Insuring what matters most

Economic shocks such as droughts and floods have disproportionate impacts on assets, nutrition, and health for women and girls. This paper develops a model to study investment under risk in settings where women are principally responsible for expenditures on household public goods. It predicts that when women increase their investment, men will share less of their income, which means that women’s assets are the first to be liquidated in the event of a negative shock. Insurance linked to expenditures within women’s traditional sphere within the household has the potential to insulate women’s assets and consumption, increasing their expected returns to investment by increasing their share of household income in the event of a negative shock. To test this prediction, I conduct a lab-in-the-field experiment in Sambury County, Kenya using a tablet-based insurance game and find that women buy significantly more insurance when it is linked to household expenditures. ∗Thanks to Michael Carter, Travis Lybbert, Katrina Jessoe, Ashish Shenoy, Mark Agerton, Jamie Hansen-Lewis, and seminar participants at UC Davis, the University of San Francisco, UC Berkeley, and the 2019 AAEA annual meeting in Atlanta for their helpful feedback. All errors are my own.