Key factors affecting earnings management the case of real estate listed firms in vietnam

According to the findings, three variables consolidated financial statements, financial performance, and financial leverage have a positive effect on earnings management, whereas auditor

INTRODUCTION

Rationale of the study

Earnings management has emerged as a significant topic of research in Vietnam, particularly as the economy grapples with challenges stemming from the Covid-19 pandemic Businesses are increasingly seeking investment capital to recover, yet they struggle to attract it due to diminished production and operational efficiency In response, managers often resort to profit management techniques to present more appealing profit figures to potential investors This practice involves manipulating accounting estimates and policies to align profits with managerial objectives, ultimately distorting the true financial health of the company Such actions can mislead investors regarding the firm's performance and growth prospects, leading to substantial financial losses and undermining the transparency and stability of the stock market.

Earnings management, as defined by Schipper (1989), refers to intentional interventions in the financial reporting process aimed at securing private benefits Healy and Wahlen (1999) elaborated that profit management occurs when managers exercise discretion in financial statements and engineer transactions to alter these statements, potentially leading to misinterpretations of the company's financial health and impacting contracts linked to accounting profit targets.

Ronen and Yarri (2008) categorized earnings management into three groups based on the objectives of information disclosure These include the white profit management group, which aims to enhance the quality of financial statements, and the earnings management group, focused on improving financial statement quality.

Earnings management is prevalent among both domestic and international companies, as noted by WangJianHui (2009) This practice can ultimately diminish the quality and reliability of financial statements Accurate accounting information regarding profits is crucial for investors and creditors, as it helps them assess the strengths and weaknesses of enterprises.

A study by Nguyen et al (2020) utilizing data from the Vietnamese stock exchange and the Beneish score model found that the real estate industry exhibited the highest levels of earnings management The research examined how ownership structure influences earnings management within real estate firms in Vietnam Their findings align with previous studies (Chen et al., 2008; Nguyen et al., 2020), indicating a positive correlation between state ownership and earnings management Additionally, similar to the findings of Matteo and Francesco (2018), Nguyen et al (2020) revealed an inverse relationship between management and board ownership and earnings management.

Recent positive changes in Vietnam have created numerous investment opportunities and boosted income levels, driving a surge in both investment demand and home purchases This growing interest is fostering the development of real estate enterprises across the country.

The high profit potential in the real estate sector has prompted businesses to expand their operations, leading to a "virtual fever" in the market.

"bubble" of real estate appearing is inevitable

Real estate companies play a crucial role in stimulating the growth of several related industries, such as banking, finance, interior design, insurance, and housing production for economic investment Consequently, investors are consistently attracted to real estate stocks, seeking to capitalize on favorable opportunities to enhance their stock market portfolios.

Investment psychology in Vietnam is expected to improve as the government effectively manages the COVID-19 pandemic The high-quality and stable policy reforms implemented by the Vietnamese government are fostering a more sustainable real estate market, which will continue to attract investors.

The management of the profitability of listed real estate businesses in Vietnam is one of the major worries for cautious investors, even though the disease-related anxiety has diminished

Profit is essential for the existence and growth of a firm, serving as a key motivator for its production and commercial activities To ensure effectiveness and showcase the production management capabilities of the team, profitability must be consistently improved A profitable business indicates its ability to adapt to market dynamics, making profit the primary objective for enterprises.

Business managers can enhance profitability and attract investments through effective earnings management Various factors influence this process, with business-related elements being the most prominent To mitigate challenges in earnings management, organizations must identify and address these issues proactively.

Research on internal issues impacting the management of publicly listed real estate firms benefits not only corporate leaders but also investors interested in real estate equities Understanding these dynamics is crucial for informed investment decisions in the industry.

Analyzing the factors influencing earnings management in Vietnam's real estate sector is essential for addressing investor interests and enhancing the scientific understanding of this critical issue.

Previous studies

Saleh et al (2005) studied the impact of the Board of Directors on the management of profits of 561 companies listed in Malaysia in 2001 Using the Jones

According to the 1991 model, factors positively influencing profit management include the return on total assets (ROA) and the concurrent holding of the Chairman of the Board and CEO positions Conversely, factors negatively affecting profit management are the size of the company and the manager's ownership ratio.

A study by Alves (2012) examined the impact of ownership structure on earnings management in 34 publicly listed Portuguese companies from 2002 to 2007 The findings revealed an inverse relationship between managers' ownership and shareholders' ownership regarding earnings management Additionally, profit management was found to decrease with higher cash flow from business contracts, while it increased in the presence of greater political costs, a higher coefficient, and a larger number of Board of Directors members.

Nico Alexander and Hengky (2017) investigated the impact of earnings management on various factors, including growth, leverage, fixed asset sales, profitability, firm size, firm age, industry, audit quality, and auditor independence, using a sample of 309 data points The findings revealed that among the nine variables analyzed, only profitability had a positive influence on earnings management The other factors, such as growth, leverage, fixed asset sales, firm size, age, audit quality, auditor independence, and industry, showed minimal impact The study suggests that managers engage in earnings management primarily to secure bonuses from investors driven by enhanced profits.

Nguyen Thi Uyen Phuong (2014) conducted a study on profit management in companies that issue additional shares, analyzing data from 75 non-financial sector companies listed between 2010 and 2012 The research concluded that businesses tend to increase profits prior to launching more shares to capture investor interest, particularly from large-scale enterprises, thereby enhancing their profit adjustment levels.

(2014), board independence (61.7%), business size (15.53%), auditors (14.56%), and financial leverage (8.74%) are the characteristics that have the greatest impact on earnings management

Previous studies have introduced various models to assess corporate profit quality, with Nguyen Thi Phuong Hong (2017) proposing 11 models to identify earnings management behaviors However, many Vietnamese companies have not been publicly listed long enough for comprehensive analysis Earlier research by Pham Thi Bich Van (2012), Hoang Khanh and Tran Thi Thu Hien (2015), and Nguyen Cong Phuong (2017) utilized only a selection of these models to evaluate profit quality Additionally, most prior research, both domestic and international, has primarily relied on the Jones model for this purpose.

(1991) model As a result, this thesis draws on earlier research and employs Jones's

(1991) model to conduct the study

A study by Susanto et al (2019) analyzing data from 132 non-financial companies listed on the Indonesia Stock Exchange revealed that institutional ownership and tax aggressiveness positively and significantly influence earnings management, while factors such as firm size, director independence, audit quality, and managerial ownership showed no significant effects Conversely, Tran and Dang (2021) found that firm size, financial leverage, profitability, and audit quality do impact earnings management practices, highlighting a contrast with Susanto et al.'s findings.

The topic of studying the behavior of forest management has attracted many researchers in Vietnam such as Tran and Dinh (2017), Bui and Ngo (2017), Ngo

In the context of global integration, Vietnam has been significantly impacted by the Covid-19 pandemic, yet there is a lack of local studies addressing this issue Research by Lassoued and Khanchel (2021) examined the effects of the pandemic on forest management practices, analyzing 2,031 observations across 15 European countries Additionally, Hsu and Yang (2022) investigated how Covid-19 influenced the quality of financial statements and affected profit management behaviors among UK-listed companies.

Research objectives and research questions

This study aims to identify factors influencing earnings management in publicly traded real estate companies, examine earnings management in the context of Vietnam, and propose recommendations to mitigate it

The following individual research objectives are described in order to fulfill the aforementioned overall goal:

• Determine earnings management evidence for real estate companies listed on the Vietnam stock exchange

• Determine the factors that influence earnings management

• Make some suggestions for minimizing earnings management for publicly listed real estate corporations.

Research subject and scope of study

Research subject: The subjects of this study are factors affecting earnings management in the case of real estate-listed firms in Vietnam

Scope of content: The research examines five factors which are: (1) Consolidated financial statements, (2) independent auditor, (3) financial performance,

(4) firm size and (5) financial leverage Due to numerous challenges in data collection, factors are assessed using information from financial statements, and discretionary accruals measure the quality of earnings (Jones, 1991)

Scope of time: The data was collected from 2018 - 2022

This study analyzes 102 selected real estate companies listed on the Vietnam Stock Exchange, focusing on the challenges of data collection and the assessment of earnings quality through discretionary accruals, as established by Jones (1991) Utilizing FiinPro, a reputable database widely used by domestic researchers, the research highlights the significant role of real estate enterprises in fostering growth across various ancillary industries, such as banking, finance, interior design, insurance, and housing production Understanding these dynamics offers valuable insights into the economic impact of the real estate sector.

Research methodology

This thesis integrates desk research with a quantitative approach to explore earnings management and its influence on economic decision-making Initially, the author conducts desk research utilizing secondary sources, including scientific themes, journal articles from both domestic and international organizations, and organizational profiles Additionally, it references empirical research models and findings from previous studies The second method employed to identify factors influencing earnings management is a quantitative analysis.

Research structure

The thesis consists of five chapters in addition to a table of contents, an acronyms list, a list of tables and figures, an abstract, references, and an appendix:

The first chapter encompasses the study's justification, related research, objectives, research questions, topic scope, methodology, and structure It also provides a comprehensive summary of existing research on earnings management.

Chapter 2: Literature review on factors affecting earnings management

The second chapter provides a comprehensive analysis of the theoretical foundations of earnings management It explores research hypotheses related to earnings management, its objectives, global research frameworks, and factors influencing earnings management, all grounded in a thorough literature review.

Chapter three outlines the study model and hypotheses derived from the previous literature review, which also informs the sampling and data collection methods Additionally, this chapter presents the findings from the data analysis.

Chapter 4: Research results and discussion

In the fourth chapter, descriptive statistics are introduced, followed by an analysis of the correlation matrix Key findings from the regression results are then emphasized, along with their justifications and interpretations.

The final chapter conveys conclusions and suggestions for key players, publicly traded companies, and the State Agency Additionally, limitations and suggestions for further research are also provided.

Literature review on factors affecting earnings management

Concept of earnings management

Earnings management is a critical factor economists consider when assessing the transparency of information presented in a company's annual newsletter, as it is a prominent topic in the financial sector In today's digital age, where the Internet plays a crucial role, the accuracy of this information significantly influences consumers' investment decisions This article aims to present definitions from renowned researchers in the field to provide a deeper understanding of earnings management.

