Government Accounting Supervision and Corporate Demand Uncertainty

Abstract

This paper investigates whether government accounting supervision shapes firms’ demand uncertainty. Using a sample of Chinese A-share listed firms from 2011 to 2024, we construct a dual framework capturing both direct inspections and industry spillovers. The results show that government inspections whether a firm is directly inspected or operates in an industry where peers are inspected significantly reduce sales revenue uncertainty. To uncover the underlying mechanisms, we examine two channels and find that government supervision lowers demand uncertainty by reducing consumer complaints and suppressing earnings manipulation, thereby improving information transparency and stabilizing market expectations. Heterogeneity analyses indicate that the effect is stronger for firms previously involved in financial fraud, those that actively undertake corrective actions, and those not subject to monetary penalties. Overall, the findings highlight the governance value of government accounting supervision and provide policy implications for improving supervisory resource allocation and promoting a virtuous cycle of supervision, rectification, and industry standardization.

JEL Classification number: M48, D81, H83.

Keywords

Government accounting supervision demand uncertainty earnings management consumer complaints

Introduction

Government accounting supervision plays a central role in rectifying accounting practices and enhancing the quality of financial information (T. Zhang et al., 2026). Through regular or ad-hoc inspections, fiscal authorities identify and correct violations and distortions in firms’ accounting treatments, internal controls, and information disclosures, thereby improving the external information environment. Government accounting supervision has become an important external governance force that influences corporate expectations, shapes industry norms, and enhances information quality in capital markets (R. Chen & Chen, 2024). However, despite the relatively mature legal foundation and enforcement mechanisms of the supervisory system, its specific economic consequences for firms’ operating environments and market behavior remain insufficiently explored.

Demand uncertainty refers to the state in which enterprises are unable to accurately forecast the quantity, timing, or structure of market demand for their products or services due to the complex and dynamic interplay of internal and external factors—such as market conditions, consumer preferences, competitor strategies, macroeconomic fluctuations, and policy changes (Banker et al., 2014; Goldberg & Kolstad, 1995; Gomez-Trejos, 2025; Guiso & Parigi, 1999). This uncertainty may manifest in various forms, including sudden surges or drops in demand, disruptions in demand cycles, or frequent fluctuations in customer orders (Baron, 1971; Cohen et al., 2016; Han & Liu, 2025). Demand uncertainty can have wide-ranging impacts on enterprises. From an operational perspective, it may cause imbalances in production planning (Sun et al., 2025). Overestimating demand may lead to inventory buildup, increased capital lock-up, and rising storage costs, and in some cases, asset impairments due to unsold products. Underestimating demand, on the other hand, could result in insufficient production capacity and delayed deliveries, potentially leading to lost market opportunities and damaged customer trust (Begen et al., 2016). From the standpoint of resource allocation, uncertainty complicates decision-making in areas such as raw material procurement, workforce deployment, and capital investment, increasing the likelihood of resource waste or shortages and reducing operational efficiency (Ni et al., 2025). Strategically, persistent demand fluctuations heighten business risk, undermine the stability of long-term planning, and may force enterprises to frequently revise their operational strategies, thereby raising management costs and weakening competitive advantage (Carlton, 1978). Moreover, demand uncertainty may ripple through upstream and downstream segments of the supply chain, triggering broader disruptions and further intensifying operational pressure on the enterprise (Huang et al., 2025; Saoud et al., 2025).

Accurately predicting market demand has long been a challenging task (Leland, 1972). When the quality of a firm’s internal financial information is distorted, it can further intensify external investors’, suppliers’, and customers’ cognitive biases regarding the firm’s future prospects. This, in turn, amplifies fluctuations in market expectations and undermines the stability of corporate operations. Meanwhile, government accounting supervision—an institutional mechanism designed to enhance information transparency and the credibility of market signals (Lyu, 2025)—is increasingly becoming an important tool for stabilizing the external environment in which firms operate. Therefore, investigating whether and how government accounting supervision affects corporate demand uncertainty is not only conducive to deepening the understanding of the economic consequences of external governance mechanisms, but also has significant practical value for optimizing regulatory resource allocation and enhancing firms’ capacity for resilient development. Based on this motivation, this study uses a sample of A-share listed companies in China to systematically examine the impact of government accounting supervision on firms’ demand uncertainty and its underlying mechanisms. The findings show that such supervision significantly reduces demand uncertainty and exhibits a clear industry-level “deterrent spillover effect.” This effect operates through two main channels: “complaint constraint” and “earnings management suppression.” On the one hand, it reduces negative consumer feedback; on the other, it enhances the authenticity of financial reporting—both contributing to the stabilization of market demand expectations. Moreover, the regulatory effects are found to be heterogeneous, with more pronounced impacts observed among firms engaged in financial fraud, those that have actively implemented corrective measures, and those not subjected to monetary penalties.

The contributions of this study are threefold. First, at the theoretical level, this paper directly addresses the gap in existing research within regulatory theory and information economics. Prior literature primarily examines how government accounting supervision affects firms’ internal financial behaviors, governance mechanisms, or capital allocation (Fan et al., 2025; T. Zhang et al., 2026), but pays limited attention to how external regulatory interventions shape firms’ market-side risks by improving the information environment. By incorporating government supervision into the framework of demand uncertainty, this study uncovers the mechanisms through which regulation stabilizes the external business environment. In doing so, it extends the application of regulatory economics and corporate risk management theory, and provides new theoretical evidence for understanding the linkage between “regulation–information transparency–market expectations.” Second, this study develops a two-dimensional supervision indicator system—capturing both direct supervision and industry spillover effects (Supervision1 and Supervision2)—to systematically identify regulatory transmission at both the firm and industry levels. This approach remedies the common neglect of industry demonstration and deterrence effects in prior research and helps bridge the literature on external governance with operations management studies on demand volatility. Third, this paper further uncovers substantial heterogeneity in regulatory effects across different types of firms. The impact of supervision is particularly pronounced among firms involved in financial fraud, firms that actively undertake rectification, and firms that are not subject to monetary penalties. These findings offer empirically grounded policy implications for enhancing the efficiency of regulatory resource allocation and for designing more targeted, firm-specific supervision strategies.

Regulatory Background

Government accounting supervision refers to the oversight and inspection carried out by financial authorities and related institutions. These actions, grounded in national laws and regulations, target the quality of accounting information and the professional conduct of enterprises, public institutions, and accounting firms (Pan et al., 2023). In China, the Ministry of Finance (MOF) is the core supervisory body. It conducts routine inspections through its regional supervisory offices and, when needed, coordinates with other regulators such as the China Securities Regulatory Commission and the National Financial Regulatory Administration. Together, they form a cross-agency and cross-regional supervisory framework (R. Chen & Chen, 2024).

The supervision focuses mainly on the quality of corporate accounting information (Jiang, 2024; Liu et al., 2025). It covers three areas. First, it checks whether accounting records are truthful and complete, and whether issues such as inflated revenue, underreported expenses, or delayed write-offs exist. Second, it identifies accounting fraud, including fabricated transactions, forged vouchers, and manipulated related-party dealings. Third, it evaluates the strength of internal controls, such as the adequacy of financial management and the implementation of internal procedures.

When violations are found, financial authorities may adjust accounts, recover taxes, or impose fines. They can also penalize responsible individuals by issuing fines or revoking accounting qualifications. In severe cases, they will transfer the matter to judicial authorities or other regulators. The MOF also publishes inspection announcements that disclose violations and corresponding penalties. These disclosures increase public oversight and create a deterrent effect. By 2025, the MOF had issued 47 special inspection bulletins.

