INTRODUCTION
Research topic
Vietnam officially joined the World Trade Organization (WTO) in 2007, leading to significant economic growth across various sectors, with a GDP growth rate of 7.13% by the end of that year, according to the World Bank However, following the Global Financial Crisis (GFC) of 2008/2009, the world economy faced substantial challenges, resulting in bankruptcies and a fragile banking system Vietnam also encountered difficulties during this period, experiencing turbulence in financial markets, a rise in bad debts, and troubles within commercial banks Consequently, financial risk management has become a critical issue for corporate governance, the government, and banks This underscores the urgent need for developing prediction models of corporate financial distress, which are essential for banks, investors, asset managers, rating agencies, and distressed firms.
The assessment of corporate financial distress and bankruptcy has been a prominent topic for decades, leading to the development of various predictive models globally In Vietnam, many researchers focus on accounting-based models, particularly the Altman Z-Score, which effectively distinguishes between distressed and non-distressed firms The Z-Score model is renowned for its reliability in predicting financial distress (Bemmann, 2005), making it a crucial tool in financial analysis.
Market-based models, such as the Merton and KMV models, utilize market data, including asset volatility, to predict corporate financial distress by measuring the Distance to Default (DD) These models provide valuable insights into a firm's financial stability, helping investors and analysts assess potential risks.
There is a notable gap in research regarding the simultaneous use of market-based and accounting-based models to predict corporate financial distress among listed firms in Vietnam Specifically, a thorough evaluation of these two modeling approaches has yet to be undertaken, especially across various timeframes, including both crisis and non-crisis periods.
On this ground of this urgent and important need, this study entitled “Corporate
This article compares two common financial distress models used to predict corporate financial distress among Vietnam's listed firms By analyzing these approaches, the study aims to identify their effectiveness and applicability in the context of Vietnam's unique market environment.
Research Objectives
This study aims to evaluate the effectiveness of accounting-based and market-based models in predicting financial distress within the Vietnamese stock market The research focuses on achieving specific objectives related to the comparative performance of these models in this unique financial context.
Presenting the best possible determinants which can be used in the accounting- based models of predicting financial distress for listed firms in the context of Vietnam
Estimating the Distance to Default (DD) and Z-Score using the market-based and accounting-based models for listed firms in the context of Vietnam
Comparing and contrasting differences between the two score with particular focus on various periods including the crisis and non-crisis periods.
Research Questions
In order to achieve the following research objectives, the following research questions have been raised:
What are the best possible determinants to be utilized to predict the financial distress for Vietnam’s listed firms?
What are the Distance to default (DD) and the Z-Score for Vietnamese listed firms?
How has financial distress level been changed during various periods including crisis and non-crisis periods for listed firms in Vietnam?
Research object and scope
This study investigates the financial distress of companies by analyzing a sample of 631 non-financial listed firms on the Vietnamese Stock Market, specifically the Hochiminh City Stock Exchange (HOSE) and the Hanoi Stock Exchange (HNX), over a period from 2006 to 2016.
In which, both financial statement and market information will be collected to evaluate the research problem.
Research Contribution
This study brings some new points in the field of research in the context of Vietnamese market These are:
(i) Providing an empirical evidence of the factors that affect the financial distress of the business
(ii) One of the first studies in Vietnam using both accounting-based and market- based models to forecast the company's financial distress.
Research Structure
The structure of research as follows:
Chapter 1 aims at presenting an overview of research topic This chapter also identifies the objectives, scope of the research, the research questions and the contribution of research
Chapter 2 will explore key theories pertinent to the research topic, including fundamental concepts of financial distress in companies, forecasting models for financial distress, and relevant studies from both domestic and international sources.
Chapter 3: Methodology and Research Design
Chapter 3 discusses how to select research data, select models, build variables in the model and econometric methods that will be used in the study
Chapter 4: Research Result and Discussion
Chapter 4 presents the empirical result of the study The result from the proposed model in Chapter 3 will be implemented and compared with the hypotheses proposed Subsequently, discussions will be made to evaluate and interpret the results
Chapter 5 is the final chapter of the research The result of the study will be summed up and recommendations for research results will be made for the relevant stakeholders Chapter 5 also outlines some of the remaining constraints and recommends further research directions
LITERATER REVIEW
Corporate financial distress
Financial distress refers to a firm's inability to meet its financial obligations, encompassing issues such as bankruptcy, overdrawn accounts, and defaults on bonds and preferred stock (Beaver, 1966) Gordon (1971) characterized financial distress negatively, emphasizing the challenges firms face when experiencing temporary liquidity shortages that hinder timely and complete fulfillment of obligations He proposed that financial distress is part of a broader process that includes failure and restructuring, advocating for a definition that considers financial structure and security valuation.
