This paper tests two hypotheses that examine whether choice of auditor can mitigate the agency costs associated with low management ownership. Based on the Warfield et al. {1995) paper, firms with low management ownership are expected to be associated with poorer earnings informativeness and higher discretionary accruals. However, a higher quality auditor would improve earnings quality, and this improvement would he more dramatic for potentially high-agency-cost (i.e.. low-management-ownership) firms, since one way in which an auditor improves earnings quality is by reducing contracting-driven accruals.
The first hypothesis predicts that higher quality auditors are expected to weaken the positive association between earnings Informativeness and management ownership. The second hypothesis predicts that higher quality auditors will weaken the negative association between management ownership' and the magnitude of discretionary accruals.
Warfield et al. (1995, henceforth WWW) use an agency-theory argument to formulate and test the hypothesis that the informativeness of accounting earnings in explaining stock returns varies systematically with the level of management ownership in a company. They find that this informativeness is lower for firms with lower levels of management ownership. The logic that WWW use to justify this result can be summarized as follows: First, high management ownership mitigates some of the potential confiicts between shareholders and managers (specifically those related to moral hazard situations).^ It follows that firms with lower management ownership are likely to rely to a greater extent on earnings-based compensation for managers, that is, the contracting role of the earnings number becomes relatively important vis-k-vis its market-information role (Dhaliwal et al. 11982]; Ayres 119861). Finally, because the contracting role of earnings is more important, compensation-driven discretionary accruals are likely to be higher in absolute magnitude in low-management-ownership firms, driving down the perceived informativeness of earnings. WWW bolster this argument by further results showing that the unsigned magnitudes of their proxies for discretionary accruals are higher for low-management-ownership firms.
In this paper, we show that the WWW result is likely to depend on perceived auditor quality in terms of Big 6 versus non-Big 6 auditors (DeAngelo [1981]; Dopuch and Simunic [1982]; Teoh and Wong [1993]; Gul [1999]). Further, the results obtained in the WWW study would hold more strongly for non-Big 6 auditees than Big 6 auditees. This is under the assumption that Big 6 audit firms are more conservative (Basu et al. [2000]) and perceived by the market to provide higher quality audits (and therefore better monitoring) thereby mitigating the negative information-usefulness effects that arise for firms with low management ownership. Similarly, the observation that higher discretionary accruals are associated with low management ownership is expected to be weaker for firms with Big 6 auditors. However, before testing this primary hypothesis, the WWW study is replicated on our sample. The replication provides further evidence of the external validity of the WWW results.
Earnings reports have multiple users with different information needs, and these may differ on what constitutes a high-quality audit (Demski [1973]). In this paper, the viewpoint is that of the investor. Teoh and Wong (1993) show that earnings-returns coefficients are higher for Big 8 auditees. This is evidence of higher audit quality for these firms, under the assumptions that (a) a high-quality audit is one that translates into high earnings quality and (b) earnings quality is positively related to the association between reported earnings and market returns. Evidence on the validity of our measure in the Australian context comes from Craswell et al. (1995), who report an average audit fee premium of more than 30 percent for Big 6 firms, that they suggest is due to both a reputation effect and industry specialization.
Download
PDF Ebook Audit Quality, Management Ownership, and the Informativeness of Accounting Earnings
