Rationale
In recent years, Vietnam has witnessed several significant accounting scandals, including those involving Truong Thanh Furniture Corporation, Vien Dong Pharmaceutical Joint Stock Company, and Bach Tuyet Cotton Joint Stock Company, underscoring the critical importance of financial reporting quality and the challenges of earnings management Notably, Truong Thanh Furniture Corporation reported a staggering VND 980 billion in non-existent inventory in its second-quarter financial statements for 2016, which led to a dramatic increase in the cost of goods sold and a reported loss of VND 1,100 billion Following the revelation of these discrepancies, the company's share price plummeted by approximately 80%, resulting in substantial losses for investors Alarmingly, an unqualified audit report for the year ending 2015 was issued by DFK, yet the core issue at Truong Thanh was not the inventory misstatement as initially claimed, but rather a decade-long pattern of inflated revenue reporting.
"processing" into inventories worth up to VND 2,200 billion (Tap chi tai chinh,
In 2018, a significant incident revealed not only the disappearance of VND 980 billion in inventory but also the presence of fictitious revenue over an extended period This situation raises concerns about the reliability of earnings quality in accurately representing an entity's performance and its utility in decision-making Financial reporting aims to furnish valuable information to current and potential investors, lenders, and creditors, assisting them in making informed decisions regarding resource allocation to the entity (IASB, 2010).
Earnings serve as a crucial performance indicator of a company's success during an accounting period In the Vietnamese stock market, firms face delisting if they report losses for three consecutive years Consequently, managers prioritize avoiding earnings losses by adjusting accounting figures to assure investors, lenders, and creditors of the company's operational stability and profitability This raises the question of whether loss avoidance earnings management is prevalent among Vietnamese listed firms.
In Vietnam, the Ministry of Finance has initiated a roadmap towards IFRS adoption to enhance financial transparency and comparability, beginning with the issuance of Circular No 200/2014/TT-BTC and Circular No 202/2014/TT-BTC in 2014 These circulars aim to align the accounting regime more closely with IFRS standards Previous studies in various countries have indicated a trend of earnings management aimed at positive outcomes and highlighted IFRS's role in mitigating loss avoidance behaviors Consequently, it remains crucial to investigate the presence of loss avoidance earnings management in Vietnam's stock market, along with the factors influencing this behavior and the impact of the new accounting framework on reducing such practices.
Therefore, the research topic “ Loss avoidance earnings management in an emerging market – evidence from Vietnam ” is chosen for discussion.
Objectives
Firstly, the study will focus on examining whether there is an existence of loss avoidance earnings management in Vietnamese listed firms for the period 2010-2019
Secondly, some changes to the accounting and financial reporting rules in two
Since 2014, the Ministry of Finance (MOF) has issued circulars that play a crucial role in shaping accounting practices This paper explores evidence demonstrating how the adoption of new accounting regimes has effectively mitigated the tendency for management to manipulate profits, thereby promoting more transparent financial reporting.
Lastly, identifying and analyzing the factors (the enforcement of Big 4 auditors and leverage ratio) influencing loss avoidance earnings management of Vietnamese listed companies.
Scope of study
This study examines earnings management aimed at avoiding reported losses in Vietnamese listed companies, particularly assessing how changes in accounting regulations influence the prevalence of loss avoidance It identifies key factors that drive managers to report positive profits, with a focus on the unique impacts of audit firm enforcement and leverage ratio within the Vietnamese context Previous research indicates that factors such as audit type, auditor industry specialization, business sector, and investor protection also play roles in loss avoidance, but this study prioritizes the effects of audit firm enforcement and leverage ratio on earnings management practices.
- Sample used: Data used for this thesis is collected in the period from 2010 to
2019 It includes available data from financial statements of more than 100 Vietnamese listed companies.
Research methodology
This study employs a mixed-methods approach, primarily emphasizing quantitative research while utilizing qualitative methods in the literature review to identify research gaps and formulate hypotheses The hypotheses are subsequently tested using quantitative techniques, including table data analysis, histogram analysis, and the binary logistic regression model Detailed discussions on the proposed research models, variable identification, and data collection methods will be provided in Chapter 2 - Methodology.
Research structure
Apart from the introduction, conclusion and reference, this paper is divided into four chapters as follows:
Chapter 1: Background and literature review
BACKGROUND AND LITERATURE REVIEW
IFRS harmonization in Vietnam
Harmonization and convergence of accounting standards are essential for enhancing the transparency and comparability of financial information globally The adoption of International Financial Reporting Standards (IFRS), developed by the International Accounting Standards Board (IASB), serves as a key mechanism for this purpose As a result of the convergence process, IFRS is increasingly utilized worldwide, establishing itself as a common framework for financial reporting.
According to the AICPA website (April 2020), around 120 nations and reporting jurisdictions either permit or require the use of IFRS for domestic listed companies, with approximately 90 countries fully conforming to IFRS as set by the IASB, as reflected in their audit reports The IFRS Foundation's 2018 report, "Use of IFRS Standards around the world," indicates that 27,000 domestically listed companies across 88 major stock exchanges globally utilize IFRS Standards, with Asia Oceania having 25 jurisdictions that require IFRS and 3 that permit it.
