Jheelum Sarkar

In less developed economies, frequent climate events continue to threaten both gender equality and stability of local labor markets. This study examines how climate shocks affect gender gap in paid labor outcomes. Using the catastrophic 2018 floods in Indian coastal state, Kerala, this paper employs a difference-in-difference design using a population weighted flood exposure to define treatment districts. Key findings indicate that the flood shock significantly reduces paid labor hours of both men and women, with a disproportionately larger impact on women. These effects are especially pronounced in rural areas. Casual labor is most vulnerable to the flood while self-employment emerged as a key coping mechanism, particularly, among men. Moreover, marital status and dependency burden further shape the extent to which the flood affects paid labor outcomes differently between men and women.

This paper examines whether household internet access affects maternal and reproductive healthcare consumption in India. Using the fourth and fifth rounds of the National Family Health Survey, we study the relationship between household connectivity and healthcare utilization during a period of rapid expansion in low-cost internet access. Results indicate that internet access is associated with higher rates of institutional delivery, skilled birth attendance, postnatal care, and modern contraceptive use. We also find changes in the composition of healthcare consumption: connected households are more likely to use private facilities for delivery and less likely to use public facilities. The results are consistent with internet access reducing information and search frictions, as connected households are more likely to obtain antenatal care from skilled providers, complete at least four antenatal visits, and initiate care earlier in pregnancy. Internet access is also associated with a shift away from spouse-only healthcare decision-making toward greater joint or female participation. The effects are heterogeneous across socioeconomic groups: the association with institutional delivery is stronger among poorer households, while substitution toward private care is more pronounced among wealthier, urban, and better-educated women. These findings suggest that digital connectivity affects both the level and composition of maternal healthcare consumption, with distributional consequences that differ across households.

As digital technologies become more deeply integrated into everyday life, they may shape fertility through competing channels. On one hand, digital activities can displace face-to-face interaction, offline leisure and family time. On the other hand, digital technologies can increase flexible work arrangements, reduce information costs and ease coordinating family life. Does greater digital life correspond to higher or lower fertility trends, and for whom? I construct a state-year Digital Life Index for the United States that captures three dimensions of digitalization: remote work, digital leisure, and access to internet services. Using a two-way fixed effects model, I examine how the lagged digital life index is associated with birth outcomes across reproductive age groups. A one-standard-deviation increase in the lagged index is associated with approximately 2.1 fewer births per 1,000 women ages 30–39 years. The estimates for women in their twenties and forties are statistically insignificant. This negative relationship is more pronounced among married women living with their spouses, women in occupations with higher potential for remote work, and women with postsecondary education less than bachelor’s degree. Spouse’s occupation also plays an important role: the relationship goes in opposite direction depending on whether husbands are in jobs with high potential for remote work or jobs that require face-to-face interaction. The findings suggest that digital life has a negative relationship with fertility trends, but this relationship is heterogeneous by work and family circumstances.

Several developing countries lack task-based surveys at detailed occupation levels. This paper develops new occupation-level telework scores for emerging economies, where the non-pecuniary cost of working outside is very high for women. Using the textual descriptions of 3,448 detailed occupations in the Indian occupation manual, I train a supervised text classifier to estimate a telework score for each occupation. I show that 18.9% of occupations have moderate to high telework potential. Software engineering, product design, management, and research-based occupations exhibit the highest telework potential, while physically intensive occupations such as machine operators and construction workers rank the lowest. In 2022--24, 25.5% of India's workforce was employed in telework-compatible occupations, including 27.9% of men and 19.9% of women. This ranking remains consistent with alternative measures derived from large language model (LLM) analysis. Extending the text classifier framework to two other large emerging economies with detailed occupation manual, namely Brazil and China, I find similar occupations ranking at the top and bottom of the distribution. Applying this occupational classification to 96 other countries, I find that teleworkable jobs are concentrated in countries with higher urbanization, internet connectivity, and income per capita. These findings demonstrate that occupation manuals can substitute for task-based surveys in measuring telework potential, an important input for studying the implications of telework shaping labor markets in emerging economies.