This study examines the impact of product market threats on the cost of equity capital. Utilizing product market fluidity as a proxy for firm-level competitive threats, we find that heightened competition is associated with a lower cost of equity. We identify operating efficiency and investment efficiency as key channels through which this negative relationship operates. This negative association is more pronounced among firms with severe agency problems, suggesting that competition serves as an external governance mechanism. Furthermore, our results reveal that firm-level competition has an incremental negative effect on the cost of equity, even after controlling for industry-level competition. Overall, our findings highlight that product market competition encourages managerial stewardship, effectively reducing slack and enhancing operational efficiency.
This study examines the effect of increased stock liquidity on the speed of corporate leverage adjustment toward the optimal leverage. We find that overleveraged firms with high liquidity reduce their leverages at a lower speed than that of their low-liquidity counterparts. In contrast, we find that underleveraged firms with high liquidity adjust leverage at a higher speed than that of their low-liquidity counterparts. These empirical results are attributed to the fact that both overleveraged and underleveraged firms with high liquidity face a lower cost of debt when managers can make more informed investment decisions from enhanced liquidity. Our empirical findings shed new light on the importance of stock liquidity in firms’ dynamic capital structure adjustments.
This study examines the association between competitive threats to firms’ major customer relationships and real earnings management (REM). Using the presence of customer-connected peer firms (CCPs) to measure competitive threats, we document a positive association between CCP presence and REM. The CCP/REM association strengthens with poorer external monitoring and financial reporting readability, weakens when firms have major government customers, and is robust to the inclusion of multiple measures of product market competition. Difference-in-differences analysis shows that firms with CCPs decrease their use of REM following CCP bankruptcy filings, which supports our claim that CCP presence motivates REM practices. Additional tests show that REM becomes less responsive to CCP threats when there are higher numbers of CCPs, and REM decreases terminations of major customer relationships involving CCP-shared customers. Our findings are consistent with firms using REM to favorably influence major customer perceptions of their relationship with the firm when customers engage with multiple suppliers.
Generative Artificial Intelligence (AI) is increasingly used for zero-shot text classification in social science, yet its outputs exhibit inherent stochasticity. Because reliability is a necessary condition for validity in content analysis methodology, this stochasticity poses a fundamental challenge, yet no systematic framework exists for quantifying and governing classification reliability prior to validity evaluation. This study proposes the Semantic Stability Protocol, which conceptualizes repeated large language model (LLM) outputs as structured groups of “AI coders” and applies traditional intercoder reliability metrics to assess classification consistency. Using DeepSeek Reasoner to classify 424 Chinese news articles into five categories within a single-model, single-language, single-domain configuration (100 runs per article), we find that raw outputs already exhibit high internal consistency (Krippendorff’s α = 0.8485) and that approximately 20 runs suffice for α > 0.94 after aggregation. Central to the protocol is a stability-stratified escalation framework: two diagnostic indicators, the Majority Rate and the Confidence Gap, partition each classification into High-, Moderate-, or Low-stability strata, triggering differentiated procedures: High-stability cases accept aggregated decisions directly, Moderate-stability cases undergo additional runs to reassess consistency, and Low-stability cases are flagged for human review. This study illustrates that generative model stochasticity can be governed within established reliability frameworks, providing researchers with actionable guidance (minimum run counts, aggregation strategy selection, and stability diagnostics) for transforming zero-shot classification into a transparent, auditable procedure.
This study examines the impact of corporate disclosure on price efficiency by using a comprehensive ranking system that evaluates over 100 measures across five categories of information disclosure for all publicly listed firms in Taiwan. Addressing sample selection bias, we find that firms exceeding a specific disclosure threshold exhibit enhanced stock price efficiency. Furthermore, our path analysis illustrates that domestic institutions play a more pronounced role than foreign institutions in this relationship. These findings from a quasi-natural experiment have noteworthy policy implications, highlighting the necessity for policymakers to consider the extent of corporate disclosure to improve market quality.
