Effect of Earnings Management on Earnings Quality and Sustainability: Evidence from Gulf Cooperation Council Distressed and Non-Distressed Companies

Abstract

This study evaluates the effect of earnings management on earnings quality and sustainability in the GCC region, particularly in distressed and non-distressed companies. Studies on earnings quality and sustainability have mostly concentrated on developed markets, with little attention paid to emerging markets like the GCC region. This research is the first to examine how manipulating earnings impacts the quality and sustainability of earnings in distressed and non-distressed companies. This study utilized a unique dataset that represents the GCC region, which has a specific socio-cultural context. We collected data from 839 publicly listed companies in the GCC region between 2011 and 2022 using DataStream^®, WorldScope (WS), and Refinitiv Eikon. To test our hypotheses and ensure accuracy, we used three types of regressions (the fixed effects model, OLS, and 2SLS) and conducted robustness and endogeneity tests. The results of this study indicate that accruals-based earnings management has a negative impact on earnings quality for distressed and non-distressed firms but a positive effect on earnings sustainability for both types of companies. The results of this study also find variations in earnings management practices across industries. These findings provide valuable guidance for auditors, investors, and other stakeholders to evaluate the earnings quality and sustainability of distressed and non-distressed companies, benefiting the GCC economy and similar economies.

1. Introduction

This study examines whether the connection between earnings management (EM), earnings quality, and sustainability varies between distressed and non-distressed companies. The concepts of earnings management, earning quality, and earning sustainability, along with the methods used to measure them, have consistently sparked engaging and thought-provoking debates. Dechow et al. (2010) state that researchers have employed a range of criteria to measure the quality of earnings, encompassing factors such as persistence, accruals, smoothness, timeliness, loss avoidance, and investor responsiveness. They added that researchers have also considered external indicators, such as restatements and SEC enforcement releases, to gauge earnings quality. Dechow et al. (2010) do not arrive at a definitive conclusion regarding the definition of earnings quality, as they recognize that the concept of “quality” is dependent on the specific decision-making context. Additionally, they highlight that the “quality” of earnings is closely linked to the fundamental performance of the firm. This study makes a clear distinction between the three concepts (i.e., EM, earnings quality, and earnings sustainability) and measures each one differently. A clear understanding of the quality and sustainability of a company’s earnings is essential in accounting (Dichev et al. 2013). Understanding these two major factors can provide valuable insights into a company’s financial health and future prospects. Analyzing these metrics is crucial for making informed decisions regarding investments, partnerships, and strategic planning. Such an analysis would necessitate an examination of financial statements, which can provide a thorough picture of a company’s current situation, prospects, and potential hazards. This information can be utilized to discover areas for improvement, maximize financial performance, and assure long-term success.

Earnings quality (EQ) refers to the accuracy with which a company reports its income and predicts future earnings (DeFond 2010). This is because EQ accurately reflects a firm’s financial statements and provides stakeholders with a reliable perception of its performance (Dechow et al. 2010). By contrast, earnings sustainability (ES) clearly shows a firm’s ability to maintain its earnings performance over time (Chen et al. 2014). However, EQ and ES are multifaceted intersections that present several complexities. Although high-quality earnings are generally assumed to indicate sustainability, this relationship is not always straightforward. Indeed, one can assume that the notion of earnings “quality” is contextual and can mean different things to different users of financial statements (Dechow and Schrand 2004). A company may occasionally report high-quality earnings based on specific metrics; however, these earnings may not be sustainable if they are the product of one-time events or favorable economic conditions unlikely to persist over an extended duration (Penman 2003). As such, it is critical to examine numerous aspects when assessing the quality and durability of a company’s earnings. Companies’ financial health is mostly dependent on their capacity to sustain a high EQ and ES. It is crucial to handle earnings appropriately to ensure that a company’s financial performance is accurately represented and to prevent any manipulation of earnings information.

EM is the deliberate action taken by management to meet desired earnings levels or smooth earnings, which can potentially obscure the true financial performance of a company (Schipper 1989; Dechow et al. 2012). Research has shown that companies with a high EQ are less likely to engage in aggressive EM (Dechow et al. 2010; Lo 2008). Conversely, a lack of ES may indicate impending financial distress. If a company cannot generate sustainable earnings, it may face competitive disadvantages, operational inefficiencies, or other serious problems that could lead to financial distress or bankruptcy (Altman et al. 1977; Altman et al. 2017; Charitou et al. 2011; Howe and Houston 2016; Elmassri et al. 2020). Studies have shown that companies in financial distress have a greater chance of going bankrupt and, as such, may resort to manipulating their earnings which would ultimately result in a lowered EQ (Trombetta and Imperatore 2014; Habib et al. 2013). However, the effect of financial distress on ES is not fully understood. Therefore, it is crucial to examine whether the connection between EM, EQ, and ES varies between distressed and non-distressed companies. Therefore, it is essential to study the EQ and ES to identify any warning signs of EM and evaluate the risk of financial distress. This can help companies avoid financial distress and bankruptcy while promoting greater transparency and accountability in financial reporting.

EM has been extensively researched, leading to a discussion of its impact on the quality of earnings and sustainability. While researchers such as Akers et al. (2007) and Healy and Wahlen (1999) believe that EM reduces EQ, others such as Subramanyam (1996) and Lee et al. (2006) argue that it can enhance EQ and ES. This ongoing discussion highlights the complexity of the issue and the need for continued research to gain a complete understanding of the effects of EM on EQ and ES (Mcnichols 2002). The main objective of this study is to comprehensively analyze EM’s effects on the EQ and ES of distressed and non-distressed companies operating in the Gulf Cooperation Council (GCC) region. It is important to note that the GCC plays a significant role in the global economy; therefore, understanding the impact of EM on businesses in this region is crucial for ensuring sustainable economic growth.

Most studies examining the quality and sustainability of earnings have focused on developed markets such as the US and Europe (Francis et al. 2005; Spohr 2005), focusing little on emerging markets, particularly the GCC region. Consequently, there is a significant research gap in the evaluation of the relationship between EQ and ES in distressed and non-distressed companies operating in the GCC region. GCC countries are categorized as emerging markets, and research on EM practices in these markets can provide unique insights not found in more developed markets. Several distinctive institutional and operational features make the GCC region an important setting in which to examine EM and whether such manipulative strategies differ between distressed and non-distressed GCC companies. First, the extant literature documents that GCC financial markets are characterized by high ownership concentration levels, where firms are controlled by a small number of majority controlling investors (Al-Amri et al. 2017; Al-Sehali and Spear 2004; Dalwai et al. 2015). Second, GCC markets are influenced by various factors such as corporate governance structures, regulatory frameworks, cultural practices, and regional characteristics which provides an interesting research environment in which to examine EM, EQ, and ES. Third, markets in the GCC region suffer from high levels of information asymmetry and economic uncertainty coupled with suboptimal levels of regulatory oversight mechanisms, which may induce firms operating in turbulent environments to manipulate their earnings.