The overall goal of government accounting supervision is to enforce the Accounting Law of the People’s Republic of China and related regulations. It aims to restore order in the accounting system, reduce information distortion, and improve the quality and transparency of financial reporting (Y. Li et al., 2023; Lyu, 2025). The supervision system also helps regulate the behavior of accounting firms and supports the healthy development of the CPA profession. By improving the credibility of the accounting information environment, it promotes stable capital markets, protects national economic security and public interests, and strengthens overall social trust (Sarwar et al., 2025).

Literature Review and Research Hypotheses

Literature Review

Existing studies have examined the economic consequences of government accounting supervision from multiple perspectives. Overall, the evidence shows that such supervision significantly influences firms’ internal governance, financial reporting quality, and external market performance. For example, inspection activities curb executives’ excessive perk consumption (Pan et al., 2023), reduce stock price crash risk in listed firms (R. Chen & Chen, 2024), and accelerate corporate financialization by strengthening external governance pressures (Fan et al., 2025). In addition, government accounting supervision improves the transparency and coordination of supply chains by standardizing financial reporting and disclosure practices, thereby reducing supply chain risks (Lyu, 2025). At the operational level, supervision promotes better internal control and resource allocation, contributing to higher total factor productivity (T. Zhang et al., 2026). In capital markets, enhanced information accuracy and verifiability resulting from supervision improve analysts’ earnings forecast precision (Liu et al., 2025). Taken together, the literature consistently highlights the important role of government accounting supervision in strengthening governance, regulating corporate behavior, and improving the information environment.

In contrast, studies on demand uncertainty mainly focus on its economic consequences (Belloni et al., 2017; Chang et al., 2024; Fuss & Vermeulen, 2008). Research on the determinants of demand uncertainty is relatively limited. Existing evidence suggests that internet water armies distort information environments and disrupt consumer perceptions, significantly increasing firms’ demand uncertainty (Xin et al., 2025). However, the overall understanding of how demand uncertainty forms remains insufficient.

More importantly, prior literature has not investigated the role of government accounting supervision in shaping firms’ external operating environments, and research on how such supervision affects demand uncertainty is virtually absent. This paper aims to fill this gap.

Research Hypotheses

Government accounting supervision aims to improve the quality of accounting information and enhance the transparency of financial reporting, thereby mitigating information asymmetry between firms and the market. When firms are inspected by the Ministry of Finance and required to make corrections, the authenticity and compliance of their financial information improve, leading to more stable market expectations. In addition, the institutional pressure generated by accounting supervision encourages firms to strengthen their operational practices and reduce compliance risks, which helps align market participants’ assessments of future business prospects. Therefore, we propose:

Hypothesis H1: Government accounting supervision significantly reduces firms’ demand uncertainty.

Government accounting supervision may reduce firms’ demand uncertainty by lowering the number of consumer complaints. First, supervision improves the truthfulness and transparency of financial disclosures (Lyu, 2025), reducing complaints triggered by misleading information. Once misconduct such as falsified accounting, concealed costs, or earnings manipulation is identified and corrected, the firm’s external information environment becomes clearer, helping restore trust among consumers and the broader market. Second, the volume of complaints itself reflects the stability of a firm’s operations. Frequent complaints often indicate weak product quality or unreliable service, which increases demand volatility (Grosz & Raval, 2025; Y. Zhang & Huang, 2025). When complaints decline, customer stickiness and market expectation stability increase, and product demand becomes more predictable (Arora et al., 2025; Y. N. Li et al., 2025; Nugroho & Wang, 2024). Based on this logic, we propose:

Hypothesis H2a: Government accounting supervision reduces firms’ demand uncertainty by lowering the number of consumer complaints.

Government accounting supervision may also influence demand uncertainty by curbing earnings manipulation. The Ministry of Finance’s inspections focus on identifying practices such as fabricated revenue, premature revenue recognition, and delayed expense recognition. Supervisory pressure encourages firms to enhance financial discipline and reduces both the incentive and the ability to engage in earnings manipulation (Roychowdhury, 2006; Y. Wang et al., 2023). Moreover, earnings manipulation distorts external information and causes investors, consumers, and supply-chain partners to misjudge a firm’s true operating conditions, which amplifies market fluctuations and increases demand uncertainty (Adebambo & Yan, 2018; Magnuson, 2011; Modigliani & Cohn, 1979; Walker, 2013). When manipulation decreases and information becomes more accurate and credible, market expectations about the firm’s future operations become more stable and rational, resulting in less volatile demand. Therefore, we propose:

Hypothesis H2b: Government accounting supervision reduces firms’ demand uncertainty by suppressing earnings manipulation.

Data and Model

Data

This study focuses on A-share listed companies on the Shanghai and Shenzhen stock exchanges from 2011 to 2024, using panel data for empirical analysis. To ensure the representativeness of the sample and the validity of the data, the following steps were taken for data selection and processing: financial firms were excluded; companies under special treatment (ST) were removed; firms with missing values for key variables were dropped. In addition, all continuous variables were winsorized at the 1st and 99th percentiles to mitigate the influence of outliers on regression estimates, thereby enhancing the stability of variable distributions and the economic interpretability of the results. The final sample comprises 35,404 firm-year observations.

The dependent variable in this study is the firm’ s demand uncertainty (Uncertainty), which measures the degree of unpredictability in market demand fluctuations faced by the firm during its business operations. Given the significant difficulty in directly observing demand uncertainty, this study follows existing literature by identifying the unpredictable component from the volatility of the firm’s sales revenue to construct a robust measurement. Specifically, we first use quarterly financial data to build a regression model where historical sales revenue, inventory changes, and seasonal factors serve as explanatory variables to capture the predictable portion of the firm’s quarterly sales revenue. This model effectively filters out sales fluctuations caused by the firm’s own operational patterns and cyclical factors, treating the regression residuals as the uncontrolled and unpredictable part of sales revenue. Then, at the annual level, we calculate the standard deviation of the quarterly regression residuals for each firm, defining this metric as Uncertainty1 to represent demand uncertainty. A higher standard deviation indicates more severe and uncontrollable demand fluctuations faced by the firm within that year, that is, higher demand uncertainty. Through this construction method, Uncertainty1 captures well the instability in firm-level demand caused by external market fluctuations, changes in consumer behavior, and other factors, thereby helping to reveal the level of environmental uncertainty encountered by firms in their decision-making processes.

The core explanatory variable of this study is government accounting supervision, measured through two constructed indicators—Supervision1 and Supervision2—which capture the direct regulatory effect and the industry-level spillover effect, respectively. First, Supervision1 identifies whether a firm is directly subject to inspection pressure from the Ministry of Finance. This variable is essentially a staggered DID indicator. Following Lyu (2025) and Fan et al. (2025), we define a post-inspection dummy Post, which equals 1 for the year in which a listed firm is inspected and publicly reported to have accounting problems, and remains one for all subsequent years; it equals 0 beforehand. We also define a treatment dummy Treat, which equals 1 if the firm is ever reported to have accounting issues during the sample period, and 0 for firms that are never inspected or never found to have problems. The core variable Supervision1 is constructed as the interaction Treat × Post, reflecting the institutional pressure triggered by direct regulatory intervention.

Second, to capture potential spillover effects of government accounting supervision on non-inspected firms within the same industry, we construct Supervision2. When a firm belonging to a corporate group is inspected in a given year and publicly disclosed to have problems, we assign Supervision2 = 1 to all other firms in the same industry and in the same year; otherwise, it is set to 0. This variable is an industry-year dummy that captures the “warning effect” or “deterrence effect” generated by regulatory actions—that is, whether peer firms adjust their strategies or managerial behaviors in response to supervision targeting comparable firms.

