ABSTRACT This paper examines firms' voluntary disclosure of subsidy information under socio‐political scrutiny. In 2015, China enforced a regulation to crack down on “zombie firms” relying heavily on government subsidies to survive, which aroused widespread socio‐political scrutiny of the provision of subsidies. Using a difference‐in‐differences analysis, I find that the zombie firms significantly reduce their voluntary disclosure of subsidies received after the regulatory shock. Cross‐sectional analyses show that the effect is stronger for firms receiving more public attention, firms whose performance is more vulnerable to subsidies, and politically connected firms. Overall, these findings suggest that subsidy recipients tend to adopt a more conservative disclosure strategy on subsidy information to mitigate the impact of socio‐political scrutiny.
Using data from a sample of Chinese listed companies spanning 2014 to 2023, we construct an evaluation index of corporate digitalization using textual analysis and test the impact of digitalization on firms' employee stock ownership plans ("ESOPs"). We find that firm digitalization predicts ESOP adoption. Mechanism tests show that the probability of implementing an ESOP increases in digital corporations that emphasize intellectual capital. In additional tests, we find that firms in the process of digitalization with lower salary levels have a higher probability of implementing ESOPs, and ESOPs lead to higher firm value and levels of innovation. This study highlights the role of employee motivation in the digital economy, providing insights into how firms strategically utilize ESOPs to manage intellectual capital and increase firm value.
Extensive research in earnings management has focused on understanding managerial bias and its impact on disclosure quality. This study extends this literature by introducing a model of earnings management with dual oversight mechanisms. The first is internal supervision, such as internal controls and audit committees, which operates before reports are released. The second is external supervision, such as SEC investigations and litigation, which occurs after they are released. The model shows that although stronger internal supervision reduces managerial bias, it may paradoxically reduce the informativeness of disclosures for investors. In addition, internal and external penalties interact; internal penalties enhance the effectiveness of internal supervision, whereas external penalties weaken it. Together, these results show that disclosure quality depends on the joint operation of supervisory tools and that strengthening internal controls in isolation may not improve outcomes.
We explore the empirical strength of the relation between working capital accruals and working capital utilization efficiency. Using the approach in financial statement analysis textbooks, we model accruals (i.e., investment in working capital) as a function of sales and working capital efficiency (WCE) in the previous and the current periods. We build a model to predict current-period WCE based on firm fundamentals. Our results reveal that these factors correlate with a firm's WCE. We find that incorporating WCE variables and their interactions with sales significantly improves the explanatory power of the accrual model. We also find that the model incorporating these variables does better than other models in predicting future operating income.
We provide evidence that common auditors among lenders and borrowers mitigate the aggravating effect of COVID-19 on syndicated loan pricing. Specifically, a common auditor alleviates lenders' COVID-19 exposure constraints, resulting in a 2.5% decrease in the offered loan spread. This easing effect is magnified by the length of the auditor-lender tenure; it is concentrated in loans between highly exposed lender-borrower pairs and, notably, further facilitates access to loan financing for auditor-connected borrowers. Nonetheless, this does not constitute irresponsible lending behavior based on a comparison of ex post loan performance for borrowers with common auditors versus their non-common-auditor counterparts. Our results highlight an important yet overlooked function of common auditors: their ability to act as a broker between lenders and borrowers during periods of heightened stress.
An important stream of literature spanning decades has drawn conflicting inferences regarding whether accrual accounting serves the purposes prescribed by the Conceptual Framework by conveying information about future cash flows. We help reconcile this literature by demonstrating that tests of the relative abilities of current earnings and cash flows to explain future cash flows require careful interpretation. Tests of relative predictive ability can lead to the inference that accrual accounting does not achieve the aims of the Conceptual Framework precisely because accruals do covary with future cash flows. This phenomenon arises from the statistical regularity that cash flows and accruals covary unconditionally with future cash flows in opposite directions. We show that this opposing covariances phenomenon does not affect the inferences drawn in tests of earnings' (1) incremental predictive ability or (2) relative ability to explain contemporaneous returns. After accounting for this phenomenon, the conclusion is clear and consistent: accrual accounting does indeed contribute to the purposes laid out in the Conceptual Framework. We suggest that future research studying how accrual accounting affects cash flow prediction should exercise care in designing tests and forming conclusions.
