Sage Journals: Discover world-class research

Abstract

Earnings management (EM) is the practice of adjusting profits or earnings on financial statements at the discretion of the management. This study undertook a regression analysis of the earnings management model, with data from 300 non-financial listed companies on the Vietnamese stock exchange during the period 2016 to 2021 to determine whether the COVID-19 pandemic has had an impact on earnings management. To estimate earnings management, compared with the original model, we add the leverage variable (LV) was used as earnings management proxy in this research. The results show that the COVID-19 pandemic positively and significantly impact the earnings management of non-financial listed companies. Moreover, it was found that managers are more involved in real earnings management than accrual earnings management, implying that financially distressed firms need to reassure investors to raise more capital during the pandemic. However, there is no evidence of a trade-off between these two techniques during the pandemic compared to before the pandemic. Additionally, this study provides evidence that the model by Roychowdhury, following Cohen et al., and the performance-matched model by Jones, following Kothari et al. are more suitable for detecting EM behavior when factoring in the financial leverage (LV) variable for firms with a high debt ratio in developing countries.

JEL Classification: H8, M42

Plain language summary

Purpose. This study determines whether the COVID-19 pandemic has impacted earnings management (levels, trends, and techniques) using emerging markets datasets. Methods. The regression analysis of the earnings management (EM) model, with data from 300 non-financial listed companies on the Vietnamese stock exchange from 2016–2021. We use the model by performance-matched Jones, following Kothari et al. and the model by Roychowdhury, following Cohen et al. to measure the dependent variable is EM through the choice of method—optimal estimation between Pool OLS, FEM, and REM. The results of measuring EM with the above models are not statistically significant with our research data. Therefore, we have added leverage (LV) as an EM proxy. After adding the LV, the models estimate the EM dependent variable with statistical significance. Conclusions. The results show that the COVID-19 pandemic positively and significantly impact the EM. Moreover, it was found that managers are more involved in REM than AEM, implying that financially distressed firms need to reassure investors to raise more capital during the pandemic. However, there is no evidence of a trade-off between these two techniques during the pandemic compared to before the pandemic. Additionally, this study provides evidence that he original EM estimator model are more suitable for detecting EM behavior when factoring in the financial leverage (LV) variable for firms. Implications. In developing countries with large debt-bearing firms, leverage impacts earnings management. Limitations. The analysis of the income management behavior of enterprises by industry has yet to be conducted.

Keywords

earnings management financial reporting COVID-19 ROA

Introduction

Prolonged pandemics often cause both financial crises and economic downturns (Gormsen & Koijen, 2020), negatively affecting the production and business of most enterprises (Shen et al., 2020). Within such contexts, managers often practice earnings management (EM) to maintain earnings goals (Matsumoto, 2002), gain rewards and compensations (Bergstresser & Philippon, 2006), and receive financial support through diminishing accrual earnings management (H. DeAngelo et al., 1994) or inflating future earnings after a crisis by adopting methods such as the “big bath” strategy during the crisis (Cohen et al., 2008; Kirschenheiter & Melumad, 2002). There are at least two different perspectives commonly used to explain the tendency to perform EM in the direction of increasing or decreasing profits (Habib et al., 2022). Such as the motivations that lead to lower EM, lower investor return expectations, and higher legal risk to the business (Cimini, 2015). Several studies provide empirical evidence that accrual earning management (AEM) decreased during the 2008 financial crisis (Cimini, 2015; Ebrahimi et al., 2017; Trombetta & Imperatore, 2014) due to increased management and control in order to protect investor interests (Smales, 2021) or less EM in periods of economic distress (Ali et al., 2022; Arthur et al., 2015; Filip & Raffournier, 2014). There is also a perception that crisis periods provide greater opportunities to engage in EM, as investor tolerance increases during these periods (Ahmad-Zaluki et al., 2011).

Since the beginning of COVID-19 in December 2019, several studies across countries have attempted to understand the impact of the pandemic on corporate governance behavior, such as

Ryu and Chae (2022) in Korea, Ljubisavljević and Jakobsson (2022) in Sweden, Klomp (2022) across nine European countries, Kazemi (2022) and Lassoued and Khanchel (2021) in European countries, Yan et al.(2022) in China, Ali et al. (2022) in G-12 countries, Liu and Sun (2022) in the US, Garfatta et al. (2023) in Tunisia, and da Silva Flores et al. (2023) in Brazil. The results of empirical studies show that the COVID-19 pandemic impacts the EM level but the trend is inconsistent between studies. On the other hand, the above studies are yet to explain the differences in managers’ choice of EM strategies in the context of the pandemic.

Vietnam has been among some of the world’s fastest-growing and most dynamic emerging economies in recent decades (World Bank, 2023). Vietnam’s stock market consists of two exchanges, Ho Chi Minh City (HOSE) and Hanoi (HNX), with about 747 listed companies in 2021, including approximately 348 non-financial listed companies. COVID-19 has had substantial negative effects on the performance of stock markets worldwide, and “the VN-index declined dramatically in the first 3 months of the year from 31 December 2019 to 30 March 2020” (Hung et al., 2021, p. 1). After imposing social distancing, the government successfully controlled the pandemic, and the stock market showed signs of recovery, becoming one of the four best-performing stock markets in the world (Hung et al., 2021). Nevertheless, from May 2021, the pandemic seriously affected Vietnam.

This article examines changes in the EM of non-financial listed companies in Vietnam before and during the pandemic. First, through an overview of previous studies, we confirm that the COVID-19 pandemic affected the level of EM. However, trends (increasing or decreasing) and EM techniques (accrual or real) are inconsistent across studies. These studies mainly used data from developed countries. Our study is one of the few examining the impact of COVID-19 on EM levels, trends, and techniques, using emerging markets datasets. We contribute to existing literature by providing country-specific results.

Second, a number of studies explore the EM strategies of managers by testing both AEM and REM models. However, there has been little research on the choice of EM strategy in the context of the pandemic to explain the influence of financial shocks on EM behavior. Therefore, we add to the literature on the possible trade-off between AEM and REM in the context of the pandemic to highlight policy implications for the government.

Third, although the modified Jones model by P. Dechow and Dichev (2002) and the performance-matched model by Jones (1991), following Kothari et al. (2005) and Roychowdhury (2006), have been confirmed by many researchers to be the most suitable EM estimation models, these models have some defects that reduce reliability (Stubben, 2010). Therefore, before determining the level of EM for non-financial listed companies in Vietnam in the sample, we tested the suitability of the selection model. The research results are intended to supplement the test evidence of the data fit for the estimation models. The addition of the financial leverage (LV) factor to the above models, although not directly referring to the sample data, indicate that the above estimation model ensures reliability (Supplemental Appendix 1 and 2). We opted for LV to add to the estimation model because Vietnamese enterprises mainly depend on bank loans to reduce debt and credit costs or to expand investment. Therefore, they must find ways to maintain a healthy financial picture. According to signal theory, managers can oversee earnings to ensure this goal.

This paper is structured as follows: Section “Literature Review and Hypothesis Development” provides background definitions, theories, and motivations for earnings management; Section “Methodology” presents the research model, hypothesis, and data; and Section “Research Results and Discussion” provides research findings and discussions, followed by conclusions and policy implications.

Literature Review and Hypothesis Development

Literature Review

EM is a dominant term used in research to describe the motives and causes of EM by managers. Although there are multiple definitions of EM, it generally places the financial picture of a business within real contexts (Healy & Wahlen, 1999). EM has become the most discussed topic in accounting research. According to Dokas et al. (2021), general EM studies include: (a) the EM estimation model; (b) the motivation of EM; and (c) the factors affecting the behavior of EM.

There are two standard techniques of EM: (a) Accruals earnings management (AEM), through the selection of estimates and accounting policies to adjust current period earnings to the will of management, and (b) REM, by changing business strategies and decisions, such as inflating revenue, overproduction, or adjusting discretionary costs (Chen et al., 2020; Roychowdhury, 2006). According to Habib et al. (2022, p. 4280), “Researchers initially investigated the determinants of AEM following the seminal paper of Jones (1991). However, after the passage of the Sarbanes Oxley Act (SOX) of 2002, research on the determinants and consequences of REM proliferated.”

