Analysis of financial risk in telecommunications companies listed on Vietnam's stock market

analysis. They were applied to assess the impact of independent variables on the main dependent variable FRit in Bathory model. The results showed that financial risk has a negative correlation with solvency, profitability and capital structure. Financial risk also has no clear linear correlation with debt structure and management ability in small and medium-sized enterprises in China as well as in India, (Gang & Dan, 2012), (Bhunia & Mukhuti, 2012). Studies reviewed five groups of factors affecting financial risk including debt structure, solvency, profitability, operational performance and capital structure. Table 2.1. Defintion of the model’s variables Variable Acronyms Related definition Financial risk FR Bathory Data from Bathory model Debt structure X1 Debt structure Solvency X2 Short-term solvency X3 Fast solvency X4 General solvency Profitabilty X5 Profitability of sales X6 Profitability of total assets Operational ability X7 Inventory turnover X8 Fixed-asset turnover X9 Total assets turnover X10 Receivables turnover ratio Capital structure X11 Self-finance ratio X12 Fixed assets investment ratio Source: Gang & Dan, 2012 Research in Kenya with data collected in 2012 applied and modified the Bathory model with new scales. The new model is presented as follows: FR = β0 + β1(LEV) + β2(ACCESS) + β3(CAPS) + β4(COSC) + β5(PRUD) + α In which: 9 FR: Financial risk; LEV: Leverage; PRUD: Prudent; ACCESS: Ability to access financial information; CAPS: Capital structure; COSC: Capital cost; β: coefficient of the model; βo: constants; α: Random error The results demonstrated that leverage affects financial risk of companies listed on NSE more positively than financial information, capital cost, capital structure and prudential supervision as shown in non-standard beta coefficient. Access to financial information may adversely affect tcompanies listed on NSE and the degree is not the same as that of leverage, capital cost, capital structure and prudent regulations positively impact on financial risk (Okelo, 2015). 2.3.2.2. Logit model Some studies suggested that in reality, logit models are able to provide more effective assessment of bankruptcy risk than the MDA model. Some Logit models can be listed: - Logit model of Ohlson, J: Ohlson (1980) is said to be the first to develop a model that uses the Logit Regression to build a probabilistic model to predict bankruptcy. Ohlson conducted research from 1970 to 1976 with a sample of 2,163 companies, of which 105 were bankrupt and 2,058 were not bankrupt. The study achieved an accuracy of 96.12% for the period of one year, 95.55% or the period of two years and 92.84% or the period of three years. Four indicators are statistically significant when considering the relationship with bankruptcy or the period of a year: net income/total assets, size, financial structure (total debt/total assets) and capital/total assets. In 2002, Kolari applied the logit model to evaluate factors affecting the breakdown of the US banking system in the 80s. He collected a sample of 1,000 non-bankrupt banks and 55 bankrupt banks between 1989 and 1992. Results showed that the bankruptcy of US banks was mainly due to following factors: net income/total assets, ROA, total equity total assets, interest rate/total assets (Kolari, 2002). - Logit model by Haydarshina G.A.: Haydarshina (2008) applied the Logit model in bankruptcy prediction for companies in energy and fuel industry and service industry. Nguyen Thi Nga (2018) applied the logit model in bankruptcy risk analysis at real estate companies listed on Vietnam's stock market. The author studied 14 variables affecting bankruptcy risk and divided them into five groups which are solvency, profitability, financial leverage, stability and operational capacity. The research was conducted from 2008 to 2015 with 45 real estate companies listed on HNX and HOSE. Research results showed that solvency, ROA have a negative correlation with bankruptcy risk and financial leverage has a positive correlation with bankruptcy risk. The application of MDA or Logit model in financial risk analysis is not carried out as much often as the application of models related to bankruptcy risk analysis. Many studies used both models simultaneously in analyzing risks and had different conclusions. Pongsatat et al (2004) applied the model of Altman and Ohlson in analyzing bankruptcy risk of 120 large and small companies in Thailand, of which there are 60 bankrupt companies and 60 non-bankrupt companies from 1988 to 2003. Findings showed that Altman model had higher rate of accuracy when predicting bankruptcy than Ohlson model. Research by Ugurlu & Aksoy (2006) had the opposite results. Ohlson logit model had higher rate of accuracy than Altman MDA model. The study was conducted from 1996 to 2003 in Turkey with 27 bankrupt and 27 non-bankrupt companies. This study also added the economic environment factor and drew important conclusion stating that the economic environment is unstable as well as errors in management increase the risk of bankruptcy. Xu & Zhang (2009) studied the bankruptcy of companies listed on the Japanese stock market from 1992 to 2005. The authors conducted analysis according to Altman's MDA and Ohlson's Logit model. They incorporated both models and found that the accuracy rate of prediction increased when applied both models with the support of both the banking system and major business sectors in the Japanese economy. 