Audit quality information risk and information asymmetry between traders

Publication Type:
Thesis
Issue Date:
2008
Filename Description Size
01Front.pdfcontents and abstract477.94 kB
Adobe PDF
02Whole.pdfthesis32.27 MB
Adobe PDF
Full metadata record
NO FULL TEXT AVAILABLE. Access is restricted indefinitely. ----- This study investigates the effects of audit quality information risk on the information asymmetry between informed and uninformed traders. Audit quality information risk is defined as the risk of errors, misstatements; omissions and frauds in accounts arising from the quality of the audit delivered on these accounts. This research suggests that that audit quality information risk reduces the reliability of a firm's accounting information, which reduces the quality of disclosed accounting information. This in turn increases the information asymmetry between traders by increasing the amount of private information relative to public information, reinforcing informed traders' informational advantage (Levitt, 1998), making traders' beliefs more heterogeneous (Diamond, 1985) and stimulating traders to search for private information (Diamond, 1985; Verrecchia, 2001 ). This study utilises traditional audit quality proxies most supported by the extant literature, based on audit firm size, auditor industry specialisation, audit effort and audit opinion to capture the audit quality information risk and to investigate the influence of this risk on information asymmetry between traders. Using both Australian and US data, this research measures information asymmetry between traders in the stock and options markets based on absolute price difference, absolute volatility difference and absolute difference of long/short ratio of trades between the two markets. In a sample of230 firm-year observations from years 1999 to 2004 in the Australian market, the study finds that industry specialist auditors and unexpected audit fee (the ratio of actual audit fees to expected audit fees) are significantly negatively correlated with information asymmetry between traders. Results suggest that firms choosing industry specialist auditors or firms having higher unexpected audit fees enjoy lower information asymmetry between traders. This supports the hypothesis that lower audit quality information risk (AQIR} is associated with lower information asymmetry between traders. The test results for a sample of 4715 firm-year observations from years 2002 to 2005 in the US market show that Big n and industry specialist auditors are significantly negatively correlated with information asymmetry between traders. This suggests that firms choosing Big n or industry specialist auditors have lower information asymmetry between traders, which supports the hypothesis that lower AQIR is related to decreased information asymmetry between traders. However, unexpected audit fee (UAF) is significantly positively correlated with all information asymmetry measures. Moreover, in both the Australian and US markets, effects of the key AQIR proxies on information asymmetry are unchanged after controlling for the effect of earnings quality (EQ), suggesting AQIR and EQ have separate effects on information asymmetry between traders. Finally, motivated by Khurana and Raman (2004) who conduct a cross-country study to investigate the two drivers of audit quality (an information driver that is tied to the financial reporting credibility notion and an insurance driver that is tied to the litigation exposure notion), the investigation of both Australian and US settings shows that two common AQIR proxies (industry specialist and UAF) are significant in both countries. This suggests that audit quality is driven by information demand as opposed to insurance demand. Overall, the results offer strong support for the proposition that audit quality information risk primarily related to the choice of industry specialist auditors and the level of unexpected audit fees is information driven and lower AQIR lowers information asymmetry between traders.
Please use this identifier to cite or link to this item: