Bank Competition and Firms’ Financial Reporting Quality: Evidence from a Bank-Based Economy

Abstract

We empirically investigate the impact of bank competition on corporate financial reporting quality. Using data from Vietnam for the period 2007 to 2023, we reveal a significant association between increased bank competition and firms’ better information disclosure. Specifically, heightened competition prompts firm managers to curtail upward earnings management practices. Our channel analysis highlights the alleviation of financial constraints as a key mechanism driving this relationship. Furthermore, heterogeneity tests reveal that the impact of bank competition is less pronounced for firms with strong bank-firm relationships, while smaller firms, non-state-owned entities, and those with limited market power exhibit greater responsiveness to bank competition in terms of manipulating financial reporting. Besides, our findings also suggest that the influence of bank competition on earnings manipulation is more pronounced during unfavorable economic conditions, underscoring the importance of financing constraints in shaping firms’ financial reporting practices under competitive banking environments.

Introduction

Banking market structure has a notable impact on firms’ operational landscapes, thereby shaping their choices of investment and financing. Under this notion, many studies have shown relevant results. After the deregulation that heightens the competitiveness of the US banking sector, firms exhibit a notable reduction in their reliance on external debt and a corresponding decrease in their investment activities (Zarutskie, 2006). Aligning with the market power view, multiple authors claim that high bank concentration, also known as low bank competition, can lead to financing obstacles (Beck et al., 2004; Leon, 2015; Saeed & Vincent, 2012). Other studies also offer evidence supporting this view, as Love and Pería (2015) highlight reduced access to finance in low-competition scenarios, Z. Zhang et al. (2019) find that growing competition among Chinese banks reduces financing constraints on listed firms, and Khan and Kutan (2023) observe lower likelihoods of credit constraints for small and medium-sized enterprises (SMEs) in competitive banking environments. Contrary to this line of research, however, Ratti et al. (2008) find more available funding in highly concentrated banking sectors in Europe, which is supported by Álvarez and Bertin (2016), who show that banking competition increases financial constraints in Latin America. Besides, Chauvet and Jacolin (2017) note that competitive banks favor firm growth, particularly with high financial inclusion. Cao and Li (2024) show that bank competition promotes corporate financialization, while Cañón et al. (2022) reveal that loans from banks with higher market power are significantly more expensive for firms. This study extends the existing literature by examining whether and how bank competition influences firms’ financial reporting quality.

Financial reporting quality reflects the extent to which financial statements faithfully represent firms’ underlying economic performance and provide decision-useful information to market participants (Biddle et al., 2009). A central dimension of financial reporting quality is earnings quality, defined as the degree to which reported earnings map into firms’ future cash flows with minimal estimation error (Dechow & Dichev, 2002). Earnings management weakens this mapping by introducing managerial distortions into reported earnings, thereby reducing their informational content and transparency. Accordingly, the literature commonly interprets higher levels of earnings management as indicative of lower financial statement quality or greater earnings opacity (Biddle et al., 2009; Dechow & Dichev, 2002). In line with this perspective, we treat earnings management as an inverse proxy for financial reporting quality in this study. Given the critical importance of financial reporting quality, numerous studies have explored its determinants at both firm and industry levels (Aljughaiman et al., 2023; Bermpei et al., 2022; Jiang et al., 2022; Kurt, 2018; Q. K. Nguyen, 2024; Qi et al., 2021; Thanh et al., 2020). However, the explicit examination of the influence of bank competition on financial reporting quality remains absent in the existing literature, making our study the first analysis in this area.

In this study, we empirically investigate the influence of bank competition on the earnings management practices of nonfinancial firms in Vietnam. We select the Vietnamese market for several specific motivations as follows. First, despite recent advancements in the Vietnamese stock market, many firms still heavily rely on bank credit as their primary financing source (Dang & Huynh, 2025). This underscores the significant role of commercial banks in the country’s financial system. Second, Vietnam’s entry into the World Trade Organization in 2007 was a key milestone, prompting substantial reforms in the banking sector as part of the nation’s commitment to global integration. Consequently, measures such as bank privatization and the liberalization of foreign bank entry were actively pursued, resulting in profound shifts in the overall banking market structure (T. N. Nguyen et al., 2018). In subsequent years, particularly around 2015, comprehensive banking reforms were witnessed in response to challenges such as rising non-performing loans and inefficiencies among several weaker banks. Third, given the centrality of the banking system, any fluctuations in its market dynamics directly impact firms, influencing their access to financing. Notably, Vietnamese firms tend to be smaller and financially less healthy compared to their counterparts in advanced economies and even other major emerging markets, rendering them more vulnerable to financing constraints (Huynh, 2025). This characteristic provides a natural framework for examining how bank competition shapes firms’ behavior through financing challenges. Fourth, Vietnam’s local stock market still features lower-quality accounting information relative to more mature markets (Thanh et al., 2020). Additionally, concerning the operations of Vietnamese firms, existing studies underscore that corporate governance practices, accounting standards, and financial transparency are even behind those of firms in other emerging markets, catalyzed by regulatory inconsistencies and ineffective enforcement mechanisms (Q. Nguyen et al., 2024).

We perform our analysis by employing a comprehensive approach to gage bank competition, encompassing structural metrics and non-structural measures. Focusing on discretionary accruals as a proxy for corporate earnings management, we empirically investigate our hypothesis using panel data from nonfinancial firms in Vietnam from 2007 to 2023. Based on a generalized method of moments (GMM) model that addresses endogeneity and dynamic reporting characteristics, our analysis reveals a negative association between bank competition and firms’ earnings management practices. To ensure the strong validity of our regression finding, we undertake a set of robustness assessments.

We also explore a potential pathway through which bank competition influences firms’ information disclosure. Specifically, intensified competitive dynamics prompt banks to relax lending criteria, facilitating greater access to credit for firms. Therefore, firms face reduced pressures to inflate earnings to enhance their market standing, as they can rely on readily available credit resources. This study delves deeper into this mechanism, particularly examining its manifestation through firms’ financial constraints. Our analysis reveals a propensity for upward earnings manipulation, driven by managerial efforts to bolster performance amidst capital-raising endeavors or in response to financing constraints. These findings corroborate the insights gleaned from the channel analysis, thereby reinforcing the hypothesized link between bank competition, financial constraints, and earnings management.

This study further extends its analysis through cross-firm heterogeneity analyses to provide deeper insights into the association between bank competition and firms’ financial reporting quality. The rationale behind this expansion stems from the understanding that firms experiencing greater financial constraints are prone to resorting to earnings management tactics. Hence, we explore the moderating role of bank-firm relationships, hypothesizing that firms with stronger ties to banks may experience alleviated financing obstacles, thus potentially mitigating the impact of competition on earnings opacity. Interestingly, our empirical tests affirm this notion, revealing that firms with bank-firm relationships exhibit a dampened susceptibility to the effects of bank competition on information disclosure practices. Additionally, we look into the influence of other firm characteristics, namely firm size, ownership structure, and market power. Motivated by the notion that smaller firms, non-state-owned firms, and firms with limited market power encounter heightened financing constraints, our analyses uncover a more pronounced impact of bank competition on earnings opacity among these entities, underscoring the significance of their financial challenges.

Given that the principal pathway through which bank competition curtails firms’ earnings management is by alleviating financing constraints, the influence of bank competition on firms’ earnings management is anticipated to be less pronounced in favorable economic environments characterized by diminished financing hurdles. Put differently, the reduction in financing constraints induced by competitive dynamics in the banking sector is expected to carry less weight in environments with low financing obstacles compared to those with heightened financial constraints. Following these arguments, our additional analysis reveals that the impact of bank competition on earnings manipulation is more conspicuous during recessionary times than in favorable economic periods.

Our study contributes to the literature on bank competition and corporate financial reporting in several meaningful ways. First, while prior studies such as Huang et al. (2023) investigate the impact of banking deregulation on earnings management by exploiting exogenous policy shocks in developed economies, we examine an emerging market context where bank competition evolves through changes in market structure rather than regulatory interventions. By employing a comprehensive set of both structural and non-structural measures of bank competition, our analysis captures direct and continuous variation in competitive conditions, ensuring that the results are not driven by any single competition proxy. Second, our study departs methodologically from the predominantly static and deregulation-based frameworks in the existing literature by adopting a dynamic panel system GMM approach. This framework explicitly accounts for persistence in earnings management behavior and mitigates endogeneity concerns, which are particularly relevant in bank-dominated financial systems where firms’ reporting choices adjust gradually over time. Third, we identify financing constraints as a key transmission channel through which bank competition affects corporate earnings management. While Huang et al. (2023) emphasize credit supply expansion following deregulation, our findings highlight how intensified banking competition alleviates firms’ financing frictions, thereby disciplining managers’ incentives to engage in upward earnings manipulation. Finally, we document substantial heterogeneity in the effect of bank competition across firms and macroeconomic conditions. Specifically, the disciplining role of bank competition is weaker for firms with strong bank-firm relationships, whereas it is more pronounced for smaller firms, non-state-owned enterprises, and firms with limited market power. We further show that these effects are amplified during unfavorable economic conditions, underscoring the importance of financing constraints in shaping firms’ financial reporting behavior under competitive banking environments.