Financial distress is a critical event that marks the transition from a company's financial health to a state of financial illness, necessitating corrective actions for recovery (Andrade & Kaplan, 1998; Brown, James & Mooradian, 1992) Andrade and Kaplan (1998) identified two primary forms of financial distress: defaulting on debt payments and attempting to restructure debt to avert default According to Brown, James, and Mooradian (1992), a company is considered financially distressed if it seeks to implement restructuring measures to prevent default or as a response to existing financial challenges.
Asquith et al (1994) identified the interest coverage ratio as a key indicator of financial distress, classifying a firm as distressed if its EBITDA falls below 80% of its interest expenses for two consecutive years This benchmark reflects the common challenges faced by financially distressed companies, including declining profitability, excessive leverage, and inadequate cash flow to meet current obligations.
Turetsky and MacEwen (2001) effectively characterized financial distress as a structured, continuous process comprising distinct stages marked by specific adverse financial events They describe financial distress as a series of intervals between distress points, beginning with a significant drop from positive to negative cash flow This decline is followed by a reduction in dividends, indicating a transition to the next stage, which can lead to default The progression includes technical defaults on debt and often culminates in troubled debt restructuring, aimed at mitigating the risk of potential bankruptcy.
Purnanandam (2005) explored financial distress from a solvency perspective, developing a theoretical model for corporate risk management that incorporates financial distress costs He defines financial distress as a transitional phase between solvency and insolvency, occurring when a company misses interest payments or breaches debt covenants The shift from solvency to insolvency is marked by the maturity date, contingent on the company's asset value falling below its debt obligations This definition effectively differentiates financial distress from default and bankruptcy, highlighting that a company can experience distress without defaulting, while default and bankruptcy cannot occur without prior financial distress.
Purnanandam's model (2005) aligns with the financial distress concept introduced by Gilbert et al (1990), who emphasized that financial distress possesses distinct financial traits compared to bankruptcy Notably, financial distress is marked by a pattern of negative cumulative earnings sustained over several consecutive periods.
Financial distress can lead to various outcomes for a company, including bankruptcy, but not all companies facing such challenges will necessarily fail Many may recover by restructuring their debt to achieve solvency, merging with other entities, or strategically filing for bankruptcy as a response to financial difficulties.
Financial distress and failure occur when a company experiences chronic losses, leading to a significant rise in liabilities and a decrease in asset value This situation arises when a firm struggles to meet its financial obligations to creditors The likelihood of financial distress escalates for companies with high fixed costs, illiquid assets, or revenues vulnerable to economic downturns.
Financial distress is defined through key indicators that are essential in empirical research for predicting the financial health of firms These indicators help analyze the performance of distressed companies, their restructuring efforts, and the effects of legal proceedings on their capital structure Financial ratio analysis is a crucial method for detecting signs of financial distress, and accounting-based indicators remain popular among researchers despite criticisms regarding their historical focus Although these ratios may not fully capture future company dynamics, they are effective in predicting financial distress and default probabilities A mixed approach that combines both accounting and market-based determinants enhances the identification of financial distress, addressing its complex nature by linking both internal and external factors.
In summary, different procedures to the definition of the term “Financial Distress” show how complex, versatile and maybe even controversial this economic category is
Corporate failure has significant economic implications, particularly for stakeholders of public companies, as firms often experience distress before bankruptcy Identifying corporate financial distress early is crucial for lenders and investors This importance has spurred extensive research into predicting corporate bankruptcy and financial distress This study will examine two primary approaches: accounting-based models and market-based structural models.
Financial distress and bankruptcy
Bankruptcy serves as a viable option for businesses in Vietnam that are unable to meet their debt obligations promptly, as outlined in the Bankruptcy Law of 2014 This law defines a bankrupt enterprise as one that experiences significant operational difficulties or losses despite implementing necessary financial strategies, ultimately leading to an inability to pay off debts The determination of bankruptcy hinges on two critical conditions that assess the financial health and payment capabilities of the business.