According to Ronen and Yaari (2008), Schipper (1989) and Healy and Wahlen (1999) provide definitions that exemplify negative earnings management They argue that Healy and Wahlen's perspective most effectively illustrates the manipulation of financial findings Healy and Wahlen (1999) define earnings management as the use of judgment in financial reporting and transaction structuring to mislead stakeholders regarding a company's economic performance or to influence contractual outcomes reliant on reported accounting figures.

Earnings management occurs when managers exercise discretion in financial reporting to influence stakeholders, often adjusting account values to meet contractual obligations based on accounting information (Healy & Wahlen, 1999) A recent study by Halaoua, Hamdi, & Mejri (2017) indicates that companies in France and Britain engage in profit management strategies to avert losses and declines in earnings.

In earnings, Ronen and Yaari (2008) classify earnings management into three distinct groups:

White – White earnings management (beneficial) enhances the transparency of reports

Gray – Managing reports within the boundaries of compliance with bright line standards (gray), which could be either opportunistic or efficiency-enhancing (Ronen

Black – Black earnings management involves absolute misrepresentation It assumes practices intended to misrepresent or reduce transparency in financial statements

In summary, the definitions discussed align with established guidelines, yet there is no universally accepted definition of account manipulation in accounting literature Instead, concepts such as earnings management and creative accounting are utilized to address similar issues Within the scope of this thesis, earnings management refers to the practice of adjusting profits within the accounting framework to meet management goals, achieved through a flexible interpretation of accounting rules to fulfill organizational objectives.

Earnings management aims to achieve targeted profits and is influenced by various incentives at different stages of a business In Vietnam, this practice is commonly employed to manage contracts, salaries, and bonuses, balance profits across accounting periods, avoid violations of credit contracts, attract investments, and facilitate bond issuance Publicly available stocks meet the expectations of market analysts and adapt to the needs of change managers.

To maintain stable stock prices and enhance market value, enterprises aim to minimize profit fluctuations between accounting periods High volatility increases investment risk, leading to a decline in share prices compared to companies with consistent earnings Consequently, managers of publicly traded firms often adjust profits to achieve stability, ensuring sustainable long-term profitability.

Burgstahler and Dichev (1997) found that managers often engage in earnings management to avoid reporting losses and to reduce profits Similarly, Nelson et al (2002) identified various motivations for earnings management, such as aligning with market analysts' forecasts, impacting stock prices, and meeting management compensation targets Additionally, Athanasakou et al (2009) analyzed data from the UK from 1994 onwards, further supporting these findings.

In 2002, research indicated that UK-listed companies engage in earnings management to meet market analyst expectations and prevent loss declarations Similarly, Peek et al (2004) demonstrated that Dutch corporations utilize earnings management strategies to present favorable financial statements and satisfy shareholder expectations.

Earnings management, as outlined by Cohen and Zarowin (2010), is categorized into two main types: manipulating profits through actual transactions and managing earnings via accrual accounting variables A key metric that draws investor interest is the profitability index, which indicates a company's performance and growth potential Investors are inclined to support firms with strong growth prospects and high economic efficiency, prompting publicly traded companies to often inflate their financial results during crucial periods Consequently, corporate leaders consistently seek innovative strategies to shift revenues across fiscal years through earnings management.

Managers often utilize earnings management strategies to navigate challenges related to loan agreements A study by Prevost et al (2008) on US corporations from 1993 to 2005 found that profits are strategically adjusted to avoid breaching loan covenants The authors identified two primary methods for profit adjustment: shifting profits from the current year to future years and bringing future profits into the current year Additionally, research by Perez and Hemmen (2010) on Spanish companies revealed a correlation between liabilities and earnings management, indicating that higher liabilities increase the incentive for executives to manipulate financial results.

Stock market experts express concerns regarding a company's production and business activities, primarily assessed through revenue and net profit metrics Consequently, managers face pressure to align with analyst expectations, often leading to profit adjustments aimed at enhancing performance for the reporting period.

Identifying earnings management can be achieved through various methods, as empirical studies reveal Managers often manipulate earnings by changing accounting methods, engaging in real transactions, and utilizing total or discretionary accruals Additionally, specific accruals and strategies for income distribution and smoothing play significant roles in this process.

Numerous studies conducted in the 1970s and early 1980s concluded that managers can exercise discretion by selecting the accounting procedures or policies they wish to use

Managers tend to choose accounting policies that minimize tax payments, secure favorable regulations, and enhance profitability, as noted by Watts and Zimmerman (1978) Their positive accounting theory suggests that these choices are driven by the desire to maximize personal wealth Additionally, Hagerman and Zmijewski (1979) highlight that decisions related to inventory methods, depreciation, investment tax credits, and pension cost amortization are significantly influenced by the presence of incentive compensation plans.

Managers can influence earnings through real transactions and accounting adjustments For instance, they may boost sales by offering price discounts or extending credit terms, leading to higher earnings in the current period Additionally, increasing production allows managers to distribute fixed overhead costs over more units, effectively lowering the fixed cost per unit and enhancing operating margins by reducing the cost of goods sold Other strategies include selling fixed assets and cutting R&D expenses Schipper (1989) introduced the concept of real earnings management, suggesting that managers can time investment or financing decisions to manipulate reported earnings.

The distribution approach to identifying earnings management is a relatively new concept in financial literature Research by Burgstahler and Dichev (1997) analyzed the revenue distribution patterns influenced by management, revealing that managers often strive to meet earnings benchmarks This results in a scarcity of observations just below the earnings threshold and an increase just above it, indicating a discontinuity known as the earnings kink, which serves as evidence of earnings manipulation According to Burgstahler and Dichev (1997) and Degeorge et al (1999), managers are typically concerned about three psychological barriers: achieving positive earnings, matching the previous year's earnings, and meeting analysts' consensus predictions.

Some theories relating to earnings management

Signaling theory explores how organizations navigate information asymmetry with external stakeholders by employing signals to bridge knowledge gaps (Steigenberger & Wilhelm, 2018; Mavlanova, Benbunan-Fich, & Koufaris, 2012) In uncertain situations or when information is lacking, companies utilize signals to convey important information and reduce disparities These signals often manifest as actions that communicate unobservable qualities, such as a job candidate showcasing a diploma from a prestigious university to demonstrate their competence (Bergh, Connelly, Ketchen, & Shannon, 2014).

The intuitive nature of signaling theory enhances its widespread application As noted by Spence (2002), signals serve as observable mechanisms for interpersonal interaction Espeland and Hirsch (1990) emphasize that signals can manifest as written words and numbers, as well as in more abstract symbolic forms They further argue that management and companies manipulate various financial metrics—such as earnings, growth, profits, and debt—demonstrating how these figures serve as symbolic authority in attracting investor interest through accounting technology.

Financial economists have developed various models to demonstrate the connections between business debt and dividends as indicators of a company's quality For example, Ross (1973) and Bhattacharya (1979) suggest that only high-quality firms can maintain long-term dividends and interest payments, while low-quality companies struggle to meet these financial obligations Consequently, these signals influence how external parties, such as lenders and investors, assess a company's quality This foundational research has led to the emergence of key concepts and frameworks within signaling theory, as highlighted by Riley (2001).

The agency model, as discussed in management and economics literature, highlights the challenges that arise from the separation of ownership and management in businesses (Daily, Dalton, & Rajagopalan, 2003; Wasserman, 2006) Agency theory focuses on the governance techniques necessary to regulate agent behavior in jointly owned firms, particularly in the context of the modern corporation in the USA, where ownership is dispersed and control is separated (Berle and Means, 1932) Shareholders, as principals, delegate authority to managers, the agents, to operate the business, raising concerns about whether these managers prioritize the interests of the owners or their own (Jensen & Meckling, 1976; Ross, 1973).

Although agency theory is widely accepted and efficient, it still has a number of shortcomings, which have been noted by numerous authors including Eisenhardt

The theory proposed by Shleifer and Vishny (1997) suggests that an unpredictable future and contractual agreements between principals and agents can address the agency problem However, practical challenges such as knowledge asymmetry, rationality issues, fraud, and transaction costs often hinder this process Shareholders primarily aim for the highest possible returns, yet their involvement in company operations is limited Meanwhile, directors are primarily responsible for overseeing managers, with their additional duties remaining unclear.

In the 1970s and 1980s, the economic theory of asymmetric information emerged as a significant explanation for market failures, highlighting how discrepancies in information between buyers and sellers can disrupt market efficiency Market failure is characterized by an ineffective distribution of goods and services in a free market, where prices are influenced by supply and demand dynamics.

Michael Spence's 1973 study "Job Market Signaling" highlights the uncertainties organizations face when hiring new workers, viewing candidates as risky investments due to knowledge asymmetries He compares the hiring process to a lottery, where employers are unsure of a candidate's true productivity Meanwhile, Joseph Stiglitz popularized the concept of information asymmetry, contributing significantly to market screening theory and examining its implications, particularly in insurance markets His research illustrates how asymmetric information can lead to adverse selection, negatively impacting market dynamics in general equilibrium models.

George Akerlof, along with Michael Spence and Joseph Stiglitz, played a crucial role in developing the concept of asymmetric information in economics In his landmark 1970 study, "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism," Akerlof highlighted how automobile buyers possess different information compared to sellers, which incentivizes sellers to offer low-quality cars, or "lemons," without adjusting prices He argued that buyers often struggle to differentiate between high-quality vehicles and lemons, leading to a market where sellers of quality cars cannot achieve fair prices.

Many successful businesses, such as Johnson & Johnson, Google, and AES, exemplify stakeholder theory by prioritizing values and stakeholder relationships over mere profit These companies, highlighted in "Built to Last" and "Good to Great," demonstrate that long-term success is rooted in understanding and applying the principles of stakeholder theory They address the moral dimensions of management by recognizing the importance of purpose and human connections, showing that profitability can coexist with a commitment to broader stakeholder interests.

