From a theoretical perspective (Vishny and Shleify, 1993), corruption in tax reduces costs. Using firm-level cross-country data, an important paper by Alm, et al. (2016) provides evidence that it is corruption that largely drives higher levels of tax-avoidance. According to the above conclusions, when a country or region takes anti-corruption and tax inspectors no longer dare to accept bribes, will the degree of tax-avoidance of firms naturally decline? The conclusions of this paper show that things are not that simple. 　　In China, private firms can be divided into conservative tax-avoidance private firms and aggressive tax-avoidance private firms according to the income tax rate and the inspection and supervision from the government and other firms. Aggressive tax-avoidance private firms achieve higher degree of tax-avoidance through rent-seeking before the anti-corruption campaign since the 18th Party Congress in China. Using the sample of Chinese listed private firms during 2009-2016, we choose the private firms which are conservative in tax-avoidance as a control group and use the DID method to assess the impact of the recent anti-corruption campaign in China on tax-avoidance of aggressive tax-avoidance private firms. The empirical results show that the recent anti-corruption campaign in China enhances rather than lowers the degree of tax-avoidance of aggressive tax-avoidance private firms. 　　We employ and develop the theory model initiated by Mookherjee and Png (1995) to explain this conflict. We find it is obvious that the degree of tax-avoidance of firms will decline as tax inspectors become less corrupt when information is symmetric between firms and tax inspectors or the inspector intensity of tax inspectors keeps constant before and after the anti-corruption campaign. However, the symmetry of information between the two parties is a very strong assumption. Under the condition of information asymmetry, if the performance incentives for the honest inspection and the penalty for the collusion set by the government meet the conditions of tax corruption, bribes will give tax inspectors stronger incentives to inspect. When the government increases the penalty for corruption of tax inspectors, the latter will choose a lower intensity of inspection, providing a more tax-free environment for aggressive tax-avoiding private firms. In spite that firms choose a lower degree of tax-avoidance in a certain intensity of tax inspection, the equilibrium of firms and tax inspectors may still be manifested as an increase in the degree of firms’ tax-avoidance. 　　Bribery is an inefficient way to motivate tax inspectors. By increasing the penalty for corruption of tax inspectors and providing higher performance rewards or greater promotion prospects, the government can simultaneously achieve the dual goals of eliminating tax corruption and reducing firms’ tax-avoidance. 　　During the anti-corruption campaign in China, the " ouster” of senior tax inspectors will provide unexpected promotion for junior tax inspectors. Using this " natural experiment” condition of the position vacancies of the " ouster” of senior tax inspectors as a measurement of promotion for junior tax inspectors, this paper also examines the impact of promotion prospects of tax inspectors on the degree of tax-avoidance of aggressive tax-avoidance private firms. The empirical results show that ceteris paribus, the promotion prospects of tax law enforcement personnel are negatively related to the degree of tax-avoidance of aggressive tax-avoidance private firms. Promotion prospects have a reverse adjustment effect on the degree of tax-avoidance of aggressive tax-avoidance private firms due to the anti-corruption campaign in China. 　　The main contributions of this paper are as follows: First, the existing literature is less empirically examining how the dynamic changes in the level of corruption of tax inspectors in a country or region affect the degree of tax-avoidance of firms. Using the exogenous shocks provided by the anti-corruption campaign in China, we make up for this deficiency and enrich the relevant literature on anti-corruption (corruption) economics. Second, based on the Mookherjee and Png (1995) model, this paper expands the scope of incentives for client granted by the agent, from simple material incentives to material and political (promotion prospects) incentives. In China’s political system, promotion is a very important incentive for government employees. Our empirical results also show that promotion prospect is an effective incentive to improve tax inspectors’ inspection intensity. This treatment has a stronger explain ability to Chinese economic phenomena.