Vietnam is currently in the preparation stage for adopting International Financial Reporting Standards (IFRS), with the Ministry of Finance (MOF) having issued a draft IFRS roadmap for public comment on March 23, 2019 On March 16, 2020, the MOF approved Decision 345/QD-BTC, outlining a scheme to implement Financial Reporting Standards (FRS) aimed at enhancing the transparency and effectiveness of corporate financial information The scheme focuses on developing accounting frameworks, creating plans for IFRS application among selected entities, and aligning Vietnamese Financial Reporting Standards (VFRS) with international practices The implementation plan for IFRS is structured into three distinct stages.
From 2020 to 2021, the Ministry of Finance (MOF) laid the groundwork for the implementation of the International Financial Reporting Standards (IFRS) scheme, aimed at supporting enterprises in adopting IFRS starting in 2022 Key actions included publishing the Vietnamese translation of IFRS standards by March 2021, offering training programs for businesses, and amending the financial regime related to IFRS adoption, all to be completed by November 15.
From 2022 to 2025, Phase I of IFRS implementation allows enterprises, including parent companies of State-owned economic groups, listed and non-listed parent companies, and large unlisted public companies, to voluntarily prepare consolidated financial statements under IFRS after notifying the Ministry of Finance (MOF) Additionally, foreign direct investment (FDI) companies with the necessary resources may adopt IFRS for their separate financial statements.
In Phase II, set to commence after 2025, the Ministry of Finance (MOF) will utilize the results from Phase I to evaluate the demand and capacity for implementing IFRS among the specified enterprise groups.
During the period, the scheme for the application of VFRS also takes place with 2 phases:
− Preparation phase (from 2020 to 2024): development and issuance documents providing guidelines for VFRS to replace the current VAS The expected to issue of VFRS is by November 15, 2024
− VFRS application phase (from 2025): VFRS is applicable for all entities; whereas, MOF would provide Accounting Standards for Super-Small entities
The implementation of IFRS in Vietnam is significantly supported by strong leadership from the Ministry of Finance (MOF) Since 2014, the MOF has issued Circular No 200/2014/TT-BTC, which provides guidance for the enterprise accounting regime, and Circular No 202/2014/TT-BTC, which focuses on the preparation and presentation of consolidated financial statements These circulars reflect amendments to the accounting regime, aligning it more closely with IFRS standards and are applicable to all entities in Vietnam for fiscal years beginning in January.
Circular No 200/2014/TT-BTC -“Guidance for enterprise accounting regime” replaced the enterprise accounting regime issued under Decision No 15/2006/QD- BTC dated March 20, 2006 and Circular no 244/2009/TT-BTC dated December 31,
2009 with some changes to the accounting and financial reporting rules, such as:
Enterprises engaged in transactions using foreign currencies can choose a foreign currency for accounting purposes, provided they notify the relevant tax authorities Once a currency is selected, it cannot be changed except under special circumstances In addition to preparing financial statements (FSs) in the chosen foreign currency, businesses must convert these statements into Vietnamese dong (VND) before submission to regulatory authorities Unlike the previous Circular 244/2009/TT-BTC, which allowed only foreign-invested enterprises to select a currency under specific criteria, all enterprises must now convert their financial statements at the end of the accounting period using the average interbank exchange rate The principles for converting financial statements under Circular No 200/2014/TT-BTC align more closely with IFRS reporting methods than earlier regulations.
According to Circular No 200, revenue from the sale of goods and services must be recognized only when the transaction obligations are fulfilled, particularly for real estate enterprises where this includes the completion and transfer of ownership risks to customers Consequently, revenue should not be recorded for advance payments received from customers Additionally, revenue from real estate sales is governed by the accounting standards for the sale of goods, rather than construction contracts Article 79 of Circular 200 further clarifies that revenue recognition for goods eligible for return should only occur once the goods are no longer returnable.
Enterprises have the flexibility to utilize accounting book formats provided in the Appendix of Circular 200 or to create their own, provided they comply with the Law on Accounting and its amendments This indicates that the use of specific accounting book forms is not mandatory.
Recent changes in inventory measurement have eliminated the last-in-first-out (LIFO) approach due to concerns over profit manipulation Additionally, Circular 200 has introduced new accounts for recording investments, replacing short-term and long-term investment accounts with Account 121 for trading securities, Account 128 for held-to-maturity investments, and Account 228 for other investments.
On March 21, 2016, the Ministry of Finance (MOF) issued Circular No 53/2016/TT-BTC, which amends and supplements certain articles of Circular No 200/2014/TT-BTC regarding the accounting system for enterprises This new circular focuses on adjustments related to foreign exchange rates and the disposal of trading securities, among other issues.
Circular 202/2014/TT-BTC introduces updated guidelines for the preparation and presentation of consolidated financial statements, superseding the previous instructions outlined in Part XIII of Circular No 161/2007/TT-BTC issued by the Ministry of Finance on December 31, 2007 Key changes include enhanced clarity in reporting requirements and improved standards for financial disclosure.