Against this backdrop, this study aims to fill this gap in the literature by investigating whether managers of GCC distressed companies are more inclined to engage in EM than those in non-distressed companies. This study also aims to determine the impact of this behavior on EQ and ES. The issue is rooted in the fact that some businesses may manipulate their earnings to achieve or exceed their financial goals, influence the perspectives of various stakeholders, including investors, lenders, and regulatory bodies, or enhance their overall financial standing (Peasnell et al. 2000; Athanasakou et al. 2011; Walker 2013). Furthermore, this study examines industry-specific differences in the EQ and ES of distressed and non-distressed companies, addressing whether managers of distressed companies behave differently from those of non-distressed companies. This study is the first to investigate the impact of manipulating earnings on the quality and sustainability of earnings in both distressed and non-distressed companies. This study utilized a unique dataset that specifically represents the GCC region, which has a distinct socio-cultural context.

Utilizing data from a sample of 839 publicly listed companies in the GCC region from 2011 to 2022, obtained from DataStream^®, WorldScope (WS), and Refinitiv Eikon, this study focuses on gaining a complete understanding of the decision-making process used by companies in managing their earnings. Abnormal discretionary accruals are used as a proxy for the EM approaches of distressed and non-distressed firms. By analyzing the changes in discretionary accruals over specific years, this study provides insights into the choices made by companies regarding income-increasing or -decreasing accounting choices. Notably, no previous study has examined the effect of EM on EQ and ES in distressed and non-distressed companies in the GCC region. The results of this study provide valuable insights into the relationship between EM, EQ, ES, and financial distress. The findings of this study provide a deeper understanding of financial statements and the vital role that auditors play, especially in times of economic difficulty. It is worth noting that this study’s findings are particularly significant when considering the distinct socio-cultural settings prevalent in the GCC region.

The findings of this study may serve as a guide for investors and other capital providers to evaluate the EQ and ES of distressed and non-distressed companies. In conclusion, the findings of this study hold great promise for bolstering the GCC economy and other comparable markets. This could help increase investor trust, maintain regulatory structures, enhance corporate governance, and ultimately make the region and other similar regions more attractive to foreign investment.

The remainder of this paper is organized as follows: Section 2 provides a background, and Section 3 reviews the relevant literature and develops the hypotheses. The research design and empirical models are discussed in Section 4. Section 5 presents and discusses the empirical findings. Finally, a conclusion is presented in Section 6.

2. Background

The GCC region, including Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates (UAE), offers a complex environment to study the relationship between EM, EQ, and EQ due to regulatory changes and economic initiatives. In recent years, the GCC region has seen significant changes in its regulatory landscape, particularly in financial transparency and accountability. Regulatory bodies like the Emirates Securities and Commodities Authority (ESCA) in the UAE and the Capital Market Authority (CMA) in Saudi Arabia, as well as similar institutions in other GCC countries, have played a crucial role in implementing stricter financial reporting standards. These new regulatory frameworks are specifically designed to ensure that companies maintain high levels of financial disclosure and reporting, ultimately working to reduce the occurrence of EM.

After the global financial crisis, the GCC countries have reformed corporate governance codes and guidelines to strengthen auditor roles, increase board independence, and empower shareholders. The aim is to create a more resilient governance framework, reduce the risks of EM, and promote higher-quality earnings. For instance, in 2020, the UAE introduced new regulations to enhance transparency and accountability in the corporate sector. Similarly, the Saudi Vision 2030 emphasizes transparency and accountability, encouraging companies to adopt best governance practices to foster integrity and responsible governance throughout the region.

The GCC countries are diversifying their economies to reduce reliance on oil revenues, requiring investment in non-oil sectors and sustainable financial practices. High-quality earnings are crucial for attracting investment and maintaining economic balance. This diversification is happening alongside traditional values and rapid modernization, shaping corporate culture and financial reporting. A high ownership concentration and adherence to Islamic principles significantly influence EM and financial reporting decisions. Additionally, the region’s financial markets are marked by significant information imbalance and economic uncertainty, creating opportunities for manipulating earnings. Studying the relationship between information imbalance, regulatory oversight, and EM practices is valuable for understanding their combined impact on earnings quality and sustainability.

In summary, the GCC region provides an evolving regulatory environment, corporate governance reforms, and distinct socio-cultural characteristics, making it an excellent area to study EM, EQ, and ES. This research aims to enhance our understanding of these issues to improve financial transparency and stability in the region.

3. Literature Review and Hypotheses Development

This section analyzes recent research on the connections between EM, financial distress, EQ, and ES and explains the methods of EM and their impact on EQ and ES. Additionally, we investigate the factors that contribute to the occurrence of EM, with an emphasis on the financial distress of companies. Our literature review offers a comprehensive overview of these topics, culminating in a thorough summary of the impact of EM on the EQ and ES of both distressed and non-distressed firms.

This section is organized into several subsections. In Section 1, various EM studies are presented. Section 2 reviews selected studies that examine the link between EM and EQ. Section 3 provides relevant studies on the connection between EM and ES. Section 4 explores the correlation between EM and distressed and non-distressed companies. In Section 5, we explain the relationship between EM, EQ, and ES across various industries. Finally, Section 6 offers a conclusion for this section.

3.1. EM Studies

Earnings management is the practice of altering financial statements to achieve specific objectives, which may include improving reported earnings or reducing fluctuations in earnings (Hall et al. 2013). This practice often involves accounting accruals, as highlighted by Alfadhael (2021), Dechow et al. (2012), Aljifri and Moustafa (2007), and Healy (1985). Accrual changes are a preferred method for earnings manipulation because of their lower cost and subtlety compared to other methods. In general, managers often use discretion over accrual items to manipulate reported earnings for their own benefit, either to avoid contracting and political costs or to increase their interests (Watts and Zimmerman 1978).

Healy and Wahlen (1999) argue that incentives such as bonuses and stock options tied to earnings performance can motivate firms. According to alternative theories, firms engage in EM to maintain a steady earnings growth pattern, which can attract investors and have a favorable impact on stock prices, as Dechow et al. (1995) point out. Additionally, firms have been found to manage their earnings in an attempt to meet or surpass financial analysts’ projections, which can strengthen a company’s reputation. Furthermore, external factors such as debt covenants or the need for inexpensive external financing may also contribute to EM (Alfadhael 2021). Several theories have been proposed to explain the motives and characteristics of EM, as summarized by Iatridis and Kadorinis (2009). Theories of EM emphasize the importance of understanding the prevalence, legality, ethicality, and potential negative consequences of this practice, as noted by Dechow et al. (1995). This suggests that while EM practices have been widely studied, further critical examination of their ethical implications is necessary. Expanding research geographically can also lead to a more comprehensive understanding of global accounting practices.

