[Abstract]:This research investigated whether the formation of audit committees and their characteristics are associated with improved financial reporting quality. Modified versions of the models developed by Jones (1991) and Dechow and Dichev (2002) provided three measures of earnings quality, which were used to proxy for financial reporting quality. The audit committee characteristics investigated were: independence, expertise, activity, size and tenure.Several contributions to knowledge are made by this research. First, this research examined the association between audit committee formation and financial reporting quality. This could not be done in many of the prior studies that used data on companies in the United States (Klein 2002a; Xie, Davidson and DaDalt 2003a; Bedard, Chtourou and Courteau 2004; Vafeas 2005; Yang and Krishnan 2005; Dhaliwal, Naiker and Navissi 2006), where audit committees have been mandatory for companies listed on the New York Stock Exchange since 1978. A large number of public and private sector groups have recommended mandatory audit committee establishment for all Australian listed companies. However, there has been a lack of empirical support for these recommendations and this research provides evidence regarding this association.Second, audit committees are more heavily regulated in the United States than Australia. Given the relative lack of audit committee regulation for Australian companies, Australia represented a richer empirical setting for the examination of the association between audit committee characteristics and financial reporting quality. The use of Australian company data for the selected time period, avoided the confounding effect of regulation on this association.Third, this research used both a modified version of the traditional Jones (1991) discretionary accruals model and the more recently developed accrual estimation error model from Dechow and Dichev (2002) to estimate proxies for financial reporting quality. Most of the prior studies predominantly used the Jones (1991) model, which has been subject to criticism in the literature. Therefore, the use of multiple models provides more powerful tests of the association between audit committees and financial reporting quality. Finally, this research included changes tests in addition to cross-sectional tests to reduce the likelihood of problems with omitted variables.Several conclusions can be drawn from the results. First, there was some evidence that earnings quality measured using the modified Jones (1991) model significantly reduced in the year following audit committee formation, thus providing some support for the notion that the formation of audit committees improves financial reporting quality. However, a comparison of these results with those of tests using earnings quality measures based on Dechow and Dichev (2002) indicates that audit committees appear more effective at reducing opportunistic earnings management, rather than total accrual estimation errors. Second, there was little evidence of a significant association between the characteristics of audit committees and improved financial reporting quality. Consequently, it can be suggested that, once audit committees are established, variations in their characteristics do not significantly affect financial reporting quality.These conclusions provide support for the mandatory audit committee requirement under the Australian Stock Exchange (ASX) listing rules, which became effective from 1 January 2003. However, there are doubts over the usefulness of several aspects of the ASX Corporate Governance Council's recommendations concerning the composition and size of audit committees.
| Identifer | oai:union.ndltd.org:ADTP/220938 |
| Date | January 2007 |
| Creators | Baxter, Peter J. |
| Publisher | University of Southern Queensland, Faculty of Business |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | http://www.usq.edu.au/eprints/terms_conditions.htm, (c) Copyright 2007 Peter J. Baxter |
Page generated in 0.002 seconds