− Using the similar terminology with IFRS 3- Business Combinations: Circular 202 use “non-controlling interest” instead of “Minority interest” like previous Circular and use the term “ gain from a bargain purchase”
− For preparing the consolidated financial statements, the parent company is allowed to revalue the value of previous investments at fair value at the acquisition date (Article 16)
− The deferred tax recording requirement at the acquisition date is due to the difference in fair value and book value of the acquired net assets (Article 18)
Circular 202 mandates that goodwill must be amortized over a period not exceeding 10 years, similar to Circular 161, while also requiring regular assessments for goodwill impairment If evidence indicates that the impairment exceeds the annual amortization, the excess must be allocated based on the amount of impairment recognized during the reporting period (Article 10, point 9).
The recent changes in accounting regulations, while still differing from IFRS, are anticipated to positively influence the quality of financial information in the Vietnamese stock market This raises the research question of whether the new accounting regime effectively enhances the transparency and comparability of financial data, particularly in mitigating the issue of loss avoidance among listed Vietnamese companies.
Loss avoidance earnings management and Earnings distribution approach
Earnings serve as a crucial performance indicator for assessing business success within a specific accounting period This raises the important question of whether the quality of earnings accurately reflects an entity's performance and aids in decision-making A comprehensive review of over 300 published papers on earnings by Dechow, Ge, and Schrand (2010) reveals that "higher quality earnings more faithfully represent the features of the firm’s fundamental earnings process that are relevant to a specific decision made by a specific decision-maker."
Ge and Schrand (2010) categorize earnings quality measurement into two primary groups: properties of earnings and earnings response coefficients Additionally, they introduce a third category that assesses earnings quality using external indicators of earnings misstatements.
Properties of earnings: according to Dechow, Ge and Schrand (2010), if earnings quality faithfully reflects the actual performance of the entity, it must have the following characteristics:
Earnings persistence, defined by the regression coefficient between future and current profits, is crucial for accurate equity valuations and enhances decision usefulness Research indicates that persistent earnings and cash flow predictability are essential for forecasting future cash flows According to Dechow, Ge, and Schrand (2010), cash flow is more persistent than accruals, with abnormal accruals being less persistent than normal accruals Firms with high total accruals often experience lower earnings persistence, more frequent restatements, and weaker internal controls, leading to diminished investor responses to earnings information (Giang, 2015) Ultimately, earnings persistence is influenced by the fundamental characteristics of a firm's operations and its accounting practices.
Earnings smoothness can arise from both inherent characteristics and managerial manipulation of accounting figures While many studies focus on the incentives for smoothing and the specific accounting choices that lead to smoother earnings, Dechow, Ge, and Schrand (2010) highlight a critical finding: commonly-used proxies for smoothness tend to show a negative correlation with other earnings quality proxies This suggests that the methods of accounting measurement do not uniformly affect the smoothness properties of reported earnings compared to other quality indicators.
Timely loss recognition, particularly within a conservative accounting framework, emphasizes the quicker acknowledgment of negative financial events compared to positive ones, a principle known as "conditional conservatism." This approach is influenced by various factors, including accounting standards, market incentives, and both internal and external enforcement mechanisms, all of which play a crucial role in shaping how promptly losses are recognized in financial reporting.
Benchmarking, as discussed by Dechow, Ge, and Schrand (2010), indicates that meeting or exceeding targets serves as a censored measure of earnings management Their research emphasizes analyzing profit distribution to identify abnormal points that suggest profit manipulation aimed at achieving these targets A key focus is on the "kink" in reported earnings distribution around zero, particularly among firms that report small positive profits or avoid minor losses, effectively "meeting or beating" forecasts.
Investor responsiveness to earnings: Prior studies use an earnings response coefficient (ERC) as a measure of earnings informativeness or earnings quality
A higher Earnings Response Coefficient (ERC) indicates a stronger predictability of the fundamental earnings process The concept of "investor responsiveness to earnings" evaluates how earnings quality influences investor decision-making Most studies in this area concentrate on the connection between stock prices, which are value-relevant, and accounting profits or returns from securities investments.
High-quality earnings accurately represent an entity's performance, making them crucial for investors' decision-making When earnings reflect true performance, they can significantly influence investment choices and help explain stock price volatility in the market.
Earnings quality can be evaluated through various external indicators of earnings misstatements, including the necessity to restate profit figures, the occurrence of regulatory investigations, and the issuance of qualified audit reports.
Research often employs selective criteria based on specific objectives, with prior studies suggesting that earnings with minimal management are deemed higher quality Earnings management is primarily evaluated through earnings smoothness and loss avoidance, among other factors Healy and Wahlen's (1999) widely accepted definition states that earnings management occurs when managers manipulate financial reporting and transactions to mislead stakeholders about a company's true economic performance or to affect contractual outcomes reliant on reported accounting figures.
Earnings management is often indicated by small profits or the avoidance of small losses (Dechow, Ge, and Schrand, 2010) According to Degeorge, Patel, and Zeckhauser (1999), this practice is influenced by three key thresholds: achieving positive profits, maintaining recent performance, and meeting analysts' expectations This aligns with three earnings benchmarks identified by Habib and Hansen (2009), which include the earnings level benchmark focused on loss avoidance, the earnings changes benchmark for improvements, and the analyst forecast benchmark Specifically, the earnings level benchmark targets firms near zero earnings, as loss avoidance is a prevalent measure of earnings management (Burgstahler and Dichev, 1997) This study emphasizes the importance of loss avoidance in earnings management among Vietnamese listed firms.
