Yu Sun, Wei Yuan
Abstract
We investigate a pilot program launched in 2020 by China’s Ministry of Finance, which strengthened accounting supervision in 11 local bureaus, and find that strengthened accounting supervision significantly reduces stock price crash risk in listed firms. By analyzing disclosed texts using generative artificial intelligence, this study reveals that the supervisory effect is more pronounced when pilot bureaus adopt more effective regulatory measures, including enhancing the use of information technology, leveraging external expert pools, strengthening regulatory cooperation, self-rectification of intermediary institutions and establishing closed-loop regulatory mechanisms. However, local economic pressure can weaken the supervisory effect, which is more pronounced in non-state-owned firms and non-cross-listed firms. Further analysis shows that 1) reducing earnings management and improving the accuracy of managerial information disclosure are two key mechanisms; 2) in the short term, stock price crash risk temporarily increases following the implementation of strengthened supervision; and 3) by distinguishing between the “words” and “deeds” of pilot bureaus, this study finds that bureaus with more “deeds” exhibit better supervisory outcomes. This paper provides micro-level evidence on the effectiveness of strengthened accounting supervision and the heterogeneous impacts of specific measures, offering important insights for emerging markets.
Zitationsformat
SUN, Yu; YUAN, Wei. Can strengthening accounting supervision reduce the stock price crash risk? Evidence from the ministry of finance’s pilot program. China Journal of Accounting Research, 2026: 100484.