Stakeholder theory disproves the separation thesis by starting with the premise that values are inescapably and transparently part of conducting business (Freeman

The separation thesis posits that ethics and economics can be distinctly separated, complicating the promotion of business ethics and the moral functioning of industries Proponents of shareholder primacy often emphasize financial outcomes while neglecting ethical implications and values This narrow focus results in a limited understanding that fails to encompass the complexities of human endeavors.

Management often benefits from issuing additional shares at elevated prices while aligning with investor and market analyst expectations through earnings management When managers fail to meet profit expectations, investors are inclined to buy or sell shares prematurely, leading to potential costs for management due to early transaction cancellations Consequently, this risk motivates managers to actively engage in earnings management practices.

Models to identify earnings management

In 1985, Healy introduced the first model for evaluating earnings management behavior by analyzing data from 250 of the largest industrial companies in the United States, as listed in Fortune Magazine in 1980 The study encompassed 1,527 year-over-year observations, including 94 companies that fully disclosed their bonus plans but not their long-term incentive plans The findings indicate a tendency to use accruals to enhance benefits, although Healy's methodology focuses solely on total accruals across business groupings without making direct comparisons of accurately measured data.

Healy's model simulates the earnings-management behavior of a single manager through a bonus formula with fixed parameters in a two-period framework It assumes that management sets targets and observes income prior to discretionary accruals, allowing them to make income-increasing or income-decreasing discretionary accruals based on their incentives Discretionary accruals in the Healy model are calculated accordingly.

NDAit is the non-discretionary accruals in year t of company i

TAit is the total accruals of year t of company i

A it-1 is the total assets of year t-1 of company i

The model indicates that when management does not manipulate earnings, discretionary accruals (DA) are zero, and total accruals (TA) equate to non-discretionary accruals (NDA) However, when managers alter earnings for a specific year, a certain level of DA arises These modifications are typically temporary, as managers select the accounting methods they wish to employ Consequently, if earnings are artificially inflated in the short term, it is inevitable that future profits will decline This is exemplified by the choice of depreciation method, as outlined in Clauses 3 and 4, Article 13 of Circular 45/2013/TT-BTC dated April 25.

In 2013, companies gained the flexibility to select a depreciation method for each fixed asset, allowing for a one-time change in method during the asset's useful life This change impacts costs and, consequently, profits, emphasizing the need for managers to adhere to the principle of consistency rather than making frequent alterations Furthermore, the depreciation method can only be modified once, ensuring that the average depreciation amount (NDA) remains stable over time, reflecting that it does not fluctuate between accounting periods.

This model's computational simplicity is an advantage This approach, however, is constrained as it counts on the NDA remaining constant throughout time

The Non-Disclosure Agreement (NDA) may fluctuate based on the level of business activity, adjusting accordingly as activity changes However, this method does not consider the volatility caused by factors such as growth in activity, business expansion, and economic shifts Ignoring these variables can lead to inaccuracies in the calculated NDA, ultimately resulting in erroneous outcomes for the Determined Amount (DA).

The second model discovered by DeAngelo (1986) is as follows:

DAit is the discretionary accruals of company i in year t

TAit is the total accruals of company i in year t

TA it-1 is the total accruals of company i in year t-1

A it-1 is the total assets of company i in year t-1

DeAngelo aligns with Healy's perspective, suggesting that the changes in accruals between two accounting periods reflect discretionary accruals (DA) Specifically, DA represents the variation in total accruals (TA) from year t to year t-1 Consequently, the non-discretionary accruals (NDA) are derived from the total accruals of the previous year, expressed as NDAit = TAit-1.

DeAngelo's model can be considered a specific instance of Healy's model, with the primary difference being that DeAngelo computes the Non-Decreasing Assets (NDA) based on the preceding year rather than several years ago Both models estimate the NDA section of the study period using Total Assets (TA) from either the prior year or earlier years DeAngelo's model provides an accurate NDA estimate when the NDA remains constant over time and the mean Depreciation Assets (DA) is zero during the estimation period However, NDA levels often fluctuate with the company's business activity, leading to variations in NDA percentages from year to year, particularly for companies in growth phases Consequently, both Healy and DeAngelo's models may miscalculate NDA levels in such dynamic scenarios.

To address the limitations of previous models, Jones focused on managing the impact of business activity fluctuations on the Non-Discretionary Accruals (NDA) According to Jones (1991), the NDA is influenced by revenue and the historical cost of fixed assets By analyzing changes in net revenue and historical costs from year (t-1) to year t, Jones developed a predictive approach for the NDA component This model effectively challenges the stability assumption of the NDA proposed by Healy (1985) and DeAngelo (1986), providing a more dynamic framework for understanding accruals.

𝑇𝐴𝑖𝑡 is the total accruals of company i in year t

𝐴𝑖𝑡−1 is the beginning total assets of company i in year t

∆𝑅𝐸𝑉𝑖𝑡 is the change in revenue for company i from year t-1 to t

𝑃𝑃𝐸𝑖𝑡 is the property, plant and equipment of company i in year t

Dechow et al (1995) identified the Jones model as the most effective tool for detecting earnings management by managers However, Aljifri (2007) criticized the Jones model for excluding revenue adjustments, as it treats revenue as non-discretionary In reality, managers often manipulate earnings by accelerating or delaying economic transactions to recognize revenue within specific accounting periods.

2.3.4 Model of Dechow et al (1995) – Modified Jones

Dechow et al (1995) enhanced Jones' (1991) model by removing the growth in credit sales during periods of identified data manipulation and incorporating additional variables, such as the change in accounts receivable (ΔREC) This adjustment aims to mitigate the impact of accrual sales linked to increases in accounts receivable, thereby providing a more precise representation of the enterprise's business environment within the fiscal year The revised model introduced by Jones in 1995 reflects these improvements.

𝑇𝐴𝑖𝑡 is the total accruals of company i in year t

𝐴𝑖𝑡−1 is the beginning total assets of year t

The change in revenue for company i from year t-1 to t is represented by ∆𝑅𝐸𝑉𝑖𝑡, while the change in receivables for the same company during that period is indicated by ∆𝑅𝐸𝐶𝑖𝑡 Additionally, the property, plant, and equipment of company i in year t is denoted as 𝑃𝑃𝐸𝑖𝑡.

Numerous studies have evaluated the effectiveness of earnings management estimation models, notably those by Subramanyam (2007) and Bartov et al (2002), who analyzed the Jones (1991) and Modified Jones (1995) models using both time series and cross-sectional data Their findings indicate that these models demonstrate superior explanatory power when applied to cross-sectional data compared to temporal data Essentially, the Jones and Modified Jones models, when utilized with cross-sectional data, effectively highlight the impact of industry characteristics and the study period.

2.3.5 Model of Kothari et al (2005)

Kothari et al (2005) expand on the model by Dechow et al (1995) by suggesting that a company's total accruals are influenced by its unusual performance To enhance the model, they incorporate the Return on Assets (ROA) variable, highlighting its significance in analyzing financial performance.

TA𝑖𝑡 is the total accruals of company i in year t

𝐴𝑖𝑡−1 is the beginning total assets of year t

∆𝑅𝐸𝑉𝑖𝑡 is the change in revenue for company i from year t-1 to t

∆𝑅𝐸𝐶𝑖𝑡 is the change in receivables for company i from year t-1 to t

𝑃𝑃𝐸𝑖𝑡 is the property, plant and equipment of company i in year t

ROAit is the return on assets of company i in year t

The model's variables align closely with those established by Dechow et al (1995), with the exception of the NDA variable, which is calculated as the profit from operating activities before changes in working capital, divided by the assets from the previous year.

Factors affecting earnings management

Various micro and macro variables have an impact on earnings management

To make informed economic decisions, stakeholders such as investors, shareholders, auditing firms, and government agencies must understand and analyze the factors influencing profit management Key factors impacting earnings management include market conditions, regulatory changes, and internal company policies.

Figure 2.1: Key Factors affecting earnings management

Consolidated financial statements, which integrate the financial data of parent and subsidiary companies, must be presented in a consolidated format It is essential to offset specific transactions, including internal transactions and internal receivables and payables, when preparing these statements.

Managers often employ acquisitions as a strategy for earnings management, as highlighted by Omar et al (2014) By purchasing another company, they secure a reliable future stream of earnings, which, when consolidated into financial statements, leads to higher reported integrated earnings.

Research indicates that employing higher-quality auditors is associated with reduced earnings management This connection, supported by the work of De Angelo (1981) and Healy and Wahlen, highlights the importance of audit quality in promoting transparency and integrity in financial reporting.

(1999) De Angelo (1981) asserts that because larger firms are thought to be more independent and qualified, their size can be employed as a proxy for the quality of their audits

The Big Four auditors may not always provide superior audit quality compared to auditors who are not a part of the Big Four, according to a number of studies (Yasar,

In the context of Korea, research by Jeong and Rho (2004) suggests that there is no significant difference in audit quality between Big Six and non-Big Six auditors Their findings support the hypothesis that auditor intervention does not influence earnings management practices To promote high-quality audits, it is essential to establish and maintain a supportive institutional environment, a sentiment echoed by Tsipouridou and Spathis.

A study conducted in 2012 revealed that, in Greece, there is no significant difference in the earnings management practices between companies audited by the Big Four auditors and those audited by non-Big Four firms.

AlKhaddash et al (2013) argue that the size of an audit firm does not impact auditor selection They highlight that multiple factors influence audit quality, suggesting that it is not solely determined by the firm's size.

Managers engage in earnings management to either mislead stakeholders regarding a company's performance or to manipulate contract outcomes based on accounting data, as outlined by Healy and Wahlen.

Earnings management can be influenced by a company's auditor, as highlighted by Memis and Cetenak (2012) When subjected to a higher-quality audit, companies are likely to adopt a more conservative stance in their financial reporting.

Profitability plays a significant role in earnings management, as highlighted by Iatridis and Kadorinis (2009), who found that organizations with low profitability can effectively manipulate profits to improve their financial appearance Additionally, Sun and Rath (2008) identify profitability and firm size as the primary factors affecting earnings management, noting that lower profit levels provide greater flexibility for altering earnings.