- Loss of ability to pay due debts
- The phenomenon of losing the ability to pay due debt is no longer a temporary but that is very intrinsic nature and incurable
Upon the appearance of the above-said bankruptcy, enterprises shall have to apply necessary financial measures as follows: To overcome the insolvency situation such as:
- Planning for reorganizing production and business, strict management of expenses, seeking markets for products
- Taking measures to handle the unsold goods and supplies
- Recovery of debts and assets appropriated
- Negotiating with creditors to defer debt, buy debt, debt guarantee, write off debt
- Finding new grants and loans to invest in technology innovation, maintain operations to generate revenue
After having applied the above-said financial measures but still having difficulties and failed to overcome the situation of insolvency, the enterprises have fallen into
Bankruptcy occurs when a business is unable to pay its due debts, marking the final step after unsuccessful attempts to resolve financial difficulties In such cases, assets must be disposed of in accordance with legal provisions.
There are many ways to classify bankruptcy Based on the nature of the bankruptcy and the person filing for bankruptcy filing Based on the nature of the bankruptcy:
- Honest bankruptcy: Bankruptcy due to real causes
- Fraudulent bankruptcy: A bankruptcy committed by owner in advance of fraudulent tricks to appropriating the assets of creditors
Based on the person filing for bankruptcy filing
Voluntary bankruptcy occurs when a business submits a self-filing request for bankruptcy declaration due to its inability to repay outstanding debts, signaling that it has exhausted all options to resolve its insolvency.
- Mandatory bankruptcy: The creditor applying for the court declaring bankruptcy of the indebted enterprise, the business itself does not want to be declared bankrupt
This study emphasizes corporate financial distress rather than bankruptcy, highlighting that while firms in distress often face potential bankruptcy, not all distressed firms ultimately fail; many can recover (Faelten and Vikrova, 2014) Therefore, developing effective strategies to identify financial distress promptly is crucial for lenders and investors.
Models of corporate financial distress
Measuring distress risk has been a longstanding question in financial literature, with numerous researchers developing various approaches to predict financial distress and bankruptcy over the decades This prediction has been analyzed from multiple perspectives, including economic, financial, accounting, statistical, and informational viewpoints Existing forecasting models can be classified chronologically or based on their underlying statistical or mathematical methodologies (Altman and Hotchkiss, 2005).
With chronological classification, Altman and Hotchkiss (2005) divide all methods for the prediction of corporate financial distress into the following groups:
- Univariate analysis (Use of accounting-based ratios or market indicators for the distress risk assessment: Beaver, 1966)
Multivariate analysis encompasses various statistical techniques such as Discriminant Analysis, Logit and Probit models, Non-Linear Models, Neural Networks, and Recursive Partitioning Analysis These methods utilize accounting and market information to assess financial health and risk, exemplified by Altman’s Z-Score and Ohlson’s O-Score, as well as Shumway’s Simple Hazard Model By integrating these approaches, analysts can effectively evaluate and predict financial outcomes.
- Discriminant and Logit Models in Use (Z-Score for manufacturing, ZETA- Score for Industrials, Private Firm Models – Z’’-Score, EM (Emerging Markets) Score, etc
- Artificial Intelligence Systems (Expert Systems, Neural Networks Credit Model of S&P)
- Contingent Claim Models (KMV Credit Monitor Model, Risk of Ruin)
- Mixed Ratios / Market Value Models (Moody’s Risk Calc, Z-Score / Market Value Model)
This study categorizes widely used empirical models into two main groups based on the type of data utilized: accounting-based models, which rely on accounting information, and market-based models, which leverage capital market data Each approach is examined in detail below.
Accounting-based models evaluate the significance of financial statement information, such as balance sheets and profit and loss statements, to assess the likelihood of a company experiencing financial distress These models utilize key financial data, including profitability, liquidity, and solvency ratios, to predict potential financial challenges While accounting information is generally accessible, annual auditor reports can delay the timely availability of this crucial data.
The delayed release of financial reports until the following year has contributed to the popularity of various analytical tools for assessing financial distress The ease of access to financial data has made these techniques essential in empirical research over the decades.
