Enterprise bankruptcy prediction is a critical task in financial risk management. Traditional methods, such as logistic regression and decision trees, rely heavily on structured financial data, which limits their ability to capture complex relational networks and unstructured industry information. Heterogeneous graph neural networks (HGNNs) offer a solution by modeling multiple relationships between enterprises. However, current models struggle with financial risk graph data challenges, such as the oversimplification of internal financial features and the lack of dynamic imputation for missing external topological features. To address these issues, we propose HGNN-EBP, an enterprise bankruptcy prediction algorithm that integrates both internal and external features. The model constructs a multi-relational heterogeneous graph that combines structured financial data, unstructured textual information, and real-time industry data. A multi-scale graph convolution network captures diverse relationships, while a Transformer-based self-attention mechanism dynamically imputes missing external topological features. Finally, a multi-layer perceptron (MLP) predicts bankruptcy probability. Experimental results on a dataset of 32,459 Chinese enterprises demonstrate that HGNN-EBP outperforms traditional models, especially in handling relational diversity, missing features, and dynamic financial risk data.

Reducing stock price crash risk is vital for capital market stability, particularly in emerging economies such as China. This study investigates whether Employee Stock Ownership Plans (ESOPs) can mitigate crash risk by analyzing panel data from A-share listed firms between 2014 and 2022. In contrast to prior research that has largely centered on managers and controlling shareholders, we highlight employees as active participants in corporate governance. Employing firm, year, and industry fixed effects, together with propensity score matching and instrumental variable techniques, we find robust evidence that ESOPs significantly reduce crash risk. Mediation analyses indicate that this effect operates through reduced agency costs both between managers and shareholders and between controlling and minority shareholders, as well as enhanced corporate productivity. Moderation tests further show that ESOPs are most effective when investor attention is high and when exit threats from non-controlling major shareholders are stronger. Heterogeneity analyses reveal that ESOPs exert greater influence in non-state-owned enterprises, in eastern regions, in firms with higher employee participation, and when shares are sourced from the secondary market. By extending the observation window to nearly a decade and deploying multiple robustness checks, this study provides one of the most comprehensive examinations of ESOPs and crash risk to date. It contributes to the literature by reframing employees as central actors in market stability and offers actionable insights for managers, investors, and regulators seeking to enhance corporate governance and reduce systemic risk.