For AEM, Healy (1985) defines accruals as the difference between reported earnings and cash flows from operating activities. This study compares the effect of predicting a reward plan based on business plan data and actual results. The research results show that the reward mechanism and business plan affect the level of EM. L. E. DeAngelo (1986) investigated the EM behavior of managers when they intend to convert a public-traded firm to a private enterprise. This research shows that the manager has adjusted earnings downward to reduce the buyout compensation. M. McNichols and Wilson (1988) proved that firms use income-decreasing accruals when earnings are extremely high or low. Meanwhile, the research results of Jones (1991) showed that the policy of supporting imports stimulates managers to implement EM in the direction of reducing income.

REM has been defined as “departures from normal operational practices, motivated by managers’ desire to mislead at least some stakeholders into believing certain financial reporting goals have been met in the normal course of operations” (Roychowdhury, 2006, p. 337). According to Graham et al. (2005), managers prefer to use REM rather than AEM because REM is difficult to detect, especially in the post-SOX period (Cohen et al., 2008).

At least two different perspectives have been commonly used to explain the tendency to perform EM (Habib et al., 2022). On a positive note, “the efficiency view, which is informed by signaling theory, suggests that managers engage in EM to signal private information to capital market participants” (Habib et al., 2022, p. 4282). EM can reduce information asymmetry in the capital market, attract new investors, and strengthen corporate confidence (Jiraporn et al., 2008). REM is more expensive than AEM, but managers choose REM in cases where the positive effect is greater from information transmitted to capital markets (Zhao et al., 2012). From the efficiency perspective, Habib et al. (2022, p. 4282) argue that “firms that have a potential to show better future performance use REM to signal such potential to the capital markets.”Gunny (2010) claimed that managers engage in REM tactics to meet benchmarks (e.g., the prior year’s earnings, zero profits) in the attempt to influence the stakeholders’ predictions for the future performance.

The agency theory-based opportunistic view posits that managers use EM to provide misleading information to achieve financial reporting goals to gain private benefits (Roychowdhury, 2006). This could include the use of opportunistic activities that increase information risk, such as cutting research and development costs, advertising costs, and training costs to maintain profitability at current levels; influencing the company’s sales policy to increase profits by increasing the credit period for customers, reducing the listed selling price if buying in large quantities; or overproduction. These activities mask the real economic performance of the business. Accordingly, EM creates information asymmetries and triggers inter-agency conflicts in the form of adversarial selection and issues of moral hazard (Habib et al., 2022).

Managerial incentives can be categorized into three groups: contractual, capital market, and third-party (Walker, 2013). Dokas et al. (2020), in a review of studies on EM up to 2020, identified five groups of EM motivators: (a) stock market incentives, (b) revealing private information, (c) regulations and political costs, (d) private benefits, and (e) lending contracts.

According to Ljubisavljević and Jakobsson (2022), the first and most common managerial incentive is to meet market expectations and provide better information to stakeholders, mainly shareholders (Kothari et al., 2016). The second reason is to maintain business performance to gain rewards. Moreover, depending on the circumstances and the manager’s motivation, earnings can be managed upward (income-increasing), downward (income-decreasing), or not managed at all (Liu & Sun, 2022). This motivation comes from fulfilling commitments to raising capital from credit institutions or investors to avoid breach of agreements (Clikeman, 2003; Healy & Wahlen, 1999). Although there are differences of opinion, most researchers generally agree that EM has more negative consequences because of distorted financial information provided to shareholders. Consequently, the financial information provided is less helpful in analyzing and forecasting the company’s profitability and future growth trends (Bernstein & Siegel, 1979).

Several recent studies have recommended choosing EM techniques based on their cost comparisons (Cohen & Zarowin, 2010; A. Y. Zang, 2012). Managers often use both AEM and REM techniques to avoid stakeholder detection and produce more convincing results (Dang & Nguyen, 2021; Fields et al., 2001). A. Y. Zang (2012) proved that, after implementing REM, a manager may implement AEM. The extent to which these two techniques are used depends on the cost of implementing EM. In situations with increased external control and supervision pressure or reaction to market signals (market incentives), REM tends to overwhelm AEM (A. Zang, 2006; Zhao et al., 2012). These statements have been verified by studies such as Cohen et al. (2008), Cohen and Zarowin (2010), and A. Y. Zang (2012). Few studies have shown that managers perform REM in the context of a disaster or a pandemic (e.g., Klomp, 2022; Xiao & Xi, 2021; Yan et al., 2022). However, these studies did not examine how the EM strategy played out during the pandemic.

Hypothesis Development

There is a correlation between a crisis and EM behavior. Financial crises often harm a company’s financial position, such as lowering business performance, causing a drop in stock prices, or increasing input prices due to scarcity as the crisis lingers. In these rare cases, managers often consider an income management strategy based on the severity of the crisis and the resilience of the business.

Empirical studies about the financial crisis and EM in firms have been conducted by researchers the world across, with several studies indicating that the level of EM declines during crises (Cimini, 2015; Filip & Raffournier, 2014; Kjærland et al., 2020). For instance, studies have found that managers are less likely to engage in EM because of the market and the possibility of increased litigation risk (Filip & Raffournier, 2014); the level of supervision by investors through auditors, creditors, and other stakeholders will be higher (Kumar & Vij, 2017); and there will be increased efforts to strengthen investor protection (Cohen et al., 2008; Persakis & Iatridis, 2016; Smales, 2021). As a result, income adjustment will become more difficult and complex (Filip & Raffournier, 2014; Kumar & Vij, 2017). This statement is also supported by empirical data from nine European countries and shows that countries with higher levels of investor protection have lower levels of EM because managers are afraid of detection and punishment (Klomp, 2022).

At less severity, management needs to reassure investors about the company’s growth prospects or how to attract potential investors. Then, the EM strategy is applied to inflate profits until the economy recovers (Graham et al., 2005). Popular earnings adjustment techniques include changes in inventory accounting methods, bad debt estimates, assumptions regarding retirement assets and credit lines, and changes in the volume of products produced during the year. In the case of China, Chen et al. (2020) found that financially distressed firms apply EM techniques to avoid government scrutiny. In a severe financial crisis, the level of EM during the crisis is lower, which often leads to increased use of income-decreasing EM since managers want to blame the objective cause (Kjærland et al., 2020; Kousenidis et al., 2013).

Accordingly, firms with weak investor protection are often more inclined to use AEM because they are less legally bound (Leuz et al., 2003). Several other studies have provided evidence that AEM increased during a financial crisis (e.g., Bornemann et al., 2012; De Luca & Paolone, 2019 in Europe; Ahmad-Zaluki et al., 2011 and Masruki & Azizan, 2012 in Malaysia). The evidence from the Asian financial crisis (1997–1998) showed a higher level of income-increasing earnings management during the crisis years (Ahmad-Zaluki et al., 2011). Some studies suggest that income management practices toward reducing income may be more common in times of crisis (Kjærland et al., 2020; Ming Chia et al., 2007). Managers may be inclined to use earnings reduction techniques because poor results are inevitable, allowing managers to take advantage of times of crisis by reporting larger losses than necessary (Kjærland et al., 2020).

Ljubisavljević and Jakobsson (2022, p. 15) state that, “during macroeconomic difficulties, the typical market-based incentives that would encourage earnings management techniques to increase earnings, may instead drive managers to alter the reported financial information by reporting larger losses in current periods, and consequently allowing for reporting an enhanced financial picture to shareholders in future periods that surpasses their expectations.”

Table 1 shows a summary of eleven studies published from 2021 to 2023 on the impact of COVID-19 on EM in European, US, and Asian countries; nine out of the eleven findings indicate that the level of EM increased during the pandemic. There is a split in EM theory in times of crisis, with compelling arguments on both sides. From the above perspective, the first hypothesis is as follows:

Table 1.

Results of Studies on the Impact of COVID-19 on Earnings Management.

Study	Survey country	Earnings management (EM) during COVID-19
Study	Survey country	Level of EM	AEM trend	REM trend	EM strategies
Ljubisavljević and Jakobsson (2022)	Sweden	+	−	N	Accrual
Klomp (2022)	Nine European countries	+	−	+	Real
Xiao and Xi (2021)	China	+	+	−	Accrual
Liu and Sun (2022)	United State	−	−	NC
Yan et al. (2022)	China	+	+	+	Accrual
Kazemi (2022)	European Union	+	+
Lassoued and Khanchel (2021)	Fifteen European countries	+	+	NC
Garfatta et al. (2023)	Tunisia	+	−	−	Accrual
Ali et al. (2022)	G-12 countries	−	−	NC
Ryu and Chae (2022)	Korea	+	+	NC
da Silva Flores et al. (2023)	Brazil	+	+	N

Note.“+” increasing; “−” decreasing; “N” insignificant; “NC” non-check.