Beaver (1966) pioneered modern distress risk assessment by utilizing univariate statistical analysis to predict corporate failure, selecting 30 financial ratios to evaluate 79 firms that failed within five years His research identified the Cash flow/Total Debt ratio as the most effective bankruptcy indicator However, recognizing the limitations of his univariate approach prompted the advancement of the Z-Score, which employs multiple discriminant analysis (MDA) for a more robust evaluation of financial distress.
Z-Score model for public firms
In 1968, Edward Altman introduced the Z-score formula, employing Multiple Discriminant Analysis (MDA) to create a predictive model for financial distress By applying multivariate discriminant analysis, he identified a linear combination of financial ratios that effectively distinguishes between financially distressed and non-distressed firms His study utilized a sample of 33 bankruptcies from 1946 to 1965, paired with 33 non-distressed firms of similar size and industry Altman selected 22 financial ratios, categorizing them into five groups: liquidity, profitability, leverage, solvency, and activity Through extensive statistical testing, he assessed the interrelations among these variables, ensuring the model's statistical significance and predictive accuracy.
Out of 22 variables, 5 have been identified as the most significant indicators of distress risk The overall assessment can be quantified using Altman’s Z-Score, derived from a specific discriminant function.
(where the first four variables are expressed in decimals, e.g., 20.0%)
Working capital is defined as the difference between current assets and current liabilities, serving as a key indicator of a company's liquidity Altman analyzed various liquidity ratios, including the current ratio and quick ratio, but determined that the working capital to total assets ratio demonstrated greater statistical significance in both univariate and multivariable analyses This measure effectively compares net liquidity against total capitalization, providing valuable insights into a company's financial health.
The RE/TA ratio assesses a firm's cumulative profitability over time in relation to its total assets, indicating that younger firms typically exhibit lower ratios due to insufficient accumulated profits, making them more susceptible to bankruptcy This measurement tends to discriminate against young companies, as research by Altman suggests they face a higher likelihood of financial distress compared to their more established counterparts Additionally, it is important to adjust this ratio for paid dividends to shareholders, as they can impact the overall assessment.
X 3 = Earnings before interest and taxes / Total assets
The ratio is determined by dividing a firm's total assets by its earnings before interest and tax, serving as an indicator of how effectively a company utilizes its assets to generate earnings, excluding tax and interest considerations This measure is particularly significant for assessing the productivity of companies, especially those struggling with profitability, as it highlights the importance of asset efficiency for long-term survival in the market.
X 4 = Market value of equity / Book value of total liabilities
The ratio measures a firm's total market value, encompassing all shares, in relation to its total debt, which includes both current and long-term liabilities This metric indicates the extent to which a firm's assets can decline in value—assessed through the market value of equity plus debt—before its liabilities surpass its assets, potentially leading to insolvency issues Altman's research suggests that this ratio is a more effective tool for assessing financial stability.
13 predictor of bankruptcy than a similar, more commonly used ratio: Net worth / Total debt (book values)
The measurement reflects the sales value generated by the firm in relation to its total assets, indicating management's effectiveness in navigating competitive conditions, as noted by Atman This ratio is crucial due to its connections with other financial metrics, although it holds less significance when evaluated individually.
Companies with a Z-Score below the cutoff are considered financially distressed, while those above are financially stable, with a lower Z-Score indicating a higher likelihood of default Altman (1968) demonstrated that the MDA model could predict financial failure with 95% accuracy one year ahead, making it a valuable tool for researchers and banks to assess and mitigate credit risk.
Z’-Score and Z’’-Score models for private firms
The original Z-Score Model, developed by Altman in 1983, was specifically designed for publicly traded firms, relying on their market value Recognizing this limitation, Altman proposed a revised approach by substituting the book value of equity for the market value in the model's X4 variable This led to the creation of the updated Z’-Score Model, which maintains the same data parameters while enhancing its applicability to a broader range of firms.
- X3 = Earnings before interest and taxes/Total assets
- X4 = Book value of equity/Book value of total liabilities
Previous empirical studies
Altman et al (2014) conducted a comprehensive review of the Altman Z-Score bankruptcy prediction model, analyzing 33 scientific papers published since 2000 in top financial and accounting journals Their research utilized a diverse international sample of firms from 31 European and three non-European countries to evaluate the model's effectiveness in predicting bankruptcy and distressed firms The study focused on the Z’’-Score Model, originally developed for private manufacturing and non-manufacturing firms Findings revealed that while the international model achieved an average prediction accuracy of approximately 75%, some countries exceeded 90% accuracy Moreover, the classification performance can be significantly enhanced through country-specific estimations and the inclusion of additional variables, which further improve prediction accuracy.
