By Maurizio Bussolo, lead economist, and Jonah Rexer, economist, at the World Bank Chief Economist Office for South Asia
At roughly 30 percent, female labour force participation (FLFP) in South Asia is among the lowest in the world. In 2023, about 400 million working-age women in the region were out of the labour force. Despite sustained economic growth over the past several decades, women’s engagement with the labour market has remained persistently low. Convergence in education levels, higher labour market returns to education for women, and declining fertility rates have all failed to close the gap vis-à-vis the participation of men. This gender gap represents an immense misallocation of resources, costing between 13-50% of regional GDP.
Read more: Urbanisation vs. tradition: The paradox of female labour force participation in South AsiaStructural change and women’s participation
A recent report from the Office of the Chief Economist for South Asia at the World Bank, Women Jobs and Growth, shows that structural transformations – urbanisation, tertiarisation, and trade openness – support the creation of more job opportunities for women. However, these transformations have not been deep enough, and the job opportunities they have created have not been taken up by women in large numbers. Stubborn supply-side barriers have restricted women from taking advantage of these possibilities.
In historical development experiences, urbanisation has been accompanied by increases of labour demand in historically female-friendly service sectors (Petrongolo and Ronchi 2020). Jobs in these sectors value “people skills” or “brain” over the “brawn” of agriculture and manufacturing. In addition, cities exhibit less restrictive gender norms and therefore less gender discrimination in employment (Hyland et al. 2020). And urban firms are also exposed to greater product and labour market competition, reducing incentives for gender discrimination (Becker 1995; Ashenfelter and Hannan 1986; Cooke et al. 2019). Indeed, we show that in a cross-country sample of 192 countries from 1992-2022, urbanisation is associated with increased relative participation for women (Figure 1A). For every one percentage-point increase in the urban population share, female participation increases by 0.75%, relative to men. However, this relationship is reversed in South Asia: urbanisation is negatively associated with relative female labour force participation (Figure 1B).
This negative relationship is perhaps unsurprising when we consider the large rural-urban gap in female labour force participation. For every country in South Asia, the female participation rate is markedly lower in urban areas. In pooled labour force survey data covering the period 1983-2022, the regional average gap in employment rates between rural and urban women is roughly 16 percentage points, after controlling for age and education. Nor have these gaps closed over time.
In South Asia, administrative definitions of rural and urban can vary substantially across countries and jurisdictions, creating measurement challenges. To account for this, we conducted an additional analysis that defines urbanisation using satellite data on population density and travel times, yielding similar results.
Figure 1. Urbanisation and FLFP globally and in South Asia
Sources: A – Global data from the International Labor Organization (ILO) 1992-2022 database from World Development Indicators (WDI) DataBank. Female sector shares averaged across 209 countries, excluding the SAR countries. Data at the country-year level. B – Data on SAR come from the Global Labor Database (GLD) labour force survey microdata. Data at the subnational unit (state)-survey year level. GLD Survey waves cover Bangladesh, Nepal, India, Pakistan, and Sri Lanka from 1983-2022.
The Urbanisation Paradox: Why are Women Less Likely to Participate in Urban Areas?
This sizable rural-urban gap in FLFP across the region has been the subject of substantial investigation in recent years (Klasen and Pieters 2015; Chatterjee et al 2015). Here, we offer a new explanation: the interaction between social norms and labour market structure. Urban areas offer high-wage opportunities, particularly to educated women, but social norms, even if less restrictive in cities, are actually more binding because women must leave the home and participate in the wage labour market to access these opportunities.
This novel explanation remains relevant even after considering the income effect, or the fact that women can afford to stay at home because of the higher incomes in urban areas. Controlling for the husband’s wage in a comparison of employment rates among urban and rural women only reduces the rural-urban female employment gap to 12 percentage points. Almost three-quarters of the gap is still unaccounted for. This remaining gap is even more puzzling since wage differences between women and men are lower in cities than in rural areas.
To understand this puzzle, consider the decision in a household of who participates in the labour market – whether the husband alone, or both spouses – as the result of a bargaining process where the spouses care about both consumption of private goods and production of a household public good (child rearing). When the man has a comparative advantage in the labour market, the bargaining process produces a division of labour where he participates in the labour market and the women takes care of child rearing. However, when that comparative advantage decreases, the specialisation decreases, too. In other words, when women’s wages move closer to that of men, women’s participation in the labour market should increase. The fact that it does not in urban areas in South Asia suggests the presence of a mechanism that is counteracting the price signal.
Cities, Social Norms and the Structure of Labour Markets
The most plausible mechanism is the interaction between social norms and the urban labour market structure. While social norms in urban areas tend to be less restrictive toward women (Figure 2), this looseness of the urban social environment is deceptive. This is because participating in economic activities in urban areas is different than in rural areas: to work in cities, women must enter more formal wage employment relationships outside the home, while in rural areas they supply their labour to subsistence farming or household enterprises, thus largely at home. That is where norms, even if looser, may “bite” more in urban areas: leaving the house for waged employment opportunities entails potentially contravening strict household domesticity norms (Jalota and Ho 2024), as well as being subjected to unsafe transportation or work environments (Field and Vyborny 2022), street harassment (Amaral et al. 2023), and potential employer discrimination (Buchman et al. 2023, Bussolo et al. 2024, Gentile et al. 2024).
Figure 2. Rural-urban norms differences by country
Sources: Demographic and Health Surveys (DHS). Note: Average level of the DHS gender norms index by country and DHS round. Higher values indicate more conservative gender attitudes. The norms index, which ranges from 0 to 1, measures the strength of conservative gender norms by averaging female responses to four DHS questions on son preference, household decision-making, contraceptive use, and attitudes toward domestic violence.
In other words, the social and physical risks associated with work may well be much higher for women in South Asian cities than in rural areas. That is why are they are less likely to participate in these activities and only partially absorbed into the paid employment sector. This is clearly shown in Figure 3 which highlights that while men’s participation rates remain high in both settings, women’s employment levels and patterns differ significantly across rural and urban areas.
Figure 3. Urban-rural differences in employment status
Sources: The World Bank Global Labor Database (GLD) labour force survey microdata (database). Note: GLD Survey waves cover Bangladesh, Nepal, India, Pakistan, and Sri Lanka from 1983-2022. Figures show the composition of employment for rural/urban and male/female cells, as a share of the total working age population.
Gender inequality in the labour market will not resolve itself as South Asia urbanizes. Women experience substantially higher wages, returns to education and intergenerational social mobility in cities (Asher et al., in progress), but while gender attitudes are more liberal overall than in rural areas, their stronger discouraging effects may explain why female employment remains lower there. Realising the potential of urbanisation to empower women requires broader shifts in social attitudes to ease constraints to gender roles.