We examine how regulatory suspensions of initial public offerings (IPOs) affect venture capitalists' (VCs') post-IPO monitoring of portfolio firms in China. Using a difference-in-differences design that exploits three major IPO suspension episodes as quasi-natural experiments, we provide causal evidence that these shocks divert VCs' attention and are followed by increased tunneling by controlling shareholders in VC-backed firms. Specifically, we document significant increases in related-party intercorporate loans and in the incidence of capital occupation violations after IPO suspension episodes. Evidence from VC interviews, surveys, and board-meeting data further indicates that IPO suspensions reduce VCs' oversight of portfolio firms. The effects are stronger when VCs are more exposed to IPO-related distractions, when VC monitoring would otherwise be more effective, and when controlling shareholders have stronger incentives to tunnel, and they are attenuated when alternative governance mechanisms are present. Overall, our findings highlight the post-IPO governance role of VCs and reveal an unintended governance cost of IPO suspension policies.
We examine the association between CEO narcissism and the likelihood of engaging in related party transactions (RPTs) and its influence on the value implications of these transactions. Furthermore, we investigate the effect of board monitoring on the relationship between CEO narcissism and the value effects of RPTs. We find that CEO narcissism is positively associated with the propensity for RPT engagements. We also highlight the detrimental effect of RPTs conducted by narcissistic CEOs by documenting that RPT engagements have a negative (positive) effect on firm performance at high (low) levels of CEO narcissism. In addition, we show that strong board monitoring moderates the negative influence of CEO narcissism on the value implications of RPTs. Overall, we find evidence of the opportunistic behavior of narcissistic CEOs engaging in value-reducing RPTs. We also underscore the significance of board monitoring in mitigating the adverse effects of CEO narcissism on the RPT-firm performance association.
We examine whether the disclosure and quality of non-GAAP earnings are influenced by CEO political ideology in the United States. We find that Republican-leaning CEOs are less likely to disclose non-GAAP earnings than non-Republican-leaning CEOs. The lower likelihood of non-GAAP disclosure from Republican-leaning CEOs is mitigated in the presence of losses and transitory gains. The CEO political ideology is not associated with the magnitude of total or special items non-GAAP exclusions. However, we find that politically conservative CEOs exclude fewer other expense items among firms that disclose non-GAAP earnings. CEO political ideology does not significantly moderate the association between non-GAAP exclusions and both future operating earnings and future operating cash flows. CEOs' propensities to use non-GAAP disclosure to meet or beat analysts' street earnings expectations when GAAP earnings fall short of analysts' GAAP earnings expectations are not moderated by political ideology. Our results shed light on the relation between non-GAAP reporting practices and CEO political ideology.
We examine the role of intrinsic motivations-psychologically based, non-economic factors-in earnings benchmark beating by focusing on owner-managed firms that are largely free from external pressures from shareholders, analysts, and the media. We observe benchmark beating as instances in which owner-managers decrease their salaries to transform a loss into a profit. Two key results emerge. First, managers' earnings benchmark beating correlates with their private, non-economic benchmark beating, such as earning more than their spouses and marrying near base-ten-year ages. Second, even when accounting for various extrinsic motivations, such as reporting pressures from debt-holders, employees, and suppliers, we find that benchmark beating remains highly prevalent when these motivations are negligible. Overall, our results suggest that reference-dependent preferences from psychology literature complement economic arguments in explaining earnings benchmark beating.