10 2.3.3. Research on control of financial risk Studies on financial risk control were conducted by Beasley et al. (2006), Kleffiner et al. (2003), Hoyt et al. (2011). These studies all evaluated the function of Chief Risk Officer and consider this as an important factor in the decision to implement risk management in enterprises. Another research with a survey sample of 89 enterprises in Malaysia carried out bt Pagach et al. (2010) showed that Chief Risk Officer (CRO) is an important factor for businesses to accept and perform risk management. According to a survey by Deloitte (2014) with the participation of over 192 non- financial enterprises in the US, very few non-financial firms have CROr. Meanwhile, 66% of surveyed companied said that financial risk has been increasing recently, proving that CFOs face more difficulties in their jobs when they have to take more responsibilities for risk management. Research by Paulin (2015) shared similar results. The use of derivative financial instruments in prevention and control of financial risk is mainly deployed by large enterprises, not small and medium enterprises due to the relatively high costs. Research by Hanschel (2008) suggested that small and medium enterprises should apply other risk prevention methods that are more simple, appropriate and effective. He recommended these company use insurance contracts. Allayannis & Weston (2001) surveyed the use of derivative financial instruments in exchange rate risk management in 720 large non-financial enterprises in the US from 1990 to 1995. The results confirmed the positive relationship between exchange rate risk management and the value of companies. Allayannis also conducted another study with his colleagues in 2004. Their research was to examine the relationship of exchange rate risk management via derivative tools and enterprises' value. The sample consisted of 279 US companies and data was collected in the period of 10 years, from 1990 to 1999. Findings mentioned the positive effect when using derivative tools in risk management and companies' value (Allayannis et al 2004). Kim et al. (2004) studied 424 enterprises from 1996 to 2000. Their research aimed at examining the impact of financial risk management and operational risk on the change in companies' value. The study found that risk management increases the value of a firm. Nain (2004) studied exchange rate risk management in companies using derivatives and those do not. Using quantitative methods, the survey sample consists of 548 enterprises using derivative tools and 2,711 enterprises not using derivative tools in risk management. The data collection period is three years, from 1997 to 1999. Results showed that enterprises applying derivative tools in financial risk management will increase their value (Tobin's Q measure) compared with their competitors. Enterprises that do not apply derivative instruments in risk management will not increase their valuecompared with their competitors (Tobin’s Q ratio is lower than other opponents). Carter et al. (2004) reviewed risk management of gasoline prices in 26 US airlines. The study was conducted from 1994 to 2000. Findings confirmed that airlines applying derivative tools in managing the risk of price fluctuations will increase their own value. Callahan (2002 deployed an empirical study of 20 enterprises in the gold mining industry in North America on gold price risk management. The period was five years from 1996 to 2000. Results suggested the risk management and stock prices have a negative relationship. The study by Loolman (2004) in 125 US oil and gas manufacturing companies with two survey periods, from 1992 to 1994, and from 1999 to 2000. The research also shared similar findings with that of Callahan (2002) as it concluded that the value of enterprises decreases when companies do not diversify goods and apply risk management. On the other hand, in businesses diversifying goods and applying risk management, their value will increase. 11 Nevertheless, according to Jin & Jorrion (2004), corporate value has no relationship with risk management. The authors studied risk management of 119 US petroleum enterprises in the period of 1998-2001. In 1958, M&M theory mentioned financial risk through research of corporate loans. Researchers found that the value of a debt-borrowing enterprise is equal to the value of a debt- free enterprise in the absence of tax. By 1663, the theory was added by the researchers with the case of enterprises being subjected to tax and the value of debt-borrowing companies increased by an amount equal to the tax shield. However, firms who use too much debt will have a large risk of financial distress. Trinh Thi Phan Lan (2016), in her thesis named "Financial risk management in non- financial enterprises listed on Vietnam's stock market, had used quantitative and qualitative methods. Research in 158 enterprises in the period of 2010 - 2014 showed that the control of financial risk in companies was not paid full attention. Concurrently, the author pointed out the positive correlation of risk management on the value of companies and offered recommendations based on her findings. Vu Thi Hau (2013) also mentioned the control of financial risk through the use of derivatives, insurance contracts and reserve funds. Interviews with 21 industrial enterprises listed on Vietnam's stock market illustrated that: 28.57% have not used forward contracts to controlling exchange rate risk, 23.81% have never heard of interest rates and 61.9% of enterprises have heard but never used interest rate contracts in controlling financial risk. Nguyen Minh Kieu (2014) presented the financial risk management tool and derivative financial instruments in her studies. The author had proposed solutions to manage each type of financial risk by using derivative tools in each scenario. Management solutions are clearly separated when applied to two groups of subjects including enterprises and banks. 