We organize the remaining sections of the paper as follows. Section 2 develops the theoretical arguments and presents the study hypotheses. Section 3 details the data collection process and the research methodology employed. Section 4 presents all empirical findings derived from our baseline and extended analysis. Finally, Section 5 offers concluding remarks and implications stemming from our results.

Hypothesis Development

While the existing literature does not provide a definitive forecast or conclusion regarding the impact of bank competition on firms’ earnings opacity, it does offer some insights into possible mechanisms through which bank competition influences firm opacity. Due to the issues of adverse selection and moral hazard stemming from information asymmetries between lenders and borrowers, banks are motivated to enhance their activities of monitoring and information acquisition (Love & Pería, 2015). Banks with greater market power (i.e., experiencing lower competitive conditions) are more likely to establish enduring relationships with borrowers. This tendency arises from their enhanced ability to internalize the costs associated with such activities and their reduced risk of borrowers switching to other banks (Petersen & Rajan, 1995). In concentrated markets where firms undergo close and sustained monitoring by banks, they may be required to maintain high earnings quality, leaving minimal room for earnings management. Conversely, heightened competition within the industry may lead banks to relax their monitoring and supervision, consequently facilitating firms’ manipulation of earnings. Hence, according to this information channel, increased bank competition is anticipated to amplify firms’ opaque behaviors.

However, alternative streams of literature propose that bank competition reduces earnings management by reducing managerial incentives to manipulate financial statements. This effect can be elucidated as follows. In less competitive markets, banks tend to constrict loan supply and impose higher lending rates to secure monopoly profits (Beck et al., 2004). Additionally, greater market concentration correlates with increased collateral requirements (Hainz et al., 2013). In contrast, heightened competition within the banking sector is expected to augment credit at lower prices, thereby alleviating corporate financing constraints. Empirical studies extensively substantiate the mitigating effect of bank competition on firms’ financial constraints (Khan & Kutan, 2023; Leon, 2015; Love & Pería, 2015; Saeed & Vincent, 2012). Meanwhile, another line of literature posits that more financial constraints may intensify managers’ inclination toward upward earnings management. By inflating earnings figures, managers aim to signal positive firm prospects to the market, potentially improving firms’ access to external financing and alleviating market-imposed financial constraints. This viewpoint aligns with prior empirical findings illustrating a positive association between financial constraints and upward earnings management (Bermpei et al., 2022; Iatridis & Kadorinis, 2009; Kurt, 2018).

The arguments above suggest that bank competition may influence firms’ financial reporting quality through competing channels, leading to contrasting theoretical predictions regarding its effect on earnings management. Consequently, the net impact of bank competition on earnings management remains an empirical question. Based on the two alternative mechanisms discussed above, we develop the following competing hypotheses:

• Hypothesis A: Bank competition is positively associated with firms’ earnings management.

• Hypothesis B: Bank competition is negatively associated with firms’ earnings management.

Methodology and Data

Variables

Measurement for Earnings Management

Various methods are available for assessing accrual-based earnings management, with no universally acknowledged optimal approach. In the primary estimations, we determine discretionary accruals through the utilization of the performance-matched modified Jones model (Kothari et al., 2005). This proxy is derived from the following equation:

$$\begin{array}{cc}\multicolumn{1}{c}{{\mathrm{TC}}_{i,t}/T{A}_{i,t-1}}& \multicolumn{1}{c}{={\alpha }_0+{\alpha }_1\times (1/{\mathrm{TA}}_{i,t-1})}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+{\alpha }_2\times (\mathrm{\Delta }{\mathrm{SR}}_{i,t}/T{A}_{i,t-1})}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+{\alpha }_3\times ({\mathrm{PPE}}_{i,t}/T{A}_{i,t-1})}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+{\alpha }_4\times ({\mathrm{IN}}_{i,t}/T{A}_{i,t-1})+{\epsilon }_{i,t}}\end{array}$$

(1)

in which $\mathrm{TC}$ denotes the aggregate accruals, determined by the equation: (Changes in current assets − Changes in current liabilities − Changes in cash and cash equivalents + Changes in short-term debt included in current liabilities + Depreciation costs). $\mathrm{TA}$ represents total assets, and $\mathrm{\Delta }\mathrm{SR}$ captures the change in sales minus the change in accounts receivables. $\mathrm{PPE}$ is tangible fixed assets (plant, property, and equipment), and $\mathrm{IN}$ is income before extraordinary items. The residual term, denoted as ${\epsilon }_{\mathrm{i},\mathrm{t}}$ , reflects discretionary accruals. It is essential to note that we perform estimations of this model considering each industry in each year separately. Industries are classified according to the ICB Level 1 classification, excluding financial firms and utilities in line with standard practice, and all industry-year groups contain a sufficient number of firm observations to ensure reliable estimation. We use the beginning-year value of total assets as a deflator for all components to address heteroscedasticity issues. Finally, to proxy for accrual-based earnings management, we employ the absolute value of discretionary accruals, where a larger value implies greater earnings management.

Measurement for Banking Market Structure

Notwithstanding certain constraints associated with the structural approach, we still utilize bank-level data for the creation of two widely employed indicators of bank competition, commonly referred to as concentration ratios. Specifically, we employ the proportion of assets held by the five largest banks as an indicator of market concentration, in conjunction with the Herfindahl-Hirschman index, calculated as the sum of the squared market shares of individual banks. Within this route, it is established that measures of concentration are reverse competition metrics.

We enhance our understanding of market dynamics by supplementing the banking market structure information obtained through the structural approach, which imperfectly captures bank conduct. This supplementation is achieved through the incorporation of three competition indicators rooted in the non-structural approach, namely the Lerner index, the Boone indicator, and the H-statistic. These distinct measures offer varied insights into the banking competition, capturing different facets of the competitive landscape (Leon, 2015).

The Lerner index quantifies a bank’s market power, where elevated Lerner index values signify increased market power and reduced market competition. In an ideally competitive market, where equilibrium prices align with marginal costs, the Lerner index attains a value of 0. Accordingly, for each bank denoted as i in year t, the Lerner index is obtained as follows:

$${\mathrm{Lerner}}_{\mathrm{it}}=\frac{P_{\mathrm{it}}-M_{\mathrm{it}}}{P_{\mathrm{it}}}$$

(2)

where $P_{\mathrm{it}}$ is the output price and computed by total income scaled by total assets, while $M_{\mathrm{it}}$ represents the marginal cost of banks and is computed using the following translog cost function (Berger et al., 2009):

$$\begin{array}{cc}\multicolumn{1}{c}{{\mathrm{lnC}}_{\mathrm{it}}}& \multicolumn{1}{c}{={\beta }_0+{\beta }_1{\mathrm{lnA}}_{\mathrm{it}}+\frac{{\beta }_2}{2}{\left({\mathrm{lnA}}_{\mathrm{it}}\right)}^2}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+\sum _{k=1}^3{\gamma }_{\mathrm{it}}{\mathrm{lnW}}_{k,\mathrm{it}}+\sum _{k=1}^3{\delta }_k{\mathrm{lnA}}_{\mathrm{it}}{\mathrm{lnW}}_{k,\mathrm{it}}}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+\sum _{k=1}^3\sum _{j=1}^3{\mathrm{lnW}}_{k,\mathrm{it}}{\mathrm{lnW}}_{j,\mathrm{it}}+{\epsilon }_{i,t}}\end{array}$$

(3)

where $C_{\mathrm{it}}$ represents the total costs incurred by banks, encompassing the sum of interest and non-interest costs, while $A_{\mathrm{it}}$ denotes total assets. The marginal cost equation is computed considering three input factors denoted by W _k , which include the cost of human (W₁)—measured as personnel expenses divided by total assets, the cost of physical assets (W₂)—non-interest expenses divided by fixed assets, and the cost of funds (W₃)—interest expenses to total deposits. Ultimately, $M_{\mathrm{it}}$ is yielded based on the following formula:

$$M_{\mathrm{it}}=\frac{C_{\mathrm{it}}}{A_{\mathrm{it}}}[{\beta }_1+{\beta }_2{\mathrm{lnA}}_{\mathrm{it}}+\sum _{k=1}^3{\delta }_k{\mathrm{lnW}}_{k,\mathrm{it}}]$$

(4)

The Lerner variable for regression is computed as the weighted mean (with weights assigned based on bank assets) of individual Lerner indices for all sample banks.

