Can Annual Report Vagueness Forecast Stock Price Crash Risk?

Abstract

The clarity of annual reports is vital for stakeholder decision-making, but companies may intentionally use vague language to obscure unfavorable information, reflecting their motives for information manipulation. This study investigates the relationship between textual vagueness in annual reports and the risk of subsequent stock price crashes, emphasizing its implications for financial risk management. Employing a novel methodology based on linguistic theories and machine learning to quantify vagueness in Chinese corporate annual reports, we find a significant positive relationship between vagueness and future stock price crashes, supported by robustness checks using the linguistic specificity index. Furthermore, such vagueness leads to negative earnings surprises and deteriorating asset quality, indicating shifts in expectations and a decline in asset quality. Our analysis reveals that these effects are more pronounced during heightened market sentiment and are particularly significant among non-state-owned enterprises. This study demonstrates that vagueness in annual reports can penetrate the potential manipulation of corporate information, highlighting the critical role of vagueness valuation in protecting investor interests and improving market transparency.

Keywords: Annual report vagueness, Stock crash risk, Information manipulation, Text risk indicators