Do catastrophic floods change the gender division of labor? Panel data evidence from Pakistan

    Research output: Contribution to journalArticle


    This study examines how large-scale environmental shocks alter the gender division of labor in traditional rural societies. Using individual-level time allocation data from the Pakistan Rural Household Panel Survey 2012, 2013, and 2014, I compare hours spent per week across productive and reproductive activities of more than 8000 adult men and women after the 2011, 2012, and 2013 floods in 76 villages from three provinces of Pakistan. Individual-level panel data is combined with village-level flood inundation depth data collected using remotely sensed (MODIS) satellite imagery. The panel fixed-effects regression results reveal a significant shift in the gender division of labor for both paid and unpaid work in the flood-affected villages. For a 1-m flood inundation depth, men's and women's weekly time use as hired labor increased by ~2 and ~1 h, respectively, women's weekly time allocation on child and elderly care decreased by ~1.5 h, and men's weekly labor allocation for reproductive work and fuel collection increased by ~2 h. Overall, a 1-m inundation depth increased the weekly time allocation for all activities for men by ~6 h and women by ~2 h. Women's time allocation changes were most pronounced for landless and marginal agricultural landowners (<0.5 ha), when compared with medium and large landowners. The favorable shift in the gender division of reproductive work was positively associated with women's decision-making power. Men who believed that both men and women should equally participate in income generation activities were more likely to increase their reproductive work time.
    Original languageEnglish
    Pages (from-to)1-9
    JournalInternational Journal of Disaster Risk Reduction
    Publication statusPublished - 2021


    Dive into the research topics of 'Do catastrophic floods change the gender division of labor? Panel data evidence from Pakistan'. Together they form a unique fingerprint.

    Cite this