By incorporating both Supervision1 and Supervision2, this study systematically identifies the direct impact of government accounting supervision as well as its indirect industry-level spillover effects, providing a more comprehensive and nuanced perspective on the transmission mechanisms of regulatory policies.

This study incorporates a set of firm-level control variables that prior literature has widely shown to be closely related to corporate behavior and firms’ responses to external environments. Specifically, the controls include firm size (Size), leverage (Lev), return on assets (ROA), cash flow ratio (Cashflow), inventory ratio (INV), fixed asset ratio (FIXED), proportion of independent directors (Indep), CEO duality (Dual), and ownership concentration measured by the shareholding ratio of the top 10 shareholders (Top10). Descriptive statistics are reported in Table 1, and detailed variable definitions are provided in Appendix Table 1.

Table 1.

Descriptive Statistics.

VarName	Obs	Median	Mean	sd	Max	Min
Supervision1	35,404	0.000	0.050	0.218	1.000	0.000
Supervision2	35,404	1.000	0.521	0.500	1.000	0.000
Uncertainty1	35,404	0.131	0.184	0.172	2.239	0.000
Size	35,404	22.062	22.236	1.207	26.309	19.854
Lev	35,404	0.404	0.411	0.193	0.904	0.051
ROA	35,404	0.037	0.038	0.056	0.215	−0.226
Cashflow	35,404	0.047	0.048	0.060	0.235	−0.155
INV	35,404	0.109	0.131	0.105	0.663	0.000
FIXED	35,404	0.182	0.208	0.144	0.674	0.002
Indep	35,404	36.360	37.627	5.059	57.140	33.330
Dual	35,404	0.000	0.302	0.459	1.000	0.000
Top10	35,404	0.583	0.577	0.146	0.904	0.225

Model

To investigate the impact of government accounting supervision on firms’ demand uncertainty, this study constructs the following regression model:

\begin{matrix} Uncertaint y_{it} = α + β^{*} Supervisio n_{it} + λ Control s_{it} \\ + fir m_{i} + yea r_{t} + ε_{it} \end{matrix}

(1)

Uncertainty represents the firm’s demand uncertainty, and Supervision denotes government accounting supervision. Standard errors are clustered at the firm level. Our main focus is on the coefficient of Supervision. If this coefficient is significantly negative, it indicates that, controlling for other variables, government accounting supervision can significantly reduce firms’ demand uncertainty. This result would thus support our Hypothesis 1, which posits that government accounting supervision plays a role in stabilizing market expectations and mitigating external uncertainty faced by firms.

Regression Results

Baseline Regression

The baseline regression results are presented in Table 2. Columns (1) and (2) use Supervision1 as the core explanatory variable, while Columns (3) and (4) use Supervision2. In Columns (1) and (2), the coefficient of Supervision1 is significantly negative, indicating that when a firm’s parent group is inspected by government fiscal authorities and publicly reported for violations, the firm’s demand uncertainty declines significantly. This finding supports Hypothesis 1, which posits that government accounting supervision mitigates corporate demand uncertainty.

Table 2.

Baseline Regression.

	(1)	(2)	(3)	(4)
Variables	Uncertainty1	Uncertainty1	Uncertainty1	Uncertainty1
Supervision1	−0.045*** (−4.180)	−0.045*** (−4.348)
Supervision2			−0.011** (−2.071)	−0.013** (−2.288)
Size		−0.006* (−1.812)		−0.006* (−1.787)
Lev		0.040*** (2.907)		0.041*** (2.932)
ROA		−0.091*** (−3.379)		−0.092*** (−3.406)
Cashflow		−0.095*** (−4.913)		−0.094*** (−4.906)
INV		−0.080*** (−3.440)		−0.080*** (−3.447)
FIXED		−0.019 (−1.093)		−0.020 (−1.150)
Indep		−0.001*** (−4.095)		−0.001*** (−4.097)
Dual		−0.001 (−0.449)		−0.002 (−0.472)
Top10		0.082*** (4.838)		0.082*** (4.820)
_cons	0.186*** (348.117)	0.331*** (4.201)	0.190*** (66.126)	0.334*** (4.225)
Year-fixed	Yes	Yes	Yes	Yes
Firm-fixed	Yes	Yes	Yes	Yes
N	35,404	35,404	35,404	35,404
R ²	.455	.458	.455	.458

Note. Columns (1) and (3) control only for firm and year fixed effects, while columns (2) and (4) additionally include firm-level control variables.

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Moreover, the effect is not only statistically significant but also economically meaningful. According to the descriptive statistics in Table 1, the standard deviation of Uncertainty1 is 0.172. The baseline coefficient of Supervision1 is −0.045, which corresponds to approximately one quarter (about 25%) of the standard deviation of demand uncertainty. This suggests that the regulatory pressure generated by government accounting supervision has a substantial stabilizing effect in practice and can materially improve the demand information environment faced by firms.

In addition, Columns (3) and (4) examine Supervision2, which captures industry-level spillover effects. The coefficient of Supervision2 is significantly negative, indicating that even when a firm is not directly inspected, regulatory actions targeted at other firms within the same industry can still indirectly reduce its demand uncertainty. This finding reveals that government accounting supervision generates demonstration and deterrence effects at the industry level by improving the overall information environment, thereby stabilizing market expectations for non-inspected firms.

Taken together, the baseline results consistently show—across both direct supervision and indirect spillover channels—that government accounting supervision effectively reduces corporate demand uncertainty, with effects that are both statistically significant and economically meaningful.

Robustness Tests

Parallel Trends Test

To ensure that the Ministry of Finance’ s random inspections of accounting information quality serve as a valid and exogenous policy shock, this study further conducts a parallel trends test. The purpose of this test is to verify whether the treatment and control groups exhibited similar trends in demand uncertainty before the policy implementation, thereby ruling out potential biases caused by other unobserved factors. Figure 1 presents the results of the parallel trends test within the event study framework. Specifically, the figure shows the coefficients of the time dummy variables for several years before the inspection and their 95% confidence intervals. It is clearly observed that the coefficients in all pre-inspection periods fluctuate around zero, and their confidence intervals cover zero, indicating no significant difference in the levels or trends of demand uncertainty between the two groups prior to the policy implementation. This satisfies the core requirement of the parallel trends assumption. Meanwhile, within the 3 years following the inspection, the coefficients of the time dummy variables are significantly negative and their confidence intervals do not include zero, demonstrating that the Ministry of Finance’s random inspections of accounting information quality have indeed produced substantive effects, effectively reducing firms’ demand uncertainty.

Figure 1.

Parallel trends test.

Bacon Decomposition

This study employs the Bacon decomposition method to break down the overall estimator. The Bacon decomposition can decompose the total effect estimated by the staggered DID into a weighted average of several subgroup comparisons, helping to reveal the estimation weights and corresponding effect differences across different time periods and treatment statuses, thereby assessing whether the estimation results contain potential biases. Table 3 presents the detailed results of the Bacon decomposition. The “Treatment vs. never treated” comparison accounts for the largest share, with a weight of 95.8%, and its corresponding average difference-in-differences estimate is −0.045. This is highly consistent with the coefficient of the Supervision1 variable in the baseline regression, indicating that the main identification effect comes from the comparison between firms that have been regulated and those that have never been regulated, and this part of the estimate is robust and representative.

Table 3.

