Firms that face more agency conflicts will smooth dividends more, such as cash cows, firms with low growth prospects, with weaker corporate governance, and firms controlled by institutional investors.
+ Jeong (2013) used a research sample of 279 Korean listed companies during the period 1981 - 2012 and gave quite interesting research results. Size, risk, growth and large shareholder ownership were found to be important determinants of dividend smoothing behavior of Korean companies. In addition, taxes and interest rates all had positive effects on the level of stability. These findings suggest that institutional factors of financial markets may play an important role in dividend decisions in emerging stock markets. Therefore, theories based on asymmetric information and agency problems are not suitable to explain the dividend smoothing phenomenon in Korea. This may be due to the characteristics of the Korean financial market, where the ownership structure is characterized by cross-ownership, there is no capital gains tax but a 15.4% dividend tax rate, and Korea has had a rapidly growing economy over the past four decades.
+ Javakhadze et al. (2014) used a sample of companies from 24 countries during the period 1999 - 2011, using Lintner's (1956) SOA measure for each company as the dependent variable of the factor model. Also with the goal of verifying the ability to explain the dividend stabilization phenomenon of "signaling theory and agency theory"; at the same time, due to the difference in the level of legal protection for investors, the characteristics of national ownership structures, tax systems and national cultures can affect the explanatory ability of these theories. Therefore, the study approached in 4 groups: (i) Group of factors belonging to company characteristics. The study found that concentrated ownership and corporate governance have an inverse relationship with dividend smoothing. Therefore, dividend stabilization as a substitute for corporate governance, that is, dividend stabilization tends to be associated with weaker shareholder rights. (ii) Competition and information effects. Firms operating in more competitive industries are more likely to smooth dividends in an effort to attract investment. The results support this hypothesis, but do not find evidence of a relationship between dividend smoothing and the information environment. (iv) Legal regime. Firms in countries with weak investor protection are more likely to smooth dividends. (v) National culture. A novel point of this study is to investigate the relationship between national culture and the degree of dividend smoothing, since the dividend policy depends not only on a firm's agency problem but also on the
depends on the subjective assessment of corporate managers and investors on this issue. Therefore, the study suggests that countries with low power distance and protected individual rights tend to have more stable dividends.
Thus, all three studies above aim to test the factors affecting the level of dividend stabilization, the explanations are approached based on the BCXTT problem and the agency problem. The research results have contributed to providing evidence on the phenomenon of dividend stabilization in the Korean and US markets and based on a multinational sample. However, the use of only the SOA measure to measure the level of dividend stabilization in the study of Jeong (2013) and the study of Javakhadze et al. (2014) makes it difficult to compare the research results of these works due to the difference in data exploitation and other spatial or temporal characteristics. Besides, in terms of methodology, if these studies had controlled for endogeneity in the model, the research results would have been more reliable.
Most of the studies used Lintner’s (1956) SOA measure to measure dividend stability. A few recent studies that used both SOA and Relative Volatility measures to measure dividend stability are Ali et al. (2019) to test the influence of ownership and board structure on dividend stability behavior in Pakistan.
Table 2.3: Summary of research on factors affecting the level of dividend stabilization
S
tt
Author (year) | Research sample | Method study | Secondary variable belong | Main independent variable | Main research results | |
1 | Leary and Michaely (2011) | America (1985 – 2005) 1,335 non-financial companies (excluding financial companies, companies participating in M&A, paying dividends < 10 years, companies with missing data > 5) year) | - Simulation problem, - 2-step regression determines influencing factors: + Calculate SOA and RelVol for each company + OLS regression on cross-sectional data for factors. Quantile regression | SOA, RelVol | M/B, AGE, SIZE, TANGIBLE, RISK, INST, PAYOUT, TAX, FCF, ANALYST | - Younger, smaller firms, firms with lower dividend yields, firms with higher earnings volatility, and firms with less accurate forecasts: adjust dividends less (opposite to signaling theory) - Firms with excess cash, low growth prospects, weak corporate governance and firms controlled by institutional shareholders: higher degree of dividend stabilization (in the same direction) with representation theory) |
2 | Javakhadze et al. (2014) | 2,219 non-financial companies from 24 countries (1999-2011) | Two-step regression: - SOA regression by Lintner model - OLS regression on cross-sectional data for each factor group | SOA (RelVol gives similar results) | + M/B, AGE, SIZE, TANGIBLE, RISK, INST, LARGE, PAYOUT, CASH, FCF, TAX + HHI, RISK_IND + National law and culture | - Companies with low M/B ratio and less excess cash: adjust dividends more - Companies with highly concentrated ownership structures and good corporate governance: less likely to adjust dividends - Companies operating in industries facing high levels of competition: tend to have more stable dividends (Consistent with signal theory and theory) representational theory) |
Maybe you are interested!