Burgstahler and Dichev (1997) highlighted that companies often manipulate profits to avoid reporting losses, with 8% to 12% of firms with low earnings making upward adjustments Furthermore, 30% to 44% of companies experiencing losses engage in earnings manipulation Supporting this notion, Degeorge et al (1999) noted that achieving a positive profit indicator is a critical initial step for businesses employing earnings management strategies.

The fragmentation of ownership in large firms has led to directors holding significant executive power, as they are appointed by shareholders to manage the business This separation of ownership and management becomes more pronounced with larger companies, raising concerns about directors prioritizing their own interests over those of shareholders Potential conflicts arise when managers engage in practices such as selecting accounting policies to meet profit targets tied to their compensation, manipulating earnings through accruals, or influencing actual transactions to enhance reported performance.

Research by Bushman et al (2004) indicates that firm size significantly influences the tendency and extent of earnings management within enterprises Additionally, studies by Lobo and Zhou (2006) and Shen and Chih (2007) reveal that larger companies possess both the opportunity and incentive to manipulate profits, often aiming to stabilize or inflate their earnings The inherent flexibility in business operations, coupled with the challenges faced by information users, makes such manipulations difficult to identify.

Businesses can enhance their profits by effectively managing fixed assets Under Vietnamese accounting standards, enterprises cannot revalue assets while they are in use However, if these assets are utilized to contribute to working capital and are valued at market price, they can provide a substantial source of profit Additionally, liquidating valuable fixed assets presents another straightforward method to boost profitability.

Accountants in large enterprises, particularly publicly listed companies, typically possess specific qualifications that enable them to comprehend accounting policies and regimes This expertise allows managers to engage in earnings management, supported by accountants' knowledge and the use of accounting software, which facilitates the creation of various earnings "scenarios" with relative ease.

Research hypothesis

Five hypotheses are provided to evaluate the effect of factors on profits management in the case of real estate-listed corporations in Vietnam

H1: Consolidated financial statements have a positive relationship with earnings management

Consolidated financial statements provide a comprehensive view of the economic and financial health of a parent company and its subsidiaries, presenting them as a single economic entity This approach allows users to access more accurate and insightful information about the overall business than they would obtain from examining only the parent company's financial documents.

(2018), when researching the factors affecting the earnings management of companies listed on the stock exchange, proved that consolidated financial statements have a positive relationship with earnings management

According to Article 5 of Circular 202/2014/TT-BTC, the parent company is responsible for preparing the consolidated financial statements for the entire group at the end of the accounting period This requirement applies particularly to publicly listed companies, large public enterprises, and state-owned parent companies, which must produce annual consolidated financial statements, semi-annual consolidated financial statements in full form, and quarterly consolidated financial statements, with the option to prepare full quarterly statements if necessary.

H2: Auditor size has a negative relationship with earnings management

Agency theory posits that principals must implement controls to mitigate conflicts of interest and information asymmetry with their agents Employers bear the costs of agency expenses, including audits, which serve as a key oversight mechanism Consequently, the role of independent auditors is crucial in enhancing the accuracy and fairness of financial reporting.

Kinney and Martin (1994) analyzed nine studies and discovered a positive correlation between audit activities and the net earnings and assets of audited entities Users of financial statements exhibit greater trust in audited reports compared to unaudited ones As a result, when financial statements are audited by leading firms such as the Big 4 (KPMG, EY, Deloitte, and PWC), management is inclined to minimize earnings management These large audit firms utilize rigorous auditing processes and techniques to identify significant misstatements, including fraud and errors, which enhances audit standards and fosters positive evaluations from major auditors.

In the Australian context, Gallery and Cooper (2008) discovered evidence demonstrating that the disclosure accuracy of financial statements audited by the Big

The audit quality of Big 4 firms is notably superior compared to non-Big 4 auditors, as mandated by ISA 220, which requires the establishment of rules and processes to ensure audit quality However, both auditors and financial statement users struggle to assess the actual quality of audits This evaluation often hinges on the size of the auditing firm, with a clear correlation identified between auditor size and earnings management practices.

H3: Financial performance has a positive relationship with earnings management

Investors rely on stock exchange information, including share prices and audited financial accounts, to make informed financial decisions According to Charfeddine et al (2003), poor economic performance leads to a decline in stock prices and company value To uphold a favorable reputation among stakeholders, earnings management strategies were introduced by Charfeddine et al (2013).

Change and Warfield (2005) found that stock price increases are often utilized for earnings management, benefiting the stock market Conversely, Chen et al (2006) revealed that entities with poorer financial performance tend to engage more in earnings management However, studies by Rahman and Ali (2006) in Malaysia and Alves (2012) in Portugal indicated no significant relationship between financial performance and earnings management Various metrics such as ROA, ROE, ROS, revenue growth, cash flow, and stock fluctuations are commonly employed to assess financial performance.

H4: Firm size has a positive relationship with earnings management

Firm size significantly impacts how users interpret financial statements, as larger companies often experience a greater separation between ownership and management This separation raises concerns about the accuracy of financial statements, which may be tailored more for management than for external users Additionally, larger firms face heightened profit expectations, leading to pressure to present inflated financial performance to please analysts Research by Barto and Simko (2002) indicates that these companies may engage in earnings management to mislead investors regarding projected profits Furthermore, the sheer volume of economic transactions conducted by larger firms means that changes in accounting practices can substantially influence reported earnings Evidence from Myers and Skinner (2000) supports the notion that large corporations tend to overstate their earnings, highlighting firm size as a critical factor in earnings management.

H5: Financial leverage has a positive relationship with earnings management

Borrowings, alongside capital from shareholders, are essential for funding business operations and reducing agency costs associated with management and shareholders (Jensen and Meckling, 1976) The liability-to-owner's equity ratio serves as a key indicator of financial structure and helps assess financial leverage, demonstrating that borrowings increase asset size (Charfeddine et al., 2013; Alsharairi and Salama, 2012) Additionally, the ratio of liabilities to total assets reflects an entity's financial independence (Shen and Chih, 2007; Fathi, 2003) According to Watts and Zimmerman (1990), funding decisions can signal corporate governance quality to external investors, allowing firms with high liability proportions to choose accounting policies that adjust profits, thereby mitigating the risks of excessive financial leverage.

METHODOLOGY

Research process

The research process begins with defining the subject and objectives, followed by compiling existing studies to establish a robust theoretical framework Next, suitable models and hypotheses are chosen, and data is collected, with key characteristics analyzed through descriptive statistics The results of the estimated regression model are then presented, offering insights and recommendations for decision-makers, businesses, and financial statement users A visual summary of this research process is provided in Figure 3.1.

Research methodology

The thesis combines the use of many research methods, including the following two methods:

Desk research involves utilizing existing data that has already been collected, summarized, and published This type of research, also known as secondary research,

This thesis undertakes a comprehensive examination of earnings management and its impact on economic decision-making It begins by identifying the research subject and objectives, followed by a thorough review of prior literature The research methodology includes collecting data through online sources, public libraries, and existing surveys, as well as utilizing data from governmental and non-governmental organizations Descriptive statistics are performed to analyze the data, and a regression model is estimated and tested to validate the findings Ultimately, the study formulates research hypotheses and models based on the insights gathered from secondary materials.

Quantitative research plays a vital role in identifying factors influencing earnings management by employing statistical analysis and multivariable regression models This approach relies on data extracted from management reports, financial reports, and annual reports of publicly traded companies The quantitative research process involves two key steps: data gathering and the application of regression models to analyze the collected data effectively.

Research model

Hypotheses are tested by utilizing the model following NH Dang, TVH Hoang and MD Tran (2017):

𝐷𝐴 𝑖𝑡 is the discretionary accruals of firm i for year t, measured by equations 1 and 2 as below

CONS is the consolidated financial statement

AUDIT is the auditor size

ROA is the financial performance

SIZE is the firm size

LV is the financial leverage

This study will explore the correlation between discretionary accruals and various influencing factors to validate the proposed hypotheses According to Chen, Hope, Li, and Wang (2011), the Jones model (1991) and its subsequent adjustments are the most effective frameworks for analyzing earnings management behavior The Jones model is favored in accrual-based earnings management research for several reasons: it is widely adopted in academic studies, all newer models are derived from it, and while it has certain limitations, it remains a benchmark for comparison Therefore, the Jones model (1991), which comprises two equations, will be utilized to accurately measure discretionary accruals.

𝑇𝐴 𝑖𝑡 is the total accruals of firm i for year of t, calculated by the equation based on cash basis accounting, as follows:

𝑇𝐴 𝑖𝑡 = Net profit – Cash flow from operating activities

𝐴 𝑖𝑡 −1 is the beginning total assets of year t;

∆𝑅𝐸𝑉𝑖𝑡 is the change in revenue for firm i for year t-1 to t;

𝑃𝑃𝐸𝑖𝑡 is the property, plant and equipment of firm i for year t

From equation 1, coefficients 1, 2 and 3 are obtained to estimate DA (discretionary accrual) by using equation 2 as follows:

Table 3.1: Summary of variables in the model

Name of variables Type of variables

Earnings management Dependent DA Discretionary accrual (DA) is measured by the model of Jones (1991)

Independent CONS 1 if consolidated financial statements

Auditor size Independent AUDIT 1 if audited by Big 4 auditors

0 if not audited by Big 4 auditors

Financial performance Independent ROA Profit after tax/ Total assets

Firm size Independent SIZE Net profit

Financial leverage Independent LV Total liabilities/ Total assets

Research sample

Real estate plays a crucial role in the national economy, directly impacting various financial markets, including the money and labor markets As a vital sector, real estate investment and business activities significantly contribute to budget revenues and represent a substantial portion of a country's gross domestic product However, the current development of Vietnam's real estate market faces several limitations and challenges, necessitating effective solutions to promote stable and sustainable growth in the future.

The ongoing crisis between Russia and Ukraine has heightened inflation concerns, prompting individuals to turn to gold, USD, and real estate as safe havens, thereby increasing liquidity in the real estate market Notably, domestic economic growth (GDP) reached 5.03% in the first quarter of 2022, surpassing figures from the same period in 2020 and 2021 As a result, the real estate sector has become increasingly vibrant, emerging as a favored investment avenue.

