With the development of Internet technology and the maturity of cloud storage and computing frameworks, the era of big data is quietly coming. While big data technology improves the speed and convenience of information acquisition, it is also profoundly affecting the economic development of various countries and the strategic layout of the digital economy. Since the Fourth Plenary Session of the 19th Central Committee of the Communist Party of China, China has been committed to modernizing the national governance system and governance capabilities, in response to the advent of the digital economy era. In 2013, the “Third Stage of Golden Tax” big data tax administration system, which is part of the “Twelve Gold” project of China’s e-government affairs, was put into trial operation for the first time in Chongqing, Shandong, and Shanxi, and gradually spread across the country after system optimization and improvement. As an important part of the modernization of national governance in the field of tax, the “Third Stage of Golden Tax” has improved the tax compliance and information environment of companies. It has produced some additional policy spillover effects. However, existing research pays little attention to the governance effect of big data tax administration in corporate finance. Therefore, exploring the impact of the “Third Stage of Golden Tax” big data tax administration system on the stock price crash risk and its mechanism has important theoretical significance and practical value. This paper takes A-share listed companies in Shanghai and Shenzhen Stock Exchanges from 2010 to 2017 as a research sample, chooses the “Third Stage of Golden Tax” tax system reform as a quasi-natural experimental scenario, and uses a multi-period difference-in-differences model to test the impact of big data tax administration on the stock prices crash risk empirically. It is found that big data tax administration significantly suppresses the stock price crash risk. Further exploration of its impact mechanism finds that, through extensive and mutually confirmed information sources and centralized information processing on big data information platforms, big data tax administration has a governance effect by reducing agency behavior behind the camouflage of tax avoidance and improving information quality by reducing manipulate earnings. Research based on the macro and micro governance mechanisms of the capital market finds that the governance effect of big data tax administration is more significant in the sample group with poor audit supervision and poor legal environment. This paper studies on the influencing factors of stock price crash risk from the perspective of big data tax administration, and provides important empirical data support for further exerting the policy spillover effect of big data tax administration.