Altman, et al (2007) developed a particular model called to ZChina Score support identification of potential distress firms in China:
- X2 = Net profit/ Average total assets
A study analyzing 120 listed companies on the Shenzhen and Shanghai Stock Exchanges from 1998 to 1999 developed a robust model featuring 15 variables across four key factors: asset liability, working capital, return on total assets, and retained earnings ratios This model closely resembles the Z"-Score four-variable version, known as the Emerging Market Scoring Model, created in 1995 The findings revealed that the model demonstrated high accuracy, achieving 80% forecasting power over three years for firms identified as Special Treatment (ST), indicating their problematic status.
In their 2014 study, Meeampol et al analyzed financial distress among companies listed on the Stock Exchange of Thailand using the Z-score model and the Emerging Market Z-Score model Focusing on firms marked with a Non-Compliance (NC) sign in 2012, the research found that both models effectively predicted potential bankruptcy risks Notably, the accuracy improved when utilizing two years of data instead of one Additionally, the Z-score model demonstrated a better fit for the Thai Stock Market, despite the country's classification as an emerging economy, suggesting a stronger alignment with the Emerging Market Z-Score model.
A study by Thai et al (2014) in ASEAN utilized the MDA approach to forecast financial distress in Malaysia, analyzing a sample of 30 companies, which included 15 experiencing financial distress and 15 that were not, over a specified period.
Between 2009 and 2013, an analysis of the Altman Z-score model was conducted using five financial ratios, which were evaluated through Discriminant Analysis The findings revealed that the working capital-to-total-assets ratio emerged as the most significant variable for distinguishing between different financial outcomes.
23 distressed and non-distressed companies Moreover, discriminant analysis reached an accurate rate of 76.7% in predicting corporate bankruptcy
In a study conducted by Le and Nguyen (2012) in Vietnam, the Z”-score model was utilized to predict financial distress among 293 companies listed on the Ho Chi Minh Stock Exchange (HOSE) The findings revealed that the Z”-score model accurately forecasts firms facing financial distress, achieving a prediction accuracy of 91% for one-year forecasts and 72% for two-year forecasts.
Afik et al (2012) conducted an empirical analysis on the sensitivity of the Merton model's default predictability, utilizing data from the merged CRSP-COMPUSTAT database spanning 1988 to 2008 Their study highlights various model specifications found in the literature, including those commonly used by practitioners However, recent research indicates that these methods yield less accurate estimates compared to simpler alternatives The findings reveal that while the prediction quality is only marginally affected by different default barrier choices, the selection of expected asset returns and asset volatility plays a crucial role Based on these insights, the authors propose improved specifications to enhance the model's accuracy using information from the equity market.
Allen, et al (2013) also had a research to compare the market risk and credit risk of Agriculture and Mining Sectors to other sectors in Indonesia during the period from
From 2005 to 2011, Value at Risk (VaR) and Conditional Value at Risk (CVaR) were employed to assess market risk, while the Drawdown (DD) model and Conditional Distance to Default (CDD) model were used for evaluating credit risk The findings reveal that market risk in both the Agriculture and Mining sectors consistently exceeds that of the overall market, attributed to significant discrepancies between VaR and CVaR Notably, the Agriculture sector exhibits a considerably lower credit risk compared to the market, whereas the Mining sector's credit risk closely aligns with market levels This disparity is largely due to the high equity ratios in these sectors, which help mitigate overall credit risk.
In Vietnam, there are lack of studies that use Market – based models in their empirical studies to measure the risk, especially the probability to default and the
24 distance to default Some studies have employed KMV model as a measure of default risk
Lam and Phan (2009) utilized the Merton-KMV model to assess credit risk by analyzing the effects of three key variables: the maximum ratio of borrowed funds to the market value of mortgage assets, the borrower's choice in utilizing these funds, and the frequency with which the borrower re-mortgages assets purchased with prior borrowed funds.
