【作者】 沈玉清；

【导师】 曾勇； 戚务君；

【作者基本信息】 电子科技大学 ， 企业管理， 2009， 博士

【摘要】 基于过长的审计任期可能会损害审计独立性、从而降低盈余质量的观点，许多国家实行了审计师定期轮换制度。基于相同的考虑，我国也于2004年1月1日开始实施签字注册会计师定期轮换制度。然而该制度的提出还缺乏实证支持，而且轮换的实施效果也有待检验。本文对此进行了相关研究，主要研究内容和结论包括如下三个方面。首先，从签字注册会计师任期(审计师任期)和事务所任期两个层面，实证检验了强制轮换政策实施前审计任期与审计独立性、审计质量之间的关系。研究结果表明审计师任期的延长并非会降低审计独立性，而只有在非十大会计师事务中过长的事务所任期才会伤害到审计独立性。在以审计师对财务危机的公司是否出具持续经营的查核意见作为审计质量良莠的代理变量，我们没有发现审计师任期或事务所任期与审计质量有关联性的证据。然而，一旦以异常营运资本作为分析对象时，我们则发现较长的审计师任期有助于提高审计质量。除此之外，当企业有增加盈余的动机时，我们还发现非四大会计师事务所内的审计师之间有审计经验传承的效果。这种审计经验的累积对于弥补国内会计师事务所自身能力的不足、提高审计质量起到十分重要的作用。我们的研究结论不支持审计师的强制轮换。然后，对审计师强制轮换效果进行检验。利用异常应计和非经常性损益两个指标作为实质的盈余质量的替代变量，我们发现强制轮换后第一年的盈余质量高于自愿性事务所轮换后第一年，但比强制轮换前的盈余质量要低，与当年无须轮换公司、自愿性更换审计师的公司相比无显著差异。利用盈余反应系数来衡量投资者感知的盈余质量，我们发现强制轮换公司的盈余价值相关性低于当年无须轮换公司，而与强制轮换前和自愿性更换审计师的公司相比无显著差异。因此，在我国实施的签字注册会计师强制轮换制度并未实现提高盈余质量的效果。最后，本文对公司治理与审计任期对盈余质量的联合影响进行了较全面的探讨。当以异常应计和盈余稳健性作为衡量盈余质量的指标时，我们发现，对于治理水平较高的公司，事务所任期的延长对盈余质量有正面的影响；而对于治理水平较低的公司，事务所任期长短对盈余质量没有影响；当用盈余反应系数来衡量投资者感知的盈余质量时，我们发现会计师事务所任期对盈余质量没有影响。此外，虽然公司治理水平对盈余稳健性没有显著的影响，但在以异常应计和盈余反应系数作为盈余质量的替代变量时，我们发现公司治理水平越高，盈余质量越高。因此，我们认为应加强对治理水平较差公司的监管，完善其治理结构，通过提高公司治理综合水平来提高盈余质量才是切实可行的治标之本，而仅仅依靠审计师强制轮换制度反而会适得其反。更多还原

【Abstract】 According to that lengthy auditor tenure may impair auditor independence and thusdecrease earnings quality, several countries have implemented audit-partner rotationrequirement. Based on the same argument, mandatory audit-partner rotation is enactedin China on Jan 1, 2004. However, the regulation of mandatory rotation is lack ofempirical support, and the effect of it has yet to be tested. This dissertation makesrelative studies on it including the following main contents and conclusions from threeaspects.Firstly, this dissertation makes an empirical testing on the relation between thetenure of Chinese Signature CPA (audit-partner), audit firm tenure and auditindependence or audit quality prior to the implementation of mandatory auit-partnerrotation. We find that the longer audit firm tenure instead of audit-partner tenure harmsauditor independence, and the negative impact mainly comes from non-big 10 auditfirms. Taking probability of issuing a going-concern audit opinion for distressedcompanies as proxy for audit quality, we find no linear relation between audit-partnertenure or audit firm tenure and audit quality. However, when we turn to use the directionand amount of abnormal working capital accruals as the proxy variable, we find thatlonger audit-partner tenure helps to produce higher audit quality. Moreover, we findevidence that there exists knowledge spillover effect within non-big 4 audit firms whenthe clients have incentive to increase their earnings. Auditor’s accumulated knowledgeand expertise with longer audit firm tenure are more important for non-big 4 audit firmsto compensate their inability and improve audit quality. Our findings do not supportmandatory audit-partner rotation.And then, this dissertation makes a positive test on the effect of mandatoryaudit-partner rotation. Using abnormal accruals and unusual earnings as proxies forearnings quality, we find the earnings quality over the first year of mandatoryaudit-partner rotation is lower than before, but higher than that of voluntary audit firmswitch, not significantly different from that of non-mandatory rotation in the same yearand voluntary audit-partner switch. Taking the earnings response coefficient as a proxy for investor perceptions of earnings quality, we find that the earnings responsecoefficient from returns-earnings regressions subject to mandatory audit-partner rotationis lower than not subject to, not different from that of the year before mandatoryaudit-partner rotation and the voluntary audit-partner switch. So, the implementation ofmandatory audit-partner rotation has not get the achievement of enhancing audit qualityand earnings quality.Finally, this dissertation makes a comprehensive discussion of the joint impact ofcorporate governance and audit firm tenure on earnings quality. Under the condition ofgood corporate governance, taking abnormal accruals and earnings conservatism asproxies for earnings quality, our findings show that longer audit firm tenure can enhanceearnings quality, while under the condition of poor corporate governance, auditor tenuremakes no impact on earnings quality. However, taking earnings response coefficient as aproxy for investor perceptions of earnings quality, audit firm tenure makes no impact onearnings quality. The level of corporate governance makes no significant influence onearnings conservatism. But, using abnormal accruals and earnings response coefficientas proxies for earnings quality, we find earnings quality is higher with higher level ofcorporate governance. Therefore, we think the supervision to the companies of poorcorporate governance should be strengthened and their corporate governance structureshould be improved. Improving earnings quality through raising the level of corporategovernance is a practical strategy to solve the symptoms, but merely relying on themandatory audit-partner rotation maybe come to the opposite.更多还原