1.2.2 Earnings distribution approach (EDA) in searching for evidence on loss avoidance earnings management
Researchers investigating behavior management for positive earnings primarily employ the earnings distribution approach (EDA), analyzing the statistical properties of earnings to pinpoint critical thresholds, as noted by Burgstahler and Dichev.
Reported earnings can be assessed through various metrics, including annual net income relative to the beginning-of-year market value, earnings per share, net return on sales, return on equity (ROE), and return on assets (ROA) ROE is particularly relevant in studies focused on loss avoidance in banks and financial institutions, while other metrics are typically applied to samples excluding these highly regulated sectors When managers manipulate earnings to prevent reporting a loss, the distribution of reported earnings often reveals an unusual concentration of small profits.
Recent studies utilizing Exploratory Data Analysis (EDA) reveal a bell-shaped distribution of net income with a notable abnormality at zero, characterized by fewer observations just below zero and a high concentration just above Dechow, Richardson, and Tuna (2003) analyzed 47,847 firm-years from global companies between 1988 and 2001, identifying a "kink" in earnings distribution where fewer firms report small losses and more report small profits, particularly pronounced among newly listed firms Similarly, Suda and Shuto (2005) found a significant abnormal discontinuity at zero in the annual net income distribution of firms listed on Japanese stock exchanges from 1990 to 2000 In Southeast Asia, Saleh, Iskandar, and Rahmat (2005) examined firms on the Kuala Lumpur Stock Exchange from 1989 to 2001, confirming a single-peaked distribution of profit before tax scaled by total assets, with a concentration of positive earnings reported They also noted a decline in earnings management practices following the introduction of a corporate governance code in Malaysia in 2001.
In a study conducted by Pham et al (2019), which analyzed 1,115 net profit distributions from Vietnamese listed companies, it was found that the majority of firms reported a net profit ratio within the range of 0-7% Notably, very few companies reported negative profits, attributed to the implementation of an extended loss avoidance threshold.
(Hamdi and Zarai, 2012)) suggested that “in the emerging stock market like Vietnam, profitability may be higher and more fluctuated”
Logistics regression model and factors affecting loss avoidance
Previous research has employed binary logistic regression models, utilizing a dummy variable as the independent variable to analyze firms' earnings within small profit margins This approach is particularly effective for identifying factors influencing management behavior towards achieving specific financial targets Studies across various disciplines, such as those by Lang, Raedy, and Wilson (2005), Barth, Landsman, and Lang (2008), and Francis and Wang (2008), as well as in the banking and finance sectors, including Beatty, Ke, and Petroni (2002) and Kanagaretnam, Lim, and Lobo (2010), have successfully implemented this methodology.
In particular, Francis and Wang (2008), focus on listed firms from 42 countries over the period 1994-2004, test the loss avoidance behavior with dependent variable
The variable LOSS indicates whether a firm reports negative income before extraordinary items, with a value of 1 for negative income and 0 otherwise Research shows that differences in accounting standards and regimes have a minor impact on the prevalence of loss reporting across countries compared to the enforcement differences by Big 4 auditors In a study by Kanagaretnam, Lim, and Lobo (2010), which analyzed data from 29 countries over 14 years, the authors examined the existence of target management in the banking industry by focusing on two benchmarks: loss avoidance and meeting or exceeding the previous year's earnings Their findings indicate that the type of auditor (Big 5) and auditor industry specialization significantly restrict the ability to meet these benchmarks, as evidenced by negative coefficients associated with these factors.
In Vietnam's emerging market, Pham et al (2019) found that firms with a net profit ratio between 0% and 7% adjust their profits, using a binary logistics regression model to identify independent influencing factors Key findings indicate that business sectors, auditing firms, and debt-to-equity ratios significantly impact profit management behaviors Notably, 13 out of 14 business sectors exhibited a high propensity for profit adjustment Additionally, the debt ratio suggests that borrowing pressures lead listed companies to modify their financial statements Lastly, firms audited by the Big 4 firms demonstrate a lower tendency to engage in loss avoidance.
The criteria for defining a small profit margin vary across studies Kanagaretnam, Lim, and Lobo (2010) classify a small Return on Assets (ROA) as an interval between 0 and 0.002, assigning a value of 1 in their logistic regression model for this range, while Barth, Landsman, and Lang (2008) define it as a net income scaled by total assets within the range of [0; 0.01] In contrast, Pham et al (2019) focus on Vietnamese listed firms, identifying a small net profit ratio as 0%-7% of profit after tax scaled by net sales, where firms within this range receive a value of 1, and those outside it are assigned a value of 0.
Research gaps and research questions
This study examines loss avoidance earnings management among Vietnamese listed companies, particularly in light of the recent amendments to the accounting regime, specifically “Circular 200” and “Circular 202,” which align more closely with IFRS standards To date, the only relevant research by Pham et al (2019) has explored the existence of loss avoidance and its influencing factors, yet there is a notable gap regarding the impact of the new accounting regime on this behavior Previous studies have indicated a trend toward managing earnings for positive outcomes and the influence of IFRS adoption on loss avoidance, but the effectiveness of the new accounting regulations in mitigating such practices remains uncertain Pham et al (2019) suggest that firms with net profit ratios between 0-7% tend to adjust profits, citing that profitability in emerging markets like Vietnam can be higher and more volatile However, the broad profit margin interval proposed by Pham et al (2019) may obscure the distinction between inherent small profits and those manipulated by management, potentially skewing the analysis of factors affecting loss avoidance earnings management.
