In recent years, the accounting profession has faced increased scrutiny because of scandals involving management fraud (e.g., Enron, WorldCom). In response, Statement on Auditing Standards (SAS) #99 has expanded auditors' responsibility for detecting fraud, requiring auditors to gather significantly more information in their assessment of fraud. In addition, the Public Company Accounting Oversight Board (PCAOB) will focus on fraud detection through their inspections of registered accounting firms. In light of the increased emphasis on auditors' responsibility for detecting fraud, public accounting firms face the challenge of improving their fraud detection process, including their assessment of management fraud risk. Decision aids are one way for auditors to improve their assessment of management fraud risk. In fact, several studies from the decision aid literature suggest that aids are useful tools for a variety of tasks, including fraud risk assessment. At the same time, another stream of the decision aid reliance literature, which looks at people's willingness to rely on decision aids, suggests that individuals tend to be reluctant to accept the output given by an aid. Thus, the primary focus of this paper is on uncovering factors that would encourage one to voluntarily use and rely upon a decision aid. Toward that end, 132 senior-level auditors participated in an experiment that examined how several factors (confidence, perceived usefulness, client size, and conformity pressure) affect decision aid usage and reliance. The results show that perceived usefulness and decision aid reliance are significantly related. Further, the results suggest that perceived usefulness affects reliance more than variables examined in prior studies (e.g., confidence). Finally, the results suggest that decision aid usage mediates the relationship between perceived usefulness and reliance. The results of the current study have important implications for research in both the information systems and decision aid reliance areas. First, the study shows that perceived usefulness, a significant construct from the technology acceptance model (TAM) literature, is also a significant factor in determining decision aid usage and reliance. Second, the study further delineates how certain factors affect decision aid reliance.

在美國，會計師界多主張受到不公平的法律對待，一直到現在，他們對於未偵測到的重大財務報表誤述，多需負起「連帶賠償」的責任，因此近二十年來會計師團體一直極力爭取更為寬鬆的「比例賠償」制度，終於在最近幾年得到逐步地改善；但另一方面，投資人則擔心「比例賠償」制度的實施，將會使得投資人在面臨被告經理人沒有能力償還賠償時，無法從被告會計師身上取得補償而降低對其之保護。有鑑於此項議題之爭議不斷，本研究欲藉由檢視會計師法律責任會如何影響經理人的舞弊行為以及會計師的查核努力，以釐清會計師法律責任對審計失敗的影響。 本研究修正Patterson and Wright (2003)的分析模型，進一步以經理人策略性選擇舞弊金額的情況加入賽局模型，探討在審計品質只受會計師查核努力與其同時受到會計師查核努力和經理人舞弊金額的兩種審計情境下，不同會計師法律賠償責任對於會計師最適查核努力和經理人舞弊金額選擇之影響。 □ 本研究發現，無論在何種審計情境下，若會計師被裁定應負擔法律賠償比率與其早先的查核努力水準間有相當程度之敏感性時，則在比例賠償制度下，會計師的查核努力水準將會大於連帶賠償制度下所投入的水準，如此將使得財務報表的查核品質提升，增加對投資人事前的保護。本研究之結果希冀能提供國內主管機關在制訂會計師法律賠償責任時的參考依據。 / In America, the audit profession contends legal practices treat auditors unfairly. Until recently, they have been held jointly and severally liable for undetected material misstatements and have had to pay their own legal fees whether or not they prevail in court. Hence, they have asked for proportionate liability regime in decades and have got some success. A primary concern about proportionate liability is its potential effects on stockholders. Opponents claim that proportionate liability would decrease investor protection when managers are bankrupt due to the reduction in compensatory payments from auditors. Until now, this controversy still exists. To understand how auditor’s legal liability affects audit failure, this study focuses on the impacts of auditor’s legal liability on management fraud and audit effort. Based on Patterson and Wright (2003) analytical model, this study investigates the effectiveness of proportionate liability in reducing the amount of management fraud and the audit failure rate relative to joint and several liability in two strategic audit settings: one that audit quality is affected by audit effort and one that audit quality is affected by both audit effort and the amount of management fraud. My results show that a proportionate liability rule with large marginal liability relief for audit quality can effectively decrease the amount of management fraud and the audit failure rate relative to joint and several liability in both strategic audit settings. This result would provide valuable reference in auditor’s legal liability determination for policy-makers and regulators.