Purpose – The purpose of this paper is to investigate the relationship between board composition and firm performance variance in the context of recent corporate governance reforms, based on the agency and organisational literatures. Design/methodology/approach – This paper uses 384 of the top 500 Australian companies as its dataset. Board composition measures include the percentages of affiliated, executive and independent members on the board. Firm risk is represented by the standard deviation of shareholder return. Several control variables are introduced in the regression analysis. Findings – The results show a negative impact of executive directors on subsequent risk. Affiliated and independent directors, however, have no significant effect on the level of performance variance. Blockholders give a positive influence on firm risk. Moreover, companies with poor dividend payout or low managerial shareholdings tend to be riskier. Research limitations/implications – This paper does not examine the actual risk preference of individual directors, which could involve an attitudinal survey of board members. Future research may also examine the specific attributes towards risk for each type of affiliated directors. Practical implications – The findings cast doubts on the hope that promoting board independence would reduce agency conflicts relating to managerial risk aversion, and support the proposition that, although firms may comply with the demands for more independent directors, they could employ a number of tactics to neutralize the power of outsiders. Originality/value – The empirical work surrounding this topic has been scant. This study may present the first Australian empirical evidence on the relationship between board composition and firm performance variance.