This study aims to address the research gap by examining the prevalence of loss avoidance in Vietnamese listed companies during two distinct periods, 2010-2014 and 2015-2019 Additionally, it analyzes the factors influencing the management's behavior towards achieving positive earnings The research employs an interval width that aligns with previous global studies and poses four key research questions.
1 Vietnamese listed companies manage accounting numbers to avoid reporting a loss or not?
2 Do stricter and detailed accounting regulations help to restrain loss avoidance earnings management among Vietnamese listed companies?
3 Does the enforcement by Big 4 audit firms affect the loss avoidance earnings management of listed companies in Vietnam?
4 Does leverage ratio affect the loss avoidance earnings management in Vietnamese listed companies?
This research investigates whether Vietnamese listed companies engage in earnings management aimed at achieving positive earnings If the findings reveal the presence of loss avoidance in earnings management, the study will further explore the impact of changes in accounting regulations and identify the factors that influence this behavior.
Chapter 1 indicates IFRS convergence in Vietnam through the IFRS road map which approved by the MOF and discusses some changes to the accounting regime to be more suitable for IFRS Besides, this chapter presents the prevalence of loss avoidance and typical method in searching for evidence on loss avoidance earnings management – Earnings distribution approach (EDA) In addition to EDA, the binary logistics regression model is applied to affirm the existence of managing toward positive earnings and identify factors that affect this behavior These are the basis for the method in searching for evidence on loss avoidance earnings management, evaluating the effect of the implementation of new standards and the proposed research model mentioned in chapter 2.
METHODOLOGY
Hypothesis development
Previous research highlights the prevalence of loss avoidance in earnings management, particularly in Vietnam A study by Pham et al (2019) analyzed 1,115 net profit distributions across 14 industries, revealing that most firms reported net profit ratios within the 0-7% range, with very few listed companies showing negative profits Additionally, Giang (2015) examined the empirical distribution of Return on Assets (ROA) and Return on Equity (ROE) among Vietnamese commercial banks from 2004 to 2017, confirming the existence of loss avoidance in earnings reporting This leads to the formulation of the first hypothesis regarding loss avoidance earnings management among Vietnamese listed firms.
Hypothesis H1: Vietnamese listed companies manage accounting number to avoid reporting a loss
Anticipated changes in the accounting regime suggest that new regulations will enhance earning quality by reducing the tendency for loss avoidance Therefore, we propose hypothesis 2.
Hypothesis H2: Stricter and detailed accounting regulations help to restrain loss avoidance earnings management among Vietnamese listed companies
Previous research indicates that loss avoidance is influenced by various factors, including audit type, auditor industry specialization, leverage ratio, business sector, and investor protection In the context of Vietnam, however, the enforcement of audit firms and the leverage ratio are identified as particularly significant factors This study will specifically examine the impact of these two elements on loss avoidance in earnings management The following hypotheses, H3 and H4, have been formulated to explore these relationships.
Hypothesis H3: The enforcement by Big 4 audit firms helps to constrain the pervasiveness of loss avoidance earnings management in Vietnamese listed firms
Hypothesis H4: Leverage ratio affects the loss avoidance earnings management in Vietnamese listed companies.
Proposed research model
This section builds upon four research hypotheses to develop a research method and propose a model aimed at investigating loss earnings management in Vietnamese listed firms It will explore the impact of new accounting regimes on the prevalence of loss avoidance and identify the factors influencing this behavior.
2.2.1 Earnings distribution approach (EDA) in searching for evidence on loss avoidance earnings management and the effect of changes in accounting regime in the prevalence of loss avoidance
This study employs the Earnings Distribution Approach (EDA) to investigate loss avoidance earnings management among Vietnamese listed non-financial companies, following the methodology established by Burgstahler and Dichev (1997) By analyzing the distribution of net income relative to total assets, measured as Return on Assets (ROA), the research aims to identify patterns of loss avoidance behavior ROA serves as a critical profitability metric, reflecting the efficiency of a company's management in generating earnings from its assets The study posits that if managers manipulate earnings to achieve positive profits, the ROA distribution will reveal an unusual concentration of small profits, aligning with findings from prior research that focuses on ROA measures.
The concept of small profit margins is crucial in financial analysis, particularly within the interval range of [0;0.01] Studies such as those by Jeanjean and Stolowy (2008) highlight the significance of this range, focusing on income distribution before extraordinary items relative to lagged total assets In contrast, Pham et al (2019) examined loss avoidance in Vietnam, identifying a net profit ratio primarily between 0-7%, which is considerably broader than the small profit margins observed in earlier global studies This wider interval can obscure the true nature of a firm's small profit, complicating the distinction between genuine small profits and those manipulated by management Thus, this study defines a Vietnamese listed firm as having a small reporting profit when its return on assets lies within the narrow range of 0 to 0.01.
This study investigates the influence of changes in accounting regulations on loss avoidance earnings management practices With new Circulars applicable to all entities in Vietnam starting January 1, 2015, the research compares the distributions of Return on Assets (ROA) between the periods 2010–2014 and 2015–2019 The analysis aims to determine how the implementation of the new accounting regime affects the skewness of the ROA distribution.