Le and Le (2013) introduced a novel model for measuring credit risk during both crisis and non-crisis periods by integrating Conditional Value at Risk (CVar) with the Merton-KMV Model (DD) Their study analyzed data from 179 publicly listed firms on the Ho Chi Minh City Stock Exchange (HOSE) and the Ha Noi Stock Exchange (HNX) from 2007 to 2011 The findings indicate that this new approach offers greater consistency in calculating Default Distance (DD) and Probability of Default (PD), while also highlighting the relationship between risk and equity ratios.
Nguyen and Pham (2014) employed the KMV – Merton model to assess the credit risk of corporate clients and potential losses faced by banks Analyzing data from 6,398 firms borrowing from the Joint Stock Commercial Bank for Foreign Trade of Vietnam (Vietcombank) between 2008 and 2013, the study revealed a Default Probability (DP) of 2.6% for Vietcombank’s loan portfolio, indicating potential losses of 6.319 billion Vietnam Dong Notably, Small and Medium Enterprises (SMEs) exhibited a lower DP compared to larger firms.
2.4.3 Comparison of accounting – based and market – based models
Various studies have been conducted to provide a comparison of the performance of bankruptcy prediction between accounting-based and market-based models
Hillegeist et al (2004) analyzed a substantial dataset of 65,960 firm-year observations, including 516 bankruptcies from 1979 to 1997, to evaluate the effectiveness of two prominent accounting-based measures, the Z-Score and the O-Score, alongside a market-based measure derived from option-pricing theories (Black and Scholes, Merton - BSM) Their findings indicate that the market-based measure demonstrates greater explanatory power compared to the accounting-based metrics.
25 of the two Scores, even when the Scores are decomposed to reflect industry differences or annual changes
Vassalou and Xing (2004) applied the Merton (1974) option pricing model to assess monthly default likelihood indicators for individual firms and their impact on equity returns, using annual financial data from COMPUSTAT spanning 1971 to 1999 Their findings reveal that equally-weighted portfolios of stocks with high default probabilities yield significantly higher returns compared to those with low default probabilities Additionally, they highlight that the Merton model outperforms accounting models like Altman’s (1968) Z-score and Ohlson’s (1980) model in forecasting default risk Unlike accounting-based models, which suggest similar default probabilities for firms with comparable financial ratios, Merton’s model indicates that firms with identical debt and equity levels can exhibit different default probabilities due to variations in asset volatility This underscores the importance of asset volatility in determining default risk, as Merton’s model utilizes market value estimates for debt, contrasting with the book value approach of accounting models.
Das et al (2008) examined the performance of accounting-based models, such as the Altman Z-Score and Ohlson O-Score, against market-based models like Merton’s model Analyzing a dataset of 2,860 quarterly CDS spreads, they found that accounting-based distress models perform similarly to market-based structural default models Additionally, a combined model that incorporates both accounting and market information outperforms each model individually These findings indicate that accounting and market-based information are complementary in assessing and pricing distress.
Agarwal and Taffler (2008) performed a comparison of market-based and accounting-based bankruptcy prediction models which cover all non-finance industry
UK firms fully listed on the London Stock Exchange (LSE) at any time during the period 1985-2001, and find that traditional models based on financial ratios are not inferior to
KMV-type, option-based models for credit risk assessment purposes They conclude that “in terms of predictive accuracy, there is little difference between the market-based and accounting models”
Miller (2009) identified that practitioners and researchers frequently employed two distinct approaches to assess the financial health of various companies, extending beyond manufacturing to include non-manufacturing firms in their analyses This expansion of the testing universe led Miller to conclude that the DD model was applicable across a broader range of industries.
“superior ordinal and cardinal bankruptcy prediction power”, and it had a more durable bankruptcy signal, but it generated less constant ratings than the Z-score
METHODOLOGY AND RESEARCH DESIGN
Data
This study analyzes annual financial statement data and daily stock prices from all non-financial firms listed on the Vietnamese stock market, specifically the Hochiminh Stock Exchange (HOSE) and the Hanoi Stock Exchange (HNX) Data will be sourced from publicly accessible websites like www.cophieu68.vn and www.cafef.vn, covering a decade from 2006 to 2016 to capture both crisis and non-crisis periods The focus on non-financial firms is due to their publicly disclosed, easily collectible, and audited data, ensuring reliability, while financial firms exhibit distinct financial structures and accounting practices A total of 631 firms were identified for analysis and categorized into 10 major sectors based on the Global Industry Classification Standard (GICS).
