This study examines the linkages between discretionary accruals, management ownership and remuneration and non-audit service fees. All findings of the study are based on an extensive analysis of 351 publicly listed firms of Singapore for the fiscal year 2001. Inferential statistics results, using Ordinary Least Squares (OLS) and Two-Stage Least Square (2SLS), reveal three key findings. First, there is a negative association between discretionary accruals and non-audit service fees. Second, managerial ownership positively affects the negative association between discretionary accruals and non-audit service fees. Third, this positive affect is weaker amongst firms with high accounting-based management remuneration. It is also documented that when single-equation estimates are used, audit committee effectiveness has a significant positive (negative) influence on audit coverage (purchase non-audit service fees). Even after fee endogeneity is controlled, the evidence shows that audit committee effectiveness is not associated with the purchase of either audit or non-audit service fees. Findings from single-equation models of audit and non-audit service fees confirm prior research showing a knowledge spillover effect. Consistent with emerging literature, it has been observed that when simultaneous-equations are used, the association between audit and non-audit service fees suffers from simultaneous-equation bias. Thus, consistent with Whisenant et al., (2003) the authors conclude that there is no knowledge spillover between the audit and non-audit service fees. Another key feature of this article is the expansion of very limited literature, investigating linkages between audit committee effectiveness and the audit and non-audit service fees. Finally, the authors infer that failure to control feedback relationship between the audit and non-audit service fees would probably produce spurious findings and inferences.

Several motivations underlie our study. The initial motivation stems from the intense international interest and concern (particularly amongst regulators and corporate governance reformists) about the affect of non-audit services purchased from the incumbent auditor on the quality of the financial reporting process. Levitt (2000, p.1), for example, states that non-audit services “shorten the distance between the auditor and management” such that “independence—if not in fact, then certainly in appearance—becomes a more elusive proposition”.4 Another primary objective extends from the expanding global attention with an audit committee’s oversight role. Various international corporate reform committees (e.g., Blue Ribbon Committee, 1999; Cadbury, 1992; King, 1994, 2002; Corporate Governance Committee (CGC), 2001) stress that the primary responsibility of the audit committee is to formally and regularly review auditor independence. Increasingly, many perceive the provision of non-audit services by the incumbent auditor to its client may inescapably impair the auditor’s independence. The question of how (or even if) audit committees act to preserve auditor independence by balancing the provision of audit and non-audit service fees by the incumbent auditor essentially remains open and largely unexplored. A final motivation for our study is the emerging evidence using single-equation model specifications that indicate knowledge spillovers between audit and non-audit service fees, and vice versa, are overstated due to simultaneous-equation biases.