Bacon Decomposition.

	(1)	(2)
	Uncertainty1	Uncertainty1
Variables	Weight	Average DD estimator
Earlier treatment vs. later control	0.002	−0.052
Later treatment vs. earlier control	0.003	−0.075
Treatment vs. never treated	0.958	−0.045
Treatment vs. already treated	0.037	−0.052

Replacement of the Dependent Variable

To test the robustness of the baseline regression results, this study draws on the methodologies of J. Wang et al. (2025) and Ghosh and Olsen (2009) by adopting an alternative measure of firms’ demand uncertainty—namely, the industry-adjusted standard deviation of sales revenue over the past 5 years, denoted as Uncertainty2. This indicator not only reflects the volatility of a firm’s sales revenue over a longer time horizon but also removes industry-level systematic effects, thereby more comprehensively capturing the demand uncertainty faced by the firm itself.

Table 4 presents the regression results using Uncertainty2 as the dependent variable. The results show that whether Supervision1 is used as the core explanatory variable in columns (1) and (2), or Supervision2 in columns (3) and (4), the coefficients are significantly negative. This indicates that government accounting supervision, whether through direct regulation or industry demonstration effects, can significantly reduce firms’ demand uncertainty.

Table 4.

Replacement of the Dependent Variable.

	(1)	(2)	(3)	(4)
Variables	Uncertainty2	Uncertainty2	Uncertainty2	Uncertainty2
Supervision1	−0.270** (−2.477)	−0.244** (−2.295)
Supervision2			−0.101** (−2.167)	−0.100** (−2.177)
Size		0.150*** (3.453)		0.150*** (3.470)
Lev		0.465*** (3.222)		0.466*** (3.228)
ROA		0.021 (0.093)		0.010 (0.046)
Cashflow		−0.286* (−1.718)		−0.288* (−1.734)
INV		−0.396* (−1.793)		−0.392* (−1.777)
FIXED		−0.291 (−1.438)		−0.295 (−1.458)
Indep		−0.007*** (−2.796)		−0.007*** (−2.795)
Dual		−0.024 (−0.771)		−0.024 (−0.782)
Top10		1.832*** (8.788)		1.832*** (8.781)
_cons	1.244*** (200.217)	−2.917*** (−2.972)	1.288*** (47.323)	−2.881*** (−2.947)
Year-fixed	Yes	Yes	Yes	Yes
Firm-fixed	Yes	Yes	Yes	Yes
N	23,971	23,971	23,971	23,971
R²	.442	.459	.442	.459

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Propensity Score Matching

To further eliminate sample selection bias and verify the robustness of the impact of government accounting supervision on firms’ demand uncertainty, this study employs the Propensity Score Matching (PSM) method for treatment effect estimation. Specifically, radius matching is used to ensure comparability between the treatment and control groups on key covariates. Table 5 reports the regression results after matching. The coefficients of Supervision1 and Supervision2 in columns (1) and (2) are both significantly negative, indicating that the negative effect of government accounting supervision on reducing firms’ demand uncertainty remains significant.

Table 5.

Propensity Score Matching.

	(1)	(2)
Variables	Uncertainty1	Uncertainty1
Supervision1	−0.045*** (−4.329)
Supervision2		−0.013** (−2.295)
Size	−0.006* (−1.838)	−0.006* (−1.808)
Lev	0.040*** (2.853)	0.040*** (2.926)
ROA	−0.092*** (−3.404)	−0.093*** (−3.434)
Cashflow	−0.094*** (−4.880)	−0.094*** (−4.885)
INV	−0.080*** (−3.447)	−0.081*** (−3.494)
FIXED	−0.019 (−1.060)	−0.021 (−1.162)
Indep	−0.001*** (−4.105)	−0.001*** (−4.087)
Dual	−0.001 (−0.353)	−0.002 (−0.481)
Top10	0.084*** (4.904)	0.082*** (4.809)
_cons	0.334*** (4.207)	0.336*** (4.255)
Year-fixed	Yes	Yes
Firm-fixed	Yes	Yes
N	35,265	35,397
R ²	.458	.458

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Two-Stage Heckman Test

To further verify whether the impact of government accounting supervision on firms’ demand uncertainty is affected by sample selection bias, this study employs the classic two-stage Heckman test for robustness checking. The Heckman method effectively controls for endogeneity caused by non-random sample selection, thereby enhancing the reliability of the estimation results.

In the first-stage selection model, the number of local listed companies is introduced as an instrumental variable (IV) to explain the likelihood of a firm being randomly inspected by the Ministry of Finance (Fan et al., 2025). The rationale behind this variable is that within the same region, the more listed companies there are, the more limited the fiscal inspection resources become, which reduces the probability of any single firm being inspected. Therefore, this variable is correlated with the probability of inspection but does not directly affect the firm’s demand uncertainty, meeting the requirements for an instrumental variable.

The test results are presented in column (1) of Table 6, showing that a higher number of local listed companies corresponds to a lower probability of a single company being inspected. The inverse Mills ratio (IMR) estimated in the first stage is included in the second-stage model for further testing. The results in column (2) indicate that the IMR coefficient is not significant, suggesting no obvious sample selection bias in this study. Meanwhile, the coefficient of Supervision1 remains significantly negative in the second-stage regression, indicating that the Ministry of Finance’s random inspections of accounting information quality can significantly reduce firms’ demand uncertainty.

Table 6.

Two-Stage Heckman Test.

	(1)	(2)
	First Stage	Second Stage
Variables	Supervision1	Uncertainty1
IV	−0.588** (−1.973)
Supervision1		−0.104** (−2.362)
IMR		0.030 (1.355)
Size	0.214*** (6.424)	−0.006 (−1.481)
Lev	0.230 (1.102)	0.036** (2.459)
ROA	−0.498 (−1.115)	−0.057** (−2.034)
Cashflow	0.441 (1.092)	−0.077*** (−3.716)
INV	−0.126 (−0.356)	−0.065*** (−2.582)
FIXED	−0.747** (−2.401)	−0.032 (−1.623)
Indep	0.002 (0.319)	−0.001*** (−4.675)
Dual	−0.122* (−1.853)	−0.003 (−0.844)
Top10	−0.794*** (−3.522)	0.090*** (4.930)
_cons	−6.024*** (−6.495)	0.325*** (3.757)
Year-fixed	Yes	Yes
Firm-fixed	Yes	Yes
N	29,364	29,072
R ²	.118	.471

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Mechanism Analysis

Mechanism Analysis: Complaint Constraints

In recent years, with the development of digital supervisory tools, consumer protection channels have continued to expand. Among them, the “Black Cat Complaint” platform launched by Sina has become one of the leading online platforms for consumer rights protection in China, serving as an important channel for the public to voice dissatisfaction and for firms to undergo reputation monitoring (Y. Chen et al., 2025; Shang et al., 2024). Following the approach of Lu et al. (2025), this study uses web-scraping techniques to collect all valid complaint records from the Black Cat platform since its launch. We then match these raw complaint data to Chinese A-share listed firms based on firm names and industry classifications, constructing the annual number of complaints received by each firm (Complain).

Table 7 presents the results of the mechanism analysis. In column (1), the coefficient of Supervision1 on Complain is significantly negative, indicating that when a firm (or its parent group) is inspected by the Ministry of Finance and publicly reported for violations, its annual number of consumer complaints decreases notably. This suggests that government accounting supervision may encourage firms to improve accounting transparency, standardize business processes, and enhance service quality, thereby strengthening consumer trust. In column (2), after including Complain in the regression for demand uncertainty, its coefficient is significantly positive, implying that firms with more complaints face higher demand uncertainty. Frequent complaints often reflect poor consumer experience or weakened trust, which may lead potential customers to delay purchases, damage brand reputation, and increase future sales volatility.

