Unveiling gender disparities in corporate board career paths using deep learning

Abstract

In this study, we investigate the relationship between professional networks and gender disparities in corporate board appointments, focusing on publicly traded Canadian companies. Using data from over 19,000 senior managers and board members in more than 700 firms from 2000 to 2022, we combine social network analysis with a causal learning framework and long short-term memory (LSTM) models to examine how networks act as both enablers and barriers to achieving gender diversity in leadership. Our findings highlight a clear glass-ceiling effect: female board members must build wider and more influential networks than men to reach similar positions of influence, even when their demographics and career paths are comparable. Gender-specific personalized PageRank further reveals the strong role of female-to-female connections in supporting women’s advancement. This research contributes to a broader discussion of corporate governance and gender diversity, highlighting the need for inclusive networking and mentorship initiatives to reduce existing barriers.