The Boone indicator, grounded in the connection between profits and efficiency, is quantified as the elasticity of profit to marginal cost (Boone, 2008). This indicator inherently assumes a negative value, and a greater value indicates a lower level of competition. To enter regressions for this study, the Boone indicator is computed for the banking system each year (Cañón et al., 2022) through the application of the following equation:

$${\ln (p}_i)=\alpha +\beta {\ln (M}_i)$$

(5)

where $pi$ and $Mi$ correspond to the profits and marginal costs of bank i, respectively. The coefficient associated with marginal costs serves as the Boone indicator. In our primary analysis, we adhere closely to the methodology outlined by Schaeck and Cihák (2014), employing the system GMM to regress our Boone model.

Panzar and Rosse (1987) highlight that, subject to certain assumptions, the transmission of input price changes varies with market competition. The H-statistic, measuring the elasticity of bank revenues to input prices, assesses competition. In collusive markets, the H-statistic is ≤0, while it equals 1 in perfectly competitive markets, and ranges between 0 and 1 in monopolistic competition. The index is computed by regressing the following model for Vietnam (Claessens & Laeven, 2004):

$$\begin{array}{cc}\multicolumn{1}{c}{\mathrm{lnIR}\mathrm{it}}& \multicolumn{1}{c}{={\beta }_0+{\beta }_1{\mathrm{lnW}}_{1,\mathrm{it}}+{\beta }_2{\mathrm{lnW}}_{2,\mathrm{it}}}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+{\beta }_3{\mathrm{lnW}}_{3,\mathrm{it}}+\gamma B_{\mathrm{it}}+\delta T_t+{\epsilon }_{i,t}}\end{array}$$

(6)

Considering three inputs—labor, physical capital, and deposits—denoted as W₁, W₂, and W₃, the model incorporates time fixed effects (T) and control variables ( $B_{\mathrm{it}}$ )—bank equity, loan share, and bank size. The summation of the estimated coefficients for each input price yields the H-statistic value. Under the Panzar-Rosse framework, the H-statistic is valid only under long-run equilibrium; this condition is satisfied based on the conventional revenue-elasticity test. The estimation of this indicator is conducted using a rolling window of 3 years, in line with the prior approach (Park, 2009). For the main estimations reported in the paper, we employ generalized least squares (GLS) regression to conduct the regression work (Claessens & Laeven, 2004).

Model Specification and Estimation Technique

We empirically test how the earnings management of firms respond to the banking market structure using the following model:

$$\begin{array}{cc}\multicolumn{1}{c}{{\mathrm{EM}}_{i,t}}& \multicolumn{1}{c}{={\alpha }_0+{\alpha }_1\times {\mathrm{EM}}_{i,t-1}+{\alpha }_2\times {\mathrm{COM}}_{i,t-1}}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+\beta \times {\mathrm{FC}}_{i,t-1}+\gamma \times {\mathrm{MC}}_{t-1}+{\alpha }_j+{\epsilon }_{i,t}}\end{array}$$

(7)

In the specified model, EM represents the measure of earnings management, denoted by discretionary accruals, while COM reflects the variable capturing bank competition. FC incorporates firm-level control variables potentially associated with earnings management: firm size (SIZ), return (ROA), market-to-book ratio (TOQ), financial leverage (LEV), tangible fixed assets (PPE), and state ownership (GOV). MC includes macro-level control variables that might influence firms’ financial reporting quality, including economic cycles (GDP), money supply (M2), and two macro shocks reflecting the financial crisis and the coronavirus pandemic (CRI and COV). All control variables are adopted based on the related literature (Jiang et al., 2022; Kim & Yasuda, 2021; Yung & Root, 2019). Our model encompasses industry-fixed effects to account for variations in earnings management attributable to cross-industry disparities, thus mitigating time-invariant unobserved industry biases. Time-fixed effects are not added, given that each firm i in year t is assigned the common bank competition of year t; thus, incorporating time-fixed effects would automatically destroy the explanatory power of competition variables. All explanatory variables, including bank competition, firm-level controls, and macroeconomic variables, are lagged by 1 year to mitigate potential reverse causality and to reflect the delayed adjustment of firms’ reporting behavior to changes in banking conditions, firm fundamentals, and the macroeconomic environment. Robust standard errors are clustered at the firm level.

For the econometric method, we adopt the two-step system GMM as the primary estimation technique to address the dynamic panel issue and endogeneity bias (Blundell & Bond, 1998; Roodman, 2009). This method has been widely employed in various recent studies focusing on corporate earnings management (El Diri et al., 2020; Fan et al., 2019; Hu, 2021). It is essential to highlight that the GMM approach is better than ordinary least squares (OLS) estimations, which may produce incorrect inferences when applied in a two-step procedure for analyzing the discretionary accruals model (Chen et al., 2018). In implementing this approach, we introduce earnings management indicators with a 1-year lag to the baseline model. Bank competition and macroeconomic variables are treated as exogenous, while firm-specific characteristics are considered endogenous (Cantero-Saiz et al., 2014). To address potential instrument proliferation, we restrict the maximum lag depth to four periods (Bouvatier et al., 2014; Demir & Danisman, 2021). This approach ensures that the number of instruments remains well below the number of firms and mitigates the “too many instruments” problem.

Data

This study explores the influence of bank competition on corporate opacity in Vietnam. To this end, the metric for bank competition is established through a rigorous examination of bank-level data, sourced from 42 commercial banks and 601 bank-year observations. This dataset represents a comprehensive coverage, constituting over 99% of the total banking sector assets in any sample year. The firm-level data originates from listed companies on the Ho Chi Minh (HOSE) and Hanoi (HNX) stock exchanges. A common approach is adopted to adhere to established sampling procedures in the corporate finance literature, resulting in the exclusion of financial firms and utilities, which is also based on our research objective to test the influence of bank competition on corporate behaviors. Firms with missing data crucial for variable calculations are also excluded. The final sample is characterized by 623 unique firms, contributing to 7,589 firm-year observations from 2007 to 2023. This specific sample period, commencing in 2007, is chosen to ensure the inclusion of wholly audited financial data, a guarantee not universally available before this timeframe. The FiinPro database provides the source for data on listed firms and commercial banks, while the World Development Indicators database offers the remaining macro data.

Table 1 provides an overview of all main variables, presenting their detailed definitions along with descriptive statistics. As a precaution against the impact of outliers, all continuous firm-level variables undergo winsorization at the 1st and 99th percentiles. Preliminary findings reveal that the proxy for earnings management has a mean value standing of 0.195 and a noteworthy deviation of 0.116. These statistics suggest that Vietnamese firms exhibit a higher propensity for engaging in earnings management practices compared to their counterparts in other major countries, which yield smaller values of earnings management observed through the same accrual-based method (Jiang et al., 2022; Kim & Yasuda, 2021; Qi et al., 2021).

Table 1.

Summary Statistics of the Regression Sample.

Variables	Mean	SD	Min	Max	Definitions
EM	0.195	0.116	0.016	0.533	Accrual-based earnings management, estimated by Equation 1
CR	0.535	0.135	0.370	0.743	Market shares of the top five largest banks in the sector
HHI	0.051	0.038	0.038	0.061	Sum of the squares of all banks’ market shares
LER	0.390	0.108	0.325	0.527	Difference between price and marginal cost, divided by price, estimated by Equations 2–4
BOO	−0.078	0.012	−0.094	−0.065	Elasticity of profits to marginal costs, estimated by Equation 5
HIN	0.523	0.214	0.385	0.746	Elasticity of bank revenues relative to input prices, estimated by Equation 6
SIZ	27.235	1.575	23.798	31.569	Natural logarithm of total assets
ROA	0.063	0.070	−0.122	0.329	Net income scaled by total assets
TOQ	1.147	0.649	0.283	4.623	Market value of equity divided by its book value
LEV	0.484	0.223	0.038	0.908	Total liabilities as a share of total assets
PPE	0.178	0.178	0.000	0.787	Gross property, plant, and equipment, as a share of total assets
GOV	0.224	0.253	0.000	0.818	Fraction of ownership by the government
GDP	0.058	0.015	0.026	0.080	Growth rate of GDP from year to year
M2	0.166	0.059	0.061	0.297	Growth rate of money supply from year to year
CRI	0.090	0.287	0.000	1.000	Dummy variable taking 1 for observations from 2007 to 2009, and 0 otherwise
COV	0.161	0.367	0.000	1.000	Dummy variable taking 1 for observations from 2020 to 2021, and 0 otherwise
WW	−1.246	0.731	−2.536	1.311	An accounting-based measure of financing constraints, estimated by Equation 8
KZ	0.224	1.827	−4.426	7.672	An accounting-based measure of financing constraints, estimated by Equation 9

Concerning bank competition, the outcomes presented in Table 1, coupled with moderate correlations among different competition proxies (ranging from 0.14 to 0.62, not exhaustively detailed for brevity), indicate the volatility in the banking market structure and supports the notion that each measure captures a specific facet of bank competition. These findings align with those reported by prior research on the Vietnamese banking system (Huynh, 2023), lending credibility to the calculation of our regression sample. Besides, it is noteworthy that, to prevent perfect multicollinearity in regressions, we exclude the dummy variable for the COVID-19 pandemic due to its high correlation with economic growth (.89); other correlation coefficients remain low.