3.2. EM and EQ

A broad range of studies has addressed the relationship between EM and EQ, with a particular focus on their implications for investors and regulators. Examples of such studies include those conducted by Jiang (2020), Man and Wong (2013), and Leuz et al. (2003). Researchers are divided into two factions to explore the connection between EM and EQ. The first faction argues that EM practices can improve EQ and is justified by legitimate business decisions and accounting practices. The second faction believes that these practices can negatively impact EQ and reflect unethical behavior or manipulation. Studies conducted by Graham et al. (2005), Burgstahler and Eames (2006), and Subramanyam (1996) support the former argument, while the findings of studies conducted by Dichev et al. (2013), Ettredge et al. (2010), Francis et al. (2005), and Dechow et al. (1995) support the latter argument.

In summary, some studies suggest that EM can harm EQ by creating inaccuracies and undermining investors’ trust. On the other hand, other studies indicate that EM can be beneficial in certain contexts, providing valuable information to investors and aligning them with their expectations. As such, we posit our first hypothesis as follows:

H1: 

EM has a significant negative effect on EQ.

3.3. EM and ES

Researchers and professionals in accounting and finance have recently paid considerable attention to both EM and ES. Earnings management can have either positive or negative effects on ES. Some levels of EM can help stabilize earnings fluctuations, making them more predictable and sustainable. This can provide stakeholders with a more dependable and stable view of a company’s performance, leading to increased confidence and investment in the company (Ali and Zhang 2015; Barua et al. 2010; Ghosh et al. 2005; Graham et al. 2005; Davis-Friday and Frecka 2002; Healy and Wahlen 1999). However, the adverse effects of EM on sustainability were significant. When companies excessively manage their earnings, they can distort the true financial health of the business, resulting in misleading financial statements. This can lead to a loss of trust among stakeholders and investors, ultimately damaging a company’s long-term sustainability (Guay et al. 2016; Kothari et al. 2016).

In summary, these studies highlight that although EM can provide short-term benefits, it often poses risks to long-term financial health and sustainability. The twofold effect of EM on ES highlights the need to wisely implement EM strategies and prioritize long-term prosperity over short-term profits. This serves as the foundation for the second hypothesis:

H2: 

EM has a significant positive effect on ES.

3.4. EM, EQ and ES, and Distressed and Non-Distressed Companies

Various studies have demonstrated that distressed companies often adopt accounting practices that increase reported income to fulfill debt covenants and reduce the costs associated with financial distress (Tulcanaza-Prieto et al. 2020; Aljifri and Taylor 2002; Beneish 2001; DeAngelo et al. 1994). On the other hand, EM can worsen financial distress by presenting an inaccurate image of stability and concealing real issues within the company (Jiang 2020). Moreover, financial distress can also increase the pressure on managers to engage in EM as they strive to maintain their job security and compensation (Strakova 2021). Financial distress may lead to EM as managers attempt to overcome the challenges and pressures they encounter (Kamal and Khazalle 2021). Essentially, financial distress heightens the likelihood of managers engaging in EM to manipulate reported earnings and create a more favorable financial outlook (Kurniawan and Hermawan 2017). Such practices during periods of financial distress can adversely affect the quality and sustainability of reported earnings (Humeedat 2018). These findings suggest that EM significantly impacts the quality and sustainability of earnings in companies with higher debt levels, leading to our third hypothesis:

H3: 

The impact of EM on EQ and ES varies significantly in distressed and non-distressed companies.

3.5. EM, EQ, and ES across Industries

The prior literature suggests that industry-specific factors may influence EM practices, thereby affecting the quality and sustainability of earnings. Traditionally, industries such as hospitality, airlines, and car-rental firms have utilized revenue or yield management techniques to enhance profitability and long-term sustainability (Denizci Guillet and Mohammed 2015; Jiang et al. 2010). Industries sensitive to economic fluctuations, such as oil, gas, and real estate, may use EM to stabilize their earnings (Ball and Shivakumar 2008). In contrast, heavily regulated industries may have a limited scope for EM. Additionally, the varied adoption of IFRS among GCC companies introduces further discrepancies in EM practices, affecting earnings across different industries (Daske et al. 2008). This analysis supports the hypothesis that EM’s impact on EQ and ES varies notably between industries, emphasizing the need for industry-tailored assessments in the GCC context.

H4: 

EM varies significantly across industries in the GCC region.

3.6. Conclusions

The impact of EM on EQ and ES has been debated in the literature. While some studies suggest positive effects, most indicate that it could negatively affect them. Excessive or opportunistic EM can harm financial reporting quality and investor confidence, leading to negative consequences for firms and stakeholders. It is crucial to exercise prudence and caution when dealing with such practices and ensure that they align with ethical principles and standards. Therefore, it is imperative to maintain a balance between maximizing profitability and ensuring ethical conduct during EM. Overall, most evidence suggests that EM has a negative impact on EQ but positively influences ES. These effects vary according to the company’s financial health and industry characteristics.

4. Research Methodology

4.1. Sample Selection

Our study explores the impact of EM on the EQ and ES of distressed and non-distressed companies within the GCC region from 1 January 2011 to 31 December 2022. To gather relevant data, we retrieved information on all publicly listed firms from the Thomson Refinitiv Eikon database. Additionally, financial data for GCC publicly listed firms were collected from the DataStream database. We restricted our sample to publicly listed non-financial firms having a complete dataset. Financial firms (SIC Codes 60-69) were excluded from the sample for several reasons documented in the literature. First, as compared to other sectors, the financial sector is extremely regulated which may impact the performance and reporting practices of firms operating in this sector (Elrazaz et al. 2021; Elmassri et al. 2024). Second, the extant literature has not yet documented the reasonableness of EM models, such as the Jones model and its variations, in detecting accruals management in financial firms (Vasilescu and Millo 2016). Finally, the exclusion of such financial firms is a common practice in the literature (Lo et al. 2017; Kothari et al. 2016; Campa and Hajbaba 2016; Raman et al. 2013; Botsari and Meeks 2008; Louis 2004; Erickson and Wang 1999; Elrazaz 2019). The exclusion of the financial sector resulted in a final sample of 839 listed companies for the study period. It is also noted that the sample was heterogeneous in terms of the level of distress, where some firms, for example, are already in payment and/or technical default, while others are in pre-breach negotiations or exhibiting some indicators of financial distress.

