CFO(Chief Financial Officer)意指公司首席财政官或财务总监,负责公司财务监控，并参与公司经营政策的制定和执行，在公司战略的实施和公司治理中起着积极的作用。在市场经济发达的国家，CFO已经从传统的财务功能演变成为企业战略伙伴，一起承担企业的财务报告责任和生产经营责任。
进入21世纪以来，美国证券市场相继爆发的一系列财务造假丑闻均涉及上市公司管理层的欺诈行为。作为公司高级管理人员的CFO，其在公司财务报表真实性方面所扮演的角色引起了广泛的关注。由于国内外大量研究高管人员更换的文献一般集中于CEO（首席执行官）更换与业绩相关性的检验，以及CEO更换影响因素的探讨，对CFO更换的研究较少，因此本文通过研究CFO更换对可操纵应计利润的影响，为我国高层管理人员中CFO更换对财务报表成果的影响提供经验证据。
本文首先对国内外相关文献资料进行总结研究，并运用委托代理理论、契约理论和公司控制机制等理论进行理论分析。然后选择有效的模型，对我国上市公司CFO更换与可操纵应计利润的关系进行了实证研究。选取了与CFO更换样本相匹配的控制公司样本数据与CFO更换公司样本数据进行纵向分析，并综合运用了描述性分析、t检验和Wilcoxon检验等分析方法对样本进行假设检验，得出以下结论：(1)在更换当年，与未发生CFO更换的公司相比，发生CFO更换公司显著地调低了可操纵应计利润。(2)在更换后一年，与未发生CFO更换公司相比，发生CFO更换公司显著地调增了可操纵应计利润。(3)而在更换前一年，发生CFO更换公司并没有如假设那样显著的调增可操纵应计利润。

CFO(Chief Financial Officer) participates in management,and plays an active role in the implementation of corporate strategies and corporate governance. In the developed country, CFO has already developed from the traditional function to the enterprise strategic partnership. CFO has now been elevated to the same level of financial oversight responsibility as was formerly viewed as the responsibility of the CEO.
Since the beginning of 21st century, a series of scandals about the financial fruad have been disclosed that relate to top managers in listed companies. CFOs and the role they play in creating and disseminating accurate or fraudulent financial information have recently attracted wide attention. There has been considerable research regarding the reason for CEO(Chief Executive Officer) turnover and relationship between the top managers turnover and operating performance, yet there has been very little investigation regarding the relationship between new CFOs and reported financial information. Thus, an examination of changes in discretionary accruals around the hiring of a new CFO lends empirical evidence to the ability of the individual CFOs to significantly effect their firms`reported financial statements.
Firstly, this paper investigates the related literatures about top managers turnover at home and abroad, and carries out theoretical analysis based on the Principal-agent theory, contract theory and corporate governance mechanism. Then, we study the relationship between the CFO turnover and discretionary accruals with the expanded Johns model. We choose sixty-seven listed companies that changed CFO from 2003 to 2005 and sixty-seven listed companies as control samples. We draw some conclusions from the statistical tests and the regression analysis: Companies that appoint a new CFO report siginificant reductions in discretionary accruals compared to other non-hiring, and siginificantly increase discretionary accruals in the last year.