We employ georeferenced data and longitudinal household panel survey data to investigate the impact of the dramatic flooding that hit Bangladesh from August-September 2014 on women's employment and empowerment. Development economics models suggest an increase in household members labour supply as a shock-coping strategy. Our difference-in-differences estimates confirm this assumption: women's employment probability increases by approximately 13 percentage points. Correcting for selection bias due to the initial employment status of women, we also find significant increases in the probability of non-employed women entering employment, in the average monthly income of employed women and in the probability of women engaging in autonomous wage-earning activities. Finally, we show that the greater earning capacity of employed women - instrumented by the intensity of flooding in the villages where women live - contributes to raising their bargaining power within the household as measured by the Women's Empowerment in Agriculture Index and by economic decision-making indicators.