Between 2015 and 2020, the real estate market experienced significant growth due to a stable low-interest rate environment and prior accumulation, which facilitated easier access to credit for home buyers and real estate businesses However, in the latter half of 2022, a sharp increase in interest rates led to a sudden drop in market liquidity and poor absorption rates Prolonged reliance on financial leverage has restricted the ability of these businesses to finance and refinance effectively.

The real estate market has experienced a silent decline in recent years, which remained unnoticed under normal business conditions However, the outbreak of the Covid-19 pandemic, coupled with rising inflation and political issues, revealed underlying problems that businesses struggled to address As a result, short-term liquidity pressures surged dramatically in the third and fourth quarters of 2022, leading to a significant downturn in the real estate market.

The Chairman of the Vietnam Real Estate Brokers Association highlights that despite current challenges, Vietnam's real estate market holds significant long-term potential.

Vietnam stands out as Asia's most dynamic and innovative market, boasting an impressive average annual growth rate of 15% The real estate sector is poised for a promising future over the next two decades, driven by ongoing urbanization, which remains relatively low but is progressing rapidly.

Vietnam's urban population currently stands at 44 million, making up 45% of the total population, with 862 urban areas By 2025, this figure is expected to rise to 52 million, representing 50% of the population and approximately 1,000 urban areas, including at least one mega-city with over 10 million residents and five cities housing 5-10 million people Projections indicate that by 2050 to 2070, the urban population could reach 70-75% Furthermore, the OECD forecasts that urbanization in Southeast Asia will exceed 90% by 2070.

The State plays a crucial role in shaping the real estate market by acting as the primary regulator of its activities To ensure the market operates effectively, the State must oversee that all real estate transactions adhere to legal standards and comply with relevant laws.

The State's influence significantly contributes to the imperfections in the real estate market, as it intervenes at various levels, particularly concerning land use to achieve collective development objectives Given the unique nature of real estate, where market information is often scarce and primary land availability is determined by government decisions, the real estate sector operates as an incompletely competitive market.

Real estate enterprises typically recognize revenue and profit later than companies in other industries This year's profits reflect sales from previous periods, leading to an unclear impact on the current profit outlook for these businesses However, the effects are expected to materialize in the following year.

The real estate industry is heavily influenced by macroeconomic factors like GDP, credit growth, and foreign investments, including Foreign Institutional Investment (FII) and Foreign Direct Investment (FDI) Currently, credit remains the primary source of capital in Vietnam's real estate market, making it a crucial driver of industry growth.

The real estate market often demands significant capital for investment projects, leading many firms to seek loans and attract investors due to insufficient funds Historically, businesses needed only to secure a project to profit during market booms However, the influx of companies into the real estate sector, drawn by substantial earnings, has resulted in an oversupply across various segments, including industrial properties, apartments, office spaces, and vacation homes Despite a decline in real estate prices, they remain excessively high compared to the actual needs of most consumers.

In challenging economic conditions, financial leverage can become a significant burden for real estate businesses When the housing market is sluggish, high interest rates make it difficult to secure loans, leading to increased pressure on companies to meet their interest and principal debt obligations As a result, these businesses struggle to generate cash flow, impacting their overall financial stability.

The real estate industry is currently experiencing a growth rate and gross profit margin that surpass the GDP growth rate and profitability of the overall Vietnamese market This trend is expected to persist in the near future, indicating that the real estate sector remains in a growth phase.

In summary, acquiring panel data from 102 real estate enterprises in Vietnam is feasible due to the industry's rapid growth, highlighting the significance of this thesis Furthermore, the choice of the research sample is shaped by time and budget limitations.

Data collection

The data collection process involves selecting real estate companies listed on the Vietnam Stock Exchange that provide comprehensive information necessary for evaluating earnings management behavior To assess the factors influencing earnings management, data is gathered from annual reports, financial statements, management reports, and the Fiinpro platform Consolidated statement variables are assigned a value of 1 for years when companies present consolidated financial statements and a value of 0 for those issuing single financial statements Additionally, auditor size is classified as 1 if the financial statements are audited by a BIG4 firm and 0 otherwise.

Data analysis method

Descriptive statistics are essential for gaining a comprehensive understanding of data, as they highlight key characteristics through measures such as frequencies, means, medians, and standard deviations This analysis distinguishes between qualitative variables, like gender and socioeconomic status, and quantitative variables, such as age and weight By examining independent and dependent variables of real estate companies listed on the Vietnam Stock Exchange from 2018 to 2022, the study calculates average, maximum, and minimum values, along with the variance range, providing valuable insights into the dataset.

A correlation matrix is utilized to identify the relationships between pairs of research variables, focusing on the connection between DA and other variables These matrices are crucial in multivariate research as they quantify the pairwise relationships among components of a random vector Their significance is particularly evident in methods like Principal Component Analysis and Factor Analysis, which produce different outcomes compared to covariance matrices Additionally, this approach assesses the independence of variables or subsets of variables (Johnson, 1998).

In this study, GLS (Generalized Least Squares) regression models were implemented using Stata 17.0 to analyze the impact of various factors on the dependent variable (DA) The analysis employed several models, including Pooled OLS, Fixed Effects Model (FEM), Random Effects Model (REM), and GLS Ordinary Least Squares (OLS) is used to estimate parameters in a linear regression model by minimizing the sum of the squared differences between observed and predicted values Pooled OLS combines cross-sectional and time series data, while the author developed this model using panel data Model comparison was conducted using the Breusch-Pagan test for Pooled OLS and REM, and the Hausman test for REM and FEM Although OLS provides unbiased estimates for the random effects model, it is ineffective due to the autocorrelation of the error component To address this issue, GLS estimation was employed to account for autocorrelation and heteroscedasticity, ensuring objective and effective estimation outcomes.

RESEARCH RESULTS AND DISCUSSION

Descriptive statistics

Descriptive statistics play a crucial role in analyzing the independent variables within a multivariate regression model focused on factors influencing earnings management This study utilizes a dataset comprising 510 firm-year observations derived from 102 publicly traded real estate companies.

2018 to 2022 Table 4.1's descriptive statistics findings reveal that in Vietnam:

DA has mean of -0.107, minimum of -4.623 and maximum of 1.834 This shows that the average discretionary accrual is -0.107, the lowest and highest value is -4.623 and 1.834 relatively

The average score for consolidated financial statements among Vietnamese listed real estate companies is 0.663, indicating that approximately 47.3% of these companies utilize consolidated financial reporting The scale ranges from 0.00 to 1.00, with 1 representing companies that have consolidated financial statements and 0 for those that do not.

The AUDIT variable has an average value of 0.278, with a minimum of 0.00 and a maximum of 1.00, indicating that approximately 44.9% of Vietnamese listed real estate companies are audited by Big4 firms.

The average ratio of ROA is 0.036, with the lowest value being -0.483 and the highest value being 0.998

SIZE has mean of 0.209, minimum of -0.890 and maximum of 3.907, which means that the average profit of Vietnamese companies is 0.209, the lowest is -0.890 and the highest is 3.907

The average total debt to total assets ratio for Vietnamese real estate companies stands at 0.518, with values ranging from a minimum of 0.001 to a maximum of 1.169.

Table 4.1: Descriptive statistics for variables included in the model

Variable Obs Mean Std dev Min Max

Correlation matrix

This test investigates the linear relationship between variables affecting earnings management and earnings management itself A multivariate regression model will be utilized, accepting factors if significant correlations are identified In cases of significant correlation, it is essential to consider potential multicollinearity issues Detailed test results are illustrated in Figure 4.2 below.

Figure 4.2: Correlation matrix for variables included in the model

| DA CONS AUDIT ROA SIZE LV

Figure 4.2 illustrates the correlation analysis of the variables, revealing that the coefficients exhibit appropriate connections Notably, most correlation coefficients among the independent variables are below 0.8, indicating that the model does not suffer from collinearity.

Checking for multicollinearity for pooled OLS model

Figure 4.3 displays the variance magnification factor (VIF) for each independent variable, indicating an average VIF coefficient of 1.20 across all variables, with none exceeding 10 According to the guidelines established by Prof Dr Nguyen Quang Dong and Prof Dr Nguyen Thi Minh in their 2013 work "Giao trinh Kinh te luong," a VIF coefficient below 10 suggests that multicollinearity is not present in the model.

Checking for the Heteroskedasticity

Heteroskedasticity occurs when the variance of errors in a linear regression model is asymmetrical and does not follow a normal distribution, violating the assumption that error variances should be consistent This inconsistency leads to unreliable model results, particularly affecting the accuracy of tests and regression coefficients To identify this issue, the author utilizes the Wald test (1943) in conjunction with the OLS model, which involves testing two hypotheses for each method employed.

The results of the model testing of the dependent variables are shown in the figure below

According to the figure, all models have a p-value of 0.0024 As a result, we discard thesis H0 and accept thesis H1 These demonstrate that all models of profitability metrics suffer from heteroskedasicity

Cameron & Trivedi's decomposition of IM-test

Check for autocorrelation

Autocorrelation can lead to inaccuracies in coefficient differences and diminish the precision of models To identify the unique effects of autocorrelation in Ordinary Least Squares (OLS) models, Wooldridge's test (Wooldridge, 1991) is employed by researchers.

H0: The model lacks first-order autocorrelation

H1: The model exhibits first-order autocorrelation

The following figure display the outcomes of the model testing of the dependent variables We accept H0 and reject H1 based on the p-value in the figure that is higher than 5%

Wooldridge test for autocorrelation in panel data

Fix model

The model's return coefficients are currently imprecise and significantly affected by infringement issues To address these shortcomings, the author employs the generalized least squares (GLS) method developed by Aitken in 1936 to replicate the random effects models (REM) This approach aims to enhance the accuracy of the model's elements, leading to more reliable results.

Regression results

The analysis of Equation 1 using Generalized Least Squares (GLS) reveals that all independent variables exhibit regression coefficients below 0.05, indicating their statistical significance and impact on the model The estimated coefficients for these independent variables are detailed in Column 2 of Figure 4.6.

