that it maintains that productive work can be performed in any kind of economic unit, including the family or household. Employment—defined as “work for pay or profit”—therefore becomes a subset of work. Figure 3 provides a graphical representation of this, which clarifies those forms of work that are seen as part of the System of National Accounts (SNA) production boundary.

Figure 3: Forms of Work and the System of National Accounts 2008

In the early rounds of the National Sample Survey (NSS), until the 22nd Round (1967-68), the term “work” was used as the basic category, defined in general terms as any work done for production of goods and services. However, domestic work by the family members was not considered as “work”.

Thereafter, from the 32nd (1977-78) to the 49th (Jan-June 1993) Rounds, the NSS shifted to the concept of “gainful activity”, which was defined as any activity pursued for pay, profit or family gain or, in other words, any activity which adds value to the national income. While normally these would be activities that result in production of goods and services for exchange, the NSS also included within this category of gainful work all the agricultural activities in which a part or whole of the agricultural production was not sold but used for own consumption.

However, some activities that did result in earnings, such as prostitution, begging, and gambling, were not considered to be “gainful”. It was argued that this was “by convention”, although that is hardly a precise conceptual basis on which to define work, especially as these activities do generate incomes that are then included in national income.

It is worth noting that the NSS explicitly notes that the activity status of a person is to be judged irrespective of the situation whether such activity is carried out illegally or not, say, in the form of smuggling. So clearly legality is not the basis for exclusion of activities from the realm of “economic” activities—rather it appears to be some inchoate and non-explicit notion of morality. The entire range of activities involved in social reproduction (described simply as “the execution of household chores”), along with voluntary work for the community or for meeting social commitments, was also not considered to be part of gainful activities.

Once again, all domestic work has been excluded, as have been miscellaneous activities such as prostitution, begging, and gambling even though they generate earnings.

From the 50th Round (1993-94) onwards, the NSS moved to a notion of “economic activity” rather than “gainful” activity. As in the previous Rounds, this included market activities and non-market activities in agriculture directed towards own consumption, including not only cultivation and post-harvest activities but also gathering of uncultivated crops, forestry, collection of firewood, hunting, fishing, and the like.

In addition, it included another set of activities that were previously not included—those related to the production of fixed assets on own account, such as the construction of own houses, roads, and wells, and of machinery, tools, and so on for household enterprises and also for the construction of any private or community facilities free of charge. Involvement in such own account construction either as labourer or supervisor has been construed as economic activity. However, the processing of primary products for own consumption has not been considered as an economic activity. Once again, all domestic work has been excluded, as have been miscellaneous activities such as prostitution, begging, and gambling even though they generate earnings.

As a result, the NSSO even in the latest labour force surveys includes the following as working or being engaged in economic activity (employed):

• Code 11– worked in household enterprise (self-employed) as own account worker (11)
• Code 12 – worked in household enterprise (self-employed) as employer (12)
• Code 21 – worked as helper in household enterprise (unpaid family worker) (21)
• Code 31 – worked as regular salaried/wage employee (31)
• Code 41 – worked as casual wage labour in public works (41)
• Code 51 – worked as casual wage labour in other types of works (51)

Those that the NSSO puts together in the broad category that it calls “neither working nor available for work (or not in labour force)” includes the following:

• Code 91: attended educational institutions
• Code 92: attended to domestic duties only
• Code 93: attended to domestic duties and was also engaged in free collection of goods (vegetables, roots, firewood, cattle feed, and so on), sewing, tailoring, weaving, and the like. For household use
• Code 94: rentiers, pensioners, remittance recipients, and so on.
• Code 95: not able to work owing to disability
• Code 96: beggars, prostitutes, and so on
• Code 96: others

Codes 92 and 93 are the activity classifications that cover unpaid work done within households, which are in fact economic activities, in that they can potentially be performed for remuneration, and indeed often are. However, because the NSSO classifies those engaged in Codes 92 and 93 as “not in the labour force”, they are excluded from the employment data. (It is worth noting that the exclusion of Code 96 from the defined work force is an enigma since these are clearly activities that are economic in that they involve exchange and/or remuneration. This exclusion appears to be based on some notion of morality rather than any economic rationale.)

It has very important implications because it turns out that it is precisely in the category of “unpaid helpers in family enterprises” in which most of the increase in women’s “employment” has occurred.

But matters are further muddied by yet another sub-classification within those it chooses to define as “workers”; that is, “unpaid helpers in family enterprises”. But the dividing line between such activities and those of persons working as “unpaid helpers in family enterprises” is thin and can even be indistinguishable. This is certainly in the case in agriculture as well as in many non-agricultural informal enterprises. It is possible, and even likely, that some similar activities are arbitrarily placed in one or the other category.

Note that definitions of work and employment by the International Labour Organisation (ILO) make it clear that while work can comprise all activities that generate goods and services, whether for exchange or for consumption within the household or for own consumption, employment is only that subset of work for which remuneration is received. By that token, since “unpaid helpers” are clearly not the primary workers who would receive remuneration, they should not really be classified as “employed” at all.

Note that “other work” in Figure 3, which includes compulsory work without pay for others, is not considered as “employment”. This could well describe much of the work done by “unpaid helpers in family enterprises”, given the patriarchal social relations and unequal access to monetary incomes that we know dominate in Indian society.

This is not a random foray into definitions and conceptual grey areas. It has very important implications because it turns out that it is precisely in the category of “unpaid helpers in family enterprises” in which most of the increase in women’s “employment” has occurred.

Growing importance of self-employment—and its problematic nature

Figures 4 and 5 describe this disaggregation of women’s activities by status of employment, and the changes over this recent period, for rural and urban Indian women separately.