-
Research on factors affecting the decision to purchase gypsum board of Huy An private trading enterprise from institutional customers in Ho Chi Minh City - 13 -
Research on factors affecting the behavior of using international bank cards in Vietnam - 21 -
Research Contents Identify Factors Affecting Slow Growth In The Study. -
Building a Research Model of Factors Affecting Agribank's Brand Value -
Building a Scale and Research Model of Factors Affecting Customers' Decision to Choose a Bank to Deposit Savings at
S
tt
Author (year) | Research sample | Method study | Secondary variable belong | Main independent variable | Main research results | |
3 | Jeong (2013) | South Korea (1981 – 2012) 279 non-financial companies | Two-step regression: - SOA regression by Lintner model - OLS regression on cross-sectional data for each factor group | SOA | + AGE, SIZE, RISK, LARGE, GROWTH, SLACK + Macro factors: INTEREST, TAX | - Larger and lower growth company: More stable - During the period 1982-2012, riskier firms tended to be more stable in their dividends. Safer firms tended to be more stable in the post-liberalization period (1995-2012) - The more concentrated the ownership structure of a company (large shareholder ownership), the higher the level of stabilization. - Tax and interest rates are correlated with dividend stability (Does not support signaling theory and agency theory) |
4 | Kighir et al. (2015) | Malaysia (1999- 2012) 131 non-financial companies | Applying the model of Andres et al. 2009 modified from the model of Lintner. Using GMM technique (panel data) | SOA | DPS t-1 , EPS t , EPS t-1 , CFO t , CFO t-1 , FCF t , FCF t-1 (study of the relationship between cash flows and dividend changes) | - EPS t is more important than CFO t , CFO t-1 is more important than EPS t-1 in dividend payment decision (Supporting Lintner 1956 model, not supporting Andres et al. 2009 model) - FCF is the key to explaining dividend stabilization phenomenon in Malaysia (Supporting agency theory) |
S
tt
Author (year) | Research sample | Method study | Secondary variable belong | Main independent variable | Main research results | |
5 | Shinozaki and Uchida (2015) | 9968 non-financial companies from 44 countries (2003-2013) | - Applying Lintner model (but TPR calculated according to Leary and Michaely) - OLS regression, FEM selection | SOA | Studying the relationship between CTSH and dividend stabilization: LARGE | - Companies with concentrated ownership structures with controlling shareholders adjust their dividends more quickly (the more concentrated the ownership of a company, the lower the degree of dividend stabilization). - No clear evidence of correlation between dividend stability and ownership structure variables was found. other |
6 | Tran Thi To Uyen (2015) | Vietnam (2009-2013) 144 listed companies on HOSE and HNX, (filtered to 52 companies, 260 observations) | Two-step (OLS) regression - Step 1: Calculate SOA for each company according to Lintner's model - Step 2: Inheriting Jeong's MHNC (2013) | SOA | + AGE, SIZE, RISK, LARGE, GROWTH, SLACK + Macro factor: INTEREST | - There is a correlation between the RISK variable (income volatility) and the level of dividend stabilization. - No clear evidence of correlation between dividend stability and the remaining variables was found. (The small sample and inappropriate research methods make the research results (research is not reliable enough) |
Source: Author's synthesis
CSCT in Vietnam has attracted the attention of many researchers, as shown through works in various genres such as scientific articles, master's theses, doctoral dissertations, and ministerial-level research topics. The authors mainly focus on the following research issues:
(i) Identifying factors affecting the capital structure (Tran Thi Tuan Anh, 2016; Nguyen Thi Minh Hue et al., 2014; Dinh Bao Ngoc and Nguyen Chi Cuong, 2014; Ngo Thi Quyen, 2016). The authors believe that “profit” is the main factor determining the capital structure of listed enterprises in Vietnam, however, the factor “past dividends” is not really consistent in these studies.
(ii) Testing the impact of ownership structure on corporate governance (Vo Xuan Vinh, 2014; Tran Thi Hai Ly and Do Thi Bay, 2015), the results all confirm the existence of a relationship between ownership structure and corporate governance.
(iii) Research on the impact of dividend policy on stock prices and enterprise value. Nguyen Thi Minh Hue (2015) using event research method conducted with a sample of 20 enterprises with the largest capitalization listed on HOSE showed that each announcement of increasing, decreasing or maintaining dividends has an impact on the market value of stocks in a certain period of time.
