National Council of Applied Economic Research
It is a national tragedy that women unable to find work are dropping out of the labour force
If labour force survey data are to be believed rural India is in the midst of a gender revolution in which nearly half the women who were in the workforce in 2004-5 had dropped out in 2017-18. The 61st round of the National Sample Survey Office (NSSO) recorded 48.5% rural women above the age of 15 as being employed either as their major activity or as their subsidiary activity — but this number dropped to 23.7% in the recently released report of the Periodic Labour Force Survey (PLFS). Is this part of a massive transformation of rural lifestyles or are our surveys presenting a skewed picture? If this change is real does it offer a cause for worry?
Incremental decline
Before we turn to examining these changes it is important to note that the drop in work participation by rural women is not sudden. The latest data from the PLFS simply continue a trend that was well in place by 2011-12. Worker to population ratio (WPR) for rural women aged 15 and above had dropped from 48.5% in 2004-5 to 35.2% in 2011-12 and then to 23.7% in 2017-18. In contrast the WPR for urban women aged 15 and above declined only mildly changing from 22.7% in 2004-5 to 19.5% in 2011-12 and to 18.2% in 2017-18.
One can view this drop in the rural female WPR both positively and negatively. If rising incomes lead households to decide that women’s time is better spent caring for home and children that is their choice. However if women are unable to find work in a crowded labour market reflecting disguised unemployment that is a national tragedy.
If the WPR is declining due to rising incomes we would expect it to be located in richer households — households with higher monthly per capita expenditure and among women with higher education. A comparison of rural female WPRs between 2004-5 and 2017-18 does not suggest that the decline is located primarily among the privileged sections of the rural population. Between 2004-5 and 2017-18 women’s WPR declined from 30.6% to 16.5% for the poorest expenditure decile and from 31.8% to 19.7% for the richest expenditure decile. More importantly most of the decline in the WPR has taken place among women with low levels of education. For illiterate women the WPR fell from 55% to 29.1% while that for women with secondary education fell from 30.5% to 15.6%.
This broad-based decline with somewhat higher concentration among the least educated and the poorest is consistent with the industries and occupations in which it has occurred. Decomposing the 24.8 percentage point decline in women’s WPR between 2004-5 and 2011-12 the decline in work on family farms and allied activities contributed the most (14.8 percentage points) followed by casual wage labour (8.9 percentage points) and in work on family enterprises in other industries (2.4 percentage points). These were counter-balanced by a 0.7 percentage point increase in regular salaried work and a 0.5 percentage point increase in engagement in public works programmes such as Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA). Most of the decline — 23.1 percentage points out of 24.8 — came from reduced participation in agriculture and allied activities.
Men’s participation in agriculture has also declined. Among men aged 15 and above 56.1% participated in agriculture in 2004-5 while only 39.6% did so in 2017-18. However men were able to pick up work in other industries whereas women reduced their participation in other industries as well as agriculture — resulting in a lower WPR. Therein lies the conundrum for rural women. Mechanisation and land fragmentation have reduced agricultural work opportunities for both men and women. Other work opportunities except for work in public works programmes are not easily open to women. This challenge is particularly severe for rural women with moderate levels of education. A man with class 10 education can be a postal carrier a truck driver or a mechanic; these opportunities are not open to women. Hence it is not surprising that education is associated with a lower WPR for women; in 2016-17 29.1% illiterate women were employed compared to only 16% women with at least secondary education.
Another clue to the decline in women’s work opportunities rather than women’s desire to work is reflected in the fact that women who are counted as being out of labour force are not simply content to be homemakers but often engage in whatever economic activities they find. Women’s work and family responsibilities rarely fit in neat compartments but household responsibilities do not prevent women from working. Many rural women raise chickens as well as children; husk paddy for sale while daal simmers; and sell vegetables in a market while caring for babies.
The NSSO and PLFS survey design relies on two questions. First interviewers assess the primary activity in which respondents spent a majority of their prior year. Then they note down the subsidiary activity in which individuals spent at least 30 days. If individuals are defined as working by either primary or subsidiary criteria they are counted among workers.
This is a categorisation that serves well in cases where agriculture is the primary activity and various agriculture-related tasks can be grouped together to comprise the 30-day threshold. But as demand for agricultural work declines and women engage in diverse activities their work tends to become fragmented. A woman who spends 15 days on her own field during the sowing period 10 days as a construction labourer and 15 days in MGNREGA work should be counted as a worker using the subsidiary status criteria but since none of the activities exceed the 30 days threshold it is quite possible that interviewers do not mark her as being employed. On-going experimental research at the National Council of Applied Economic Research’s National Data Innovation Centre (NCAER-NDIC) suggests a tremendous undercount of women’s work using standard labour force questions particularly in rural areas.
This is not to suggest that fixing the problem of undercount in surveys is the solution to declining WPRs. The undercount is a symptom of the unfulfilled demand for work. Although women try to find whatever work they can they are unable to gain employment at an intensive level that rises above our labour force survey thresholds. This suggests an enormous untapped pool of female workers that should not be ignored.
Possible solutions
Establishment of the Cabinet Committee on Employment and Skill Development is a welcome move by the new government. It is to be hoped that this committee will take the issue of declining female employment as seriously as it does the issue of rising unemployment among the youth. Not all policies need to be gender focussed. One of the most powerful ways in which public policies affect rural women’s participation in non-agricultural work is via development of transportation infrastructure that allows rural women to seek work as sales clerks nurses and factory workers in nearby towns. If the cabinet committee were to focus on multi-sectoral reforms that have a positive impact on women’s work opportunities the potential gender dividend could be far greater than the much celebrated demographic dividend.
Sonalde Desai is Professor of Sociology University of Maryland U.S. and Professor and Centre Director NCAER-National Data Innovation Centre. The views expressed are personal