Research Projects

In the context of the remarkably low participation of female candidates in the labor market of India, we are running a randomized control trial experiment to test a method to increase female labor force participation without requiring any costly policy change or any additional burden to the hiring firms. In collaboration with the Confederation of Indian Industries (CII) and a prominent staffing agency (SA), we are exogenously varying the gender composition of the pool of recommended candidates – where recommendations are made to the different hiring firms. The SA recommends a pool of potential applicants to the client firms as and when there are vacancies in the client firms. The treatment is going to be a "balanced gender pool". A "balanced gender pool" is created by the research team with support from the SA where we exogenously vary the pool to increase the ratio of females in the pool while ensuring a “quality threshold” by considering the prior gender distribution of the applicant pool. The males and females so chosen – are expected to be equally competent to perform the task associated with the vacancy. The control will be a business-as-usual recommendation by the staffing agency. The staffing agency has a large repository of candidate databases from where they procure suitable candidates for different job vacancies. We expect this experiment to result in a larger number of female candidates being hired for different job titles by the client firms. Also, the SA has an applicant tracking system (ATS) that helps them capture necessary details throughout the recruitment process. The ATS system captures a large number of outcome variables which are the outcomes of interest for the research. It is expected that the female candidates, thus hired, will continue to perform equally, if not better, than their male counterparts, in different male-stereotyped or non-female stereotyped jobs. Then, we model the process of bias formation of the hiring firms that explains how the intervention may influence the hiring decision of the firm. Finally, we model the performance of the female candidates in our experiment using their “willingness to pay” and their “outside options”. If this experiment is successful – we will be able to question the so-called “statistical discrimination” which is a dominant logic used so far to justify gender discrimination in different jobs. We have derived this intuition from various recent anecdotal evidence where firms have hired females in stereotypically male-dominated jobs and the hired females have outperformed their male counterparts (details below).