This study is based on an initial sample of 9324 firm-year observations distributed into 11 industrial sectors on the basis of The Refinitiv Business Classification (TRBC) industry classification in Eikon. In addition, several firm-year observations were excluded due to a lack of full accounting data for the various EM models used and/or data required for control variable construction. As a result of these exclusions, the final sample consisted of 3900 firm-year observations. All continuous data variables used in the regression models and analysis were winsorized at the 1% and 99% percentiles to reduce the influence of extreme observations and outliers.

4.2. Empirical Models

To examine the first research hypothesis on the impact of EM on EQ and whether managers of distressed companies are more inclined to manipulate earnings than those of their counterparts, we run the empirical model (1) stated below. All our empirical models are estimated while clustering at the “firm” level to produce Rogers (1993) robust standard errors, which are heteroskedasticity and autocorrelation consistent. Moreover, industry dummies are added to the regressions to control for industry fixed effects, and following Petersen (2009), year dummies are added to control for time effects1. For all models, industry and year fixed effects are not reported for the sake of brevity.

EQ_it = β0 + β1(MJM)_it + β2(EV)_it + β3(ROA)_it + β6(Firm Size)_it + β4(Industry Type)_i + β5(Distress)_it +
β7(MJM*Distress)_it + β8(EV*Distress)_it + β8(ROA*Distress)_it + β9(Firm Size*Distress)_it + ε

(1)

In the above model, EQ is calculated as cash flow from operations (CFO) divided by operating profit. The main explanatory variable of interest is the Modified Jones Model (MJM) which is our proxy for EM. The control variables are earnings volatility (EV) calculated as the standard deviation of the change in earnings levels over a 2-year period, firm performance measured by return on assets (ROA), firm size (Firm Size) measured as the natural logarithm of total assets, and Industry Type, which is industry dummies to control for industry fixed effects. “Distress” is an indicator variable that takes the value of 1 for firms with a debt ratio equal to or exceeding 62%2.

To examine the second research question on the impact of EM on ES and whether managers of distressed companies are more likely to conduct EM than those of non-distressed companies, we run the following empirical model (2):

ES_it = β0 + β1(MJM)_it + β2(Growth)_it + β3(ROA)_it + β4(Firm Size)_it + β5(Industry Type)_i + β6(Distress)_it +
β7(MJM*Distress)_it + β8(ROA* Distress)_it + β9(Firm Size*Distress)_it + ε

(2)

In the above model, ES is calculated as the standard deviation of the annual operating income growth rate for a firm over the period 2011–2022 on a rolling window basis. Revenue growth rate (GROWTH) is likely to affect EM since a surge in revenues would affect certain accruals and inflate expectations of future growth and cash flows. The revenue growth rate is also included as a control variable in the model and calculated as the percentage change in revenues. All other variables are as previously described. Appendix A provides detailed definitions of these variables.

In testing our fourth hypothesis to examine whether companies in particular industries demonstrate a greater tendency to manipulate their earnings, we employ both parametric (ANOVA) and non-parametric (Kruskal–Wallis) tests.

EM Measurement

Following the literature on EM, the cash flow approach is used to calculate total accruals (TACCs) while employing the Modified Jones Model (Dechow et al. 1995) in its cross-sectional form. Total accruals computed from the statement of cash flows are determined as the difference between net income before extraordinary items (NI) and cash flows from operations (CFOs) as follows:

The WorldScope item codes are in parentheses.

TACC_t = NI_jt (WC04001) − CFO_jt (WC04201 + WC04831)

where the following variables are used:

NIjt = net income before extraordinary items for firm j at time t;

CFOjt = operating cash flows as from the statement of cash flows for firm j at time t.

For the purposes of this study, when the Modified Jones Model (Dechow et al. 1995) is used to estimate abnormal discretionary accruals, nondiscretionary accruals (NDACCs) are estimated as follows:

NDACC_t = α₁ (1/A_t−1) +α₂ (ΔREV_t − ΔREC_t) + α₃ (PPE_t)

where the following variables are used:

ΔREVt	= revenues in year t less revenues in year t − 1 scaled by total assets at t − 1;
PPEt	= gross property, plant, and equipment in year t scaled by total assets at t − 1;
At−1	= total assets at t − 1;
ΔRECt	= net receivables in year t less net receivables in year t − 1 scaled by total assets at t − 1 and all other variables as previously defined;
α1, α2, α3	= firm-specific parameters.

The estimates for firm-specific parameters αˆ1, αˆ2, and αˆ3 are generated using the following OLS regression:

TACC_t = a₁ (1/A_t−1) +a₂ (ΔREV_t − ΔREC_t) + a₃ (PPE_t) + γ_t

where the following variables are used:

TACCt	= total accruals scaled by lagged total assets;
a1, a2, a3	= the ordinary least squares estimates of α1, α2, and α3;
γt	= measurement error in year t and all other variables as previously defined.

Following Botsari and Meeks (2008), our cross-sectional estimation consists of constructing industry–event period-matched portfolios for each firm in the sample. Subsequently, regression (3) is run for each industry-year portfolio to yield industry-year-specific estimates of α, β₁, and β₂. As a second stage, these estimates are combined with firm-specific data in Equation (4) to generate estimated discretionary accruals for each firm3.

TACC_ijp/A_ijp−1 = α_jp + β_1jp(ΔREV_ijp/A_ijp−1) + β_2jp(PPE_ijp/A_ijp−1) + ε_ijp

(3)

where the following variables are used:

TACCijp	= total accruals for estimation portfolio j for firm i in event year p;
ΔREVijp	= change in revenue (total sales) for estimation portfolio j for firm i in event year p;
PPEijp	= gross PP&E for estimation portfolio j for firm i in event year p;
Aijp−1	= total assets at beginning of period for estimation portfolio j for firm i in event year p;
εijp	= measurement error for estimation portfolio j for firm i in event year p;
i	= 1, …, N firm index;
j	= 1, …, J estimation portfolio index;
p	= 1, …, P year index.