H₁: During the COVID-19 outbreak, the level of earnings management of the listed companies in Vietnam increased.

According to Kim et al. (2018), managers can manipulate earnings reported through AEM and REM. While AEM occurs before the release of financial statements to inflate (or mask) reported earnings, REM occurs throughout the financial year, since it requires changing a company’s operations (A. Y. Zang, 2012). Therefore, REM implementation costs more than AEM due to the adjustment of business strategies but is more difficult to detect (Kim et al., 2018). The research results of Ali et al. (2022) and Xiao and Xi (2021) show that AEM rather than REM techniques were more commonly used during the pandemic. Since REM focuses on manipulating cash flow through operating, investing, and financing activities throughout the financial year, it is more difficult to implement compared to AEM. Klomp (2022, p. 1) investigated nine European countries and stated that “firms with strong investor protection turn to more real earnings management during COVID-19 supportive of the second hypothesis.”Table 1 shows five of eleven studies examining both earnings management strategies; four of five studies provide evidence that managers prefer to use AEM. In the first year of the pandemic, Vietnam controlled the effects of the pandemic very well. However, in the second year, that is, 2021, Vietnam suffered a severe shock—the social distancing was prolonged, the economy stalled, and GDP dropped. Moreover, most of the non-financial listed companies in Vietnam are small and have high debt ratios. Do these challenges pressure managers to adjust earnings upward in order to provide a positive financial picture of the business according to market signal theory? Alternately, does it compel managers to implement income management in the direction of reduction based on agency theory to achieve personal benefits? Therefore, the second hypothesis is as follows:

H₂: Listed companies performed more accrual earnings management than real earnings management during the COVID-19 pandemic compared to before the pandemic.

Methodology

The Research Model

The dependent variables in the AEM model are measured on the basis of two models: the modified Jones model by P. Dechow and Dichev (2002), following M. F. McNichols (2002), and performance-matched Jones model (2005), following Kothari et al. (2005) and Cohen et al. (2020). For the dependent variable of REM, the author uses the proxies of Roychowdhury (2006).

The independent variable, COVID-19, is a dummy variable equal to 0 for observations before COVID-19 and 1 for observations during COVID-19. Control variables in the model inherited from previous studies to control the difference between enterprises include the following: (a) Following Burgstahler et al. (2006), who use firm growth (Growth) to control for growth opportunities; (b) Including ROA to control for differences in firm performance affecting the level of AEM and REM (Cohen et al., 2020; Kothari et al., 2005); (c) Following previous studies (Cohen et al., 2008) to control for risk by using leverage (LV) measured as the percentage of long-term liabilities to total assets; (d) Controlling for the possible impact of a size (Size) by including the natural logarithm of the firm’s total assets in the model; (d) Controlling for audit quality (Big4), captured by a dummy variable that equals 1 if the firm’s auditor is one of the Big 4 audit firms (i.e., PricewaterhouseCoopers, Deloitte Touche Tohmatsu, Ernst & Young, and KPMG), and 0 otherwise; and (e) Controlling the equation including a dummy variable controlling for Net Income (NEIC) (Ljubisavljević & Jakobsson, 2022)

As multiple linear regressions are more sensitive to the normality assumption, the continuous variables are subject to a 1st and 99th percentile winsorizing prior to running the regression to ensure that no outliers drive the results.

$\begin{matrix} AEM_D D_{it} = b_{0} + b_{1} COVID - 1 9_{it} + b_{2} RO A_{it} \\ + b_{3} GROWT H_{it} + b_{4} L V_{it} + b_{5} SIZ E_{it} + b_{6} Au D_{it} \\ + b_{7} NETI C_{it} + e_{it .} \end{matrix}$ (1)

$\begin{matrix} AEM_P M_{it} = b_{0} + b_{1} COVID - 1 9_{it} + b_{2} RO A_{it} \\ + b_{3} GROWT H_{it} + b_{4} L V_{it} + b_{5} SIZ E_{it} + b_{6} Au D_{it} \\ + b_{7} NETI C_{it} + e_{it .} \end{matrix}$ (2)

$\begin{matrix} RE M_{it} = b_{0} + b_{1} COVID - 1 9_{it} + b_{2} RO A_{it} \\ + b_{3} GROWT H_{it} + b_{4} L V_{it} + b_{5} SIZ E_{it} + b_{6} Au D_{it} \\ + b_{7} NETI C_{it} + e_{it .} \end{matrix}$ (3)

In which i = 1, 2, 3,…,300 (representing 300 listed companies) and t = 1, 2, 3,… (6 year period from 2016 to 2021). Here, AEM_it is equal to either the net level of AEM or the absolute level of AEM in year t (AAEM). They represent two models of accrual earnings management: (a) AEM_DD for Modified Jones model of P. Dechow and Dichev (2002) and (b) AEM_PM for Performance-matched Jones (1991).

Here, REM_it is equal to either the net level of REM or the absolute level of REM in year t. They represent the total REM level in year t, where the three proxies are combined into an aggregate score following Cohen et al. (2008).

The COVID-19 variable in multiple linear regressions is a dummy variable used to test the sign of the dependent variable during and before the pandemic. If the coefficient of the test variable is positive, it can be interpreted as an increase in the level of EM or income managed, and vice-versa (Kazemi, 2022). The absolute value of AEM or REM could be in both directions, that is, income-increasing earnings management and income-decreasing earnings management. The sample is divided into two sub-samples based on the sign of the firm’s EM (Lassoued & Khanchel, 2021).

Earnings Management Measures

This study uses AEM and REM to determine whether the pandemic impacts EM.

Accruals Earnings Management

Many models have been published and applied to detect the AEM behavior of managers. Jones’s (1991) model has been an advanced, effective, and widely used model in detecting EM. The models by Cohen et al. (2020), P. M. Dechow et al. (1995), P. Dechow and Dichev (2002), and Kothari et al. (2005) are considered improved models of Jones (1991). In this research, to test EM, the AEM technique was applied to identify discretionary accruals (DA) on financial statements by P. Dechow and Dichev (2002) and Kothari et al. (2005). DA is the difference between the total accruals (TA) and non-discretionary accruals (NDA) (Table 2).

Table 2.

Accruals and Real Earnings Management Model.

Accruals earnings management (AEM)	Real earnings management (REM)
Modified Jones model of P. Dechow and Dichev (2002) following M. F. McNichols (2002), hereafter AEM_DD	Roychowdhury (2006) model add ROA and LV, hereafter REM
$\frac{{ACCA}_{i t}}{A_{i t - 1}} = \propto \frac{1}{A_{i t - 1}} + β_{1} CFO i t - 1 + β_{2} CFO i t + β_{3} CFO i t + 1 + β_{4} \frac{Δ {SALES}_{i t}}{A_{i t - 1}} + β_{5} \frac{{PPE}_{i t}}{A_{i t - 1}} + ε_{it} .$	Determine abnormal levels (ε_it) of operating cash flow $\frac{CF O_{it}}{A_{it - 1}} = \propto \frac{1}{A_{it - 1}} + β_{1} \frac{SALE S_{it}}{A_{it - 1}} + β_{2} \frac{Δ SALE S_{it}}{A_{it - 1}} + β_{3} RO A_{it} + β_{4} L V_{it} + ε_{it}$
Performance-matched Jones (1991) following Kothari et al. (2005) and Cohen et al. (2020), hereafter AEM_PM add ROA and LV, hereafter AEM_PM	Determine the industry specif normal levels (ε_it) of discretionary expenses (DISC), accordingly: $\frac{Dis c_{it}}{A_{it - 1}} = \propto \frac{1}{A_{it - 1}} + β_{1} \frac{SALE S_{it}}{A_{it - 1}} + ε_{it}$
$\frac{{ACCA}_{i t}}{A_{i t - 1}} = \propto \frac{1}{A_{i t - 1}} + β_{1} \frac{Δ {REV}_{i t} - Δ {REC}_{i t}}{A_{i t - 1}} + β_{2} \frac{{PPE}_{i t}}{A_{i t - 1}} + β_{3} {ROA}_{i t} + β_{4} {LV}_{i t} + ε_{i t}$

Note. Calculate the accrual sum (ACCR) of each enterprise in each year according to the following formula: ACCR_it = NIC_it-CFO_it.