2.4 Establishment of research topic The above overview showed that most previous studies in developed and developing countries used MDA or Logit model in measuring bankruptcy risk as well as financial risk which include Z, Z' and Z" model by Altman; Fulmer's H model; Bathit's FRit model, Ohlson's Logit model, Haydarshina's model In these researches, the nature of factors' influence on financial risk has not been consistent depending on many issues such as research conducted in developed or developing countries, the stage of economic development in each nation, specific characteristics of each industry.... In Vietnam, previous research only focused on Z and its adjusted models in financial risk analysis in companies. Nhu Vu Thi Hau (2013) applied Z model Z in her study on bankruptcy risk in Vietnamese enterprises. Trinh Thi Phan Lan (2016) applied the Z" model in assessing the risk of bankruptcy in Vietnamese firms from 2010 to the end of 2014. Her research studied the following groups: real estate, construction, transport, industry, agriculture - forestry - fishery. In addition, the new Z model only mentions groups of factors that affect bankruptcy risk such as solvency, profitability, financial leverage and operational performance. It has yet to evaluate other financial and non-financial factors such as debt structure, financial structure, interest rates, size of the enterprises, years of operation and specifically, telecommunications companies. Fulmer H.'s H model is a bankruptcy classification model applied to small businesses with five groups of factors: profitability, solvency, operational performance, capital structure and size of operation. The H model is mainly applied in European countries but not in Asia and Vietnam. Concurrently, the model does not present the influence of factors such as debt structure, asset structure, years of operation on the degree of bankruptcy. Model H has not yet been considered to be applied for a specific industry group such as telecommunications. Financial risk analysis model by Alexander Bathory was applied in China in the study by Gang & Dan (2012), in India in the study by Bhunia & Mukhuti (2012). Findings from two 12 research indicated that financial risk has a negative correlation with short-term solvency, profitability of net revenue and fixed assets coefficient. Additional, financial risk does not have a clear linear correlation with debt structure and management ability in small and medium enterprises. Research by Vu Thi Hau (2017) also demonstrated that financial risk correlates negatively with solvency and capital structure. Financial risk does not have connections with debt structure, operational performance and profitability in listed real estate companies. Influence of control variables synthesized from previous studies including interest rate, years of operation, company size on financial risk is found to be inconsistent. The reason may be due to characteristics of each country's economy as well as each economic sector. However, there has not been a research thatapply the above models and control variables in financial risk analysis in telecommunications companies. In Vietnam, research on financial risk as well as factors affecting financial risk, especially in a specific industry are very few. Therefore, the fellow decided to carry out the study of "Analysis of financial risk in telecommunications companies listed on Vietnam's stock market". Data of indicators used for analysis in the research model was collected from financial reports of telecommunications companies. The indicators demonstrating the debt structure in previous studies were short-term debt over long-term debt did not reflect the true nature of debt structure. In the thesis, the author used short-term debt indicator/total liabilities as a variable to reflect debt structure. In addition, the fellow absored and added a number of control variables such as interest rates, years of operation, company size to evaluate the impact of these variables on inancial risk in telecommunications companies. Other studies related to financial risk have a relatively short research period and does not fully presentthe economic fluctuations. Therefore, the research period of this thesis is seven years from 2010 to 2016 which is quite a long period compared to previous studies in domestic and international scale. Chapter 3: METHODOLOGY 3.1. Data collection and processing 3.1.1. Selection of research sample - Hanoi Stock Exchange (HNX): In the first stage, based on HNX's industry classification criteria, the research filtered and selected listed companies belonging to the category-I industry group which is Information and Communications. Next, the author reviewed the category I, selected telecommunications companies after excluding the rest. Then, the fellow examined selected telecoms companies, if they have completer financial statements from 2010 to the end of 2016, they will be listed on the sample list. Through this procedure, six out of seven telecommunications enterprises listed on HNX were chosen as they have been operating continuously from 2010 to 2016. The data in the study was collected according to each criteria in financial statements of telecommunications companies within seven years from 2010 to 2016. Thus, there were 42 observations collected from HNX. - Ho Chi Minh Stock Exchange Firstly, based on the HOSE's classification criteria, the author identified listed telecommunications companies that belong to category-I industry group: Information Technology. The next procedure was performed at HNX and HOSE and the fellow obtained a sample of 42 observations. After summarizing, the thesis achieved 84 observations with high transparency. The ratio betwen observations and overall sample reached 85.71% - this is a fairly high ratio. 3.1.2. Data processing The data collected was conducted in the Excel office software for the research model. Next, the author used Stata14 software to process data and run research models. 13 3.2. Research model and theories 3.2.1. Research model Figure 3.4 Financial risk analysis model in telecommunications companies Source: The author summarized based on research results. Variables in the model are presented in Table 3.1. Table 3.1. Description of variables in the model No Group of variables Varia- bles Formula References Dependent various 01 Dependent variables FRit Value of financial risk Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Vu Thi Hau (2013). Independent variables 01 Debt Structure DS Short-time debts Liabilitíe Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Vu Thi Hau (2013). 