Hence, the second stage estimates discretionary accruals (DACCs), as shown in the following equation:

DACCs_ip = TACC_ip/A_ip−1 − [a_jp + b_1jp(ΔREV_ip/A_ip−1 − ΔREC_ip/A_ip−1) + b_2jp(PPE_ip/A_ip−1)]

(4)

Using the cross-sectional coefficients estimated from Equation (3) above with the residuals representing estimated discretionary accruals, abnormal accruals for a firm in the sample are determined in the second stage, as shown in Equation (4). Accordingly, discretionary accruals as a proxy for EM are calculated as shown in the following Equation (5):

DACC_ip = [TACC_ip/A_ip−1] − NDACC_ip

(5)

where the following variables are used:

TACCip	= total accruals for firm i in event year p;
DACCip	= discretionary accruals for firm i in event period p;
NDACCip	= nondiscretionary accruals for firm i in event period p;
i	= 1, …, N firm index;
p	= 1, …, P year index.

5. Empirical Findings

5.1. EQ and EM

Table 1. Descriptive statistics and correlation matrix.

Panel A. Descriptive Statistics
Variables		Mean	Std. Dev.	Min		Max	1st percentile	99th percentile
EQ		0.949	6.347	−35.567		31.130	−35.567	31.034
EM		0.000	0.087	−0.330		0.393	−0.273	0.348
EV		−19.009	361.630	−2227.358		1606.701	−2089.268	1594.212
ES		4.689	1.526	1.372		9.038	1.539	9.038
Growth		12.333	60.906	−100.000		444.444	−98.038	386.707
ROA		0.044	0.086	−0.303		0.306	−0.298	0.306
Firm Size		13.307	2.539	8.047		19.472	8.072	19.472
Distress		0.464	0.499	0.000		1.000	0.000	1.000
Observations		3621
Panel B. Correlation Matrix
Variable	EQ	ES	EM	EV	ROA	GROWTH	Firm Size	Distress
EQ	1.000
ES	−0.058 **	1.000
EM	−0.077 **	−0.019	1.000
EV	0.013	−0.030	0.125 **	1.000
ROA	0.055 **	−0.309 **	0.367 **	0.171 **	1.000
Growth	−0.004	0.071 **	0.151 **	0.079 **	0.167 **	1.000
Firm Size	0.030	−0.062 **	0.010	0.046 *	0.125 **	0.049 *	1.000
Distress	−0.007	0.133 **	−0.071 **	−0.052 *	−0.174 **	0.005	0.178 **	1.000

Panel B presents the Pearson Correlation Matrix for key variables for the sample firm-year observations. *, **—correlation coefficient is significant at 5% level and 1% level or better, respectively.

reports summary statistics and the correlation matrix for variables used in our models. First, Panel A displays the descriptive statistics for all continuous variables used in our models. Our main explanatory variable of interest (EM) which is a proxy for EM has a mean value of 0.000, consistent with the prior literature (Lee and Lu 2014) which indicates that, on average, firms in our sample do not manipulate their earnings in the period under study. Nevertheless, this should be interpreted with caution since univariate analysis is susceptible to a high likelihood of bias due to omitted factors. Interestingly, the maximum value for EM is 0.393 which indicates an upward direction of EM. Moreover, the minimum value for EM is −0.330 which is an indication of a downward direction of EM. As such, this is an indication that some firms in the sample manipulate earnings upwards while others manipulate earnings downwards. Accordingly, multiple regression analyses are needed to further support or refute the results of the univariate analysis. The mean for “Distress” which is our indicator for distressed firms is 0.464. Notably, the firms in our sample are large in size which might have a strong impact on the EQ and ES.

Panel B in Table 1 presents the correlation matrix among the variables used in our study. Our main explanatory variable of interest (EM) is negatively correlated with EQ (−0.077) indicating that EM practices negatively impact the EQ for firms in our sample. The correlation between the ROA and EQ (0.055) is positive and highly significant indicating that highly profitable firms display high quality earnings. “MJM” which is our proxy for EM is negatively correlated with our distress measure, Distress ((correlation coefficient −0.071) significant at better than the 1 percent level) indicating that distressed firms in our sample engage in downward EM.

Table 2. Regression analysis of earnings quality on earnings management proxies and controls.

Dependent Variable	Earnings Quality (EQ) Modified Jones Model Discretionary Accruals (1)	Earnings Quality (EQ) Kothari et al. (2005) Performance-Matched Discretionary Accruals (2)
Variable
Intercept	0.637 **	0.675 **
	(3.49)	(3.58)
Earnings management proxy	−5.583 **	−4.557 **
	(−15.80)	(−12.99)
Earnings volatility	−1.203	−1.450 *
	(−1.93)	(−2.21)
Return on assets (ROA)	0.004 **	0.002 **
	(6.09)	(2.64)
Firm Size	−0.007	−0.012
	(−0.59)	(−1.09)
Distressed Firms	−0.565	−0.587
	(−1.67)	(−1.65)
EM*Distressed Firms	0.469	−0.395
	(0.64)	(−0.53)
Earnings volatility * Distressed Firms	0.714	0.857
	(1.12)	(0.84)
ROA * Distressed Firms	0.017 **	0.003
	(2.60)	(0.39)
Firm Size * Distressed Firms	0.027	0.035
	(1.18)	(1.47)
Observations	2459	2459
F-statistic	14.71 **	10.93 **
Year dummies	Yes	Yes
Industry dummies	Yes	Yes

All variables are winsorized at the top and bottom 1 percentile. (**, * represent statistical significance at 1% and 5%, respectively). T-statistics (in parentheses below the coefficients) are calculated using Rogers (1993) clustered robust standard errors clustered at the firm level and are robust to both heteroskedasticity and serial autocorrelation (Petersen 2009). Industry fixed effects and year fixed effects are not reported for the sake of brevity.

presents the main regression results of EQ on the EM proxies and various controls. Model 1 uses discretionary total accruals derived from the Modified Jones Model as a proxy for EM. To provide further robustness to the results, we performed an additional analysis using Kothari et al.’s (2005) measure of EM, Model 2. We reran our main baseline multivariate regression using this alternative measure as a proxy for EM. Overall, the results reported under the performance-adjusted accruals model (Kothari et al. 2005) are qualitatively similar to those obtained in our main analysis. Several studies in the EM literature emphasized the importance of considering performance as a key factor when testing for EM (McNichols 2000; Kothari et al. 2005). Indeed, a plethora of published research examining EM documents the relationship between firm performance and the degree of discretionary accruals (Christensen et al. 2022).

To ensure that the results derived from our main analysis are robust to alternative measures of EM, we follow Kothari et al. (2005) and include contemporaneous return on assets (ROA) as an additional regressor in Model 2. In both models, the indicator variable Distressed Firms is used as a proxy for distressed firms. Firms with debt ratios equal to or higher than 62% are considered distressed. In both models, the EM coefficient is negative and statistically significant (statistical significance at 1%), indicating that EM practices negatively impact the quality of earnings for firms in our sample. In Model 2, we find that the coefficient of our main explanatory EM variable (Kothari Discretionary Accruals) is negative and statistically significant at the 1% level. These results are consistent with and provide further robustness to those reported in Model 1, indicating that EM practices negatively impact the EQ for the firms in our sample. After thoroughly analyzing these results, it is evident that Hypothesis 1 is supported. These results indicate that EM negatively affects the EQ, leaving little room for doubt or disagreement.