Results

Main Regression Results

This part presents the primary regressions concerning the link between bank competition and firm information disclosure. As illustrated in Table 2 (and subsequent GMM results tables in this study), we have confidence in the validity of the dynamic GMM framework adopted in our analysis.

Table 2.

Bank Competition and Corporate Earnings Management: Baseline Estimations.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.105** (0.041)
HHI		0.610*** (0.175)
LER			0.125*** (0.035)
BOO				1.451*** (0.523)
HIN					−0.040*** (0.012)
SIZ	−0.007*** (0.001)	−0.008*** (0.001)	−0.007*** (0.001)	−0.007*** (0.001)	−0.007*** (0.001)
ROA	0.026 (0.026)	0.030 (0.026)	0.022 (0.026)	0.039 (0.026)	0.023 (0.026)
TOQ	−0.005* (0.003)	−0.005* (0.003)	−0.006** (0.003)	−0.006** (0.003)	−0.005 (0.003)
LEV	0.022*** (0.008)	0.023*** (0.008)	0.023*** (0.008)	0.022*** (0.008)	0.021** (0.008)
PPE	0.016* (0.009)	0.016* (0.009)	0.016* (0.009)	0.015* (0.009)	0.017* (0.009)
GOV	−0.036*** (0.006)	−0.035*** (0.006)	−0.036*** (0.006)	−0.038*** (0.006)	−0.037*** (0.006)
GDP	−0.298*** (0.078)	−0.291*** (0.078)	−0.279*** (0.078)	−0.236*** (0.079)	−0.343*** (0.081)
M2	−0.096*** (0.022)	−0.064*** (0.023)	−0.034 (0.027)	−0.084*** (0.022)	−0.053** (0.025)
CRI	−0.004 (0.008)	0.006 (0.006)	0.005 (0.006)	0.008 (0.006)	0.009 (0.006)
Lagged DV	−0.052*** (0.012)	−0.051*** (0.012)	−0.055*** (0.012)	−0.052*** (0.012)	−0.055*** (0.012)
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	132	132	132	132	132
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.899	0.912	0.797	0.917	0.802
Hansen test	0.168	0.245	0.237	0.166	0.210

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

We commence our analysis by scrutinizing the estimates related to market concentration, as indicated by the CR and HHI variables. Columns 1 to 2 of Table 2 present statistically significant positive coefficients on these concentration variables. These findings imply that firms operating within highly concentrated banking markets tend to engage in accrual earnings management to a greater extent compared to those in less concentrated banking markets. Furthermore, considering that greater competition corresponds to lower market concentration, these results suggest diminished employment of accrual earnings management in more competitive banking environments. The economic significance of bank competition is noteworthy, as evidenced by the calculation that a one-standard-deviation change in the level of bank concentration (CR or HHI) results in a variation of 0.014 to 0.023 unit points in the level of accrual earnings management, equivalent to approximately 7.27%–11.89% of the mean value of EM. This magnitude indicates that changes in banking market structure can generate economically meaningful variations in firms’ financial reporting behavior. From a policy perspective, these results suggest that fostering a competitive banking environment may indirectly enhance financial reporting quality by alleviating firms’ financing pressures and reducing managerial incentives to inflate earnings. Such implications are particularly relevant for bank-based emerging economies, where firms rely heavily on bank credit and where improvements in banking competition may contribute to stronger financial transparency.

Remarkably, the finding above remains consistent even after employing three supplementary non-structural competition metrics. In detail, the coefficient associated with the Lerner and Boone indicators is positive and statistically significant. Results from these indicators, which capture the inverse measure of competition, indicate that heightened banking competition diminishes firms’ inclination to manipulate financial reporting. Furthermore, the coefficient linked to the Panzar-Rosse H-statistic emerges as negative and significant, affirming that firms tend to exhibit reduced opacity within banking markets characterized by greater competitive pressure. In magnitude, the impacts of various non-structural competition measures may not appear insignificant. A one standard deviation alteration in the three additional metrics (LER, BOO, or HIN) triggers a fluctuation of 0.009 to 0.017 unit points in the level of earnings management, amounting to approximately 4.39%–8.93% of the mean value of EM. Taken together, these results are more consistent with Hypothesis B than with Hypothesis A, suggesting that the primary effect of bank competition operates through the alleviation of financing constraints rather than through the monitoring channel. The findings also highlight the broader regulatory relevance of maintaining effective competition within banking systems in emerging markets.

Robustness Checks

Alternative Measures of Bank Competition

To measure bank competition, we employ various metrics, namely the Lerner index, the Boone indicator, the H-statistic, and the concentration index. However, the lack of a standardized calculation method for each measurement introduces potential variability in research outcomes. In this subsection, we apply alternative methods to compute each bank competition measure. To this end, we select total loans as a representation of banks’ market shares, replacing total assets when calculating two concentration variables. We prevent potential overestimation of the Lerner index due to market power by adopting a straightforward solution that excludes the cost of funds from the cost function (Turk Ariss, 2010). Furthermore, we employ pooled OLS estimation as the basis for the Boone and H-statistic computations. The original competition measures are replaced with these newly derived proxies as the main independent variable to replicate all baseline tests, and the results in Table 3 remain unchanged.

Table 3.

Robustness Tests with Alternative Measures of Bank Competition.

Variables	(1)	(2)	(3)	(4)	(5)
CR_loan	0.117*(0.071)
HHI_loan		0.511***(0.147)
LER_adj			0.122***(0.036)
BOO_adj				1.317***(0.509)
HIN_adj					−0.052***(0.014)
SIZ	−0.007***(0.001)	−0.008***(0.001)	−0.007***(0.001)	−0.007***(0.001)	−0.007***(0.001)
ROA	0.023(0.026)	0.033(0.026)	0.028(0.027)	0.030(0.026)	0.024(0.026)
TOQ	−0.006**(0.003)	−0.006**(0.003)	−0.005*(0.003)	−0.006**(0.003)	−0.005*(0.003)
LEV	0.021**(0.008)	0.023***(0.008)	0.026***(0.009)	0.023***(0.008)	0.022***(0.008)
PPE	0.016*(0.009)	0.016*(0.009)	0.016*(0.009)	0.016*(0.009)	0.016*(0.009)
GOV	−0.036***(0.006)	−0.036***(0.006)	−0.038***(0.006)	−0.037***(0.006)	−0.036***(0.006)
GDP	−0.322***(0.084)	−0.296***(0.078)	−0.282***(0.080)	−0.349***(0.084)	−0.265***(0.078)
M2	−0.109***(0.023)	−0.059**(0.024)	−0.041(0.028)	−0.061**(0.025)	−0.028(0.028)
CRI	0.000(0.008)	0.009(0.006)	0.005(0.007)	0.012*(0.007)	0.008(0.006)
Lagged DV	−0.052***(0.012)	−0.051***(0.012)	−0.052***(0.013)	−0.052***(0.012)	−0.055***(0.012)
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	132	132	132	132	132
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.927	0.933	0.855	0.901	0.793
Hansen test	0.129	0.241	0.107	0.161	0.251

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Alternative Measures of Earnings Management

For another step of robustness checks, we incorporate two additional methods to compute accrual-based earnings management. First, we adopt the methodology employed in the Hribar and Collins (2002) model, which is expressed as follows:

$$\begin{array}{cc}\multicolumn{1}{c}{{\mathrm{TC}1}_{i,t}/{\mathrm{TA}}_{i,t-1}=}& \multicolumn{1}{c}{{\alpha }_1\times (1/{\mathrm{TA}}_{i,t-1})+{\alpha }_2}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{\Delta }{\mathrm{SR}}_{i,t}/T{A}_{i,t-1})+{\alpha }_3}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times ({\mathrm{PPE}}_{i,t}/T{A}_{i,t-1})+{\epsilon }_{i,t}}\end{array}$$

(8)

where $\mathrm{TC}1$ denotes the aggregate of accruals, calculated as the disparity between operating income and operating cash flow. The remaining factors are consistent with the modified Jones model depicted in Equation 1. The absolute value of the residuals gives us the first alternative proxy of accrual-based earnings management (EM1).