The decrease in ROA during the later period compared to the previous one suggests improved management towards achieving positive earnings Additionally, the reduced ratio of small profits to small losses reflects a decline in loss avoidance tendencies, as noted by Leuz, Nanda, and Wysocki.
(2003), Jeanjean and Stolowy (2008) and Miková (2014)) For the same variable, this study assumes that small profit corresponds to ROA is in the range [0;0.01] and small loss corresponds to the range [-0.01;0)
2.2.2 The binary logistics regression model and factors affecting loss avoidance earnings management
This study aims to identify the factors influencing firms' management strategies towards achieving positive profits The dependent variable, loss_avoid, is analyzed using a binary logistics regression model, where it is assigned a value of 1 if the firm reports a small Return on Assets (ROA) within the range of [0,0.01], and 0 if the ROA falls outside this range To uncover the influencing factors, the regression model incorporates independent variables such as big4, leverage, size, and period.
As the previous studies indicate that enforcement of Big 4 auditor (Deloitte,
The involvement of Big Four auditors (EY, KPMG, and PWC) plays a crucial role in limiting earnings management, as evidenced by studies from Kanagaretnam, Lim, and Lobo (2010) and Pham et al (2019) In this model, the "big4" variable is a dummy variable, assigned a value of 1 if the firm's auditor is one of the Big Four and 0 if not Additionally, the need to persuade investors, lenders, creditors, and shareholders of a firm's profitability drives managers to manipulate accounting figures to secure more resources, indicating that the leverage ratio may influence positive earnings management Pham et al (2019) specifically highlight that the debt-to-equity ratio significantly impacts loss avoidance in firms listed on the Vietnamese stock market This study also examines the effect of changes in the accounting regime on loss avoidance earnings management, utilizing a "period" variable that indicates whether the firm-year observation falls within 2015-2019 This serves as a further test of the impact of accounting regime changes, corroborating findings from the earnings distribution approach (EDA) Additionally, following the methodology of Kanagaretnam, Lim, and Lobo (2010) and Lang, Raedy, and Wilson (2005), the model incorporates control variables, including firm size, measured as the natural logarithm of total assets at the end of the period.
As a result, the binary logistics regression model used in this graduation thesis as the table follows
Table 2.1: Logistics regression model for identifying factors affecting loss avoidance earnings management
The article posits that Vietnamese listed firms manipulate accounting numbers to evade reporting losses, with the enforcement of Big4 auditors expected to negatively influence this behavior (negative β1) Additionally, it anticipates a positive relationship between leverage and profit management, suggesting that managers are incentivized to achieve positive profits According to the second hypothesis, the "period" variable is expected to have a negative correlation with the dependent variable, while the inclusion of "size" as a control variable means no directional predictions are made regarding its coefficients.
Data and sample
This study utilizes secondary data from the financial statements of Vietnamese listed companies from 2010 to 2019, sourced from their official websites, cophieu68.vn and cafef.vn The analysis excludes banks and financial institutions to focus on other sectors.
Model Logistics regression model for identifying factors affecting loss avoidance earnings management
The independent variables used in the model are defined as follows: loss_avoidit an indicator variable, taking the value 1 if the firm i reports a small
The study focuses on the Return on Assets (ROA) within the interval [0, 0.01] for year t, with a binary indicator for Big 4 auditors assigned a value of 1 if applicable, and 0 otherwise Leverage is defined as total liabilities scaled by total equity, while size is represented by the natural logarithm of total assets for firm i in year t An additional indicator variable, period, is set to 1 for firm-year observations from 2015 to 2019 and 0 otherwise The intercept, β0, is acknowledged as being highly regulated, suggesting distinct motivations for loss reporting Due to time constraints, the sample was narrowed to over 100 companies, excluding those lacking sufficient data for 10 consecutive years Ultimately, the analysis included 108 Vietnamese listed companies across various non-financial sectors, with detailed sample structure presented in subsequent tables.
Table 2.2: The sample structure categorized by industry and audit firm
Industry Number of observations audited by Big 4
Number of observations audited by non-Big 4
Number of firm-year observations
A total of 108 Vietnamese listed companies, representing 1,080 firm-year observations across 12 industries, were analyzed based on their market capitalization as of April 2020 The real estate sector had the highest number of observations, totaling 350, followed by the energy, oil & gas, and food production sectors In contrast, the manufacturing and telecommunications industries recorded the fewest observations Overall, non-Big 4 audit firms conducted more audits than Big 4 firms, with 462 observations (42.78%) audited by Big 4 firms and 618 by non-Big 4 firms.
Table 2.3: The sample structure categorized by audit firm and stock exchange
Audit firm Number of observations listed on HNX
Number of observations listed on HOSE
Number of observations listed on UpCOM
Audited by Big 4 audit firms
Table 2.3 presents a breakdown of the sample structure by audit firm and stock exchange, highlighting 108 Vietnamese listed firms across three exchanges: HNX (Hanoi Stock Exchange), HOSE (Ho Chi Minh Stock Exchange), and UpCOM (Unlisted Public Company Market) Notably, the number of firms listed on HOSE surpasses that of the other exchanges, with a higher likelihood of these firms being audited by Big4 audit firms compared to non-Big4 firms and those on other exchanges Conversely, HNX and UpCOM show a trend where a majority of firm-year observations are audited by non-Big4 firms, comprising 78.75% and 71.67%, respectively.