One explanation for this result is that managing earnings can provide inaccurate information about a company’s financial health, which can adversely affect the company’s overall economic standing. This can also lead to a decrease in the EQ, making it challenging for stakeholders and investors to make informed decisions. Another possible reason for this is that managing earnings can involve manipulating financial results in the short term to meet specific goals, such as analysts’ predictions or performance-based compensation. This can lead to a drop in the long-term EQ, as companies may prioritize these goals when making strategic decisions. Additionally, managing earnings may include practices that violate Generally Accepted Accounting Principles (GAAP) or International Financial Reporting Standards (IFRS), which are intended to ensure the accountability and comparability of financial information. These violations can decrease the EQ and cast doubt on the credibility of the financial reporting process (Leuz et al. 2003). Moreover, in both models, the coefficient on ROA is positive and statistically significant (statistical significance at 1%), indicating that firms with high profitability display a higher EQ.

Based on our analysis, it appears that our Hypothesis 3 regarding the relationship between a company’s debt level (financial distress), EM, and the quality of its earnings may not hold true. Examining the relationship between company distress, EM, and EQ is a complex issue influenced by various factors, including corporate governance structures, regulatory frameworks, cultural practices, and regional characteristics. In GCC countries, both distressed and non-distressed companies demonstrate similar levels of control mechanisms and regulations and comparable degrees of EM and EQ. In GCC countries, family-owned or government-involved companies may have similar financial management practices, resulting in less variation between distressed and non-distressed companies in terms of EM and quality. Distressed firms may avoid manipulating earnings to protect their reputation, while non-distressed firms may have similar motivations (Aldamen et al. 2012). Another possible reason for the lack of distinction could be the strong regulatory monitoring in GCC nations. For example, entities such as the Emirates Securities and Commodities Authority (ESCA) enforce stringent rules and policies, reducing the possibility of fraudulent activities among all companies (Aljifri and Moustafa 2007). In addition, cultural practices and economic factors in the GCC can promote uniformity in business practices and financial reporting, which can help reduce discrepancies between distressed and non-distressed companies. Additionally, adherence to Islamic principles of fairness can encourage a more cautious approach to financial reporting (Baydoun and Willett 2000). Furthermore, publicly traded companies in the GCC region are mandated to subject their financial statements to thorough and meticulous audits by accredited auditors. This essential practice is a vital tool in ensuring the integrity and accuracy of earnings reports, thus reducing any differences in this matter between distressed and non-distressed companies.

The inclusion of the interaction terms between the EM discretionary accruals, ROA, Firm Size, and EV with Distressed Firms did not significantly affect the relationship between the EQ and these variables. However, the positive and significant coefficient of ROA* Distressed Firms in Model 1 suggests that the positive relationship between the ROA and EQ may be stronger for firms with high debt ratios. Overall, the regression models are statistically significant and have explanatory power (F-statistic for both models is statistically significant), and year and industry dummies were included to control for unobservable time and industry-specific effects. These fixed effects help improve the accuracy of the estimates and reduce the risk of omitted variable bias. These results provide further robustness to the inferences obtained from the descriptive statistics and the correlation matrix presented in the previous section.

5.2. ES and EM

Table 3. Regression analysis of earnings sustainability on earnings management proxies and controls.

Dependent Variable	Earnings Sustainability (ES) Modified Jones Model Discretionary Accruals (3)	Earnings Sustainability (ES) Kothari et al. (2005) Performance-Matched Discretionary Accruals (4)
Variable
Intercept	5.281 **	5.211 **
	(30.46)	(30.02)
Earnings management proxy	1.895 **	2.164 **
	(5.26)	(5.98)
Revenue growth rate	0.003 **	0.003 **
	(6.78)	(6.48)
Return on assets (ROA)	−6.257 **	−5.171 **
	(−18.80)	(−15.55)
Firm Size	−0.041 **	−0.040 **
	(−3.19)	(−3.14)
Distressed Firms	−0.630	−0.639
	(−1.70)	(−1.72)
EM * Distressed Firms	−0.442	−0.600
	(−0.62)	(−0.85)
ROA * Distressed Firms	0.005	0.004
	(0.64)	(0.58)
Firm Size * Distressed Firms	0.070 **	0.070 **
	(2.69)	(2.70)
Observations	3621	3621
F-statistic	10.957 **	10.949 **
Year dummies	Yes	Yes
Industry dummies	Yes	Yes

All variables are winsorized at the top and bottom 1 percentile. (**, * represent statistical significance at 1% and 5%, respectively). T-statistics (in parentheses below the coefficients) are calculated using Rogers (1993) clustered robust standard errors clustered at the firm level and are robust to both heteroskedasticity and serial autocorrelation (Petersen 2009). Industry fixed effects and year fixed effects are not reported for the sake of brevity.

presents the results of a regression analysis exploring the relationship between ES and EM, as well as various control variables. Model 3 uses discretionary total accruals derived from the Modified Jones Model as a proxy for EM. To provide further robustness to the results, we performed an additional analysis using Kothari et al.’s (2005) measure of EM, Model 4. The findings show that both models reveal a positive and statistically significant relationship between EM and ES, indicating that an increase in EM is associated with an increase in ES. This suggests that firms with higher EM tend to have more sustainable earnings. Based on these findings, Hypothesis 2 is supported, indicating that EM plays a crucial role in maintaining sustainable earnings with a positive impact. Additionally, the revenue growth rate variable has a positive and significant relationship with ES, while the ROA has a negative and significant relationship with the ES in both models. The relationship between the ROA and ES is not as simple as commonly believed. While it is widely believed that there is a positive correlation, various other factors suggest otherwise.

The concept of “Mean Reversion in Profitability” is a well-documented phenomenon in the finance literature showing that high-profit companies may experience a decline in profitability over time, while low-profit companies may see an increase due to market pressures and self-correction (Davis et al. 2000). Therefore, a firm with a high ROA may not be sustainable because it is more likely to revert to the mean. Moreover, competition and market saturation can negatively affect companies with a high ROA, leading to a decrease in their ROA and a lower ES (Porter 1979). Additionally, a high ROA may encourage firms to take on more risks to maintain or enhance profitability, leading to increased earnings volatility. Finally, companies that achieve a high ROA may prioritize short-term profitability over long-term sustainability, leading to short-termism among managers and actions that could compromise their long-term success.