Figure 4.6: Regression results of equation 1

Cross-sectional time-series FGLS regression

Correlation: common AR(1) coefficient for all panels (0.2556)

Estimated covariances = 102 Number of obs = 510 Estimated autocorrelations = 1 Number of groups = 102 Estimated coefficients = 4 Time periods = 5 Wald chi2(3) = 51.49

- TAit | Coefficient Std err z P>|z| [95% conf interval]

The full estimated equation 1 is:

𝐴 𝑖𝑡−1 Using results from equation 1, discretionary accruals (DA) are estimated by equation 2:

Discretionary accruals will be estimated for 510 firm-year observations in the collected panel data

The discretionary accrual variable (DA) is calculated using the Jones model (1991) to explore the factors affecting earnings management, with DA serving as the dependent variable in the research model.

Conduct GLS regression using the previously selected independent variables The variable DA indicates earnings management, which is influenced by independent factors including BOARD, DUAL, AUDIT, and LV The findings are illustrated in Figure 4.7.

Figure 4.7: Regression results of the model

Cross-sectional time-series FGLS regression

Correlation: common AR(1) coefficient for all panels (0.5481)

Estimated covariances = 102 Number of obs = 510 Estimated autocorrelations = 1 Number of groups = 102 Estimated coefficients = 6 Time periods = 5 Wald chi2(5) = 15.68 Prob > chi2 = 0.0078

DA | Coefficient Std err z P>|z| [95% conf interval] -+ - CONS | 0512033 01765 2.90 0.004 01661 0857965 AUDIT | 0044923 007256 0.62 0.536 -.0097292 0187138 ROA | 2091928 0804281 2.60 0.009 0515566 3668291 SIZE | -.0045012 0039209 -1.15 0.251 -.012186 0031836

Hypothesis H1 proposes that consolidated financial statements have a positive relationship with earnings management

The analysis reveals a positive correlation between the CONS and DA variables, with a CONS coefficient of 0.051, supporting hypothesis H1 (probability = 0.004, which is less than 0.05) This indicates that within a parent-subsidiary framework, earnings management is often executed through complex methods in the preparation of consolidated financial statements, aligning with findings from Omar et al (2014) and Ha and Hung.

In their 2018 study, Ha and Hung found that enterprises with complex structures, particularly those operating under a parent company-subsidiary model, are inclined towards accruals-based earnings management This tendency arises from managers' objectives and the sophisticated techniques used in preparing and presenting consolidated financial statements, which ultimately aid in profit management.

The analysis presented in Figure 4.7 indicates that Big4 audit clients exhibit higher adjusted earnings compared to others, suggesting a positive relationship between the variables DA and AUDIT However, this relationship lacks statistical significance, with a probability of 0.536, leading to the rejection of hypothesis H2 This finding is further contextualized by the observation that approximately 72.15% of the 510 companies analyzed in this study do not undergo independent audits Supporting this notion, Jeong and Rho (2004) assert that there is no significant difference in audit quality between Big Six and non-Big Six auditors in the Korean context, positing that auditor intervention does not influence earnings management practices.

High-quality audits are influenced by the institutional environment in which they operate Research by Tsipouridou and Spathis (2012) indicates that there is no significant difference in earnings management practices between companies audited by Big Four firms and those audited by non-Big Four firms in Greece Additionally, AlKhaddash et al (2013) found that the size of the auditing firm has minimal impact on an auditor's decision-making process.

Real estate companies faced significant challenges in the past year, with financial statements reflecting a struggle that cannot be easily remedied through technical accounting practices It is evident that financial revenues are insufficient to support core operations, leading businesses to confront difficult realities and accept unfavorable evaluations from audit units regarding their overall performance.

PwC has been auditing Novaland Group's financial statements since 2015, but for the first time, the firm highlighted concerns in Novaland's 2022 financial statement While PwC did not provide a definitive opinion, it emphasized that a material uncertainty could significantly impact Novaland's ability to continue as a going concern.

Independent audits appear to have little impact on earnings management in real estate companies, as these firms openly communicate past declines in business performance to shareholders With a clear understanding of ongoing market challenges, most businesses are prioritizing risk management and restructuring over growth in their future plans.

For publicly listed companies, disclosing accounting information is mandatory, and the accuracy and transparency of these reports must be validated by auditors Companies audited by the Big Four firms—KPMG, Ernst & Young, Deloitte, and PwC—benefit from a team of highly qualified and independent professionals, ensuring that the financial statements are more reliable compared to those audited by other firms.

The study's findings, which are represented in Figure 4.7's regression coefficient of 0.209, demonstrate a positive correlation between ROA variable and

DA variable with a statistically significant relationship (prob.= 0.009 < 0.05) In other words, hypothesis H3 is accepted This result agrees with those of Barto and Simko

Alves (2012) examined the relationship between ownership structure and profit management in 34 companies listed on the Portuguese exchange from 2002 to 2007, finding that financial performance does not influence earnings management Furthermore, previous studies indicate that larger companies are more likely to engage in earnings management compared to smaller firms This tendency arises as large companies strive to uphold their image and investor trust, often manipulating profits to present an inflated business scale and enhanced efficiency, thereby reassuring their investors.

The analysis depicted in Figure 4.7 indicates a negative association between firm size (SIZE) and discretionary accruals (DA), although this relationship lacks statistical significance (p = 0.251) Consequently, hypothesis H5 is rejected These findings align with similar studies conducted in Nigeria (Egbunike, 2015) and Iran (Farzaneh, 2006) While larger firms tend to demonstrate operational efficiency and enhanced risk protection, they also face challenges related to cash flow and accruals impacting profitability The size of a company can influence decision-making, as larger organizations often experience a greater separation between management and ownership, leading managers to potentially prioritize personal interests over those of shareholders Additionally, larger firms are expected to achieve higher profit targets, which complicates the role of firm size in earnings management This complexity arises from the principles of double-entry bookkeeping, where an increase in profit in one period necessitates a decrease in subsequent periods, limiting managers' ability to manipulate earnings over the long term.

Pincus and Rajgopal (2002) highlight a positive correlation between financial leverage and earnings management, indicating that higher debt levels increase the likelihood of engaging in earnings management activities Research findings, illustrated in Figure 4.7, support this hypothesis, showing that increased financial leverage correlates with heightened enterprise earnings management, as evidenced by a regression coefficient of LV = 0.051 and a significance level of prob 0.048, confirming hypothesis H6.

Financial leverage involves acquiring assets for a company through borrowed funds instead of equity As leverage increases, so does the debt ratio, which can deter investors due to perceived financial risks such as potential crises, insolvency, and bankruptcy (Andrade & Kaplan, 1998) Consequently, companies with high leverage ratios often engage in profit management to mitigate these risks.

CONCLUSIONS AND RECOMMENDATIONS

Conclusions

This experimental study analyzes 510 observations from 102 real estate companies listed on the Vietnam Stock Exchange between 2018 and 2022, utilizing panel data and independent variables such as consolidated financial statements, auditor size, financial performance, firm size, and financial leverage through generalized least squares (GLS) analysis The research finds that earnings management, measured using Jones' (1991) models, is positively correlated with consolidated financial statements, return on assets, and financial leverage The complexity of preparing consolidated financial statements aids management in profit manipulation, while more profitable companies tend to actively manage their earnings Additionally, firms with high financial leverage are more likely to engage in earnings management to mitigate financial risks However, auditor size and firm size do not significantly influence the earnings management phenomenon This study aims to enhance theoretical foundations for future research on earnings management, particularly in underexplored areas, to help prevent managerial engagement in such behaviors.

Recommendations

Organizations with complex structures, particularly those following a parent-subsidiary model, must consolidate their financial statements Research shows a positive correlation between earnings management and these types of firms Consequently, stakeholders should be more vigilant when making economic decisions.

Investors are significant consumers of data from public firms' financial statements, but they must be cautious as company managers may manipulate these numbers To navigate this landscape effectively, investors should enhance their knowledge in areas such as accounting, finance, securities analysis, portfolio management, risk management, and market analysis, particularly focusing on how listed companies manage their earnings This understanding will enable investors to anticipate risks and make informed decisions when selecting their stock portfolios Therefore, colleges and institutions should include courses on corporate governance and earnings management in their Accounting, Auditing, and Finance programs Additionally, self-education through online resources and literature is essential for investors to stay informed.

Unrealized revenue is crucial for real estate companies, as it can take the form of one-time or multiple revenue recognition due to the sector's unique characteristics Investors must understand the principles of revenue recognition and unrealized revenue as outlined in the financial statements' notes A higher unrealized revenue allows a business to lease more land, leading to significant future revenue and profit This situation can affect the company's profit value, potentially influencing management's decisions regarding earnings management to present appealing financial figures to investors.

5.2.2 Recommendations for real estate listed firms in Vietnam

Real estate-listed companies must develop criteria that balance both long- and short-term financial indicators when determining managers' salaries and bonuses It's essential to avoid the temptation of prioritizing short-term profits at the expense of long-term growth Publicly traded firms should recognize the importance of providing investors with accurate and transparent information, as this is vital for building long-term credibility and trust.

To enhance leadership and transparency in Vietnam's real estate listed companies, it is essential for the Supervisory Board and the Board of Directors to collaborate more closely This partnership will ensure adherence to accounting and corporate finance standards while emphasizing the management of earnings The Supervisory Board plays a crucial role in overseeing the Board of Directors' decisions and representing the General Meeting of Shareholders By focusing on earnings management techniques, organizations can significantly improve the transparency of their financial image.

Management awareness is crucial for minimizing earnings management, as transparency relies on business leaders valuing and respecting it Real estate companies should adopt a bottom-up approach to corporate governance, focusing on the specific activities that define their operations This strategy fosters an open mindset among executives, ultimately promoting transparent and honest disclosure of information, especially financial data, to the public.

The real estate market faces significant challenges in 2023-2024, necessitating comprehensive business restructuring Key strategies include debt restructuring, revising product portfolios, and awaiting the removal of legal bottlenecks, particularly through the updated Decree 65/2022/ND-CP.