This study provides a comprehensive overview of the operational status of selected Vietnamese listed firms, focusing on their total assets and debt-to-asset ratios Total assets serve as a key criterion for evaluating the size of a listed firm, alongside total equity and performance metrics The growth rate of total assets among the Vietnamese listed firms in the sample is detailed in the accompanying table.
Table 2.4: Total assets of selected Vietnamese listed firms form 2010-2019
Total assets of selected Vietnamese listed firms
N Mean Std.Dev Min Median Max
Source: The author calculates from published data of listed companies
Over a 10-year period, selected Vietnamese listed firms exhibited a steady increase in total assets, with significant fluctuations among enterprises The standard deviation of total assets rose dramatically from VND 6,567,550 million in 2010 to VND 41,434,970 million in 2019, reflecting the diversity in size among the firms, including both small and medium-sized enterprises as well as large corporations Notably, the growth in total assets for larger firms outpaced that of smaller firms Additionally, there was a remarkable surge in average total assets between 2018 and 2019, with the average reaching VND 11,361,688 million in 2019, marking over a 20% increase from 2018 and a staggering 343.1% rise compared to 2010.
Figure 2.1: Debt-to-asset of selected Vietnamese listed firms from 2010 to 2019
Source: The author calculates from published data of listed companies 1
Figure 2.1 illustrates the debt-to-asset ratios of various Vietnamese listed companies over a decade, revealing that the distributions of these ratios—encompassing their centers, variations, and shapes—exhibit notable similarities throughout the entire period.
The box plot, which excludes the top 1% of observations to mitigate the impact of extreme values, illustrates that the median debt-to-asset ratio remains consistently around 0.5 across the years Notably, the medians are situated closer to the upper quartile, indicating that over 50% of the observations fall below this median value Additionally, the analysis reveals an increased range in the debt-to-asset ratio during the period from 2015 to 2019, marked by a higher maximum value compared to earlier years This trend suggests a rise in financial leverage, potentially contributing to an increase in accounting data manipulation.
Chapter 2 clarifies the research hypotheses for the topic loss avoidance earnings management in Vietnamese listed firms In addition, this chapter also presents the method of data collection and data analysis, including a detailed analysis of sample structure and clarifying the distribution characteristics of the selected samples through total assets, debt-to-asset ratio Three hypothesizes and logistics regression model for identifying factors affecting loss avoidance earnings management are developed Based on the selected sample, this study will find the evidence for loss avoidance among listed firms on the Vietnamese stock market, after that, measure the impact of the implementation of new accounting regime on loss avoidance and identify factors influencing this behavior The empirical results are presented in the next chapter.
RESEARCH FINDINGS
Descriptive statistics and correlation between variables
This study aims to investigate the presence of loss avoidance earnings management among Vietnamese listed companies, focusing on return on assets (ROA) as a key indicator of management efficiency in utilizing assets to generate earnings ROA is determined by dividing the net income for the year by the total assets at the end of the fiscal year Descriptive statistics for ROA are presented in Table 3.1, which retains the originality of the data by not excluding extreme value firm-year observations.
Table 3.1: Descriptive statistics by year for ROA
Year N Mean Std.Dev Min Median Max
Source: The author calculates from published data of listed companies using R
The analysis includes 1,080 firm-year observations, revealing that both the mean and median Return on Assets (ROA) remain positive over a decade, with values of 0.0580 and 0.0480, respectively The low standard deviation of 10.36% suggests that this sample is representative of the overall population Consequently, it is plausible that Vietnamese listed companies may manipulate accounting figures to prevent reporting losses.
The binary logistics regression model was employed to identify factors influencing loss avoidance in earnings management This analysis utilized financial statement data from 108 listed companies in Vietnam, covering the period from 2010 onwards.
2019 Descriptive statistics for independent variables as the table below
Table 3.2: Descriptive statistics for loss avoidance test
Source: The author calculates from published data of listed companies using R
The "size" variable exhibits a mean of 6.1153 and a median of 6.0345, indicating a close correlation between the two measures Additionally, the average total liabilities to total equity ratio stands at 143.44%, with a minimum of -451.71% and a maximum of 22.67 times The extreme values are attributed to the presence of a listed company in the sample that has significant negative cumulative retained earnings, resulting in negative total equity and a decrease in total assets.
Variables N Mean Std.Dev Min Median Max leverage 1080 1.4344 1.7540 -4.5171 1.0447 22.6718 big4 1080 0.4278 0.4950 0.0000 0.0000 1.0000 size 1080 6.1153 0.7675 4.4848 6.0345 8.6061 period 1080 0.5000 0.5002 0.0000 0.5000 1.0000
Figure 3.1: Correlations for loss avoidance test
Source: The author calculates from published data of listed companies using R
Figure 3.1 illustrates the correlations among variables in the binary logistics regression model, highlighting that "size" and "leverage" significantly correlate with the dependent variable "loss_avoid" at 1% and 0% significance levels, respectively However, these variables do not account for all factors influencing loss avoidance in earnings management Consequently, multivariate analysis is employed for more accurate inferences Notably, aside from the correlation between "size" and "big4," all other correlations among the independent variables remain below 0.5, suggesting minimal risk of multicollinearity in the model, though further testing will be conducted to confirm this.
