The effect of board diversity on earnings management: evidence from an unregulated capital market

This study investigates the effect of board diversity on accrual and real earnings management in Ethiopia’s unregulated capital market. Before its official launch in 2025 under Proclamation No. 1248/2021, Ethiopia’s capital market was largely unregulated. During this period of limited external oversight, the internal governance mechanism plays a significant role. Given this context, the study draws on agency, resource dependency, and upper echelon theories to conceptualize board diversity into five dimensions of gender, age, tenure, nationality, and educational background. We analyze panel data of manufacturing companies over the period 2014–2023. Earnings management is measured using both accrual earnings management estimated by the Modified Jones Model and real earnings management estimated by the Roychowdhury model. For the data analysis, we employ fixed and random effects regression as the baseline estimation technique. To ensure the robustness of the result, we further implement a two-step Generalized Method of Moments (GMM) and Propensity Score Matching (PSM). The results show that gender, age, tenure, and educational diversity have a significant negative effect on earnings management. Therefore, these findings indicate that boardroom diversity mitigates earnings management by integrating a rich diversity of perspectives, experiences, and expertise. This study makes a meaningful contribution by examining the multidimensional effect of board diversity on both accrual-based and real earnings management in an unregulated capital market. Furthermore, this study has important implications for policymakers, corporate leaders, and investors in emerging markets.