The relationship between Firm Size and sustainability, as well as the relationship between the Distressed variable (D/E) and ES, are inconclusive based on the presented models. Furthermore, the interaction variables between the MJM, ROA, Firm Size, and Distressed Firms did not significantly affect the relationship between the ES and these variables. The results on the relationship between financial distress, EM, and ES indicate that Hypothesis 3 cannot be substantiated. Finally, both models show a highly significant F-statistic, indicating that the overall models are statistically significant and have explanatory power.

5.3. EM across Industries

To test our fourth hypothesis, which posits that the level of EM and its impact on EQ and ES vary across industries, we further classify our sample firms according to the Thomson Reuters Business Classification (TRBC). We hypothesize that firms in certain industries may be more inclined to engage in EM as a result of various factors such as economic and financial conditions and high regulation. Moreover, because firms in the same industry face similar pressures and challenges and are subject to comparable idiosyncratic shocks, they are more likely to adopt similar accounting practices. Therefore, such firms may share similar incentives to manage earnings.

Table 4. Sample distribution by TRBC industry.

TRBC	Description	Frequency	%	Cumulative %	Firm-Year Observations
50	Energy	36	6.56	6.56	432
51	Basic materials	98	17.85	24.41	1176
52	Industrials	75	13.66	38.07	900
53	Consumer cyclicals	67	12.20	50.27	804
54	Consumer non-cyclicals	85	15.48	65.76	1020
56	Healthcare	26	4.74	70.49	312
57	Technology	39	7.10	77.60	468
59	Utilities	25	4.55	82.15	300
60	Real estate	86	15.66	97.81	1032
63	Academic and educational services	12	2.19	100.00	144
	Total	549	100.00	100.00	6588

presents the distribution of firms by industry according to a two-digit TRBC classification. Overall, ten industry sectors are presented in the sample of firms.

The distributions presented in Table 4 indicate that the highest concentration of firms is in the basic materials sector (TRBC 51), constituting 17.85% of the total sample. The energy industry sector (TRBC 50) comprises 432 firm-year observations from industries such as coal, oil, and gas. The highest concentration sector, basic materials (TRBC 51), consists of 1176 firm-year observations from industries such as metals and mining, chemicals, and construction materials.

In testing our fourth hypothesis, ANOVA and Kruskal–Wallis tests were employed to see if there are significant differences in EM practices across ten industries.

Table 5. Industry earnings management univariate analysis.

		Observations	Parametric (ANOVA) Test		Non-Parametric (Kruskal–Wallis) test
TRBC	Industry	N	Mean	Std. Deviation	Rank Sum	Rank Mean
50	Energy	284	−0.0163	0.0611	504,612	1776.80
51	Basic materials	815	−0.0004	0.0650	1,620,000	1987.09
52	Industrials	539	0.0020	0.1050	1,080,000	2004.19
53	Consumer cyclicals	464	−0.0125	0.0936	860,603	1854.75
54	Consumer non-cyclicals	584	0.0058	0.0939	1,220,000	2088.88
56	Healthcare	173	0.0228	0.0768	415,277	2400.45
57	Technology	244	−0.0423	0.1011	302,873	1241.28
59	Utilities	172	−0.0049	0.0478	322,898	1877.31
60	Real estate	700	0.0184	0.0730	1,670,000	2390.72
63	Academic and educational services	52	0.0062	0.0424	110,969	2134.02
	p-value			0.0000		0.0001

suggests that there is a statistically significant difference (p < 0.01) in EM practices among the industries in both tests. This result supports Hypothesis 4 and confirms that EM varies significantly across industries in the GCC region. Table 5 shows that the mean rank (2400.45) of the healthcare industry (TRBC 56) has the highest mean ranking that corresponds to abnormal discretionary accruals, which indicates that this sector is more likely to conduct EM. This was supported by the abnormal discretionary accruals mean (0.0228) of this sector, which is the highest among the other industries. On the other hand, the table shows that the technology industry (TRBC 57) has the lowest mean rank (1241.28) which implies that this sector is less likely to conduct EM. The results reported in Table 5 support this inference where the abnormal discretionary accruals mean (−0.0423) of this sector is the lowest among other industries. These findings strongly emphasize the critical role of considering industry-specific factors when examining earnings management practices. Industries are influenced by a wide array of unique economic pressures, regulatory environments, and market conditions, all of which have a deep impact on their financial reporting behaviors. Therefore, customized assessments and regulations tailored to individual industries are important to gaining a comprehensive understanding of and effectively mitigating the risks associated with earnings management. In short, the extended discussion and analysis of Hypotheses 3 and 4 offer a deeper understanding of the intricate relationship between earnings management, financial distress, and industry-specific factors in the GCC region. These findings highlight the need for sophisticated approaches in financial regulation and corporate governance to tackle the specific challenges presented by earnings management in different scenarios.

5.4. Endogeneity Tests

The preceding empirical analyses and models consider EQ, ES, and EM exogenously determined. Nevertheless, there is a possibility that the relationship between EM and EQ and/or between EM and ES is endogenous. Although we include several control variables in our empirical models, one endogeneity concern in our analyses might be that some omitted variables drive EM, EQ, and ES. Another plausible concern might be that companies that already maintain a high level of EQ may not have the motive to manipulate their earnings and vice versa. The same plausible notion also holds for firms with smooth and sustainable earnings where they might not be inclined to manage their earnings to meet a desired earnings target and vice versa. In other words, reverse causality might be one form of endogeneity concern that we aim to address in our analyses. Accordingly, we conducted several tests to account for this possible endogeneity.

First, we attempt to rerun our main baseline regressions while using lagged values for all explanatory variables of interest, specifically our EM proxies to mitigate the concerns of reverse causality and simultaneity bias. We also adopt the recommendations of DeFond and Park (2001), Cohen et al. (2008), and Kothari et al. (2016) and include the lagged value of total discretionary accruals to control for accruals reversal in subsequent periods. This modification aims to mitigate the offsetting mechanism common to accruals where income-decreasing accruals in one period would be offset by income-increasing accruals in another, and vice versa. These modifications imposed on our baseline models yield better specified EM proxies while reducing any serial correlations concern in the models. The unreported results (for the sake of brevity) of these modifications further reinforce our findings that EM is associated with a lower EQ and a higher ES for firms in our sample.

