Despite rapid economic growth, declining fertility, and an increase in education of women in India over the past three decades, the female workforce participation rate – proportion of women who are working – in the country continues to remain low. In fact, it has shown a precipitous and persistent decline since 1987.

Figure 1 plots the workforce participation rates for women and men aged 25-60 in India’s rural (Panel A) and urban (Panel B) regions from 1987-2017. This age group is chosen since educational attainment is largely completed by 25 years of age for almost all individuals, thus, ruling out any direct effects of an increase in enrollment in higher education on the workforce participation rates.

Notably, the levels and patterns differ between the two panels. In both rural and urban India, the male workforce participation is much higher than that of females. Male employment rates in this age group have declined slightly from 96% to 94% in rural and from 94% to 91% in urban India.

The female workforce participation in urban India has also fallen marginally from 26% to 24%. However, the most significant decline in the female employment rate comes from rural India, where the female workforce participation has fallen from 54% in 1987 to 31% in 2017.

Supply-side factors

Existing research shows that even within the group of rural women aged 25-60, the fall has occurred mainly among those currently married – almost 90% of all women. Supply-side factors like changing demographic characteristics – increasing education qualifications of married women and men in their household – and rising household income explain the entire decline in female workforce participation from 1987-’99.

However, these account for 14% to at most half of the fall from 1999–2011. A U-shaped relationship between female education and female employment and a negative relationship between education or income of male household members and female employment drives this decline.

A deeper analysis of the change in employment by type of work shows that the decline has only occurred in agriculture. Figure 2 plots the female and male workforce participation in panels A and B, respectively, across sectors and over time in rural India.

The female workforce participation in agriculture fell from 46% in 1987 to 33% in 2011 and further decreased to 23% in 2017. Female workforce participation has also declined slightly from 3.5% to 2.5% in manufacturing. The exception is construction and services, where it has risen by almost 1-1.5%. Male workforce participation has also fallen in agriculture from 77% in 1987 to 64% in 2011, indicating that the reduction has been far greater for women than for men.

Demand-side factors

Given that the decline in female workforce participation between 1999-2011 remains unexplained by supply-side factors and that the fall occurred in agriculture, demand-side factors in the agriculture sector may have contributed.

There is a gendered division of labour in Indian agricultural operations. Female labour is less likely to be used in operations that require physical strength (tilling) and more likely to be utilised in tasks that require precision (sowing, transplanting, and weeding). This results in limited substitutability between men and women in agriculture.

When male and female labour are imperfect substitutes, technological change can have disproportionate gender impacts, such as the threefold rise in agricultural mechanisation (primarily in tilling land) from 1999-2011. For instance, while male labour can be directly substituted in tilling, female labour can be substituted in weeding due to mechanised tilling, leading to restricted weed growth.

In a recent paper, my co-authors and I show that mechanisation led to a significantly greater decline in women’s than men’s labour on Indian farms from 1999–2011. We accomplished this by comparing changes in labour force participation of men and women in regions more suited to mechanization to those less so.

To ensure that our results are not biased due to unobserved factors driving mechanisation, we use exogenous variation in the extent of loamy soil, which is more amenable to deep tillage than clayey soil and therefore more likely to see the adoption of tractor-driven equipment for primary tilling. In terms of magnitude, a 10% increase in mechanised tilling led to a 5% fall in women’s farm labour use, with no accompanying increase in their non-farm employment.

Given an increase in mechanisation by 32% from 1999-2011, the estimates show that almost the entire decline in female labour use in agriculture during this period can be explained by a rise in mechanisation, with other factors remaining constant. We find that this decline is driven by reduced demand for labour in weeding, a task often undertaken by women.

Women missing structural transformation

The aggregate and sectoral trends in WFP, along with a greater decline of WFP for women in agriculture, point to a worrying future for India – women are missing out on structural transformation. While men who exited agriculture have found jobs in other sectors – particularly construction and services – women have not.

Evidence from National Sample Surveys shows that lack of access to jobs near one’s home is one of the significant hurdles women face in gaining access to remunerative employment outside agriculture. Lower female mobility restricts women’s access to non-farm jobs in construction and low-skilled services since these jobs are usually located at a great distance from the village.

Employment data from International Crops Research Institute for the Semi-Arid Tropics surveys show that 32% of males work outside the village, but only 5% of women in the workforce do so. At the same time, less-than-graduate education restricts their access to higher skilled service sector jobs.

Lower female mobility constricts their access to non-farm jobs, limiting the extent to which female workers can cope with adverse income shocks like droughts. Our research shows that women’s workdays are 19% lower than men’s when a drought occurs, driven by the former’s lack of diversification to the non-farm sector.

The urban context

Although there has been only a marginal decline in female workforce participation in urban India, the overall participation of urban women in the labour market remains conspicuously low at 24% vs 90% for men. Research shows that a combination of supply and demand factors explains this low level.

Some of the factors that substantiate the low levels of urban women in the labour market include social norms around the gendered division of domestic work, concerns regarding family status if a woman is working, a higher increase in home productivity of women vis-à-vis their market productivity with an increase in their education levels, and a lack of job opportunities near their homes.

In addition, employed women in India face a penalty in the marriage market. Recent research shows that employed women are 14%-20% less likely to receive proposals from potential suitors in India. Such a penalty is especially prevalent among the higher castes of northern India, where gender norms are more patriarchal.

How can a country of 1.5 billion people fully realise its economic and social potential if 40% of its working-age population do not engage in productive jobs? While the low levels are often attributed to social norms around the gendered division of domestic work and status concerns, the declining involvement of women in agriculture with little diversification to other non-farm sectors points to a lack of mobility and skills.

The limited mobility of women itself can stem from norms around household domestic responsibilities – in which women spend at least seven hours of work per day as opposed to men, who spend only 30 minutes per day. Changing social norms is a slow process, but other barriers like providing safe and accessible transport for women could perhaps induce some women to enter the labour market.

Another avenue to improve female employment rates in India could be providing women with skills that align with the demand in the labour market. As it stands, the low and further declining female employment rates in India not only reduce female agency and bargaining power within the household, but are also detrimental to the country’s income and social growth.

Kanika Mahajan is an Assistant Professor at Ashoka University, India. The research discussed in this article is co-authored with Farzana Afridi, Monisankar Bishnu, Diva Dhar, Taryn Dinkelman, and Nikita Sangwan.

The article was first published on India in Transition, a publication of the Center for the Advanced Study of India, University of Pennsylvania.