ACCR = total accruals in year t; NIC: net income in year t; CFO = cash flow from operations in year t; A_it-1 = total assets in year t-1; ΔREV_it = difference in revenue between year t and year t-1; PPE_it = gross value of property, plant, and equipment in year t; ROA_it = net income in year t divided by total assets in year t; ΔREC_it = difference in accounts receivable between year t and year t-1; Sales_it = net sales of the company i in year t. The data is collected from the business report of the enterprise; ΔSales_it = change in company i’s net sales in year t compared to year t-1; A_it-1 = total assets at the beginning of the year, collected from the balance sheet of the business; DISC_it = calculated from the enterprise’s total selling and administrative expenses, collected from each enterprise’s income statement in each observation year. PROD = determined by the cost of goods sold plus the change in net inventory at the end of the period compared to the beginning of the period. The costs of goods sold are collected from the income statement of the enterprise. Net inventories are collected from the balance sheet.

Real Earnings Management

This study estimates REM using the model proposed by Roychowdhury (2006) with three representative variables: cash flow from operations (CFO), production costs (PROD), and discretionary expenses (DISC). This research model added the ROA variable according to Cohen et al. (2020) as an independent variable to match the firm’s performance as well as ensure consistency in comparison to Klomp’s results (2022). In addition, we add a financial leverage (LV) variable in the estimation model (Table 2).

Research Data

Among 748 enterprises listed on stock exchanges in Vietnam as of 2021, 348 are non-financial. However, more data is required for 48 non-financial listed companies over a 2 year period or more during 2016 to 2022. Therefore, our research used 300 non-financial listed companies on the Ho Chi Minh City Stock Exchange (HSX) and the Hanoi Exchange (HNX) from 2016 to 2021. Financial data has been derived from financial statements, with 1,800 observations that were used for analysis. The research data is divided into pre-pandemic (2016 to 2019: 1,200 observations) and during the pandemic (2020 to 2021: 600 observations).

Data Analysis Methods

Applying the least squares regression (OLS) method, the author conducted a model selection test and model defect test for the three models (AEM_DD, AEM_PM, REM). There is no statistical significance after performing multivariate regression estimation and defect correction using strong standard OLS (Robust Standard Errors). We added the ROA variable, as suggested by Cohen et al. (2020) and added an LV variable to ensure that it is consistent with market data. Vietnamese enterprises are mainly small- and medium-sized enterprises that use loans, and so, the LV index is relatively high. When the measurement model is suitable, the dependent variables ensure reliability (statistically significant). We conducted a regression analysis of the impact of the pandemic on the dependent variable (AEM_DD, AEM_PM, and REM) based on the choice of method—optimal estimation between Pool OLS, FEM, and REM. All of the results of the above estimates encounter heterogeneity of variance. Accordingly, FGLS is applied to overcome the variable variance and ensure the selection of an efficient estimation model between FEM and REM (Supplemental Appendix 1 and 2).

Research Results and Discussion

Research Results

Descriptive Statistics and Correlations

Table 3 presents descriptive statistics of the research sample on AEM and REM for the first period (2016–2019), and the pandemic period (2020–2021). Average accrual management in the modified Dechow and Dichev model (AEM_DD) and the performance-matched Jones model (AEM_PM) is positive, and the mean of REM is negative (before the pandemic). Interestingly, all models’ level of EM was lower than before the pandemic, except for AEM_PM. According to the above data, managers’ use of income-decreasing earnings management increased during the pandemic (average AEM_PM changes from positive to negative). At the descriptive level, firms use arbitrary accrual and real decision changes for rising and falling earnings.

Table 3.

Descriptive Statistics for Earnings Management.

Description	Obs	Mean	Std. Dev.	Min	Max
Data in 2016–2019
AEM_DD	1,200	0.056	0.077	(0.108)	0.339
AAEM_DD	1,200	0.064	0.068	0.000	0.339
AEM_PM	1,200	0.036	0.186	–0.663	0.752
AAEM_PM	1,200	0.127	0.148	0.001	0.835
REM	1,200	(0.254)	0.317	(1.202)	0.767
AREM	1,200	0.313	0.265	0.003	1.250
Data in 2020–2021
AEM_DD	600	0.038	0.063	(0.108)	0.339
AAEM_DD	600	0.052	0.054	0.000	0.339
AEM_PM	600	(0.090)	0.219	(0.663)	0.752
AAEM_PM	600	0.172	0.175	0.001	0.835
REM	600	(0.185)	0.308	(1.202)	0.767
AREM	600	0.273	0.253	0.003	1.250

As per Table 3, for the AEM_ DD model, both the pre-pandemic period and the pandemic year have a positive mean value for the magnitude of DA. This indicates that using income-increasing accrual-based earnings management practices are most common throughout the sample. The REM has a negative mean value for the amount of REM. This data indicates that using income-decreasing REM practices are most common throughout the sample. Meanwhile, the mean for the AEM_PM model with positive pre-pandemic but negative post-pandemic values indicates two opposite income adjustment trends, that is, adjusting for the direction of income increases before the pandemic and vice-versa.

The other primary variable of interest is the absolute value of discretionary accruals (AAEM) and the absolute value of real earnings management (AREM). We use the absolute value to study potential changes in the magnitude of EM. The research results show that AAEM_DD and AREM for the before-pandemic period have greater magnitude than the pandemic period sample, contrary to the model for AAEM_PM.

To examine if there is a significant change in the amount of EM in the sample between the pre-pandemic period and the pandemic period, we use two-sample t-tests. Table 4 presents the results of the t-test.

Table 4.

Two-sample t-tests for Earnings Management.

Period	AEM		REM	Absolute AEM (AAEM)		Absolute REM (AREM)
Period	Modified DD	Performance match PM	REM	Modified DD	Performance match PM	Absolute REM (AREM)
2016–2019	0.057	0.036	(0.254)	0.064	0.127	0.313
2020–2021	0.038	(0.090)	(0.185)	0.051	0.172	0.273
Difference	0.019***	0.127***	(0.070)***	0.013***	(0.045)***	0.040***

Note. t statistics in parentheses.

***

p < .01.

There is a statistically significant difference between the observations for the three EM models before and during the pandemic. We find that the magnitude of the AEM_ DD variable decreases from 0.057 pre-pandemic to 0.038 during the pandemic, and the decrement of 0.019 is significant at the 1% level (t = 5.39). The level of REM decreases from −0.254 in the pre-pandemic to −0.185 in the pandemic, and the decrement of 0.070 is significant at the 1% level (t = −4.4). However, the mean difference of −0.127 corresponds to a significance at the 1% level (t = 12.8) for AEM_PM variable change from 0.036 pre-pandemic to −0.090 during the pandemic.

Furthermore, the purpose of the study is to capture both income-increasing and income-decreasing uses of EM. If both the mean difference of the test and its absolute are positive, the mean of the second group of observations (i.e., the pandemic) is lower than that of the first group, and vice-versa. Table 4 shows two possibilities: increasing income-decreasing earnings management or decreasing the level of income management in the second observation group (AEM_DD), and vice-versa (REM). However, AAEM_PM is negative. There are two possibilities: increasing income-decreasing earnings management or increasing the level of earnings management in the second observation group.

Despite our ambiguous univariate results, the analysis in Table 4 corroborates our earlier evidence and the predictions in H₁ and H₂ on the impact of the pandemic on accruals and real earnings management.

Table 5 shows that the correlation coefficients between the discretionary accrual measures (AAEM_ DD and AAEM_PM) and AREM measures are positive and statistically significant at the 1% confidence level. This high positive correlation can be explained by firms simultaneously engaging in accruals and real earnings management. This is consistent with prior research (i.e., Alissa et al., 2013; Cohen & Zarowin, 2010; Lin & Shen, 2015), indicating that managers can manipulate earnings to achieve the target by using both REM and AEM, since REM occurs during the fiscal year while AEM occurs after the end of the accounting period but within the confines of a generally accepted accounting system. The correlation values of COVID-19 with AAEM_ DD, AAEM_PM, and AREM are (−0.094***), 0.135***, and (−0.070***), respectively.