It is important to note that these numbers describe different categories of female workers in the age group 15 years and above, and leave out an important category: openly unemployed workers. In other words, these figures refer only to those whom the NSSO considers to be workers: self-employed on own account or as employers (Codes 11 and 12), unpaid helpers in household enterprises (Code 21), regular workers (Code 31) and casual workers (Codes 41 and 51) who together are added up to show total workers. They also show the unpaid and unrecognized workers who engage in domestic duties and related activities (Codes 92 and 93). The somewhat vague category of Code 96 is excluded from this, but they account for less than 2 per cent of women above the age of 1 years. This disaggregation enables an understanding of the change in pattern of activities of women over this period, while excluding that part of the labour force that is looking for but unable to find work (unemployed).

Consider rural women first. By 2023-24, there was a dramatic decline in unpaid workers (Codes 92 and 93) to half of the 2017-18 level. At the same time, there was a significant increase in self-employment, which amounted to 95% of the increase in recorded “employment”.

So the decline in the proportion of unpaid women workers (Codes 92 and 93, who are not included in the labour force by the NSSO) is almost completely explained by the increase in self-employment, including both those serving as unpaid helpers in family enterprises and those working for remuneration on their own account. The share of both regular and casual workers barely increased at all, and, in any case, together they account for less than one-tenth of rural women of 15 years and above.

Figure 4: Rural Women by Status of Work in India, 2017-18 and 2023-24

Figure 5: Urban Women by Status of Work in India, 2017-18 and 2023-24

In urban areas, as shown in Figure 5, the pattern is different. Here, there was an 8 percentage point increase in women’s recorded work participation and a 6 percentage point decrease in unpaid workers. In this case, the increase in women’s recorded employment was split equally between regular workers and self-employed women. This is a positive development, but note that regular workers still account for less than 13% of females aged 15 years and above, and self-employed women only 7.4%.

There was a decline in the proportion of both casual workers and unpaid helpers in family enterprises. And aggregate employment rates for urban women (at 26%) still remain shockingly low by global standards and relative to comparator countries.

This immediately raises two important questions. First, has there simply been some shifting across the two categories (of unpaid helpers in family enterprises or Code 21, and unpaid workers not included in the labour force, or Codes 92 and 93)? This explains around half of the increase in women’s “employment” rates, so this is not an irrelevant question but it is certainly one that is hard to answer since so much depends on the training and instructions given to the field investigators and the actual implementation of the guidelines in what is admittedly a very grey zone. Second, what is the quality of the self-employment that is actually remunerated, which more women seem to be moving into?

Further disaggregation is required to identify the nature of the regular work that has engaged more women before assessing its quality. We know that a significant amount of urban regular work for women is in less desirable and relatively poorly paid activities such as domestic work and precarious informal work without any legal or social protection.

An indicator of the quality of work would be the average wages—and Table 1 points to some disturbing tendencies in terms of the observed trends in real wages.

Table 1: Workers’ Remuneration in India at 2023-24 Prices

Real wages for women regular workers fell in both rural and urban areas, suggesting that much of this was distress employment with likely poor conditions. This could well have been in domestic work and similar occupations, which have been significant sources of employment in urban India for poor women. However, the decline in real incomes from self-employment for women was even greater, once again suggesting a crowding of more women workers into relatively limited types of activities as part of a survival strategy to ensure at least some incomes for their households.

As a result of these unfortunate tendencies, the gender gap in earnings grew significantly over this period, particularly in rural areas, a shown in Figure 6. The gap is shockingly high for self-employed workers, whereby women receive only around a third of the incomes from self-employment that their male counterparts earn.

Figure 6: Female Wages as Share of Male Wages in India

The broad conclusions from these aggregate data are as follows. Real wages declined for both urban and rural regular female workers between 2017-18 and 2023-24. The decline in real incomes from self-employment for women was even greater. The gender gap in earnings grew significantly over this period, particularly in rural areas.

But the main conclusion is important—recent labour force data do not provide a picture of “an equitable, vibrant labour market where women play a leading role in shaping the nation’s future”.

Among self-employed workers, women receive only a little more than one-third of what their male counterparts earn. There is the possible crowding of more women workers into relatively limited types of activities (as regular or self-employed workers) as part of a survival strategy to ensure at least some incomes for their households. There are also substantial increases in unpaid family helpers.

Clearly, much more work needs to be done to delve deeper into unpacking these figures by examining the sectoral decomposition of such employment and work, looking at the regional and state-wise variations (which are large), and so on. But the main conclusion is important—recent labour force data do not provide a picture of “an equitable, vibrant labour market where women play a leading role in shaping the nation’s future”.

If the country is really to experience positive change in this respect, policymakers would do well to move beyond the fairy tale designed to impress public perception to address the realities of stagnating incomes and uncertain, fragile, and low-paying job opportunities, especially for women in India.

Jayati Ghosh taught economics at Jawaharlal Nehru University, New Delhi, for more than three decades and is now professor at the University of Massachusetts at Amherst.

This is a lightly edited version of a paper presented at “Peasants and Workers in Structural Transformations: Perspectives on the Challenge of Industrialisation in the Global South, An International Seminar in Honour of Late Professor Ajit Ghose” organised by the Institute for Social and Economic Change, Bengaluru, and the Institute for Human Development, New Delhi, on 7-8 January 2025 in Bengaluru.

I was honoured to be part of this tribute to Ajit Ghose, a friend of many decades who also was an inspiration in terms of how to conduct economic research. His research was in the best tradition of economics, being analytically grounded yet painstaking and meticulously careful in its empirical assessments and claims. I cannot claim to the same degree of punctiliousness that he achieved, but that was the motivation for this examination of the recent trends in women’s employment in India, particularly self-employment.