Table 7.

Mechanism Analysis: Complaint Constraints.

	(1)	(2)
Variables	Complain	Uncertainty1
Supervision1	−0.064*** (−12.024)	−0.058*** (−5.456)
Complain		0.007** (2.053)
Size	0.018** (2.345)	−0.006 (−1.615)
Lev	0.005 (0.241)	0.034** (2.488)
ROA	−0.185*** (−3.546)	−0.099*** (−3.680)
Cashflow	0.024 (0.644)	−0.090*** (−4.684)
INV	−0.047 (−1.084)	−0.076*** (−3.287)
FIXED	−0.014 (−0.451)	−0.018 (−1.054)
Indep	0.000 (0.312)	−0.001*** (−3.847)
Dual	−0.001 (−0.199)	−0.001 (−0.326)
Top10	0.042 (1.108)	0.074*** (4.411)
_cons	−0.373** (−2.178)	0.319*** (4.003)
Year-fixed	Yes	Yes
Firm-fixed	Yes	Yes
N	34,887	34,887
R ²	.356	.457

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Taken together, Table 7 shows that government accounting supervision reduces demand uncertainty by lowering the number of consumer complaints and improving market trust and demand stability. Hypothesis H2a is supported.

Mechanism Analysis: Earnings Manipulation

To further uncover the pathway through which government accounting supervision affects firms’ demand uncertainty, this study examines the information transparency mechanism from the perspective of financial information quality. When facing external market pressures or internal governance deficiencies, firms may manipulate accounting earnings to embellish their financial statements, thereby concealing their true operating conditions. Earnings manipulation not only undermines the trust of investors and consumers but also weakens firms’ accurate judgment of market demand, thus exacerbating demand uncertainty. Therefore, if government accounting supervision can effectively restrain earnings manipulation, it may enhance information transparency, stabilize market expectations, and consequently reduce firms’ demand uncertainty.

In the empirical implementation, we follow Qi et al. (2021) and Lee and Jeong (2024) and employ the widely used modified Jones model to estimate firms’ discretionary accruals (DA), which serve as a proxy for the extent of earnings manipulation. Table 8 reports the main results of the mechanism analysis. The regression shows that Supervision1 is significantly negatively associated with DA, indicating that government accounting supervision effectively restrains earnings manipulation and enhances the authenticity of financial information. Furthermore, when DA is included in the regression as a mediating variable, its coefficient is positive and significant, suggesting that a higher level of earnings manipulation indeed leads to greater demand uncertainty.

Table 8.

Mechanism Analysis: Earnings Manipulation.

	(1)	(2)
Variables	DA	Uncertainty1
Supervision1	−0.010** (−2.322)	−0.050*** (−4.777)
DA		0.032* (1.781)
Size	0.004*** (3.354)	−0.008** (−2.386)
Lev	−0.010** (−2.221)	0.027** (2.151)
ROA	1.019*** (118.686)	−0.119*** (−3.810)
Cashflow	−1.059*** (−144.923)	−0.057** (−2.149)
INV	0.026*** (4.465)	−0.068*** (−3.129)
FIXED	0.068*** (11.923)	−0.040** (−2.444)
Indep	0.000 (0.683)	−0.001*** (−3.943)
Dual	−0.002 (−1.608)	−0.002 (−0.571)
Top10	−0.021*** (−4.444)	0.087*** (5.477)
_cons	−0.068*** (−2.905)	0.365*** (4.804)
Year-fixed	Yes	Yes
Firm-fixed	Yes	Yes
N	33,828	33,828
R ²	.710	.458

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Taken together, these results imply that government accounting supervision reduces demand uncertainty by curbing earnings manipulation and improving financial reporting transparency, which stabilizes market expectations regarding firms’ future operating performance. Hypothesis H2b is supported.

Heterogeneity Analysis

Heterogeneity Analysis Based on Reasons for Announced Violations

To further explore the heterogeneous mechanisms through which government accounting supervision affects firms’ demand uncertainty, this study examines whether firms respond differently to supervision depending on the nature of their violations. Specifically, if a firm engages in typical financial fraud violations—such as inflating revenues, fabricating transactions, or manipulating profits to embellish performance—this is classified as Reason = 1. Conversely, if a firm’s violations mainly involve tax evasion, improper accounting practices, failure to disclose information as required, or lax internal controls without profit manipulation, it is categorized as non-financial fraud violations, denoted as Reason = 0. The sample firms are grouped according to Reason, and regression analyses are conducted separately to test whether the supervisory effects vary by violation type. The results are reported in Table 9.

Table 9.

Heterogeneity Analysis Based on Reasons for Announced Violations.

	(1)	(2)	(3)	(4)
	Uncertainty1	Uncertainty1	Uncertainty2	Uncertainty2
Variables	Reason = 1	Reason = 0	Reason = 1	Reason = 0
Supervision1	−0.048*** (−3.457)	−0.031 (−1.373)	−0.375*** (−2.804)	−0.023 (−0.194)
Size	−0.005 (−0.607)	−0.006* (−1.729)	−0.207** (−2.361)	0.153*** (3.576)
Lev	−0.021 (−0.667)	0.033** (2.466)	0.313 (1.125)	0.371** (2.565)
ROA	0.010 (0.148)	−0.101*** (−3.804)	0.198 (0.343)	−0.116 (−0.507)
Cashflow	−0.047 (−0.868)	−0.094*** (−4.875)	−0.005 (−0.016)	−0.319* (−1.846)
INV	0.089 (1.583)	−0.072*** (−3.191)	0.590* (1.790)	−0.326 (−1.481)
FIXED	−0.097** (−2.470)	−0.012 (−0.684)	−0.348 (−0.828)	−0.372* (−1.813)
Indep	−0.001 (−1.461)	−0.001*** (−3.347)	−0.008 (−1.375)	−0.006** (−2.341)
Dual	−0.008 (−0.964)	−0.002 (−0.512)	0.010 (0.196)	−0.036 (−1.144)
Top10	0.018 (0.414)	0.067*** (4.026)	1.147** (2.428)	1.674*** (8.183)
_cons	0.347* (1.685)	0.323*** (4.205)	5.402** (2.588)	−2.878*** (−2.975)
Year-fixed	Yes	Yes	Yes	Yes
Firm-fixed	Yes	Yes	Yes	Yes
N	1,393	33,612	1,075	22,572
R ²	.472	.456	.500	.459

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

Table 9 shows that for the Reason = 1 group, the coefficient of Supervision1 on demand uncertainty is significantly negative regardless of whether Uncertainty1 or the alternative measure Uncertainty2 is used, both at the 1% significance level. This indicates that for firms involved in financial fraud, government accounting supervision effectively curbs information manipulation, significantly enhances financial transparency and corporate credibility, and thereby reduces the market demand uncertainty they face. In contrast, for the Reason = 0 group (firms with tax evasion or general accounting irregularities), the coefficient of Supervision1 is not significant, suggesting that government supervision has a relatively limited impact on these firms. This may be because such violations have a lesser effect on external financial statements; although compliance issues exist, the resulting information distortion for the market is limited, leaving little room for supervision to improve demand uncertainty. This heterogeneity highlights the dependence of government supervision on the “information repair” pathway. For firms committing financial fraud, distorted financial reports directly interfere with the perceptions of investors, suppliers, consumers, and other external stakeholders regarding the firm’s true operating condition, significantly exacerbating market uncertainty. Government accounting supervision plays a crucial role in “information disclosure” and “signal reconstruction” in these firms, effectively reducing market doubts, strengthening coordinated expectations between firms and external parties, and thereby mitigating demand-side volatility. Conversely, for firms with only “internal operational” violations such as tax evasion or non-standard accounting, although their financial reports contain compliance flaws, these do not necessarily mislead external demand parties directly. Therefore, the marginal governance effect of supervision on demand uncertainty in these firms is relatively weak.

