India’s workforce has fewer women than it did six years ago: no more than 18% in rural areas are employed, compared to 25% in 2011-’12 and 14% in urban areas from 15% in 2011-’12. However, in urban areas, the percentage of women in salaried jobs has increased from 35.6% in 2004 to 52.1% in 2017, but continue to be under-represented compared to their presence in self-employed or casual work, according to an analysis of latest government employment data by researchers at the Azim Premji University.

Both men and women in rural and urban India are out of work, but there is a continuing decline of the number of women in the workforce, the latest government data, collected by the Periodic Labour Force Survey 2017-’18, found.

Four months after a leaked newsbreak of national employment data revealed that India’s current unemployment rate was at a 45-year high, the government released the official Periodic Labour Force Survey report on May 31.

Understanding the survey

The Periodic Labour Force Survey report, released after many controversies, provided quarterly estimates on labour market indicators by gender, education and caste. In 2017, the Periodic Labour Force Survey had replaced the National Sample Survey Office’s Employment Unemployment Survey that had been conducted every five years since 1972-’73.

The Periodic Labour Force Survey report compares current data with the National Sample Survey Office’s Employment Unemployment Survey, but the two differ in terms of methodology as well as the questions asked of the 433,339 persons surveyed in both rural and urban areas. Some argue that the two surveys remain broadly comparable, while others think it is like comparing apples to oranges.

While the sampling strategy has changed between the two surveys, statistical techniques ensure that the estimates of the National Sample Survey Office’s Employment Unemployment Survey and Periodic Labour Force Survey are comparable. Moreover, the broad questions used to capture labour force status remain unchanged between the surveys.

Women on the job

In 1993, nearly 33% of rural women were employed. By 2011-’12, the last time the National Sample Survey Office’s Employment Unemployment Survey was conducted, this had fallen by eight percentage points to about 25%. The proportion of urban women dropped one percentage point to 15% in the two decades to 2011-’12.

The 2017-’18 Periodic Labour Force Survey estimates indicate a further decline. Rural women’s workforce participation has fallen to 18% and the participation of urban women is down to 14%.

More women in regular employment

The vast literature on the falling female workforce participation provides some reasons for this decline. One explanation is that women withdraw from the workforce due to an income effect, that is an increase in the husband’s income which pushes up the household income as written by Kapsos et al, in 2014. Another reason is that women’s domestic and childcare responsibilities constrain them from participating in the labour market as proposed by Chaudhary and Verick in 2014.

Some, such as Kingdon and Unni in a 2001 paper, argue that women are withdrawing from the workforce to pursue higher education, while others like Chatterjee et al, in a 2015 paper contend that there is a lack of good jobs for women.

However, there is some evidence which suggests that educated women with higher skills are demanding better jobs in cities. The few women who do participate in the urban workforce are increasingly employed in regular, salaried jobs.

In 2004, of all the women in the workforce, 35.6% were in the regular salaried category in urban areas. By 2017, their representation had increased to 52.1%. Thus, the Periodic Labour Force Survey report shows that an increasing share of women is now being employed as regular salaried workers, rather than self-employed or casual workers. For self-employed workers, their share in the workforce fell from 47.7% to 34.7% in urban areas between 2004 and 2017. In the same period, the share of women workers in casual labour also declined from 16.7% to 13.1%.

Rural areas saw a similar pattern during this period, with a rise in the share of women in the regular, salaried category and a decline in casual labour and self-employed work. This rising share of women in regular work is a welcome development. Despite this rise, women continue to be under-represented in regular, salaried employment.

We use the representation index – the ratio of the share of men or women in an employment type divided by the share of that gender in the entire workforce – to estimate men and women’s representation in employment categories.

If the ratio is less than one, it means that the group is under-represented and a ratio greater than one indicates the group’s over-representation in that employment category.

We find that women are over-represented in the self-employed category with a representation index of 2.23 and in the and casual work category with a representation index of 1.61. In regular employment type, however, women are under-represented with a representation index of 0.9.

Thus, even if women’s participation as salaried workers has increased over time, they continue to be under-represented. And, when women do work, they are more likely to work as self-employed or casual workers.

Pay gap

The Periodic Labour Force Survey introduced a question to capture data on earnings from self-employment. As a majority of workers in India are self-employed, it was an important addition the previous surveys had overlooked. These data on income now allow a comparison between how much men and women earn across employment categories.

The raw gender gap, measured as the difference in the average earnings of men and women in the regular, salaried category, shows that men earn Rs 4,594.5 more in rural settlements and Rs 3,429.75 more in urban settlements. That is, women earn Rs 70 for every Rs 100 that men earn in rural areas and Rs 80 for every Rs 100 that men earn in urban areas.

We found that among self-employed workers, men earn more than two times what women earn in both rural and urban areas. However, the information on earnings from self-employment is prone to reporting errors and has to be approached with caution. The estimates’ reliability can be ascertained only after comparing it with alternate data sources.

Work hours

The information on hours worked is another addition to the Periodic Labour Force Survey questionnaire. This is measured as the average number of hours worked in a week during the four quarters of the survey period from July 2017 to June 2018.

There are no significant differences between the percentage of men and women available for additional work in rural areas. In urban settlements, however, women workers in the self-employed and regular work categories report more hours available for work than men.

When self-employed, men work 12 hours more than women per week. In regular and casual employment, men work seven hours more than women in a week. However, this does not necessarily imply that women work for fewer hours since women’s unpaid work is not accurately captured. The data on the number of hours worked can be better utilised after understanding women’s unpaid contribution inside and outside the household.

We are now awaiting the National Sample Survey Office’s Time Use Surveys’ data. Once released, it will help present insightful observations on how men and women spend their time on economic activities.

Janaki Shibu and Rosa Abraham are researchers at Azim Premji University.

This article first appeared on IndiaSpend, a data-driven and public-interest journalism non-profit.