The existence of loss avoidance earnings management among Vietnamese listed companies- evidence from earnings distribution approach
Between 2010 and 2019, a total of 108 Vietnamese listed companies across 12 industries were analyzed, resulting in 1,080 observations This study focused on the density distribution of net income in relation to total assets.
(ROA) Figure 3.2 is a histogram of ROA for the range -0.3 to +0.3 with an interval width of 0.01
Figure 3.2: Empirical Distribution of ROA for the period 2010-2019
Source: The author calculates from published data of listed companies using R
The analysis reveals a single-peak distribution of Return on Assets (ROA) with a noticeable abnormality at zero, indicating fewer observations below this point and a higher concentration just above it Most ROA values fall within the range of [0, 0.1], with an average of 0.0580, and very few enterprises report negative profits These empirical findings align with previous research, including the studies by Burgstahler and Dichev.
Research by Pham et al (2019) on Vietnamese stock market firms, alongside findings from 1997 regarding U.S listed companies, supports the hypothesis H1, indicating that Vietnamese listed companies engage in accounting practices to prevent reporting losses.
The effect of changes in accounting regime in loss avoidance among
Vietnamese listed firms – evidence from earnings distribution approach
The analysis of Return on Assets (ROA) indicates a trend of positive earnings management among Vietnamese listed companies Consequently, this study aims to investigate the impact of the newly implemented accounting regime on decreasing the occurrence of loss avoidance practices.
The recent changes in accounting regulations in Vietnam, while still having some differences from IFRS, are anticipated to enhance the quality of financial information in the Vietnamese stock market This includes a reduction in the prevalence of loss avoidance earnings management The new Circulars will be applicable to all entities in Vietnam starting from the fiscal year that begins on January 1.
2015, the impact of changes in accounting regime can be assessed through the comparison of distributions of ROA for two periods 2010–2014 and 2015–2019
Figure 3.3: Distribution of ROA for two the periods 2010–2014 and 2015–2019
Source: The author calculates from published data of listed companies using R
The histograms reveal persistent discontinuities in the distribution of net income scaled by total assets, both prior to and following the introduction of the new accounting regime Notably, there is a slight increase in firms reporting small profits within the interval just above zero [0,0.01], while the number of firms experiencing small losses, indicated by the interval just below zero, shows a moderate decline over the two periods.
Researchers, including Leuz, Nanda, and Wysocki, have conducted assessments by examining the small profits/small losses ratio through odds ratios This method allows for a comparison of the frequencies of small profits against small losses, highlighting the tendency to report small profits when they occur.
Research by Jeanjean and Stolowy (2008) and Miková (2014) indicates that a Vietnamese listed firm experiences a small profit when its Return on Assets (ROA) is between 0 and 0.01, while it incurs a small loss when ROA falls within the range of [-0.01; 0) Table 3.3 illustrates the ratio of small profits to small losses, calculated using odds ratios The analysis of these odds changes suggests that variations in the accounting regime significantly impact loss avoidance in earnings management.
Table 3.3: Measurement: small profits/small losses ratio
Case (small profit) Controls (small loss) Odds
Test for trend of odds Chi2
Source: The author calculates from published data of listed companies using R
The significant odds ratio observed in both periods indicates a widespread occurrence of reporting small profits Furthermore, the rise in odds from 9.43 to 18 in the post-period contradicts the hypothesis that loss avoidance earnings management diminishes following the introduction of a new accounting regime Additionally, the Chi-Square statistic's p-value of 0.31 suggests that the statistics are independent at a 95% confidence level.
In summary, the analysis of histograms and asymmetry measures indicates that there has been no decrease in the prevalence of loss avoidance among Vietnamese listed companies following changes in the accounting regime, contradicting hypothesis H2.
Factors affecting loss avoidance earnings management – results from the
With 1,080 observations of 108 Vietnamese listed companies in 10 years (from 2010 to 2019), after using the binary logistics regression model, the results of loss avoidance test is reported as the table below
Table 3.4: Regression results for the loss avoidance test Panel A: Regression Results
Panel B: Testing for reliability of model
Source: The author calculates from published data of listed companies using R
Logistic regression differs from linear regression by not adhering to key assumptions such as linearity, normality, and homoscedasticity, while still applying some other assumptions The variance inflation factors (VIF) for nearly all independent variables are below 2, indicating that multicollinearity is not an issue Although homoscedasticity isn't a primary assumption for logistic regression, the Goldfeld-Quandt test shows a p-value greater than 5%, confirming constant variance in the error term According to Atmathew (2015), a logistic regression model is considered a better fit if it shows improvement over a model with fewer predictors To assess model fit, this study employs the likelihood ratio test, which evaluates the goodness of fit between two statistical models A p-value less than 0.05 leads to the rejection of the null hypothesis that the reduced model is true The likelihood ratio test results in panel B (pr=0.0001672 at a 0% significance level) indicate that model 2, which includes additional predictors, is significantly more accurate than the reduced model (model 1).
The coefficients in panel A show that except “period”, other variables have significant influence (p-value