Heterogeneity Analysis of Firms’ Subsequent Response Degree

To further investigate whether the marginal effect of government accounting supervision varies with firms’ degree of subsequent response, this study conducts a heterogeneity analysis based on whether firms have complied with required rectifications. Specifically, we manually identify firms’ feedback and handling results from publicly released accounting information quality inspection announcements, constructing a dummy variable Correction. If an announcement explicitly states that a firm has completed rectification as required by regulatory authorities after accounting information quality issues were discovered, Correction is set to 1; otherwise, if the firm has not yet completed the rectification, Correction is set to 0. Subsequently, regression analyses are conducted on sub-samples grouped by the Correction variable to test whether firms’ cooperation in rectification affects the governance effect of government accounting supervision on their demand uncertainty. The results are shown in Table 10.

Table 10.

Heterogeneity Analysis of Firms’ Subsequent Response Degree.

	(1)	(2)	(3)	(4)
	Uncertainty1	Uncertainty1	Uncertainty2	Uncertainty2
Variables	Correction = 1	Correction = 0	Correction = 1	Correction = 0
Supervision1	−0.032*** (−2.799)	−0.025 (−1.210)	−0.264** (−2.166)	0.110 (0.464)
Size	−0.013* (−1.723)	−0.001 (−0.838)	−0.122 (−1.488)	0.109*** (4.387)
Lev	0.019 (0.587)	0.005 (0.686)	0.303 (1.524)	0.187** (2.088)
ROA	0.020 (0.322)	−0.087*** (−5.895)	0.701 (1.625)	−0.588*** (−3.799)
Cashflow	−0.021 (−0.471)	−0.057*** (−4.867)	−0.152 (−0.572)	−0.088 (−0.753)
INV	0.053 (1.204)	−0.002 (−0.124)	0.355 (1.166)	−0.147 (−0.957)
FIXED	−0.095*** (−3.052)	−0.010 (−1.047)	−0.280 (−0.899)	−0.324*** (−2.915)
Indep	−0.001** (−2.315)	−0.000 (−1.063)	−0.009** (−2.002)	−0.002 (−1.124)
Dual	−0.008 (−1.032)	−0.000 (−0.182)	−0.037 (−0.794)	−0.005 (−0.211)
Top10	0.035 (1.199)	0.002 (0.190)	0.820*** (3.145)	0.580*** (4.846)
_cons	0.496*** (2.690)	0.190*** (4.928)	3.583* (1.774)	−1.558*** (−2.772)
Year-fixed	Yes	Yes	Yes	Yes
Firm-fixed	Yes	Yes	Yes	Yes
N	1,645	30,430	1,271	19,934
R ²	.490	.452	.450	.479

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

As seen in Table 10, within the Correction = 1 subsample (firms that have completed rectification as required), the coefficients of Supervision1 on both Uncertainty1 and the alternative measure Uncertainty2 are significantly negative. This indicates that government accounting supervision can effectively reduce firms’ market demand uncertainty only when firms actively respond to regulatory requirements and carry out rectification. In contrast, for the Correction = 0 subsample, the coefficient of Supervision1 is not significant, suggesting that supervisory interventions have limited substantive improvement effects on firms that have not completed rectification. This heterogeneity reflects that the effectiveness of government supervision largely depends on firms’ subsequent response behaviors. Although supervision itself transmits external pressure, its ultimate success hinges on whether firms are willing to actively cooperate, standardize financial practices, and enhance information transparency. For firms willing to rectify, supervision plays a dual role of governance and guidance: it not only corrects existing problems but also improves external trust expectations through disclosure of rectification information, thereby reducing market demand fluctuations caused by information asymmetry. Conversely, for firms that have yet to complete rectification, even if issues are identified through supervision, passive or uncorrected subsequent behaviors may lead the market to adopt a more pessimistic view of their compliance and future prospects, weakening or even offsetting the potential positive impacts of supervision.

Heterogeneity Analysis of Different Governance Intensity

To further explore the actual effects of government accounting supervision under varying governance intensities, this study introduces whether firms have been subject to administrative penalties as a grouping dimension for heterogeneity analysis. Specifically, based on accounting information quality inspection announcements issued by the Ministry of Finance and local financial supervision departments, we identify that some firms with more severe issues receive administrative penalties such as fines. Accordingly, a dummy variable Punish is constructed: Punish = 1 if a firm is fined or otherwise administratively penalized by the Ministry of Finance due to inspection results; Punish = 0 if problems are identified but no penalties are imposed. The sample is then grouped by the Punish variable, and regressions on firms’ demand uncertainty (Uncertainty1 and Uncertainty2) are performed separately. The results are presented in Table 11.

Table 11.

Heterogeneity Analysis of Different Governance Intensity.

	(1)	(2)	(3)	(4)
	Uncertainty1	Uncertainty1	Uncertainty2	Uncertainty2
Variables	Punish = 1	Punish = 0	Punish = 1	Punish = 0
Supervision1	−0.008 (−0.250)	−0.048*** (−4.523)	−0.173 (−1.132)	−0.272** (−2.410)
Size	−0.029 (−1.600)	−0.006* (−1.782)	0.101 (0.856)	0.149*** (3.418)
Lev	−0.109* (−1.762)	0.042*** (3.000)	−0.731* (−2.018)	0.477*** (3.270)
ROA	−0.100 (−0.598)	−0.091*** (−3.354)	−1.709 (−1.540)	0.031 (0.137)
Cashflow	−0.122 (−1.071)	−0.094*** (−4.831)	0.618 (1.185)	−0.297* (−1.763)
INV	0.185 (1.453)	−0.081*** (−3.489)	0.375 (0.407)	−0.401* (−1.800)
FIXED	−0.197* (−1.923)	−0.018 (−1.018)	−0.895 (−1.424)	−0.286 (−1.406)
Indep	0.000 (0.210)	−0.001*** (−4.110)	−0.006 (−0.861)	−0.007*** (−2.766)
Dual	0.009 (0.496)	−0.002 (−0.485)	−0.024 (−0.275)	−0.024 (−0.757)
Top10	−0.059 (−0.829)	0.084*** (4.852)	0.187 (0.402)	1.848*** (8.761)
_cons	0.907** (2.181)	0.330*** (4.150)	−0.800 (−0.298)	−2.914*** (−2.950)
Year-fixed	Yes	Yes	Yes	Yes
Firm-fixed	Yes	Yes	Yes	Yes
N	414	34,989	325	23,645
R ²	.593	.457	.422	.458

*, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

The regression results show that in the Punish = 0 subsample (firms found with issues but not fined), the coefficient of Supervision1 is significantly negative, indicating that accounting supervision effectively reduces demand uncertainty in such cases. However, in the Punish = 1 subsample (fined firms), the coefficients are insignificant. This heterogeneity reflects differences in the marginal governance effects of supervision under distinct regulatory contexts. For unfined firms, the identified problems are likely minor, and supervision mainly serves as guidance and a warning. These firms are more inclined to undertake substantive rectification and improve disclosure quality, thereby restoring market confidence and reducing uncertainty about future performance.

