Though small loans were once considered a silver bullet for women’s empowerment, existing evidence shows varying impacts on borrowers across contexts*. In Nigeria, the research team evaluates the impacts of a digital credit product on women clients based on (i) traditional indicators of economic welfare such as women’s employment and business activity, as well as (ii) measures of women’s empowerment such as decision-making authority and control over household finances. This study also investigates how credit scoring could be reoptimized to increase the benefit to women. While financial service providers have traditionally employed loan officers to decide who can borrow, and how much given anticipated repayment rates, digital credit products make lending decisions automatically using algorithms. This lowers the costs of lending, and raises new possibilities to inspect and re-optimize lending decisions to achieve objectives beyond repayment. The research team extends their ongoing work to design and test new “welfare-sensitive” machine learning methods to better understand how lending decisions can balance the predicted (i) economic welfare benefits, (ii) women’s empowerment, and (iii) lender profits. They will then compare their algorithms’ performance to other lending approaches that focus on removing women’s barriers to access, for example by simply “gender-blinding” to prevent discrimination. Results forthcoming.
*A. Banerjee, Karlan, & Zinman, 2015; A. V. Banerjee, 2013; Meager, 2019
Copyright 2019. All Rights Reserved
Design & Dev by Wonderland Collective