(iv) Studying the relationship between dividend policy and profitability. Truong Thu Huong (2019) provided evidence of the causal relationship between dividend changes and corporate profitability.
Although these studies are not related to the research topic of the thesis, the research results have raised the author's doubts about CSCT in Vietnam.
- Is the current dividend policy in Vietnam still governed by State regulations (Vu Van Ninh, 2008) or is it heavily dependent on business results, in the style of high profits being distributed more, low profits being distributed less or not distributed at all (Nguyen Minh Kieu, 2012)? Are past dividends a factor affecting the company's dividend decision for the following year?
- Dividend announcements or dividend increases or decreases all have an impact on stock prices and corporate value (Nguyen Thi Minh Hue, 2015). So is signaling theory suitable to explain CSCT in Vietnam?
- There is a relationship between ownership structure and dividend decisions, ownership structure potentially creates conflicts of interest among stakeholders. So is agency theory suitable to explain CSCT in Vietnam?
The first research work that the author learned about with a research topic close to the topic of the thesis is the master's thesis of Tran Thi To Uyen (2015). With
With a research sample of 52 enterprises over a period of 5 years (2009-2013), the thesis uses Lintner's model to calculate the stability level and applies Jinho's model (2013) to test the influencing factors. With a short research time, inappropriate research methods and no standard theoretical framework for the research problem, the research results are not reliable enough. Therefore, a quantitative study for a large sample over a long period of time with a rigorous research method will give more accurate conclusions.
2.3. Research gap
From the overview of the above research works, the following gaps can be pointed out:
First, about the research context. Most of the research on “dividend smoothing” is done at DFMs and a few at EFMs. In Vietnam, according to the author’s overview, this is still an open research issue.
With the characteristics of a small-scale, low-liquidity, and limited market access (such as: market openness and equal rights with foreign investors, limited information disclosure, limited information infrastructure, etc.), Vietnam's stock market is currently classified by international rating organizations as a frontier market awaiting upgrade to an emerging market. Based on differences in institutions, legal constraints, macroeconomic environments, and corporate governance, the agency issues and potential BCXTT issues for businesses in these markets will be different. The use of agency theory and signaling theory to explain the phenomenon of dividend stabilization in Vietnam is necessary to provide a new perspective in the new context.
In addition, there are two main measures of dividend stabilization: the speed of adjustment (SOA) proposed by Litner (1956) and relative volatility (RelVol) proposed by Leary and Michaely (2011). RelVol is a measure with many advantages that the SOA measure does not have, such as: Avoiding bias when the sample is small, can be performed with short-term sample sizes and limited observations. However, most previous studies use the traditional SOA measure to measure the adjustment level. Therefore, validating the RelVol measure in another context, with a more modest data set, is also a research gap. Filling this gap by comparing and contrasting the use of RelVol and SOA measures would also enrich the research context on dividend stabilization, and could also serve as a basis for
Other studies may use this measure as an alternative to measure the degree of dividend stabilization.
Second, on the influence of ownership structure . Most previous studies have acknowledged the influence of ownership structure on the dividend stabilization behavior of enterprises, however, these studies mainly refer to the role of ownership concentration and institutional ownership. Meanwhile, the ownership structure in Vietnam is characterized by concentrated ownership and State ownership. The listed enterprises which were previously State-owned enterprises after equitization, the State owns controlling shares or holds a certain percentage of ownership in the enterprises. With this characteristic, in addition to the variables of concentrated ownership and institutional ownership, it is necessary to evaluate the influence of State ownership on the phenomenon of dividend stabilization in Vietnam.
In addition, to test the influence of factors on the level of dividend stabilization, it seems that previous studies have only solved some weaknesses of the model and have not mentioned the endogeneity issue, leading to unconvincing estimation results. Meanwhile, empirical research in the field of corporate financial management, in addition to the risk of autocorrelation due to the indicators being calculated back and forth, also raises the endogeneity issue. Therefore, in addition to the conventional OLS estimation method, it is really necessary to use the GMM method or the IV, 2SLS, 3SLS methods to control this issue.
Thus, the expected research gap discovered when studying the phenomenon of dividend stabilization in Vietnam aims to provide useful information for investors, corporate managers and policy makers. Moreover, the validation of the new RelVol measure is also very useful to confirm that a new measure is suitable for the situation of small research samples and short observation periods - which are very common in data problems in developing countries.