Table 5.

Pearson’s Correlation Coefficient.

Pearson correlation	AAEM_DD	AAEM_PM	AREM	COVID_19	GROWTH	LV	ROA	SIZE	AuD	NETIC
AAEM_DD	1.000
AAEM_PM	.140***	1.000
AREM	.524***	.285***	1.000
COVID_19	–.094***	.135***	–.072***	1.000
GROWTH	.106***	–.132***	–.175***	–.400***	1.000
LV	–.221***	.153***	–.098***	–.003	.137	1.000
ROA	.859***	.073***	.544***	–.197***	.298***	–.164***	1.000
SIZE	–.068***	.173***	.185***	.039*	.066***	.189***	.000	1.000
AuD	.123***	–.159***	.108**	–.437***	.396***	–.013	.190***	.260***	1.000
NETIC	.029	–.023	.161***	–.135***	.323***	.027	.380***	.088***	.109***	1.000

Note. t-statistics in parentheses.

p < .1. **p < .05. ***p < .01.

Multivariate Analyses

Table 6 presents the estimation results for Equations (1) to (3). Our hypotheses (H₁ and H₂) are tested by looking at the b₁ coefficient, which captures the impact of the pandemic on EM. The dependent variable is the regression residual from two AEM models (AEM_ DD and AEM_PM variables) and REM models (CFO, PROD, and DISC variables).

Table 6.

The Pandemic as Determinant of Accruals and Real Earnings Management (FGLS).

	Panel A—AEM		REM	Panel B—Absolute AEM		Absolute REM
Variables	Modified DD	Performance match-PM	REM	Modified DD	Performance match-PM	Absolute REM
1	2	3	4	5	6	7
COVID_19	0.003***	–0.040***	–0.020**	0.004***	0.026***	0.027***
	(5.61)	(–7.48)	(–2.54)	(6.59)	(5.19)	(3.68)
GROWTH	–0.008***	0.083***	0.000	–0.006***	–0.034***	0.006
	(–18.45)	(19.26)	(0.07)	(–12.84)	(–8.72)	(1.16)
LV	–0.007***	0.092***	0.190***	–0.005***	0.105***	–0.004
	(–8.43)	(10.26)	(14.19)	(–4.76)	(12.65)	(–0.39)
ROA	0.812***	0.404***	–1.602***	0.769***	0.321***	1.660***
	(267.05)	(12.62)	(–30.53)	(196.69)	(10.58)	(36.21)
AuD	0.001***	0.013***	0.027***	–0.001**	–0.035***	–0.072***
	(2.80)	(3.96)	(5.23)	(–2.53)	(–11.74)	(–15.09)
NETIC	–0.001***	0.008**	–0.031***	0.000	–0.005	0.027***
	(–2.76)	(1.99)	(–5.13)	(0.84)	(–1.25)	(4.71)
SIZE	0.013***	–0.036***	0.011	–0.069***	–0.018**	–0.034***
	(13.16)	(–3.88)	(0.84)	(–29.15)	(–2.07)	(–2.76)
_cons	–0.026***	–0.197***	–0.470***	0.0795***	0.458***	0.978***
	(–6.99)	(–5.31)	(–7.98)	(16.09)	(13.62)	(17.74)
N	1,794	1,797	1,800	1,794	1,797	1,800

Note. All three regression models evaluate the fit between Pool OLS, FEM, and REM. The FEM model is the most suitable, but there is a phenomenon of heteroskedasticity or autocorrelation, which the study overcame by using FGLS regression. t-statistics in parentheses.

p < .1. **p < .05. ***p < .01.

Panel A of Table 6 indicates that, in the AEM_DD model, the coefficient b₁ (pandemic) is slightly positive (0.003) and significant (p < .001), as shown in column (2). Both AEM_PM and REM models have a significant negative coefficient (−0.04 and −0.02, respectively) and are highly significant (p < .001,05, respectively), as shown in columns (3) and (4). There is a statistically significant difference in EM before and during the pandemic.

Next, we ran a sensitivity analysis considering the absolute value of EM as our dependent variable. Results are reported in Panel B of Table 6. As mentioned above, we use the absolute value, the alternative dependent variable, to measure the magnitude of EM rather than the signed. The coefficient b₁ (pandemic) is significant when discretionary accruals (AEM_DD and AEM_PM) and REM are measured for samples. This means that firms exhibit evidence of EM after the COVID-19 pandemic. The coefficient of COVID-19, as shown in columns (5) to (7), is positive and significant in the sample firms, indicating that the level of the absolute value of EM in 2020 and 2021 (Crisis Period) had increased (Kazemi, 2022).

The above conclusions demonstrate that the COVID-19 shock intensified EM behavior, and managers use both accruals and real earnings management to manipulate earnings.

This finding is in line with other recent studies, such as Xiao and Xi (2021) and Yan et al. (2022) studying Chinese firms; Kazemi (2022), Klomp (2022), Lassoued and Khanchel (2021), and Ljubisavljević and Jakobsson (2022) studying European firms, Ryu and Chae (2022) in Korea, and da Silva Flores et al. (2023) studying Brazil firms on the impact of COVID-19 on the level of EM. Scholars found that the level of EM has increased during the pandemic. However, this finding is inconsistent with the findings of other researchers, such as Ali et al. (2022), who studied G-12 firms; and Liu and Sun (2022), who studied US firms; and, prior to the pandemic, studies such as Filip and Raffournier (2014), who studied European firms; and Cimini (2015), all of which found decreasing levels of EM during crises.

Furthermore, Table 7 shows the use of income-increasing earnings management, and the coefficient values of 0.006, 0.032, and 0.140 of the independent variables of the three models include positive AEM_DD, positive AEM_PM, and positive REM, respectively, suggesting that the pandemic affects the magnitude of accrual and real earnings management as found in the financial statements of the firms in the study sample (da Silva Flores et al., 2023; Kazemi, 2022; Lassoued & Khanchel, 2021; Ryu & Chae, 2022; Xiao & Xi 2021; Yan et al., 2022). Therefore, the data supports hypothesis H₁.

Table 7.

Accruals and Real Earnings Management.

	Positive AEM		Positive REM	Negative AEM		Negative REM
Variables	Modified DD	Performance match		Modified DD	Performance match
1	2	3	4	5	6	7
COVID_19	0.006***	0.032**	0.140***	–0.002	–0.042***	–0.029**
	(4.81)	(2.13)	(3.90)	(–0.76)	(–3.90)	(–2.17)
_cons	0.006	0.657***	1.548***	–0.081***	–0.584***	–1.017***
	(0.57)	(6.64)	(6.59)	(–3.08)	(–8.29)	(–9.68)
N	1,476	881	307	318	916	1,493
R-sq	0.938	0.203	0.316	0.732	0.153	0.374

Note. Robust t-statistics in parentheses.

p < .05. ***p < .01.

This contradicts studies that observed a decrease in AEM during financial crises (Cimini, 2015; Filip & Raffournier, 2014) but supports studies that observed an increase in accrual earnings-decreasing management during financial crises, and vice-versa (Habib et al., 2013; Persakis & Iatridis, 2016).

Further, Panel B of Table 6 shows that most control variables have a statistically significant correlation with EM models, except for the NETIC variable for the two accrual-based income management models (columns 5, 6) together with the GROWTH and LV variables for REM (column 7). For instance, the negative and significant estimated coefficient for SIZE shows that smaller firms managed higher earnings than larger firms during the pandemic. This result is consistent with the findings of Lassoued and Khanchel (2021) and Liu and Sun (2022). The other two control variables have positive relationships with the EM proxy. The leverage (LV) coefficient of 0.105 is positive and significant, which shows that firms with higher levels of debt-to-asset ratio are more likely to perform income-increasing EM during the pandemic. The positive sign of ROA is in line with the findings of Kothari et al. (2005), which is interpreted as a higher level of income-increasing earnings management for highly profitable firms.

The study conducted a regression for the dependent variables, income-increasing earnings management and income-decreasing earnings management variables, to test the EM technique during the pandemic (H₂ hypothesis). The statistical significance of the beta coefficient (b₁) of the regression equations indicates whether there is a corresponding EM technique. Comparing the coefficient b₁ of the accrual regression Equations (1) and (2) with the real regression Equation (3) shows the dominant EM technique during the pandemic.