We examine the effect of managerial overoptimism on discretionary disclosure of subjective information, such as earnings forecasts. The market applies a discount upon disclosure to capture the possibility that the revealed subjective expectation is too optimistic. While this discount is correct on average, it is too high (low) for a truly objective (overoptimistic) manager. Consequently, overoptimistic managers disclose more frequently, and their firms are overvalued. We show that higher levels of overoptimism or a greater fraction of overoptimistic managers amplify the market discount, which ultimately reduces overall disclosure in equilibrium.
This study examines whether disaggregated credit ratings offer incremental information about corporate credit risk beyond what is conveyed by aggregated ratings. Using a novel hand-collected dataset of ratings by Morningstar Credit Research, we examine the information value of four disaggregated rating components-business risk, distance to default, cash flow cushion, and solvency score. Controlling for aggregated credit ratings, we find that disaggregated components add value by predicting both the magnitude of future credit risk and key aspects of issuers' future financial fundamentals. Our results further show that the predictive ability of disaggregated ratings depends on issuer-level characteristics, including ex ante financial constraints and the surrounding information environment.
Pay transparency can affect workers' behavior and, consequently, firm outcomes. This study exploits the state-level, staggered enactment of pay transparency laws to examine how pay transparency affects firms' product recalls. We find that firms headquartered in pay-transparency-law states see an increase in product recalls. This outcome suggests that pay transparency may lower workforce quality. Lending further support to this theory, our evidence shows that the increase in recalls is more pronounced for more labor-intensive firms, and those that are more at risk of losing key employees. An additional analysis reveals that pay transparency increases both the likelihood and severity of product recalls. Finally, we document that when pay transparency negatively impacts firms' product quality, the firms' future performance also declines.
Using recently developed proxies for firm-level political risk and earnings manipulation, we test the limited attention theory. Contrary to Hirshleifer and Teoh's core prediction that investor attention is associated with less managerial manipulation, we find that firm-level political risk, serving as a proxy for investor attention, is positively associated with manipulative earnings management, using both accruals and real activities. The results are robust to alternative proxies for political risk and earnings manipulation, various techniques addressing endogeneity concerns, and subsamples of firms with different earnings manipulation incentives. Moreover, we find that the negative relation between earnings manipulation and subsequent operating performance is more pronounced among firms exposed to more firm-level political risk, suggesting that firm-level political risk is associated with managerial incentives for manipulation more than its association with the monitoring that comes with greater attention.
We investigate how narrative disclosures during earnings conference calls (ECCs) provide context for quantitative numeric disclosures in quarterly earnings releases, enhancing their informativeness. Drawing on a large sample of 34,918 quarterly ECCs from 1621 US-listed firms from 2007 to 2020, we extract interpretable textual attributes-such as tone, specificity, novelty, and thematic content-from introductions, analyst questions, and management answers. Using stochastic gradient boosting, we model the nonlinear and interactive effects of these attributes with firm fundamentals and financial metrics. Our results reveal that narrative disclosures substantially improve the ability of firm fundamentals and financial metrics to explain market responses, both immediately and over longer horizons. We complement our analysis with Shapley additive explanations (SHAP) to identify which textual features most influence investor reactions and how their effects vary across contexts. Immediate market responses are primarily attributable to direct cues, such as the tone, while longer term responses increasingly reflect more complex content, such as strategic themes and support activities. These effects are especially pronounced when earnings surprises are either very negative or close to zero, and for firms with lower institutional ownership. They have also become more salient over time, suggesting that shifts in investor attention and analytical sophistication may be reshaping how financial narratives are processed. Our results contribute to understanding the contextual role of narrative disclosures in financial communication and illustrate the value of explainable AI in accounting research.