Second, we apply the framework introduced by McNichols (2002), an adaptation of the Hribar and Collins (2002) model crafted for accrual estimation. This revised model integrates additional variables to consider past, present, and future operating cash flows, as articulated below:

$$\begin{array}{cc}\multicolumn{1}{c}{{\mathrm{TC}1}_{i,t}/{\mathrm{TA}}_{i,t-1}=}& \multicolumn{1}{c}{{\alpha }_1\times (1/{\mathrm{TA}}_{i,t-1})+{\alpha }_2}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{\Delta }{\mathrm{SR}}_{i,t}/T{A}_{i,t-1})+{\alpha }_3}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times ({\mathrm{PPE}}_{i,t}/T{A}_{i,t-1})+{\alpha }_4}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times ({\mathrm{OCF}}_{i,t-1}/T{A}_{i,t-1})+{\alpha }_5}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times ({\mathrm{OCF}}_{i,t}/T{A}_{i,t-1})+{\alpha }_6}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times ({\mathrm{OCF}}_{i,t+1}/T{A}_{i,t-1})+{\epsilon }_{i,t}}\end{array}$$

(9)

where ${\mathrm{OCF}}_{i,t-1}$ , ${\mathrm{OCF}}_{i,t}$ , and ${\mathrm{OCF}}_{i,t+1}$ reflects the cash flows from operations of firm i at time $t-1$ , t, and $t+1$ , respectively. All other variables maintain identical definitions as previously outlined. Following the same spirit, we take the absolute value of the residuals as the second alternative proxy for earnings management (EM2).

In Table 4, we substitute the original opacity variable with two newly-created earnings management variables, and the results remain unaltered (though the significance fades away in columns 4 and 5). Bank competition continues to discourage firms from participating in activity manipulations, regardless of the various models employed to measure accrual-based earnings management.

Table 4.

Robustness Tests with Alternative Measures of Earnings Management.

	Dependent variable: EM1 (columns 1–5)					Dependent variable: EM2 (columns 6–10)
Variables	(1) CR	(2) HHI	(3) LER	(4) BOO	(5) HIN	(6) CR	(7) HHI	(8) LER	(9) BOO	(10) HIN
Competition	0.032*** (0.009)	0.080** (0.031)	0.021*** (0.006)	0.012 (0.089)	−0.002 (0.002)	0.022** (0.010)	0.084** (0.041)	0.037*** (0.010)	0.234** (0.113)	−0.004* (0.002)
SIZ	−0.002*** (0.000)	−0.002*** (0.000)	−0.002*** (0.000)	−0.002*** (0.000)	−0.002*** (0.000)	−0.001*** (0.000)	−0.001*** (0.000)	−0.004*** (0.001)	−0.002*** (0.001)	−0.001*** (0.000)
ROA	−0.016** (0.007)	−0.013* (0.007)	−0.011* (0.007)	−0.011* (0.007)	−0.011* (0.007)	−0.011* (0.007)	−0.011 (0.007)	0.001 (0.022)	−0.008 (0.008)	−0.012* (0.007)
TOQ	−0.004*** (0.001)	−0.004*** (0.001)	−0.004*** (0.001)	−0.004*** (0.001)	−0.004*** (0.001)	−0.003*** (0.001)	−0.003*** (0.001)	−0.009*** (0.003)	−0.004*** (0.001)	−0.003*** (0.001)
LEV	−0.003 (0.002)	−0.003 (0.002)	−0.002 (0.002)	−0.002 (0.002)	−0.002 (0.002)	0.008*** (0.002)	0.008*** (0.002)	0.013 (0.013)	0.006* (0.003)	0.008*** (0.002)
PPE	0.035*** (0.003)	0.036*** (0.003)	0.035*** (0.003)	0.036*** (0.003)	0.036*** (0.003)	0.022*** (0.003)	0.022*** (0.003)	0.065*** (0.013)	0.024*** (0.004)	0.022*** (0.003)
GOV	−0.006*** (0.001)	−0.006*** (0.001)	−0.006*** (0.001)	−0.006*** (0.001)	−0.006*** (0.001)	−0.003** (0.001)	−0.003** (0.001)	−0.007 (0.008)	−0.002 (0.003)	−0.003** (0.001)
GDP	−0.054*** (0.013)	−0.031*** (0.011)	−0.030*** (0.011)	−0.030** (0.012)	−0.031*** (0.011)	−0.040** (0.016)	−0.034** (0.015)	−0.061*** (0.019)	−0.016 (0.017)	−0.025* (0.014)
M2	0.004 (0.004)	0.004 (0.004)	0.016*** (0.005)	0.007* (0.004)	0.006 (0.004)	0.007 (0.006)	0.004 (0.006)	0.003 (0.007)	0.021*** (0.007)	0.008 (0.006)
CRI	0.006*** (0.001)	0.004*** (0.001)	0.004*** (0.001)	0.004*** (0.001)	0.004*** (0.001)	0.007*** (0.002)	0.006*** (0.002)	0.005*** (0.002)	0.006*** (0.002)	0.006*** (0.002)
Lagged DV	0.231*** (0.015)	0.229*** (0.015)	0.234*** (0.015)	0.231*** (0.015)	0.230*** (0.015)	0.287*** (0.017)	0.288*** (0.017)	0.241*** (0.022)	0.274*** (0.019)	0.291*** (0.017)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966	6,354	6,354	6,354	6,354	6,354
Firms	623	623	623	623	623	612	612	612	612	612
Instruments	132	132	132	132	132	132	132	132	132	132
AR(1) test	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.208	0.213	0.204	0.209	0.212	0.313	0.184	0.280	0.243	0.268
Hansen test	0.122	0.201	0.102	0.101	0.200	0.113	0.115	0.155	0.111	0.165

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. Competition measures are shown at the top of the columns. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Further Addressing Endogeneity with External Instruments

Despite implementing strategies such as (i) lagging variables to mitigate reverse causality, (ii) adding many control variables and fixed effects to address the issue of omitted variables, and (iii) utilizing the dynamic GMM setting with internal instruments, one may call for further efforts in dealing with endogeneity. Therefore, we perform a two-stage least squares (2SLS) regression analysis, relying on past levels of financial development and bank outcomes in Vietnam as instruments, in line with prior research (Fungáčová et al., 2017). These lagged variables capture past conditions of the banking system that are closely related to the current level of bank competition, thereby satisfying the relevance condition. At the same time, transformations in banking market structure and financial development typically evolve gradually over time, implying that past values of these aggregate banking indicators are unlikely to directly affect firms’ contemporaneous earnings management decisions, except through their impact on current banking conditions. This characteristic supports the exclusion restriction. In this regard, we use the 5-year lags of financial development, represented by domestic credit to GDP (L5_CREDITGDP), and bank return, represented by net income to equity (L5_BANKROE). These two instruments are designed at the country level.

Note that, in the 2SLS analysis presented in Table 5, other terms of no main interest are suppressed (though they are always used in regressions). The first-stage estimates (columns 1, 3, 5, 7, and 9) indicate significant associations between the instrumental variables and bank competition. Statistical tests also validate the instrumental variables. The second-stage results (columns 2, 4, 6, 8, and 10) demonstrate that measures of bank competition maintain a significant linkage with corporate opacity in an unaltered manner. In summary, the results of the 2SLS specification addressing endogeneity concerns continue to support our established findings.

Table 5.

Further Addressing Endogeneity with External Instruments.

	(1) First stage	(2) Second stage	(3) First stage	(4) Second stage	(5) First stage	(6) Second stage	(7) First stage	(8) Second stage	(9) First stage	(10) Second stage
Variables	– CR	– EM	– HHI	– EM	– LER	– EM	– BOO	– EM	– HIN	– EM
L5_BANKROE	−0.005*** (0.000)		−0.001*** (0.000)		−0.004*** (0.000)		0.001*** (0.000)		−0.017*** (0.000)
L5_CREDITGDP	0.001*** (0.000)		0.001*** (0.000)		0.001*** (0.000)		−0.001*** (0.000)		−0.004*** (0.000)
CR		0.384*** (0.096)
HHI				0.680** (0.296)
LER						1.031*** (0.257)
BOO								6.683** (2.653)
HIN										−0.077*** (0.029)
Firm controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966
R-squared	0.707		0.718		0.612		0.490		0.564
Anderson canon. corr. LM statistic	2,952.61***		3,845.52***		175.15***		435.6***		1907.08***
Cragg-Donald Wald F statistic	2,757.2		4,874.89		89.91		233.45		1,361.12

Note. This table reports the first- and second-stage results from the 2SLS approach. The sample consists of nonfinancial listed firms in Vietnam. The dependent variables (DV) are shown at the top. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

**

p < .05. ***p < .01.

Underlying Mechanisms

The Direction of Earnings Management

To further explore how bank competition affects firms’ earnings management behavior, we divide the sample into upward and downward subsamples based on the sign of residuals from the modified Jones model (positive for upward and negative for downward earnings management). Importantly, the dependent variable remains the absolute value of discretionary accruals in all regressions, with the distinction implemented solely through sample partitioning. We then re-estimate the baseline model for each subsample.