Furthermore, we conduct additional post-estimation tests for endogeneity and over-identifying restrictions and estimate a two-stage least square (2SLS) regression. Our post-estimation endogeneity tests were performed under the null hypothesis that the dependent variable in our model and the EM proxy are exogenous. Accordingly, we conduct the Durbin–Wu–Hausman test (Hausman 1978) for endogeneity. According to this specification test, endogeneity exists in the model if the results (reported p-values) are statistically significant, implying a rejection of the null hypothesis that the dependent variable in our model and our EM proxy are exogenous. Additionally, if the null hypothesis cannot be rejected, it implies that our model is well fitted and endogeneity is not a problem. The finding of the Durbin–Wu–Hausman test (Hausman 1978) indicates that the null hypothesis is not rejected (two-tailed Durbin test p-value = 0.1319, two-tailed Wu–Hausman test p-value = 0.1327) indicating that the variables under consideration are not endogenously determined. Further details are provided in

Table 6. Tests of endogeneity.

Tests of Endogeneity and Over-Identifying Restrictions	Test Specification
Durbin (score) chi2	F = 2.27	p-value = 0.1319
Wu–Hausman	F = 2.26	p-value = 0.1327
Sargan (score) chi2	X2 = 0.45	p-value = 0.9288
Basmann (score) chi2	X2 = 0.45	p-value = 0.9294

.

To test for over-identifying restrictions in estimating our 2SLS regression to ensure that our model is correctly identified and that the instruments used are valid and can be estimated with consistency, we rely on Sargan’s (1958) and Basmann’s (1960) specification tests. These specification tests were used to examine whether certain instrumental variables used in the 2SLS estimation regression are correlated with the error term. The results for both tests indicate that the instruments used in our model are valid, and as such, our model does not entail an endogeneity problem. Further details are provided in Table 6 above.

6. Conclusions

This study thoroughly analyzes EM practices and their impact on EQ and sustainability. It sheds light on the practices employed by distressed and non-distressed companies in various industries across the GCC region. This study reveals that EM can positively and negatively affect EQ and ES. It introduces an accurate approach to measuring EQ by balancing cash flow and accruals and using operating income to gauge EQ. Financial distress is measured using the debt ratio. Robustness tests were conducted to ensure the accuracy and reliability of the results. This study provides valuable insights into the effects of EM on EQ and ES. It highlights the need for stricter accounting regulations and corporate governance policies to ensure transparency and fairness in financial reporting. This is crucial for building trust in a business community. GCC investors, regulators, and policymakers should take note of and implement ethical earnings practices for financial stability and healthy competition, benefiting all stakeholders.

This study examines the impact of EM on EQ and ES. The limited impact of the interaction between EM and financial distress on EQ and ES can be explained by the similar financial management practices among companies in the GCC, whether they are experiencing financial distress or not. This means that the way companies manage their earnings seems to have a consistent impact on EQ and ES, regardless of their financial health. In the GCC, publicly traded companies are subject to rigorous financial audits, which helps to minimize differences in financial reporting between companies facing various degrees of financial health. Additionally, the strict adherence to comprehensive financial regulations and corporate governance standards in the region guarantees a uniform level of financial reporting quality. This, in turn, plays an important role in mitigating the significant effects of EM on EQ and ES, which are primarily associated with financial distress.

The findings of this study have significant practical implications for the GCC region, specifically in terms of enhancing the overall financial stability and attractiveness of the region’s markets. First, these findings could improve the regulatory framework by emphasizing the need for strong regulatory oversight in the GCC. Regulatory bodies, such as the ESCA in the UAE and the CMA in Saudi Arabia, should continue to strengthen their oversight mechanisms. This includes enhancing transparency requirements and implementing strict penalties for non-compliance to prevent unethical EM practices that could harm the region’s economic and financial stability. Moreover, GCC regulators need to consider aligning their accounting and reporting standards more closely with international benchmarks like the IFRS. This alignment can potentially enhance the comparability and reliability of financial statements, thereby strengthening investor confidence both within the region and on an international scale.

Second, this study’s findings emphasize the critical role of corporate governance in addressing EM within GCC firms. This study implies the specific actions that GCC firms should undertake to strengthen their corporate governance practices. It is recommended that these firms focus on enhancing CG mechanisms, such as the independence and financial expertise of their boards of directors. This involves appointing independent directors with robust accounting and financial backgrounds who can provide more effective oversight of management practices. By doing so, this study suggests that the firms can expect to see improvements in EQ and ES. Moreover, this study focuses on the importance of strengthening internal controls and audit functions within companies. It stresses the need for robust measures to detect and prevent earnings manipulation. The suggestion is for firms to invest in advanced auditing technologies and provide ongoing training for their internal audit teams. This continuous improvement approach is necessary for avoiding potential manipulation techniques. In addition, this study emphasizes the significance of transparent and comprehensive disclosure practices for GCC firms. It recommends that these firms should aim to provide clear, detailed, and timely financial disclosures. It is further underlined that this transparency should extend beyond mandatory financial reports to include voluntary disclosures, which can offer deeper insights into the firm’s operations and financial health. Such comprehensive disclosure practices are expected to significantly enhance investor confidence in GCC firms.

Third, this study’s findings suggest that while EM might offer short-term benefits, it can harm long-term sustainability. GCC companies, therefore, should focus on long-term strategic planning and sustainable business practices. This shift can help build a more resilient and diversified economy that is less dependent on volatile sectors such as oil and gas. Integrating ESG criteria into business practices can enhance earnings sustainability. ESG practices attract socially responsible investors and contribute to long-term financial stability and performance. GCC firms should adopt and report on ESG initiatives to demonstrate their commitment to sustainable development. Finally, this study indicates that the impact of EM varies across industries. Policymakers in the GCC should consider developing sector-specific regulations and guidelines that address the unique challenges and risks associated with different industries. By paying attention to these practical implications, the GCC region can work towards improving the quality and sustainability of earnings among its companies. This would have the effect of boosting the region’s economic resilience and increasing its attraction to global investors.

It is important for future research to extensively analyze the comparative impact of EM across different regions, shedding light on the distinctive characteristics of each region. To strengthen the evidence, future research could develop more methods for estimating EM and include additional factors that could affect EQ and ES. In addition, there is a need to thoroughly investigate the effectiveness of recent regulatory changes in the GCC on EM practices, as well as the specific governance mechanisms that effectively limit EM. Moreover, it is essential to conduct in-depth sector-specific analyses and longitudinal studies to track changes in EM practices over time. Furthermore, exploring the role of emerging technologies such as AI and blockchain in detecting and preventing EM and understanding the influence of ESG integration and macroeconomic factors on EM would provide invaluable insights. Finally, gaining a comprehensive understanding of the behavioral and cultural factors influencing EM practices and the impact of financial crises on financial reporting quality will significantly enrich the literature, contributing to the advancement of more resilient financial systems.