Generally accepted accounting principles (GAAP) requires dichotomous classification of financial claims as liabilities or equity. Classifying claims is challenging when instruments have attributes of both liabilities and equity (i.e., hybrid instruments). We examine the conditions under which common shareholders assess one class of hybrid instrument-preferred stock-as similar to liabilities or equity. Preferred stock is similar to liabilities because it is senior to common and thus payments to preferred shareholders reduce net assets available to common shareholders, negatively affecting their expected cash flows (dilution perspective). Preferred stock is similar to equity because its payments cannot cause bankruptcy; thereby, having no obligatory effect on cash flows to common (solvency perspective). We identify expected financial distress costs as the entity-level economic characteristic that distinguishes conditions when each of these perspectives is more important to common shareholders' assessments. We predict and generally find that common shareholders change their assessments of preferred stock from negative to zero as expected financial distress costs change from low to high, consistent with a dichotomous classification, first as liabilities, then equity. We also find a positive association between preferred stock and common equity prices for high expected financial distress firms (1) issuing additional preferred stock and (2) 1 year prior to delisting. In both situations, common shareholders would expect an increase in cash flows from preferred stock. Last, we find that common shareholders do not assess preferred stock consistent with US GAAP classification guidance that is based on economic characteristics (i.e., contractual provisions) of the instrument. Standard-setting implications are discussed.
This study investigates the influence of crowdsourced employee reviews on insider trading behavior. Leveraging the staggered timing of first-time employee reviews about their employers on Glassdoor.com, we employ a stacked difference-in-differences (DID) model comparing firms with review initiations to those without reviews during the same period. Our findings reveal a statistically and economically significant decrease in insider trading activities for firms with review initiations. The results hold across a series of robustness tests, including propensity score matching, entropy balancing, and alternative time windows for the DID setting. Further analyses suggest that insiders scale back their informed trading activities and earn lower abnormal trading profits following review initiations. The effect is stronger for firms with higher litigation risk, greater reputational cost, stronger labor market pressure, recent employee exploitation incidents, and when the reviews are more negative. Overall, these findings suggest that crowdsourced employee reviews discipline unethical insider trading behavior.
In this study, we examine the impact of firm-level political risk on the demand for voluntary disclosure. When firms face higher levels of political risk, they not only face greater uncertainty relating to political outcomes but also a number of potentially significant negative consequences that can stem from them. We investigate if firms facing higher levels of political risk experience greater demand for disclosure from investors who are concerned with protecting against costs associated with this risk. Our results indicate that firms facing higher levels of political risk not only experience greater demand for disclosure, in general, but also experience greater demand for political disclosure, in particular. In a subsequent analysis, we investigate whether these findings are concentrated in firms that publicly disclose a relation with a public official (or a relation with someone closely tied to the political circle). We find that the demand for disclosure is concentrated in firms that provide such public disclosure.
A firm's financial disclosures can (i) influence its own investment (own-firm effects), (ii) influence peers' investment directly through the information they convey (contextual peer effects), and (iii) influence other firms indirectly through a chain of strategic investment responses that propagate through the network (endogenous peer effects). Each channel carries distinct implications for disclosure economics, making it essential to quantify its relative influence. We employ a network-based empirical design and financial restatements within a unified framework that addresses well-known challenges in estimating peer effects. We find that firms' investment decisions are tightly linked to their own disclosures. Moreover, disclosure-induced investment spillovers operate predominantly through the endogenous channel (peers' strategic investment responses that propagate through the network) while direct informational spillovers (contextual effects) are economically modest at most. Our estimates of the magnitudes of all three channels differ considerably from prior research, thereby altering the understanding of how financial reporting quality relates to investment.
This study examines the relationship between commercial insurance coverage and corporate innovation activities. Using firm-level data from 2010 to 2019 in China, we find that the coverage of commercial insurance is positively associated with both the quantity and quality of corporate innovation. We further identify three plausible channels through which this effect operates: the risk-sharing channel, the financing channel, and the corporate governance channel. Notably, our results show that firms with higher levels of commercial insurance coverage are more inclined to undertake risky innovation projects. Additionally, the positive relation between corporate insurance and the quantity and quality of innovation is particularly strong for firms in high-tech industries, those facing significant financing needs and greater financial constraints, as well as those with weaker corporate governance. Our findings offer new insights into how the development of the commercial insurance market can promote economic growth by fostering corporate innovation.