In Table 7, for income-increasing EM, the coefficient values of 0.006, 0.032, and 0.140 for the COVID-19 variables of the three models include positive AEM_DD, positive AEM_PM, and positive REM, respectively (columns 2–4). For income-decreasing EM, the coefficient values of −0.002, −0.042, and −0.029 of the COVID-19 variables of the three models are negative AEM_DD, negative AEM_PM, and negative REM, respectively (columns 5–7). Furthermore, the pandemic variable in all models is statistically significant, well below the 5% threshold, except for negative AEM_DD (column 5), confirming that Vietnamese firms engage in more income-increasing REM during the pandemic.

The value of the pandemic variable coefficient is compared according to each EM model. With income-increasing EM, the value of the AEM_DD and AEM_PM coefficient (columns 2 and 3, respectively) are significantly lower than the REM coefficient (column 4). For income-decreasing EM, the value of the AEM_PM coefficient (column 6) is slightly higher than that of the REM coefficient (column 7). In addition, the coefficients of all the above models are statistically significant. The results of this analysis indicate that managers of listed companies in Vietnam practiced a combination of both EM techniques during the pandemic, consistent with the findings of other researchers, such as Garfatta et al. (2023), Klomp (2022), Xiao and Xi (2021), and Yan et al. (2022) (Table 1). However, more REM techniques were practiced compared to AEM during the pandemic following the study of Klomp (2022), but in contrast to other similar studies in Table 1. Therefore, the data does not support hypothesis H₂.

Discussion

This study contributes to the literature by providing results that allow a better understanding of EM strategies in times of crisis. Specifically, this study aims to examine how uncertainties caused by the COVID-19 pandemic have affected the quality of AEM and REM of non-financial listed companies in Vietnam. Our evidence shows that there have been significant changes in accrual and real-based earnings management among Vietnamese companies when comparing pre-crisis and pandemic periods.

AEM significantly changed during the crisis period. However, the downward adjustment trend in income dominated during the pandemic (Tables 4 and 6, column 3) because managers want to generate reported earnings that appear worse during times of crisis to enhance future profits. Our empirical results reveal that companies tended to adopt “big bath” activities more eagerly. This finding reinforces the efficiency view, which is informed by signaling theory, suggesting that managers use EM to signal private information to capital market participants (Habib et al., 2022). This aligns with prior studies such as Ming Chia et al. (2007), who focused on the Asian 1997 crisis, and Liu and Sun (2022) for US companies during the pandemic.

Our study also reports on REM during the pandemic in Vietnam. Research results show that Vietnamese companies have decided on REM strategies in the direction of upward adjustment rather than downward, contrary to the trend of EM before the pandemic. Due to financial difficulties before the crisis, managers had to use policies such as increasing discounts and reducing selling prices to increase sales in order to recover capital to pay debts and maintain operations in the near future. Such conclusions corroborate the evidence of Hsu and Yang (2022), who have shown that UK firms are more engaged in earnings management through actual operations during the period. Our findings are inconsistent with the conclusions of Xiao and Xi (2021) during the COVID-19 pandemic in China. These findings imply that managers use EM to provide misleading information to achieve financial reporting goals to gain benefits (Roychowdhury, 2006) from the perspective of the agency theory-based opportunistic view posits.

Additionally, we provide unique evidence that the magnitude of real-based earnings adjustment is higher than that of accrual-based earnings adjustment, which is different from the argument of Ljubisavljević and Jakobsson (2022) and Papadaki and Tzovas (2017). They suggest no effect in using REM during the pandemic (Papadaki & Tzovas, 2017). Our evidence supports the efficiency view that managers choose REM in cases where the positive effect is greater from information transmitted to capital markets (Zhao et al., 2012). However, there is no evidence that there is a trade-off between these two techniques during the pandemic versus pre-pandemic.

The results of this study indicate two points of view that need further evidence. First, the level of EM during the pandemic was higher than before, the primary reason being that EM is higher during the pandemic. As Vietnam is a developing country, listed companies are small and medium enterprises, and borrowed capital being the primary source of capital (financial leverage is relatively high). While control and supervision mechanisms are still relatively weak, the most likely reason for this difference is technical problems or managers’ fear of punishment if discovered. This implies that the theory of market incentives or the degree of investor protection is likely the primary motivation for managers to adjust earnings, as they managed earnings upward to avoid catastrophic results by reporting an acceptable level of losses to restore investor and stakeholder confidence and signal that the position of the firm is not worse than its competitors and consequently avoid significant deterioration of the firm’s stock price (Lassoued & Khanchel, 2021; Ozili, 2020).

Second, EM was higher during the pandemic, and managers preferred applying REM techniques compared to AEM. Because of the nature of EM techniques that use changes in performance to manage earnings in a way that would not usually occur in a typical business cycle, financial constraints likely limited the decision-making ability. This leaves less room for management discretion in operational decisions. REM implementation is more than AEM to avoid government scrutiny (Kim et al., 2018). Another reason is that, in the context of multiple uncertainties, managers want to sell more goods, liquidate assets to recover capital to cover debts, as well as maintain normal business operations to ensure personal benefits. This could explain why the results are different between the techniques.

Conclusion

The pandemic has forced most companies to adjust their operations to overall slowdown in the economy. Accordingly, there will be fewer opportunities to alter production costs or manipulate sales. In addition, the expectation of business performance and the reduced level of competition is also an incentive for managers to choose an income management strategy that is easier to implement. Based on a sample of 300 firms listed on the Vietnam stock exchanges during the pre-pandemic period (2016–2019) as well as the pandemic period (2020–2021), two discretionary accruals and a REM were determined to measure EM.

The results of this study provide stakeholders with a deeper understanding of the consequences, drivers, and trends of EM when uncertain situations such as the COVID-19 pandemic occur. In the context of small-and medium-sized enterprises, the primary source of business capital is loans; the pandemic has pushed earnings managers to show an acceptable level of loss and, therefore, lessen the impact of the pandemic in the eyes of investors and stakeholders. On the other hand, managers will focus on management to implement short-term instead of long-term goals. Accordingly, they will prefer an income management strategy that is easier to implement and also avoids government scrutiny. In this regard, the research results support protecting managers’ interests and image rather than safeguarding investors according to the theory of market incentives. In addition, through testing the estimation models AEM and REM, we propose to add the variable financial leverage (LV) because the estimated results in our selected sample are reliable and the estimated model has a higher R2 than the original model. Although more studies are required to further test the suitability of the above proposal, we expect the EM estimation model used in the study to be suitable for emerging economies.

In terms of application, this study shows that the quality of financial statements of non-financial listed enterprises in the context of the pandemic is lower, managers tend to use accruals to adjust or reduce income and use business strategies such as increasing sales, liquidating assets, and downsizing to recover capital to have reserve for future response.

This study contributes to the debate on whether earnings quality is impacted during crises, using evidence from Vietnamese firms during the COVID-19 pandemic. Several limitations in this study may point to opportunities for other studies in the future. First, the study did not conduct industry-specific or sector-specific data analysis. Second, we only examined the impact of COVID-19 on the earnings management behavior of non-financial listed companies in Vietnam. Therefore, future research could compare the effects of different market downturns or research using data from other listed companies. Finally, although the EM level estimation results when adding the LV variable are statistically significant (i.e., meet the reliability requirements) and have a higher R2 than the original model, we only examine business data in Vietnam. Accordingly, there should be studies to verify the suitability of the above proposal in other markets.

Footnotes

Author Contribution

We confirm that the manuscript has been read and approved by all authors. We also guarantee that the authorship order has been agreed upon and the criteria for authorship as defined in the “Uniform requirements for manuscripts” have been met.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research,authorship,and/or publication of this article.

ORCID iD

Anh Tuan Dang

Data Availability Statement

The data that support the findings of this study are available from the corresponding author,[Anh Tuan Dang],upon reasonable request.