Debt restructuring involves divesting certain projects, seeking a financially strong strategic partner, and optimizing operational costs by reducing personnel and discontinuing less profitable business segments Notably, the updated Decree 65/2022/ND-CP aims to alleviate the pressure of maturing bonds by facilitating debt rescheduling and asset swaps through negotiation.

Implications to the State Agency

The real estate market faces significant inadequacies and legal challenges that hinder its development To foster growth, the State must amend and enhance legal policies, facilitating resource mobilization for land investment to support socio-economic progress Currently, the market operates with considerable informality and uncertainty, largely due to inadequate state management Strengthening state oversight is essential for regularizing the market and enhancing its role in the country's industrialization and modernization Furthermore, while the real estate sector contributes greatly to economic growth, its rapid expansion can lead to adverse effects, such as market disruptions and inflated "virtual" prices that do not reflect actual values, ultimately impacting investor confidence.

The government must strategically balance the national financial market to direct funds into areas that align with the goals of socioeconomic development while maintaining control to prevent inflation It is essential to ensure that cash flow remains transparent and does not hinder investments in the real estate market Financial policy management should prioritize redirecting cash flows towards production factors, particularly in the creative and processing industries that possess competitive advantages Additionally, support should be extended to emerging sectors like smart cities, smart rural areas, and the circular economy to foster sustainable growth.

To promote the transparent development of capital and money markets, it is essential to enhance mechanisms and policies while addressing the activities related to real estate bond pledging This approach aims to mitigate risks and foster societal engagement through online mortgage services, online property valuation, and an official information network for the land and real estate market.

The government is seeking guidance for the real estate sector, with the Ministry of Finance proposing a draft Decree to amend certain articles of Decree 65/2022/ND Key suggestions include postponing the enforcement of regulations regarding the classification of professional securities investors and delaying mandatory credit ratings for one year Additionally, the Ministry recommends allowing enterprises to extend their deadlines by up to two years If approved, these measures could facilitate capital flow into real estate businesses starting in Q3/2023.

Limitations and future research

This study focuses exclusively on the real estate sector listed on the Vietnamese stock exchange, which may limit the applicability of the findings to other industries in Vietnam Additionally, the research identifies only five factors influencing earnings management within the real estate industry, indicating a constraint in the development of independent variables for the model.

The balanced-panel research design is used, and the DA estimation approach is the GLS regression model Additionally, the thesis only uses the model Jones

(1991) and accrual methods In future studies based on the same data set, other researchers can utilize the tests to select regression approaches that will be appropriate for the data

1 Akerlof, G.A (1970), ‘The Market for "Lemons": Quality Uncertainty and the Market Mechanism’, The Quarterly Journal of Economics, Vol 84, No 3, pp

2 Alexander, Nico & Hengky (2017), ‘Factors Affecting Earnings Management in the Indonesian Stock Exchange’ J Bank Fin Review 2 (2): 8–14

3 Aljifri, K (2007), ‘Measurement and Motivations of Earnings Management: A Critical Perspective’, Journal of Accounting - Business & Management, Vol.14, 75-95

4 Al-Khaddash, H., Al-Nawas, R & Ramadan, A (2013), ‘Factors affecting the quality of Auditing: The Case of Jordanian Commercial Banks’, International Journal of Business and Social Science, 4 (11), 206-222

5 Alves, Sandra (2012), ‘Ownership Structure and Earnings Management: Evidence from Portugal’, Australian Accounting Business and Finance Journal 6.1, 57-74

6 Athanasakou, Norman Strong & Martin Walker (2009), ‘The market reward to achieving analyst earnings expectations: Does expectations or earnings management fool investors?’, Working paper, The University of Manchester

7 Bartov, E., Dan Givoly & Carla Hayn (2002), ‘The rewards to meeting or beating earnings expectations’, Journal of Accounting and Economics, vol 33, issue 2, 173-204

8 Bergh, D D., Connelly, B L., Ketchen, D J., Jr., & Shannon, L M (2014),

‘Signaling theory and equilibrium in strategic management research: An assessment and a research agenda’, Journal of Management Studies, 51(8),

9 Berle, A & Means, G (1932), ‘The Modern Corporation and Private Property”,

Commerce Clearing House, New York

10 Bhattacharya, S (1979), ‘Imperfect Information, Dividend Policy, and “The Bird in the Hand” Fallacy”’, The Bell Journal of Economics, 10, 259-270

11 Bui, D V., & Ngo, D H (2017), ‘Board characteristics and earnings management in listed companies of Vietnam's stock market’, Science magazine

City Open University, HCM- Economics and Business Administration, 12(2),

12 Burgstahler, D., & Dichev, I (1997), ‘Earnings management to avoid earnings decreases and losses’, Journal of Accounting and Economics, Vol 24, No 1, 99-

13 Bushman, R., Chen, Q., Engel, E et al (2004), ‘Financial Accounting Information, Organizational Complexity and Corporate Governance Systems’,

Journal of Accounting and Economics, 37, 167-201

14 Collins, J (2001), ‘Good to Great: Why Some Companies Make the Leap and Others Don’t”, New York: William Collins

15 Collins, J and Porras, J.I (1994), ‘Built to Last: Successful Habits of Visionary Companies’, Harper Collins Publishing, New York

16 Daily, C.M., Dalton, D.R & Cannella, A.A (2003), ‘Corporate Governance: Decades of Dialogue and Data’, Academy of Management Review, 28, 371-382

17 Daniel A Cohen & Paul Zarowin (2010), ‘Accrual-based and real earnings management activities around seasoned equity offerings’, Journal of Accounting and Economics, vol 50, issue 1, 2-19

18 DeAngelo, L.E (1981), ‘Auditor Size and Audit Quality’, Journal of Accounting and Economics, 3, 183-199

19 DeAngelo, L.E (1986), ‘Accounting Numbers as Market Valuation Substitutes:

A Study of Management Buyouts of Public Stockholders’, The Accounting Review, Vol 61, No 3, 400-420

20 Dechow, P., Sloan, R & Sweeney, A (1995), ‘Detecting earnings management’,

21 DeFond, M L., & Jiambalvo, J (1994), ‘Debt Covenant Violation and Manipulation of Accruals’, Journal of Accounting and Economics, 17, 145-176

22 Degeorge, F., Patel, J., & Zeckhauser, R (1999), ‘Earnings Management to Exceed Thresholds’, The Journal of Business, Vol 72, No.1, 1-33

23 Eckel, N (1981), ‘The income smoothing hypothesis revisited Abacus’, 17(1), 28-40

24 Efendi, A Srivastava, E.P Swanson (2007), ‘Why do corporate managers misstate financial statements? The role of option compensation and other factors’, Journal of Financial Economics, 85 (3), pp 667-708

25 Eisenhardt, K M (1989), ‘Building theories from case study research’, Academy of Management Review, 14, 532-550

26 Freeman, R.E (1994), ‘The Politics of Stakeholder Theory’, Business Ethics Quarterly, 4, 409-421

27 Gallery, G., Cooper, E & Sweeting, J (2008), ‘Corporate disclosure quality: lessons from Australian companies on the impact of adopting International Financial Reporting Standards’, Australian Accounting Review, 18, 257-273

28 Healy P.M (1985), ‘The Effect of Bonus Schemes on Accounting Decisions’,

Journal of Accounting and Economics, Vol 7, 85-107

29 Healy, Paul M; Wahlen, James M (1999), Accounting Horizons; Sarasota Vol 13, Iss 4, 365-383

30 Hoang Khanh, Tran Thi Thu Hien (2015), ‘Detecting errors in financial statements of listed construction companies’, Journal Development Economics,

31 Hong, N.T.P (2017), ‘Analyze the current state of the level of financial management of companies listed on the stock market of Vietnam by the Jones model correction’, Economic & Development Magazine, No 245

32 Hsu Y L., Yang Y C (2022), ‘Corporate governance and financial reporting quality during the COVID-19 pandemic’, Finance Research Letters, 47 (B): 1–

33 Iatridis, G., & Kadorinis, G (2009), ‘Earnings management and firm financial motives: A financial investigation of UK listed firms’, International Review of

34 Imhoff, E.A (1977), ‘Income smoothing, a case for doubt’, Accounting Journal, 85-100

35 Jensen, MC & Meckling (1976), ‘Theory of Firm: Managerial Behaviour, Agency Costs and Ownership Structure’, Journal of Financial Economics, Vol

36 Jeong & Rho (2004), ‘Big Six auditors and audit quality: The Korean evidence’,

The International Journal of Accounting, Volume 39, Issue 2, Pages 175-196

37 Jones, J.J (1991), ‘Earnings Management during Import Relief Investigations’,

38 Kothari, S.P., Lcone, A.J., & Wasley, C.E (2005), ‘Performance-Matched Discretionary Accruals’, Journal of Accounting and Economics, Vol 39, 163-

39 Lassoued, N., & Khanchel, I (2021), ‘Impact of Covid-19 pandemic on earnings management: An evidence from financial reporting in European firms’, Global

40 Lobo & J Zhou(2006), ‘Did Conservation Financial Reporting Increase after the Sarbanes-Oxley Act’, Accounting Horizons, Vol 20, No 1, pp 57-74

41 Manh Dung Tran & Ngoc Hung Dang (2021), ‘The Impact of Ownership Structure on Earnings Management: The Case of Vietnam’, SAGE Open, vol 11(3), pages 21582440211, September

42 Matteo, P., & Francesco, P (2018), ‘Earnings Management in Developing Countries The Case of Brazilian Real Estate Industry’, Academy of Accounting and Financial Studied Journal, 22(10), 1528-2635

43 Mavlanova, T., Benbunan-Fich, R., & Koufaris, M (2012), ‘Signaling theory and information asymmetry in online commerce’, Information & Management, 49(5), 240-247

44 Memis, M U., & Cetenak, E H (2012), ‘Earnings Management, Audit Quality and Legal Environment: An International Comparison’, International Journal of

