Cao Thi Nhan Anh

Type of the article: Research Article

Abstract
Voluntary disclosure theory predicts that managers delay bad-news disclosure and accelerate good-news disclosure when firm-specific uncertainty rises. However, this prediction may not hold in low-trust frontier markets. This study aims to determine whether lagged idiosyncratic volatility changes the timing of good- and bad-news voluntary disclosure in Vietnam and whether environmental, social, and governance performance, analyst coverage, and state ownership moderate that relation. The study uses hand-collected voluntary disclosures from 210 Vietnamese non-financial listed firms over 2018–2024 and estimates probit models with firm and year-month fixed effects on 5,122 firm-month observations. The results show a reversal of the developed-market pattern. A one-standard-deviation increase in lagged idiosyncratic volatility raises the probability of bad-news disclosure by about 12 percentage points and lowers the probability of good-news disclosure by more than 17 percentage points. Higher environmental, social, and governance performance does not weaken this asymmetry and instead amplifies it; analyst coverage provides no mitigating effect, and the reversal is stronger in state-owned enterprises. Robustness tests using two-way clustering, paired cluster bootstrap, random subsample splits, and placebo volatility confirm the pattern. The findings indicate that disclosure incentives under uncertainty depend on institutional trust and ownership structure, so governance mechanisms effective in mature markets cannot be assumed to operate similarly in Vietnam.

Acknowledgment(s)
This research was funded by Ho Chi Minh City University of Technology and Engineering (HCMUTE), Vietnam, under grant number T2025-143.

Type of the article: Research Article

Abstract
Financial reporting quality is important for maintaining investor confidence, but earnings management remains a persistent concern in developing markets where corporate governance mechanisms are still being strengthened. Vietnam offers a suitable context for this issue because listed firms operate in an environment marked by evolving governance practices, uneven disclosure quality, and concentrated ownership structures. This study examines whether board quality helps limit real earnings management (REM) and accrual-based earnings management (AEM) among Vietnamese non-financial listed firms. The dataset includes 3,697 firm-year observations for companies listed on the Ho Chi Minh Stock Exchange and the Hanoi Stock Exchange from 2017 to 2023. Board quality is captured by an unweighted Board Characteristics Index based on ten board-related attributes. AEM is proxied by performance-matched discretionary accruals, while REM is derived from abnormal cash flows from operations, abnormal production costs, and abnormal discretionary expenses. Panel regression models are estimated, and Feasible Generalized Least Squares (FGLS) is applied to address heteroskedasticity. The main results show that board quality is negatively and significantly related to REM, with a coefficient of –0.0701 and a z-value of –3.19 at the 1% level. In contrast, the relationship between board quality and AEM is negative but statistically insignificant, with a coefficient of –0.0194. These findings indicate that boards are better able to constrain earnings manipulation through operating activities than through accrual choices. For Vietnamese listed firms, stronger board monitoring over real business decisions may help improve the transparency of financial reporting.