References

Ahmad-Zaluki

N. A.

Campbell

Goodacre

(2011). Earnings management in Malaysian IPOs: The East Asian crisis, ownership control, and post-IPO performance. The International Journal of Accounting, 46(2), 111–13. https://doi.org/10.1016/j.intacc.2011.04.001

Ali

Amin

H. M.

Mostafa

Mohamed

A. K.

(2022). Earnings management and investor protection during the COVID-19 pandemic: Evidence from G-12 countries. Managerial Auditing Journal, 37(7), 775–797(23). https://doi.org/10.1108/maj-07-2021-3232

Alissa

Bonsall

S.B.

Koharki

Penn

M.W.

Jr. (2013). Firms’ use of accounting discretion to influence their credit ratings. Journal of Accounting and Economics, 55, 129–147. https://doi.org/10.1016/j.jacceco.2013.01.001

Arthur

Tang

Lin

Z. S.

(2015). Corporate accruals quality during the 2008–2010 global financial crisis. Journal of International Accounting, Auditing and Taxation, 25, 1–15. https://doi.org/10.1016/j.intaccaudtax.2015.10.004

Bergstresser

Philippon

(2006). CEO incentives and earnings management. Journal of Financial Economics, 80, 511–529. https://doi.org/10.1016/j.jfineco.2004.10.011

Bernstein

L. A.

Siegel

J. G.

(1979). The concept of earnings quality. Financial Analysts Journal, 35(4), 72–75. http://www.jstor.org/stable/4478260

Bornemann

Kick

Memmel

Pfingsten

(2012). Are banks using hidden reserves to beat earnings benchmarks? Evidence from Germany, 36(8), 2403–2415. https://doi.org/10.1016/j.jbankfin.2012.05.001

Burgstahler

D. C.

Hail

Leuz

(2006). The importance of reporting incentives: Earnings management in European private and public firms. The Accounting Review, 81(5), 983–1016. http://www.jstor.org/stable/4093095

Chen

Jory

Ngo

(2020). Earnings management under different ownership and corporate governance structure: A natural experiment with master limited partnerships. The Quarterly Review of Economics and Finance, 76, 139–156. https://doi.org/10.1016/j.qref.2019.05.005

10.

Cimini

(2015). How has the financial crisis affected earnings management? A European study. Applied Economics, 47(3), 302–317. https://doi.org/10.1080/00036846.2014.969828

11.

Clikeman

P. M.

(2003). Where auditors fear to tread - Internal auditors should be proactive in educating companies on the perils of earnings management and in searching for signs of its use. The Internal Auditor: Journal of the Institute of Internal Auditors, 60(4), 75. http://dx.doi.org/10.25212/lfu.qzj.2.3.08

12.

Cohen

D. A.

Dey

Lys

T. Z.

(2008). Real and accrual-based earnings management in the pre- and post-sarbanes-oxley periods. The Accounting Review, 83(3), 757–787. http://www.jstor.org/stable/30244500

13.

Cohen

Pandit

Wasley

C.E.

Zach

(2020). Measuring real activity management. Contemporary Accounting Research, 37, 1172–1198. https://doi.org/10.1111/1911-3846.12553

14.

Cohen

D. A.

Zarowin

(2010) Accrual-based and real earnings management activities around seasoned equity offerings. Journal of Accounting and Economics, 50, 2–19. http://dx.doi.org/10.1016/j.jacceco.2010.01.002

15.

da Silva Flores

Sampaio

J. O.

Beiruth

A. X.

Brugni

T. V.

(2023). Earnings management during the COVID-19 crisis: Evidence from the Brazilian and American capital markets. Journal of Accounting in Emerging Economies, 13, 760–783. https://doi.org/10.1108/JAEE-10-2021-0317

16.

Dang

V. C.

Nguyen

Q. K.

(2021). Internal corporate governance and stock price crash risk: Evidence from Vietnam. Journal of Sustainable Finance & Investment, 14(1), 1–18. https://doi.org/10.1080/20430795.2021.2006128

17.

De Luca

Paolone

(2019). The impact of the financial crisis on earnings management: Empirical evidence from Italian and Spanish listed companies. Corporate Ownership & Control, 16(2), 121–130. http://doi.org/10.22495/cocv16i2art12

18.

DeAngelo

Skinner Douglas

(1994). Accounting choice in troubled companies. Journal Accounting and Economics, 17(1–2), 113–142. https://doi.org/10.1016/0165-4101(94)90007-8

19.

DeAngelo

L. E.

(1986). Accounting numbers as market valuation substitutes: A study of management buyouts of public stockholders. Accounting Review, 61(3), 400–420. https://www.jstor.org/stable/247149

20.

Dechow

Dichev

(2002). The quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77(Supplement), 35–59. https://doi.org/10.2308/accr.2002.77.s-1.35

21.

Dechow

P. M.

Sloan

R. G.

Sweeney

A. P.

(1995). Detecting earnings management. The Accounting Reveiw, 70(2), 195–225. http://www.jstor.org/stable/248303

22.

Dokas

Leontidis

Eriotis

Hazakis

(2021). Earnings management. An overview of the relative literature. Bulletin of Applied Economics, 8(2), 25–55. https://doi.org/10.47260/bae/823

23.

Ebrahimi

S. K.

Bahraminasab

Mamashli

(2017). The effect of financial crisis on earnings quality. Journal of Accounting and Auditing Review, 23(4), 415–434. https://doi.org/10.22059/acctgrev.2017.60853

24.

Fields

T. D.

Lys

T. Z.

Vincent

(2001). Empirical research on accounting choice. Journal of Accounting and Economics, 31(1–3), 255–307. https://doi.org/10.1016/s0165-4101(01)00028-3

25.

Filip

Raffournier

(2014). Financial crisis and earnings management: The European evidence. The International Journal of Accounting, 49, 455–478. https://doi.org/10.1016/j.intacc.2014.10.004

26.

Garfatta

Hamza

Zorgati

(2023). COVID-19 outbreak and earnings management practice: Case of Tunisia. Asian Journal of Accounting Research, 8, 307–318. https://doi.org/10.1108/AJAR-04-2022-0129

27.

Gormsen

N. J.

Koijen

R. S.

(2020). Coronavirus: Impact on stock prices and growth expectations. The Review of Asset Pricing Studies, 10(4), 574–597. https://doi.org/10.1093/rapstu/raaa013

28.

Graham

J. R.

Harvey

C. R.

Rajgopal

(2005). The economic implications of corporate financial reporting. Journal of Accounting & Economics, 40, 3–73. http://dx.doi.org/10.1016/j.jacceco.2005.01.002

29.

Gunny

K. A.

(2010). The relation between earnings management using real activities manipulation and future performance: Evidence from meeting earnings benchmarks. Contemporary Accounting Research, 27(3), 855–888. https://doi.org/10.1111/j.1911-3846.2010.01029.x

30.

Habib

Ranasinghe

J. Y.

Biswas

P. K.

Ahmad

(2022). Real earnings management: A review of the international literature. Accounting & Finance, 62, 4279–4344. https://doi.org/10.1111/acfi.12968

31.

Habib

Uddin Bhuiyan

Islam

(2013). Financial distress, earnings management and market pricing of accruals during the global financial crisis. Managerial Finance, 39(2), 155–180. https://doi.org/10.1108/03074351311294007

32.

Healy

P. M.

(1985). The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics, 7(1–3), 85–107. https://doi.org/10.1016/0165-4101(85)90029-1

33.

Healy

P. M.

Wahlen

J. M.

(1999) A review of the earnings management literature and its implications for standard setting. Accounting Horizons, 13, 365–383. http://dx.doi.org/10.2308/acch.1999.13.4.365

34.

Hsu

Y. L.

Yang

Y. C.

(2022). Corporate governance and financial reporting quality during the COVID-19 pandemic. Finance Research Letters, 47, 102778. https://doi.org/10.1016/j.frl.2022.102778

35.

Hung

D. V.

Hue

N. T. M.

Duong

V. T.

(2021). The impact of COVID-19 on stock market returns in Vietnam. Journal of Risk and Financial Management, 14(9), 441. https://doi.org/10.3390/jrfm14090441

36.

Jiraporn

Miller

G. A.

Yoon

S. S.

Kim

Y. S.

(2008). Is earnings management opportunistic or beneficial? An agency theory perspective. International Review of Financial Analysis, 17(3), 622–634. https://doi.org/10.1016/j.irfa.2006.10.005

37.

Jones

J. J.

(1991). Earnings management during import relief investigations. Journal of Accounting Research, 29(2), 193–228. https://doi.org/10.2307/2491047

38.