45 Michael Spence (1973), ‘Job Market Signaling’, The Quarterly Journal of Economics, Vol 87, No 3, pp 355-374

46 Mohd Saleh, N., Mohd Iskandar, T Rahmat, M.M (2005), ‘Avoidance of Reported Earnings Decreases and Losses: Evidence from Malaysia’, Malaysian

47 Nelson, M W., J, A Elliott, & R L Tarpley (2002), ‘Evidence from Auditors about Managers and Auditors Earnings Management Decisions’, The Accounting Review, Vol 77 (Sulement), 175-202

48 Ngo, D H (2019), ‘Factors affecting the profit management behavior of managers in listed companies of Vietnam's stock market’, University of Economics Ho Chi Minh City, Vietnam

49 Ngo, Diem Nhat Phuong, & Anh Thi Hong Le (2021), ‘Relationship between the audit committee and earning management in listed companies in Vietnam’,

Journal of Asian Finance Economics and Business 8: 135–42

50 Nguyen Cong Phuong (2017), ‘Applying Cross-Sectional Modified Jones: Overstating earnings before seasoned equity offerings’, Journal of Economic Development, 28(3), p.32-49

51 Nguyen Cong Phuong, Nguyen Thi Uyen Phuong (2014), ‘An investigation on earnings management of seasoned equity offerings by listed companies on Vietnam Stock Exchange’, Journal of Economic Studies 2, 91-101

52 Nguyen et al (2020), ‘Ownership structure and earnings management: empirical evidence from Vietnam real estate sector’, Real Estate Management and Valuation, 28(2), 37-51

53 Peek, Coens, Laurent & Erik (2004), ‘An Analysis of Earnings Management by European Private Firms’

54 Perez, G R., & Hemmen S (2010), ‘Debt Diversification and Earnings Management’, Journal Public Policy, Vol 29, 138-159

55 Pham Thi Bich Van (2012), ‘Model for Identifying Profit Adjustment of Enterprises Listed on Ho Chi Minh Stock Exchange’, Journal of Economic Development, No 212, pp 28-35

56 Prevost, A K., Skousen, C J., & Rao, R P (2008), ‘Earnings Management and the Cost of Debt’, Working paper, Ohio State University, Department of Finance, Stillwater

57 Riley, John, G (2001), ‘Silver Signals: Twenty-Five Years of Screening and Signaling’, Journal of Economic Literature, 39 (2): 432-478

58 Robert L Hagerman & Mark E Zmijewski (1979), ‘Some economic determinants of accounting policy choice’, Journal of Accounting and Economics, vol 1, issue 2, 141-161

59 Ronen, J., Yaari, V (2008), ‘Maximization and Minimization In: Earnings Management’, Springer Series in Accounting Scholarship, vol 3 Springer,

60 Ross, S (1973), ‘The Economic Theory of Agency: The Principal's Problem’,

American Economic Review, vol 63, issue 2, 134-39

61 Sameh Halaoua, Badreddine Hamdi & Tarek Mejri (2017), ‘Earnings management to exceed thresholds in continental and Anglo-Saxon accounting models: The British and French cases’, Research in International Business and

62 Schipper, K (1989), ‘Commentary on Earnings Management’, Accounting Horizons, 3, 91-102

63 Shen, C., & Chih, H (2007), ‘Earnings management and corporate governance in Asia’s emerging markets’, Corporate Governance: An International Review, Vol 15 (5): 999-(Iss 5), pp 999-1021

64 Shleifer, A & Vishny, R (1997), ‘A Survey of Corporate Governance’, Journal of Finance, 52, 737-783

65 Spence (2002), ‘Signaling in Retrospect and the Informational Structure of Markets’, American economic review, vol 92, no 3

66 Steigenberger, N., & Wilhelm, H (2018), ‘Extending signaling theory to rhetorical signals: Evidence from crowdfunding’, Organization Science, 29(3), 529–546

67 Sun and Rath (2008), ‘Fundamental Determinants, Opportunistic Behavior and Signaling Mechanism: An Integration of Earnings Management Perspectives’,

International Review of Business Research Papers, Vol.4 No.4, Pp 406-420

68 Susanto Y K., Pirzada K., Adrianne S (2019), ‘Is tax aggressiveness an indicator of earnings management?’, Polish Journal of management studies, Vol.20 No.2

69 Tran, T M T (2014), ‘Factors affecting earnings management of listed firms on

Ho Chi Minh Stock Exchange’, Master thesis, Ho Chi Minh University of Economics

70 Tran, T T G., & Dinh, T N (2017), ‘The presence of women on Board and in top management and earnings management behavior - Evidence from Ho Chi Minh City listed companies’, Science Journal of the Open University of Ho Chi

71 Tsipouridou, Maria & Spathis, Charalambos (2012), ‘Earnings management and the role of auditors in an unusual IFRS context: The case of Greece’, Journal of

International Accounting, Auditing and Taxation, Elsevier, vol 21(1), pages 62-

72 Wang, Z., & Williams, T H (1994), ‘Accounting Income Smoothing And Stockholder Wealth’, Journal of Applied Business Research (JABR), 10(3), 96–

73 WangJianHui (2009), ‘The advantages and disadvantages of the use of earnings management analysis’, From: www.yourpaper.net, Posted: 2009-10-06 21:50:05

74 Wasserman, Larry (2006), ‘All of Nonparametric Statistics’, New York: Springer,

75 Watts, R L & Zimmerman, J L (1990), ‘Positive Accounting Theory: A Ten- Year Perspective’, Accounting Review, Vol 65, No 1, 131-156

76 Watts, R.L & Zimmerman, J.L (1978), ‘Toward a Positive Theory of the Determination of Accounting Standards’, The Accounting Review, 53, 112-134

77 Wendy Nelson Espeland & Paul M Hirsch (1990), ‘Ownership changes, accounting practice and the redefinition of the corporation’, Accounting, organizations and society, Vol.15(1), pp.77-96

78 Xiao Chen A, Chi-Wen Jevons Lee (2008), ‘Government assisted earnings management in China’, Journal of Accounting and Public Policy, Volume 27, Issue 3, Pages 262-274

79 Yasar, A (2013), ‘Big Four Auditors’ Audit Quality and Earnings Management: Evidence from Turkish Stock Market’, International Journal of Business and Social Science, 4 (17), 153-163

Cameron & Trivedi's decomposition of IM-test

3 Check for autocorrelation Pooled OLS model

Wooldridge test for autocorrelation in panel data

Fixed-effects (within) regression Number of obs = 510 Group variable: ID Number of groups = 102

Within = 0.0422 min = 5 Between = 0.0012 avg = 5.0 Overall = 0.0087 max = 5 F(5,403) = 3.55 corr(u_i, Xb) = -0.2785 Prob > F = 0.0037

DA | Coefficient Std err t P>|t| [95% conf interval] -+ - CONS | -.1476763 1250587 -1.18 0.238 -.3935252 0981725 AUDIT | 0154643 1018623 0.15 0.879 -.1847836 2157121 ROA | 7202406 2332153 3.09 0.002 2617701 1.178711 SIZE | -.0137017 0835355 -0.16 0.870 -.1779214 1505181

LV | 470613 1867961 2.52 0.012 1033965 8378295 _cons | -.2800516 1316845 -2.13 0.034 -.538926 -.0211772 -+ - sigma_u | 33397351 sigma_e | 35276205 rho | 47266125 (fraction of variance due to u_i)

Random-effects GLS regression Number of obs = 510 Group variable: ID Number of groups = 102

Within = 0.0293 min = 5 Between = 0.2035 avg = 5.0 Overall = 0.1126 max = 5

Wald chi2(5) = 35.81 corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000

DA | Coefficient Std err z P>|z| [95% conf interval] -+ - CONS | 1456543 0601181 2.42 0.015 027825 2634836 AUDIT | 0591365 0624129 0.95 0.343 -.0631906 1814636 ROA | 7659422 2161531 3.54 0.000 34229 1.189594 SIZE | -.009519 0560819 -0.17 0.865 -.1194376 1003996

LV | 3719229 1133021 3.28 0.001 1498548 593991 _cons | -.4380463 0748153 -5.86 0.000 -.5846815 -.2914111 -+ - sigma_u | 24027925 sigma_e | 35276205 rho | 31691539 (fraction of variance due to u_i)

6 Test for suitable model between REM model and FEM model by Hausman test

| (b) (B) (b-B) sqrt(diag(V_b-V_B)) | fem rem Difference Std err

- b = Consistent under H0 and Ha; obtained from xtreg

B = Inconsistent under Ha, efficient under H0; obtained from xtreg Test of H0: Difference in coefficients not systematic chi2(5) = (b-B)'[(V_b-V_B)^(-1)](b-B)

7 Test for suitable model between POOLED OLS và REM by Lagrange multiplier test

Breusch and Pagan Lagrangian multiplier test for random effects

DA[ID,t] = Xb + u[ID] + e[ID,t]

Wooldridge test for autocorrelation in panel data

Cross-sectional time-series FGLS regression

Correlation: common AR(1) coefficient for all panels (0.5481)

Estimated covariances = 102 Number of obs = 510 Estimated autocorrelations = 1 Number of groups = 102

Estimated coefficients = 6 Time periods = 5 Wald chi2(5) = 15.68 Prob > chi2 = 0.0078

DA | Coefficient Std err z P>|z| [95% conf interval] -+ - CONS | 0512033 01765 2.90 0.004 01661 0857965 AUDIT | 0044923 007256 0.62 0.536 -.0097292 0187138 ROA | 2091928 0804281 2.60 0.009 0515566 3668291 SIZE | -.0045012 0039209 -1.15 0.251 -.012186 0031836

10 Comparison between models: POOLED OLS, FEM, REM, FGLS

DA DA DA DA - CONS 0.196*** -0.148 0.146** 0.0512*** [4.42] [-1.18] [2.42] [2.90] AUDIT 0.0538 0.0155 0.0591 0.00449 [1.09] [0.15] [0.95] [0.62] ROA 0.880*** 0.720*** 0.766*** 0.209*** [3.85] [3.09] [3.54] [2.60] SIZE -0.0185 -0.0137 -0.00952 -0.00450 [-0.40] [-0.16] [-0.17] [-1.15]