Table 6 presents the subsample findings. Columns 1 to 5 indicate that the coefficients on competition measures remain significant and consistent with those observed in the baseline estimation for the entire sample. This suggests that bank competition significantly reduces firms’ propensity for upward earnings management. Conversely, columns 6 to 10 demonstrate that bank competition does not exert a notable impact on downward earnings management. In contexts characterized by lower competition or greater monopolization within the banking sector, firms may encounter heightened financing constraints (Khan & Kutan, 2023; Leon, 2015). In response to such constraints, managers may be incentivized to engage in upward earnings management to bolster the firm’s reputation and facilitate easier access to external financing (Bermpei et al., 2022; Iatridis & Kadorinis, 2009). Accordingly, heightened bank competition may mitigate managers’ inclination to manipulate earnings upward.

Table 6.

Decomposing into Income-Increasing and Income-Decreasing Earnings Management.

	Subsample A: Income-increasing management (columns 1–5)					Subsample B: Income-decreasing management (columns 6–10)
Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
CR	0.107*** (0.041)					0.160 (0.150)
HHI		0.691*** (0.177)					0.932 (0.608)
LER			0.164*** (0.034)					−0.007 (0.092)
BOO				1.428*** (0.508)					−0.810 (1.407)
HIN					−0.058*** (0.012)					0.006 (0.028)
Lagged DV	−0.046*** (0.014)	−0.044*** (0.014)	−0.048*** (0.014)	−0.046*** (0.014)	−0.047*** (0.014)	−0.046** (0.020)	−0.048** (0.019)	−0.046** (0.020)	−0.047** (0.021)	−0.050** (0.022)
Firm controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	5,315	5,315	5,315	5,315	5,315	1,651	1,651	1,651	1,651	1,651
Firms	521	521	521	521	521	179	179	179	179	179
Instruments	132	132	132	132	117	117	117	117	117	117
AR(1) test	0.000	0.000	0.000	0.000	0.000	0.002	0.001	0.002	0.001	0.002
AR(2) test	0.824	0.842	0.644	0.805	0.772	0.941	0.933	0.929	0.931	0.906
Hansen test	0.138	0.180	0.124	0.136	0.121	0.311	0.310	0.269	0.279	0.278

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

**

p < .05. ***p < .01.

Tests for the Mechanism of Financial Constraints

Based on the observed direction in earnings management, we hypothesize that financial constraints may serve as a significant economic mechanism through which less bank competition prompts managers to engage in upward earnings management. To substantiate this hypothesis, we undertake the following steps. First, we examine the impact of bank competition on corporate financial constraints – the mediating variable. Second, we proceed to evaluate the influence of firms’ financial constraints on their earnings management. Consistent with prior research, we utilize the WW index to assess a firm’s level of financial constraints (Whited & Wu, 2006), supplemented by the KZ index for robustness assessments (Kaplan & Zingales, 1997). Specifically, we construct these indices using the following formulas:

$$\begin{array}{cc}\multicolumn{1}{c}{\mathrm{WW}}& \multicolumn{1}{c}{=-0.091\times (\mathrm{operating}\mathrm{cash}\mathrm{flow}/\mathrm{total}\mathrm{assets})}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{-0.062\times \mathrm{dividend}\mathrm{dummy}+0.021}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{long}-\mathrm{term}\mathrm{debt}/\mathrm{total}\mathrm{assets})-0.044}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times \ln (\mathrm{total}\mathrm{assets})+0.1021}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times \mathrm{annual}\mathrm{industry}\mathrm{growth}-0.035}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times \mathrm{firm}\mathrm{annual}\mathrm{growth}}\end{array}$$

(10)

$$\begin{array}{cc}\multicolumn{1}{c}{\mathrm{KZ}}& \multicolumn{1}{c}{=-1.002\times (\mathrm{operating}\mathrm{cash}\mathrm{flow}/\mathrm{total}\mathrm{assets})+0.283}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{market}\mathrm{value}\mathrm{of}\mathrm{assets}/\mathrm{book}\mathrm{value}\mathrm{of}\mathrm{assets})}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{+3.139\times (\mathrm{total}\mathrm{debt}/\mathrm{total}\mathrm{assets})-39.368}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{dividend}\mathrm{payments}/\mathrm{total}\mathrm{assets})-1.315}\\ \multicolumn{1}{c}{}& \multicolumn{1}{c}{\times (\mathrm{cash}\mathrm{and}\mathrm{cash}\mathrm{equivalents}/\mathrm{total}\mathrm{assets})}\end{array}$$

(11)

Table 7 presents the estimations concerning the channel of financial constraints, as captured by the WW index. Initially, we investigate the impact of bank competition on financial constraints, incorporating relevant control variables as recommended in existing literature (Khan & Kutan, 2023; Leon, 2015; Love & Martínez Pería, 2015; Saeed & Vincent, 2012). The results depicted in columns 1 to 5 offer support for the market power hypothesis (with the exception of HHI), indicating that less bank competition amplifies firms’ financial constraints. This assertion is evidenced by the significantly positive coefficients associated with CR, LER, and BOO, alongside the significantly negative coefficient on HIN.

Table 7.

Mechanism Tests for Financial Constraints (as Proxied by the WW Index).

	Dependent variable: WW (columns 1–5)					Dependent variable: EM (columns 6–10)
Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
WW						0.217** (0.110)	0.301*** (0.113)	0.273** (0.116)	0.473** (0.189)	0.446** (0.178)
CR	0.082*** (0.013)					0.099* (0.055)
HHI		0.061 (0.070)					0.681*** (0.235)
LER			0.075*** (0.010)					0.160*** (0.060)
BOO				1.147*** (0.172)					2.432*** (0.730)
HIN					−0.027*** (0.004)					−0.039** (0.019)
Lagged DV	0.239*** (0.019)	0.242*** (0.018)	0.224*** (0.018)	0.250*** (0.019)	0.246*** (0.018)	−0.053*** (0.015)	−0.052*** (0.015)	−0.051*** (0.015)	−0.059*** (0.016)	−0.071*** (0.016)
Firm controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623	623	623	623	623	623
Instruments	132	132	132	132	132	132	132	132	132	132
AR(1) test	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.442	0.483	0.656	0.408	0.449	0.262	0.281	0.289	0.201	0.256
Hansen test	0.133	0.115	0.164	0.129	0.159	0.173	0.140	0.132	0.225	0.356

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. Bank competition affects financing constraints contemporaneously, while financial constraints enter the earnings management regressions with a one-period lag, thereby mitigating simultaneity and reverse-causality concerns. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Subsequently, we introduce financial constraints into the earnings management regressions in columns 6 to 10, where both bank competition and the intermediate variable are jointly included. Importantly, to address potential simultaneity concerns, our empirical design explicitly accounts for timing: bank competition affects firms’ financing constraints contemporaneously, reflecting that banking market conditions shape firms’ access to credit within the same period, whereas financing constraints enter the earnings management regressions with a one-period lag. This structure mitigates reverse causality and simultaneity between earnings management and accounting-based constraint measures. Consistent with prior studies, the coefficients on lagged financial constraints are uniformly positive and statistically significant, corroborating the view that tighter financing conditions strengthen managers’ incentives to engage in earnings management (Bermpei et al., 2022; Iatridis & Kadorinis, 2009; Kurt, 2018). Taken together, we document that reduced bank competition or increased bank concentration intensifies financial constraints, prompting managers to inflate earnings strategically to facilitate firms’ access to external funding.

We observe a consistent pathway in Table 8, employing the KZ index to gage financing constraints. Overall, our findings affirm that financing constraints serve as a mediator between bank competition and corporate opacity. Concretely, bank competition attenuates corporate earnings management by alleviating corporate financial constraints.

Table 8.

Mechanism Tests for Financial Constraints (as Proxied by the KZ Index).

	Dependent variable: KZ (columns 1–5)					Dependent variable: EM (columns 6–10)
Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
KZ						0.010*** (0.002)	0.009*** (0.001)	0.008*** (0.001)	0.010*** (0.001)	0.009*** (0.001)
CR	9.847*** (2.358)					0.186*** (0.052)
HHI		62.017*** (13.617)					0.566*** (0.188)
LER			5.030*** (1.745)					0.106*** (0.037)
BOO				111.313*** (30.111)					1.527*** (0.528)
HIN					−1.713*** (0.596)					−0.036*** (0.013)
Lagged DV	0.417*** (0.018)	0.427*** (0.017)	0.423*** (0.018)	0.423*** (0.017)	0.428*** (0.018)	−0.043*** (0.013)	−0.045*** (0.012)	−0.048*** (0.012)	−0.044*** (0.013)	−0.047*** (0.012)
Firm controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623	623	623	623	623	623
Instruments	132	132	132	132	132	132	132	132	132	132
AR(1) test	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.436	0.411	0.417	0.428	0.404	0.883	0.907	0.823	0.859	0.801
Hansen test	0.132	0.149	0.121	0.147	0.142	0.106	0.183	0.170	0.169	0.174

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. Bank competition affects financing constraints contemporaneously, while financial constraints enter the earnings management regressions with a one-period lag, thereby mitigating simultaneity and reverse-causality concerns. Figures in parentheses are standard errors.

***

p < .01.