Table 3.1 Sectors and industry groups
Commercial and Professional services Transportation
4 Consumer Discretionary Automobile and Component
Consumer Durables & Apparel Consumer Services
5 Consumer Staples Food & Staples Retailing
Food, Beverage & Tobacco Household & Personal Products
6 Health Care Health Care Equipment & Services
7 Information Technology Technology Hardware & Equipment
Source: Global Industry Classification Standard (GICS)
Table 3.2: List of listed firm by sectors
Year/Sectors M CS CD I HC U TS E IT RE Summary
Table 3.2 indicates that the number of firms in the Industrials Sector has consistently increased over the years, representing a significant portion of the market Following this sector are Materials and Consumer Discretionary, which also show growth In contrast, the Information Technology and Telecommunications Services sectors account for the smallest share of firms.
To determine crisis and non-crisis periods of data, the study based on history of the Annual GDP Growth of Vietnam:
Figure 3.1 Annual Vietnam's GDP growth
The chart illustrates that Vietnam's Annual GDP Growth experienced a decline due to the Global Financial Crisis (GFC), starting in 2008 and continuing until 2012 Consequently, the author has selected this crisis period, specifically from 2008 to 2012, for the analysis.
1 https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=VN
GDP GROWTH OF VIET NAM
Selection of models
Two models have been selected in this study: (i) the ZChina-Score model developed by Altman (2007) which is an accounting-based model; and (ii) the KMV
DD model which is market-based model Both models have been around for decades and adopted in many studies as mentioned in literature review part
The ZChina-Score model which Altman developed in 2007 is very similar to the Z’’-score four-variable version – Emerging Market Scoring Model – developed in
The KMV model, introduced in 1995, is applicable to various types of companies and has proven effective in the Chinese stock market since 2007, making it relevant for the current Vietnamese stock market's market capitalization, number of listed firms, and transaction size Its simplicity makes it an attractive choice for analysis, and it raises the question of whether this market-based model can outperform the classic z-score model (Daniel, 2010) Both models will be discussed in detail below.
3.2.1 Accounting-based model (Z China -Score model)
The ZChina-Score model utilizes Multiple Discriminant Analysis (MDA), a statistical technique designed to classify observations into distinct groups based on individual characteristics MDA aims to model a quantitative dependent variable as a linear combination of independent variables, ultimately predicting a qualitative variable from these predictors Typically, the dependent variable consists of two classifications, and MDA formulates an equation that best discriminates between these groups, known as the discriminant function The model assigns weights to each independent variable, accounting for their interrelationships, to enhance predictive accuracy.
33 to as discriminant coefficients The discriminant equation can be expressed as follows:
F : a latent function by the linear combination of the dependent variable
𝛽 𝑖 : a discriminant weight (coefficient) for particular independent variables
𝑛 = number of independent variables ε : the error term
First, this study apply the formula of The ZChina-Score model to calculate the Z- Score The higher the Z-Score, the more secure company is The formula as follow:
- X2 = Net profit/ Average total assets
Then, based on the discriminant equation above, this study is to adopt the following regression equation:
X 2 X 3 : the independent variable β0: a constant of the function
34 β 1 β 2 β 3 : the coefficient of independent variable t for the 𝒕 𝒕𝒉 time period ε: the error term
The Interest Coverage Ratio (ICR) will be utilized as a proxy for financial distress in this study, serving as the dependent variable This approach aligns with previous research conducted by Faelten and Vikova (2014), Asquith (1994), and Rajan et al (1995) The ICR is calculated by dividing Earnings Before Interest, Tax, Depreciation, and Amortization (EBITDA) by Net Interest Expense.
ICR = (Earnings Before Interest,Tax,Depreciation,and Amortization
In 2007, Altman identified four independent variables for the ZChina-Score, which will be utilized in this study The author plans to employ a panel data regression model using Stata, focusing on three common estimation methods: Pooled Ordinary Least Squares (OLS), Fixed Effects Model (FEM), and Random Effects Model (REM) To determine the most suitable panel model, the research will utilize specific tests to identify the optimal approach.
The interest coverage ratio (ICR) assesses a company's capacity to fulfill its interest obligations on debt, indicating the balance between interest expenses and profits A higher ICR signifies a lower debt burden, suggesting better financial health (Altman, 2007) The firm's ability to manage interest payments is primarily influenced by its profitability and overall debt levels.
In case the interest cover rate is