In contrast, for fined firms, penalties may signal severe violations or even opportunistic behavior, intensifying concerns over future compliance, operational stability, and governance capacity. As a result, market distrust may persist despite supervisory intervention, weakening or offsetting its stabilizing effect. In other words, when firm credibility is already severely damaged, supervision is less likely to produce immediate governance benefits. Moreover, fined firms may have deeper governance flaws or distorted incentives, leading to merely “passive compliance,” which limits real improvements in information quality and fails to meaningfully reduce operational uncertainty.

Conclusion

Using Chinese A-share listed firms from 2011 to 2024 as the research setting, this study examines from an external governance perspective how government accounting supervision affects corporate demand uncertainty and through which mechanisms. The findings can be summarized in four main points, which also highlight the paper’s core theoretical contributions to the literatures on regulatory economics, information economics, and external risk management. First, government accounting supervision significantly reduces firms’ demand uncertainty. Whether a firm itself (or its parent group) is inspected and publicly reported for violations, the unpredictability of its demand fluctuations decreases. Moreover, inspections of peer firms within the same industry generate a “demonstration–deterrence effect,” which improves the industry-wide information environment and indirectly stabilizes market expectations for non-inspected firms. This indicates that government accounting supervision not only shapes internal financial behavior but also influences the external operating environment through enhanced transparency, thereby enriching theoretical frameworks linking external governance to demand-side risk.

Second, multiple robustness checks—including parallel trends tests, Bacon decomposition, alternative dependent variables, PSM matching, and the Heckman two-stage model—consistently confirm the reliability of the baseline results, strengthening the causal interpretation. Third, mechanism analyses show that government accounting supervision reduces demand uncertainty through two channels: Complaint-constraint mechanism, Supervision improves transparency and promotes compliance, reducing consumer complaints and stabilizing market demand. Earnings manipulation suppression mechanism, Supervisory pressure effectively constrains financial misreporting, mitigating market misjudgments caused by distorted information and making demand more predictable. These mechanisms shed new light on how government regulation influences firms’ external market environments and provide fresh empirical evidence for the information economics logic linking transparency to expectation stability.

Fourth, heterogeneity analyses reveal substantial variation across firms. The corrective effect of supervision is strongest among firms involved in financial fraud. Among firms that actively implement rectification measures, the expectation-stabilizing effect of supervision is more pronounced. In contrast, for firms receiving administrative penalties, the effect becomes insignificant, likely because penalties themselves create negative signals that offset the stabilizing impact of supervision. This suggests that corporate governance foundations and the nature of violations shape the effectiveness of regulatory interventions, offering new theoretical insights into the marginal governance effects of government supervision.

Despite providing a systematic examination of government accounting supervision and corporate demand uncertainty, this paper still has several limitations that future research may address. First, the empirical setting is based on China’s unique fiscal supervision system, which raises concerns about external validity. Whether the findings generalize to other countries or regulatory environments remains uncertain. Future studies could use cross-country data or compare different regulatory frameworks to assess the applicability of government supervision across governance contexts. Second, the mechanism analysis in this paper focuses primarily on two channels. However, government supervision may influence demand uncertainty through other unobserved pathways. Future work could employ more structural models or granular data to capture these additional mechanisms. Third, it is difficult to fully rule out alternative explanations. For instance, accounting inspections may coincide with cycles of heightened media scrutiny, industry-wide demand shocks, or regional policy adjustments, all of which could jointly affect firms’ demand uncertainty. Although this study attempts to control for these factors through robustness tests, future research could incorporate additional variables such as media attention indices to further strengthen causal identification. Fourth, the mechanism tests rely mainly on traditional mediation models, which may not fully reflect the complexity of regulatory transmission. Future studies may improve mechanism identification by integrating text analysis, event-study designs, or machine learning–based approaches. Finally, the measurement of demand uncertainty is based on annual data, which may not capture higher-frequency operational dynamics. Future research could explore quarterly or even monthly data to obtain a more refined depiction of demand fluctuations.

The policy recommendations of this study are as follows. First, regulators should adopt tiered and categorized supervision based on the nature of violations. For high-risk firms involved in financial fraud, inspection intensity should be increased. At the same time, rotating industry-wide inspections can strengthen regulatory spillover effects and improve the efficiency of supervisory resource allocation. Second, the post-inspection rectification mechanism should be improved by establishing a “inspection–rectification–follow-up” closed loop. Incorporating rectification outcomes into credit evaluations can ensure that supervisory pressure is effectively translated into improvements in information quality. Third, transparency in regulatory disclosures and consumer complaint data should be enhanced. Regulators should also encourage firms to strengthen their complaint-handling systems to stabilize market expectations and reduce demand-side volatility. Fourth, regulators should distinguish between minor and severe violations. Flexible regulatory tools may be applied to minor violations to avoid unnecessary market disruption, while more stringent and continuous governance measures should be implemented for severe violations, thereby promoting compliant corporate behavior and fostering the overall stability of the market environment.

Footnotes

Appendix

Table 1.

Variable Definitions.

Variables	Measurement
Supervision1	A staggered DID variable capturing whether a firm is directly inspected by fiscal authorities. Following Lyu (2025) and Fan et al. (2025), Post equals 1 for the year in which a firm is inspected and publicly reported for violations, and all subsequent years, and 0 otherwise. Treat equals 1 if the firm is ever identified as having accounting problems during the sample period, and 0 otherwise. Supervision1 is defined as the interaction Treat × Post, measuring the regulatory pressure triggered by direct government accounting supervision. Source: We manually collect the data from the Accounting Information Quality Inspection Announcements of China’s Ministry of Finance.
Supervision2	An industry-year indicator capturing spillover effects of inspections on firms not directly inspected. If any firm within a corporate group is inspected and publicly reported for violations in a given year, Supervision2 takes the value 1 for all other firms in the same industry-year, and 0 otherwise. This variable reflects the “warning” or “deterrence” effect of government supervision at the industry level, indicating whether peer inspections influence the strategic or operational behavior of non-inspected firms.
Uncertainty1	Constructed to measure the unpredictability of a firm’s market demand. Following prior literature, demand uncertainty is captured using the unexplained volatility in quarterly sales revenue. Specifically, quarterly sales are regressed on historical sales, inventory changes, and seasonal controls to extract the unpredictable component of sales. The standard deviation of the regression residuals across four quarters within a fiscal year is computed as Uncertainty1. A higher value indicates greater volatility in unexplained sales movements and thus a higher level of demand uncertainty faced by the firm.
Size	The natural log of market capitalization (Ln (market capitalization)), measured at the end of fiscal year t. Source: CSMAR
Lev	The sum of long-term debt and debt in current liabilities divided by total assets (Total liabilities/total assets). Source: CSMAR
ROA	Return on assets, defined as net income divided by total assets. Source: CSMAR
Cashflow	Cash flow ratio, defined as net cash flow from operating activities divided by total assets. Source: CSMAR.
INV	Inventory ratio, measured as total inventories divided by total assets. Source: CSMAR.
FIXED	Fixed asset ratio, calculated as net fixed assets divided by total assets. Source: CSMAR.
Indep	Proportion of independent directors on the board, calculated as the number of independent directors divided by the total number of board members. Source: CSMAR
Dual	CEO duality, equal to 1 if the chairman also serves as the CEO, and 0 otherwise. Source: CSMAR.
Top10	Ownership concentration, measured as the shareholding ratio of the top ten shareholders. Source: CSMAR

ORCID iDs

Lin Zhang

Fan Yong

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data are available from the authors upon request. This paper does not involve human experimenters.