Additional Analysis

The Moderating Role of Bank-Firm Relationship

Given the validated impact of the competition of banks on the earnings management of nonfinancial firms, it is imperative to explore the influence of the bank-firm relationship on this impact. Firms that maintain close ties with banks often receive credit assistance (Tran, 2020), thereby mitigating the impact of banking market dynamics, particularly when the mechanism linking competition to earnings management involves financial constraints. Hence, we predict that the link between bank competition and earnings management is attenuated for firms with a bank-firm relationship.

To verify this prediction, we introduce an interaction term between bank competition and the bank-firm relationship. We gage the bank-firm relationship (Bankfirm) using a dummy variable indicating whether a firm has had a long-term loan in the previous year (G. Zhang et al., 2015). Companies with access to long-term debt face fewer financial constraints compared to those relying solely on short-term debt for financing (Pál & Kozhan, 2009). We expect the coefficient of this interaction to exhibit an opposite sign compared to the individual competition variables. The regression results reported in Table 9 agree with our expectation, although the interaction terms created by HHI and BOO are not statistically significant. We also apply an alternative proxy of the bank-firm relationship using the bank debt ratio (captured by the ratio of a firm’s bank debt to its total assets) and obtain findings that do not change substantially. We do not report these estimates to save space. Overall, our findings provide sufficient evidence to suggest that firms with established bank-firm relationships are less influenced by bank competition, likely due to their assured access to credit from their banks.

Table 9.

Bank Competition and Corporate Earnings Management: The Moderating Role of Bank-Firm Relationships.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.096** (0.046)
CR × Bankfirm	−0.022*** (0.008)
HHI		0.613** (0.250)
HHI × Bankfirm		−0.001 (0.357)
LER			0.082** (0.041)
LER × Bankfirm			−0.036*** (0.006)
BOO				1.012* (0.554)
BOO × Bankfirm				−0.316 (0.256)
HIN					−0.038*** (0.013)
HIN × Bankfirm					0.016* (0.009)
Bankfirm	−0.047** (0.019)	−0.040** (0.018)	−0.035* (0.019)	−0.045** (0.019)	−0.045*** (0.017)
Lagged DV	−0.066*** (0.013)	−0.063*** (0.013)	−0.064*** (0.013)	−0.065*** (0.013)	−0.067*** (0.013)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.578	0.603	0.621	0.638	0.513
Hansen test	0.194	0.265	0.208	0.161	0.233

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Heterogeneity Effect by Firm Size, State Ownership, and Market Power

We further investigate the impact of firm size, state ownership, and market power on the interplay between bank competition and earnings management. Small firms often grapple with pronounced financial constraints, relying heavily on the banking sector for funding due to more limited access to external financing (Khan & Kutan, 2023). In contrast, larger enterprises are typically less affected by such constraints, benefiting from better access to external financing owing to their established market presence (Bates et al., 2009).

Research on firms’ financial constraints has revealed that ownership structure can significantly impact a firm’s ability to secure external financing. State-owned enterprises, in particular, are often found to face fewer credit constraints and enjoy greater access to external financing compared to their non-state-owned counterparts (Poncet et al., 2010; Shi & Zhang, 2018). In the context of Vietnam, where government influence over banks is pronounced, state-owned firms may benefit from more favorable financing terms, particularly when dealing with state-owned banks.

We also explore the potential impact of market power on our findings. Firms with significant market power often wield influence in negotiating favorable credit terms with their suppliers (Love & Zaidi, 2010). Additionally, dominant firms possess alternative avenues for securing financing beyond simply seeking extended credit periods from suppliers (Jory et al., 2020). Given their strong bargaining position, these firms can negotiate advantageous loan terms with banks. In less competitive banking markets, such firms may find it easier to access bank credit at favorable terms, potentially avoiding heightened financing constraints (Fabbri & Klapper, 2016).

We perform the work by introducing interaction terms between bank competition and each moderator (i.e., SIZ, GOV, and POW). Following standard practice, we proxy market power (POW) using firms’ gross profit margin, as higher margins indicate stronger pricing power and a greater ability to pass costs onto customers, thereby capturing firms’ relative competitive position in the product market. The results regarding the moderating effect of firm size are summarized in Table 10. Notably, across all five specifications, the interaction term exhibits a sign opposite to that of the standalone competition measures. This suggests that firm size mitigates the influence of bank competition on firm opacity. In other words, the effect of banking market structure on corporate earnings management is more pronounced in smaller firms. Turning to Table 11, where we explore the conditioning role of state ownership, we find that the coefficient of the interaction terms is significantly negative in the columns corresponding to CR, HHI, and BOO, while it is significantly positive in the case of HIN. Similar to the findings for firm size, our results regarding state ownership indicate that this factor diminishes the impact of competition on corporate earnings management. Stated differently, the influence of banking market structure on firm opacity is stronger for non-state-owned enterprises. Lastly, in Table 12, all estimates consistently reveal a pattern: bank competition exerts more pronounced effects on firms’ earnings manipulations when corporate market power is lower.

Table 10.

Bank Competition and Corporate Earnings Management: The Moderating Role of Firm Size.

Variables	(1)	(2)	(3)	(4)	(5)
CR	5.604*** (1.033)
CR × SIZ	−0.215*** (0.038)
HHI		49.636*** (9.205)
HHI × SIZ		−1.867*** (0.340)
LER			4.145*** (1.161)
LER × SIZ			−0.155*** (0.042)
BOO				34.728*** (7.642)
BOO × SIZ				−1.343*** (0.281)
HIN					−3.515*** (0.805)
HIN × SIZ					0.130*** (0.029)
SIZ	−0.121*** (0.020)	−0.103*** (0.017)	−0.068*** (0.016)	−0.102*** (0.020)	−0.087*** (0.018)
Lagged DV	−0.054*** (0.013)	−0.058*** (0.013)	−0.061*** (0.013)	−0.054*** (0.013)	−0.062*** (0.013)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.828	0.684	0.668	0.926	0.142
Hansen test	0.177	0.281	0.191	0.184	0.114

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

***

p < .01.

Table 11.

Bank Competition and Corporate Earnings Management: The Moderating Role of State Ownership.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.241*** (0.059)
CR × GOV	−0.234* (0.122)
HHI		2.552* (1.303)
HHI × GOV		−1.433*** (0.364)
LER			0.124** (0.049)
LER × GOV			−0.042 (0.154)
BOO				2.169** (1.026)
BOO × GOV				−1.734*** (0.539)
HIN					−0.044*** (0.013)
HIN × GOV					0.020** (0.008)
GOV	−0.059** (0.024)	−0.024 (0.027)	−0.031 (0.027)	−0.079*** (0.022)	−0.050** (0.025)
Lagged DV	−0.047*** (0.012)	−0.047*** (0.012)	−0.052*** (0.012)	−0.048*** (0.012)	−0.052*** (0.012)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.949	0.963	0.842	0.952	0.860
Hansen test	0.138	0.178	0.143	0.188	0.129

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Table 12.

Bank Competition and Corporate Earnings Management: The Moderating Role of Firm Market Power.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.076* (0.044)
CR × POW	−0.184* (0.097)
HHI		0.702*** (0.185)
HHI × POW		−2.079* (1.104)
LER			0.104*** (0.036)
LER × POW			−0.307** (0.122)
BOO				3.494*** (0.932)
BOO × POW				−20.144** (9.467)
HIN					−0.023* (0.014)
HIN × POW					0.189** (0.085)
POW	−0.008 (0.043)	0.092* (0.049)	−0.042 (0.043)	1.560** (0.682)	−0.033 (0.045)
Lagged DV	−0.055*** (0.013)	−0.066*** (0.014)	−0.053*** (0.013)	−0.067*** (0.014)	−0.054*** (0.013)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	134	134	134	134	134
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.603	0.636	0.509	0.898	0.459
Hansen test	0.143	0.220	0.223	0.285	0.173

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

As anticipated, our findings in this subsection consistently indicate that the impact of bank competition on corporate opacity is notably more pronounced in highly financially constrained firms compared to those with lower financial constraints. Variations in firm size, ownership structures, and corporate market power lead to divergent responses to bank competition and its influence on earnings management. Large firms, state-owned enterprises, and those with substantial market power enjoy greater access to financial resources, enabling them to navigate environmental shifts effectively. As a result, these firms are less susceptible to significant effects from bank competition on their earnings management practices and less inclined to alter their financial reporting behaviors.

Do Macroeconomic Conditions Matter?

The influence of bank competition on firms may rely not solely on firm-level attributes but also on country-level factors. It has been posited that banks possessing some degree of monopoly power tend to accrue greater profits than their competitive counterparts. Consequently, they exhibit a heightened aversion to risk and are less inclined to extend credit during tough times (Matutes & Vives, 2000). In such a scenario, it is anticipated that banking competition may play a greater role in alleviating financial constraints during crises than normal economic conditions. Moreover, competitive banks may adopt a less restrictive stance during financial upheavals, driven by the imperative of fierce competition to undertake greater risks to attract new borrowers and bolster market shares (Allen & Gale, 2004).