02 Solvency CR Short-time assets Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Short-term debts 03 QR Short-term assets Inventory Short-time debts 04 IGS Total assets Total liabilities 05 Profitabilit -y ROS Profit after tax Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Vu Thi Hau (2013). Net revenue 06 ROA Profit after tax Average total assets 07 Operatio- nal Performan- ce IT Cost of goods sold Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Vu Thi Hau (2013). Average inventory 08 FAT Net revenue Average fixed assets 09 TAT Net revenue Average assets FRit Debt Structure- DS Control Variables - SIZE - IR - AGE Financial Structure - ES - FASProfitability - ROS - ROA Operational Performance - IT - FAT - TAT - RT Solvency - CR - QR - IGS 14 No Group of variables Varia- bles Formula References 10 RT Net revenue Average receivables 11 Financial structure ES Equity Gang & Liu Dan (2012), Bhunia et al (2012), Fu et al (2012), Okelo (2015), Gunarathna (2016), Vu Thi Hau (2013). Total capital 12 FAS Value of fixed assets Total assets Control variables 01 Interest IR Average loan interest by State Bank Defang & Muli (2005), Vu Thi Hau (2013). 02 Years of company’s operation AGE Calculated from the period when the company went public to the time of research The author inclued it in the model 03 Company size SIZE Ln(Total assets) The author inclued it in the model Source: The author summarized based on research results 3.2.2. Research theories Based on empirical and related theories, the author proposes the following research hypotheses: - H1: Debt structure of telecommunications companies has a positive correlation with financial risk. According to the trade-off theory, when the company's value reaches its optimal point, the optimal capital structure can be determined (ratio of optimal debt ratio and equity). However, if the firm continues to increase their debts, the bankruptcy cost will increase until the bankruptcy cost is greater than benefits of the tax shield, the company's value will start to decrease leading to an increase in financial risk. - H2: Solvency of telecompanies has a negative correlation with financial risk. Solvency shows quite clearly the enterprises' financial situation. Telecommunications companies with effective operations often have healthy financial status and vice versa. When the firm's solvency is not guaranteed - the enterprise cannot pay due debts - then its operation will face challegnes. If insolvency continues and the financial situation is not guaranteed, companies will easily bankrupt. - H3: Profitability of telecommunications has a negative correlation with financial risk. Profitability reflects the capacity of a unit of cost to make profits or the capacity of input or an output to demonsttrate business results. When the operation of telecommunications companies are successful that can increase profits, their profitability will also rises. Concurrently, firms have the opportunity to increase their accumulated profits and equity, improve their solvency, coverdue debts, and reduce financial risk. - H4: Operational performance of telecommunications companies is negatively related to financial risk. Operational performance is telecommunications enterprises' capacity to achieve operational results when consuming the inputs during the operation. There are many performance indicators such as the rotation speed of inputs or payment speed. The growth of turnover or payment speed shows the development in firms' operation and low financial risk and vice versa. - H5: Financial structure of telecommunications companies has an negative correlation with financial risk. Financial structure reflects the asset structure, capital structure, the relationship between assets and the sources of assets. In this thesis, when reviewing the 15 relationship between financial risk and financial structure, the author mainly evaluated the relationship between asset structure, capital structure and financial risk. Capital structure presents the proportion of each capital source in total capital. The fellow used an indicator named "self-financing ratio" to reflect the capital structure of telecommunications enterprises. When the high self-financing ratio is equivalent to the ratio of debt to total low capital, the company's solvency is more easily guaranteed, the creditors will be in a safer position and the company's financial risk will decrease. Asset structure represents the proportion of each asset type in total assets. The thesis used the indicator named "Rate of investment in fixed assets" to reflect the asset structure. The higher the investment rate of fixed assets is, the more guaranteed the creditors' debts will be, financial risk will reduce and vice versa. - H6: Interest rate of telecommunications companies has a positive relationship with financial. According to M&M theory, expected profits (capital cost) of firms not paying corporate income tax or beingsubject to corporate income tax are affected by the debt cost (interest rate). The higher the interest rate is, the lower the expected profit will be and this will affect the accumulated profits and companies' opportunity to increase capital. In addition, the higher the interest rate is, the more negative impact it will have on solvency (both in short-term and long-term), financial risk in telecommunications enterprises will rise and vice versa. - H7: Telecommunications companies' years of operation is negatively related to financial risk. The operating time is calculated from the period when the company went public to the time of research. According to Stinchcome (1965), the longer companies operate in the more experience they obtain in organizing their businesses. Concurrently, when enterprises are eligible to develop their scale, establish their brand and credit, it can serve as the basis for companies to avoid risks and increase acces