Kazemi

(2022). Impact of COVID-19 pandemic on earnings management: The European Union empirical evidence. (Master’s thesis. Hanken School of Economics, Department of Accounting and Commercial Law, Accounting). https://helda.helsinki.fi/dhanken/handle/10227/509524

39.

Kim

Jeon

(2018). Critical success factors of a design startup business. Sustainability, 10, 2981. https://doi.org/10.3390/su10092981

40.

Kirschenheiter

Melumad

(2002). Can Big Bath and earnings smoothing co-exist as equilibrium Financial reporting strategies? Journal of Accounting Research, 40(3), 761–796. https://doi.org/10.1111/1475-679X.00070

41.

Kjærland

Kosberg

Misje

(2020). Accrual earnings management in response to an oil price shock. Journal of Commodity Markets, 22, 100138. https://doi.org/10.1016/j.jcomm.2020.100138

42.

Klomp

(2022). COVID-19, investor protection and earnings management: A cross-country analysis. Master’s Thesis, Nijmegen School of Management, 1–45.

43.

Kothari

S. P. A

Leone

A. J.

Wasley

C. E.

(2005). Performance-matched discretionary accruals. Journal of Accounting & Economics, 39(1), 163–197. https://doi.org/10.1016/j.jacceco.2004.11.002

44.

Kothari

S. P.

Mizik

Roychowdhury

(2016). Managing for the moment: The Role of earnings management via real activities versus accruals in SEO valuation. The Accounting Review, 91(2), 559–586. http://www.jstor.org/stable/43867623

45.

Kousenidis

D. V.

Ladas

A. C.

Negakis

C. I.

(2013). The effects of the European debt crisis on earnings quality. International Review of Financial Analysis, 30, 351–362. https://doi.org/10.1016/j.irfa.2013.03.004

46.

Kumar

Vij

(2017). Earnings management and financial crisis: Evidence from India. Journal of International Business and Economy, 18(2), 84–101. https://doi.org/10.51240/jibe

47.

Lassoued

Khanchel

(2021). Impact of COVID-19 pandemic on earnings management: An evidence from financial reporting in European firms. Global Business Review, 0(0), 1–25. https://doi.org/10.1177/09721509211053491

48.

Leuz

Nanda

Wysocki

P. D.

(2003). Earnings management and investor protection: An international comparison. Journal of Financial Economics, 69, 505–527. https://doi.org/10.1016/S0304-405X(03)00121-1

49.

Lin

Y. M.

Shen

C. A.

(2015). Family firms’ credit rating, idiosyncratic risk, and earnings management. Journal of Business Research, 68(4), 872–877. https://doi.org/10.1016/j.jbusres.2014.11.044

50.

Liu

Sun

(2022). The impact of COVID-19 pandemic on earnings management and the value relevance of earnings: US evidence. Managerial Auditing Journal, 37(7), 850–868. https://doi.org/10.1108/MAJ-05-2021-3149

51.

Ljubisavljević

Jakobsson

(2022). Earnings Management during the Covid-19 Pandemic: Evidence from Sweden. In A. Ljubisavljevic & C. Jakobsson. Earnings management during the COVID-19 Pandemic: Evidence from Sweden. Master’s Thesis, Department of business studies, Uppsala University, 1–50.

52.

Masruki

Azizan

(2012). The impact of Asian financial crisis to earnings management and operating performance in Malysia. The Social Sciences, 7, 510–516. http://dx.doi.org/10.3923/sscience.2012.510.516

53.

Matsumoto

D. A.

(2002). Management’s incentives to avoid negative earnings surprises. The Accounting Review, 77, 483–514. https://doi.org/10.2308/accr.2002.77.3.483

54.

McNichols

M. F.

(2002). Discussion of the quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77(s-1), 61–69. https://doi.org/10.2308/accr.2002.77.s-1.61

55.

McNichols

Wilson

G. P.

(1988). Evidence of earnings management from the provision for bad debts. Journal of Accounting Research, 26, 1–31. https://doi.org/10.2307/2491176

56.

Ming Chia

Lapsley

Lee

H. W.

(2007). Choice of auditors and earnings management during the Asian financial crisis. Managerial Auditing Journal, 22(2), 177–196. https://doi.org/10.1108/02686900710718672

57.

Ozili

Arun

(2020). Spillover of COVID-19: Impact on the global economy. MPRA Paper 99317, University Library of Munich.

58.

Papadaki

Tzovas

(2017). Financial crisis and accrual and real earnings management in Europe. Corporate Ownership & Control, 14(3), 8–19. https://doi.org/10.22495/cocv14i3art1

59.

Persakis

Iatridis

(2016). Audit quality, investor protection and earnings management during the financial crisis of 2008: An international perspective. Journal of International Financial Markets, Institutions & Money, 41, 73–101. https://doi.org/10.1016/j.intfin.2015.12.006

60.

Roychowdhury

(2006). Management of earnings through the manipulation of real activities that affect cash flow from operations. Journal of Accounting and Economics, 42(3), 335–370. https://doi.org/10.1016/j.jacceco.2006.01.002

61.

Ryu

Chae

S. J.

(2022). The impact of COVID-19 on earnings management in the distribution and service industries. Journal of Distribution Science, 20(4), 95–100. https://doi.org/10.15722/jds.20.04.202204.95

62.

Shen

Liu

Wang

Chai

Cheng

(2020). Novel insights into impacts of the COVID-19 pandemic on aquatic environment of Beijing-Hangzhou Grand Canal in southern Jiangsu region. Water Research, 193, 116873. https://doi.org/10.1016/j.watres.2021.116873

63.

Smales

(2021). Investor attention and global market returns during the COVID-19 crisis. International Review of Financial Analysis, 73, 101616. https://doi.org/10.1016/j.irfa.2020.101616

64.

Stubben

S. R.

(2010). Discretionary revenues as a measure of earnings management. The Accounting Review, 85(2), 695–717. https://doi.org/10.2308/accr.2010.85.2.695

65.

Trombetta

Imperatore

(2014). The dynamic of financial crises and its non-monotonic effects on earnings quality. Journal of Accounting and Public Policy, 33(3), 205–232. https://doi.org/10.1016/j.jaccpubpol.2014.02.002

66.

Walker

(2013). How far can we trust earnings numbers? What research tells us about earnings management. Accounting and Business Research, 43(4), 445–481. https://doi.org/10.1080/00014788.2013.785823

67.

World Bank. (2023). The World Bank in Vietnam. Retrieved May 16, 2023, from https://www.worldbank.org/en/country/vietnam/overview

68.

Xiao

(2021). The COVID-19 and earnings management: China’s evidence. Journal of Accounting and Taxation, 13(2), 59–77. https://doi.org/10.5897/JAT2020.0436

69.

Yan

Liu

Wang

Zhang

Zheng

(2022). How does COVID-19 affect earnings management: Empirical evidence from China. Research in International Business and Finance, 63, 101772. https://doi.org/10.1016/j.ribaf.2022.101772

70.

Zang

(2006). Evidence on the trade-off between real activities manipulation and accrual-based earnings management. The Accounting Review American Accounting Association, 87(2), 675–703. https://doi.org/10.2308/accr-101962012

71.

Zang

A. Y.

(2012). Evidence on the trade-off between real activities manipulation and accrual-based earnings management. The Accounting Review, 87, 675–703. https://doi.org/10.2308/accr-10196

72.

Zhao

Chen

K. H.

Zhang

Davis

(2012). Takeover protection and managerial myopia: Evidence from real earnings management. Journal of Accounting and Public Policy, 31(1), 109–135. https://doi.org/10.1016/j.jaccpubpol.2011.08.004

The Impact of Covid-19 on Earnings Management: Empirical Evidence from Vietnam

Abstract

Plain language summary

Keywords

Introduction

Literature Review and Hypothesis Development

Literature Review

Hypothesis Development

Methodology

The Research Model

Earnings Management Measures

Accruals Earnings Management

Real Earnings Management

Research Data

Data Analysis Methods

Research Results and Discussion

Research Results

Descriptive Statistics and Correlations

Multivariate Analyses

Discussion

Conclusion

Footnotes

Author Contribution

Declaration of Conflicting Interests

Funding

ORCID iD

Data Availability Statement

References