Theoretical underpinnings also suggest that financial institutions adopt a more cautious stance during economic contractions due to heightened risks of loan defaults among corporations and consumers. This cautious approach often manifests in tighter credit conditions, exacerbating firms’ challenges in seeking access to necessary funds. Indeed, studies have demonstrated that countries experiencing pronounced financial and health crises tend to exhibit more constrained external financing (Ivashina & Scharfstein, 2010; Çolak & Öztekin, 2021). Conversely, during periods of economic expansion, the overall business context tends to improve, facilitating easier access to financing for firms and reducing corporate financing frictions. Though in a less competitive banking sector, firms are less likely to encounter financing difficulties amid economic expansion; however, they may experience heightened financing challenges during economic downturns (Khan & Kutan, 2023).

Considering the arguments outlined above, it is plausible that the influence of banking market structure on firms’ earnings management becomes less pronounced under more favorable economic conditions, when financing constraints for firms may carry less significance. However, an alternative hypothesis warrants careful consideration regarding this impact. If the link between bank competition and upward earnings manipulation is indeed rooted in firms’ financing constraints, one might anticipate a weaker relationship in unfavorable economic conditions because, during such periods, firms typically scale back their investment and financing activities (Demirgüç-Kunt et al., 2020), thereby diminishing the importance of financing constraints relative to periods of economic prosperity.

We verify our conjectures by introducing the interaction of macroeconomic conditions with banking competition. The results presented in Table 13 shed light on the relationship between bank competition and corporate earnings management amidst the global financial crisis. We observe that across all estimates, the interaction terms display significant and similar signs compared to the competition measures, indicating an amplification of the impact of bank competition on firm opacity during the financial crisis. This observation aligns with findings from the COVID-19 pandemic, as evidenced by the results outlined in Table 14. Here, we document a heightened effect of bank competition on corporate opacity during periods of health crisis. Remarkably, this pattern persists across various proxies for bank competition. In unreported robustness checks, we exclude all periods of the financial and health crises, focusing instead on the shorter period from 2010 to 2019. We observe diminished responses in this non-crisis subsample, further corroborating our findings derived from the interaction terms.

Table 13.

Bank Competition and Corporate Earnings Management in the Context of the Financial Crisis.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.248*** (0.078)
CR × CRI	0.109** (0.043)
HHI		3.042*** (0.976)
HHI × CRI		0.640*** (0.182)
LER			0.481*** (0.076)
LER × CRI			0.134*** (0.037)
BOO				2.732*** (0.788)
BOO × CRI				1.475*** (0.538)
HIN					−0.042*** (0.013)
HIN × CRI					−0.036*** (0.006)
CRI	0.013 (0.027)	0.017 (0.027)	0.010 (0.027)	0.025 (0.028)	0.011 (0.027)
Lagged DV	−0.059*** (0.013)	−0.058*** (0.013)	−0.058*** (0.013)	−0.058*** (0.014)	−0.061*** (0.013)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.813	0.833	0.789	0.823	0.721
Hansen test	0.113	0.199	0.192	0.110	0.154

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Table 14.

Bank Competition and Corporate Earnings Management in the Context of the COVID-19 Pandemic.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.199*** (0.074)
CR × COV	0.173*** (0.034)
HHI		0.551* (0.318)
HHI × COV		0.018** (0.009)
LER			0.232*** (0.051)
LER × COV			0.031* (0.017)
BOO				1.129* (0.641)
BOO × COV				0.848*** (0.164)
HIN					−0.040** (0.016)
HIN × COV					−0.016* (0.009)
COV	−0.008 (0.005)	−0.003 (0.005)	−0.032*** (0.008)	−0.001 (0.004)	−0.012** (0.006)
Lagged DV	−0.050*** (0.012)	−0.055*** (0.014)	−0.055*** (0.014)	−0.053*** (0.012)	−0.051*** (0.012)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.881	0.889	0.850	0.822	0.879
Hansen test	0.549	0.403	0.434	0.453	0.295

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

To bolster the evidence regarding macroeconomic shocks, we conduct estimations allowing for the interaction terms between bank competition and the GDP variable, which reflects economic cycles. As depicted in Table 15, across all regressions, the interaction term exhibits a sign opposite to that of the competition variable. These estimates indicate that the association between bank competition and accruals is less pronounced during economic upturns. Remarkably, the findings from this analysis support those observed when investigating the significance of crisis periods.

Table 15.

Bank Competition and Corporate Earnings Management: The Moderating Role of Economic Cycles.

Variables	(1)	(2)	(3)	(4)	(5)
CR	0.453*** (0.138)
CR × GDP	−0.134*** (0.044)
HHI		4.037*** (1.263)
HHI × GDP		−0.945*** (0.203)
LER			0.626*** (0.187)
LER × GDP			−0.199*** (0.042)
BOO				3.319*** (1.099)
BOO × GDP				−1.697*** (0.524)
HIN					−0.393*** (0.115)
HIN × GDP					0.045*** (0.013)
GDP	−0.088*** (0.024)	−0.193* (0.100)	−0.089*** (0.024)	−0.035*** (0.006)	−0.006** (0.003)
Lagged DV	−0.053*** (0.012)	−0.052*** (0.012)	−0.053*** (0.012)	−0.052*** (0.012)	−0.054*** (0.012)
Firm controls	Yes	Yes	Yes	Yes	Yes
Macro controls	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Observations	6,966	6,966	6,966	6,966	6,966
Firms	623	623	623	623	623
Instruments	133	133	133	133	133
AR(1) test	0.000	0.000	0.000	0.000	0.000
AR(2) test	0.858	0.856	0.838	0.930	0.806
Hansen test	0.134	0.233	0.226	0.153	0.122

Note. The sample consists of nonfinancial listed firms in Vietnam. The dependent variable (DV) is accrual-based earnings management. All explanatory variables are lagged by one period. Figures in parentheses are standard errors.

*

p < .1. **p < .05. ***p < .01.

Overall, our findings indicate that during periods of economic prosperity characterized by fewer financing constraints, the impact of bank competition on earnings management is diminished. Differently, we observe a heightened impact during times of crisis and economic downturn. Potentially, when financing constraints intensify in economic downturns or crises, competition that alleviates these constraints assumes a more substantial role.

Conclusion

Utilizing a dataset comprising Vietnamese commercial banks and listed firms from 2007 to 2023, our analysis documents a negative correlation between the degree of bank competition and corporate earnings management. Our further analysis offers evidence supporting financial constraints as an economic mechanism mediating the association between bank competition and firms’ earnings management. We demonstrate that decreased bank competition heightens firms’ financial constraints; in response, firm managers are likely to adopt earnings manipulation, primarily shown by upward adjustments.

Our analysis also reveals that the effect of bank competition on corporate earnings management is particularly more significant for smaller firms, non-state-owned enterprises, and those with lower market power. Moreover, firms with stronger bank-firm ties may attenuate the impact of bank competition on their financial information disclosure. Interestingly, our analysis also suggests that the impact of bank competition on corporate opacity tends to be more pronounced during unfavorable economic conditions.

This study offers significant policy implications for economies with a bank-centric financial system. We advocate for the promotion of bank competition as a means to mitigate financial constraints on businesses, curb earnings management practices, and enhance financial transparency in the stock market. Future research avenues could delve into specific policy measures, such as reducing barriers to entry and exit in the banking sector or fostering competition from non-bank entities, to bolster competition among banks. Our findings underscore the importance of policy frameworks facilitating firms’ access to financing, which is pivotal in improving financial reporting quality. Furthermore, this study provides valuable insights for investors, creditors, and other stakeholders interested in evaluating the reliability of firms’ financial disclosures. Given the potential challenges posed by monopolistic tendencies in the banking sector, particularly during adverse economic conditions, regulators should intensify their oversight of firms’ disclosure practices under such circumstances.

This study is subject to several limitations that should be acknowledged. First, our analysis relies on data from Vietnam, and differences in national contexts, including the degree of banking dependence, financial infrastructure, accounting standards, and the distinction between bank-centric and capital-market-oriented financial systems, may influence the prevalence and nature of earnings management, thereby limiting the generalizability of our findings. Second, this study employs national-level measures of bank competition, which are standard in the literature but provide limited variation in a single-country setting and may abstract from heterogeneity in regional credit conditions or firm-bank relationships. Future research using regional banking indicators or firm-bank matched data could introduce additional sources of variation and further strengthen empirical identification. Third, while we focus on financing constraints as a key channel through which bank competition affects corporate earnings management, we do not claim that this mechanism is exclusive. Other channels may operate concurrently but fall beyond the scope of the present analysis, and future research could explore these alternative pathways to provide a more comprehensive understanding of how bank competition shapes firms’ financial reporting behavior.