The influence of women’s leadership on the work–life balance of Latin American women microentrepreneurs: a neural network analysis

Purpose – This study aims to explore how four leadership styles – transformational, transactional, servant and authentic – influence the work–life balance of women microentrepreneurs in Peru, a context representative of emerging economies in Latin America. It also examines how age and years of experience moderate these effects. Design/methodology/approach – A quantitative mixed-methods approach was used, combining partial least squares structural equation modeling and artificial neural networks (ANN). The sample consisted of 345 women microentrepreneurs operating in three Peruvian regions. ANN models were used to uncover non-linear and context-sensitive patterns that complement traditional linear modeling. Findings – Transformational leadership significantly reduced both work–family conflict and family–work conflict (FWC), with the highest predictive significance in the ANN analysis. Authentic leadership reduced FWC, especially when combined with greater entrepreneurial experience. Transactional leadership only reduced FWC, while servant leadership showed no significant effect and was associated with increased FWC among older participants. ANN analysis confirmed the relative predictive power of leadership styles, highlighting the superiority of transformational and authentic leadership in mitigating conflict. Originality/value – This study contributes to the underexplored intersection between leadership, gender and work–life conflict in informal and resource-constrained business contexts. By incorporating ANN, it brings methodological innovation, capturing complex and non-linear interactions that are often overlooked in leadership research. These results offer both theoretical insights and practical implications for gender-sensitive entrepreneurship support policies in emerging markets.