References

Adebambo

B. N.

Yan

(2018). Investor overconfidence, firm valuation, and corporate decisions. Management Science, 64(11), 5349–5369.

Arora

S. D.

Chakraborty

Narayanamurthy

(2025). Why and when consumers post complaint messages on social media? Conceptualizing social voice as a distinct complaining behaviour. British Journal of Management, 36, 1746–1766.

Banker

R. D.

Byzalov

Plehn-Dujowich

J. M.

(2014). Demand uncertainty and cost behavior. Accounting Review, 89(3), 839–865.

Baron

D. P.

(1971). Demand uncertainty in imperfect competition. International Economic Review, 12, 196–208.

Begen

M. A.

Pun

Yan

(2016). Supply and demand uncertainty reduction efforts and cost comparison. International Journal of Production Economics, 180, 125–134.

Belloni

Lopomo

Wang

(2017). Resource allocation under demand uncertainty and private information. Management Science, 63(12), 4219–4235.

Carlton

D. W.

(1978). Market behavior with demand uncertainty and price inflexibility. American Economic Review, 68(4), 571–587.

Chang

Kwok

W. C.

Wong

(2024). Demand uncertainty, inventory, and cost structure. Contemporary Accounting Research, 41(1), 226–254.

Chen

(2024). Government accounting supervision and risk of stock price crashes of listed companies: Evidence from the ministry of finance in China. Finance Research Letters, 69, 106248.

10.

Chen

Liao

S. S.

Nie

(2025). Behavior-based optimal refund policy under advance selling: An analysis of the impact of loss fairness concerns. Journal of the Operational Research Society, 76(4), 804–827.

11.

Cohen

M. C.

Lobel

Perakis

(2016). The impact of demand uncertainty on consumer subsidies for green technology adoption. Management Science, 62(5), 1235–1258.

12.

Fan

Chen

Lei

(2025). How government accounting supervision affects corporate financialization? Evidence from random inspections by China’s ministry of finance. International Review of Economics & Finance, 103, 104462.

13.

Fuss

Vermeulen

(2008). Firms’ investment decisions in response to demand and price uncertainty. Applied Economics, 40(18), 2337–2351.

14.

Ghosh

Olsen

(2009). Environmental uncertainty and managers’ use of discretionary accruals. Accounting, Organizations and Society, 34(2), 188–205.

15.

Goldberg

L. S.

Kolstad

C. D.

(1995). Foreign direct investment, exchange rate variability and demand uncertainty. International Economic Review, 36, 855–873.

16.

Gomez-Trejos

(2025). A note on how demand uncertainty matters for emissions regulation in energy markets. Economics Letters, 254, 112434.

17.

Grosz

Raval

(2025). Amplifying consumers’ voice: The Federal Trade Commission’s Report Fraud Website Redesign. Marketing Science, 44(3), 525–545.

18.

Guiso

Parigi

(1999). Investment and demand uncertainty. Quarterly Journal of Economics, 114(1), 185–227.

19.

Han

Liu

(2025). A competitive supply chain model considering consumer green preference under demand uncertainty. Operational Research, 25(2), 57.

20.

Huang

Tan

Chan

H. K.

(2025). Financial contract invisibility in a competing supply chain. International Journal of Production Research, 63, 8611–8631.

21.

Jiang

(2024). Government accounting supervision and ESG performance. Applied Economics Letters, 1–5. Advance online publication. https://doi.org/10.1080/13504851.2024.2402482

22.

Lee

Jeong

S. W. T.

(2024). Economic policy uncertainty, audit quality, and earnings manipulation. Asia-Pacific Journal of Accounting & Economics, 31(2), 269–299.

23.

Leland

H. E.

(1972). Theory of the firm facing uncertain demand. American Economic Review, 62(3), 278–291.

24.

Liu

Zhang

Liu

(2025). How does governmental accounting and financial supervision affect the quality of analysts’ forecasts? Finance Research Letters, 85, 107926.

25.

Zhu

Qian

(2023). Financial accounting supervision, fiscal decentralization, and national governance. Transformations in Business and Economics, 22(3), 909–932.

26.

Y. N.

Jiang

Wang

Wei

(2025). XGBoost-based risk prediction model for massive vehicle recalls using consumer complaints. Risk Analysis, 45(12), 4408–4422.

27.

Chang

Zhou

(2025). The impact of consumer complaints on corporate digital transformation. International Review of Economics & Finance, 101, 104203.

28.

Lyu

(2025). Government accounting supervision and corporate supply chain risk. Finance Research Letters, 74, 106785.

29.

Magnuson

(2011). The role of expectations in value and glamour stock returns. Journal of Behavioral Finance, 12(2), 98–115.

30.

Modigliani

Cohn

R. A.

(1979). Inflation, rational valuation and the market. Financial Analysts Journal, 35(2), 24–44.

31.

Deng

Tudi

(2025). The impact of foreign demand uncertainty on within-firm pay inequality. Applied Economics Letters, 1–9. Advance online publication. https://doi.org/10.1080/13504851.2025.2530152

32.

Nugroho

Wang

W. T.

(2024). Applying justice theory to investigate the effects of consumer complaints and opportunistic intention on brand reputation and consumer repurchase behavior. Journal of Electronic Commerce Research, 25(1).

33.

Pan

Chen

Zhang

Fang

(2023). Government accounting supervision and excessive perk consumption of executives: Evidence from China. Finance Research Letters, 57, 104139.

34.

Zhou

Chen

(2021). Corporate site visits and earnings management. Journal of Accounting and Public Policy, 40(4), 106823.

35.

Roychowdhury

(2006). Earnings management through real activities manipulation. Journal of Accounting and Economics, 42(3), 335–370.

36.

Saoud

Kourentzes

Boylan

J. E.

(2025). The importance of forecast uncertainty in understanding the bullwhip effect. International Journal of Production Research, 1–22. Advance online publication. https://doi.org/10.1080/00207543.2025.2527957

37.

Sarwar

Baig

Rahi

Sattar

(2025). Fostering Green Behavior in the workplace: The role of ethical climate, motivation states, and environmental knowledge. Sustainability, 17(9), 4083.

38.

Shang

Han

(2024). Quality improvement decisions in service supply chains with collaborative and free-riding behaviors. Systems, 12(10), 419.

39.

Sun

Xing

(2025). To business or consumer? Competing choices of business models for online fresh stores. IEEE Transactions on Engineering Management, 72, 2180–2195.

40.

Walker

(2013). How far can we trust earnings numbers? What research tells us about earnings management. Accounting and Business Research, 43(4), 445–481.

41.

Wang

Zhang

Sun

(2025). Geopolitical risk and corporate environmental investment. Applied Economics, 57, 10951–10965.

42.

Wang

J. J.

Wang

(2023). Earnings manipulation and similarity of annual report disclosure: Evidence from China. Accounting and Finance, 63, 1137–1156.

43.

Xin

Liu

Zhang

(2025). Can public opinion manipulation disrupt markets?—A study on the impact of Internet Water army on corporate demand uncertainty. International Review of Economics & Finance, 104, 104748.

44.

Zhang

Huang

(2026). Government accounting supervision and corporate productivity: Evidence from market-oriented regulatory reform. International Review of Financial Analysis, 109, 104796.

45.

Zhang

Huang

(2025). Unraveling how poor logistics service quality of cross-border E-commerce influences customer complaints based on text mining and association analysis. Journal of Retailing and Consumer Services, 84, 104237.