Does audit committee gender diversity moderate the relationship between audit committee size and earnings management? Evidence from South Africa

This study examines how audit committee gender diversity (ACGD) moderates the relationship between audit committee size (ACSIZE) and earnings management (EM) in non-financial firms listed on the Johannesburg Stock Exchange (۲۰۲۰–۲۰۲۴). Using a quantitative quasi-experimental, ex post facto design, data were drawn from the Osiris database and integrated annual reports. ACGD was measured by the proportion of women on the committee, ACSIZE by member count, and EM via both accrual (AEM) and real earnings management (REM). Fixed effects models based on curvilinear analyses tested the hypotheses. Unmoderated curvilinear quadratic models showed ACSIZE had an insignificantly positive effect on AEM and REM, while ACSIZE² showed an insignificantly negative effect, indicating a partial inverted U-shape. The ACGD-moderated model revealed that ACGDACSIZE significantly positively influenced REM, while ACGDACSIZE² significantly negatively affected it, confirming an inverted U-shaped relationship. Graphically, the unmoderated model required ACSIZE of five to six members before AEM and REM declined, whereas the moderated model needed only two to three. Overall, South African boards should enhance women’s representation on audit committees, as gender-diverse, smaller committees appear more effective at limiting earnings manipulation.

کلیدواژه ها:

Audit committee gender diversity ، Audit Committee Size ، Earnings Management ، Curvilinear Quadratic Models